TREATISE ON GENERAL AND INDUSTRIAL ORGANIC CHEMISTRY ALSO BY DR. MOLINARI TREATISE ON GENERAL AND INDUSTRIAL INORGANIC CHEMISTRY TRANSLATED BY DR. ERNEST FEILMANN B.Sc., Ph.D., F.I.C. With 280 Text-Figures and 3 Plates TREATISE ON GENERAL AND INDUSTRIAL ORGANIC CHEMISTRY BY DR. ETTORE MOLINARI Professor of Industrial Chemistry to the Society for the Encouragement of Arts and Manufactures and of Merceology at the Luigi Bocconi Commercial University, Milan TRANSLATED FROM THE SECOND ENLARGED AND REVISED ITALIAN EDITION BY THOMAS H. POPE B.Sc., A.C.G.I., F.I.C. School of Malting and Brewing, University of Birmingham WITH 506 ILLUSTRATIONS PHILADELPHIA P. BLAKISTON'S SON & CO. 1012 WALNUT STREET 1913 T P I Ml**. MM O Printed in Great Britain ' * . .* - -> TRANSLATOR S PREFACE FOR the purposes of this English translation of his " Trattata di Chimica Organica," the author has made a number of alterations in and additions to the text of the second Italian edition, these consisting principally in ampli- fications of the statistical data referring to Great Britain and the United States. It has been deemed undesirable to convert the metric weights and measures into those of the English system, but, in general, prices are given in British currency, twenty-five lire being taken as the equivalent of one pound sterling. Where quantities are given in tons, the latter are to be read as metric tons of 1000 kilograms or 2204' 6 Ib. avoirdupois. The abbreviations employed for the different units of weights and measures are those in common use, and temperatures are expressed in degrees Centigrade in all cases. THOMAS H. POPE BIRMINGHAM 267544 PREFACE TO THE SECOND ITALIAN EDITION THE first edition of this treatise on Organic Chemistry was published in two volumes in 1908 and 1909, and rapidly exhausted, the second edition being now published in one volume. The distribution of the matter is similar to that of the first edition, but many chapters have been brought up to date, others have been considerably amplified and others again have been introduced for the first time. The largest additions have been made in the chapters dealing with the treatment of tar, with colouring-matters, with alkaloids, &c. The statistics of production, exportation, and importation have been brought up to the year 1910 and, where possible, to 1911. Special attention has been devoted to this characteristic feature of the book, as experience has shown the author that among the most important factors in deciding the possibility or convenience of starting new or of extending existing industries are those governed by the laws of economics and statistics. The author will be grateful to any readers or colleagues who may point out omissions or errors, which are unavoidable in a work of this character with such varied contents in so condensed a form. This second edition is in course of translation into English and German. E. MOLINARI MILAN PREFACE TO THE FIRST ITALIAN EDITION A NEW treatise on Organic Chemistry might, in view of the existence of the excellent works of Bernthsen and Holleman, be considered superfluous. But both of these books, which differ little in the manner in which the subject is developed, are confined to a theoretical and systematic exposition of the many organic compounds, the industrial side of the question and the applications of these compounds being almost entirely neglected. It is hence difficult for the student to ascertain which of the thousands of substances described are really of practical importance. Modern teaching of chemistry adheres in a too one-sided manner to the old but fruitful idea of Liebig, that " to obtain a sound practical man it is necessary to train a good theorist." This conception was taken too literally, although it gave good results when chemical industry was in its infancy, since in those days any theorist could easily introduce new and important methods. But to-day, when the industry has attained the adult stage has advanced to such an extent and become so varied and complex, being stimulated incessantly by keen national and international competition, which demands rapid changes and improvements the valuable time of the young technician cannot be wasted in a protracted and sometimes sterile apprenticeship. Present- day conditions require, therefore, some such expansion of Liebig's maxim as the following : In order to produce, rapidly and with increased certainty, a sound, practical man, it is necessary to train a good theorist and to initiate him into both the theoretical and practical study of the more salient industrial problems. It does not suffice that the young chemist, about to begin his industrial or teaching career, should have a thorough knowledge, for instance, of the various syntheses and constitutional formulae of the sugars. He should also be acquainted with at least the general outlines of the industrial processes and of the technique of the manufacture of sugar, beginning with the slicing of the beets and proceeding to the exhaustion of the pulp, defecation, saturation, filtration with filter- presses, boiling, and vacuum concentration in multiple- effect apparatus, refining and centrifugation of sugar crystals, utilisation of residues, and so on. He should, indeed, understand the plant and chemical processes of the more important industries, as these often find application in the manufacture of products of a secondary or entirely new character. What would avail a study of the wonderful artificial colouring-matters derived from coal-tar, with the inexhaustible syntheses composing their theo- retical basis, if it were limited to a simple mnemonic exercise for the student and no notice were taken of the interesting practical applications to the dyeing of the various textile fibres ? Nor should the young student ignore statistics of production ; he should be able to appreciate the importance of variations in the exportation and importation of the principal chemical products, and to judge of the economic and social conditions with which such variations correspond. After a brief novitiate, he should be in a position to point out the more striking technical defects and the more marked difficulties met with in par- ticular industrial processes and to suggest rational and not fanciful remedies viii PREFACE It is this space, the vacant region representing a suitable fusion of theoretical with applied chemistry, which requires filling. This I have attempted in the present work, which of itself is certainly insufficient to cover the whole of the ground. The difficulties encountered in preparing the volume on Inorganic Chemistry are multiplied in dealing with Organic Chemistry, and this is the case not only as regards the collection and confirmation of the statistical data but of the chemical processes giving the best results in practice. For in any particular industry it has often been found that the results of investigations are in such disaccord with the practical data as to render it a matter of great uncertainty what conclusions should be presented to the reader. Inquiries addressed to manufacturers resulted in aggravation of this uncertainty, what was confirmed on the one hand being denied on the other, and plant guaranteed by one firm to be the best being decried by a competing firm. It hence became necessary to apply directly to the operatives working a given process and to draw conclusions from the whole of the data and information thus obtained. It is thus that readers may explain the contradictions between different authorities on one and the same subject, and also the fact that the conclusions reached by the author with reference to certain industrial processes are not always in accord with those given in other treatises. The intention has certainly not been to prepare a complete treatise on technological chemistry and still less on chemical technology. The work having to be restricted within limits of space approximating to those of vol. i, the author has descended to details only with some of the principal industries and especially with those best adapted to give a general idea of the different applications of chemical processes and of chemical technics. To this end the author has dwelt preferably on the industries of illuminating gas, sugar, alcohol, beer, acetic acid, dyeing, textile fibres, fats and soaps, explosives, &c. From these examples the student may gather much instruction applicable to many other industries not dealt with in detail. Repetition has been avoided and time and space saved by frequent references to arguments already developed in vol. i, " Inorganic Chemistry." Advice and collaboration are desired from readers and colleagues in order that gaps in the present work may be filled and inaccuracies and defects remedied. E. MOLINARI MILAN CONTENTS PART I. GENERAL PAOE PURIFICATION OF ORGANIC COMPOUNDS 2 Crystallisation, 2 ; sublimation, boiling-point, fractional distillation, 2 ; rectification, 3 ; melting-point, 5 ; specific gravity, 6. ANALYSIS OF ORGANIC COMPOUNDS 6 Qualitative composition, 6 ; quantitative estimation : of carbon and hydrogen, 7 ; of nitrogen, 10 ; of halogens, 11; of sulphur and phos- phorus, 12. CALCULATION OF EMPIRICAL FORMULA 12 DETERMINATION OF MOLECULAR WEIGHT BY CHEMICAL MEANS 13 POLYMERISM 14 VALENCY OF CARBON, CONSTITUTIONAL FORMULA, ISOMERISM 14 Theory of radicals and types, 14 ; structural formulae, 16 ; rational formulae, 17. METAMERISM, PSEUDOISOMERISM, TAUTOMERISM, DESMOTROPY 17 STEREOISOMERISM OR SPACE ISOMERISM 18 Stereoisomerism in derivatives with doubly linked carbon (alloisomerism), 21 ; Stereoisomerism of nitrogen, 22 ; separation and transformation of stereoisomerides, 22. HOMOLOGY AND ISOLOGY 23 PHYSICAL PROPERTIES OF ORGANIC COMPOUNDS IN RELATION TO THE CHEMICAL COMPOSITION AND CONSTITUTION 23 Crystalline form, 24 ; solubility, 24 ; specific gravity, 24 ; molecular volume, 24 ; melting-point, 24 ; boiling-point, 24 ; heat of combustion and of formation, 25 ; heat of neutralisation, 25. Optical Properties : colour, 26 ; refraction, 26 ; influence on polarised light, 26 ; magnetic rotatory power, 27. Electrical conductivity, 27. CLASSIFICATION OF ORGANIC COMPOUNDS 28 OFFICIAL NOMENCLATURE 28 PART II. DERIVATIVES OF METHANE AA. HYDROCARBONS (a) SATURATED HYDROCARBONS 30 Natural formation and general methods of preparation, 30 ; table of satu- rated hydrocarbons, 31 ; Methane, 32 ; properties, preparation, fire-damp, detonating mixtures, industrial preparation, 33-34 ; Ethane, 34 ; Propane, 35 ; Butanes, 35 ; Pentanes, 35 ; Hexanes, 35 ; Higher Hydrocarbons, 36. x CONTENTS PA OK Illuminating Gas Industry : history, 36 ; components, 38 ; pro- perties, 38 ; retorts, 38 ; furnaces, 41 ; hydraulic main, 43 ; washing, 44 ; purification, 46 ; exhausters, 48 ; pressure regulators, 48 ; gasometers, 48 ; pressure regulators, 49 ; gas-meters, 50 ; yield, 51 ; statistics, 52 ; physical and chemical testing of illuminating gas, 53 ; comparison of different sources of light, 57 ; Oil Gas, 57. Petroleum Industry : localities of production, 58 ; hypotheses on the origin of petroleum, 59 ; composition and properties of crude petroleum, 62 ; industrial extraction and working of petroleum, 64 ; distillation, 66 ; chemical purification, 68 ; storage tanks, 69 ; statistics, 70. Treatment of crude benzine, 73. Treatment of Petroleum Residues : A. Mineral lubricating oils, 74 ; requirements and analysis of lubricating oils, 77. B. Vaseline, gelatinised vaseline oil, 80. C. Paraffin wax, 80; different sources: (1) pyropissite, 81 ; tar, 81 ; photogen, 82 ; tar, asphalt, pitch, and bitumen, 83 ; (2) bituminous shale, 83 ; (3) ozokerite and cerasin, 85. (ft) UNSATURATED HYDROCARBONS 87 T. Ethylene Series (alkylenes or olefines), C ra H 2n , 87 ; official nomen- clature, 87 ; constitution, methods of preparation, 88. Ethylene, propylene, butylenes, amylenes, cerotene, and melene, 89-90. II. Hydrocarbons of the Series C w H 2n _ 2 : A. With two double Unkings (diolefines or allenes) : allene, erythrene, isoprene, piperylene, diallyl, conylene, 90. B. With a triple linking (acetylene series) : metallic acetylides, acety- lene, 90-94. III. Hydrocarbons of the Series C n H 2M _ 4 and C n ~H 2 n-6' 94. BB. HALOGEN DERIVATIVES OF HYDROCARBONS Table of the halogen derivatives f)5 I. Halogen Derivatives of Saturated Hydrocarbons : properties, 94 ; preparation, 95-96. Methyl chloride, 96. Methyl iodide, 97. Ethyl chloride, 97. Isopropyl iodide and butyl iodides, 97. Methylene, ethylene, and ethylidene halogen derivatives, 98. Chloroform, 98-100. lodoform, 100. Polychloro-derivatives, 101. II. Halogen Derivatives of Unsaturated Hydrocarbons, 102; allyl chloride, 102. CC. ALCOHOLS I. SATURATED MONOHYDRIC ALCOHOLS 103 Nomenclature, 102. Methods of formation of monohydric alcohols. 104. Table of monohydric saturated alcohols, 105. Methyl Alcohol, 106-108. Ethyl Alcohol, 108. Solid alcohol, 109. Bacteriology, 110. Enzymes, 111. Oxydases, peroxydases, 112. Biogen hypothesis, toxins, liquid crystals, origin of life, 114. Industrial preparation of alcohol : prime materials, 116. Alcoholic fermentation, 121. Yeasts and ferments, 122. Factors facilitating or retarding fermentation, 126. Losses and yields, 128. Table for the calculation of the attenuation of fermented saccharine worts, 130. Dis- tillation of fermented liquids, 132. Rectification of alcohol, 138. Other raw materials for alcohol manufacture, 140. Alcohol from fruit, 141. Alcohol from woody matter, 142. Alcohol from wine, lees, withered grapes, 143. Refining and depuration of spirit, 144. Fusel oil, 144. Alcohol meters, 146. Alcoholometry and tests for alcohol, 146. Windisch's table, 148. Alcoholism and alcohol-free wines, 150. Statistics, 149. Denatured alcohol for industrial purposes, 152. Distillery residues, 153. CONTENTS xi PAGE Alcoholic Beverages : Wine, 155. Marsala, 159. Vermouth, 159. Cider,"] 159. Liqueurs, 159. Fermented milk (kephir, koumiss), 160. Beer, 161 : barley, hops, water, germination, kilning of malt, mashing, Balling's table, 161-168 ; infusion and decoction mashing, 168 ; boiling of the wort with hops, 170 ; fermentation, 171 ; attenuation, 174. The Nathan-Bolze rapid process, 175 ; racking, pitching of casks, 176 ; pasteurisa- tion, 177 ; alcohol-free beer, 178 ; composition of beer, 178 ; analysis of beer, 178; statistics, 179. Higher Alcohols, 180 ; propyl, butyl, amyl, &c., 180. II. UNSATURATED MONOHYDRIC ALCOHOLS : vinyl, allyl, propargyl, &c., 182. III. POLYHYDRIC ALCOHOLS. (A) Dihydric Alcohols or glycols, 182. (B) Tri- hydric alcohols.- glycerol, 183. (C) Tetra- and poly-hydric alcohols: acetyl number, 188. Erythritol, arabitol, mannitol, dulcitol, sorbitol, 189-190. DD. DERIVATIVES OF ALCOHOLS (A) DERIVATIVES OF MONOHYDRIC ALCOHOLS 190 I. Ethers, 190 ; methyl ether, 192 ; ethyl ether : properties, industrial preparation, 192. II. Thioalcohols and Thioethers, 195. III. Alkyl Derivatives of Inorganic Acids, 196: (1) of sulphuric acid, 197 ; (2) of sulphurous acid, 197 ; (3) of nitric acid, 197 ; (4) of nitrous acid, 197 ; (5) nitro-derivatives of hydrocarbons, 197 ; (6) various acids, 198 ; (7) Derivatives of hydrocyanic acid : (A) Nitriles ; (B) Isonitriles, 198-199. IV. Alkyl Nitrogenous Basic Compounds (amines), 200 ; methylamine, dimethylamine, ethylamine, 201 ; triethylamine, 202 ; alkylhydrazines, azoimides, a- and /3-alkylhydroxylamines, diazo- compounds, 202. V. Phosphines, Arsines, and Alkyl-metallic Compounds. Grignard reaction, 202-203. VI. ALDEHYDES AND KETONES 204 (a) Aldehydes : Functions, constitution, chemical properties, 204. Aldoximes, hydrazones, semicarbazones, hydroxamic acid, 206. Form- aldehyde : preparation, properties and analysis, 206. Acetaldehyde, acetal, 208. Higher aldehydes, 209. Chloral and its hydrate, 209. Aldehydes with unsaturated radicals : acrolei'n, crotonaldehyde, citral, &c., 209. (6) Ketones : Properties, preparation, 210. Acetals, sulphonal, thioketones, ketoximes, phenylhydrazones, isonitrosoketones, 210. Acetone, 211 ; mesityl oxide, phorone, butanone, 212. JCetenes, 212. (B) DERIVATIVES OF POLYHYDRIC ALCOHOLS 213 Glycolsulphuric acid, Ethylenecyanohydrin, Ethylene oxide, 213; Taurine, Glycide alcohol, Glycerophosphoric acid, 214. Nitric ethers of glycerol, 215. Explosives : Theory of explosives, 215. Chemical reactions of explosives : heat of explosion, 216 ; temperature of ignition, 217 ; mechanical work of ' explosives, 217 ; pressure of the gases, 218 ; charging density, 218 ; crushers, 219; specific pressure, 219. Velocity of explosion, 219; shattering and progressive explosives, 219 ; velocity of combustion, 219 ; initial shock and course of explosion, 220 ; determination of explosion, 220 ; explosive wave, 221 ; explosion by influence, 221. Classification of explosives, 222. Nitroglycerines, 222. Trinitroglycerine, 223. Manufacture of nitroglycerine, 225. Dynamites, 229 : with inactive bases, 230 ; with active bases, 230. xii CONTENTS PAGE Nitrocellulose, 232. Guncotton : preparation, manipulation, compression, 233-239. Collodion cotton for gelatines, 239. Smokeless powders, 240. Powders with picrate bases, 245. Explosives of the Sprengel type, 245. Safety explosives, 246. Black powder, 248. Various powders, 254. Deto- nators, 255. Mercury fulminate, 255. Caps, cartridges, fuses, 255. Destruc- tion of explosives, 256. Storage and preservation of explosives, 258. Analysis of explosives, 259. Ballistic tests of explosives, 261. Uses, 263. Statistics, 263. EE. ACIDS I. MONOBASIC SATURATED FATTY ACIDS, C n H 2n 2 264 Table, 265. General methods of preparation, 264. Coefficients of affinity, 266. Separation, 267 ; constitution, 268. Formic Acid, 268. Acetic Acid, 270 : Oudemann's table of specific gravity, 271 ; tests and manufacture, 272 ; distillation of wood, 272 ; utilisation of wood-waste, 274 ; pyroligneous acid, 276 ; calcium acetate, 277. Uses, statistics, and price of acetic acid, 279. Manufacture of vinegar, 280. Analysis of vinegar, 284. Salts of Acetic Acid : potassium, sodium, ammonium, calcium, ferrous and ferric acetates, 285 ; neutral and basic aluminium acetates, silver acetate, neutral and basic lead acetates, chromic, stannous, and copper acetates, 286-287. Propionic Acid, 288. Butyric Acids : (1) Normal butyric acid, 288 ; (2) isobutyric acid, 288. Valeric Acids : (1) Normal valeric acid ; (2) isovaleric acid ; (3) ethylmetliyl- acetic acid ; (4) trimethylacetic acid, 288. Higher Acids : Caproic, heptylic, caprylic, nonoic, undecoic, lauric, myristic, 289. Palmitic Acid, 289. Margaric acid, 293. Stearic acid, 290. Cerotic acid, 290. II. MONOBASIC UNSATURATF.D FATTY ACIDS 291 A. OLEIC OR ACRYLIC SERIES : Table, 291. General method of formation, 291 : general properties, 292. Acrylic Acid, C 3 H 4 2 , 294. Crotonic Acids, C 4 H 6 2 : (a) vinylacetic acid 294 ; (bn) solid crotonic acid, 295 ; (bft) liquid crotonic acid, 295 ; (c) methylmethyleneacetic acid, 296. Pentenoic Acids, C 5 H 8 O 2 : (a) angelic acid, 296 ; (b) tiglic acid, 296. Pyroterebic Acid, C 6 H 10 2 , 297. -y-Allylbutyric Acid, C 7 H 12 2 , 297. Teracrylic Acid, C 8 H 14 2 , 297. Citronellic Acid, C 10 H 18 O 2 : rhodinic acid, 298. Undecenoic Acid, C 11 H 20 2 , 298. Hypogseic Acid, C 16 H 30 O 2 , 298. Oleic Acid, C 18 H 34 O 2 , 298 ; Elaidic Acid, 298 ; Iso-oleic Acid, 299 ; A a 0-oleic acid, 299. Erucic Acid, C 22 H 42 2 , 300 ; Brassidic Acid, 300 ; Isoerucic Acid, 300. B. UNSATURATED ACIDS OF THE SERIES C n H 2n _ 4 O 2 300 (a) Acids with a Triple Linking (propiolic series) : Table, 300. Pre- paration, 300 ; properties, 301. Propiolic Acid, C 3 H 2 2 . Tetrolic Acid, C 4 H 4 O 2 . Dehydroundecenoic Acid, C U H 18 2 , 301. Undecolic Acid, 302. Stearolic Acid, C 18 H 32 O 2 . Tariric Acid. Behenolic Acid, C 22 H 40 2 , 302. (b) Acids with two Double Linkings (diolefine series), 302. /3-Vinyl- acrylic Acid, C 6 H 6 O 2 . Sorbinic Acid, C 6 H 8 2 . Diallylacetic Acid, C 8 H 12 2 . Geranic Acid, C 18 H 32 2 . Linolic Acid ; Drying oils, 303. a-Elaeostearic Acid, 304. C. ACIDS WITH THREE DOUBLE LINKINGS, C n H 2w _ 6 2 . Citrylidene- acetic Acid, C 12 H 18 O 2 . Linolenic and Isolinolenic Acids, C 18 H 30 2 . Jecorinic Acid, C 28 H 30 2 , 304. ITI. POLYBASIC FATTY ACIDS 304 A. SATURATED DIBASIC ACIDS, C M H 2w (C0 2 H) 2 , 304; Table, 305; preparation, properties, 305. Oxalic acid, C 2 H 2 4 , 306. Salts of oxalic acid, 307. Malonic Acid, C 3 H 4 O 4 , 308. Table of malonic acid derivatives, 308. CONTENTS xiii PACE Ethyl Malonate, its use in syntheses, 308. Succinic Acid, C 4 H 6 O 4 , 310. Homologous derivatives, 310. Isosuccinic Acid, 311. Pyrotartaric acids, C 4 H 9 O 4 : glutaric acid, pyrotartaric acid, 311. Higher Homologues, 311. ft-Methyladipic and azelaic acids, 311. B. UNSATURATED DIBASIC ACIDS, C w H 2n _ 4 O 2 312 OLEFINEDICARBOXYLIC ACIDS: Table, 312. Fumaric Acid, 313. Maleic Acid, C 4 H 4 4 . Itaconic Acid, C 5 H 6 4 . Mesaconic Acid, C 5 H 6 O 4 . Citraconic Acid, C 5 H 6 4 . Glutaconic Acid, C 5 H 6 O 4 . Pyrocinchonic Acid and Anhydride, C 6 H 8 4 . Korner and Menozzi reaction of amino- acids. Hydromuconic Acid, C 6 H 8 O 4 . Diolefinedicarboxylic Acids. Acetylenedicarboxylic Acids, 313-315. C. TRIBASIC ACIDS, &c. 315 Tricarballylic Acid, C 3 H 5 (COOH) 3 . Camphoronic Acid, C 9 H 14 O 6 . Aconitic Acid, C 6 H 6 6 , 315. D. TETRABASIC ACIDS 31 G FF. DERIVATIVES OF ACIDS I. HALOGEN DERIVATIVES 310 (a) Halogenated Acids, 316. Cyano-acids. Monochloracetic Acid, 317. Table of the halogenated acids, 318. (b) Acid Halides : chloranhydrides ; acetyl chloride ; acetyl iodide, &c., 317-319. II. ANHYDRIDES 319 Properties, preparation, Table, 319-320. Acetic Anhydride, 320. III. HYDROXY-ACIDS 320 A. SATURATED DIVALENT MONOBASIC ACIDS Preparation, properties, constitution ; lactides ; lactones, 321. Glycollic Acid, OH-CH 2 -COOH, and its derivatives (anhydride, glycollide, &c.), 322. Glycocoll, 322. Lactic Acids, C 2 H 4 (OH)(COOH) : (1) i-Ethylidenelactic acid (of fermen- tation), 323 ; Alanine, 325. (2) d-Ethylidenelactic (or sarcolactic) acid. (3) 1-Ethylidenelactic acid. (4) Ethylenelactic acid, 325. Hydroxybutyric Acids, C 3 H 6 (OH)(COOH) : a-Hydroxybutyric acid. a-Hydroxyisobutyric acid. /3-Hydroxybutyric acid, 326. Higher Hydroxy-Acids : Hydroxy valeric, hydroxycaproic, hydroxy- myristic, hydroxypalmitic, hydroxystearic, 326. B. UNSATURATED MONOBASIC HYDROXY-ACIDS 32(5 a-, ft-, y, and 8-Hydroxyolefinecarboxylic acids : Ricinoleic acid ; ricino- leinsulphonic acid and Turkey-red oil (sulphoricinate), 326-328. C. POLYVALENT MONOBASIC HYDROXY-ACIDS 328 Glyceric Acid, 2 H 3 (OH) 2 (COOH). Dihydroxystearic acid, C 17 H 33 (OH) 2 - COOH, Erythric Acid, C 3 H 4 (OH) 3 -COOH. Penfonic acids. Arabonic Acid. Hexonic Acids, 328. Heptonic Acids, 329. D. MONOBASIC ALDEHYDIC ACIDS (Aldehydic Alcohols and Dialdehydes) 329 Glyoxylic Acid C0 2 H -CHO. Glycuronic, Forrnylacetic, and /3-Hydroxy- acrylic Acids, 329. xiv CONTENTS PAGE Glycollic Aldehyde, OH -CH 2 -CHO. Glyceraldehyde. Aldol. Glyoxal, 321). E. MONOBASIC KETONIC ACIDS (Keto-alcohols, Diketones, and Keto-aldehydes) 330 General properties. Methods of preparation, a-, ft-, and y-Ketonic acids. Syntheses with ethyl acetate, 330-331. Pyruvic Acid, 331. Acetoacetic Acid. Ethyl Acetoacetate, 332. Levulinic Acid, 333. KETONIC ALCOHOLS : Acetonealcohol. Dihydroxyacetone. Butanolone, 333. DIKETONES : Diacetyl. Acetylacetone, 333-334. KETO-ALDEHYDES : Pyruvic Aldehyde and Acetoacetaldehyde. Hydroxymethyleneacetone. 334. F. POLYVALENT DIBASIC HYDROXY-ACIDS AND THEIR DERIVATIVES 334 Tartronic Acid, 334. Malic Acid and higher homologues, 335. TARTARIC ACIDS: (1) d-Tartaric Acid, 335. (2) 1-Tartaric Acid. (3) Racemic Acid. (4) Mesotartaric Acid, 336. TARTAR INDUSTRY : Manufacture of Tartar, 337. Analysis of tartar, 337. Statistics, 340. Manufacture of tartaric acid, 341 ; uses and statistics, 343. Artificial tartaric acid, 343. Trihydroxyglutaric Acid, 343. Saccharic and Mucic Acids, 344. DIBASIC KETONIC ACIDS, 344. Mesoxalic Acid. Oxalacetic Acid. Acetonedicarboxylic Acid. Dihydroxytartaric Acid, 344. G. POLYVALENT TRIBASIC HYDROXY-ACIDS 345 Tricarballylic Acid. Aconitic acid. Citric Acid and its Industry, 345. Tests for citric acid, 346. Salts of citric acid, 346. Citrates, 346-347. Citrus industry, 347. Statistics, 349. Higher polybasic hydroxy-acids, 351. IV. THIO-ACIDS AND THIO-ANHYDRIDES 351 Thioacetic Acid. Ethanthiolic Acid. Acetyl Sulphide. Ethyl Thioacetate. V. AMIDO-ACIDS, AMINO-ACIDS, IMIDES, AMIDINES, THIOAMIDES, IMINO-ETHERS AND ANALOGOUS COMPOUNDS 351 A. Amido-Acids and their Derivatives : Primary, secondary, and tertiary amides ; alkylated amides. Preparation and properties of amides, 351-352. Formamide ; Acetamide, diacetamide ; Oxamic Acid ; Oxamide ; Succinamic Acid ; Succinamide ; Glycollamide, diglycollimide ; Malamic Acid, malamide, 352-353. B. IMIDES AND IMINO-ETHERS: diacetamide, iminohydrin of glycollic acid ; Oximide, Succinimide, pyrrole, pyrrolidine, succinanil ; Glutarimide, 353-354. C. AMINO-ACIDS AND THEIR DERIVATIVES: Glycocoll, sarco- sine, betaine, aceturic acid ; Serine ; Leucine ; Aspartic Acid, glutamic acid ; Ethyl Diazoacetate ; Lysine, ornithine, putrescine, taurine, cysteine, cystine ; Asparagine, Aspartamide, homoaspartic acid and homoasparagine, 354-356. D. AMIDO- AND IMIDO-CHLORIDES : acetamido-chloride, acetimino- chloride, 356. E. THIOAMIDES : thioacetamide, 367. CONTENTS xv PAGE F. IMINOTHIOETHERS : acetiminothioinethyl hydriodide, 357. G. AMIDINES : acetamidine, 357. H. HYDRAZIDES AND AZIDES : diaccthydrazide, 358. I. HYDROXYLAMINE DERIVATIVES OF ACIDS : hydroxamic acids, amidoximes, isuret, 358. VI. CYANOGEN COMPOUNDS 358 Cyanogen : paracyanogen ; rubeanhydric acid and flaveanhydric acid. Cyanogen Chloride, 358-359. Cyanic Acid : potassium and ammonium cyanates, 359. Ethyl Isocyanate, 359. Cyanuric Acid : Ethyl cyanurate and isocyanurate, 359. Fulminic Acid, 360. THIOCYANIC ACID AND ITS DERIVATIVES, 360. Potassium, Ammonium, Mercuric, Silver, and Ferric Thiocyanates, 361. Ethyl Thiocyanate. Allyl Thiocyanate, 361. MUSTARD OILS : methyl, ethyl, propyl, Allyl, 361. CYANAMIDE AND ITS DERIVATIVES, 362. Calcium cyanamide, 362. Diethylcyanamide. Dicyanodiamide. Melams : Melamine, Ammeline, Ammelide, 362. VII. DERIVATIVES OF CARBONIC ACID 362 Esters of carbonic acid. Ethyl carbonate, ethylcarbonic acid, 363. Chlorides of Carbonic Acid. Chlorocarbonic acid, ethyl chlorocarbonate and chloroformate, 363. Amides of Carbonic Acid. Carbaminic acid, urethane, urea, semicarbazide, acetylurea, allophanic acid, ureides, biuret, hydantoic acid, hydantoin, 363-364. DERIVATIVES OF THIOCARBONIC ACID : thiophosgene, trithio- carbonic acid, potassium xanthate, xanthonic acid, dithiocarbamidic acid, diethylthiourea. Thiourea, 364-365. GUANIDINE AND ITS DERIVATIVES : nitroguanidine, aminoguani- dine, diazoguanidine, hydrazo- and azo-dicarbonamide, glycocyamine, sar- cosihe, creatine, creatinine, 365-366. URIC ACID AND ITS DERIVATIVES : ureides, uro-acids, diureides ; parabanic acid, barbituric acid, dialuric acid, alloxan, oxaluric acid, alloxanic acid, cholestrophane, methyluracil, alloxanthine, murexide, allantoin, purine, dimethylpseudouric acid, theophylline, caffeine, theobromine, hypoxanthine, xanthine, adenine, guanine, uric acid, adenine, 366-369. VIII. ESTERS (Oils, Fats, Waxes, Candles, Soaps) 309 Preparation : theory of the formation of esters ; fruit essences : ethyl formate, ethyl acetate, amyl acetate, ethyl butyrate, isoamyl isovalerate, cetyl and melissyl palmitates, ceryl cerotate, 369-372. Glycerides, Fatty Oils, Waxes, Candles, Soaps, 372. Tripalmitin, tristearin, triolein, lecithin ; serum-lipase, drying oils, varnishes ; rancidity of oils, blown oils, 372-376. Waxes : beeswax, virgin wax, white wax, carnauba wax, Japanese wax, 376-377. Saponifi cation of Fats and Waxes, 377. Table of physical and chemical constants, 378. ANIMAL OILS AND FATS, 379 : tallow, 380 5 oleomargarine, 382 ; margarine, 383; butter, 385; milk, 385; bone fat, 388; lard, 388; fish oils, sperm oil, cod-liver oil, spermaceti, 389 ; degras : wool-fat, 389. VEGETABLE OILS, 390 : Table, 391 ; extraction by pressure, hydraulic press, extraction by solvents, refining, emulsor-centrifuges and centrifugal separators, 391-395 ; olive oil, 395 ; castor oil, 398 ; linseed oil, 39$ ; oil xvi CONTENTS varnishes and lacs, 400 ; palm oil, palm-kernel oil, 401 ; coco-nut oil, 402 ; vege- table tallow, 403 ; cotton-seed oil, 403 ; maize oil, 403 ; sesame oil, 404 ; arachis, soja bean, grape-seed and tomato-seed oils, 404-405. TREATMENT OF FATS FOR CANDLES AND SOAPS: (1) saponi- fication with lime, magnesia, or zinc oxide ; (2) decomposition with sulphuric acid ; oleine of distillation ; transformation of oleic acid into solid fatty acids ; (3) saponification with water ; (4) biological process ; (5) catalytic process (Twitchell), 405-411. MANUFACTURE OF CANDLES, 412. De Schepper and Geitcl's Table, 413. MANUFACTURE OF SOAP, 415. Theory of saponification, 416. Fatty acid or oleine soap, 419. Resin soap, 420. Mottled soap, 421. Transparent soap, 422. Soft soap, 422. Statistics, 424. Analysis of soap, 425. GG. ALDEHYDIC OR KETONIC POLYHYDRIC ALCOHOLS CARBOHYDRATES A. Monoses 426 Aldohexoses, ketohexoses, osazones, hydrazones ; general methods of formation of monoses, 426-428. Tetroses and Pentoses : pentosans, arabinose, xylose, rhamnose, &c., 429-431. Hexoses : glucose, caramel ; fructose ; mannose ; galactose, 431-437. Glucosides, 437. B. Hexabioses 438 Maltose, lactose, 438. Sucrose : calcium sucrate, 440. C. Trioses. Raffinose 442 INDUSTRIAL PREPARATION OF SUCROSE 442 I. Acer saccharinum nigrum, 443. II. Sugar-cane, 443. III. Sugar beet, 445 : cultivation, composition, 446 ; extraction of sugar from beet, 448 ; extraction by diffusion, 450 ; extraction by the Steffen process, 456 ; filter-presses, 459 ; concentration of the juice, 461 ; boiling of the concentrated juice, 466 ; centri- fugation of the, massecuite, 468 ; refining, 470 ; revivification of animal charcoal, 470 ; utilisation of molasses : osmosis, lime, strontia, baryta processes, 473. Yield, 476. Statistics, 477. Fiscal relations, 477. Density table, 482. Quantitative analysis of saccharine materials, 481. Stammer's table, 482. Polarimeters and saccharimeters, 483. Scheibler's table, 483. Chemical tests, 486. Non-sugar, apparent and real densities, quotient of purity, 487. Purifi- cation of waste-liquors from sugar factories, 489. D. Tetroses 489 E. F. Higher polyoses . 489 Starch, 489. Analysis, 500. Dextrin, 501. Gums, 502 ; glycogen, 503. Cellulose, 503. Hydro- and oxy- cellulose. 504-506. Artificial parchment, 506. Paper industry, 506. Statistics, 517. PART III. CYCLIC COMPOUNDS A A. ISOCYCLIC COMPOUNDS 520 I. Cydoparajfins and cyclo-olefines or polymethylene compounds, 520. II. Benzene derivatives or aromatic compounds : the formula of benzene, 521. Isomerism in benzene derivatives, 523. General character and forma- tion of benzene derivatives, 524-525, CONTENTS xvii PAGE A. AROMATIC HYDROCARBONS 526 Distillation of tar, 526. Table, 527. Lampblack, 528. Tar oils, 531. Preservation of wood, 532. Benzene, 533. Toluene and xylenes, 534. Hydrocarbons with unsaturated side-chains, 535. B. HALOGEN SUBSTITUTION DERIVATIVES OF BENZENE 536 Table, 537. C. SULPHONIC ACIDS 538 Benzenesulphonic acid, 538. ' D. PHENOLS 539 (a) Monohydric phenols, 539. Table, 540. Carbolic acid, 541. Antiseptics, 541. Homologues, 543. (b) Dihydric phenols : pyrocatechol, resorcinol, hydroquinone, orcinol, eugenol, isoeugenol, 543. (c) Trihydric phenols : pyrogallol, hydroxyhydroquinone, phloroglucinol, 545. (d) Polyhydric phenols : hexahydroxybenzene, quercitol, inositol, 546. E. QUINONES 546 F. NITRO-DERIVATIVES OF AROMATIC HYDROCARBONS 547 Table, 548. Nitrobenzene, dinitrobenzenes, nitrotoluenes, trinitrotoluenes, phenylnitromethane, pseudo-acids, 649-654. G. AMINO-DERIVATLVES OF AROMATIC HYDROCARBONS 554 Table of Aromatic Amines, 555: (1) primary monamines ; (2) secondary monamines ; (3) tertiary monamines; (4) quarternary bases; (5) diamines, Iriamines, tetramines. Aniline, nitraniline, methylaniline, diphenylamine acetanilide, exalgin, phenylsulphaminic acid, 554560. Homologues of aniline : toluidines, xylidines, benzylamine, phenylenediamine, 560-562. H. NITROPHENOLS, AMINOPHENOLS, AND THIOPHENOLS 562 Nitrophenol, picric acid, aminophenols, thiophenols, 562-564. I. AZO- DIAZO- AND DIAZOAMINO-COMPOUNDS AND HYDRAZINES 565 (1) Azo- derivatives ; (2) diazo-derivatives ; (3) diazoamino- derivatives (4) hydrazines, 565-570. L. AROMATIC ALCOHOLS, ALDEHYDES, AND KETONES 570 Benzyl alcohol, benzaldehyde and its homologues, cinnamaldehyde, 570-572. Aromatic ketones, 572. Aromatic oximes, 572. Beckmann rearrangement, 573. M. HYDROXY - ALCOHOLS, HYDROXY - ALDEHYDES, AND KETONIC ALCOHOLS 673 Salicylaldehyde, anisaldehydei vanillin, 573. Aromatic hydroxy-alde- hydes, 674. N. AROMATIC ACIDS 57 General methods of preparation and properties, 575. (a) MONOBASIC AROMATIC ACIDS, 576. Table, 677. Benzole anhydride, benzoyl chloride, benzamide, hippuric acid, chlorobenzoic acid, m-nitrobenzoic acid, azobenzoic acids, aminobenzoic acids (anthranilic acid), diazobenzoic acids, anthranil, sulphobenzoic acids, saccharin, toluic acids, phenylacetic acid, xylic acids, cuminic acid, cinnamic acid, phenylpropiolic acid, 578-580. II b xviii CONTENTS PAGE (6) DIBASIC AND POLYBASIC AROMATIC ACIDS, 580 . phthalic acid, phthalic anhydride, phthalide, phthalophenone, phenolphthalein, fluorescein, eosin, phthalimide, isophthalic acid, terephthalic acid ; polybasic acids : mellitic acid, 580-581. (c) HYDROXY- ACIDS OR PHENOLIC ACIDS, 582; salicylic acid, m- and p-hydroxybenzoic acids, gallic acid, ink, cumaric and mandelic acids. Tannin, tanning of skins, commercial data, 582-591. 0. HYDROGENATED BENZENE COMPOUNDS 591 Hydrophthalic acids. Terpenes, cymene, carvene, 1-limonene, sylvestrene, terpinolene, terpinene, dihydrocymene, phellandrene, menthene, menthane, 591-596. Complex Terpenes, pinene, oil of turpentine, colophony, camphene, fenchene, camphane ; rubber, ebonite, guttapercha, ionone, 596-600 ; camphors, terpane, menthol, pulegone, carvone, terpenol, terpineol, terpin, cineol ; camphor, artificial camphor, 600-605. P. CONDENSED BENZENE NUCLEI 605 (1) Diphenyl and its derivatives : diphenyl, benzidine, carbazole, di- hydroxyphenyl, &c., 605. (2) Diphenylmethane and its derivatives : dihy- droxybenzophenone, diphenylethano, tolylphenylmethane, benzoylsulphonic acids, fluorene, 606. (3) Triphenylmethane and its derivatives : leuco-bases, malachite green, pararosaniline, rosaniline, fuchsine, methyl violet ; rosolic acid, phthalophenone, hexaphenyle thane, and pentaphenylethane, 607. (4) Dibenzyl and its derivatives, 609. (5) Naphthalene and its derivatives : naphthalene, hydronaphthalene, perinaphthalenecarboxylic acid, a-nitro- naphthalene, a-naphthylamine, naphthalenesulphonic acids, o- and /3- naphthols, dinaphthol, betol : naphthionic acid, a- and /3-naphthaquinones, naphthalic acid, dinaphthyl, acenaphthene, 610-614. Addition products of naphthalene. Indene, 614. (6) Anthracene group : anthracene, carbazole, anthraquinone, alizarin, 614-618. Phenanthrene, phenanthraquinone, fluor- anthrene, pyrene, chrysene, picene, retene, 618-619. Q. HETEROCYCLIC COMPOUNDS 6 d .# 9 FIG. 13. a 5 cm. free ; b => 12 cm. spiral of oxidised copper gauze ; c = 8-10 cm. for the boat ; d 3 cm. copper spiral ; e = 40-45 cm. granulated cupric oxide ; / = 3 cm. oxidised copper spiral or 12 cm. of reduced copper spiral for nitrogenous substances ; g = 5 cm. free. under a bell-jar ; while Gay-Lussac, Thenard, and Berzelius successively improved this process by burning the substance in presence of potassium chlorate. Gay-Lussac, how- ever, showed that certain nitrogenous substances cannot be burned with the chlorate, and suggested as a general and more certain oxidising agent cupric oxide, which when hot gives up its oxygen, transforming the carbon and hydrogen of any organic compound into carbon dioxide and water respectively, while the nitrous compounds are reduced to free nitrogen by passing the products of combustion over red-hot copper turnings. But it is to Liebig that the credit is due of rendering this method of organic analysis simple and exact and of devising simple and ingenious forms of apparatus for absorbing the products of combustion. Even to-day disregarding improvements in combustion furnaces and modifications of the absorption apparatus the determination of carbon and hydrogen (the oxygen is estimated by difference) is carried out by what is virtually the method employed by Liebig. The method most commonly used is as follows : 0-15-0-30 grm. of the substance is weighed in a small porcelain boat, which is then filled with powdered cupric oxide, previously heated to redness and perfectly dry ; the boat is then introduced into the position c of the hard glass combustion tube (Fig. 13), this being 70-90 cm. long, or 10-12 cm. longer than jthe combustion furnace,*, which is heated by 25-30 gas flames (Fig. 14). The other parts of the tube are reserved for the previously heated copper spirals and granulated cupric oxide (Fig. 13). When a fresh combustion is to be made, all that it is necessary to do is to remove the spiral b and the boat and to introduce the new substance into the tube, which is already charged in d, e, and / and is not allowed to cool below 40-60. The combustion is carried out in the furnace shown in Fig. 14, the tube being closed at a with a good cork and a glass tap which can be connected at will with a gasometer containing air or one containing oxygen, which should, however, before reaching the combustion tube, pass through tubes containing potassium hydroxide to remove the 8 . carbon dioxide, and then through drying tubes containing calcium chloride. At the other end the combustion tube communicates at b, first with a tared tube, c, containing granulated calcium chloride to absorb the water formed during combustion ; then follows the tared apparatus d, containing potassium hydroxide solution (30-35 per cent.), which absorbs the carbon dioxide from, the burnt substance and is furnished with a calcium chloride tube to retain the moisture given off by the potassium hydroxide solution. Finally follows a calcium chloride tube, e, which is not weighed; and prevents moisture entering the apparatus from the air. Before the combustion is started the apparatus is tested to ascertain if it is perfectly air-tight. This is done by closing the tap a, and sucking into e eight or ten bubbles of gas ; the slight rarefaction produced in the interior of the combustion tube causes the potash solution to rise in the first large bulb to a level which should remain constant for some minutes. The burners at the end b are then gradually lighted until the portion / and almost all of the portion e are heated to redness. The spiral b is then gradually heated from the a end, the heating being gradually extended under the boat so that the substance is completely burnt. During the combustion bubbles of air are passed into the tube from the gas-holder so as to transport the gases produced into the absorption apparatus ; during the last 10-15 minutes a gentle current of oxygen is passed through, and then the flames are extinguished and air again passed .for 10-15 minutes. In this way all the gases from the combustion are removed from the combustion apparatus and the copper oxide is completely reoxidised, so that the tube is ready for the next combustion. FIG. 14. The increases in the weights of the potash and calcium chloride apparatus give the amounts of carbon dioxide and water respectively formed during the combustion, and, since 44 parts of carbon dioxide correspond with 12 parts of carbon and 18 parts of water with 2*of hydrogen, the quantities or percentages of carbon and hydrogen in the substance can^be^calculated. The sum of these two percentages, when subtracted from 100, gives that of the oxygen, excepting where the substance, contains nitrogen, which is determined directly by methods given later. In this way the percentage composition is determined. For determining carbon and hydrogen in nitrogenous substances the above method is modified only by inserting in the combustion tube, in place of the spiral / (Fig. 13), one of reduced copper gauze x about 15 cm. long, this serving to fix the oxygen from the oxides of nitrogen resulting from the combustion and to liberate the nitrogen, which passes unchanged through the absorption apparatus. If the substance to be analysed contains sulphur or a halogen, the combustion is made with lead chromate in place of the granular copper oxide, and the heating is more gentle to avoid fusion of the chromate. By this means the sulphur remains fixed in the tube as lead sulphate and the halogens as halogen salts of lead. Halogens can also be fixed on a spiral of silver foil about 10 cm. long placed at / (Fig. 13), the substance being combusted as usual with cupric oxide ; if both nitrogen and a halogen are present the copper and silver spirals are used together. A new apparatus, which admits of the combustion of organic substances being very rapidly carried put, is that devised by Carrasco and Plancher (1904-1906). It consists of 1 The reduction is effected in a separate glass tube, through which a current of hydrogen is passed while the spirals are heated ; when the copper has assumed its characteristic red colour, the flames are extinguished and the spirals allowed to cool in the current of hydrogen, being afterwards kept in desiccators ready for use ; or, better, when reduction is complete and the spirals are still hot, the tube is exhausted and is kept so until cold, so as to a void t,ho danger of hydrogen being occluded by the copper. RAPID ELEMENTARY ANALYSIS 9 a small external combustion tube, c (Fig. 15), of hard glass and about 20 cm. long and 2 cm. wide and slightly expanded at the lower closed end. The tube is closed at the top by a rubber stopper, /, through which passes a porcelain tube, e, wound round with an electric resistance formed of platinum-iridium wire, d ; along the interior of the porcelain tubes passes a thick silver wire, which starts from d, the negative pole, and ends in a small platinum wire loop and serves to convey the current (3 amps, at 20 volts). The oxygen for the combustion traverses OS and the upright tube of the stand, and passes through the porcelain tube to the bottom of the combustion tube. In the stopper, /, is fastened a piece of nickel tube, &, which is united to the + pole and to the platinum spiral, d, and serves at the same time for the escape of the gases formed by the combustion to the tube r. The gases are absorbed by the usual tared apparatus (M=calcium chloride, FIG. 15. p= concentrated potassium hydroxide solution), but with nitrogenous or halogenatcd substances the gases are first passed through a U -tube containing lead dioxide heated to 180 by means of a small furnace, TO. The connections a and b are insulated from one another by porcelain and rubber. When the current passes through the resistance the glass tube is heated to redness, and the substance (0-12-0-20 grm.), mixed with cupric oxide or, better, with platinised porous porcelain powder, and placed at the bottom of the glass tube, is burned by heating the outside of the tube directly with a Bunsen flame. The combustion is very soon completed, the platinum-iridium spiral apparently accelerating the oxidation catalytically ; apart from the time occupied by the weighings, this method requires 15-20 minutes, and usually gives good results. For the analysis of fairly volatile liquids or of substances which readily sublime, the lower part of the combustion tube is drawn out almost horizontally, and the substance is mixed with platinised porcelain powder (2-3 per cent, of platinum) ; liquids can be heated in a separate tube and the vapour then injected into the combustion tube. An electrical method for determining carbon, hydrogen, and sulphur in organic subtances was also proposed by Morse and Gray in America in 1906. 10 ORGANIC CHEMISTRY QUANTITATIVE DETERMINATION OF NITROGEN. (1) Dumas' Method. The nitrogenous organic substance (0-2-0-3 grm.) is heated in a hard glass tube similar to that shown in Fig. 1 3, but closed at the end, a. The portions a and 6 contain sodium hydrogen carbonate or magnesium carbonate ; between b and c is placed a small plug of copper gauze, in c granulated copper oxide, and in d powdered copper oxide. Then follows a space 10 cm. in length in which is placed the substance to be analysed, this being weighed and mixed with powdered cupric oxide ; next comes granulated cupric oxide, and in / a spiral of reduced copper, 10-12 cm. long. 1 The extremity, g, of the tube is connected by means of a gas delivery tube with a graduated tube (25 or 50 c.c.), placed upside down in a basin of mercury and filled half with mercury and half with concentrated potassium hydroxide solution. This graduated tube may have the form devised by Dumas and shown in Fig. 16 ; the gas from the combustion tube passes into the tube a, furnished with a clip, m, thence through a little mercury in the bottom of the tube b, which is filled with potassium hydroxide solution and is in com- munication with a reservoir, c, of this solution. The operation is begun by heating the combustion tube at the point where the magnesium carbonate lies ; the carbon dioxide thus evolved expels the air from the apparatus into b, whence it is driven by raising the reservoir, c, and opening the cock at the top of b. The carbon dioxide is absorbed by the potash solution, and when no more air collects in 6 the magnesium carbonate is no longer heated. The copper spiral and the copper oxide are now gradually heated in the same way as for the estimation of carbon and hydrogen, the heating being slowly extended until it reaches the substance itself. Oxides of nitrogen are decomposed by the copper spiral, so that all the nitrogen is evolved in the free state and collects in b. Finally the nitrogen remaining in the com- bustion tube is driven into b by means of carbon dioxide formed by again heating the magnesium carbonate. At the end of the operation, in order to measure the nitrogen, a graduated tube filled with water is inverted over d, and the cock at the top of b having been opened, the reservoir, c, is raised until all the gas passes into the graduated tube. The latter can then be removed to a large cylinder full of water and when, after a few minutes, the gas has assumed the temperature of the water (shown by an accurate thermometer) the tube, grasped by a clip (the hand would warm it), is arranged so that the level of the liquid inside it coincides with that outside and the volume (v) of the gas read off. At the same time the atmospheric pressure (b) is read, and the exact tempera- ture (t) of the water. The percentage of nitrogen (p) in the substance is then calculated by means of the following formula : v. (b -w). 0-12511 P = s.760(l + 0-00367-0 where s indicates the weight of substance taken, w the pressure of water vapour expressed in mm. of mercury (see vol. i. p. 34) and 0-0012511 grm. the weight of 1 c.c. of moist nitrogen at and 760 mm. (Rayleigh and Ramsay). When several determinations of nitrogen are to be carried out the procedure is some- times simplified by using a combustion tube open at both ends, like that of Fig. 13, the magnesium carbonate or sodium bicarbonate being omitted and the combustion tube being connected at a with a small Kipp's apparatus for the evolution of carbon dioxide (marble and hydrochloric acid), care being taken to free the apparatus from all air by a prolonged current of carbon dioxide. |(2) KjelddhVs Method (Dyer's modification). 0-5-1 grm. of the substance is placed in a hard glass flask (200-300 c.c.) with a long neck, into which penetrates the stem of a funnel used to cover the flask (Fig. 17). 20 c.c. of concentrated sulphuric acid (66 Be.) and 1 In this case the copper spiral can be rapidly reduced by heating it over a large non-luminous gas flame and dropping it into a thick-walled test-tube containing \ c.c. of ethyl or, better, methyl alcohol ; the tube is immediately closed by a rubber stopper through which passes a glass tube. The latter is connected with a pump until the spiral is cold. FIG. 16. ESTIMATION OF NITROGEN 11 a drop of mercury (which acts as a catalytic oxidising agent) are added, and the contents of the flask are heated, at first gently and finally more strongly, until vigorous boiling sets in. 10 grms. of potassium sulphate are then added, a little at a time, the heating being continued until the liquid is decolorised, by which time the whole of the nitrogen is transformed into ammonium sulphate. After the flask has been allowed to cool, its contents are washed out with water into a flask already containing 200-300 c.c. of water. 3-4 grms. of zinc dust (which decomposes ammoniacal compounds of mercury and prevents bumping by the evolution of hydrogen) are then added, and the flask closed with a rubber stopper through which passes a tapped funnel containing 120-160 c.c. of concentrated sodium hydroxide solution (30-35 per cent.) and a glass bulb (Figs. 18 and 19) communicating with a simple condensing tube dipping into a flask containing a mea- sured volume of standard sulphuric acid and a drop of methyl orange. In order to prevent spurting of the caustic soda and its introduction into the condenser tube, the glass bulb is fitted with a delivery tube curved towards the wall of the bulb ; it is, however, as well to push into this tube, almost as far as the bulb, a small plug of glass-wool or asbestos. Solutions of soda more concentrated than 35 per cent, often lead to spurting. About one-half the liquid is distilled and the excess of sulphuric acid remaining in the collecting flask determined]by titration with alkali. Hence the amount of ammonia fixed by the acid can be'calculated and so the percentage of nitrogen in the substance analysed. In Figs. 17 and 19 are shown forms of apparatus with which it is possible to carry out several determina- tions simultaneously. FIG. 17. FIG. 18. FIG. 19. Kjeldahl's method cannot be used for the analysis of organic substances which contain nitrogen either united to oxygen (nitro compounds) or forming part of a pyridine or similar nucleus (quinoline, &c.) (3) Will and Varrentrapp's Method. This method is based on the principle that almost all nitrogenous organic substances (which do not contain nitrogen linked to oxygen, such as the nitro-compounds), when they are heated with an alkali hydroxide or, better, with soda lime (see vol. i, p. 490), yield hydrogen, which transforms the nitrogen into ammonia. Little use is made of this method to-day. QUANTITATIVE DETERMINATION OF THE HALOGENS. The method most commonly used is that of Carius. The substance (0-15-0-2 grm.) is weighed out^in a small tube, which is then introduced into a large, hard glass tube 30-40 cm. long and 2-3 cm. wide, closed at one end and containing about 2 c.c. of fuming nitric acid and about 0-5 grm. of solid silver nitrate ; this introduction is effected in such a way that the acid does not enter the small tube. The large tube is then softened near the open end by heating in the blow-pipe flame andjjradually drawn out to a point, the walls of the tube being 12 allowed to thicken during the fusion (Fig. 20, B, shows the upper part of the tube on a larger scale). After being allowed to cool in a vertical position, the tube is introduced into a thick -walled iron sheath, which is closed with a screw-cap. It is then safe to incline the tube and introduce it into a bomb-furnace (Fig. 21), which holds four or more tubes and is raised slightly at one end ; this is heated for 4-6 hours, the temperature being raised gradually to about 250. Sometimes the tubes burst owing to the great internal pressure, but without danger from flying fragments of glass owing to the protection of the iron sheaths and of the folding shutters at the ends of the furnace, these being lowered during the heating. At the end of the operation, when the tube is cool, it is taken from the iron sheath, held in a vertical position and its point (Fig. 20, A a) softened in a Bunsen flame. When the pressure in the tube has been thus relieved, a scratch is made with a file at the point marked 6, and the file-mark touched with a red-hot glass, with the result that the upper part of the tube breaks off. The tube is then carefully emptied and washed out into a beaker with water, the small tube, held in pincers or a piece of platinum wire, being well washed inside and outside before removal. The liquid is heated and the precipitated silver FIG. 20. FIG. 21. halogen compound is then collected on a filter, washed, dried in an oven, detached from the filter and heated in a weighed porcelain crucible until it just begins to melt. After being allowed to cool in a desiccator, the crucible is weighed and the amount of halogen contained in the organic substance calculated from the weight of silver haloid. QUANTITATIVE DETERMINATION OF SULPHUR AND PHOSPHORUS. This is carried out by the Carius method in the same way as for halogens, except that no silver nitrate is introduced into the tube. At the end of the heating, the sulphur is obtained as sulphuric acid or the phosphorus as phosphoric acid, estimation of the amounts of these acids being effected by the ordinary methods. The halogens, sulphur and phosphorus, may also be determined after fusion of the substance with pure sodium peroxide. ' CALCULATION OF THE EMPIRICAL FORMULA. From the results of the elementary analysis of an organic substance can be calculated the percentage composition, i.e. the quantity of each component in 100 parts of substance. To deduce the chemical formula, that is, the proportions in which the different atoms enter into the molecule, the percentage weight of each component is divided by the corresponding atomic weight, the numbers MOLE. CULAR WEIGHTS 13 thus obtained giving the proportions between the numbers of atoms of the different elements. These numbers sometimes give directly the numbers of atoms contained in the molecule, but in other cases they represent multiples or submultiples of the real numbers of atoms. If, for example, lactic acid is analysed, the percentage composition is found to be : C, 40 per cent. ; H, 6-6 per cent. ; 0, 53-4 per cent ; by dividing these numbers by the corresponding atomic weights, the following numbers are obtained: C, 3-3 (i.e. -*-); H, 6-6 (*-); and 0, 3-3 (^*-). These pro- portions have a common factor, 3-3, and division by this gives 1C, 2H, and 10, i.e. CH 2 O, which is an empirical minimum or formula, the simplest formula expressing the proportions between the numbers of atoms of the different elements. This minimum formula does not, however, represent the molecular magni- tude, and, in fact, analyses of formaldehyde, acetic acid, grape sugar, &c., give the same percentage composition and the same minimum formula, CH 2 0, which must hence be a submultiple of the formulae of these substances. ^ knowledge of the percentage composition is not sufficient to determine the true molecular formula ; the molecular magnitude, i.e. the molecular weight, must also be known in order to permit of a choice between the various multiples. By making use of one of the methods described in vol. i, " Inorganic Chemistry " (pp. 39, 81 et seq.), the molecular weight of lactic acid is found to be 90, so that, of the various possible formulae, CH 2 (mol. wt. 30), C 2 H 4 2 (mol. wt. 60), C 3 H 6 3 (mol. wt. 90), C 4 H 8 4 (mol. wt. 120) C 6 H 12 6 (mol. wt. 180), &c., only C 3 H 6 3 corresponds with lactic acid. But even this formula and the empirical formula tell nothing concerning the grouping of the atoms in the molecule which, as is explained in the following pages, is given by the constitutional formula. DETERMINATION OF THE MOLECULAR WEIGHT BY CHEMICAL MEANS In lactic acid one-sixth of the hydrogen can be substituted by a metal, so that there must be at least six (or a multiple of six) atoms of hydrogen in the acid, the empirical formula being necessarily at least trebled, giving C 3 H 6 3 . To ascertain if this is the true formula, a derivative of the acid is prepared, such as the silver salt, which can easily be obtained pure. Analysis of this salt shows it to contain 54-8 per cent, of silver, and the atomic weight of silver being 107-7,. calculation indicates that the residue of the lactic acid combined with 107-7 parts of silver weighs 89. Assuming that only 1 atom of silver has entered the lactic acid in place of 1 of hydrogen (as can, indeed, be deduced from the fact that the quantity of hydrogen in the salt is five -sixths of that originally present in the acid), the weight of the lactic acid would be 89 + 1, or 90. The true formula of the acid would hence be that. corresponding with a molecular weight of 90, i.e. C 3 H 6 O 3 . For acid substances in general this chemical method may be employed for determining the molecular weight, making use of the silver salt and determining if the acid is mono-, di-, or tri-basic (that is, ascertaining if the silver replaces 1, 2, or 3 atoms of hydrogen), the calculation being then based on the presence of 1, 2, or 3 atoms of silver in the salt. For basic substances, the molecular magnitude may be determined chemically by analysing the platinichlorides, the formulae for which are always of the type of that of ammonium platinichloride : PtCl^NHg-HCl^, the ammonia being replaced by the organic base, which is mono- or di-acid, according as it replaces one or two molecules of ammonia in the platinichloride. For other (indifferent) organic substances derivatives are prepared by substituting chlorine atoms for one or more hydrogen atoms, the proportion of chlorine being then estimated ; the calculation is then similar to that described above. The chemical method for determining the molecular magnitude does not always give certain results : experimental difficulties sometimes occur and often entail great labour. 14 ORGANIC CHEMISTRY Consequently the determination of molecular weights is usually effected by physical methods : vapour density method, cryoscopic method, ebullioscopic method, &c., these being all described and illustrated in vol. i (Part I). POLYMERISM It sometimes happens that the analysis of different substances shows them to have the same percentage composition, although their chemical and physical properties are different ; thus, for example, acetic acid, lactic acid, glucose, &c., contain the same elements, C, H, and 0, in the same proportions, there being 2n hydrogen atoms and n oxygen atoms for every n carbon atom. Accurate study of these compounds and determination of the molecular magnitude (molecular weight) shows that the differences depend on the true formulae being multiples of the minimum or empirical formula. Thus, whilst the molecule of acetic acid is represented by C 2 H 4 2 , that of lactic acid corre- sponds with C 3 H 6 3 , and that of glucose with C 6 H 12 6 . These molecules are hence all multiples of a hypothetical complex CH 2 0, the ratios (but not the absolute quantities) between carbon, hydrogen, and oxygen being the same (1:2:1) in all cases. These compounds are termed polymerides and the phenomenon is known as polymerism. In some instances, however, it happens that the molecular magnitude is not sufficient to differentiate certain compounds, which, besides containing the same elements in the same proportions (equal percentage compositions), have also the same molecular magnitudes, although differing in their physical and chemical properties. To explain the existence of these isomeric compounds, the chemical nature of carbon must be studied more in detail. VALENCY OF CARBON, ISOMERISM, AND CONSTITUTIONAL FORMULA On the foundation of multivalent radicles, 1 discovered by Odling, and of the investigations of Frankland (1852), which showed that nitrogen, phos- phorus, and other elements easily formed compounds with three or five equivalents of other elements, Kekule, in 1857 and 1858, accurately developed the true conception of valency, showing the constant tetravalency of carbon and thus widening the horizon of organic chemistry and originating the remarkable theoretical and practical development of the past half -century. 1 Theory of Radicles and Types. In the first twenty years of last century, various compounds were discovered which stood in apparent contradiction to the electro-chemical theory of dualistic formulae, put forward by Berzelius (vol. i. p. 44) ; in fact, in certain compounds, the hydrogen (electro-positive) was replaced by chlorine (electro-negative) without appreciably changing the chemical characters of the original compounds. It was then that chemical combinations came to be represented by unitary formulae, no account being taken of the grouping of the atoms in the molecule. But gradually, as the number of new organic substances increased, certain analogies became evident in their chemical behaviour. In studying cyanogen Gay-Lussac (1815) had indeed met, in various reactions and in various substances, the residue or radicle CN, which behaved as a monovalent element (like the halogens), combining with one atom of different monovalent metals, &c. In 1832 Liebig and Wohler discovered and studied a monovalent atomic group or radicle, benzoyl, C,H 6 O, which was found in oil of bitter almonds combined with an atom of hydrogen (C 7 H,O) ; on oxidation by the air, this essence became transformed into benzoic acid, C,H 6 O 2 , which with PC1 5 gave benzoyl chloride, C 7 H 6 OC1, and this, in its turn, gave the aldehyde C 7 H 6 O, when treated with nascent hydrogen, or benzoic acid under the action of water. All these compounds contain the monovalent benzoyl nucleus, C 7 H 5 O, which passes unchanged from one to the other by combining with monovalent atoms or groups. In 1833, in a classic work, Bunseu studied another radicle, cacodyl, which is a monovalent organic arsenic residue, As; (3) HI c Hexavalent Tetravalent Divalent HI ; (4) C=; (5) C ; (6) C ; (7) C ; c- \/ \/ \/ \ / fir P P r 1 / \ C \ \ \ (8) (9) C Of- , &c. (10) / c \ C C (11) II I f"1 f-\ V (12) _ P __ pi _ || II ; C C (13) >c >c C C (14) C C A c (15) - c x c c- II I C C Among these chains are two (Nos. 8 and 9) containing four carbon atoms and having equal numbers of free valencies. By saturating these ten free valencies with ten H atoms two compounds are obtained (these have actually been prepared) which contain equal numbers of C and H atoms, and have therefore the same percentage composition and the same molecular weight. The physical and chemical differences of these two compounds, termed isomerides, are explained by the different grouping or linking of the atoms in the molecule. In their chemical transformations, isomerides give up or exchange quite different atomic groups or atoms, owing to the different functions and positions occupied by these atoms or groups in the molecule. It is hence not sufficient to represent organic compounds by an empirical molecular formula, the structural or constitutional formula, deducible from the graphic representation of the chains illustrated above, being necessary in many cases to distinguish between isomerides. To decide which of two isomeric formulae should be assigned to a given substance, various chemical reactions are carried out with the substance, study of the new products indicating the constitutional formula. An example will render these ideas clear : It is found that ethyl alcohol (ordinary liquid alcohol) and gaseous methyl ether have different physical and chemical properties, although they possess the same percentage composition and the same molecular magnitude, represented by the formula C 2 H 6 0. The constitutions or internal molecular structures of the two compounds are determined by a study of the following chemical reactions : treatment of the alcohol with hydrochloric acid gives first a compound C 2 H 6 C1 (ethyl chloride), one atom of monovalent chlorine having replaced one atom of oxygen and one of hydrogen or a hydroxyl residue, OH. By means of nascent hydrogen, the chlorine atom of ethyl chloride can be replaced by a hydrogen atom, giving the compound C 2 H 6 (ethane). These reactions are hence expressed by the following equations : CHEMICAL CONSTITUTION 17 (1) C 2 H 6 .OH + HC1 = H 2 + C 2 H 5 C1 ; (2) C 2 H 5 C1 + H 2 = HC1 + C 2 H 6 ; but ethane H \ / H can have only the constitution, H-^C C~ H, i.e. CH 3 CH 3 , so that the alcohol will H/ \H Hv ,OH have the constitution H-^C 0^ H H/ \E On the other hand, it is found, by various reactions, that the six hydrogen atoms of methyl ether present no difference one from another, and, no matter under what conditions hydriodic acid acts on the ether, it eliminates the oxygen as water, and another product is obtained which contains only one carbon atom in the molecule : The reaction hence takes place according to the equation : G 2 H 6 + 2HI = 2CH 3 I + It is evident, then, that in methyl ether the six hydrogen atoms are united homo- geneously to the two atoms of carbon and that the carbon atoms are joined, not directly, but indirectly, by means of an oxygen atom, which is readily eliminated. The constitu- tional formula of methyl ether will hence be : H H-)C OH or CH 3 O CH 3 . H/ \H Use is not always made of constitutional formulae, since they are not simple and are often inconvenient to write ; hence attempts are made to simplify them by indicating the more important groups or residues contained in the molecule and giving at the same time an idea of the constitutions and of the functions of these groups ; this is done by means of the so-called rational formulce. The rational formula of ethyl alcohol will be C 2 H 5 *OH, in which the monovalent OH residue, characteristic of all the alcohols, is separated ; that of acetic acid will be CH 3 'COOH, the group COOH being characteristic of and common to all organic acids, &c. METAMERISM. Constitutional and rational formulae explain clearly isomerism in general and also the special case bearing the name metamerism. When, to an atom of a polyvalent element are united one or more groups in their different isomeric forms, we have special cases of isomerism for definite groups of substances. xC 3 H 7 For example, in the compound, N^-H , the monovalent group C 3 H 7 may be \H xCH 3 present in its isomeric forms, i.e, either as CH 2 CH 2 CH 3 or as C^-H . Although there is considerable resemblance between these two compounds, their different con- stitutions are manifested in certain chemical and physical properties. The following xCH 3 /CH 3 compounds are also metameric isomerides : N^-GyB* and N^-CH 3 ; in fact, although the percentage compositions and molecular magnitudes are the same in both cases, the sub- stituent groups of the ammonia molecule are different and the compounds belong to different categories disubstituted and trisubstituted ammonias. PSEUDOISOMERISM, TAUTOMERISM, DESMOTROPY. A sub- stance sometimes contains atomic groups that occupy a very precarious (labile) position, since they exert certain influences one on the other and under certain given conditions can react in different ways, giving now one new substance and now another ; this explains how it is that some compounds having a well-defined chemical character can, under some conditions, behave like substances with other chemical characters, without it being necessary to II 2 18 assume a true change of constitution. Thus, for example, some of the deriva- tives of cyanic acid, CN OH, behave like derivatives, sometimes of the formula N=C OH and sometimes of the formula NH = C = O, when the hydrogen atom is replaced by a given radicle. The same is the case' for derivatives of cyanamide, N^?C NH 2 , and of carbodiimide, NH = C=NH ; and of the two non-nitrogenous types, C(OH) = C CO and CO CH CO, where a hydrogen atom oscillates between, the two carbon atoms. These compounds exist usually in only one form, the more stable one, but in the derivatives this stable form, simply on heating, is transformed into the labile one. For this phenomenon Baeyer proposed the name pseudoisomerism, and others that of desmotropy. These forms can be distinguished sometimes by chemical reactions, but more generally by the molecular refraction, dielectric constant, magnetic rotation, electrical conductivity, &c. In various substances, where several hydroxyls are present in more or less adjacent positions, there is often a tendency for intramolecular transformation to take place with condensation of two of these groups and separation of a molecule of water, giving rise to isomeric anhydrides, ethers, ketones, or alcohols, &c. In their turn, these derivatives or isomerides, which can be transformed one into the other, give rise to distinct classes of compounds, and this species of isomerism is called tautomerism. STEREOISOMERISM OR ISOMERISM IN SPACE. We have already seen that, by the tetra valency of carbon and its property of uniting with itself to form various chains, it is possible, in certain cases, to explain the existence of isomerides, which have the same percentage composition and molecular magnitude, but different groupings within the molecules. Many cases of isomerism, foreseen from theoretical considerations, have since been actually met with and different isomerides have been prepared artificially after their existence had been foretold. For a long time, however, certain compounds were known for which ordinary isomerism did not provide any explanation j among these the most important, from an historical point of view also, are the four dihydroxysuccinic acids (tartaric acids), of which two (ordinary tartaric acid and racemic acid) were studied by Berzelius as long ago as 1830. To these must be added laevo -rotatory tartaric acid and meso tartaric acid discovered by Pasteur. All these compounds have the same internal grouping of the atoms, although they are isomerides ; it is not possible to distinguish between them by chemical reactions, but they can be clearly differentiated by their physical behaviour : they form hemihedral, i.e. symmetrical, but non-superposable crystals (related as an object to its image in a mirror) : they have, too, different actions on polarised light, the plane of which is turned to the right by some and to the left by others. These acids are hence known as physical or optical isomerides. Pasteur attempted to explain this isomerism by supposing the atomic groups to be arranged unsymmetrically in the molecule, in some cases in a dextro-rotatory spiral and in others in a laevo -rotatory spiral, or arranged at the vertices of an irregular tetra- hedron. When other similar isomerides the lactic acids had been discovered, J. Wislicenus, in 1873, suggested that isomerism of this kind could be explained only by regarding the groups or atoms of these compounds as arranged in space so as to form distinct configurations. This isomerism in space (stereoisomerism) was explained by van 't Hoff and Le Bel (1874), independently, by means of the hypothesis of the asymmetric carbon atom. The starting-point of this hypothesis was Kekule's idea (1867) of regarding, for the sake of convenience, the carbon atom as situated at the centre of a regular tetrahedron, and its four affinities as directed towards the four vertices, i.e. arranged homogeneously in space (Figs. 22, 23). If these affinities are satisfied at the vertices by monovalent atoms or atomic groups, the following cases present themselves : no isomerism is possible in the com- pounds Ca 3 b, Ca z b z , Ca z be, and Ca b z c, where a, b, and c indicate either atoms other than carbon or groups of atoms (I, H, OH, &c.) ; the compound CH 2 I 2 exists in only one form, and if we put the four atoms (H 2 and I 2 ) at tiie apices of the carbon tetrahedion, no matter STEREOISOMERISM 19 how their positions may be changed, it is not possible to find two different, i.e. non-super - posable arrangements. If, however, the four groups or atoms combined with the carbon atom are all different, e.g. Cabcd, two isomerides are possible and in this case the carbon atom is termed asymmetric; in fact, if these atoms or groups are arranged, in one case, so that the circle a, b, c has a sense opposite to that in which the hands of a clock move (Fig. 24, I) (called, therefore, dextro-rotatory isomerides, and indicated by d- or by the 8tgn +) and, in the other, in the opposite sense (Fig. 25 II) (termed Icevo-rotatory isomerides, like levulose and indicated by I- or ), two non-congruent configurations are obtained ; these cannot be superposed, one on the other, so that the same groups occupy the same positions in the two cases. These two figures represent two different isomerides and are related in the same way as an object to its mirror-image or as the left hand to the right. This isomerism is called enantiomorphism. These two different arrangements of the atoms round the asymmetric carbon atom also explain how it is that when polarised light traverses these molecules, its plane of polarisation is rotated, in one case to the right and in the other to the left. Van 't Hoff and Le Bel pushed their deductions still further, and showed that the dextro-optical deviation should be numerically equal to the laevo -optical deviation of the corresponding isomeride. This has been confirmed practically, and it also follows that when a pair of such isomerides are mixed in equal proportions, there should result an optically neutral mixture, thus giving rise to a special inactive or racemic isomeride. A substance with only one asymmetric carbon atom always gives three stereoisomerides (for example, three lactic acids). It has also been deduced theoretically and proved practically that all optically active compounds contain at least one asymmetric carbon atom, 1 although not all compounds containing asymmetric carbon atoms are optically active, since the molecules may contain groups which neutralise each other's activity. Many examples illustrating these principles will be discussed later in the special part of this book ; meanwhile mention may be made of the most important of these com- FIG. 27. FIG. 28. pounds : leucine, asparagine, coniine, the lactic acids (hydroxypropionic acids), &c., which contain one asymmetric carbon atom and give, in each case, three stereoisomerides. These cases of stereoisomerism, and those which follow, will be understood more easily if studied by means of cardboard tetrahedra with differently coloured vertices. When the substance contains two asymmetric carbon atoms, the number of stereo- isomerides increases as follows : If we take two tetrahedra like that shown in Fig. 26 I and Fig. 28 II, representing two similar molecules containing only one asymmetric carbon atom in which the groups 1 Or else an asymmetric atom of nitrogen (see later) or sulphur, tin, &c. The exceptions to this rule are very rare and uncertain, one of the cases most discussed during recent times (1909-1910) being \-methylcyclohexylidene- 1-acetic acid, which does not appear to contain an asymmetric carbon atom, but is optically active. 20 ORGANIC CHEMISTRY a, 6, and c, satisfying three of the^ valencies, are arranged in a dextro-rotatory sense, and superpose one tetrahedron on the other, so that the free valencies satisfy one another, there results a new isomeride, i.e. a molecule with two dextro-rotatory asymmetric carbon atoms, as shown in Figs. 27 and 29. 1 If we join two Isevorotatory carbon atoms (Fig. 28 II), that is, the mirror images of Fig. 26 I, a laevo -rotatory isomeride (Fig. 30 II) is obtained. Finally, if one dextro-rotatory (Fig. 26 I) and one Isevo -rotatory asymmetric carbon atom (Fig. 28 II) are united, a third stereoisomeride is obtained, which is permanently optically inactive (Fig. 31 III), the effect produced on polarised light by one asymmetric carbon atom being destroyed by the effect of the other. FIG. 29. FIG. 30. In order to understand these stereochemical speculations better, we will apply them to the isomerism of tartaric acid, which has the formula C 4 H 6 6 , and contains two asym- metric carbon atoms (marked with asterisks) to which are joined the groups OH, C0 2 H, andH: CO 2 H CO 2 H If, for the letters a, 6, and c of the tetrahedra considered above, we substitute the groups OH, C0 2 H, and H, and if the tetrahedron of Fig. 26 I (which we will call + A) be represented as if projected on to a plane, thus: a C c or OH C H (dextro- \1 \ I 6 X C0 2 H I I rotatory), and that of Fig. 28 II ( A), thus : c C a or H C OH (Isevo -rotatory), I/ I / 6 C0 2 H we arrive at the following stereoisomerides of tartaric acid : I. By joining two + A atoms, we get d-tartaric acid (Fig. 29 or 32 I). II. By joining two - A atoms, we get Z-tartaric acid (Fig. 30 or 32 II). III. By joining one + A atom with one A atom, we have the permanently inactive mesotartaric acid (t-tartaric acid), as can be seen in Fig. 31 III, or 32 III. IV. By mixing, mechanically, equal parts of acid I ( + ) and II ( - ), there results racemic acid, apparently inactive, but from which, by mechanical means (by hand with the aid of a lens), the two forms of crystals can be separated. It is often assumed that the two asymmetric carbon atoms can rotate independently, on the common axis joining them, so that if the groups of one asymmetric carbon atom exert an attraction or influence on those of the other, a most favourable position could be attained, a chemical reaction being sometimes possible between one group and another 1 Looking at the order in which the letters a, b, and c come in the two asymmetric carbon atoms, it would seem that these are not dextro-rotatory, but this is because the upper carbon atom has been turned through ISO" from its position m Fig. 26 , if itn base is brought down, its identity with the other dextro-rotatory atom becomes evident. ALLOISOMERISM 21 with separation of, say, water and loss of the freedom of rotation ; to the new isomerism thus created we shall refer shortly. STEREOISOMERISM IN DERIVATIVES WITH DOUBLY LINKED CARBON (ALLOISOMERISM). By means of the tetrahedra, we can show a double linking between two carbon atoms by arranging one side of one tetrahedron (carbon atom) in][contact with a side of the other (Fig. 33). With such an arrangement, even without asymmetric carbon atoms, isomerism is possible. In fact, a compound />C=C<\ forms the following isomerides : (1) that CO,H CO,H CT>,H H-'C-OH HO-*C_H BO-'C-B H-'C.OH co,n CO,H I II FIG. 32. CO,H H-'C-OH H-'C-OH CO,H III FIG. 33. shown in Fig. 34, where the two similar atoms or groups of atoms, e.g. a and a, although. a C b united to two different carbon atoms, occupy adjacent positions : || , or ci's-posi'tions aCb (cis-isomerism) ; such a molecule exhibits plane-symmetry, the two pairs of similar groups lying to the left and right, respectively, of the perpendicular plane containing the common side (double linking) ; (2) that shown in Fig. 35, where two similar groups occupy non- a C 6 adjacent or diagonally opposite or trans-positions , this form exhibiting centro- bCa Similarly, a compound of the type, ^>C=C<%, forms two isomerides, the cis-form, aCb aCb , and the trans-torm, a C c c C a _ HC C0 2 H ~ HC.CO.jH __ HO,C.CH HO.C0 2 H FIG. 36. FIG. 37. FIG. 38. The best illustration of this type of isomerism is afforded by the two isomerides : maleic acid (cis-form, Fig. 36) and fumaric acid (trans-iorm, Fig. 37). From these figures it is seen that the cis-form, maleic acid, should lend itself to the ready formation of anhydrides (condensation of two molecules or acid groups with separa- tion of one molecule of water), since the two acid groups, CO 2 H, are very near to one another, and it is, indeed, found that maleic acid easily gives an anhydride with separation of one molecule of water (Fig. 38), whilst no anhydride of fumaric acid is known. Isomerism of this kind is exhibited by various substances, e.g. crotonic and isocrotonic acids (CH 3 CH : CH COOH); metacommand citraconic acids [CH 3 C(COOH ) : CH COOH], &c. Baeyer found that cases of isomerism similar to those just described occur also with cyclic compounds (see Part III), i.e. closed-chain compounds with simple linkings between 22 ORGANIC CHEMISTRY the carbon atoms. He distinguishes with the sign T compounds containing true asym- metric carbon (absolute asymmetry), adding the sign + or if the compound is optically active ; while he gives the name relative asymmetry to that shown by compounds with doubly linked carbon atoms (alloisomerism) or by cyclic compounds with simple linkings, the term cis or trans being added to the JP. Thus, to the name tartaric acid would be added the sign F + or T according as the acid is dextro- or Isevo -rotatory, and to the name maleic acid _T cis , to fumaric acid J 7 " 3 " 5 , &c. STEREOISOMERISM OF NITROGEN. Le Bel attempted to explain the isomerism of certain nitrogen compounds (e.g. methyl-ethyl-propyl-isobutyl-ammonium chloride) by assuming absolute asymmetry for the nitrogen atom. A more plausible explanation seems, however, to be afforded by the idea of relative asymmetry of the nitrogen, analogous to that of carbon atoms when united by double linking ; in this way V. Meyer, Hantzsch, Werner, and others easily explained the isomerism of the oximes, hydroxamic acids, Cab phenylhydrazones, &c. In general, a substance of the constitution || should give two Nc isomerides which can be represented as shown in Fig. 39 ; the s^n-series (Fig. 39 I) and the anti-series (Fig. 39 II). *. ^^^^ These investigators also studied those cases of isomerism in which the nitrogen behaves as (II) s"~& a P en t ava l en * element. SEPARATION AND TRANSFORM A- TION OF STEREOISOMERIDES. Stereo- isomerides and, in general, compounds contain- ing asymmetric carbon atoms, when prepared p . a C b artificially in the laboratory from inactive it || substances, are inactive, the racemic configura- ^ jj N c tion, composed of a mixture of the optical YIG. 39. antipodes in equal quantities, being formed. When, however, these substances are elabo- rated in the animal or vegetable organism, they are usually optically active. The transformation of one of these optical antipodes into the other corresponding with it may sometimes be effected by passing through halogen derivatives, separation of the halogen from which results in the formation of the isomeride of opposite optical activity. The separation of the antipodes, or of one of them, from the racemic isomeride was carried out by Pasteur (1848) in various ways. The following are the methods used at the present time : (1) By fractional crystallisation (see above) of the racemic isomerides or of some of their salts at various temperatures and from various solvents, the antipodes can be sepa- rated directly or else they crystallise in hemihedral forms which can be readily separated. For some substances, it is convenient to prepare compounds with alkaloids (optically active basic compounds, e.g. strychnine, cinchonine, &c.), which, even when they do not form well-defined hemihedral crystals, can be easily separated by fractional crystallisation. (2) By means of enzyme action (maltase, emulsin, &c. ; see section on Fermentation), Fischer succeeded in resolving certain racemic glucosides. Much earlier than this, Pasteur discovered that certain bacteria or moulds (Penicillium glaucum, &c.) are capable of developing in a solution of the racemic substance at the expense of one of the optical antipodes, the other being left unchanged. This phenomenon is explained by the fact that bacteria owe their activity to certain substances which they produce (enzymes), and which are optically active and behave analogously to optically active solvents. Indeed, in many cases, stereoisomeric antipodes are separated by virtue of their different solubilities in an optically active solvent. (3) With certain racemic compounds, the antipodes are separated by taking advantage of their different velocities of esterification in presence of an optically active alcohol ; e.g. for racemic mandelic acid, menthol (which is an active alcohol) is used. For inactive alcohols, the velocity of esterification is the same for the two antipodes composing the racemic compound. (4) When an optically active substance is heated within certain definite limits of temperature (transformation point, see vol. i, p. 190), it is often converted, to the extent HOMOLOGY AND ISOLOGY 23 of one-half, into the oppositely active isomeride, so that an inactive mixture (racemic compound) is obtained ; this takes place readily, for example, with the lactic acids. Above the transformation point the racemic substance may form inseparable mixed crystals (see vol. i, p. Ill), the substance being then called pseudo-racemic. On the other hand, it has been shown that, with certain halogenated compounds, the transformation occurs even at ordinary temperatures, but with a minimum velocity ; thus, with isobutyl bromopropionate, about three years is required. (5) R. Stoermer (1909) found that the more stable form with the higher melting-point is often converted into the more labile form by means of the ultra-violet rays. HOMOLOGY AND ISOLOGY Turning to the more simple compounds, those formed from only carbon and hydrogen, we can easily see what procedure is necessary to arrive at those containing longer and more complex chains of carbon atoms. If we start from the most simple compound, methane (or marsh gas), CH 4 , we can substitute an atom of hydrogen in it by other elements or even condense two of the monovalent CH 3 residues into one compound, CH 3 'CH 3 , thus obtaining ethane (C 2 H 6 ). But in this compound we can also replace an atom of hydrogen by another CH 3 residue, forming propane, CH 3 CH 2 CH 3 or C 3 H 8 , and by continuing this process we arrive at butane, CH 3 CH 2 CH 2 CH 3 , i.e. C 4 H, , ; pentane, C 5 H 12 ; hexane, C 6 H 14 , &c. All the compounds of this series have analogous structures and have also many analogous chemical and physical properties ; such a series is called a homologous series. This series of the derivatives of methane can be represented by the general formula C w H 2w + 2 , each term being the higher or lower homologue of the pre- ceding or following term and differing from it by having one CH 2 complex more or less. If in all the simple compounds of this homologous series of methane we replace successively one hydrogen atom of the CH 3 group by the hydroxyl residue OH (characteristic of the alcohols) we obtain a homologous series of alcohols : CH 3 OH, methyl alcohol; C 2 H 5 OH, ethyl alcohol, the values approach one another as the number of carbon atoms increases : at about C^H^ and for higher terms, the specific gravity becomes about 0-78. The specific gravity of the monobasic fatty acids is greater than 1 for the first terms of the series, but it diminishes with augmentation of the number of carbon atoms in the molecule. MOLECULAR VOLUME. It was thought for many years that certain important rules could be deduced from the molecular volumes of organic compounds, that is, from the quotients, M/P, obtained by dividing the molecular weights (M) by the specific gravities (P), In 1842 Kopp had found that, for liquids at the boiling-point, the molecular volume is very approximately equal to the sum of the atomic volumes of the component elements. For homologous compounds, the molecular volume increases by about 22 for every added CH 2 group. More recent studies (Lessen, R. Schiff, Horstmann, Traube, &c.) show, however, that these regularities are only relative and that isomeric compounds do not possess equal molecular volumes. In unsaturated series, every double linking increases the molecular volume and, with closed-chain compounds, the molecular volume is less than those of the corresponding open-chain compounds with double linkings. So that, in general, the molecular volume depends not only on additive factors (e.g. the sum of the atomic volumes), but also on constitutive factors (different linkings between the carbon atoms). MELTING-POINT. Of two isomerides, that with the more symmetrical structure has the higher melting-point. The members of a series have varying melting-points, those with odd numbers of carbon atoms having lower melting-points than those imme- diately below them with even numbers. There are, in addition, other less important rules, but all present exceptions. A mixture of two substances, in suitable proportions, often has a melting-point lower than that of either of the components. BOILING-POINT. In compounds of the same series, the boiling-point rises with increase of molecular weight, the amount of the increase being about 20 per CH 2 in the methyl alcohol or formic acid series and about 30 for benzene derivatives with methyl THERMAL RELATIONS 25 groups in the nucleus. The boiling-points of isologous hydrocarbons, that is, those of the same number of carbon atoms but of different series (derivatives of methane, ethylene, and acetylene) are very close to one another. Of the isomeric compounds of the aliphatic series, the normal one boils at the highest temperature and the boiling-point is increasingly lowered by increase in the branchings. The substitution of hydrogen by halogens and by hydroxyl groups raises the boiling- point. The ethers boil at lower temperatures than the corresponding isomeric alcohols. HEAT OF COMBUSTION AND HEAT OF FORMATION FROM THE ELEMENTS (see vol. i, pp. 60, 109, 372). The Hess-Berthelot law states that the difference between the heats of combustion of two equivalent chemical systems is equal to the heat developed in the transformation of one system into the other, that is, is equal to the heat of formation from the elements of this latter. In general, we can hence calculate the heat of formation from its elements of an organic compound by subtracting its heat of combustion from the sum of the heats of combustion of the elements composing it. As an example : the heat of combustion of methane, CH 4 , at constant volume is 211,900 cals. ; the heat of combustion of carbon (C + 2 = C0 2 ) being 97,000 cals. and that of hydrogen (H 2 + O '= H 2 0) 68,400 cals., the complete combustion of methane is given by the following equation : CH 4 + 2O 2 = C0 2 + 2H 2 O = 97,000 + (2 x 68,400) = 233,800 cals., the sum of the heats of combustion of the component elements of methane. The heat of formation of methane will then be given by : 233,800 - 211,900 = 21,900 cals., which also represents the heat necessary to resolve methane into its elements in order to initiate its combustion. The heat of combustion of ethyl alcohol being 340,000 cals., that of acetic acid 210,000, and that of ethyl acetate 554,000, the heat of formation of the last named from the first two will be : 340,OQO + 210,000 - 554,000 = - 4000 cals. In the analogous paraffin and olefine series, a difference of CH 2 corresponds with a variation of 150,000-160,000 cals. in the molecular heat of combustion. The heats of combustion of isomeric compounds are equal, if they are chemically similar, for example, methyl acetate (CH 3 -CO 2 CH 3 ) and ethyl formate (H'C0 2 C 2 H 5 ), but different if the compounds are of different molecular character (for example, allyl alcohol, CH 2 : CH-CH 2 -OH, and acetone, CH 3 -CO-CH 3 ), compounds with multiple linking in the fatty series having higher heats of combustion than the corresponding cyclic isomerides. These calculations are also of importance for the evaluation of the energy produced in organisms by the transformations of various foods (see also later in the section on Explosives 1 ). HEAT OF NEUTRALISATION. With the organic acids this is the same for all, namely, 13,700 cals. (see vol. i, p. 97), as long as the resulting salts are not decomposed by water ; with the phenols (cyclic compounds containing OH) the heat of neutralisation is about one-half the above value, or more if the acid character is intensified by the presence of the NO 2 group ; with the alcohols it is almost zero. 1 The following are the heats of formation from the elements of certain organic compounds, expressed in large calories per gramme-molecule : Naphthalene, C 10 H 8 : solid . . . -42 Cals. Nitronaphthalene, C 10 H,NO 2 : solid . 14-7 Dinitronaphthalene, C 10 H 6 (NO 2 ) i : solid 5-7 Trinitronaphthalene, C 10 H 6 (NOji)3 solid. 3-3 Acetylene, C 2 H 2 : gas .... 61-4 Ethylene C 2 H 4 : gas . '..'*". . 15-4 Benzene, C,H, : gas . . .... . 10-2 Nitrobenzene, C,H 6 N0 2 : liquid . . 4'2 Dinitrobenzene, C,H 4 (NO 2 ) 2 : solid . 12-7 Mannitol, C e H 14 : solid . . . 320 Nitromannitol, C e H 8 N,O 18 : solid . . 179' 1 Mercury fulminate, C 2 N 2 Os,Hg : solid . 62-9 Anthracene, C 14 H 10 : solid . . Methyl alcohol, CH S OH : liquid . . 62 Cals. Ethyl alcohol, C,H 5 OH : liquid . . 70-5 Phenol, C,H 6 OH : liquid . . . 34-5 Trinitrophenol (picric acid),C,H 2 OH(NO 2 ), solid 49-1 Sodium picrate, CH 2 ONa(NO,) 8 : solid 117-5 80-] 65-3 72 Glycerol, C 3 H 6 (OH), : liquid . . . 165-5 Trinitroglycerol, C,H,(ONO 2 ) 3 : liquid . 196 ,, Cellulose (cotton), C 6 H 10 O 5 : solid . . 227 Ammonium picrate, solid Ether, (C 8 H.) 2 o{** s uid ; Nitrocellulose, solid 624-696-706 -42-4 and the heats oj combustion of various organic compounds are as follow : ethyl alcohol, 340 cals. ; methyl alcohol, 182-2 ; mannitol, 727 ; cellulose, 680 ; terephthalic acid, 771 ; diphenyl, 1494 ; cane sugar, 1355 ; acetic acid, 210 ; benzole acid, 772 ; ethyl acetate, 554 ; urea, 152 ; benzene, 779-8 ; dihydrobeniene, 848 ; tetra- hydrobenzene, 892 ; toluene, 933 ; hexane, 991-2 ; methane, 211-9 ; ethane, 370-4 ; propane, 529-2 ; trimethyl- methane, 687-2 ; ethylene, 333-4 ; propylene, 492-7 ; trimethylene, 499-4 ; isobutylene, 650-6 ; methyl chloride, 164-7 ; ethyl chloride, 321-9 ; propyl chloride, 480-2 ; chloroform, 70-5 ; dinitrobenzene (o-, m-, and p-), about 700 ; trinitrobenzene, 666 681 ; succinic acid, 357 ; azelaic acid, 1141 ; erucic acid, 3297 ; tribrassidinic acid, 10,236 ; glucose, 674 ; oxalic acid, 60-2 ; formic acid, 62-8 ; hydrocyanic acid, 152-3 ; naphthalene. 1233-6 ; phenol, 732 ; pyrogallol, 639. 26 ORGANIC CHEMISTRY OPTICAL PROPERTIES. (1) Colour. The majority of organic compounds are colourless, but if they contain iodine or the nitro -group or doubly linked nitrogen atoms ( N=N ), or two oxygen atoms directly united, they are generally coloured, especially in the aromatic series. In the section on Dyes are given detailed illustrations of the remarkable relations between the chemical constitution of organic compounds and their colour. (2) Refraction. This is the deviation produced in the direction of a ray of light (homogeneous ; for example, sodium light) on passing through a transparent liquid, and varies with the substance. The index of refraction n varies with the temperature, and hence with the specific gravity (d) of the substance. The relation between these two 1 1 R. values which gives the refraction constant R (or specific refractivity) is : which is almost independent of the temperature. n 2 +2.' d By multiplying by the molecular n 2 1 P weight P, the molecular refraction is obtained : M = -r ~ -5 , this being constant for n ~\~ d true isomerides and changing by a constant amount for a constant change in the composition. The molecular refraction of a compound is approximately equal to the sum of the elementary atomic refractions, but here double and triple linkings have an influence, so that these can be detected in an organic compound by means of the refraction (true double linkings of the aliphatic series are often distinguished in this way from the cyclic linkings of benzene). (3) Polarised Light. Owing to the importance of this phenomenon for whole groups of organic substances, it will be useful to recall briefly in a note 1 the fundamental ideas on polarised light. FIG. 40. 1 The luminous waves of white light are propagated in the cosmic ether with velocity of about 300,000 kilo- metres per second, and there are physical instruments which admit of the measurement of the time required for a ray of light to traverse a few metres ; indeed, Foucault measured the time taken by light to pass over a distance of 120 metres. By studying the phenomena of interference of light rays, it can be shown that the vibrations of the ether in them are not longitudinal, i.e. along the direction of propagation of the ray, but that the et'her particles vibrate in all directions in a plane perpendicular to. the direction of the ray (a transverse section of a ray is shown in Fig. 40), whilst the propagation of sound is effected by means of longitudinal vibrations in the direction of the path traversed by the sound. A ray that enters a liquid or a non-crystalline solid, or a crystal of the regular system (cube or octahedron) gives only one refracted ray ; when it enters a crystal of the rhombohedral system, two refracted rays are formed, one extra- ordinary and the other ordinary ; when a ray enters a crystal of any other system, two refracted rays are formed, but these rays both behave likp the extraordinary ray, and, like the latter, they do not obey the laws of refraction, according to which an incident ray, perpendicular to a medium with parallel faces, should not be deviated or refracted. If a ray of light, J i (Fig. 41) strikes a rhombohedral crystal of Iceland spar perpendicularly to the face ABCD, the ray divides into two. The one, ioO, continues in the same direction, the other, ie, is deviated, but when it emerges from the crystal assumes the direction e E, parallel to the original direction. The two parallel rays leaving the crystal have equal luminosities, but o follows the FIG. 41. O FIG. 42. FIG. 43. ordinary laws of refraction (vide supra) and is called the ordinary ray, whilst the other, eE, does not obey these laws and is termed the extraordinary ray. If the crystal is rotated about the incident ray Ji as an imaginary axis, the position of the ray o does not change, whilst the ray e E moves in the sense in which the crystal is rotated. The extraordinary ray i E always lies in the plane of the principal axis of the crystal dbBD, which passes through the principal axis of the crystal b O and is parallel to it. These two rays emerging from the crystal have, however, properties different from those of the incident ray J i ; in fact, if either of the two refracted rays (eE or o O) is passed into a second rhombohedron of Iceland spar, two new rays (double refraction) are obtained, but the intensities of the two rays vary according to the relative positions of the two crystals. Thus, if a ray emerging from the first crystal passes porpendiculaily into the second crystal, the principal section cf which is parallel to that of the first, no double refraction is observed, only one ray leaving the second crystal (s in Fig. 42, the second hypothetical ray n not being visible and marked black in the figure). If, however, the second crystal is rotated round the imaginary axis, o O, a second ray (extia- ordinary) suddenly appears, i.e. double refraction takes place, and whilst the luminosity of the new lay increases, OPTICAL ROTATION 27 Those organic substances are called optically active which rotate the plane of polarised light. Some substances are optically active in the crystalline state (not in the amorphous state or in solution), and hence the action on polarised light is due in these cases to the peculiar arrangement of the molecules ; very few are active in both the crystalline and amorphous states, the majority exhibiting activity only in a dissolved condition (sugars, &c.), where the phenomenon depends on the arrangement of the atoms or groups of atoms in the molecule. This holds also for camphor and oil of turpentine, which are active even in the form of vapour. The longer the layer (I) and the greater the concentration of the solution (p = grammes of dissolved substance in 100 of solution) traversed by the polarised light, the greater will be the rotation of the plane of polarisation. Referring the observed rotation a to a length of 10 cm. of a solution containing 1 grm. of pure substance in 1 c.c. ( = pdjlOO, where d is the specific gravity of the solution), we get the specific rotatory power of the solution for the yellow sodium light (D line of the spectrum x ) by means of the following 1 00 formula : fain = ; ; For active liquid substances examined without solvent, I'd- p [ a ] D = - . The molecular rotation (for a molecular weight M) is given by : [M] = For a definite solvent and given concentration and temperature, every active substance (and such are almost all those containing asymmetric carbon, see p. 26) has a constant and characteristic specific rotatory power, either to the right ( +) or to the left ( - ). This varies with the nature and degree of electrolytic dissociation of the solvent, and increases with dilution and diminishes with rise of temperature ; for purposes of comparison, it is usually determined at 20, and is then indicated thus : [a] D. By repeating the determina- tions and using moderately high concentrations, the influence due to the solvent is deter- mined and, on subtracting this, the true specific rotation is obtained. Freshly prepared solutions of certain sugars exhibit the phenomenon of muta-rotation, which, however, disappears after a time or on boiling the liquid, the normal rotation then being given. This important property of optically active compounds is studied by means of special apparatus termed polarimeters, which are used particularly in the analysis of sugars (and hence often called saccharimeters), and will be described in the section dealing with this group of substances. MAGNETIC ROTATORY POWER. All liquids in a magnetic field produce a greater or less rotation of the plane of polarised light, according to their chemical composition and in conformity with the laws governing the refractivity of light. In many cases the constitution of a substance has been determined or confirmed by determining the molecular magnetic rotation. ELECTRICAL CONDUCTIVITY. We must refer the reader to the detailed treat- ment of electrolytic dissociation and the theory of ions in vol. i (pp. 91 et seq.), as the same is directly applicable to organic compounds, especially as regards the conductivity of salts, acids, bases, &c. that of the first ray becomes weaker and when the principal sections of the two crystals form an angle of 45, the two rays have equal intensities (', n') ; if the crystal is rotated still more, the extraordinary ray becomes more luminous, whilst the first (ordinary) decreases in luminosity, and when the principal sections are perpendicular to one another, the intensity of the ordinary ray (*") is aero (i.e. it is not seen), only the extraordinary ray being seen with its maximum intensity (n"). The light rays emerging from the second rhombohedron are hence different from those emerging from the first, these latter not varying in intensity when the prism is rotated, whilst the others do so. The rays leaving the first prism are called polarised, and are distinguished from ordinary light rays, since, on passing through a second prism, they undergo the changes described above. A polarised ray passes as an ordinary ray through a second rhombohedron only when its plane of polarisation is parallel to the principal section of the new rhombohedron It ia found, then, that the plane of polarisation of the polarised ordinary ray is perpendicular to the plane of polarisation of the extraordinary ray. Hence, the rays E and O vibrate in planes perpendicular to one another (Fig. 43). POLARISATION BY REFLECTION. Polarised light rays are obtainable, not only by double refraction, but also by reflection under special conditions, namely, when a light ray falls on a plate of glass at an incident angle of 54 35'. Polarised light is also obtained by simple refraction, by passing a ray of light through a series of superposed parallel plates or sheets of tourmaline. 1 The angle of rotation varies with the length of the light- wave and is greater for violet rays (which have a smaller wave-length and are hence refracted more) and less for red rays (which have a greater wave-length and are hence less refrangible). 28 ORGANIC CHEMISTRY CLASSIFICATION OF ORGANIC SUBSTANCES These are usually divided into two large series : (1) That of the open-chain carbon compounds or methane derivatives, termed also compounds of the fatty or aliphatic series, as all the fats and many of their derivatives belong here. This series embraces two groups of substances ; that of the saturated compounds or derivatives of the paraffins (C n H 2n + 2 ) and that of the unsaturated compounds (olefines, C n H 2w and derivatives of acetylene, (2) That of the closed-chain carbon derivatives, this being subdivided into : (a) The isocyclic or carbocyclic compounds, which have the closed chain formed either of nuclei of six carbon atoms with six available valencies to every nucleus (C n H 2n _ 6 , benzene derivatives or aromatic compounds] or from nuclei with different numbers of carbon atoms, but more highly hydrogenated (cycloparaffins, cyclo-olefines, and polymethylene derivatives). (b] The heterocyclic compounds, the closed chain of which contains atoms (N, P, S, 0, &c.) other than carbon. The hydrogenated compounds of' carbon are called hydrocarbons and are termed saturated when the carbon atoms are joined by single valencies, and the other valencies are all satisfied by hydrogen. These saturated hydro- carbons cannot combine with a further quantity of hydrogen. Hydrocarbons containing carbon atoms united by double or triple linkings are called unsaturated hydrocarbons, and these can combine with further quantities of hydrogen, thus becoming saturated. Other important hydro- carbons are those with closed chains, which we shall study in Part III of this book. Usually in homologous series, with increase in the number of carbon atoms, the compounds pass from the gaseous to the liquid and solid states ; e.g. formic acid, with one carbon atom, is a liquid and boils at 99, while the homologous acid with six carbon atoms is a solid and boils at over 300. OFFICIAL NOMENCLATURE With the continuous development of organic chemistry and the multiplication of new compounds, the need was often felt for a rational method of naming compounds which would facilitate the treatment of these vast numbers of compounds. And for the new nomenclature to be the more efficacious it needed to be international, because everywhere there reigned the greatest confusion in the naming of chemical compounds, this referring either to the starting substance or to the new group to which they belonged, or to the use for which they were intended, or to the molecular constitution, and so on, so that the same substances often had four or five names. f". In 1892, at an International Convention of Chemists at Geneva, a general system of nomenclature of organic compounds was agreed on. This is gradually being introduced into chemical literature, and, although not always felicitous, it has helped to simplify the naming of compounds and to reduce the confusion. Following only in part the ideas proposed by Kolbe many years before, the new nomenclature derives the names of all compounds from the names of the fundamental hydrocarbons to which the compounds can be referred, taking into account the number of carbon atoms present as well as the nature of the linking. Thus, to the fundamental names of the saturated hydrocarbons : methane, ethane, propane, butane, pentane, hexane, heptane, &c., the addition of the suffix ol indicates the presence of the hydro xyl group OH, and thus an alcohol, for example, methanol (methyl alcohol), efhanol (ethyl alcohol), &c. ; the suffix al serves to denote the aldehyde group ( C^ ) , thus, e.g. methanal = formaldehyde, ethanal = acetaldehyde, &c. ; the suffix one indicates the OFFICIAL NOMENCLATURE 29 ketonic group ( CO ), thus, propanone (commonly called acetone), &c. The suffix oic is used to indicate the organic acids, which all contain the characteristic carboxyl ( / \ group I C0 2 H, i.e. C45 ^22^46 Docosane .... 44 224 0-778 f a ^23^8 Tricosane .... 48 234 ft rr. 0-779 ^WHso Tetracosane . . . 51 243 00 0) 0-779 "^ C 25 H 52 Pentacosane .... 53-5 ft a ^26^54 Hexacosane .... 58 ^a C 27 H 56 Heptacosane .... 60 270 a >o 0-780 C2sH 58 Octocosane .... 60 11 ^31^64 Hentriacontane 68 302 0-781 ^32^66 Dotricontane (Dicetyl) 70 310 0-781 ^35^12 Pentatricontane . " . 75 331 ) 0-782 / C 60 H 12 2 Hexacontane .... 101 32 ORGANIC CHEMISTRY Gas) ; also when petroleum residues are strongly heated under pressure (cracking), hydrocarbons similar to petroleum and also gaseous ones are formed. Of the numerous synthetical methods of preparation of the saturated hydrocarbons, the following more important ones may be mentioned : (a) Any member of the series can be obtained by reducing the halogen derivatives of the hydrocarbon (obtained from the alcohols and the halogen hydracids) by means of nascent hydrogen (generated by sodium amalgam, or by a solution of sodium in absolute alcohol, or by zinc and hydrochloric acid, or by heating zinc and water at 160) or by hydriodic acid, especially in the presence of red phosphorus (which transforms the iodine into hydriodic acid) : C 2 H 5 I + H 2 = HI + C 2 H 6 ; C 2 H 5 I + HI = I 2 + C 2 H 6 (see Table of the halogen derivatives of the hydrocarbons). (b) The alcohols give paraffins on being heated with hydriodic acid : C 2 H 6 -OH + 2HI = H 2 O + I 2 + C 2 H 6 . (c) By the interaction of zinc alkyls and water : Zn(C 2 H 6 ) 2 + 2H 2 = Zn(OH) 2 + 2C 2 H 6 . (d) From unsaturated hydrocarbons by the action of hydrogen, e.g. by heating acetylene and hydrogen at 400*-500, or in presence of platinum -black. (e) By eliminating a molecule of carbon dioxide from the organic acids and salts by heating with soda-lime or sodium alkoxide : CH 3 .COONa (sodium acetate) + NaOH = Na 2 C0 3 + CH 4 . (/) By the action of sodium or of zinc on the zinc alkyls or alkyl iodides in ethereal solution in a closed tube (Wurtz), two alkyl groups, even different ones, being condensed : (1) 2CH 3 I + Na 2 = 2NaI + C 2 H 6 . (2) C 2 H 6 I + C 4 H 9 I + Na 2 = 2NaI + C2H 5 .C 4 H 9 . (3) 2CH 3 I + Zn(CH 3 ) 2 = ZnI 2 + 2C 2 H 6 . (g) During the last few years it has been shown that magnesium is much more active than zinc in many organic syntheses (see later, Grignard Reaction), and with alkyl iodides dissolved in absolute ether, magnesium forms magnesium alkyl salts which, on decomposi- tion by means of water or dilute acid or ammonia with solid ammonium chloride, yield the saturated hydrocarbons : C 2 H 6 I + Mg = C 2 H 5 MgI, and this + H 2 O = C 2 H 6 + Mg(OH)I. In part, however, the magnesium fixes the halogen, and then two alkyl residues condense, forming a hydrocarbon of double the number of carbon atoms : 2C 2 H 6 I + Mg = MgI 2 + C 4 H 10 . (h) Sabatier and Senderens' catalytic process, for which see pp. 34 and 59. (i) By electrolysing acetic acid : CH 3 COOH CH 3 + 2C0 2 + H 2 CH 3 COOH CH 3 the hydrogen going to the negative pole and the hydrocarbon and carbon dioxide to the positive one. METHANE (MARSH GAS), CH 4 This is a gas which is often found ready formed in nature, and in former times it was always confused with hydrogen .(inflammable air). Pliny refers to the gases which exude from the earth in certain regions and are inflam- mable (these are probably the sacred fires of the ancient Chaldeans). Basil Valentine (1500) records fires in mines preceded by the emanation of asphyxi- ating, poisonous vapours, which are dispersed and rendered innocuous by the fire issuing from the rock. Also Libavius (1600) speaks of the inflammable and explosive gas of mines, and in 1700-1750 history records numerous explosions, especially in coal-mines. In was Volta who, in 1776, when studying METHANE 33 the same gas, which is also evolved in marshes, showed that it differed from hydrogen, since in burning it requires double its volume of oxygen and forms carbon dioxide. In 1785 Berthollet proved that the gas is formed of carbon and hydrogen, and later Henry, Davy, and Berzelius determined its true composition. It occurs abundantly as exhalations from the earth near the Caspian Sea (sacred fires of Baku) and in the peninsula of Apsheron is used for heating purposes. At Pittsburg there are great wells of pure methane, and it is found also at Glasgow, in the Crimea, and also in Italy, at Pietra Mala (Bologna), in Ferrarese, in Piacento (Salsomaggiore), &c. It always occurs in coal-mines, being formed from the coal by slow decomposition and remaining occluded in the coal under great pressure, together with carbon dioxide and nitrogen. It is invariably developed in marshy places, where there is organic matter putrefying under water. It is found in the gas of the intestines of man and, still more, of the ruminants (about 50 per cent. CH 4 ), being produced by the action of enzymes on the cellulose of vegetable matter. Illuminating gas contains up to 40 per cent, of it. PROPERTIES. It is one of the permanent gases (vol. i, p. 28) ; it liquefies at 164 and solidifies at 186. It has no colour or taste, but a faint garlicky odour. It dissolves slowly but appreciably in fuming sul- phuric acid, but only very slightly in water (0-05 per cent.). It is readily inflammable and burns with a faintly luminous flame ; mixed with oxygen it forms a detonating mixture (inflammable at 667, whilst the mixture with ethane inflames at 616 and that with propane at 547), the maximum effect being obtained with 1 vol. of methane and 2 vols. of oxygen (CH 4 + 20 2 = C0 2 + 2H.JO). 1 Mixed with air, it forms the firedamp of coal-mines, which is very dangerous owing to its explosibility, 2 although it is not poisonous since miners can withstand an atmosphere containing 9 per cent, of methane ; if there is not more than this proportion, it produces a kind of pressure at the forehead, which ceases immediately on breathing pure air. By an electric charge or in a red-hot tube, it decomposes into carbon and 1 Explosive gas mixtures (Teclu, 1907) : Minimum effect With excess With deficit Vols. Vols. Vols. 100 volumes of air + hydrogen 40 170 8-10 , + methane 10 3-6 , + coal gas . 17-20 31 4-7 + acetylene 8-3 130 2-4 + ether vapour 3-3 8 1-5 , + alcohol vapour 6-5 3-4 2 Since the methane is occluded under great pressure between the layers of coal, its development and hence also the danger is greater when the atmospheric pressure diminishes or when the temperature rises. To prevent explosions of firedamp, the miners use the Davy lamp (vol. i, p. 377). Considerable danger of explosion more often exists in mines owing to the coal dust suspended in the air of the galleries and behaving Jike a pyrophotic substance (vol. i, p. 174) ; as a precautionary measure, air is continually circulated through the galleries by powerful fans, and the air and the walls are moistened by means of pulverisers. Hardy has constructed an apparatus which allows of the quantity of methane being determined from the sound produced by the mixture of air and methane in traversing an organ pipe. Mines containing much dust are dangerous even if the atmosphere is moist and the Davy lamp is used, since the particles of coal passing through the gauze into the lamp may issue in a red-hot condition. When mines are being excavated, safety explosives (which see) are used to avoid fires and explosions. Sometimes the coal ignites in certain parts of the mine ; in such cases, work is not suspended, but these parts are isolated by walls and if the fire becomes threatening, recourse is had (usually with success) to the sealing of the mine and subsequent inundation with water or filling of the galleries with carbon dioxide. When an explosion occurs in a mine, a large amount of carbon monoxide is formed which poisons the workers, who can, however, sometimes be rescued if they can be made to breathe, sufficiently promptly, under a bell containing compressed air (Mosso's Method ; vol. i, p. 175). n 3 34 ORGANIC CHEMISTRY hydrogen, and a few unsaturated hydrocarbons, with traces of benzene and naphthalene. PREPARATION IN THE LABORATORY. Besides by the general methods given above, methane is formed by passing a mixture of carbon monoxide or dioxide with hydrogen over reduced nickel (catalyst) heated at 250 (Sabatier and Senderens) : CO + 3H 2 = H 2 + CH 4 . Attempts have recently been made to put this method on an industrial basis, by transforming the carbon monoxide and dioxide of water-gas into methane (Ger. Pat. 183,412). Pure methane is formed by passing a mixture of carbon disulphide vapour and hydrogen sulphide over red-hot copper (Bert helot) : CS 2 + 2H 2 S -f 8 Cu = 4 Cu 2 S + CH 4 ; also by treating aluminium carbide with water : C 3 A1 4 + 12H 2 O = 4A1(OH) 3 + 3CH 4 . In the laboratory it is usually prepared from an intimate mixture of one part of crys- talline sodium acetate with four parts of soda lime (or better, with four parts of baryta or with a mixture of anhydrous sodium carbonate and dry powdered calcium hydroxide). This is heated in a retort or in a hard glass tube until gas begins to be evolved, the tem- perature being then kept constant. As impurities, it contains a little hydrogen and acetylene, so that, before collecting the methane, the gas is passed over pumice moistened with concentrated sulphuric acid. Chemically pure, it can be obtained, by the general method, from zinc ethyl and water. INDUSTRIAL USES. For several centuries, the inflammable gases issuing from the earth and from petroleum have been utilised at Baku for heating lime-kilns. In North America, as far back as 1821, these natural emanations were used as illuminating gas. The most important discoveries, made at Pittsburg in 1882, resulted in 98 per cent, of the American production being obtained from this source in 1900 ; after 1905, the wells of Louisiana also acquired importance. The utilisation of the gas at the present day is carried out rationally and on a vast industrial scale, the gas (issuing from suitably constructed wells) passing to large gasometers which distribute it directly to over 500 factories and 40,000 houses, where it is employed for power, heating, and lighting (with the Auer mantle), the price being about 3^ cents per cubic metre. In Canada, 400 wells are being used, giving, in 1907, gas of the value of 120,000. In England, wells have been sunk since 1900 which yield 400,000 cu. metres of gas per day. The spring at Wels, in Austria, which gave 57,000 cu. metres of gas per day in 1894, yielded only 500 cu. metres in 1901. The gas utilised in the United States of America represents the following values in pounds sterling : in 1882, 40,000 ; in 1890, 1,400,000 ; in 1894, 2,800,000 ; in 1899, 4,000,000 ; and in 1906, 9,600,000. These gases have the sp. gr. 0-624-0-645, and a calorific value of 9000-10,000 cals. per cu. metre. The composition varies between the following limits : CH 4 , 80-95 per cent. ; H, 0-5-1-5 per cent, (some- times 15 per cent.) ; C 2 H 4 , 0-3-4 per cent. ; CO, 0-0-6 per cent. ; C0 2 , 0-3-2-5 per cent. ; 0, 0-35-0-80 per cent. ; N, 0-5-3-5, together with traces of H 2 S. The amounts of natural gas used at Baku were 46-5 million cu. metres in 1905, 96-3 million in 1906, and 117 million in 1907, the composition being: C0 2 , 3-3-8 per cent. ; C,,H W , 1-2-2-6 per cent. ; 0, 7-7-6 per cent. ; CH 4 , 54-8-60-2 per cent. ; H, 13-58-0-8 per cent. ; and N, 20-4-25 per cent. The gas which is used at Salsomaggiore (Piacenza) for public lighting purposes and which issues from the earth together with petroleum and saline waters containing iodine, has a specific gravity of 0-692, and the following composition (Nasini and Anderlini, 1900) : CH 4 , 68 per cent. ; C 2 H 6 , 21 per cent. ; heavy hydrocarbons, 1 per cent. ; N, 8 per cent. In Italy, 1 ,520,000 cu. metres of these gases, of the value 2280, were used altogether in 1902, 6,737,500 cu. metres in 1908, and 8,270,000 cu. metres, of the value 8760, in 1909. Important sources of these gases have been recently discovered in Hungary, England (at Heathfield a well gave as much as 500,000 cu. metres per day), and in Denmark (since 1872). ETHANE, C 2 H 6 This gas is found dissolved in crude petroleum and is one of the principal constituents of the North American gas-wells of Delamater, near Pittsburg. It is a gas which can be liquefied at by means of a pressure of twenty-four atmo- spheres and then has a sp. gr. 0-446 ; at the ordinary pressure it becomes liquid and boils PROPANE, BUTANES, PENTANES, ETC. 35 at 84 and is solid and melts at -172. It is almost insoluble in water ; 1 vol. of absolute alcohol dissolves 1^ vol. of it. It burns with, a faintly luminous flame, and is more readily soluble than methane. In the laboratory it is prepared by the general methods already given (p. 32). PROPANE, C 3 H 8 (METHYLETHYL, CH'C 2 H 5 or DIMETHYLMETHANE, CH 2 (CH 3 ) 2 ) This is a gas like ethane and becomes liquid at - 44, or at under five atmospheres pressure, the liquid at having a sp. gr. 0-535 ; it solidifies and melts at 45. It is slightly soluble in water, and absolute alcohol dissolves 6 vols. of it. With water under pressure and at temperatures below it forms a solid hydrate, which decomposes at + 8-5. The illuminating power of propane is about 1^ times that of ethane. It is best prepared, in the laboratory, by reducing isopropyl iodide by means of the copper- zinc couple, or by reducing acetone or glycerol with hydriodic acid : C 3 H 5 (OH) 3 + 6H = 3H 2 + C 3 H 8 or CH 3 -COCH 3 + 4H = H 2 + C 3 H 8 . glycerol acetone BUTANES, C 4 H 10 (Two Isomerides) (a) Normal Butane, CH 3 CH 2 ' CH 2 CH 3 (diekhyl), is a gas which liquefies at + 1, and at has a sp. gr. 0-600. It is found in Pennsylvanian petroleum, and is prepared in the laboratory by the ordinary methods (p. 32). PTT (b) Isobutane, CH 3 CH-c^Tj 3 (trimethylmethane or methylpropane), is a gas which t^ 3 becomes liquid at 115 ; it is contained in petroleum and is prepared by the usual methods in the laboratory. PENTANES, C 5 H 12 These hydrocarbons are found especially in the petroleum products boiling a little above 0, and are placed on the market under the names of rhigolene and cymogen for anaesthetic purposes and for the manufacture of artificial ice. The three isomerides predicted by theory are known : (a) Normal pentane, CH 3 - [CH 2 ] 3 'CH 3 , is a colourless, mobile liquid boiling at + 37-3, having a sp. gr. 0-454 at 0, and solidifying only at about 200 ; it is hence used for making low-temperature thermometers, and as a lubricant in the Claude liquid air machine (vol. i, p. 298). It occurs abundantly in Pennsylvanian petroleum. (b) Isopentane, CH 3 CH CH 2 CH 3 (methyl-2-butane or ethylisopropyl), is a light CH 3 colourless liquid boiling at 30-4, and having a sp. gr. 0-622 at 20. It is found in large quantities in petroleum, and can be prepared artificially from isoamyl iodide by the ordinary methods (p. 32). pTT PTT (c) Tetramethylmethane, riT] 3 ^>V<^ nri 3 (dimethyl-2-propane), ie found in the gases from petroleum, and is liquid at + 9 and solid at 20. It can be obtained in the laboratory either by chlorinating acetone, CH 3 'CO'CH 3 , by means of phosphorus pentachloride and treating the dichloropropane thus formed with zinc-methyl : CH 3 Cl CH 3 _ CH 3 CH 3 CH 3 > < CH 3 .CH 3 , CH 3 CH 3 I + Zn 8 to 13 per cent, on the capital. 54 ORGANIC CHEMISTRY To estimate the ammonia in the purified gas, 200 litres of it are passed through 10 c.c. of an N/10 solution of hydrochloric acid, the excess of which is subsequently determined by titration. The determination of the total sulphur compounds can be simply effected, according to F. Fischer, as follows : About 50 litres of the gas (measured by a good meter) are burned in a small Bunsen burner, g (Fig. 72), in the drawn-out bulb, A, of a bulb-con- denser arranged as shown. All the sulphur of the sulphur compounds burns, forming sulphurous and sulphuric acids with the water from the combustion of the gas, this condensing in the bulbs of the condenser and being collected at the bottom in a beaker by means of the tube e. The com- bustion is regulated so that gas containing 4-6 per cent, of oxygen escapes at o. Water FIG. 72. enters the condenser at z and leaves at n. At the end of the operation, the bulbs are rinsed out with water and the sulphurous acid in the liquid oxidised by means of pure, neutral hydrogen peroxide solution ; the. sulphuric acid is then titrated with N/10 sodium hydroxide solution. If the sulphuric acid is estimated gravimetrically with barium chloride, the oxidation of the sulphurous acid must be effected with hydrogen peroxide free from sulphates. The quantity of sulphuric acid found gives the total sulphur-content of the gas. A well -purified gas f contains less than 0-5 grin, of sulphur per cubic metre. The hydrogen sulphide is estimated separately by passing a known volume of the gas through ammoniacal silver nitrate solution, which is afterwards acidified with a little nitric acid, the silver sulphide being filtered off, washed, dried at 100, and weighed. The calorific power can be determined fairly rapidly by means of the Junker calorimeter (Fig. 73, section, and Fig. 74), which consists of a metal cylinder, C (the letters refer in all cases to Fig. 74), which is mounted on three feet, and inside which a known volume of the gas is burned by means of the Bunsen burner, n. The hot products of combustion pass several times up and down the calorimeter and issue at the outlet S, which is furnished with a valve and also regulates the air- draught. Passing in a direction opposite to that of the gases of combustion and in alternate adjacent chambers is a current of water which enters by w the small reservoir m, the excess being carried off by the overflow, 6, while a regular stream passes through the tap e (furnished with an indicator) into the calorimeter at a temperature given by the thermometer x, and flows away at c at a higher temperature, shown by the ther- mometer y. When the combustion is started, the entry of water is regulated by means of e, so that the thermometers, x and y, indicate a temperature difference of 10-20 ; when the flow of both gas and water is constant, the thermometer y soon shows a constant temperature. The gas is measured by the meter, G, and then passes through the regulator, P, to the Bunsen burner, n, which is drawn from the calorimeter to be lighted and is then pushed in again to the height q (about 6 in. up). If the apparatus is in order, no water should fall from d into the cylinder, v. When water is discharging from b and from c, and the thermometer remains stationary, as soon as the index of the meter reaches the zero mark or a definite number of litres, the rubber tube c is instantly placed from t into V, which is a graduated cylinder placed quite close to the discharge-funnel, t. In the cylinder V is collected all the water which is discharged during the combustion of a definite volume of gas (in the proportion of FIG. 73. CALORIFIC POWER OF GAS 55 100 to 200 litres of illuminating gas or 400 to 800 litres of suction gas or Dowson gas per hour). Exactly at the moment when the meter indicates the volume of gas fixed upon, the rubber tube, c, is removed from V to t. During the course of the experiment the small variations in the indications of the thermometer y are noted at intervals, the mean temperature being subsequently determined. The graduated cylinder, v, contains the condensed water (a c.c.) formed during the combustion of the gas, and this, in condensing, has given up to the water of the calori- meter a certain quantity of heat, which must be subtracted before calculating the net calorific power. The gross calorific power, U, expressed in calories per cubic metre, is A. T. 1000 calculated by means of the formula : U = , where A indicates the quantity of water in litres collected in V, and Q the volume of gas burned. If, for example, Q = 3 litres, A = 0-900, T = 18 (that is, 26-77, the mean of six readings of the thermometer y, less 8-77 shown by the thermometer x to be the temperature of the water enter - 0-900.18.1000 ing at e), we have U = = 5400 Calories per cubic metre of gas. In cases where the gas is used in engines or other apparatus from which the products of combustion issue at a temperature above 65, the water-vapour does not condense and the gross calorific power must be diminished by. the heat FIG. 74. due to the condensation of the water -vapour produced by the combustion of the gas in the calorimeter. From U must hence be subtracted a value obtained by multiplying by 60 the number of c.c. of water condensing during the combustion of 10 litres of gas. This net calorific power, U 1 , is, for illuminating gas, usually 10 per cent, lower than the gross calorific power, U. The specific gravity sometimes serves to test the constancy in composition of gas or to compare two different gases ; it also gives a rough idea of illuminating power, since the specific gravities of the more highly light-giving hydrocarbons acetylene (0-920), ethylene (0-976), propylene (1-490), and benzene (2-780) are higher than those of the non-luminous components hydrogen (0-0695), methane (0-559), &c. The specific gravity can be determined rapidly and exactly with the Bunsen effusiometer (see vol. i, p. 39). ILLUMINATING POWER. There is no absolute measure of the power of different sources of light, but these can be compared when a conventional unit has been chosen. This standard of light has been differently chosen in different countries and has been 56 ORGANIC CHEMISTRY. continually modified. Thus in England spermaceti candles are used of such size that six weigh 1 lb., while, when burned, they lose 7-78 grms. (120 grains) per hour with a flame 45 mm. in height. In Germany in 1872 a paraffin candle 20 mm. in diameter was employed, the wick having 24 threads and weighing 0-668 grm. per metre and the flame being 50 mm. high ; six of these candles weighed 1 lb. Use is now made in Germany of the more rational Hefner-Alteneck lamp, fed with a liquid of constant composi- tion, namely, amyl acetate, the compact wick, 8 mm. in diameter, protruding 25 mm. FIG. 75. FIG. 76. from the metallic sheath holding it ; the flame is 40 mm. high. In France and Italy the Carcel lamp is used, this consuming 42 grms. of purified colza oil per hour and having a wick which is 23-5 mm. in diameter, is formed of 75 threads, and weighs 3-6 grms. per 10 cm. The relative values of these different units is as follows : 1 Carcel = 9-600 English candles (spermaceti) = 8-768 German candles (paraffin) = 10-526 Hefner-Alteneck flames. The luminous unit being fixed, different sources of light and their illuminating powers can be compared by means of photometers. Of these, the one most largely used is that of Bunsen, which is based on the principle that the intensity of light produced on a definite surface by a source of light is inversely proportional to the square of the distance. If the distance between the source of light and the surface illuminated is trebled, the intensity of the illumination is diminished to one-ninth of its previous value. The luminosities of two flames, / and /!, which illuminate equally a given screen and are at the respective dis- tances, L and L l9 from it, are directly proportional to the squares of these distances : / : /j L z : Lj 2 , and if I I is the unit of measurement, the intensity of the other source L 2 of light will be : / = -=-' The Bunsen photometer (Fig. 75) consists of a horizontal iron L \ photometer bench 3 metres long and divided decimally (into half -centimetres or milli- metres) ; at one end is placed the comparison electric or candle lamp or the Carcel lamp, the consumption of oil in which is regulated by a small pump actuated by a clockwork mechanism, weighing on a balance the consumption in a given time (indicated by a bell) this corresponding with 42 grms. of oil per hour. A screen of paper can be moved backwards and forwards along the bench and normally to it, the middle of the screen being rendered translucent by means of a grease spot (spermaceti) ; at the other end FIG. 77. I L - G A S 57 of the bench is placed the light to be examined. When the screen is equally illuminated on its two faces, the grease-spot is no longer perceptible. The intensities of the two sources of light are then proportional to the squares of their distances from the screen. The measurement is made in a dark room and, in order to render more evident the disappearance of the spot on the two surfaces, the screen is placed between two mirrors arranged at an angle (Fig. 76). An improvement on the Bunsen photometer has been made by Lummer and Brodhun, who substitute for the screen with the grease-spot a closed box, h (Fig. 77), in which are two opposite circular apertures, these illuminating the two faces of a white screen, /, by means of light from the standard lamp, and that to be tested placed at the two extreme ends of the photometer bench. By means of a system of prisms, A B, the two faces of the white screen reflect the light on to two concentric zones of the field of the eye-piece, r. When the two faces of the screen are equally illumi- nated, the two zones of the field also appear uniformly lighted . x OIL-GAS In cases where the installation of a plant for the carbonisation of coal would be inexpedient, owing to the small consumption of illuminating gas, it may be convenient to prepare oil-gas by dropping into a red-hot retort (see later, " Cracking " Process in the Petroleum Industry) fatty residues, tar oils, resins, and petroleum. This destruction by heat produces a gas which can be readily compressed without separation of liquid, and, enriched with 25 per cent, of acetylene, is used for the illumination of railway carriages. Oil-gas can also be prepared easily and abundantly by dropping oil into gasogens con- taining red-hot coke. As early as 1815 public lighting with oil-gas was attempted (Liverpool used it for some years), but it was only after 1860-1870 that this industry assumed importance. From 100 kilos of lignite paraffin oil are obtained 60 cu. metres of gas, and with a consumption of 35 litres of the gas per hour, 7-5 normal candles (German) are obtained ; its illuminating power is four times as great as that of ordinary lighting gas. If a greater yield of the gas is obtained, it loses in illuminating power. The purification of oil -gas is carried out in practically the same way as that of coal-gas. Mineral oil for gas and for engines is produced in large quantities in Galicia, where it is sold for less than 19-5d. per cwt. (4 lire per quintal) ; Germany alone imported 30,000 tons of it in 1909. 1 Comparison between Various Sources of Light. To produce the luminous intensity of a Hefner candle- hour (HK), the following quantities of lighting materials must be consumed : ,-Stearine, first quality ,, third ... - Paraffin Two parts of paraffin and one of \. stearine ..... 'Carcel: colza oil .... CH 2 ) form 80 per cent, of Baku petroleums X CH 2 CH/ and occur abundantly in those of Galicia, together with about 10 per cent, of hydrocarbons of the aromatic series (recently (1910) hexahydrocumene has been identified). In a Russian petroleum and also in a Roumanian one, Molinari and Fenaroli (1908) found hydrocarbons derived from naphthenes with two double linkings and having the general formula C n H 2M _ 14 (for example, C 17 H 20 ). In certain petroleums small quantities of acetylene derivatives occur. It is found that petroleums produced in localities relatively near to one another often have different compositions ; according to David Day this is due to the fact that the unsaturated hydrocarbons diffuse less easily through sandy or other soils, and this system of natural filtration gives rise to various types of petroleum, with preponderance of saturated hydrocarbons in some and of unsaturated hydrocarbons in others. This explanation is more reason- able than that the separation has been effected by distillation. The products that distil below 180 are almost exclusively saturated and those distilling about 200 mostly unsaturated. The very small quantities of oxygenated substances contained in petroleum (often less than 1 per cent, and rarely 5 per cent.) are composed of phenols and organic acids (e.g. in Galician petroleum). The traces of nitrogenous substances found in various petroleums (see above) support the hypothesis of the organic origin of petroleum. Almost all petroleums contain sulphur, which is very difficult to remove and imparts an unpleasant odour and bad colour. Usually the proportion of sulphur is about 0-10-0-15 per cent., but the petroleum of Terra dl Lavoro contains as much as 1-3 per cent., while still more is found (up to 3 per cent.) in those of Texas, Ohio, Indiana, and Virginia, from which it has to be separated (see later). The nature of the sulphur compounds present has not yet been completely denned, but the presence of mercaptans, thio-ethers, thiophene, and its homblogues (methyl- and dimethyl-thiophene) has been detected. According to Heusler it is only necessary to heat a little of the petroleum with a granule of aluminium chloride to detect the presence of sulphur, hydrogen sulphide being then developed. Also by fractional distillation and partly by the specific gravity, the four princpal types of petroleum can be distinguished. The products distilling below 150 form the b?nzinis (see later), then up to 280 are obtained illuminating petroleums or solar oil (or kerosene), and after 300 remain products used for the extraction of paraffin and vaseline (American) or for the preparation of mineral lubricating oils (Russian) : Crude petroleum Specific gravity Benzine Solar oil Residue Pennsylvania . . 0-79-0-82 10-20% 55-75% 10-20% Ohio . . . 0-80-0-85 10-20% 30-40% 35-50% Caucasus . Roumania Galicia Piacenza . Alsace 0-85-0-90 0-2-5 % 25-30 % 60-65 % 0-85 3-10 o/ 70-80 % 10-15% 0-82-0-90 5-30 % 35-40 % 30-50 % 0-74-0-79 25-40 % 55-65 % 4-8 % 0-912 5% 35-70 % 55-60 % 64 ORGANIC CHEMISTRY In some of the islands of the Caspian Sea (Tscheleken) is found a petroleum resembling the American type, with a large proportion of paraffin (5-5 per cent.), and in Columbia (S. America) petroleums like those of Russia (Caucasus) occur. The Italian 'petroleums vary considerably in composition and those of Emilia and Piacenza are so pale and so rich in benzine and poor in residues that it is supposed that they are the condensed more volatile products of more important deposits not yet dis- covered. In the distillation of the Velleia petroleums at Fiorenzuola d'Arda the little residue obtained is added to the crude petroleum to be refined and thus becomes distributed in the lighting oil, so that the less remunerative residues are never placed on the market. The absence of optical activity in the petroleums of Montechino and Velleia (see above) seems to confirm the view that they are derived from more important deposits, in which optically active products would probably be found. EXTRACTION AND INDUSTRIAL TREAT- MENT OF PETROLEUM. From the most remote times petroleum has been raised in China by means of wells similar to the pre- sent artesian ones, which the Chinese used many centuries before Europeans for obtaining drinking water. In other regions in times gone by the petroleum flowing at the surfaces of the water- courses began to be sepa- rated and used ; then wide, shallow wells were dug and the petroleum raised to the surface in buckets. Nowadays, how- ever, petroleum is every- where obtained by wells bored into the earth like artesian wells, and some- times the petroleum flows up to the surface under great pressure, so that it forms a fountain (see Note, p. 65, and Fig. 78). It is supposed that the deposits of petroleum in the interior of the earth's crust are situated in large cavities or pockets, where there is often a lower layer of salt water (Fig. 79, W ) and on this floats a more and less abundant layer of petroleum, E ; and, in general, the upper part of the pocket is filled with inflam- mable gas, G, which exerts great pressure. If the boring, B, reaches one or the other layer, one or the other product is obtained in preponderance or even exclusively, and, after exhausting the aqueous layer, the same well may yield only petroleum. x The sinking of a well is begun with a boring 35-40 cm. in diameter by means of suitable boring tools worked by long rods and toothed gearing, or by compressed-air drills mounted BORING FOR PETROLEUM 65 FIG. 79. on wooden structures termed derricks (Fig. 80) ; the detritus of the bored rock is con- tinually carried away from the boring by a current of water, whilst in former times the much slower dry boring was preferably employed. When the petroleum layer is approached, the water of the well or tube begins to show drops of petroleum. The power is often supplied by portable steam-engines, which should not be placed too near the boring, since if the petroleum or gas escapes accidentally in any quantity during the boring, it may ignite and cause considerable damage by fire or explosion. In such cases it is hence advisable to transform the energy on the site, for instance, with electric motors. And even then fires and explosions have been caused by the accidental ignition of the gas mixed with air, by sparks formed by stones, issuing violently from the well along with sand and petroleum and striking the iron framework or the rails of the woodwork. 1 When the petroleum is not exuded under pressure, it is often raised by means of pumps, but this is not possible where much sand (up to 30 per cent.) is also extracted and has to be allowed to deposit ; this is the case at Baku, where, however, one-third of the petroleum issues under pressure. During recent years there have remained relatively few " fountains " at Baku, and the petroleum of the sandy wells, which cannot be raised by pumps, is extracted by special " bailers " made of a cylinder of sheet- metal terminating in a cone and fitted in the lower portion with a valve which opens when the bailer (called a shalonka) becomes immersed in the petroleum and closes on raising by means of pulleys and windlass, the steel rope carrying the bailer being wound round a large drum a short distance from the well. The shalonka, containing some hecto- litres of petroleum, is discharged by inverting it over a channel. From the large reservoir near the wells, the petroleum passes by means of iron pipes to the refineries or to the despatching stations (suitable trains or vessels), which at Baku are very near, but in America some hundreds of kiloms. from the wells ; these pipes then traverse plains, mountains, and valleys, and in the same way and with the help of powerful pumping-stations, the refined petroleum is despatched to the place of loading. In 1905, the Standard Oil Company began the construction of another such pipe (pipe-line) to connect FIG. 80. the works at Kansas City with the coast ; the distance is about 1700 miles and the construction cost 880,000 and served to transport daily from 10,000 to 15,000 barrels of petroleum. 1 Artesian wells for extracting petroleum have an average diameter of 25 to 50 cm., and vary in depth according to the region ; at Baku they were first of all 60 to 150 metres deep. But of recent years wells have usually been sunk to a depth of 250 to 350 metres (occasionally 1000 metres). In the Washington district of the United States the wells are from 700 to 850 metres, and near Pittsburg is the deepest of all, 1820 metres. The wells are 100 to 200 metres apart according to the locality, and they remain active for five to ten years. The expense of boring varies with the district, that is, with the nature of the subsoil, and, under favourable conditions and for wells not too deep, each boring costs about 400. Those made in the Washington district cost even 1400 to 1600. At, Velleia in the province of Piacenza, the wells are little more than 100 metres deep, whilst at Salsomaggiore they have been bored to a depth of 400 metres, and in one case of 700 metres, in order to utilise for medical purposes the iodine-salt water whiphis obtained, together with alittle petroleum. Jn America the ii 5 66 ORGANIC CHEMISTRY Boiling-point Specific gravity 40-70 . . 0-635-0-660 70-80 . . 0-660-0-667 80-100 . . 0-667-0-707 100-120 .. 0-707-0-722 120-150 .. 0-722-0-737 150-200 200-250 250-300 above 300 0-753-0-864 FIG. 81. DISTILLATION. Crude petroleum cannot be used as it is for lighting, as it has a bad smell and colour, contains many impurities, and is composed partly of too volatile products, which might easily cause explosions or fires in the lamps. In order to avoid these dangers, the petroleum is subjected to exact refining, which is controlled by legal enactments and with special apparatus (see later). The refining is carried out in a manner which varies with the nature of the petroleum and usually consists of a fractional distillation and a chemical purification. The fractional distillation in the laboratory is carried out in Engler flasks (Fig. 81 ), which are of definite size and shape and permit of concordant results being obtained in all labo- ratories ; the following fractions are then weighed separately : I. Light or readily volatile petroleums : (a) Petroleum ether ..... (b) Gasolene ...... (c) Benzine , . . . (d) Ligroin (burnt in special lamps for lighting) (e) Petroline (used for de-fatting or cleaning) . II. Petroleum for lighting : I quality ....... II quality ...... III quality ....... III. Residues of the distillation : (a) Heavy oils : lubricating oils . . (b) Paraffin oil . . . . (c) Coke The industrial refining of petroleum consists in separating the crude petroleum into these three groups, I, II, and III. Apparatus is used for periodic or alternate distillation, or for continuous distillation. Periodic distillation is conveniently carried out in the so-called waggon-still largely used in America and at Baku. It holds as much as 2500 barrels at a time (Figs. 82 and 83). It is made of wrought iron 10-14 mm. in thickness, and has a corrugated bottom ; it is commonly 7 metres long, 4 metres wide, and 3 metres deep. The top is fitted with three flanged elbows which carry off the vapour. In thirty hours three distillations can be carried through, the residues being discharged through the three orifices, c. The heating is effected by means of these residues, which are forced into perforated pipes, r, in the double-arched hearth ; rational circulation of the products of combustion results in effective utilisation of the heat. More profitable use is made to-day of simpler, cylindrical boilers, which, although of larger dimensions, correspond almost exactly with the various types of steam-boilers, the heating being external, or lateral, or internal, or two of these together. Such boilers well is widened at its lowest jpoint,' where it| meets ,the] petroleum, by 'exploding a dynamite cartridge ("tor- pedoing ") A well sunk in 1891 at Balakhany, 270 metres deep, gave an uninterrupted jet producing 3276 tons of petroleum per twenty-four hours, and the mass of sand expelled covered the whole neighbourhood. A little distance away one of the Nobel Company's wells, in 1892, gave 13,000 tons per day. In February 1893 a well was sunk at Romany, near Baku, which for several weeks yielded 10,000 tons of petroleum per day ; the oil issued from the earth with such violence that the movement of the air broke the windows of neighbouring houses, and, as at first it was not possible to guide the jet into horizontal channels, all the iron plates used for this purpose being pierced, 250,000 tons of petroleum were lost in five weeks In 1909 a new well at Baku gave, for a long time, 3500 tons of naphtha per day. A well bored at Maikop (70 kiloms. from the Black Sea), on September 12, 1910, to a depth of 70 metres, gave a jet 64 metres above the surface of the ground and a production of 6000 tons in- twenty-four hours ; on September 18 the fountain caught fiie and five days passed bef6re it could be -extinguished. Fountains as rich as this are exceptional ; usually wells yield much less, and at Baku a well is generally abandoned when it gives less than four tons in twenty-four hours. In Italy, however, wells are used which give only a few hundredweights of petroleum per day ; some of the Italian wells produce only 60 litres a day, others as much as 2500 litres or more 0-7446-0-8588 0-8588-0-9590 DISTILLATION OF PETROLEUM 67 of 600-700 or more barrels capacity are commonly used even in America, where, however, both the more complex and more perfect Lugo apparatus and the Rossmassler apparatus, in which the heating is effected by superheated steam, are used. For the condensation of the vapours that distil over, complicated iron coils are arranged in cisterns through which cold water circulates continuously, the bore of the pipe being 20-25 cm. at first and gradually diminishing to 5-8 cm. The distillate with specific gravity not exceeding 0-750 and b.pt. 150 forms the crude benzine and is collected and worked up separately. The distillate with sp. gr. 0-750-0-860 forms the lighting oil, and the residue is treated separately. FIG. 82. FIG. 83. Continuous distillation is employed more especially at Baku, with large plants con- sisting of boilers arranged in series so that each boiler is maintained at a definite, constant temperature, the vapours passing from one boiler to the other only depositing in a con- densed form those portions corresponding with a given boiling-point and a given specific gravity. By feeding the first boiler which is at the highest temperature continuously, the others are also fed indirectly and kept full, each of them discharging a fraction of a definite, constant specific gravity. Naturally the higher temperature boilers are furnished with dephlegmators (Fig. 84), which cause ready deposition of the heavy oil carried over with the very hot vapours. In these boilers the heating or distillation is effected by FIG. 84. FIG. 85. means of superheated steam, which is usually obtained by passing steam from a boiler (D, Fig. 85) through a series of iron pipes heated in a furnace by direct-fire heat. In addition to other advantages, continuous distillation gives an increase of 30 per cent, in the amount of solar oil. The residue left after distilling the crude petroleum up to 280 bears the Tartar name of masut or the Russian one of astatki (ostatki). The amount of petroleum distilled in twenty-four hours corresponds with four times the capacity of all the boilers in the battery. The Nobel Company at Baku has boilers which distil 1000 tons of petroleum in twenty- four hours. During recent years rectifying columns similar to those used for alcohol have been employed, these admitting of a large production without the use of large boilers. In the " Black Town " near Baku, there are 200 refineries which treat the whole of the 68 petroleum of the district. The odour of petroleum is perceptible at a great distance, and the town is always covered and surrounded with dense, black smoke. The most important refinery is that of Nobel Brothers, which refines half of the annual output of the Caspian, although this firm possesses only one -eighth of the total number of wells. CHEMICAL PURIFICATION OF PETROLEUM. The petroleum distilling between 150 and 300 is not yet suitable for lighting purposes, as it has a marked, rather un- pleasant odour ; it has a faint yellow colour, and contains substances which detract from its value. It was Eichler at Baku who first suggested purification by means of concen- trated sulphuric acid. This is carried out in large iron tanks with conical bases (Fig. 86), the petroleum being treated with several separate quantities (altogether 1-3 per cent.) of concentrated sulphuric acid of 66 Be. (nowadays the mono- hydrate obtained by the catalytic process), the mixture being vigorously agitated by compressed air blown in at the bottom of the tank, and each quantity of the acid separated after half an hour's rest. The sulphuric acid acts especially on the aromatic hydrocarbons (forming sul- phonic acids), the defines and the oxy- genated acid compounds, as well as on the colouring and sulphur substances. A small part (1-3 per cent.) of the petro- leum is resinified and the acid is turned black, but can still be used for the manu- facture of superphosphates. 1 In order to weaken the action of the acid somewhat, it is mixed with sodium sulphate ; further, in order that yellowing of the petroleum may be avoided, sulphuric acid contain- ing less than 0-01 per cent, of nitrous acid should be employed. After the action of the acid, the petroleum is washed thoroughly with water and then with 1-1-5 per cent, of concentrated caustic soda solution (30-33 Be.), air being FIG. 86. passed in from beneath to effect mixing ; in this way the traces of acid remaining and also the phenolic compounds are removed. After the alkali has been separated, the oil is again well washed with water. The remaining petroleum is not clear, as it is emulsified with a little water, but it clarifies on standing and on being filtered rapidly through sawdust and salt, which remove all traces of emulsion. The alkaline petroleum residues are now used in some places to impregnate and preserve railway sleepers ; but sometimes they are subjected to dry distillation, which regenerates the soda and gives coke and unsaturated hydrocarbons and ketones (acetone, &c.). The heating of these alkaline residues also yields naphthenic acids (tridecanaphthenic acid), from which cheap antiseptic soaps are prepared. Some crude petroleums give a rather yellow solar oil, which is decolorised by exposure for some time to the sun in shallow tanks covered with sheets of glass. Sometimes the yellow tint is removed by dissolving in the petroleum traces of complementary blue or violet dyes ; as, however, nearly all commercial dyes are insoluble in petroleum, it is necessary to obtain from the manufacturers the bases of these colouring-matters, these being soluble. In certain cases, decolorisation is attained with infusorial earths, clays, or natural magnesium hydrosilicates. A most important operation for petroleum rick in sulphur (present especially as H 2 S) 1 According to Ger. Pat. 221,615 of 1909 this black acid, containing sometimes as much as 2-5 per cent, of complex organic substances, may be purified by causing it to fall into pure, boiling sulphuric acid through which a current of air is passed ; all the acid distilling over is then pure and colourless. J. Fleischer (1907) obtains colour- less acid (45 to 50 B$.) by causing the black acid to diffuse through porous partitions washed by a little wate r . DESULPHURISING OF PETROLEUM 69 and hence dark and of unpleasant odour (like those from Canada, which can be used only as a combustible and not for lighting purposes) is that of a desulphurising according to the process proposed by Frasch (1888-1893) ; this consists in distilling the petroleum with an excess of a mixture of metallic oxides powdered copper oxide, 75 per cent. ; lead oxide, 10 per cent. ; iron oxide, 15 per cent. This operation reduces the sulphur- content from 0-75 per cent, or more to 0-02 per cent. It is calculated that, by this method, about 50 tons of sulphur are extracted daily from Ohio petroleum, most of it being lost. The operation is carried out by simple mixing or by means of vapour. In the first case 6800 kilos (68 quintals) of the oxide mixture are added to 200 tons of petroleum in a large tank, the mixture being sub- jected to prolonged agitation by mechanical stirrers, which keep the oxidising mass at the bottom of the tank in continual motion. The petroleum is then decanted off into the fractional distilling apparatus, a second quantity of 200 tons of petroleum, together FIG. 87. with 4500 kilos (45 quintals) of oxides being added to the residue in the tank ; the operation is repeated four or five times before renewing the oxides completely. The Frasch process of desulphurising the vapour is far more rational and rapid ; it consists in passing the petroleum vapours from the distillation vessel (from 100 tons of petroleum) (A, Fig. 87) successively into two communicating cylinders, B and C, placed one over the other and enclosed by a metal casing, D, above the boiler. The vapours pass first into the casing, next into the lower cylinder, and then into the upper one, coming into intimate contact with the mixture of metallic oxides, which are kept moving and subdivided in both cylinders by means of rotating reels, h, provided with peripheral brushes, H. The oxidising mixtures in the two cylinders are renewed alternately, while the purified vapours, after traversing a gravel filter, G, which retains particles of the oxides carried over, are condensed in ordinary coils, F. By this process, some refineries are able to purify as much as 11,000 tons of petroleum per day. FIG. 88. FIG. 89. Recently petroleum has been desulphurised by means of metallic sodium, and treatment with aluminium chloride in the hot and under pressure is also recommended. V. Walker (U.S. Pat. 955,372, 1910) passes the vapours into columns fitted with perforated plates and containing anhydrous cupric chloride, the last traces of hydrogen sulphide being removed by passing the vapours into a solution of lead oxide in caustic soda. Robinson (1909) separates the sulphur by treating the petroleum with highly concentrated sulphuric acid. In well-refined petroleums, the proportion of sulphur is always less than 0-06 per cent., usually 0-02 per cent. PETROLEUM TANKS. The refined petroleum is preserved in large cylindrical sheet-metal tanks (Fig. 88), situated near the works ; they are whitened outside to reflect 70 ORGANIC CHEMISTRY the heat of the sun, and are furnished with charging and discharging pipes communicating with the pumping-station by which all the liquids in the works are circulated. For transport by land and sea, wooden casks holding 159 litres (about 145 kilos) were at one time exclusively used, but to-day land transport is effected by tank-cars (Fig. 89), which are now numbered in hundreds of thousands. For sea transport, tank-steamers are used (there are now 360 of these of the total capacity of 630,000 tons) (Fig. 90) ; when they arrive at their destinations in the ports of different countries, they are discharged by means of pumps into storage-tanks or directly into tank-cars. From these stores (there are tanks of 2000 tons capacity at Leghorn, Savona, Genoa, and Venice) it is dispatched inland in wooden or iron casks or in cans holding 14 kilos (17 litres) and packed in pairs in wooden cases. ' FIG. 90. Carbone, coal ; j>etrolio, petroleum ; macchine e caldaie, engines ancfboilers. USES AND STATISTICS. The greater part of refined petroleum is still used for lighting purposes, either in the old lamps with flat wicks or in those with cylindrical wicks and flame -spreaders or in lamps with incandescent Auer mantles ; it can be used advantageously for household illumination in town and country. Part of it is employed for power purposes, as in internal -combustion engines it gives an efficiency of 25-37 per cent., whilst coal yields only 12 per cent. However, while in Russia large quantities of petroleum were used in the past in factories and for locomotives, nowadays it is being replaced by coal ; in America, on the other hand, the opposite is the case, and the Mexican Railway alone consumed more than 4000 barrels of petroleum per day for its locomotives in 1908. Its use on fast ships has the advantage of 28 per cent, saving in space. In America, about 19,000,000 barrels of petroleum were used altogether for railway loco- motives in 1907. Lastly, it is used as a disinfectant and for lubricating engines, &c. The production of petroleum has increased in a surprising manner, in spite of the growing development of the gas and electrical industries. The following figures illustrate this for the two great petroleum -producing regions : In 1874 1884 1894 1903 1905 1908 1910 Caucasus (Russia) Tons 100,000 1,500,000 5,000,000 9,902,000 7,969,239 8,292,000 9,500,000 United States Tons 1,500,000 3,400,000 7,000,000 13,160,006 17,636,000 23,940,000 26,000,000 In America to-day petroleum is monopolised by huge " trusts," especially the Vacuum Oil Company and the Standard Oil Company of New Jersey, to which are affiliated seventy companies with a total capital of 18,000,000 and employing 60,000 workmen and monopolising about 60 per cent, of American petroleum. The Standard Oil Company, founded in 1872, paid in dividends from 1882 to 1892 a total of 94,400,000, and from 1894 to 1903 paid to its shareholders dividends of 33 to 48 per cent. 1 In 1906 President Roosevelt, under pressure of public opinion, waged war against this colossal [trust by rupturing the connection between the steel ring and the interests bound up with it and making them liable to a fine of over 6,000,000. In consequence of this commercial war of 1906 the Standard Oil Company lost 25,000,000, of which 12,900,000 fell on Rockefeller, the well-known millionaire president of the company. The sentence was then annulled on appeal, but the result was that the company fought its competitors by lowering prices (petroleum that previously cost 30 centesimi (2-9d.) per litre has been lowered in price during the last few years to 15 centesimi (l-45rf.) ), and in 1908 made a net profit of 16,000,000, and proposed raising its capital to 100,000,000. This explains how Rockefeller has been able, without any great sacrifice, to make benefactions of so many millions during the past few years, especially for the extension of university study in America. The last sentence of the Supreme Court of Washington (May 15, 1911) gave judgment against the Standard Oil Company, for contravention of the law against trusts, and ordered dissolution of this powerful company within six months. PETROLEUM STATISTICS 71 One-third of the American production has been given by California, more than one- fourth by Texas, and one-sixth by Ohio, and now one-sixth is given by Illinois, one- fourth by California, and one-fourth by Oklahoma. In 1910 the Calif ornian production reached almost 10 million tons. The total production of the. world was about 12,000,000 tons in 1894, 31,000,000 in 1905, and 38,000,000 in 1908. The following Table gives, in the first three columns the production in thousands of tons of each of the petroleum -producing countries of the world, and in the last three columns the percentages of the total amounts yielded by each country : 1903 1905 1908 Per cent, of the total production 1900 1902 1908 United States . Russia Dutch East Indies . 13,600 9,902 870 20,000 7,969-2 1,126-3 23,940 8,292 1,143 51-50 42-41 1-83 52-51 37-12 3-50 63-0 21-8 3-0 Galicia 728 835-8 1,754 1-97 2-90 4-6 Roumania 1 381 641-0 1,148 0-85 1-74 3-0 British India . 329 599-8 672 0-87 1-42 1-7 Japan Canada . 126 194-4 91-9 276 70-5 0-86 0-18 0-50 0-24 0-75 0-20 Germany Italy . . Peru 58-9 2-5 81-3 6-1 5-4 141-9 7-08 135 0-35 0-02 0-35 Various other countries 79-0 In 1890, Germany produced only 15,000 tons of crude petroleum. The country that consumes the most petroleum, after the United States and Russia, is Germany, where, in 1904, 970,600 tons were used for lighting, 143,000 tons for lubri- cating purposes, and 110,000 tons for various other uses ; in 1909, it imported about 950,000 tons of refined petroleum and 31,400 tons of crude petroleum, of a total value of 3,600,000. The importation into Italy has been as follows : 1884 1890 1900 1904 1906 1907 1908 1909 1910 Tons . 73,361 72,000 73,000 69,233 61,588 72,714 82,373 88,930 84,748 and whilst in 1907 two-thirds of this came from the United States, one-fourth from Russia, and little from Roumania, after the new commercial treaty with the last two nations, the proportions changed considerably, Roumania alone sending 29,000 tons in 1909. Almost the whole of the Italian industry is in the hands of one company, and the production is very small and almost stationary. The consumption of petroleum by different countries is quite different proportionately from the production, as is shown in the following Table, which gives the mean consump- tion per inhabitant in 1904 : United States .. Germany England . . France Russia (140,000,000 inhabitants) Japan . . . Total consumption Tons 2,016,700 970,600 520,933 312,210 1,050,787 299,370 Annual consumption per inhabitant Kilos 25-21 16-72 11-84 8-22 7-51 6-65 1 In 1910 the production was 1,352,300 tons, 339,300 tons of distilled petroleum and 125,750 tons of benzine being exported. In 1903 the refineries of Roumania treated altogether 314,748 tons and in 1904 391,387 tons of crude petroleum, which yielded 62,218 tons of benzine, 109,510 tons of lighting oil, 30,214 tons of mineral oil, and 173,661 tons of residues. In 1909 Koumania exported 420,000 tons of petroleum benzine, and mineral oils. ORGANIC CHEMISTRY Roumania . . . Austria -Hungary . . Italy . . . . India (300,000,000 inhabitants) ghina (300,000,000 inhabitants) Annual consumption Total consumption per inhabitant Kilos 4-50 Tons 27,025 215,546 73,000 503,930 254,464 4-31 2-21 1-70 0-85 The units of measure of petroleum in different countries have already been given on p. 58. In view of the enormous and increasing consumption of petroleum, it may be interesting to know how much longer the known stock of petroleum in the earth will last. According to the calculations made in 1909 by the Geological Survey Office, the known deposits of petroleum would last until 1990 if the annual consumption remained at its present amount, but if the consumption increases in the same proportion as it has been doing during the last few years, the deposits will be exhausted in 1935. The price of rectified petroleum at Batoum is about 7s. 2d. per quintal, and the trans- port to Genoa Is. 5d., and, making allowance for all taxes, Russian petroleum costs at Genoa 16s. per quintal, including the cask ; the American costs 16s. I0d., and at the present time Russian petroleum is beginning to oust the American product from the European markets. In the free port of Hamburg, Russian and American petroleums cost 16s. Wd. per quintal in 1879, 13s. Id. in 1890, and 14s. Q%d. in 1904. TESTS FOR LIGHTING PETROLEUM. A good petroleum is limpid and colourless, does not turn brown with sulphuric acid (sp. gr. 1-53), and has a specific gravity of 0-820- 0-825 (Russian) or 0-780-0-805 (American) ; the specific gravity is determined with an aerometer at 15 (corrected by 0-0007 for each degree) and referred to water at 4. It should not have an acid reaction ; when 10 c.c. of the petroleum is dissolved in a mixture of alcohol and ether previously rendered neutral to phenolphthalein, an immediate violet coloration should be produced on addition of a single drop of N/10 alcoholic caustic soda. When subjected to fractional distillation in the Engler flask (p. 66), it should not yield products distilling below 110, only 5 per cent, or at most 10 per cent, up to 150, and less than 10 per cent, or at most 15 per cent, above 300 ; in the distillation products the difference in specific gravity between Russian and American petroleums is increasingly marked. American petroleum is distinguished from the Russian (see p. 62 et seq.) also with the refractometer and by the different solubilities of the fractions of equal specific gravity in a mixture of chloroform and aqueous alcohol (Riche-Halphen test) 1 . The viscosity determined with the Engler viscosimeter (see later, Mineral Oils) should not be greater than 1-15 at 20. The luminosity is determined with the Bunsen photometer (p. 56) and, in general, 3-5-5 grms. are consumed per candle-hour. The determination of the temperature at which a petroleum gives off inflammable vapours is of great importance, and in order to obtain concordant results, the Abel apparatus modified by Penski (Figs. 91 and 92) is employed in all laboratories. The petroleum to be examined is placed in a brass receiver, G, up to the level-index, h ; the cover, D 8, carries a thermometer, t, which dips into the petroleum, and a clockwork mechanism, T b, which, when it is released (by a lever, h), opens automatically a small window in the cover ; at the same instant a small oil-flame passes through the window and is immediately withdrawn, the window then closing. The petroleum receiver is surrounded by an air- chamber, A, which is heated to 55 in the reservoir, W, regulated by the thermometer t 2 . For every 0-5 increase of temperature of the petroleum, the spring is released, this being continued until the flame ignites and explodes the mixed petroleum vapour and air. The slight explosion sometimes extinguishes the flame. The temperature shown at this 1 Of each fraction with specific gravity higher than 0-760, 4 grms. is weighed into a beaker, and from a burette a mixture in equal parts of anhydrous chloroform and 93 per cent, alcohol is run in until the tuibidity fiist foimed suddenly disappears : Density . . . . 0-760 0-770 0-780 0-790 0-800 0-810 0-820 0-830 0-850 0-880 American petroleum (cubic centimetres solvent). . 4-3 4-6 5-2 5-9 6-6 7-7 9-5 11-3 Russian petroleum (cubic centimetres solvent). . 4-0 3-d 4-1 4-2 4-0 4-2 4-5 5-0 6-4 11-9 Italian petroleums behave like the Russian, but this reaction does not serve to distinguish between the other European petroleums (Utz, 1905). FLASH-POINT OF PETROLEUM 73 moment by the thermometer ^ is that of inflammability (flash-point), which is, however, influenced by the atmospheric pressure and should be corrected by + 0-035 for every mm. of pressure above 760 mm. In Italy, Germany, and Austria the sale of petroleum for lighting purposes is prohibited if it shows a flash-point below 21 in the Abel apparatus ; otherwise explosive vapours could be formed in ordinary lamps, even at 30 or 32, which would be dangerous. A petroleum inflammable at above, 60 (Abel) cannot be used for lamps. A rough-and-ready test to detect if a petroleum is dangerous consists in pouring a little into a glass and throwing into it a lighted match ; if the latter is extinguished, the petroleum is safe. FIG. 91. FIG. 92. The illuminating power is determined with the Lummer and Brodhun photometer (see Fig. 77, p. 56). To determine the moisture or water, which does not separate well in the distillation of certain Calif ornian petroleums, Robert and Fraser (1910) proposed adding calcium carbide and measuring the quantity of acetylene formed, this depending on the amount of water present. TREATMENT OF CRUDE BENZINE The portion of crude petroleum distilling below 150 forms crude benzine, which can be separated by fractional distillation into various qualities for different commercial uses. The crude benzine is redistilled in small horizontal or vertical boilers, usually heated by superheated steam either in a jacket or in closed coils inside the boiler, the condensed water being collected outside the boiler. In some cases moderate fire-heat is used in addition. When there are many volatile products, an apparatus similar to that used in the rectification of spirit is employed. Such a system of rectifying columns is to-day in general use, and the condensation of the vapours and the cooling of the condensed benzine are effected by the crude benzine, which is thus fractionated and fuel at the same time economised. A special apparatus for condensation and rectification, devised by Veith, consists of five iron double-walled cylinders (with water-circulation), connected in series and .ter- minating in a sixth cylinder containing a coil with many turns for the condensation of the vapour from the preceding cylinder. The coil is cooled by ice and cold water, which then passes successively into the jackets of the other five cylinders and gradually becomes heated. These five cylinders are full of pure iron turnings free from oil. The vapours from the boiler in which the benzine is distilled pass through cylinders 1-5, in each of which 74 ORGANIC CHEMISTRY that part condenses which is liquefied at the temperature of the water circulating in the jacket. The least volatile products condense in the first cylinder and the most volatile ones in the final coil. At the bottom of each cylinder is a pipe with a tap communicating with a tank. The apparatus for distilling and rectifying benzine are so constructed that the vapour above the boiling liquid which is mixed with air is separated from the liquid, e.g. by metal gauze, so that in case of fire or explosion the liquid does not ignite. Baku petroleums give only 0-2 per cent, of benzine, those of Grosny (Russia) about 4-5 per cent. In 1902, 341,000 tons of naphtha were distilled at Grosny, 14,000 tons of benzine (about 4 per cent.) being obtained. Pennsylvanian petroleums give up to 12 per cent, of benzine, and those from Campina (Roumania) 3-5 per cent. ; a petroleum from Anapa (Caucasus) gave 28 per cent. Italian petroleums from Emilia yield 30-35 per cent, of benzine. After the fractional distillation of the benzine the separate portions are often refined by treating with concentrated sulphuric acid mixed with 0-2 per cent, of potassium dichromate and 0-01 per cent, of lead oxide. Fuming sulphuric acid also gives good results, but animal charcoal and magnesium hydrosilicates are not very satisfactory. The treatment is carried out in closed vessels with mechanical stirrers, the use of com- pressed air being inapplicable here. The majority of the benzine is produced at Baku and in Pennsylvania, but some is refined in Germany and large quantities are sent to Europe from the East Indies from Java, Sumatra, and Borneo ; Galicia and Roumania also yield large quantities. The consumption of benzine is to-day tending to increase, not only as a solvent for fats (benzine boiling between 60 and 80), but also for automobiles, aeroplanes, and dirigible balloons, its calorific value (about 11,000 cals.) being high. That used for cleaning fabrics should boil at a higher temperature, otherwise it evaporates too easily and leaves an annular mark round the spot (other varieties, see p. 66). The consumption of benzine in the various countries of Europe amounted in 1908 to : 115,000 tons in. Germany, 130,000 tons in Trance, 100,000 tons in England, 10,000 tons in the Netherlands, 110,000 tons in Russia, 20,000 tons in Roumania, 10,000 tons in Austria and Galicia, and 25,000 tons in other European countries. The United States produced 800,000 tons of benzine in 1908 and the Dutch Indies 260,000 tons. TREATMENT OF PETROLEUM RESIDUES A. Lubricating Oils. B. Vaseline. C. Paraffin. (A) LUBRICATING OILS. The crude petroleum residue remaining in the boilers at 300 (astatki or masut 1 ) forms a brownish black mass with a greenish reflection, dense and sometimes semi -solid at ordinary temperature, and often with a burnt, faintly creosotic smell ; it has a specific gravity of 0-900-0-950 and a coefficient of expansion of 0'00091, and gives inflammable vapour even at 120-160 ; that of Baku contains no paraffin and hence does not freeze. When these residues are discharged from the boiler, in order to cool them and so prevent them taking fire they are passed through the tubes which serve to heat the crude petroleum before introducing it into the boiler. At Baku the residues, which form almost two-thirds of the crude naphtha, are largely used as a combustible for the distillation vessels and also for locomotives and marine engines, the calorific power being 9700-10,800 cals. and 1 kilo being able to evaporate as much as 14-15 kilos of water. 2 1 Masut contains, on the average, 87-5 per cent. C, 11 per cent. H, and 1-5 per cent. O ; it has a mean specific gravity of 0-91, an ignition temperature of 110 and a calorific value of 10,700 cals. When used as a combustible it is gasified, the vapours, mixed with compressed air, burning completely ; it is often burnt directly after pulverisation with compressed air or steam. In view of the great calorific value of petroleum residues and their increasing production, new outlets have been sought for them ; they should have a great future as a substitute for coal in the heating of boilers, steam-engines, ships, &c. But, as has been already stated, this use of it is diminishing in Russia, although continually extending in the United States. In Italy attempts have recently (1911) been made to burn it, after pulverisation, directly under boilers, and it can be used advantageously if it does not cost "at the factory more than about 5s. per quintal, coal giving 8000 cals. costing 2s. lOd. ; the cost of transport is hence excessive, increasing the price from lOd. or 15rf. at the refinery to 5s. in Italy. The Customs duty (Italy) is 20 centesimi (just under 2CH 2 !_CH, \c-cy II., II., CH, Trimethylene Hexamethylene The carbon atoms in these last compounds are all in the same conditions and cannot be differentiated. The cyclic compounds will be studied in a separate section of the aromatic series (Part III). The following Table gives the more important members of the olefine series (the number in parentheses representing boiling-points under reduced pressure) : Melting- point Boiling- point Melting- point Boiling- point Ethylene, C 2 H 4 . . -169 -103 Decylene, C 10 H 20 . 172 Propylene, C 3 H 6 . -48 Endecylene, C 11 H 22 195 Butylene (3 isoms.),! , OH V 5 + 1 Dodecylene, C^H^ Tridecylene, C^H^ -31 (96) 233 41 8 [y ' -6 Tetradecylene, C^Hog -12 (127) Amylene (5 isoms.), Pentadecylene, Ci 5 Tl 30 '247 C 5 H 10 ; normal - Hexadecylene C 16 H 32 1 4 ( 274 amylene + 35 (Cetene) } \ (165) Hexylene, C 8 H 12 . 68 Octadecylene, C^gHgg . + 18 (179) Heptylene, C 7 H 14 98 Eicosylene, C 20 H 40 . Octylene, C 8 H 16 124 Cerolene, C 27 H 54 . +58 Nonylene, C 9 H 18 153 Melene, C 30 H 60 . . + 62 The official nomenclature of the defines is the same as that of the paraffins, excepting that the final ane is changed into ene (thus ethylene, which is isologous with ethane, is called ethene, and so on ; see also p. 28). These unsaturated hydrocarbons differ little in their physical properties from the corresponding saturated homologues. The first terms up to C 4 H 8 are gases, and after C 5 H 10 come liquids with increasing boiling-points, these gradually approaching one another as in the paraffins ; the higher members are solid and, like the paraffins, have a sp. gr. 0-63-0-79, are insoluble in water, but soluble in alcohol or ether. The chemical properties differ somewhat from those of the saturated compounds. Thus, they readily take up HC1, HBr, HI, Cl, Br, I, fuming H 2 S0 4 , hypochlorous acid (giving chloro-alcohols or chlorhydrins, e.g. 88 ORGANIC CHEMISTRY CH 2 : CH 2 + HC10 = CH 2 C1 CH 2 OH), hyponitrous acid, ozone, &c., forming compounds of the saturated series. Cl is added more easily than I, Br occupying an intermediate position, whilst HI is added more easily than HBr, and this more easily than HC1. With these acids, the halogen is added to the carbon atom with which the least hydrogen is combined. Ethylene unites with fuming sulphuric acid at the ordinary temperature and with the ordinary acid at 165, forming ethylsulphuric acid, C 2 H 5 S0 3 H ; with higher compounds, the acid radicle passes to the less hydrogenated carbon atom. They often polymerise under the action of sulphuric acid or zinc chloride ; for example, amylene, C 5 H 10 forms C 10 H 20 , and C 15 H 30 gives C 20 H 40 . They are readily oxidisable, for example, with potassium permanganate or chromic acid (not with nitric acid in the cold), the chain being then broken at the double linking, with formation of oxygenated compounds (acids) con- taining less numbers of carbon atoms in the molecule. Careful use of per- manganate results initially in the addition of two hydroxyl groups without breaking the chain and forming dihydric alcohols (glycols), for example, OH-CH CH-OH. 1 Almost all compounds with a double linking between atoms of carbon give Baeyer's reaction, that is they rapidly discharge the violet colour of a dilute solution of potassium permanganate and sodium carbonate, with forma- tion of a reddish brown flocculent precipitate of hydrated manganese peroxide. This reaction is not given by reducing substances like aldehydes or by certain aromatic compounds (phenanthrene, &c.). All compounds with doubly linked carbon atoms give the ozone reaction (Harries, 1905, and Molinari, 1907), that is, when dissolved in a suitable solvent they fix, quantitatively and in the cold, the ozone contained in a current of ozonised air passed through the solution ; in this property they differ from compounds with either a triple linking or a benzene double linking (E. Molinari, Ann. Soc. Chim., Milan, 1907, 116). METHODS OF PREPARATION. (1) They are formed, together with petroleum, in the dry distillation of wood, lignite, coal, paraffin (" cracking," &c.). (2) By eliminating water from the alcohols, C n H 2w +iOH, by heating them with dehydrating agents (H 2 S0 4 , P 2 5 , ZnCl 2 , &c.) ; a stable intermediate product is sometimes formed, e.g. ethyl-sulphuric acid, C 2 H 5 -HS0 4 , which at a higher temperature gives ethylene and sulphuric acid. Higher alcohols and ethers are resolved, merely on heating, into defines and water. (3) From saturated halogen derivatives C w H 2n+1 X (X = halogen), especially from secondary and tertiary bromo- and iodo-derivatives, by heating them with alcoholic potash, or by passing their vapours over heated lime or lead oxide, &c. C 5 H U I + C 2 H 5 OK = KI + C 2 H 5 -OH + C 5 H 10 . The mixed ether, C 5 H U C 2 H 5 , may also be formed to some extent. 1 From what has been said up to the present, it is obvious that a double Unking does not signify a firmer union between carbon atoms ; it is simply a convention. And the breaking of the chain, by oxidising agents, at the double linking is to be attributed to the ease of formation of intermediate products (e.g. dihydric alcohols) rather than to a less attraction existing between carbon and carbon at that point. Such readiness to react may, according to Baeyer, be explained by regarding the affinities of the carbon atom as orientated or grouped at four poles arranged like the vertices of a regular tetrahedron (see p. 18 et seq.). If two carbon atoms unite by a double linking the poles at the surface of the carbon atoms become displaced and approach one another, so that there results a certain tension which tends to restore the poles to their original positions (Baeyer's tension hypothesis of valency), and which explains the readiness with which the double linking reacts or opens. After the initial oxidation leading to these intermediate products, further action of the oxidising agent, as a general rule, oxidises or breaks the chain at a point where oxygen already exists, that is where the oxidation is already begun (see Part III, The Hypothesis of the Partial Valencies of the Beniene Nucleus). ETHYLENE 89 (4) From dihalogenated compounds by heating with zinc : C 2 H 4 Br a + Zn = ZnBr 2 + C 2 H 4 . (5) By electrolysis of dibasic acids of the succinic acid series : C 2 H 4 (COOH) 2 = C 2 H 4 + 2C0 2 + H 2 . (6) Unsaturated compounds are obtained by heating the condensation products of the ketenes (q.v.). CONSTITUTION OF THE OLEFINES. In this group it is assumed that between two carbon atoms there exists a double linking: H 2 C=CH 2 , H 2 C=CH CH 3 , &c., the presence of two free valencies, thus, H 2 C CH 2 or HC CH 2 , being excluded for the following reasons : f\ In unsaturated compounds the addition of halogen does not take place at a single carbon atom, so that ethylene chloride, C 2 H 4 C1 2 , has not the formula CH 3 CHC1 2 , which is that of ethylidene chloride obtained from acetaldehyde, CH 3 CHO, by replacement of the O by C1 2 (by the 'action of PC1 6 ). Since ethylene chloride is chemically and physically different from ethylidene chloride, the former must have the constitutional formula, CH 2 C1 CH 2 C1, and the third formula for ethylene, CH 3 CH< is thus excluded. The second formula is not probable because, if the existence of free valencies is assumed, they could also occur in non-adjacent carbon atoms, and thus give rise, in the higher hydro- carbons, to numerous isomerides which have, however, never been prepared (if propylene had two free valencies, four isomerides should exist, instead of only one) ; further, the addition of halogen always takes place at two contiguous carbon atoms (see Note on preceding page). Finally, the admission of free valencies in organic compounds is inadmissible in view of the unsuccessful attempts to prepare methylene (or methene), CH 2 , for instance, by eliminating HC1 from methyl chloride, 2CH 3 C1 = 2HC1 + 2CH 2 <^ ; the two methylene residues always condense, forming ethylene, as the two valencies cannot remain free. ETHYLENE, C 2 H 4 (Ethene), H 2 C = CH 2 . This is a gas, becoming liquid at 103 and solid at 169, or liquid at under 44 atmos. pressure. It is very slightly soluble in water or alcohol. It has a somewhat sweet smell and burns with a luminous flame ; indeed, illuminating gas, which contains 2 to 3 per cent, of ethylene, owes part of its luminosity to this gas. When mixed with 2 vols. of chlorine it burns with a dark-red flame, carbon being deposited and HC1 formed. At a red heat it yields C, CH 4 , C 2 H 6 , C 2 H 2 , &c. ; with hydrogen in presence of spongy platinum or, better, powdered nickel at 300, it is converted into ethane. It is prepared in the laboratory by heating alcohol with excess of sulphuric acid ; as an intermediate product, ethyl-sulphuric acid is formed, this giving ethylene when heated : C 2 H 5 -OH + H 2 S0 4 = H 2 + C 2 H 5 HS0 4 ; C 2 H 5 HSO 4 = H 2 S0 4 + C 2 H 4 . Pure ethylene is obtained (1) by passing a mixture of carbon monoxide and hydrogen over finely divided nickel or platinum at 100 : 2CO + 2H 2 = C 2 H 4 + 2H 2 ; (2) by dropping alcohol on to phosphoric acid at 200-220 ; .or (3) from ethylene bromide and a copper zinc couple. PROPYLENE, C 3 H 6 (Propene), CH 2 =CH CH 3 . This can be prepared by heating glycerol with zinc dust or from isopropyl iodide and potassium hydroxide. It is a gas which liquefies at 48 and is isomeric with trimethylene. BUTYLENES, C 4 H 8 (Butenes). Three isomerides, the a, ft, and y, are known, and are obtained by treating normal, secondary, and tertiary butylene iodides respectively with potassium hydroxide : CH 2 =CH CH 2 CH 3 CH 3 CH =CH CH 3 3 >C =CH 2 Butene-l (-butylene) Butene-2 (0-butylene) Methylpropene (isobutylene) Tetramethylene or cyclobutane is isomeric with the butylenes. 90 ORGANIC CHEMISTRY AMYLENES, C 5 H 10 (Pentenes). Of the various isomerides theoretically possible several Have been prepared. By heating fusel oil (of distilleries) with zinc chloride, pentanes and various isomeric amylenes are formed which can be separated by means of the different velocities with which HI is added to them, or by the property possessed by some of them of dissolving in the cold in a mixture in equal parts of concentrated sulphuric acid and water, forming amylsulphuric acid, whilst the others either do not react or give condensation products (di- and triamylenes). CEROTENE, C 27 H 54 , and MELENE, C 30 H 60 , are similar to paraffin, and are obtained by distilling Chinese wax or beeswax. II. HYDROCARBONS OF THE SERIES, C,,H 2 ,, _ 2 A. With Two Double Linkings (Diolefines or Allenes) Of the few known terms of this series, the first and best investigated is ALLENE, H 2 C : C : CH 2 (propandiene) : this is a colourless gas which differs from its isomeride allylene in not forming metallic derivatives ; it is obtained by eliminating one atom of bromine from tribromopropane by means of potassium hydroxide and the remaining two by zinc dust, its constitution being thus rendered evident : CH 2 Br CHBr . CH 2 Br-+ CH 2 : CBr . CH 2 Br-+ CH 2 C : CH 2 . ERYTHRENE, C 4 H 6 (Pyrrolilene or Butane-1 : 3-diene), CH 2 : CH-CH : CH 2 , is a gas found in illuminating gas, and when heated with formic acid gives erythritol. ISOPRENE, C 5 H 8 , boils at 37 and is obtained by distilling rubber. On the other hand, with concentrated HC1, it condenses, regenerating rubber or forming terpenes, riTj CioH^g, CifrH-zi, &c. Since dimethylallene, r ,TT 3 ">C : C : CH 2 , gives, with 2HBr, a di- bromide, nT , 3 ^>CBr CH 2 CH 2 Br, which is identical with that obtained from isoprene U1 3 CH 2 . + 2HBr, the constitution of isoprene must be : ^.C CH : CH 2 , CH 3 / The normal isomeride PIPERYLENE, CH 2 : CH-CH 2 .CH : CH 2 (Pentane-1 : 4-diene) boils at 42 and is obtained from piperidine. DIALLYL, C 6 Hio (Hexine), is prepared by the general reaction the action of sodium on allyl iodide which indicates its constitution : 2CH 2 : CH-CH 2 I + 2Na = CH 2 : CH - CH 2 CH 2 CH : CH 2 + 2NaI. CONYLENE, C 8 H 14 (1 : 4-octadiene), CH 2 : CH-CH 2 -CH : CH 2 -CH 2 .CH 3 , boils at 126 and is obtained from coniine. B. Hydrocarbons with Triple Linkings (Acetylene Series) The most important members of this series are : Acetylene, C 2 H 2 (ethine), HC=sCH, gas. Allylene, C 3 H 4 (propine), CH 3 -C=-CH, gas. Crotonylene, C 4 H 6 (2-butine or dimethylacetylene), CH 3 -C I C-CH 3 , boils at 27. Ethylacetylene, C 4 H 6 (3-butine), CH 3 -CH 2 -C : CH, boils at 18. Methyleihylacetylene, C 5 H 8 (3-pentine), CH 3 -CH 2 -C I C-CH 3 , boils at 55. n-Propylacetylene, C 5 H 8 (4-pentine), CH 3 -CH 2 -CH 2 -C CH, boils at 48. Isopropylacetylene,C 6 H. s (3-methyl-l-butine), *:5 3 >CH-C I CH, boils at 28. Ui 3 Several of these compounds (the first three) are formed during the dry distillation of coal and other complex substances, and are hence found in lighting gas. In the laboratory they are obtained by the following methods : {a) By electrolysing acids of the fumaric acid series (see later) : COOH-CH CH-COOH = H 2 + 2C0 2 + HC ' CH. ACETLYENES 91 (6) By heating with alcoholic potash the halogenated compounds (best the bromo-derivatives) C M H 2w X 2 and C n H., w _2X 2 , gradual elimination of halogen hydracid (of HBr or, in presence of KOH, of KBr and H 2 O) occurs : C 2 H 4 Br 2 = HBr + C 2 H 3 Br HBr + C 2 H 2 . In general, starting from the saturated hydrocarbons, C M H 2n + 2 , the action of halogen and elimination of halogen hydracid gives an unsaturated hydrocarbon, C H H 2n ; addition of halogen to this and subsequent removal of halogen hydracid gives a still less saturated hydrocarbon, C M H 2M _ 2 , and so on. Elimination of 2HC1 from the compounds C n H 2n Cl 2 , obtained from alde- hydes or from certain ketones (methylketones, C M H 2M+1 -CO-CH 3 ) by the action of PC1 5 , yields always a trebly linked compound, in which, however, one of the carbon atoms is always united to a single, characteristic hydrogen atom : C = CH ; for example, acetaldehyde gives ethylidene chloride, CH 3 CHC1 2 , which then yields 2HC1 + CH i CH ; while acetone, CH 3 CO CH 3 , gives chloroacetone CH 3 -CC1 2 -CH 3 , and this 2HC1 + CH 3 -C = CH, the elimination of halogen hydracid never occurring in such a way as to give compounds with two double linkings, such as CH 2 : C : CH 2 . Acetylene derivatives are also obtained by heating the acids of the pro- piolic series (see later), Compounds with this characteristic hydrogen atom C^CH have a feebly acid character and form solid metallic derivatives (acetylides) when treated with an ammoniacal solution of copper chloride or silver nitrate : copper acetylide, Cu-C : C-Cu, H 2 0, having a reddish brown colour and apparently the constitution, Cu 2 CH-CHO, since with hydrogen peroxide it gives acetaldehyde, CH 3 CHO (Makowka, 1908) ; and silver acetylide, AgC : CAg, which is white and insoluble in water or ammonia and, in the dry state, is extremely explosive, simple rubbing being sufficient to explode it. With hydrochloric acid it regenerates acetylene in a pure state. The proof that it is the characteristic hydrogen atom which is replaced by metals lies in the fact that acetylene derivatives from other ketones (not from methylketones) do not give metallic acetylides : CH 3 CH 2 CO CH 2 CH 3 >CH 3 CH 2 CC1 2 CH 2 CH 3 > 2HC1 + CH 3 -C i C-CH 2 -CH 3 . Four atoms of a halogen or of hydrogen can be added to the hydrocarbons of the acetylene series, saturated compounds being formed ; but as a rule only two atoms are readily added, although under the action of light four halogen atoms can be added almost always. The compounds of the olefine series can, however, be distinguished from those of the acetylene series by means of the ozone reaction, since compounds with a triple linking do not fix ozone at all (Molinari). The hydrocarbons of the acetylene series take up a molecule of water in presence of mercury salts, giving rise to complex mercuric compounds, which, with HC1, give as final product an aldehyde or ketone ot the satu- rated series CH.J-C i CH (aUylene) + H 2 O = CH 3 -CO-CH 3 (acetone) or CH i CH + H 2 = CH 3 -CHO (acetaldehyde). This last reaction serves to illustrate the transformation of inorganic into organic substances (see later, p. 108). In the acetylene series, also, condensation or polymerisation is possible, three molecules of acetylene, on heating, yielding benzene C 6 H 6 ; three mols. of dimethylacetylene, C 4 H 6 , giving, with concentrated sulphuric acid, hexa- methylbenzene, C 6 (CH 3 ) 6 , and allylene C 3 H 4 similarly,, yielding trimethylbenzene (mesitylene) C 6 H 3 (CH 3 ) 3 . 92 ORGANIC CHEMISTRY In the higher compounds, the position of the triple bond is deduced from the oxidation products, since, as with substances with a double linking, the breaking of the chain occurs at the multiple linking. When certain acetylene derivatives, e.g. XC=ssC-CH 3 , are heated with sodium, the triple bond changes its position, the products being sodium derivatives of isomeric hydrocarbons, X-CH 2 -C : CH (these give metallic acetylides, but the original compounds do not) ; when these are heated with alcoholic potash, the reverse change occurs. ACETYLENE, C 2 H 2 (Ethine), HC CH. Without having isolated or characterised this compound, Davy obtained it in 1839 in a very impure condition, by treating with water the product obtained by heating together potassium carbonate and carbon, which should yield potassium. Bert helot first obtained it pure (and named it) in 1859, by passing ethylene or alcohol or ether vapour through a red-hot tube ; he prepared it also by means of a voltaic arc passing between two carbons in an atmosphere of hydrogen. In 1862, Wohler prepared it by treating calcium carbide (obtained by heating carbon with an alloy of zinc and calcium) with water. It is formed in the incomplete combustion of various hydrocarbons and of illuminating gas (e.g. in the flame of a bunsen burner alight at the bottom). But the industrial preparation of acetylene has assumed great and unfore- seen practical importance since 1870, when it became possible to prepare calcium carbide on an enormous industrial scale by means of the electric furnace (see Calcium Carbide Industry, vol. i, p. 504) : c x \] >Ca + 2EUO = Ca(OH) 2 + HC \ CH. C/ Acetylene is a colourless gas, sp. gr. 0-92 (1 litre weighs 1-165 grm.) with a pleasant odour when pure and a disagreeable one when impure (as usually obtained). At + 1 under a pressure of 48 atmos. it forms a highly refractive, mobile, colourless liquid, sp. gr. 0-451, and, on evaporating rapidly, partially solidifies in the form of snow, m.pt. 81. One volume of acetylene gas dissolves in 1-1 vol. of water, or in |- vol. of alcohol or in 20 vols. of saturated salt solution ; 1 litre of acetone dissolves 24 litres of acetylene, or 300 litres at 12 atmos., or about 2000 litres at 80, its volume being then increased fourfold. Permanganate oxidises it giving oxalic acid, and chromic acid acetic acid. It is an endothermic compound, requiring for its formation from its elements, 61,000 cals. ; it is hence very unstable and is readily decomposed by the detonation of a mercury fulminate cap or by an electric discharge, developing as much heat as an equal volume of hydrogen on conversion into water. The explosion takes place much more readily and is much more dangerous with the compressed gas and still more so with the liquid. Acetylene decomposes at 780 and, when mixed with air, ignites at 480. One cubic metre (1-165 kilo) of acetylene, in burning, develops 14,350 Cals. (12,300 Cals. per kilo), whilst ordinary coal-gas gives about 5000 Cals. When mixed with air or, better, with oxygen it forms a detonating mixture which explodes with great energy in contact with an ignited body. The explosion is violent even with 1 vol. of acetylene and 40 vols. of air ; it reaches its maximum violence with 1 vol. of the gas and 12 vols. of air (2-5 vols. of oxygen), whilst scarcely any explosion but mere burning takes place with 1 vol. of acetylene and 1-3 vol. of air (as has been already stated on p. 33, ordinary illuminating gas only explodes when at least 1 vol. is present to about 20 vols. of air). Explosive mixtures of acetylene are more dangerous than those of coal-gas ACETYLENE 93 owing to the greater speed of propagation of the explosion (e.g. with 1 vol. of acetylene and 40 of air), the explosive force being thus increased (see section on Explosives) ; further, acetylene contains less hydrogen and hence forms less water, the condensation of the gases resulting from the explosion being consequently smaller. The wide limits of the explosive mixtures (from 2-4 to 130 vols. of acetylene per 100 vols. of air) are explained by the fact that this gas, being an endothermic compound, reacts or decomposes with great facility. In contact with copper, bronze, silver, &c., acetylene readily forms explosive acetylides (see p. 91). 1 It was at first thought that acetylene, like carbon monoxide, was poisonous, but experiments made during the last few years have shown that animals do not die in an atmosphere containing 9 per cent, or, in some cases, even 20 per cent, of the gas. When, however, the acetylene is highly contaminated with sulphides and phosphides, it may be poisonous. With an ordinary gas-jet, acetylene burns with a reddish, smoky flame ; but by passing the gas at a pressure of 60 mm. through two jets nearly meeting at an angle, a white, highly luminous, fan-shaped flame is obtained without the dark middle portion of the ordinary bat's-wing coal-gas flame. One kilo of chemically pure calcium carbide should yield theoretically 349 litres of acetylene, and good commercial carbide yields practically 300 litres. The luminosity of acetylene in comparison with that of other substances has already been referred to on p. 57. A proportion of 2 vols. of air to 3 of acetylene gives the maximum luminosity, and at the present time special incandescent mantles are made for use with acetylene. The impurities present in ordinary acetylene (98-99 per cent, purity) are : N, NH 3 , CO, H 2 S and PH 3 , the last three of which are poisonous. The gas is purified by passing it through an acid solution of a metallic salt. Lunge and Cederkreutz recommend chloride of lime (hypochlorite) for purifying acetylene, care being taken that the mass does not heat, as this would be dangerous. Latterly it has been suggested to fix the PH 3 by passing the gas through concentrated sulphuric acid (64 Be.) saturated with As 2 3 . A good purifying material is made by preparing a paste of calcium hypochlorite, quicklime, sodium silicate and powdered calcium carbide, this remaining porous when allowed to dry in the air. The use of liquid acetylene would be very convenient, but is highly dangerous, since a sharp blow or other accident might easily produce a terrible explosion. It is still too expensive to employ in place of benzene for carburetting coal-gas. Dissolved in acetone, which dissolves a large quantity of it (vide supra), it is used to great advantage for the oxy-acetylene blowpipe in place of oxy-hydrogen. With the latter, for every cubic metre of oxygen 4 cu. metres of hydrogen are used practically (theoretically 2 cu. metres), whilst the same amount of oxygen burns with 600 litres of acetylene (theoretically 400 litres), which costs much less than 4 cu. metres of hydrogen. The oxy- acetylene flame exhibits at the centre a shining point, which has a temperature of 2800-3000, and to fuse iron sheets 1 mm. thick requires 50-75 litres of acetylene, while in an hour sheets 5 mm. in thickness can be melted. With a slight excess of oxygen large tubes are easily cut and steel blocks perforated. Acetylene dissolved in acetone, especially if the solution is absorbed by porous material, is not at all dangerous and can be transported in iron cylinders. The hope of manufacturing synthetic alcohol economically from acetylene has died out. Even for motors it is still too dear to use. Acetylene can, however, be used conveniently with a rational plant and relatively small gasometers connected with iron tubes which carry the gas direct to the burners (when prepared from pure carbide) ; but it is necessary to avoid the use of copper or bronze in any part of the gasometers, pipes 1 The ready formation of metallic acetylides, especially that of copper, led Brdmann (1907) to devise a rapid and exact analytical method for the direct quantitative precipitation of copper from any solution and in presence of any metals (except Ag, Hg, Au, Pd, and Os, which must be previously eliminated) ; the feebly alkaline solution of the copper salt is reduced until decolorised with hydroxylamine hydrochloride, C 2 H 2 being then passed through and the precipitated copper acetylide collected on a filter, washed with water and pumped off ; together with the filter-paper it is introduced into a porcelain crucible, treated with 10 to 15 c.c. of dilute nitric acid CH, Jlethylene, CH,X, 42 97 180 -CH 2 -CH 2 Ethylene 84 131 CH, CH 2 < Ethylidene (or ethydene) 57 108 (3) Trisubstituted CHX, (chloroform, 61" 161" solid bromoform, iodoform) mpt.H9 CH.-CC1, methyl chloro- 74 188 form (a-trichloroethane CH.C1-CHC1, (/3-tri- 114 220 chloroethane) CH.X-CHX-CH,X (tri- 168 chlorohydrin, tri- bromohydrin) (4) Polysvbftituted CX 4 (carbon tetra- 77 solid chloride, iodide) C a Cl, perchloroethane solid m.pt. 187 (&) UNSATURATED DERIVATIVES (1) Ethylenic series CH, :CH-X Vinyl chloride, &c. - 18 23 56 C,H,-X AUyl 46 70 101 CVH,:X S Dichloroethylene 55 G,H : X, Trichloroethylene 88 C, : X 4 Tetrachloroethylene 121 (2) Acetylene tenet HC: CX Monochloro- and mono- gas gas _ bromo-acetylene METHODS OF PREPARATION, (a) By the action of halogens on saturated h3 r drocarbons : chlorine and bromine react directly at the ordinary temperature on the gaseous hydrocarbons, and on heating with the liquid ones. The first halogen atom is fixed more readily than the succeeding ones, and the addition of iodine facilitates the reaction with bromine and chlorine, since the iodine forms, for example, IC1 3 , which readily gives nascent chlorine, IC1 3 = IC1 + C1 2 (i.e. it acts like SbCl 5 , which yields SbCl 3 + Cl ? ). By saturating with chlorine and heating under pressure energetic chlorinations may be effected. Methane, ethane, propane, &c., exchange their hydrogen atoms one by 96 ORGANIC CHEMISTRY one for chlorine atoms, the completely substituted compounds (C 2 C1 6 , C 3 C1 8 , &c., and especially the higher ones), on further energetic chlorination, being resolved into other completely chlorinated compounds containing less numbers of carbon atoms : C 2 C1 6 + C1 2 = 2CC1 4 ; C 3 C1 8 + C1 2 = C 2 C1 6 + CC1 4 , a little hexachlorobenzene, &c., being always formed as well. Iodine scarcely ever acts directly on the hydrocarbons, since the HI, formed acts in the opposite sense on the iodo-products. The reaction proceeds only in presence of iodic acid or mercuric oxide, which fixes the hydrioclic acid as it is formed. The iodo -compounds are easily obtained from zinc-alkyls and iodine. When the halogens act directly, the more energetic (F or Cl) replaces the weaker (Br or I). The iodo-compounds may, however, be easily obtained by preparing first the magnesium compounds of the alkyl chlorides or bromides and treating these with iodine : Alkyl-Mg-Cl + I 2 = Alkyl-I + MglCl. (b) Unsaturated hydrocarbons, with the halogen hydracids, give saturated monosubstituted derivatives : C 2 H 4 + HBr = C 2 H 5 Br, ethyl bromide, &c. ; if the halogens act directly, disubstituted saturated products are obtained : C 2 H 4 + C1 2 = C 2 H 4 C1 2 , ethylene dichloride. Propylene, CH 3 CH : CH 2 , reacts with HI giving isopropyl iodide, CH 3 -CHI-CH 3 , which is decomposed by alcoholic potash, yielding propylene ; but normal propyl iodide, CH 3 -CH 2 -CH 2 I, which also yields propylene when HI is removed from it, can thus be converted into isopropyl iodide. Similar behaviour is exhibited by the butyl iodides. The halogen always goes to the carbon atom united with the lesser number of hydrogen atoms : CH 3 -CH : CH 2 + HI = CH 3 -CHI-CH 3 . (c) The alcohols C w H 2w+1 OH with the halogen hydracids give : C w H 2n+1 OH + HBr = H 2 + CjjHg^+jBr, but the reverse action also pro- ceeds and to limit this, excess of the halogen hydracid is used and the water formed is fixed, e.g. by addition of zinc chloride. Further, the chlorine of the phosphorus chlorides also replaces hydroxyl : PC1 3 + 3C 2 H 5 OH = P(OH) 3 + 3C 2 HsCl, or, better, PC1 5 + C 2 H 5 OH = POC1 3 + HC1 + C 2 H 5 C1. This reaction is of importance for the preparation of the bromo- and iodo-compounds : 3CH 3 -OH + P + 31 = 3CH 3 I + H 3 P0 3 ; the bromine or iodine first acts on the phosphorus to form PBr 3 or PI 3 , this then reacting with the alcohol. The polyhydric alcohols act in the same way ; for example, glycerol, C 3 H 5 (OH) 3 reacts with PC1 5 giving trichlorohydrin, CH 2 C1 CHC1 CH 2 C1. The resulting halogenated products are easily separated by distillation, as the phosphoric acid does not distil. In these, as in most other chemical reactions, secondary products are always formed ; these are often very com- plex and form viscous resins of unknown composition. (d) The aldehydes and Tcetones yield disubstituted products : for example, ethylidene chloride, CH 3 -CHC1 2 , is obtained from acetaldehyde, CH 3 -CHO, and dichloropropane, CH 3 CC1 2 CH 3 , from acetone, CH 3 -CO-CH 3 , by the action of PC1 5 . METHYL CHLORIDE (Chloromethane), CH 3 C1. This is prepared by passing hydrogen chloride into boiling methyl alcohol containing half its weight of zinc chloride in solution, or by heating 1 part of methyl alcohol with 3 parts of concentrated sulphuric acid and 2 parts of concentrated hydro- chloric acid. Industrially it can be obtained by heating methyl alcohol and crude, concentrated hydrochloric acid together in an autoclave. It is also obtained to-day in appreciable quantity, by the old Vincent process, from the final residues of the beet-sugar industry, which are evaporated ALKYL HALOGEN COMPOUNDS 97 and then dry-distilled. In this way an abundant quantity of trimethylamine is formed ; this is neutralised with HC1, and the hydrochloride distilled at 300. A regular evolution of methyl chloride and trimethylamine is thus obtained : 3N(CH 3 ) 3 HC1 = 2CH 3 C1 + 2N(CH 3 ) 3 + CH 3 -NH 2 -HC1. Triinethylamine hydro- Trimethyl- Methylamine hydro- chloride amine chloride (residue) The chloromethane, distilled as a gas, is purified with HC1, dried with CaCl 2 , and liquefied in steel cylinders under pressure, just as is done with carbon dioxide (vol. i, p. 382). It is a colourless gas of ethereal odour, and at 23-7 becomes liquid, then having a sp. gr. 0-952 (at 0). Water dissolves one-fourth of its volume, and alcohol rather more. It burns with a green-edged flame. In the liquefied condition it is used as a local anaesthetic ; it is used also to extract perfumes from flowers, and in considerable quantities for the manufacture of dyestuffs (methyl green), especially for methylation ; but the greatest amount is employed in cooling machines. In France there are about 100 ice-machines which use methyl chloride instead of liquefied NH 3 , C0 2 , or S0 2 . In brass cylinders containing from 1 to 30 kilos it is sold at 11s. to 14s. Qd. per kilo, in addition to the cost of the cylinder, which is 20s. for the 1-kilo, 25s. Qd. for the 3-kilo, and 3 16s. for the 30-kilo size. METHYL IODIDE, CH 3 I, is prepared from methyl alcohol, phosphorus, and iodine as described later for ethyl iodide. It is a liquid of sp. gr. 2-293, boiling at 45 ; with excess of water at 100 it is decomposed into hydrogen iodide and methyl alcohol. ETHYL CHLORIDE (Chloroethane), C 2 H 5 C1, was termed by Basil Valentine " Spiritus salis et vini," or spirit of sweet wine. It is obtained from ethane and chlorine, or by passing hydrogen chloride into a solution of zinc chloride and ethyl alcohol. It is also formed as a secondary product in the manufacture of chloral. It boils at + 12-2 and burns with a flame having green edges. It is a local anaesthetic and is soluble in alcohol, but only slightly so in water. It costs from Is. Id. to 4s. per kilo in metal cylinders containing 1 to 30 kilos. ETHYL IODIDE, C 2 H 5 I, is prepared by digesting 10 grms. of red phosphorus with 80 grms. of absolute alcohol for 12 hour's and gradually adding 100 grms. of iodine ; the mix- ture is then heated for 2 hours under a reflux condenser and the ethyl iodide distilled on the water-bath, washed with dilute alkali and with water, and dried by means of calcium chloride. According to Ger. Pat. 175,209, ethyl iodide is obtained quantitatively if diethyl sulphate is slowly added to the calculated amount of hot potassium iodide solution. It boils at 72-3 and has the sp. gr. 1-944 (at 14) ; it is highly refractive and dissolves in alcohol or ether. It decomposes when heated with water at 100. Chlorine converts it into ethyl chloride and bromine into ethyl bromide. In the light it slowly decomposes with separation of iodine, which colours the liquid brown, but it remains colourless in presence of a drop of mercury. It is used as an inhalation for the treatment of asthma. It costs about 28s. to 32s. per kilo. ETHYL FLUORIDE, C 2 H 5 F, is liquid at -48, burns with a blue flame, and does not attack glass. From PROPANE two series of isomeric compounds are derived: CH 3 CH 2 CH 2 X, prepared from normal propyl alcohol, and CH 2 -CHX-CH 3 , derived from isopropyl alcohol, and hence from acetone. ISOPROPYL IODIDE (Iodo-2-propane), CH 3 -CHI.CH 3 , is obtained from glycerol, phosphorus and iodine, small amounts of allyl iodide and propylene being also formed. The butyl compounds occur in four isomeric modifications : NORMAL BUTYL IODIDE (Iodo-i-butane), CH 3 - CH 2 - CH 2 - CH 2 I. SECONDARY BUTYL IODIDE (Iodo-2-butane), CH 3 . CH 2 - CHI - CH 3 . ISOBUTYL IODIDE (Methyl-2-iodo-3-propane), ^ 3 >CH-CHI. /-ITT TERTIARY BUTYL IODIDE (Methyl-2-iodo-2-propane), CH 3 >CI-CH 3 . The constitutions of the four isomerides are deduced from those of the corresponding butyl alcohols from which they are obtained by the action of hydriodic acid. Of the AMYL derivatives eight isomerides are known. ii 7 98 ORGANIC CHEMISTRY METHYLENE CHLORIDE (Dichloromethane), CH 2 C1 2 , bromide and iodide (see Table, p. 95). ETHYLENE COMPOUNDS, CH 2 X-CH 2 X, are formed from ethylene by the addition of halogens or from glycol, C 2 H 4 (OH) 2 and halogen hydracids. ETHYLIDENE (or Ethydene)COMPOUNDS, CH 3 CHX 2 , are obtained by substituting the oxygen of the aldehydes by halogens. ETHYLENE CHLORIDE (Dichloro-i : 2-ethane), CH 2 C1-CH 2 C1 (Dutch liquid), boils at 84. The IODIDE, BROMIDE, and CHLORIDE with alcoholic potaeh give acetylene and glycol. ETHYLIDENE CHLORIDE (Ethydene chloride or Dichloro-i : i -ethane), CH 3 .CHC1 2 , is obtained from aldehyde and phosgene: CH 3 -CHO + COCJ 2 = C0 2 + CH 3 -CHC1 2 , chloral (which see) being also formed ; it boils at 57. CHLOROFORM (Trichloromethane), CHC1 3 . Chloroform was discovered by Liebig and Souberain and its constitution shown by Liebig in 1835. It is prepared from (1) ethyl alcohol or (2) acetone, by heating with chloride of lime and water : (1) 4C 2 H 6 OH + 16CaOCl 2 = 3H 2 C a 4 Ca (calcium formate) + 13CaCl 2 + 8H 2 + 2CHC1 3 ; in this reaction there is always an appreciable evolution of C0 2 , which appears to originate in the oxidation of the alcohol, and liberates HC10 and so forms aldehyde and hence chloral, this, in presence of lime, yielding chloroform : 3C 2 H 5 OH + 8Ca(OCl) 2 = 2CHC1 3 + 3CaC0 3 + C0 2 + 8H 2 + 5CaCl 2 . (2) 2CH 3 -CO-CH 3 + 3Ca(OCl) 2 = 2CH 3 -CO-CC1 3 (trichloro-acetone) + 3Ca(OH) 3 ; 2CH 3 -CO-CC1 3 + Ca(OH) 2 = Ca(C 2 H 3 O 2 ) 2 (calcium acetate) + 2CHC1 3 . In a very pure form for pharmaceutical use it is obtained by treating chloral with aqueous caustic soda solution, sodium formate being also formed : / H CC1 3 C( + NaOH = CHC1 3 + H C0 2 Na. % Chloroform can also be obtained industrially by reducing carbon tetra- chloride with hydrogen in the hot : CC1 4 + H 2 = HC1 + CHC1 3 ; the hydrogen necessary to treat 75 kilos of CC1 4 is given by 60 kilos of HC1 at 22 Baume and 50 kilos of zinc. To obtain very pure chloroform from the impure product, Anschtitz treats the latter with salicylic anhydride, C 6 H 4 C0 2 , which forms a crystalline mass only with chloroform, (C 6 H 4 C0 2 ) 4 , 2CHC1 3 ; this, after separation from the mother-liquor, is heated on the water-bath, when pure chloroform distils off. It is a colourless liquid with a sweet ethereal smell and taste ; it dissolves only to a slight extent in water (0-7 per cent.), but is soluble in alcohol or ether. It boils at 61-2, and its vapour pressure at 20 is 160 mm. of mercury ; its specific gravity is 1-5263 at and 1-500 at 15, referred to water at 4. It is non-inflammable, and it dissolves resins, rubber, fats, and iodine, with the last of which it gives violet solutions. Exposed to light and air, it decomposes partially into Cl, HC1, and COC1 2 , but it can be kept in yellow bottles, while that for pharmaceutical use keeps better if 1 per cent, of absolute alcohol is added. It is the most efficacious anesthetic (Simpson, 1848), but in some cases may cause death if not used with great care, since it acts on the heart ; to diminish this effect, it is mixed with atropine or morphine. 1 1 From coal-tar products various anaesthetics or hypnotics are produced synthetically, and these have been of great service to medicine, especially to surgery, rendering possible the execution of the most complicated opera- tions without any pain to the patient. At first substances were used which produced general ancesthesia of the organism, but they were accompanied by many inconveniences, sometimes by fatal results. Indeed, the anaesthetic is transported by the blood into contact with the higher nervous centres by which pain is felt, producing poisoning and paralysis of them often lasting for some time ; at the same time an influence is felt by the centres controlling the action of the heart and of respiration, this being the cause of the danger and disturbance produced by general anaesthesia. The nerve-currents start from the periphery, from the points where the surgical operation is to begin, and are transmitted to the brain, which transforms them into painful sensations, MANUFACTURE OF CHLOROFORM 93 In America, chloroform is used to render pigs insensible so as to kill them painlessly and to skin them more easily. Also, in fattening them, they are subjected to periodic inhalations of chloroform, which renders them more restful. Chromic acid transforms chloroform into phosgene (COC1 2 ), whilst potassium amalgam gives acetylene. With potassium hydroxide, it gives potassium formate and chloride : CHC1 H-C0 2 K 2H 2 0. FIG. 102. 3 4KOH = 3KC1 With ammonia at a red heat, it gives hydrocyanic and hydrochloric acids : CHC1 3 + NH 3 = HCN + 3HC1. Pictet Chloroform is pure chloroform obtained from the commercial product by freezing it at -80 to -120 ; the impurities remain in the liquid, the crystals giving pure chloro- form. INDUSTRIAL PREPARATION. A considerable amount of chloroform is pre- pared even to-day from chloride of lime and alcohol, but the latter should not contain fusel oil. The reaction takes place in a double -bottomed iron boiler, A (Fig. 102), which contains a mechanical stirrer, M, and into which the chloride of lime, water, and alcohol are introduced through a large aperture, F, at the top. The heat- ing is effected by a steam-coil, Pp, and cold water can be circulated through the jacketed bottom, when necessary, by means of another pipe not shown in the figure. To produce 100 kilos of chloro- form 100 kilos of alcohol and 1300 kilos of chloride of lime (with 36 per cent. Cl) are actually used ; but in practice a large excess of alcohol about ten times that really required by the reaction is employed, but the excess is used up, since it is added all at once and the process then continued by gradually replacing the quantity that reacts. An apparatus for producing 125 kilos of chloroform daily with four charges of the apparatus in 24 hours is charged first of all with 300 kilos of alcohol (96 per cent.) and 1300 litres of water, 400 kilos of chloride of lime being then added, in small quantities and with constant stirring ; the aperture F is then covered and the temperature raised to 40 by steam -heating. The steam is then shut off and the stirring continued until the temperature rises spontaneously to 60 (if this is exceeded, cold water is passed through the jacket). The mixing is then stopped and the chloroform, mixed with a little alcohol, begins to distil. The vapours are cooled and condensed in a coil, Z, placed in the lank, K, through which cold water circulates from V to ms. The mixed chloroform and alcohol is collected in a reservoir, L, with a graduated standpipe. When about 30 kilos of chloroform have distilled over, the stirrer is started again, and a little of the distillate and it is by influencing the cerebral centres by anaesthetics that pain is avoided ; but anaesthesia ceases to be dangerous if the peripheral nervous centres at the beginning of the nerve-currents are paralysed without the latter reaching the brain. Thus local anaesthesia is much more rational and less dangerous, since the insensibility extends only to one organ or one region of the subject of the .operation. So that, to chloroform, ether, &c., was added, in 1885, cocaine, which paralyses only the sensitive peripheral nerves without influencing the motor nerves. By studying anaesthetic and hypnotic substances chemists were able to determine what specific atomic groups produced anaesthetic properties in a molecule. Thus, with many of these substances, it was found to be the hydroxyl group which induced sleep, especially when it is united to carbon joined at the same time to several alkyl groups ; replacement of the hydroxyl by other groups resulted in the disappearance of the anaesthetic properties. Also various amino-acid gioups, under certain definite condi- tions, give rise to anaesthetics. To enumerate all the members of the vast group of anaesthetics which chemistry has placed at the disposal of surgery would be out of place here, but the following few examples may be mentioned : a-eucaine, ^-eucaint, orthoform, alipine, holocaine, and, on the other band, ttUphonal, trional, dormiol, hedonal, Veronal, &c. Other properties of anaesthetics are described in Part III, in the section on alkaloids. 100 collected from time to time from the tap, y, at the bottom of the condensing coil ; when the addition of water to this no longer causes separation of chloroform at the bottom of the liquid, the remainder of the distillate obtained finally the contents of the boiler are again heated with steam is collected at y, communication with the reservoir, L, being shut off and the tap, O, closed. More or less dilute alcohol now distils over and the distillation is stopped when the distillate contains less than 2 to 2ij per cent, of alcohol. The total amount of alcohol (usually 260-265 kilos) in the alcoholic distillate (500-600 litres) is determined, and sufficient pure alcohol added to bring the total quantity up to 300 kilos ; this dilute solution serves for the next operation, allowance being made for the water it contains. In this way the loss of alcohol is small. The crude chloroform is washed and agitated with water (30 litres per 100 kilos) to remove the alcohol present, or, better, with lime-water or a weak soda solution, which removes also the small quantity of HC1 that always forms. Finally, the liquid is agitated with concentrated sulphuric acid, thoroughly rewa?hed with water, dried over CaCl 2 and redistilled, the chloroform, passing over at 62-63, being collected. Instead of alcohol, acetone is used by some manufacturers when it can be bought cheaply, and in that case 100 kilos of acetone yields up to 170 kilos of chloroform. According to Ger. Pat. 129,237, a good yield and continuous formation of chloroform are obtained by heating, in a vessel divided into a number of cells communicating at the bottom, alcohol (35 Be.) which has been previously chlorinated by means of chloride of lime and alkali in the hot. During recent years the industrial preparation of chloroform has again been attempted by electrolysing an aqueous solution of KC1 (20 per cent.) into which alcohol or acetone is slowly introduced. In this process 1 h.p.-hour is consumed to produce 40 grms. of chloroform. Erlworthy and Lange (Fr, Pat. 354,291, 1905) propose to produce chloroform from methane and chlorine diluted with indifferent gases (N, C0 2 ) by subjecting the mixture to the action of light in suitable retorts : CH 4 + 6C1 = 3HC1 + CHC1 3 . TESTS FOR CHLOROFORM. Minute quantities of chloroform can be detected by gently heating a little of the liquid with a few drops of aniline and of alcoholic potash solution, the characteristic repulsive odour of phenylcarbylamine (phenyl isocyanide) being formed. Pure chloroform for medicinal use should not be acid or give a precipitate with silver nitrate solution or redden potassium iodide solution ; on evaporation it should not leave a residue of water or odorous substances, and it should not darken with concentrated sulphuric acid. To test for the presence in it of carbon tetrachloride, 20 c.c. are treated with a solution of 3 drops of aniline in 5 c.c. of benzene ; a turbidity or separa- tion of crystals of phenylurea indicates with certainty the presence of the tetrachloride. To ascertain if it contains alcohol it is treated with a very dilute potassium permanganate solution, which is decolorised in presence of this impurity. Its estimation is effected by treating a given weight with Fehling's solution (see under Sugar Analysis) and heating the mixture in a closed bottle on a water-bath for some hours (until the odour of chloroform disappears) ; the cuprous oxide, formed according to the equation CHC1 3 + 2CuO + 5KOH = K 2 CO 3 + 3H 2 O + 3KC1 + Cu 2 O, being weighed. One molecule of chloroform corresponds with 2 atoms of copper. It can also be determined by heating with alcoholic potash in a reflux apparatus on the water-bath ; it is then diluted with water, the alcohol distilled off, and the potassium chloride formed (together with potassium formate, see preceding page) titrated with a standard silver nitrate solution. This method serves for the estimation of all alkyl-halogen compounds. The price of industrial chloroform is about 8 per 100 kilos ; redistilled costs 2s. Wd. per kilo ; the pharmacopceial preparation 2s. 2d. ; puriss. from chloral, 6s. 5d. to 9s. Id. ; Pictet's, 12s. per kilo, and that of Anschiitz Wd. per 50 grms. Part of the chloroform consumed in Italy is imported from abroad ; in 1906 this amounted to 12,200 kilos ; in 1907, 10,100 ; in 1908, 7000 ; and in 1909, 9000 kilos -of the value 680. IODOFORM (Tri-iodomethane), CHI 3 , was discovered by Serullas in 1822, and its constitution was established by Dumas who, unlike his predecessors, did not overlook the very small proportion of hydrogen (0-25 per cent.) present. IODOFORM It is formed by heating ethyl alcohol or acetone with iodine and sufficient alkali hydroxide or carbonate to decolorise the iodine (Lieberis reaction) : C 2 H 5 OH + 81 + 6KOH = CHI 3 + H-COOK + 5KI + 5H 2 0. This reaction (separation of yellow crystals and formation of a character- istic odour) is so sensitive that it serves for the detection of minute traces (1 : 2000) of ethyl alcohol or acetone in other liquids (waiting 12 hours for the separation of crystals if the amount of alcohol is small) ; the same reaction is, however, given by isopropyl alcohol, acetaldehyde (and by almost all compounds containing the group CH 3 -CO-), but not by methyl alcohol, ether, or acetic acid. For the practical preparation of iodoform 32 parts of K 2 CO 3 are dissolved in 80 parts of water and 16 parts of alcohol, the mixture being heated to 70 and 32 parts of iodine gradually added. The separated iodoform is filtered off and the iodine of the potassium iodide in the nitrate utilised as follows : 20 parts of HC1 are added and 2-3 parts of potassium dichromate, the liquid being then neutralised with K 2 CO 3 , mixed with a further 32 parts of K 2 CO 3 , 16 parts of alcohol and 6 parts of iodine. On heating, a second quantity of iodoform separates, and after this or another similar operation the mother -liquor is treated to recover the iodine from the potassium iodide. It has been proposed to prepare iodoform by treating the metallic acetylides (see p. 91 ) with iodine and caustic soda. It seems that practical use is now made of the old electrolytic process, using a bath of 6 parts KI, 2 parts soda, 8 vols. alcohol, and 40 of water at 60-65. The iodine to be used in the reaction is set free at the anode and to avoid the formation of a little iodate with the KOH formed at the cathode the latter is enclosed in parchment paper. When pure, iodoform crystallises in hexagonal, yellow plates (sp. gr. 2), insoluble in water but soluble in alcohol or ether. It has a penetrating and persistent odour, ' recalling partly that of saffron and partly that of phenol. It melts at 119, readily sublimes, and is volatile in steam. On heating with either alcohol or reducing agents, it gives methyleiie iodide. It is used in surgery as an important antiseptic, which, however, acts indirectly on bacteria by means of the decomposition products formed from it under the action of the pus of wounds or of the heat of the body. Owing to its disagreeable odour, it has been to some extent replaced latterly by Xeroform, which is a tribromophenoxide of bismuth, C 6 H 2 Br 3 OH, Bi 2 2 , obtained by the action of bismuth chloride on sodium tribromophenoxide and forming a tasteless, odourless, yellow powder insoluble in water or alcohol ; it is used also as a disinfectant for the intestines, and costs 44s. to 48s. per kilo, whilst iodoform costs only 24s. to 28s. a kilo. TESTS FOR IODOFORM. It should leave no residue on sublimation and should dissolve completely in alcohol or ether. It is estimated by heating about 1 grm. with about 2 grms. of silver nitrate and 25 c.c. of concentrated nitric acid (free from chlorine) in a reflux apparatus so that the liquid does not boil ; when the nitrous vapours have disappeared the liquid is diluted with water to 150 c.c. and heated, the silver iodide being collected on a tared filter, dried and weighed : 1-789 grm. Agl corresponds with 1 grm. iodoform. CARBON TETRACHLORIDE (Tetrachloromethane), CC1 4 (see vol. i, p. 378). POLYCHLORO-DERIVATIVES OF ETHYLENE AND ETHANE. 1 Asymm. HEPTACHLOROPROPANE was prepared in 1910 by Boeseken and Prins from tetra- chloroethylene and chloroform in presence of aluminium chloride as catalyst. 1 Since 1908 (Ger. Pats. 196,324, 204,516, 204,883, &c.), the Chemische Fabrik Griesheim-Elektron of Frankfort, and the Usines electriques de la Lonia of Geneva have placed on the market, as non-inflammable solvents for industrial purposes, six chlorinated compounds obtained as colourless liquids by the action of chlorine on acetylene. They are all good solvents for fats, resins, rubber, &c., and can replace advantageously benzene, carbon disulphide, and alcohol, since they are not inflammable and their vapours do not form explosive mixtures with air ; ovci ; 102 r.$>Jf.GANIC CHEMISTRY 1 *f ' C r " II. HALOGENATED DERIVATIVES OF UNSATURATED HYDROCARBONS These are obtained from saturated halogen derivatives by partial elimination of the halogen hydracid : C 2 H 4 Br 2 = HBr + C 2 H 3 Br. They are formed by incomplete satura- tion, with halogens or halogen hydracids, of the less saturated hydrocarbons : C 2 H 2 + HBr = C 2 H 3 Br (see Table in footnote). The allyl compounds, C 3 H 5 X, are formed from allyl alcohol by the action of halogen hydracid or of phosphorus and halogen. ALLYL CHLORIDE (Chloro-3-propene-i), CH 2 : CH.CH 2 C1, the bromide and iodide having analogous constitutions. They are related to the natural allyl compounds (garlic oil and mustard oil). Two stereoisomerides are known : H C Cl H C Cl a-chloropropylene, and iso-a-chloropropylene, H C CH a CH 3 C H CC. ALCOHOLS These form an important group of organic compounds containing one or more characteristic hydroxyls, the hydrogen of which has pronounced reactive properties, so that numerous series of other compounds are derived from the alcohols. They have a neutral reaction, although their chemical behaviour is analogous to that of the inorganic bases which always contain the anion OH'. The majority of these alcohols are colourless liquids, but those of high molecular weights are oily, solid, and sometimes of a yellowish colour. The first members of the series are soluble in water, but with increase of molecular weight the solubility decreases and the smell, generally slight, also tends to disappear. They are often found in nature either free or combined with organic acids, in the fats, waxes, fruits, essential oils, &c. According to the number of hydroxyl groups they contain, they are divided into mono-, di-, . . . polyhydric alcohols, and may belong either to the satu- rated or to the unsaturated series already studied in connection with the hydrocarbons of which they retain the fundamental characters ; added to the latter are those characteristic of the alcoholic group, which we shall study generally with the monohydric alcohols. carbon tetrachloride they have the advantage of not attacking the metal parts of the extraction apparatus, and the loss on extraction varies from 0-3 to 0-8 per cent. ; they are, however, dearer than the ordinary solvents and seem to be injurious to health. The properties of these compounds are given in the following Table : DlCHLORO- TRICHLORO- TETRA- TETRA- PENTA- HEXA- ETHYLENE ETHYLENE CHLORO- CHLORO- CHLORO- CHIORO- ETHYLENE ETHANE ETHANE ETHANE C,H 2 C1, CjHCl, C 2 C1 4 CjH^Cli CjHCl 6 C 2 C1. Common name Didine Trieline EtUine Tetraline Pentaline Specific gravity . 1-278 1-471 1-628 1-600 1-685 2 Boiling-point 52 85 119 144* 159 (185') Vapour pressure at 20 205 mm. 56 17 11 7 3 Specific heat at 18 0-270 0-233 0-208 0-227 0-207 . Heat of evaporation 41 cals. 57-8 50 52-8 45 Freesing-point -70 30 Uses and properties Readily dis- Dissolves Serves well Dissolves Readily dis- Has an solves rub- fats, paraf- for remov- resins and solves cellu- odour like ber fin, and va- ing spots varnishes, lose acetate camphor, seline hotter like turpen- for artificial and serves than ben- tine and al- silk and as an insec- 'ne cohol and dissolves cellu- cinemato- ticide lose acetate for films and graph films artificial silk MONOHYDRIC ALCOHOLS 103 I. SATURATED MONOHYDRIC ALCOHOLS The specific gravity of these is always lower than that of water and up to the C 16 member they distil unchanged at the ordinary pressure ; beyond that reduced pressure must be employed. That alcohols always contain a hydroxyl group OH can be shown by the following chemical reactions : The alcohols can be obtained by the action of silver hydroxide, AG-OH (which cer- tainly contains the group OH), or even of the alkalis or hot water, on halogenated hydro- carbons : C w H 2n+1 I + AgOH = Agl + C re H 2w+1 OH. With the halogen hydracids the hydroxyl separates from the alcohols in the form of water : C re H 2n+1 OH + HBr = H 2 + C n H 2w+1 Br ; and the same happens with oxy- acids, the so-called esters being formed : C n H 2w+1 OH + HN0 3 = H 2 + C n H 2n+1 NO 3 . Just as sodium and potassium react with water, liberating hydrogen, so do they act on the alcohols, from which only the typical hydrogen (hydroxylic), not united directly to carbon, is eliminated : C n H 2n+1 OH + Na = C n H 2n+1 ONa (sodium alkoxide) + H. Magnesium alkoxides are also easily obtained. With phosphorus trichloride, however, the hydroxyl group is eliminated : 3C n H 2n+1 OH + PC1 3 = 3C n H 2n+1 Cl = P(OH) 8 . On p. 16 the difference in constitution between ethyl alcohol and methyl ether has been demonstrated. If the hydroxyl group occurs in place of a hydrogen atom in the methyl group ( CH 3 ) at the extremity of the hydrocarbon chain, the primary alcohols are obtained, / ^ z \ all containing the characteristic group CH 2 -OH (i.e. Cf 1, e.g. propyl alcohol, CH 3 CH 2 CH 2 OH, and by oxidation of these alcohols are formed first aldehydes with the characteristic group ( X G'f ), and then acids with the characteristic carboxyl group COOH (i.e. Cf ). Substitution of a hydroxyl for a hydrogen atom in an intermediate methylene group ( = CH 2 )ln the saturated hydrocarbon chain yields secondary alcohols, which have the characteristic group ^>CH-OH (i.e. ^>C<^QTT) and on oxidation give ketones containing the special group ^>CO. Finally the substitution of the hydrogen of a branched hydrocarbon may take place in the methinic group (=CH), giving tertiary alcohols with the characteristic grouping =C-OH, the other three valencies of the carbon being united to three carbon atoms. When the secondary alcohols are oxidised they cannot give either acids or ketones with an equal number of carbon atoms, but, if the oxidation is energetic, the chain breaks, and then acids and ketones may be formed, but with less numbers of carbon atoms. According to B. Neave (1909), primary, secondary, and tertiary alcohols can be distinguished by the Sabatier and Senderens reaction (see p. 34), by passing the vapours of the alcohol over finely divided copper heated at 300 ; the primary alcohols form hydrogen and aldehydes (recognisable by Schiff s reaction ; see section on Aldehydes), the secondary ones give hydrogen and ketones (detectable by semicarbazide hydrochloride solution) and the tertiary alcohols give water and unsaturated hydrocarbons (which decolorise bromine water). The primary alcohols and the corresponding ethers have the highest boiling- points, the tertiary ones and, in general, those with branched chains showing the lowest boiling-points. In the group of alcohols the isomerism and the number of isomerides are 104 O R GA N.I C CHEMISTRY similar to those of the halogenated derivatives of the hydrocarbons, since the halogen atom is here replaced by a hydroxyl group. * The names of the primary alcohols are made from those of the corresponding hydro- carbons (see p. 31) with the termination ol, and those of the secondary and tertiary alcohols are derived from the names of the hydrocarbons with the longest non-branched chains ; or the secondary and tertiary alcohols may be regarded as derivatives of methyl alcohol or carbinol, CH 3 OH, formed by substitution of the hydrogen atoms of the methyl group. We have, hence, two different, but still equally clear, systems of nomenclature. For example : (1) Normal butyl alcohol: CH 3 CH 2 CH 2 CH 2 OH = butan-1-ol or n-propylcarbinol. 2 1 (2) Secondary butyl alcohol: CH 3 .CH 2 -CH(OH).CH 3 = butan-2-ol or methylethyl- carbinol. 2 l 123 (3) Isobutyl alcohol: CH 3 CH CH 2 OH = 2-methylpropan-3-ol or isopropylcarbinol. CH, 1 23 (4) Tertiary butyl alcohol : CH 3 C CH 3 = 2-methylpropan-2-ol or trimethylcarbinol. CH 3 OH PROCESSES OF FORMATION OF MONOHYDRIC ALCOHOLS. As well as from the halogen derivatives, the alcohols can usually be obtained by decomposing esters with acids, alkalis, or superheated water. This reaction is termed saponification or hydrolysis : C 2 H 5 0-N0 2 + KOH - KN0 3 + C 2 H 5 -OH. Tn a general way, the primary alcohols are formed by reducing the acids (C n H 2n 2 ) or aldehydes (C n H ow O) with nascent hydrogen : 7 CH 3 CC (acetaldehyde) + 2H = CH 3 CH 2 OH. H Since the acids, in their turn, can be prepared from the alcohols with one carbon atom less, we have at our disposal a general reaction for preparing synthetically any higher alcohol. The secondary alcohols are formed by reducing the ketones,'C w H 2W 0, e.g. : CH 3 CO CH 3 + H 2 = CH 3 CH(OH) CH 3 Acetone Isopropyl alcohol (see later, Aldehydes and Ketones). The tertiary alcohols are formed by the prolonged action of zinc methyl on acid chlorides, the intermediate compounds thus formed being decomposed with water. For the secondary and tertiary alcohols Grignard's reaction may also be employed (see later, Alkylmetallic Compounds). Of more industrial importance, however, is the preparation of some of the more common of these alcohols by the distillation of wood or the fermentation of certain carbohydrates (see later). In addition to the properties of the alcohols given above, namely, their behaviour towards acjds, halogens (which oxidise them), chlorides, and oxidising agents in general (which give aldehydes and acids), it may be mentioned that the higher alcohols (primary) are transformed into the corresponding acids by simple heating with soda lime. Traces of primary alcohols are detectable by oxidising with permanganate and sulphuric acid and then testing for aldehyde with a sulphurous acid solution of fuchsine. CONST ANTS OF ALCOHOLS 105 ft PHYSICAL CONSTANTS OF THE MONOHYDRIC ALCOHOLS Name and Formula Specific gravity Melting- point Boiling- point 1. Methyl alcohol, CH 3 -OH 0-812 (0) -94, -98 66 2. Ethyl alcohol, C 2 H 6 -OH 0-806 -112 -117 78 3a. Normal propyl alcohol (prim.) CH 3 -CH 2 -CH 2 -OH 0-817 -127 97 36. Iso propyl alcohol (sec.) CH 3 -CH(OH)-CH 3 0-789 (20) 81 4a. Normal butyl alcohol (prim.), C 4 H 9 -OH . 0-810 -80(-122) 117 46. Normal butyl alcohol (sec.), C 4 H 9 -OH 0-808 100 4c. Isobutyl alcohol, C 4 H 9 - OH . . ; : . 0-806 (20) 107 4:d. Tertiary butyl alcohol (trimethylcarbinol), C 4 H 9 -OH 0-786 (20) + 25 83 5a. Normal amyl alcohol (prim.) CH 3 -[CH 2 ] 3 -CH 2 -OH 0-817 (20) 138 56. Amyl alcohol of fermentation or isobutyl- carbinol, (CH 3 ) 2 CH-CH 2 -CH 2 -OH . 0-810 (20) 130 5c. Active amyl alcohol or sec. butylcarbinol, CH 3 -CH(C 2 H 5 )-CH 2 -OH 0-816 (20) 128 5d. Trimethyl- or tertiary butyl-carbinol, (CH 3 ) 3 C-CH 2 -OH 0-812 (20) 49 113 5e. Diethylcarbinol, C 2 H 5 -CH(OH)-C 2 H 5 0-831 (0) 117 5/. Methylpropylcarbinol, CH 3 .[CH 2 ] 2 -CH(OH)-CH 3 0-824 (0) 119 5g. Methylisopropylcarbinol, (CH 3 ) 2 CH-CH(OH)-CH 3 0-819 (0) 112-5 &h. Dimethylethylcarbinol, (CH 3 ) 2 C(OH).C 2 H 5 0-814 (15) -12 102 6. Normal hexyl alcohol (prim.), C 6 H 13 -OH . 0-833 (0) 157 7. Normal heptyl alcohol (prim.), C 7 H 15 -OH 0-836 175 8. Normal octyl alcohol (prim.), C 8 H 17 -OH 0-839 191 9. Normal nonyl alcohol, C 9 H 19 -OH 0-842 5 213 10. Decyl alcohol, C^^-OH . . . 0-839 + 7 231 11. Undecyl CuH 28 -OH . , . . + 19 131(15rrm.) 12. Dodecyl C^H^-OH . . 0-831 24 143 13. Tridecyl C^H^-OH . . . 30-5 156 14. Tetradecyl alcohol, Ci 4 H 29 - OH 0-824 38 167 15. Pentadecyl alcohol, C^H^-OH 45-46 16. Hexadecyl (cetyl) alcohol, C 16 H 33 -OH 0-818 50 190 17. Octodecyl alcohol, C^H^-OH . 0-813 59 211 18. Ceryl C 26 H 53 -OH . . . 79 19. Myricyl C 30 H 61 -OH . 85 corre- By the behaviour of the nitro-compounds (prepared from the spending iodides and silver nitrite) and also by the initial velocity and degree of esterification, primary alcohols can be differentiated from the secondary and tertiary ones. Various primary normal alcohols enter inorganic compounds as alcohol of crystallisation, e.g. BaO, 2CH 3 -OH; CaCl 2 , 4CH 3 -OH; KOH, 2C 2 H 5 -OH MgCl 2 6C 2 H 5 OH ; CaCL, 4C,H 5 -OH, &c. ; it is hence evident why calcium chloride cannot be used for drying alcohol, although it serves well in the case of ether. 106 ORGANIC CHEMISTRY METHYL ALCOHOL, CH 3 -OH (Methanol or Carbinol) This is called wood-spirit, since it was obtained by Boyle in 1661 from wood-tar, and is to-day prepared in large quantities by distilling wood. Its chemical composition was not determined until 1834 (by Dumas and Peligot). In nature it occurs in the form of its salicylic ester, in Gaultheria pro- cumbens (in Canada) and as butyric ester in the bitter seeds of Heracleum giganteum. PROPERTIES. When pure it is a colourless liquid, b.pt. 66, with a faint alcoholic smell ; it burns with a non-luminous flame, solidifies at very low temperatures and melts at 94. When 1 kilo is burned, 5310 cals. are developed. It dissolves in all proportions in water, alcohol, ether, or chloroform. Its specific gravity at 15 is 0-7984, and in aqueous solutions the amount of the alcohol present can be determined from the specific gravities. 1 Like spirits of wine (ethyl alcohol) it is intoxicating, dissolves fats, oils, &c., and when it is anhydrous it dissolves also calcined copper sulphate forming a bluish green solution. Jfrn It is more poisonous to the human organism than ethyl alcohol, since it produces fatty degeneration of the liver and undergoes changes quite different from those of ethyl alcohol, passing only in minimal quantities into the urine and being mostly oxidised in the organism. When heated with soda lime or with oxidising agents it readily yields formaldehyde and formic acid, and sometimes carbon dioxide ; r [when distilled with zinc dust it gives CO and H. With potassium it forms a crystalline alcoholate, CH 3 -OK, CH 3 -OH. INDUSTRIAL PREPARATION. In the laboratory methyl alcohol can be prepared by saponifying methyl chloride or iodide. Industrially, if wood is heated in retorts out of contact with air, after all the water has distilled over, gradual decomposition commences at 150, and between 150 and 280 acetic acid (about 5 per cent, of the weight of wood), acetone (0-1 to 0-2 per cent. ), methyl alcohol (0-5 to 0-8 per cent. ), certain ammonia bases, &c., distil over in the form of a reddish brown aqueous liquid of empyreumatic odour, termed wood-spirit, and containing about 10 per cent, of acetic acid, 1 to 2 per cent, of methyl alcohol and 0-1 to 0-5 percent, of acetone. Between 300 and 400 the distillate is mainly black, oily, dense wood-tar (about 10 per cent, of the wood), and at the fame time gases (about 6-5 per cent.) are developed which are utilised for heating the retorts. At the end of the distillation, charcoal (about 18 per cent.) remains in the retort?. If the distilla- tion is rapid, a greater yield of charcoal is obtained, whilst with slow heating more volatile and liquid products are obtained and only 9 to 10 per cent, of charcoal. As the principal product of the distillation of wood is acetic acid, the description of the apparatus employed in this industry will be left until later. Per Per Per Per Per Specific cent, by Specific cent, by Specific cent, by Specific cent, by Specific cent, by- gravity weight gravity weight gravity weight gravity weight gravity weight at 15-56" of at 15-56 of at 15-56 of at 15-56 of at 15-56 of alcohol alcohol alcohol alcohol alcohol 0-99729 1 0-96524 22 0-93335 42 0-89358 62 0-84521 82 0-99554 2 . 0-96238 24 0-92975 44 0-88905 64 0-84001 84 0-99214 4 0-95949 26 0-92610 46 0-88443 66 0-83473 86 0-98893 6 0-95655 28 0-92237 48 0-87970 68 0-82938 88 0-98569 8 0-95355 30 0-91855 50 0-87487 70 0-82396 90 0-98262 10 0-95053 32 0-91465 52 0-87021 72 0-81849 92 0-97962 12 0-94732 34 0-91066 54 0-86535 74 0-81293 94 0-97668 14 0-94399 36 0-90657 56 0-86042 76 0-80731 96 0-97379 16 0-94055 38 0-90239 58 0-85542 78 0-80164 98 0-97039 18 0-93697 40 0-89798 60 0-85035 80 0-79589 100 0-96808 20 METHYL ALCOHOL 107 To separate the methyl alcohol from the liquid products of the distillation these are subjected to fractional distillation in copper boilers with a Pistorius rectifier (see Ethyl Alcohol), and when the specific gravity of the distillate has increased from 0-9 to 1 all the methyl alcohol (crude wood-spirit) has passed over and forms a greenish yellow liquid with a disagreeable odour. To eliminate the majority of the impurities the liquid is mixed with about 2 per cent, of lime, left overnight, and then distilled with the Pistorius rectifying apparatus, the acetic acid remaining fixed by the lime. The crude methyl alcohol thus obtained has a specific gravity of about 0-816 (93 per cent.) and is colourless, but it turns brown on standing in the air and becomes turbid on mixing with water. To purify it, it is diluted with water to the sp. gr. 0-935 (about 40 per cent. ), left for several days, and after the superficial tarry layer which collects has been removed it is treated with 2 per cent, of lime and distilled almost completely. The distilled product is mixed with 0-1 to 0-2 per cent, of sulphuric acid and rectified, the concentrated alcohol distilling at 64 to 66, being collected separately ; this is used for many industrial purposes, although it contains a small proportion of acetone. The latter can be removed almost completely by transforming the alcohol into an ester (e.g. the oxalate, by treatment with concentrated sulphuric acid and potassium dioxalate), which is easily separated from the impurities ; by hydrolysing the ester with KOH, distilling and rectify- ing, pure methyl alcohol is obtained. The acetone can also be got rid of by combining the alcohol with CaCl 2 , giving the compound CaCl 2 , 4CH 3 -OH, which is stable at 100, so that the acetone can be distilled off at 56 together with the other impurities ; the residue is then decomposed with water and the pure methyl alcohol distilled. To ascertain if the alcohol still contains acetone, 10 c.c. of it are treated with caustic soda and an aqueous solution of iodine in potassium iodide ; no turbidity due to iodoform should be formed for some time. 1 According to Farkas's patent (Ger. Pat. 166,360, 1904) alcohol of 92 to 95 per cent, purity is obtained direct if the vapours from the distillation of wood, while still hot, are passed through hot NaOH solution (15 to 20 B6.) and then into hot fatty acids, r the alcoholic condensate being rectified by passing the vapours into milk of lime. USES AND STATISTICS. Methyl alcohol is used for the manufacture of formaldehyde and various aniline dyes, for the preparation of different varnishes and for the denaturation of spirit (ethyl alcohol). 1 Tests for Methyl Alcohol. When pure it should leave no residue on evaporation, should not have an acid reaction towards litmus, and should not contain ethyl alcohol, which can be detected as follows : a little of the liquid is heated with sulphuric acid, diluted with water and distilled, the distillate being treated with permanganate, then with sulphuric acid, and finally with sodium hydrogen sulphite ; if ethyl alcohol is not present this liquid will not give a violet coloration with fuchsine solution. Acetone and ethyl alcohol can also be detected by the iodoform reaction (Lieben's reaction : see below and also p. 101). Proportions of 2 to 3 per cent, of methyl alcohol can be detected by Scudder and Rigg's reaction (1906), which consists in treating 10 c.c. of the liquid at 25" with 5 c.c. of concentrated sulphuric acid and 5 c.c. of saturated permanganate solution, decolorising (after two minutes) with sulphurous acid solution and boiling until all smell of sulphur dioxide or acetaldehyde disappears. This liquid is then tested for formaldehyde by adding a few centigrams of resorcinol to 2 c.c. and pouring 1 c.c. of pure concentrated sulphuric acid to the bottom of the liquid ; a blue ring, due to the formaldehyde formed from the methyl alcohol, forms at the surface of separation of the two liquids. Deniges (1910) detects as little as 1 per cent, of ethyl alcohol by heating the methyl alcohol with bromine water and testing for the acetaldehyde formed with fuchsine solution decolorised with SO, (see Aldehydes). Estimation of the methyl alcohol in the commercial product is effected by the Krell-KrSmer method : 30 grms. of phosphorus tri-iodide is placed in a flask furnished with a long reflux condenser, down which is poured, drop by drop, 10 c.c. of the'methyl alcohol ; after a short time the methyl iodide formed is distilled from a water-bath into a graduated cylinder containing a little water ; when the distillation is completed, the condenser is rinsed out with water and the volume of the methyl iodide under the water measured at 15" ; 5 c.c. of pure methyl alcohol give 7-19 c.c. of methyl iodide. The acetone is estimated by Kramer's method : in a 50 c.c. graduated cylinder with a ground stopper are placed 10 c.c. of a 2N-caustic-soda solution, then 1 c.c. of the alcohol, and, after shaking, 5 c.c. of a 2N-iodine solution. After a short time 10 c.c. of ether free from alcohol are added, the liquid shaken and then allowed to stand ; the volume occupied by the ether is read off, an aliquot part of it evaporated to dryness on a tared watch-glass and the iodoform crystals dried in a desiccator and weighed : 394 parts CHI, correspond with 58 of acetone. A good commercial methyl alcohol should contain not more than 0-7 per cent, of acetone and at least 95 per cent, of the alcohol ; it should distil within 1 ; 5 c.c. of 0-1 per cent, permanganate solution should not be decolo- rised immediately when treated with 5 c.e. of the alcohol, and 25 c.e. of the alcohol, mixed with 1 c.c. of an acetic acid solution of bromine (1 part Br in 80 parts of 50 per cent, acetic acid) should give a yellow solution. Detection of Methyl Alcohol in Ethyl Alcohol. To 0-1 c.c. of the alcohol, in a test-tube, are added 5 c.c. of 1 per cent, potassium permanganate solution and 0-2 c.c. (not more) of pure, concentrated sulphuric add. The liquid is shaken and left at rest for 2 or 3 minutes, 1 c.c. of 8 per cent, oxalic acid solution being then added. The mixture i? again shaken and when it has assumed a brownish yellow coloration, 1 c.c. of concentrated sulphuric acid is added, decolorlsation then occurring in a few seconds. Five c.c. of rosaniline bisulphite are then mixed with the liquid, which is afterwards allowed to stand. With ethyl alcohol alone, an intense greenish to violet coloration is obtained, but this disappears after a few minutes. But if the alcohol contains even as little as 1 per cent, of methyl alcohol, the more or less blue coloration persists for several hours. 108 ORGANIC CHEMISTRY In 1902, Germany produced 5000 tons of the pure spirit, of which 1151 tons was exported, and imported 4273 tons of the crude product. In 1910 England imported 448,500 galls, of methyl alcohol and exported 47,290 galls. The United States exported 1,691,000 galls, in 1910 and 2,040,000 (179,600) in 1911. Pyroligneous alcohol of 90 per cent, strength (French) is sold at 4 12s. per 100 kilos ; that of 92 to 93 per cent, strength (English) at 4 17s. 6d. ; and that of 95 to 96 per cent, strength for lacs at 5 Is. Qd. ; the purest methyl alcohol, free from acetone, costs 7 per 100 kilos. ETHYL ALCOHOL, C 2 H 5 -OH (Ethanol, Spirit of Wine) This is found rarely in nature (as butyric ester in Pastinaca sativa) and sometimes as an abnormal product in certain vegetables and animals, whilst it is easily formed by the alteration (fermentation) of various organic vegetable substances (saccharine juices, fruits, &c.)- It has hence been known from the most remote times. Aqua vitae or spirit of wine, obtained by distilling alcoholic beverages, was used as early as the eighth century and gave rise to an industry which acquired great renown in the province of Modena in the fourteenth century. Various European races learnt the use of aqua vitse from the custom introduced everywhere by the soldiers, who consumed large quantities of it during the wars of the Middle Ages. But very soon the northern peoples, who did not produce aqua vitae from wine, began to prepare alcohol by suitable transformations of the starch in the cereals abounding in their countries. By the beginning of the nineteenth century alcoholic liquors (exciting and enfeebling the nervous system and the brain) were spread over the whole of the civilised world and produced the terrible social scourge of alcoholism, much more disastrous in its material and moral consequences than all the other maladies that afflict humanity (see later, Alcoholism). Later, however, alcohol gradually acquired agricultural and industrial importance owing to its increasing practical applications in the arts and industries. Since 1830 Germany has extended the manufacture of potato spirit, and in many districts great agricultural advantages have followed the culture of this vegetable, since the waste products from the distilleries serve as nourishment for large numbers of cattle a source of great direct and indirect profit owing to the abundance of stable manure, which increases the fertility of the land and hence also the crops. SYNTHESIS OF ALCOHOL. In the laboratory alcohol can be obtained synthetically by hydrolysing ethylsulphuric acid, prepared from ethylene and concentrated sulphuric acid (Faraday and Hennel, 1828). Alcohol is formed by hydrolysing ethyl chloride, and, since ethyl chloride is prepared from ethane, which, in its turn, can be obtained from acetylene and hydrogen at 500 (or in presence of platinum black), the synthesis of alcohol from acetylene can be effected (Berthelot, 1855). Further, acetylene can be obtained from so-called inorganic substances, from C and H (Berthelot) ; by decomposing calcium carbonate with an acid, carbon dioxide is obtained, and magnesium, burnt in this gas, gives carbon, which, with lime, gives calcium carbide, and this, with water, acetylene ; there is hence a transformation of mineral sub- stances into organic substances. In 1907, Jonas, Desmonts, and Deglotigny (Fr. Pat. 360,180) proposed preparing alcohol by first forming acetylene in mercurous nitrate and then heating the mass to boiling ; the precipitate decomposes, regenerating the mercury salt and evolving vapours of acetaldehyde, which are condensed and converted into alcohol by means of sodium amalgam (nascent hydrogen). PROPERTIES. When pure, it is a colourless liquid with a characteristic odour, sp. gr. 0-7937 at 15, 0-80625 at ; it boils at 78-3 (or at 13 under 21 mm. pressure), and its vapour is stable at 300 ; at a very low temperature it gives a glassy mass, which at 135 is converted into another solid mass m.pt. 117 (enantiotropy, vol. i, p. 191). When concentrated (absolute) it is extremely hygroscopic, and it mixes PROPERTIES OF ALCOHOL 109 with water or ether in all proportions. To obtain absolute alcohol, i.e. absolutely free from water, fractional distillation is not sufficient, since at 78-15 an aqueous alcohol containing 95-57 per cent, of alcohol by weight distils ; the higher alcohols also give mixtures with water which boil at lower temperatures than the corresponding alcohols. If benzene is mixed with alcohol, the latter can be obtained pure although a mixture of water and benzene first distils over, then alcohol (at 64-8), then alcohol and benzene (68-2) and finally pure alcohol. Usually absolute alcohol is obtained by distilling the ordinary 90 to 96 per cent, alcohol over calcined potassium carbonate or over anhydrous (i.e. calcined) copper sulphate, redistilling over lime and finally over baryta or a little sodium or calcium ; or it may be left over powdered aluminium until hydrogen ceases to be evolved. The aldehydes of the alcohol can be separated by boiling with 5 per cent, of caustic soda. If alcohol contains a little water, it becomes turbid on mixing with benzene, carbon disulphide, or paraffin oil, and turns white, calcined copper sulphate blue, and barium hydroxide is precipitated on addition of baryta, the latter dissolving only in the absolute alcohol. A mixture of 53-9 vols. of alcohol with 39-8 of water gives 100 vols., the contraction of 3-7 per cent, being due to the formation of a labile compound, (C 2 H 5 -OH) 18 ,H 2 (or 2H 2 0, &c.). It is a good solvent for resins, oils, colour- ing-matters, varnishes, ethereal essences and many other substances, and dis- solves sulphur and phosphorus to a slight extent ; it coagulates proteins and diffuses through porous membranes more rapidly than water. It dissolves and gelatinises soaps. 1 It unites with various salts and alkalis as alcohol of crystallisation (KOH, LiCl, CaCl 2 , MgCl 2 ) (see p. 107). It oxidises easily, giving aldehyde and acetic acid, e.g. with potassium dichromate, Mn0 2 or even H 2 S0 4 , or oxygen in presence of platinum, or with micro-organisms if the solution is dilute. With concentrated nitric acid, it gives various oxidation products and with the dilute acid, glycollic acid. Alcoholic solutions of caustic alkalis turn brown, since they are partially resinified by the aldehyde which forms first and which acts as a reducing agent. Chlorine gives acetaldehyde and various intermediate chlorinated products. In a red-hot tube it decomposes, giving hydrogen and many hydrocarbons and acids. With sodium it gives sodium ethoxide in the form of a white powder. Absolute alcohol, which plays an important part in organic syntheses, is poisonous and rapidly produces death when injected into the blood. The complete combustion of 1 kilo of pure alcohol (C 2 H 5 -OH + 60 = 2C0 2 + 3H 2 0) generates 7193 cals. and 96 per cent, alcohol, about 6750 cals. Alcohol can be detected even in traces (1 : 2000) by means of Lieben's iodoform reaction (see pp. 101 and 107). This reaction is also given by acetone, isopropyl alcohol, and the aldehydes ; according to Buchner (1905) it is preferable to heat the alcoholic liquid with a little paranitrobenzoyl chloride, which forms crystals of ethyl paranitrobenzoate, N0 2 -C 6 H 4 -C0 2 C 2 H 5 , m.pt. 57. In Rimini's reaction, the liquid is heated with sulphuric acid, and a dilute solution of potassium dichromate : the green colour of the solution and the odour of acetaldehyde are sufficiently characteristic, but the reaction can be confirmed by distilling a few drops of the liquid and treating the distillate 1 Solid Alcohol is nothing but a soapy mass formed from about 20 per cent, of water, 20 per cent, of soap (sodium stearate) and 60 per cent, or more of alcohol ; it burns like liquid alcohol but leaves a residue. A richer product can be prepared by heating and stirring 100 parts of 96 per cent, alcohol at 60, dissolving 1 part of stearine and adding 0-5 part of a 30 per cent, aqueous sodium hydroxide solution just sufficient to make it redden phenolphthalein. Some use a sodium soap charged with silicate (500 per cent.). A solid alcohol that burns without leaving a residue can be obtained by dissolving 20 to 40 parts of collodion in 100 parts of alcohol ; others add, instead, 25 parts of a 25 per cent, solution of cellulose acetate in acetic acid, and shake, the crust of solid alcohol which separates being squeezed out. 110 ORGANIC CHEMISTRY with a little sodium nitroprussiJe and a drop of piperidine, a beautiful blue coloration being obtained if acetaldehyde is present. The manufacture of alcohol became One of the great chemical industries when a scientific explanation was obtained of the phenomena governing the transformation of starch and sugar. Fermentation, although known from the most ancient times, remained unexplained up to the nineteenth century, and it is solely, or largely, owing to the studies of Caignard de Latour and Schwann, Turpin, Schroeder, Liebig, Pasteur, Nageli, Cohn, de Bary, and, more recently, Duclaux, Buchner, &c., that the phenomena of fermentation are now completely explained and rationally regulated. In 1836 Caignard de Latour and Schwann found that the fermentation of wine and beer is strictly dependent on the germination of microscopic fungi which multiply in the must or wort. Turpin supposed that these fungi are nourished by the sugar, producing, as the excreta of their vital action, alcohol and carbon dioxide. In 1838 Liebig held that this transformation of sugar is caused by a special inter molecular movement due to substances contained in the ferment itself "(microscopic fungus). Pasteur, in 1872, showed that certain ferments that live at the expense of the oxygen of the air and can decompose sugar into water and carbon dioxide, when they are immersed in saccharine liquids, being no longer able to assimilate oxygen from the air, extract it from the sugar, resolving the molecule of the latter into alcohol and carbon dioxide. Although Nageli, in 1879, had attempted to reconcile the hypotheses of Liebig and Pasteur, yet up to a few years ago all fermentative phenomena were interpreted on the basis of the ideas enunciated by Pasteur. Progress in the fermentation industry proceeded, part passu, with that of bacteriology. 1 1 Bacteriology is the science which studies morphologically and biologically the smallest, unicellular, vegetable organisms which are propagated with immense rapidity by segmentation. The cell is formed, as in the other organisms, of an extremely thin membrane which permits all the osmotic phenomena (see vol. i, p. 77), and encloses the protoplasm in which no central nucleus is visible, but in which there occur scattered granules (of starch and other substances), fat globules, vacuoles containing cell-sap, and sometimes crystals (e.g. of sulphur), while in certain bacteria the protoplasm holds various colouring-matters in solution. The temperature most favourable to their vitality varies, according to the species, from 5 to 40 ; they live, however, in a latent con- dition, at very low temperatures, although they do not reproduce, and they usually die at about 70 (excepting the spores, tee below). As, in general, they do not contain chlorophyll, they are nourished by complex organic substances already elaborated by other organisms and hence soluble or capable of being rendered soluble (sugars, organic ammonium salts, ammonium compounds, &c.) ; and in this they are clearly differentiated from vegetable organisms and approximate more to the animals. Nutrient matter for bacteria always contains phosphorus, sulphur, potassium, and calcium, and, in certain cases, magnesium and manganese. They live well and reproduce rapidly in meat-broth or nutrient gelatine. They tolerate more easily alkaline than acid media and direct sunlight kills many species of bacteria, even pathogenic ones. As a result of their vital actions, sub- stances are sometimes formed which kill the bacteria themselves. Different antiseptics have various actions on different bacteria, or else only a specific action on certain of them. The reproduction of bacteria takes place ordinarily by segmentation, that is, when the cell has reached a certain length a thin wall forms in the middle and divides the cell into two new ones ; these divide, in their turn, so that the reproduction of these organisms, which increase in geometrical proportion (1, 2, 4, 8, 16, 32, &c.), proceeds with prodigious rapidity and yields millions of individuals in a few hours. The universal distribution of bacteria is thus easily understood. When the vital conditions are rendered abnormal or difficult for bacteria, in many of them there occurs a contraction of their proto- plasm into a more compact mass (at the centre or laterally, according to the species), which forms a separate individual, the spore, much more resistant to cold (180) and heat (130-140), and even to antiseptics than the corresponding bacterial cell ; the spores can retain life even for some years. Under favourable conditions, the spore breaks its envelope and gives a cell which reproduces by segmentation like the original one. Only certain rare species of bacteria are provided with chlorophyll or other colouring-matters capable of assimilating carbon dioxide under the action of sunlight. These micro-organisms, termed bacteria or schizomycetes or microbes, are those which produce putrefaction and infectious diseases (cholera, carbuncles, typhus, tuberculosis, small-pox, diphtheria, &c.) ; they are classified, according to their form, into: (1) Desmobacteria (bacillus or vibrio forms like small rods); (2) Sphere bacteria (cocci and micrococci of spherical shape and termed diplococci if united in twos, staphylococci if joined in bunches, and streptococci if in strings) ; (3) Spirobacteria (spirilla of twisted shape). To give a concrete idea of their forms de Bary described them as analogous to a pencil, a billiard ball, and a corkscrew. On the basis of their different activities and physiological properties Cohn divided all the species of bac- teria into three characteristic groups : (1) zymogenic, or those which produce all the non-alcoholic feimentations ; (2) ehromogenic, which produce various colouring-matters (red, violet, yellow, &c.) ; (3) pathogenic, which cause diseases of man and animals. To recognise the latter given the difficulty of distinguishing them morphologicaly under the microscope, since different species often have the same form and the same species sometimes several forms they are inoculated into the blood of living rabbits, rats, guinea-pigs, &c., the pathogenic character being deduced from the effects produced in the animals in two or three days, or sometimes even after a few hours. The lesser diameter (width) of these unicellular bacteria measures a few tenths of a micron (1 micron or /* = 0-0001 mm.), and, in rare cases, as much as 1-7 M ; the greater diameter (length) is usually several microns. If we wish to indicate bacteria in a wider sense of the term, and not to limit them to the pathogenic or sapro- phylic (non-pathogenic) but still to those that produce all putrefactions and widen the limits of their dimensions, FERMENTATION 111 During recent times, however, new facts have been discovered which have profoundly shaken the fundamental basis of this theory, according to which no fermentation is possible, except in the presence of certain species of living micro -organisms. In reality certain special fermentations are already known which are produced by enzymes, i.e. substances of complex chemical compositions which do not manifest anything in the nature of living micro-organisms ; for example, diastase transforms starch into maltose 2(C 6 H 10 5 ). V + H 2 = aAaH^Ou. 1 In 1900 Buchner succeeded in showing, by careful experiment, that some of these fermentations, which in the past could only be induced by the living micro-organisms, could also be effected by using the extract of the bacteria obtained by squeezing out, under great pressure, through special unglazed porcelain niters, the extract of the ferment- cells previously ground with quartz-sand. In this way Saccharomyces cerevisice yields maltase (which is an enzyme occurring also in germinating barley or maize and contained in Saccharomyces octosporus), which hydrolyses maltose, transforming it into glucose ; f i'om beer-yeast is obtained invertase (or invertin) capable of resolving saccharose or cane- sugar (not directly fermentable) into fructose and glucose (fermentable) ; fresh yeast calls yield zymase, the enzyme capable of effecting the alcoholic fermentation of various six-carbon-atom sugars (glucose, fructose, &c.). The action of the enzyme cannot be attributed to the still living protoplasm derived from the cells of the ferment, since the protoplasm can easily be killed in a mixture of alcohol and ether, and after this treatment the enzyme retains its activity. The action of ferments is hence due to the enzymes that they are able to produce, rather than to the biological phenomena of the life of the organisms. To-day numerous enzymes are known which, are of great importance in many vital functions of vegetable and animal organisms. It is not certain if the enzymes, with^their large and complex molecules, are true proteins, since up to the present they have not been obtained chemically pure ; all of them contain nitrogen but, as they are purified more and more, the nitrogen content continually diminishes and to-day it is held by some that the com- position of each enzyme approaches that of the substance it transforms ; so that diastase would be a substance similar to starch and poor in nitrogen, whilst the enzymes that transform the proteins would be of true protein nature. Proteolytic (decomposition of proteins) and fermentative actions only occur we can logically divide these micro-organisms into two other similar groups of similar beings, namely, the Hyphomycetes (moulds) and the Blastomycetes (ferments). The Hyphomycetes form groups of branched filaments (mycelia), which often subdivide into portions similar to bacteria, but the width of these always exceeds 2/x, and often 5/x ; they multiply by means of spores and four principal species are distinguished according to the mode of formation of these spores (conidia) : (1) the Aspergittus species which form, at the extremities of the fruit-bearing filaments (spore-bearing hyphce), a swelling in the form of a club covered with series of spores attached by means of intennediaFe steriymata ; (2) the Mucor species (or Mucedinece), in which the spore-bearing hyphse which start from the mass of mycelia carry sporangia (species of capsule) in which the spores develop ; (3) the Oidium species in which the spores are formed directly in the spore- bearing hyphae without any special organ of fructification ; (4) the Penicillium species, which is very common and has branched spore-bearing hyphse in the form of a brush containing series of spores AsperyiUus and Oidium are, however, not separate species but special sporifying forms of Eurotium and Erysiphw belonging to the order of Ascomycetes. The most important of these micro-organisms for industrial purposes are the Blastomycetes, i.e. the ferments or unicellular fungi which usually multiply by gemmation (budding), that is, by excrescences forming on the cells and becoming detached when they have reached a certain size, forming new cells which live independently of the mother-cells ; under abnormal conditions, however, the ferments multiply also by means of spores, four nuclei being usually formed inside the cell, these then becoming covered with membranes and dividing the mother-cell into four parts forming four new cells. The cells of the ferments have often a magnitude greater than 5/x, and the most important for alcoholic fermentation form the family of the Saccharomycetes (see later). The extraordinary beneficial influence of the bacteria and ferments in nature (apart from the pathogenic action of certain of them on some of the higher organisms) is manifested in the wonderful destructive activity they exert on the refuse and remains of all the higher organisms, converting the complex substances composing them into continually more simple substances until they give CO.,, H a O, NH,, and HNO,. These are the simplest materials which can be used by vegetable organisms to recommence the life-cycle, since in nature nothing is destroyed or created, but everything is transformed and thus life itself rendered eternal. 1 Starch, which is formed in the green leaves, of plants under the action of sunlight and of chlorophyll, although an insoluble substance and very resistant to various reagents, emigrates during the night and accumulates in the seeds, roots (tubers), medulla, &c. We can, however, stop the starch in its path, and can explain how it can be transported by the juices into other parts of the plant. In fact, various enzymes occur distributed through plants, and among these is diastase or amylase, which renders the starch soluble by transforming it into soluble (and hence transportable by the juices) sugar (maltose), to be regenerated by an inverse process unknown to us in the form of insoluble starch in other parts of the plant. 112 ORGANIC CHEMISTRY between certain limits of temperature (0-65) and are retarded or prevented by certain poisons (e.g. by traces of prussic acid or by metallic salts that act on proteins, like HgCl 2 , &c., although they are more, and sometimes completely, resistant to the action of antiseptics that kill ferments, such as salicylic acid, boric acid, ether, &c.). The various enzymes produce one or other of the following general reactions : hydrolysis (amylases, sucrases), coagulation (enzyme of rennet), decomposition (zymase of alcoholic fermentation), oxidation (laccase oxidises the juice of the lac-tree), &c. Enzymes exhibit different behaviour towards the stereoisomerides of certain hydrolysable and ferment- able substances (see section on Sugars). 1 1 The following are some of the more important enzymes : Diastase (or amylase) occurs abundantly in malt (germinating cereals) but is found also in plants, the pancreas, the saliva, the liver, the bile, the blood, the kidneys, and the mucous membrane of the stomach and of the intestines ; it transforms starch into maltose and dextrin. Maltase transforms maltose into glucose, and is found in malt, in Saccharomyces cerevisice, and in plants and animals. Zymase causes alcoholic fermentation of glucose and is contained in yeast and the alcoholic ferments (Saccharomyces). Lactase decomposes milk-sugar. Melibiase resolves rafflnose (or cane-sugar) into molecules of more simple sugars. Invertase (sucrase, saccharase, or inverting decomposes saccharose into glucose and levulose, and is obtained from beer-yeast. Cytase or Cellase attacks cellulose. Maltodextrinase ferments maltodextrin. Dextrinase ferments dextrins. Peptase governs the important digestive functions of the stomach, and peptonises proteins. Tryptase is found in the pancreas and contributes to the peptonisation and decomposition of the proteins, Lipase is also found in the pancreas and renders the fats soluble (hydrolyses them). Emulsin, contained in bitter almonds, and capable of decomposing amygdalin. Ptyaiiii is contained in the saliva and initiates the digestion of starchy foods. Reductase is capable of effecting reduction phenomena, especially in presence of aldehydes, and is hence also known as aldehydo-catalase ; it decolorises Schardinger' s reagent (mixture of methylene blue and formalin). Reductase is widespread in the animal kingdom and occurs in unboiled milk (boiled milk is detected by the lack of this enzyme ; it does not decompose water or decolorise guaiacol). The Oxydases form a group of enzymes (laccase, tyrosinase, aenoxydase, catalase, &c.) capable of effecting oxidations by fixing the free oxygen of the air and transferring it, in the nascent state, to the substances to be oxidised. They occur widespread in the vegetable kingdom and are also found in the animal kingdom, and their oxidising action is comparable to that of platinum black (catalyst). In fact the catalase found in the blood is capable of decomposing H 2 O 2 , giving nascent oxygen and water (Loew, 1901). It is now found that the oxydases are formed of mixtures of oxygenase and peroxydase. Euler and Boliu (1909) obtained a laccase of the Medicago type in a chemically pure state, and found it to be composed of calcium salts and a small amount of iron salts of mono-, di-, and tri-basic hydroxy-acids, especially citric, malic, mesoxalic, and glycollic acids. Peroxydases and Oxygenases. Schonbein (1856) had observed that certain vegetable and animal organisms contain substances analogous to ferments and capable of decomposing hydrogen peroxide catalytically with liberation of oxygen, and also of accelerating catalytically this decomposition (i.e., the oxidising action) in the same way that ferrous sulphate does. Loew (1901) showed that the first action is due to a special enzyme, catalase (oxygenase). Linossier, in 1898, succeeded in separating from pus an enzyme free from oxydase (oxygenase), yet capable of accelerating but notof initiating the decomposition of hydrogen peroxide ; this he called peroxydase. The oxydases and peroxydases often occur together and they may be separated by heating the mixture to 70, the oxydase being thus killed, or, as was proposed by Aso of Tokio (1902), by dissolving the peroxydase in alcohol which does not dissolve the oxydase, or by poisoning the latter with sodium fluoride or fluosilicate. There are also several plants that contain only peroxydases, among them pumpkins and horse-radish roots (Bach and Chodat, 1903, 1906). The peroxydases are nitrogenous but non-protein substances, and, when heated with NaOH give NH 8 ; they always contain about 6 per cent, of ash, 0-8 to 1-4 per cent, being aluminium and 0-2 to 0-6 per cent, manganese. The peroxydases dialyse, whilst the oxygenases do not. The specific action of the peroxydases consists in activating in a remarkable manner the oxidising action of H 2 O, on organic substances, e.g. gallic acid, pyrogallol, &c. ; they activate also the action of the peroxides that form in organic substances by the action of the oxygen of the air (e.g. ethereal oils, turpentine, &c.). In 1897 Bertrand introduced the following hypothesis to explain the action of the oxydases : the latter are regarded as hydrolysable manganous protein compounds, in which the manganese, in the mauganous condition, is the transmitter of oxygen from the air to the oxidisable substance ; the manganese dioxide formed would then be again reduced by the protein acid radical, the original manganous protein compound being regenerated. Bach and Chodat have, however, found manganese in the peroxydases, although these are not direct oxidising agents. The peroxydases have no oxidising action, unless a peroxide is present. They do not turn fresh guaiacol tincture blue, but after some hours this change does occur, the tincture having formed peroxide, which can be detected by starch and potassium iodide solution. Whilst the peroxydases accelerate the decomposition of very dilute H 2 O 2 , this kills them if concentrated. In 1908 J. Wolff obtained the reactions of the peroxydases by traces of ferrous sulphate or copper sulphate. The oxidising action of the oxygenases (which have, however, not yet been obtained free from peroxydases, although the latter are known free from oxygenases) is only weak and is strongly activated by addition of peroxydase. On the other hand, it seems established that there are two species of peroxy- dases existing, the one activating strongly the oxygenases and feebly the decomposition of H 2 O 2 , and the other behaving in the opposite way. The character of the oxydases themselves is indicated by the specific action of one or the other species of peroxydase. Indeed, Bertrand had in 1896 extracted from certain plants, e.g. young potato tubers) an oxydase which differed from all others in not oxidising phenols or the aromatic amines, whilst it oxidised and blackened tyrosine, which is not altered by the ordinary oxydases or even by the presence of H,O a alone. Bach (1906) succeeded in separating the specific peroxydase from tyrosinase and in showing that this peroxydase ENZYME ACTIONS 113 But still more interesting is the fact that during an ordinary fermentation the amount of sugar fermented does not depend closely on the quantity of living ferment or enzyme ; thus large quantities of sugar can be decomposed by small quantities of ferment or enzyme. The action of the enzymes and of the ferments may be logically compared with that of the inorganic catalysts (vol. i, p. 67), which only produce an enormous increase in the velocity of reaction, in our case, of the decomposition of sugar. And that these organic catalysts have an action really similar to that of the inorganic catalysts can be shown by certain other interesting facts. Some years ago Duclaux succeeded in producing alcoholic fermentation by dilute alkali ; Traube in 1899 transformed sugar into alcohol by means of finely divided platinum alone at 160 ; while Schade in 1906 converted an alkaline solution of glucose, in absence of enzyme, quantitatively into acetaldehyde and formic acid (C 6 H 12 O 6 =2C 2 H 4 O +2CH 2 O 2 ), and these products, under the catalytic influence of rhodium, are transformed quantitatively into C0 2 and alcohol (perhaps the formic acid first gives C0 2 and H 2 , the latter, in the nascent state, reducing the aldehyde to alcohol). 1 Further, as in chemical equilibria (vol. i, p. 62), the action of catalysts in reversible reactions is regulated by conditions of temperature and of con- centration different from those met with in the case of enzymes : indeed, when diastase has converted a certain quantity (dependent on the temperature) of starch into maltose, the hydrolytic change is arrested (i.e. equilibrium is reached in the reversible reaction : starch t; maltose) ; but if part of the maltose is fermented into alcohol and C0 2 , the equilibrium is disturbed and the diastase hydrolyses a further quantity of starch. Also at temperatures above 55, diastase forms dextrin in preference to maltose. An analogous phenomenon is observed in the hydrolysis of amygdalin by emulsin. It has already been mentioned that maltase transforms maltose first into glucose, but when a certain proportion between these two products is reached, the hydrolysis ceases owing to equilibrium being attained : C 12 H 22 U + H 2 ^ 2C 6 H 12 6 , and the transformation proceeds only when the glucose is removed by alcoholic fermentation ; Emmerling has realised the inverse reaction by displacing the equilibrium by addition of glucose (in which case isomaltose is produced). is only capable of causing the oxidation of tyrosine when mixed with the corresponding oxygenase or in presence of H 2 O a alone. Hence the action of tyrosinase is due to the specific action of its peroxydase. Bach holds further that in the phenomena of respiration of organisms, oxidation due to oxydases plays no part, since this leads to true condensations, to syntheses of more complex products ; for respiratory phenomena there should exist euzymes of a type not yet known and capable of decomposing and oxidising there serve materials of the organism (fats, carbohydrates, &c., which are not oxidised by oxydases). At the present day the catalytic action of the enzymes is explained as due to small quantities of metal which they contain ; thus the important action of the haemoglobin of the blood (which fixes the oxygen in the lungs in a labile condition and transports it to all parts of the organism) appears to be due to the small quantities of iron present, this inducing the decomposition of the food materials ; thus the synthetic action of the peroxydases is perhaps due to the manganese they contain (see above), just as the important synthetic functions of chlorophyll, according to Willstatter's recent work, appears to be owing to the magnesium present in it. Recently (1910) Bach has, however, succeeded in preparing very active oxydases and peroxydases free from iron and manganese, so that the true explanation of the activity of these enzymes remains to be discovered. 1 Buchner and Meisenheimer (1909) explain the action of ferments, from the chemical point of view, by the addition of a molecule of water to the sugar and abstraction of an atom of oxygen by the ferment, so that there results, as an unstable intermediate product, a dihydric alcohol, which, in its turn, is immediately decomposed into H 2 and 2 mols. of dihydroxyacetone ; the last product is able to decompose into CO 2 and alcohol, while the hydrogen continues to transform fresh quantities of sugar into the dihydric alcohol, and so on. Boysen-Jensen (1909) finds that the reactions for dihydroxyacetone are given by fermentations ; the decomposition would hence take place thus : CH 2 OH CH 2 OH CH 2 OH CH, CHOH CHOH CO - CO. + C'llj,- UJ1 CHOH CHOH CH,OH + H,0 = O + . - H 2 + p H ' OH CHOH CHOH CH, CHOH CHOH CO ~* C0 2 + - CH 2 -UH CHO CH,OH CH.OH Glucose Dihydric alcohol 2 mols. of 2 mols. Dihydroxyacetone of Alcohol ii 8 114 ORGANIC CHEMISTRY Also in the action of maltase on amygdalin, Emmerling succeeded in producing the reverse reaction, and at the St. Louis Exhibition in 1904 he showed a fine specimen of amygdalin prepared synthetically by an enzymic process. 1 So that with one and the same enzyme, analytic and synthetic processes can be effected. Cremer obtained glycogen (C 6 H 10 5 ) y from levulose by means of an extract of yeast, and Hanriot, Kastle, and Loewenhart prepared monobutyrin and butyl acetate synthetically by means of lipase. The enzymes also effect the so-called asymmetric syntheses, i.e. they give optically active compounds containing asymmetric carbon (1908). Also interesting is the fact that a single ferment may contain various enzymes ; thus, from Saccharomyces cerevisice, maltase and invertase can be extracted easily and also zymase, though with more difficulty. These recent discoveries on the reversibility of the reactions effected by enzymes are of great importance, as it was at first thought that enzymes or ferments in general were capable of causing only decompositions and not synthetical reactions, whereas their analogy with inorganic ferments is now complete. But the discoveries are all the more remarkable, since the same phenomenon of vitality in the single cell as in more complex organisms can be reduced to an enzymic phenomenon ; that is to say, the exchange of material in the organism (decomposition, recomposition, growth) takes place by means of these organic catalysts, which cause the decomposition of food, preparing various complex materials which form the organism itself, and at the same time generating the energy manifested in the vitality, enzymic phenomena being always exothermic. This hypothesis can, with advantage, be substituted for the too abstract biogen z hypothesis, to explain vital phenomena. 1 C.Hs-CHfCI^-C.HnOo + C.H 1S 8 ; C 20 H 27 NO n + H 2 O Glucoside of phenylgly- Maltase Amygdalin collie nitrile or, more completely : 2CH 12 O, + HCN + C 6 H,,-CHO ^ 2H 2 O + C 20 H 2 ,XO U Glucose Hydrocyanic Benzal- acid dehyde Hypotheses of Biogen, Toxins, and Genesis of Life. The physical and physiological basis of life resides especially in the protoplasm, the semi-fluid, almost always colourless, refractive substance insoluble in water which everywhere constitutes the essential part of the cell. Protoplasm is formed principally from protein sub- stances, whilst it is thought that the fats and carbohydrates are not active components. To the protoplasm is attributed the fundamental property of vitality, i.e. the exchange of material, but it is not known how its com- ponents the proteins can have such properties or in what physico-chemical aggregation of the proteins (the plasti- dules and bionomads are regarded as morphological components or units of protoplasm) they have their orign. In animals one of the principal functions of the blood is that of supplying the respiratory needs of the tissues in virtue of the haemoglobin contained in the blood of vertebrates [besides fibrinogen, serum-albumin, and para- globulin; whilst with the invertebrates there are echinochrom, chlorocruorin, hcemoerythrin, hcemocyanin (con- taining copper), and pinnoglobin (containing manganese), which have the same functions as hsemoglobin] ; it is formed of a protein substance united with a ferruginous compound, which takes up oxygen at the respiratory surfaces of the organism (skin, bronchi, and lungs), and brings it into close contact with the tissues. The vital processes of the organism being due to the exchange of material in the cells full of protoplasm, the biogenic hypothesis assumes that this is brought about by a very complex, labile compound, which, by being con- tinually decomposed and reconstituted, maintains the interchange uninterruptedly. By many this compound is called living albumin, but Max Verworn (1895 and 1902) regards this as an unsuitable name and does not think it has been shown to be a true albuminoid, although it is a nitrogenous substance ; there arc possibly several sub- stances in a state of labile combination and these he calls molecules of biogen. It has been observed that in organisms, as in parts of them, vitality ceases when oxygen is eliminated, many of them subsequently (the frog even after twenty-five hours) recovering it in presence of oxygen. From this arise two hypotheses : (1) the molecule of biogen becomes labile, and hence gives rise to decompositions and recompo- sitions, that is, to the vital process since it unites transitorily with oxygen ; (2) oxygen serves only to oxidise or eliminate the decomposition products of the biogen (admittedly labile), and when there is no oxygen, these products are not eliminated, so that the decomposition and recomposition of the biogen arc arrested. By experiments on the frog Max Verworn has shown that the foimer hypothesis is the more probable. Since, in the vital process, under the action of oxygen, it is especially the carbon dioxide that is eliminated, often along with lactic acid, water, &c., whilst the elimination of nitrogenous substances does not increase, it may be assumed that biogen is constituted of a benzene nuckus with lateral chains of carbohydrate and aldehydic character and with an oxygen-carrying nitrogenous group which fixes the oxygen of the air (just as NO gives NOj in the lead-chambers of sulphuric acid works) and gives it up to the lateral chain, which is oxidised (Ehrlich's side-chain hypothesis, 1882-1902) to CO 2 , lactic acid, H.O, substances, 0-880 parts have disappeared, i.e. have been fermented. From the degree of apparent fermentation (B), the degree of apparent attenuation can, of course, be obtained : thus, = B gives p m = Bp ; and from the factor a mentioned above, the amount of alcohol resulting from such degree of apparent fermentation is known. The real attenuation (A') is determined by distilling a certain quantity of the fermented wort until its volume is reduced to one-third, the residue being made up to the original volume with water and the density, n, measured ; the real attenuation then = p n. But, since the residue always contains unfermented matter, in order to calculate the alcohol, a factor, b, is determined in the same way as the factor, a, i.e. by distillation of a part of the fermented wort ; the quantity of alcohol can then always be determined from the density of the fermented wort, for, since A' = (p n) b, b = . Similarly, the degree of real fermentation will be B' = -which p n p expresses the fraction of the extract (dissolved substance without alcohol) really fermented, the manufacturer being thereby able to judge if the fermentation proceeds normally and to establish comparisons with previous fermentations, &c. The apparent attenuation (alcohol being present) is always greater than the real (derived after elimination of the alcohol) and the attenuation difference, D, is obtained by subtracting one from the other, (p m) (p n) = D- This magnitude,!), is therefore equal to n mand increases as the fermentation proceeds towards completion; also here the quantity of alcohol already formed is found by determining experimentally a factor, c, in the usual way, A p m so that ^_ = c, or A = (n m).c. The ratio of the apparent to the real attenuation, n = q, gives a quo- tient of attenuation which varies with the concentration of the liquid but becomes constant towards the end of the fermentation and shows how much the apparent fermentation is greater than the real ; by its means, almost all the saccharometric calculations can be made : ~ = the alcohol factor for the real attenuation, and if this is q divided by q diminished by unity [i.e. by (q 1)], the factor, c, for the difference of attenuation is obtained D The factor, r, is used for the analysis of liquids for which the value of p is unknown also = B' (degreeof real fermentation). The following illustrates a practical calculation : the original saccharometric degree of a wort was p = 16-2, and that after fermentation m = 1, and that after boiling n = 3-9 ; applying any one of the three factors (a, II 9 130 ORGANIC CHEMISTRY way of maltose and dextrin. Of these moulds, Amylomyces Rouxii, discovered by Calmette in 1892, and the Mucors B and C discovered by Collette, Boidin, and Mousain, are of most importance industrially. 1 Of the first two, the forms observed under the microscope in different stages of development are shown in Fig. 125 (A, B, C, D, and E). b, and c) given in the appended Table, the apparent attenuation becomes A = (p m) a (where p = 16-2, a = 0-4267) = 6-4858 per cent, of alcohol. Calculating according to the real attenuation, A = (p n) b (where p 16-2, n = 3-9, and 6 = 0-5274) = 6-4870 per cent, of alcohol. Lastly, calculating from the attenuation difference, D, A (n m) c (where c = 2-2350) = 6-4815 per cent. Hence the fermented wash consists of 6-48 per cent, of alcohol, 3-9 per cent, of unfermented extract (n), and 89-62 per cent, of water. TABLE FOB, CALCULATING THE ATTENUATION IN FERMENTED WORTS Alcohol factors for Saccharo- meter degrees of the wort the attenuation Factors for the attenuation difference Attenuation quotient c Values of r Apparent Real b P a b c 9 6 . 0-4073 0-4993 2-2096 1-226 4-4247 7 . 0-4091 0-5020 2-2116 1-227 4-4052 8 . 0-4110 0-5047 2-2137 1-228 4-3859 9 . 0-4129 0-5074 2-2160 1-229 4-3668 10 . 0-4148 0-5102 2-2184 1-230 4-3478 11 . 0-4167 0-5130 2-2209 1-231 4-3289 12 . 0-4187 0-5158 2-2234 1-232 4-3103 13 . 0-4206 0-5187 2-2262 1-233 4-2918 14 . 0-4226 0-5215 2-2290 1-234 4-2734 15 . 0-4246 0-5245 2-2319 1-235 4-2553 16 . 0-4267 0-5274 2-2350 1-236 4-2372 17 . 0-4288 0-5304 2-2381 1-237 4-2194 18 . 0-4309 0-5334 2-2414 1-238 4-2016 19 . 0-4330 0-5365 1M!U,S 1-239 4-1840 20 . 0-4351 0-5396 2-2483 1-240 4-1660 21 . 0-4373 0-5427 2-2519 1-241 4-1493 22 . 0-4395 0-5458 2-2557 1-242 4-1322 23 -. 0-4417 0-5490 2-2595 1-243 4-1152 24 . 0-4439 0-5523 2-2636 1-244 4-0983 25 . 0-4462 0-5555 2-2677 1-245 4-0816 26 . 0-4485 0-5589 2-2719 1-246 4-0650 27 . 0-4508 0-5622 2-2763 1-247 4-0485 28 . 0-4532 0-5636 2-2808 1-248 4-0322 29 . 0-4556 0-5690 2-2854 1-249 4-0160 30 . 0-4580 0-5725 2-2902 1-250 4-0000 B. Wagner, F. Schultze, and J. Rub (1908) suggest the Zeiss immersion refractometer as a means of deter- mining the attenuation : exact results are obtained rapidly and with a small quantity of liquid (20 to 30 c.c.). A little of the wort is well shaken to get rid of carbon dioxide, and filtered through a covered filter, 5 c.c. of the filtrate being used to determine the refractometer reading, A, at a temperature of 17-5 ; a further 20 c.c. are evaporated to one-half the volume in a porcelain dish to expel the alcohol, the volume being then made up exactly to 20 c.c. with water and the refractometer reading, B, taken. From the difference, AB=C, 15 (the refracto- meter reading for water) is subtracted, giving E ; the corresponding alcohol degree (by volume), V, is then found in the following Table and can be subsequently corrected for the density of the wort : E: V : 16-2 17-5 18-8 20-1 21-4 22-8 24-2 25-6 27-1 28-6 30-1 31-7 33-3 34-9 36-4 38-0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 Among the Ilyphomyceles (moulds, p. Ill) in the Mucor and Mucedirue Pasteur found certain varieties (Mucor racenwsm) capable of transforming sugar into alcohol and carbon dioxide when they live immersed in the liquid out of contact of air (like the yeasts) ; in presence of air, they convert the sugar directly into water and carbon dioxide. These are called facultative anaerobic organisms. In 1887 Gayon studied other varieties which behave similarly (Mucor alternant, spinosus, and cirrinettoides), and Prinsen Geerligs investigated Chlamy- domucor oryzce, which is used in Java to ferment molasses. In 1892 Calrnette imported from China, studied, and named Amylomyces Rouxii, the Mucor isolated from the rice-ferment used by the Chinese (which is more active than the Japanese l-oji) for the preparation of spirit ; later he found this Mucor in rice-husks. At Tokyo in 1894, Takamine studied, and applied practically to the saccharification of rice, Aspergillus oryzce (separated from Japanese koji, which is a mixture of yeasts and moulds used in Japan for producing alcoholic fermentation), but it did not meet with success, owing to its action being too energetic. Boidin, Collette, and Mousain investi- gated Mucor /3, which is another Mucor separated from Japanese koji and is different from, and more important industrially than, that of Takamine ; Mucor y, which was separated at the same time from Tonkin rice, is of still greater practical value than Mucor /s. These moulds have the special property of saccharifying starch and of fermenting the sugar thus formed. Their saccharifying and fermentative activity is, however, influenced by the acids that they produce. Thus, Amylomyces Rouxii, which was the first to be used in practice in 1898, was abandoned later, as it transforms rather too much sugar into carbon dioxide and water and, owing to the production of 1-45 grms. of acid per litre of wort (at AMYLO PROCESS 131 Collette and Boidin patented in 1897 (Eng. Pat. 19,858) a process for the industrial utilisation of Amylomyces Rouxii for manufacturing alcohol directly from the starch of cereals, &c., and later they utilised Mucor ft. At the present time this process is employed on an enormous scale in varioiis distilleries in France, Belgium, and Italy (at Savona). As it is necessary to work with perfectly aseptic worts, the starch-paste prepared in the ordinary way with the Henze apparatus is passed into closed metal cylinders holding 200 to 1000 hectols. and furnished with vertical stirrers. When the temperature reaches 65, 1 per cent, of malt (on the amount of maize used) is added to render the mass rather more liquid ; after an hour the mash is slightly acidified by the addition of 0-1 grm. of sulphuric acid per litre, and is then rendered completely sterile by passing steam in at the bottom and boiling the wort until the steam issues freely from the upper aperture. The apparatus is then closed hermetically, a vacuum being produced by the condensation FIG. 125. A. Colonies of Amylomyces Rouxii in wort-gelatine. B. Mycelial conidia of Amylomyces Rouxii in aerobic cultures. C. Segmentation into gemmae of the mycelium of Amylomyces in anaerobic culture. D. Hyphse of Mucor /3 (1 : 100) with sporangia in aerobic culture. E. Mycelium of Mucor with spores in different stages of development in anaerobic culture : 1, spores just sepa- rated ; 2, turgid spores ready to germinate ; 3, germinating spores ; 4, mycelium (1 : 600). of the steam. The vacuum is relieved by allowing sterilised air filtered through cotton- wool (see p. 124) to enter ; the maintenance of a slight pressure inside the vessel prevents the entry of germs. By stirring the starch and running cold water down the outer walls of the cylinder 1000 hectols. of boiling wort may be cooled in five hours to 38 ; this is the most suitable temperature for the Mucor fermentation, but a great part of the 16 Balling), complete attenuation is obtained only in very dilute worts (7 to 8 Balling, these giving 4 to 4-5 per cent, alcohol) ; Mucor ft, on the other hand, forms only 0-75 grm. of acid, and can ferment worts at 10 to 17 Balling (which give 8 to 9 per cent, of alcohol) without oxidising completely more than a very small proportion of sugar. Calmette studied more particularly the saccharifying properties of Amylomyces Rouxii, but in 1897 Boidin and Bolants, and simultaneously Sanguinetti (Institut Pasteur, 1897) found that this mould is also capable of transforming sugar and dextrin into alcohol ; it was found later that Mucor racemosus, which had been already studied by Pasteur, behaved similarly. In 1895 Professor Saito, of Tokyo, isolated Rhizopits oligotyorus, which acts like Amylomyces Rouxii. 132 ORGANIC CHEMISTRY sulphuric acid added must first be neutralised. A vat of 1000 litres capacity contains 150 to 200 quintals (15 to 20 tons) of maize and six times as much water. The Amylomyces is cultivated in the laboratory on 100 grms. of rice and 200 c.c. of sterile wort, so that preferably spores are developed. Every culture-flask contains a total of about 0-1 grm. of spores, and this quantity is sufficient to inoculate 1000 hectols. of wort. The Mucor is introduced, under aseptic conditions, into the vats from above and the stirrer set in motion ; a little air is introduced, this issuing by an upper tube with a hydraulic seal. In the course of 24 hours the wort is attacked by an abundant growth of the Mucor. The mass is then cooled to 33 and, in order to com- plete the alcoholic fermentation more rapidly, a small quantity of ordinary yeast (500 c.c. of a wort culture, corresponding with 3 to 4 grms. of pressed yeast) is added. After 3 to 4 days, the alcoholic fermentation is complete (the carbon dioxide passes out at the top through the water-seal). Fig. 126 shows dia- grammatically a plant with five large fermentation vessels. FIG. 126. The advantages of the amyio-process are : (1) a considerable saving in malt, only about 1 per cent, being used instead of 12 to 15 per cent, by the old process ; further, air-dried malt is difficult to keep in hot countries ; (2) the reduction of the amount of yeast required to a minimum. The yield of alcohol is also sensibly increased, one quintal of maize containing 57 to 58 per cent, of starch yielding 37-5 litres of alcohol, i.e. 65 (often 66) litres of pure alcohol per 100 kilos of starch ; the old method of working gives only 60 to 61 litres. The increase in the alcohol-yield is naturally due to the fermentation taking place in a wort uncontaminated with extraneous micro-organisms ; on rectification, 4 to 5 per cent, more good spirit (bon gout) are obtained than by the old process. Finally, the spent wash (residue after distillation) filters better, since it contains less dextrin and does not block the filter-presses. DISTILLATION OF THE FERMENTED LIQUID. As has already been stated, the fermentation is rendered the more complete by using worts which are not too con- centrated and yield 9 to 10 per cent, of ethyl alcohol. These fermented liquids contain also RECTIFICATION 133 FIG. 127. small quantities of various other substances, such as aldehydes, organic acids (acetic, propionic, butyric, lactic, succinic, &c.), certain higher alcohols (amyl, propyl, butyl ; glycerol), &c., besides the solid residues of cereals and yeast and small amounts of unfermented dextrin and starch. It was formerly not easy to separate the ethyl alcohol from these products, in spite of the great differences in boiling-point in some cases (amyl alcohol, 132 ; ethyl alcohol, 78-4), and, as already explained on p. 109, this separation cannot be effected with the most exact fractional distillation, so that recourse must be had to rectification (see p. 3). 1 Every distillation apparatus is now composed of four parts : (1) the boiler in which the alcoholic liquid is heated; (2) the rectifier ; (3) the dephlegmator ; and (4) the condenser. The liquid collecting in the dephlegmator returns to the column (hotter), where alcohol vapours are formed richer than those from which it was formed in the first distillation ; so that the alcohol vapours of the dephlegmator, uniting with the other vapours before the condenser is reached, contribute to form a more con- centrated alcohol. Apparatus with continuously working columns and with re- covery of the heat have been studied and applied since 1867 (Savalle). The action of a rectifying column may be understood from Fig. 127, showing part of the column, which is divided into a number of chambers communicating by means of tubes and placed above the boiler. The mixture of alcohol and water vapours from the Jj IBilg^aa boiling fermented wash below ascends the column from chamber to chamber through the central tubes, which are covered with ffl ^ caps dipping below the surface of the liquid in the chambers ; by this arrangement the mixed vapours are obliged to pass through the hot, condensed liquid, which slowly descends the column through the drop-tubes, when it reaches a certain level in each chamber. The vapours give up to the liquid mainly water-vapour, and the liquid gives up to the vapours preferably the alcohol it contains, so that the alcohol -vapour reaches the top of the column mixed with only a little water-vapour and passes to the condenser, whilst water almost free from alcohol flows downwards, forming vinasse or spent wash. With this column, 8 to 10 metres high and containing 20 to 25 plates and chambers, one distillation and partial rectification yields directly a crude 50 to 65 per cent, alcohol, and when this is subjected to a second similar distillation and rectification a concentration of 90 per cent, or even 96 per cent, is attained ; each apparatus gives a high output. This is the procedure often adopted in France. Taller columns (14 to 18 metres) are, how- ever, used, especially in Germany, and these with efficient dephlegmators give 90 per cent, or even 96 per cent, alcohol in one continuous, although slower, operation. The cylindrical columns are advantageously replaced by square ones, which are less easily stopped up and more easily cleaned and repaired ; in place of the costly copper columns, cheaper cast-iron ones are now largely used. A square plate of such a Savalle column is shown diagrammatically in Fig. 128, the apertures and 1 The first forms of distillation apparatus were used in the times of the ancient Arabs, and were termed alembics. The alchemists made improvements in the shape, especially of the part used for condensing. Simple distillation apparatus, like that used for obtaining distilled water (vol. i, p. 225), yield a highly aqueous spirit, termed phlegm. Argand, and later Adam (about 1800), utilised the heat of the aqueous alcoholic vapours distilling over to heat the liquid to be distilled. Solimani and Berard (1805) improved the apparatus so as to allow a distillate moderately rich in alcohol to be obtained in a single operation. Before the condenser was placed a vessel called a dephlegmator, which condensed part of the water-vapour and part of the alcohol (phlegm), more concentrated alcohol vapours passing to the condenser. The first really rational and complete apparatus for the fractional distillation of alcohol was constructed by Cellier-Blumenthal (1815), who used dephlegmators and the first rudi- mentary rectifiers ; but as early as 1813, A. Baglioni had placed semi-rectifying dephlegmators directly above the boiler. , The first column rectifying dephlegmator was devised by Derosne and Cail in 1817, and shortly afterwards widespread use was made of the very convenient Pistorius apparatus, with its flat, lenticular dephlegmators, which allows of 60 to 75 per cent, alcohol being obtained directly, and is still used in some of the smaller distilleries. FIG. 128. 134 ORGANIC CHEMISTRY tubes being sufficiently wide to avoid obstructions when dense fermented worts, rich in solid matters, are distilled. The heating of the column and of the liquid is no longer effected by direct steam, as this causes useless dilution ; indirect steam is employed with FIG. 129. a tubular heater, to be described later. In order to obtain regularity of working and constancy in the alcoholic strength an automatic steam regulator is used (see below), and the supply of fermented wash to the apparatus is so controlled that the yield and strength of the alcohol remain uniform. The heat of condensation of the alcohol vapours is recovered to heat the wash, and the latter, before being introduced into the top of the CONTINUOUS DISTILLATION 135 column, is passed through tubular heaters so as to utilise also the heat of the spent wash before this is discarded. Fig. 129 shows the whole of a Savalle continuous distilling apparatus. The wash to be distilled passes from large constant -level tanks, situate on the upper floors, through the tube m, furnished with a regulating cock, 2, into the bottom of the heater, C, from which it issues at the top, after serving to condense the alcohol vapours coming from the column by the tube k ; these vapours, however, first yield a little condensed spirit in B, this being carried to the column by the tube r. The heated wash passes along the pipe q to the top of the column and slowly descends, meeting meanwhile the ascending vapour current, to which it gradually gives up its alcohol, as stated above (see Fig. 127). The alcohol condensed in the wash -heater is cooled in the condenser, D, below, through which cold water circulates. If the wash is heated in the wash -heater sufficiently to form vapour this passes into the small dephlegmator, H, whence the condensed alcohol and water are led by the tube S r to the column, whilst the alcohol vapour which is not condensed proceeds through t to the condenser along with the other alcohol. When all the plates of the column are covered with wash, steam is passed in from below by heating the exhausted vinasse by pipes from the heater, G, in which superheated steam from suitable boilers circulates ; this steam is regulated by the tap j, which in its turn is controlled by the automatic regulator F. When the distilled alcohol issues from the test-glass, E, the access of wash through 2 is regulated so that the alcoholic strength remains con- stant. In the column the wash traverses a path more FIG. 130. than 125 metres in length, the total absorptive surface being more than 200 metres, so that every litre of wash, before exhaustion, meets a surface of vapour 200 metres long. In this way 30,000 kilos or more of wash can be distilled per day without interruption of the working for months. Fig. 130 shows Savalle 's tubular heater more in detail. Steam under pressure from ordinary boilers traverses the regulator, E, and passes through the tube i to a large metallic cylinder, G, which contains a series of vertical tubes connecting the upper chamber, 0', with the lower one, Q" ; the latter is filled with almost exhausted vinasse supplied from the lower part of the Savalle column by the pipe x. The spent wash, which is already very hot, is thus easily brought into a condition of vigorous ebullition and loses the last traces of alcohol, which rise with a large quantity of steam through the pipe y into the Savalle column. The exhausted spent wash is discharged continuously from the tube 7, whilst the condensed steam issues from the tap 8. Fig. 131 shows the automatic regulator of the pressure and steam in the distillation and rectifying column. In order that it may pass through all the layers of liquid on the plates of the column the steam must be at a certain pressure in the column itself ; this pressure increases or diminishes according as the quantity and temperature of the steam rise or fall, and the greater the supply of steam the more dilute will be the alcohol. If the column is connected with the pressure regulator by means of the tube F(f in Fig. 129), FIG. 131. 136 ORGANIC CHEMISTRY then, when the pressure increases, the water in the lower chamber, A, of the regulator is forced along the tube B to the upper chamber and raises a float, C, which operates the lever D, and so partially closes the tap (or valve) E controlling the supply of steam to the heater, G ; owing to the diminished supply of steam the pressure falls. In the opposite case, when the pressure in the column is smaller than that necessary for regular distilla- tion, so that the concentration of the alcohol (measured in E, Figs. 129 and 132) becomes too high and the yield too small, the water of the upper chamber of the regulator descends to the lower one, the float, C, hence falling and the steam-cock, E, opening a little. With these regulators, which are sensitive to variations of one -thousandth part of an atmo- sphere, the distillation is automatically regulated and requires very little personal control. '" The constancy of the strength of the alcoholic distillate is controlled by the test- glass, E (see Fig. 132), which is situated in the alcohol discharge tube and contains an alcoholometer fitted with a thermometer, so that the concentration and temperature are indicated continuously. Of the variously highly perfected forms of apparatus (Ilges, Coffey, Pampe, the last of. which gives very pure spirit by distillation under reduced pressure) used in England, Germany, Russia, &c., which allow of the continuous and direct pro- duction of 90 to 96 per cent, alcohol without special rectification and refining (when the first and last products of distillation foreshots and tailings are kept separate ; see later), we shall refer only to the apparatus of Siemens Brothers, which is largely used in Germany (Fig. 133). The column is com- posed of three principal parts : the heater (or pre- heater), A, the distillation column, B, and the recti- fier, C ; the whole is formed of superposed cast-iron discs or rings fitted with pasteboard packing and held tightly together by bolts extending from the top to the bottom. Inside are plates arranged FIG. 132. spirally round a central tube, D, which passes about half-way up the column to / ; the liquids thus tra- verse a long path, so that a large production is possible with a relatively small tower-space. The apparatus is also economical since it is not necessary to construct it of copper. The heater, A (see also Fig. 134, A), contains, in the chambers a and o, hot spent wash which comes from the top of the column. Between these hot chambers are arranged alternately others in which circulates the cold wash or wine to be distilled ; this is supplied through the pipe d by means of high -pressure pumps, and begins to be heated as it descends the spiral chambers between the hot ones containing the spent wash. When it reaches the bottom the hot wash passes into the central pipe D, and rises to the higher level,/, in the distillation column, B (which embraces the space between d and E). The pipe D empties on to the perforated spiral plates (see Fig. 134, B) and, as it descends, the wash meets a current of steam rising from the tube o through B. In this way the alcohol liberated from the wash rises with the steam through the perforations of the spiral plates and thus continually meets fresh quantities of wash and becomes continually richer in alcohol, as is shown in Fig. 134, B. The wash, thus deprived of alcohol, reaches the bottom as very hot spent wash, which, before leaving the column, traverses the chambers of the heater (shown in Fig. 134, A) and is then discharged continuously from the pipe J K, at a lower level than /. The mixed alcohol and water vapours enter the rectifying compartment, E, 1 which is formed of plain discs and is filled with wash, the level of which can be seen through suitable glass windows. The alcohol vapours rise into the rectifier, C (more properly termed a fractionator or dephlegmator, see p. 133), formed of non -perforated and hence non-communicating spiral chambers (Fig. 134, C), in some of which circulate the ascending vaporous mixture, whilst the alternate ones are traversed by a descending current of water ; the latter is not very cold, as it comes from the top of the condenser, S (by means of the pipe t), so that it condenses mainly steam and only a little alcohol vapour, 1 Pampe (Ger. Pat. 199,142, 1908) suggests placing, before the rectifying compartment, a steam-turbine with rapidly rotating vanes, which separate all the suspended drops or impurities from the vapours. RECTIFYING COLUMNS 137 which falls into the distilling column again. The alcohol vapours gradually become more and more highly concentrated and pass through the tube F to the refrigerator, S, where they condense and are cooled by water flowing in at s and out at t. By means of a sample taken from the column B by the tube p and examined in the tester, T, it can be ascertained if the spent wash is completely free from alcohol. In some cases it is observed that the spirit from such a cast-iron apparatus absorbs traces of hydro- carbons and of hydrogen sulphide which are formed FIG. 133. ConC.atco* . f nol vapoun JLyueous alcoltol vapour: FIG. 134. from the iron and give an unpleasant taste and smell to the alcohol ; this may, perhaps, depend on the quality of the metal and on the newness of the apparatus. We shall mention finally the attempts which have been made, first by Perrier in 1875, to transform the vertical column into a horizontal distilling and rectifying column with a central rotating axis carrying helically arranged blades, which transport even a very dense wash from one end to the other, whilst the opposing current of steam removes the whole of the alcohol. The process was perfected by Sorel and Savalle (1891), who arranged the numerous vertical chambers of the horizontal column in a more rational manner. These forms are not yet free from disadvantages, but they have the advantage of being 138 ORGANIC CHEMISTRY considerably more economical to construct and of bringing all the taps conveniently to hand on the ame level. Lastly, Guillaume eliminated various defects of these columns and at the same time retained all their advantages by employing very simple and convenient inclined columns (made by Egrot, of Paris), which allow of very dense washes being employed without danger of obstruction. Fig. 135 shows the complete Guillaume-Egrot apparatus, and the description of the various parts given underneath will indicate the way in which it works. The cross-section shown in Fig. 136 gives an idea of the internal arrangement of the inclined column, and Fig. 137 represents the ground plan of the column, the arrows indicating the horizontal, zigzag course followed by the liquid from the highest part of the column, whilst the vapours ascend the column in a zigzag vertical path and bubble through the liquid in all the chambers formed by the numerous vertical partitions. With relatively small plant, which can be mounted on portable cars (see later), 30,000 litres or FIG. 135. A, distilling column ; a, entrance of the wash into the heater or refrigerator ; B, condenser and heater; b, hot wash pipe; C, adjustable steam regulator; c, exit for spent wash;.D, hot wash extractor used as heater ; d, steam-tap ; E, test-glass giving the strength of the alcohol ; e, valve regulating flow and hence strength of the alcohol ; h, entrance of water into refrigerators in case of need. more of wash, containing 10 per cent, of alcohol, can be distilled per 24 hours, 90 per cent, alcohol being produced. In the modern distillery the consumption of steam should not exceed 25 kilos (about 3 kilos of coal) per 100 kilos of wash, and the consumption of water in the condenser should not exceed 80 litres. RECTIFICATION OF ALCOHOL. The alcohol obtained with the ordinary Savalle apparatus is not sufficiently concentrated or pure to be placed on the market, and even that obtained with other forms from washes which have nqt been fermented with selected yeasts should be freed by rectification and refining from various impurities which impair the colour, smell, and taste. These impurities may be more volatile than alcohol (i-_uch as aldehydes and certain esters) or less volatile (as acetic and butyric acids ; propyl, isopropyl, and amyl alcohols ; various esters, &c.), and they are separated from the true alcohol if, in the redistillation and rectification, the portions which distil most readily (foreshots) and also the least volatile portions (tailings or fusel oil, which has a very PROCESS OF RECTIFICATION 139 disagreeable odour if obtained from potatoes, molasses, or maize, but a pleasing odour if derived from grapes, fruit, &c. ) are kept apart. Beatification apparatus usually consists of a large copper or iron boiler, A (Fig. 138), which is heated with an indirect steam-coil and on which is mounted the copper rectifying column, B. Above this and to one side is a large dephlegmator, G, which serves as a heater, and is of importance not so much for condensing the less volatile products (water, amyl alcohol, &c.) as for furnishing a continuous and abundant supply of a suitable alcoholic liquid to wash the vapours arriving at the top of the column ; it is, however, quite useless to employ several dephlegmators, as was erro- neously done in the past. The foreshots, which have a con- centration up to 94 per cent, and boil at 85, are collected separately. Then from 85 to 102 alcohol passes over. The tailings, boiling above 102, are collected in the bottom of the column by shutting off the PIG. 136. steam and thus emptying the plates. The quantities of these products vary according to the quality of the alcohol required ; thus 20 per cent, of foreshots and tailings may be obtained and 80 per cent, of alcohol (bon gout extra), or 5 per cent, of foreshots and tailings and 95 per cent, of alcohol (bon gout). This apparatus does not work continuously, the boiler requiring to be discharged and recharged. Attempts y*T>tv j^ 9 to render the process con- tinuous were met with suc- cess in 1881 (E. Barbet) in spite of the difficulty of separating the pure alcohol from an impure product that boils below it and another that boils above it. This is effected by carrying out the operation in two phases, which are, however, continuous ; in the first phase the foreshots are driven off and the alcohol distilled from the remaining liquid, the tailings being left behind. The boiler is then replaced by a rectifying column, which receives the impure product and distils the foreshots, passing the residue continuously at a certain height to a second lower column at the side ; this distils and rectifies the pure alcohol and retains in the lowest chamber of the column the tailings, which are continuously discharged. In the Savalle rectifiers 45 kilos of coal are consumed per hectolitre of pure rectified alcohol. Continuous rectification results in a saving of almost 50 per cent, of fuel compared with the discontinuous process. During rectification the loss of alcohol is 1 to 2 per cent., and the cost of rectification varies from 3 to 3-5 lire (2*. 6d. to 3*.) per hectolitre. The firm of Savalle holds that it is more economical to use cold air than water in the refrigerators of the condensers, FIG. 137. 140 ORGANIC CHEMISTRY Attention may lastly be drawn to the ingenious although complicated Perrier distilling and rectifying apparatus, in which the vapours of alcohol, water, higher alcohols, and aldehydes are pacsed successively into columns filled with glass beads and surrounded by a jacket containing a liquid boiling at a constant temperature, the latter bsing hence assumed by the whole of the tower. In one of these, having a tempera- ture of 85 to 90, only water and the tailings are condensed ; the vapours then pass into a second tower, kept at 75, where all the ethyl alcohol (which can be recti- fied in another tower) separates ; the vapours from this form the foreshots and are condensed in a succeeding tower. OTHER PRIME MATERIALS FOR THE MANUFACTURE OF ALCOHOL. (1) Beetroot and Mo- lasses. It is especially in France that considerable quantities of beet are used for the manufacture of alcohol instead of sugar ; this is never done in Germany or Italy. The beets are washed, minced, and the pulp exhausted by pressure, maceration, or diffusion with water. This treatment is described in the section on sugar. The spirit obtained from the beet is less pure than that from potatoes, containing more propyl and butyl alcohols but less amyl alcohol. Of more importance in Italy and various other countries is the utilisation of beet-molasses. 1 The complete fermentation of molasses has presented many diffi- culties, which have now been over- come. Formerly, after the molasses was diluted to 8 to 10 Be. (this was carried out in vats provided with stirrers, see Fig. 139), it was slightly acidified with sulphuric acid (2-5 grms. of free H 2 SO 4 per litre), as the reaction is usually alkaline. The liquid was then boiled for some 1 These are the dense, viscous, and FlG. 138. blackish mother-liquors which remain from the final crystallisation of the sugar (which e) and from which no further sugar will crystallise although 45 to 50 per cent, are present (see explana- tion in the section on Sugar) ; it has a density of 40 to 45 Be. (74 to 84 Balling). The composition of beet- is as follows : water, 16 to 20 per cent. ; sugar, 44 to 52 per cent. ; non-nitrogenous extractive matters, 10 to 15 per cent, (largely pentoses) ; nitrogenous compounds, 6-5 to 9-5 per cent, (of which only one-third consists [ proteins, the rest being amino-acids) ; ash (deducting CO 2 ), 8-5 to 11 per cent. In Italy the working-up of nolasses has assumed considerable importance during the last few y^ars, owing to a change in the method of taxing sugar ; previous to 1903, sugar recovered from molasses by somewhat expensive processes (see Sugar) was exempt from taxation, whilst nowadays all sugar produced is taxed uniformly, so that the manufacturers nnd it advantageous to sell the molasses to the distillery at 4s. 9d. to 6s. 5d. per quintal. In Germany, Belgium, and part of France, it is found to be more convenient and rational to utilise a large proportion of the molasses as cattle-food after absorbing it by highly porous vegetable substances. In Italy, tumelina, patented by E. Molinari, and sanguemelassa (blood-molasses), patented by L. Fino, are manufactured ; the residues of dried tomatoes (Squassi, Bono) and various other dried industrial products are now used as absor- bents. In Germany more than 1,500,000 quintals of molassic fodder are consumed ; Italy produced 400,000 quintals in 1908 and more than 480,000 in 1909 ALCOHOL FROM MOLASSES, FRUIT, ETC. 141 hours in a current of air in order to eliminate the volatile acids (nitric, &c.) liberated, and, after cooling it to 15, alcoholic fermentation was initiated by the addition of vigorously fermenting liquid ; the excess of acid which forms is gradually neutralised with chalk. The spirit thus obtained is difficult to purify as it contains an aldehyde and various acids which boil at a very low temperature. To-day, however, the process is much more simple, as Jacquemin and Effront have devised various methods of preparing races of yeast capable of living actively in worts rich in salts (nitrates, carbonates, &c.), such as those prepared from beet-molasses. In the past the difficulty of fermentation was attributed to the presence of nitrates, but it appears from Fernbach and Langenberg's experiments (1910) that nitrates, even in proportions as great as 0-3 per cent., facilitate fermentation. (a) In the Jacquemin process the fermentation is initiated in small quantities of wort in suitable vessels (see Fig. 123, p. 125), and the wort of the last rather larger vessel (into which is also placed a little hydrofluoric acid, to which the yeast has been previously " acclimatised ") serves to pitch a 200-hectol. vat containing diluted, non-sterilised molasses, to which has been added 8 to 10 kilos of calcium hypochlorite, this preventing the de- velopment of heterogeneous organisms during the first few hours without damaging the yeast already adapted to chlorine. By means of this vat two other 500-hectol. vats of similar diluted molasses can be brought into a state of vigorous fermentation ; the fermentation takes place so rapidly (and this is the most specific action of these yeasts) that in three days the Avhole of the molasses is fermented, there being thus no time for the development of extraneous germs causing harmful secondary fermentations. (b) The Effront process is still more simple, and is based on the use of selected yeasts specially adapted to molasses worts and endowed with exceptionally rapid fermenting properties ; these yeasts are placed under such conditions that they easily overcome deleterious bacteria (namely, the addition of resin) 1 and complete the fermentation before these become harmful. To the molasses simply diluted with water and not sterilised are added these special yeasts together with 1 kilo of colophony per 10 hectols. of wort ; in three days the fermentation is complete. In 1903 almost 1,000,000 kilos of colophony were used in France for this purpose. (2) Alcohol from Fruit. This is not of great industrial importance, although in certain districts and in certain years it assumes considerable magnitude. In Italy, dried figs of little commercial value, carobs, &c., are used ; and, in other countries, plums, apples, pears, &c. These fruits often give an irregular, and seldom a complete, fermentation, owing to conditions similar to those encountered with beet-molasses. Hence, as in the latter case, use is made of very active yeasts adapted, where possible, to these special worts. The alcohol obtained from these worts has a characteristic odour indicating its origin. * Effront observed that the law of the strongest, which is often verified in bacteriology the most numerous and powerful bacteria rendering life impossible to weaker ones scarcely ever holds in the case of .alcoholic fer- mentation, where, even though the harmful bacteria are less numerous than the yeasts, the latter are seldom victorious, the bacteria often entirely arresting alcoholic fermentation even when the conditions are favourable for the latter. According to Effront, this is owing to the different specific gravities possessed by yeasts and bacteria, which hence live in different, relatively distant strata, so that there is no opportunity for the application of the law of the strongest -which consists in the production by certain micro-organisms of poisonous substances preventing other forms from developing. Effront hence proposes to add suitably emulsified resin (colophony) to the worts at the beginning of Uie fermentation ; this has the property of coagulating only the bacteria, which become denser and are brought into more intimate contact with the yeast, the latter then being in the most favourable condition for the annihilation of the bacteria. The resin itself is not the cause of the death of the bacteria, as Effront states that these can be readily cultivated in the pure state in presence o"f resin (private communication). FIG. 139. 142 ORGANIC CHEMISTRY (3) Alcohol from Woody Substances. This is a subject which has aroused con- siderable interest during about the last twenty years. Many attempts have been made to transform a part of the wood (sawdust, peat, &c.) into fermentable sugar by the action of acids on the matter (lignin) encrusting the 'wood and not on the cellulose. In Chicago the process was applied on a vast industrial scale according to A. Classen's patents (Ger. Pats. 130,980, 1899, and 161,644, 1904). 100 kilos of wood (with 25 per cent, of moisture) are treated in an autoclave for an hour with about 100 kilos of aqueous sulphur dioxide and sulphuric acid in presence of steam at 6 to 7 atmos. pressure (150 to 165). The excess of sulphur dioxide is eliminated by means of a current of air, the residue being boiled with water or extracted in diffusers, and the liquid neutralised with calcium carbonate and fermented ; x about 8 litres of pure alcohol are thus obtainable, and the residues are partially utilisable for making paper. It is not improbable that in the near future wood and the more economical wood refuse will replace cereals and potatoes in spirit factories. 2 In France, England, and the United States there were in 1910 four factories making alcohol from wood and obtaining yields of 7 per cent. (4) Alcohol from Wine, Lees, Vinasse, and Withered Grapes. In seasons when 1 Wood thus yields a product contaiuing 35-36 per cent, of solid residue, 34-63 per cent, of water, 10-97 per cent, of fermentable reducing sugar, 3-21 per cent, of non-fermentable reducing sugars (pentoses : xylose, &c.) : 0-35 per cent, of sulphuric acid, and 0-77 per cent, of other acids. As early as 1820, Braconnot observed that sugar is formed when wood or even cotton cloth is treated with sulphuric acid. Later on Melsens obtained a good yield by treating cellulose with dilute sulphuric acid in an autoclave under pressure. In 1860 Pettenkofer investigated this process and showed that it could, at that time, compete with the use of potatoes. Still later, Basset prophesied a yield of 32 per cent, of alcohol from the similar treatment of wood (I). Simonson, in 1889, treated wood under pressure with dilute sulphuric acid, transforming 25 per cent, of it into sugar (78 per cent, of which was fermentable) and obtaining a practical yield of 6 to 7 litres of pure alcohol (Third International Congress of Applied Chemistry, Berlin, 1903). Ileiferscheidt (1905) overcame the resistance of the wood to penetration by liquid acid (met with also by Classen) by causing sawdust to absorb two-thirds of its weight of sulphuric acid (sp. gr. 1-65) and subjecting the mass to the maximum pressure of a hydraulic press ; simple digestion of the mass with water and nitration gave a fermentable liquid and a yield of 6-5 per cent, of alcohol on the weight of wood (pine, containing 53 per cent, of cellulose) taken. A similar yield is obtained by treating the wood with five times its weight of 1 per cent, sulphuric acid solution at a pressure of 8 atmos. for fifteen minutes. He confirmed the observation that the pentosans of the wood do not ferment, and with pure cotton he obtained as much as 13 per cent, of alcohol. According to Xh. Korner, the addition of oxidising agents or of ozone, as was suggested by Both and Gentzen (1905), is of no advantage. He obtained the best yields by heating sawdust, straw, &c., with 0-5 per cent, sul- phuric acid for 2 hours in an autoclave at 6 to 8 atmos. ; only a small part of the molecular complex of the cellulose is converted into fermentable sugar, and he obtained a yield of alcohol equal to 15 to 18 per cent, of the weight of the true cellulose in the wood. Without the addition of sulphuric acid, the yield was about one-fourth less. P. Ewen and H. Tomlinson, of Chicago (U.S. Pat, 938,308, 1909) treated 400 kilos of sawdust, straw or stems of various cereals (with 30 per cent, of moisture) in autoclaves with 5 kilos of sulphuric acid of 60 Be 1 , diluted with 20 litres of water ; after complete digestion and agitation the temperature of the mass is brought in fifteen minutes to 135 to 160 by means of steam under pressure ; after half an hour the temperature is lowered rapidly to 100 by allowing the steam to escape, and the sulphuric acid then separated in the usual way. By this means 20 to 30 per cent, of the weight of the cellulose is transformed into fermentable sugar. A similar process is that of Eckstrom (Norw. Pat. 17,634, 1907). Classen's process, which has been tried on a large scale in North America, has exhibited various disadvantages : the time required for treating 2 tons of wood was as much as six hours, the consumption of sulphuric acid was large, part of the sugar was destroyed, and frequent repairs were necessary. The process was improved by Ewen and Tomlinson, and was worked in a factory near Chicago. Less acid was used and the treatment main tained only for forty minutes, the autoclave being rotatable and made of steel protected outside with fireclay This was filled with sawdust, sulphur dioxide (1 part per 100 of dry wood) being then passed in, and subsequently steam at 7 atmos. After forty minutes, the vapours of water, acetic acid, terpenes, and sulphur dioxide are passed into washing or absorption vessels, while the residual darkened sawdust is extracted with hot water : the aqueous extract is neutralised with chalk, filtered, fermented, and distilled. Rectification yields 94 per cent, alcohol free from methyl and higher alcohols, and containing only traces of furfural and other aldehydes, The cost of this alcohol seems to be less than three-halfpence per litre of 90 per cent, concentration. J. Ville and W. Mestrezat (1910) state that, whilst cellulose resists dilute solutions (up to 30 per cent.) of hydrofluoric acid, with 50 per cent, solutions, 100 grms. of cellulose yield 50 grms. of glucose 1 According to the Swedish patents of J. H. Vallin and of Eckstrom, alcohol is obtained by treating the waste sulphite liquors of paper-mills in the hot with sulphuric acid and fermenting the liquid containing the glucose formed. The hot acid liquid has to be neutralised almost completely with chalk and decanted, the residue being then pressed in a filter-press ; the liquid is then cooled on piles to 30, pitched with yeast, aerated during fermentation (5 to 6 hours) and the dilute alcoholic liquid (0-7 to 0-8 per cent, alcohol) distilled. From 10 cu. metres of the sulphite liquors are obtained 60 litres of 100 per cent, alcohol (which is, however, of bad flavour and is used for denaturation). For a factory producing 60 tons of cellulose per day, i.e. 600 tons of waste sulphite liquors, the cost of tanks, pumps, piles, distilling apparatus, filter-presses, &c., may be taken as about 6000, and the alcohol produced (36 hectols. per day) would cost (including all expenses, but excluding taxation) 10s. to 11. per hectolitre at 100 per cent, strength. The problem of the disposal of the waste liquors (which con- taminate the rivers) of paper-mills is not, however, solved in this way, since the liquid still contains much decom- posable organic matter after the distillation of the alcohol. Before starting such an industry, it is also necessary to consider the condition of the market, so that there may not be an over-production of alcohol and hence depression of prices. In 1910 there were two factories in Sweden for the manufacture of alcohol from these waste sulphite liquors : that at Billingfors prepared methyl alcohol (15 kilos per ton of wood pulp) by H. Bergstrom and H. Fahl's process ; the other at Skutskiir manufactured ethyl alcohol. For every ton of cellulose there are obtained 8 to 9 tons of ALCOHOL FROM WINE RESIDUES 143 wine is abundant and prices low and in general when there are spoilt wines (at 6s. to 8s. per hectolitre), it is convenient to extract the alcohol from them, this being of use in the preparation of liqueurs and spirits. The distillation presents no difficulty and is carried out either in the large distilleries or with a Guillaume-Egrot apparatus (see p. 138), which is mounted on a car so as to be readily transportable, and can be used in places where there is little available water, since the coolers and condensers act as heaters and are fed with the wine to be distilled. It gives directly 90 to 94 per cent, alcohol. In the same way as wine, fresh lees or bottoms from wine vats (containing 4 to 6 per cent, of alcohol) and dried grapes l are treated. The distillation of vinasse, containing 2-25 to 3-5 per cent, of alcohol, is of considerable importance in Italy ; if this were all distilled it would yield about 250,000 hectols. of pure alcohol annually (for a production of 40 million hectols. of wine). Of the various forms of apparatus for the distillation of vinasse only those of Villard-Rottner and of Egrot will be described, as they are the commonest and differ little from other good types. The generator, K (Fig. 140), of the Villard-Rottner apparatus sends steam from the dome, M, into the three boilers, A, in succession, the steam entering at the bottom and FIG. 140. FIG. 141. issuing at the top of each. These three boilers contain the vinasse mixed with an equal volume of water. The vapours, which are rich in alcohol, pass through the pipe, E, to the dephlegmator, G, and are then condensed in the coil, 7, at a concentration little exceeding 50 per cent. When the first boiler is exhausted it is emptied and again charged, the steam passing meanwhile through the second and third ; the first boiler now becomes the third, the second being then emptied, so that two boilers are always in use. The hot water from the boilers is treated separately for the extraction of tartar (see this). In the Egrot apparatus (Fig. 141) the boilers, A, are arranged on pivots, so that they sulphite liquors containing, either dissolved or suspended, as much as 12 per cent, of organic substances and yielding alcohol at loss than \\d. per litre. Considerable interest was aroused in 1901 by the English patent of Dornig and Pratorius, according to which human faces yielded about 9 per cent, of alcohol, but it proved to be a fraud. There has been much discussion recently (1906-1907) concerning a process for extracting spirit from peat in a manner similar to that described for wood. These attempts date from 1870, and various patents were filed in 1882-1891. The most important tests were made in Norway in 1906 by the Reynaud process (1903), in which 300 kilos of peat were treated in the hot with 700 kilos of water containing 7 kilos of sulphuric acid (66 Be.) under 3 atmos. pressure ; 600 litres of liquid were thus obtained and this was fermented with specially selected yeasts (Saccharomyces ellipsoideus), the yield being 25 litres of burning spirit at an inclusive cost of about 4-5<7. per litre, which is about double the cost of that obtained from ordinary starchy materials. In 1905, the Danif b Government offered a prize for the improvement of this process, but the yield was not increased although it varies somewhat (6 to 8 per cent.) with the quality of the peat ; in all cases the alcohol obtained in this way is too costly. 1 In some countries at certain times in Italy dried grapes are used for the production of alcohol, especially Greek grapes, which are received from viticulturists by the Greek Government in payment of taxes, and are dried and placed on the European markets. These grapes are first macerated in tepid water, then crushed and fermented in the usual way ; the wine obtained may be used for mixing with other wines or for distillation. In 1905-1907, in order to help the crisis in the South, the Italian Government granted a considerable rebatement of taxation on the alcohol obtained from grapes The Italian distillers then began to import large quantities -of Greek grapes (containing 50 to 55 per cent, of sugar), which could be delivered in the factory at about 13*. per ' quintal, so that the southern viticulturists reaped no advantage from the rebate, which was hence abolished. 144 ORGANIC CHEMISTRY can be inverted and rapidly emptied. Steam from the boiler, D, extracts the alcohol from the three boilers, which are arranged in series, as before, so that two are always in use while the third is being emptied and recharged. The alcohol vapours pass into the dephlegmator, B, and thence into the spherical rectifier, C ; R acts as a condenser and is cooled by water from the tank, K. The condensed alcohol passes along the tube, m, to the test-glass, M, and from there to the casks, t, at a concentration of 55 to 60 per cent. With the first apparatus, to treat 100 quintals of vinasse, yielding about 8 hectols. of brandy at 51 per cent., roughly 13 quintals of coal are consumed, whilst the Egrot apparatus uses much less than this for an equal yield. The brandy thus obtained has almost always a rather unpleasant flavour and is often used for rectification in the ordinary way (if too dilute it becomes opalescent) and is then left to age in oak casks so as to acquire a pleasing aroma. This result is obtained more rapidly by pasteurisation, that is, by passing the brandy through a coil surrounded by water at 60 to 65, or by passing a current of ozonised air through it (artificial maturation). The name cognac is given to the finest old French brandies. Alcohol from cereals can be distinguished from that obtained from wine, &c., as the latter always contains aldehydes (see later, Rimini's Reaction and Schiff's Reagent). REFINING AND PURIFICATION OF SPIRIT. After the introduction of rational methods of fermentation with selected yeasts and of more perfect rectifying appliances, the quantity of actual alcohol was considerably increased and it was generally sufficiently pure for ordinary commercial purposes. But when it became recognised that the harmful effects of alcoholism are aggravated by the presence in commercial alcohols for liquors, &c., of even minimal quantities of aldehydes and amyl alcohol, recourse was sometimes had to a special purification or refining of rectified spirits in order to give them a slight ethereal odour, which is greatly valued. Of the many and varied substances suggested for the purification, mention need only be made of charcoal in lumps calcined and cooled out of contact with air and placed in batteries of tall cylinders through which the alcohol is passed ; when the charcoal becomes inactive it is revivified by means of superheated steam at 600. The charcoal has an oxidising, esterifying, and decolorising action, but it does not fix the amyl alcohol. Treatment with fatty oils (which retain the aldehydes) and subsequent distillation are also used, as also are carbonates of the alkalis and alkaline earths. Treatment with oxidising agents ozonised air, potassium permanganate or dichromate, nitric acid, chloride of lime, &c. has the disadvantage of forming acetic acid and ethyl acetate. Consequently Naudin prefers reducing the aldehydes with nascent hydrogen formed in the liquid itself by means of a copper-zinc couple. R. Pictet has devised a totally different process : owing to the variations (at different temperatures) of the maximum vapour pressure of volatile liquids, he ascertained that the vapours obtained from a mixture of water or other substances with alcohol are the richer in alcohol the lower the temperature to which the mixture is heated. He boils the mixture at 50 to 60 in a vacuum and then rectifies the vapours in a column at a tempera- ture of 30 or 40, obtained by means of a sulphur dioxide refrigerating machine. The apparatus is somewhat complex, but it yields a well-refined pure spirit. TESTS FOR THE PURITY OF ALCOHOL. The tests mentioned on p. 109 will detect traces of water in so-called absolute alcohol. If alcohol is highly purified (puriss.), 10 c.c. of it, mixed with 1 c.c. of water and 1 c.c. of 0-1 per cent, potassium permanganate solution, should maintain its red colour for 20 minutes, or for at least five minutes if the alcohol is termed pure ; it should not become turbid on dilution with water, should give neither an acid nor an alkaline reaction (with phenolphthalein), and should remain unchanged with ammoniacal silver nitrate solution. To test for aldehydes the alcohol is diluted with water and a few drops distilled and tested by Rimini's reaction (see p. 109) ; or, for aldehydes in general, by Schiff's reagent(fuchsine solution decolorised with sulphur dioxide : 0-5 grm. of fuchsine is dissolved in 500 c.c. of water and decolorised with 10 c.c. of sodium hydrogen sulphite solution of sp. gr. 1-26 and 10 c.c. of concentrated HC1) ; a few c.c. of this reagent are coloured red when shaken with a few drops of alcohol containing traces of aldehydes. Of more importance is the quantitative estimation of the fusel oil 1 (always formed in 1 Fusel oil has a varying composition : 14-24 per cent, of water, 15-45 per cent, of ethyl alcohol, 6-14 per cent, of normal propyl alcohol, 10-25 per cent, of isobutyl alcohol, and 10-40 per cent, of amyl alcohol of fer- mentation. Traces of fusel oil may be detected by Kamarowsky's reaction, i.e. with salicylic aldehyde and ESTIMATION OF FUSEL OIL 145 alcoholic fermentation), which is made with Herzfeld and Windisch's modification of Rose's apparatus (Fig. 142) ; the method is based on the property possessed by chloro- form of dissolving the higher alcohols and a very little ethyl alcohol, at the same time increasing in volume. The alcohol is first diluted to a concentration of 30 per cent, by volume or, better, to the sp. gr. 0-9656 at 15-5 (see Table, p. 148 ; if the alcohol has a concentration, v, less than 30 per cent., then 10 (30 v) 1 c.c. of absolute alcohol should be added). The Rose tube (washed with alkali, acid, water, alcohol, and ether and well dried) has a cylindrical expansion at the bottom containing 20 c.c. up to the first mark ; then comes a tube 18 cm. long, holding 2-5 c.c. and graduated in O'Ol c.c. ; at the top is a pear-shaped bulb of about 200 c.c. capacity, closed with a ground stopper. The tube is placed in water at 15 and into it are introduced by a long funnel reaching to the lower bulb 20 c.c. of pure chloroform at 15, and then 100 c.c. of the alcohol diluted to 30 per cent, at 15 and 1 c.c. of sulphuric acid of sp. gr. 1-2857 (38 per cent. H 2 SO 4 ). The tube is then closed, inverted so that all the liquid passes into the pear-shaped bulb, shaken vigorously for a minute (150 shakes) and placed erect in the water-bath at 15, where it is left for 15 minutes after a rotatory movement has been imparted to the liquid so as to collect the drops of chloroform adhering to the walls. The increased volume of the chloroform is then compared with that obtained in a similar test with pure alcohol of the same concentration. If no blank experiment is made, 1-64 c.c. is subtracted from the increase in volume as being due to the ethyl alcohol dissolved. Each 0-01 c.c. increase hi volume of the chloro- form corresponds with 0-006634 per cent, by volume of fusel oil. For an alcohol rich in fusel oil which gave a final volume of chloroform of 22-14 c.c. the true increase in volume will be 22-14 1-64-20=0-5 c.c. The percentage, /, of fusel oil by volume on the original alcohol (not on that diluted to 30 per cent.) is calculated by the following formula : _ (c-6)(100 + a) * = 150 ' where c is the uncorrected increase in volume of the chloroform, 6 is the correction, 1-64, due to the ethyl alcohol, and a indicates the number of c.c. of water or absolute alcohol added to 100 c.c. of the original spirit to bring it to 30 per cent. Example : If 80 per cent, alcohol is used, 171-05 c.c. of water must be added to 100 c.c. to break it down to 30 per cent. ; 100 c.c. of this then increases the volume of the chloroform from 20 to 21-94 c.c., so that: / = (1-94 -1-64) (100 + 171-05) 150 FIG. 142. = 0-54 per cent, by volume of fusel oil. The furfural is determined in 10 c.c. of distilled alcohol, to which are added 10 drops of colourless aniline and 2 c.c. of acetic acid ; if a red coloration appears after 20 to 30 minutes furfural is present. 1 sulphuric acid; H. Kreis's modification (1907) of this colorimetric reaction yields moderately accurate results. Fusel oil is now largely used for the preparation of amyl alcohol, which is used in the manufacture of fruit essences, for obtaining nitrous and other ethers, and for gelatinising explosives (nitrocellulose) ; during the last five years the price of fusel oil has risen from 65 to 170, and even 195 lire per quintal. Pasteur thought that the amyl alcohol (iso- and d-amyl) arose from the action of specific bacteria on the sugar. But in recent years F. Ehrlich has thrown doubt on the formation of an alcohol with a branched chain from a sugar with a direct chain, and has now shown that it is the proteins of the malt and their decomposition products which furnish nitrogen to the yeast for the synthesis of its protein constituents and at the same time form amyl alcohol. In fact, in the fermentation of a pure sugar, Ehrlich obtained a quantity of fusel oil proportional to the quantity of leucine added ; he was also able to obtain an amount of fusel oil equal to 7 per cent, of the alcohol formed (the usual amount being 0-4 to 0-6 per cent.) and, further, he succeeded in reducing the formation of fusel oil considerably by the addition of ammonium salts. The United States imported 2350 tons (122,000) of fusel oil in 1910 and 2900 tons (255,000) in 1911. 1 The estimation of small quantities of benzene in denaturated alcohol can be carried out by means of Rose's apparatus (for more than 1 per cent, of benzene). The best method is to dilute 100 c.c. of the alcohol to a con- centration of 24-7 per cent, by weight and to distil the whole ; the first 10 c.c. of the well-cooled distillate are diluted to 20 to 25 c.c* with water in a graduated cylinder ; the volume of the benzene which separates is increased by 0-3 per cent., which is a constant error of the method. This method of Holde and Winterfeld (1908) is based on the fact that when the alcohol is diluted with water, the pressure of the benzene is considerably augmented, whilst that of the alcohol is diminished. To ascertain if methyl alt-olio as present in alcohol, 1 c.c. of it is treated with 1 c.c. of chromic acid solution II 10 146 ORGANIC CHEMISTRY FIG. 143. ALCOHOL METERS OR MEASURERS. These are important instruments, as in nearly all countries the manufacture of alcohol is subject to taxation which is calculated on the quantity of alcohol passing through a sealed meter indicating automatically the corresponding amount of pure alcohol (100 per cent.). The Siemens measurer is the one most used (Figs. 143 and 144) and somewhat resembles the gas-meter (see p. 50) even in its registration. The alcohol, which enters laterally by the tube I, is discharged into the inner central part of the drum, B, i.e. into D, this being divided longitudinally into three small chambers furnished with apertures, r 1 , r 2 , r 3 ; when the small chamber is about half full the alcohol falls into the large lower chamber (e.g. I), which has a capacity of 4 Hires. When this chamber is filled with alcohol the level of the latter reaches the chamber D, the alcohol then falling through r 2 into // and displacing the equilibrium, so that the drum, B, is forced round in the sense of the arrow. At the same time the first 4 litres of alcohol are discharged into the vessel C, which communicates with the storage reservoir by means of the tube G. The compartment II then occupies the position of 7, and so on. The axis of the drum is con- nected with a suitable automatic regis- tering device. At the same time, in the cylinder A in front of the drum, the alcohol which passes through raises the float, P, more or less according to its strength, and a screw, Q, operates the lever, T, and so moves the index, 8, the point of which registers the alcoholic strength on a paper ribbon moving along a carefully calcu- lated curve, X. In order that alcohols of different concentrations may be well mixed and so influence the float correctly, they are delivered at E, where there are two tubes ; one of these, a, collecting the lighter alcohol, rises and then descends (c), dis- charging into the bottom of A by the per- forated tube, e ; the denser alcohol passes preferably along 6 and is discharged through the perforated tube, d, at the top of A, so that mixing is rapid and com- plete. The registration is also independent of the temperature of the alcohol, as its expansion (or contraction) is allowed for by that of the float. ALCOHOLOMETRY AND TESTS FOR ALCOHOL. As a rule alcohol is sold practically by volume and not by weight ; 1 litre of absolute alcohol weighs 0-7937 kilo or 1 kilo measures 1-2694 litre. Industrially alcohol is stated to be of so many litre-degrees ; thus 100 litres of 2 per cent, alcohol would contain 200 litre-degrees (100 x 2), and 100 litres of 50 per cent, alcohol would indicate 5000 litre -degrees, which would also be given by 1000 litres of 5 per cent, alcohol ; so also 75-48 litres of 100 per cent, alcohol would be expressed as 7548 litre-degrees. Alcohol is taxed on the basis of the number of litres of absolute alcohol. The alcohol-content of an aqueous alcoholic solution is deduced from the specific gravity determined by the Westphal balance, or directly by the Gay-Lussac alcoholometer (at 15) in France, or by the Tralles official alcoholometer (at 15-56) in Italy and Germany, these giving the percentage of alcohol by volume contained in 100 vols. of the aqueous and 5 c.c. of water, the whole being then carefully distilled until only 0-5 c.c. remains. The distillate is condensed in a long air-cooled tube and collected in a test-tube, the condenser-tube being washed out with 2 c.c. of distilled water. One drop of ferric chloride and two of albumin solution are added to the test-tube, which is shaken ; 5 c.c. of concentrated sulphuric acid are then cautiously added. The immediate appearance of a violet ring at the zone separating the two layers indicates that the original alcohol contained more than 5 per cent, of methyl alcohol ; if the coloration appears after a minute, the proportion is 1 to 5 per cent, and if after two minutes less than ]per cent.(A. Vorisek, 1909). FIG. 144. ALCOHOLOMETRY 147 FIG. 145. alcohol. The reading on the alcoholometer is made at the point of the stem coincident with the lower meniscus, which is well seen by looking rather below the surface of the liquid (Fig. 145) ; to avoid error, the alcoholometer must be so im- mersed that the whole of the graduated stem is not wetted (see vol. i, p. 75). To determine the percentage by weight contained in 100 vols. the percentage by volume is multiplied by 0-7937 (specific gravity of absolute alcohol) and divided by the specific gravity of the alcohol examined (see Table on p. 148). To correct the alcohol reading determined at a temperature different from 15 (or 15-56 for the Gay-Lussac alcoholometer), the following moderately exact formula of Pranco3ur is used : x = c 0-392, where x is the number of Gay-Lussac degrees at 15, c the number of degrees found at the non-normal tempera- ture, and t the number of degrees the latter is above or below 15 ; the + sign of the formula is used if the temperature is below 15 and the sign if it is above 15. Thus an alcohol showing 72 on the Gay-Lussac alcoholometer at a temperature of 28 would have : x = 72 - 0-39 x 13 = 66-93 Gay-Lussac at 15. With dilute alcoholic liquids of complex composi- tion (wine, beer, spirits, &c.) the alcoholic degrees cannot be deduced from their specific gravities. But, if a given volume, e.g. 100 c.c., is taken and distilled (Fig. 147) until all the alcohol has passed over (about 70 c.c.), the distillate can be made up to the original volume with distilled water and its specific gravity and alcoholic strength determined in the usual manner. In order to prevent frothing during the distillation of beer and wine a piece of tannin or a few drops of oil are added. In some cases the alcohol of wines and other liquors is determined by the Geissler vapoiimeter, which indicates the pressure of the vapours from the liquid heated at 100, By means of a Table the alcoholic strength may be read off, knowing the vapour pressure ; the latter is measured on a special barometric U-tube, B (Fig. 146), to one end of which is fixed the bottle, O, containing mercury and the alcoholic liquid and placed in the jacketed vessel, D, filled with steam from the boiler, A. This apparatus gives results which are influenced by several factors (dis- solved carbon dioxide, salts, &c.), so that little use is made of it. In more general use is the ebullioscope devised in 1823 by Groningen and subsequently improved by Tabarie (1833), Brossard-Vidal (1842), Malligand (1874), Salleron (1880), and Amagat (1885). Malligand's form (Fig. 148) is the most commonly used and is based on the different boiling-points possessed by alcoholic liquids of different concentrations. The reservoir, F, is provided with a cover, through which pass a thermometer, T, bent at a right-angle and a tube surrounded by the condenser, R. This cover is unscrewed and water poured into the reservoir as far as the lowest mark inside, the cover being then screwed on (the bulb of the thermometer does not touch the water). The burner, L, is then lighted under the small chamber, S, which is traversed by a brass tube communicating with the reservoir ; the part a' being rather higher than a, circulation of 146. FIG. 147. 148 ORGANIC CHEMISTRY WINDISCH'S TABLE FOR CALCULATING THE STRENGTH OF AQUEOUS ALCOHOL SOLUTIONS Sp. gr. at 15 C. Grms. of alcohol in 100 grms. C.c. of alcohol in 100 c.c. Grms. of alcohol in 100 c.c. Sp. gr. at 15 C. Grms. of alcohol in 100 grms. C.c. of alcohol in 100 c.c. Grms. of alcohol in 100 c.c. 0-9999 0-05 0-07 0-05 0-9550 31-66 38-06 30-21 0-9992 0-42 0-53 0-42 0-9535 32-55 39-07 31-01 0-9985 0-80 1-00 0-80 0-9520 33-42 40-06 31-79 0-9978 1-17 1-48 1-17 0-9505 34-28 41-02 32-55 0-9970 1-61 2-02 1-60 0-9490 35-11 41-95 33-30 0-9963 2-00 2-51 1-99 0-9470 36-21 43-17 34-26 0-9956 2-39 3-00 2-38 0-9455 37-01 44-06 34-96 0-9949 2-79 3-49 2-77 0-9440 37-80 44-93 35-66 0-9942 3-19 4-00 3-17 0-9420 38-84 46-07 36-56 0-9935 3-60 4-51 3-58 0-9405 39-61 46-90 37-22 0-9928 4-02 5-03 3-99 0-9385 40-62 47-99 38-09 0-9922 4-39 5-48 4-35 0-9365 41-61 49-06 38-93 0-9915 4-81 6-01 4-77 0-9345 42-59 50-11 39-76 0-9909 5-19 6-47 5-14 0-9330 43-31 50-88 40-38 0-9902 5-63 7-02 5-57 0-9305 44-51 52-14 41-38 0-9896 6-02 7-50 5-95 0-9290 45-22 52-89 41-97 0-9889 6-48 8-07 6-40 0-9265 46-39 54-12 42-95 0-9884 6-81 8-48 6-73 0-9245 47-32 55-08 43-71 0-9877 7-29 9-06 7-19 0-9225 48-24 56-03 44-47 0-9872 7-63 9-48 7-53 0-9205 49-16 56-97 45-21 0-9866 8-05 10-00 7-94 0-9180 50-29 58-13 46-13 0-9860 8-48 10-52 8-35 0-9160 51-20 59-05 46-86 0-9854 8-91 11-05 8-77 0-9140 52-09 59-95 47-57 0-9849 9-28 11-50 9-13 0-9115 53-21 61-06 48-46 0-9843 9-72 12-05 9-56 0-9095 54-10 61-95 49-16 0-9838 10-10 12-50 9-92 0-9070 55-20 63-04 50-03 0-9832 10-55 13-06 10-36 0-9050 56-09 63-91 50-71 0-9827 10-94 13-53 10-74 0-9025 57-18 64-98 51-56 0-9822 11-33 14-01 11-12 0-9000 58-27 66-03 52-40 0-9817 11-72 14-48 11-49 0-8975 59-36 67-08 53-23 0-9811 12-20 15-07 11-96 0-8955 60-23 67-91 53-89 0-9807 12-52 15-46 12-27 0-8930 61-31 68-94 54-71 0-9801 13-00 16-04 12-73 0-8905 62-39 69-95 55-51 0-9796 13-41 16-54 13-13 0-8880 63-47 70-96 56-31 0-9791 13-82 17-04 13-52 0-8855 64-54 71-96 57-10 0-9786 14-23 17-54 13-92 0-8830 65-61 72-94 57-88 0-9781 14-65 18-04 14-31 0-8805 66-67 73-92 58-66 0-9776 15-06 18-54 14-71 0-8775 67-95 75-07 59-57 0-9771 15-48 19-04 15-11 0-8750 69-01 76-02 60-33 0-9766 15-90 19-55 15-51 0-8725 70-06 76-97 61-08 0-9761 16-32 20-05 15-91 0-8695 71-33 78-08 61-97 0-9756 16-73 20-55 16-31 0-8670 72-37 79-00 62-69 0-9751 17-15 21-06 16-71 0-8640 73-63 80-09 63-56 0-9747 17-49 21-46 17-03 0-8615 74-67 80-99 64-27 0-9741 17-98 22-06 17-50 0-8585 75-91 82-05 65-11 0-9736 18-40 22-55 17-90 0-8555 77-15 83-10 65-94 0-9731 18-81 23-05 18-29 0-8530 78-17 83-96 66-63 0-9726 19-22 23-54 18-68 0-8500 79-40 84-97 67-43 0-9721 19-63 24-02 19-07 0-8470 80-62 85-97 68-23 0-9716 20-04 24-51 19-45 0-8440 81-83 86-95 69-00 0-9710 20-52 25-08 19-91 0-8405 83-23 88-08 69-90 0-9705 20-92 25-56 20-28 0-8375 84-42 89-02 70-65. 0-9695 21-71 26-50 21-03 0-8340 85-80 90-09 71-50 0-9685 22-49 27-42 21-76 0-8310 86-97 90-99 72-21 0-9675 23-25 28-32 22-47 0-8275 88-31 92-01 73-02 0-9665 24-00 29-20 23-17 0-8240 89-64 93-00 73-80 0-9655 24-73 30-06 23-86 0-8200 91-13 94-09 74-66 0-9645 25-45 30-91 24-53 0-8165 92-41 95-00 75-39 0-9630 26-51 32-14 25-50 0-8125 93-85 96-00 76-19 0-9620 27-19 32-93 26-13 0-8080 95-43 97-08 77-04 0-9605 28-19 34-10 27-06 0-8040 96-79 97-99 77-76 0-9590 29-17 35-22 27-95 0-7990 98-46 99-05 78-61 0-9580 29-81 35-95 28-53 0-7925 100-00 100-00 79-36 0-9565 30-74 37-02 29-38 There are also Tables by Hehner, Haas, Tralles-Brix, Gay-Liissac, &c., which differ little (at most 0-1 to 0-2 per cent.) from that of Windisch. For any specific gravity not given in the Table the corresponding alcoholic degree can be obtained easily and with sufficient accuracy by proportional interuolat ALCOHOL STATISTICS, ETC 149 liquid takes place through the tubes and reservoir. When the mercury thread of the thermometer remains stationary owing to the water boiling and the steam hence having a constant temperature, the scale is adjusted by the screw, E, so that the zero-point corre- sponds with the end of the mercury column. The reservoir is then emptied, rinsed out with the wine, &c. (containing less than 15 per cent, of alcohol), and then filled with the wine to the upper mark, so that the thermometer bulb dips into the liquid when the cover is screwed on. The condenser is filled with cold water, the burner lighted, and the heating continued until the thermometer again shows a constant reading ; the corre- sponding scale-reading then gives directly the percentage of alcohol by volume. In tho case of sweet wines or beers it is advantageous to dilute with an equal volume of water, the result given by the instrument then being doubled. An ingenious and simple ca pillar imeter, recently devised by Bosla and constructed by the Italian (Enological Agency, Milan, gives the alcoholic strength of wines or spirits with sufficient accuracy in three or four minutes. The Table given in the footnote 1 indicates the volume of water to be added to 100 c.c. of alcohol of known strength in order to bring it to a definite lower concen- tration. This Table is calculated from the formula : /S'x. v \ x = 100 (. ~r- - 8} where v is the strength of the more concentrated alcohol, S its specific gravity, S' and V the specific gravity and alcoholic strength required, and x the quantity of water to be added to 100 c.c. STATISTICS, FISCAL REGULATIONS, DE- NATURED ALCOHOL. The annual production of alcohol is now about 21,000,000 hectols., 2 and of this 23 per cent, is made in Germany (the taxation amounting to 7,600,000 in 1907 and 8,000,000 in 1909 ; in 1911 the consumption of alcohol in Germany fell to 3,650,000 hectols.), 20 per cent, in European Russia, 16 per cent, in Austria-Hungary, 14 per cent, in France, 15 per cent, in the United States, 10 per cent, in England, and 1-4 per cent, in Italy. In 1908 Turkey imported about 175,000 hectols. of alcohol (one-half from Russia), FIG. 148. Concen- GIVEN ALCOHOL AT tration desired 95% 90 % 85% 80 % 75% 70% 65% 60 % 55 % 50% by vol. by vol. by vol. by vol. by vol. by vol. by vol. by vol. by vol. by vol. 90% 6-4 85 13-3 6-56 80 20-9 13-79 6-83 75 29-5 21-89 14-48 7-20 70 39-1 31-10 23-14 15-35 7-64 65 50-2 41-53 33-03 24-66 16-37 8-15 60 63-0 53-65 44-48 35-44 26-47 17-58 8-76 55 78-0 67-87 57-90 48-07 38-32 28-63 19-02 9-47 50 95-9 84-71 73-90 73-04 52-43 41-73 31-25 20-47 10-35 45 117-5 105-34 93-30 81-38 69-54 57-78 48-09 34-46 22-90 11-41 40 144-4 130-80 117-34 104-01 90-76 77-58 64-48 51-43 38-46 25-55 35 178-7 163-28 148-01 132-88 107-82 102-84 87-93 70-08 58-31 43-58 BO 224-4 206-22 188-57 171-05 153-53 136-34 118-94 101-71 84-54 67-45 25 287-0 266-12 245-15 224-30 203-61 182-83 162-21 141-65 121-16 100-73 20 381-8 355-80 329-84 304-01 278-26 252-58 226-98 201-43 175-96 150-55 15 539-5 505-27 471-00 436-85 402-81 368-83 334-91 301-07 267-29 233-64 10 859-0 ^ 804-50 753-65 702-89 652-21 601-60 551-06 500-50 460-19 399-85 _^ c.c. of water to be added to 100 c.c. of the more concentrated alcohol. For example, if an alcohol of 90 per cent, by volume is to be diluted to 50 per cent, by volume, to 100 c.c. of the former must be added 84-71 c.c. of water. 1 See Table on next page. 150 ORGANIC CHEMISTRY For every 100 litres of alcohol consumed as beverages the following amounts are used for industrial purposes : 54 litres in Germany, 19 in Austria, 18 in France, and 14 in England. These figures indicate the countries where alcoholism is causing the greatest amount of harm. 1 PRODUCTION OF ALCOHOL IN THOUSANDS OF HECTOLITRES 1902-3 1904-5 1905-6 1907-8 1908-9 Observations Germany 3383 3791 4020 4500 Austria-Hungary 2318 2480 2700 2650 Russia . 3855 4196 4500 2700 United States . 2900 2900 2900 France . 1800 2500 2700 2538 England 1297 1300 1284 1400 Holland . 354 367 351 Belgium 328 329 389 Sweden . 186 195 220 (Imports in 1909 : 12,000 hectolitres) Italy 173 300 293 463 Denmark 169 155 154 In Germany the exportation varies considerably : 313,400 hectolitres in 1902, 14,000 in 1904, 194,000 in 1906, and 9700 in 1908. 1 Alcoholism. The abuse of alcoholic beverages is leading to the ruin and decadence of certain nations, since it is largely the cause of depopulation and produces actual decay of the human organism. Alcoholism produces a diminution in stature, as is shown by the increased numbers of those unfit for military service ; it quickly leads to crime and folly, and renders the organism easily attackable by all kinds of disease, its effects being felt to the third generation. Alcohol acts as a poison which first excites and exalts, then intoxicates and depresses the psychic faculty more or less permanently. The abuse of wine and spirits is the real cause of much intestinal catarrh and of certain visceral lesions, and sometimes leads to chronic nephritis, heart-injury, enlargement and inflammation of the liver, hepatic cirrhosis, cerebral apoplexy, progressive paralysis, and often to madness. Among the industrial classes it is thought that alcohol warms, prevents cold, and gives greater strength during work, but this is a great error based on appearances. Almost as soon as it is swallowed, the alcohol of wine and spirits is absorbed by the blood by means of the capillaries and brought into contact with all parts of the organism, the nervous centres are then more or less paralysed, and the numerous capillaries under the skin dilate, since an increased amount of blood rushes to the skin itself. The drinker has, indeed, a red face ; but the sensation of great heat is only superficial ; if the surroundings are cold, the heat of the body is more easily dispersed. This explains why drunken men, sleeping on the roads in the winter, readily die of cold. Nansen, the famous Polar explorer, withstood temperatures 52 below zero without using alcoholic liquors. The International Congress on Industrial Diseases, held at Milan in 1906, declared that the use of alcohol " is unnecessary for the nourishment of the workman, and becomes harmful where the work is heavy or long. As regards useful effects in the food rations of the worker, alcohol may be advantageously replaced by sugar, coffee, and tea." Alcohol may diminish the using-up of fat in the organism and hence the consumption of proteins, but as a food it is very costly and of little effect. During the last few years alcohol-free wines have been prepared by crushing grapes from the best vineyards and subjecting the must to filtration and pasteurisation (heating to 60) so as to render it clear and prevent fermentation ; the wine is then stored in hermetically sealed, sterilised bottles. These wines retain the taste and fragrance of the grape and have considerable nutritive value since the sugar of the grape remains unchanged (15 to 20 per cent.). Alcohol also has a harmful effect on the reproduction of man, this explaining the slowness or absence of the increase in population of nations consuming much alcohol ; as in France, where 6,000,000 was spent in 1898 on so-called aperitives (absinthe, bitters, Ac.) alone. In England 60,000,000 is spent annually on spirits, and in Switzerland even 6,000,000. Drink causes the direct or indirect death of about 45,000 people annually in France, 40,000 in Germany, 50,000 in England, 20,000 in Belgium, and 100,000 in Russia. In Italy, L. Ferriani stated that 627 cases of death in 1904 were evidently due to acute alcoholism. Dr. Marambat affirms that in France 72 per cent, of the criminals and 70 per cent, of the individuals (121,688) appearing annually before the courts make excessive use of alcoholic liquors. In Germany, A. Baer found that 41-7 per cent. (13,706) of the prisoners (32,837) were addicted to drink ; in Switzerland, it is 41 per cent. ; and in England, 33 per cent, of those sentenced at the Assizes. In Holland, four-fifths of the crime is attributed to alcohol, and in Sweden three- fourths. Similar figures to the above have been giyen for Italy. In various countries it has been found that 25 per cent, of the lunatics are excessive alcohol drinkers. In the Salpi'triere Hospital of Paris, 60 out of 83 babies afflicted with epilepsy had alcoholic parents. In Germany, 30,000 persons are attacked every year by alcoholic delirium and other cerebral disturbances due to abuse of alcohol. Alcoholism in Germany was a national calamity as early as the fifteenth and sixteenth centuries, when to the enormous consumption of beer was added that of brandy and, after 1550, of cereal and potato spirit. After the eighteenth century, when the production of cereal and potato spirit., became a great industry, their consump- tion as beverages increased enormously. In 1905 the annual expenditure for alcoholic drinks amounted to 47*. per head, or 8 for every person over fifteen years old, making a total of 120,000,000 for the whole of Germany, or about 80,000,000 for the working classes corresponding with 12 per cent, of their wages. Every year there are 200,000 cases of inebriety, and 75 per cent, of the crimes against the person are the result of drunkenness. The question of alcoholism is closely connected with the social problem, as it is especially among the working classes and the ignorant and ill-nourished that the victims are found. Abstainers are less liable to illness and usually live longer, as is shown by the following statistics. The Tables of the Sceptre Life Association for eleven years (1884-1894) show that the mortality in the temperance section ALCOHOL STATISTICS 151 In 1874 the average consumption of alcohol per inhabitant in Italy amounted to 6-5 litres, and hi 1898 to 10-23 litres, to which must be added about 100 litres of wine. 1 In Italy the production was 80,000 hectolitres in 1878 ; 165,000 in 1888 ; 187,000 in 1898-9 ; 306,700 in 1904-5, 90,000 being from cereals, 72,600 from molasses, 59,000 from wine, 83,000 from vinasse, and 1725 from fruit. In 1907-8 Italy produced 463,000 hectolitres and exported 64,000 in 1908 (half in bottles) ; 134,000 in 1909, 40,000 being in bottles and 7000 sweetened or rendered aromatic for beverages, and 95,000 in 1910. In 1903 there were 3275 distilleries in Italy employing 8670 workmen. In 1904-5 spirit factories consumed 234,000 quintals of maize, 6000 of durra, and 17,000 of barley, rye, millet, and rice ; also 280,000 quintals of molasses and sugar and 53,000 of other materials. To these must be added 575,000 hectolitres of wine, 2,600,000 quintals of vinasse, and 13,700 of fruit. In Germany 80 per cent, of the alcohol comes from potatoes (the cultivation of which occupies 3,300,000 hectares out of a total area of 26,000,000 hectares capable of cultiva- tion) ; in Austria 60 per cent., in Russia 50 per cent., and in France 20 per cent. ; the rest is obtained from cereals and saccharine products. The origin of the alcohol produced in France is as follows, the numbers representing hectolitres : From starchy matters From molasses From beetroot From wine From cider Total 1877 . . 163,204 642,709 272,883 157,570 9,468 1,308,881 1885 . 567,768 728,523 465,451 23,240 - 20,908 1,864,514 1897 . . . 484,637 734,819 798,484 83,719 26,579 2,208,140 1901 . 269,074 1,006,933 578,628 330,966 115,220 2,437,964 1904 . 380,710 626,722 992,149 88,509 2,181,362 f about 1908 . 362,500 448,000 1,260,000 468,000 ~ \ 2,600,000 In Italy the tax for manufacturing alcohol was 21s. per hectolitre at 100 per cent, in 1871, 4 in 1883, 6 in 1885, and 7 4s. in 1887 ; to this the sale-tax of 2 8s. was added (abstainers) was 57 per cent, and that in the general section (non-abstainers) 81 per cent. In times of epidemics nine out of ten non-abstainers die and only two out of ten abstainers. The introduction of the alcoholic tendency into Africa, as a result of colonisation, wrought such havoc qmong the natives that the International Congresses against Alcoholism held in Brussels in 1899 and 1906 adopted various prohibitive and fiscal measures to save the black race of Africa from the terrible plague. Many remedies for alcoholism have been proposed, but singly they are almost all inefficacious, though more useful if combined. Increase of the price of drinks and diminution of the number of shops have proved almost useless in France, Belgium, and England. In England, however, the latest increase in taxation has diminished by one-third the consumption of spirit ; the amount of beer has fallen from 31-4 to 25-8 litres per head per annum, whilst the consumption of tea and wine has increased. In the United States the enormous taxes on alcohol have not diminished the consumption of liquors. Sweden has obtained good results by making a State monopoly of alco- holic drinks, by granting licence to sell only to trustworthy persons, by giving them special facilities for, and large profits on the sale of other beverages and of food, by abolishing profit on alcoholic drinks and by making the licencees responsible for cases of drunkenness on their premises. This example has been partially followed in America and England, and many temperance associations have helped by opening establishments where good food and drink are obtainable at low prices, alcohol being banned. Another effective factor against alcoholism is education and the explanation of the harm done by it : in schools, churches, barracks, the streets, workshops, books, reviews, newspapers, 'advertisements indeed everywhere should an intelligent campaign be waged against alcoholic liquor which, as Gladstone said in the House of Commons, commits more slaughter in our days than the three historic plagues : famine, pestilence, and war, since it decimates more than famine and pestilence and kills more than war, and is in all cases a disgrace often lowering man below the level of the brute. 1 The average annual consumption per head in litres of absolute alcohol in the form of different beverages is as follows : Germany Austria-Hungary France . England . Belgium . Denmark Sweden . . Russia . United States . Italy Beer 4-8 1-7 1-3 8-3 8-7 2-6 2-3 0-2 3-4 0-1 Wine 0-66 2-1 17-5 0-2 0-6 0-06 0-28 12-0 Spirits 4-1 5-1 3-5 2-3 3-7 7-0 3-9 2-5 2-7 2-0 Total 9-5 8-9 22-3 10-8 13-0 9-6 6-26 2-7 6-38 14-1 In Sweden 27 litres of alcohol in the form of spirits were consumed per inhabitant in 1830. 152 ORGANIC CHEMISTRY in 1888 (so that the consumer paid about 23 pence per litre in taxation alone !) ; the sale-tax was abolished in 1904. A rebate of 90 per cent, of the tax is made on exported alcohol (added to marsala, vermouth, &c.). In 1903 alcohol obtained by distilling wine and vinasse and destined for industrial use was exempted of all taxation, and to alleviate the crisis in the wine industry it was proposed, but in vain, to grant a substantial bounty to the distillers of wine and vinasse. In 1911 the tax was raised to 10 16s. per anhydrous hectolitre at 15-56. In 1910 the Italian exchequer received nearly a million sterling in alcohol taxes. In Germany the manufacturing tax of ordinary non-denatured alcohol varied prior to 1909 from 64s. to 72s. per hectolitre, this being entirely repaid on exported alcohol, which in certain cases also enjoyed a bounty of 9s. Before 1909 the tax was based on the volume of the wort, so that all distillers tried to work with concentrated worts (up to 25 Brix). Nowadays the payment is made on the volume of anhydrous alcohol produced, and the tax varies according to the production, which is established every ten years for each factory (contingent production). On this contingent quantity the tax is 105 marks (shillings) per anhydrous hectolitre, excess production paying 125 marks. There are then supplementary taxes of 4 to 14 marks to protect the small factories, so that a hectolitre of alcohol, costing of itself 28s. to 32s., with taxes, costs 7 4s. to 8 8s. The German Government received about 8,000,000 in alcohol taxes in 1908-9 and expect in the future to raise this to 14,000,000 ; but increase in the taxation has been followed by a diminution of 25 per cent, in the consumption. Potato spirit is made in 6400 large factories, that from cereals in 730 large and 6600 small factories, that from molasses in 27 special distilleries, and that from wine, fruit, and yeast by about 60,000 small dis- tilleries. In Germany, besides the concession of untaxed denatured alcohol to all indus- tries, non-denatured alcohol is also allowed free of taxes to scientific laboratories and for medicinal uses and military explosives. The alcohol of spirituous beverages imported into Germany pays a Customs tax of about 14 16s. per quintal. In England the spirit duty amounted to about 30,000,000 in 1907. In Prance alcohol for drinking pays a tax of 10 per hectolitre, whilst industrial spirit is untaxed (as in Germany), and is sold at about 4-5 pence per litre. Denatured Alcohol. In several countries denatured alcohol is allowed free of tax to manufacturers, and in Italy in 1903 this spirit was taxed 12s. per hectolitre (100 per cent.) instead of 8 (which is subject to 25 to 40 per cent, bonus if made from vinasse or wine). Denaturation is, however, allowed only for the manufacture of ether, collodion, mercury fulminate, varnishes, photographic papers, artificial silk, and alcohol for heating or illuminating purposes. In 1905 Italy also abolished the tax of 12s. for denatured alcohol of whatever origin (cereals, vinasse, &c.), but there remains the cost of the denaturant, which sometimes amounts to 2s. Qd. or more per hectolitre for about 3 per cent, of de- naturant composed of methylene, acetone, pyridine, and benzene. In order that alcohol intended for various industries may not be used for beverages (wines, liqueurs, &c.), the Government denatures it by the addition of various substances 1 stinking, coloured, or of unpleasant taste which cannot be separated from the alcohol DENATURANTS Crude wood spirit Crude pyridine Acetone Benzene Crude benzine per cent. per cent. per cent: per cent. per cent. France . . 7-5 2-5 o-r. Germany . . 1-5 0-5 0-5 __ ,, (motors) 0-75 0-25 0-25 2-0 . Austria . 3-75 0-5 1-25 . ,, (motors) 0-5 traces traces 2'5 Russia 10-0 0-5 VO . Switzerland 5-0 0-32 2-2 In the United States methyl alcohol and pyridiue are used, and, for special purposes, ether, cadmium iodide, ammonium iodide, &c. In France denaturation costs about 10 fr. (8s.) per hectolitre, and the Government makes a rebate of 9 fr. In Germany it costs only 2 marks (shillings) since much less, although sufficient, denaturant is added. In Italy denaturation is possibly excessive and too expensive. DENATURED ALCOHOL 153 by any of the ordinary means (distillation, &c.), but which do not damage the alcohol for its industrial use. The denaturant should vary according to the use to which the spirit is to be put. There are hence in all countries a general denaturant for alcohol as fuel, for motors, &c., and special denaturants. As colouring- matter, traces of crystal violet (hexamethyl-p-rosaniline hydrochloride) are used in Germany. Alcohol intended for the manufacture of ether, collodion, and artificial silk is denatured by the addition of ether and sometimes of a little acetone ; in Italy, for varnishes, 2 per cent, of methylene, 2 per cent, of light acetone oils, and 20 per cent, of a 50 per cent, solution of sealing-wax are used. It has also been proposed to use part of the stinking products obtained on distilling certain bituminous shales. In 1906-7, 41,000 hectolitres of alcohol were denatured in Italy with the general denaturant for fuel, motors, lighting, &c. (16,790 in 1903-4, about 18,500 in 1904-5, over 30,000 in 1905-6, and almost 83,000 in 1910), 1031 hectolitres for making ether (about 1100 in 1904-5 and 8120 in 1910), 38 hectolitres for collodion (63 in 1910), 130 hectolitres for the manufacture of mercury fulminate (140 in 1910), 1625 hectolitres for artificial silk in 1910, 50 hectolitres for photographic paper (1910), 995 hectolitres for lacquer according to the Dermoid patent, and 1364 hectolitres for other lacquers (1910). In France 23,000 hectolitres out of a total of 1,488,000 were denatured in 1879 ; in 1901 153,000 hectolitres with the general denaturant were used for motors and lighting, and 98,130 hectolitres with special denaturants for chemical industries ; in 1904, 290,000 hectolitres with the general denaturant and 133,500 hectolitres with special denaturants, the total production being 2,180,000 hectolitres ; in 1907, 600,000 hectolitres were de- natured altogether ; and in 1908, about 626,670 hectolitres 442,758 for heating and lighting, 12,054 for varnishes, 21,300 for celluloid, 1147 for dyes, 359 for collodion, 194 for chloroform, 950 for tannin, 490 for chloral, 138,346 for ether, fulminate of mercury, and explosives, 6972 for pharmaceutical products, 587 for scientific purposes, and 1514 for other uses. In the United States, 126,000 hectolitres were denatured in 1908 and 173,000 hecto- litres in 1909 (after the law of 1907). In 1910-1911 the United States consumed 250,000 hectolitres of denatured alcohol. In Norway, in 1910, 400 hectolitres were denatured, and the consumption of spirits, which was 40,000 hetcolitres in 1874, diminished to 15,000 hectolitres in 1910. In Germany, 1,400,000 hectolitres of denatured alcohol were sold in 1904-5 (1,582,000 hectolitres in 1908), of which 36,000 were for motors (in 1903 only 24,000 hectolitres were used for this purpose, 12,500 horse-power being developed). 1 In 1909-1910, 1,883,000 hectolitres of alcohol were denatured in Germany. Denatured 90 per cent, alcohol now costs 465. per quintal in Italy, whilst in Germany it costs only about half this, namely, 25 marks (shillings) per hectolitre (after 1909, with the new tax, 48s.), in Austria 26s., in Switzerland 24s. (retail), and in Belgium 25s. UTILISATION OF DISTILLERY RESIDUES. All the components of the prime materials used in the production of alcohol are found (excepting the carbohydrates : starch and sugar) in the residues (grams, spent wash) left after the distillation of the alcohol. These residues formerly formed inconvenient refuse, since they readily undergo putre- faction and, if discharged into rivers or canals, contaminate the water. In exceptional cases, when the distilleries are in large agricultural centres, the residues are used in the wet state for cattle-food, but more commonly they are evaporated and dried, these dried grains being highly valued as a concentrated fodder, rich in proteins 2 and having a restricted (1 : 3 to 1 : 5) nutritive ratio (ratio between nitrogenous and non -nitrogenous substances). 3 In the fresh residues two-thirds of the part which is not water is dissolved 1 An automobile weighing 1200 kilos, on a journey of 174 kiloms. (109 miles) at 30 kiloms. (19 miles) per hour, consumed 11-3 litres of alcohol ; under similar conditions, 10 litres of petrol are required. For an 8 h.p. car, 350 grms. of alcohol or 500 of petrol are used per horse-power hour. For automobiles and explosion motors in general, the Paris Omnibus Company uses alcohol mixed with 50 per cent, of benzene, this giving a better thermal efficiency (34 per cent.). A domestic 25-candle lamp with an Auer mantle uses about 2 grms. of alcohol per candle-hour. The use of alcoholene, a mixture of alcohol and ether, has now been proposed, and from a technical standpoint presents advantages over alcohol and other mixtures. * The average percentage compositions of the principal residues will be found in the Table on page 154. 3 For fodder, the nutritive values of the proteins, fats, and non-nitrogenous digestible substances are in the proportions 3 : 2 : 1, so that the commercial value of a fodder, expressed in nutritive units, is given by : nitrogenous substances x 3 -f fatty substances X 2 + non-nitrogenous substances, given by the percentage composition of the digestible components. 154 ORGANIC CHEMISTRY and the remaining third suspended in the water. Potatoes give about 10 per cent, of dried residue, malt about 40 per cent., and maize 45 to 50 per cent. It will hence be understood how distilleries have greatly increased the raising of cattle and consequently production of stable manure, thus contributing to the fertilisation of formerly unfertile lands. The economics of the drying of these residues has always constituted a difficult problem owing to the presence of more than 90 per cent, of water in which part of the nutritive products is dissolved and to the fact that the dried residues sell at 8s. to 11s. per quintal. In many cases the liquid portion is abandoned and the solid part separated by filter- presses or centrifuges ; but if the liquid part cannot be got rid of, even after addition of lime, ferrous sulphate, &c., it is best to evaporate it by means of the hot fumes from the flues, the operation being hastened with disc-stirrers of large surface and with fans. The evaporation is sometimes carried out in a vacuum apparatus (see Sugar) furnished with stirrers, by which means a marked economy in fuel is effected (see also vol. i, pp. 442- 444). FIG. 149. Of the various drying systems (Hatschek, Meeus, Porion and Mehay, Venuleth and Ellenberg, Theisen, Biittner and Meyer, &c.), we shall only deal with that of Donard and Boulet, which has been applied with advantage in France and recently also in Italy. The solid residue from the filters or centrifuges (perhaps mixed with the evaporated residue of the liquid portion), still containing more than 50 per cent, of water, is carried by mechanical transporters into the vacuum drying apparatus (Fig. 149), consisting of a horizontal cast-iron cylinder rotatable about a hollow axis through which the steam enters or issues ; its length and diameter are 2-5 metres. Inside are a number of tubes (heating area about 60 sq. metres) into which steam is passed from D, the condensed water being discharged without coming into contact with the mass to be dried. At the TABLE OF AVERAGE PERCENTAGE COMPOSITIONS OF THE PRINCIPAL RESIDUES Beetroot Potato Rye Maize Durra Barley liquid dried liquid dried liquid dried liquid dried liquid dried liquid dried Water . 91-0 10-12 94-0 8-10 91-0 10-12 90-6 10-12 90-3 10-12 75-0 14-0 Proteins 0-9 6-7 1-3 18-24 1-9 22-28 2-0 24-26 2-0 24-26 4-0 20-0 Non-nitrogenous -\ matter I 7-2 60-65 2-6 45-55 5-2 48-52 4-9 35-40 5-1 30-34 10-0 46-0 Fatty matter ( 1-3-1-6 0-2 3-4 0-3 5-6 1-0 12-16 0-7 12-14 1-7 7-0 Cellulose > 13-15 0-9 9-11 1-0 5-7 1-0 10-12 1-1 14-16 5-0 16-0 Ash 10-12 0-5 1-2 0-6 4-6 0-5 5-6 0-8 7-8 1-3 5-0 ALCOHOLIC BEVERAGES: WINE 155 other end, by means of the perforated axis, G', the interior of the cylinder communicates with a double-action exhaust pump to carry away the vapour from the grains which are hsated in a vacuum of 700 mm., while the cylinder slowly rotates (three turns per minute). The charge consists of 25 to 30 quintals of solid grains, which are dried (to 15 per cent, moisture, it then keeping well) in less than four hours, the coal consumption being about 150 kilos. By thus drying at a relatively low temperature (in a vacuum) and out of contact with air, the oil of the grains does not become rancid. Since maize-grains contain as much as 15 to 18 per cent, of fat, it is sometimes' convenient to extract them in one of the forms of apparatus described in the section on Fats. Special interest attaches to the residues from Molasses and Beet, since these contain special nitrogenous compounds (amino-acids) and a large proportion of potassium salts utilisable for fertilisers or for chemical products. The evaporation of the liquid part of these residues may be carried to a certain stage in the ordinary vacuum plant, the mass being subsequently completely evaporated and the residue calcined in suitable furnaces (Porion model in France and Belgium) which are similar to the reverberatory furnaces or muffles used in the preparation of sodium sulphate (see vol. i, p. 161). Care must be taken not to fuse the mass, which, when discharged, should still be carbonaceous and, indeed, sufficiently so to cause it to burn when placed in heaps outside the furnaces ; the greyish mass thus obtained known in France as salin contains : water, 0-3 to 6 per cent. ; KC1, 6 to 10 per cent. ; K 2 SO 4 , 10 to 14 per cent. ; potassium phosphate, 0-5 to 1 percent. ; K 2 C0 3 , 53 to 58 per cent. ; Na 2 CO 3 ,6to 9 per cent. ; soluble substances, 9 to 14 per cent. By this treatment, however, all the nitrogen compounds are lost ; but in some cases these are used for the extraction of methyl chloride (see p. 96). The process for extracting pure potassium carbonate, ammonia, and sodium cyanide is referred to in vol. i, p. 435. During recent years the utilisation of these nitrogenous substances has assumed great importance : according to the Effront patents (1907), the amino-acids are utilised by enzymic processes 1 for the preparation of organic acids and ammonium sulphate (with each hectolitre of alcohol produced correspond 25 kilos of ammonium sulphate and 35 grms. of organic acids, principally acetic, propionic, and butyric). Since 1902, the Dessau Sugar Refinery, and since 1904 the Ammonia Company of Hildesheim, have utilised the nitrogen compounds as potassium cyanide and ammonium sulphate. In 1907 the Ammonia Company utilised 60 per cent, of the nitrogen of the residues, producing potassium cyanide to the value of 80,000 and ammonium sulphate to the value of 20,000. ALCOHOLIC BEVERAGES WINE. Only the liquid obtained by the spontaneous alcoholic fermentation of the must of fresh grapes, without any addition, should be called wine. The fermentation i^ spontaneous owing to the presence on the grapes of Saccharomyces cerevisice. Grape must has the sp. gr. 1-08 to 1-10 and contains 70 to 86 per cent, of water, 16 to 36 per cent, of sugar (glucose and levulose, which reduce Fehling's solution) ; 1 to 3 per cent, of cream of tartar, tartaric, malic, and tannic acids ; 0-4 to 1 per cent, of colouring, aromatic, extractive, gummy, and protein substances, and mineral salts. If the musts have to be transported over long distances, either they are concentrated in a vacuum or by freezing, or the fermentation is interrupted for a time by filtering them. One quintal of grapes gives 60 to 70 litres of must and 30 to 35 kilos of residue (marc). By fermentation in open vats the sugar is transformed, more or less completely, in 7 or 8 days into alcohol, large quantities of carbon dioxide being developed and a little glycerol, succinic acid, &c., always being formed. With more than 25 per cent, of sugar, sweet wines are obtained, and with less, dry wines. Fermentation cannot yield more than 15 to 16 per cent, of alcohol, as with more than this proportion the yeast dies. After the principal fermentation, when the wine, without the marc, is placed in casks of chestnut or oak, a slow fermentation goes on, this ceasing in the winter ; with increase in the alcohol -content and lowering of the temperature, the yeast and part of the tartar 1 Ehrlich was the first to show that the fermentation of amino-acids is produced by amidages. Effront (1908) found that amidases occur especially in top beer-yeasts and in aerobic yeasts which, in seventy-two hours at 40" are able to transform, e.g. all the nitrogen of an alkaline asparagine (fee this) solution, and almost all the nitrogen of the yeast itself into ammoniacal nitrogen, organic acids being formed at the same time. 156 ORGANIC CHEMISTRY (slightly soluble in alcoholic liquids) are deposited. In the spring, the clear wine is decanted into clean (sulphured?) casks, which are kept full. It can now be placed on the market, or it can be further matured by clarifying it in the cask (by shaking with albumin and a little tannin and allowing to stand) and by decanting and filtering it several times during the course of a year or more before placing in well -cleaned bottles ; the latter are corked by machinery with paraffined corks. As time goes on, the wine acquires a pleasing aroma, this process being hastened sometimes by pasteurisation, which consists in passing the wine rapidly through coils heated to about 60 ; this process ako arrests certain incipient diseases, which would otherwise end by spoiling the wine (acidity, &c.). Sparkling wines are obtained by saturating the cold wine with carbon dioxide during bottling or by bottling sweet wines, the fermentation of which continues slowly in the corked bottle ; in the latter case, however, a deposit forms at the bottom of the bottle. In order to obtain wines of constant type on a large scale, co-operative wineries have been recently instituted in France, these collecting the grapes or must from a whole district, mixing it and preventing it from fermenting by saturating it in the cold with sulphur dioxide (70 grms. liquid SO 2 per hectolitre) ; in this way, not only the yeasts, but also the moulds, bacteria, and unpleasant odours are destroyed and the must can then be kept for months in closed vessels. When part of the must is to be converted into wine, it is heated at 50 to 60 in a vacuum by allowing it to pass down a kind of recti- fying column (Barbet, Ger. Pat. 195,235, 1906), the sulphur dioxide thus removed being recovered ; selected yeast or other wine rich in yeast is then added, the resulting wines being of uniform and improved character, although somewhat rich in sulphates. These desulphurated musts might well be used as non-alcoholic wines. There are also special yeasts capable of destroying SO 2 in the musts and of starting fermentation. In Italy much has been said in favour of co-operation in 1909 and 1910, but no trial has been made on a large scale. The proportions of the most important components of wine vary between wide limits, owing to variation of the vines, soil, climate, system of wine-making, and season (certain wines contain manganese, sometimes as much as 27 mgrms. per litre). It is hence difficult to ascertain if there has been an artificial addition of constituents similar to those naturally present in the wine, so that considerable dilution with water and addition of alcohol, glycerol, tartar, sugar, &c., are not easy to detect if they do not exceed such limits. Natural wines may contain 8 to 16 per cent, of alcohol, 1-6 to 4 per cent, (for dry wines, and as much as 20 per cent, or more for sweet wines) of dry extract (obtained by evaporating a definite volume to dryness on a water-bath and drying in an oven at 100), 0-5 to 1-5 per cent, of various acids and tartar (expressed as tartaric acid) and 0-15 to 0-45 per cent, of mineral substances (ash, obtained by calcining the dry extract) ; the glycerol varies from one-seventh to one -fourteenth part of the alcohol. Naturally these variations are much smaller for wines of a certain quality and year and obtained from one and the same district, for which the results of numerous analyses have been collected. In Italy the following minimum legal limits have been recently (Ministerial Circular, 1907) established as those which must be reached for a wine to be called natural (except in cases where genuine, wines of the same origin and year are shown to give lower limits) : alcohol, 8 per cent, by volume in white wines, 9 per cent, in red ; dry extract without sugar, 1-6 (white), 2-1 (red) ; total acidity expressed as tartaric acid, 0-5 (white), 0-6 (red) ; ash, 0-15 (white), 0-2 (red) ; alkalinity of the ash in c.c. of normal alkali per litre, 11 (white), 16 (red) ; the glycerol should be from one-seventh to one-fourteenth by weight of the alcohol, and the relation between ash and extract (for dry wines or for sweet wines after deducting the sugar) should be about 1 : 10 ; plastering, 1 expressed as sulphuric acid, should not exceed 0-02 per cent. In France, and now also in Italy, watering of a wine is detected by adding the per- centage of alcohol by volume to the total acidity per litre expressed as sulphuric acid ; this should give 13-5 for red and 12-5 for white wines (in Milan, 12-5 is allowed for red and 11-5 for white wine). Wines weak in alcohol or tartar do not keep well in the warm 1 In order to prevent certain diseases to which southern wines low in acidity are liable, recourse is had to plastering, i.e. the addition of sulphites or bisulphites, which increase the quantities of sulphuric acid and sulphates. Thus some wines remain clear in the bottle, but become turbid and dark on exposure to the air ; this disease, termed casse, is prevented by addition of potassium bisulphite, which also arrests secondary fermentation. To make certain weak wines keep better in summer in tapped casks, calcium sulphite is added, this giving a slow evolution of sulphur dioxide. TESTING OF WINE 157 weather. A weak wine can be improved by either mixing with stronger wines or con- centrating by freezing, water then separating in the form of ice (this method, in use even in the Middle Ages, has recently been patented in Italy !) New wine has sometimes the smell and taste of rotten eggs, i.e. of hydrogen sulphide ; this can be remedied by decanting it into casks in which sulphur has been burnt : 2H 2 S + S0 2 = 2H 2 + 3S. 1 From the vinasse remaining after the wine is drawn off a little rather rougher wine can still be obtained by subjecting it to considerable pressure, and from the pressed vinasse alcohol (see above) and tartar (see later) can be extracted. The testing or analysis of wine is usually limited to determining the alcohol (by the method described on p. 146), dry extract, ' ash (see above), glycerol, plastering, and total acidity, and to testing for the addition of colouring -matter and other adultera- tions. 2 Statistics. The countries which produce the most wine are France, Italy, and Spain. For Italy the statistics are very contradictory, and even the official ones are erroneous ; for instance, the production for 1909, which was given officially as 40,000,000 hectols. was officially corrected in 1910 to 60,000,000 hectols. 1 To desulphur musts and wines, use is sometimes made of a small quantity of urotropine (hexamethylene- tetramine) ; such addition can be detected, according to Fonzes-Diacon and Bouis (1910) by distilling 25 c.c. of the wine with 3 drops of sulphuric acid, acidifying the first 5 c.c. of distillate with 1 c.c. of sulphuric acid, and observing if it colours a solution of fuchsine decolorised with sulphur dioxide. The residue from the distillation is rendered alkaline with magnesium hydroxide and distilled, the vapours distilling over being condensed in a known volume of N/10 sulphuric acid, which is titrated back to ascertain how much ammonia distils over from the urotropine. 1 The Total Acidity is estimated by titrating 10 c.c. of the wine, diluted with water, with N/10 sodium hydroxide solution, using blue litmus paper as indicator ; multiplication of the number of c.c. of NaOH by 0-75 gives the total acidity in 100 c.c., expressed as tartaric add. A volatile acidity (acid that distils in a current of steam) exceeding 0-1 per cent., expressed as acetic acid, indicates a sour wine. The Cilycerol is determined by evaporating 100 c.c. of wine to about 10 c.c. on the water-bath, then adding sand and milk of lime until it is strongly alkaline and evaporating to dryness ; the residue is taken up in 50 c.c. of 95 per cent, alcohol, the solution boiled and filtered, and the residue washed with 150 c.c. of hot alcohol ; the filtrate is then evaporated on the water-bath to a syrup, which is well mixed with 10 c.c. of absolute alcohol and 15 c.c. of ether, allowed to deposit, filtered into a tared dish, the residue on the filter being washed with a mixture of equal volumes of alcohol and ether. Evaporation of the solvent leaves the glycerol, which is dried in a steam- oven and weighed. Plastering is allowed bylaw up to a maximum quantity of total sulphuric acid (of sulphates) corresponding with 2 grms. of normal potassium sulphate per litre. Hence, on adding to 50 c.c. of the wine, 50 c.c. of a solution of BaCl 2 (2-8 grms. of the crystallised salt and 50 c.c. of HC1 to a litre), boiling, allowing to stand, and filtering, the filtrate should give no further precipitate with barium chloride solution, that already added being exactly sufficient to precipitate the maximum allowable amount of potassium sulphate. Excessive sulphuration of wines is sometimes masked by the addition of urotropine (see above), which decomposes into ammonia and formal- dehyde, the latter fixing the sulphurous anhydride ; this can, however, be detected by Schiff's reaction (see Aldehydes). Artificial Coloration. 100 c.c. of the wine are evaporated to about one-third the volume, 3 to 4 c.c. of 10 per cent. HC1 and 0-5 grm. of well defatted white wool being then added and the liquid boiled for five minutes ; the solution is then poured off, and the wool, after being thoroughly rinsed in running water, is repeatedly boiled with fresh quantities of water acidified with HC1 until the latter no longer becomes coloured ; the wool is again well washed with water and boiled for ten minutes with 50 c.c. of water and 15 to 20 drops of concentrated ammonia solution, the wool being then removed and the boiling continued to expel the ammonia ; the liquid is then slightly acidified with HC1 and boiled for five minutes with fresh wool. If the latter, after washing, remains distinctly red, the presence of artificial coal-tar colours in the wine may be certified ; but if the colour of the wool is faint or indefinite, the colour is removed with water and ammonia, and the solution acidified and boiled with fresh wool ; even a faint red coloration of this confirms the presence of coal-tar dye. L. Bernardini (1910) finds that if the lower end of a strip of filter-paper is dipped into wine coloured with vegetable or animal substances, these rise to a greater height than the cenocyanin, which is more tenaciously fixed ; hence, after the paper has been dried, different parts can be tested for artificial colouring-matters by the characteristic general reactions (see Table of Colouring-Matters in Part III). It has been observed recently that the natural colours are slowly decolorised (in 48 hours) by hydrogen peroxide, whilst the artificial ones are not, Salicylic Acid and Saccharin are detected as in beer (see later). Added water is difficult to recognise if it does not bring the constituents of the wine below the legal limits (see above), and sometimes as much as 40 per cent, of water can be added to strong wines without reaching these limits. However, since natural wines never contain nitrates, which are present in almost all waters, the following test may be made : 100 c.c. of the wine are treated with 6 c.c. of lead acetate solution and filtered. To the filtrate are added 4 c.c. of concentrated magnesium sulphate solution and a little pure animal charcoal, the liquid being shaken, allowed to stand a short time, and then filtered. To a few drops of the decolorised liquid are added a few crystals of diphenylamine and 1 to 2 c.c. of concentrated sulphuric acid. If a blue coloration is formed, the presence of nitrates is demonstrated the reagents being assumed to be pure. If the wine is watered with distilled or condensed water, or pure rain water, the reaction for nitrates is not given. The addition of Glucose to wine or liqueurs is detected by adding to the wine a little pure yeast (pressed yeast) so as to ferment completely any grape-sugar still present as well as the added glucose. Commercial glucose, prepared from starch, always contains a small quantity of unfermentable, dextro-rotatory substances, so that if the wine, after fermentation is complete (when no more CO, is evolved) and after decolorisation with animal charcoal or with a little lead acetate, still exhibits a dertro-rotation greater than 0-5" in a 20 cm. tube, the presence of, glucose is proved. 158 ORGANIC CHEMISTRY The following figures represent hectolitres (1 hectolitre = 22 gallons) France. Italy Production Production Exportation 1875 83,000,000 28,000,000 363,000 1879 25,000,000 34,000,000 1,075,000 (phylloxera) * 1884 34,781,000 21,000,000 2,380,000 1887 24,333,000 34,000,000 3,603,000 (2,800,000 to France) 1889 23,000,000 22,000,000 1,440,000 (commercial treaty with France broken in 1887) 1893 . 59,000,000 32,000,000 2,362,000 (750,000 to" Austria-Hungary ; 300,000 to Switzerland, 1 and 426,000 to America) 1897 32,000,000 28,000,000 2,400,000 1901 58,000,000 44,000,000 1,334,000 1902 35,000,000 2,164,000 (976,300 to Austria-Hungary). 1903 39,000,000 1904 42,000,000 1,200,000 (Austro -Hungarian market lost by new commercial treaty) 1905 57,000,000 28,000,000 980,000 (worth 1,400,000) 1906 62,000,000 30,000,000 710,000 1907 66,070,000 54,000,000 920,000 1908 66,500,000 52,000,000 1,200,000 1909 66,000,000 60,000,000 1,450,000 (France has a vine area of 1,625,630 and Italy of 3,500,000 hectares) 1910 42,000,000 2,000,000 In the Italian exportation is included that of marsala, vermouth, and bottled wine, this amounting in 1885 to 1,200,000 bottles and flasks (including vermouth and marsala), in 1894 to 3,000,000, in 1897 to 4,720,000, in 1904 to 8,120,000, and in 1905 to 9,000,000 (worth 440,000), whilst in 1891 France exported 33,000,000 bottles (worth 3,800,000) and to-day has an enormous export. The world's 'production of wine in 1902 was 126,000,000 hectolitres: 16,000,000 in Spain, 5,200,000 in Austria, 2,000,000 in Hungary, 5,000,000 in Portugal, 3,500,000 in Algeria, 2,000,000 in Germany (formerly 3,700,000 ; in 1906 1,636,000 and in 1907 2,492,000 from 118,600 hectares of vineyards, in 1910 846,139 hectolitres), 2,300,000 in Russia, almost 2,000,000 in Turkey and Cyprus, nearly 1,000,000 in Greece and its islands, 2,300,000 in Bulgaria, 2,700,000 in Roumania, 500,000 in Servia, 1,100,000 in the 1 The following is a statistical resume of the wine imported into Switzerland from 1906 to 1910 (in hectolitres) : From 1906 1907 1908 1909 1910 Italy . i, 137,843 300,208 531,776 651,726 828,559 France 273,731 581,163 363,769 386,486 216,909 Spain 123,587 219,666 415,052 352,347 422,775 Austria 53,411 78,104 69,634 91,034 110,608 Greece 9,370 10,120 12,209 42,234 64,874 Algeria-Tunis 19,520 52,501 17,342 11,118 10,714 Germany 10,009 8,870 7,619 7,650 5,504 Turkey . . 5,743 5,183 3,637 2,865 2,451 Other countries 50 50 62 457 Total hectolitres 633,566 1,255,865 1,421,290 1,546,027 1,675,427 Total value . . "^ 1,430,800 2,255,840 VARIOUS WINES, ETC. 159 United States, 1,500,000 in Argentine, 2,500,000 in Chili, 350,000 in Brazil, 327,000 in Australia, &c. The total production in 1909 was estimated at 160,000,000 hectolitres. The average annual consumption per head is 144 litres in France, 121 in Italy, and 116 in Spain. In Milan in 1909 duty was paid on 1,000,000 hectolitres, the Corporation receiving 420,000, and the consumption per head being 200 litres. In 1905 Italy exported to Germany 124,000 quintals of dessert grapes, whilst France exported only 78,000 quintals. In 1892 Italy exported about 260,000 hectolitres of wine to Germany, but less amounts in subsequent years. 1 MARSALA. This is a liqueur wine made for the first time at Trapani in 1773 by J. Woadhouse of Liverpool to compete with the world-famous Madeira. In 1812 another large establishment was started by the Englishman, Benjamin Ingham, and in 1840 Vincenzo Florio's factory which has since become the most celebrated. The prime material for the manufacture of Marsala is white Trapani wine with 13 per cent, of alcohol, to which is added (in quantity varying for different types of Marsala) the must (cotto) of very mature white grapes, concentrated in open boilers until two-thirds have evaporated ; then is added, in varying amount, the sifone, obtained by filling a cask to the extent of three-fourths with clear, must from a very ripe white grape, and one-fourth with pure alcohol (free from tax if for export), mixing and allowing to age so as to develop the Marsala aroma. Mixtures of these three components in different proportions give the various brands of Marsala : the Italian brand is the least alcoholic (16 to 17 per cent.) ; the original English brand, the strongest (up to 24 per cent, of alcohol) ; while the Margherita and Garibaldi brands are of intermediate strengths and are sweeter, In 1904 Italy exported in cask 30,540 hectolitres of Marsala, worth 92,000 ; in 1905,29,765 hectolitres, worth 83,280, and 51,000 bottles, value 2040 ; in 1906,26,800 hectolitres ; in 1907,27,677 hectolitres ; in 1908,24,900 hectolitres ; and in 1909,24,800 hectolitres, of the value of 97,600, together with 136,000 bottles. In 1910 the exportation was 32,500 hectolitres. VERMOUTH. This was prepared formerly in Tuscany, but nowadays almost exclusively in Piedmont, where the industry was started in 1835 by Giuseppe Cora and A. Marendazzo. The prime material for manufacturing vermouth is the muscat wine of Asti and of the Monferrato heights, which contains 6 to 11 per cent, of alcohol and 2 to 4 per cent, of sugar ; with this is mixed 2 to 5 per cent, of a vinous infusion of aromatic drugs in which wormwood predominates and which contains also sweet flag, juniper, gentian, &c. ; finally alcohol is added to bring the strength up to 15 to 18 per cent, and sugar to the density of 6 to 9 Be. (if for exportation, 90 per cent, of the alcohol and sugar taxes are refunded). Sparkling vermouth is made by saturating it with CO 2 in the cold under pressure. It cannot be said that in the manufacture of Marsala and vermouth all the rational methods prescribed by modern oenotechnics are followed. The production of vermouth in Piedmont is now about 250,000 hectolitres, the exports (especially to America) being 12,400 hectolitres in cask and 31,214 hectolitres in bottle in 1902 ; 10,000 hectolitres (24,000) in cask and 53,500 hectolitres (224,000) in bottle in 1905 ; 8960 hectolitres in cask and 64,980 hectolitres in bottle in 1906 ; 8600 hecto- litres in cask and 77,800 hectolitres in bottle in 1907 ; 7874 hectolitres in cask and 83,300 hectolitres in bottle in 1908 ; 10,176 hectolitres in cask (27,680) and 100,000 hectolitres in bottle (464,920) in 1909 ; 20,400 hectolitres in cask (53,040) and 173,670 hectolitres in bottle (760,000) in 1910. CIDER. This is an alcoholic drink obtained by the partial fermentation of the juice of apples and pears. It is largely used in the north of France, in Germany, and in Swit- zerland. It is consumed almost immediately it is made. In France the production varies from 8,000,0000 to 30,000,000 hectolitres, part of which is distilled to produce alcohol (30,000 to 70,000 hectolitres of alcohol). LIQUEURS. These contain 40 to 70 per cent, of alcohol. The finest are those obtained by collecting the first, more highly alcoholic distillate from other fermented liquors. Such are brandy (prepared by distilling vinasse or wine and containing 45 to 55 per cent, of alcohol), cognac, kirschwasser (obtained especially from the cherries of the Black Forest ) f 1 The import dutie elevied by different countries on Italian wines are as follows : Germany, 29*. per quintal ; Belgium 18. &d. Holland, 34*. ; England, 23s. for wines with less than 14-84 per cent, of alcohol, and 54.*. M. for strosger one li -. i 45s. ; United States, 54s. 6d. ; and British India, 33*. 6d. 160 ORGANIC CHEMISTRY rum (prepared principally in Jamaica by distilling fermented cane-sugar molasses), maraschino (prepared from small Zara cherries), gin (from juniper berries), atole or chica of South America, arrack of the Arabs and Indians (prepared from rice, cane-sugar, and coco-nuts), schnapps of the Germans (potato spirit), &p. The other class of liqueurs comprises those obtained from aromatic substances, sugar, and more or less concentrated pure alcohol. In this way are obtained rosoli, anisette, absinthe (alcoholic decoction and distillation with wormwood) much used in France and the principal cause of the terrible effects of alcoholism (p. 150) creme de menthe, creme de cafe, &c. ; ratafia from fruit must, spirit, and sugar ; Chartreuse (the most cele- brated was that prepared by the Carthusian monks, before their expulsion from France in 1904, from balm-mint, cinnamon, saffron, hyssop, angelica, sugar, alcohol, and other ingredients), coca (from Bologna), curacao (first prepared from two kinds of orange in the island of Curasao in the Antilles), kummel (in Russia the best kinds are obtained by distilling brandy or alcoholic liquids with Dutch cumin seeds and dissolving pure sugar in the highly alcoholic distillate). It is unnecessary to mention that all liqueurs, even the most celebrated, are more or less poorly imitated in all countries with mixtures in no way resembling the original types, but the latter always command very high prices. Cognac is a brandy prepared especially in Charente by very carefully distilling weak wines of special vintages and refining and maturing the product in casks of Angouleme or Limousin oak, which gradually imparts to the spirit a pale yellow colour and a characteristic aroma. The finer and older brands sell at as much as 40 per hectolitre. FERMENTED MILK. This bears the following names according to the locality and method of its preparation and the nature of the milk from which it is made : kephir, koumis, galazin, leben (Egypt), and mazun. The first three of these are the best known. KEPHIR, or Kefir, is of very ancient origin among the Caucasian highlanders, who nowadays make enormous use of it and jealously keep the secret of its preparation. There is a legend to the effect that Allah was the first to make it, and that he recommended it as a remedy for various diseases. Kephir is simply cows' milk (fresh or skim) fermented by the addition of a special ferment in the form of granules, which the Russians call "fungi " and the Tartars " grain or millet of the Prophet," as they regard it as discovered by Mahomet. It was only in 1882 that Dr. Dmitrieff called the attention of the rest of Russia and of Europe to kephir and its great recuperative properties in cases of lung diseases. Kern and, later, Freudenreich showed that the alcoholic fermentation of milk with millet of the Prophet is due to the simultaneous action (symbiosis) of the new Saccharomyces kephiri (similar to ordinary Saccharomyces ellipsoideus), a streptococcus, and a bacillus. The alcoholic fermentation of milk-sugar with evolution of C0 2 takes place rapidly and is always accompanied and followed by acid fermentation (lactic acid), which partially dissolves the casein (propep tones) and forms a very fine coagulation, almost a frothy emulsion. In practice the kephir granules are softened with tepid water (30 to 35) for a couple of hours, the milk being then added and the mixture shaken every hour for eight hours ; it is then sealed up in clean bottles fitted with mechanical stoppers and is shaken now and then, the temperature being maintained at 15 to 20 ; in 24 hours' time the kephir is ready ; it forms a slightly alcoholic and acidulated dense, frothing liquid. If the kephir is left in closed bottles for two days, the pressure increases and the mass becomes more acid and more liquid ; by the third day it becomes extremely acid and contains up to about 2 per cent, of alcohol, and after this it is inadvisable to drink it. In Italy kephir or kephir-extract is placed on the market by the Borgosatollo Dairy (Brescia) and kephir dried in vacua is also prepared (Rosemberger, Ger. Pat. 198,869, 1907). KOUMIS is similar to kephir, and of equally ancient origin, but is prepared from mares' milk. In Russia there are various sanatoria which make efficacious use of large quantities of koumis. The composition of the latter has been found to be : Water, 94 per cent. ; CO 2 , 0-9; ethyl alcohol, 1-7; lactic acid, 0-7; lactose, 1-3 (before fer- mentation 5-5) ; fats, 1-3 ; proteins, 2-3 (largely peptonised) ; salts, 0-3. GALAZIN is obtained by placing skim (cows') milk, with 2 per cent, of sugar and 0-3 per cent, of beer-yeast in strong, tightly stoppered bottles, and allowing fermentation to proceed for twenty -four hours at 16 ; from the second to the sixth day the proportion of alcohol rises from 0-3 to 1-5 per cent. Galazin is less nutritious than kephir or koumis. BREWING 161 BEER This is another alcoholic liquor saturated with C0 2 and is obtained by fermenting aqueous decoctions of barley-malt and hops. The ancient Egyptians were acquainted with the manufacture of beer and held it in great regard. Later it became known to the Ethiopians and the Hebrews, but the Greeks never acquired a taste for beer. The industry was taken by the Armenians from Egypt into the interior of Asia, and still later beer was manufactured in Spain and France, but it was never consumed by the Romans. In Germany beer has been made from time immemorial. A marked improvement in the manufacture of beer dates from the time of Charles the Great, when hops were first used. FIG. 150. FIG. 151. Lager beer (see later) was prepared as early as the thirteenth century, and its use has since been greatly extended in various countries. In England the manufacture has flourished since the fifteenth century, the famous porter being first made at the beginning of the eighteenth century. The improvements made in brewing operations by the introduction of scientific methods have led to a very considerable development of the industry in Germany and elsewhere. The prime materials for the manufacture of beer are barley, rice, maize, &c., hops, water, and yeast. LA. BARLEY 1 should satisfy the following requirements : 1 Barley (botanical species Hordeum) used for making beer is of two types : two-rowed (Fig. 150), in which the corns are arranged in the ear in two rows, one on each side, and six-rowed (Fig.151), in which there are three rows of corns on each side of the ear. Different kinds of barley can, to some extent, be recognised by the form of the small basal bristle found at the base of the corn inside the longitudinal furrow. The value of barley for brewing purposes is largely influenced by the nature of the soil, climate, methods of cultivation, and manuring. Barley is cultivated in all countries and in all climatesin Holland and also in Sicily. It is difficult to keep II II 162 ORGANIC CHEMISTRY (a) When moistened and kept at 25 to 30, 80 per cent, of the corns should germinate in 48 hours and 90 to 95 psr cent, in 72 hours. (b) Those are preferred which are heaviest (60 to 70 kilos per hectolitre) and contain about 62 per cent, of starch, about 10 per cent, of protein, and 12 to 14 per cent, of moisture. (c) The skin should be thin and the colour pale yellow, the ends of the corns not being brown. Barley starch swells at 50, and with water forms a paste at 80. With diastase it begins, unlike potato starch, to saccharify as soon as it is completely transformed into paste. ! B. Wheat is sometimes used, together with barley, for pale beers. C. Maize is used in America after being skinned and degermed, the germ being rich in oil. D. Rice is used in America and Scandinavia with the barley. 2. HOPS. The female flowers, dry and mature, of Humulus lupulus (Fig. 152) are used, these containing 10 to 17 per cent, of a powder (which can be separated by shaking and sieving) possess- ing the aromatic and bitter principles which bestow on the beer its aroma and keep- ing qualities. 1 varieties pure, since they become modified during growth owing to crossing. Only by the rational system of selection initiated by Dr. Nilsson at the Svalof Institute is it possible to fix different varieties with constant, well-marked charac- ters suited to the various districts in which at one time they originated. From a commercial point of view, the weisjht of a barley is of importance and good qualities give a weight of 40 grms. per 1000 corns, or 62 to 67 kilos per hectolitre for thin barleys and as much as 70 kilos per hectolitre for the larger ones. The grains should have a floury and not a vitreous appearance when cut through, and there should be few broken corns as these do not ger- minate and become mouldy on the malting floor. Germination tests, made on 500 or 1000 corns, should show at least 95 per cent, of ger- m nated corns in 5 to 6 days. With barley harvested under wet conditions, the ends of the corns are darkened. The world's production of barley in 1906 amounted to 315,000,000 quintals ; in France, in 1909, 10,800,000 quintals (or 17 million hectols.) were grown on an area of 737,300 hectares ; Italy imported 104,000 quintals in 1907, 124,000 in 1908, 176,000 (value 126,120) in 1909, and 178,000 (value 128,120), mostly from Austria- Hungary, in 1910. 1 The best hops are cultivated in Bohemia (at Saaz), Bavaria, Posen, Wiirtemberg, Baden, and Alsace- Lorraine, where they are picked towards the end of August. If they are too ripe the bracts of the hop-cones open and lupulin is lost. The hop should have a yellowish green, and not a brown, colour, and the bracts should not be opened ; a too green colour indicates that the hops have been picked in an unripe condition. The seeds have no value for brewing purposes, but the largest hops are of least value. They should not have an unpleasant odour. Since the fresh hops contain 75 to 85 per cent, of moisture, so that they will not keep, it is necessary to dry them in the air or in ovens at 25 to 30 with a strong current of dry air, until they contain only 12 to 15 per cent, of moisture; they will then keep well, even for a year or more. Their keeping qualities may be improved by sulphuring them (with S0 2 ) either when dry or during the drying. Sulphuring is, however, often applied to inferior hops to mask their defects. The better qualities are seldom sulphured and, when they are well dried, are kept tightly compressed in large sacks or in evacuated metal cylinders. They may also be kept in a very cool place (cold store). The bitter flavour and keeping properties imparted by hops depend on their content of a- and p-bitter acids, which varies from 6 to 18 per cent, and is determined by Lintner's method as follows : 10 grms. of an average sample of the hops are heated in a flask graduated at 505 c.c., with 350 c.c. of light petroleum (b.pt. 30 to 50) for six hours on a water-bath at 40 to 45, an efficient reflux condenser being fitted to the flask. When the latter Is cold, it is filled to the mark with light petroleum and shaken, the contents then being filtered. 100 c.c. of the filtrate, mixed with 80 c,c. of alcohol, we titrated with a drcinormal potassium hydroxide solution in Fia. 152. STEEPING 163 3. WATER. Formerly water for brewing purposes was invested with a mysterious importance, but nowadays the water is tested in a much more rational and rigorous manner. Preference used to be given to moderately soft water, but now waters of medium hardness are regarded as best, as it is found that a certain quantity of calcium sulphate aids fermentation ; but if the water is too hard, less extract is obtained from the malt and hops. Iron is also harmful, and especially so are waters contaminated with bacteria. 1 The principal operations in the manufacture of beer are as follow : (1) CLEANING OF THE BARLEY, to remove dust, soil, stones, damaged and light corns, &c. by means of sieves, fans, &c. (2) STEEPING OF THE BARLEY for 2 or 3 days in water at 11 to 12 in order that it may absorb the water necessary for germination. For this purpose use is generally made of the Neubecker tank (Fig. 153) made of iron plates, opan at the top and cone-shaped at the bottom. In the middle is a wide perforated pipe, E, which is surrounded by the barley (500 to 3000 kilos). The water is supplied by the pipe W, and is discharged through the perforations of E, thus covering the barley ; it is then discharged from the top of the tank through the pipe U, the lighter floating corns being carried away. After 7 or 8 hours the water is run off through the tap C, and the moist barley left exposed to the air for 5 or 6 hours. Fresh water is then introduced and left for 10 to 12 hours, after which it is run off and the grain exposed for 5 or 6 hours, and so on. This procedure is continued for 30 to 50 hours in summer or 70 to 100 hours in winter, the corns having in that time absorbed about 40 per cent, of water. Steeping of the barley in lime-water has been suggested as a means of preventing abnormal fermentations (Windisch, 1901). In some cases steeping is preceded by washing the barley in running water in rotating cylinders ; or else compressed air is forced into the steeping vessels at frequent intervals, so as to stir the barley. The of 10-15 drops of phenolphthalein solution. If much fat is present an aliquot part of the light petroleum solution is evaporated and the residue extracted with methyl alcohol, which does not dissolve the fat and, on evaporation, gives the bitter acids ; these can then be weighed. The quality and commercial value of hops are influenced largely by the nature of the soil and the quality of the manure used, as well as by the variety of the hop itself. Chemical composition does not always give satisfactory indications for judging of the value of hops, and this is almost always done by men experienced in valuing hops. Hops give up to alcohol 22 to 30 per cent, of extract, about two-thirds of which is composed of a resin giving the bitter flavour and acting as an antiseptic towards certain bacteria injuriously affecting the keeping of the beer, although it has no influence on the yeast. The flavour of the beer is also considerably affected by the tannin contained in the hop to the extent of 2 to 4 per cent. The determination of the ethereal extract is also employed in judging of the quality of hops ; with good qualities, after evaporation of the ether, 27 to 28 per cent, of residue is left (see above, Lintner Test). The total area of the earth's surface under hops in 1909 was 97,421 hectares (of which 29,000 hectares in Germany) and the production varied from 10 to 15 quintals per hectare. Germany imported 28,000 quintals of hops in 1908 and 36,360 in 1909, but exported 124,000 quintals in 1908 and 88,000 in 1909. The hops imported by the United States were valued at 247,200 in 1910 and at 427,400 in 1911, and those exported at 461,400 in 1910 and at 851,600 in 1911. 1 The compositions of various waters are as follows : FIG. 153. Good Medium Cad Dry residue ...... 250-450 450-550 550-700 A Ferric oxide and alumina (Fe 2 O a ,Al 2 O 3 ) 0-1-5 1-5-2-5 3 1 .- Lime (CaO) 120-150 150-200 200-300 1 .2 Magnesia (MgO) ..... 20-50 50-80 80-120 1 3 Sulphuric acid (SO,) 20-60 60-80 100-200 j Ammonia ....... trace-1-5 / <> 1 Q< Nitrates ....... 0-0-5 0-5-1-5 j Organic matter (as oxygen absorbed) . 0-4-1-5 1-5-2-0 2-3 | Hardness (French degrees) .... 15-25 25-35 35-50 / PH Number of bacteria per 1 c.c. 50-500 500-4000 4000-10,000 These numbers are only indicative apd mupt not be tefcen too strictly, 164 ORGANIC CHEMISTRY steep-water becomes yellowish brown and acid, and after some time undergoes lactic and butyric fermentations. At the end of the operation, the barley is discharged through the lower aperture, A, by undoing the screw, B, and raising the tube, E, by means of the lever, D. (3) GERMINATION OF THE BARLEY. The steeped barley is carried to the spacious malting floor, which is fitted with numerous windows to allow of the renewal of the air when desired, and is arranged so that the temperature can be maintained constant at 15 to 20. On the impermeable floor (of cement or asphalte), the barley is spread out in a layer 50 to 60 cm. deep, and on the second day the mass is moved with wooden shovels so as to reduce the depth to 30 to 35 cm., this being further reduced to 15 cm. on the third day. Every 8 Or 10 hours the grain is turned, the floor being kept well ventilated. The temperature gradually rises, but should - not be allowed to exceed 20 ; if necessary it can be modified by turning more often and thinning out the barley. After the second day the radicles begin to sprout and later the plumule. In eight to ten days the rootlets become twice or three times as long as the corn and the transformation of nitrogenous material into diastase is at its maximum (Fig. 154 shows the various stages in the ger- mination of barley). The germination should then be interrupted so as not to lose any part of the diastase formed, the green malt then containing about 40 per cent, of moisture. A floor of 20 sq. metres is sufficient for only 1000 litres of steeped grain. If the piece dries too much, it is moistened by sprinkling with water. In order to prevent mould- growth when the floor is free, it, and also the walls, are washed with 1 per cent, calcium bisulphite solution, the floors being then well dried by ventilation. FIG. 154. - v ; ;:.,..,-. '' -. ?>>'-... -K^': .o>v,' ' v . ; ^- x ;.- -..-.> ^u.^.:..^^^""- ';v.\ ,;- ,;A G=r^ ""tSS? L "^l i "Q "D~ D ~D"D B "Q n T D D"C" B "n o ~G~D a a Ta o"o~ B"o t I C L D i FIG. 159. FIG. 160j and fitted with a perforated false bottom of sheet-iron, for each day that the germination lasts. These vessels, B, communicate under the false bottom with a channel containing a fan which draws moistened'air through the mass of barlej' in the vessel (50 cm. deep). 166 ORGANIC CHEMISTRY Above each vessel is a mechanical turner, A, with a number of screws which rotate in the barley as the turner passes along the vessel. The turner can be transported from one vessel to another and is put into operation twice a day at first (the temperature of the barley being 12 to 14), then four times a day (at 15 to 18), and finally twice a day (at 18 to 50"). In some maltings a saving is effected by operating the fan only at intervals when the temperature rises. Dry air, drawn along the channels, S, is finally passed through the malt. The advantages of the various mechanical processes over the old system of malting are that they can be worked continuously and at any season of the year, while they occupy less space, allow of efficient regulation, of temperature, economise labour and general exp3nses and diminish the percentage of waste. (4) KILNING OF MALT. The germinated barley is too moist to keep sound, and as breweries require large stocks of malt this must be dry and capable of being kept. If the moisture is reduced to 6 per cent, by air alone the germination process is stopped, and on subsequently raising the temperature to 60 a slight diastatic saccharin- cation occurs, this being greater in amount if the moisture is kept at 12 to 15 per cent. ; beyond 70 the diastase is destroyed and certain substances formed which give good flavour, aroma and fullness of taste to the beer and at the same time furnish food for the yeast. When the tempera- ture exceeds 100 part of the maltose is caramelised for the making of dark beers and a considerable amount of nitrogenous substances, which would cause the beer to keep badly, thrown out of solution. In order not to destroy too much of the diastase and to make malt suitable for pale beers the drying must first be conducted with warm air. When the proportion of mois- ture has reached 5 to 6 per cent, the diastase can withstand a temperature of 60 to 70 without losing much of its activity ; whilst if the malt is heated when it contains too much moisture (15 to 20 per cent.) the diastase is rapidly destroyed. The drying is carried out in a current of warm air (or of air mixed with the hot gases from a coke or anthracite fire), which passes through the green malt placed in layers 1 5 to 20 cm. deep on wire or tile floors, often arranged one above the other. Above the upper floor is a chimney, which increases and facilitates the draught started by suitable fans. The air is heated by passing directly over a fire or through batteries of tubes heated in the usual way. During the drying the malt is turned by a suitable mechanical device, at first every 2 hours and later on continuously. The temperature of the air gradually rises, during the course of 84 to 90 hours, by 30 to 35 (during the first few hours germination still proceeds feebly, causing increase in the diastase), and ends at 100 to 110 (for dark beers). Drying is usually effected in less than 48 hours, and it is only beyond 80 that the diastase partially loses its saccharifying properties (at 90 it loses 50 per cent, and at 100 85 per cent.) ; this loss is, however, an advantage, since a too highly diastatic malt leads to excessive saccharification and hence to increased attenua- tion in the subsequent fermentation, so that the beer tastes less full. The peptases also are destroyed beyond 90, so that the nitrogenous substances are dissolved to a less extent and the beer hence keeps better. Fig. 161 shows diagrammatically a section of a two -floor malt -kiln in which the air is heated in the tubing, t, surrounding the ducts carrying the hot fumes from the coal burning on the grate, F. The hot air then traverses the malt on the floors, B and C, and issues from the chimney, D, the turning apparatus, a, being kept in motion meanwhile. To obtain 100 kilos of dry malt in 24 hours (maximum temperature 90 to 100) 20 kilos of coal are required. For making dark beers of the Munich type part of the kilned malt is further roasted at about 200 in suitable rotating iron cylinders heated by direct fire ; this treat- ment leads to the formation of caramel, which colours the beer, the malt being then called coloured malt. The temperatures on the malting floors and kiln are registered by auto- FIG. 161. MALT ANALYSIS 167 matic devices which construct diagrams showing the temperature at any particular moment. Nowadays malt for pale beers is sometimes heated only to 25 to 30. The kilned malt leaves the kiln with 2 to 5 per cent, of moisture and is then cooled and stored in silos or large bins. A malt kept for only 1 or 2 months is to be preferred to an older one. 1 1 The commercial value of a malt is determined largely by its yield of extract, which is measured as follows : 45 grins, of ground malt are placed in a tared flask with 200 c.c. of water, the temperature being kept at exactly 45 for half an hour and then raised 1 per minute up to 70, this temperature being maintained until the liquid no longer gives a blue colour with iodine ; the time required at 70 to reach this point is noted (saccharification test). The mass is then cooled and water added to bring its total weight up to 450 grins. ; after mixing and filtering through a dry filter, the density of the liquid is determined at 15 and by Windisch's or Schulze's tables the corresponding quantity of extract deduced. The latter can also be obtained from Balling's tables (see below). note being taken that they yield low values, the deficit being 0-08 grm. per cent, for specific gravities up to 1-01 ; 0-345 for specific gravities up to 1-05 ; 0-48 for specific gravities up to 1-06 ; and 0-4 for specific gravities up to . 1-08. If the maltose is to be determined directly, 10 grms. of the filtered saccharine liquid (corresponding witli 1 grm. malt) are diluted to 100 c.c., various quantities of this liquid being then titrated with Fehling's solution, 1 c.c. of which corresponds with 0-0075 grm. of maltose. C. Lintner (1886-1908) has modified the Kjeldahl method for determining the diastatic power of malt as follows : 25 grms. of the ground malt are extracted for 6 hours with 500 c.c. of water at the ordinary temperature, the mixture then being filtered ; 2 c.c. (for pale malts) or 8 c.c. (for dark malts) of the filtrate are added to 100 c.c. of 2 per cent, soluble starch solution and the mixture left for exactly half an hour, at the end of which time 10 c.c. of caustic soda solution are added. Into a number of test-tubes, each containing 5 c.c. of Fehling's solution, are introduced varying quantities of the saccharified starch solution (e.g. from 1 to 6 c.c.) ; the tubes are next immersed for ten minutes in a boiling water-bath and then taken out, and the precipitated cuprous oxide allowed to settle ; it can then be seen in which of the tubes the Fehling's solution is just completely reduced and in which it is just not reduced. A more exact result can be obtained by using quantities of the saccharified starch solution intermediate to those corresponding with these two tubes. When 0-1 c.c. of the cold water malt extract, acting for one hour on 10 c.c. of 2 per cent, soluble starch solution, forms just sufficient maltose to reduce 5 c.c. of Fehling's solution, the malt is said to have the diastatic power 100 ; if 0-2 c.c. of the malt extract is required, the diastatic power is taken as 50, and so on. BALLING'S TABLE Degrees Degrees Degrees Degrees Balling or Balling or Balling or Balling or Sp. gr. grms. of Sp. gr. grms. of Sp. gr. grms. of Sp. gr. grms. of at 17-5 saccharose at 17-5 saccharose at 17-5 saccharose at 17-5 saccharose per 100 per 100 per 100 per 100 grms. liquid grms. liquid grms. liquid grms. liquid 1-0010 0-250 1-0210 5-250 1-0410 10-142 1-0610 14-904 1-0020 0-500 1-0220 5-500 1-0420 10-381 1-0620 15-139 1-0030 0;750 1-0230 5-575 1-0430 10-G19 1-0630 15-371 1-0040 1-000 1-0240 6-000 1-0440 10-857 1-0640 15-604 1-0050 1-250 1-0250 6-244 1-0450 11-095 1-0650 15-837 1-0060 1-500 1-0260 6-488 1-0460 11-388 1-0660 16-070 1-0070 1-750 1-0270 6-731 1-0470 11-571 1-0670 16-302 1-0080 2-000 1-0280 6-975 1-0480 11-809 1-0680 16-534 1-0090 2-250 1-0290 7-219 1-0490 12-047 1-0690 16-767 1-0100 2-500 1-0300 7-463 1-0500 12-285 1-0700 17-000 1-0110 2-750 1-0310 7-706 1-0510 12-523 1-0710 17-227 1-0120 3-000 1-0320 7-950 1-0520 12-761 1-0720 17-454 1-0130 3-250 1-0330 8-195 1-0530 13-000 1-0730 17-681 1-0140 3-500 1-0340 8-438 1-0540 13-238 1-0740 17-909 1-0150 3-750 1-0350 8-681 1-0550 13-476 1-0750 18-136 1-0160 4-000 1-0360 8-925 1-0560 13-714 1-0760 18-363 1-0170 4-250 1-0370 9-170 1-0570 13-952 1-0770 18-590 1-0180 4-500 1-0380 9-413 1-0580 14-190 1-0780 18-818 1-0190 4-750 1-0390 9-657 1-0590 14-428 1-0790 19-045 1-0200 5-000 1-0400 9-901 1-0600 14-666 1-0800 19-272 Correction of Degrees Balling for Various Temperatures Determina- tion made at tempera- ture of Correction of degrees Balling Determina- tion made at tempera- ture of Correction of degrees Balling Determina- tion made at tempera- ture of Correction of degrees Balling Determina- tion made at tempera- ture of Correction of degrees Balling Deg. Deg. Deg. Deg. 4 -0-43 11 -0-22 17-5 24 + 0-27 5 - 0-40 12 - 0-19 18 + 0-02 25 + 0-32 6 - 0-37 13 - 0-16 19 + 0-05 26 + 0-37 7 - 0-34 14 0-13 20 + 0-09 27 + 0-42 8 - 0-31 15 -0-10 21 + 0-13 28 + 0-48 9 -0-28 16 -0-06 22 + 0-17 29 + 0-54 10 -0-25 17 0-02 23 + 0-22 30 + 0-60 168 ORGANIC CHEMISTRY Malt kilned with fumes direct from a coal fire communicates to the beer a certain flavour from the smoke. Also, when coal is employed which contains arsenic, the latter becomes deposited on the malt and hence finds its way into the beer. Arsenic may also be present in the glucose often used in brewing ; in this case it is introduced by the employment of arsenical sulphuric acid in the manufacture of the glucose from starch. (5) CLEANING AND GRINDING. Before the malt is mashed it is freed from dust and rootlets by means of rotating drums of metal gauze (a kind of sieve) furnished with fans. It is then ground but not too finely, the husks being kept whole as far as pos- sible, since they serve in the subsequent operations as filtering material ; if the malt is ground too fine it cannot be exhausted, as the liquid will not drain off. A suitable form of mill is the Excelsior Mill, made by Messrs. Krupp (Figs. 162 and 163). The shaft, g, fitted with fast and loose pulleys, s and t, can be shifted from right to left or vice versa through the stuffing- boxes, m, by means of the lever, d. One toothed disc, a, is fixed, whilst the FIG. 162. other, 6, rotates with the axis, g, and is so adjusted that the teeth pass through the tooth spaces of the other disc. The barley from the hopper, /, falls between the two discs, where it is ground, the ground malt (grist) being discharged at n. For the sake of economy the discs are toothed on both faces, so that when one face is worn the other can be used. The total loss in weight suffered by the barley during steeping, germination, kilning, cleaning, and grinding amounts to about 20 per cent. (6) MASHING. This consists in subjecting the ground malt to the action of warm water so that the diastase may act on the starch and convert it into soluble products. The temperature at which the maximum extract is obtained is about 65, whilst at 55 the starch is only very slightly attacked by diastase, and above 70 diastase loses its saccharifying pro- perties very largely and the wort filters through the grains (husks ; see later) with difficulty this effect is aggravated by coagulation of part of the proteins. The quantity and quality of the water have an influence on the mashing, the presence of calcium sulphate facilitating the forma- tion of maltose and maltodextrins and increasing the amount of nitrogenous substances dissolved. 3 hectols. of beer are made. There are two systems of mashing : the infusion method (at 65 to 72), used only in top-fermentation breweries, and the decoction system, used for bottom-fermentation and sometimes for top-fermentation beers, and with highly diastatic malt or when unmalted barley is used with the malt. (I) The infusion process, used largely in England and Scotland, less in France and still less in Germany, is usually carried out in one of two ways : (i) rising infusion, where the malt is first mixed to a paste with 10 per cent, of cold water and then with hot water in the ratio of two parts of water to one part of malt, so that a temperature of 40 is attained. To raise the temperature of 1 kilo of malt (which has a specific heat of about FIG. 163. From 1 quintal of malt 2 to MASHING 169 0-5) from 20 to 40 requires 10 Calories, which can be supplied by 2 litres of water at 45, the latter falling to 40 on losing 10 Calories ; owing, however, to unavoidable loss of heat, water at 48 to 50 should be used. This mixing is done in a circular mash-tun of metal or wood, furnished with a perforated false bottom several centimetres above the true bottom, in which are fitted the pipes supplying the hot water (Fig. 164). The mashing and subsequent mixing are effected by efficient mechanical stirrers or rakes. As soon as the mash has reached the temperature of 40 water at 80 is gradually introduced, the temperature being raised to 63 to 65 in half an hour. It is next mixed for 60 to 70 minutes, the liquid being then discharged by^opening the taps under the false bottom so that the liquid passes through the grains and is conducted to the copper. The residue in the tun is mixed for 15 minutes with water at 75, the liquid being run off and the grains finally washed with water at 80, all these extracts passing to the copper. In this way almost complete saccharification is attained and the subsequent fermentation produces considerable attenuation. If a less attenuation is desired, either a higher temperature (72 to 73) is used in place of 65, or high-dried malt is used. (ii) Descending infusion, which is rarely used, consists in bringing the mass directly to a temperature of 65 to 70 with very hot water and then allowing it to fall slowly to 35 to 40. Neither of these methods admits of the use of rice or maize, the starch of which is attacked by diastase only after it has been heated with water to 80 to 85. Hence with such material the decoction process is used. (II) Decoction Process. This is largely used in North Germany, Austria, and Belgium, and allows of the use of F IG . 164. unmalted barley, rice, maize, wheat, &c. The malt grist is first mixed to a paste with cold water so as to dissolve the diastase, this being carried out in a metal vessel without a false bottom ; by the addition of small quantities of boiling water the temperature of the mass is raised gradually to 35, From one-third to one-half of the turbid wort (Dickmaische) is transferred to a double-bottomed copper heated with steam. In many cases coppers with direct-fire heat are used, these being furnished with chains which scrape on the bottom and so prevent caramelisation of the mass which settles (Fig. 165 shows a complete decoction or infusion plant). The wort transferred to the copper is boiled for 20 to 40 minutes and is then returned to the original tun, where it raises the temperature to about 55. Another one-third or one-half is similarly removed, boiled, and returned, the temperature being thus raised to 65 ; the saccharification has then reached a maximum and the mash become thinner. The complete disappearance of starch is controlled by the reaction with iodine. About one-half of the wort is again removed, boiled, and returned, the temperature being thus raised to 75. During all these operations continual stirring is maintained. The greater the number of decoctions made the greater will be the density of the wort and the darker the beer. The turbid wort is either allowed to deposit in tuns with false bottoms as shown in Fig. 164, or passed through filter-presses (see Sugar Industry) to clarify it, the grains remaining in the form of cakes being well washed. 1 When considerable quantities of other cereals are to be used with the malt use 1 The grains are composed of the whole of the husks of the ' malt coagulated proteins, pentosans, fat, maltose and dextrin. They serve as excellent cattle-food, but if not consumed in the course of 24 hours, they undergo change ; they may, however, be placed in silos and dried in a suitable apparatus (see Tig. 149, p. 154). Wet grains contain 70 to 80 per cent, of water, 4 to 6 per cent, of protein, 1 to 3 per cent, of fat, 8 to 14 per cent, of extractive substances, 1 to 3 per cent, of ash, and 3 to 9 per cent, of cellulose. Dried grains contain 6 to 18 per cent, of water, 17 to 26 per cent, of protein, 4 to 9 per cent, of fat, 35 to 55 per cent, of extractive substances, 3 to 12 per cent, of ash, and 9 to 20 per cent, of cellulose ; they have a brown colour, a pleasing odour of new bread and a sweet taste ; they make a good food to follow wheat or oat bran. 170 ORGANIC CHEMISTRY is made of a Henze pressure apparatus, as described under Distilling (Fig. 104, p. 119). The wort thus obtained is boiled with a certain quantity of hops until a certain amount of concentration has been effected. This boiling finally destroys the diastase, intensifies the colour of the wort and aerates it, and oxidises various substances producing acid bodies ; it completely sterilises the liquid, which is also clarified owing to the precipitation of nitrogenous substances, partly by the tannin of the hops. The decoction of the hops is carried out in a separate vessel, the boiling liquid being continually circulated until the hops are exhausted. The decoction is then added to the boiling wort, principally towards the end of the operation ; if added earlier the hop extract loses some of its aroma. The direct addition of the hops to the copper is still used, although the method is not a very rational one ; it is better to pass the boiling wort from time to time into a separate vessel containing the hops and then back to the copper, this procedure being repeated until the hops are exhausted. FIG. 165. In general, 400 to 500 grms. of hops are used per hectolitre of beer, or 2-5 to 5 kilos for every quintal of malt mashed. More hops are usually employed for beers to be kept for some time (lager beer, stock ale) than for draught beer. The lupulin powder contained in the hop gives up resins and essential oils, while the leaves give tannin and the stalks somewhat bitter substances ; the whole gives the bitter taste and aroma of the beer, and causes the latter to keep better. A temperature of 75 (Pasteur) is sufficient to sterilise a hopped beer, since the resins have a marked antiseptic action. The boiling of the wort is carried out in copper vessels (see Fig. 165, a) heated by direct fire or by indirect steam (passed through coils or through the double bottom of the copper), the boiling being continued for 4 to 6 hours with dilute worts (infusion) and only 1^ to 2 hours with the more concentrated decoction worts ; as a rule boiling is continued until the density reaches a certain value for the particular kind of beer to be made (see later). The temperature during boiling should be gradually raised and registered. In many modern breweries there are automatic registering thermometers which show the whole course of these operations. When the boiling is finished the wort is allowed to stand for a time, and the Inland Revenue officials then generally make their first measurements (they calculate that 1 kilo of dry malt should give 25 litres of wort with a density of 1 Balling, 5 litres at 5, &c., an allowance being made of 10 per cent.). The copper is then dis- charged, the hops being strained off, and the wort pumped to the cooler, which is usually at the top of the building. These coolers are large shallow vessels of iron (or copper or wood) in whicfy the coagulated proteins are deposited ; the temperature here is not allowed COOLING OF WORT 171 to fall below 60 to 65, otherwise contamination with harmful organisms (butyric, lactic, &c.) might occur. In Italy the tax on the manufacture of beer is calculated from the volume, temperature, and specific gravity of the wort in the cooler (see later). The wort is next cooled rapidly by suitable refrigerators to 2 to 3 (for bottom fermentation) or 12 to 15 (for top fermentation). One form of refrigerator which is much used consists of a number of superposed, communicating horizontal tubes (Fig. 166). In the tubes of the upper half water circulates, and in those of the second half brine at a temperature of -6 or -8 from a refrigerating machine (see vol. i, p. 231). The wort flows down in a thin skin over the outside of the tubes, meanwhile dissolving an appreciable quantity of air. The cooled and aerated wort flows down to the fermenting vessels placed in cool rooms ; for bottom fermentation these are cooled to about by pipes conveying cold brine. The wort from the coolers is turbid and should be filtered through conical cloth bags or filter -presses. In some modern breweries the coolers are omitted in order to avoid any possible contamination (which is, however, difficult with hopped wort at 60) and the wort is passed direct from the copper to the closed refrigerator and the filter- press, aeration being afterwards effected with air filtered through cotton -wool. The refrigerators consume considerable quantities of water, and where this is scarce the warm water from the refrigera- tors is cooled by means of pulverisers or by causing it to flow down over twigs, the evaporation thus caused often lowering the temperature below that of the air (see section on Sugar). The boil- ing of the wort has hence effected a con- centration, the preparation of a sterile (aseptic) liquid, and the extraction of the useful principle of the hop, the tannin of which has partially precipitated the pro- teins. If pale beer is to be brewed the wort can, if necessary, be clarified during the boiling by the addition of a little tannin. During the cooling on the cooleis the wort takes up the oxygen necessary for the oxidation of the resins, for clarify- p IG ^gg ing it and, more especially, for aiding the development and multiplication of the yeast during the initial stages of the fermentation Contact of the wort with tin, e.g. tinned vessels, is avoided, as turbidity of the beer may be caused thereby, especially if the wort is acid or rich in carbon dioxide. FERMENTATION. From the density (degrees Balling) or the dry extract of the wort, the extract yielded by the materials may be deduced, and, under favourable conditions, the dry extract amounts to about 70 per cent, of the weight of the malt, whilst with bad working it may be as low as 45 per cent. When ready for fermentation the wort contains mainly maltose, malto- dextrins, dextrins, a little saccharose, glucose, and levulose, besides nitro- genous substances partially peptonised and transformed into amino-acids ; also lactic acid and potassium phosphates. Fermentation with yeast converts the carbohydrates more or less completely into alcohol and carbon dioxide. 1 1 In addition to what has been said on pp. Ill and 123 on ferments and yeasts in general, the following i& of interest, especially to the brewing industry : All yeasts which attack only saccharose, maltose, glucose, and levulose, giving alcohol and carbon dioxide, are feebly attenuating yeasts of the so-called Saaz type (e.g. the beer-yeasts of Li6ge, which yield fairly full-tasting sweet beers containing little alcohol). Other yeasts are also capable of fermenting the combined maltose of maltodextrins by means of a special enzyme studied by Delbriick, maltodextrinase ; these yeasts give the maximum attenuation and form the so-called Frohberg type, producing alcoholic, highly attenuated beers even from weak worts. Between these types Saaz and Frohberg there exist intermediate ones giving in 4 days at 25 to 27, well-defined attenuations in a normal wort. Certain other yeasts are capable of fermenting dextrin combined as maltodextrins, since they contain an enzyme which Delbriick has termed dextrinase. Such is the Schizosaceharomycet Pombi : , separated from the millet beer of the Egyptians. These yeasts constitute the so-called Logos type. Wild yeasts are all strongly attenuating and may produce turbidity in finished, slightly fermented beers, which they referment. The yeasts 172 ORGANIC CHEMISTRY The concentration of the wort most favourable to the multiplication of yeast is 15 Balling (corresponding with a specific gravity of 1-06). 1 A too dilute wort or one prepared with an excessive proportion of non-germinated grain has not sufficient assimilable nitrogenous food (amino-acids), and this is remedied^ by the addition of zymogen, which is a commercial product. During the period when the yeast develops (first stage of the fermentation) little alcohol and much carbon dioxide are produced. Two distinct methods of fermentation are in use : top fermentation, used generally in England, Belgium, and Holland, and largely in France, and also, at one time, exclusively in Italy ; and bottom fermentation, usually employed in Germany, Austria, and Denmark, and in general use in countries where beers of the Munich and Pilsen types are made. In hot countries it is easier to regulate bottom fermentation (by refrigeration) than top fermentation, since in summer the temperature of the air is often high enough to have an injurious effect on top fermentation. So that, as a refrigerating plant is necessary, the bottom fermentation system is preferable. The difference between bottom and top yeasts is that the latter are covered with viscous, mucilaginous substances and readily stick together and carry bubbles of carbon dioxide developed in the wort to the surface and so produce a rapid feimentation ; the former, however, fall to the bottom of the fer- menting vessel and, even under the microscope, are not found in large masses. Top yeasts develop well only at temperatures above 12 best at about 24 and effect complete fermentation in 4 to 6 days, whilst the bottom yeasts develop below 10 and, after the vigorous primary fermentation of 8 to 12 days at 6 to 8, continue the maturation of the beer for two or three months by a secondary fermentation at a low temperature (0 to 2) ; this procedure gives beers of less attenuation which can be produced or con- sumed even in summer (lager beer). Top -fermentation beers are almost always more highly attenuated, are consumed at once (draught beer), and are made more especially in the cold weather ; they can, however, be kept, and in some cases stock beers are made on this system. The advantages and disadvantages of the two processes are as follow : Top fermentation does not require costly refrigerating plant, and hence lends itself to the construction of small breweries ; further, the beer can be sold immediately, and the capital, although small, thus frequently renewed each year. The control and successful working of top fermentation are, however, more difficult owing to ready contamination with numerous harmful bacteria which find at 15 to 20 the most favourable conditions for their development, especially in the summer ; in bottom-fermentation beers only yeasts can develop at to 2. With top fermentation, in which at first yeasts of the Saaz type and those intermediate to the Saaz and Frohberg types predominate, there develop later bacteria and also Frohberg yeasts (especially during the secondary fermentation), and both of these render difficult the preparation of a clear beer which does not become turbid after fermentation ; on the other hand, a bright beer is easily and naturally obtained by bottom fermentation. In summer, then, unless an abundant supply of cold water and also cool cellars are avail- intermediate to the Saaz and Frohberg types and also Frohberg yeasts themselves are especially active in the secondary fermentation ; they increase the apparent fullness of the beer, even when this is light, arid maintain a continuous and desirable evolution of carbon dioxide by slowly fermenting the maltodextrius and even dextrins. In order to grow and multiply, yeasts generally require, in addition to carbohydrates and free oxygen, nitro- genous substances, but they cannot make use of nitrates, or ammonium salts, or even the true proteins ; they can, however, utilise the decomposition products of the latter, namely, the amino-acids (such as asparagine) pro- duced by the proteolytic enzymes secreted by healthy yeasts. They-.require also mineral substances, e.g. calcium and potassium phosphates. The oxygen of the air is, as has been said, indispensable to the development and multiplication of yeast, and well-aerated worts facilitate the multiplication during the first few days, when only CO. and H,O are produced when, however, the supply of free oxygen diminishes or ceases, the yeast produces more especially alcohol and carbon dioxide. There are also saccharomyces which are solely aerobic and form membranes on the surface of the wort, producing only carbon dioxide and water and destroying the alcohol produced by other yeasts. 1 The strengths of the worts for different types of beer are : 9 to 10 Balling for light beers ; 12 to 13 for draught beers (Schenkbier) ; 15 to 20 for double beers (Bock or Salvator beer) ; and up to 25 for table beers. FERMENTATION 173 able, and rigorous precautions and disinfection are resorted to, it is very difficult to prepare top-fermentation beer, whilst the low temperature required for bottom fer- mentation can be attained at any season of the year by refrigerating plant. Bottom fermentation gives beers of a more constant type, since the mother -yeast from succes- sive fermentations does not become contaminated so easily as, and hence requires renewal less frequently than, with top fermentation. 1 When a large amount of yeast is added to a wort the fermentation is initiated and completed more rapidly ; with small quantities the same result is obtained, but after a longer time, so that there is more danger of contamination. Usually 250 to 300 grms. of pressed yeast are used per hectolitre of wort rather more for strong worts. Especially with top, but also with bottom fermentation, it is most important that all instruments, vessels, and rooms should be kept clean and disinfected. For this purpose boiling water is used and also dilute solutions of hydrofluoric acid, ammonium fluoride, ammonium fluosilicate, calcium bisul- phite, and calcium hypochlorite. In all cases, however, great care must be taken to remove the disinfectant com- pletely with abundant supplies of hot water, in order that the yeast may not be injured. Chloride of lime is elimi- nated by rinsing first with bisulphite solution and then with hot water. Even traces of bisulphite (sometimes added during mashing to prevent the action of lactic ferments) must be completely eliminated, otherwise, during the alcoholic fermentation, which is a process of reduction, they may yield hydrogen sulphide and so give a bad taste and odour to the beer. (Bacteria capable of producing hydrogen sulphide sometimes develop in beer.) Whatever system of fermentation is used, it is always divided into two phases : the primary or vigorous, and the secondary. The primary fermenta- FIG. 167. tion begins 12 or 24 hours after pitching, when the yeast has grown to some extent at the expense of the dissolved oxygen, and continues for 3 or 4 days in the case of top fermentation or for 10 to 12 days with bottom fermentation ; considerable quantities of carbon dioxide are developed, these forming a dense, white, frothy head on which can be seen brownish spots of hop resin or agglutinated bacteria. In top fermentation, this first head is removed, the next darker one being collected for pitching purposes. In the bottom fermentation system and in large modern breweries in general, in order that the yeast may be kept as pure as possible, the pitching is carried out in the manner described on p. 127 for distilleries. 1 With top fermentation, the type and taste of the beer are determined by the united activity of a number of different yeasts and bacteria which are present in given equilibrated proportions, these becoming modified as contamination increases. When the yeast is renewed, the pure yeast naturally gives a different taste to the beer, and this inconvenience cannot be avoided by preparing a mixture of yeasts and bacteria similar to that normally present in the partially contaminated top fermentation. New pure yeasts are less resistant to con- taminating surroundings than old ones are. Attempts are made to-day to keep the fermentation pure as long as possible by the use of good, hops, the resins of which exert an agglutinating and paralysing action on the bacteria, so that these can be removed from the tun with the first scum forming on the surface of the fermenting wort ; the purer yeast of succeeding heads is then collected for pitching subsequent worts. When the collection of the yeast is delayed, that of the Frohberg type increases. With the object of maintaining the cultures naturally pure and constant, Effront has proposed the addition of abietinic acid a component of lupulin and of colophony to agglutinate and render innocuous the bacteria in fermenting worts (see also p. 141). Thus, after elimination of the bacteria with the first scums, purer yeast can be collected and washed with pure water or, better, with water containing a little hydrofluoric acid or ammonium fluoride (5 to 10 grms. per hectolitre), which attacks the bacteria, but not the yeast. It cannot, however, be^denied that, in general, washing produces considerable weakening of yeast, which can be reinvigorated by preliminaryjgrowth in sterilised, unhopped wort. 174 ORGANIC CHEMISTRY During the primary fermentation, a considerable quantity of heat is evolved, and to prevent the temperature exceeding 22 to 25 in top or 7 to 8 in bottom fermentation, attemperating coils,, through which cold water (top) or brine (bottom) passes, are used to cool the fermenting wort (F, Fig. 167). Each fermenting vat is provided with a slate, &c., on which are noted, each day, the temperature and the specific gravity of the wort ; the attenuation should reach 58 to 62 per cent, in the primary fermentation and 70 to 75 per cent, in the secondary fermentation, in order that the beers may keep in the warmer rooms of the consumers. 1 When the vigorous fermentation is ended, the head falls and almost disappears, carrying to the bottom of the wort the suspended yeast ; in this way the secondary fermentation is started, this being allowed to proceed for 15 to 20 days in the trade casks placed in cellars at 10 to 12 (for top fermentation) ; the beer is then cleared, filtered, and sold. In bottom fermentation, on the other hand, the secondary fermentation is completed in large tuns pitched inside (-see later) ; these are not quite filled and are kept for 1 to 3 months in cellars maintained continually at to 2, where the beer acquires the desired attentuation and its characteristic flavour. The yeast which is deposited in the fermenting vessels can be collected, pressed (p. 125) and sold to bakers or small brewers. In some breweries the carbon dioxide is now drawn off from the fermenting vats, which are fitted with covers, by pumps and, after being passed through potassium permanganate solution to purify it, the gas is then liquefied (see vol. i, p. 382) ; it can be either utilised in the brewery itself or sold. The fermenting vessels and the storage casks are constructed of oak or pitch-pine. The use of glass vats has been proposed, as these retain the pure flavour of the beer ; such a vat to hold 42 hectols. costs about 40. The cellars have walls and floor of concrete (1 metre higher than the first aqueous border of the subsoil) so that they can be washed when necessary ; the roof is of brickwork. These cellars are furnished with draughts to remove the carbon dioxide, with double doors (always on the north side) to prevent the 1 Determinatiort of the Attenuation and of the Apparent and Real Extracts of Beer. The apparent extract is deduced from the density of the well-shaken (to remove CO 2 ) beer and the corresponding number of degrees Balling (see p. 167). The real extract is deduced from the specific gravity (and Balling's tables) of the beer freed from alcohol by evaporating it to one-third of its volume and making the residue up to the original volume. The original extract of the wort may he calculated with moderate accuracy by adding to the real extract the amount of alcohol (determined as in wine, p. 147) multiplied by 1-92. The degree of real attenuation (A) is referred to 1 hectolitre of wort and indicates how many parts per 100 of the extract of the wort are transformed into alcohol and carbon dioxide : it is obtained by means of the following formula : A = ^^ X 100 where D represents the percentage of extract in the wort and d the percentage of real e.rtract of the beer. In practice, the percentage of extract is sometimes replaced by the degrees Balling, but the results thus obtained are not very exact. If we make D = 15 Balling and d = 5, the real attenuation becomes : 15-5 A = x 100 = 66-66 per cent. 15 But it cannot be denied that Balling degrees refer to kilos of sugar or of extract in 100 kilos of solution, so that a wort showing 15 Balling (sp. gr. 1-0615) contains 15 kilos of extract per 100 kilos of wort, or 15-922 kilos (i.e. 15 X 1-0615) in a hectolitre of wort ; the beer, free from alcohol, showing 5 Balling, has a sp. gr. 1-020, and 1 hectolitre contains 5-100 kilos of extract, so that 10-822 kilos of extract have been fermented and the true 10-822 attenuation is TcTooo x ^ ~ ^'* " Practical brewers find it more convenient, in considering the degree of attenuation of a wort, to calculate the D-d degree of apparent attenuation (A') from the apparent extract of the beer d bymeans of the formula, A = x 100; for example, a wort of 16 Balling has the sp. gr. 1-0658 and 1 hectolitre contains 17-05 kilos of extract, while the beer, with 7 Balling of apparent extract, has the sp. gr. 1-0281, corresponding with 7-20 kilos of extract per 17.05 _ 7-20 hectolitre. The apparent attenuation is hence - -r-rr - X 100 = 57-9 per cent, per hectolitre. L i *Ui> The attenuation can be deduced in a rather less exact manner if instead of degrees Balling are used degrees of the legal densimeter (i.e. the figures in the second decimal place of the specific gravity, a value of 1-063 for the latter thus corresponding with 6-3 on the legal densimeter). In the above example, 16 Balling corresponds with sp. gr. 1-0658, hence with 6-58 on the densimeter ; similarly, 7 Balliog corresponds with 2-81 densimeter degrees. Hence the apparent attenuation is given by : A' = - - X 100 = 57-3 per cent. 6-58 which differs little from the value calculated above from the degrees Balling, and is sufficiently exact for practical purposes. Hence, both for real and apparent attenuation, Balling's tables can be dispensed with, it being sufficient to determine the specific gravity. It should be noted that the lesjal density expresses the weight of wort contained in the volume occupied hy 1 kilo of water measured t 17-5, NATHAN-BOLZE PROCESS 175 entry of warm air from outside and with electric lighting so that windows, which dissipate the cold may be avoided. The vats and casks are raised 50 to 60 cm. from the ground and are inclined slightly forward so that they can be emptied completely and easily cleaned from outside. Along the ceiling run pipes for the circulation of cold brine (bottom fermentation), which maintain a temperature below 60 in the fermentation cellars and one of to 2 in the lager cellars. Ten or fifteen days before the beer is run off from the lager vessels which have been several times filled up to avoid contact of the beer with the air and consequent danger from acetic ferments the bung-hole is tightly closed so as to supersaturate the beer under slight pressure with carbon dioxide, which is still developed more or less feebly according to the state of ma- turity of the beer. If a beer contains, say, O'l to 0-2 per cent, of C0 2 before the bung- hole is closed, it will sub- sequently contain six or seven times that proportion. Nathan-Bolze Rapid Process (Ger. Pat. 135,539, 1900). This process was tested on an industrial scale in 1904 in the Fermentation Institute at Berlin, and gave satisfac- tory results. But the applica- tion of the process has not progressed as rapidly as was hoped for a process which allows of mature beer being prepared in 8 or 10 days, and works under conditions of sterilisation formerly attain- able only in the laboratory or in the manufacture of spirit by the amylo -process (p. 129). The hot, sterile wort from the copper passes into a large hermetically sealed, sterile vessel of enamelled iron (a special resistant enamel being employed) surrounded by an F IG> igg. iron jacket through which water can be passed. These vessels have a capacity of 125 hectols. or more and are called Hansena vessels. They are provided with powerful stirrers (Fig. 168), which keep the wort in continual motion during the fermentation and thus accelerate the transforma- tion of the maltose into alcohol and carbon dioxide. After the temperature of the wort has been lowered to 50 by passing water through the jacket and the diminution of pressure (owing to the condensation of steam) compen sated by the admission of sterilised air, the latter (which has served also to aerate th wort) is replaced by carbon dioxide, the cooling being continued to 10. The pure yeast is then introduced through suitable pipes, the mass being slightly stirred at intervals of an hour. The gas developed is removed in order to hasten the fermentation, and is washed with permanganate, part of it then being compressed (see p. 174). The carbon dioxide which is not compressed is utilised to remove the new beer flavour from beer already fermented in the Hansena vessels ; the gas is passed in at the bottom (after removal of the yeast sediment) at the ordinary temperature, the mass being continually stirred 176 ORGANIC CHEMISTRY meanwhile, it being the carbon dioxide which effects the elimination from the boor of the volatile products to which the disagreeable taste and odour of new beer are due. The gas issues from the top of the vessel, passes to the purifiers and is again conducted through the beer, this process being continued for 10 hours on end. The primary fermentation is finished in less than 3 days, and, after the passage of gas through the beer is completed, the temperature is lowered to and the beer saturated for 24 hours with slightly com- pressed carbon dioxide. The beer is finally filtered and delivered to the trade casks, where it keeps well even in the hot weather. Such a process, simple, rapid, and economical (the cost of the beer being diminished by about 2s 6d. per hectolitre), although it does not give a very delicate flavoured beer, should be suitable to hot countries and to small breweries. Several European breweries already work on these lines and recently (1907) one has been constructed at Milan to employ a modification of the Nathan patent, consisting of a system intermediate to the old process with open fermenting vessels and that devised by Nathan ; in this case enamel] ed iron vessels are used both for the primary fermentation and for the maturation (3 to 4 weeks). These vessels cost about 1 for each hectolitre of capacity. If to the Nathan process is added the Meura system of mashing (1891) which has rendered the preparation of the wort as simple as possible by mashing the finely ground malt in a horizontal cylinder fitted with stirrers so that the mash can be rapidly cooled or heated and wort ready for passing to the filter-press and thence to the copper can be obtained in an hour it will be understood how the manufacture of ordinary beer has been shorn of those practical and theoretical difficulties long regarded as insurmount- able. RACKING OF BEER. Beer is delivered to the consumer in bottles and in casks, and should be perfectly bright, cold, and super- saturated with carbon dioxide. To render it bright, the old method of clarification with gelatine or of filtration through bags has now FIG. 169. been largely replaced by the use of the filter- press, which acts more rapidly and yields brilliant beer. The filtration is carried out in suitable frames through filter-cloths or, better, through finely divided cellulose (such as is used in paper-making) under a pressure of about half an atmosphere. These filter-presses are the same in principle as, and little different in form from, those which are used for the filtration of saccharine liquids and are described in the section on Sugar. (In England, beer in cask is clarified by mixing with the beer a small quantity of finings, which consist of isinglass " cut " or dissolved in an acid, such as tartaric, sulphurous, &c. ; these finings are gradually deposited on the bottom of the cask and carry down with them any suspended protein substances, hop- resins, &c.). Bottling is to-day carried out with all the care employed in the preparation of sparkling wines. A few lines may be devoted to the preparation of beer-casks, since the methods employed are peculiar to the brewing industry. In order that beer for retail consumption may retain its flavour, it must be kept cool and saturated with carbon dioxide up to the moment when it is drawn off into the customers' glasses, and for this purpose the use of liquid carbon dioxide with the arrangement shown in vol. i, p. 389, is well adapted. RESINING OR PITCHING OF CASKS. The keeping of beer sound depends largely on the cleanliness of its surroundings and of the vessels in which it is stored. Hence the casks, returned empty from the customers, are first well scrubbed and washed both inside and outside with water under pressure by means of automatic plant (Fig. 169), and are then disinfected by means of formalin vapour or other antiseptics, or, better still, by pitching the internal surface with natural or artificial resins, which should be transparent and have a melting-point of about 50 ; in this process, which was first used in Bavaria, and is nowadays largely employed all over the Continent, aromatic resins are no longer used, mixtures of colophony with other residues from the distillation of turpentine being prepared by fusion and then rendered more elastic by the addition of resin oil (10 per cent.). To free the casks from the old resin and coat them again every time they are PASTEURISATION 177 returned to the brewery, they are heated inside by means of air supplied from a Roots blower, B (Fig. 170), and heated bypassing through red-hot coke, the hot air being forced into the casks through the tubes, D, for 5 minutes. The old pitch is discharged and the new pitch (about 200 to 250 grms. per hectolitre), fused and heated to 250, introduced into the sterile cask. The bung-hole is then closed, the cask rotated automatically for a FIG. 170. few minutes, the excess of pitch poured out, and the rolling of the cask continued until it is cold. The lager -vessels used for the maturation of the beer are treated in a similar way. PASTEURISATION. Beer, more than wine, is subject to numerous changes and diseases (turbidity due to inferior materials, incomplete saccharification or excess of proteins ; acidity caused by acetic or lactic acid ; stinking fermentation produced by FIG. 171. various bacteria, &c.), and it is difficult to remedy these inconveniences except by improve- ment in the methods of working. In order that beer may remain unchanged when kept for a long time in bottle or when sent to hot places, it is advisable to pasteurise it. The bottles are tightly stoppered and placed in vessels containing cold water, which is then gradually heated to a maximum of 60 to 65, this temperature being maintained for 10 minutes ; the vessels should be covered so as to avoid danger from breakages. The water-bath is subsequently allowed to cool slowly to the ordinary temperature. Top- fermentation beers are rarely pasteurised, as they sometimes acquire an unpleasant flavour II 12 178 ORGANIC CHEMISTRY under this treatment ; bottom-fermentation beers, however, undergo no change and keep good even for ten years. In large breweries, very efficient pasteurising apparatus is employed, the bottles being moved automatically in suitable vessels in which the water moves in the opposite direction. Of the many improved forms in use at the present time, the Gasquet circular type is shown in Pig. 171. Here the chambers are filled successively with baskets of bottles, which are raised by suitable cranes. The water, at a gradually increasing temperature, is drawn from each chamber by means of a tube communicating with a pump, heated by a central thermo-syphon, and then passed on to the succeeding chamber. A bell rings every five minutes as a signal for the bottles of a cool chamber to be removed and replaced by fresh ones. The bottles are made of a special glass, which diminishes the proportion of breakages to less than 1 per cent. ALCOHOL-FREE BEER. A proposal has recently been made to manufacture beer containing no alcohol by treating wort directly at with yeast which has previously been, subjected to special treatment effecting the destruction of almost all of the zymase but not that of the peptase and other proteolytic enzymes ; the carbohydrates hence give no alcohol, the proteins alone being decomposed. These yeasts remove the flavour of fresh wort, the beer being used before alcoholic fermentation begins (Ger Pat. 180,128). COMPOSITION AND ANALYSIS OF BEER. The most varied types of beer are found in different countries, and of each type there are usually the two qualities pale and dark. 1 The density varies from 1-010 to 1-030, and the amount of alcohol usually from 3-5 to 4-5 per cent, by volume, although export beers often contain 5 to 5-5 per cent, of alcohol, and certain special beers still more. The amount of extract also varies considerably, being as 1 The compositions of some of the best-known beers are as follow : Alcohol Extract Ash Real attenuation per cent, by vol. per cent, by vol. per cent, by vol. per cent, by vol. Pale Berlin beer .... 3-91 4-85 0-14 60-50 Berlin lager beer ..... . 4-00 6-15 0-20 54-70 Export Bavarian beer . . . . . 4-78 1067 0-29 45-44 Munich Spaten beer (at Munich) . 3-23 6-61 0-28 48-40 ,, ,, (at Milan) . . 5-23 Salvator beer' . . . 4-64 9-08 0-28 49-00 Spaten table beer . . . 7-0 10-35 57-40 Bock ..... 4-20 7-10 54-20 white beer .... 3-51 4-73 59-58 Vienna lager beer .... 3-62 6-01 54-50 Pilsen beer ...... 3-47 4-97 59-00 North of France beer .... 3-20 4-04 61-20 Amsterdam beer ..... 4-30 7-0 36-40 Brussels Iambic ..... 5-94 3-30 78-00 Belgian faro ..... 4-33 5-1 62-80 Bass's pale ale ..... 6-15 6-87 64-00 Scotch pale ale ..... 8-50 10-90 59-9 Dublin stout ..... 7-23 6-15 70-64 London porter .:.... 5-40 6-00 ' 63-3 American beer ..... 5-89 6-45 63-15 Milan beer : Pilsen type ... 3-92 5-43 0-21 57-91 ,, Munich type ... 3-50 5-58 0-20 54-63 Porretti beer (Varese) .... 3-98 5-66 0-22 57-45 Italia beer (made at Milan by the modified Nathan- Bolze process) ....... 4-78 6-00 0-22 59-43 The real attenuation (or degree of fermentation, see p. 174) is calculated by multiplying the percentage of nlcohol by 1-92 (= d'), and adding to this product the extract of the beer, d ; this gives the extract, D, contained in the D d wort prior to fermentation and then the attenuation or percentage of extract fermented = X 100. Some English breweries make stout from a mixture of 65 per cent, of pale malt, 10 per cent, of black malt (for colour), 10 por cent, of caramelised malt and sometimes 10 per cent, of cane-sugar and 5 per cent, of maize. This very dark beer is attenuated to a relatively small extent, and retains a full, sweet taste, this being partly due to the almost entire absence of gypsum in and the small total hardness of London water ; these beers also contain few hops. Export stout is made from worts having gravities as high ag 25 Balling, whilst porter is lighter jn character, J'he pale beers of Berlin are made with a gggd proportion (75 per 00nt-) Q* malted wheat, BEER STATISTICS 179 much as 12 per cent, for certain types of beer ; for ordinary beers it lies between 5 and 6 per cent. (1 per cent, being maltose). The proportion of ash is generally less than 0-3 per cent. The amount of carbon dioxide dissolved varies from 0-15 to 0-40 per cent. The analysis qf beer is carried out in a similar manner to that of wine (p. 157), but the carbon dioxide is eliminated by heating the beer to 40 and shaking for several minutes before the specific gravity and acidity are determined ; the latter does not exceed 0-3 per cent, and is expressed as lactio acid (1 c.c. N/10-alkali = 0-009 grm. lactic acid) or as cubic centimetres of normal alkali used per 100 c.c. of beer. To avoid frothing during the distillation of the alcohol, 1 a little tannin is added. The nitrogenous substances are deter- mined on the extract of 40 c.c. of the beer by Kjeldahl's method (p. 10), the proportion of nitrogen being multiplied by 6-25 to give the corresponding amount of proteins. The reducing sugar is determined by means of Fehling's solution and is calculated as maltose (see Note, p. 167). 2 STATISTICS. In Italy the brewing industry has never been in a flourishing condition, owing to the abundance and cheapness of wine possibly more commonly drunk than water. The beer manufactured from remote epochs in Italy was made by the top -fermentation process and was of poor quality ; it did not keep well in summer, was stored carelessly by the retailers and was consumed for only about a couple of months in the year close to where it was produced. Technical improvements have been introduced tardily, but nowadays the industry is largely concentrated into a few large breweries using the most modern methods and controlled by technical experts from other countries. About one -half of the beer imported into Italy is supplied by Austria -Hungary, about one -third by Germany, and one -tenth by Switzerland : PRODUCTION, IMPORTATION, AND CONSUMPTION OF BEER IN ITALY Consumption Production Imports Total Per head hectols. hectols. in cask hectols. litres 1880 . ". 116,000 . . 46,900 . . 163,000 . . 0-57 1890 . . 160,900 . . 99,500 . . 260,000 . . 0-86 1894-95 . . 95,500 .. 60,000 .. 156,000 .. 0-50 1900 . . 154,000 . . 54,750 . . 209,000 . . 0-66 1903 .' . 185,000 . . 70,000 . . 255,000 . . 0-79 1904 . . 220,000 . . 80,000 . . 300,000 . . 0-92 1905-06 . . 304,000 . . 90,000 . . 394,000 . . 1-20 1906-07 . . 360,000 .. 94,494 .. 455,000 .. 1-50 1907-08 . . 400,000 .. 95,213 .. 495,000 .. 1-60 1908-09 . . 473,000 .. 88,100 .. 561,000 .. 1-80 1909-10 . . 563,000 . . 89,737 . . 651,000 . . 2-00 1 The proportion of alcohol can be calculated indirectly by means of the formula, A = (s/S) -4- 8, where A indicates the percentage of alcohol, s the specific gravity of the beer, S the specific gravity of the beer freed from alcohol and made up to the original volume ; the alcohol Table (p. 148) gives the percentage by weight corresponding to the value of s/S and division of this percentage by S gives the true percentage of alcohol. 8 The determination of sulphurous add (only traces are allowed in beer) derived from sulphites or sulphurous acid added to preserve the beer, is effected by distilling 200 c.c. of the beer, previously acidified with 5 c.c. of syrupy phosphoric acid, in a current of carbon dioxide and passing the distillate through 50 c.c. of iodine solution (5 grms. I + 7-5 grms. KI made up to 1 litre with water) ; the iodine solution is then acidified with hydrochloric acid, boiled to expel excess of iodine and precipitated with barium chloride, the filtered, washed, and ignited barium sulphate being weighed ; multiplication of this weight by 1-372 gives the amount of SO 2 per litre of beer. For the detection of boric and, 100 c.c. of beer are evaporated to dryness and the residue calcined ; a little sulphuric acid and alcohol are then added to the resulting ash and the mixture ignited and stirred ; the appearance of a green colour at the edges of the flame indicates the presence of boric acid. The quantitative determination of boric acid is difficult and is only rarely carried out, Rosenbladt and Gooch's method being then used. For the detection of fluorides, sometimes (although prohibited) added as preservative, 100 c.c. of the beer, rendered alkaline with ammonium carbonate, are boiled, mixed with 3 to 4 c.c. of calcium chloride solution, boiled again for 5 minutes and filtered, the residue being washed and calcined in a platinum crucible. One cubic centi- metre of concentrated sulphuric acid is then added and the crucible, covered with a watch-glass partly coated with paraffin wax, gently heated. 1 In presence of fluorides, the glass is attacked in the unprotected parts. The degree of attenuation or of fermentation is calculated as indicated in the Note on the preceding page. Adulteration with salicylic acid is detected by acidifying 100 c.c. of the beer with 5 c.c. of hydrochloric acid and shaking with 50 c.c. of ether and 50 c.c. of light petroleum. The ethereal solution is separated and evaporated to dryness, the residue being taken up in water and filtered. If the liquid gives a violet coloration with a little dilute ferric chloride solution and a red one with MiKon'9 mgent (aqueous mercuric nitrate containing a little nitrous acid), the presence of salicylic acid is certain, Saccharin \ determiped by evaporating an. ethereal fjrfrapt obtained as abpve, dissolving fte resjdup i s a Jltyfe 180 ORGANIC CHEMISTRY The consumption of beer in Italy takes place mostly in the towns of the north and centre, and the average consumption per head in Milan, Turin, or Rome is at least ten times that for the whole country. The production of beer in Japan was 362,000 hectols. in 1907 ; 294,100 in 1908 ; 271,500 in 1909, and 280,000 in 1910. The production of beer in other countries in 1900 was as follows : Germany, 67,000,000 hectols. or 118 litres (in 1907, 70, and in 1910, 64 litres) per head. England, 59,000,000 hectols. (57,000,000 or 150 litres per head in 1909). Austria -Hungary, 20,000,000 hectols. (72 litres per head) or 19,000,000 in 1909. Belgium, 14,000,000 hectols. (213 litres per head). France 9,000,000 hectols. (25 litres per head ; but here, too, the consumption is localised, the annual consumption per head in Lille being 360 litres) ; in 1909 France produced 11,000,000 hectols. The United States, 48,000,000 hectols. (63 litres per head) in 1900 and 70,000,000 in 1909. Spain, about 1,000,000 hectols., and Russia, 6,200,000 hectols. in 1909. In 1900, Germany, with 10,000 breweries, produced twice as much beer as in 1880, and in 1885 exported 1,500,000 hectols. One large brewery in Germany makes more beer than the whole of Italy consumes. (Italy has 93 breweries at the present time.) In 1881, England produced 45,000,000 hectols. ; Austria -Hungary, 12,000,000 ; Belgium 9,000,000 ; France, 8,000,000 ; Switzerland, 1,000,000 (now 1,500,000), and the United States, 19,000,000. The world's production of beer in 1910-11 was 271,000,000 hectols. In Italy the brewing tax was 5fd. up to 1891, when it was raised to ll^d. (causing a temporary diminution in the consumption at that time) per saccharometer degree per hectolitre, measured with the decimal saccharometer at 17'5 on the wort from the cooler, an allowance of 12 per cent, being made for loss during the subsequent operations ; the tax varied from a minimum of 115d. to a maximum of 184d. per hectolitre, according to the strength of the beer. Imported beer pays 29d. more, or the importers can demand the tax to be levied on the extract degrees, these being increased by twice the number of alcohol degrees. The exchequer collected 180,000 in 1905-6 and 211,800 in 1906-1907 as tax of manufacture. In Germany beer costs about 12s. per hectolitre, or rather more with the extra taxation of 1910. In Italy the cost is about 32s. (that imported from well-known breweries about 40s. per hectolitre). ALCOHOLS HIGHER THAN ETHYL PROPYL ALCOHOLS, C 3 H 8 O. The two isomerides theoretically possible are known : (1) Normal, CH 3 CH 2 CH 2 OH (propanol-1 or ethylcarbinol). This can be obtained from fusel oil (p. 122) by fractional distillation or from its bromo -derivative. It has an agreeable odour, b.pt. 97, sp. gr. 0-804, and is readily soluble in water. On oxidation it gives propionic acid, which proves its constitution. (2) Sec. or Iso-Propyl Alcohol, CH 3 -CH( OH)- CH 3 (propanol-2 or dimethylcarbinol), is a colourless liquid, b.pt. 81, sp. gr. 0-789. It is obtained from isopropyl iodide and hence indirectly from glycerol, or by reducing acetone with sodium amalgam, the constitution attributed to it being thereby confirmed. BUTYL ALCOHOLS, C 4 H 10 O. The four isomerides, predicted by theory, are known : (1) Normal Butyl Alcohol, CH 3 -CH 2 - CH 2 -CH 2 - OH (l>utanol-\ or propylcarbinol), is a liquid, b.pt. 117, sp. gr. 0-810, and has an irritating odour ; 12 vols. of water at 22 dissolve only 1 vol. of it, this being separated from the solution by the addition of a soluble salt. sodium carbonate solution, evaporating in a silver dish and fusing the residue with solid caustic soda ; the white mass is dissolved in water, the solution acidified with hydrochloric acid, and the sulphuric acid (derived from the sulphonic group of the saccharin) precipitated quantitatively as barium sulphate. The weight of the latter, multiplied by 0-785, gives the weight of saccharin. Caramel, rjdded to colour the beer is recognised by shaking 20 c.c. with about 30 to 40 grms. (i.e. until saturated) of solid sodium sulphate and 60 c.c. of 05 per cent, alcohol. If the lower liquid is markedly coloured and forms a greenish brown deposit, the presence of caramel is indicated ; beer .containing no caramel becomes decolorised and gives only a greenish or dark greenish brown deposit if it contains coloured malt. Picric acid is detected by evaporating a litre of the beer to a syrupy consistency, extracting with boiling absolute alcohol, filtering and evaporating the alcoholic liquid, dissolving the residue in water, adding a few drops of hydro- chloric acid and heating for an hour with a few strands of wool ; if the latter are coloured yellow, picric acid is present. Extraneous bitter substances are tested for by evaporating 2 litres of beer to half its volume and precipitating the residue in the hot with lead acetate ; the hot liquid is filtered rapidly and the lead then precipitated with ammonium sulphate and filtered off. The filtrate should have no bitter taste. BUTYL, AMYL, AND HIGHER ALCOHOLS 181 It is found in fusel oil and can be obtained by fermenting glycerol or mannitol (yield 8 to 10 per cent.) with Bacillus butylicus (contained in the excreta of cows). It can also be prepared synthetically by the various general processes (p. 104). Its constitution is indicated by its syntheses and by the possibility of transforming it into normal butyric acid by oxidation. (2) Secondary Butyl Alcohol, CH 3 CH 2 CH(OH) CH 3 (butanol-2 or ethylmethylcarbinol) is a liquid with an intense, peculiar odour, b.pt. 100, sp. gr. 0-808. It can be obtained by treating the tetrahydric alcohol, erythritol, C 4 H 6 (OH) 4 , with hydriodic acid or by the interaction of normal butylene and hydriodic acid and hydrolysis of the resulting iodide. /ITT (3) Isobutyl Alcohol, ^^^CH CH 2 OH (methylpropanol), is termed also butyl alcohol of fermentation, since it abounds in the fusel oil of potatoes, from which it can be extracted by forming the corresponding iodo -compound. It is a colourless liquid, b.pt. 107, sp. gr. 0-806, and has a characteristic alcoholic smell. Its constitution is determined by the fact that, on oxidation, it yields isobutyric acid, the constitution of which is known. C*TT (4) Tertiary Butyl Alcohol, r , T4 - 3 >C(OH)-CH 3 (trimethylcarbinol or methyl-2-propanol), Or 3 occurs in small proportion in fusel oil, and can be prepared by the action of hot 75 per cent. sulphuric acid on isobutylene, which thus takes up 1 mol. of water. When pure, it forms rhombic prisms or plates, m.pt. 25-5, sp. gr. 0-786 (solid), b.pt. 83. On oxidation it gives acetic acid, acetone, and carbon dioxide. AMYL ALCOHOLS, C 6 H U .OH. The eight isomerides theoretically possible are known, the most important being : (1) Normal Amyl Alcohol, CH 3 - CH 2 CH 2 - CH 2 CH 2 - OH (pentanol-l), b.pt. 138, sp. gr. 0-817, is of little importance, and is obtained by reducing normal valeraldehyde or by the other general methods. (2) Amyl Alcohol of Fermentation, 3 >CH-CH 2 - CH 2 - OH (methyl-3-butanol-I or L>1 3 isobutylcarbinol), is a liquid, b.pt. 130, sp. gr. 0-810, and is solid at 134. It imparts its characteristic smell and burning taste to fusel oil, in which it abounds. It is to this alcohol that the poisoning effect of spirits is principally due. It occurs naturally in Roman chamomile oil. (3) Active Amyl Alcohol, /Lr^CH CH 2 OH (methyl-2-butanol-l or 2-methylbutan- L>rL 3 l-ol), boils at 128, has the sp. gr. 0-816, and is found with the amyl alcohol of fermentation. It contains an asymmetric carbon atom (see p. 19) and is laevo -rotatory, whilst the halogen compounds and the valeric acid derived from it are dextro-rotatory ; also the dextro- isomeride of this acid yields a laevo -rotatory iodide. PTT (4) Tertiary Amyl Alcohol, ^TT 3 >C( OH)- CH 2 -CH 3 (methyl-2 -butanol-2 or amylene CM 3 hydrate or dimethylethylcarbinol) is an oily liquid with a faint odour of mint. It boils at 102 and is prepared from amylene by the indirect addition of water under the influence of sulphuric acid. It exerts a soporific action. HIGHER ALCOHOLS. Of these may be mentioned : Primary normal hexyl alcohol or hexanol, CH 3 - [CH 2 ] 4 -CH 2 -OH (14 of the 18 hexyl alcohols predicted by theory are known), can be obtained from caproic acid, C 6 H 12 2 , and is found as butyric and acetic esters in the ethereal oil of the seeds of Heracleum giganteum and in the fruit of Heracleum spondylium : it boils at 158 (under 740 mm. pressure), and has a specific gravity of 0-820. Caproyl or isohexyl alcohol, (CH 3 ) 2 : CH CH 2 CH 2 CH 2 OH, b.pt. 150, is f ound in vinasse and in fusel oil. Heptyl (or oznanthyl) alcohol, C 7 H 16 O ; of the ^38 possible isomerides, 13 are known. Normal octyl alcohol, C 8 H 18 O, is contained in Heracleum spondylium and H-eracleum giganteum ; secondary octyl alcohol (or capryl alcohol or methylhexylcarbinol) is formed on distilling castor oil. Other higher alcohols are obtained by reducing the corre- sponding aldehydes with zinc dust and acetic acid ; they are almost solid, like paraffin wax. Cetyl or normal hexadecyl alcohol, C^H^O, combined with palmitic acid, forms the principal component of sperm oil. Ceryl alcohol (cerotin), C 26 H 53 OH, occurs as cerotic ester in Chinese wax and in wool-fat ; it melts at 76 to 79. Melissyl or myricyl alcohol, C 30 H 61 OH, is found as the palmitic ester in beeswax and carnauba wax and is obtained free by saponi- fication with alcoholic potash. 182 II. UNSATURATED MONOHYDRIC ALCOHOLS These are similar to the saturated alcohols, but, as they contain one or .two double linkings, they behave like the olefines and diolefines in taking up two or four atoms of hydrogen, halogens, &c., to give saturated compounds. If they contain a triple linking, C ^ CH, they form explosive metallic compounds, as does acetylene (p. 91 ). VINYL ALCOHOL, CH 2 : CH-OH (Ethenol), appears to be present in commercial ether, but it has never been isolated, attempts to synthesise it leading, as is the case with other similar compounds, to an isomeride acetaldehyde, CH 3 -CHO ; the formation of the latter is explained by the addition of a molecule of water to the alcohol, and immediate loss of a molecule of water from the compound thus formed. ALLYL ALCOHOL, CH 2 : CH-CH 2 -OH (Propenol), is a liquid of pungent odour, b.pt. 97, and readily soluble in water. It is formed in small quantity in the distillation of wood, but is more easily obtained by heating glycerol at 26 with oxalic acid and a little ammonium chloride. Cl, Br, CN, and HC10 can be added on to it directly, but not H. When cautiously oxidised, it takes up O and H 2 O, giving glycerol or even acrolein (ally! aldehyde) and acrylic acid, which shows it to be a primary alcohol. CITRONELLOL, C 10 H 2 oO, is found in attar of roses. PROPARGYL ALCOHOL, CH C-CH 2 -OH (Propinol), is a liquid with a pleasant odour, lighter than water, b.pt. 114. GERANIOL, C 10 H l8 Oor(CH 3 ) 2 : C : CH-CH 2 -CH 2 -C(CH 3 ) : CH-CH 2 -OH, is a pleasant- smelling oil, b.pt. 121 under 17 mm. pressure. It is obtained from geranium oil, and on oxidation gives citral (the corresponding aldehyde) which occurs in mandarin oil and in essences of orange and lemon and to a very considerable extent (60 per cent.) in verbena oil. III. POLYHYDRIC ALCOHOLS (a) DIHYDRIC ALCOHOLS OR GLYCOLS, C W H 2 ,XOH) 2 Substitution of two hydrogen atoms joined to different carbon atoms by two hydroxyl groups gives dihydric alcohols, containing two alcoholic groups. It is not, however, possible to have two hydroxyl groups united to the same carbon atom although similar compounds are known for the ether derivatives known as Acetals (see later) since even if they could be formed they would immediately lose a molecule of water, forming aldehydes or ketones. The dihydric alcohols, owing to their sweet taste, were called Glycols by Wurtz, who prepared them by transforming a dihalogenated hydrocarbon into the corresponding diacetyl-ester by means of silver acetate and then saponi- fying the diacetyl compound either by baryta or sodium hydroxide or by boiling with water and lead oxide or sodium carbonate solution : CH 2 Br CH 2 .0-COCH 3 1 +2CH 3 -COOAg = 2AgBr+ I CH 2 Br CH 2 -0-COCH 3 Ethylene bromide Diacetylglycol CH 2 .0-COCH 3 CH 2 -OH + 2KOH = 2CH 3 -COOK + I (glycol) CH 2 -0-COCH 3 CH 2 -OH A special group of glycols, the pinacones, containing two adjacent tertiary alcohol groups (=C-OH), are formed by reducing the ketones with sodium and water, or, better, together with isopropyl alcohol, by electrolysing a dilute solution of sulphuric acid and acetone, the latter being reduced at the negative pole : CH 3 -C(OH)-CH 3 3CH 3 CO CH 3 + H 4 = CH 3 CH(OH) CH 3 + | CH 3 -C(OH)-CH 3 POLYHYDRIC ALCOHOLS 183 this pinacone (2 : 3-dimethyl-2 : 3-butandiol), melts at 38, boils at 172 and crystallises with 6H 2 O. When distilled with dilute sulphuric acid, it is transformed into pinacotine, (CH 3 ) 3 C-CO- CH 3 , with separation of H 2 and transposition of an alkyl group. The glycols have an almost oily appearance ; their solubility and sweetness increase with the molecular weight ; the specific gravity and boiling-point are much higher than those of the monohydric alcohols with equal numbers of carbon atoms. The hydroxyl groups of the glycols behave like those of monohydric alcohols, so that the glycols can give rise to ethers and esters, alkoxides (sodium, &c.), halogen compounds (e.g. the chlorohydrins), aldehydes and acids, besides which they may give up 1 mol. of H 2 forming anhydrides. ETHYLENE GLYCOL (Ethan-1 : 2-diol), C 2 H 4 (OH 2 ), is a dense liquid, b.pt. 198, and, on oxidation, yields glycollic acid, CO 2 H CH 2 OH and oxalic acid, C0 2 H C0 2 H. PROPYLENE GLYCOLS. Two isomerides are known : a-Propylene Glycol, OH.CH 2 -CH(OH)-CH 3 (propan-1 -. 2-diol), boils at 188 and is formed in the distillation of glycerolwith sodium hydroxide. It contains an asymmetric carbon atom and, by the action of certain ferments, the Isevo -rotatory isomeride can be isolated. /3 -Propylene Glycol boils at 216 and is formed by the bacterial decomposition of glycerol, as well as by the usual synthetical methods. In the higher glycols, when the two hydroxyl groups have four carbon atoms between them (y-glycols), water is readily separated and furan derivatives, analogous with pyrrole and thiophene compounds, formed. (6) TRIHYDRIC ALCOHOLS, C,,H 2w . 1 (OH) 3 These are colourless, dense liquids with a sweetish taste and readily soluble in water ; they contain at least three carbon atoms and three hydroxyl groups, and are hence capable of forming three series of esters by combination with a monobasic acid. GLYCEROL, C 3 H 5 (OH) 3 , or OH-CH 2 -CH(OH)-CH 2 -OH (Propantriol), was discovered by Scheele in 1779. Chevreul and Braconnot (1817) found it as a component of all oils and fats. Its formula and constitution were estab- lished later (Pelouze, Wurtz, and Berthelot). It occurs abundantly in nature, not in the free state, but combined with higher fatty acids in the form of esters (glycerides), which form the fats and oils ; these contain 9 to 11 per cent, of combined glycerol. It exists free in rancid fats and is formed in small proportions in the fer- mentation of sugar (all wines contain 0-98 to 1-67 per cent.). Industrially glycerol is obtained principally from factories where fats are decomposed (stearine- and soap-works). Synthetically it can be obtained by transform- ing propylene (from isopropyl iodide), by means of chlorine in the hot, into dichloropropane, C 3 H 6 C1 2 , which, with iodine chloride, gives the trichloro- derivative C 3 H 5 C1 3 ; the latter, when heated with water at 170, gives glycerol : CH 2 C1-CHC1-CH 2 C1 + 3H 2 O = 3HC1 + OH-CH 2 -CH(OH)-CH 2 -OH. This formation of glycerol and also that by the oxidation of allyl alcohol, CH 2 : CH-CH 2 -OH, demonstrate the constitution of glycerol. On the other hand, it is possible to prepare glycerol synthetically from the elements by way of acetylene, acetaldehyde (p. 91), acetic acid, acetone (by distillation of calcium acetate), isopropyl alcohol (by reduction), propylene, and thence, as above to glycerol (Friedel and Silva). PROPERTIES. Glycerol (also termed glycerine) is an oily, colourless, dense (sp. gr. 1-265 at 15) liquid, with a sweet taste ; it is very hygroscopic and dissolves in all proportions in water and alcohol, heat being developed on mixing 58 parts of glycerol with 42 parts of water. It is insoluble in ether and chloroform ; it dissolves to the extent of 5 per 184 ORGANIC CHEMISTRY cent, in dry acetone and to a greater degree in aqueous acetone. It boils at 290 with partial decomposition, but it can be distilled unchanged in a vacuum (at 10 mm. pressure it boils at 162). It crystallines at 40 or at a higher temperature if it contains water ; the separated crystals melt only at 22. When heated for a long time at 130 to 160 in presence of sulphuric acid, glycerol loses one or more molecules of water, giving anhydrides or ethers of glycerol or polyglycerines (A. Nobel, 1890) ; W. Will (1904) arrived at the same result by heating glycerol for 7 to 9 hours at 290 to 295 and distilling off the water formed. This treatment yields about 60 per cent, of diglycerol, C 3 H 5 (OH) 2 -0-C 3 H 6 (OH) 2 , and a little tri- and polyglycerols ; all these pro- ducts can be esterified like glycerol and yield, e.g. tetranitrodiglycerine, which does not congeal even at 20 and has an explosive power like trinitroglycerine (see also C. Claessen, Ger. Pats. 181,754 and 198,768, 1907). According to U.S. Pats. 978,443 (1910) and 13,234 (1911), glycerol readily polymerises when heated at 275 in presence of 0-5 to 1-0 per cent, of sodium acetate, 70 per cent, being polymerised in an hour. - When it is heated rapidly and strongly it decomposes, yielding partly acrolein with the characteristic pungent odour. Also when heated with P 2 5 or KHS0 4 , it loses 2H 2 0, giving acrolein, CH 2 : CH- CHO. One hundred parts of glycerol dissolve the following quantities of mineral salts : 98 of sodium carbonate, 60 of borax, 50 of zinc chloride, 40 of potassium iodide, 10 of boric acid, 50 of tannin ; bromine, ammonia, ferric chloride, &c., are also dissolved. Glycerol has the refractive index 1-476 at 13 and in aqueous solution the index varies proportionally with the dilution. By means of Lenz's table, the concentration of glycerol solutions can be determined from either the specific gravity or the index of refraction : Per- centage of glycerol Degrees Bailing, Beck, Gerlach Sp. gr. at 12 to 14 Index of refraction at 12-5 to 12-8 Per- centage of glycerol Degrees Baum, Beck, Gerlach Sp. gr. at 12 to 14 Index of refraction at 12-5 to 12-8 98 30-1 1-2637 1-4729 48 16-2 1-1265 1-3979 96 29-6 1-2584 1-4700 46 15-5 1-1210 1-3950 94 29-1 1-2531 1-4671 44 15-0 1-1155 1-3921 92 28-7 1-2478 1-4642 42 14-3 1-1100 1-3890 90 28-2 1-2425 1-4613 40 13-6 1-1045 1-3860 88 27-7 1-2372 1-4584 38 13-0 1-0989 1-3829 86 27-1 1-2318 1-4555 36 12-3 1-0934 1-3798 84 26-6 1-2265 1-4525 34 11-5 1-0880 1-3772 82 26-1 1-2212 1-4496 32 11-0 1-0825 1-3745 80 25-6 1-2159 1-4467 30 10-3 1-0771 1-3719 78 . 25-1 1-2106 1-4438 28 9-6 1-0716 1-3692 76 24-5 1-2042 1-4409 26 9-0 1-0663 1-3666 74 24-0 1-1999 1-4380 24 8-3 1-0608 1-3639 72 23-5 1-1945 1-4352 22 7-6 1-0553 1-3612 70 23-0 1-1889 1-4321 20 6-9 1-0498 1-3585 68 22-3 1-1826 1-4286 18 6-1 1-0446 1-3559 66 21-6 1-1764 1-4249 16 5-6 1-0398 1-3533 64 21-0 1-1702 1-4213 14 4-9 1-0349 1-3507 62 20-3 1-1640 1-4176 12 3-8 1-0297 1-3480 60 19-8 1-1582 1-4140 10 3-4 1-0245 1-3454 58 19-2 1-1530 1-4114 8 2-8 1-0196 1-3430 56 18-6 1-1480 1-4091 6 2-1 1-0147 1-3405 54 18-0 1-1430 1-4065 4 1-3 1-0098 1-3380 52 17-4 1-1375 1-4036 2 0-7 1-0049 1-3355 50 16-9 1-1320 1-4007 GLYCEROL 185 Glycerol has the interesting property of preventing the precipitation of various metallic hydroxides (i.e. it keeps them dissolved) ; for instance, in presence of glycerol, potassium hydroxide does not precipitate salts of chromium, copper, &c. With alkalis it forms slightly stable soluble alkoxides. It does not reduce silver or cupric salts, and hence cannot contain aldehyde groups ; it is not coloured by concentrated sulphuric acid or by sodium hydroxide on boiling. The halogens act on glycerol, not as substituting, but as oxidising agents. It inverts cane-sugar and renders starch soluble ; 100 parts of glycerol and six of starch at 190 give starch soluble in water, and the starch can be separated from the glycerol, when cold, by precipitation with alcohol. Like the other polyhydric alcohols (glycols, erythritol, and its isomerides, also glucose and its isomerides galactose, &c. but not cane-sugar, quercitol or dextrin) glycerol, when added in sufficient quantity, transforms the alkaline reaction of borax solutions in an acid reaction, thus allowing of the determina- tion of boric acid and borax by titration. Under the action of certain schizomycetes, glycerol yields normal butyl alcohol, butyric acid and, partly, ethyl alcohol. Being a trihydric alcohol, glycerol is able to form esters of three types (mono-, di- and tri-), according as one, two, or three hydroxyl groups are replaced by inorganic or organic acid residues. In this way the glycerides can be regenerated ; for example, when excess of stearic acid is heated with glycerol at 200 under reduced pressure until no more water separates, tristearin is formed. When cautiously oxidised, glycerol forms first glyceric acid, OH-CH 2 - CH(OH)-COOH, which undergoes further oxidation to tartronic acid, COOH-CH(OH)'COOH, so that it is proved that glycerol contains two primary alcohol groups, (-CH 2 -OH); also, as tartronic acid still exhibits alcoholic characters, it must contain a secondary alcohol group. The con- stitution of glycerol is hence completely proved. USES OF GLYCEROL. The majority of the glycerol manufactured is used for the preparation of nitroglycerine and hence of dynamite (see later). It is also used to give body to light wines (termed Scheelisation, after Scheele, the discoverer of glycerol). It is employed in the manufacture of liqueurs, syrups, preserves, and sweetmeats, since it is sweet and dense, and, to some extent, anti-fermentative. It is added to chocolate, tobacco, cosmetics, textiles to be dressed, and leather goods, since it does not dry and keeps them soft or pliable. It is also used in extracting from flowers and herbs delicate perfumes which would undergo change if extracted by distillation. It is employed as a non-congealing and lubricating liquid (a solution of $p. gr. 1-13) in gasmeters ; for greasing iron objects to prevent them from rusting ; for making copying-ink, soap, and shoe-polish ; for preserving anatomical preparations, &c. INDUSTRIAL PREPARATION. Glycerol is almost exclusively obtained as a secondary product in the treatment of fats. Until the year 1885 only the aqueous residues of stearine works were worked up (the fats are decomposed with lime, sulphuric acid, steam, or ferments), but nowadays almost all the alkaline lye of soap factories (where the fats are treated directly with caustic soda and then with salt) 1 are utilised. Of the 9 to 11 per cent, of glycerol contained in fats, 8 to 10 per cent, can be recovered (only 4 per cent, when the decomposition is effected by sulphuric acid, the maximum yield being obtained when water or ferments are used). The treatment of the dilute solutions of crude glycerol varies with their origin : soap- lyes (which are sometimes concentrated in the soap-works and sold to the glycerol refiners) 1 These lyes have an alkaline reaction and, on analysis, one of them gave the following results : water, 61 per cent.; glycerol, 16-5 per cent, salts; 22 per cent, (eight-tenths of which were NaCI, one-tenth Na 2 SO4, and one-thirtieth Na 2 CO,). The specific gravity varies from 3 to 7 B6., and the proportion of glycerol usually from 6 to 12 per cent. 186 ORGANIC CHEMISTRY are treated with 0*1 to 0*2 per cent, of lime or ferrous sulphate and mixed by means of an air-jet ; the liquid is decanted, slightly acidified with hydrochloric acid and skimmed ; a small quantity of aluminium sulphate is then added, the liquid being decanted, rendered slightly alkaline, passed through a filter-press and concentrated in open boilers furnished with stirrers until sodium chloride begins to separate ; subsequent concentration to the sp. gr. 28 Be. is carried out in a vacuum, the salt deposited being gradually removed. This crude glycerol contains 85 to 90 per cent, of glycerol and 1 per cent, of salts, and has a dark yellow or brownish colour. Sometimes the alkali is removed from the soap lyes by adding a little resin and boiling, so that the resin soap formed is carried to the surface and can be decanted (to be utilised by adding to ordinary soap). The free lime may also be pre- cipitated with an oxalate or with carbon dioxide. The concentration is not carried out in open vessels, as, when the aqueous solutions are vigorously boiled, the steam given off carries away appreciable quan- tities of glycerol. The concentration is hence carried to a certain point FIG. 172. in an apparatus (Fig. 172 shows the Droux apparatus and Fig. 173 that of Morane), fitted with rotating coils or hollow discs, in which steam under pressure circu- lates. The apparatus is covered in and the steam from the solution .issues rapidly through a tube communicating with an aspirator. When the density reaches 18 to 20 Be. the solution is decanted or filtered and then further concentrated in a vacuum to 27 to 28 Be. In some cases the glycerol thus obtained, while still boiling, is decolorised by adding animal charcoal and filtering through a filter-press. This glycerol always contains a small quantity of dissolved salts. To purify it, its temperature is raised to 110 to 120 by means of superheated steam, the acids or more volatile products being thus eliminated. It is then distilled with superheated steam at 170 to 180, at which temperature all the pure glycerol passes over. This is rectified in one apparatus to 22 Be. and in a second, under diminished pressure and wit h superheated steam, to 28 Be., at wh:'ch . concentration almost all the salt separate' . The vacuum distillation is sometiirto effected by a triple-effect apparatus (Pick type, see vol. i, p. 453 ; also section c n Sugar), with which it is easy to remove the salt as it separates without interrupt- ing the distillation. These forms of apparatus for purifica- tion and distillation are named after their inventors (Hagemann, Scott, Jobbins, van Ruymbeke, Lehmann, Heckmann, &c.). The Heckmann process consists in distilling the aqueous glycerine, already concentrated to beyond 20 Be., in a boiler, A (Fig. 174), into which steam superheated to 200 to 220 and under half an atmosphere pressure is passed by means of a perforated coil. In order to prevent the scum being carried over with the steam and glycerol, a perforated disc, a, fitted with a vent-pipe is fixed two-thirds of the way up the boiler. The vapours issue by the pipe B, and are condensed in the reservoir, C, which is heated to 80 to 90 with indirect steam circulating in the jacketed bottom, D. Above the reservoir is a rectifying column, with a dephlegmator, K, similar to, but. much lower than, that used for the rectification of alcohol (see p. 136). During the distillation, a slight vacuum is maintained in the whole apparatus by means of a suction pump, V, so that principally water-vapour and only a little glycerol are evolved from the reservoir, C. The glycerol vapour separates in the column and returns to the reservoir, whilst the condenser, M , condenses only the water-vapour, which is controlled FIG. 173. PURIFICATION OF GLYCEROL 187 by its density, colour, and taste in the test-glass, N, and is then collected in the tank, 0. In C the glycerol finally reaches a concentration of 95 to 99 per cent. The rectifying column is sometimes replaced by a series of communicating, vertical copper tubes (Fig. 175) which fractionally condense the glycerol- and water-vapours from the boiler, B (heated partly by d : rect fire), into which passes steam from v, superheated in the furnace, T. By means of the pump, Z, a vacuum is maintained in the whole apparatus, so that, as the distillation proceeds, fresh glycerine from the reservoir, A, can be drawn into the boiler. In the first cylinder or condensing tube, which soon reaches a temperature of 100, almost pure glycerol separates, whilst in the succeeding tubes, cooled only by the surrounding air, more and more dilute glycerine and finally water separate. Below each tube is a horizontal cylinder, these serving to collect the glycerols of different concen- trations, some of which are subjected to redistillation. In this way is obtained the best dynamite, glycerine, which must have a specific gravity of 1-263 (98 to 99 per cent.), and should not contain lime, sulphuric acid, chlorine, or arsenic. The final decoloration may also be effected by sodium hydrosulphite. Very pure FIG. 175. glycerol has been obtained by maintaining it at for some time and then inducing crystal- lisation by a few pure crystals obtained separately by cooling to 40 (Kraut's process). The degree of purity is increased by a second crystallisation. Purification by an osmotic process has also been attempted but with unsatisfactory results. During the last few years the glycerine liquids from the biological or catalytic decom- position of fats (see section on Fats) have also been worked up : they are first neutralised or, better, rendered slightly alkaline with milk of lime and, after being left for some time, the liquid is decanted or filtered off, concentrated to 15 Be. in vacuo, again allowed to stand to deposit a further quantity of lime, decolorised by passing through a carbon filter and again concentrated to 28 Be. Various attempts have also been made to recover the glycerine from the waste liquors from the manufacture of alcohol, but as yet without much success (Ger. Pats. 114,492, 188 ORGANIC CHEMISTRY 125,788, 129,578, 141,703, and 147,558). Separation of glycerine by dialysis does not give good results. STATISTICS AND PRICES. In 1890, the world's production of crude glycerine amounted to 26,000 tons from candle factories and 14,000 tons from soap factories, the amounts due to the principal nations being : France, 6000 tons (candles), 3500 tons (soap) ; Germany, 3000 and 2000 ; England, 1200 and 5500 ; Italy, 180, &c. In 1900 the production rose to 80,000 tons (equally divided between soap and candle factories) and Germany, with a production of about 10,000 tons, exported 2730 tons (value about 140,000) in 1900 and 1580 tons in 1909, against importations of 5373 tons in 1908 and 3530 tons in 1909. In 1890 France exported 3856 tons (value 156,000), in 1900 about 7450 tons (value 308,000), and in 1909 as much as 7000 tons out of a total production of 12,000 tons ; 9000 tons were made at Marseilles, where the most important refinery produces more than 2000 tons per annum. The French exportation is now directed especially to the United States (more than 4000 tons in 1910). According to the official statistics (!) Italy produced 190 tons of distilled glycerine (worth 8660) in 1905 and 215 tons (value 12,040) in 1908 ; the imports were 198 tons in 1907 and 1908, 160 tons in 1909 and 270 tons in 1910; and the exports 833 tons in 1908, 1145 tons (worth 59,540) in 1909, and 1763 tons (value 126,920) in 1910. In 1910 Spain produced 2500 tons of glycerine and exported 893 tons. In 1905 the United States produced 23,000 tons (1,040,000), of which 13,500 tons were obtained from soap-works ; the imports amounted to 16,000 tons in 1909 and to more than 20,000 tons in 1910. England exported 10,500 tons (one-half in the crude state) in 1909 and about 12,500 tons (1,040,000) in 1910 ; in 1911 the output was 16,000 tons, one-half of which was refined. Two main qualities of glycerine are distinguished : * (a) Crude glycerine from the candle or soap works ; (b) Refined glycerine, which is subdivided into : pale, white, for dynamite, and chemically pure. In 1905-1909 the price of No. II dark brown crude glycerine at 24 Be. was 30s. 6d. per quintal, and at 28 Be. 36s. per quintal ; for the light brown quality, 46s. 6d. per quintal at 28 Be., and for the pale at 28 Be. 4. Yellow refined at 28 Be. cost 93s. ; white refined No. I, 96s. at 28 Be. and 108s. at 30 Be. ; free from lime for soap, 5 at 28 and 108s. at 30 Be. Finally the purest double distilled glycerine for nitro- glycerine at 31 Be. cost 6 per quintal. At the beginning of 1910 these prices were increased by 25 per cent, and towards the end of 1910 by 50 per cent, or even 70 per cent. At the beginning of 1911 they were still higher mainly owing to the large amount required in North America for making dynamite for the Panama Canal and other public works. C. TETRA- AND POLY-HYDRIC ALCOHOLS These are usually sweet, crystalline substances which decompose near their boiling-points. They are distinguished one from another by the crystalline forms of their phenylhydrazine derivatives. 1 Tests for Glycerine : the crude, pale at 28 Be., contains 0-5 per cent, of ash and is not rendered turbid by HC1, and only faintly so by lead acetate ; that separated from sulphuric acid saponifications, besides having a bad smell and taste, gives 3 to 5 per cent, of ash and 84 to 86 per cent, of glycerine, a turbidity (fatty acWs) or precipitate being produced by HC1 or lead acetate. The glycerine to be used for nitroglycerine and dynamite is subjected to the following tests : the water is calculated from the loss in weight of 20 grms. heated for 10 hours at 100 and for a few hours at a slightly higher temperature. Five grammes, after being heated in a platinum dish at 180 until no further evolution of vapour takes place, are weighed, and should then undergo no further diminution in weight when again heated for a short time ; it is then ashed in the usual way and the ash tested for metals and salts. Glycerine for nitroglycerine should have been distilled at least once, should not contain sugar or fatty acids, should have a neutral reaction and should contain no lead, calcium, or other metals or foreign metalloids ; only traces of Cl, As, and Fe are allowed : the specific gravity should exceed 1-262. The purest glycerine (puriss.) does not contain more than 0-03 per cent, of ash and as much organic impurity, and for dynamite these two should not exceed 0-25 per cent. Oxalic acid is detected by neutralising with ammonia, acidifying with acetic acid and precipitating with CaCl,. The glycerine content is determined from the density (the air- bubbles being removed by heating), use being made of the Table on p. 184 ; in Germany a special Berthelot scale is used indicating one degree higher than the Baume' scale, 26 Berthelot corresponding with a specific gravity of 1-210, 28 with 1-230, 29 with 1-240, and 30 with 1-250. The index of refraction is determined at the temperature indicated in the Table. In many cases the glycerine is estimated directly by means of the acetyl number (see (succeeding Note), but the method in which the glycerine is oxidised by hot permanganate and potassium hydroxide to oxalic acid and the latter precipitated as calcium oxalate should be rejected. The fairly rapid Hehner- Richardson-Jaffe method is used more successfully : the glycerine is destroyed with dichromate and sulphuric acid, and the amount of dichromate used up (or, according to Gautter and Schulze, how much CO 8 is evolved) measured by titration with sodium thiosulphate, or, better, ferrous ammonium sulphate. This method assumes that the glycerine contains no chloride, nitrate, or extraneous organic matter ; these impurities can, in any case, be eliminated by means of silver oxide (chlorides), and lead acetate and calcium carbonate (organic matter), decoloration being then effected by heating with animal charcoal. POLYHYDRIC ALCOHOLS 189 They do not reduce Fehling's solution and hence differ from the carbo- hydrates, but are derived from these by reduction. The valency of an alcohol is given by the number of alcoholic hydroxyls it contains, and hence by the number of monobasic acid residues it can fix to form a neutral ester. Acetic anhydride serves well for this purpose, the hydrogen atoms of the hydroxyl groups being replaced each by an acetyl group, CH 3 CO : 1 C 6 H 8 (OH) 6 + 6(CH 3 -CO) 2 O = 6CH 3 -COOH + C 6 H 8 (0-CO-CH 3 ) 6 . Mannitol Hexacetylmannitol Esters can also be prepared with bromobenzoic acid, the bromine in the resultant product being determined and the number of hydroxyl groups deduced therefrom. Well-defined compounds are also formed with benzal- dehyde and are employed in separating the constituents of different mixtures. ERYTHRITOL(Butantetrol),OH-CH 2 -CH(OH)-CH(OH)-CH 2 -OH, is found in nature in the free state in Protococcus vulgaris, and as orsellinic ester (erythrin) in lichens and algae. It forms crystals, m.pt. 112, b.pt. 330, and is slightly soluble in alcohol and insoluble in ether. It is obtained by decomposing rf-glucose or synthetically from crotonylene, and its constitution is deduced from the fact that it yields secondary normal butyl iodide on reduction with hydriodic acid. A similar reaction takes place with the higher polyvalent alcohols with normal chains. The four possible stereoisomerides are known, the most common being the one now described which is optically inactive. PENTA-ERYTHRITOL has the formula C(CH 2 -OH) 4 , and melts at 253. ARABITOL, C 5 H 7 (OH) 5 (Pentahydroxypentane), crystallises in acicular prisms, m.pt. 102, has a sweet taste and is formed by reducing the corresponding sugar, arabi- nose, with nascent hydrogen ; reduction of xylose similarly yields xylitol. MANNITOL, C 6 H 8 (OH) 6 (Hexanhexol), occurs abundantly in the vegetable kingdom (the larch, sugar-cane, Agaricus integer containing 20 per cent, of mannitol, &c.), but espe- cially in the manna ash (Fraxinus ornus), the dried juice of which forms ordinary wanna; 2 1 In this way is determined the so-called acetyl number which is so widely used in the analysis of fats and oils. With these, the test is made on the insoluble fatty acids obtained by saponifying 40 to 50 grms. of the fat with 40 c.c. of KOH solution (sp. gr. 1-4) and 40 c.c. of alcohol, this mixture being heated for half an hour on the water-bath, after which it is diluted with a litre of water and boiled for three-quarters of an hour in an open beaker to eliminate the alcohol. The liquid is acidified with sulphuric acid and boiled until the fatty acids separate in a transparent condition, when they are removed with a tapped funnel, washed twice with hot water and dried in an oven at 100 to 105". To determine the acetyl number, a few grammes of the substance containing the hydroxyl groups (or about 20 grms. of hydroxylic fatty acids) are treated with two or three times their volume of acetic anhydride and a few drops of concentrated sulphuric acid (formerly in place of the sulphuric acid fused sodium t acetate, in quantity equal to the acetic anhydride, was used, the mixture being heated for two hours on the water- bath in a reflux apparatus). The mass heats spontaneously, and in a few minutes acetylation takes place ; it is then allowed to cool, calcium carbonate being added to precipitate the sulphuric acid and the liquid filtered. The filtrate is distilled or evaporated to separate the acetate in a liquid or crystalline condition. In the case of the fatty acids, the filtrate is, however, diluted with 600 to 700 c.c. of water and boiled for 30 to 40 minutes in an open beaker to remove the acetic acid, a slow current of CO 2 being passed into the bottom of the liquid to prevent bumping. The supernatant liquid is then siphoned from the acetyl compound, which is boiled with another 500 c.c. of water and so on, this operation being repeated until the washing water no longer has an acid reaction. The acetylated derivative is then collected on a filter, washed and dried in an oven. Of this compound, 0-5 to 1 grm. is dissolved in pure, neutral alcohol, and the solution heated for 45 minutes on the water-bath in a 150 c.c. flask with a definite volume (30 to 50 c ; c.) of seminormal alcoholic potash. When cold, the liquid is titrated with seminormal hydrochloric acid in presence of phenolphthalein to determine the excess of alkali which has not taken part in the splitting of the acetic ester. One hydroxyl group for every grm.-mol. of substance corresponds with 56 grms. of KOH fixed. With the fatty acids, which contain also the carboxyl group, the procedure is as follows : 3 to 4 grms. of the acetyl derivative are dissolved in pure, neutral alcohol and the acidity of the carboxyl group (acetyl acid value) determined by titration with N/2-alkali ; the neutralised liquid is boiled with a known volume in excess of N/2-alcoholic potash for a short time on the water-bath, retitration with N/2-hydrochloric acid given the excess of alkali not combined with acetyl groups. The alkali combined (after the first neutralisation), expressed in mgrms. of KOH per 1 grm. of acetyl compound gives the acetyl number. With the fatty acids the sum of the acetylated acid number and the acetyl number is termed the acetyl saponiftcation value. From the acetyl number (N), the molecular magnitude 56,100 (M ), of the alcoholic substance can be deduced by the formula : M - - 42. 1 Manna is extracted more particularly from Fraxinus ornus and Fraxinus rotundifolia, which are widespread in Sicily and Calabria and from which it readily flows through long vertical incisions made in summer and autumn. It seems to occur in the rising sap before this reaches the leaves and is thought by some to be produced by enzyme actions. Crude, commercial manna contains 12 to 13 per cent, of water, 10 to 15 per cent, of sugar, 32 to 42 per cent, of mannitol, 40 to 41 per cent, of mucilaginous substances, organic acids and nitrogenous matter, 1 to 2 per cent, of insoluble substances and 1 to 2 per cent, of ash. Australian manna (from Myoporum platycorpum) contains as much as 90 per cent, of mannitol. The manna tree grows in fertile, rocky soiljind is incised iu its tenth year and in the following 10 or 15 years. It 190 ORGANIC CHEMISTRY from this alcohol extracts pure mannitol, which can be decolorised by repeated treatment with charcoal. In manna it was discovered by Proust in 1806. It is obtained synthe- tically by reducing fructose or glucose : C 6 H 12 O 6 + H 2 = C 6 H 14 O 6 . The optically inactive, laevo- and dextro-rotatory forms are known, the last being the most common ; the optical activity is slight but is rendered more apparent by the addition of borax. When heated it loses water giving anhydrides (mannitan, C 6 H 12 5 , and mannide, C 6 H 10 O4) ; in a vacuum it distils unchanged. One hundred parts of water dissolve 16 parts by weight of mannitol at 16. From alcohol it crystallises in triclinic acicular prisms and from water in large rhombic prisms having a sweet taste and melting at 160. Stereoisomeric with mannitol is DULCITOL, C 6 H 8 (OH) 6 , which occurs in a number of plants and in Madagascar manna. It forms sweet, monoclinic prisms, m.pt. 188, and is almost insoluble in water, even in the hot. Synthetically it can be prepared by reducing lactose and galactose. It is optically inactive even in presence of borax. Another stereoisomeride of mannitol is sorbitol, which melts at 104 to 109, or at 75 when crystallised with 1H 2 O. It can be obtained synthetically by reducing d-glucose or ^-fructose. In presence of borax it shows a slight dextro-rotation. Other stereoisomerides are TALITOL and IDITOL ; these isomerides are usually separated by means of the acetals they form with benzaldehyde. DD. DERIVATIVES OF THE ALCOHOLS A. DERIVATIVES OF MONOHYDRIC ALCOHOLS i. ETHERS These are generally formed by eliminating 1 mol. of water (for example, by concentrated sulphuric acid or by hot hydrochloric acid) from 2 mols. of alcohol, which condense to form 1 mol. of ether in the same way as 2 mols. of an acid give an anhydride : C 2 H 5 OH _ TT o , C 2 H 5 \ O CH 3 -OH CH 3 / L Ethers are not formed by secondary or tertiary alcohols. The first term of the series, methyl ether, is gaseous, and the succeeding terms become liquid and then solid as the molecular weight increases, the ethereal odour of the first members being gradually lost. is then cut back and the new branches incised in the seventh year and the succeeding 10 or 15 years. It is then again cut back, this procedure being continued for 80 or 100 years. J One hectare with 4500 trees gives as much as 100 kilos of manna per annum. It is harvested in August and September. Manna is used as a mild purgative for children. It has a sweetish taste, is soluble in' water or alcohol, and, besides mannitol, contains various sugars such as stachyose and mannatrwse. To extract the mannitol, the crude manna is dissolved in half its weight of water containing white of egg. The solution is boiled for a few minutes and strained, and the filtered mass, solidified by cooling, pressed in bags, or, better, centrifuged and washed at the same time with a large quantity of cold water. It is redissolved in water and the solution boiled with animal charcoal, filtered under pressure, crystallised and centrifuged. The mother-liquors are used to dissolve fresh quantities of the crude manna. The fineness of the crystals depends on the concentration and on the temperature of the air ; in some cases the crystallisation is disturbed by continually stirring the mass. Sometimes the manna solutions are first subjected to lactic fermentation, by which means considerable quantitie of calcium lactate are obtained ; the mannitol is then extracted from the residual liquors. Mannitol is not fermented by beer-yeast, but with chalk and sour cheese it gives a considerable amount of alcohol, volatile acids, carbon dioxide, and hydrogen. When cautiously oxidised with nitric acid, it forms d-mannose and d-fructose, whilst with the Sorbose bacterium it gives only the latter sugar. Mannitol has a slight laevo-rotation ( 0-15) which is increased by alkali and changed in sign by borax. It dissolves in 6-5 parts of water at 18, in 80 parts of 60 per cent, alcohol at 15, or in 1400 parts of absolute alcohol ; it is insoluble in ether. Manna in casks costs 3s. to 5s. per kilo ; assorted, 1. 7d. ; in lumps, 9i is found ready formed in certain plants, e.g. in the dog- mercury weed (Mercurialis perennis). It is formed in the distillation of wood and occurs in beetroot and bone residues and in herring brine. It is a gas like ammonia and precipitates various metallic salts, but, when added in excess, does not dissolve nickel and cobalt hydroxides ; it is more highly basic and more soluble in water than ammonia, and has a strong odour of ammonia and rotten fish. It becomes liquid at 6 and at 11 has the sp. gr. 0-699. With sodium hydroxide and bromine it gives acetamide. Its hydrochloride, CH 3 NH 2 ,HC1, is a crystalline, deliquescent substance extremely soluble in alcohol. With aluminium sulphate its sulphate forms an alum containing 24H 2 O. DIMETHYLAMINE, (CH 3 )NH, is a liquid boiling at + 7, and is formed, together with acetic acid, in the distillation of wood. TRIMETHYLAMINE, (CH 3 ) 3 N, is a gas which liquefies at +3, and has an intense odour of rotten fish. It is found in various plants (Arnica montana, shoots of the pear- tree, &c.), and in herring brine. It is formed by the decomposition of betaine during the distillation of beetroot molasses (p. 96). ETHYLAMINE, C 2 H 5 -NH 2 , is a liquid, b.pt. + 19, and smells strongly of ammonia, which it surpasses in basicity. It dissolves very readily in water with generation of heat. It dissolves aluminium hydroxide, and to a small extent cupric hydroxide but not ferric or cadmium hydroxide. DIETHYLAMINE, (C 2 H 5 ) 2 NH, is a liquid, b.pt. 56, and does not dissolve zinc hydroxide. 202 ORGANIC CHEMISTRY TRIETHYLAMINE, (C 2 H 5 ) 3 N, is an oily liquid which precipitates metals from their salts but does not redissolve the precipitates. It has a strongly alkaline reaction and boils at 89. It is extremely soluble in cold water, but above 20 it becomes completely insoluble, separating from the water in an oily layer. A group of nitrogen compounds which may be considered as formed by the condensation of ammonia (hydrazine, azoimide, hydroxylamine, &c.) has been already mentioned in vol. i, pp. 327 and 332. The alkyl derivatives of hydroxylamine, NH 2 -OH, are divided into two groups : a-alkylhydroxylamines, in which the alkyl replaces the hydroxylic hydrogen NH 2 OR, and which hence have an ether character and do not reduce Fehling's solution ; and ft-alkylhydroxylamines, in which the alkyl radical replaces an amino -hydrogen and is therefore joined to the nitrogen, R NH OH ; these reduce Fehling's solution even in the cold and on energetic reduction yield primary amines. Also the Alkylhydrazines, RNH NH 2 , R 2 N-NH 2 , &c., unlike amines, reduce Fehling's solution in the cold and give characteristic reactions with aldehydes and ketones. The Diazo-compounds of the methane series are of slight importance, whilst those of the aromatic series are a very important class of compounds ; the former differ from the latter in that the characteristic divalent nitrogen group, N=N , has its valencies saturated by only one carbon atom. Diazomethane, CH 2 N 2 , which is a yellow, poisonous gas, is prepared from hydroxylamine and dichloromethylamine. V. PHOSPHINES, ARSINES, AND ALKYL METALLIC COMPOUNDS Like ammonia, the hydrogen derivatives of phosphorus, arsenic, antimony, &c., give rise to alkyl compounds which have a very feebly basic character and a very unpleasant odour. 1. PHOSPHINES. These are gases or colourless liquids with repulsive odours. Their basic properties and their stability towards water become more marked as the number of alkyl groups increases. They are readily oxidisable with nitric acid, the remaining hydrogen atoms of the PH 3 being transformed into hydroxyl groups. The quaternary phosphonium bases are very strongly basic, and, unlike the corresponding ammonium bases, they lose an alkyl group in the form of a saturated hydrocarbon when heated, the residue being a trialkylphosphonium oxide. C w H 2n+1 PH 2 (C n H 2n+1 ) 2 PH (C M H 2n+1 ) 3 P (C n H 2n+1 ) 4 P-OH Primary phosphine Secondary phosphine Tertiary phosphine Tetralkylphosphonium hydroxide C n H 2n+1 PO(OH) 2 (C n H 2n+1 ) 2 PO. OH (C n H 2n+1 ) 3 PO Alkylphosphonic acid Dialkylphosphonic acid Trialkylphosphine oxide The primary and secondary phosphines are formed by heating phosphonium iodide with alkyl iodides and zinc oxide, whilst the tertiary phosphines and phosphonium deri- vatives are obtained from hydrogen phosphide, PH 3 , and alkyl halogen compounds. 2. ARSINES. Well-known primary and secondary compounds are : methylarsenic dichloride, CH 3 AsCl 2 (liquid, b.pt. 135) ; dimethylarsenic chloride, (CH 3 ) 2 AsCl (b.pt. 100) ; dimethylarsine, (CH 3 ) 2 AsH (b.pt. 36) ; dimethylarsenic acid or cacodylic acid, (CH 3 ) 2 AsO OH, &c. The tertiary arsines are obtained by the action of sodium arsenide, AsNa 3 , on alkyl iodides : 3C 2 H 5 I + AsNa 3 = 3NaI + As(C 2 H 5 ) 3 ; they are liquids slightly soluble in water, with which they do not form bases. The quaternary arsonium compounds, e.g. (CH 3 ) 4 AsI (tetramethylarsonium iodide), obtained from the tertiary arsines and alkyl iodides, are, however, very energetic and are able to give, with moist silver oxide, tetramethylarsonium hydroxide. The cacodyl [(CH 3 ) 2 As - As(CH 3 ) 2 ] compounds were studied by Bunsen (1837-J843), who obtained cacodyl oxide, (CH 3 ) 2 As-O.As(CH 3 ) 2 , " by distilling arsenic trioxide with potassium acetate (this reaction serves as a delicate test for acetates in mixtures) : As 2 O 3 + 4CH 3 -COOK = 2C0 2 + 2K 2 C0 3 + [As(CH 3 ) 2 ] 2 0. With hydrochloric acid, cacodyl oxide gives cacodyl chloride, (CH 3 ) 2 AsCl. ORGANOMETALLIC COMPOUNDS 203 Many of these cacodyl compounds are liquids which ignite in the air and have nauseating odours ; the cacodyl behaves like a true electro- positive element. 3. Various alkyl derivatives are known of antimony (stibines), boron, silicon, bismuth, tin, &c., but these are of little practical importance. 4. ALKYLMETALLIC (Organometallic) DERIVATIVES. These are obtained from various metallic chlorides or from the metals themselves (Zn, Hg, Mg, Al, &c.) by the action of halogen derivatives of the hydrocarbons. They are generally colourless liquids with low boiling-points, and some of them are violently decomposed by water and ignite in the air. Of importance for many organic syntheses are the zinc-alkyls (see pp. 32, 96, and 149). ZINC METHYL : Zn(CH 3 ) 2 , is a colourless, highly refractive liquid, sp. gr. 1-39, b.pt. 46, and has an intense, repulsive odour ; it ignites in the air, forming zinc oxide, and with water gives methane and zinc hydroxide. It is formed in two phases, as follows, and is separated by distillation : (a) CH 3 I + Zn = Zn(CH 3 )I (zinc methyl iodide, solid) ; (b) 2Zn(CH 3 )I = Znl + Zn(CH 3 ) 2 . GRIGN ARD ' S REACTION. Mention has already been made of the use of this reaction in synthesising the saturated hydrocarbons (p. 32). One molecule of a monohalogen (Br or I) compound, in presence of absolute ether, combines with an atom of magnesium : Mg + C 2 H 5 Br = C 2 H 5 MgBr (ethyl magnesium bromide), and with compounds containing several carbon atoms there is always formed, as a secondary product, a saturated hydro- carbon. The ether probably takes part in the reaction, forming an intermediate product C 2 H 5 .Mg.Br[(C 2 H 6 ) 2 0]. The latter, and also the alkyl magnesium halogen compounds, when dissolved in ether, are highly reactive and form additive compounds with aldehydes, ketones, and even esters of mono- and poly-basic carboxylic acids ; with water these additive compounds then give the corresponding secondary and tertiary alcohols, the reaction occurring in the following two phases (R = alkyl) : /0-MgI R-CHO + R'Mgl = R-C^-R' -> + H 2 O = I-Mg-OH + R.CH(OH)-R' Aldehyde Alkyl mag- H Secondary nesium iodide alcohol /OMg-Br H - COOC 2 H 5 + C 2 H 5 MgBr = C 2 H 5 - C^H -> + C 2 H 5 . MgBr = Ethyl formate X OC 2 H 5 /OMgBr Br.Mg.OC 2 H 5 + C 2 H 5 .CH -> + H 2 O = BrMgOH + C 2 H 5 .CH(OH).C 2 H 5 Diethylcarbinol If esters of other monobasic acids are used instead of a formic ester, tertiary alcohols are obtained, whilst esters of dibasic acids give dihydric alcohols. Hence, by means of the Grignard reaction, the carboxylic oxygen of any acid (starting from the corresponding ester) is ultimately replaced by two alkyl residues. Similar behaviour is also shown by acid chlorides and anhydrides, which also contain carbonylic oxygen ( CO ). With nitriles, Tcetonimides and ketones are obtained : yN-Mgl R-CN + R'Mgl = R-Cf -* + H 2 O = X R' IMgOH + R.C ( : NH).R' -> + H 2 = NH 3 + R.CO-R'(ketone). Further, with dry CO 2 , alkyl magnesium compounds give organic acids : R'Mgl + CO 2 = R'-COOMg-I -v + HX = IMgX + R'-COOH (acid). Other most varied organic syntheses have been rendered possible of late years by the Grignard reaction. 204 ORGANIC CHEMISTRY VI. ALDEHYDES AND KETONES, C w H 2n O The elimination of two atoms of hydrogen by means of an oxidising agent (e.g. potassium dichromate and dilute sulphuric acid, or sometimes even the oxygen of the air), from a primary or secondary alcohol yields an aldehyde or a ketone : R-CH 2 -OH + = H 2 + R-CHO (aldehyde), or R-CH(OH)R' + = H 2 + R-CO-R' (ketone). The aldehydes have a strong reducing action, as they fix oxygen and become converted into acids with the same numbers of carbon atoms, whilst the ketones resist oxidising agents, and, if these are very energetic, are oxidised to acids containing fewer carbon atoms than the original ketones. (a) ALDEHYDES The first members of this series are neutral liquids with pronounced and often disagreeable odours (formaldehyde is a gas) and are soluble in water, whilst the higher ones gradually become solid and insoluble. Their boiling- points are much lower than those of the corresponding alcohols. The aldehydes are formed when a calcium or barium salt (or even two salts) of a monobasic organic acid is dry distilled with calcium or barium formate (reducing agent): (R-COO) 2 Ca + (H-COO) 2 Ca = 2CaC0 3 + 2R-CHO. They are also obtained on heating with water compounds containing two halogen atoms united to the same carbon atom : CH 3 -CHC1 2 (ethylidene chloride) + H 2 = 2HC1 + CH 3 -CHO. The constitution of the aldehydes can be deduced from their methods of 7 formation (e.g. the latter) and the characteristic aldehyde group is OC X H PROPERTIES. They are substances of considerable and varied reac- tivity. With oxidising agents they are transformed into acids, and this re- ducing property is readily manifested in their reduction of ammoniacal silver nitrate solution (22 per cent, ammonia solution and 10 per cent, of dilute silver nitrate diluted with its own volume of 10 per cent, sodium hydroxide solution ; or 1 grm. of silver nitrate dissolved in 30 c.c. of water and dilute ammonia added as long as no precipitate forms) or of Fehling's solution (the latter, how- ever, is not reduced by aldehydes containing as many as 8 or 9 carbon atoms). In their turn, the aldehydes are converted back into the primary alcohols when reduced with nascent hydrogen ; with PC1 5 , they give ethylidene chlorides again. Hydrocyanic acid, ammonia, sodium hydrogen sulphite, and sometimes alcohol and acetic anhydride (also the alkyl magnesium halogen compounds : see above, Grignard's Reaction) form characteristic additive products with the aldehydes : CH 3 -CHO + 2C 2 H 5 -OH (+ a little HC1) = H 2 + CH 3 -CH(OC 2 H 5 ) 2 (acetal), which is an ether of the hypothetical glycol (dihydric alcohol), CH 3 - CH(OH) 2 ; the latter, however, does not exist in the free state, since two hydroxyl groups cannot remain joined to one and the same carbon atom, excepting in the case of chloral hydrate (see later) and a very few other substances. 1 They combine with sodium and ammonium bisulphites (very concentrated 1 See Table on opposite page. ALDEHYDES 205 solutions) forming crystalline bisulphite compounds soluble in water and slightly so in alcohol : OH S0HNa = CH / -CO- H H S0 2 Na, and these compounds, when heated with dilute acid or with alkali (even Na 2 C0 3 ), liberate the aldehyde again. This reaction hence renders possible the separation of aldehydes from other substances. The aldehydes combine with ammonia forming crystalline aldehyde- ammonias soluble in water and slightly so in alcohol but insoluble in ether, for example, CH 3 -CH(OH)(NH 2 ), which gives the aldehyde again when heated with a dilute acid. But formaldehyde, with ammonia, readily forms poly- merised derivatives, e.g. hexamethylenetetramine, (CH 2 ) 6 N 4 . With hydrocyanic acid they form cyanohydrins (p. 199). An interesting change is the aldol condensation, that is, the condensation of 2 mols. of an aldehyde brought about by prolonged heating with dilute mineral acids, dilute alkalis, or even aqueous solutions of sodium acetate. Possibly a molecule of water is first added to one of the aldehydes : CH, X ,0 H H 2 = CH 3 N OH this hypothetical hydrate then condensing with another molecule of aldehyde, with separation of water and formation of a hydroxyaldehyde (aldol) : ,0 //O CH 3 CHX CH C H = H 2 CH 8 CH(OH) CH 2 V H (fi-hydroxybutyraldehyde). These aldols in their turn readily lose a molecule of water, forming an unsaturated aldehyde, which can also be obtained directly (aldehyde condensation) by heating the original aldehyde with a dehydrating agent such as zinc chloride : CH,-C / CH-C H H =H 2 CH,-CH:CH-C H DERIVATIVES OP ACETALS Name formula Boiling-point Specific gravity ALKYL DERIVATIVES Mcthylal CH 2 (OCH a ) 2 41-3 -41-7 (749-8 mm.) 0-862 (18) Diethylmethylal . CH 2 (OC 2 H 6 ) 2 87 0-834 (20) Dipropylmethylal CH 2 (OC,H,) 2 136 0-834 (20) Diisopropylmethylal CH 2 (OC a H,) 2 118 0-831 (20) Diisobutylmethylal CH 2 (OC 4 H 9 ) 2 164 0-824 (20) Di i soamylmcthylal CH 2 (OC 5 H n ; 2 + H 2 O 206 0-835 (20) Dihexylmethylal CH 2 (OC,H 1S ) 2 174-175 0-822 (15) Dioctylacetal CH 8 -CH(OC,H 17 ) 2 289 0-848 (15) Dimethylacetal CH a -CH(OCH 3 )j 63 0-865 (22) Diethylacetal CH 3 -CH(OC 2 H 5 ) 2 102-9 0-831 (20) Dipropylacetal CH 8 -CH(OC,H,) 2 147 0-825 (22) Diisobutylacetal CH 3 -CH(OC 4 H ! ,) a 170 0-816 (22) Di isoamylacetal CHa-C^OCjHnJj 211 0-835 (15) ACID DERIVAT VES Methylencdiacetate CH 2 (0-CO-CH a ) 2 170 Ethylencdiacetate CH 3 -CH(O-CO-CH 3 ) 2 169 1-073 (15) Ethyleneclipropionate CH3-CH(O-CO-C 2 H 5 ) 2 192 1-020 (15) Ethylencdibutyratc CH 3 -CH(O-CO-C 3 H,) 2 215 0-985 (15) Ethylenediisovalerate CH a -CH(O-CO-C 1 H,) 2 225 0-947 (15) 206 ORGANIC CHEMISTRY The aldehydes, especially form-, acet-, and prop-aldehydes, &c., exhibit a tendency to polymerise, in the mere presence of a little hydrochloric or sulphuric acid, sulphur dioxide, zinc chloride, &c. Acetaldehyde, for example, gives two isomerides : paraldehyde, m.pt. 10, b.pt. 124, and metaldehyde, which sublimes at 100 : /0-CH(CH 3 K 3C 2 H 4 = CH 3 -CH/ )>O. X 0-CH(CH 8 r These no longer react with ammonia, sodium bisulphite, silver nitrate, and hydroxylamine, but they yield the aldehyde again when distilled in presence of a small quantity of dilute sulphuric acid. With alkalis, even dilute alkalis, many aldehydes, especially the more simple ones of the fatty series, resinify, whilst some give rise to an alcohol and an acid : X 2HC- f (formaldehyde) + H 2 = CH 3 OH + H- C0 2 H (formic acid). ^U With halogens the aldehydes give substitution products, and with hydrogen sulphide various complex products (thioaldehydes, &c.) with characteristic odours. With hydroxylamine, aldehydes form aldoximes, which are resolved into their components when boiled with acids, and yield nitriles when treated with dehydrating agents : CH 3 - CHO + NH 2 - OH = H 2 + CH 3 - CH : N- OH. A similar action is exhibited by the hydrazines (as hydrochloride or acetate in acetic acid solution containing sodium acetate ; the most suitable is phenyl- hydrazine), which give characteristic, stable, and often crystalline compounds, termed hydrazones : CH 3 -CHO + C 2 H 5 -NH-NH 2 (ethylhydrazine) = H 2 O + CH 3 -CH : N-NH-C 2 H 6 (acetaldehyde ethylhydrazone) ; by nascent hydrogen (4H) this is converted into 2 mols. of primary amine : 2CH 3 -CH 2 -NH 2 . Characteristic of the aldehydes is also the formation of crystalline semi- carbazones by the action of the hydrochloride of semicarbazide, NH 2 CO NH-NH 2 (obtained by the interaction of potassium cyanate and hydrazine hydrate) : R-CHO + NH 2 -CO-NH-NH 2 = H 2 + R-CH : N-NH-CO-NH 2 . Both the hydrazones and semicarbazones serve for the separation of the aldehydes from other substances and for their quantitative determination. Finally a characteristic qualitative reaction which is given generally by the aldehydes and is very sensitive is that of Schiff. It consists in shaking the liquid to be tested with a solution (0-02 per cent.) of fuchsine previously decolorised by a current of sulphur dioxide. Traces of an aldehyde produce a reddish violet coloration (it is uncertain if pure ketones also give this reaction). Another reaction characteristic of the aldehydes and not given by ketones is that with benzosulphinehydroxamic acid or with nitrohydroxylaminic acid, /OH OH-NO:N'OH, which forms hydroxamic acids, R-C^ the latter producing a cherry-red coloration with ferric chloride. FORMALDEHYDE (or Methanal), H-CHO, is a gas which liquefies at 20 to a mobile, colourless liquid having the sp. gr. 0-8153 and solidifying at 92. It is very soluble in alcohol o^ water, and is placed on the market FORMALDEHYDE 207 in the form of 40 per cent, aqueous solution 1 under the name of formalin or formal ; the commercial product often contains^l2'to 15 per cent, of methyl alcohol to prevent separation of polymerised compounds. Indeed, even in the cold, formaldehyde readily forms paraformaldehyde, (CH 2 0) 2 , a white solid soluble in water, or the crystalline trioxymethylene (or metaformaldehyde), (CH 2 0) 3 . Both of these give the aldehyde when volatilised by heat, and they are used thus as disinfectants under the names triformol and paraformol. Formaldehyde may also give rise to a mixture of saccharine compounds (formose). With ammonia it gives, not an aldehyde -ammonia, but hexa- methylenetetramine, C 6 H 12 N 4 , which is crystalline and of feebly monobasic character. 2 With potassium hydroxide it does not resinify,but yields methyl alcohol and formic acid (p. 206). A question which has been under discussion for many years is the possible formation of formaldehyde as the first product in the natural synthesis of carbohydrates (see Sugar) in the leaves of plants, from carbon dioxide under the influence of chlorophyll. Numerous sensitive reagents have been employed to detect microscopically the transitory formation of formaldehyde in living leaves ; but almost all these reagents are poisonous to plants and no decisive results have been obtained, even those of Pollacci (1907), who distilled the leaves with water and tested for formaldehyde in the distillate, being doubtful. Schryver (1910) has succeeded in establishing the formation of aldehyde in green plants in sunlight, by making use of a very sensitive reagent (detecting 1 part of aldehyde per million) consisting of a solution of phenylhydrazine, potassium ferricyanide, and hydrochloric 1 The concentrations of commercial aqueous solutions of formaldehyde can be deduced from the specific gravities by means of the following table (Auerbach, 1905) : Sp. gr. at 1-0054 1-0090 1-0126 1-0172 1-0218 1-0311 1-0410 1-0568 1-0719 1-0853 1-1057 1-1158 Grms. of CH 2 O in 100 c.c. of solution 2-24 3-50 4-66 6-51 8-37 11-08 14-15 19-89 25-44 30-17 37-72 41-87 Grms. of CH 2 O in 100 grms. of solution 2-23 3-45 4-60 6-30 8-0 10-74 13-59 18-82 23-73 27-80 34-11 37-53 If the aldehyde is pure and leaves no residue, the percentage by volume, if greater than 23, should be increased by about 5. The analysis of commercial formalin is based on the following reaction of Blank and Finkenbeiner : 2CH 2 O + 2NaOH + H 2 O 3 = H 2 + 2H 2 O + 2H-CO 2 Na. Three grammes of the formaldehyde solution are poured into a long-necked flask containing 25 c.c. of 2N-caustic-soda solution (free from carbonates), the liquid being mixed and 50 c.c. of hydrogen peroxide solution (neutralised or of known acidity) carefully added, 3 minutes being taken to make this addition. After 7 to 8 minutes, the excess of alkali remaining is titrated with 2N-sulphurie acid. With every cubic centimetre of the 2N-alkali that has reacted corresponds 0-06 grm. of formaldehyde. Litmus purified several times with alcohol should be used as indicator. The estimation of the aldehyde may also be carried out with ammonia (see succeeding Note). Brautigam (1910) suggested determining formaldehyde by adding to it excess of clear calcium hypochlorite solution. After a time the solution deposits calcium carbonate, which is filtered, washed, and weighed ; 1 mol. of CaCO 3 corresponds with 1 mol. of formaldehyde. To determine the methyl alcohol which may be present, 5 c.c. of the solution, diluted with 100 c.c. of water, are distilled with an excess of ammonia (about 10 c.c. of concentrated ammonia), 50 e.c. of the distillate being collected in a 100 c.c. flask and made up to volume with water. The methyl alcohol in 5 c.c. of this solution, which contains only negligible traces of formaldehyde, is determined by the iodine method (see p. 107). 2 This reaction was proposed by L. Leger in 1883 as a means of estimating formaldehyde in commercial solutions : 6CH Z O + 4NH, = (CH 2 ) e N+6H 2 O ; the reaction is, however, slow and the method not very accurate. F. Hermann (1911) has rendered it more rapid and exact in the following manner. Pour cubic centimetres of the formalin are weighed into a 150 c.c. flask with a ground stopper, and about 3 grms. of pure powdered ammonium chloride and exactly 25 c.c. of 2N-caustic soda (equivalent to 50 c.c. of normal soda) added. The flask is stoppered and shaken, and, when the mass is cool, 50 c.c. of water and 4 drops of 1 per cent, methyl orange are added and the excess of alkali titrated with normal sulphuric acid. Deduction of the volume of acid required from 50 c.c. gives the volume of soda used in liberating, from the ammonium chloride, a corresponding amount of nascent ammonia, which instantly transforms the aldehyde into hexamethylenetetramine. The latter is, however, monobasic and reacts with part of the sulphuric acid, and, in order to obtain the number of grammes of formaldehyde in the quantity of formalin taken, the number of cubic centimetres of soda arrived at above must be multiplied by the factor 0-06. If the formalin be acid initially, the acidity must be determined separately by titration with soda in presence of phenolphthalein and the 50 c.c. of soda increased accordingly. 208 ORGANIC CHEMISTRY acid ; this reagent gives a magenta-red coloration with formaldehyde or with the methylene derivative which chlorophyll would form with the aldehyde. MANUFACTURE. Formaldehyde is obtained by passing a mixture of methyl alcohol vapour and air over copper or platinum gauze or the finely divided metals, which act as catalysts (O. Blank, Ger. Pat. 228,697, 1908, obtained quantitative yields by using silver precipitated on asbestos). The product is rectified in a column filled with pieces of clay. Patents have been taken out for the preparation of formaldehyde by the oxidation of methane with oxide of iron, hydrogen peroxide, &c. It is also formed by the electrolysis of dilute methyl alcohol, and some years ago a patent was granted for its preparation by passing a mixture of formic acid vapour and hydrogen through a tube containing pieces of metal (e.g. lead, iron, zinc, nickel, silver, &c.), heated to 300. Formaldehyde has considerable antiseptic power, even in aqueous solution. It is largely used at the present time as a disinfectant in houses and for the preservation of readily putrescible substances (meat, beverages, &c.). Its vapour has an acute and penetrating odour and irritates the eyes. On account of the property possessed by formaldehyde of combining with proteins to form insoluble and stable products, it is used in the manufacture from casein of articles of a horny consistency and in making imitation pegamoid ; also in preparing photographic films with gelatine, for rendering insoluble or hardening the coloured gelatine for textile printing, and for hastening the tanning of skins. Owing to its great reactivity, it is largely used in organic syntheses, e.g. in the manufacture of aniline dyes. Various solid and liquid disinfectants containing free aldehyde are prepared by means of soaps (soap solutions are also on the market under the names of lysoform and ozoform, the starting product in the case of the latter being sulphoricinoleic acid). A characteristic and very sensitive reaction of formaldehyde is that proposed by Rimini, according to whom a mixture of phenylhydrazine hydrochloride, sodium nitroprusside and caustic soda is coloured blue even by minimum traces of the aldehyde. Formaldehyde gives Schiffs reaction even in presence of a certain amount of sulphuric acid, whilst acetaldehyde does not. The price of commercial 40 per cent, formaldehyde is about 4 per quintal, while pure, powdered paraldehyde costs 4s. to 5s. per kilo. ACETALDEHYDE (Ethanal), CH 3 -CHO, is a colourless, mobile liquid, sp. gr. 0-801 (at 0), b.pt. 21, and solidifies at 121. It has an agreeable but suffocating odour, and it polymerises with moderate ease, giving the paraldehyde and metaldehyde (see above). It dissolves in water, alcohol, or ether, and is readily converted into acetic acid by oxidising agents. It is prepared by pouring a mixture of 3 parts of 90 per cent, alcohol and 4 parts of concentrated sulphuric acid slowly into a solution of 3 parts of potassium bichromate in 12 of water, the liquid being kept cool meanwhile. The solution is then heated in a reflux apparatus on a water-bath and subsequently distilled. The mixture of alcohol, aldehyde, and acetal thus obtained is heated to 50 and the aldehyde vapour passed into cold ether. On passing ammonia into this solution, crystallised aldehyde-ammonia, CH 3 CH(OH) NH 2 , separates ; this, when pressed and distilled with dilute sulphuric acid, gives pure acetalde- hyde. The commercial aldehyde is obtained from the foreshots of alcohol distillation, from which it is separated by simple fractional distillation. It is of importance in many organic syntheses and in .the production of silver mirrors. The price of 50 per cent, solutions is 2s. per kilo, that of the 95 to 99 per cent, product 3s. 6d., and that of the purest aldehyde 15s. 1 1 The estimation of acetaldehyde is based on the following reaction (Seyewetz and Bardin) : 2Na 2 SO, + 2CH.-CHO + H a SO 4 = Na a SO 4 + (CH.-CHO, NaHSO,) z . The aldehyde is diluted to 7 to 8 per cent, and about 10 c.c. of this solution is poured into 40 c.c. of 10 per cent, pure sodium sulphite solution. After the addition of a few drops of neutralised alcoholic phenolphtbaJc.in solution UNSATURATED ALDEHYDES 209 METHYLAL, H.CH(OCH 3 ) 2 , and ACETAL, CH 3 .CH(OC 2 H 6 ) 2 (see p. 204). PROP ALDEHYDE, C 2 H 5 -CHO, is found among the tarry products from the distillation of wood. Valeraldehyde, C 4 H 9 CHO, boils at 92 and begins to show a diminu- tion in solubility in water. Normal heptaldehyde (oenantaldehyde), C 6 H 13 -CHO, is found among the products of decomposition of castor oil when this is subjected to distillation in a vacuum. Nonyl aldehyde, C 8 H 17 CHO, occurs in the oxidation products of oleic acid or, better, in the decomposition products of the ozonide of oleic acid (Harries, Molinari, &c.) ; it boils at about 192. CHLORAL (Trichloroethanal), CC1 3 -CHO, is the most important halo- genated derivative of the aldehydes. It is a dense liquid with a peculiar, penetrating odour and boils at 94-4. It is prepared by passing chlorine into pure alcohol (96 per cent.) for some days, the hydrochloric acid formed being collected. The liquid is then heated with sulphuric acid in a reflux apparatus until no further evolution of hydrogen chloride occurs, the chloral being distilled and subsequently purified by rectification. Within recent times it has also been prepared electrolytically : the bath contains potassium chloride solution at 100 and is fitted with a diaphragm ; alcohol is passed into the anode chamber, where chlorine is formed, and the hydrogen chloride produced at the anode is neutralised by the potassium hydroxide formed (1 h.p.-hour yields 50 grms. of chloral). Chloral gives the reactions of the aldehydes and is used in medicine as an anaesthetic and soporific, being first treated with water to form the crystalline OTT CHLORAL HYDRATE, CC1 3 CH< ::;, which is readily soluble in water (m.pt. 57) ; this is one of the few compounds having two hydroxyl groups united to the same carbon atom. The crystalline alcoholates or Acetals, CC1 3 -CH(OH)-OC 2 H 5 and CC1 3 -CH(OC 2 H 5 ) 2 , corresponding with this hydrate are known. Chloral costs about 6s. per kilo. ALDEHYDES WITH UNSATURATED RADICALS ACRYLIC ALDEHYDE (Propenal, Acrolem, or Allyl Aldehyde), CH 2 : CH-CHO, is formed when fats are burned, owing to loss of water by the glycerol present ; a similar change takes place when glycerol is heated with potassium hydrogen sulphate or boric acid. Acrolein, which can also be obtained by the oxidation of allyl alcohol, is a liquid, b.pt. 52-4, and has a characteristic pungent odour. When oxidised, it yields acrylic acid and, when reduced, allyl alcohol. It has all the chemical properties of the aldehydes and polymerises in the course of a few hours. With ammonia, it gives a solid, basic condensation product, soluble in water : 2C 3 H 4 + NH 3 = H 2 O + C 6 H 9 ON (acrole'ir.- ammonia, which gives picoline on distillation). Owing to its double linking, acroleiin unites with 2 mols. of sodium bisulphite and the resulting product, when boiled with acid, gives up only one bisulphite molecule, namely, that combined with the aldehyde group. CROTONIC ALDEHYDE, CH 3 .CH : CH-CHO, is obtained by distilling the corre- sponding aldol, CH 3 .CH(OH).CH 2 .CHO, at 140 or by the dehydrating action of zinc chloride or sodium acetate on the saturated aldehyde. It is a liquid boiling at 104 and possessing a penetrating odour, and its constitution is shown by the fact that it yields crotonic acid when oxidised with silver oxide. CITRAL L (or Geranial), (CH 3 ) 2 C : CH.CH 2 .CH 2 -C(CH 3 ) :CH-CHO, is a liquid of pleasant odour, b.pt. 229, and occurs in many essences (of mandarin, citron, lemon, orange, and most abundantly 60 per cent. in that of Verbena Indiana or lemon -grass, from which it is separated by means of its bisulphite compound). It may also be obtained by the gentle oxidation of the corresponding alcohol, gerianol, which boils at 230. It exists in two stereo - isomeric forms, the cis- and trans-modifications. When oxidised with potassium bisulphate at 170, citral is transformed into cymene (with a closed ring) with separation of water. the liquid is cooled to 4 to 5 and titrated with normal sulphuric acid until it is decolorised. This occurs when no further combination of aldehyde and sulphurous acid takes place. This determination is not affected by the presence of alcohol, acetal, or paraldehyde. II 14 , 210 ORGANIC CHEMISTRY CITRONELLAL, (CH 3 ) 2 C : CHC-H 2 .CH 2 .CH(CH 3 ).CH 2 .CHO, is found with citral in citron oil and boils at 208. PROPARGYL ALDEHYDE, CH C-CHO, is a solid, m.pt. 60, and is obtained from dibroinoacroleiin by way of the acetal. As it contains the group CH : C, it forms metallic derivatives (see p. 91). (b) KETONES (R CO R') These have the carbonyl group attached to two alcohol radicals and, if the latter are similar, are known as simple ketones and, if different, mixed ketones. The first member must contain at least three carbon atoms. They present the same cases of isomerism as the secondary alcohols, and are metameric with the aldehydes. Up to the C u -compound they are liquid and beyond that solid, but all are less dense than water. They resist feeble oxidation but energetic oxidising agents (dichromate and dilute sulphuric acid) break the chain of the ketone at the carbonyl group, thus forming an acid with a lower number of carbon atoms: CH 3 -CO-CH 3 + 40 = H 2 + CO 2 + CH 3 -C0 2 H. In mixed ketones, however, the carboxyl is mainly united to the smaller alkyl radical (R or B'), but the acid with the higher alkyl is always formed to some extent. With ammonia, the action is different from that in the case of aldehydes : water is eliminated from 2 or 3 mols. of ketone and di- and tri-ketonamines (or acetona- mines), e.g. C 6 H 13 ON, formed. Further, the ketones do not polymerise, but they form condensation products. They do not react with ammoniacal silver solutions or with Fehling's solution, and are hence not reducing in character (difference from aldehydes). With phosphorus pentachloride they give the corresponding dichloro- derivatives ; for instance, acetone gives 2-dichloropropane, CH 3 -CC1 2 -CH 3 . On reduction, they yield secondary alcohols, and with very energetic oxidis- ing reagents (H 2 2 , &c.), they form characteristic polymerised ketonic per- oxides, e.g. [(CH 3 ) 2 C0 2 ] 2 , [(CH 3 ) 2 C0 2 ] 3 . With ethyl orthoformate they give acetals, (CH 3 ) 2 C(OC 2 H 5 ) 2 , and similarly with mercaptans they form Mercaptols, e.g. (CH 3 ) 2 C(SC 2 H 5 ) 2 , which, when oxidised with permanganate, gives Sulphonal, (CH 3 ) 2 C(S0 2 C 2 H 5 ) 2 . Ketones, which generally contain the group CH 3 -CO- form, with sodium bisulphite, compounds which are crystalline and hence readily separable from other substances : (CH 3 ) 2 CO + S0 3 HNa = (CH 3 ) 2 C ,-p,>>C 2 H4, melts at 104 and boils at 146. Tetramethylenediamine is called also Putrescine. TAURINE (Aminoethylsulphonic Acid), NH 2 -CH 2 -CH 2 -S0 3 H, is found in com- bination with cholic acid (as Taurocholic Acid) in the bile of various^animals and also in the lungs and kidneys. It forms monoclinic prisms soluble in hot water but insoluble in alcohol, and has a neutral reaction, the basic and acid groups neutralising one another. It is not hydrolysable. Of the derivatives of Olycerol, the CTilorTiydrins or esters of hydrochloric acid are of interest ; they are liquids soluble in alcohol or ether, and, to a less extent, in water. With hydrochloric acid, glycerol forms the Monochlorhydrin, C 3 H 5 (OH) 2 C1, of which two isomerides (- and ft-) are known, and the Dichlorhydrin, C 3 H 5 (OH)C1 2 , also existing in two isomeric forms. Either of these, when treated with PO 5 , gives the trichloro- derivative, C^sCl^ 1 At the present time interest attaches also to the formins and acetins, which are used in the manufacture of non -congealing explosives. 2 GLYCIDE ALCOHOL, CH 2 -CH-CH 2 .OH, is a liquid, b.pt. 162, soluble in alcohol \0/ or ether, and also in water, with which it gives glycerol again ; with hydrochloric acid it gives the chlorhydrin. It may be regarded as derived from glycerol by the removal of a molecule of water, and is prepared by the separation of HC1 from the a-monochlorhydrin by means of baryta. It is isomeric with propionic acid and reduces ammoniacal silver solution. Separation of hydrogen chloride from the dichlorhydrin yields Epichlorhydrin, CH 2 C-HCH 2 C1, which may be regarded as the hydrochloric ester of glycide alcohol. \0/ It boils at 117, has an odour like that of chloroform and is insoluble in water. It is isomeric with propionyl chloride and monochloroacetone. GLYCEROPHOSPHORIC ACID, OH,CH 2 .CH(OH).CH 2 .0-PO(OH) 2 , is optically active, as also are its calcium and barium salts (laevo -rotatory). It is interesting from the fact that when the hydroxyl-groups are esterified with palmitic, stearic, or oleic acid, and 1 According to Ger. Pat. 180,668, the monochlorhydrin is made by heating for 15 hours in an autoclave at 120 a mixture of 100 parts of glycerol with 150 parts of hydrochloric acid (sp. gr. 1-185). The water is distilled off and the residue subjected to fractional distillation in a vacuum (15 mm. pressure) ; after the acid and water have been eliminated, the monochlorhydrin distils over at 130 to 150, and the unaltered glycerine at 165 to 180. If it is to be nitrated and used for explosives, it is sufficient to get rid of the water and acid. According to Fr. Pat. 370,224, the monochlorhydrin may also be obtained by shaking glycerine with the calculated quantity of sulphur chloride at a temperature of 40 to 50 ; the water formed is distilled off in a vacuum at 60 to 70. The a-Monochlorhydrin, CH 2 C1-CH(OH)'CH 2 -OH, is obtained (according to Fr. Pat. 352,750) bypassing hydrogen chloride into glycerine heated to 70 to 100. Like glycerine itself, the chlorhydrins are easily nitrated, yielding non-congealing explosives (gee Inter). 3 MonoaceMn, C.3H 6 (OH) ? (O'COCH3),isobtained by heatingfor lOto 15 hoursat 100 a mixture of 10 parts of glycerol with 15 parts of 40 to 100 per cent, acetic acid, the weak acetic acid (25 to 30 per cent.) that distils over being condensed separately. Ten parts of 70 per cent, acetic acid are then added and the weak acid up to 40 per cent., which distils at 120 collected apart. After this the temperature is raised in 3 hours to 250, the weak acid still being kept separate. The crude monoacetin remaining contains about 44 per cent, of combined acetic acid and about 0-8 per cent, of the free acid. This acetin is soluble in water and serves well for the manu- facture of explosive and non-congealing nitroacetins (see Explosives) and for gelatinising the nitrocellulose of smokeless powders (Vender Ger. Pat. 226,422, 1906), EXPLOSIVES 215 the phosphoric residue united to choline, it gives rise to the important group of lecithins, which are optically active : OR CH 2 CH(OR') CH 2 . PO(OH) - O CH 2 CH 2 N(CH 3 ) 3 OH, where R and R' are fatty acid residues. Lecithins are found in the brain, yolks of eggs, and many seeds and are soluble in alcohol, and, to a less degree, in ether ; they give salts with acids and with bases and yield solid compounds with chloroplatinic acid or cadmium chloride. They are saponified by baryta, with formation of choline, fatty acids, and glycerophosphoric acid. Of the nitric esters of glycerine, the most important is trinitroglycerine, or trinitroglyceric ester, C 3 H 5 (ON0 2 ) 3 , which is one of the most powerful explosives. We shall hence study it from the industrial standpoint, first dis- cussing certain general notions concerning explosives. The manufacture of the latter constitutes one of the most interesting industries of organic chemistry, partly because of the varied mechanical appliances which it requires. EXPLOSIVE SUBSTANCES The name explosive substances, or explosives, is given, in general, to those solid and liquid bodies which, under the influence of heat, percussion, electrical discharge, &c., are transformed instantaneously and completely or nearly so into a gaseous mass with an enormously increased temperature. If the reaction takes place in a closed space, the gases thus produced and heated exert a very considerable pressure which can be immediately trans- formed into mechanical work, the enclosing substance and all the surround- ing objects being shattered with great violence and noise. Such a phenomenon (or effect) constitutes a so-called explosion, and if it attains very great rapidity and power it is termed a detonation. For a constant quantity of gas produced in an explosion, the effect will be the greater the higher the temperature developed in the reaction. THEORY OF EXPLOSIVES. The chemical reactions and physical phenomena of explosives are produced under conditions differing greatly from those in which physical and chemical properties of substances are usually studied. The pressures, temperatures, and velocities with which we have to deal in ordinary phenomena are of a very different order from that of the enormous pressures of the gases in the interior of the earth's crust, which are measured in hundreds of thousands or millions of atmospheres. So also the temperatures in various stars, e.g. in the sun, reach thousands of degrees, and the velocities of the planets hundreds of kilometres per second. The phenomena now to be considered, although they do not attain these enormous magnitudes, still do approach them. Indeed, explosions give pressures of tens of thousands of atmospheres, temperatures of thousands of degrees, and velocities (of projectiles) of thousands of metres per second. Almost all explosive substances contain oxygen (furnished by chlorates, nitrates, &c.), only very few, such as nitrogen chloride and iodide, and aniline fulminate, being without it. Mixtures of oxidising agents with readily combustible substances (sulphur, carbon, sugar, &c.) are explosive, but they are less powerful than those composed of single com- pounds which explode by themselves. This is because the elements necessary for complete combustion are in much greater proximity, being present in the molecule of the explosive itself ; examples of such explosives are nitroglycerine, guncotton, mercury fulminate, picric acid, &c. The determination of the theoretical power of an explosive requires a knowledge of : (a) the chemical reaction accompanying the explosion, so that the heat of the reaction, the temperature, and the volume and relative pressure of the gases formed can be deduced ; (b) the velocity of the reaction. In order to understand the theory of explosives, it is indis- pensable to call to mind the fundamental principles of thermochemistry and of thermo- dynamics, for which the reader is referred to the brief account given in vol. i, pp. 49 and 57. 216 ORGANIC CHEMISTRY (a) The chemical reaction is deduced from the difference in composition between the explosive and the products resulting from the explosion. When there is sufficient oxygen in the explosive to produce complete combustion, the nature and quantities of the gases can be calculated a priori, and from their heats of formation their temperature can be deduced. The total combustion of nitroglycerine, when exploded in a closed space, gives the following products (a) : 2C 3 H 5 (NO 3 ) 3 = 6C0 2 + 5H 2 + 3N 2 + 0. When there is deficiency of oxygen, as in guncotton and other substances, it is not easy to foretell the products of the reaction, as these vary with the conditions in which the explo- sion occurs, and usually several reactions take place simultaneously. Further the gases found after the explosion of such products are probably not always those formed at the instant of the explosion, as at such high temperatures certain substances (H 2 O, CO 2 , &c.) may undergo dissociation with absorption of heat. 1 (b) The heat developed in the explosion is deduced by calculation from the thermo- chemical data of the equation, but the practical result is not in accord with the theoretical calculation, since part of the heat (25 to 30 per cent.) that should theoretically be developed is transformed into mechanical work, which is what is utilised in practice. In calculating theoretically the heat of explosion, the heat of formation of the explosive (from the elements) is subtracted from the heat that should theoretically be developed in the reaction. The heat of explosion varies, however, according as it is determined at constant volume or at constant pressure ; in the latter case the explosion of nitroglycerine, for example, is effected in the open air, since then the volume varies, but the pressure is only that of the atmosphere. The heat cf formation of nitroglycerine from its elements (see p. 25) is given by the following equation (6) : C 3 + H 5 + N 3 + O 9 = C 3 H 5 (NO 3 ) 3 + 98 Gals. The heat of reaction of nitroglycerine can be calculated from equation (a) given above, from which it is seen that 2 mols. or 454 grms. of nitroglycerine yield 6C0 2 + 5H 2 + 3N 2 + O. The heat of formation of 6C0 2 is 6 x 97 = 582 Cals., and that of 5H 2 0, 5 x 68-5 = 342-5 Cals. For the nitrogen and oxygen there is no development of heat since they are not combined, so that the total heat of reaction calculated on the gases formed in the explosion of 2 grm.-mols. of nitroglycerine will be 924-5 (i.e. 582 + 342-5) Cals. From this must be subtracted the heat of formation from the elements of 2 mols. of nitroglycerine, since on decomposing under these conditions of temperature the explosive first of all liberates its atoms, absorbing as much heat as is evolved in its formation from its elements (reaction b), i.e. 196 (98 x 2) Cals. per 2 mols. The atoms thus liberated combine immediately to give the gases which result from the explosion, the heat of formation of which has already been calculated. The true theoretical heat of explosion at constant pressure for 454 grms. of nitroglycerine will hence be 728-5 (i.e. 924-5 - 196) Cals., or for a kilo, 1603 Cals. The heat of reaction at constant volume the explosion occurring in a closed vessel is rather higher, the heat corresponding with the expansion of the gas (see vol. i, pp. 26 and 50) not being absorbed as no expansion takes place ; theoretically the heat at constant volume is calcu- lated to be 1621 Cals. per kilo. 2 Serrau and Vieille, by direct practical measurements, found the heat of explosion of nitroglycerine at constant volume to be 1600 Cals., which confirms the accuracy of the calculation. With substances which themselves contain sufficient oxygen for complete combustion The following Table gives the percentage compositions of the gases resulting from the normal explosion of various explosives in the calorimetric bomb : CO CO 2 O 2 CH 4 Hj N 2 Nitrocellulose powder 46-87 16-8 0-08 1-26 20-44 14-9 Gelatine dynamites. Carbonite Picric acid Trinitrotoluene 34-0 32-68 0-75 10-0 21-0 36-0 19-2 2-8 27-6 14-4 61-05 3-46 0-34 1-02 13-18 21-1 57-01 1-93 0-11 . 20-45 18-12 For every gramme-molecule of a substance passing from the solid or liquid to the gaseous state, owing to the new volume occupied, 590 small calories (vol. i, pp. 26 and 50) are absorbed. In the explosion of 2 mols. of nitroglycerine, 14-5 mols. of gas (6C0 2 + 5H 2 O -f 3N 2 + O) are formed, and these, on expanding, will absorb 14-5 x 590 = 8550 small calories, or 8-5 Cals. per 454 grms. of nitroglycerine, i.e. 18 Cals. per kilo. This, added to 1603, the heat of reaction at constant pressure, gives 1621 Cals. as the. heat of reaction at constant volume. 217 during explosion, it is not easy to calculate theoretically the heat of explosion, since the products of the reaction are not exactly known ; in such cases, various direct practical determinations must be made. It is not easy to calculate theoretically the temperature of the gases at the moment of explosion, since the specific heat of the gases at such high temperatures cannot be deter- mined, but is certainly rather higher than the ordinary value. Further, at such tem- peratures dissociation phenomena occur which cannot be defined ; these, however, lower the temperature, although not greatly, since with the great pressures developed the disso- ciation is minimal. On the other hand, with the means we possess, it is not possible to measure these temperatures directly and only approximately can they be determined for black powder. In general, however, they are very high and in some cases exceed 4000 (for instance, by burning ballistite in the air, platinum withm.pt. 1800 is easily melted), but even these temperatures, deduced indirectly, are much lower than those calculated theoretically. 1 The temperature of ignition does not usually coincide with the temperature of explosion since explosion is caused not so much by the temperature as by the pressure and other factors to be considered later ; so that for explosion to occur, special conditions (detonators) are necessary. But for some substances, e.g. black powder, non -compressed guncotton, &c., the temperature of ignition, given in the following Table, is identical or almost so with that of explosion : Fulminate of mercury . . . 200 Non -compressed guncotton . . 220 to 250 Nitroglycerine . . . .218 (explodes'at 240 to'250) Black powder *>'' 288 There are thus explosives which explode when merely ignited with a match and others which are exploded indirectly by means of detonators^ The mechanical work, in kilogram-metres, yielded by an explosive is calculated by multiplying the number of calories developed in the explosion of 1 kilo of the substance by the mechanical equivalent of heat (= 425, see vol. i, pp. 50 and 51). For various explosives this mechanical work (or potential energy) is given in the following Table : Nitroglycerine . (1 kilo) = 1600 Gals, x 425 = 680,000 kilogram -metres Explosive gelatine . . . = 1530 = 650,000 Dynamite . . . . = 1178 = 500,000 Guncotton . . . . = 1074 = 456,000 Fine sporting powder . . = 849 = 360,000 Potassium picrate . . . = 780 = 330,000 Fulminate of mercury . . = 403 = 170,000 Nitrogen chloride . ' . . = 339 = 144,000 Owing to various causes, the total theoretical energy of explosives cannot be utilised practically ; e.g. the expansion of the gases at the moment the projectile leaves the cannon or gun, the friction, the heating of the barrel, &c., all constitute losses of the useful effect of the explosive. The volume of the gases formed in the explosion can be calculated with reference to and 760 mm., taking account of the fact that at the moment of explosion the water is in the state of vapour. But in practice it is of more importance to calculate the volume at the temperature of explosion, when a knowledge of the gases formed is possible, as is the case with nitroglycerine, and, in general, with explosives containing sufficient oxygen 1 Indeed, water-vapour, formed from H a + 0, should have theoretically a temperature of 7927 (see Calcula- tion, vol. i, p 378), but in the most favourable theoretical conditions the oxy-hydrogen flame does not exceed 2500. For carbon dioxide the heat of formation is 97,000 cals., and the specific heat 0-217, so that for 44 grms. 97 000 of C0 2 gas (grm.-mol.) the temperature attainable would be TJTTiESJS = 10,160, and allowing for the fact that along with the 6 mols. of CO 2 and 5 of H 2 O, the 3 mols. of N 2 and half a mol. of oxygen formed in the explo- sion of nitroglycerine are also to be heated the theoretical temperature of the gases from the explosion would ft be about 7000. This theoretical temperature is determined in general by the formula t = - TTTT ^-^ where p, p', p" . . . are the weights of the gases formed in the explosion, s, s', s" . . . their specific heats, and C the total heat in caloriesi 218 ORGANIC CHEMISTRY for their complete combustion. It is, however, not easy to calculate the volume of gas formed by products containing an insufficiency of oxygen, like guncotton, &c., with which the gases vary quantitatively and qualitatively according to the type of explosion ; in such cases the volume must be determined directly. The volume of gases is calculated (see vol. i, p. 34) by means of the general formula, 7 (1 + 0-00367 1) .. v t - p where V f is the required volume at the temperature of explosion t, V is the volume at and 760 mm. pressure (which can be found from the weight of the gaj?es formed), and 0-00367 is the coefficient of expansion for all gases. For such high temperatures and pressures, however, the coefficient of expansion is rather higher than that resulting from Gay-Lussac's and Boyle's laws, but this difference is compensated for by the somewhat higher specific heat of the gas at high temperatures, in consequence of which more than the theoretical quantity of heat is absorbed. The pressure of the gas is deduced from the general formula given above, V f being diminished by the volume v of the mineral, non-gasifiable residue (in the case of dynamite or other mixtures), so that : + 0-00367 Q V t -v with nitroglycerine, guncotton, &c., v = 0. P is the maximum theoretical force of an explosive, starting from its volume (solid) at the ordinary temperature, but the effect of a given explosive will be the greater as its density increases, that is, the greater the weight for the same volume ; and for guncotton, for example, the effect will be the greater for the same volume, the more it is compressed. Thus the relative specific gravities of different explosives are of importance, and in fact fulminate of mercury, which has a high specific gravity (five times that of ordinary powder and three times that of nitroglycerine), has a maximum rapidity of reaction and is the most powerful detonator, being capable of exerting a force of about 27,000 kilos per square centimetre (atmospheres), this being about treble that given by any other known explosive. In practice, pressures higher than any imaginable may be attained when the volume occupied by a given weight of explosive in a closed vessel is less than the critical volume of the gas developed, since this critical volume (vol. i, p. 28) cannot be diminished by any pressure, however great. If we term charging density the ratio between the weight of the explosive in grammes and the volume in cubic centimetres occupied by it in absolutely filling its envelope (as though it were liquid or fused), this charging density corresponds with the specific gravity of the explosive ; if this density equals or exceeds the reciprocal of the limiting volume ( - I into which the gases developed (critical volume) \v' can be compressed, the pressure attained will be infinitely great and will rupture any enclosing vessel, no matter how resistant it may be. The reciprocal of the critical volume of the gases produced in the explosion is termed the critical specific volume (or limiting density), and comparison of this with the density of charge leads to consequences of practical importance. Limiting density Specific gravity of of the gases the explosive Black powder . . . .2-05 .. 1-75 Nitroglycerine . . .1-40 . . 1-60 Powdered guncotton . . . 1-16 .. 1-20 Picric acid . . . .1-14 .. 1-80 Fulminate of mercury . .3-18 . . 4-42 .Thus, black powder has a charging density (or specific^gravity) of 1-75 to 1-82, which does not reach the limiting density, so that even if it is exploded in its own volume it does not break the envelope if the latter is strong enough to withstand the pressure developed, namely, about 29,000 kilos per square centimetre. For granular powder, the density of which is 1 , the pressure is only 6000 kilos. The real density (specific gravity) of compressed guncotton is 1 -2, that of nitroglycerine 1 -6, and that of picric acid 1 -8, all of these being VELOCITY OF EXPLOSION 219 superior to the limiting densities of the corresponding gases ; so that when they explode in their own volume, all of these explosives burst the most resistant envelope, and, in such cases, the velocity of the explosive wave becomes infinitely great. Fulminate of mercury, although it has the high limiting density 3-18 (owing to the low critical volume, v), has a specific gravity of 4-42 (to which the density of charge approximates) and behaves like nitroglycerine, &c. As it is difficult to calculate a priori the pressures exerted by explosives, it is preferable to determine them relatively by measuring certain effects of the ga?es at the instant of explosion ; this is done, for instance, by observing the crushing or deformation of small cylinders of copper or lead, which are termed crushers (Fig. 179). The total pressure depends on the character of the explosive and on the nature of the explosion (see later), but more especially on the density of charge. The specific pressure of an explosive is a constant (a), given by the ratio of the pressure P (p) to the corresponding density of charge (d) of the explosive itself : a = . This specific a pressure a is characteristic of any explosive and expresses the pressure developed ~by unit weight (1 grm.) of an explosive in unit volume. The specific pressure is not always the maximum pressure that can be exerted, this depending, as we have seen, on the charging density in its relation to the critical volume. Velocity of reaction. The duration of the explosion is of great importance, since on it depends the greater or FIG. 179. less utility of the explosive for different purposes. The more rapid the explosion the better is the heat developed utilised, so that this can be used almost entirely in heating and expanding the gases and so increasing the pressure considerably. If, however, the reaction is slow, a large portion of the heat is dissipated by radiation and conduction. Explosives with an extremely rapid reaction produce special effects, as they shatter the envelope or rock in immediate contact with the explosive into minute fragments an effect often not desired. These are termed shattering or detonating explosives and their properties are utilised in certain cases, as, for example, where a small cavity is to be made in a rock so that a large quantity of a progressive explosive may be subsequently introduced. If the reaction, although rapid, is not instantaneous, the explosion produces other effects, for instance, the cleaving of large stones or rocks and the projection of fragments nearer to the exolosive ; this progressive or rending action is the effect usually desired by miners. According as the gasification takes place more or less instantaneously (and the one or the other effect can be obtained with the same substance by adding inert materials to, say, dynamite, or mixing paraffin with guncotton), explosives are more or less shattering. Thus, panclastite is more shattering than guncotton, the latter more than dynamite, and this more than smokeless powder, which is a progressive explosive. Many substances explode only with detonators (of fulminate of mercury) and the cause of the explosion in such cases is not only the high temperature produced by the explosion of the detonator, but more especially the great immediate pressure resulting from the instantaneous production of gas, this pressure and the sudden shock provoking the decomposition of the molecules of the explosive (Berthelot, Abel, Vieille). The duration of explosion or of gasification of the detonator is 500 times less than that of the explosive material, and the greater relative amount of heat developed in "a certain time by detonators explains their greater shattering power compared with that of progressive . explosives. The most highly shattering materials are : fulminate of mercury, panclastite, compressed guncotton, and nitroglycerine. The duration of reaction for detonators is only about T ^ of a second, the extraordinary effect of these explosives being due to the enormous amount of energy developed (1600 Cals. for nitroglycerine) in this short time and in the small space containing them. 1 1 The velocity of combustion (or of deflaaratron) is sharply distinguished from the velocity of the explosive reaction and is made use of it in certain cases, e.g. in the throwing of projectiles (expansive and progressive action)j The velocity of combustion of explosives depends on, and increases with increase of, the pressure at which they decompose. Another factor influencing the mode of combustion of explosives is the maximum velocity with which the pressure develops. The exponent of the power of the pressure, which admits < f passing from one value to the other in the increase 220 ORGANIC CHEMISTRY As has been already stated, the shattering effect of a substance is rendered evident by exploding a few grammes of it on a cylinder of metal (crusher) and the actions of different explosives are compared by means of these deformed and disfigured crushers. Fig. 180 B shows a leaden cylinder before the explosion, whilst A shows the same cylinder after 10 grms. of dynamite (a progressive explosive) have been exploded on it and C the result of the explosion of 10 grms. of panclastite (from nitrotoluene). One and the same explosive substance may be made to give either a shattering or a progressive effect by varying the velocity of the reaction, this usually depending on the power of the initial shock which causes the explosion. The more powerful the initial shock the greater is the amount of kinetic energy transformed into heat and hence the higher the temperature developed ; therefore, also, the greater is the pressure of the resultant gases and the more rapid and powerful the effect of the explosive. The effects vary considerably with the manner in which the explosion is induced ; thus, if a flame is brought near to non-compressed guncotton, the latter burns rapidly but does not explode ; whilst if it is compressed and subjected to the action of a cap (detonator) of fulminate of mercury, a real and very powerful ex- plosion occurs ; similar pheno- mena are observed with nitro- glycerine and dynamite. 1 DETERMINATION OF THE EXPLOSION. In order to induce the explosive re- action of a substance, it is sufficient to bring it at a single point to a certain initial decom- A position temperature (by percus- FIG. 180. sion, detonation, &c.), the sharp decomposition at this point then producing a new shock which heats the neighbouring points to the decomposition temperature, and so on, the explosion being thus communicated to the whole mass by a true explosive wave, which is enormously more rapid than simple burning. From this will be understood the great importance of detonators, which do not serve merely for ignition ; and the difference will also be apparent between an ordinary explosion by ignition and percussion and that induced by fulminate of mercury detonators. When the phenomenon of explosion is studied more closely, it becomes evident that the gases produced at the point of ignition tend to expand and hence to diminish the pressure at that point and also the rapidity of explosion, but if this initial expansion is impeded the pressure and hence the velocity of decomposition increase rapidly. In practice miners obtain this effect by filling the cavity containing the explosive with a of the pressure, is called the modulus of progressirity, and serves to characterise the various explosives. Thus, this modulus varies from 1-25 to 1-50 for black powders and from 1-86 to 1-87 for smokeless powders, whilst that of picric acid is 2-82 and that of "Earner's explosive (12 per cent, of dinitronaphthalene -f 88 per cent, of ammonium nitrate) 3-25. As will hence be seen, these last two explosives have the dangerous property of furnishing accidental superpressures, owing to undulatory phenomena which always accompany the combustion of substances in- flammable with difficulty. In smokeless powders, the moderate progressivity compared with the great power constitutes a valuable safeguard in their use in firearms ; in this they are surpassed only by black powders, which are, however, much less powerful. 1 The percussive force (kinetic energy) of an explosive serves best to establish the shattering power and is calcu- lated by C. E. Bichel by means of the formula -, where m denotes the mass of the gases formed in the explosion, or the weight of the explosive, divided by 9-81 and v is the velocity of detonation (i.e. the time elapsing from the beginning of the explosion to its completion throughout the whole mass). For 1 kilo of an explosive gelatine (92 per cent, of nitroglycerine and 8 per cent, of collodion cotton) with the charging density, 1-63, Bichel gives a velocity of detonation of 7700 metres per second, so that the percussive force in absolute units will be : 1 X 7700 2 "9^1 x 2 = 3>021 ' 916 kilogram-metre-seconds ; for black powder (with a charging density 1-04) exploded under the same conditions in a closed vessel with a detonating cap, the velocity of detonation is 300 metres per second, 1 x 300 so that 9 gl x 2 = 4587 kilogram-metre-seconds ; for kieselguhr dynamite (35 per cent, nitroglycerine) the velocity of detonation is 6818, and hence the percussive force, 2,369,272 kilogram-metres per second ; for a gdatine-dynamite (63-5 per cent. 1 nitroglycerine, 1-5 per cent, collodion cotton, 27 per cent, sodium nitrate, 8 per cent, wood meal), with a charging density of 1 67, the velocity of detonation is 7000 and the percussive force 2,497,452 ; for trinitro- toluene, with a [charging density of 1-55, the velocity was 7618, and the percussive force 2,957,896 ; guncotton, with a charging density of 1-25, had a velocity of 6383, the percussive force being 2,076,589 ; and picric acid, with a charging density of 1-55, gave the velocity of detonation 8183, and the percussive force 3,412,920 'kilogram- metre-seconds. EXPLOSION BY INFLUENCE 221 tamping of earth or stone. The same end may also be attained by increasing considerably the mass of the explosive and the surface of ignition, and this explains why certain sub- stances burn, without exploding, in small quantities (guncotton, nitroglycerine, &c.), or when the ignition is confined to a limited area, whilst a powerful explosion may occur when a large quantity of explosive is used or when it is surrounded by a source of considerable heat. For shattering explosives (e.g. fulminate of mercury) no tamping is used, since the reaction is so rapid that the atmospheric pressure, that is, the air itself with its inertia, is sufficient to maintain the pressure of the gases. Even fulminate of mercury, if ignition is effected by an electric contact (which heats a platinum wire to redness) and under an evacuated bell-jar, burns without exploding, thus confirming the tamping action of the air in the case of detonators and even of ordinary explosives ; in fact, if a roll of dynamite is exploded on a bridge, the latter is cut in two owing to the tamping action of the air. The explosive, wave produced in the explosion of gaseous mixtures and of liquids and solids is only slightly related to waves of sound. The latter is transmitted from crest to crest with but little kinetic energy, with a small excess of pressure and with a velocity depending only on the nature of the medium in which it is propagated and of equal magnitude for all kinds of vibrations. The explosive wave, on the contrary, propagates the chemical transformation through the mass of the explosive substance, communicating from point to point of the decomposing system an enormous amount of potential energy and a great excess of pressure. The sound-wave is propagated in a mixture of hydrogen and oxygen with a velocity of 514 metres per second at 0, but the velocity of the explosive wave in the same mixture (exploded at a point) is 2841 metres. With guncotton, the velocity of this wave varies from 3800 to 5400 metres per second according to the compression ; with nitroglycerine it is 1300, with dynamite 2700, with picric acid 6500, and with nitromannitol 7700 metres per second. This velocity depends only on the nature of the explosive and not on the pressure, but it varies to some extent with the nature of the envelope. For instance, in a rubber tube having a thickness of 3-5 mm. and an internal diameter of 5 mm. and covered with cloth, ethyl nitrate gives a velocity of 1616 metres ; whilst in glass tubes of various diameters and thicknesses the value is 1890 to 2480 metres. The propagation of the explosive wave bears no relation to that of ordinary combustion (which is much slower). The former occurs when the inflamed gaseous molecules acquire the maximum velocity or energy of translation, i.e. act with the whole of the heat developed in the chemical reaction. Explosion by Influence. If a long row of dynamite cartridges are arranged on a flat solid at distances of 30 cm. or on a metal disc at a distance of 70 cm., explosion of the first with a fulminate cap results in the rapid and successive explosion of the remaining ones simply by influence and without the need of detonators or fuses. Air does not conduct the wave of explosive influence as well as solids, and if the cartridges are suspended in the air by wires such explosion by influence does not occur. Water conducts the explosive wave to a certain distance, but the influence gradually diminishes with increasing distance from the centre of explosion (there have been cases in which the shock of a large charge of guncotton has exploded neighbouring torpedoes ; to avoid these inconveniences, so-called safety explosives are now used). These explosive waves are first propagated through the explosive itself, not by a single shock which would gradually weaken as it advanced but by a very rapid series of such shocks produced by the propagation of the explosion from point to point of the whole mass of the explosive, the kinetic energy being thus regenerated along the whole course of the wave in the exploding substance. An explosive wave is thus distinguished from an ordinary sound-wave by the fact that the latter becomes enfeebled as it advances, whilst the former is characterised by the uniformity of the energy transmitted from point to point by a series of numerous and successive explosions throughout the exploding mass. Only the last of these explosions is transmitted with its energy to the surrounding air and to the matter on which the explosive rests, and, since it is no longer reinforced (by other shocks), it weakens as it becomes more remote. Hence explosion by influence is not due to the fact that the distant explosive transmits or propagates the explosive wave through its own mass, but is owing to the arrest and transformation, at the point of impact, of the mechanical energy it being capable 222 ORGANIC CHEMISTRY of similar (but not all) waves into heat energy, able to cause decomposition and explosion of the substance itself. The effects of large charges of dynamite (25 to 1000 kilos) when freely exploded are dangerous to buildings and to life for a distance of 500 metres and are felt as far away as 3 kilometres (L. Thomas, 1904). CLASSIFICATION OF EXPLOSIVES. Explosives are to-day so numerous and are prepared from such different mixtures and serve such a variety of purposes that a rigorous or rational classification is difficult or impossible. Also with a large number of classes there would be many substances which might belong to more than one of them. It will hence be preferable to limit ourselves to a description of the various explosives without any prearranged classification. They will be taken in the following order : (1) Dynamites with a basis of nitroglycerine; (2) Nitrocellulose ; (3) Various smokeless powders ; (4) Picrate powders ; (5) Explosives of the Sprengel type (the components are explosive only when mixed) ; (6) Sundry explosives ; (7) Black nitrate and other powders ; (8) Chlorate and Perchlorate powders. NITROGLYCERINES This name is given improperly to nitric esters of glycerine since they do not contain true nitro-groups (N0 2 ) united directly with carbon as is often the case in benzene deriva- tives. On the contrary, the union is effected through an intermediate oxygen atom, so that these compounds should rather be called nitrates of glycerine. Being a trihydric alcohol, glycerine can form three such compounds, the only one known until quite recently being trinitroglycerine containing 18-5 per cent, of nitrogen and having very considerable industrial importance. In 1903, Mikolajczak prepared also pure DINITROGLYCERINE, C 3 H 5 .OH(ONO 2 ) 2 , containing 15-4 per cent, of nitrogen, and he proposed to use it as an explosive, as it possesses almost all the ballistic advantages of trinitroglycerine and is not easily frozen ; it is, however, very hygroscopic and readily soluble in water and in acids. Dinitroglycerine is prepared by nitrating 100 parts of glycerine with 400 parts of nitric sulphuric mixture containing 8 to 12 per cent. H 2 O, 60 to 70 per cent. H 2 S0 4 , and 15 to 32 per cent. HNO 3 ; at the end of the reaction, the mass is poured into an equal volume of water, and the acid neutralised with calcium carbonate, when the dinitroglycerine separates as a dense, floating oil. During the reaction, the temperature is maintained at 18 to 20 by cooling with ice. Dinitroglycerine is also formed by dissolving trinitroglycerine in sul- phuric acid and then diluting the solution with a little water. In whatever way it is pre- pared (e.g. by treating 1 part of glycerine with 2 parts of sulphuric acid, separating by means of lime the glycerinedisulphuric acid formed and treating this with nitric acid, as proposed by Escales and Novak, 1906), a mixture of the two possible isomerides is always obtained : dinitroglycerine K (i.e. ay-), N0 3 -CH 2 -CH(OH)-CH 2 -N0 3 , and dinitroglycerine F (i.e. a (3-), NO 3 CH 2 CH(NO 3 ) CH 2 . OH, which was studied by W. Will (1908). The mixture forms an almost colourless, faintly yellow oil, sp. gr. 1-47 at 15, which freezes at below 30 to a glassy mass, this distilling almost undecomposed at 146 under reduced pressure (15 mm.) ; at 15 it is soluble to the extent of 8 per cent, in water and at 50 to the extent of 10 per cent. In dilute sulphuric or nitric acid it dissolves in all proportions and by sulphuric acid (up to 70 per cent.) it is transformed into mononitroglycerine and then into glycerine. It is very hygroscopic and, when dry, dissolves or gelatinises nitrocellulose (guncotton or collodion-cotton) very well. The two isomerides can be separated by taking advantage of the fact that, in the air, the F compound absorbs 3 per cent, of water and is transformed into a crystalline hydrate, 3(C 3 H 6 7 N 2 ) + H 2 0, whilst the other remains liquid. The J^-form gives a nitrobenzoyl-derivative melting at 81, the corresponding compound of the Jf-isomeride melting at 94. In the dry state, the dinitroglycerines are as useful for explosives as the trinitro -compound, but when moist they are much inferior. A mixture of 50 per cent, dinitro- and 50 per cent, trinitro -glycerine freezes below 20. Of MONONITROGLYCERINE, C 3 H 5 (OH) 2 .NO 3) the pure a- and /3-isomerides are known (W. Will, 1908). These are not true explosives and dissolve to the extent of 70 per cent, in water. The n -compound melts at 58 and boils at 155 to 160 under 15 mm, pressure. NITROGLYCERINE 223 Nitrochlorhydrin, C 3 H 5 C1(NO 3 ) 2 , and Tetranitrodiglycerine (see p. 184) have also been proposed as non-congealing explosives, but better still for this purpose are the nitroacetins (V. Vender) (see later). 1 CH 2 O-NO 2 TRINITROGLYCERINE, CHO-N0 2 or C 3 H 5 (O.N0 2 ) 3 . CH 2 O.NO 2 This was discovered in 1846 by Ascanio Sobrero, 2 who called it Pyroglycerine and established its explosive properties but regarded its industrial manufacture as too dangerous. Its chemical composition was determined by Williamson in 1854. At first it was used only in small doses as a medicine, owirg to its marked power of inducing dilatation of the blood-vessels. Later, after various unavailing attempts, Alfred Nobel succeeded in applying it industrially, and in 1863 established two nitroglycerine factories in Sweden, these rapidly prospering owing to the great demands of various nations for this powerful explosive. Nevertheless, owing to the neglect of precautions by consumers in the handling of nitro- glycerine, various terrible explosions occurred which almost resulted in the abandonment and prohibition of this substance. Fortunately just at this time Nobel discovered a very happy solution of the problem which completely eliminated this danger, by mixing the nitroglycerine with inert substances (kieselguhr or infusorial earth) and thus obtaining dynamite, this being to-day at the head of the great explosives industry. PROPERTIES. When pure it is a dense almost colourless or faintly yellow liquid of sp. gr. 1-6 at 15, and when it freezes its density increases by almost one-tenth, it is odourless and has a sweetish, burning taste. It is almost insoluble in water (0-16 to 0-20 per cent, being dissolved at 15), is not hygroscopic, and dissolves easily in concentrated alcohol, ether, benzene, chloroform, glacial acetic acid, toluene, nitrobenzene, acetone, olive oil, and concentrated sulphuric acid, and to a less extent in nitric acid and still less in hydrochloric acid ; it is, however, insoluble in carbon disulphide, glycerine, petroleum, vaseline, turpentine, benzine, and carbon tetrachloride. In solu- tion it will not explode. It evaporates spontaneously and in very small quantities even at 50, and if gradually heated to 109 it begins to decompose with evolution of brown nitrous vapours. Its specific heat is 0-356, and its heat of solidification 23 to 24 Cals. Dinitromonochlorhydrin is obtained, according to F. Roewer (1906), by nitrating the monochlorhydrin in the same manner as glycerine is nitrated (see later), and is then quickly separated from the top of the nitric-sulphuric acid mixture as an oil which is easily rendered stable by washing with water and soda. It forms a faintly yellow, mobile oil of aromatic odour, sp. gr. 1-541 at 15, soluble in alcohol, ether, acetone, or chloroform, but insoluble in water and in acids. At 180 it gives yellow vapours, and at 190 boils without detonation or deflagration, and with only slight decomposition ; under a pressure of 15 mm. it distils unchanged at 121 to 123 as an almost colourless oil. It is much more stable towards pressure than nitroglycerine, although possessing almost the same explosive properties. It does not freeze even at 30 and is not hygroscopic. It dissolves nitrocellulose, forming explosive gelatine, and mixes readily with nitroglycerine, giving non-congealing dynamites (with 5 to 20 per cent, of nitrochlorhydrin, Ger. Pat. 183,400), these being prepared by nitrating directly -a mixture of glycerine and chlorhydrin. In order to avoid the inconvenient effects on miners of the hydrochloric acid formed in the explosion of nitrochlorhydrin, potassium nitrate is added ; during the explosion this is transformed into potassium carbonate, which neutralises the acid. Dinitroacetylglycenne, C 3 H 6 (ONO 2 ) 2 (OCOCH 8 ), is obtained by nitrating the monoacetin in the same apparatus as is used for nitroglycerine, but using an acid mixture containing a preponderance of nitric acid, e.g. 65 per cent. HNO, and 35 per cent. H-jSO,. The dinitroacetylglycerine being somewhat soluble in water, it is lost to some extent during the washing. It is a yellowish oil, sp. gr. 1-45 at 15, and is soluble in alcohol, acetone, ether, nitroglycerine, or nitric acid, and almost or quite insoluble in water, benzene, or carbon disulphide. It contains 12-5 per cent, of nitrogen and with double its weight of nitroglycerine gives a mixture with 16-5 per cent, of nitrogen, which has a lower freezing-point (below 20) than any other mixture of these substances. It serves well for preparing non-congealing dynamites, and as it dissolves nitrocellulose easily it can be used for gelatinising smokeless powders. Uinitroformylglycerine, C 3 H 6 (ONO 2 ) Z (O- CHO), is prepared in a similar manner to the preceding com- pound, or, together with nitroglycerine, by nitrating the product obtained by heating 2 parts of glycerine with 1 part of oxalic acid for 20 hours at 140. Nitroformin and nitroacetin have explosive powers rather inferior to that of nitroglycerine. 1 Ascanio Sobrero was born at Casalmonferrato on October 12, 1812. He first studied medicine and then chemistry. In 1840 he went to complete his chemical studies in the laboratory of the celebrated Pelouze at Paris, where he stayed two years, and in 1843 he worked in Liebig's laboratory at Giessen. In 1845 he became Professor of Applied Chemistry at Turin, where he taught until 1883. He died on May 26, 1888, after a modest life, during which he filled various honorary social positions. It was always his aim that science should not be made a pretext or means of dishonorable undertakings or of business speculations. 224 ORGANIC CHEMISTRY At a red heat it evaporates without decomposing, but if it begins to boil vigorously during the heating, there is danger of explosion. According to Champion, pure nitroglycerine in small quantities boils, giving yellow vapours, at 185, evaporates slowly at 194, and rapidly at 200, burns quickly at 218 and detonates with difficulty at 241, violently at 257, feebly at 267, and feebly with flame at 287 (being in the spheroidal state). When heated in small quantities in the Bunsen flame, it burns without exploding, and if spread in a thin layer on paper it ignites with difficulty and burns only partially. Explosion of nitroglycerine can be induced either by violent percussion at a temperature of 250, or by energetic detonation (e.g. by explosion of fulminate of mercury). Nitroglycerine may be easily supercooled below its solidifying point. Kast (1906) showed that nitroglycerine represents a case of monotropic allo- tropy (see also vol. i, p. 191), i.e. it has two freezing-points, + 12 and + 13-5, corresponding with different crystalline forms. 1 When frozen, nitroglycerine explodes with more difficulty than in the liquid form. Pure nitroglycerine will not redden blue litmus paper or turn starch paste and potassium iodide blue, unless it contains free acids or nitrous compounds due to partial decomposition. Impure nitroglycerine readily decomposes and may explode spontaneously, whilst in the pure state it keeps indefinitely. A sample of nitroglycerine (200 grms.) prepared by Sobrero in 1847 is still kept under water in the Nobel factory at Avigliana. When decomposing, nitroglycerine turns green owing to the formation of N 2 and N 2 3 ; C0 2 , CO, H 2 O, N, and (see also p. 216) are also successively formed. In exploding, 1 litre of nitroglycerine produces 1298 litres of gas, which, at the temperature of explosion, occupies a space of 10,400 litres. L In large doses nitroglycerine is poisonous and its vapour causes headache (especially at the back of the head), giddiness, and vomiting. These effects are produced even by working with or simply touching nitroglycerine and are cured by means of cold compresses on the head, by breathing fresh, pure air, and by drinking coffee and taking suitable doses of morphine acetate. Workmen who handle the nitroglycerine paste during the manufacture of the various dynamites become habituated to it in two or three days and afterwards feel no ill -effects. Nitroglycerine is moderately easily decomposed by alcoholic potassium hydroxide (with separation of glycerine), and, when necessary, this reaction is employed to destroy and render harmless small quantities of nitroglycerine ; similarly benches or floors on which nitroglycerine is spilt are washed with caustic alkali solutions : C 3 H 6 (ON0 2 ) 3 + 5KOH = KNO 3 + 2KNO 2 + CH 3 .COOK + H-COOK + 3H 2 ; a little ammonia is also formed. With reducing agents it gives ammonia and glycerine, whilst with concentrated sulphuric acid it yields nitric acid and glycerinsulphuric acid. 1 Both nitroglycerine and dynamites and smokeless powders prepared from it are liable to solidify, and although they are then more stable the thawing is accompanied by danger, and when not carried out with great precautions has often led to fatal explosions, these being sometimes caused by the mere rubbing of the crystals. Indications will be given later of the precautions taken in magazines to prevent freezing, and mention may be made here of the attempts which have been made to render nitroglycerine non-congealable. As early as 1895 it was proposed to add nitrobenzene to nitroglycerine to lower the freezing-point, and later the use of orthonitrotoluene was suggested ; but the practical results were not very satisfactory in either case, the depression of the freezing-point being very small. Substances were required which were almost as explosive as trinitroglycerine, and were insoluble in water and stable on heating, and, in addition, were good solvents for nitrocelluloses (for making smokeless powders). These conditions were well satisfied by the nitroformins and nitroacetins tested by Nobel as early as 1875 but rendered practically useful in 1906 by V. Vender. The best results are given by dinitromonoacetin which is obtained from the monoacetin of glycerine prepared by the ordinary method used for esterifying alcohols with acids (see later, Esters). Forty parts of the monoacetin are introduced slowly into a mixture of 100 parts of nitric acid (sp. gr. 1-530) and 25 parts of oleum or Nordhausen sulphuric acid (containing 25 per cent, of free SO 3 , see vol. i, p. 275), the mass being cooled so that the temperature does not exceed 25. The whole is then poured into water and washed with cold and afterwards with hot (70) dilute soda. By this means an oil is obtained having sp. gr. 1-45 and containing 12-5 per cent, of nitrogen ; it is insoluble in water, carbon disulphide or benzene, but dissolves un- changed in nitric acid, nitroglycerine, methyl or ethyl alcohol, acetone, acetins, &c. Even in the cold, it has considerable solvent and gelatinising power for collodion-cotton and guncotton (with 13-4 per cent, of nitrogen) and the resulting explosive gelatines do not freeze even at 20. Naukhorf (1908) has proposed the addition of nitromethane or nitroethane to dynamite to lower its freezing-point, and at the present time liquid dinitrotoluene is largely used for the same purpose. MANUFACTURE OF NITROGLYCERINE 225 Characteristic Reactions. According to Weber, small quantities of nitroglycerine are detected by treatment with aniline and concentrated sulphuric acid : a reddish purple coloration is obtained which turns green on addition of water. To establish the purity and keeping qualities of nitroglycerine, the nitrogen is determined and Abel's heat test carried out (see later, Testing of Explosives) ; if it is satisfactory, 2 c.c. of it withstands 20 to 30 minutes' heating at 82 without giving sufficient nitrous vapours to be detectable by means of starch and potassium iodide paper. This reaction is, however, given by nitroglycerine* kept for a few days at a temperature exceeding 45, or for a long time below this temperature. PREPARATION. It is obtained by the action of a mixture of nitric and sulphuric acids on glycerine : C 3 H 5 (OH) 3 + 3HN0 3 = 3H 2 + C 3 H 5 (N0 3 ) 3 . The mono- and dinitro-compounds are probably formed as intermediate products of this reaction. The presence of sulphuric acid, which plays no apparent part in the change, is usually regarded as being necessary to maintain the nitric acid at a high concentration, i.e. to decompose the hydrates formed by nitric acid with the water from the reaction (KN0 3 , H 2 HN0 3 , 3H 2 0) and so regenerate mono- hydrated nitric acid, which acts on the glycerine (Kullgren, 1908). If the func- tion of the sulphuric acid were merely to fix the water, phosphoric acid could be used in its place ; but if this is done no nitroglycerine is obtained. The excess of the nitric-sulphuric mixture which is always used helps to produce a moderately complete separation of the nitroglycerine, which has a slightly lower density, so that it is possible to recover the acids employed. Although nitroglycerine is soluble in sulphuric or nitric acid alone, it does not dissolve in the mixed acids. But if one of the two acids is in large excess, a con- siderable amount of nitroglycerine remains in solution and is lost. In the nitra- tion, the whole of the glycerine cannot be added at one time, since sufficient heat would in that way be developed to produce decomposition and explosion of the nitroglycerine instantaneously formed. It is also not convenient to reverse the operation, that is, to add the mixed acids gradually to the glycerine, the greater density of the latter rendering rapid and homogeneous mixing difficult ; it is hence preferable to run the glycerine slowly into the acid mixture and to keep the latter continually and thoroughly stirred and cooled. MANUFACTURE. The theoretical proportions of the reacting substances 1 would be 100 parts by weight of glycerine and 205-43 of pure nitric acid, the theoretical yield of trinitroglycerine being then 246-74 parts. But on a large scale the whole of the nitric acid does not come into immediate contact with the whole of the glycerine, and it is hence better to use a slight excess of nitric acid (240 parts or even more) ; the amount of sul- phuric acid employed always exceeds that of the nitric acid (about if times). In modern factories the following proportions are often used : 100 kilos of glycerine, 240 to 260 kilos of nitric acid (98 per cent.), and 340 to 360 kilos of sulphuric acid (96 to 98 per cent.). In the best factories, the practical yield is 215 to 232 kilos of nitroglycerine per 100 of glycerine, but in some cases it amounts to only 205 to 210 kilos. Good yields are obtained by cooling the acid mixture during nitration by means of solutions from cooling machines, the temperature of reaction being kept down to about 10. The low value of the practical compared with the theoretical yield (246-7) is due to the fact that towards the end of the reaction there is very little free nitric acid and the last 1 The prime materials used in the manufacture of trinitroglycerine should be subjected to rigorous control ; the glycerine should be pure and distilled and should satisfy the requirements indicated on p. 188. The nitric acid should have a specific gravity of 1-500 (48 B6. or about 95 per cent. HNO 3 ) and should not contain more than 1 per cent, of nitrous acid, i.e. it should not be yellow, as otherwise an increased amount of heat is evolved during nitration and the yield is lowered. The sulphuric acid should be pure, with a sp. gr. of 1-8405 (i.e. at least 96 per cent. H 2 SO 4 ) and acid containing more than 0-1 per cent, of arsenic should be avoided ; lead and iron should also be absent as they might lead to reduction. When nitrations are carried out with nitric-sulphuric acids almost free from water (1 to 2 per cent.) the sulphuric acid is replaeed by oleum or Nordhausen acid (see vol. i, p. 275), i.e. acid containing 20 per cent, or more of dissolved sulphur trioxide. 15 226 ORGANIC CHEMISTRY portions of glycerine added are nitrated only with difficulty and hence remain dissolved in the sulphuric acid. The mixture of nitric and sulphuric acids, which is prepared separately, is ir.ade by pouring the sulphuric acid slowly into the nitric acid (not vice versa) in an iron vessel, the mixture being kept well cooled and stirred. With this procedure there is no danger of the acid spurting, and no production of nitrous fumes, since the development of heat is gradual. This mixture is forced by means of elevators (Montejus) or pulsometers working with compressed air (vol. i, p. 264) into tanks which feed the leaden apparatus in which the glycerine is nitrated. During recent years, many vitriol and explosives works have made considerable use of Kuhlmann emulsors (or Mammoth pumps) for raising concentrated acids, which are rendered lighter by emulsification with air (see illustration, vol. i, p. 265). The leaden nitration apparatus is shown in Fig. 181. It is surrounded by a wooden jacket inside which water circulates. Inside the vessel are peri- pheral leaden coils through which large quantities of cold water are continually passed by means of the two tubes D. The tubes C lead dry compressed air to the bottom of the liquid, which is thus kept thoroughly mixed. The tube F serves as exit for the air, and for any nitrous vapours which may be evolved and may be observed through the window, / ; these vapours are recovered in small condensation towers sprinkled with a little water. The cold acid mixture is first introduced through the pipe G. The glycerine, at a temperature of 20 to 25 (if colder it would be too viscous), is measured in the reservoir, M, and is passed, by means of compressed air supplied through O, slowly into the tube H, and thence into a perforated circular pipe at the bottom of the appa- ratus. Two thermometers, E, show the temperature of the reacting mass at any moment. The bottom of the apparatus is slightly inclined and at the lowest part is inserted a large stoneware tap, K, with an ebonite screw containing an aperture of at least 5 cm. It is convenient to have two of these taps so that, in case of danger, the whole of the mass may be rapidly discharged into a vessel of water underneath (drowning of the nitroglycerine). In such an apparatus, the same quantity of nitro- glycerine is produced each time and the treatment of 100 kilos of glycerine requires less than half an hour. 1 In America as much as 2000 kilos of glycerine are worked at one time in open vessels provided with stirrers, but the risk, in case of explosion, is greatly increased. At the conclusion of the operation the nitro- glycerine (sp. gr. L-6) floats on the acids (sp. gr. 1-7) and is separated by means of a suitable decanting apparatus (Fig. 182) to the bottom of which the whole mass is passed through the tube K. The apparatus consists of a leaden tank with its base sloping towards 1 The temperature during the reaction should not exceed 25 to 30, and it can be regulated by passing the cooling water more or less rapidly through the coils, and, if necessary, through the wooden jacket ; increase of the air-current also helps to lower the temperature. Rise of temperature and consequent explosion were at one time due principally to the use of impure glycerine, but nowadays it is generally due to slight escape of water from the coils. In order to avoid such danger, the apparatus and coils are tested at least once a day, usually in the evening when the plant is free ; water under pressure is forced into the coils and jacket and left until the morning, when any leak can be detected. Although the apparatus is constructed of vety thick plates, the lead corrodes in time ; tests made with aluminium apparatus (proposed by Guttler) have not been very successful. Some works now employ more solid vessels of wrought or cast iron, which arc more easily cooled. Boutmy and Faucher avoid the dangers of violent reactions by first dissolving, e.g. 100 parts of glycerine in 320 of sulphuric acid and then pouring the solution into a mixture of 280 parts of nitric and 280 of sulphuric acid. After 12 hours the reaction is complete, the yield being 190 per cent., calculated on the weight of the glycerine taken. This method did not give good results in England, but lias bei-n applied in France. Kurtz increases the yield and accelerates the reaction by emulsifying the glycerine with air and passing it under the acid mixture, a more intimate mixture being thus obtained. FIG. 181. ITROGLYCERINE PLANT 227 FIG. 182. the centre and supported by a wooden structure ; the cover, C, is raised on wooden joists, B. The tube D, with the glass window, E, serves to carry off any gas which may be evolved ; a thermometer is inserted into the vessel at t. The tube shown at the bottom and in the centre of the appa- ratus communicates with two or three taps, H, and is also fitted with a window, F. After half an hour, the nitro- glycerine in this vessel separates into a distinct layer, as may be seen through /. The surface of separation of the two layers coincides very nearly with the tap J ', so that the nitroglycerine can be discharged almost" com- pletely through the tube J into the lead-lined wooden tank, L. The acid that remains is dis- charged through one of the taps, H, it being noted through F when a turbid layer appears, as this separates the acid from the nitro- glycerine and contains various nitro -products and certain im- purities. 1 The tap, H, is then closed and this liquid is passed through other taps into suitable washing and decanting vessels (see later). The nitroglycerine in L is washed with water 1 The acid separated from the nitroglycerine and containing about 72 per cent. H 2 SO 4 , 9 per cent. HN0 3 , 16 per cent. H 2 O, and 3 per cent, of dissolved nitroglycerine, is collected in leaden tanks in which it remains for one or two days, during which time a small quantity (about 0-5 per cent.) of nitroglycerine separates at the surface. The dangers of this slow separation are some- timrs avoided by neglecting the nitroglycerine which separates after 4 to 5 hours ; to avoid danger in succeeding nitrating operations, a large proportion of the nitroglycerine remaining dissolved is decomposed by adding cautiously 4 to 5 per cent, of water so as to raise the tem- perature to 35 to 40 and then again mixing the mass by means of air (part of the trinitro- glycerine is thus transformed into soluble dinitro- glyeerine). These recovered acids, which are utilised again, are first denitrated in the appa- ratus shown in Fig. 183. This consists of a cylinder of earthenware or volvic stone filled with fragments of silica (quartz) or glass, on to which the acid from the tank, D, is sprayed ; a current of steam from the cock, a, together with a little air are passed upwards through the tower. As the temperature rises the organic matters are oxidised at the expense of the nitric acid, which thus gives oxide of nitrogen, this passing with the other nitrous vapours into the tube, H, which is supplied with a current of air from the injec- tor, //. The mixed vapours are divided between a double battery of long vertical earthenware pipes, G, where nitric acid of 38 to 40 B6. condenses, any vapour escaping being finally condensed in a Lunge-Rohrmann tower. The sulphuric acid at last reaches the bottom of the tower, A, where it collects in the basin, E, and thence passes through the leaden cooling coil, F. The acid thus obtained is darkened by the impurities present and has a density of about 56 to 58 B6 ; it is usually concentrated in cascade apparatus of the Negrier type or in Gaillard towers (see vol. i, p. 269). - During recent years, instead of the sulphuric and nitric acids being recovered and concentrated separately, it has been found preferable to send the acid mixture after decomposition of the dissolved nitroglycerine (see above) directly but carefully into the boilers (already containing the sodium nitrate) in which nitric acid is made. Some prefer to revivify the acid mixture, i.e. to bring it up to its original strength by adding the necessary quantities of fuming nitric and sulphuric acids, so that it can be used again for the production of fresh quantities of nitro- glycerine ; for this purpose, sulphuric acid or oleum is added slowly to the required amount of concentrated nitric acid and the mixture then poured into the weak acid. For this process of recovering the weak acids (by which the 2 ."> percent, or so of nitroglycerine dissolved in the acid is recovered) to be employed, a cheap supply of sulphuric anhydride or oleum must be available (oleum at less than 4s. per quintal). ^ FIG. 183. 228 ORGANIC CHEMISTRY and is then agitated by passing compressed air through the perforated pipe, N, for about fifteen minutes ; the nitroglycerine is allowed to settle and the water decanted off by means of the upper tap, M. The washing with water is repeated two or three times, all the washing water being collected in a single tank. 1 Finally the nitroglycerine is passed into a similar vat where it is stabilised, i.e. washed alternately with very dilute soda solution and water until the wash -water no longer has an acid reaction towards litmus and the nitroglycerine has a feeble alkaline reaction (0-01 of alkalinity, which disappears later). In the British Government factory at Waltham Abbey, Nathan, Thomson, and Rintoul (Eng. Pats. 15,983, 1901 ; and 3020, 1903) prepare nitrogly- cerine in large leaden vessels (a, Fig. 184) with inclined bottoms ; in these 300 to 500 kilos of glycerine are treated at one time and at the end of the operation, after 50 to 60 minutes' rest, the acid recovered from a preceding operation is passed from the tank, c, to the bottom of a. FIG. 184. In this way the nitroglycerine is displaced and caused to discharge through s into the washing vessel, e, exit for the vapours being supplied by the tube o. When all the nitro- glycerine has been forced out, a little of the acid mixture is drawn off by the pipe i, 2 to 3 per cent, of water being then slowly added to the remainder, which is mixed meanwhile with a current of air. By this means the dissolved nitroglycerine is decomposed and the dangers of slow separation in any of the vessels avoided (see preceding Note). 'The acid is immediately denitrated, after sufficient has been passed into the tank, c, to displace the nitroglycerine of the succeeding operation, b is the tank in which fresh acid is mixed, / the vessel for drowning, g that for stabilising, and h the filter for the nitro- glycerine, these two being at a considerable distance from e so that the nitroglycerine may be conducted to them as soon as it has undergone its initial rough washing. Yields of as much as 230 per cent, are obtained with this Nathan -Thomson process. In Italy it is used at the Villafranca (Tuscany) dynamite factory. FILTRATION. The washed nitrogly- cerine is carried in hardened rubber or ebonite buckets to the filters, which are merely wooden frames covered with woollen cloth or felt to retain the impurities, scum, gummy matters, &c. By covering these cloths with a layer of dried salt, the emulsified water can also be held back. The cloths rapidly become blocked and are frequently renewed. The filtration is often, especially in England, effected by means of the apparatus shown in Fig. 185. This consists of a lead-lined tank, A, with inclined base. In the lid is inserted a leaden cylinder, G, with a metal gauze bottom on which rests a filtering cloth, N, and on this a layer of sodium chloride, O, covered by another filtering cloth kept stretched by a leaden ring, Q ; the free part of this cloth is folded, stretched, and fixed by a conical leaden weight, E. In place of salt, a sponge may be employed to retain the water. In some cases complete separation of the water from nitroglycerine is obtained by leaving the latter at rest for a couple of days in a tepid place (30) ahd then decanting it ; but there 1 The wash-waters from all the preceding operations are collected in an inclined lead-lined tank called the labyrinth, which is divided into a number of chambers by vertical leaden walls perforated alternately at the top and bottom. The wash-waters enter slowly at one end of the tank and traverse a long up-and-down course, gradually depositing the emulsified or suspended drops of nitroglycerine before the opposite end of the tank is reached. The nitroglycerine collected at the bottom is discharged through suitable taps and added to that in the washing apparatus. FIG. 185. DYNAMITES 229 is then some risk owing to the prolonged accumulation of large quantities of nitro- glycerine, i In the working of nitroglycerine, each operation is usually carried out in a separate building, that in which the explosive is produced being at a very high elevation, the nitro- glycerine then flowing to lower points for the succeeding operations. All these buildings are of wood so as to diminish the damage in case of explosion. The floors of the sheds in which the nitroglycerine is produced and of those where it is treated in the liquid state are covered with sheet-lead with raised edges so that the material may be caught in case of breakage. Where the nitroglycerine is worked in a pasty state (for dynamites) the flooring is of wood free from crevices. If nitroglycerine is accidentally spilled, it should be immediately wiped up with sponges. The channels through which nitroglycerine passes from one shed to another are in the form of gutters furnished with removable covers and are fitted with a longitudinal pipe through which warm water can be circulated in winter and the danger of freezing avoided. A disadvantage attending the use of these channels is that an explosion in one shed is propagated along the channels to all the other sheds. So that the precaution is taken of disconnecting one section of a channel when not in actual use. In many factories the nitro- glycerine is transported in rubber pails (see above). The windows of the sheds are smeared with whitening, as the presence of curved parts in the naked glass might possibly result in the focusing of light on the explosive material and the explosion of the latter. USES OF NITROGLYCERINE. Small quantities are sometimes used in medicine to induce dilatation of the blood-vessels, but practically the whole of the production is used as an explosive. In America it has been long in use in the pure state for large mining operations ; Mowbray freezes it and trans- ports it in large quantities on trains from the factory to the place of consump- tion, as he regards it as less sensitive in the frozen state ; but this view is generally contested. It has also been transported without danger in solution in methyl or ethyl alcohol, from which it is reprecipitated with water at its destination. Almost all the nitroglycerine made is used in the manufacture of various kinds of dynamites, dynamite gelatines, explosive gelatines, smokeless powder, &c. DYNAMITES. This generic name is given to explosives obtained by gelatinising or absorbing nitroglycerine by various other substances. We have already mentioned that Alfred Nobel, the father of dynamite, had from 1860 to 1864 various explosions of nitroglycerine, sometimes of that recovered from the alcohol in which it had been transported (see above). In his attempts to diminish the dangers of nitroglycerine by diluting it with inert substances, Nobel discovered in 1866 that it is absorbed by kieselguhr (infusorial earth) in considerable proportions (up to 81 per cent.), and that in this state its power is diminished but little, while it can be safely handled and transported. He found further that this dynamite is exploded only by means of a fulminate of mercury cap. Kieselguhr is found in a very pure state in the Liineburg moors, near Unterluss in Hanover, and in an inferior quality in Scotland, Norway, and Italy. It consists almost exclusively of the siliceous remains of diatoms, and contains also traces of iron and organic matter. Its particles are formed of empty tubes perforated in all directions, and it is this structure which renders kieselguhr so highly absorbent. Under the microscope, it presents the appearance shown in Fig. 186. At the present time kieselguhr dynamite has been almost entirely replaced by new types (gums or gelatines) described later. If the absorbing substances are inert, like infusorial silica (kieselguhr), sawdust, cellulose, &c., they form dynamites with inactive absorbents, which contain about 72 to 75 per cent, of nitroglycerine, 24-5 per cent, of kieselguhr, and 0-5 per cent, of soda for the No. 1 quality, and less nitroglycerine in the Nos. 2 and 3 qualities. But in the new type of dynamite the solid matter consists of active substances, e.g. nitrocellulose, which take part in the explosion. These are dynamites with active absorbents, 230 ORGANIC CHEMISTRY the absorbents or bases being again divided into nitrates or inorganic oxidising bases and organic nitro-absnrbents (collodion-cotton, &c.). I. MANUFACTURE OF DYNAMITE WITH INACTIVE ABSORBENTS. The kieselguhr used must be suitably prepared. It is first spread out in furnace chambers TIG. 186. and gently heated to eliminate moisture and organic matter, and is then more strongly calcined in reverberatory or muffle-furnaces, excessive heating being avoided as it may destroy the absorbing properties. It is then ground into fine powder and sieved. The flour thus obtained should not contain more than 1 per cent, of moisture and should be immediately filled into sacks and consumed the same day, as otherwise it might absorb moisture. It consists of silica with traces of oxides of iron and aluminium. The nitroglycerine is weighed in buckets of hard gutta- percha or lacquered compressed wood-pulp and is carefully taken to the mixing-house, where it is poured into wooden troughs lined with sheet-lead, and containing the absor- bent. Skilled workmen then mix the mass rapidly by hand ; sometimes rubber gloves are worn, but usually the men prefer to do without gloves, as the hands become accustomed to the action of the nitroglycerine in two or three days. It is important to obtain a homogeneous mixture, so . that not the least portion of the kieselguhr remains free from nitroglycerine. After this hand-mixing the mass is rubbed through brass-wire sieves (2 to 3 meshes per centi- metre) arranged above lead-lined wooden troughs. The dynamite is placed on the sieve with a wooden spatula and pressed through with the palm of the hand ; here, too, the use of rubber gloves is not popular with the operatives. In the troughs the dynamite is in the form of fine grains, which should not be too dry or too greasy. If too dry, it is passed again through the sieve or mixed with more nitro- glycerine, whilst if too greasy it is mixed with a further amount of kieselguhr. It is then placed in small portions in iiidiarubber bags or in .wooden boxes lined with sheet- zinc and is removed to the building where the cartridges, used especially in mines, are pre- pared. Here the dynamite is transformed by simple presses into rolls, 19, 23, or 26 mm. in diameter. A very simple press devised by O. Guttmann is shown in Fig. 187. The dyna- mite is introduced into the cloth bag, m, and falls into the tube, I, being pressed into this by the lignum vitse or ivory piston, p, at the end of the bar, d, which is actuated by the FIG. 187. COMPOSITIONS OF DYNAMITES 231 ICVCT, i ; the cylinder of dynamite issuing from the bottom of the tube, Z, is broken by hand into definite lengths, which are wrapped in parchment paper or paraffined paper. The ordinary length is 10 cm. (discharge cartridges) or 2-5 to 5 cm. (primers). These cartridge machines are sometimes worked by pulleys and motors. In some cases the boudineuses illustrated later are used. After the dynamite is wrapped up, packets of 2-5 kilos are placed in cardboard boxes, which are wrapped and tied round and filled in tens into wooden cases. For military purposes the cartridges are put directly into metal boxes with a socket in the lid for inserting the detonator. For use under water these metal boxes are sometimes used, and sometimes sausage-skins or rubber bags. These cartridge buildings are usually small with light walls and roof ; only two or three operatives work in each, high earthen banks separating one man from the next so that the effects of an explosion may be mitigated. In place of kieselguhr various other absorbents are used at the present time, e.g. wood meal (cellulose) mixed with inert mineral salts (calcium carbonate, sodium bicar- bonate, &c.). First in America and then in -Austria fulgurite was prepared with 60 per cent, of nitro- glycerine, the remaining 40 per cent, consisting of wheaten flour and magnesium carbonate. At Cologne, Miiller prepared a Wetter- dynamite (safety dynamite, for use in mines containing firedamp ; see later) by mixing 10 parts of ordinary dynamite with 7 parts of crystalline sodium carbonate ; the water-vapour formed on explosion surrounds the flame and the explosive gases and thus prevents explosion of the firedamp. Many varieties of these dynamites are used to a greater or less extent in practice, e.g. carbodynamite containing 90 per cent, of nitroglycerine and 10 per cent, of carbonised cork, sebastine, lithoclastite, carbonite, &c. Properties of Dynamite with Inert Bases. This forms a pasty mass of reddish yellow, red, or grey colour according to the quality of the infusorial earth employed ; to ensure a uniform colour about 0-25 per cent, of burnt ochre is often added. It is odourless and has the sp. gr. 1 -4 and the pasty consistency of wet modelling clay ; the inside of the wrapper should show no traces of nitroglycerine (sweating). It is much less sensitive to pressure and percussion than nitroglycerine and, in small portions, can be lighted and burned without exploding. It can, however, be exploded by powerful percussion or detonation, or by red-hot metal, or by heating suddenly to a high temperature or for a long time at 70 to 80. Dynamite freezes at temperatures below + 8 and then becomes less sensitive ; before being used it must be carefully thawed in warming-pans, surrounded by water at a tem- perature not exceeding 60 ; it must never be thawed on a heated metal plate. Thawed dynamite should be used carefully as a little nitroglycerine exudes during thawing. Most of the dynamite made is used as an explosive in mines and for firearms ; for cannon it has little use, owing to the danger caused by sweating during the thawing, so that for military purposes explosives are used which are safer to transport and not so sensitive to shock or to discharge (explosion by sympathy). II. DYNAMITES WITH ACTIVE BASES, (a) Pulverulent Dynamites with Inorganic Nitrates. Immediately after the discovery of dynamite with a silica base came the idea of replacing the inactive substance, which diminished the force of the nitroglycerine, by active substances so that the explosive power of the dynamite might be increased. In America such dynamites are often made with 40 per cent, of nitroglycerine, 45 per cent, of sodium nitrate, 14 per cent, of wood-pulp, and 1 per cent, of magnesium carbonate ; these dynamites are well suited for mines where no great power but considerable safety is required. In Europe mixtures of nitroglycerine, ammonium nitrate, fine sawdust, sodium nitrate, carbon, &c., are made ; e.g. 20 per cent, nitroglycerine, 36 per cent, sodium nitrate, 25 per cent, ammonium nitrate, 18-5 per cent, roasted rye flour, and 0-05 per cent. soda. In Austria Trauzl in 1867 prepared a pasty mixture of nitroglycerine with guncotton, which was not affected by water and was exploded only by fulminate of mercury detonators. This product was not successful, but similar and improved preparations were subsequently made. About this time Abel in England prepared glyoxiline by soaking defibred guncotton and potassium nitrate in nitroglycerine ; this was also unsuccessful. (b) Blasting Gelatine and Gelatine Dynamite. Since these contain nitrocellulose, they 232 will be mentioned later (see Smokeless Powders), after the manufacture of nitrocellulose has been described. Statistics of dynamite : see later at the end of the chapter on Explosives. NITROCELLULOSE (Guncotton or Pyroxyline and Collodion-Cotton) This substance should, to be in order, be described later, after cellulose (which is a carbohydrate with many alcoholic groups and with a molecular formula polymeric with C 6 H 10 5 ) has been studied, but as its properties and uses are closely connected with those of explosives, it is considered opportune to include it in the present section. 1 CONSTITUTION OF NITROCELLULOSE. The relation C w H 2w O w of the com- ponents of cellulose being expressed by the more simple formula (C 6 H 10 O 6 ) W , it is found that the maximum degree of nitration consists in the introduction of three nitric acid residues per molecule of C 6 H 10 O 6 , so that guncotton was. given the name trinitrocellulose, and was repre- sented by the formula C 6 H 7 5 (N0 2 )3. Since the use of more dilute acids results in the combination of a less proportion of nitric acid residues, it is supposed that a mononitro- and a dinitro-cellulose are also formed. It was found later by Eder that there exist nitrocelluloses with compositions intermediate to those of tri- and di-nitrocellulose, and others between the mono- and di-nitro-compounds, so that it must be supposed that cellulose has a formula at least double that of the simple one given above ; but the mononitrocellulose corresponding with this doubled formula, C 12 H 20 10 , i.e. (C 6 H 10 H 5 ) 2 , has not yet been prepared. Still later Vieille, by accurate study of the nitrocelluloses prepared with acids of various concentrations, succeeded in preparing eight different types of nitrocellulose, this result indicating that Eder's formula, which predicted only six, could no longer serve. Vieille then proposed for cellulose a formula double that of Eder, i.e. C 24 H 40 20 or (C 6 H 10 5 ) 4 . according to which twelve nitrocelluloses are theoretically possible ; eight of these, from endeca- to tetra -nitrocellulose have been actually prepared. Mendelejeff, having found nitrocelluloses intermediate to or identical with these twelve, but different from those studied by Vieille in being soluble in a mixture of alcohol and ether, proposed the doubling of Vieille 's formula, so that cellulose becomes C 48 H 80 40 or (C 6 H 10 O 5 ) 8 . To-day, however, it is thought that these differences are due to mechanical mixtures of the various nitro- 1 In 1833 Braconnot observed that when starch or wood is treated, with concentrated nitric acid, a mucilaginous solution is obtained which, on addition of water, yields a white powder soluble in a mixture of alcohol and ether ; this powder, which burns vigorously, he called xyloidin. In 1838, by subjecting cotton to the same treatment, Pelouze obtained a product which exploded on percussion and was indeed nothing but xyloidin ; he recommended it as highly suitable for the manufacture of fireworks. In 1845 Schonbein at Basle, and some months later, and independently, Bottger at Frankfort discovered that the nitration of cellulose takes place much more easily and completely if the cotton is treated with a mixture of concentrated nitric and sulphuric acids. In order to utilise in- dustrially the guncotton thus obtained, the two discoverers combined and kept their process secret. After the initial difficulty in getting this new explosive taken up in practice, the extraordinary power of guncotton and its great advantages over black powder aroused considerable enthusiasm. But scarcely had it come into general use in various countries than a number of spontaneous and fatal explosions in guncotton factories and magazines, by which whole buildings were razed to the ground, created such a panic that its manufacture was everywhere abandoned. The process of nitration was then already known to Knop and Karmarsch, and to others, who manufactured guncotton by this simple process. And in 1846 Sobrero made use of the nitric-sulphuric mixture for the preparation of nitroglycerine. In 1853 the Austrian, Captain von Lenk, ascertained how to render guneotton safe. The Austrian Government acquired from Schonbein and Bottger the process of manufacture (at a price stated to be 30,000 florins, or 2500) and maintained the secret of avoiding the spontaneous decomposition of guncotton until 1862. Then von Lenk communicated the secret to the French and English Governments, and in 1864 patented the process in America. Whilst in America the manufacture was undertaken on an enormous scale, in Austria and England it was again suspended on account of further terrible explosions in the factories themselves ; these were explained by the English workers as due to the insufficient purification of the nitrocellulose by von Lenk. In 1865 Abel discovered the method of bestowing absolute safety and keeping qualities on the nitrocellulose. He used first of all the process of washing proposed in 1862 by the Englishman J. Tonkin, which consisted simply in complete washing with abundant supplies of water ; the nitrated cotton was then defibred or pulped in machines similar to those em- ployed in paper factories and the wet pulp subjected to considerable pressure. From that time the manufacture extended to all countries, in spite of an English factory being blown up in 1871 (apparently a criminal act), and in recent times it has acquired new and increased importance owing to the discovery of smokeless powder. The process universally used at the present time in the manufacture of guncotton is that indicated by Abel. Before the discovery of smokeless powder, guncotton had limited applications and was not used in mines since, in the form in which it was prepared, it had an excessive shattering action, whilst in mines progressive explosives are usually required. GUNCOTTON 233 celluloses rather than to separate chemical combinations, and further, that the nitration is gradual and leads from the more simple to the more complex forms. Nitrocelluloses with more than 12-83 per cent, of nitrogen were at one time regarded as being insoluble in alcohol-ether, but Abel showed that there are nitrocelluloses with 13-2 per cent, of nitrogen and still soluble in this mixture, whilst others with only 12-8 per cent, are insoluble ; and that this depends on the method of preparation the duration of action of the mixed acids, the ratios and concentrations of the latter, and the temperature at which they act and on the nature, purity, and dryness of the cotton. Only by following exactly the directions is it possible to obtain a constant percentage of nitrogen and complete solubility or insolubility in the mixture of alcohol and ether. It was also once thought that guncotton was a nitro-compound in the true sense of the word, i.e. that the N0 2 groups were united directly to carbon. But first Bechamp and then others showed that it is a true nitric ester which can be saponified, with regenera- tion of . the cellulose, by alkalis, alkaline salts, ammonium sulphide, or ferrous chloride. It has been further shown that with the maximum of nitrogen an oxynitrocellulose is obtained [oxycellulose is (C 6 H 10 5 ) 3 + (C 6 H 10 6 ) n , so that the nitrocellulose will be C 6 H 7 (N0 2 ) 3 O 5 + C 6 H 7 (NO 2 ) 3 O 6 , and this with ferrous chloride gives oxycellulose ; nitro- mannitol, treated similarly, gives mannitol and not oxymannitol]. PROPERTIES OF GUNCOTTON. Under the microscope nitrocellulose has the same appearance as ordinary cotton, but in polarised light it appears iridescent. When moistened with a solution of iodine in potassium iodide and then with sulphuric acid, nitrocellulose becomes yellow and cellulose blue. It is somewhat less white than ordinary cotton, is rather rough to the touch and crackles when pressed with the fingers ; it becomes electrified when rubbed and then appears phosphorescent in the dark. It is soluble in ethyl acetate, nitro- benzene, benzene, acetone, &c., but insoluble in water, alcohol, ether, acetic acid or nitroglycerine, although a mixture of nitroglycerine and nitrocellulose is soluble in acetone, forming a jelly, cordite (see later). It resists the action of dilute acids, but is decomposed slowly by concen- trated sulphuric acid or hot alkali, and completely by hot sodium sulphide. Decomposition is also effected by iron and acetic acid or by ammonium sul- phide or ferrous chloride (Bechamp). Flocculent, loose guncotton has the sp. gr. 0-1, whilst the powder (pulp) has the sp. gr. 0-3 and is exploded by shock or percussion only at the point where it is struck, the explosion not being propagated to the whole mass. When ignited, it burns so rapidly that even when it is placed on black powder, the latter does not burn. In the form of cord, it burns more slowly and may be used as a rapid fuse. When wet or compressed it has a specific gravity varying from 1 to 1-3 (the abso- lute sp. gr. is 1-5) and it then burns slowly and cannot be exploded by percussion or by ordinary detonators ; explosion can, however, be induced by detonating a little dry guncotton with a fulminate of mercury cap. The decomposition proceeds according to the equation : 2C 6 H 7 2 (ON0 2 ) 3 = 500 + 7C0 2 + 8H + 3H 2 + 6N. Less compressed guncotton gives more CO and H in comparison with the C0 2 and H 2 0, and hence has a less effect, the development of heat being smaller. No ash or smoke is formed, and 1 kilo of guncotton yields 741 litres of gas (the water being liquid, or 982 litres if the water is in the state of vapour), which is inflammable and, owing to the presence of carbon monoxide, poisonous. The temperature of combustion has been given as 6000 (i.e. 1071 Cals. are developed by 1 kilo). Unless guncotton is carefully prepared, it undergoes gradual change and may explode spontaneously, especially in the light, and to this are probably 234 O R G A N I C C H E M I S T R Y due the great explosions which occurred formerly (1848-1862). Even dry, granulated guncotton becomes harmless and safe to handle if it is immersed for a moment in ethyl acetate, as it becomes coated with a gelatinous layer which dries it and preserves it, if moist, from further evaporation. Guncotton is transported in large quantities in the wet state in wooden boxes placed in others of zinc which are sealed hermetically to retain the moisture. It is stored in dry magazines which are situate at least 150 metres from any habitation and are not surrounded by earthworks so that the more serious effects due to projection of debris may not be added to those of an explosion. MANUFACTURE OF GUNCOTTON. Hanks of purified cotton, free from impurities, are employed. This cotton should fulfil certain requirements. 1 The pure cotton is placed loose on trays which are arranged in a drying-stove heated by means of gilled pipes through which steam circulates ; the heating is continued until the proportion of moisture is less than 0-5 per cent., after which the cotton is allowed to cool for 12 to 15 hours in hermetically sealed boxes. If not pure the cotton is best defatted by boiling it for 2 to 3 minutes with 2 per ce'nt. sodium hydroxide solution ; it is then washed with water and subsequently treated with very dilute nitric acid in the hot. In some cases it is also bleached with a weak solution of hypochlorite, well rinsed with water and dried in a hot-air oven as above. When almost dry it is carded, dried completely, and, while still hot, placed in hermetically sealed boxes so that no more moisture may be absorbed. Nitration is then effected with a mixture of con- centrated nitric and sulphuric acid, as follows : 3 parts of pure sulphuric acid of sp. gr. 1-841 (96 per cent.) a ic poured into 1 part of pure nitric acid of sp. gr. 1-516, mixing taking place immediately and completely without the aid of stirrers. The mixture is then delivered with the help of an acid elevator (Monte/jus) into the nitration apparatus, consisting of a cast-iron FIG. 188. vessel, A (dipping pot) (Fig. 188), standing in a larger vessel, G, through which cold water circulates from H to J. The cotton is immersed in small portions (300 to 800 grms.) in the acid-bath and is stirred with an iron fork. In England 1 kilo of cotton is used per 160 kilos of the acid mixture, while in Germany 1 kilo of cotton is taken for every 40 kilos of acid ; after a short time (15 to 30 minutes) the nitrated cotton is removed with iron forks and is placed to drain on a cast-iron grid (grate), B and C, arranged on one side above the vessel ; before it is taken away, it is pressed with a cast-iron plate, F, connected with a lever, D E. The acid mixture is renewed when it has treated 30 to 50 per cent, of its weight of cotton ; also after each portion of cotton is removed from the bath, fresh acid mixture, equal to ten times the weight of the cotton taken out, is added in order to make up for what has been absorbed and combined. On the German system (where less acid is used) renewal takes place more frequently. Above each of the nitrating vessels is a hood with a strong draught to carry off the nitrous vapours which are always evolved. Excessive rise of temperature gives a guncotton which contains less nitrogen and shows less complete insolubility in alcohol-ether than are required in practice. In some factories the nitration is carried out in a number of small, deep and narrow, hemispherical vessels of cast-iron mounted on trolleys. These are charged in order with certain weights of acid mixture (30 to 50 kilos) and dry cotton (2 to 4 kilos), and, after thorough mixing, the trolleys are pushed into an oblong lead-lined chamber provided with 1 Cotton for nitrocellulose should be pure white and should not contain dust or fibres of jute, hemp, or flax, or woody matter or pods ; these impurities, when separated by hand from 200 grms. of the cotton, should not exceed 0-5 grm. The filaments should not be too short, otherwise they form a paste during nitration. A small piece thrown into water should sink in two minutes. It should not contain more than 6-9 per cent, of substances soluble in ether (fats, &c.) ; in many factories 0-5 per cent, is not allowed. In England the amount of fat allowed is 1-1 per cent, extracted with ether in 4 hours in a Soxhlct apparatus (sea Analysis of Fats). The moisture, determined by heating the cotton in an oven at 100 until its weight remains constant, should not exceed 6 per rent, and the cotton merchant is debited with any excess and also with the cost of drying. When moistened with a few drops of water, the cotton should maintain a neutral reaction. The ash, estimated by heating a few grammes of the cotton to redness in a platinum capsule until it becomes quite white and of constant weight, should not amount to more than 0-3 per cent. MANUFACTURE OF GUNCOTTON 235 as many doors as there arc trolleys. A powerful aspirator draws the nitrous vapours into a wooden flue. A battery of soaking-pots is used in such a way that w.hen the last is introduced into the chamber the first has already finished reacting (30 to 40 minutes), and as the pots are of metal and relatively small and are in a strong draught, the heat developed is readily dispersed. The pets are removed from the chamber and taken to the neighbouring centrifugal machines, into which the contents of the pots, which are mounted on pivots, are tipped. The centrifuges are similar to those employed in sugar factories (see Sugar) and have a steel or leaded steel rotating basket ; aluminium ones have also been tried but not with great success. A few minutes' centrifugation at 1000 revolutions per minute removes the greater part of the acid from the guncotton ; the latter is immediately taken to the washing machines, while the acid recovered is revivified in the manner described on p. 227. If drops of water or lubricating oil fall on to the cotton during centrifugation, the mass sometimes undergoes sudden decomposition with formation of a dense cloud of brown vapour ; this does not constitute a serious danger, since usually it is not accompanied by explosion. In some factories the nitration is nowadays carried out directly in the centrifuges, which may be of naked or leaded steel or even of earthenware, although these are heavier FIG. 189. FIG. 190. and more fragile (see Figs. 189, 190). The latter consist of a double-walled earthenware basket, the inner wall, d and a, but not the outer one, being perforated ; the two walls being a slight distance apart, an annular space, c, is left, which has an outlet above in a number of holes, s, in the edge of the bush. The whole is bound with steel hoops, t, to prevent danger from projection in case of fracture. The dry cotton (7 to 8 kilos or more) is arranged peripherally inside the perforated basket, the acid being supplied by the tube, m ; the basket, surrounded by the jacket, 6, and the cover, z, both of earthenware, is set in motion by the shaft, p, driven by the belt, r. The acid is driven uniformly through the cotton by centri- fugal force, rises through the ring space, c, and issues from the holes, s, into the channel, e, whence a pipe, /, carries it to an elevator to be again circulated. The operation is of short duration, and the red vapours are emitted from the tube, g. During nitration, the velocity of the drum is relatively low, but at the end the velocity is increased ; the nitrated cotton can then be taken away at once to be washed. Use is, however, preferably made of steel centrifuges with circulation of the acid, as proposed by Selwig and Lange ; the basket, d, is perforated (Fig. 191) and the cover is of aluminium and hinged and is furnished with a large tube, o, communicating with the pipe of an aspirator, n. The basket is moved slowly and filled with the nitric-sulphuric mixture (e.g. 70 per cent. H 2 S0 4 , 23 per cent. HN0 3 and 7 per cent, water) up to the top edge ; the cotton is then introduced in packets (1 kilo per 50 kilos of acid) and the basket given a velocity of 20 to 30 turns per minute. This movement causes the acid to circulate continuously through the cotton, and in half an hour, the nitration of 6 to 8 kilos of cotton is complete ; the acid is then discharged and the velocity increased to remove as much acid as possible from the cotton, which is taken out and washed in the ordinary way. 236 ORGANIC CHEMISTRY Excessive prolongation of the centrifugal ion and excessive velocity are not only without advantage but involve increased danger of explosion. From August 1905 in the Royal Gunpowder Factory at Waltham Abbey (where 2000 tons are produced per annum), guncotton has been made by the displacement process of J. M. and W. Thomson and Nathan, which is briefly as follows. FIG. 191. Into the earthenware basins, which are furnished with aluminium covers (Fig. 1 92) and are connected in groups of four by means of leaden pipes and also communicate with exhausters, 600 litres of the nitric -sulphuric mixture are placed ; about 10 kilos of cotton are then introduced in small portions into each vessel and are pressed with perforated wrought -iron plates. The nitration lasts two and a half hours, and at the end water is introduced above the perforated plate, this displacing at the bottom a corresponding quantity of the acid ; the acid thus recovered is reinforced with oleum and strong nitric acid. The displacement lasts 3 hours, after which the mass is centrifuged and the cotton washed, rendered stable, pulped, &c. FIG. 192. Since guncotton should have a very definite nitrogen content, different from that of collodion-cotton used to gelatinise nitroglycerine (see later), the process of nitration is care- fully followed by numerous rapid analyses until suitable conditions are found for obtaining a constant product ; after this has been done, the final control is sufficient. It has been proposed to follow the extent of nitration of cellulose by observing its behaviour towards polarised light. In recent years it has been shown that guncotton of more constant type and more readily rendered stable is obtained if the acid mixture is renewed for each nitration ; the last processes described are hence to be preferred. WASHING. The nitrocellulose from the centrifuge is passed directly into the oval washing vessel (see. Fig. 193), which has a longitudinal partition down the middle (like the hollander machines us-ed in paper-making), and in which a shaft furnished with beaters PULPING 237 mixes the whole mass with water ; the latter is constantly renewed and the washing continued until the acid reaction towards litmus paper disappears (2 to 3 hours). The washed guncotton is either centrifuged again or put to drain in wooden baskets. Although it no longer exhibits an acid reaction, yet, as was shown by J. Tonkin in 1862 and by Abel in 1865 (in England), it still contains acid or rather unstable sulphuric esters in the small channels of the fibres. To separate these remaining traces of acid, the nitrocellulose is rendered stable by boiling it for two consecutive periods of 12 hours each with water in wooden vats fitted with perforated false bottoms (one vat holds even more than 1000 kilos of the cotton), beneath which steam is passed. Then follow four more boilings of 4 hours each with water (formerly one or two boilings with calcium carbonate were also carried out), and finally two or three boilings each of 2 hours with fresh water. This system of washing was proposed by Dr. Robertson and employed with advantage in the Government Factory at Waltham Abbey ; it lasts altogether 48 hours, is preferable to that in which the boilings are short at the beginning and long at the end, and especially to that where boiling with, soda is interposed, as the soda hydrolyses the nitrocellulose and transforms it partially into collodion -cotton poor in nitrogen and soluble in alcohol-ether. Some of the boiling may be dispensed with if the nitrocellulose is steamed in closed vats. FIG. 193. PULPING. In spite of all the washing and boiling to which it is subjected, the guncotton persistently retains a trace of acid, and to remove this, the cotton is thoroughly defibred (pulped) as was proposed by Tonkin and by Abel in 1865. This operation is carried out in hoi- landers similar to those used for the preceding washing and identical with those used in the manufacture of cellulose for paper (see later, section on Paper, for figures and cross -sections). Pulping lasts from 5 to 8 hours, according to the fineness required, but if it is incomplete, inconveniences are met with in the subsequent compression, the desired density not being attainable ; also if pulping is carried too far, the compression is disturbed in another way. Guttmann proposed the use of hot water in pulping, and this possesses several advantages in addition to saving time. In the large wooden vats, 'as much as 200 kilos of guncotton can be treated at one time. In some cases a little calcium carbonate is added to guncotton to preserve it and to neutralise any residual acid ; it is added in powder just before the completion of pulping. If the guncotton thus prepared does not answer the rigorous tests to which it is sub- jected (see later, Tests of Stability), it is rendered stable by again boiling it for some hours with water in large wooden tanks (sometimes lined with lead), jets of steam, and also of air to keep the mass moving, being passed in. In order to separate the water, the mass is placed in suitable centrifuges fitted with drums of fine metal gauze entirely surrounded by linen ; in other cases, the water is separated as in paper-mills by placing the mass in chambers having perforated brass floors covered with cloth, the pulp drained in this way being finally centrifuged. The water separated from the pulp is allowed to stand in suitable vessels to deposit the finer fibres it has carried away. After centrifugation, the pulp contains about 25 to 30 per cent, of water and in this state it can be kept safely in zinc boxes, in 238 ORGANIC CHEMISTRY which it can be transported if it is slightly compressed and the cover of the box soldered. If properly prepared, guncotton should not contain more than 3-5 to 4 per cent, of collodion- cotton (soluble in alcohol-ether), but in England 7 to 8 per cent, is allowed. COMPRESSION OF GUNCOTTON. For military purposes, that is, for cartridges and for the blocks used for charging torpedoes, the still moist guncotton is strongly com- pressed to render it safer and more powerful owing to the increased charging density (see above), which reaches the value 1-2 with pressures of 500 to 1000 atmos. Fig. 194 shows in section a Taylor and Challen hydraulic press used for this purpose ; this is set up in an isolated room and can be controlled from a distance so as to avoid any great amount of damage in case of explosion during the compression, this mostly happening if any hard foreign body chances to be present in the guncotton. To obtain the greatest density, the pulp is first washed with hot water and slightly compressed in the mould, d, by means of the lever, h, the water being drawn away under the perforated base, c (covered with steel gauze), by a pump connected with the tube, I. FIG. 194. The partitions, I, are raised and the mould passed through an aperture in the wall, M (which serves as a protection for the workmen), and thus above the plate, n, of the hydraulic press ; this plate is kept horizontal by four columns, S'. The mould is raised by the piston, t , of the press so that it is pressed against a die, r, fixed to the cover, q. This cover is held fast by the four columns so that the die penetrates the mould and compresses the cotton under a pressure of 800 to 1000 atmos. The degree of humidity after the compression is about 10 per cent., and at each operation a block of 1 kilo is made, the shape being adapted to that of the projectile. Thus compressed, guncotton is so hard and compact that it can be worked quite safely with the plane, saw, or boring tool, a fine jet of water being directed at the point where the cutting is taking place. To prevent compressed guncotton from losing moisture and from becoming mouldy, it is dipped in molten paraffin wax ; or, better, it is immersed for a moment in ethyl acetate (or acetone), which dissolves a little nitro- cellulose at the surface and forms a kind of impermeable varnish. The theoretical yield of dry guncotton is 185 kilos per 100 kilos of dry cotton ; practically 171 to 176 kilos are obtained. USES OF GUNCOTTON. For the charging 'of torpedoes, moist com- pressed guncotton has replaced all other explosives. It is used also for rilling grenades, which are then covered with molten paraffin wax to unite the grenade and the explosive ; explosion is effected by a detonator of dry guncotton. It is made also into compressed cartridges for use in mines, a cavity being left for the detonating cap and the fuse. COLLODION-COTTON 289 Mixtures of granulated guncotton and nitrates are placed on the market undsr the namss of tonite, potentite, &c. Abel obtained beautiful, pyrotechnic effects by saturating guncotton with solutions of various mineral salts capable of imparting different colours to the flame. It is sometimes used for filtering acids, alkalis, and solutions of permanganate, being resistant to these reagents in the cold. Also it is in some cases employed as an electric insulator and for bandaging purulent sores and wounds, being first saturated with potassium permanganate. COLLODION-COTTON FOR GELATINE DYNAMITE, DYNAMITE, AND SMOKELESS POWDERS. During recent years, a different, less nitrated nitrocellulose, collodion-cotton, has assumed very great importance in the manufacture of smokeless explosives. On the other hand, guncotton itself has, of late years, been largely replaced by compressed, crystalline, or fused trinitrotoluene (see Part III), especially for military and naval purposes. Collodion -cotton was at one time thought to be dinitrocellulose, soluble in a mixture of alcohol and ether, but it has now been shown to be a mixture of various soluble nitro -compounds, which are formed under different conditions from those yielding guncotton. Collodion -cotton should have a constant nitrogen -content, and it should be readily soluble in a mixture of alcohol (1 part) and ether (2 parts), giving a dense viscous solution. If it answers these requirements, it gelatinises nitroglycerine well and dissolves completely in it ; attention is, however, also paid to the time necessary for gelatinisation. For photographic plates, extensive use was formerly made of ethereal -alcoholic solutions of soluble nitrocellulose (collodion), and in this case importance was attached not so much to the viscosity as to the proportion of nitrocellulose which would yield an elastic film of marked resistant properties. For this purpose, the nitration is carried out at a temperature of at least 40 to 50, so that the resulting collodion is less viscous ; also the nitrocellulose is not pulped. The cotton is immersed for 60 to 90 minutes in a mixture of 1 part of 96 per cent, sulphuric acid (sp. gr. 1-840) and 1 part of 75 per cent, nitric acid (sp. gr. 1-442) at a temperature of about 40. The more concentrated the acid and the more prolonged its action, the higher will be the nitrogen -content, but the viscosity will not be decreased ; a high temperature, however, results in diminution of the proportion of nitrogen and also of the viscosity. The nitration can be effected in the cold, but more concentrated acids and more pro- longed action are then required. After nitration collodion -cotton intended for the manufacture of gelatine dynamite goes through all the operations of washing, pulping, and boiling employed with guncotton. Collodion cotton for gelatine dynamite or smokeless powder must be subjected to a drying process. Since the centrifuged pulp still contains 30 per cent, of water, whilst nitrocellulose begins to decompose at 70 (or even at 50 if badly prepared) and in the dry state is very sensitive to shock or percussion, the drying of collodion-cotton con- stitutes a very dangerous operation. At one time it was dried by means of indirect steam, but nowadays it is placed on iron plates heated to 40 to 50. When dry, it sometimes becomes electrified on rubbing, and this phenomenon explains the frequent spontaneous fires formerly occurring in the drying ovens. Guttmann prefers to dry the collodion-cotton on copper plates connected with the earth by wires (to discharge the electricity). These plates are perforated with conical holes 0-25 mm. wide at the top and 1 mm. at the bottom ; strips of leather are used to prevent rubbing of the metal parts. In these ovens, the pulp is spread out and is subjected to the action of a current of air heated to 40 (in some case," also dried) and in two days the mass is dry, not more than 0-1 per cent, of moisture being then present. The dried material is then carefully placed in rubber bags and stored in air-tight boxes. The drying ovens are provided with alarm-thermometers, which also regulate the temperature automatically. Drying in a vacuum is also employed (especially with fulminate of mercury) and is then more rapid and takes place at a lower temperature, while the danger of an explosion is diminished owing to the absence of the tamping effect of the atmospheric pressure (see p. 221). 240 ORGANIC CHEMISTRY Collodion -cotton for making ballistite (see later) should contain 11-75 to 11-95 per cent, of nitrogen, whilst that for ordinary gelatine dynamites contains as much as 12 per cent. SMOKELESS POWDERS. Even 50 years ago attempts were made to diminish the smoke produced by ordinary gunpowder by diminishing the amount of sulphur present, but its relations to the nitre and carbon cannot be greatly altered. Potassium nitrate was then replaced by ammonium nitrate, but this was found to be too hygroscopic ; yet later, ammonium picrate was employed with better, but still not satisfactory, results. In 1864 Schulze prepared a smokeless powder from nitrocellulose obtained from pure wood-cellulose. It gave good results with sporting guns, but was too shattering for use in warfare, and the same was the case with a smokeless powder prepared in 1882 by Walter Reid by granulating nitrocellulose and gelatinising it superficially with alcohol and ether. The true solution of this important problem is due to Vieille, who in 1884 found that the shattering action of guncotton could be transformed into a progressive (or propellant) action by destroying the fibrous structure with suitable solutions. To attenuate the rapidity of explosion of guncotton it must be made as dense as possible (theoretically the fibre free from interstices has the density 1-5) and this cannot be done practically with fibrous cotton (even when pulped) as a pressure of 4000 atmospheres would be necessary. Vieille, however, dissolved or gelatinised the nitrocellulose and then recovered it by evaporating the solvent. With the smokeless powder prepared by Vieille in 1885 the velocity of projectiles from cannon was increased by 100 metres per second over that obtained with ordinary powder, the pressure in the cannon being the same in the two cases ; hence guns of smaller calibre could advantageously be employed. This amounted to a revolution in the region of ballistics, since, in addition to the advan- tages of no smoke or ash, and of the use of lower calibres, there was also the possibility of charging empty projectiles with these explosives, which are made and kept so safely. Gelatinisation is effected by solvents of nitrocellulose, i.e. by ether, acetone, ethyl acetate, nitroacetylglycerine, &c. (see p. 223). I. SMOKELESS POWDERS OF PURE NITROCELLULOSE. The quantity of dry nitrocellulose (6 to 10 per cent, of the weight of solvent) decided on is introduced into the kneading machine (see p. 243), which is furnished with a cover, the necessary quantity of solvent being then added and the kneading continued for 6 to 8 hours ; no danger of explosion attends this process. If a mixture of alcohol and ether is employed as solvent, less highly nitrated cellulose (collodion-cotton) may be used ; the 30 per cent, of water in the moist, centrifugated collodion-cotton is first displaced by alcohol and the mass then centrifugated again, the amount of alcohol remaining in the cotton being calculated so that the quantities of ether and alcohol required in the kneading machine may be known. This procedure offers the great advantage of avoiding the very dangerous drying of the collodion-cotton. When the gelatine in the kneading machine is homogeneous and cold, it is taken to the rolls, which are similar to those employed for ballistite (see later). The principal object of rolling is to increase the density of the gelatine and to give it a uniform composition. It is carried out between ordinary cylindrical rolls with increasing pressure, so that with repeated rollings between different rolls, sheets varying in thickness from half a centimetre to a fraction of a millimetre can be obtained. The rolls are heated by means of steam to a temperature not exceeding 60, so that the solvent is gradually eliminated from the whole mass. One of the most commonly used rolling machines for thick sheets is shown in Fig. 195 and one for thin sheets in Fig. 196. In France preference is given to hydraulic presses which give a still more uniform product. The thin sheets can then be cut into fine strips by means of rollers with superposed knives, as shown in Figs. 197 and 198. Some machines give a product like cut tobacco. If the strips, as they issue from the machine, are passed under other cutters perpendicular to the first, pieces of various lengths or cubes can be obtained which are convenient to carry and to use. Since these smokeless powders still contain small quantities of. free solvent the cut pieces are dried in a well -ventilated oven at about 40. This drying is now carried out more rapidly and with less danger in a vacuum (see p. 239). II. SMOKELESS POWDERS OF NITROCELLULOSE AND NITROGLYCERINE. A. As we have already seen, in dealing with the theory of explosives, the explosion of nitroglycerine is accompanied by the liberation of unused oxygen ; on the other hand, it GELATINE DYNAMITES, ETC. is known that guncotton does not contain sufficient oxygen for the complete combustion of the carbon and hydrogen present in the nitrocellulose molecule. In 1875, A. Nobel conceived the happy idea of associating the two substances by dis- solving in nitroglycerine a certain quantity of soluble nitrocellulose, that is, that used in the manufacture of collodion. This procedure gives gelatines of varying consistency FIG. 195. FIG. 196. according to the quantity of nitrocellulose (collodion-cotton) dissolved. Blasting gelatine is made from 90 to 93 per cent, of nitroglycerine and 7 to 10 per cent, of dry collodion- cotton ; gum dynamites, on the other hand, contain about 97 per cent, of nitroglycerine and 3 per cent, of collodion-cotton, and when they are mixed with about one-third of their weight of absorbent substances (wood-meal, rye-flour, sodium or ammonium nitrate) they FIG. 197. FIG. 198. form the gelatine dynamites, which are still plastic, although less so than the gum dynamites, and are also less violent and hence serve well for mining purposes. A common type of gelatine contains, for instance, 62-5 per cent, of nitroglycerine, 2-5 per cent, of collodion- cotton, 25-5 per cent, of sodium nitrate, 8-75 per cent, of wood meal, and 0-75 per cent, of sodium carbonate ; it has a specific gravity of 1-5, is exploded with a No. 1 fulminate of mercury cap, and is sold in Austria for No. I dynamite, whilst gelignite is sold for No. II dynamite and contains 45 to 50 per cent, of gum dynamite and about 50 per cent, of absorbents as above. II 16 242 ORGANIC CHEMISTRY At Christiania a non -congealing gum dynamite is made from blasting gelatine and a little nitrobenzene and ammonium nitrate ; it has a specific gravity of 1-49 and is less effective than the gelatine dynamites. For military purposes (torpedoes, cannon, &c.), as much as 4 per cent, of camphor is added in Italy, Austria, and Switzerland ; these gelatines are thus rendered insensitive and very safe, and they require special detonators (e.g. a mixture of 60 per cent, of nitro- glycerine and 40 per cent, of collodion-cotton or compressed guncotton). In certain commercial products the collodion -cot ton is replaced by nitrated wood or straw, while nitro benzenes, nitrotoluenes (especially liquid dinitrotoluene), &c., are used instead of nitroglycerine. 1 Gelatine and gum dynamites have the appearance of plastic masses, the latter, which has the sp. gr. 1-6, being especially horny and translucent. Gum dynamite sometimes exudes a'little nitroglycerine and so loses in shattering force ; when heated for a long time at 70, it swells up, becomes spongy and decomposes with formation of red, nitrous vapours ; it sometimes ignites in metal boxes when exposed to the sun. It is less sensitive even than dynamite (about six times less) to percussion and special caps of gelatine dynamite are required to explode it. It serves well for use in war, since it is insensitive to discharges, and to render it still less prone to detonation by influence it is mixed with a little camphor. When 20 per cent, of collodion -cotton is dissolved in nitroglycerine, a gum dynamite is obtained which is not exploded by the most powerful caps. And ballistite, which contains 30 to 50 per cent, of collodion-cotton, requires special detonators. After freezing and thawing, it becomes more sensitive and dangerous, as is the case with dynamite. It has a greater shattering power than dynamite and acts better than this under water, which does not wash away the nitroglycerine so easily. Exudation of nitroglycerine occurs more readily than with dynamite and causes some degree of danger. It is used as a basis for the manu- facture of smokeless powder. Gelatine dynamite is safer to handle and store than ordinary dynamite, which it is largely replacing. The manufacture of these gelatinised dynamites requires collodion-cotton, which is very carefully prepared and is completely soluble in a mixture of alcohol and ether, in addition to which it must possess as great a proportion of nitrogen as possible. When it 1 It is impossible at the present time to compare the various commercial brands of dynamite of different countries: or even of one country, so varied are the types and the ratios of the components, sometimes when the commercial name is the same. Thus No. 1 ammonia dynamite (French) contains 40 per cent, of nitroglycerine, 45 per cent, of ammonium nitrate (this, when pure, is not hygroscopic), 5 per cent, of sodium nitrate, and 10 per cent, of wood-meal or wheat-flour ; the No. 2 quality of the same brand contains 20 per cent, of nitroglycerine, 75 per cent, of ammonium nitrate, and 5 per cent, of wood-meal In Germany, the name Gelatine Dynamites is given to all mixtures prepared from explosive yum (96 per cent, nitroglycerine gelatinised with 4 per cent, collodion-cotton and a nitre-base as absorbent. In England, however, No. 2 gelatine ^dynamites are called gelignites, and are often formed of 65 per cent, of the gum and 35 per cent, of absorbents (75 p*er cent, nitre, 24 per cent, wood-meal wood- pulp used for paper, in a dried state and 1 per cent, of soda). In Austria, dynamite I is made from 65-5 per cent, of nitroglycerine, 2-1 per cent, of collodion-cotton, 7-41 per cent, of wood-meal, 24-85 per cent, of nitre, and 0-26 per cent, of soda ; dynamite II contains 46 per cent, of nitroglycerine, &c., and dynamite II A, 38 per cent, of nitroglycerine, &c. In France, gelatine dynamites are called gums, and are prepared in very varied forms, e.g. gum MB with 74 per cent, of nitroglycerine, gum D with 69-5 per cent., and gum E with 49 per cent. ; then there are dynamite gelatini- 1, 2, 26, and 2c (the last with 43 per cent, of nitroglycerine, &c.), &c. In Belgium, gelatine dynamites are called forcites ; forcite extra contains 74 per cent, of nitroglycerine, superforcite 64 per cent., forcite No. 2, 36 per cent., Ac. In England the types most commonly used are : dynamite JVo. I, with 75 per cent, of nitroglycerine ; gelignite with 65 per cent, of gelatinised nitroglycerine (97 per cent, of nitroglycerine), 25 per cent, potassium nitrate and 10 per cent, wood-meal ; blasting gelatine with 93 per cent, of nitroglycerine and 7 per cent, of collodion-cotton ; gelatine dynamite with 80 per cent, of gelatinised nitroglycerine (with 3 per cent, of collodion -cotton), 15 per cent, of potassium nitrate and 5 per cent, of wood-meal. In Italy there is dinamite-gomma A (or simply gomma A, corresponding with the French gomme extra-forte) formed from 92 per cent, of nitroglycerine and 8 per cent, of collodion-cotton ; gomma B (corresponding with the French gomme a la soude) with 83 per cent, of nitroglycerine, 5 per cent, of collodion-cotton, 8 per cent, of sodium nitrate, 3-7 per cent, of wood-meal, and 0-3 per cent, of sodium or calcium carbonate or ochre. Commercially, however, the strength is given in terms not of nitroglycerine but of gelatine, that is, the starting material is taken as a gelatine formed by gelatinising 94 per cent, of nitroglycerine with 6 per cent, of collodion-cotton, to which are then added the various absorbents ; thus gomma B contains 88 per cent, of gelatine (equivalent to 83 per cent, of nitroglycerine). In Italy the old kieselguhr dynamite is no longer used and is replaced by the so-called gelatine- dinamiti, which are marked with various letters and numbers ; thus No. 0, containing 74 per cent, nitroglycerine, 5 per cent, collodion-cotton, 15-5 per cent, sodium nitrate, 5 per cent. Vood-meal, and 0-5 per cent, carbonates ; Q.D.No. 1, with 70 to 72 per cent, nitroglycerine, &c. ; G. D. No. 2, with about 43 percent. nitroglycerine; and dinamit 2fo. 3, with 25 per cent, nitroglycerine, 54 per cent, sodium nitrate, 19 per cent, wood-meal and cellulose, and 2 per cent, soda and yellow ochre. During recent years there have also been prepared in Italy gelatine- dinamiti - suggested by Dr. Leroux with 8 to 10 per cent, of the nitroglycerine (of No. 1) replaced by as much liquid (. initrotoluene, which gelatinises cotton well and gives non-congealing dynamites, more economical and almost as powerful as, sometimes more powerful than, the corresponding gelatine dynamites ; these act well in open mines, but give large quantities of unpleasant fumes, and hence are unsuitable for use in galleries, Ac. KNEADING 243 FIG. 199. is not well prepared, although it dissolves in alcohol and ether, it is not readily and entirely soluble in nitroglycerine, or it does not retain the latter completely for a long time. The quality of the collodion-cotton depends, then, on the choice of a good cotton and on exact conditions of nitration duration, purity of acids, temperature. This should then be finely subdivided (pulped) and dry, so that it can be passed through a fine sieve before being mixed with the nitroglycerine, in which it does not dissolve well if moist. This operation of gelatinisation and kneading is termed in French petrinage. The necessary quantity of nitroglycerine is placed in wide, shallow vessels of copper or lead heated externally by hot water (50 to 60). After 30 to 60 minutes, when the temperature has reached 45 to 50, the required amount of dry, powdered collodion - cotton is added in small quantities and mixed now and then with a wooden spade. It is then left for a couple of hours and afterwards thoroughly mixed by hand, just as dough is mixed, so as to form a homogeneous, soft paste ; this, on cooling, forms a more or less hard, elastic, translucent gelatine which constitutes the gelatine or explosive gum. If, instead of col- lodion-cotton alone, absorbents are also used, gelatine dynamites are obtained ; these are converted into rolls and cartridges with the machines already described (p. 230). When the gelatine is not intended for the manufacture of ballistite (see later), the conversion into cartridges is effected by means of an Archimedean screw machine (boiidineuse), similar to sausage-making machines (Fig. 199). The mixing for causing gelatinisation, especially if other substances besides collodion- cotton are added, can be carried out in mechanical kneading machines (Fig. 200) mounted on a wooden platform, b, which can be raised by screws and cog-wheels, g, e, and h, resting on supports, a a ; on this platform is a double -walled, copper pan, ij, which can be surrounded with hot water and can be moved on rollers. Above are the bevel-wheels and pulleys for working the stirrers, q and r. The nitroglycerine is first heated to 50 by raising the temperature of the water in the jacket of the copper pan, the latter being then raised so as to submerge the stirrers ; the ingredients necessary to give the required type of gelatine dynamite are then added. Mixing is complete in an hour. Other forms of kneading machine are used, e.g. the Werner-Pfleiderer ma- chine, which is employed for smokeless powders and for bread-making. After cooling, the plastic dynamite, which has a yellowish, translucent appearance, is removed from the kneading machine to a separate building to be converted into cartridges. This is done in special machines (boudineuses) (Figs. 201 and 202), furnished with endless screws, which force the dyna- mite or gum from a hole, B, in continuous rolls, these being collected in definite lengths in a casing of parchment paper or paraffined paper, C. B. Military Smokeless Powders. These approach the gum dynamites in character, but contain more collodion -cotton, so that they are safer towards shock and useful as propellants (only slightly shattering). FIG. 200. 244 ORGANIC CHEMISTRY FIG. 201. The most important type is that prepared by Nobel in 1888 under the name of ballistite (after he had been preparing since 1878 gum dynamite by gelatinising nitroglycerine with 6 to 10 per cent, of collodion-cotton). Ballistite contains about 50 per cent, of nitroglycerine and 50 per cent., or even more, of collodion-cotton (with 11-2 to 11-7 per cent. N). To incorporate these two substances thoroughly and so that there is no danger in the subse- quent operations, use is made of Lundholm and Sayer's process, by which the constituents are united in presence of a liquid able to dissolve neither of them. This liquid is merely water, 0-5 to 1 per cent, of aniline being added to fix the acids liberated and thus increase the stability of the ballistite. The pulped collodion-cotton, contain- ing 30 per cent, of water, as it comes from the centrifuges (after boiling) is introduced into a cylinder of sheet-lead containing water at 60. The mass is well stirred by compressed air and the finely divided nitroglycerine passed in by means of a compressed-air injector. The agitation is continued until all the nitroglycerine is incorporated with the cotton, none remaining suspended in the water. The mixture is left in this condition for some weeks and is then centrifuged. It is next rolled at 40 to 50 in various machines similar to those shown in Figs. 195 and 196 on p. 241. The sheets thus obtained are then cut into strips, wires (flite, cordite *), cubes, granules, or shreds (lanite). The granulated smokeless powder thus obtained is sieved, and if in large strips these are sifted by hand ; it is then placed in the drying oven, while the scraps are softened in a warm bath and again pressed. Ballistite is almost brown in colour, has a sp. gr. 1-63 to 1-65, and is practically unaffected by moisture ; it inflames at 180 without exploding. The gases formed in its explosion contain no nitrous vapours and do not corrode the firearms. With some smokeless powders, attempts have been made to replace the nitrocellulose by nitrated starch and the liquid solvents by the corresponding va- pours, but no advantage has yet been procured in this way. Explosive gelatines can also be obtained by adding metallic nitrates (of barium or potassium) to collodion -cotton ; these have diminished power but possess the advantage of being readily inflammable. Mixtures of collodion - cotton and nitropenta-erythritol have recently been prepared for the use of large-bore artillery. FIG. 202. PROPERTIES OF SMOKE- LESS POWDERS. Those formed of nitrocellulose alone are hard ; ballistite 1 CordUe is a smokeless powder in filaments like hollow twine. Modern cordites contain 65 per cent, of guncotton (not collodion-cotton), 30 per cent, of nitroglycerine, and 5 per cent, of vaseline. Guncotton, which is Jin- soluble (to the extent of 90 per cent.) in alcohol, ether, or even nitroglycerine, can also be gelatinised by the action of a common solvent, e.g. acetone, which gives a colloidal solution persisting even after evaporation of the solvent. The dry guncotton is first mixed by hand with nitroglycerine, the mass being then introduced into an ordinary kneading machine, which is of bronze and is jacketed to allow of water-cooling ; the acetone (20 kilos per 100 kilos of the paste) is then added and kneading continued for at least 3 hours, after which the vaseline is mixed in for some time. The mass tends to heat and must be cooled. At the end of the operation, lumps of the paste, roughly cylindrical in form, are introduced into the cylinder of the cordite press, whichis similar to that used for making macaroni. The threads of varying thickness, length, and shape of cross-section thus obtained are then dried at 40 for 5 to 8 days. MELINITE, LYDDITE 245 is not so hard, and even in thick strips can be bent and then broken like a very hard paste. They are very resistant to the action of water and so have a great advantage over ordinary powders, which are destroyed by water. They also possess the advantages of a high density, 1-6 or more (see p. 218). Whilst Vieille's smokeless powder (gelatine of pure guncotton) withstands all ordinary conditions of temperature and moisture, ballistite, on the other hand, gradually loses nitroglycerine and so undergoes change of its properties if the humidity of the atmosphere oscillates much. But these conditions are rarely met with in practice, and ballistite is used not only by the Italian army and navy, but also by other Governments, and is in some ways superior to the cordite used in the English and also, to a certain extent, in the Italian armies. In 1906, the proportions of the components of cordite were varied slightly and the form altered to that of ribbons, being then known as axite. Smokeless powders withstand the blow of a projectile and require special detonators, fulminate of mercury not giving good results. They are exploded by compressed guncotton caps, which in their turn are exploded by fulminate of mercury. If accidentally ignited, smokeless powders are not very dangerous, since they do not explode, but regard must be paid to the very high temperatures (above 3000) produced, as these will melt iron, stone, &c. POWDERS WITH PICRATE BASES. As early as the fifteenth century an alchemist obtained an explosive substance by treating a kind of tar with aqua regia, but this acquired no importance in comparison with ordinary gunpowder. The explosive properties of picric acid and its salts were studied in the second half of the nineteenth century and assumed considerable importance when, in 1886, Turpin prepared melinite from 70 per cent, of picric acid and 30 per cent, of collodion -cotton previously rendered soluble with alcohol and ether ; this was regularly used for some years by the French army in place of dynamite. At the present time fused picric acid (m.pt. 122; sp. gr. 1-6) is poured into cartridges containing a fulminate of mercury cap and powdered picric acid. Weight for weight, picric acid is less effective than dynamite, but, measured by volume, its power is greater than that of dynamite, the specific gravity of which is 1-5. It has also the advantage over dynamite in that it does not freeze, being already in the solid state. In England, melinite was followed in 1888 by lyddite, containing about 87 per cent, of picric acid, 10 per cent, of nitrobenzene, and 3 per cent, of vaseline. This is poured in a molten state into the cartridges and is exploded with ammonium picrate detonators ; it is highly resistant to shock. It undergoes decomposition fairly readily, giving poisonous gases. These and other picric acid or ammonium picrate explosives have suffered considerably in importance since the introduction of the smokeless powders described above. The pro- perties and manufacture of picric acid will be described in the section dealing with benzene derivatives. SPRENGEL EXPLOSIVES. In 1 871 H. Sprengel, starting from the fact that explosion is nothing but instantaneous combustion, conceived the idea of preparing explosives by mixing a readily combustible substance with one possessing considerable oxidising pro- perties ; the substances separately are not explosive but become so when mixed, mixing taking place only at the spot where the explosive is to be used. This idea was taken up later by Hellhoff, who mixed nitric acid with nitrated hydro- carbons, and more effectually by Turpin and by R. Pictet, who mixed nitrogen peroxide (N 2 O 4 ) with various nitrated organic compounds and also with CS 2 (panclastite, fulgurite, &c. ) ; but these explosives never came into practical use. Another form of explosive of the Sprengel type is that with ammonium nitrate as base ; this has been largely used during recent years and is five or six times as powerful as gunpowder. The most important of these explosives is Favier powder, which, in its different forms, usually consists of a mixture of ammonium nitrate and nitronaphthalenes and sometimes contains also sodium nitrate (see later, Chlorate Powders). 246 ORGANIC CHEMISTRY SAFETY EXPLOSIVES (for Mines Rich in Firedamp). 1 Firedamp (see p. 33) is a mixture of methane and air and is formed particularly in coal-mines. It burns at 450 and inflames at 650 ; in presence of spongy platinum it burns even at 200. For ignition to occur, a cortain time at least some seconds is necessary. For instance, at 650 about 10 seconds elapse before the explosive mixture ignites, whilst at 1000 ignition occurs in 1 second. This explains why, for example, the gases produced at a temperature of 2000 by shattering explosives do not always fire the explosive mixture, the explosion occurring with enormous rapidity (scarcely measurable). The danger of ignition is diminished by decrease of the quantity of gas formed, i.e. of explosive used for each charge ; the very hot gase3 produced expand rapidly and become cooled, so that they are unable to cause ignition of the firedamp. Further, if the heat of the gases is efficiently utilised to give the maximum amount of mechanical work (splitting of the rock), the risk of firing is dimi- nished ; hence follows the necessity of a good tamping for each charge in order that the escape of the gases without performing work may be prevented. Explosion in the open is more likely to ignite firedamp. The use of a powerful detonator is advantageous, in order that the explosion may be sharp and rapid. Mine explosives should contain sufficient oxygen to produce only C0 2 and not the poisonous CO. Instead of calculating the temperature and duration of explosion, it is preferable in practice to make direct experiments with small cannons placed against a rock at the bottom of a long, Avide iron tube or test-chamber (20 cu. metres), containing an explosive gas. Discharge of the cannons should not ignite the gas if the explosive is safe. In France it is prescribed by law that in mines explosives must be used which give gases of maximum degree of oxidation but no inflammable gas (CO, H 2 ) or solid carbide : further, the calculated temperature of detonation must not exceed 1500 (or for certain piercing operations, 1900). Gunpowder, dynamite, and blasting gelatine readily cause explosion of firedamp in mines, their temperatures of explosion exceeding 2200 (as shown by Maillaid and Lc Chatelier). 2 In order to meet the requirements of a safety explosive, various ingenious processes are employed to lower the temperature of the gases from the explosion sufficiently to prevent them giving a flame. The charges are wrapped up and the tamping made wilh 1 The frequent explosions occurring in mines have led scientific men during the past thirty years to make attempts to mitigate their effects and to render them less common. Commissions for this purpose have been ap- pointed in France (1880), Russia (1887), Austria (1891), and other countries. In England the question has been studied by Macnab (1876) and Abel (1886) ; in France by Mallard and Le Chatelier (1883), Watteyne, &c. ; in Germany by Winckhaus (1895), very systematically by C. E. Bichel and Mettegang (1904-1907), who devised various ingenious forms of indicating apparatus, Beyling (1903-1907) and Heise (1898) ; and in Austria by Siersch (1896), Bohm (1886), Mayer (1889), and Hess (1900). The studies of Bichel more especially have shown that the safety of an explosive for use in mines (especially coal-mines) depends simultaneously on several factors, each of which must lie within definite limits, excess of one of these not being able to compensate for deficiency of another. Thus, for example, ordinary Mack poicder, which lias almost all the requisites of a safety explosive, cannot be employed for the sole reason that the duration of its flame is too long and so renders it dangerous. The principal factors establishing the safety of an explosive are : velocity of explosion, temperature of the gases formed, length of the flame, duration of the ilame, quantity of explosive used in each explosion, &c. 2 Dynamite and especially gunpowder, if exploded without tamping, will certainly ignite firedamp or even the coal-dust suspended in the air of coal-mines. The danger is diminished but not excluded by tamping, so that even in 1853 the Englishman Elliot suggested replacing the explosives by quicklime, a large compressed charge of which is placed in a cavity in the rock ; the pipe from a pressure water-pump is then introduced and a good tamping effected. The water, coming into contact with the lime, increases the volume of the latter 2 to 5 times, and, with the steam formed at the same time, pressures of 500 atmos. can be obtained. In 1880 the use of lime cartridges was fairly general in mines, but they were abandoned later owing to the unsatisfactory results given. No better fortune befell cartridges of quicklime, water, and sulphuric acid, or powdered aluminium and sulphuric acid (which develop hydrogen), or chlorine and ammonia compressed separately and then united, or compressed explosive mixtures of oxygen and hydrogen. In 1876, the Englishman Macnab suggested tamping gunpowder charges with water, but this did not always prevent explosion of the firedamp ; the same system applied to dynamite by Abel in 1886 gave better results. In some cases, the water is replaced by moist substances (sand, moss, a 0-862 bC Ci 5 H 30 O 2 Pentadecoic 51 257 a Cj 6 H3 2 O 2 Palmitic 62-6 268 o o 0-853 " Ci 7 H 34 O 2 Margaric 60 277 +3 a C 18 H 36 2 Stearic 69-3 287 0-845 <3 C 19 H 38 O 2 Nonadecoic 66-5 298% C 20 H 40 2 Arachidic 77 C 22 H 44 2 Behenic 84 360/60 mm. C 2 4H 48 2 Lignoceric 80-81 C 26 H 52 O 2 Cerotic 78-5 C 30 H 60 2 Melissic 91 oxidised, the chain is broken and acids with a less number of carbon atoms are obtained. (b) Hydrolysis of the nitriles (see these) in the hot with potassium hydroxide or with mineral acids yields the amides (see these) as intermediate compounds, and then the acids with one carbon atom more than the alcohols from which the nitriles originate : CH 3 -CN 2H 2 - NH 3 CH 3 -C0 2 H. (c) The interaction of a zinc-alkyl with phosgene gives : Zn(CH 3 ) 2 + 2COC1 2 = ZnCl 2 + 2CH 3 -COC1 (Acetyl chloride), which, on decomposition with water gives : CH 3 -COC1 + H 2 = HC1 + CH 3 -CO 2 H. (d) When a hydroxy-acid is heated with hydrogen iodide, separation of water and iodine occurs and a fatty acid is formed. (e) Other general reactions are those of Grignard (see p. 203), those of ethyl acetoacetate and ethyl malonate (see these), and those of elimination of C0 2 from dibasic acids (containing two carboxyls, CO -OH) and of addition of hydrogen to unsaturated acids, &c. PROPERTIES. In aqueous solution the acids are electrolytically dis- sociated into the cations H and the anions R- C0 2 (see vol. i, p. 91). Substitution of this ionic hydrogen by a metal yields salts, which in aqueous solution (when they are soluble) are almost completely dissociated, whilst the 266 ORGANIC CHEMISTRY hydrogen of the alcohols is also replaceable by a metal (alkoxide), but the resulting alkoxide is decomposed by water (hydrolysed). The strength of an acid (or its power) can always be determined from the degree of dissociation (vol. i, p. 98) ; this decreasing in the following order : formic, acetic, propionic, normal butyric, valeric, &c. ; thus, with rise of the molecular weight the dissociation diminishes. The hydroxyl group of the carboxyl group, CO -OH, can sometimes be substituted by halogens (especially by chlorine, by the action of PC1 5 , which forms acid chlorides or chloroanhydrides, e.g. acetyl chloride, CH 3 -COC1). Substitution of the hydroxyl, (1) by SH, gives thio-acids, and (2) by NH 2 yields the amides, e.g. acetamide, CH 3 -CO-NH 2 (by heating ammonium acetate) ; under certain conditions these compounds all give the acids from which they originate. It has already been mentioned that the saturated hydrocarbons are formed by the electrolysis of the alkali salts of the corresponding acids, with elimination of C0 2 , H, and (the last two from. the water present as solvent) and also of secondary products (unsaturated ethers and hydrocarbons) ; if the electrolysis is effected without a diaphragm, alkaline carbonate and bicarbonate are formed, and hence also a lower alcohol. Carbon dioxide can also be eliminated, and hydrocarbons thus formed, from the alkali salts of the acids by heating in presence of soda-lime or baryta, or by reducing the acids with hydriodic acid and phosphorus. But if the calcium salts of the acids are distilled, with or without P 2 5 , the principal product is a ketone formed from two molecules of the acid : (CH 3 -COO) 2 Ca = CaC0 3 + CH 3 -CO-CH 3 ; if the calcium salt is heated in presence of calcium formate, the aldehyde corresponding with the higher acid is formed. The halogens also replace the hydrogen of the alkyl residues of the acids, giving products which surpass in acids properties the acid from which they are formed. 1 By heating the acids homologous to acetic acid (which are very 1 Besides referring to what has been stated in vol. i, p. 91 et seq., we may here quote the very clear considera- tion of this question given by Professor Miolati on the Affinity Constants of Acids. That diffeient acids possess different strengths follows, for example, from the phenomenon of displacement of one acid from its salts by another acid. When sulphuric acid is added to a solution of sodium acetate, the characteristic odour of acetic acid is perceived, since the sulphuric acid is transformed into sodium sulphate and a certain amount of acetic acid is liberated. This quantity and, in general, the quantity of any acid displaced by a second acid, is not equivalent to the amount of the latter added, but the two acids divide the base according to their strengths, i.e. according to their affinity constants, and also to their quantities. The effect of the latter factor may be eliminated by using equivalent quantities of the two acids and of the base, e.g. by causing an equivalent of an acid to act on an equiva- lent of neutral salt, so that the distribution of the base between the two acids depends only on their strengths. A chemical equilibrium is then established which is represented by the equation : (1- z)[NaX + HX'] ^ z[NaX'+HX]. In order that this method may give exact results, it is of course necessary that the bodies formed in the condi- tions of the experiment be not eliminated either as gas, or solid, or complex molecules, &c., but that they remain to take part in the equilibrium. To determine this, Thomson made use of the thermal change and Ostwald the changes of volume and the indices of refraction, these methods leading to the same results. In general, any physical property may be used for the analysis of the equilibrated system. If, for example, a is the heat-change observed on neutralising an equivalent of the first acid with a base, b the corresponding quantity for the second acid, and c that observed on adding an equivalent of base to an equivalent of the mixed acids, it is evident that c will be equal to the thermal effect of the neutralisation of a certain frac- tion of an equivalent of the first acid (1 x) plus the thermal effect of the neutralisation of the complementary fraction of the second acid : f a o a o x -r is a measure of the relative affinities of the two acids. The following Table gives certain val x ) lined by Ostwald, x indicating the fraction of the molecule of base taken up by the acid given first : x es certain values of x deter- HNO, HC1 CCVCOOH CCVCOOH CCVCOOH CHCVCOOH CHCVCOOH CHCVCOOH CH.C1-COOH H-COOH 0-76 0-74 o-7i 0-92 0-97 H-COOH: CH,-COOH H-COOH: C 2 H5-COOH H-COOH : C,H,-COOH I, 1 ' CH.-COOH : C 3 H,-COOH (norm ) 0-76 0-79 0-80 0-81 0-53 AFFINITY CONSTANTS 267 resistant to oxidising agents) with concentrated sulphuric acid, C0 2 is evolved, whilst acids with carboxyl united to a tertiary carbon atom (e.g. formic or trimethylacetic acid) evolve CO and are transformed by oxidising agents into hydroxy-acids : (CH 3 ) 2 : CH-COOH gives (CH 3 ) 2 : C(OH)-COOH. Separation of the fatty acids from mixtures of them is not always easy and is sometimes effected by taking advantage of their greater or less volatility either in steam or in a vacuum, or by precipitating with magnesium acetate or barium chloride, since in alcoholic solution the higher acids are precipitated first. Use is also made of the fractional solution of the calcium, barium, or lead salts in various solvents (alcohol, ether, &c.), or of fractional neutralisa- tion followed by distillation of the acids not neutralised. From an aqueous mixture of formic, acetic, butyric, and valeric acids, the last two can be separated by extraction with benzene, from which they can be isolated by shaking with baryta water. Further separation can then be effected as above. If we calculate 1 - x Nitric acid . Hydrochloric acid Trichloroacetic acid Dichloroacetic acid Monochloroacetic acid Acetic acid . , making nitric acid equal to 100, we obtain the following values : 100 98 80 33 7 1-23 Formic acid . . . . .3-9 Propionic acid .... 1-04 Butyric acid. .... 0-98 Glycollic acid . ... 5-0 Lactic acid . . . 3-3 The acids arrange themselves in the same order and almost with the same coefficients, if other properties are studied. All acids possess, for example, the property of accelerating certain hydrolyses, such as that of ethyl acetate and the inversion of cane sugar : CHj-COOC a H 5 -f H 2 C 12 H 22 U C 2 H 6 -OH + CH 3 -COOH; H 2 = 2C 6 H 12 6 . In these reactions the acid added acts only by its presence (catalysis), since at the end of the reaction it remains unchanged. But, on the addition of equivalent quantities of various acids, the reactions take place with greater or less velocities, i.e. the same quantity of ethyl acetate or cane-sugar is transformed in a longer or shorter time according to the acid added. The velocity of the reaction is proportional to the affinity constant of the acid. Finally, the acids are arranged in the same order if we compare their electrical conductivities. According to the theory of electrolytic dissociation, the value of the conductivity depends on the number of molecules of the dis- solved acid which are dissociated into their ions, i.e. into hydrogen ions on the one hand, and acid ions on the other. The possibility of furnishing hydrogen ions in aqueous solution would hence be characteristic of the acid nature of a substance, the amount of these hydrogen ions in unit volume being a measure of the acidity. With equivalent solutions of different acids, the strong acids will be those which contain, in a given volume of the solution, a large number of hydrogen ions, and the weak ones those containing only a small number of such ions. The condition of an acid in solution may hence be represented by the expression : AH A' + H' and we may term the fraction of the equivalent which is dissociated, the degree of dissociation, a. Without entering into further details it may be mentioned that Cu(C 2 H 3 O 2 )OH] + 5H 2 O, is obtained by arranging sheets of copper with flannel saturated with hot vinegar, acetic acid, or acid vinasse, in between. The crust of acetate is detached from the plates and sold in cakss either dry or with 30 par cent, of moisture. It forms blue needles or scales, which effloresce in the air and become green, owing to loss of water. It dissolves only slightly in water and, when heated in the dry state, gives off acetic acid and water. When pure, it is completely soluble in excess of ammonium carbonate solution. It was formerly used as a colouring-matter, but is now used for the preparation of Schweinfurth's green (copper aceto-arsenite), Cu(C 2 H 3 O 2 ) 2 , 3CuAs 2 4 , by mixing with the requisite proportion of arsenious anhydride solution ; this gives a beautiful green colouring- matter, which is still used to some extent, although it is very poisonous owing to the evolution of hydrogen arsenide in the air. In cakes or balls, the basic acetate costs 6 per quintal, whilst the refined powder costs 8 10s. 1 Analysis of lead acetate is effected by dissolving 5 grms. of it in water, precipitating the lead with a known quantity, in slight excess, of normal sulphuric acid, making up to 250 c.c. and then adding a volume of water about equal to that of the precipitate. In 50 c.c. of the nitrate, the sulphuric acid is precipitated with barium chloride, the weight of the resulting barium sulphate giving the quantity of sulphuric acid which has remained in combination with the lead. Another 50 c.c. of the filtrate is titrated with normal caustic potash, the total acidity thus found being due to acetic acid and excess of sulphuric acid ; deduction of the latter then gives the amount of acetic acid existing in combination in the lead acetate 288 ORGANIC CHEMISTRY PROPIONIC ACID, C 3 H 6 2 orCH 3 -CH 2 -COOH This acid is obtained by hydrolysing ethyl cyanide (see p. 199) and also by the action of certain micro-organisms on calcium lactate. It is, however, usually prepared by fermenting wheat-bran, or is extracted from crude pyroligneous acid, being formed in small quantity in the dry distillation of wood ; it can also be easily obtained by oxidising normal propyl alcohol with chromic anhydride. For some years it has been manufactured by the Effront process (see p. 155) from the residues of beetroot molasses : 1000 kilos of molasses yield 75 kilos of ammonium sulphate and 95 to 120 of fatty acids consisting largely of propionic acid (see also section on Sugar). It is a liquid of sp. gr. 0-992 and resembles acetic acid in odour and in physical and chemical properties. It boils at 141 and solidifies at 22. It forms crystalline salts soluble in water, and its esters have a fruity aroma. Chemically pure propionic acid costs 32s. per kilo, and the commercial acid formerly cost 14s. 6d., but nowadays the Effront process yields a much cheaper commercial acid, which in practice may be used to replace formic and acetic acids, these being more difficult to purify. BUTYRIC ACIDS, C 4 H 8 O 2 Two isomerides are known of different structures, their constitutional formulae being deduced from their methods of synthesis. (1) NORMAL BUTYRIC ACID (Butanoic or Propylcarboxylic Acid or Buty- ric Acid of Fermentation), CH 3 CH 2 CH 2 COOH, is the more important of the two isomerides and exists in butter in the form of glyceric ester to the extent of 4 to 5 per cent. It is formed also in sweat and occurs in solid excreta and in decomposing cheese, as well as among products of fermentation of glycerine. It is obtained, not by synthesis (see p. 265), but by the butyric fermenta- tion of starch-paste in presence of a little tartaric acid, putrefied meat or cheese being added after a few days (pure cultures of special bacteria are also used at the present time) ; it is also obtained from acid skim milk by treat- ment with powdered marble and converting the calcium lactate into calcium butyrate, then into the sodium salt, and finally, by means of H 2 S0 4 , into the free acid. It is also obtained from molasses residues by Effront's process (see above). It forms an oily liquid, sp. gr. 0-958 at 14, boiling at 162 and solidifying in scales at 19. It dissolves in water, alcohol, or ether, burns with a bluish flame, and gives crystalline, slightly soluble salts. Calcium Butyrate, (C 4 H 7 2 ) 2 Ca + H 2 O, is less soluble in hot than in cold water. The esters have pleasant, fruity odours, and are used to produce artificial rum. Commercial concentrated butyric acid costs 4s. per kilo ; the 50 per cent, acid 2s. 6d. ; and the chemically pure (100 per cent.) 5s. I0d. The concentrated esters are sold at 2s. Qd. to 5s. per kilo. PT-T (2) ISOBUTYRIC ACID (2-Methylpropanoic or Dimethylacetic Acid), ;::: 3 >CH- Crla COOH, resembles the preceding acid, but is less soluble in water. It boils at 154 and solidifies at 79, and occurs free in arnica and carob roots and as ester in chamomile oil. It can be obtained by the ordinary synthetic processes and is less resistant than the normal acid to oxidising agents. The pure acid costs 40s. per kilo and the commercial acid about one-half as much. The Calcium Salt, Ca(C 4 H 7 O 2 ) 2 , is more soluble in hot water than in cold. VALERIC ACIDS, C 5 H 10 O 2 The four isomerides predicted by theory are known. (1) NORMAL VALERIC ACID (PentanoicorPropylacetic Acid),CH 3 - [CH 2 ] 3 -COOH, is a dense liquid (sp. gr. 0-956 at 0), boiling at 185 and solidifying at 58-5. It is obtained synthetically from propylmalonic acid or butyl cyanide and is met with in HIGHER FATTY ACIDS 289 pyroligneous acid ; it is slightly soluble in water. The pure product costs 5d. per gramme. CH (2) ISOVALERIC ACID, ): u 3 >CH.CH 2 -COOH, is found free or in the form of Crl 3 esters in animals (fat of the dolphin, sweat of the feet, &c.) and vegetables (roots of Valeriana officinalis), and from the latter can be extracted by boiling with solutions of soda or by distilling with water containing phosphoric acid. It is a liquid (sp. gr. 0-947 at 0), bailing at 174 and solidifying at 15 ; it has a disagreeable odour of stale cheese. It is often obtained by oxidising fusel oil with dichromate and sulphuric acid. The pure acid costs 96s. per kilo. Its esters are used as artificial fruit essences and cost from 10s. to 16s. per kilo. (3) ETHYLMETHYLACETIC ACID (Methyl-2-butanoic or Active Valeric Acid), PTT s ^>CH-COOH, is optically active as it contains an asymmetric carbon atom (see C2H 5 p. 18) ; it occurs naturally with iso valeric acid. The inactive mixture of the two oppositely active acids can be resolved into its active components by means of the brucine salts. It boils at 174. (4) TRIMETHYLACETIC ACID (Dimethyl - 2 - propanoic or Pivalic Acid), (CH 3 ) 3 : C-COOH,is a solid, m.pt. 35, b.pt. 163. It has an odour resembling that of acetic acid, and it can be obtained from tertiary butyl cyanide. HIGHER ACIDS Of the numerous isomerides theoretically possible and of the many actually known, mention will be made only of some of the more important which occur naturally and are usually of the normal structure and with even numbers of carbon atoms. NORMAL CAPROIC ACID, C 6 H 12 O 2 or CH 3 -[CH 2 ] 4 -COOH, is a liquid boiling at 205 and solidifying at 1-5. It is volatile in steam, has an unpleasant odour like rancid butter, #nd is found free in Limburger cheese and coco -nut oil, and as glyceride in goats' butter. It is formed on oxidising proteins or higher fatty acids (unsaturated). HEPTOIC ACID (CEnantic Acid), C 7 H 14 O 2 or CH 3 -[CH 2 ] 5 -COOH, is formed on oxidation of castor oil or wnanthaldehyde. It is a liquid boiling at 220 and solidifying at 20. It differs from its lower homologues by exhibiting a slight odour of fat. . CAPRYLIC ACID (Octoic Acid), C 8 H 16 O 2 or CH 3 -[CH 2 ] 6 COOH, solidifies at 16-5 and boils at 237-5 ; it is found in coco -nut oil and as glyceride in ordinary butter and that of goats. NONOIC ACID (Pelargonic Acid), C 9 H 18 O 2 or CH 3 -[CH 2 ] 7 -COOH, is a liquid boiling at 254. It is formed by oxidising oleic acid or by decomposing the ozonide of oleic acid (Molinari and Soncini, 1905) with dilute alkali. In nature it occurs in Pelargonium roseum. DECOIC ACID (Capric Acid), C 10 H 2 oO2 or CH 3 -[CH 2 ] 8 -COOH, is a solid, melting at 31-4 and boiling at 200 under 100 mm. pressure. It also is found in coco-nut oil and goats' butter. UNDECOIC ACID, C U H 22 O 2 or CH 3 -[CH 2 ] 9 -COOH. Distillation of castor oil under reduced pressure yields the unsaturated undecenoic acid, C 11 H 20 O 2 , which gives undecoic acid on reduction with hydrogen. It melts at 28 and boils at 212 (100 mm.). LAURIC ACID, C 12 H 24 2 or CH 3 -[CH 2 ] 10 -COOH, is a solid, melting at 44 and boiling at 225 (100 mm.) ; it occurs in the form of glyceride in laurel berries. MYRISTIC ACID, C 14 H 28 O 2 or CH 3 -[CH 2 ] 12 -COOH, melts at 54 and boils at 248 (100 mm.). It is found as glyceride (myristin) in the nutmeg (Myristica moschata) and in ox-gall, and abounds in the seeds of Virola Venezuelensis. PALMITIC ACID (Hexadecoic Acid), C 16 H 32 O 2 or CH 3 - [CH 2 ] 14 -COOH, forms a moderately transparent white mass, which readily softens and melts at 62-6. It is insoluble in water and crystallises from alcohol in scales or needles. It boils unchanged at 268 under 100 mm. pressure, or, with partial decomposition, at 339 to 356 under the ordinary pressure. ii 19 290 ORGANIC CHEMISTRY It is one of the normal components of animal and vegetable fats, in which it occurs as a glyceride (palmitin), and is easily obtained, together with oleic acid, from palm oil by hydrolysing and then decomposing the soap formed ; the palmitic acid is then isolated by fractional crystallisation. Japanese vegetable wax consists almost exclusively of palmitin. The industrial treat- ment of fats and oils for the extraction of the corresponding fatty acids (palmitic, stearic, and oleic) will be described in the section dealing with the manufacture of soap and candles. Commercial palmitic acid is also known by the inaccurate name of palmitin and is likewise manufactured by melting oleic acid with potassium hydroxide (Varrentrapp's reaction) : C 18 H 34 2 + 2KOH = H 2 + CH 3 - C0 2 K + C 16 H 31 2 K (potassium palmitate, which gives palmitic acid under the action of mineral acid). Its alkali salts (soaps) are soluble in alcohol or water, but considerable dilution of the aqueous solutions results in the separation of an acid salt and liberation of alkali. Whilst in alcoholic solution these soaps show virtually normal molecular weights, the aqueous solutions show no rise in the boiling-' point, the soaps thus behaving as colloids in these solutions (see vol. i, p. 102). The other salts (palmitates) are insoluble in water and, in some cases, soluble in alcohol ; mineral acids liberate palmitic acid from them. The commercial acid costs 4 per quintal, the refined product 8, and the doubly refined 25s. Qd. per kilo. MARGARIC ACID, C 17 H 34 O 2 or CH 3 - [CH 2 ] 15 -COOH, was for a long time thought to exist in fats, but it has been shown that a mixture of palmitic (C 18 ) and stearic (C 16 ) acids was being dealt with. Synthetically it can be obtained by hydrolysing cetyl cyanide, C 16 H 33 -CN, and by other methods. It melts at 60 and distils unchanged at 277 under 100 mm. pressure. STEARIC ACID, C 18 H 36 O 2 or CH 3 - [CH 2 ] 16 -COOH, which is improperly known commercially as stearine (and is then mixed with palmitic acid), and its separation from oleic acid will be described when dealing with candles. As glyceride, it occurs with that of oleic acid as one of the principal constituents of fats and oils, and is usually prepared industrially from beef suet.- Synthetically it can be obtained by the reducing action of hydrogen on oleic acid (see this), and the constitution of the latter being known, that of stearic acid follows directly. Industrial application is now made of this process, the catalytic reaction of Sabatier and Senderens being employed. It forms a somewhat soft white mass, melting at 69-3, and crystallises from alcohol in shining scales. It boils unchanged at 287 under 100 mm. pressure or with partial decomposition at 359 to 383 under the ordinary pressure. It is insoluble in water, soluble slightly in light petroleum, and more readily in alcohol, ether, benzene, or carbon disulphide. Its salts behave like those of palmitic acid. The lead salts of these high fatty acids are obtained by boiling the fats or oils with lead oxide and water. This lead soap is used for the preparation of lead plaster, and is used in the manufacture of varnish. Stearic acid made into a paste with gypsum forms a kind of artificial ivory. Italy imported, especially from France, England, and Belgium, 17,080 quintals of stearic acid in 1906 ; 12,50P in 1908 ; and 14,450 (of the value of 63,570) in 1910. CEROTIC or CEROTINIC ACID, C 27 H 54 O 2 , is found free in beeswax (together with Melissic Acid, C 30 H 60 O 2 ), as ester in Chinese wax and as glyceride in the fat of raw wool. It melts at 78-5 and is converted by oxidising agents into various acids with lower molecular weights. PREPARATION OF UNSATURATED ACIDS 291 II. UNSATURATED MONOBASIC FATTY ACIDS A. OLEIC or ACRYLIC SERIES, C W H 2 _ 2 O 2 (Olefine-Carboxylic Acids) Empirical formula Name of acid Constitutional formula Melting- point Boiling- point C 3 H 4 2 Acrylic acid CH 2 : CH-CO 2 H 13 140 rVinylacetic acid . . CH 2 : CH-CH 2 -CO 2 H -39 163 C 4 H 8 2 r . . J Solid crotonic acid IC I Liquid crotonic acid CH 3 -CH : CH-CO 2 H (cis) CH 3 -CH : CH-CO-H (trans) 72 15-5 181 169 vMetacrylic acid . CH 2 : C(CH 3 )-CO 2 H 16 161 /'Angelic acid . CH 8 C CO 2 H (8 structural isomerides and one I II 45 185 CjH,O 2 stereoisomeride) j CH 3 C H VTiglic acid CH 3 C CO 2 H II 65 198-5 H C CH 3 C,H 10 0. 2 (Not all stereoisomcrides known) Pyrotcrebic acid . (CH 3 ) 2 :C:CH-CH 2 -CO 2 H 15 207 C,H, 2 O 2 Do. -y-Allylbutyric acid . CH 2 :CH-[CH 2 ] 4 -CO 2 H 226 C,H 12 0, Do, Teracrylic acid (CH 3 ), : C: C(CH 3 )-CH 2 -CO 2 H 218 Ci H ls O, Do. Citronellic acid CH 2 :C(CH 3 )-[CH 2 ] 4 -CH(CH 3 )-CH 2 -CO 2 H 152" (18 mm.) C U H 20 2 Do. Undeceiioic acid . CH S :CH-[CH 2 ] S -CO 2 H 24-5 213-5 (100 mm.) Ci6H 30 O 2 Do. Hypogseic acid CH 3 -[CH 2 ],-CH:CH-[CH 2 ] 6 -CO 2 H rOleic acid CH 3 -[CH 2 ],-CH:CH-[CH 2 ] 7 -CO 2 H (cis) 14 223 C 18 H 34 2 Do. (10 mm.) vElaidic acid . CH 3 - [CH 2 j,-CH:CH- [CH 2 ],-CO 2 II( CH 2 :CH-CH 2 -COOH. (2) Oxidation of unsaturated alcohols and aldehydes with mild oxidising agents (silver oxide or the oxygen of the air) which do not attack the dcuble linking ; allyl alcohol and acrolei'n give acrylic acid. (3) Of general use is Perkin's reaction applicable especially to the aromatic series, but of service also for the fatty series : when an aldehyde is heated with the sodium salt of a saturated fatty acid in presence of an anhydride (e.g. acetic anlrydride) and then treated with water, the resulting products are the saturated acid corresponding with the aldehyde used and an unsaturatc d acid, which always has the double linking between the a- and /3-carbon atoms, the a-carbon atom being that adjacent to the carbonyl group, CO. If the chain united to the aldehyde group is denoted by R, the intermediate phases of this reaction are probably as follow : (a) R CHO + CH 3 CO CO CH 3 give, by aldol condensation, R-CH(OH)-CH 2 -CO-0-CO-CH 3 ; this unstable compound immediately 292 ORGANIC CHEMISTRY separates water, giving R CH : CH CO O CO CH 3 , treatment of this product with water yielding the unsaturated acid : (6) R-CH : CH-CO-0-CO-CH 3 +H 2 = CH 3 -COOH + R-CH : CH-COOH. It is evident that, if only one hydrogen atom is united to the carbon atom adjacent to the carbonyl group of the original anhydride, the first phase of the reaction, but not the second, will be possible, so that only a saturated hydroxy-acid will be obtained : R-CHO + CH-CO-0-O-CH + H 2 = R-CH(OH)-C-COOH + (CH 3 ) 2 : CH-COOH. /\ CH 3 CH 3 The presence of the sodium salt of the iatty acid is indispensable to all these reactions, but its function has not yet been explained. (4) Similar to Perkin's synthesis is the reaction between an aldehyde (or an a-ketonic acid, R-CO-COOH, which possibly loses C0 2 and thus gives an aldehyde) and malonic acid in presence of glacial acetic acid, a mixture of unsaturated monobasic acids with the double linking in the a /3- or /3 y- position being obtained and C0 2 split off : (a] R-CH 2 -CHO + CH 2 Malonic acid (6) 2R CH 2 CH(OH) - 2C0 9 + 2H 2 + R-CH 2 -CH: CH-COOH + R-CH : CH-CH 2 -COOH. a /3-acid 7-acid It should be noted that this reaction always gives also a condensation product of 1 mol. of the aldehyde and 2 mols. of malonic acid, this product then losing C0 2 and yielding a saturated dibasic acid : CH(COOH) 2 ~ CHa COOH (5) When monohalogenated saturated fatty acids (especially those with the halogen in the j3 -position) are heated with alcoholic potash or sometimes even with water alone, a molecule of halogen hydracid is eliminated and the unsaturated acid formed (similar to the reaction giving unsaturated hydro- carbons, p. 88) : CH 2 I-CH 2 -COOH = HI + CH 2 : CH-COOH. /3-Iodopropionic acid Acrylic acid (6) By separating a molecule of water from monohydroxy -acids by means of distillation or a dehydrating agent (H 2 S0 4 , PC1 5 , P 2 5 ) or sometimes by merely heating with caustic soda solution, the unsaturated monobasic fatty acids are formed : CH 3 -CH(OH)-CH 2 -COOH = H 2 + CH 3 -CH : CH-COOH. /3-Hydroxybutyric. acid Crotonic acid GENERAL PROPERTIES. The number of double bonds is ascertained by the same methods as are applied to unsaturated hydrocarbons (see pp. 88 and 89) by addition of either halogen or ozone. These unsaturated acids are more energetic than the corresponding saturated acids with the same PROPERTIES OF UNSATURATED ACIDS 293 numbers of carbon atoms, as can be seen from their ionisation constants (vol. i, p. 92). They are more easily oxidisable than the saturated acids, powerful oxidising agents rupturing the carbon atom chain at the double linking, the position of which can hence be established by a study of the com- positions of the two acids formed. When boiled with 10 per cent, caustic soda solution, unsaturated acids with a double linking (A) in the )3y-position undergo displacement of this linking with the partial formation of unsaturated acids with a double bond in the a/3-position (Fittig, 1891-1894) ; this is formed from an intermediate hydroxy-acid, an equilibrium being established as indicated below : R-CH : CH-CH 2 -COOH + H 2 ^ R-CH 2 -CH(OH)-CH 2 -COOH ; /3 y-acid /3-Hydroxy-acid H 2 + R-CH 2 -CH : CH-COOH. a /3-acid In general this reaction preponderates towards the formation of the a /3-acid and not vice versa, the carboxyl group apparently exerting an attraction on the double linking. When an unsaturated acid is fused with caustic soda or potash, the double linking is displaced, giving an a /3-acid, the new molecule immediately splitting at the double bond, the resultant products being acetic acid and another saturated acid. This displacement of the double linking was unknown until a few years ago, so that oleic acid, for example, which gives palmitic and acetic acids quantitatively when fused with potash, was regarded as an a /3-unsaturated acid. It has, however, been shown (see later] that the double bond of oleic acid is in the middle of the molecule, the action of the potash causing displacement of this bond before the molecule is resolved : CH 3 - [CH 2 ] 7 -CH : CH- [CH 2 ] 7 -COOH * CH 3 - [CH 2 ] 14 -CH : CH-COOH Oleic acid o./3-Oleic acid CH 3 - [CH 2 ] 14 -CH : CH-COOH + 2KOH + = CH 3 - [CH 2 ] 14 -COOK -(- H 2 O + CH 3 -COOK. Potassium palmitate The acids of the oleic and acrylic series, and unsaturated compounds in general, exhibit a tendency to polymerise ; the mere action of sodium alkoxide on the crotonic esters produces, in addition to other reactions, the following change : CH 3 C0 2 - CH 3 CH 3 C0 2 'CH 3 CH CH CH CH 2 II +11 II I CH CH C CH CO,- j CH 3 CO 2 *CH 3 C.tL 3 Methyl crotonate Methyl dicrotonate Instances of stereoisomerism among unsaturated compounds have already been described on pp. 16 and 17, and it is only necessary to state here that, of the two stereoisomerides corresponding with one and the same formula, one is less stable than the other, into which it is easily and directly transformed (the inverse change only takes place indirectly) by mere heating or by the action of concentrated sulphuric acid, caustic soda, a little nitrous acid, or a trace of bromine in the presence of light. 294 ACRYLIC ACID (Propenoic Acid) C 3 H 4 O 2 or CH 2 : CH-COOH This acid was prepared for the first time (Redtenbacher, 1843) by oxidising acrolein CH 2 : CH CHO, with silver oxide. It is now more readily obtained indirectly, by the action ot gaseous hydrogen chloride, which gives /3-chloropropaldehyde,.CH 2 Cl-CH 2 - CHO, this being converted by nitric acid into the corresponding /3-chloropropionic acid, CH 2 C1-CH 2 -COOH ; when the last compound is boiled with a solution of alkali, it loses HC1, yielding acrylic acid. Another convenient synthesis is the following : allyl alcohol (a, see below) with bromine gives dibromopropyl alcohol (b), which, on oxidation, yields a/3-dibromopropionic acid(c), and this, by the action of either zinc in presence of dilute sulphuric acid (or water) or reduced copper (containing iron) loses bromine and gives acrylic acid (d) : CH 2 CH 2 Br CH 2 Br CH 2 II I I II CH - > CHBr ' > CHBr * CH .CH 2 .OH CH 2 -OH COOH COOH a b c d Acrylic acid is a liquid soluble in water and having a pungent odour almost like that of acetic acid : it has the sp. gr. 1-0621 at 16, boils and polymerises at about 140, and when cooled forms tabular crystals melting at 13. With nascent hydrogen, it is trans- formed into propionic acid, whilst, when fused with potash, it gives acetic and formic acids. CROTONIC ACIDS, C 4 H 6 O 2 Isomeric unsaturated acids of this formula are possible theoretically two stereo- isomerides and the others structural isomerides. The following acids have actually been prepared : (a) CH 2 : CH-CH 2 .COOH, vinylacetic acid ; H C C0 2 H (ba) || , cis /3-methylacrylic acid (solid crotonic acid) ; H C CH 3 H C C0 2 H (&/3) , trans /3-methylacrylic acid (liquid crotonic acid) ; CH 3 C H CH (c) CH 2 : Ck^-^-JL-p methylmethyleneacetic or a-methylacrylic acid. With the general formula, C 4 H 6 O 2 , there corresponds also ethylene-acetic cr tri- rnethylenecarboxylic acid, | NCH-COOH, but this does not belong to the olefine- CH 2 / carboxylic acids as it contains no double linking, and it will therefore be studied with the cyclic compounds. (a) VINYLACETIC ACID, CH 2 : CH-CH 2 .CO 2 H,has been prepared, only recently, by distilling /3-hydroxglutaric acid in a vacuum : O-tL 2 OO 2 -rL C.H 2 CH-OH C0 2 + H 2 + CH CH 2 -CO 2 H CH 2 -C0 2 H and also by first brominating (with bromine dissolved in CS 2 ) allyl cyanide, hydrolysing the product, and finally removing the bromine by means of zinc dust and alcohol, thus : CROTONIC ACIDS 295 CH 2 CH 2 Br CH 2 Bri CH 2 ii r r ir CH > CHBr - > CHBr - > CH CN CN C0 2 H C0 2 H Vinylacetic acid is a very hygroscopic liquid, which solidifies when cooled to a low temperature ; it melts at 39 and boils at 163. Its calcium salt, (C 4 H 5 O 2 ) 2 Ca, H 2 O. crystallises from water in shining needles. When boiled with 5 per cent, sulphuric acid solution, it is transformed into the solid crotonic acid, CH 2 : CH CH 2 . COOH > CH 3 . CH : CH COOH. This transposition of the double linking is also effected by boiling with caustic soda solution, but in this case, a preponderance of /3-hydroxy butyric acid is formed at the same time. H C C0 2 H (5a) ORDINARY or SOLID CROTONIC ACID, || (cis ft-methylacrylic H C CH 3 acid or cis ethylideneacetic acid; also wrongly known as a-crotonic acid). Its constitution follows from its synthesis from a-bromobutyric acid (or rather its ester) by the elimi- nation of HBr under the action of alcoholic potash : CH 3 .CH 3 .CHBr.CO 2 H - HBr +CH 3 .CH : CH-C0 2 H. Prom water (solubility 1 in 12) the acid crystallises in shining needles melting at 71 to 72 ; it boils at 181 to 182, has an odour resembling that of butyric acid, and is found free in crude pyroligneous acid. Its calcium and barium salts contain no water of crystallisation and are very soluble in water. When gently oxidised in alkaline solution with permanganate, it gives afi-dihydroxy- butyric acid, CH 3 -CH(OH)-CH(OH)-C0 2 H, which cannot form a lactone, so that neither of its hydroxyl groups is in the y-position ; the double linking of the crotonic acid must hence be between the a- and /3 -carbon atoms. When halogen hydracids are added to it, the halogen goes to the ft -position. With nascent hydrogen it gives butyric acid. H C CO 2 H (6/3) LIQUID CROTONIC ACID, || (trans fl-methylacrylic acid or iso- CH 3 C H crotonic or allocrotonic acid ; known improperly as/3-crotonic acid). This acid is prepared from ethyl acetoacetate, which, with PC1 5 , gives probably a chloracetic ester, the latter losing a molecule of HC1 and yielding the two stereoisomeric chloroisocrotonic acids (or the corresponding ethyl esters) ; these two isomerides can be separated, the one formed in greater proportion being readily, and the other difficultly, distilled in steam. The latter gives solid, and the other liquid, crotonic acid on reduction with sodium amalgam : (a) CH 3 CO GIL, . CO OC 2 H 5 + PC1 5 = CH 3 - CC1 2 CH 2 CO OC 2 H 5 + POC1 3 . Ethyl acetoacetate Intermediate product (b) CH 3 . CC1 2 CH 2 CO OC 2 H 5 = CH 3 CC1 : CH CO OC 2 H 5 + HC1. Two stereoisomerides (c) CH 3 . CC1 : CH . CO . OC 2 H 5 + H 2 = CH 3 CH : CH CO OC 2 H 5 + HC1. Ethyl ester of crotonic acid The isocrotonic acid thus obtained is liquid, but is not pure, as it always contains ordinary crotonic acid and a little tetrolic acid, CH 3 -C : C-CO 2 H. Only within recent 1 This |3 y-dibromobutyric acid, when boiled with water, gives a p-bromobutyrolactone : CHjBr CH 2 - O CHBr = HBr + CHBr I I CH 2 -COOH CH 2 - CO Lactones are not usually formed by acids brominated in the a- or /3-position, but only with those where th& bromine atom is in the y-position. it may hence be concluded that the double linking in vinylacctic acid is also in the /3 y-position, since its brominated derivative gives a lactone, which is formed only when there is halogen in. the y-position. 296 ORGANIC CHEMISTRY years (1895 and 1904) has it been separated from these admixtures, either by means of its sodium salt, which is more soluble in alcohol, or by means of its quinine salt, which is less soluble hi water than that of crotonic acid. After such purification it is found that pure liquid crotonic acid forms crystals melting at 15-5 and boiling at 169 under the ordinary pressure or at 74 under a pressure of 15 mm. The calcium salt, (C 4 H 6 O 2 ) 2 Ca, 3H 2 0, forms large prisms, and the barium salt, (C4H 5 O 2 ) 2 Ba, H 2 0, large plates. When heated above 100, it is converted partially into normal crotonic acid, and in order to avoid this change during distillation the operation is carried out in a vacuum ; the transformation is instantaneous and quantitative in presence of a trace of bromine in aqueous solution or of carbon disulphide under the influence of direct sunlight. That the structure of isocrotonic acid is the same as that of crotonic acid and not of vinylacetic acid is supported by the fact that isocrotonic acid gives no lactonic derivative (see above) and also by the fact that the peroxyozonides of these two acids, obtained by Harries and Langheld (1905) by the action of ozonised oxygen and having the structure CH 3 -CH CH-C^ , give with water the same decomposition products, namely, O-O-O hydrogen peroxide, acetaldehyde, and glyoxylic acid, CHO C0 2 H. (c) METHYLMETHYLENEACETIC ACID (a-Methylacrylic, Metacrylic or Methyl- CH propenoic Acid), CH 2 : CC : CH-CH 2 -COOH, ^**3 is formed by the distillation of an oxidation product (terebic acid) of oil of turpentine <( together with the lactone of isocaproic acid, see below). (CH 3 ) 2 .C.CH(C0 2 H).CH 2 | = C0 2 + (CH 3 ) 2 : C : CH . CH 2 . COOH. CO Terebic (yy-dimethylparaconic) acid Pyroterebic acid That pyroterebic acid really has this constitution is shown by the fact that/ on reduction with hydriodic acid, it gives isocaproic acid of the known constitution (CH 3 ) 2 .CH.CH 2 -CH 2 .COOH. The position of the double linking is confirmed by the great ease with which it is converted into isocaproic lactone on prolonged boiling or by the action of a small quantity of hydro bromic acid : (CH 3 ) 2 : C : CH-CH^CC-OH * (CH 3 ) 2 : C CH 2 CH 2 CO w c- Pyroterebic acid is a colourless liquid solidifying at 15 and boiling at 207 ; it is lighter than water, which dissolves it with difficulty. y-ALLYLBUTYRIC ACID (i-Heptenoic-7 Acid), CH 2 : CH.[CH 2 ] 4 .CO 2 H, is obtained from cycloheptanone (or siiberone) by Wallaces reaction, passing through the oxime, amine, &c. : CH 2 CH 2 CH 2 v CH 2 CH 2 CH 2 COOH | ' \co -* I CH 2 C.U 2 CHjj OH 2 OH : CH 2 Cycloheptanone y-Allylbutyric acid It is a liquid boiling at 226 and, on oxidation, is converted into adipic acid, COOH- CH 2 CH 2 CH 2 CH 2 - COOH. TERACRYLIC ACID (2 : 3-Dimethyl-2-pentenoic-5 Acid), (CH 3 ) 2 : C : C(CH 3 ) CH 2 -COOH, is homologous with pyroterebic acid and is also obtained by distilling an oxidation product (terpenylic acid) of oil of turpentine : CH 2 .C0 2 H (CH 3 ) 2 : C- CH - CH 2 = CO 2 + (CH 3 ) 2 : C : C(CH 3 ) . CH 2 . C0 2 H. Teracrylic acid O CO Terpenylic acid It is a liquid, b.pt. 218, and is slightly soluble in water ; with HBr it forms hepta- lactone, the y -position of the double linking being thus confirmed. CITRONELLIC ACID (Dextro-rotatory), CjoHisOjj. It has not yet been definitely decided which of the two following formulae must be attributed to this acid : CH 3 , \C- [CH 2 ] 3 .CH(CH 3 ).CH 2 .C0 2 H (2 : Q-Dimethyl-l-octenoic-8 acid). CH./ S 3 > c : CH ' [CH 2 ] 2 .CH(CH 3 ).CH 2 .C0 2 H (2 : 6-Dimethyl-2-octenoic-8 acid). Ci 3 It is a liquid boiling at 152 under 18 mm. pressure, has a faint odour of capric acid and is obtained by the oxidation of citronellal (an aldehyde, C^H^O, abundant in ethereal oils). One of the two formulae must be attributed to Rhodinic Acid (leevo-rotatory), obtained by oxidising rhodinol, C| H 20 0. An inactive i-Rhodinic Acid is also known, this being obtained by the reduction (sodium in amyl alcohol) of geranic acid, (CH 3 ) 2 : C : CHC-H 2 -CH 2 .C(CH 3 ) : CH-C0 2 H, the hydrogen beingadded only at the a /3-double linking. UNDECENOIC ACID, CH 2 : CH.[CH 2 ] 8 -CO 2 H. The position of the double linking in this acid is confirmed by the fact that, on oxidation, the acid yields considerable quantities of Sebacic Acid, COaH^CHaJg-COaH. It is obtained (about 10 per cent.) on distilling castor oil under reduced pressure. It forms crystals melting at 24-5 and boils unchanged at 213-5 (100 mm.). When reduced with hydrogen iodide, it gives normal undecoic acid, the non-branched character of the carbon atom chain being thus proved. HYPOG- Stearolic acid CH 3 [CH 2 ] 7 CH : CH [CH 2 ] 5 . CH : CH. C0 2 H > Hypothetical intermediate acid CH 3 .C0 2 H + CH 3 - [CH 2 ] 7 .CH : CH. [CH 2 ] 5 .CO 2 H, Acetic acid Hypogseic acid OLEIC ACID, C 18 H 34 O 2 (CH 3 .[CH 2 ] 5 -CH : CH.[CH 2 ] 7 -CO 2 H) This acid is very abundant in nature in the form of glyceride (triolein), especially in vegetable and animal oils and fats. After hydrolysis of the fat, the fatty acids are liberated and from these the oleic acid separates, as it is liquid, whilst the others are solid. Although Chevreul discovered oleic acid at the beginning of the nineteenth century, it was only in 1846 that its composition was definitely established by Gottlieb. In the subsequent section dealing with soap and candles, the method of preparing oleic acid on a large scale (Italy imported 47,300 quintals of "oleine," of the value of 132,000, and exported 35,541 quintals, of the value of 99,500, in 1910) will be described in detail ; at the present juncture, only the constitution and the methods of obtaining pure oleic acid will be con- sidered. Oils rich in olein (olive oil, almond oil, lard, &c.) are hydrolysecl with caustic potash in the hot, the fatty acids being separated from the trans- parent soap thus obtained by means of hydrochloric acid ; these acids are then heated for several hours with lead oxide at 100, and the resulting lead salts dried and extracted with ether. This solvent dissolves only the lead oleate, the oleic acid being freed from the lead by means of hydrochloric acid. The oleic acid thus obtained is purified by transforming it into the barium salt and crystallising it from dilute alcohol, or by repeatedly freezing (at 6, -- 7) and squeezing the solid oleic acid, which, when pure, melts at 14 and has a specific gravity of 0-900 in the liquid state ; it has neither smell nor taste and, in alcoholic solution, shows no reaction towards litmus. In the air and when kept for a long time, the acid turns brown, assumes an acid reaction and taste, and undergoes partial oxidation. Under a pressure of 10 mm., it boils unchanged at 223 ; it can also be distilled without alteration by means of steam superheated to 250. The salts of oleic acid (and of other high fatty acids) form the soaps ; the alkali salts are soluble in water and separable from this by salt (NaCl), in saturated solutions of which they are completely insoluble. The calcium, barium, lead, &c., salts or soaps are insoluble in water. The action of concentrated sulphuric acid has already been mentioned (see Iso-oleic Acid). An important and characteristic reaction of oleic acid is its almost quantita- tive transformation, by a little nitrous acid (also by heating at 200 with sulphurous acid or sodium bisulphite), into the stereoisomeric Elaidic Acdi, OLEIC ACIDS which separates from alcoholic solution in white scales melting at 51 to 52 and boiling unchanged at 225 under a pressure of 10 mm. 1 That these two isomeric acids have direct (normal) carbon -atom chains is shown with certainty by the fact that, on reduction (e.g. with hydriodic acid and phosphorus 2 ), each of them takes up two hydrogen atoms giving stearic acid, which is known to have an unbranched carbon-atom chain ; also with bromine, they give dibromo -derivatives of stearic acid, and, with permanganate, dihydroxystearic acid, C^gH^C^OH)^ The position of the double linking in the molecule was under discussion for a number of years and until quite recently this linking was held to be at the end of the chain next the carboxyl group, i.e. in the a /3 -position, CH 3 - [CH 2 ] l4 -CH : CH-COOH, since fusion of these two acids with caustic potash resulted in the formation of palmitic acid ( Varren- trapp's reaction, p. 290). But this proof no longer seemed sufficient when it was shown that fusion with potash generally displaced the double bond. On the other hand, Baruch (1894) succeeded in eliminating hydrogen bromide from the dibromide of oleic acid, thus obtaining stearolic acid, which has a triple bond in the middle of the molecule. Also the products of oxidation (by permanganate) of oleic acid consist partly of pelargonic and azelaic acids, which contain nine carbon atoms, this reaction hence indicating that the oleic acid molecule breaks in the middle of the chain, at the position of the double linking. The most certain proof of the constitution of oleic acid has been advanced only recently as a result of the study of the ozonide of oleic acid and of its decomposition products (E. Molinariand Soncini, 1905 and 1906 ; C. Harries, 1906). The ozone is added quanti- tatively at the position of the double bond (see p. 88), and, according as ozonised air (E. Molinari) or ozonised oxygen (Harries) is employed, so the simple ozonide r CH 3 - [CH 2 ] 7 CH CH [CH 2 ] 7 . COOH 000 or a peroxide of the ozonide : CH 3 [CH 2 ] 7 CH CH- [CH 2 ] 7 . C- OH I I II o o o w II o is obtained. Decomposition of these ozonides (of oleic and elaidic acids) with dilute alkali or water in the hot results in the formation of acids (nonoic, azelaic, and others) or aldehydes (nonyl and semiazelaic) with nine carbon atoms, showing that the molecule is ruptured at the position of the central double linking. ISO-OLEIC ACID, C 18 H 34 O 2 . With concentrated sulphuric acid, elaidic and oleic acids give Stearinsulphuric Acid, C 17 H 34 (O-SO 3 H) -C0 2 H, which, with hot water, loses sulphuric acid and gives hydroxystearic acid, C 17 H 3 4(OH)-C0 2 H (with the OH at the position 10). When distilled under reduced pressure (100 mm.) or with superheated steam, this acid loses water and gives a considerable quantity of a white, solid acid iso-oleic acid which is readily soluble in alcohol and slightly so in ether, from which it crystallises in plates, melting at 44 to 45. The addition of the molecule of water to oleic or elaidic acid should take place at the central double linking (provided the sulphuric acid does not previously displace this linking) and the subsequent separation of water should occur at two carbon atoms adjacent to the double linking, so that the probable constitution of iso-oleic acid is, CH 3 [CH 2 ] 8 -CH : CH- [CH 2 ] 6 -C0 2 H. But various facts are known which throw doubt on the accuracy of this formula. A a ^-OLEIC ACID (2-Octadecenoic-iAcid), CH 3 -[CH 2 ] 14 -CH : CH-CO 2 H, is pre- pared by the removal of 1 mol. of halogen hydracid from the a -halogen derivative of stearic acid. It forms white crystals melting at 58 to 59, does not take up ozone in cold chloroform solution, and gives palmitic acid when treated with permanganate. 1 From an industrial point of view the transformation of a liquid fatty acid into a solid one is of interest, but this change is not utilisable in practice as it only proceeds well with fairly pure and fresh oleic acid and not with the commercial acid, which may be old and possibly polymerised (see also later section on Hydrolysis of Pats). 2 According to Ger. Pats. 141.029 and 211,669, 1907, the reduction can also be effected by hydrogen in presence of finely divided nickel (E. Endmann, Sabatier and Senderens* reactions [see pp. 34, 59, and 103 ; also later, Hydrolysis of Fats) 300 ORGANIC CHEMISTRY ERUCIC ACID (9-Docosenoic-22 Acid), CH 3 -[CH 2 ] 7 .CH : CH-[CH 2 ] n .CO 2 H, is found as glyceride in the oil of black and white mustard, and in tho.se of grape-seed and ravison, from which it is readily extracted. It is obtained crystalline from alcohol in shining needles melting at 34, and boils at 254-5 under a pressure of 10 mm. It forms a lead salt soluble in ether, as does oleic acid, and, like the latter, it is readily transformed into its stereoisomeride, Brassidic Acid, by the action of a little nitrous acid or of sulphurous or hydrobromic acid in glacial acetic acid solution. This isomeride crystallises from alcohol in leaflets melting at 65, and boils at 256 under a pressure of 10 mm. ; its lead salt is slightly soluble in hot ether. These two isomerides are not reduced by sodium amalgam, but with hydriodic acid they yield the saturated behenic acid, C 2 2H44O2. When fused with potash, they give arachic (C 20 H 40 O 2 ) and acetic acids. They yield various oxidation, bromination, and esterification products. They have normal carbon -atom chains and the position of the double linking is indicated by the fact that nonoic acid, CH 3 - [CH 2 ] 7 -COOH, and brassylic acid, C0 2 H- [CH 2 ] 11 -CO 2 H (and also a little arachic acid), are among the products of oxidation by nitric acid, while elimination of HBr from the dibrominated product gives the corresponding behenolic acid, which has a triple bond and is of known constitution. ISOERUCIC ACID, CH 3 .[CH 2 ] 8 .CH : CH.[CH 2 ] 10 -C0 2 H (?), is obtained by adding hydrogen iodide to, and then removing it from, erucic acid (just as with iso-oleic acid), and it appears that the double linking is not displaced during these changes, since decom- position of the dibromide of this acid (i.e. elimination of HBr) gives the same behenolic acid as is given by erucic acid, while oxidation with nitric acid also gives nonoic and brassylic acids. Isoerucic acid should hence have the same constitution as erucic and brassidic acids, so that, contrary to theoretical indications, three isomerides would seem to exist ; this requires confirmation. This acid melts at 54 to 56. B. UNSATURATED MONOBASIC ACIDS OF THE SERIES C n H 2n _ 4 2 The members of this series may be divided into two groups, as is the case with the hydrocarbons of the diolefine and acetylene series (see p. 90) : acids having a triple linking (propiolic series) and those having two double linkings (diolefine series). (a) ACIDS WITH TRIPLE LINKING (Propiolic or Acetylenecarboxylic Series) Name Constitutional formula Melting- point Boiling- point C,H 2 O 2 Acetylenecarboxylic (propiolic) acid ....... HC !C-CO,H 9 83 (50 mm.) C 4 H 4 O 2 Methylacetylenecarboxylic (tetrolic) acid . CH 3 -C ;C-CO 2 H 76-5 203 C 6 H 6 O 2 Ethyl-acetylenecarboxylic acid . C 2 H 6 -C !C-CO 2 H 50 C.H 8 2 Propyl- C 3 H 7 -C IC-CO 2 H 27 125 (20mm.) C 6 H 8 O 2 Isopropyl- C 3 H,-C ;C-CO 2 H 38 115 C,H, O 2 n-Butyl- C 4 H 9 - i C-CO 2 H 135 C,H, O a tert.-Butyl- C 4 H 9 -Ci C-C0 2 H 47 116 C 8 H 12 O 2 n-Amyl- C 5 H n -Ci C-CO 2 H 5. 149 C 6 H 14 O, n-Hexyl- C 6 H I3 -Ci C-CO 2 H 10 C 10 H 10 O 2 n-Heptyl- C,H IS -C; c-co 2 H 6-10 166 (20 mm.) Ci 2 H 20 O 2 n-Nonyl- C,H 19 -C: C-CO 2 H 30 . CnHuOs Dehydroundecenoic acid . CH i C-[CH 2 ] 8 -CO 2 H 42-8 175 (15 mm.) C n H 18 O 2 Undccolic acid CH 3 -Ci C[CHoJ,.CO 2 H 59-5 C 19 H 32 O 2 Stearolic acid CH 3 - [CH,],-C : C- [CH,] 7 .C0 2 H 48 C 1S H 32 O 2 Tariric acid .... CH 3 - [CH 2 1 10 -C C-[CK] 4 -CO 2 H 50-5 C 22 H 40 O 2 Behenolic acid CH 3 -[CH 2 j,-Ci C-[CHoi u -CO 2 H 57-5 PREPARATION. These acids can be obtained by the following reactions : From a sodium alkyl acetylide (suspended in ether), by the action of C0 2 (a) or of ethyl chlorocarbonate (b) : PROPIOLICACIDS 301 (a) , CH 3 -C 1 C-Na + C0 2 = CH 3 -C C-C0 2 Na. Sodium methylpropiolate (6) C 3 H 7 -C ! C-Na + C1-CO-OC 2 H 5 = NaCI + C 3 H 7 -C C-CO-OC 2 H 5 . Ethyl propylacetylenecarboxylate Various other acids of this series having the triple linking at a distance from the carboxyl group (and hence much more stable than the preceding, which, when heated, lose C0 2 and give acetylene hydrocarbons) are obtained by the elimination of 2HBr (by the action of alkali) from acids of the oleic series, this reaction being similar to that taking place with unsaturated hydro- carbons (see p. 91) : CH 3 - [CH 2 ] 7 -CHBr-CHBr- [CH 2 ] 7 -C0 2 H = Dibromide of oleic acid 2HBr + CH 3 - [CH 2 ] 7 -C : C- [CH 2 ] 7 -C0 2 H. Stearolic acid PROPERTIES. Acids of the type R-C: C-C0 2 H, when treated with sodium in absolute alcoholic solution, take up 4 atoms of hydrogen, giving acids of the saturated series ; they also combine easily with 2 atoms of bromine, but the second pair of bromine atoms, required to produce saturation, are added only with difficulty (the action of light facilitates the reaction) ; when boiled with alcoholic potash, they take up a molecule of water, forming saturated j3-ketonic acids : R-C i C-C0 2 H + H 2 = R-CO-CH 2 -C0 2 H; with aqueous potash, however, they yield methyl ketones, separation of C0 2 also taking place ; tert.-butyltetrolic acid does not react. The amines and hydrazine also give characteristic reactions with these acids. The esters of the acids have pleasant odours and are used in perfumes. Acids with a triple bond do not unite with the ozone of a current of ozonised air (E. Molinari, 1907 and 1908), but yield peroxyozonides with ozonised oxygen (Harries, 1907 ; see p. 299). PROPIOLIC ACID (Propinoic, Propargylic, or Acetylenecarboxylic Acid), CH C'CO 2 H, is obtained by heating the aqueous solution of potassium acetylene - dicarboxylate : C C0 2 H C H III = C0 2 + III C COjK OCO 2 K Propiolic acid is a liquid with a more intense odour than acetic acid ; it has the sp. gr. 1-139, is soluble in water, alcohol, or ether, solidifies at 4, melts at 9, and distils unchanged at a pressure of 50 mm. The alkali and alkaline -earth salts are extremely soluble in water. Prom its esters, metallic acetylides (p. 91 ) are readily prepared. Under the action of light and in a vacuum, it undergoes partial polymerisation, yielding benzenetricarboxylic acid. TETROLIC ACID (2-Butinoic or Methylpropiolic Acid), CH 3 -C i C CO 2 H, is obtained by eliminating HC1 from the /3-chloro -derivative of crotonic or isocrotonic acid. It crystallises from water in rhombic plates, melting at 76-5 and boils unchanged at 203, but it distils only with difficulty in a current of steam. Under the action of sod'um in alcoholic solution (but not with sodium amalgam in aqueous solution), it takes up hydrogen with formation of butyric acid. When oxidised with permanganate in alkaline solution, it yields acetic and oxalic acids. DEHYDROUNDECENOIC ACID (i-Undecinoic-n Acid), HC C-[CH 2 ] 8 .C0 2 H, obtained by heating dibromoundecenoic acid with alcoholic potash, melts at 42-8. On oxidation, it forms sebacic acid, C0 2 H- [CH 2 ] 8 -CO 2 H. It readily forms acetylides. 302 ORGANIC CHEMISTRY Treatment with alcoholic potash at 180 converts it into the isomeric Undecolic Acid (2-undecinoic-ll acid), CH 3 -C : C- [CH 2 ] 7 -CO 2 H, melting at 59-5; the latter "is hence formed with the dehydroundecenoic acid, if the reaction referred to above takes place at a high temperature. Oxidation of undecolic acid yields azelaic acid, C0 2 H- [CH2] 7 'CO 2 H ; it does not give acetylides, owing to the absence of the charac- teristic acetylenic hydrogen atom (see p. 90). STEAROLIC ACID (g-Octadecinoic-i Acid), CH 3 .[CH 2 ] 7 -C I C-[CH 2 ] 7 -CO 2 H, is readily obtained by boiling dibromostearic acid (prepared by brominating oleic or elaidic acid) with alcoholic potash. It melts at 48, and under the influence of sulphuric acid takes up water, giving a ketostearic acid. When oxidised with permanganate, it takes up 2 atoms of oxygen, giving diketostearic acid, whilst with nitric acid it is resolved into nonoic and azelaic acids (it is, however, stable in the air, and thus differs from oleic and linolic acids) : C [CH 2 ] 7 CH 3 CO [CH ] 7 CH 3 HI __^ CH 3 .[CH 2 ] 7 .C0 2 H + CO 2 H.[CH 2 ] 7 .(X) 2 H. C [CH 2 ] 7 -C0 2 H CO-[CH 2 ] 7 -CO 2 H Nonoic acid Azelaic acid Stearolic acid Diketcstearic acid Stearolic acid unites with 2HI, and the resulting diiodostearic acid, when heated with alcoholic potash, gives stearolic acid again, but in two isomeric forms, having the triple linking in the 8 to 9 and 10 to 11 positions respectively. The constitution of stearolic acid was confirmed by Harries (1907) by decomposing the peroxyozonide of the acid, obtained by the action of ozonised oxygen (ozonised air does not yield an ozonide, see p. 299) : 0-C-[CH 2 ] 7 .CH 3 0| || /OH +2H 2 = X 0-C.[CH 2 ] 7 .C/ >0= O Peroxyozonide of stearolic acid H 2 2 + CH 3 .[CH 2 ] 7 .C0 2 H + C0 2 H-[CH 2 ] 7 .C0 2 H. Nonoic acid Azelaic acid TARIRIC ACID (6-Octadecinoic-i Acid), CH 3 .[CH 2 ] 10 -C : C-[CH 2 ] 4 -CO 2 H, is isomeric with stearolic acid and melts at 50-5 ; as glyceride, it forms the principal component of the fat of the fruit of Picramnia Camboita. It is the first compound with a triple bond resulting from the vital process of an organism. It is stable in the air and yields stearic acid on reduction with hydriodic acid, so that its molecule contains a normal carbon atom chain. Energetic oxidation with permanganate or nitric acid yields lauric acid, CH 3 - [CH 2 ] 10 -CO 2 H, and adipic acid, C0 2 H- [CH 2 ] 4 -C0 2 H. BEHENOLIC ACID (9-Docosinoic-22 Acid), CH3.tCH2j7-CiC-tCH2Jn.C02H, melting at 57-5, is obtained from the dibromide of erucic or brassidic acid, in the same way as stearolic acid is formed from oleic acid. Its constitution follows from its behaviour towards reducing and oxidising agents and from the transformation of its oxime (Beckmann). (6) ACIDS WITH TWO DOUBLE BONDS, C n H 2W _ 4 O 2 (Diolefine or Sorbine Series) These acids are prepared synthetically by methods analogous to those used for obtaining a/3-unsaturated acids, for example, by treating a /3-unatu- rate'd aldehydes with malonic acid in presence of pyiidine : CH 2 : CH-CHO + C0 2 H-CH 2 -C0 2 H = Acrolein C0 2 + H 2 + CH 2 : CH-CH : CH-CO 2 H. /S-Vinylacrylic acid The acids of 4he sorbine seiics, in which the two double linkings are DIOLEFINICACIDS 303 conjugated that is, one in the a/3- and the other in the y ^-position and there- fore separated by a simple linking can be reduced by sodium amalgam in aqueous solution (in presence of a stream of C0 2 to fix the alkali) ; only two hydrogen atoms are then added, one for each double linking, and a new double linking remains in the place of the simple linking previously separating the two original double bonds : X-CH:CH-CH:CH-C0 2 H + H 2 = X-CH 2 -CH : CH-CH 2 -C0 2 H. When these sorbinic acids are oxidised with permanganate, two hydroxyl groups enter at the a /3-double linking, while the chain is broken at the y - double linking with formation of an aldehyde (which then undergoes oxida- tion) and racemic acid : X . CH : CH CH : CH C0 2 H > X CHO + C0 2 H CH(OH) CH(OH) CO 2 H, Racemic acid When they are heated with aqueous ammonia, 2 mols. of the latter are added at the double linking and two diamino-acids formed. On heating, acids of the sorbine series readily polymerise ; hence, when they are heated with lime or baryta, not only is C0 2 removed and diolefine hydrocarbons obtained, but di- and tri-molecular condensation occurs, giving more complex hydrocarbons which are probably of cyclic structure. /3-VINYLACRYLIC ACID (i : s-Pentadienoic, Acid),CH 2 : CH-CH : CH-CO 2 H, is synthesised by the method given above. It forms prisms showing a grey reflex, and dissolves slightly in cold water, but readily in hot. At 80 it melts to a mobile liquid, which, at 100 to 115, becomes dense and syrupy and then suddenly decomposes with evolution of gas. In carbon disulphide solution it unites with four atoms of bromine. SORBINIC or SORBIC ACID (2: 4-Hexadienoic Acid), CH 3 -CH : CH-CH : CH- CO 2 H, occurs in considerable quantities in mountain-ash berries. It melts at 134-5, boils at 228 with partial decomposition, is odourless and dissolves readily in alcohol or ether. DIALLYLACETIC ACID (i : 6-Heptadiene-4-methyloic Acid), CH 2 :CH-CH 2 - CH(CO 2 H) -CHo-CH : CH 2 , is obtained synthetically from acetoacetic acid by two separate introductions of the allyl residue. It is a liquid, sp. gr. 0-950, b.pt. 219 to 222, and has an unpleasant smell. Oxidation with nitric acid leads to the rupture of the carbon atom chain at the two double linking-; and formation of tricarballylic acid : CO 2 H-CH 2 ^~- H - pp. TT C0 2 H-CH 2 > GERANIC ACID (2 : 6-Dimethyl-2 : 6-octadienoic-8 Acid), (CH 3 ) 2 : C : CH-CH 2 . CH 2 -C(CH 3 ) :CH-CO 2 H, is obtained either by oxidation of the corresponding aldehyde (citral) with silver oxide, or by elimination of water from citraloxime by the action of acetic anhydride and hydrolysis of the resulting nitrile with alcoholic potash. It has also been obtained, by a series of reactions, from methylheptenone, (CH 3 ) 2 : C : CH- CH 2 CH 2 CO CH 3 . It is a colourless liquid of not very pleasing odour and boils a't 153 under a pressure of 13 mm. When shaken with 70 per cent, sulphuric acid it yields, among other products, the iso merle a-cycloyeranic acid, melting at 106 : CH,CH, CH,CHo / C / C \ HC CH-CO 2 H - > H 2 C CH-CO 2 H I II II H 2 C C-CH 3 H 2 C C-CH 3 H 2 H Geranic acid a-Cyclogeranic acid LINOLIC ACID, Ci 8 H 32 O 2 . In the form of glyceride, this acid is an important eon.tituent of drying oils (linseed, sunflower -seed, &c,). From these oils a mixture of 304 ORGANIC CHEMISTRY unsaturated fatty acids can be obtained which gives the nitrous acid reaction (solidifica- tion, owing to the formation of elaidic from oleic acid) only to a slight degree ; it contains less hydrogen but not less carbon than oleic acid and is readily oxidised and thickened by the oxygen of the air. The salts of these drying acids are still more readily oxidisable than the acids themselves, and their lead salts, like that of oleic acid, are soluble in ether. The various components of this mixture of fatty acids with two or three double linkings have not been completely separated, but as they fix 2 mols. of ozone (Molinari and Soncini, 1905), give a tetrabromostearic acid, C 18 H 3 2O 2 Br 4 , with bromine, and with alkaline permanganate yield a tetrahydroxystearic (sativic) acid, C^H^C^OH)^ which gives stearic acid with hydrogen iodide, the mixture must contain an acid with two double bonds. This is linolic acid, C^H^O^ which has not been obtained pure, although its stereo- isomerides a-Elaeostearic Acid, melting at 43 to 44, and^Telfairic Acid, obtained from telfairia oil, m.pt. 6 and b.pt. 220 to 225 under 13 mm. pressure have been prepared crystalline. Distillation of ricinelaidic acid gives a further isomeride, which has a normal structure, contains two double linkings, and melts at 53 to 54. (c) ACIDS WITH THREE DOUBLE BONDS, C W H 2M ^ 6 O 2 CITRYLIDENE ACETIC ACID (2:6- Dimethyl -2:6:8- decatrienoic - 10 Acid), CH 3 -C(CH 3 ) :CH-CH 2 -CH 2 .C(CH 3 ) :CH-CH :CH-CO 2 H, is a mobile oil, distilling at 175 under a pressure of 18 mm., and is formed by condensing 1 mol. of citral with 1 mol. of malonic acid in presence of pyridine : CH 3 .C(CH 3 ) : CH.CH 2 .CH2.C(CH 3 ) : CH-CHO + C0 2 H CH 2 CO 2 H = Citral Malonic acid H 2 + C0 2 + CuHuA.. Citrylideneacetic acid LINOLENIC AND ISOLINOLENIC ACIDS, C 18 H 30 O 2 , are components of the mixture of drying acids, but have not yet been isolated in a pure state. But with bromine two hexabromostearic acids, C 18 H 30 2 Br 6 , and with permanganate two hexahydroxy- stearic acids, C 18 H 30 2 (OH) 6 , have been obtained and these must be derived from two acids containing three double bonds. The fatty acids of linseed oil contain 50 per cent. of these two acids, together with linolic and oleic acids, whilst the other drying oils contain linolic acid in preponderating amount. The constitution of Linolenic Acid was definitely established by E. Erdmann, Bedford, and Raspe (1909) by decomposing the corresponding tri-ozonides. The three double bonds occur in a normal chain : CH 3 CH 2 - CH : CH CH 2 CH : CH . CH 2 - CH : CH . [CH 2 ] 7 . CO 2 H. The ozonides of two stereoisomerides were prepared, their products of decomposition being : propaldehyde, malonic dialdehyde, and azelaic semialdehyde. Fish oil contains another isomeride, Jecorinic Acid, C 18 H 30 O 2 , which has been little studied. III. POLYBASIC FATTY ACIDS A. SATURATED DIBASIC ACIDS, CH 8n (CO a H) 8 These acids are dibasic, since they contain two carboxyl groups and form two series of derivatives : acid and normal. In general they are crystalline substances, which distil unchanged in a vacuum (beyond C 3 ) and are soluble in water. The members with even numbers of carbon atoms have lower melting-points than their immediate neighbours in the series with odd numbers of carbon atoms, and the differences thus shown diminish as the number of carboii atoms increases. The solu- bility in water is greater with the acids containing an odd number of carbon atoms than for the others, and in both cases it diminishes with increase of molecular weight. The dissociation constant is very high for oxalic acid, and diminishes considerably in the higher homologues, which are hence less energetic acids. SATURATED DIBASIC ACIDS TABLE OF THE NORMAL SATURATED DIBASIC ACIDS 305 Empirical formula Name Structural formula Melting- point C2H 2 04 Oxalic acid -. ... COOH-COOH 189 (anhyd.) C 3 H 4 4 Malonic ,, . - . . ' C0 2 H.CH 2 .C0 2 H 132 C 4 H 6 4 Succinic . .'' CO 2 H.[CH 2 ] 2 .CO 2 H 182 C 5 H 8 4 Glutaric ,, . . C0 2 H.[CH 2 ] 3 .CO 2 H 97-5 C6H 10 O 4 Adipic ,, . . ... C0 2 H.[CH 2 ] 4 .C0 2 H 149 C 7 H 12 O 4 Pimelic ,, ' . ' , CO 2 H.[CH 2 ] 5 .C0 2 H 103 CgH 14 O 4 Suberic ,, . . CO 2 H.[CH 2 ] 6 .OC 2 H 141 C9H 16 4 Azelaic ,, . . CO 2 H.[CH 2 ] 7 .CO 2 H 106 QoHi 8 O 4 Sebacic . . s . CO 2 H.[CH 2 ] 8 .CO 2 H 133 C 12 H 22 O 4 Decamethylenedicarboxylic acid . C0 2 H.[CH 2 ] 10 .C0 2 H 125 Ci 3 H 24 4 Brassylic acid , i . C0 2 H.[CH 2 J U .C0 2 H 112 ^14H 2 6O 4 Dodecamethylencdicarboxylic acid C0 2 H-[CH 2 ] 12 .C0 2 H 123 Ci7H 32 4 Roccellic acid . . ..';. CO 2 H.[CH 2 ] 15 .CO 2 H 132 METHODS OF PREPARATION. In addition to the usual methods of oxidising monobasic fatty acids, primary hydroxy -acids, alcohols, and glycols, an important and general method consists in hydrolysing the nitriles (see p. 199) or cyano-derivatives of the acids, these being obtained from halogen alkyls with a less number of carbon atoms. Dibasic acids always of higher molecular weight are also obtained by the condensation of 2 mols. of the esterified monopotassium salt of a lower dibasic acid by electrolysis in Hofer's apparatus : GH,-COOCJB, CH 2 -COOK .H-OH CH 2 -COOK CH 2 -COOC 2 H 5 2 mols. Potassium ethyl succinate COOC 2 H 5 - 2C0 + 2KOH + H i H 2 Ethyl adipate PROPERTIES. The constitution is deduced from the synthesis in which compounds, especially the nitriles, of known constitution are employed. Structural isomerism commences with the acids containing four carbon atoms. Those acids which have the two carboxyls united to different carbon atoms (i.e. other than oxalic and malonic acids and their derivatives), in presence of dehydrating substances (PC1 5 , COC1 2 , &c.), or on heating, lose a molecule of water and form a kind of cyclic compound, known as an internal anhydride : CH 2 -COOH CH-CO !H 2 -COOH Succinic acid CHyCOOH CH C = H 2 = H 2 H,-COOH Glutaric acid Succinic anhydride CH 2 CO H 2 O CH 2 CO Glutaric anhydride The ready formation of these anhydrides by the reaction of the two terminal carboxyl groups (w, w') is readily explained by arranging the carbon 20 306 ORGANIC CHEMISTRY atoms in space (see p. 18 et seq.), with their valencies in the directions of the angles of regular tetrahedra. Thus, with succinic acid, which contains four carbon atoms, the two hydro xy Is of the carboxyl groups are found to be moderately close together (Fig. 246), whilst in glutaric acid the two hydroxyls are almost superposed, so that water readily separates, forming a closed ring (Fig. 247). Similarly the amides (which see) or the ammonium salts of these acids readily form imides (see later), which can be hydrolysed like the amides : CH 2 -COONH 4 CH 2 -COOH Monoammonium succinate = 2H 2 CH 2 -CCK Succinimido COOH OXALIC ACID (Ethandioic Acid), I has- been known from the COOH earliest times, since it occurs frequently in nature in plants, especially in FIG. 246. FIG. 247. sorrel, in the form of acid potassium oxalate, and also as incrustations of calcium oxalate in plant-cells and in the roots of rhubarb. It is often formed in the oxidation of organic substances (sugar, wood, starch, &c.) by nitric acid or permanganate, or by fused caustic potash. It is obtained synthetically by heating sodium or potassium formate rapidly (best in a vacuum at 280): 2H-COONa = Na 2 C 2 4 + H 2 (the reverse change, from oxalic to formic acid, has already been referred to on p. 269), or by passing carbon dioxide over metallic sodium heated to about 350 : 2Na + 2C0 2 = Na 2 C 2 4 . Its industrial manufacture was, until recently, carried out exclusively by the method devised by Gay-Lussac in 1829 and applied by Dale in 1856 : sawdust (1 part) moistened with caustic soda solution (2 parts, sp. gr. 1-4) is heated at about 240 and frequently stirred with iron plates until a greenish yellow mass is formed. While still hot, this is dissolved in water and the solution filtered and concentrated to 38 Be. When cold, the solution deposits crude sodium oxalate, which is dissolved in a small quantity of boiling water and precipitated as insoluble calcium oxalate by means of lime. The precipitate is made into a paste with water and the oxalic acid liberated by addition of sulphuric acid. The liquid is decanted and concentrated until 307 the whole of the calcium sulphate separates, the oxalic acid being then allowed to crystallise out and subsequently purified by repeated recrystallisation. At the present time, the acid and also the various alkaline oxalates are prepared by Goldschmidt's process (see p. 269), which consists in heating a mixture of potassium formate or carbonate with a little potassium oxalate and a slight excess of alkali (3 to 4 per cent.). From the oxalate thus obtained the oxalic acid is liberated by means of sulphuric acid. It crystallises in odourless and transparent monoclinic prisms, H 2 C 2 4 + 2H 2 0, which have a marked acid taste, effloresce in the air, and dissolve in 13 parts of cold or 0-3 to 0-4 part of hot water. The crystals lose their water of crystallisation partially at 30 and, after dehydration, sublime. When heated moderately strongly or treated with concentrated sulphuric acid, oxalic acid decomposes into CO, C0 2 , and H 2 O. It is poisonous and is used in the dyeing and printing of textiles and of wool ; it serves for bleaching straw, removing rust stains from textiles, purifying glycerine and stearine, cleaning brass, &c. The acid is estimated by means of normal caustic soda solution in presence of phenol - phthalein, or of decinormal potassium permanganate solution in presence of sulphui.c acid in the hot : 2KMnO 4 + 5H 2 C 2 O 4 + 3H 2 S0 4 = K 2 S0 4 + 10C0 2 + 8H 2 + 2MnS0 4 . Ammoniacal impurities are detected with Nessler's reagent (vol. i, p. 539), and, when pure, the acid should leave no ash, and 0-5 grm. of it should dissolve completely when shaken with 100 c.c. of ether. The commercial crystallised acid is sold at 56*. to 60s. per quintal, whilst the doubly purified product costs 4 and the chemically pure 128*. Italy imported the following quantities at 72*. per quintal : 1160 quintals in 1907, 960 in 1908, 755 in 1909, and 1890, costing 6424, in 1910. In Russia, four factories produced about 8500 quintals of oxalic acid in 1909, by heating sawdust with alkali. In 1908 Germany exported 51,000 quintals of oxalic acid and potassium oxalate, and 44,700 quintals in 1909. In 1911 the United States imported 1600 tons of oxalic acid of the value 33,000. SALTS OF OXALIC ACID. Owing to the presence of two carboxyl groups in the molecule, oxalic acid gives both acid and neutral salts. The alkaline oxalates are soluble in water and are often used instead of the acid, especially in dyeing. NORMAL POTASSIUM OXALATE, K 2 C 2 O 4 , used to be obtained by neutralising the acid with potassium carbonate, concentrating and allowing to crystallise. Nowadays it is prepared by Goldschmidt's method (see above). It dissolves in three parts of water, crystallises with 1H 2 O and readily effloresces in the air. It costs 84s. to 88*. per quintal, or, when chemically pure, 6. ACID POTASSIUM OXALATE (or Potassium Hydrogen Oxalate), KHC 2 O 4 , is obtained by dissolving the neutral oxalate (1 mol.) and oxalic acid (1 mol.) in water, concentrating and allowing to crystallise, when it separates with 1H 2 0. It has a bitter, acid taste, is poisonous and dissolves in 14 parts of hot water. POTASSIUM TETROXALATE (Commercial Salt of Sorrel), KHC 2 O 4 + H 2 C 2 O 4 + 2H 2 O, does not effloresce or lose its water of crystallisation in the air. It is obtained by mixing a hot, saturated solution of potassuim oxalate with the calculated amount of saturated oxalic acid solution. It costs 84*. to 88*. per quintal, or, if chemically pure, 128*. CALCIUM OXALATE, CaC 2 O 4 , crystallises with 2H 2 O and is obtained from a solution of a soluble oxalate, containing either ammonia or acetic acid, by addition of a soluble calcium salt. It is insoluble in water or acetic acid. FERROUS OXALATE, FeC 2 O 4 , or, better, Ferrous Potassium Oxalate, K 2 Fe(C 2 O 4 ) 2 + H 2 O, gives a yellow aqueous solution owing to the colour of its cation. FeC 2 O 4 ". It possesses strong reducing properties and is largely used on this account, while it serves also as a good photographic developer. POTASSIUM FERRIC OXALATE, K 2 Fe 3 (C 2 O 4 ) 3 , gives a green aqueous solution owing to the colour of its cation, Fe(C 2 O 4 ) 3 '". In the light it yields C0 2 and potas'ium ferrous oxalate, and it is used in the platinotype method of photography. 808 ORGANIC CHEMISTRY CO H MALONIC ACID (Propandioic Acid), H 4 C 3 O 4 or CH 2 which is of great importance, since it allows of the synthetical preparation of the most varied higher dibasic acids, and from these, by loss of carbon dioxide, of the corresponding monobasic acids. This ester is obtained by passing gaseous hydrogen chloride into cyanoacetic acid dissolved in absolute alcohol ; it is then separated by distillation, as it boils at 198. At 15 it has the sp. gr. 1-061. The hydrogen atoms of the methylene group of this ester can be replaced by one or two atoms of sodium (or halogens) giving highly reactive sodiomalonic esters. The sodium in these can be substituted by one or two alkyl groups simply by treatment with an alkyl HOMOLOGUES OF MALONIC ACID Name of acid Formula Melting- point of acid Boiling-point of the diethyl ester Methylmalonic . . CH 3 .CH(C0 2 H) 2 about 130 190-193 Dimethylmalonic (CH 3 ) 2 : C(C0 2 H) 2 192-193 196 Ethylmalonic . .. C 2 H 5 .CH(CO 2 H) 2 112 210 Diethylmalon ic (C 2 H 5 ) 2 : C(C0 2 H) 2 124 230 Propylmalonic C 2 H 5 -CH 2 .CH(CO 2 H) 2 93-5 219-222 Dipropylmalonic (C 2 H 5 .CH 2 .) 2 :C(C0 2 H) 2 156 248-250 Isopropylmalonic (CH 3 ) 2 :CH-CH(C0 2 H) 2 86 213-214 Methyiethylmalonic . (CH 3 )(C 2 H 5 )C(C0 2 H) 2 118 207-208 Butylmalonic . C 2 H 5 -CH 2 .CH 2 .CH(C0 2 H) 2 98-5 sec. Butylmalonic C 2 H 5 CH(CH 3 ) CH(CO 2 H) 2 76 224-225 Isobutylmalonic (CH 3 ) 2 : CH-CH 2 .CH(C0 2 H) 2 107 225-226 Diisobutylmalonic . [(CH 3 ) 2 : CH-CH 2 . ] 2 C(C0 2 H) 2 145-150 245-255 Methylpropylmalonic (CH 3 )(C 2 H 5 .CH 2 .)C(C0 2 H) 2 106-107 220-223 Methylisopropylmalonic [(CH 3 ) 2 CH](CH 3 ) : C(C0 2 H) 2 124 221 Pentylmalonic . C 2 H 5 CH 2 CH 2 CH 2 CH(C0 2 H) 2 82 Isoamylmalonic (CH 3 ) 2 CH.CH 2 .CH 2 .CH(C0 2 H) 2 98 240-242 Diisoamylmalonic [(CH 3 ) 2 CH - CH 2 CH 2 ] 2 C(CO 2 H) 2 147-148 278-280 2-Methylbutylmalonic (CH 3 )(C 2 H 5 )CH CH 2 CH(C0 2 H) 2 90-91 244-246 tert. Amylmalonic (CH 3 ) 2 (C 2 H 5 )C-CH(C0 2 H) 2 238 sec. Amylmalonic (C 2 H 6 ) 2 CH.CH(C0 2 H) 2 52-53 242-245 Methylisobutylmalonic (CH 3 ) 2 CH- CH 2 - C(CH 3 )(C0 2 H) 2 122 230-235 Ethylisopropylmalonic (CH 3 ) 2 CH.C(C 2 H 5 )(C0 2 H) 2 131-131-5 232-233 Cetvlmalonio . CH 3 .[CH 2 ] 15 -CH(C0 2 H) 2 121-5-122 Dicetylmalonic [CH 3 .(CH 2 ) 15 .] 2 :C(C0 2 H) 2 - 86- 87 Dioctylmalonic [CH 3 -(CH 2 ) 7 .] 2 :C(C0 2 H) 2 75 338-340 MALONIC ACID DERIVATIVES 309 iodide, sodium iodide being separated at the same time. The resulting products are higher homologues of the malonic ester and hence yield the corresponding homologues of malonic acid on hydrolysis. The hydrolysis of the esters of dibasic acids by alkali takes place in two stages, the second ester group being hydrolysed more slowly than the first. Treatment of ethyl malonate (1 mol.) with sodium (1 or 2 atoms) results in the evolu- tion of hydrogen and the formation of the solid mono- or di-sodiomalonic ester : "NTa fTlV 2 ' 2 5 f^ ^-C0 2 C 2 H 5 < C0 2 .C 2 H 5 ' ^ 2< C0 2 .C 2 H 6 The sodium of the monosodio -compound can be replaced by an alkyl group and the remaining methylene hydrogen then replaced by sodium, which can subsequently be substituted by an alkyl group different from the first. An example of this synthesis is as follows : CH 3 I - Nal + C Hydrolysis of the final ester yields Methylethylmalonic Acid. Homologues of succinic acids can be obtained as follows : r^r\ /*i TT 4- Tir PTT^ 2 ' ^2 n 5 PO P TT - Dr '^-' Jt:1< ^-,pTj Ethyl sodiomethylinalouate Ethyl a-bromopropionate C0 2 C 2 H 5 NaBr + CH,- C0 2 C 2 H 5 When this complex ester is saponified and the acid thus formed heated to expel C0 2 from one of the carboxyl groups united to the same carbon atom, symmetrical dimethyl- succinic acid is obtained : C0 2 H CO 2 H C0 2 H CH 3 - C CH<^E 2H - CO 2 + CH 3 CH CH . CH 3 I ^"^3 CO 2 H Also 2 mols. of ethyl sodiomethylmalonate (or ethyl sodiomalonate or its homologues) can be condensed in ethereal solution by means of bromine or iodine : CO 2 -C 2 H 5 C0 2 -C 2 H 5 CO P TT 2CH 3 CNa0 melts at 200 (sublimes) 130 boils at 160 202 91 161 132 o/3-Hydromucic acid C0 2 H CH 2 CH 2 - CH : CH-C0 2 H /3y- C0 2 H.CH 2 .CH:CHjCH 2 .CO 2 H 96 boils at 223 169 (stable) 195 (labile) FUMARIC AND OLEIC ACIDS 318 As far as the carboxyl groups are concerned, these acids have chemical properties similar to those of the saturated dibasic acids (see p. 305), whilst as they are unsaturated compounds, they are able to combine with 2 atoms of hydrogen or halogen or with 1 mol. of a halogen hydracid. They are usually prepared from the mono- and di-halogen substitution products of succinic acid and its homologues by removing either 1 mol. of halogen hydracid (by heating with KOH) or 2 atoms of halogen : CH 2 C0 2 H HBr + C0 2 H CH I II CHBr-C0 2 H CH-C0 2 H Monobromosuccinic acid Fumaric acid Distillation of the saturated dibasic hydroxy-acids results in the removal of 1 mol. of H 2 and the formation of unsaturated acids. The most interesting cases of stereoisomerism were considered on p. 21. When fumaric acid is either heated or treated with PC1 5 , POC1 3 , or P 2 5 , it is converted into maleic anhydride. Maleic acid is transformed into fumaric acid by heating at 200 in a sealed tube or by the action of bromine or of various acids in presence of sunlight. . FUMARIC ACID (trans-Butendioic Acid), C 4 H 4 O 4 or CO 2 H-CH, II HC-C0 2 H forms small white prisms which have a marked acid taste and are almost insoluble in water ; it does not melt but sublimes at about 200, subsequently losing water and becoming converted largely into maleic anhydride. It is moderately widespread in certain vegetable organisms, e.g. in fungi, truffles, Iceland moss, and especially in Fumaria officinalis. It can be prepared by the ordinary synthetical methods and also by the action of phosphorus and bromine on succinic acid, the product obtained being decomposed by heating with water. It is stereoisomeric with maleic acid (see p. 21) and its reduction to normal succinic acid by means of nascent hydrogen confirms its constitution, which is also deduced from the decomposition of the corresponding ozonide (Harries). The Silver Salt, C 4 H 2 4 Ag 2 , is slightly soluble in water, and the same is the case with the barium salt, C 4 H 2 O 4 Ba + 3H 2 0, which in boiling water becomes insoluble and separates in the anhydrous form, C 4 H 2 4 Ba. MALEIC ACID (cis-Butendioic Acid), C 4 H 4 O 4 or CH-C0 2 H forms large II CH-CO 2 H prisms melting at 130 and having an unpleasant taste ; it boils at 160, losing water and becoming converted partially into maleic anhydride. It is readily soluble in water. Its ready transformation into maleic anhydride is explained by the stereo- chemical relations considered on p. 21 et seq., and in many general methods of preparing the acid, the anhydride is first obtained. The Barium Salt, C 4 H 2 O 4 Ba + H 2 O, is soluble in hot water, from which it crystallises well. Electrolysis of the alkali salts of fumaric and maleic acids yields acetylene. When heated with sodium hydroxide at 100, these two acids are converted into inactive maleic acid. ITACONIC ACID (Methylenesuccinic Acid), C 5 H 6 O 4 or CH 2 : C-CO 2 H, is a white CH 2 -C0 2 H substance melting at 161 and non-volatile in steam. It is obtained by the action of water on its anhydride, the latter being formed by the interaction of citraconic anhydride 314 ORGANIC-CHEMISTRY and water at 150. Hydrogen converts it into pyrotartaric acid and permanganate into hydroxyparaconic acid. On electrolysis it yields allene, CH 2 : C : CH 2 . MESACONIC ACID (Methylfumaric Acid), C 5 H 6 O 4 or CO 2 H-C-CH 3 , is formed by II HC-CO 2 H heating citraconic or itaconic acid with water at 200 or by treatment of citraconic acid with dilute HNO 3 or concentrated NaOH, or with traces of bromine in sunlight. It is difficultly soluble in water, melts at 202, and does not distil in steam. When electrolysed it forms allylene, CH 3 -C : CH, while with hydrogen it gives pyrotartaric acid and with permanganate, pyrotartaric and oxalic acids. It forms a barium salt, C 5 H 4 4 Ba +4H 2 O. CITRACONIC ACID (Methylmaleic Acid), C 5 H 6 4 or CH 3 -C CO 2 H, is formed from II HC CO 2 H the corresponding anhydride and water. It melts at 91, differs from the two preceding acids by being very soluble in water, distils in steam and readily gives the anhydride again. On electrolysis it yields allylene, while with hydrogen it forms pyrotartaric acid. GLUTACONIC ACID, C 5 H 6 O 4 or CO 2 H-CH : CH-CH 2 -CO 2 H, is isomeric with the three preceding acids and is obtained by hydrolysing the corresponding ester with HC1 ; it melts at 132 and the hydrogen of its CH 2 -group is replaceable by sodium (see p. 309). Of the higher homologues of these acids mention may be made of the alkylitaconic acids, with which, on heating with NaOH solution, the position of the double linking changes, giving alkylmesaconic and alkylaticonic tfcids (Fittig), e.g. isdbutylaticonic acid (CH 3 ) 2 CH-CH : CH.CH(CO 2 H).CH 2 .CO 2 H, which melts at 93 ; with alkalis these acids undergo the reverse change to some extent. The calcium and barium salts of the alkylmesaconic acids are readily soluble in water, and those of the alkylitaconic acids slightly soluble. Of these homologous acids, the following deserve mention : PYROCINCHONIC ACID (Dimethylmaleic or Dimethylfumaric Acid), C 6 H 8 O 4 or CO 2 H-C = C-CO 2 H. Of the two stereoisomerides, only dimethylmaleic acid was until I I CH 3 CH 3 recently known and then only as the anhydride, namely, pyrocinchonic anhydride (m.pt. 96, b.pt. 223). Dimethylmaleic acid cannot exist in the free state, as it immediately gives up water, forming the anhydride ; its esters are, however, known. CH 8 .C.COv The anhydride, ^)O, may be prepared in various ways, e.g. by distilling in CHg-C-CXK steam the product of the interaction of pyrotartaric acid and sodium succinate. But a better yield is obtained by first preparing the riitrile of methylacetoacetic acid and distilling this in a vacuum. According to A. Bischoff, the stereoisomeride, Dimethylfumaric Acid, CH 3 -C-CO 2 H. II C0 2 H.C-CH 3 could not, owing to stereochemical considerations, be formed in the free state. But Fittig and Kettner (1899) and also E. Molinari (1900) have succeeded in isolating it in various ways. 1 It forms white crystals, m.pt. 152 ; its amido -derivatives have also been prepared. 1 Fittig and Kettner, making use of the property of variant acids, homologous with citraconic acid, of yielding the corresponding fumaroid isomeride when simply heated'with alkali, obtained from pyrocinchonic anhydride, the two acids : one melting at 151, to which is ascribed the constitution CH 2 : C-CO 2 H (p-methylitaconic acid), and I CH,- CH-COjH another melting at 240 and regarded as CH,-C-CO a H (dimelhylfumaric acid). It is highly probable, for the II COjH-C-CH, following reasons, that the latter constitution should be attributed to the acid melting at 151. By a long series of investigations (1881 to 1896), Korner and Menozzi showed that, in general, the treatment of a-amino-acids with methyl iodide in presence of caustic potash yields the corresponding betaines (condensed alkyl- substituted amines) ; but the p-amino-acids, if similarly treated, always yield the corresponding unsaturated, non- nitrogenous acids of the fumaroid type (betaines being probably formed as intermediate products). As the same f)-amino-acid can be obtained from the two stereoisomeric unsaturated acids, this general reaction renders it possible to pass from a malenoid unsaturated acid to the corresponding fumaroid stereoisomeride By applying this reaction to pyrocinchonic anhydride, E. Molinari arrived at the expected stereoisomeride (dinxthylfumaric acid), melting at 152. TRIBASIC ACIDS Bauer (1904) made the interesting observation that dimethylfumaric acid and, in general, compounds containing carboxyl or alkyl or phenyl groups or bromine atoms united to two carbon atoms connected by a double linking do not unite with bromine. HYDROMUCONIC ACIDS, C 6 H 8 O 4 . Of these are known (1 ) the a/3 -unsaturated acid, C0 2 H-CH 2 -CH 2 -CH : CH-C0 2 H, which is stable and melts at 169 ; with permanganate S y ft a it yields succinic acid. (2) The unstable /3y-acid, C0 2 H.CH 2 -CH : CH-CH 2 -CO 2 H, which melts at 195 and is obtained by reducing muconic acid ; when heated with alkali, it is converted into the stable isomeride, whilst with permanganate it gives malonic acid, C0 2 H.CH 2 .C0 2 H. Of the DIOLEFINEDICARBOXYLIC ACIDS, only Muconic Acid, CO 2 H-CH : CH-CH : CH-CO 2 H, melting above 260, need be referred to. Of the ACETYLENEDICARBOXYLIC ACIDS, mention will be made only of Acetylenedicarboxylic (Butindioic) Acid, CO 2 H-C i C-CO 2 H, which melts and decomposes at 175 ; it crystallises with 2H 2 O. It is obtained on removing HBr from dibromo- or isodibro mo -succinic acid by means of potash. Diacetylenedicarboxylic Acid, CO 2 H-C i C-C : C-CO 2 H + H 2 O, turns dark red in the light and explodes at 177. When reduced with sodium amalgam, it yields hydro - muconic acid. Tetracetylenedicarboxylic Acid, CO 2 H-C i C-C : C-C I C-C i C-CO 2 H, forms white crystals which blacken rapidly in the light and explode viol en tly on heating. C. TRIBASIC ACIDS These have usually been obtained synthetically and are not very stable since they readily yield carbon dioxide and dibasic acids on heating ; their esters, however, exhibit increased stability. Their properties and methods of preparation vary according as the carboxyl groups are united to one or to various carbon atoms. Of the many such acids known, the following may be mentioned : TRICARBALLYLIC ACID (symm. wa w- Propanetricarboxylic or Pentanedioic-3- methyloic Acid), CH 2 -C0 2 H, occurs in the deposits left on concentrating beet-sugar juices CH-CO 2 H CH 2 -CO 2 H in vacuo. Synthetically it is obtained by converting glycerol into the tribromohydrin or allyl tribromide, which is treated with potassium cyanide to give the corresponding tricyano -compound, the latter being then hydrolysed to tricarballylic acid ; the con- stitution of the acid is thus proved. This acid forms white, prismatic crystals melting at 166. It can also be prepared by reducing unsaturated tricarboxylic acids (e.g. aconitic acid). CO 2 H CO 2 H CO 2 H CAMPHORONIC ACID (aa/3-Trimethylcarballylic Acid), CH 3 -C C. CH 2 CH 3 CH 3 is formed on oxidising camphor, of which it serves to indicate the constitution ; it melts at 135.' ACONITIC ACID is an unsaturated tribasic acid of the constitution C0 2 H-CH 2 - C(CO 2 H) : CH-CO 2 H, and is found in beetroot, sugar-cane, Aconitum napellus, &c. It is obtained synthetically by eliminating C0 2 from citric acid by the action of heat or of various reagents. It melts at 191, losing C0 2 , and forming itaconic anhydride. It dissolves readily in water and with nascent hydrogen generates tricarballylic acid, its structure being indicated by this reaction. 316 ORGANIC CHEMISTRY D. TETRABASIC ACIDS These are formed from ethyl sodiomalonate (see p. 309) by means of an unsaturated ester, e.g. of fumaric acid. When heated, they lose C0 2 , forming tribasic and, better, dibasic acids. Olefinetetracarboxylic Acids are also known. FF. DERIVATIVES OF THE ACIDS I. HALOGEN DERIVATIVES One or more of the hydrogen atoms of an alkyl group united with carboxyl can be replaced by halogens, the carboxyl group being left intact. The halogen derivatives of the acids, thus obtained, are more markedly acid in character than the original substances. They are obtained by the action of chlorine or bromine in sunlight or, better, by heating the acid with the halogen in presence of a little water or sulphur. On the other hand, the hydroxyl of the carboxyl group can be replaced, forming acid halides ; CO X (by treating the acid with phosphorus chloride or bromide). That the halogen has replaced the hydroxyl group is shown by the fact that these acid halides yield the original acids when treated with cold water, whilst halogens are not displaced from alkyl residues in this way. These acid chlorides and bromides readily give the monochloro- and mono- bromo- acids when treated further with chlorine or bromine. (a) HALOGENATED ACIDS When the carbon atom (a), to which the carboxyl group is attached, is not united directly with hydrogen [e.g. in trimethylacetic acid, (CH 3 ) 3 C-C0 2 H], bromine is not taken up (see p. 315). The constitution of a halogenated acid, or rather the position of the halogen atom, is deduced from that of the cor- responding hydroxy-acid (containing a hydroxyl -group in place of the halogen) obtained by heating the halogenated acid with sodium carbonate solution or with water and lead oxide. On the other hand, the passage from hydroxy-acid to the corresponding halogenated acid can be effected by treatment with phosphorus chloride or bromide. The acid character becomes more marked on passing from the iodo- to the bromo- and then to the chloro-compounds and also increases with the number of halogen atoms in the molecule. While the a-halogenated acids readily yield the corresponding hydroxy- acids, the /3-acids yield the corresponding unsaturated acids (see p. 292) and may even lose C0 2 , giving unsaturated hydrocarbons. But the y-halogenated acids, when heated with sodium carbonate solution or with water alone, give up a molecule of halogen hydracid and yield, not the unsaturated acids, but lactones (see p. 295). When halogenated acids are prepared by the interaction of an unsaturated acid with a halogen hydracid (e.g. HI), the halogen becomes attached to the least hydrogenated carbon atom (see p. 96). Thus, with a A^-acid, where the double linking is between the a- and |3-carbn atoms, the halogen unites with the latter. The halogenated and poly-halogenated acids exhibit isomerism, since the halogen atom may be joined to the a, ft, y, &c., carbon atom, or several halogen atoms may be united with one and the same carbon atom or with different ones. ACIDHALIDES 317 When heated with potassium cyanide, the mono-haloid acids yield cyano acids : CH 2 C1-COOK + KCN = KC1 + CN-CH 2 -COOK. Potassium chloroacetate Potassium cyanoacetate With sodium sulphite they give dibasic sulpho-acids, the sulphonic group of which is readily replaced by hydroxyl by boiling with alkali : Na 2 S0 3 + Cl-CH 2 -COONa = NaCl + S0 3 Na-CH 2 -COONa. Sodium sulphoacetate With reference to the affinities of the halogenated acids, see Note on p. 268. MONOCHLORACETIC ACID (Chlorethanoic Acid), CH 2 C1-COOH, is prepared by the general method, that is, by passing dry chlorine into hot acetic acid in presence of acetic anhydride, phosphorus, or sulphur. It forms rhombic crystals which corrode the flesh and melt at 62 ; on solidification an unstable modification is obtained which, for some time, melts at 52 ; it boils at 186. When heated with water or alkali it gives Hydroxyacetic Acid (glycollic acid), OH-CH 2 -CO 2 H; with ammonia it yields Amino- acetic Acid (glycineor glycocoll), NH 2 CH 2 C0 2 H. The properties of the other halogenated acids are given in the Table on the next page. (b) ACID HALIDES Of these compounds the most important are the chlorides of the acid radicals, which are termed acichlorides or chloranhydrides . Although acetyl chloride, CH 3 -CO-C1, is readily obtainable, it has not been found possible to prepare formyl chloride, H- CO- Cl, a mixture of CO+HC1 being always obtained. These compounds are usually colourless liquids, which have pungent odours and fume strongly in the air, the moisture in the latter liberating hydrogen chloride. Their boiling-points are below those of the corresponding acids, and they distil without decomposing ; the higher members are, however, solid and do not distil unchanged even in a vacuum. The principal methods for preparing these substances are as follows : (a) The organic acid is heated for a short time on the water-bath with PC1 5 (with higher acids.), PC1 3 (with acids below C 10 ) or, in some cases, sulphuryl chloride, S0 2 C1 2 : C U H 23 -CO-OH + PC1 5 = C n H 23 -CO-Cl + HC1 + POC1 35 Laurie acid the phosphorus oxychloride and hydrochloric acid being eliminated by distillation in vacua ; or, 3CH 3 -CO-OH + 2PC1 3 = 3CH 3 -CO-C1 + 3HC1 + P 2 3 , the acetyl chloride thus formed being separated by distillation, the P 2 3 being left in the residue. (b) With thionyl chloride the acids yield the chloranhydrides, the other products formed at the same time being volatile and hence easily removable : X-CO-OH + SOC1 2 = X-CO-C1 + HC1 + S0 2 . (c) In some cases the acid is treated simply with HC1 in presence of a dehydrating agent, (P 2 5 ) : CH 3 -CO-OH + HC1 = H 2 + CH 3 -CO-C1. CHEMICAL PROPERTIES. The great reactivity of the chlorine atom of these substances renders them of considerable importance in chemical syntheses. Water, ammonia (amines), and alcohols decompose them in the cold with great violence : CH 3 -CO-C1 + H 2 = HC1 + CH 3 -CO-OH CH 3 -CO-C1 + NH 3 = HC1 + CH 3 -CO-NH 2 (acetamide) CH 3 -CO-C1 + C 2 H 5 -OH = HC1 + CH 3 -CO-OC 2 H 6 (ethyl acetate), 318 ORGANIC CHEMISTRY ffi ^ -C 'c? > 6 ' "*"" ^2 '^ 0) r2 HH O _fi o 03 rH '3 h^H Q . K* rS cj + O o S W jl -J a o w s CJ O g -2 o cT T3 o JB Tb a 2 O.iJ w S ** O cS ^ a >i e O ^a f{ A g 0) P< 2 A T3 C IN O O " rS 5 gii ^ U> < '3 IN j h a o O b eg II S '- H c 2 hjj i p -3 r^H ^ jj[ o O * 'bo fl cS '2 _\ o 3 g . T< M HH Preparati" By heating : CC1 3 .CHO+KCN + H 2 ( When heated with alkali, The silver salt with water By oxidising chloral with hot it gives chloroform On hydrolysis, gives male -^ II o " PH O .2 " o fe cS " Acrolein + HCI = /3-chl HN0 3 = /i3-chloropropi From propionic acid + Cl (si 5 '3 efi u o /j-Bromoacrylic acid + H W '3 rs _o 1 o a 1 'o a, o o o o o Q o bD 1 1 lO 00 O i 00 1 9 O5 O5 oo O 00 i 1 1 1 1 i 1 0 ' r acetyl xide ' ( CH 3' C )20- PROPERTIES. The first members of the series are liquid, the higher ones solid ; they generally dissolve but slightly in water, their transformation into acids being very slow. They have a neutral reaction and are soluble in ether and often in alcohol. With ammonia and the primary and secondary amines, they form amides and ammoniacal salts: (CH 3 CO) 2 + 2NH 3 = CH 3 -CO-NH 2 (acetamide) + CH 3 -CO-ONH 4 (ammonium acetate). When heated with an alcohol, they give the corresponding ester and acid : (CH 3 CO) 2 + C 2 H 5 -OH = CH 3 -COOC 2 H 5 + CH 3 -COOH. With halogen hydracids in the hot they yield the halides of the acids and the free acids : (CH 3 CO) 2 O + HC1 = CH 3 -CO-C1 + CH 3 -COOH. Aldehydes combine with anhydrides, forming esters, while sodium amalgam reduces anhydrides to aldehvdes and alcohols. GENERAL METHODS OF PREPARATION, (a) By the action of acid chlorides on the dry alkali salts of the corresponding acids : CH 3 -CO-C1 + CH 3 -COONa = NaCl + cH 3 .CO >a (6) The same result is obtained by the action of phosphorus oxychloride (or phosgene COC1 2 ) on a mixture of the alkali and alkaline-earth salts of the corresponding acid, the acid chloride being formed as an intermediate product. 320 ORGANIC CHEMISTRY (c) The higher anhydrides are obtained from the corresponding acids by the action of acetyl chloride : CH 3 -COC1 + 2X-COOH - HC1 + CH 3 -COOH + (X-CO) a O. (d) The formation of anhydrides from the acids by the subtraction of water (by means of P 2 5 ) gives low yields, the best being obtained with palmitic and stearic acids (using acetic anhydride in the hot as dehydrating agent). The properties of the best-known anhydrides are given in the following Table : Formula Name Melting- point Boiling-point Specific gravity (CH 3 .CO) 2 O Acetic anhydride , 136-5 1-078 (at 21) (C 2 H 5 .CO) 2 Propionic anhydride ~ v 168-6 1-034 (at 0) (C 3 H 7 .CO) 2 norm. Butyric anhydride 192 0-978 (at 12-5) 5> Iso butyric anhydride 182 0-958 (at 16-5) (C 4 H 9 .CO) 2 Isovaleric anhydride 215 Trimethylacetic anhydride 190 (C 5 H U .CO) 2 norm. Caproic anhydride 242 0-928 (at 17) (C 6 H 13 .CO) 2 CEnanthic anhydride + 17 257 0-912 (at 17) (C 7 H 15 .CO) 2 Caprylic anhydride 1 186 (15 mm.) (C 8 H 17 -CO) 2 Pelargonic anhydride + 16 207 (C U H 23 .CO) 2 Laurie anhydride . + 41 166 (vacuum) (Ci3H 27 .CO) 2 Myristic anhydride + 51 198 (C^H^-CO^O Palmitic anhydride 55-66 (C 17 H3 B .CO) 2 Stearic anhydride . . 72 ACETIC ANHYDRIDE (Ethanoic Anhydride), (CH 3 -CO) 2 O, is of some importance industrially owing to its formation of acetyl derivatives with alcohols or with primary or secondary amines. It is a suitable reagent for determining how many hydroxyl groups an organic substance contains (see Acetyl Number, p. 189). It is a colourless, very mobile liquid, sp. gr. 1-078 at 21, b.pt. 136-5, and has a pungent odour. It is prepared by dropping 5 parts of acetyl chloride on to 7 parts of dry powdered sodium acetate, which is kept cool meanwhile. The mixture is subsequently gently heated for a short time and the anhydride then distilled off on a sand-bath. It is purified by redistillation in presence of a little anhydrous sodium acetate, the portion boiling at the correct temperature being collected. It is also prepared industrially by method (b) given above (Ger. Pats. 161,882 ; also 132,605 and 146,690). Commercial acetic anhydride costs about 18 per quintal, and the highly purified product (puriss.) 24. The anhydrides of di- and poly-basic acids are not of great importance and are con- sidered to some extent in dealing with the corresponding acids (succinic, pyrocinchonic, &c.) ; with water they yield the acids with moderate readiness (see pp. 305 and 314). III. HYDROXY-ACIDS A. SATURATED DIVALENT MONOBASIC ACIDS These may be regarded as derived from monobasic acids by the substitu- tion of an atom of hydrogen (not that of the carboxyl group) by a hydroxyl group. These acids possess, at the same time, acidic and alcoholic characters and are hence termed divalent monobasic acids or divalent alcohol acids. The hydroxyl and the carboxyl groups may be substituted at the same time, the compounds then exhibiting the general properties of the acids and alcohols, in addition to new and special characters varying with the position occupied by the carboxyl relatively to the hydroxyl (see pp. 295 and 297). HYDROXY-ACIDS 821 They are usually syrups which may undergo crystallisation ; in comparison with the corresponding fatty acids, the hydroxy-acids are more soluble in water and in alcohol, but less soluble in ether. They do not distil unchanged and often lose water, forming anhydrides. GENERAL METHODS OF PREPARATION, (a) By oxidising dihydric alcohols so as to transform the primary alcoholic group into carboxyl. (6) By boiling unsaturated acids with sodium hydroxide, so that a mole- cule of water is added at the double bond. (c) By substituting the halogen of a monohalogenated monobasic acid by hydro xyl ; this is effected by treatment with KOH or with silver acetate, the diacetate formed in the latter case being hydrolysed by heating with sodium carbonate : CH 2 C1'COOH + H 2 = OH : CH 2 -C0 2 H + HC1. Monochloracetic acid Glycollic acid (d) a -Hydroxy-acids are obtained by hydrolysing the nitriles formed on treating the aldehydes or ketones (having one atom of carbon less) with hydrocyanic acid : CH 3 -CH< + 2H 2 = NH 3 + CH 3 -CH(OH)-COOH. Ethyl idenecyanohydrin Glycolcyanohydrin, OH CH 2 CH 2 CN, yields ethylenelactic acid, OH-CH 2 -CH 2 -COOH. (e) By the action of nitrous acid on amino-acids : COOH-CH 2 -NH 2 + NO- OH = H 2 + N 2 + COOH-CH 2 -OH. Glycocoll ( / ) By reducing aldehydic or ketonic acids : CH 3 -CO-COOH + H 2 = CH 3 -CH(OH)-COOH. Pyruvic acid Lactic acid (g) By oxidation of acids containing a tertiary carbon atom, >CH-COOH, with permanganate. PROPERTIES AND CONSTITUTION. The constitutions of these acids can always be deduced from the syntheses indicated above. That they con- tain an alcoholic group is shown by the fact that the hydroxylic hydrogen can be replaced by an alkyl group, giving true non-hydrolysable ethers. Similarly the presence of a carboxyl group is shown by the formation of hydrolysable esters. The isomerism exhibited is the same as with the haloid derivatives of the acids. The number of alcoholic hydroxyl groups is determined by the acetyl number (see p. 189). The reactivity, which corresponds with the dissociation constant, increases with the proximity of the hydroxyl to the carboxyl group. a-, /3-, /-, and g-hydroxy-acids are distinguished also by the products resulting from the elimination of one or more molecules of water. Thus, a-hydroxy-acids, when heated, lose 2 mols. of H 2 per 2 mols. of acid, the hydroxyl group of the one reacting with the carboxyl group of the other ; the compound formed is called a lactide and is a double ester, which yields the acid again on hydrolysis with hot water or dilute acid : CO-CH-CH 3 COOH-CH(OH)-CH 3 | = 2H 2 + 60 CH 3 -CH(OH)-COOH I CH 3 -CH-CO 2 mols. Lactic acid Lactide II 21 322 ORGANIC CHEMISTRY Further, a-hydroxy -acids, if heated with sulphuric acid, yield the aldehydes or keton.es from which they can originate (see above), formic acid being also formed. The |3-acids, however, lose only 1 mol. of water, giving unsaturated acids (see p. 292), while, when boiled with 10 per cent, potassium hydroxide solution, they give at the same time aft- and ay- unsaturated acids a reversible reaction leading to a position of chemical equilibrium ; when heated with sulphuric acid they form acids of the acrylic series. The y- and S-acids lose 1 mol. of water, yielding lactones (internal anhydrides) : OH-CH 2 -CH 2 -CH 2 -COOH = H 2 + CH 2 -CH 2 -CH 2 -CO y-Hydroxybutyric acid o Butyrolactone which are almost always formed when attempts are made to liberate these hydroxy-acids from their salts. -The lactones are neutral liquids soluble in water, alcohol, and ether ; they dis.til unchanged and with alkali form the salts of the corresponding hydroxy-acids. When the hydroxy-acids are heated with sulphuric acid, they furnish the corresponding fatty acids. GLYCOLLIC ACID (Hydroxyacetic or Ethanoloic Acid), OH-CH 2 -COOH, crystal- lises in needles or plates melting at 80, and is soluble in water, alcohol, or ether. In nature it is found in immature eggs and in the leaves of the wild vine. It can be obtained by the general methods .given above and also by oxidising alcohol or glycol with dilute nitric acid or by reducing oxalic acid with nascent hydrogen. It is usually prepared by hydrolysing monochloracetic acid with KOH [general method (c)]. Of some interest is the formation of the various anhydrides of glycollic acid, these being formed by the removal of 1 mol. of H 2 O from two mols. of the acid as follows : (1) From the two alcohol groups, giving a true ether with two free acid groups, .- rnnw diglycollic acid, m.pt. 148; (2) from the two carboxyl groups; this OH CH CO should give the anhydride of glycollic acid, r 'nrC>0 wmc h i g n t yet known ; vJ-ii v_y-Tl 2 \j\J (3) from one alcohol and one acid group, giving a true ester, glycolglycollic acid, s gives eit CH CO * OH CH CO . ~>O. Also loss of 2H 2 O from the two alcoholic and acidic groups gives either COOH (1) Diglycollic anhydride (anhydride and ether at the same time), j-2 (melting at 97 and boiling at 240), or, when each molecule of water separates from 1 CTT CO alcoholic and 1 acidic group, (2) the isomeric glycollide, O^^ 2 ' r ^>0, melting at 86. Glycollic acid forms a calcium salt, (OH-CH 2 -COO) 2 Ca + 3H 2 0, insoluble in water. The most important derivative of glycollic acid is GLYCOCOLL (Glycine or Aminoacetic or Aminoethanoic Acid), COOH CH 2 -NH 2 , which is the first member of the amino-acid series so important to vegetable physiology. It is obtained, together with secondary products, by the action of concentrated ammonia solution on monochloracetic CHoCl-COOH + 2NH 3 = NH 4 C1 + NH 2 . CH 2 . COOH. It is always formed in the decomposition of hippuric acid (benzoylglycocoll) with HC1, or by the action of acid or alkali on gelatine. It can also be obtained by the reduction of ethyl cyanocarboxylate by means of nascent hydrogen or from cyanogen and boiling hydriodic acid. Its homologues are prepared synthetically in various ways, e.g. by treating aldehyde- ammonias with hydrocyanic acid and hydrolysing the amino -cyanides thus obtained with HC1. Glycocoll crystallises in rhombic columns soluble in 4 parts of water but insoluble in alcohol or ether ; it has a sweetish taste and melts and decomposes at 230. LACTIC ACIDS 32B The fact that the amino -group cannot be expelled by hydrolysis establishes the structure of glycocoll. It behaves as both acid and base, forming salts with acids and also with bases. Its copper salt separates in large, dark blue needles on dissolving cupric oxide in hot glycocoll solution : (C 2 H 4 O 2 N) 2 Cu + H 2 0. With ferric chloride it gives an intense red coloration. When heated with baryta, it loses C0 2 , forming methyl- amine ; with nitrous acid it gives glycollic acid. Various alkyl and other derivatives have been obtained synthetically from glycocoll : CH 2 -NH-COCH 3 CH 2 -NH-CH 3 CH 2 N(CH 3 ) 3 I ; I ; I I COOH COOH CO O Aceturic acid Sarcosine Betaine (derived from riaffeine and from creatine) (from the beet) With nitrous acid, the esters of glycocoll yield ETHYL DIAZOACETATE, N \ NH 2 -CH 2 -COO-C 2 H 5 + HN0 2 = 2H 2 + || ">CH-COOC 2 H 5 , which is -a W yellow oil boiling at 141 and readily decomposes and reacts with evolution of nitrogen ; it serves for the synthesis of pyrazole. LACTIC ACIDS, OH-C 2 H 4 -C0 2 H The two structural isomerides foreseen by theory are known : a- and /3-hydroxypropionic acids. Also the a-acid exists in two stereoisomeric forms (I = laevo- and d = dextro-rotatory) owing to the presence of an asymmetric carbon atom (p. 18) and in an inactive form (i = inactive), con- sisting of a mixture in equal proportions of the two stereoisomerides. These lactic acids form anhydrides similar to those of glycollic acid (see above). The lactic acids give Uffelman's reaction, that is, they cause the amethyst- coloured solution obtained on adding a drop of ferric chloride to a dilute salicylic acid solution to turn yellow ; this reaction is also given by citric, oxalic, and the tartaric acids. (1) i-ETHYLIDENELACTIC ACID (I + d) (2-Propanoloic or a-Hydroxy- propionic Acid or Ordinary Lactic Acid of Fermentation), CH 3 -CH(OH)-COOH, is found in milk rendered acid by the action of the lactic acid bacillus (see Fig. 114, p. 122), in milk-sugar (also cane- and grape-sugars, gum, starch, &c.) which undergoes acid fermentation (lactic fermentation) even in absence of air, although oxygen facilitates the change. Cabbage fermented with vinegar and salt (Sauerkraut), gastric juice, putrefied cheese, fresh fodder siloed, and fermented muscular juices and the brain 1 also contain free lactic acid. When pure it melts at 18 and boils at 120 under 12 mm. pressure, but usually it forms a dense syrup soluble in water, alcohol, or ether. It is opti- cally inactive, as it consists of a racemic mixture of dextro- and Isevo-acids (see pp. 19 and 20). The two modifications can be separated by crystallisation 1 It appears now to be proved that the lactic acid in the human organism is formed in proportion to the muscular and cerebral work, and, together with carbon dioxide, which is also a waste product of the cells of the organism during wakefulness, produces sleep. While we sleep, the blood carries off these waste products more easily, the cells then recovering their function and their sensibility. The connection between sleep and fatigue is well known and is shown not only by the fact that after great muscular or cerebral fatigue sleep is more profound, but by the results of the following experiment : if the blood of a very tired dog is injected into the veins of another dog in a normal state, this dog soon exhibits signs of great fatigue and goes to sleep ; these results are not observed if the blood injected is that of a non-fatigued dog. During heavy muscular labour, the air expired contains more CO 2 than in a state of repose and more still than during sleep. The carbon dioxide diminishes the oxygen so much needed by the muscles and brain, so that the activity of these remains depressed. As is well known, lactic acid has a depressing action on the nervous cells, injection of the acid into the veins of any person inducing symp- toms of fatigue and sleepiness and finally sleep. The continuance of sleep is due to the fact that the blood flows more slowly to the brain, to which it hence carries less oxygen. It appears, indeed, to be proved that in general five or six hours' sleep very deep for two hours is sufficient for the blood to wash away these waste products of active cellular work and to restore activity to all the cerebral centres. 324 ORGANIC CHEMISTRY of the strychnine salts or by cultivating in the solution Penicillium glaucum, which first destroys the Isevo-acid (see p. 22). When heated, the active acid is transformed, to the extent of one-half, into the optical eriantiomorph, so that the inactive racemic acid is obtained. If kept in a desiccator, it is partly converted into anhydride owing to loss of water. When distilled under reduced pressure, it yields water, carbon dioxide and lactide (see above). If heated with dilute sulphuric acid it decomposes, like many other a-hydroxy-acids, into acetaldehyde and formic acid. PREPARATION. Various processes have been tried for the preparation of lactic acid. For instance, 3 kilos of cane-sugar and 15 grms. of tartaric acid are dissolved in 13 litres of boiling water. In a few^days' time, after the cane-sugar has been converted into glucose and levulose, 4 litres of acid milk and 100 grms. of putrefied cheese (also 1-5 kilo of zinc carbonate to fix the lactic acid, which otherwise would arrest the lactic fermentation) are added and the mixture left for a week at a temperature of 40 to 45, by which means the maximum production of lactic acid is obtained. The acid separates as zinc lactate in crystalline crusts which, after purification (by recrystallisation), are suspended in water and decomposed with H 2 S in order to remove the zinc as insoluble sulphide. The filtered liquid is concentrated to a syrupy consistency and then extracted with ether, which does not dissolve the impurities (zinc salts, mannitol, &c.) ; on evaporation of the ether, pure syrupy lactic acid is obtained. 1 Besides the decomposition of the sugar, various secondary reactions always accompany lactic fermentation, and the yield of the acid is scarcely 20 per cent, of the weight of the sugar taken. A better yield is, however, obtained by Larrieu's process (Fr. Pat. 206,506), which con- sists in treating, say, 900 kilos of starch with 100 kilos of malt and with hot water to bring the temperature to 50, this being finally raised to 75, the mass being continually stirred. Half a kilo of ammonium nitrate is next added to the vat and then the lactic ferment, the temperature being maintained at 50 to 60 for 20 to 30 days, and the acid formed being gradually half saturated with soda. The mass is ultimately filtered and the liquid con- centrated to a sp. gr, of 1-21 (25 Be.), mixed with 500 kilos of powdered calcium carbonate and filtered. The solution of calcium lactate is decomposed with the calculated quantity of sulphuric acid, the calcium sulphate being removed by filtration and the aqueous lactic acid evaporated to a syrupy consistency. Jacquemin prepares the acid from worts similar to those employed in breweries (barley mashed at 50 with malt, then boiled to destroy the diastase and cooled to 45) by the addition of pure lactic ferment and calcium carbonate. After 5 to 6 days, the mash is filtered and concentrated, the calcium lactate being then decomposed in the usual way with sulphuric acid. Dreher works in a similar manner, but with glucose solutions containing 1 per cent, of nutrient substances for the ferment (e.g. sodium phosphate, nitre, salt, &c.). Industrially, however, lactic acid is now always obtained from milk residues (whey or molasses of milk-sugar, which remain after the removal of the butter from the milk in the separator ; also cheese by coagulation with rennet in the hot). The whey is con- centrated in open vessels or, better, in vacuum pans, to 16 Be., and is then introduced into wooden vessels in which, at a temperature of 40, the lactic ferment is added in the form either of part of the liquid from a previous fermentation or of putrefied cheese. Powdered chalk is added to neutralise the acid formed, the liquid being stirred from time to time and the fermentation allowed to continue for 10 to 12 days. After decantation, the calcium lactate is decomposed with dilute sulphuric acid, the liquid mass being well mixed. In some cases, before the calcium lactate is decomposed, it is separated by con- centrating the solution, and is recrystallised from a little hot water, which should dissolve 20 per cent, of it. The calcium sulphate formed is removed by passing the mass through 1 Kiliani treats 500 grms. of inverted sugar with 250 grms. of water and 15 grms. of sulphuric acid at 50 to 60 for 2 hours, and then adds gradually 400 c.c. of concentrated caustic soda solution (1 : 1), the liquid being kept boiling meanwhile. The soda is subsequently neutralised with 50 per cent, sulphuric acid and the solution left for 24 hours to deposit crystalline sodium sulphate. The lactic acid is extracted with alcohol which does not dissolve the sul- phate the alcohol being recovered by distillation. The crude lactic acid remaining is diluted, saturated with Zinc carbonate and evaporated ; the zinc lactate is then allowed to separate and is filtered off, redissolved in hot water and decomposed with HjS. After filtration, the liquid is concentrated in vacua, pure lactic acid being thus obtained. LACTIC ACIDS 325 a filter-press (see figure in the section on Sugar) and the clear lactic acid solution con- centra ted in a double -or triple-effect apparatus until it attains a concentration of 50 per cent. The further small quantity of gypsum which is then deposited is separated by filtration, the resulting yellowish brown liquid representing commercial, crude, 50 per cent. (by weight) lactic acid. This should not contain more than 1-5 per cent, of ash, and should not contain sulphate or reduce Fehling's solution. The lactic acid prepared from the molasses of milk-sugar factories is more impure than, the above. According to a patent filed in 1905, lactic acid is also obtained from a mixture of bran and barley, and an English patent (No. 26,415, 1907) describes the preparation of pure, concentrated lactic acid by the distillation of the commercial 50 per cent, acid in a rapid current of air or of an indifferent gas. Very pure lactic acid is obtained by extracting the crude product with amyl alcohol which does not dissolve the impurities (sugar, gum, mineral substances) and distilling in vacua. The impurities are estimated by titrating the acid with normal caustic potash solution in presence of phenolphthalein. USES AND PRICE. Until a few years ago the uses of lactic acid were limited to the preparation of soluble lactates for medicinal purposes, but its manufacture has recently been considerably extended owing to its employment in the dyeing of wool, silk, &c., in place of tartaric acid, tartar and oxalic acid, for the reduction of the chromium com- pounds with which wool to be treated with fast dyes (alizarin dyes, &c.) is mordanted. For the same reasons it is advantageously employed in the chrome tanning of skins, its value in this case being sometimes regarded as due to its ability to keep calcium salts in solution and thus prevent the formation of certain harmful deposits. The crude 50 per cent, acid is most commonly sold, and it is necessary to ascertain whether by 50 per cent, is meant 50 kilos per 100 litres or per 100 kilos of solution ; in the former case the strength of the acid is only 43 per cent, by weight (i.e. 100 kilos contain 43 kilos of acid). Commercial, brown, 50 per cent, lactic acid costs about 64s. per quintal ; the paler, yellow product of the same strength, 105s. ; the pure (sp. gr. 1-21), 3s. Id. per kilo, and the chemically pure 12s. per kilo. Italy imported the following quantities of pure lactic acid at 104s. per quintal : 996 quintals in 1907 ; 650 in 1908 ; 520 in 1909 ; and 490 in 1910, when the amount exported was 51 quintals. The import duty in Italy is 12s. per quintal. Salts of Lactic Acid are generally soluble to some extent in water. Calcium lactate, (C 8 H 5 O 3 ) 2 Ca + 5H 2 O, forms mammillary aggregates of white needles soluble in 9-5 parts of cold water, and in all proportions in hot water ; it is insoluble in cold alcohol. The water of crystallisation is evolved in a vacuum desiccator or on heating to 100. At 250 it loses H 2 0, giving calcium dilactate, which is less soluble in alcohol than the original salt. Zinc" lactate and ferrous lactate crystallise with 3H 2 0, the latter in yellowish crystals ; both are used in medicine. ALANINE, CH 3 -CH(NH 2 ) -COOH, is obtained from the corresponding aldehyde- ammonia by the action of hydrocyanic acid. From the inactive, synthetical compound, the active stereoisomerides are separated by means of the strychnine or brucine salts. The action of PC1 5 expels both the hydroxyl- and the amino-groups, giving lactyl chloride, CH 3 CHC1 COC1, which gives a-chlorpropionic acid, CH 3 CHC1 COOH, when treated with water. (2) d-ETHYLIDENELACTIC ACID (Paralactic or Sarcolactic Acid) differs from ordinary lactic acid only in the greater solubility of its zinc salt ( + 2H 2 0), and the less solubility of its calcium salt ( + 4H 2 O). It is found in Liebig's extract of meat, and it is contained in the muscular juices, and is also formed in certain lactic fermentations. (3) Z-ETHYLIDENELACTIC ACID is formed during the fermentation of aqueous cane- sugar solutions by Bacillus acidi Icevolactici. (4) ETHYLENELACTIC ACID (Hydracrylic, /3-Hydroxypropionic or 3-Propanoloic Acid), OH -CH 2 -CH 2 -COOH, differs from its isomerides in that, when heated, it loses a mole- cule of water, giving, not the anhydride, but acrylic acid, CH 2 : CH-CO 2 H. Further, with oxidising agents it gives, not acetic acid, but oxalic acid aijd carbon dioxide. It con- tains no asymmetric carbon atom and is hence optically inactive. It can be prepared synthetically from (1) /3-iodopropionic acid, or (2) ethylene, CH 2 : CH 2 , by addition of hypochlorous acid, giving OH-CH 2 -CH 2 Cl,^which is then converted into the nitrile 326 ORGANIC CHEMISTRY OH CH 2 CH 2 CN, hydrolysis of the latter giving ethylcnelactic acid. The acid is a colourless, syrupy liquid and forms a calcium salt ( + 2H 2 O) and a readily soluble zinc salt ( + 4H 2 O). HYDROXYBUTYRIC ACIDS, OH-C 3 H 6 -C0 2 H Five isomerides are theoretically possible, four being known : two a-acids, one fi-acid, and one y-acid (prepared only as salts). a -HYDROXYBUTYRIC ACID, CH 3 -CH 2 -CH(OH) -CO 2 H, melts at 43 and is syn- thesised as the inactive, racemic form, which can be resolved into its active components by means of brucine (see p. 22). a-HYDROXYISOBUTYRIC ACID (Acetonic or 2-Methyl-2-propanoloic Acid), OH -C(CH 3 ) 2 -C0 2 H, melts at 79, boils at 212, and is obtainable by various synthetical methods from dimethylacetic acid, acetocyanohydrin, a-aminobutyric acid, &c. /3-HYDROXYBUTYRIC ACID, CH 3 -CH(OH) -CH 2 -CO 2 H, is obtained by oxidising aldol or reducing acetoacetic acid, these methods of formation indicating its constitution. It forms a syrup and its Isevo-isomeride is found in the blood and in diabetic urine. HIGHER HYDROXY-ACIDS H-HYDROXYVALERIC ACID, CH 3 -CH 2 -CH 2 -CH(OH) -C0 2 H, melts at 29. ei-HYDROXYISOVALERIC ACID, (CH 3 ) 2 CH-CH(OH) -CO 2 H, melts at 86. METHYLETHYLGLYCOLLIC ACID, ,?2 3 >C + H 2 - UUjl^HLg C0 2 + C 2 H 5 -OH + CH 3 -CO-CH 2 -CHR-C0 2 C 2 H 5 . v ft Properties of Ketonic Acids. While the a- and y-ketonic acids are stable, the /3-acids readily lose C0 2 , giving the corresponding ketones. Reduction of y-ketonic acids yields, not hydroxy-acids, but y-lactones. As we saw was the case with malonic acid (see p. 309), the esters of /3-ketonic acids contain a hydrogen atom readily replaceable by metals, e.g. ethyl sodioacetoacetate, CH 3 -CO-CHNa-C0 2 C 2 H 5 . Further, ketonic acids readily form condensation products : with aniline they give quinolines ; with phenylhydrazine, pyrazoles, &c. PYRUVIC ACID, CH 3 -CO-CO 2 H, is obtained by the dry distillation of tartaric or raccmic acid, an intermediate product in the reaction being possibly glyceric acid (formed by loss of CO 2 ), which then loses water and yields 332 ORGANIC CHEMISTRY pyruvic acid. It is formed also by oxidising lactic acid with permanganate or by hydrolysing acetyl cyanide. Pyruvic acid is a liquid with an odour of acetic acid and of meat-extract, and is soluble in water, alcohol, or ether. It boils at 165 and solidifies at 9. It is a more energetic acid than propionic owing to the presence of a carbonyl group in close proximity to the carboxyl. Its constitution is indicated by the fact that with nascent hydrogen it gives ethylidenelactic acid. It readily forms condensation products (e.g. benzenes ; and with ammonia, pyridine compounds). When heated at 150 with dilute sulphuric acid, it loses CO 2 , forming acetaldehyde. Electrolysis of a con- centrated solution of potassium pyruvate results in the union of the anion of the acid, CH 3 -CO-COO' with the ion OH', cyanogen and acetic acid being formed ; at the same time two anions combine with loss of 2C0 2 and formation of diacetyl, CH 3 -CO-CO-CH 3 . This represents the general behaviour of potassium salts of ketonic acids on -hydrolysis. Of the derivatives, cysteine (a-amino-/3-thiolactic acid), SH-CH 2 -CH(NH 2 )- C0 2 H, and cystine, the corresponding disulphide, may be mentioned. a-Ketobutyric acid, CH 3 -CH 2 -CO-C0 2 H, has no special importance. ACETOACETIC ACID (/3-ketobutyric acid), CH 3 -CO-CH 2 -CO 2 H, is obtained in the free state by cautious hydrolysis of its ester, and forms an extremely acid liquid soluble in water, in which it gives a red coloration with ferric chloride ; when heated, it readily loses C0 2 with production of acetone. ETHYL ACETO ACETATE, CH 3 -CO-CH 2 -C0 2 C 2 H 5 , is of far greater importance than the free acid, owing to the very varied syntheses in which it finds application. This ester is obtained in the form of its crystalline sodium derivative by the action of sodium and alcohol (sodium ethoxide) on ethyl acetate (see above). Ethyl acetoacetate (freed from sodium by means of acetic acid) is a liquid having a pleasant, fruity odour, and dissolves readily in alcohol or ether and slightly in water, in which it gives a red coloration with ferric chloride. It has a neutral reaction and a specific gravity of 1-030 ; it boils at 181 and is found in diabetic urine. When boiled with dilute alkali or dilute sulphuric acid, it undergoes ketonic decomposition, forming C0 2 , acetone and alcohol : CH 3 -CO-CH 2 -C0 2 C 2 H 5 + H 2 = C0 2 + C 2 H 5 -OH + CH 3 -CO-CH 3 . With concentrated alcoholic potassium hydroxide, it undergoes acid decomposition, producing 2 mols. of acetic acid : CH 3 -CO-CH 2 -C0 2 C 2 H 5 + 2H 2 = C 2 H 5 -OH + 2CH 3 -C0 2 H. Its great reactivity is due to the readiness with which one of the hydrogen atoms is replaceable by metals (Ba, Al, Zn, Ag, Cu, &c., in ammoniacal solu- tion), especially by sodium. Ethyl sodioacetoacetate, CH 3 CO CHNa C0 2 C 2 H 5 , is a white solid soluble in water, while ethyl acetoacetate is soluble in alkali, from which it is reprecipitated by acids. The sodium is readily replaced by different alkyl groups by the action of the corresponding alkyl iodides (see the analogous syntheses with Ethyl Malonate, p. 309). Since the compounds thus obtained can be subjected to either acid or ketonic decomposition, it will readily be seen how very varied acids and ketones can be obtained by means of ethyl acetoacetate. For instance, the action of normal octyl iodide on ethyl sodioacetoacetate yields methyl nonyl Jcetone, a constituent of oil of rue : CH- 3 CO-CHNa-C0 2 C 2 H 5 + I-CH 2 - [CH 2 ] G -CH 3 = Nal + CH-C KETO-ALCOHOLS AND DIKETONES 333 this compound, by ketonic decomposition, giving methyl nonyl ketone, which must hence have the normal structure CH 3 -CO-CH 2 - [CH 2 ] 7 -CH 3 . Further, by eliminating the sodium from ethyl sodioacetoacetate by means of iodine, the two residues combine, forming ethyl diacetylsuccinate : 2CH 3 CO CHNa C0 2 C 2 H 5 + I a = 2NaI + CH 3 CO CH CO 2 C 2 H 5 CH 3 -CO-CH-C0 2 C 2 H 5 and this ester, on ketonic decomposition (boiling with 20 per cent, potassium carbonate solution), reacts with 2H 2 and gives 2C0 2 , 2C 2 H 5 - OH and acetonyl- acetone, CH 3 -CO-CH 2 -CH 2 -CO-CH 3 , which has a normal carbon-atom chain 4321 and is a 1 : 4 diketone. Ethyl acetoacetate also combines with formaldehyde (in presence of diethylamine), with acetone and with ammonia ; with aniline it gives di- phenylcarbamide. With one or two mols. of sulphuryl chloride it gives ethyl chloracetoacetate (b.pt. 194, sp. gr. 1-19 at 14) or ethyl dichloracetoacetate, CH 3 -CO-CC1 2 -C0 2 C 2 H 5 , which boils at 206 and has the sp. gr. 1-293 at 16. LEVULINIC ACID, CH 3 -CO-CH 2 -CH 2 -C0 2 H, is obtained synthetically by the acid decomposition of the product of reaction of ethyl acetoacetate and ethyl chloracetate. It can be prepared by boiling hexoses, cane-sugar, cellulose, gum, starch, &c., with con- centrated hydrochloric acid. It melts at 33 and boils at 239 with slight decomposition, or at 144 under 12 mm. pressure. It is sometimes used in the printing of textiles. KETO-ALCOHOLS ACETONEALCOHOL or ACETYLCARBINOL (Propanolone) , CH 3 -CO-CH 2 -OH, is formed by heating either grape-sugar with fused potassium hydroxide or a mono-halo- genated acetone with barium carbonate. It is a liquid which boils almost unchanged at 147 and reduces Fehling's solution even in the cold. DIHYDROXY ACETONE, OH-CH 2 -CO-CH 2 -OH, is formed together with glycer- aldehyde by oxidising glycerol with nitric acid. It has a sweet taste and may indeed be regarded as a triose sugar. It crystallises in colourless plates and reduces Fehling's solution in the cold. BUTAN-2-OL-3-ONE (Dimethylacetol), CH 3 >CO-CH(OH) -CH 3 , also reduces Fehling's solution and is obtained by reducing diacetyl. It is a liquid soluble in water and boils at 142. The higher homologue, acetoisopropyl alcohol, CH 3 .CO.CH 2 .CH(OH).CH 3 , b.-pt. 177, is also known and is formed by the condensation (by means of alkali) of aldehyde with acetone. Removal of water from this compound yields Ethylideneacetone, CH 3 .CO.CH : CH CH 3 , boiling at 122. DIKETONES All diketones give mono- and di-oximes, and mono- and di-hydrazones, the latter (as with the aldehydes) bearing the name of osazones and being usually yellow. They often exhibit tautomerism and give rise to various cyclic condensation products. DIACETYL or a-DIKETOBUTANE (Butandione), CH 3 -CO -CO -CH 3 , is prepared by the general method for diketones, namely, by treating methyl ethyl ketone, CH 3 CO C 2 H 5 , with amyl nitrite and a little HC1, the CH 2 group being transformed into C : NOH, giving an isonitrosoketone, thus : CH 3 .CO.CH 2 .CH 3 + N0 2 .C 6 H U > CHg-CO-C-CHg N-OH; when boiled with dilute sulphuric acid, this compound loses the hydroxyiminic group (as hydroxylamine), the diketone remaining. 334 ORGANIC. CHEMISTRY It is a yellow liquid which has a penetrating odour, dissolves in water and boils at 88, giving yellowish green vapour (sp. gr. 0-973 at 20). With hydrogen peroxide, diacetyl is converted quantitatively into 2 mol.s. of acetic acid : CH 3 .CO-CO.CH 3 + H 2 O 2 = 2CH 3 .C0 2 H. ACETYLACETONE, CH 3 CO CH 2 CO CH 3 . The best general method for preparing 1 : 3-dikctones consists in treating an ester with sodium ethoxide : /ONa R.CO 2 C 2 H 6 + C 2 H 5 .ONa = R-C^OC 2 H 5 ; X OC 2 H 6 this compound, when treated with a ketone, R'-CO-CH 3 -, loses 2 mols. of alcohol and yields /ONa R-C^ , from which the sodium is expelled by a dilute acid. This enolic form, XXH-COR' OH C^ , readily passes into the. ketonic form, CO CH 2 , thus giving the ^CH- compound R CO CH 2 . CO R'. Another general method for obtaining 1 : 3-diketones consists in treating the sodium derivatives of acetylene homologues with an acid chloride and then acting on the acetylenic product with sulphuric acid, so that it combines with water : CH 3 .[CH 2 ] 4 -C : . Na + CH 3 .CO-C1 = Nad + CH 3 - [CH 2 ] 4 -C : C-CO.CH 3 ; Amylacetylene Acetyl chloride the latter + H 2 O > CH 3 . [CH 2 ] 4 .CO.CH 2 .CO.CH 3 . As in ethyl acetoacetate and ethyl malonate, the two hydrogen atoms of the methylene group between the two carbonyl groups are here also replaceable by metals, giving volatile compounds which are soluble in chloroform, benzene, &c., and differ from true salts, their solutions exhibiting very slight electrical conductivity. Acetylacetone has a pleasant odour and boils at 137. When boiled with water it yields acetone and acetic acid. ACETONYLACETONE (y-Diketohexane or Hexa-2 : 5-dione), CH 3 -CO.CH 2 .CH 2 - CO'CH 3 , is obtained by ketonic decomposition of the product of interaction of ethyl aceto- acetate and ethyl chloracetate (see also Levulinic Acid) ; it boils at 194 and has an agreeable smell. KETO-ALDEHYDES AND HYDROXYMETHYLENEKETONES PYRUVIC ALDEHYDE (Methylglyoxal or Propanolone), CH 3 -CO -CHO, is a volatile oil obtained by decomposing its oxime (isonitrosoacetone) with dilute acid (see Diacetyl). ACETOACETALDEHYDE, CH 3 .CO-CH 2 -CHO, was formerly thought to have been obtained in the free state, but more exact study has now shown the compound in question to be the unsaturated isomeride, HYDROXYMETHYLENEACETONE, CH 3 -CO-CH : CH-OH, which has an acid character and is obtained by the interaction of acetone and ethyl formate in presence of sodium ethoxide (see Ethyl Acetoacetate). It boils at 100 and readily condenses into 1:3: 5-triacetylbenzene, C 6 H 3 (CO-CH 3 ) 3 . F. POLYVALENT DIBASIC HYDROXY-ACIDS AND THEIR DERIVATIVES TARTRONIC ACID (Hydroxymalonic or Propanoldioic Acid), C0 2 H-CH(OH)-CO 2 H, is formed by the spontaneous decomposition of nitro- tartaric acid and is obtained synthetically by oxidising glycerol with potassium permanganate, by eliminating bromine from bromomalonic acid by the action of moist silver oxide, or by reducing Mesoxalic Acid, CO(C0 2 H) 2 . It crystal- lises with 2 H 2 O and melts at 184, losing C0 2 and forming polyglycollides. It is soluble in water, alcohol, or ether. TARTARICACIDS 335 MALIC ACID (Hydroxysuccinic or Butanoldioic Acid), CO 2 H-CH(OH)- CH 2 -CO 2 H, occurs in abundance in unripe fruits (apples, grapes, quinces, and sorb-apples, from which it is extracted). Its crystals melt at 100 and it dissolves in water or alcohol and, to a slight extent, in ether. When subjected to dry distillation, it gives fumaric acid and maleic anhydride. Synthetically it is obtained from maleic or fumaric acid or asparagine, and also by the action of moist silver oxide on bromosuccinic acid, and by the reduction of tartaric acid by means of hydriodic acid. As it contains an asymmetric carbon atom, malic acid forms three optically different stereoisomerides, all of which are known. Natural malic acid is Isevo- rotatory, that derived from (^-tartaric acid dextro-rotatory, and that obtained by other syntheses inactive but resolvable into active components by fractional crystallisation of the cinchonine salt. It gives an acid calcium salt readily soluble, and a normal salt slightly soluble in water. The presence of the alcoholic group is proved by the forma- tion of acetylmalic acid (see p. 189). For the amido-derivatives, asparagine, &c., see later. Of the higher homologues of malic acid, the following are known : Four isomeric acids, C 3 H 5 (OH)(C0 2 H) 2 (a- and fi-hydroxyglutaric acids, itamalic and citramalic acids) ; diaterebinic acid, C 5 H 9 (OH)(C0 2 H) 2 , which readily forms a lactone, and terebinic acid~C 7 H 10 4 . TARTARIC ACIDS, CO 2 H-CH(OH)-CH(OH)-C0 2 H These are dibasic and tetravalent, as they contain two secondary alcoholic groups. The presence of two asymmetric carbon atoms leads to the existence of four stereoisomerides, which have already been considered on p. 20 : (1) ordinary or d-tartaric acid ; (2) Z-tartaric acid ; (3) racemic or para- or di-tartaric acid ; (4) i- or meso- or anti-tartaric acid. They are obtained synthetically from dibromosuccinic acid, C0 2 H-CHBr-CHBr-C0 2 H, and moist silver oxide, from glyoxal cyano- hydrin, from glyoxylic acid by reduction, from mannitol by oxidation with nitric, acid and from fumaric or maleic acid by oxidation. (1) d-TARTARIC ACID. This is the ordinary tartaric acid, which occurs abundantly as such, and as monopotassium tartrate (tartar) in many fruits especially in the grape, and hence in wine, from which it is extracted in a manner to be described. Dextro-rotatory tartaric acid forms hemimorphic, monoclinic prisms with a decided and pleasant acid taste. It is readily soluble in water or alcohol and almost insoluble in ether. One hundred parts of water dissolve 114 parts of the acid at 0, 125-7 at 10, 139-4 at 20, 156-2 at 30, 176 at 40, 195 at 50, 217-5 at 60, 243-6 at 70, 273-3 at 80, 306-5 at 90, and 343-3 at 100. The acid melts at 170, giving rise to various anhydrides and to pyruvic and pyrotartaric acids ; ultimately it carbonises with an odour of burnt bread or, if the temperature is raised considerably, of burnt sugar. Energetic oxidising agents convert it into tartronic acid or dihydroxy- tartaric acid and finally into formic acid, carbon dioxide, &c. In the hot, it reduces ammoniacal silver solutions (see p. 346 for a sensitive reaction for tartaric acid). Certain bacteria transform it into succinic acid. When burned, tartaric acid and tartrates emit an odour of burnt bread, thus differing from citric acid and citrates, which give a pungent odour. Owing to the presence of alcoholic groups, tartaric acid, like glycerol, hinders the precipitation by alkali of many metallic oxides, e.g. of cupric oxide in Fehling's solution (containing caustic soda, copper sulphate, and sodium potassium tartrate ; see Sugar Analysis), the intensely blue, soluble 336 ORGANIC CHEMISTRY C0 2 Na-CH-(X compound, /Cu, being formed ; this compound is not pre- CO a K CH-Cr cipitable by alkalis, since the copper no longer functions as cation, but is 0-CO-CH-Ov contained in the anion, yCu, which migrates to the positive o-co-CH-cr pole or anode when the salt is electrolysed. Tartaric acid is used in dyeing, in the wine industry, in the preparation of aerated beverages (lemonade), in medicine, &c. The following salts of tartaric acid may be mentioned, acid potassium tartrate being considered more in detail later. ACID POTASSIUM TARTRATE (Cream of Tartar), C0 2 H-CH(OH).CH(OH). CO 2 K, is slightly soluble in water or in dilute alcohol, and has a pleasant acid taste. For its commercial preparation, see Tartar Industry. NORMAL POTASSIUM TARTRATE, C 4 H 4 6 K 2 + *H 2 O, is readily soluble in water and separates from highly concentrated solutions in monoclinic prisms. SODIUM POTASSIUM TARTRATE (Rochelle Salt), C 4 H 4 O 6 NaK + 4H 2 0, is pre- pared by saturating cream of tartar solution with sodium carbonate. Copper and iron, present as impurities, are removed by means of hydrogen sulphide, the solution being then heated with good animal charcoal, filtered, concentrated and allowed to crystallise ; the Rochelle salt separates in thick columns readily soluble in water and slightly so in alcohol. It is used to reduce silver salts in the silvering of mirrors and also for medicinal purposes and to prepare Fehling's solution. It costs about 7 per quintal. CALCIUM TARTRATE, C 4 H 4 O 6 Ca + 4H 2 O, is insoluble in water but soluble in cold sodium hydroxide solution, from which it separates on heating as a jelly, which redisf-olves on cooling. It dissolves in acetic acid, thus differing from calcium oxalate. TARTAR EMETIC (or Potassium Antimony 1 Tartrate), C 4 H 4 O 6 (SbO)K + AH 2 O, is prepared by precipitating SbOCl from a solution of SbCl 3 by means of, water, boiling the precipitate with soda solution and dissolving the Sb 2 O 3 thus formed in a solution of 4 or 5 times its weight of potassium hydrogen tartrate in 50 parts of water. After nitration and concentration the solution deposits, on cooling, efflorescent, trimetric pyramids, which are soluble in water (1 : 13 at 20, 1 : 6 at 50) but insoluble in alcohol. It is poisonous and is used in medicine as an emetic and in dyeing cotton as a mordant for basic dyes (price about 9 10s. per quintal). Germany imported 2019 quintals in 1908 and 3914 in 1909, the respective exports being 10,303 and 10,899 quintals in the two years. (2) Z-TARTARIC ACID differs from the d-acid only in the opposite sign of its rotation and in the opposed hemihedry of its crystals. Mixing of the concentrated aqueous solu- tions of the two acids results in development of heat and the formation of inactive tartaric acid. (3) RACEMIC ACID (ParatartaricAcid),(C 4 H 6 6 ) 2 + 2H 2 O, represents a mixture of dextro- and laevo -tartaric acids in equal proportions, and is hence optically inactive (see p. 20). When heated alone or, better,, in presence of concentrated caustic soda solution, either the rf-acid or the meso-acid (see below) is transformed into racemic acid. The latter is obtained from the mother -liquors of ordinary d-tartaric acid. The molecular weight, determined cryoscopically or from the vapour densities of the esters, corresponds with the simple molecule, C4H 6 6 . It forms triclinic crystals which effloresce in the air, and is less soluble than the active acids. From sodium ammonium racemate crystals (C4H 4 6 ) 2 Na 2 (NH 4 ) 2 + 2H 2 O, Pasteur separated those showing dextro- from those showing laevo -hemihedry, thus resolving racemic acid into its optically active constituents. Only in the crystalline state is the molecule of racemio acid regarded as double that of tartaric acid, whilst in dilute aqueous solution it is assumed to be decomposed completely into the two optical antipodes. (4) MESOTARTARIC ACID, C 4 H 6 6 + H 2 O, is optically inactive and is not merely . a mixture of the active compounds. It is obtained by prolonged boiling of d-tartaric acid with excess of caustic soda. Its potassium salt is more soluble in water than those of the other tartaric acids, TARTAR INDUSTRY 337 THE TARTAR INDUSTRY MANUFACTURE OF POTASSIUM BITARTRATE (Cream of Tartar or Potassium Hydrogen Tartrate). Although the crude prime material of this industry is very abundant in Italy in wine residues, it is only within the last few years that the working has been placed on a rational basis, the tartar being refined and tartaric acid prepared. Although these prime materials are subject to an export duty (Is. 9d. per quintal), the exportation from Italy amounted to about 178,000 quintals, worth 480,000, in 1905, and 178,500 quintals, worth 416,000, in 1910. The treatment of these products requires, besides special tech- nical ability, also considerable quantities of fuel, and to this is partly due the difficulties of the Italian manufacturers. Cream of tartar occurs abundantly in the green extremities of vine -shoots and in the grape, and part of it remains in the pressed vinasse. The vinasse of southern grapes con- tains as much as 4 per cent, of cream of tartar, and that of other grapes from 2 to 2-5 per cent. Vinasse that has not been in contact with the fermenting must, and that of second wines have practically no commercial value. 1 Another portion of the cream of tartar which remains dissolved in the must gradually separates (lees) as fermentation proceeds cream of tartar being less soluble in alcoholic liquids (wine) and finally part of it is deposited as a crystalline crust on the walls of the casks during the winter, the solubility being less in the cold. One litre of water dissolves 3-2 grms. of tartar at 0, 4 grms. at 10, 5-7 grms. at 20, 9 grms. at 30, 13 grms. at 40, 18 grms. at 50, 24 grms. at 60, 32 grms. at 70, 45 grms. at 80, 57 grms. at 90, 69 grms. at 100, 82 grms. at 110, and 94 grms. at 120. In an alcoholic liquid (with 10 per cent, of alcohol) the solubility is reduced almost to one-half ; the solubility is also slightly diminished by the tartaric acid and increased by the mineral acids of the wine. These crude products are of different colours according as they are obtained from white wines (white tartar) or red wines (red tartar), and according to the degree of purity. Fresh wine lees (forming about 5 per cent, of the wine) are slimy, of a dirty red colour, and contain yeasts, colouring-matters, cream of tartar (10 to 25 per cent.), and calcium tartrate (6 to 20 per cent.). Italian wine lees are the richest in potassium bitartrate and the poorest in calcium tartrate. When removed from the vats, the lees are placed to drain in strong bags suspended by cords, the bags being afterwards tied up and pressed slightly in a press. They are then removed from the bags and dried in the air, being turned from time to time. When pressed and almost dry, they contain more than double as much tartar as when in the fresh state (about 10 per cent, of moisture, 6 to 10 per cent, of lime, 3 to 5 per cent, of sand, 25 to 40 per cent, of tartaric acid). In some large wineries the lees are passed directly to the filter-presses, cakes which are readily dried being thus obtained. The crude tartar contains 45 to 70 per cent, of potassium bitartrate and calcium tartrate, and, if washed and crystallised once from hot water, this content may increase to 75 to 87 per cent., 2 the product being then placed on the market under the name of crystals. 1 One quintal of grapes yields 30 to 35 kilos of vinasse and 65 to 70 of must, so that the annual Italian produc- tion of 40,000,000 quintals would correspond with 20 to 25 million quintals of vinasse, containing, on the average more than 3 per cent, of tartar. The tartar is estimated by the method of Carles : a kilo of the vinasse is chopped and mixed, and 100 grms. weighed and boiled for 10 minutes with 700 c.c. of water in a litre flask, the liquid being subsequently made up to the mark with distilled water. Five hundred cubic centimetres of the filtered solution are concentrated to about 100 c.c., 70 c.c. of saturated calcium acetate solution being then added to the boiling liquid ; after mixing, the liquid is allowed to cool for 12 hours, the precipitated calcium tartrate being then collected on a tared filter, washed with water, dried at 60, and weighed. Multiplication of the result by 2 and deduction of 5 per cent, (to allow for the volume occupied by the vinasse in the litre flask) gives the calcium tartrate per 100 grms. of vinasse ; further multiplication by 0-723 gives the corresponding amount of potassium hydrogen tartrate. Ciapetti has rendered this method more exact by transforming the calcium tartrate (by potassium bioxalate in the hot) into potassium hydrogen tartrate, filtering and washing the residue, concentrating the filtrate and adding alcohol to precipitate the potassium bitartrate ; the latter is washed, redissolved in hot water and titrated with decinormal caustic soda solution (see below, Analysis of Tartar). 2 Analysis of Tartar. Tartar being a rather expensive substance (6 to 8 per quintal), it is frequently adulterated with sand, gypsum, Ac. It is always bought and sold on its strength, the potassium bitartrate or the total tartaric acid (thus including both the calcium tartrate and the free tartaric acid) being determined. A homogeneous sample is finely ground and sieved, the residue being again ground. A test which is not very exact but is rapid and largely used is the direct titration test. The coarse impurities, sand, clay, sulphur, woody matter, yeasts, &c., arc estimated by boiling a known weight of the crude tartrate with water acidified with HC1, and collecting, washing, drying, and weighing the residue on a tared filter; the residue is sometimes ashed. The calcium carbonate is determined by treating with an acid in the calcimeter and measuring the carbon dioxide evolved, and the total lime, including that of the tartrate, by calcining a given weight of the tartar, II 22 338 ORGANIC CHEMISTRY As a rule 60 to 70 per cent, of the total tartar is extracted from the vinasse after the alcohol has been distilled off with steam in the manner indicated on p. 143. The forms of apparatus there shown give almost saturated, boiling solutions of tartar (the remaining vinasse being centrifuged or pressed to remove all the tartaric liquors), which are allowed to cool in shallow, wooden vessels. In these vessels are hung, after some time, strings studded with tartar crystals, on which less impure crystals gradually form. The deposit forming on the walls of- the vessels is of a less degree of purity, and that on the bottom contains many coloured impurities. In 5 to 6 days the crystallisation is complete, more rapid and complete separation being attained in very cold places or by the use of artificial cooling. The mother -liquors decanted from the tartar may be utilised again for extraction of vinasse, but when they become too rich in impurities or mucilaginous substances, they are either used as fertilisers, since they contain potassium salts, or, better, are treated (Carles, 1910) at boiling temperature with 60 grms. of potassium ferrocyanide per hectolitre, the iron, alumina, copper, &c., present being thus removed ; the clarified liquid is treated with lime to separate calcium tartrate, the potassium salts being recoverable from the filtered solution by the Alberti process (see later). 1 The dregs deposited during the extrac- dissolving out the potassium carbonate, treating- the residual calcium carbonate with excess of standard nitrous acid solution, and measuring the excess of the acid by titration with soda solution. The total lime is, however, best estimated by dissolving 2 grms. in HC1, neutralising with ammonia, precipitating with ammonium oxalatc, and heating on a water-bath, the precipitate being subsequently collected on a filter, ignited in a platinum crucible and weighed as CaO. The titration test of the quantity of acid potassium tartrate is carried out by dissolving a weighed amount (2 to 3 grms.) of the tartar in water and titrating the boiling solution with N/4-sodium hydroxide solution, using very sensitive litmus paper as indicator ; 1 c.c. of N/4-caustic soda solution corresponds with 0-047 grm. of potassium bitartrate. Multiplication of the amount of bitartrate by 0-798 yields the corresponding amount of tartaric acid. For international trade, the potassium bitartrate is nowadays estimated by the filtration process, which largely excludes errors due to the presence of tannin substances and other impurities which also react with litmus : 2-35 grms. of the substance (crude tartar, sludge or lees) are heated to boiling for 5 minutes with 400 c.c. of water in a 500 c.c. flask ; water is again added and the whole cooled, made up to 500 c.c., mixed and filtered through a folded filter. Of the filtrate 250 c.c. are heated to boiling and titrated with N/4-potassium hydroxide solution (standardised with pure bitartrate), sensitive litmus paper being used as indicator. For more exact determinations, Rammer's recrystallisation method is employed : 4-7025 grms. of substance (molecular weight of tartar divided by 40) are heated to boiling with 30 to 40 c.c. of water in a 100 c.c. flask, the solution being then neutralised with N/4-caustic soda, of which 1 to 2 c.c. in excess are added. The volume is made up to 100 c.c., and of the filtered solution, 20 c.c. are rendered decidedly acid with acetic acid and treated with 100 c.c. of a mixture of alcohol and ether, which separates the potassium bitartrate completely in crystals. These are collected on a filter, washed with alcohol and ether, redissolved in boiling water, and titrated with N/20-soda solution. Determination of the total tartaric acid. This gives the total content of potassium bitartarate, calcium tartrate, and free tartaric acid. The Goldenberg-Geromont hydrochloric acid process, which was formulated as follows at the Congress of Applied Chemistry at Turin in 1902, is generally used : Six grammes of the substance are treated for 8 to 10 minutes in a small beaker with 9 c.c. (for products poor in tartar) or 13 c.c. (for richer products) of cold HC1 (sp. gr. 1-10), the whole being then washed into a 100 c.c. flask, made up to volume, mixed, and passed through a dry pleated filter. Fifty cubic centimetres of the filtrate are boiled for 10 to 15 minutes in a tall 250 to 300 c.c. beaker with 5 (or 10) c.c. of concentrated potassium carbonate solution (66 grms. in 100 c.c. of water), the liquid being then washed into a 100 c.c. flask, cooled, made up to volume, and filtered through a pleated filter. Fifty cubic centimetres of this filtrate are takdfc to dryncss in a half-litre dish, redissolved in 5 c.c. of boiling water, and vigorously stirred with 4 to 5 c.c. of glacial acetic acid ; when the liquid is cold, 100 to 110 c.c. of 96 per cent, alcohol are mixed in, and the potassium bitartrate allowed to deposit. This is filtered under pressure, the dish, rod, and filter being repeatedly washed with alcohol. The filter and the bitartrate are then washed into the same dish with boiling water, with which the volume is made up to about 300 c.c. The liquid is then boiled and titrated with N/4-alkali, the end-point being determined with sensitive litmus paper. To eliminate the error due to the volume occupied by the impurities in the original flask, 0-7 per cent, is deducted from the total content in the case of low-grade tartars (containing less than 20 per cent, of bitartrate) and 0-7 (n x 0-02) in the case of those containing 20 to 50 per cent., n being the excess percentage over 20 ; beyond 50 per cent, the correction becomes almost zero. 1 In many places the vinasse is placed in vats fitted with false bottoms, steam being passed in below while a spray of mother-liquor (red liquors) falls from above ; as many hectolitres of these liquors are employed as there are quintals of vinasse. In this manner, the first solution of the tartar is obtained almost boiling and almost saturated ; it is purified by percolating slowly through the vinasse (the steam expels the air and condenses). A second extrac- tion gives a less completely saturated solution, which is utilised for further extractions. Finally, the vinasse is not pressed but is washed with water and a little hydrochloric acid to extract the calcium tartrate, exhaustion being then obtained by means of cold water (Tarulli's method). The mother-liquors from the crystallisation, when they become too impure or, in some cases, even after the first crystallisation, are treated with milk of lime to pre- cipitate all the dissolved tartar, while the liquors from the second and third extractions are used for fresh quantities of vinasse. To obtain purer solutions from the vinasse, to avoid losses occasioned by the presence of lime in the water, and to extract calcium tartrate at the same time, Ciapetti exhausts the vinasse with dilute solutions of sulphurous and hydrosulphurous'acid which do not dissolve the colouring-matters or the pectic or albuminoid substances and allows refined, white cream of tartar (!)to crystallise out; the mother-liquors are utilised further. This method was tried in various works, but did not give the results expected of it. Extraction on the Ciapetti system may also be carried out in the series of stills used for the distillation of the vinasse (see later, Tartaric Acid, Gladysz Process). The fresh vinasse or the deposits, if not to be worked up at once, may be kept for socie months if tartaric EXTRACTION OF TARTAR 339 tion of the cream of tartar from the vinasse contain, in the dry state, 30 to 60 per cent, of potassium bitartrate and 10 to 20 per cent, of calcium tartate. The refining of crude tartar from vinasse, lees, &c., is very difficult with the poorer products, and in practice rich and poor materials are often mixed so as to give a content of 60 to 65 per cent, when a highly refined tartar is not required. The crude tartar or other mixture should first be ground and then sieved to remove pieces of wood and other impurities. In order to destroy certain impurities and protein substances and hence hasten the filtration of the subsequent aqueous solutions, the lees or low-quality tartar are heated in revolving iron cylinders until the temperature reaches 160 to 180, the loss in weight being 8 to 12 per cent. It is then introduced into a per- forated copper cylinder placed almost on the bottom of a large wooden vessel furnished with copper or aluminium steam coils. Care is taken not to use an excess of water, which would involve waste of fuel, and generally 6 to 8 parts of water are taken per 1 part of tartar. A hood is usually fitted over the vessel to carry off the steam from the boiling solution. In order to transform the calcium tartrate present into potassium bitartrate, 3 kilos of hydrochloric acid (20 Be.), and 3 kilos of potassium sulphate (previously dis- solved in 20 litres of water) are added to the water per kilo of lime (CaO) contained in the calcium tartrate (see method of determination given above). The liquid is stirred occa- sionally, boiled for an hour and allowed to settle for about a further hour, after which either the bitartrate solution is decanted by means of a tap a short distance from the bottom, or the whole mass is passed through a filter-press. The liquid is left to cool in a cold place in ordinary crystallising vessels, or, better, in copper or aluminium basins, a somewhat impure and reddish brown tartar crystallising out. Rather purer tartar may be obtained by col- lecting the crystals separating while the solution is cooling to 35 to 40, small crystals being ensured by occasional stirring ; the tepid mother -liquors are then decanted and crystallised in a cold place. The brown mother -liquors are used repeatedly to dissolve fresh quantities of the crude tartar substances in the hot. The brown crystals are detached from the crystallising vessels by means of suitable spatulas and redissolved, in a wooden vessel similar to that previously used (with a perforated bottom for the crystals), in 10 to 12 times their weight of water, which is boiled by indirect steam supplied through copper or aluminium coils. Decoloration is effected by adding, after half an hour's boiling, about 1 per cent, of animal charcoal (well washed with hydrochloric acid and thoroughly rinsed ; for very impure tartar from lees as much as 6 to 8 per cent, of animal charcoal is used.) After mixing and an hour's boiling, about 1 per cent, of kaolin free from chalk (washed with HC1) is well mixed in in the hot and the liquid either filter-pressed or left for 2 to 3 hours so that the kaolin may carry down all the suspended charcoal. In some cases the solutions are clarified by adding tannin and gelatine (50 grms. of the latter and 250 of tannin dissolved separately) after the kaolin and before filtration. The pale yellow, clear solution is decanted (the first and last portions, which are rather turbid, being kept separate) and crystallised in wooden crystallising vessels or in copper or aluminium basins ; in three or four days the crystallisation is complete. The mother -liquors are used to dissolve fresh brown crystals, since they contain a little free tartaric acid which dissolves many of the impurities better than water does, and so results in the separation of purer crystals. After the removal of the mother-liquor, the crystals are washed repeatedly with very pure water (condensed), a trace of hydrochloric acid being added to the first washing water if the crystals show any superficial turbidity. They are finally dried on cloths in a desiccator at 60 with the help of a current of air. In France and also in certain Italian factories, Carles has applied a method of extracting the tartar which consists in treating 100 parts of the crude, powdered tartar with 400 parts of water at 70, containing sufficient sodium carbonate to transform all the potassium bitartrate into the highly soluble sodium potassium tartrate (1 : 1-2). This solution is decanted or filtered and treated with more than the amount of hydrochloric or sulphuric acid necessary to neutralise the soda added ; this leads again to the formation of the slightly soluble potassium bitartrate, which crystallises out (96 per cent, purity) on cooling. Vinasse is also extracted more readily by this alkaline method. fermentation which would destroy part of the cream of tartar is prevented, either by addition of 0-05 per cent, of sodium thiosulphate or by maintaining the vinasse strongly compressed, under chalk and sand, in wooden Tats (tartaric fermentation is caused principally by Bacillus saprot/enes vini). The yields of tartar and alcohol are determined oil a small^uaiitity (5 kilos) of the vinassc in a small Savalle distilling and macerating apparatus. 340 ORGANIC CHEMISTRY The recent process of Can toni-Chau terns and Degrange (1910) refines the tartar almost in the cold and effects a marked economy of chemical products. This method serves well also for poor products (20 per cent, lees), which are heated as described above and washed first with cold sodium carbonate solution in a series of vessels, next slightly with water, and finally, slowly and systematically with dilute hydrochloric acid. In the first case, sufficient soda is added to convert the tartar almost completely into sodium potassium tartrate, which can be extracted with the minimum quantity of water. The treatment with hydrochloric acid dissolves the remainder of the tartar and also the calcium tartrate. The amount of the acid used is chosen so that, when the alkaline and acid solutions are united, it neutralises the soda first used. The hydrochloric acid solution is mixed previously with the amount of oxalic acid required to precipitate the whole of the lime present in the crude tartar as calcium tartrate, while sufficient potassium chloride is added to the alkali solution to transform all the tartaric acid, existing as calcium tartrate, into potassium bitartrate. Mixing of the two solutions in the cold results in the precipita- tion of almost all the cream of tartar in a white, highly pure state, and of the calcium oxalate. After filtering, the dark mother -liquors are kept for subsequent operations, while the solid residue is treated with the calculated (on the solubility of the tartar) quantity of water at 90, a little oxalic acid being added to render the calcium oxalate less soluble. The mass is then filtered or centrifuged (if necessary, decolorised with a small amount of animal charcoal), the clear liquid, on cooling, depositing refined tartar of a purity of 99 to 99-5 per cent. The mother -liquor is used to initiate the solution of further quantities of crude tartar, &c. Oxalic acid may be recovered from the calcium oxalate. A process which allows of the conversion of calcium tartrate into potassium bitartrate consists in boiling, say, 100 kilos of the calcium tartrate (85 per cent.) with 1500 litres of water and 53-5 kilos of potassium bisulphate (or a mixture of 35 kilos of the neutral sulphate and 24-6 kilos of sulphuric acid at 60 Be.) for half an hour, decanting and filtering the liquid, from which pure potassium bitartrate (up to 98 per cent. ) crystallises on cooling. This process is a modification of that of Martignier (Fr. Pat. Nov. 23, 1889), who transforms calcium tartrate into normal potassium tartrate by decomposition with normal potassium sulphate ; after filtration and concentration and addition of the calculated amount of sulphuric acid, the solution deposits potassium tartrate. If the cream of tartar crystals are not pure but contain small proportions of calcium tartrate, they do not give perfectly clear solutions in water. When the pale mother -liquors from the final refining are somewhat impure, they are used in place of the brown liquor to dissolve the crude material, the highly impure brown liquors being used in the manufacture of tartaric acid or added in small amounts to the prime tartaric materials. STATISTICS AND USES. The exports of crude tartar substances (vat deposits, wine lees, &c.) from Italy are as follow : in 1887, 150,000 quintals, worth 1,000,000 ; in 1896, 147,566, and in 1903, 164,000 quintals. The importation of crude tartar and lees amounted to 1424 quintals in 1887, 3975 quintals in 1896, 2356 quintals in 1902, and 3356 quintals in 1903. The total Italian production of tartar, &c., in 1905 has been estimated at 1,600,000, but this is probably in excess of the truth, since the world's production was valued at about 2,800,000. Crude tartar and lees pay an export duty from Italy of Is. 9d. per quintal, no import duty being levied ; refined tartar pays an import duty of 3*. 2d. Italy con- tains about 200 crude cream of tartar works, but only very few manufacturing refined tartar. In 1905 Germany imported, for refining, 26,000 quintals of crude tartar, worth 129,600 ; 26,914 quintals in 1908, and 20,263 quintals in 1909, the exports being 12,250 and 11,535 quintals respectively. In 1890, more than 51,000 quintals were imported, and in 1896 about 57,000 quintals, the corresponding exportation being about 5000 quintals of the refined product. France produces annually 61 ,000 quintals of refined cream of tartar. The imports of tartaric acid into England were 1850 tons in 1909 and 2250 tons (203,365) in 1910, when the exports were valued at 31,300 ; in 1911the imports. amounted to 200,300, and the exports to 38,240. The imports of cream of tartar were 3550 tons in 1909, 4100 tons (307,000) in 1910, and 301,940 in 1911. Crude cream of tartar is bought and sold at ll^d. to 14|d. per unit of strength, and the refined at Is. Id. or more. Cream of tartar is largely used in dyeing, in the bichromate mordanting of fast MANUFACTURE OFTARTARIC ACID 341 wool dyes, &c., and in the printing of textiles. Considerable quantities are used in the United States for preparing powder which is added to dough to render the bread light and elastic ; this powder contains 69 per cent, of tartar and 31 per cent, of sodium bicarbonate. MANUFACTURE OF TARTARIC ACID. This acid is prepared by decomposing its salts (cream of tartar, lees, calcium tartrate, &c.), the dark mother-liquors and the deposits and sludges of cream of tartar factories being most commonly employed. Attempts were at one time made to liberate the acid from its soluble salts by treating these, in hot solutions, with hydrofluosilicic acid, which separates as insoluble potassium fluosilicate, the solution of tartaric acid remaining being filtered, concentrated and crys- tallised. The potassium fluosilicate is treated with calcium carbonate to convert it into insoluble calcium fluosilicate, which yields hydrofluosilicic acid under the action of an energetic acid. Tartaric acid has also been prepared by treating solutions of its salts with barium carbonate and then with barium chloride, the latter precipitating the neutral potassium tartrate partly formed by the former reagent. At the present time, however, the method most commonly used consists in treating boiling cream of tartar solutions with milk of lime or powdered calcium carbonate. In this way half of the tartar is converted into insoluble calcium tartrate and the other half into the soluble normal potassium tartrate, the latter being then separated as the insoluble calcium salt by addition of calcium sulphate or chloride : 2C 4 H 5 6 K + CaC0 3 = H 2 + CO 2 + C 4 H 4 6 Ca + C 4 H 4 6 K 2 ; C 4 H 4 O 6 K 2 + CaSO 4 = K 2 S0 4 + C 4 H 4 6 Ca. The acid is liberated from calcium tartrate by means of sulphuric acid : C 4 H 4 6 Ca + H 2 SO 4 = CaSO 4 + C 4 H 6 6 . According to the nature of the materials employed, various procedures are adopted. Crude tartar is freed from coarser impurities by passing through a wide -meshed sieve, heated if necessary (see above), ground to a fine granular condition and placed in a wooden vessel, where it is treated with 8 to 10 times its weight of boiling water. The mass is well mixed with a stirrer and heated to boiling by a steam jet, most of the tartar then dissolving. A paste of calcium hydroxide (sieved) is then added until a small portion of the liquid gives only slight effervescence with calcium carbonate (as a rule, 160 grms. of quick- lime, made into a 10 per cent, paste, are required for each kilo of potassium tartrate), the mixture being then boiled for 15 minutes. In this way calcium tartrate is precipitated, while normal potassium tartrate remains in solution. The latter is then precipitated as calcium salt by boiling with a slight excess of gypsum or of calcium chloride solution (300 grms. of the chloride per kilo of tartrate used) for about two hours ; a little precipitated calcium carbonate is finally added to precipitate the neutral tartrate as completely as possible. In some cases, however, the liquid is left slightly acid to prevent separation of iron and aluminium salts. It is, indeed, necessary to ensure the absence from the various reagents (lime, chalk, gypsum, &c.) of iron, aluminium, and especially magnesium, the latter forming magnesium tartrate, which is ultimately found as magnesium sulphate in the tarfaric acid after the calcium tartrate has been treated with sulphuric acid. The boiling liquid is kept mixed and is afterwards either allowed to cool to 40 and decanted or passed immediately to the filter-press ; after repeated washings with hot and cold water, the crude, dry tartrate is placed on the market and the filtrate either rejected or evaporated to obtain the potassium chloride present. But where the calcium tartrate is converted into tartaric acid, it is not necessary to filter and dry it, the tartrate, after the first decantation, being mixed with various separate amounts of water, which are drawn off when the precipitate settles. The calcium tartrate remaining can then be treated in the same vessel with sulphuric acid as described above. The procedure is rather more complicated in the case of wine lees (sediments, sludge, &c.), as the tartrate cannot be extracted merely by treatment with water and filtration, owing to the presence of considerable amounts of mucilaginous protein substances (ferments), which impede filtration. The moist lees (drained in bags and pressed) contain as much as 8 per cent, of cream of tartar, or, if dried in the sun, still more. These lees are nowadays worked by the Dietrich and Schnitzer process (1865) : they are first powdered and mixed by means of stirrers and then heated for 5 to 6 hours in iron 842 ORGANIC CHEMISTRY autoclaves (4 metres high and 1-4 wide for 15 quintals of lees) at 4 to 5 atmos. pressure, direct steam being admitted by copper coils and the air first allowed to escape. The albu- minoids are thus coagulated together with large quantities of colouring-matters, and the mass can then be easily filtered, but before this, it is discharged into a lead-lined wooden vessel (holding 10 to 12 cu. metres) containing 3 cu. metres of water and the amount of hydrochloric acid (20 to 22 Be.) corresponding with the quantity of tartar previously determined (100 kilos of potassium bitartrate require 60 kilos of hydrochloric acid at 20 Be., or 54-5 kilos at 22 Be.). The mass is well mixed and passed through the filter- press, in which it is washed with water. The tartaric acid is then separated from the solu- tion as calcium tartrate, as described above. Lees may be treated economically and well by the process of Cantoni, Chautems, and Degrange described above. To utilise the potash salts of the filtrate from the calcium tartrate, A. Albert! (U.S. Pat. 757,295, 1910) decomposes the organic substances in the hot with calcium chloride, filters and concentrates in a vacuum. The second phase of the manufacture of tartaric acid consists in the decomposition of the calcium tartrate by means of sulphuric acid (see above) and in the subsequent crystallisa- tion of the tartaric acid. The calcium salt in the decantation or washing vessels, or as cakes from the filter-presses, is broken up and suspended in 5 to 6 times its weight of water in lead-lined wooden vessels furnished with stirrers covered with lead and with coils for indirect steam-heating. After the liquid paste has been well stirred, the sulphuric acid, previously diluted, is added in such amount that, after an hour's stirring at 60 to 70, there is still a slight excess of sulphuric acid, detectable by the faint green coloration imparted to a solution of methyl violet. Too great an excess of sulphuric acid produces blackening of the tartaric acid solution during concentration, whilst deficiency of sulphuric acid results in the formation of turbid, impure tartaric acid crystals ; but when a little free sulphuric acid is present, fine shining crystals are obtained. Usually 1 kilo of sulphuric acid at 66 Be. is employed per 3 kilos of dry calcium tartrate. The solution is boiled for a couple of hours, left to cool, and the calcium sulphate which forms separated by means of a filter-press, washed with a little tepid water (this being added to the filtrate), and then with much cold water (this being used for treating subsequent quantities of calcium tartrate). The tartaric acid solu- tions were at one time concentrated in shallow, lead-lined, wooden vessels containing leaden steam-coils. But nowadays concentration is carried out in vacuum pans which are similar to those described later in dealing with the sugar industry and are of lead and of considerable thickness. The liquid is evaporated until it becomes almost syrupy, and is then discharged into wooden vessels containing stirrers, which cool the concentrated solution rapidly and thus cause the tartaric acid to separate in small crystals. The cold mass is quickly separated from the mother -liquor in a centrifuge, the crystals being washed with a fine spray of cold water. The mother-liquors are then concentrated until they give crystals, this process being carried out three times ; they are finally treated with milk of lime to separate the residual tartaric acid as calcium tartrate, which is filtered and worked up with the other calcium tartrate. The tartaric acid crystals are dissolved in one-half their weight of boiling water (if necessary, the solution is decolorised with animal charcoal and filtered) and the liquid left to crystallise, the crystals thus obtained being centrifugcd and dried on sheets of lead in a current of air at 30 ; the mother-liquors are used to dissolve fresh quantities of the small crystals. The final yield is about 90 to 95 per cent, of the total tartaric acid in the prime materials when these are poor (e.g. lees with 20 to 25 per cent, of tartar) or 97 to 99 per cent, when richer prime materials (70 to 80 per cent, of tartar) are used. A process still little used but worked successfully for many years in the Montredon factory near Marseilles is that of Gladysz (Ger. Pat. 7,352, October 15, 1885), which is based on the following observations: (1) When calcium tartrate is suspended in water and saturated with sulphur dioxide, soluble calcium bisulphite is formed and the tartaric acid liberated ; (2) when this solution is heated to 66, the sulphurous acid is expelled and all the tartaric acid precipitated as pure crystallised calcium tartrate; (3) if in (1) potassium bitartrate is used instead of calcium tartrate, potassium bisulphite and tartaric acid are formed, and on heating the liquid to 80, sulphur dioxide is evolved and pure TARTARIC ACID STATISTICS 343 potassium bitartrate separated in a crystalline state ; (4) potassium tartrate, when treated with calcium bisulphite, gives potassium bisulphite and calcium bitartrate ; the latter separates at 100, when the potassium bisulphite gives, with lime, calcium bisulphite and caustic potash, which can also be utilised. In practice Gladysz proposed to suspend the tartar in lumps in lead-lined wooden vessels, and into these hermetically sealed and arranged in a series of five or six to pass sulphur dioxide. The solutions (10 to 12 Be.) thus obtained are sent to the concentration apparatus, which communicates with towers to condense the sulphur dioxide (see vol. i, p. 245). When the latter is completely evolved, the liquid is kept at 125 for a short time to separate crystalline calcium tartrate, which is collected by means of a centrifuge, while the solution is concentrated further and allowed to cool in shallow lead-lined wooden vessels to deposit the potassium tartrate. As a rule, however, the hot potassium tartrate solution is treated directly with calcium chloride and a little lime, so that it yields insoluble calcium tartrate, which is more easily separated. From this calcium tartrate, pure tartaric acid is then obtained in the ordinary way. Although theoretically no sulphur dioxide should be lost in this process, in practice about 15 per cent, is lost in winter and 20 per cent, in summer. The Gladysz process, somewhat modified by Ciapetti, is used in Italy in the manufacture of tartar from vinasse, lees, &c. (see p. 338). USES AND STATISTICS OF TARTARIC ACID. This acid is used in considerable quantities to replace the more expensive citric acid in the preparation of beverages, liquors, and aerated waters, and in wine-making. Large quantities are consumed in the mordanting of wool and silk, to reduce chromium salts, &c., in the printing of textiles, manufacture of dyes, photography, medicine, &c. Refined tartaric acid pays an import duty in Italy of 8s. per quintal and in the United States and Spain of 25 per cent, ad valorem. Italy possesses four tartaric acid factories : at Carpi, Agnano (Pisa), Barletta, and Milan. The last two are the more important, and are able to produce together as much as 30,000 quintals per annum. The world's production in 1905 was about 110,000 quintals, of which Italy produced 6700 quintals ; England and the United States, each more than 25,000 ; Germany, about 15,000 ; France, about 8000 (13,000 in 1910) ; and Austria -Hungary about 10,000 quintals. Germany exported 17,000 quintals of the refined acid in 1908 and 19,000 in 1909, in which year the importation was 3240 quintals. In 1909, England imported 18,500 quintals and exported 3500. Italy imported 981 quintals, worth 10,200, in 1902 ; 1050 quintals in 1905 ; 1658 in 1907 ; 1574 quintals, worth 15,216, in 1909 ; 2976 quintals (one-third from Austria), worth 27,380, in 1910. Exports from Italy amounted to 17,000 quintals in 1907 ; 19,300 in 1908 ; 15,050, worth 141,480, in 1909, and 21,774 quintals, worth 195,966, in 1910. In 1907, four Russian factories, worked by a syndicate, produced 6000 quintals of tartaric acid and sold it at 20 per quintal. In 1911 the United States imported 12,500 tons (575,400) of tartar. In 1904, Argentine imported 950 quintals, and in 1909, 4650 quintals. But there is now a factory at Buenos Ayres which can produce 3500 quintals per annum. Pure tartaric acid, which formerly cost 14 per quintal, is now sold at 10 to 10 8s. ARTIFICIAL TARTARIC ACID. In 1889 a process was patented by Basset, and in 1891 a similar but improved one by Naquet for obtaining tartaric acid from starch (1 to 5 of water) by saccharifying with an equal weight of hot sulphuric acid (51 Be.), then adding double this quantity of sulphuric acid and almost as much sodium nitrate and heating at 100. When the reaction becomes very vigorous, the temperature is moderated and the heating continued at 80 to 90 for 2 to 3 days, the evaporated water being first replaced and the liquid concentrated to a syrup when evolution of gas ceases. In this manner all the saccharic acid is decomposed ; the sulphuric and oxalic acids are then neutralised with calcium carbonate and the tartaric acid subsequently worked up in the usual manner by way of its calcium salt. One hundred kilos of starch should give theoretically 140 of calcium tartrate corre- sponding with 56 kilos of tartaric acid, but in practice the yield of the acid does not exceed 55 to 60 per cent, of this theoretical amount. TRIHYDROXYGLUTARIC ACID, CO 2 H- [CH(OH)] 3 -CO 2 H, should exist in four stereoisomeric forms, of which the dextro-, Isevo- and racemic (m.pt. 127) compounds 344 ORGANIC CHEMISTRY are known. They are obtained by the oxidation of xylose or arabinose, and, on reduction, give glutaric acid, their constitution being thus confirmed. SACCHARIC ACID, CO 2 H- [CH (OH) ] 4 -CO 2 H, forms ten stereoisomerides, which are all known and are closely related to the sugars. Saccharic acid is formed as a rule in the oxidation of sucrose, glucose, mannitol, and starch (e.g. with nitric acid). It is deliquescent and soluble in water, and when heated or fused is converted into saccharone, which is a dextro-rotatory lactone melting at about 150. MUCIC ACID, CO 2 H- [CH(OH)] 4 -CO 2 H, is the stereoisomeride of saccharic acid which is constantly inactive. It is obtained on oxidising lactose, dulcitol, gum, &c., and forms a white powder very slightly soluble in water. KETONIC DIBASIC ACIDS Ethers of these acids, like those of /3-ketonic acids (ethyl acetoacetate, &c. ; see p. 332), show both ketonic and acid decompositions, and also a new method in which carbon monoxide separates. MESOXALIC ACID (Dihydroxymalonic Acid), CO 2 H-CO-C0 2 H + H 2 O or C0 2 H-C(OH) 2 -C0 2 H, shows ketonic behaviour in agreement with the first formula, but the molecule of water cannot be separated from the deliquescent prisms even at 100, and, further, derivatives (esters, &c.), are known which correspond better with the second formula, whilst the latter also explains well why mesoxalic acid, when heated with water, loses C0 2 and gives glyoxylic acid, CO 2 H-CH(OH) 2 . The structure of the acid likewise follows from its formation by the action of barium hydroxide on ethyl dibromomalonate : CBr 2 (C0 2 C 2 H 6 ) 2 + Ba(OH) 2 = BaBr 2 + C(OH) 2 (C0 2 C 2 H 5 ) 2 . Its ketonic constitution is confirmed also by the fact that it gives tartronic acid, CO 2 H-CH(OH).C0 2 H, on reduction. OXALACETIC ACID (Butanonedioic Acid), CO 2 H-CH 2 .CO-CO 2 H,is not known in the free state, but is formed as ether by condensation of ethyl oxalate and ethyl acetate in presence of sodium ethoxide (see Ethyl Acetoacetate). It also splits up in two ways according as it is treated with dilute sulphuric acid (giving pyruvic acid, CO 2 , and alcohol) or with alkali (giving oxalic and acetic acids). Being a ketone, it forms an oxime. The alcoholic solution is coloured dark red by ferric chloride and hence corresponds with the enolic form, C 2 H 5 .O-CO.CH : C(OH).C0 2 C 2 H 5 . This ester, like ethyl aceto- acetate, is used in many syntheses, the hydrogen of the methylene group being replaceable by sodium, &c. ACETONEDICARBOXYLIC ACID (Pentanonedioic Acid), C0 2 H CH 2 CO CH 2 C0 2 H, forms crystals which melt at 135, losing 2CO 2 , and giving acetone. It is formed by the action of concentrated sulphuric acid, in the hot, on citric acid : CH 2 .C0 2 H CH 2 -C0 2 H H = C0 + H + C0 CH 2 .C0 2 H CH 2 .C0 2 H Citric acid The constitution of citric acid is shown by its formation from acetonedicarboxylic acid by the action of hydrogen cyanide and subsequent hydrolysis. The hydrogens of the two methylene groups are replaceable by sodium, so that this acid can be used in syntheses similar to those effected by ethyl acetoacetate. DIHYDROXYTARTARIC ACID, CO 2 H-CO-CO-C0 2 H + 2H 2 O, or, better, C0 2 H-C(OH) 2 -C(OH) 2 -C0 2 H, melts and decomposes "at 98 and forms a sodium salt, which is sparingly soluble and decomposes readily into CO 2 , and sodium tartronate, CO 2 H-CH(OH)-CO 2 Na. It is obtained by the action of nitrous acid on an ethereal solution of pyrocatechol, guaiacol, &c., and also by the spontaneous decomposition of nitrotartaric acid. Sodium bisulphite converts it into glyoxal, while with hydroxylamine it forms the dioxime CITRIC ACID 345 corresponding with the diketonic form, With phenylhydrazine-sulphonic acid, it forms tartrazine, a beautiful yellow colouring-matter largely used in dyeing wool and silk. Of the higher ketonic acids the following may be mentioned : hydro- chelidonic acid (acetonediacetic acid), CO(CH 2 -CH 2 -C0 2 H) 2 ; diacetosuccinic CH 3 - CO* CH- C0 2 H OHfC'O'PH VCO H acid, ; and diacetylqlutaric acid, CH 2 CH 2 'O > CH 2 'CO 2 H and (CH 2 -CO-NH 2 ) 2 0, are also known, the latter, on heating, giving ammonia and digly- PTT . p/1 collimide, 0<; 2 ^r>NH, which melts at 142. vll 2 * L/U MALIC ACID, CO 2 H-CH 2 -CH(OH)-CO 2 H, forms two amides by means of its two carboxyl groups, an amine by means of its alcoholic group (aspartic acid), and also an amino -amide (asparagine). MALAMIC ACID, NH 2 -CO-CH 2 -CH(OH)-C0 2 H, is known best as its crystalline ethyl ester, which is formed by the action of ammonia on an alcoholic solution of ethyl malonate. MAL AMIDE, NH 2 -CO-CH 2 -CH(OH)-CO-NH 2 , is formed by the action of ammonia on ethyl malonate in the dry state. B. IMIDES AND IMINO-ETHERS Attention must be drawn, not so much to the secondary amides (in which two hydrogen atoms of ammonia are replaced by two acid residues, as in diacetamide, (CH 3 -CO) 2 NH, which contains the iminic group, NH) or to the tautomeric form of the primary amides (with X-C\ corresponds the X NH 2 isomeride X C< , which is well known in the form of imino-ethers, X or, in the case of the imidohydrin of glycollic acid, OH-CH 2 -CV , in the X free state) as to the imides of certain dibasic acids. CCK OXIMIDE, ) /NH (perhaps with the double formula), is formed on or elimination of water from oxamic acid (by PC1 5 ). CH a -CNH, is obtained by distilling ammonium glutarate ; it melts at 152 and gives a little pyridine when heated with zinc dust. C. AMINO-ACIDS AND THEIR DERIVATIVES In the amino-acids, it is the hydrogen in direct union with carbon that is replaced by the NH 2 -group, the carboxyl group remaining intact, so that these compounds have both acidic and basic functions and can hence be readily separated from other substances, since after the carboxyl is esterified, salts such as the hydrochlorides of the amino-group are formed. These substances and their derivatives are of considerable importance in animal and vegetable physiology, since they are found among the products of the gradual synthesis and decomposition of the proteins in the living organism ; they are also of interest theoretically, as they form intermediate products in various chemical syntheses. The a-amino-acids are readily obtained by the action of ammonia on the cyanohydrins of ketones and aldehydes and hydrolysis of the remaining nitrile group : /OH /NH 2 CH,- Cf-H + NH, = H 9 + CH, L 3 ^\ JJL i L ' J " 1 3 J - L 2 V -' i VyJ - L 3 X CN X CN Nitrile of lactic acid Nitrile of alaiiiiie sNH 2H 2 O = NH 3 -f- Clla'Cx; JH CN C0 2 H Alaninc (a-Aminopropionic acid) They are also formed generally by reducing the oximes of ketonic acids or, better, by the action of ammonia on halogenated acids : C0 2 H-CH 2 C1 + NH 3 = HC1 + C0 2 H-CH 2 -NH 2 . We may also mention the interesting Korner-Menozzi reaction (see p. 314), which allowed these authors, by inverting the reaction, to pass from the esters of unsaturated acids (fumaroid or maleinoid form) to a single form of the corresponding saturated amino-acids by simple treatment with ammonia (or even an alkylamine in alcoholic solution). With nitrous acid, the amino-acids give hydroxy-acids and they give many reactions analogous to those of the hydroxy-acids and varying with the position of the amino-group. Two molecules of an a-amino-acid readily lose 2 mols. of water, giving a kind of anhydride with an imido-ketonic character : GLYCINE, BETAINE, ASPARTIC ACID 355 C0 2 H-CH 2 -NH 2 CO-CH 2 -NH = 2H 2 O + | NH CH-C0H NH-CH-CO 2 22 2 The y-amino-acids, however, give internal anhydrides analogous to the lactones and termed Lactams : C0 2 H CH 2 CH 2 CH 2 NH 2 = H 2 + CO CH 2 CH 2 CH 2 NH The /3-amino-acids, when heated, evolve ammonia and give unsaturated acids. GLYCOCOLL (Glycine, Aminoacetic or Aminoethanoic Acid, or Amine of Glycollic Acid), CO 2 H-CH 2 -NH 2 , is formed on boiling gelatine with alkali [Ba(OH) 2 ] or acid (dilute H 2 SO 4 ) or on heating hippuric acid (benzoylglycocoll) with dilute acid : CO 2 H.CH 2 .NH.CO.C 6 H 5 + H 2 O = CO 2 H.CH 2 -NH 2 + C 6 H 5 -CO 2 H (benzoicacid). Syn- thetically it is obtained from monochloracetic acid and concentrated ammonia (see p. 322) ; if the ammonia is replaced by methylamine, sarcosine, CO 2 H CH 2 NH CH 3 , m.pt. 115, is obtained, or if by trimethylamine, betaine (see p. 323) is formed : C0 2 H.CH 2 C1 + N(CH 3 ) 3 = HC1 + CO.CH 2 .N(CH 3 ) 3 . Betaine, C 5 H H 2 N, crystallises with 1H 2 O, which it loses at 100, or in a desiccator over sulphuric acid. It dissolves in water or alcohol, from which it is precipitated by ether or as betaine hydrochloride, by hydrochloric acid. This solid hj'drochloride ; s soluble in water, which hydrolyses it to a considerable extent, the solution then behaving like hydrochloric acid. Owing to this property it is sold, under the name of acidol, in pastilles containing exact and suitable doses for stomach complaints, and replaces solutions of hydrochloric acid for this purpose ; the same effect as that of the acid is thus obtained by a solid product. Betaine is a feeble base, and is not decomposed even by boiling aqua regia ; at high temperatures it decomposes, giving trimethylamine. It occurs abundantly in beet-sugar molasses (10 to 12 per cent., besides 1 to 2 per cent, of leucine and isoleucine and 5 to 7 per cent, of glutamic acid), from which it is extracted by means of alcohol ; after evaporation of this solvent, it is separated as hydrochloride. The action of tertiary amines, other than trimethylamine, with mor.ochlo! acetic acid gives various compounds to which is given the name of BETAINES. Substitution in the amino-group of the amino-acids also yields other interesting compounds, e.g. Aceturic Acid (acetylglycocoll), C0 2 H CH 2 NH CO CH 3 , melting at 206. The properties of glycocoll and its salts are given on p. 322. In the amino-acid group is also found SERINE or a-amino-/3-hydroxypropionic acid, CO 2 H-CH(NH2)-CHg-OH, which is obtained on boiling silk gelatine with dilute sulphuric acid or syntheticaHy from glycollic aldehyde, ammonia, and hydrocyanic acid. LEUCINE (a-aminoisocaproic acid), CO 2 H-CH(NH 2 )-CH 2 -CH(CH 3 ) 2 , is obtained synthetically by hydrolysing the nitrile of isovaleraldehyde-ammonia, and is usually found with glycine among the products of decomposition of the proteins by acid or alkali, and is then optically active (the carbon atom adjacent to the carboxyl being asymmetric). ASPARTIC ACID (Aminosuccinic Acid), C0 2 H-CH 2 -CH(NH 2 )-CO 2 H, is one of the most important products obtained by the decomposition of proteins by acid or alkali. It occurs in abundance (laevo-rotatory) in beet-sugar molasses, and has been prepared by various synthetic methods, e.g. by the action of ammonia on bromosuccinic acid. Three stereoisomerides are known, two of them being optically active owing to the presence of an asymmetric carbon atom. They are obtained in small, tabular, dimetric crystals, soluble to some extent in hot \vater. Their cold solutions and also acid solutions of the dextro-rotatory acid have a sweet 356 ORGANICCHEMISTRY taste, but hot solutions or alkaline solutions of the Isevo -rotatory acid are without taste. They give the general reaction of amines and amides with nitrous acid, being converted into malic acid. 1 The higher homologue of aspartic acid is Glutamic Acid (a-aminoglutaric acid), C0 2 H CH(NH 2 ) CH 2 CH 2 C0 2 H. Among the DIAMINO-ACIDS we have Lysine, C0 2 H-CH(NH 3 )- [CH 2 ] 4 -NH 2 , which is obtained by the action of acids on proteins or by synthetical methods ; on putrefaction it gives pen tame thylenediamine. Ornithine, C0 2 H-CH(NH 2 )- [CH 2 ] 3 -NH 2 , is the lower homologue of lysine and gives tetramethylenediamine (putrescine) on putrefaction. Taurine (Ethyleneaminosulphonic Acid), SO 3 H'CH 2 'CH 2 -NH 2 , is found in ox-bile combined with cholic acids as taurocholic acid (for properties of taurine, see p. 214). Cysteine (Thioserine), C0 2 H-CH(NH 2 )-CH 2 'SH, is formed by the reduction of cystine, C0 2 H-CH(NH 2 )-CH 2 -S-S-CH 2 -CH(NH 2 )-CO 2 H, which occurs in urinary sediments (calculi). ASPARAGINE, NH 2 -CO-CH 2 -CH(NH 2 )-CO 2 H, is the amide of aspartic acid. It was first found in asparagus, but is moderately widespread in almost all vegetables (beet, potatoes, beans, vetches, peas, &c.) during the germina- tion period, and the dry seeds of certain lupins contain as much as 30 per cent. The constitution of asparagine is confirmed by the various syntheses leading to its production. It crystallises with 1H 2 in Isevo-hemihedral, trimetric prisms, soluble in hot water but insoluble in alcohol or ether. With aqueous cupric acetate solution, it forms a blue, well-crystallised copper salt (C 4 H 7 3 N 2 )2Cu, insoluble in water. It is isomeric with malamide, from which is differs in the possession of both acid and basic characters. It is Isevo -rotatory and has an unpleasant, insipid taste, but vetch seedlings contain a dextro-rotatory asparagine which has a sweet taste (Piutti, 1886), but does not unite with the Isevo -rotatory form also present in the seedlings to give the inactive modification. Pasteur stated that the substance composing the nerves of the palate behaves as an optically active combination which acts differently towards the dextro- and Isevo- asparagines. Asparagine is converted into aspartic acid by hydrolysis and into malic acid by the action of nitrous acid. ASPART AMIDE, NH 2 -CO-CH 2 -CH(NH 2 )-CO-NH 8 , is the diamide or normal amide of aspartic acid. Numerous higher homologues of aspartic acid (Homo- Aspartic Acids) and of asparagine (Homo- Asparagines) are known. D. AMIDO- AND IMIDO-CHLORIDES With both the primary amides and also the alkylated amides, the oxygen is readily replaced by chlorine by /the action of PC1 6 . Thus, acetamide gives acetamido-chloride, CH 3 -CC1 2 -NH 2 , and ethylacetamide, ethylacetamido-chloride, CH S CC1 2 ' NH C 2 H 6 . Both of these compounds readily lose HC1, forming imino-chlorides, e.g. acetimino-chloride, CH 3 -CC1:NH, and ethylacetimino-chloride, CH 3 -CC1 : N-C 2 H 6 . These imino-chlorides, like amido-chlorides, are readily decomposed by water into hydrogen chloride and amide. These chlorinated compounds react easily with aromatic substances and with hydrogen sulphide, ammonia, and amines, the chlorine being thus replaced by sulphur or by amino- residues, forming thioamides, e.g. CH 3 -CS-NHX, and amidines, e.g. CH 3 -C(NH 2 ) : NX 2 . 1 By the action of nitrons acid on tho ethyl ester of glycocoll, Curtius obtained Ethyl Diazoacetate *\ II yCH-COjCjHs, as a yellow oil with a peculiar odour; when heated with water it explodes, losing nitrogen N' and taking up water to form ethyl glycollate. IMINOTHIOETHERS, AMIDINES 357 E. THIOAMIDES These are well-crystallised compounds, more acid in character than the amides, and hence capable of forming metallic derivatives and of dissolving in alkali. Besides by the reaction just mentioned they are obtained by the addition of H 2 S to nitriles : CH 3 -CN + H 2 S = CH 3 -CS-NH 2 (thioacetamide or ethanethioamide) ; on heating, the opposite change occurs. Phosphorus pentasulphide replaces the oxygen of amides by sulphur, thus forming thioamides. With H 2 S, isonitriles give the alkylated thioamides of formic acid, CN-X + H 2 S = H-CS-NHX. Thioamides are readily hydrolysed (by alkali, hot water, &c.), with formation of H 2 S, NH 3 (or amine), and the corresponding acids : X-CS-NHX' + 2H 2 = X-C0 2 H + NH 2 X' + H 2 S. F. IMINOTHIOETHERS The thioamides (and especially their derivatives) can exist in the isomeric or tauto- meric form, X-C(SH) : NH, in which the hydrogens of both the sulphydryl and the imino- group are replaceable by alkyl groups, Iminothioethers, e.g. X-C(SX') : NH, being then formed. These are prepared by the action of alkyl iodides on the thioamides (ako from thioalcohols, nitriles, and gaseous hydrogen chloride), e.g. CH 3 -CS-NH 2 + CH 3 I = /S-CH 3 CH 3 -C^ , HI (acetiminothiomethyl hydriodide). ^NH The iminothioethers are easily hydrolysed (by HC1), forming ammonia and esters of thio -acids : yS-CH 3 CH 3 -(\ + H 2 O = NH 3 + CH 3 -CO-SCH 3 . ^NH G. AMIDINES When the amides or alkylamides are heated with amines in presence of a dehydrating agent (like PCl^), the oxygen of the amide is substituted by an imino-residue : /NH 2 X-CO-NH 2 + R-NH 2 = H 2 O + X-C/ ^NR /NHX' X-CO-NHX' + R-NH 2 = H 2 O + X-C/ X NR These compounds are obtained also from thioamides, isothioamides, iminochlorides, or iminoethers by the action of ammonia or of primary or secondary amines : ,NH CH 3 -CS-NH 2 + NH 3 = H 2 S + CH 3 -Cf (acetamidine or ethanamidine). X NH 2 NH ,NH + R-NH 2 = X-C( + HSX'. SX' \NHR When nitriles are heated with the hydrochlorides of primary (of the aromatic series also) or secondary amines (not with NH 4 C1), alkylami dines are obtained : CH 3 -CN + R-NH 2 = CH 3 -C(: NH)-NHR. Properties. The amidines (or amimides) are bases and usually of the aromatic series ; they are easily hydrolysed (by boiling with, alkali or acid), giving (when the iminic hydrogen 358 ORGANIC CHEMISTRY is not replaced by an alkyl group) ammonia (or an amine) and a nitrile ; the same change occurs on dry distillation. With H 2 S they give first an additive product : /NHX' ; X-C(: NH)-NHX' + H 2 S = X-CSH this product then decomposes in two senses, giving (a) X-CS-NHX' + NH 3 and (6)X-CS.NH 2 + X'-NH 2 . With CS 2 , amidines give thioamides and, at the same time, thiocyanic acid or an alkyl thiocyanate. H. HYDRAZIDES AND AZIDES Introduction of an acid residue into hydrazine, H 2 N-NH 2 (see vol. i, p. 327), gives the primary hydrazides or monoacylhydrazides, e.g. CH 3 -CO-NH-NH 2 (acetylhydrazide) and H-CO-NH-NH 2 (formhydrazide, m.pt. 54) ; two acid radicals give secondary hydrazides or dihydrazides, e.g. CH 3 -CO-NH-NH-CO-CH 3 (diacethydrazide, which melts at 138 and is prepared from hydrazine hydrate and acetic anhydride). They are readily hydrolysable, and reduce ammoniacal silver nitrate solution in the cold and Fehling's solution on heating. The primary hydrazides are more highly basic than the amides, and so give more stable salts. Nitrous acid acts on primary hydrazides, forming azides, which are derivatives of hydrazoic acid (see vol. i, p, 327) : /? CH 3 -CO.NH.NH 2 + HNO 2 = 2H 2 + CH 3 -CO.N/ || \N These resemble the acichlorides in many properties, but are explosive and, when heated with alcohol, give urethanes and liberate nitrogen : /N CH 3 -CON^ || + C 2 H G -OH = N 2 + CH 3 -NH-CO,C 2 H 6 , methylurethane, \N which can be hydrolysed with formation of C0 2 , alcohol, and methylamine. It is hence possible to pass from an acid to an amine with one carbon atom less, by way of the hydrazide and azide. I. HYDROXYLAMINE-DERIVATIVES OF ACIDS Hydroxylamine or its residues can be united to acid residues, forming Hydroximic (or hydroxamic) Acids, e.g. CH 3 -C(: N-OH)OH (ethylhydroxamic acid, m.pt. 59), and Amidoximes, X-C(: N*OH)-NH 2 . The hydroxamic acids have an acid character and are formed, with evolution of ammonia, by the action of hydroxylamine on amides. Also Formyloxime Chloride, CH(: N- OH)C1, is known, this being obtained by treating mercury fulminate in the cold with HC1 ; it forms needles, which are readily decom- posable, volatile, and soluble in ether. The Amidoximes are formed by the addition of nitriles to hydroxylamine, CH 3 CN + NH 2 -OH = CH 3 -C(: NOH)NH 2 . If hydrogen cyanide is employed, ISURET (Methanamidoxime or Methenylamidoxime), CH(: NOH)NH 2 , isomeric with urea, would be obtained. VI. CYANOGEN COMPOUNDS Some cyanogen compounds, especially Hydrocyanic Acid, HCN, potassium cyanide, and ferro- and ferricyanides, have already been dealt with in vol. i, pp. 397, 437, and 650. We have to consider here the numerous and varied organic derivatives of cyanogen, which are of some interest as they often exist in polymerised forms and almost always in two isomeric modifications, sharply differentiated by their chemical properties : derivatives of nitriles, X -C : N, and of isonitriles, C ;N-X (see also p. 199). CYANOGEN, (CN) 2 , is a highly poisonous gas with a pungent odour recalling that of bitter almonds ; it is liquid at 21 and solid at 34. It CYANOGEN DERIVATIVES 359 is found in the gas from blast-furnaces and occurs largely in the tail of Halley's comet, which approached the earth in May 1910. It is obtained by the elimination of water from ammonium oxalate or oxamide (NHg-CO-CO-NHg), by the action of P 2 5 in the hot, or by heating a solution of copper sulphate with potassium cyanide, and is commonly prepared by heating cyanide of silver or of mercury, Hg(CN) 2 = Hg + (CN) 2 ; as a secondary product, PARACYANOGEN, (C 3 N 3 ) 2 , or (CN) X is formed as a brown powder. It burns with a purple flame, dissolves readily in alcohol or water (4 : 1), and its solutions gradually become brown, with formation of oxalic acid, formic acid, hydrocyanic acid, ammonia, and urea, and deposition of Azulmic Acid (brown powder). With H 2 S it forms the thioamides : RUBEANHYDRIC ACID, NH 2 -CS-CS-NH 2 , and FLAVEANHYDRIC ACID, NC-CS-NH 2 . CYANOGEN CHLORIDE, NC-C1, is of importance in the synthesis of many cyanogen compound-!, and is formed by the action of chlorine on hydrocyanic acid or metallic cyanides : NC-H + C1 2 = HC1 + NC-C1. It is a colourless gas which is easily liquefied, boils at 15-5, has a pungent odour, and dissolves in water. In presence of HC1 it poly- merises, forming Cyanogen Trichloride (melts at 145, boils at 190). With KOH it forms potassium cyanato, NOOK. CYANIC ACID, NC-OH, is a liquid of penetrating odour and only slight stability, even at the ordinary temperature. It is obtained by the dry distillation of cyanuric acid (q.v.) and condensation of the vapours in a freezing mixture. It undergoes change, even at the ordinary temperature and with slight explosions, into a compact, white isomeride, which is polymerised iso- cyanic, acid or cyamelide (O : C : NH) X ; this, on dry distillation, gives cyanic acid again. Its salts are more stable, but when attempts are made to liberate the acid from these by the action of mineral acids, immediate hydrolysis occurs : NO- OH + H 2 = C0 2 + NH 3 . If it is liberated by dilute acetic acid, the isomeric cyanuric acid is obtained. The alkyl derivatives of cyanic acid exhibit two isomeric forms : Cyanates, N : C-OX, and Isocyanates, O : C : NX. Potassium Cyanate, NCOK, forms white scales soluble in alcohol or water, and is obtained by oxidising solutions of potassium cyanide by means of potassium perman- ganate or dichromate, or by fusing potassium cyanide or ferrocyanide with Pb0 2 or Mn0 2 : NCK + = NCOK. Ammonium Cyanate, NO- ONH 4 , is isomeric with urea, into which it can be converted. It is obtained by neutralising cyanic acid with ammonia and forms a moderately stable, white, crystalline mass. ETHYL ISOCYANATE, CO : NC 2 H 5 , is prepared by distilling potassium cyanate with either potassium, ethyl sulphate, or ethyl iodide. It is a liquid of penetrating odour arid boils at 60. It does not behave as a true ester (true esters of cyanic acid do not exist), since the action of acid or alkali yields, not alcohol, but ethylamine ; CO : NC 2 H 5 + H 2 O = C0 2 + C 2 H 5 -NH 2 . Hence the nitrogen is united directly to the alkyl group, so that the structure is not N : C-OC 2 H 5 , but O : C : NC 2 H 5 . Ethyl isocyanate is instantly decomposed by water, forming derivatives of urea ; the latter are also formed by the action of ammonia or amino -bases. CYANURIC ACID, (NC) 3 (OH) 3 , is a polymeride of cyanic acid and on heating urea which contains the constituents of ammonia and cyanic acid either alone or in a current of chlorine so as to eliminate the elements of ammonia, there remain those of cyanic acid, which polymerise to cyanuric acid. It crystallises with 2H 2 in prisms, effloresces in the air, and is readily soluble in hot water. When heated with HC1, it hydrolyses slowly to NH 3 and C0 2 ; with PC1 5 it gives the chloride of cyanuric acid. It is a tribasic acid and forms a violet, crystalline copper salt ; its sodium salt is insoluble in concentrated alkalis. Like cyanic acid, it gives rise to two series of derivatives, e.g. Ethyl Cyanurate, (NC) 3 (OC 2 H 5 ) 3 , which is a colourless liquid giving alcohol on hydrolysis. It is only slightly stable, and is readily transformed into the isomeride of the other series, Ethyl Isocyanurate, (CO) 3 (NC 2 H 5 ) 3 , which is formed by polymerisation of ethyl isocyanate, or by distilling the cyanurate with potassium ethyl sulphate. On hydrolysis it gives 360 ORGANIC CHEMISTRY ethylamine, this confirming its constitution, which is shown by the following closed-ring formulas to be clearly different from that of ethyl cyanurate. O OC 2 H 6 C 2 H 5 N N C 2 H 5 N II II I 0:C C:0 C 2 H 5 0-C C-OC 2 H 5 Ethyl cyanurate C 2 H 6 Ethyl isocyanurate FULMINIC ACID, C : NOH, is readily volatile but unstable, and is decomposed by concentrated hydrochloric acid into hydroxylamine and formic acid, chloroformyloxime, GHClrN-OH, being formed as intermediate product. Kekule regarded fulminic acid as a nitroacetonitrile, N0 2 'CH 2 -CN, but Nef subsequently attributed to it the consti- tution CrN-OH, the carbon being divalent. With bromine, mercury fulminate (see p. 255) gives the compound Br-C:N-0 Br C:N-0 Silver fulminate is even more explosive than the mercury salt. F. C. Palazzo (1907-1910) has prepared various additive products of fulminic acid with different acids (HBr, HI, HSCN, HN0 2> N 3 H). With hydrazoic acid at -12, he obtained two isomerides with different constitutions, probably with intermediate forma- tion of Triazoformoxime : CH N OH N 3 H + C : NOH - 5t >C : N-OH > || 3 N N : N Triazoformoxime N -hydroxytetrazole (m pt. 145) The other isomeride also is possibly a tetrazole derivative. By the action of hydrogen sulphide on mercury fulminate suspended in water, L. Cambi (1910) obtained and isolated the Formothiohydroxamic Acid predicted by Nef : H-C:NO-H H-S THIOCYANIC ACID AND ITS DERIVATIVES THIOCYANIC ACID (Rhodanic Acid), NC-SH, is a yellow liquid of penetrating odour, stable only when anhydrous in a freezing mixture or when in very dilute aqueous solution. At ordinary temperatures it polymerises to a yellow mass. It is obtained from its mercury salt (see later) by the action of hydrochloric acid. In concentrated aqueous solution, it undergoes conversion into a yellow crystalline mass of Perthiocyanic Acid, (CN) 2 S 3 H 2 . Cyanogen Sulphide, (NC) 2 S, may be regarded as a kind of anhydride of thiocyanic acid, and is obtained from silver thiocyanate and cyanogen iodide. It forms colourless plates which have a pungent odour and are readily soluble in water. Thiocyanuric Acid, (NC) 3 (SH) 3 , is polymeric with thiocyanic acid, and is obtained by the action of sodium sulphide on cyanogen chloride. ^It is a yellow powder and gives salts corresponding with a tribasic acid. Its trimethyl salt is formed by polymerisation of ethyl thiocyanate by heating at 180. POTASSIUM THIOCYANATE (or Rhodanate), NC-SK,is obtained by fusing potas- sium cyanide with sulphur, or evaporating a solution of potassium cyanide and yellow ammonium sulphide, or, better still, by fusing potassium ferrocyanide with potassium THIOCYANATES 361 carbonate and sulphur ; as prime material the mass used for the purification of illumi- nating gas is nowadays employed (vol. i, p. 651). It forms colourless prisms soluble in boiling alcohol and to a greater extent in water with absorption of heat. AMMONIUM THIOCYANATE (or Rhodanate), NC-SNH 4 , forms colourless, tabular crystals soluble in alcohol, and readily so in water. It is obtained by heating together CS 2 , NH 3 , and alcohol : CS S + NH 3 = NOSH + H 2 S, When heated it is transformed into the isomeric thiourea. It serves for the preparation of other thiocyanates, and is extracted in large quantities from the exhausted Laming mixture of gasworks (see p. 46), which contains 1 to 4 per cent, of it. MERCURIC THIOCYANATE, (NC-S) 2 Hg, is prepared from a mercuric salt and ammonium thiocyanate, and forms a white, insoluble powder which swells up to a very considerable extent when heated (Pharaoh's serpents). SILVER THIOCYANATE is precipitated as a white mass on mixing silver nitrate and ammonium thiocyanate. The latter gives, with ferric salts, a dark red coloration of FERRIC THIOCYANATE (sensitive indicator in the titration of silver with thio- cyanate) and this, with potassium thiocyanate, gives a violet double salt, (NOS) 6 FeK 3 . Hydrogen sulphide decomposes the thiocyanates, NOSH + H 2 S = NH 3 + CS 2 , while with concentrated sulphuric acid, addition of water and decomposition into ammonia and carbon oxysulphide occur NOSH + HaO = COS + NH 3 . For thiocyanic acid there are two series of isomeric derivatives, corresponding with the two general formulae: N : OSX (alkyl thiocyanate) and S : C : N-X (mustard oils). ETHYL THIOCYANATE, NC-SC 2 H 5 , is a colourless liquid with a marked odour of garlic ; it boils at 142 and is very slightly soluble in water. It is formed by the action of cyanogen chloride on mercaptides, or by distillation of potassium thiocyanate with potassium ethyl sulphate. As it has the constitution of a true ester, it is hydrolysed by alcoholic potash with formation of alcohol and potassium thiocyanate. But in certain reactions it behaves like the isomeric mustard oils. Nascent hydrogen converts it into merceptan, since the alkyl is united to sulphur, and the action of nitric acid in the hot yields ethylsulphonic acid. ALLYL THIOCYANATE, NC-SC 3 H 5 , boils at 161, and has a garlic-like odour; it undergoes change into the isomeric mustard oil, slowly at the ordinary temperature and more rapidly on distillation. The Mustard Oils (Isothiocyanates) are obtained from the corresponding thiocyanates simply by heating. They are also formed by the action of carbon disulphide on the corresponding primary amines, CS 2 + X-NH 2 = H 2 S + S : C : NX, this change taking place by way of the intermediate alkylamine salt of alkyldithiocarbamic acid (see later), which is distilled with mercuric chloride. Mustard oils are also formed when an alkylated thiourea is distilled with phosphoric or concentrated hydrochloric acid. Their structure is indicated by their formation of primary amine on hydro- lysis : S : C : NX + 2H 2 = C0 2 + H 2 S + X-NH 2 . The isothiocyanic acid, S : C : NH, from which these mustard oils are regarded as derived, is not known in the free sta.te. METHYL MUSTARD OIL (Methyl Isothiocyanate), SC : NCH 3 , melting at 34 and boiling at 119 ; Ethyl Mustard Oil, SC : NC 2 H 5 , boiling at 134 ; and Propyl Mustard Oil, SC : NC 3 H 7 , boiling at 153, are of little importance. More interesting is ALLYL MUSTARD OIL (or Ordinary Mustard Oil ; Allyl Isothiocyanate), S : C : NC 3 H 5 , which is prepared by distilling Sinapis nigra (black mustard) with water ; it is obtained synthetically by the reactions given above. It is a liquid with a pungent odour recalling that of mustard and raises blisters on the skin ; it is sparingly soluble in water and boils at 150-7. 362 ORGANIC CHEMISTRY CYANAMIDE AND ITS DERIVATIVES CYANAMIDE, NC-NH 2 , is a white crystalline substance, melting at 40 and dissolving very slightly in water, alcohol, or ether. It is obtained by passing a current of cyanogen chloride into an ethereal solution of ammonia : 2NH 3 + NOC1 = NH 4 C1 + NC-NH 2 . It is also formed by desulphurising thiourea, NH 2 -CS-NH 2 , by means of HgO, which removes H 2 S. It is obtained abundantly and in a pure state by extracting calcium cyanamide (see later) systematically with water, neutralising the saturated solution with sulphuric acid, filtering from the calcium sulphate, concen- trating in a vacuum, again filtering from the gypsum, concentrating anew, and extracting the crystalline mass formed on cooling with ether, which does not dissolve gypsum, dicyanamide, and other impurities. Evaporation of the ether yields pure cyanamide in almost theoretical yield (Baum, 1910). With lapse of time, or rapidly at 150, cyanamide changes into the poly- meric dicyanodiamide (see later). It behaves both as a weak base forming unstable crystalline salts, and as a weak acid giving metallic salts, e.g. NC-NHNa, NONAg 2 , &c. The most important of these is calcium cyan- amide, NC-NCa, which was considered in detail in vol. i, p. 309, in the dis- cussion of the utilisation of atmospheric nitrogen ; it is formed by the action of nitrogen on calcium carbide and forms an excellent nitrogenous fertiliser. In presence of dilute acid, cyanamide fixes a molecule of water, giving urea : NONH 2 + H 2 = NH 2 -CO-NH 2 ; with hydrogen sulphide it yields thiourea. Cyanamide also gives two series of isomeric alkyl derivatives of the general formulae N : C*NX 2 and XN : C : NX. Compounds of the latter formula are derived from the hypothetical carbodi-imide, NH : C : NH ; for example, carbodiphenylimide, C 6 H 5 N : C : NC 6 H 5 , boiling at 330, is well characterised. DIETHYLCY AN AMIDE, NC-N(C 2 H 5 ) 2 , is formed by the action of ethyl iodide on the silver salt of cyanamide, its structure being indicated by the products C0 2 + NH 3 + NH(C 2 H 5 ) 2 obtained on hydrolysis with dilute acid. Methyl- and ethyl-cyanamide are also known. DICYANODIAMIDE, (NC-NH 2 ) 2 , is formed, as has already been mentioned, from NH cyanamide ; certain of its reactions indicate the structure NC-NH-C^ (Bamberger). X NH 2 It forms acicular crystals or small flat prisms. When heated strongly and rapidly, it is converted into a white insoluble powder, MELAM, C 6 H 9 N n or [(NC) 3 (NH 2 ) 2 ] 2 NH, this being an imide of melamine, into which it is transformed by sulphuric acid or ammonia. MELAMINE (Cyanurtriamide), (NC) 3 (NH 2 ) 3 , is a crystalline basic substance, insoluble in alcohol or ether. When it is boiled with acid, the amino-groups are gradually replaced by hydroxyl groups, giving AMMELINE, (NC) 3 (NH 2 ) 2 OH, then AMMELIDE, (NC) 3 NH 2 (OH) 2 , and finally Cyanuric Acid, (NC) 3 (OH) 3 . As usual, the alkyl derivatives form two isomeric series, derivatives being known of a hypothetical Isomelamine, (CNH) 3 (NH) 3 , among these being the polymerised alkylcyanamides. VII. DERIVATIVES OF CARBONIC ACID True carbonic acid, : C(OH) 2 , is not knowjn in the free state, since two hydroxyl groups cannot exist in combination with the same carbon atom (see p. 182), but it is'supposed to exist in aqueous solution, and salts corre- sponding with this formula are stable and well known (carbonates and bicar- bonatesj. Also important organic derivatives, similar to those already studied for other dibasic acids (amides, chlorides, esters, &c.), are known. The acid derivatives are less stable than the normal ones. CARBONIC ACID DERIVATIVES 363 ESTERS OF CARBONIC ACID ETHYL CARBONATE, CO(OC 2 H 6 ) 2 , is a liquid which is insoluble in water, boils at 126, and has a pleasant odour. It is formedjby the interaction of ethyl chlorocarbonate and alcohol: C 2 H 5 -OH + C1-CO-OC 2 H 5 = HC1 + CO(OC 2 H 5 ) 2 , and also from silver carbonate and ethyl iodide. Mixed esters, containing different alkyls, also exist. ETHYLCARBONIC ACID, CO(OH)-OC 2 H 5 , is known only as salts, e.g. Potassium Ethylcarbonate, CO(OK)-OC 2 H 5 , which is obtained by the action of CO 2 on an alcoholic solution of potassium ethoxide and forms shining scales, giving alcohol and potassium carbonate when treated with water. CHLORIDES OF CARBONIC ACID Carbon Oxychloride (phosgene) , COC1 2 , has already been described (vol. i, p. 394). CHLOROCARBONIC ACID, COC1-OH, is the acid chloride of carbonic acid, but is not stable and, when liberated, decomposes into CO 2 and HO. Its esters are, however, well known, the action of phosgene on absolute alcohol giving, for example, ethyl chloro- carbonate (Ethyl Chloroformate), C1-CO-OC 2 H 5 , thus: C 2 H 5 -OH + COC1 2 = HC1 + C1-CO-OC 2 H 5 . This ester is a liquid, having a pungent odour, boiling at 93 and readily decomposing under the action of water ; it is used largely in organic syntheses to introduce carboxyl into the molecule. AMIDES OF CARBONIC ACID The acid amide, NH 2 CO OH, is Carbamic Acid, and the normal amide, NH 2 CO NH 2 , urea. CARBAMIC or CARBAMINIC ACID, NH 2 -CO-OH, is obtained as ammonium salt ammonium carbamate, NH 2 -CO-ONH 4 by the direct union of dry CO 2 and NH 3 ; a white mass is thus obtained which, even at 60, dissociates into C0 2 + NH 3 . In aqueous solution this salt does not precipitate solutions of calcium salts at the ordinary tem- perature, since calcium carbamate is soluble, but in the hot the salt decomposes into C0 2 and NH 3 and gives a precipitate of calcium carbonate. Ethyl carbamate or URETHANE, NH 2 -CO-OC 2 H 5 , is also well known and is obtained by the action of ammonia or ethyl carbonate, CO(OC 2 H 5 ) 2 + NH 3 = C 2 H 5 -OH + NH 2 -CO-OC 2 H 5> or, more easily, by treating ethyl chlorocarbonate with ammonia: COC1(OC 2 H 5 ) + 2NH 3 = NH 4 C1 + NH 2 CO OC 2 H 6 . It melts at 48 to 50, is soluble in water, and is used as a soporific. t The following are also known: iodourethane, NHI CO OC 2 H 5 ; ethylurethane, NHC 2 H 5 -CO-OC 2 H 5 (boils at 175); nitrourethane, N0 2 NH CO OC 2 H 5 ; carbamidyl chloride, NH 2 -CO-C1 (melts at 50 and boils at 61) ; and diethyl iminodicarfamate, NH(CO-OC 2 H 5 ) 2 , which is the imide of urethane. Urethane derivatives are readily hydrolysable with alkalis and yield ammonia and urea when heated. UREA (Carbamide), CO(NH 2 ) 2 , is the final oxidation product of nitrogenous com- pounds in the living organism, and the adult human being produces about 30 grms. of it a day ; it is found in general in the urine of carnivora (where it was first discovered) and in other animal fluids. It crystallises in shining needles soluble in water and in alcohol, but insoluble in ether ; it melts at 132 and sublimes in a vacuum. It is formed from ammonium cyanate by simple rearrangement under the action of heat (Wohler) : NC-ONH 4 = CO(NH 2 ) 2 . Escales (1911) found that when urea is distilled or sublimed in a vacuum, the reverse reaction, i.e. formation of ammonium cyanate, occurs. Urea is also obtained by the action of ammonia on ethyl carbonate or carbamic acid : CO(OC 2 H 5 ) 2 + 2NH 3 = 2C 2 H 5 -OH + CO(NH 2 ) 2 . Many other reactions give urea, e.g. oxidation of thiourea, action of water on cyanamide, &c., but in the laboratory it is prepared by treating with barium carbonate the urea nitrate obtained by evaporating urine in presence of nitric acid, or by heating ammonium sulphate solution with potassium ferrocyanide or cyanate : (NH 4 ) 2 S0 4 + 2NCOK = K 2 S0 4 + 2CO(NH 2 ) 2 . 364 ORGANIC CHEMISTRY When heated it is decomposed into ammonia, biuret (see later), cyanuric acid, and ammelide. It is readily hydrolysed by acids, alkalis, or even hot water : CO(NH 2 ) 2 + H 2 O = C0 2 + 2NH 3 , and is decomposed by nitrous acid or sodium hypo- chlorite : 2HNO 2 + CO(NH2) 2 = 3H 2 O + C0 2 + 2N 2 . It exhibits the properties of a base and of a weak acid, giving salts with acids (e.g. Urea Nitrate, CO(NH 2 ) 2 , HNO 3 , which is soluble in water and slightly so in nitric acid, and with concentrated sulphuric acid gives the highly acid Nitrourea, NH 2 -CO-NH-NO 2 ), and with bases, e.g. CO(NH 2 ) 2 , 2HgO. It also crystallises with other salts, e.g. CO(NH 2 ) 2 + NaCl + H 2 0, CO(NH 2 ) 2 + AgN0 3 , &c. Mercuric nitrate precipitates urea quantita- tively from its neutral aqueous solutions as 2CO(NH 2 ) 2 + Hg(N0 3 ) 2 + 3HgO. Urea forms various alkyl derivatives ; thus ethyl cyanate and ethylamine give symm. or vL-diethylurea, which is isomeric with unsymm. or /3-diethylurea, NH 2 -CO-N(C 2 H 5 ; ? : CO-NC 2 H 5 + C 2 H 6 -NH 2 = CO(NHC 2 H 6 ) 2 . The constitutions of these alkyl derivatives are determined by study of the products of their hydrolysis. "M'TT Readily hydrolysable alkylisoureas, NH : C<^ r v V 2 5 are also known. O-A. SEMICARBAZIDE, NH 2 -CONH'NH 2 , which is obtained from potasdum cyanale and hydrazine hydrate, may also be regarded as a derivative of urea. It has already been seen that this base (which melts at 96) gives crystalline compounds (semicarbazones) with ketones and aldehydes (seep. 206). CARBAZIDE (Carbohydrazide) , CO(NH-NH 2 ) 2 , melts at 152, and is obtained from esters of carbonic acid by the action of hydrazine hydrate. Acetylurea, NH 2 -CONH CO-CH 3 , and Allophanic Acid, NH 2 CO NH C0 2 H (not known free, but as salts), are obtained from acid chlorides and urea. The formation of ureides (compounds of urea and mono- and dibasic acids) takes place with monobasic divalent acids or with an alcohol and acid. Such a reaction gives Hydantoic Acid (glycoluric acid), NH 2 -CO-NH-CH 2 -CO 2 H, which, when evaporated in .NH-CO presence of HC1, loses water and forms Hydantoin, C0<^ | , the latter giving first \NH-CH 2 hydantoic acid and then C0 2 , NH 3 , and glycine on hydrolysis. When urea is heated at 160, 2 mols. condense with separation of ammonia and yCO-NH 2 formation of Biuret, NH<^ | , which crystallises with 1H 2 O and is soluble in water X CO-NH 2 or alcohol ; in alkaline solution it gives a characteristic violet coloration with a little copper sulphate. DERIVATIVES OF THIOCARBONIC ACID More or less complete substitution of the oxygen of carbonic acid by sulphur gives a series of unstable compounds, which form stable alkyl deriva- tives and exhibit various cases of isomerism indicated by varying products of hydrolysis. These numerous sulphur compounds are reducible to three types, according as they contain (1) the nucleus SC is only partly obtained on heating ammonium thiocyanate at 130, the reaction being reversible. It forms crystals melting at 172, and dissolving in water and in alcohol, giving neutral solutions ; it has a bitter taste. On hydrolysis it yields C0 2 + H 2 S + NH 3 . As has already been stated, it is converted into urea by permanganate, cyanamide by mercuric oxide, and potassium thiocyanate and ammonia by alcoholic potash at 100. It behaves as a weak acid and a weak base, and its derivatives, in some cases, correspond with the tautomeric formula, oTT 2 (hypothetical iminothiocarbamic acid). About 10,000 kilos of thiourea are produced annually by two factories, one French and the other German, for preserving loaded silk from corrosion, the Gianoli process (see later, under Silk) being used. Owing to this, the price of thiourea has been lowered from 2 to 5s. 6d. or 6s. 6d. per kilo. Acetylthiourea, Sulphohydantoin, &c. are also known. GUANIDINE AND ITS DERIVATIVES GUANIDINE (Iminourea or Iminocarbamide), NH : C<^jj 2 forms crystals readily soluble in water or alcohol. It is a strong base, absorbing carbon dioxide from the air, but is converted into salts by one equivalent of acid. The fatty acid salts are converted on heating into guanamines, which form crystals of peculiar shape. It is obtained by heating cyanamide with ammonium iodide : 366 ORGANIC CHEMISTRY NH 4 I + CN-NH 2 = NH : C(NH 2 ) 2 , HI ; or, better, as thiocyanate by heating thiourea with ammonium thiocyanate at 190 : NCS-NH 4 + SC(NH 2 ) 2 = H 2 S + NH : C(NH 2 ) 2 , NC-SH. It may also be obtained from dicyanodiamide by the action of aqua regia (C. Ulpiani, 1907). Guanidine is readily hydrolysed, forming first ammonia and urea and then C0 2 and NH 3 . Guanidine Nitrate, NH : C(NH 2 ) 2 , HN0 3 , is converted by concentrated sulphuric acid into nitroguanidine, NH : C(NH 2 )(NH-NO 2 ), and this, on reduction, gives amino- guanidine, NH : C(NH 2 )(NH-NH 2 ). The latter gives hydrazine, (NH 2 ) 2 , NH 3 , and CO 2 011 hydrolysis with acid or alkali, whilst with nitrous acid it yields Diazoguanidine (iminocarbamideazide), NH : C(NH 2 )-N 3 , which is resolved by alkali into hydrazoic acid (see vol. i, p. 327) and cyanamide. From aminoguanidine can be obtained Azodicarbonamide, NH 2 -CO-N : N-CONH 2 , and Hydrazodicarbonamide, NH 2 -CONH-NH-CONH 2 . GLYCOCYAMINE, NH : C(NH 2 )-NH-CH 2 -CO 2 H, is formed by the union of glycocoll .NH-CO with cyanamide, and if water is lost Glycocyamidine, NH : C<^ | , is obtained. If X NH-CH 2 however, instead of glycocoll, its methyl -derivative is taken, sarcosine, CO 2 H-CH 2 - NH-CH 3 (melts at 115 and is neutral), results, creatine and creatinine being obtained similarly. CREATINE, NH : C(NH 2 )-N(CH 3 )-CH 2 -CO 2 H, is obtained from meat-extract, being a stable component of muscle. It has a neutral reaction and is soluble in water and sparingly so in alcohol ; it crystallises with 1H 2 O, has a bitter taste and, when heated with acid, loses 1 mol. of water of constitution and forms creatinine ; on complete hydrolysis, it gives urea and sarcosine. NH -- CO CREATININE, NH : C\ j is a weak base and dissolves very readily in X N(CH 3 )-CH 2 water, giving creatine again. It is one of the constituents of urine and forms a characteristic zinc salt, 2 mols. of creatinine combining with 1 mol. of ZnCl 2 . When hydrolyscd, it gives ammonia and methylhydantoin. URIC ACID AND ITS DERIVATIVES When the two amino -groups of urea condense with the two carboxyl groups of a dibasic acid with expulsion of 2 mols. of water, ureides (see above) xNH-CO are obtained. Thus oxalic acid yields parabanic acid, C0CH 2 ; tartroriic acid, dialluric - acid, and mesoxalic acid, alloxan, C0CO. If, however, only one molecule of water is eliminated, one amino- and one carboxyl-group remaining unchanged, uroacids are obtained, e.g. oxaluric acid, NH 2 -CONH-CO-C0 2 H, and alloxanic acid (from mesoxalic acid). These ureides are usually well crystallised, and are aminic and also markedly acid in character. On hydrolysis, they give first the corresponding uroacid and then urea and free acid. They are sometimes formed on oxida- tion of diureides (see below) ; thus parabanic acid is obtained by oxidising uric acid with nitric acid. The alcohol-acids and the aldehydo-acids also give such condensations (see above), yielding, for example, Jiydantoin, Tiydantoic acid, and allanturic acid (from glyoxylic acid). THEURICACIDGROUP 367 When 2 raols. of urea take part in the condensation, diureides are obtained, these forming the uric acid group [ (5) C NIL | || )CO (8) (3) NH -- C NH X (4) O) and its derivatives : xanthine, caffeine, theobromine, guanine, hypoxanthine, alloxanthine, purpuric acid, allantoin, &c. The positions of the substituent groups are indicated by the bracketed numbers shown in the above formula for uric acid. During recent years successful attempts have been made, by means of ethyl cyanoacetate, to synthesise all these xanthine bases and to convert them, one into the other (see Berichte der deutsch. chem. Gesells., 1899, 32, p. 435). As a general rule, the ureides and diureides have a more or less marked acid character and, as they contain no carboxyl group, this acidity is explained as due to the existence of these compounds in tautomeric forms, just as is CH 2 C(k the case with succinimide, \ /NH. In the latter, it is assumed that CH 2 CCT the iminic hydrogen atom is very mobile and undergoes displacement and union with the oxygen of the neighbouring carbonyl group, a double linking between carbon and nitrogen being formed and an acid hydroxyl group capable of forming salts with metals. The tautomeric formula of Succinimide OH 2 -C(OHk /N:OOH would hence be | /N, and that of Parabanic Acid C0\ I CH 2 -- COT N N:C-OH similar formula? hold for uric acid and barbituric acid, the latter functioning as a dibasic acid (in this case, however, the acid character is perhaps to be attributed to the hydrogen of the methylene group, CH 2 ). Several diureides are found in nature, e.g. in guano, in the urine and muscles of carnivora, in the excreta of serpents, in articular concretions, and in certain plants (theobromine in cocoa, caffeine, &c). The constitutional formulae of the more important diureides are as follow : N(CH 8 )-CO /NH C0< I C0( >CH 0< | \N(CHe)-CO \NH- CO/ C CO Dimethylparabanic acid Methyluracyl Alloxanthine (cholestrophane) ~ NH CH NIL CQ - c NH \ / X NH 2 CO NH/ Murexlde Allantoin N C N v N C N . II II >o II II >ci CH C NH/ CC1 C NH/ N = CH N = CC1 Purine Trichloropurinc! 368 ORGANIC CHEMISTRY CH 3 -N CO CH 3 -N C N v II I II >CH CO CH-NH-CO-NH 2 CO C NH/ II II CH 3 -N-CO CH 3 -N-CO Dimethylpscudouric acid Theophylline CH 3 -N-C N. N C N v NH-C N. I II >H || || )CH | || \CH CO C NfCH.,)/ CH C NH/ CO C NH/ CH 3 -N CO NH CO NH CO Caffeine Hypoxanthine Xanthine N C N v NH C N. II II >H | || >CH CH C NH/ NH:C C NH/ N=C-NH 2 . NH CO Adenine Guanine URIC ACID, C 5 H 4 3 N4. Syntheses of uric acid are many and various ; from cyano- acetic acid, or glycocoll or isodialuric acid, by heating with urea ; from aminobarbituric acid and potassium cyanate ; from ethyl acetoacetate and urea, passing through methyl- uracyl, nitrouracyl, hydroxyuracyl, and isodialuric acid. The following scheme represents the various steps of the synthesis from malonic acid : * NH 2 CO- OH NH CO NH CO CO + CH 2 > CO CH 2 > CO C-.N-OH > I I I I ' II NH 2 CO -OH NH CO NH CO Urea Malonic acid Barbituric acid Violuric acid NH OC NH CO NH CO II II II CO CH-NH 2 > CO CH NH X > CO C NH V II II >o | || >co NH CO NH CO NH 2 / NH C NH/ Uramil Pseudouric acid Uric acid Uric acid is a feeble dibasic acid (see above), and forms a white, amorphous substance insoluble in alcohol or ether and almost insoluble in water. It dissolves, however, in concentrated sulphuric acid, from which it separates unchanged on dilution with water. It is extracted from guano, the excrements of serpents, and the urine of carnivora. Evaporation of uric acid with dilute nitric acid and treatment of the residue with ammonia yields murexide, which forms yellowish green crystals which give a purple aqueous solution, turning blue on addition of alkali (characteristic reaction for uric acid). THEOBROMINE (3 : 7-Dimethyl-2 : 6-dioxypurine), C 7 H 8 O 2 N 4 or CH 3 -N C N v I II ^TTT II / CH ' CO C N(CH 3 )/ NH CO is extracted from cocoa, 2 and forms white, bitter-tasting crystals, slightly soluble in water 1 The constitution ofwic add was demonstrated first by Medicus, ajid later, by various syntheses, by E Fischer (Liebig's Annalen, 1882, 215, p. 253). The presence of a chain, C C C , and of a carbonic acid residue is shown by the formation of urea and alloxan when uric acid is treated in the cold with nitric acid. The presence ot four imino-groups is deduced from the fact that, by introduction of four methyl groups and subsequent hydro- lysis, the four atoms of nitrogen are eliminated as methylamine. A large part of the uric acid molecule is rendered evident by the formation of allantoin (of known constitution) on oxidation with alkaline permanganate, and by the formation of methylurea and methylalloxan on oxidation of dimethyluric acid. * Cocoa and Chocolate. Cocoa is placed on the market in the form of large violet seeds of Theobroma cacao, which grows well in the Antilles. Mexico, Guatemala, Java, Borneo, Esmerelda (equator), &c. The red or brown ESTERS 369 or alcohol. It behaves as a weak acid and a weak base. With methyl iodide, the silver salt yields caffeine. CAFFEINE, 1 C 8 H 10 2 N 4 + H 2 O (constitution given above), is identical with theine 2 ; it forms shining nredles which readily sublime, are sparingly soluble in alcohol and in water, and have a rather bitter taste. Synthetically it is obtained, for example, from ethyl cyanoacetate or malonic acid and dimethylurea. GUANINE, C 5 H 5 ON 5 (constitution given above), is a di-acid base, but forms salts also with bases. It is a white powder insoluble in water but soluble in ammonia. On oxidation with potassium chlorate and hydrochloric acid, it gives carbon dioxide, para- banic acid, and guanidine. XANTHINE, C 6 H4O 2 N4 (constitution given above), is a white powder and acts as both acid and base. It is obtained from guanine by the action of nitrous acid, and its lead salt reacts with methyl iodide, giving theobromine. ADENINE, C 6 H 5 N 6 (constitution given above), forms shining needles and is a base occurring in tea and in ox-pancreas. It is formed by decomposition of the nuclein of the cell-nuclei and is hence of physiological importance. VIII. ESTERS (Oils, Fats, Waxes, Candles, Soaps) The compounds or esters formed by alcohols with inorganic acids have already been studied (see p. 196), and we shall now consider the esters resulting mature fruits resemble cucumbers, each containing 50 to 60 seeds like beans. Tbe seeds are separated from tbe pulp, heaped in casks for 4 to 5 days to initiate the fermentation which increases the perfume, and then dried in the sun. The chemical composition differs considerably with the variety : fatty substance (cocoa-butter), 35 to 45 per cent. ; proteins, 3 to 18 per cent. ; cellulose, 3 to 25 per cent. ; gums and starch, 3 to 15 per cent. ; ash, 3 to 4 per cent. Cocoa-butter (or cacao-butter) is extracted by pressing the seeda hot, and forms a faintly yellow mass of pleasing odour j it melts at 29 to 31, and contains the glycerides of arachic, palmitic, oleic, stearic, and lauric acids. In the manufacture of chocolate, the seeds are washed in suitable sieves and then gently and cautiously heated for 30 to 40 minutes to facilitate skinning. They are next crushed in mortars or rotating cylinders, the flour obtained being made into a paste with sugar and is worked for a long time on stone rollers, different ingredients and flavouring matters being added to give the different kinds of chocolate ; the homogeneous paste then passes to the moulds. Good chocolate contains from 40 to 60 per cent, of cocoa, the rest being sugar ; ordinary qualities contain 10 to 15 per cent, of starch. Italy imported 12,000 quintals of cocoa in 1901, 93,000 in 1908, and 24,000 (including 5200 quintals in powder or paste), worth 185,200, in 1910, in addition to 3380 quintals of cocoa-butter of the value of 48,680. Cocoa costs about 8 per quintal, and pays an import duty of 4 (in Italy). The imports of chocolate into Italy were 8000 quintals in 1901 ; 10,900 in 1908 ; 15,000, worth 180,000, in 1910 ; the import duty being 5 4*. per quintal. Italy exports, on the average, 2300 quintals of chocolate, of the value 28,000, per annum. 1 Coffee consists of the seeds of one of the Rubiacese (Coffea arabica), which grows spontaneously in Southern Ethiopia and Arabia, and is cultivated on an enormous scale in India, the Antilles, Madagascar, and South America. It is an evergreen plant, 6 to 9 metres high and of pyramidal habit, with greyish branches and lanceolated leaves, the flowers (at the base of the leaf) being white and pleasant-smelling, like jessamine. The fruit forms drupes like cherries, the epicarp passing from yellow to green to red to brownish, and the mesocarp being yellow and of agreeable taste. The endocarp is divided into two compartments surrounded by coriaceous membrane and each containing a seed, which has one convex and one flat, furrowed face, and is covered by a friable pellicle the endosperm (albumen) is yellowish or greenish and horny. The form of the seed varies with the kind of the coffee (Coffea mauritiana, laurina, llberica, &c.). Moclu coffee berries are small and the Australian ones large, whilst those from the Antilles are intermediate in size. The cultivation of coffee has received a considerable impulse in Brazil, where as much as 400,000 tons (almost half the total production of the world) are now produced. Of the Antilles coffees, the most highly valued is that from Porto Rico. Coffee berries are composed of celluloses (18 per cent.), fatty matters (12 per cent.), gummy and saccharine substances (10 per cent.), nitrogenous compounds (12 per cent.), mineral salts (4 to 5 per cent.), a tannin (caffe- tannic acid, 8 per cent.), caffeine (0-8 to 1-3 per cent.), caffearine and water (11 per cent.) . When roasted, coffee develops aroma and loses 15 to 20 per cent, in weight, but increases in volume by one-third, while the sugar caramelises and the cellulose carbonises partially, forming a brown oil which is denser than water, dissolves in ether, and constitutes the aromatic substance (caffeone). Boasted coffee contains, on the average, 1-5 per cent, of water, 13 per cent, of nitrogenous substances, 0-8 per cent, of sugars, 13-5 per cent, of fats, 4-8 per cent, of ash, 0-9 per cent, of caffeine, and 46 per cent, of non-nitrogenous substances ; to hot water this coffee gives up about 25 per cent, of its weight. In 1900 Italy imported 141,000 quintals of coffee ; in 1908, 227,600 quintals ; and in 1910, 253,000 quintals (about four-fifths from Brazil), of the value of 1,062,080 ; the former Customs duty of 6 per quintal was lowered slightly in 1909. Tea is an evergreen shrub, Thea chinensis (order Ternstrcemiaceae), cultivated in China, Japan, British India, Java, Ceylon, and Brazil. The leaves (similar to those of the white willow) are twisted, dried, and folded, prior to rapid immersion in boiling water and drying on heated plates (4 kilos of leaves yield 1 kilo of tea). Com- mercially the numerous varieties are grouped into three types : green, black, and scented, the last two being slightly fermented. The active alkaloid is Theine (see above). The best infusion of tea is obtained by macerating for thirty minutes in cold water and then adding boiling water, the liquid being poured off before it becomes very brown and excessively rich in tannin (20 grms. of tea per litre of slightly hard water). France imported 475,000 kilos of tea in 1882 and 1,160,000 kilos in 1906. Italy imported 69,000 kilos in 1908 and 73,600 kilos, of the value of 11,120, in 1910, the Customs duty being 10 per quintal, II 24 370 ORGANIC CHEMISTRY from the substitution of the typical hydrogen of organic acids by alkyl radicals. Various isomerides exist with these compounds, e.g. methyl butyrate is isomeric with ethyl propionate, butyl formate with propyl acetate. Attention will, however, more especially be paid to the esters of glycerine (glycerides), since on these are based the fat, oil, soap, and candle industries. PREPARATION. These esters can be obtained by the general methods already described (loc. cit.), e.g. by the action of the acid chlorides or anhydrides on the alcohols or sodium alkoxides : C 2 H 5 -CO-C1 + C 2 H 5 -OH = HC1 + C 2 H 5 -COOC 2 H 5 . They are also formed by the interaction of the silver salt of the acid and the alkyl iodide, and by the action of gaseous hydrogen chloride on a hot alcoholic solution of the nitrile of the acid. Further, the alcohols and acids themselves react, slowly in the cold and more rapidly although not completely in the hot, with formation of esters : C 2 H 5 -OH + CH 3 -C0 2 H = CH 3 -C0 2 C 2 H 5 + H 2 0. In practice the preparation is carried out as follows : the dry organic acid is mixed with an excess of absolute alcohol and the mixture saturated with dry hydrogen chloride gas, left for some time in a moderately warm place and then poured into water ; the ester separates in an insoluble form after neutralisation of the aqueous liquid with alkali in the cold. In this reaction the acid chloride is probably formed as an intermediate product: CH 3 -C0 2 H + HC1 = H 2 + CH 3 -COC1 ; the latter which with water might give the reverse reaction being in presence of excess of the alcohol, forms the ester (equation given above). But esterification is never complete, the reaction being a reversible one : C w H 2n+1 -OH + C n H 2n 2 ',. > H 2 + C w H 2n _ 1 2 -C w H 2n+1 . After a certain time a system is obtained which contains given quantities of alcohol (a), acid (b), water and ester (z). The same equilibrium is attained by mixing 1 mol. of ester and 1 mol. of water as by mixing 1 mol. of acid and 1 mol. of alcohol, and this equilibrium is represented by the following equation for bimolecular reactions (see vol. i, p. 67) : k(a z)(b z) = ^z 2 , where a and b represent the respective initial concentra- tions of alcohol and ester and z that of the ester, and water when equilibrium is reached, all expressed in mols. (gram-molecules) ; k and Tc^ are constants depending on the nature of the reaction and, according to a definite law, slightly on the temperature. If, for convenience, -r is made equal to K, the equation becomes : (a z)(b z) = Kz z . With 46 grms. of alcohol and 60 grms. of acetic acid (gram-molecules), it is found experimentally that K = 0-25, and, as a and b both assume the value 1, 1 mol. of each reacting, the equation becomes (1 z) 2 =0-25z 2 , i.e. 1 z0-5z or z = f . This means that when a state of equilibrium is reached, the system contains J mol. of acetic acid + mol. of alcohol + f mol. of ester + f mol. of water. Every substance partici- pating in the equilibrium acts in proportion to its mass. If the above equation is given ft ___ 2 5* the form = K , it becomes evident that, in order to displace the equilibrium z b z so as to have a greater value of z (i.e. of esterification), the value of a must be increased and that of 6 decreased, esterification being complete when a = oo. The same final result is o'btained when b is much greater than a, esterification again being complete when b = oo. In practice, almost complete esterification is attained when 1 mol. of acid is employed per 10 mols. of alcohol or vice versa. That the game result is obtained with excess of alcohol as with excess of acid is shown by the above equation, since, if instead of TO mols. of both acid and alcohol, n times as many molecules of acid are taken, the equation becomes : = K ; whilst if n times as many molecules of alcohol are PREPARATION OF ESTERS Vi Wi 2 taken, it becomes : - = K - But these two equations are identical, multiplica- tion of the terms of the former bv giving the latter. m z The limit of esterification is modified but slightly by change of temperature and amounts, in the case of acetic acid, to 62-2 per cent, at 10 and to 66-5 per cent, at 220. The esters of monohydric alcohols and monobasic fatty acids are neutral liquids lighter than water (0-8 to 0-9) and pleasant smelling (some forming artificial fruit essences) ; they are slightly soluble in water (the first members more soluble than the higher ones) and they boil undecomposed. By means of Grignard's reaction (see p. 203), they yield tertiary alcohols. The esters are hydrolysed into their components when heated with alkali, mineral acid, or aluminium chloride, or superheated with water. The mineral acid has a purely catalytic accelerating action on the following reaction due to the water, which is very slow in its action : CH 3 -C0 2 C 2 H 5 + H 2 = C 2 H 5 -OH .+ CH 3 -C0 2 H. With bases, the hydrolysis is expressed by the equation : CH 3 -C0 2 C 2 H 5 + NaOH = C 2 H 5 -OH + CH 3 -C0 2 Na. The hydrolysing velocity of acids and bases depends on their degrees of dissociation, i.e. on their strengths, so that feeble acids and bases hydrolyse far more slowly than the strong ones. In the case of acids, the hydrolysis is caused by the hydrogen ions, and in that of bases by the hydro xyl ions. In the latter instance, however, the velocity of hydrolysis is greater than with acids, and with methyl acetate, the value of K for decinormal potassium hydroxide is 1350 times that for decinormal hydrochloric acid. In the hydro- lysis of fats, the acids of which are feeble and the resultant salts therefore hydrolytically dissociated to a marked extent (i.e. even with excess of fatty acid, there always remains free base or hydroxyl ions) complete hydrolysis is obtained industrially with a quantity of base (e.g. lime) much lass than that required theoretically. As has been already mentioned, the first ethers of the monobasic acids and monohydric alcohols are, in general, substances of pleasing odour and are used with suitable admixtures as artificial fruit essences. 1 ETHYL FORMATE, H-COOC 2 H 5 , boils at 55 and is used for artificial rum or arrack. ETHYL ACETATE, or Acetic Ester, CH 3 -COOC 2 H 5 , is used in medicine and for the preparation of ethyl acetoacetate, which is of considerable import- ance in organic syntheses. It is prepared by heating alcohol with acetic and sulphuric acids under the conditions given above. It boils at 77 and has the sp. gr. 0-9238 at 0. Methyl Acetate boils at 57-5 and has the sp. gr. 0-9577. AMYL ACETATE, CH 3 -COOC 5 H n , is used in alcoholic solution as essence of pears. It boils at 148. 1 Commercial fruit essences are prepared from the following mixtures of esters, and cost from 2. 6d. to 5. per kilo : Essence of -pineapple : 25 grms. ethyl butyrate + 135 grms. amyl valerate + 5 grms. chloroform + 5 grms. aldehyde + 850 grms. alcohol. Essence of apples : 50 grms. ethyl nitrite + 50 grms. ethyl acetate + 100 grms. amyl valerate + 40 grms. glycerol + 7-5 grms. aldehyde + 7-5 grms. chloroform + 745 grms. alcohol. Essence of pears: 200 grms. amyl acetate + 50 grms. ethyl acetate + 100 grms. ethyl nitrite -f'20 grms. glycerol + 630 grms. alcohol. Essence of apricots : 35 grms. benzaldehyde + 190 grms. amyl butyrate + 10 grms. chloroform + 765 grms alcohol. Essence of strawberries : 27 grms. amyl acetate + 18 grins, amyl valerate + 9 grms. amyl butyrate + 9 grms. amyl formate + 15 grms. ethyl acetate + 7 grms. essence of violets + 915 grms. alcohol. Essence of peaches : 100 grms. amyl valerate + 100 grms. amyl butyrate -\- 20 grms. ethyl acetate -f- 10 grma. benzaldehyde + 770 grms. alcohol. ETHYL BUTYRATE, C 3 H 7 - COOC 2 H 5 , boils at 121 and is used as essence of pineapple and in rum. ISOAMYL ISOVALERATE, C 4 H 9 -COOC 5 H n , boils at 194 and is used in essence of apples. The higher esters form constituents of waxes (Cetyl Palmitate, C 16 H 31 2 C 16 H 3 3 ; Melissyl Palmitate, C 16 H 3] 2 C 30 H 61 ; Ceryl Cerotate, C 26 H 51 2 C 26 H 53 , &c.) These higher esters distil unchanged only in a vacuum ; under ordinary pressure they decompose into olefines and fatty acids. Esters of Polybasic Acids are prepare d by the general methods described above ; acid esters are obtainable if one or more of the carboxyl groups are not esterified. The esters of oxalic acid are obtained, for instance, by heating anhydrous oxalic acid with alcohols, the normal ester being separated from the acid ester by fractional distillation. The importance of Malonic Esters in organic syntheses has already been illustrated on pp. 308 et seq. ; the normal methyl ester boils at 181 and the ethyl at 198 (sp. gr. 1-068 at 18). The two hydrogen atoms united with the middle carbon atom can also be replaced by alkyl groups. Thus, for example, Ethyl Dimethylmalonate, (C 2 H 5 - CO 2 ) : C(CH 3 ) 2 , is obtained from the sodium derivative by treatment with methyl iodide. These compounds, when heated, lose C0 2 and yield alkylacetic derivatives. Similar relations are found with the alkyl derivatives of succinic acid or esters. The preparation of Ethyl Acetoacetate and its importance in organic syntheses have been dealt with on p. 332. The Normal Methyl Ester of succinic acid, CH 3 -CO 2 -CH 2 -CH 2 -C0 2 -CH 3 , melts at 19 and boils at 80 under 10 mm. pressure ; the ethyl ester boils at 216. GLYCERIDES, OILS, FATS Glycerol being a trihydric alcohol, its three alcoholic groups may be partially or wholly esterified by acid residues. It suffices, indeed, to heat glycerol with fatty acids to obtain mono-, di-, and tri-glycerides. These glycerides are also formed by the action of the tissues of the pancreas on a mixture of oleic acid and glycerol, a still better method for synthesising fats being the treatment of the sulphuric ethers of glycerol with fatty acids dis- solved in concentrated sulphuric acid. Most fats and oils are formed of triglycerides, which, according to the nature of the fatty acid saturating the three alcoholic groups of the glycerol, are termed Tripalmitin (melts at 60), Tristearin (melts first at 55 and, after resolidification, at 71-6), and Triolein (liquid, solidifying at about 0). Triolein, which is the principal component of liquid fats and especially of olive oil, is formed by the esterification of the glycerol molecule with 3 mols. of oleic acid (see p. 298) : CH 2 -OO-C 18 H 33 CH-00-C 18 H 33 CH 2 -0-0-C 18 H 33V Mono- and di-glycerides are not found in the fats (only ravison oil contains a diglyceride, dicrucin ; see also esters of polyhydric alcohols and glycerol with mineral acids, pp. 213, 222 et seq.}. Certain fats (butter, cocoa-butter) contain mixed triglycerides, i.e. with different acid radicals, some of them being of acids of low molecular weights, REICHERT AND HEHNER NUMBERS 373 soluble in water. 1 A. Griin (1906-1909) synthesised mixed glycerides con- taining three acid residues, all different. 2 The most simple glyceride is Triformin, C 3 H 5 (C0 2 H) 3 , which was obtained crystalline by P. van Romburgh (1910) by protracted heating of glycerol with 100 per cent, formic acid ; it crystallises with difficulty, melts at 18, boils at 266 (762 mm. pressure), and at 210, under ordinary pressure, decomposes. It is hydrolysed slowly by cold water, rapidly by hot. Oils and fats have coefficients of expansion greater than those of other liquids (100 litres of olein at become 101-6 at 20). Fats and, still more, waxes contain also non-glyceride components, e.g. Cetyl Alcohol, C 16 H 34 0, which, as such or as palmitic ester, forms one of the principal constituents of spermaceti fat. Cerotic Acid, C 27 H 52 O 2 , and its ester occur in large proportions in wax. Non-hydrolysable substances (cholesterol, phytosterol, isocholesterol, aromatic alcohols, &c.) are always found in small quantities in fats (olive oil, about 0-75 per cent. ; ravison oil, 1 per cent. ; cotton-seed oil, 1-6 per cent. ; lard, 0-25 per cent. ; cod liver oil, 0-5 to 3 per 1 Volatile fatly acids soluble in water. The number of c.c. of decinormal potassium hydroxide solution required to neutralise the volatile fatty acids soluble in water from 5 grms. of the fat, constitutes the so-called Reichert- Meissl-Wollny number and serves to ascertain the purity of certain fats, especially of butter. The deter- mination is made as follows : exactly 5 grms. of the fat (melted at a low temperature and rapidly filtered) are heated in a flask of about 350 c.c. capacity with 10 c.c. of alcoholic potash (20 grms. of KOH in 100 c.c. 70 per cent, alcohol) on a water-bath with frequent shaking until almost all the alcohol is evaporated ; the remainder of the alcohol is completely expelled by shaking the flask and introducing a current of air every half- minute. After about twenty minutes, when the smell of alcohol is no longer detectable, 100 c.c. of distilled water are added, the heating being continued until a clear solution is obtained (if the liquid does not become clear the test must be commenced anew, hydrolysis being incomplete). To the tepid solution are then added 40 c.c. of dilute sul- phuric acid (1 vol. cone. H SO 4 + 10 vols. water) and a few fragments of pumice, the flask being then placed on a double wire-gauze and the liquid distilled, the dimensions of the apparatus being shown in mm. in Fig. 248. In about half an hour, exactly 110 c.c. of liquid distil over ; this is mixed and filtered through a dry filter, 100 c.c. of the filtrate being titrated with decinormal KOH solution in presence of phenolphthalein. The volume of the alkali used is increased by one-tenth of its value (the volume of the distillate being 110 c.c.) and diminished by the number of c.c. of the alkali obtained from a control ex- periment made without fat as a check on the reagents employed. The result is the Reichert- Meissl-Wollny number. At the present time many laboratories employ the Leffmann- Beam-1'olenske method, which effects more rapid hydrolysis (see later, Butter). For butter the limits for this number allowed by law are 26 to 31-5 (Municipal Laboratory of Milan), the butter being suspected if it gives a value of 22 to 23, although the butter of certain districts and from certain animals may, in exceptional cases, give a number as low as 21. The value for rancid butter, even two months old, is only slightly lower (by about 2) than the normal. Insoluble fatty acids. The quantity of fatty acid insoluble in water obtainable from 100 parts of fat is called the Hehner number, and is determined as follows : into a flask of about 200 c.c. capacity are dropped, from a weighed vessel containing the dry filtered fat, 3 to 4 grms. of the substance, the vessel being then reweighed exactly. After addition of 50 c.c. of alcohol and 1 to 2 grms. of KOH, the flask is heated on a water-bath for five minutes, a clear solution being obtained. If the addition of a drop of water produces turbidity, saponification is incomplete, and the heating is continued for a further period of five minutes, the liquid being then tested as before. Evaporation is then continued until there remains a dense mass, which is taken up in 100 to 150 c.c. of water, acidified with dilute sulphuric acid, and heated until the clear fatty acids float on the surface. The liquid is then poured on a dry, tared filter (about 12 cm. in diameter and in a funnel either without a neck or with a very short one), previously half filled with hot water. The acids are washed with boiling water until the washing water ceases to show an acid reaction (as much as 2 litres of water are sometimes required). The filter is then cooled in a beaker of water so that the fatty acids solidify. The filter is then detached from the filter and introduced, with the acids, into a tared beaker, which is heated in an oven at 100 to 102 until its weight remains almost constant (difference between two weighings less than 1 mgrm.). The weight of fatty acids, referred to 100 parts of fat, represents the Hehner number. Unadulterated fats generally have Hehner numbers of 95 to 97 (for butter it is 87-5 ; for coco-nut oil, 85 to 92 ; for palm oil, 91). 1 The synthesis of triolein has been applied practically by G. Gianoli (1891) to diminish the ranc'dity of oils, especially of olive oil obtained from the husks by means of carbon disulphide. This oil contains 20 to 30 per cent., or even more, of oleic acid, and is heated in an autoclave with the corresponding quantity of glyc-erol (or even a slight excess) at 250 in a slow stream of CO 2 , or in a vacuum with a trace of oxalic acid to facilitate mixing of the liquids and avoid blackening of the mass owing to the presence of hydroxy-acids ; the distillation of the water formed in the reaction is hastened by adding fragments of tin to the mass. This procedure yields a neutral or almost neutral oil with an iodine number less than 75 and a marked viscosity, so that it can be used even for mixing with lubricating oils. Bellucci (1911) also achieved an almost quantitative synthesis by heating together the theoretical proportions of glycerol (1 mol.) and fatty acid (3 mols.) at 180 to 260 for two hours in a vacuum, so as to expel the water formed, which would otherwise produce the reverse reaction ; in a current of CO 2 , the same reaction takes place at the ordinary pressure. A. Walter (1911) obtained a mixture of tri- and di-oloins by treating glycerol and acetic acid in presence of the enzymes of castor oil seeds, which act as catalysts. Indeed, catalysts cause reversible reactions, and while in presence of water the enzymes of castor oil seeds hydrolyse fats (see p. 409) with formation of glycerol and fatty acids, if water is excluded as much as 35 per cent, of tne fatty acids can be converted into glycerides. FIG. 248. 374 ORGANIC CHEMISTRY cent. ; tallow, 0-02 to 0-6 per cent. ; bone-fat, 0-4 to 2-4 per cent. ; wool-fat more than 7 per cent.). The oils of cereals and of Leguminosece contain abundant amounts of LECITHIN, (C ls H S5 2 ) 2 C s H 5 -PO ll <^ r nw 5 wn ih is decomposed by the enzyme of the pancreas or castor oil seed, but not by that of the blood (serum-lipase) . The fat of peas contains 1-17 per cent, of phosphorus or 30-4 per cent, of lecithin, and that of wheat, 0-25 per cent, of phosphorus or 6-5 of lecithin ; the amount of lecithin is obtained by multi- plying that of phosphorus by 26. Fresh fats and oils contain minimal proportions of free fatty acids (less than 1 per cent.), these increasing with lapse of time, especially if the fats are not melted. This rancidity is facilitated by sunlight and also by the protein sub- stances of unrefined fats and oils. Coco-nut oil does not readily turn rancid, but with olive oil the proportion of free oleic acid reaches 25 per cent., and with palm oil as much as 70 per cent, of free acids may be formed. The taste and smell of fats depend, not on the glycerides, but on other substances. The specific gravity of oils and fats varies from 0-875 to 0-970 (see Table given later) and is determined by means of an aerometer or Westphal balance (see vol. i, p. 73). They are almost completely insoluble in water, acetone, or cold alcohol (this dissolves a certain amount of castor oil and of olive- kernel oil). The solubility increases in boiling alcohol and is complete in ether, chloroform, carbon disulphide or tetrachloride, petroleum or petroleum ether (in the last two, castor oil is slightly soluble, while ether dissolves a little pure tristearin). 1 When heated on a spatula held some distance above a flame, all fats give greenish flames owing to the presence of carbon monoxide and sodium ; also all fats are blackened by osmium tetroxide (sensitive reaction). Oils dissolve small quantities of sulphur or phosphorus and larger quan- tities of soaps even when they are dissolved in ether or petroleum ether. The oxygen of the air exerts a marked and rapid influence, as it is fixed by the drying oils (linseed, walnut, hemp-seed, poppy- seed, &c.), which are thus transformed into varnishes, this occur- ring more readily if the oils are boiled with oxide of lead or of manganese. With the other non-drying oils, the air (together with light) gradually causes rancidity, which, however, some attribute to the action of bacteria, or rather to hydrolysing and oxidising enzymes ; however this may be, the acidity increases owing' to formatioli of butyric, caproic, oleic, &c., acids, but the rancid taste and smell are due more especially to the formation of aldehydic, ,* ketonic, and ethereal substances, hydroxy-acids, and volatile ;<^ acids which can be eliminated by repeated washing with dilute * solution of alkali and subsequently of bisulphite (for the alde- hydes and ketones, see later, Renovated Butter). 1 To determine the quantity of fat contained in any solid substance, a weighed portion of the latter in a finely divided, dry state (5 to 15 grins, are taken and, if pasty, mixed with fragments of pumice) is introduced into a filter-paper cartridge situate in a Soxhlet apparatus (Fig. 249). The Soxhlet apparatus is connected at the bottom with a tared flask resting on a water- bath, and at the top with a reflux condenser. From 100 to 150 c.c. of petroleum ether or ether are then added and extraction continued for 2 to 4 hours in such a way that the FlG. 249. solvent siphons over 15 or 20 times per hour. A calcium chloride tube may be attached to the extremity of the condenser to prevent access of moisture from the air. The solvent is afterwards evaporated from the flask and the residual fat dried at 100 to 102 until almost constant in weight. ILc difference between the weight of fat and that of the original substance gives the solids not fat. IODINE NUMBER: REFRACTIVE INDEX 375 When fats turn rancid, the iodine number 1 is lowered and the index of refraction? the dropping or melting point (see pp. 5 and 16), and the acetyl number (see p. 189) rise. In butter rancidity is facilitated by the presence of the casein and milk-sugar, which give rise to other decompositions. Although not rigorously exact, the degree of rancidity is expressed by the number of c.c. of normal potash necessary to neutralise 100 grins, of the fat. A butter with 10 of rancidity should be rejected. The free fatty acids in fats and oils are usually determined with a decinormal alkali solution, 5 to 10 grms. of the fat being dissolved in 50 to 60 c.c. of a perfectly neutral mixture of alcohol and ether (1:2) and phenolphthalein being used as indicator. The acid number gives the number of mgrms. of KOH necessary to neutralise 1 grm. of fat. By passing a current of air through oils heated to 70 to 120, the so-called blown or oxidised oils, rich in triglycerides of hydroxy-acids, are obtained. These are dark in colour and have the density of castor oil (but are soluble in petroleum ether), but if " blown " in the cold for a longer time, they are 1 The Iodine Number is characteristic of a fat (see Table, p. 378), and expresses the percentage of iodine absorbed by the fat (i.e. by its unsaturated components, e.g. oleic acid or the corresponding glycerides, two atoms or iodine being fixed for each double linking, see p. 87). This determination requires : (1) An iodine solution obtained by mixing, 48 hours before using, equal volumes of the two- following solutions : (a) 25 grms. of iodine in 500 c.c. of pure 95 per cent, alcohol, and (6) 30 grms. of mercuric chloride in 500 c.c. of pure 95 per cent, alcohol ; (2) a sodium thiosulphate solution, prepared by dissolving 24 grms. of the pure salt in a litre of water, the titre in iodine being ascertained as follows : 3-8657 grms. of pure, dry potassium dichromate are dissolved in water at 15 and the solution made up to a litre ; exactly 20 c.c. of this solution are introduced into a flask with a ground stopper, about 15 c.c. of a 10 per cent, potassium iodide solution (free from hydroxide) being added and then 5 c.c. of concentrated hydrochloric acid. This procedure results in the liberation of exactly 0-2 grm. of iodine. The thiosulphate solution is run into this from a burette until the solution is only faintly yellow. A few drops of fresh starch-paste are then added and addition of the thio- sulphate continued until the blue colour disappears. It is thus found how much iodine corresponds with 1 c.c. of thiosulphate solution, the strength of which remains constant for several months. The iodine number is determined by dissolving a known weight of the fat or oil (0-2 to 0-5 grm. or, for drying-oils, 0-1 to 0-12 grm.), in a 500 to 800 c.c. flask with a ground stopper, in 15 c.c of pure chloroform and adding 25 c.c. of the iodine solution (prepared forty-eight hours previously, as stated above) ; if, after two hours, the liquid is no longer very brown, a further measured volume of iodine solution is added and the whole left in the dark. After six hours the excess of iodine left unabsorbed by the fat is determined by adding 20 c.c. of a 10 per cent. KI solution, diluting with 150 c.c. of water, and add- ing more KI if the reddish brown solution is not clear. The excess of iodine is then titrated with the thiosulphate solution in the manner already described. Immediately afterwards, 25 c.c. of the iodine solution employed are titrated. The difference between the two values thus obtained, expressed as grammes of iodine per 100 grms. of the fat, represents the iodine number. 1 The index of refraction is measured in the Zeiss Butyro- refractometer (Fig. 250), by observing the total reflection of a very thin layer of oil or fat situate between two prisms, p, mounted in the two chambers, A and B (the latter rotates on the hinge, C, so as to squeeze uniformly the film of oil smeared in p ; the screw, F, fixes B against A). Indirect light from the sun or from a powerful sodium lamp is passed through the prisms by means of the mirror, /, and the limit between the light and dark portions of a scale reading from to 100 is read through the eye-piece, K. A thermometer, M, indicates the temperature at which the observation is made, and this temperature can be regulated (so as to melt solid fats) by passing water, at a higher or lower temperature, in at E and through the rubber tube, D, to the outflow, e. The refraction is usually stated in the centesimal degrees of the Zeiss scale, the temperature normally 25 being indicated. Values obtained at other temperatures can be referred to the normal temperature by adding or subtracting 0-55 for each degree above or below 25 (the number 0-55 is accurate for butter, but slightly inexact for other fats). The index of refraction is obtained from the reading on the Zeiss scale by adding to the value 1-4220 as many ten -thousandths as are obtained by multiplying the scale degrees by 7-8 when the reading is between and 30 ; 7-5 if between 30 and 50 ; 7-3 if between 50 and 70 ; and 7-0 if between 70 and 100. (This procedure, too, gives accurate values for butter, but slightly inaccurate ones for other fats). Thus, 30 on the Zeiss scale would corre- 7- 8 spond with a refractive index of 1-4220 + 30 X , nnn/ . = 1-4220 + 6-0234 = 1-4454, which agrees almost 10000 exactly with the true index of refraction (1-4452) ; similarly, 60 on the scale means a refractive index of 7 3 1-4220 + 60 x = 1-4658. Inversely, the scale reading is obtained by subtracting 1-4220 from the refractive index and dividing the remainder by 7-8, 7-5, 7'3, or 7-0. The colour of the line of demarcation on the scale sometimes gives an indication of impurity in the fat, being colourless for pure butter, blue if margarine is present, and orange with admixtures of certain other fats. FiG. 250. 376 ORGANIC CHEMISTRY obtained almost colourless. Blown oils are valued as lubricants. If the blowing is continued, yellow or brown gelatinous masses are obtained. With the exception of the iodine number and the Hehner number which are lowered the chemical and physical constants of blown oils (thickened oils, &c.) are higher than those of the original oils. Oils also fix ozone in proportion to the unsaturated fatty acids they contain, and at the same time become denser (see p. 299) ; olive oil has an ozone number of 15-8 (grms. of ozone fixed per 100 grms. of oil, Fenaroli, 1906) ; maize oil, 21 ; linseed oil, 33 ; and castor oil, 16. Also sulphur is dissolved and combined in amount increasing with the proportion of glycerides of unsaturated acids present, giving very viscous, brown liquids, sometimes almost solid and gummy. Chlorine acts on fats, partly replacing hydrogen and partly combining directly. Iodine is added slowly, but the addition becomes rapid in alcoholic solution and in presence of mercuric chloride (Hiibl). Addition of concentrated Sulphuric Acid to oils results in the development of heat and the evolution of sulphur dioxide ; in the cold, sulphuric ethers of the triglycerides are formed. 1 Dilute Nitric Acid, in the hot, slowly oxidises fats, while the concentrated acid attacks them with evolution of red vapours. Nitrous Acid renders non-drying oils denser and solidifies them, the triolein being converted into trielaidin (see p. 298) ; the drying oils remain liquid, although their specific gravity, viscosity, and saponification number increase, and the iodine number and Hehner number (per cent, of insoluble fatty acids) diminish. When burnt, fats give the characteristic odour of acrolei'n, which is derived from the glycerol. On paper, fats and oils produce a translucent spot, insoluble in water (different from glycerol). All these reactions serve as qualitative and quantitative tests to establish the purity of fatty substances (see later). WAXES. Unlike fats, waxes are usually composed, not of triglycerides, but of esters derived from the higher monohydric alcohols (e.g. cetyl, myricyl, and ceryl alcohols, cholesterol, &c.), and sometimes dihydric alcohols also. They contain, in addition, the high acids (e.g. palmitic, stearic, cerotic, oleic, &c.) and alcohols in the free state. Further, beeswax contains as much as 15 per cent, of high melting-point hydrocarbons. They form homogeneous mixtures in all proportions when fused with fats and give also a greasy spot on paper, but they yield no odour of acrolei'n when burned (unlike fats) and do not become rancid when exposed to the air. The commonest waxes are beeswax, Japanese wax, spermaceti wax (from whales), and carnauba wax (from the leaves of certain palms). Beeswax forms the hexagonal cells of beehives. After the honey has been expressed, the mass is melted with water to remove im- purities ; on cooling, a solid layer of crude wax separates at the surface, and this, after melting and casting into blocks, forms virgin or yellow wax. This is placed on the market in various qualities and colours, some of them being olive-brown ; they bear the name of the place of origin and can be bleached with varying facility. FIG. 251. 1 Maumene 1 found that the rise of temperature produced by sulphuric acid of definite concentration serves to distinguish different fats (see Table given later). This constant (Maumeni' number) is nowadays determined by means of the Tortelli thermo-oleometer (1905). 20 c.c. of the oil are poured into the glass receiver, A (Fig. 251), the jacket of which has been evacuated. The oil is stirred with the thermometer, B, fitted with platinum vanes and the initial temperature read. 5 c.c. of concentrated sulphuric acid (sp. gr. 1-8413 or 66 B6.) are then added from a pipette in thirty seconds, the liquid being kept sHrred as long as the temperature rises. The rise of temperature is the Maumeni- number. If the sulphuric acid has not the density given above, but is allowed to absorb even traces of moisture, discordant results are obtained. HYDROLYSIS OF FATS 377 The European waxes have the following physical and chemical constants, which allow of the detection of the frequent adulteration to which they are subjected : melting-point, 62 to 64 ; solidification point, 60 ; specific gravity at 98 to 100, 0-822-0-847 ; saponi- fication number, 95 to 97 (rarely 88 to 105) ; acidity number, 19 to 22 ; difference between saponification number and acid number (ester number), 74 to 76 ; iodine number, 8 to 11 ; degrees on the Zeiss butyro-refractometer at 40, 44 to 45-5 (rarely 42). Foreign waxes have somewhat different constants. The bleaching of the wax is effected by melting it several times with slightly acidified water, allowing it to cool slowly so as to separate the impurities more thoroughly and then causing it to solidify in thin layers on a cylinder half immersed in water and exposing these to the sun and air for five to six weeks. A more expeditious method of bleaching consists in treatment with hydrogen peroxide or other oxidising agent (dichromate and dilute sulphuric acid), or with animal charcoal. The white wax thus obtained often improved in appearance by the addition of 4 to 5 per cent, of tallow presents almost the same physical and chemical constants as the virgin wax, the iodine number alone being lowered by 1 to 7. The wax is insoluble or only slightly soluble in cold alcohol or ether, but dissolves in the boiling solvents. It dissolves in the cold in chloroform, oil of turpentine, carbon disulphide, or fatty oils. It resists dilute caustic alkalis and concentrated alkali car- bonates. It is used for making candles, waxed cloth and paper, mastics, artificial fruit and flowers, &c. Carnauba Wax is exuded from the leaves of certain palms (Corypha cerifera) of Brazil and Venezuela. In the crude state, it is hard and brittle, and of a yellowish green colour ; it melts at 83 to 88, has an acid number of 4 to 8, a saponification number of 80 to 95, an ester number of 75 to 76, and an iodine number of 7 to 13, and contains more than 50 per cent, of non-hydrolysable substances. It is used for the manufacture of candles and, mixed with potash (soft) soap, forms the encaustic with which pavements are cleaned. Japanese Wax is the fat extracted from the fruit of certain Japanese and Chinese trees of the order Terebinthacese (Rhus succedanea, R. vernicifera, and R. sylvestris). It differs from beeswax in having an ester number of about 200 and a saponification number of about 220. It is completely hydrolysable, since it consists of glycerides of palmitic, stcaric, and arachic acids, and contains also 9 to 13 per cent, of free palmitic acid. STATISTICS. The United States imported 2100 tons (142,600) of vegetable wax in 1910 and 2200 tons (199,400) in 1911. In 1900 Italy imported about 1000 quintals of wax, almost all in the raw state, and exported about 1900 quintals of crude yellow and 1100 quintals of white, treated sorts. In 1906, the imports were 1452 quintals ; in 1908, 1015 quintals ; and in 1910, 1070 quintals (of the value of 14,000). In 1906 Germany imported more than 25,000 quintals of wax and exported more than 4000, besides 10,000 of candles, &c. England imported 3350 tons of wax in 1909 and 3070 tons (259,049) in 1910. Yellow beeswax costs up to 15 per quintal, and the bleached wax 17. Hydrolysis (Saponification) of Fats and Waxes. The term saponification is applied to the decomposition of fats into the alcohols and acids composing them, with simultaneous addition of a molecule of water (hydrolysis), by heating with water under pressure at 200 or by the action of acid or alkali (see p. 371) ; when alkali is used, the alkali salt (soap) of the fatty acid and not the free acid itself is obtained : C 3 H 5 (O.OC 18 H 35 ) 3 + 3KOH = C 3 H 5 (OH) 3 + 3C 18 H 35 2 K. Tristeariu Glycerine Potassium stearate The mechanism of the saponification of fats. was for long a matter of con- troversy. Some regarded it as occurring gradually, 1 mol. of fat first reacting with 1 of alkali (bimolecular reaction) (see vol. i, p. 67) and di- and mono -glycerides being formed as intermediate products, whilst, according to others, saponification was a single (tetramolecular) reaction. Only since the investigations of Geitel (1897), Lewkowitsch (1898-1901) and, more especially, Kremann (1906), does it appear to be established with certainty that saponification is gradual, consisting of successive bimolecular reactions, 02 Q hH o H H O ^ i i fi {2; O fc co tf O O fi 525 co H 2 CO H H QQ X O a 3 M I O i i cc Index of refraction at 60 Fatty acids 1 * ( * So 2 J" CM CM 5- CO o m - I~ OS C- CO | CO if 1 ' 1 5 l< H . TP Tr T . . r~ Butyro- refractometer K ( ts 10 o OJ tl 3 in in m o 5 00 00 00 00 --h t~ CM tb 5 to to to I- to to CO CO oo * cs a oj '" "T '? sii l l 5 n M 1 IN -i- ' a (N 5 O '" l ~ o a; 11 & g S to (N OS O -t rH in o 05 50 05 10 to , W IN , ^ 1 rH 00 1-- m^''H"H< 1 ^ 1 to 00 i IN ^ 1 4, M :> 1 Os O5 > Iodine number Fatty acids oo in m co '." co 1 t CO I 1 6 1 II CO ^ 1 *-< ' CO -* ^* K CO .^ j^-oo 10 Jgo'- | oo 00 CM 1 -HrHCOrHOOCO-fCOllosScO r- co ooosoosSi- i- 1 rH rH rH i-S OS 00 O to i < ^i-l!- tOtO^COO i 1 CO-^ r-i -* i- 1 o -t> >ra op co t- m i- i~ 1-1 I-H 1 o co 1 1 T 1 1 to CO1 i-l OOtOi-HtOO g t- 1-1 1-1 l- Saponiflcation number mgrms. KOH Fatty acids ,HH 00 f ' ( rHn CM CM to o i to O IN * C O CM I- O ,:, CM IN os I os o J i o ! ^ J ^ *' CM 2 C1 c IN . SM . > 1 ' 1 I m CM co o co os to co co CM o m o co o O5 tO Tt< 00 OS oo 4 1 1 e j. xS J3 js .S -S 1 op CM co CM I co in to .H 6 4< 1 6 6 J> 1 -* -o g CO ^ ti 6 ll\ 6- 6 "** I -235f. 8i m CM SMQO^ CO CO CM rH CO CM CM CM 1 ) CO CM 1| OS O to M os co o n i ^ in m os co 10 os os . ^ -rH .^ m in ^ e ' O to isll si O O5 CO 1 > os m ^s p o 1 oS t* 00 rH rH ' r r IT rH 1 j 1 r IT in co IN rH rH ;-- CM -? CM IN IT 5 r ^ " J J ^ 7 o s oo m co ?n oo co co oo ?H ' IN Tln ooooo CM CM O O CO to rH rH 1 1 1 03 1-1 i-HTtl t-CMOCOO OO all directions. So that, if a pressure of 1 kilo, is exerted, by means of a piston 1 sq. cm. in area, on a liquid in one arm of a U-tube, the other branch of which is closed by a piston 16 sq. cm. in area, this would require a pressure of 16 kilos to balance the first piston (Fig. 262), the pressure transmitted by the pressing surface being proportional to the area receiving the pressure. 16 K FIG. 262. FIG. 263. The hydraulic press consists of a suction pump, P (Fig. 263), which draws water from the reservoir, A, and forces it through the strong copper tube, t, into the thick-walled chamber, B, hermetically sealed at the upper part by a large piston, 6, carrying a wide plate, c, on which is placed the material to be compressed. The com- pressing surface is that of the base of the small pump-piston and the surface receiving the pressure is given by the base of the piston, b, the pressure received being dependent on the ratio of the sections of the pistons and on the ratio between the arms, OP and PR, of the pump-lever. If PR is ten times as long as PO and the force exerted at R is 30 kilos, the piston of the pump receives a pressure of 300 kilos (30 x 10) ; if, on the other hand, the section of the large piston, b, is 15 times as great as that of the small piston, the pressure exerted on the latter will be 4500 kilos (300 X 15). When the piston, b, rises, the plate presses the substance against a strong cover, d, fixed by three or four EXTRACTION OF OILS 393 chloride has also been suggested recently (see vol. i, p. 378), since it is not inflammable like the other two solvents and, further, allows of the extraction of the oil from moist substances. The extraction can be carried out by direct exhaustion or by systematic exhaustion. In the former case, the substance is treated with pure solvent, so that large quantities of dilute solutions which must be concentrated are obtained , in the other process, a number of appa- ratus are arranged in a series so that the solvent passes from one to the other and leaves the last com- pletely saturated, while the first apparatus, as it becomes exhausted, is charged with fresh material and placed last in the series (see vol. i, p. 470, and exhaustion of beet in the diffusers, under the heading Sugar, later). From the saturated solution of the oil, the solvent is distilled by means of direct or indirect steam and is thus completely recovered, while the crude fat remaining is refined. There are various forms of appa- ratus corresponding with the first method of extraction, such as the FIG. 264. Merz universal extractor, that of Pallenberg, and the Wegelin and Hubner (Fig. 264) form, which is fairly widely used. In the last of these the fatty material is placed in the vessel, A, into which solvent is introduced from D by means of the tube, r q. The solvent saturated with fat is discharged into the still, C, where, by means of indirect steam passing through the coil, y, the solvent is distilled, its vapour ascending the tube, i, and condensing in B, and the liquid collecting in D. The fat remaining in columns, e. When the pressure is to be released, the water is discharged from the chamber, B, and the piston descends. The pump is provided with a safety-valve which regulates the maximum pressure desired. The large piston is made tight by encircling it at b with a leather ring (devised by the Englishman Bramah) with an Inverted U-section ; the water, in its attempts to escape along the sides of the piston, enters the ring and forces its edges against the piston with a pressure increasing with the pressure of the water, and thus forms a true hermetic seal. Nowadays horizontal hydraulic presses, which discharge the oil and cake more easily, are also used, but these occupy more space, while at the same time the piston does not recede of itself at the end of the operation. In practice, when a substance is to be compressed with a hydraulic press, two or more pumps are used. The first, which has a long stroke, raises the piston and plate rapidly, since at first the resistance is small ; when the pressure increases, the compression is continued more slowly by means of a small pump. To avoid attention to a number of pumps and loss of energy, works employing many hydraulic presses make use of the so-called hydraulic accumulators (Armstrong, 1843), which provide a store of water under high pressure for the feeding of several presses at once (Figs. 265 and 266). A piston, L, moving in a cylinder, B, just as in an ordinary hydraulic press, receives pressure from below by means of compressed water from a pump, passing through p and Vi ; the upper part of the piston is fixed to the centre of a plate, C, which, by means of three columns, S, supports the plate, E, carrying the heavy iron discs, D. When the piston is raised by the compressed water entering A, the whole accumulator, E, C, and the discs, D, are raised. When Vi is closed, A contains a store of water under great pressure which transmits pressure to a number of hydraulic presses simultaneously when the cock, D 2 , communicating with these presses is opened. In order to prevent the piston, L, from being raised too much and so forced out of the cylinder, B, the lower part of the piston is provided with a small vertical channel with a lateral exit ; when the latter is forced from the top of the cylinder, B, the water escapes, the pressure is lowered and the piston falls. Large works are supplied with two or more accumulators, so that when FIG. 265. FIG. 266. 394 ORGANIC CHEMISTRY C can then be drawn off through the tap, x, but if it retains solvent tenaciously, it is first heated by a current of direct steam, solvent and water then condensing together in the condenser, B ; owing to their mutual insolubility, these two liquids can be separated by means of a suitable separator situated at w between B and D, the water being thrown out. If the solvent saturated with fat, instead of being drawn off by the tube, u, is caused to rise to the top to the tube, I, whence it falls into the tube, v, the extraction is effected with continuous circulation of the solvent until the substance is exhausted. To expel and recover the solvent retained by the substance remaining in A, a current of direct steam is passed into the latter ; this carries off the vaporised solvent along the tube, k, through the valve, n, to the cooling coil, B, the condensed water and oil being passed through the separator, w, before the latter liquid is collected in D. Large works, however, always use batteries of extraction apparatus arranged in series. In a good extracting plant, the loss of solvent does not usually exceed 0-5 per cent, of the weight of oil extracted and is always less than 1 per cent. REFINING of oils, to separate as far as possible the tannin, protein, and colouring- matters extracted from the oily seeds and fruits, is generally effected by means of dehydrating or oxidising agents .(the latter attack the colouring -matters more especially). In order that sulphuric acid may not act on the glycerides (forming ethers) and heat and partially carbonise the oil, it must be used at a concentration of about 60 Be. and in small quantity (1 to 2 per cent.) with oil heated to 50 to 60, or with the cold oil ; under these conditions the few impurities are first carbonised and the oil becomes coloured, but after filtration it is obtained paler, purer, and clear. Zinc chloride often gives almost the same results as sulphuric acid, and is added in concentrated solution (sp. gr. 1-85) and in amounts up to 1-5 per cent, of the oil ; the black flocculent matter formed separates on standing or filtration. In some cases it is sufficient to leave the oil in large closed tanks of tinned iron with conical bases fitted with taps so that the impurities which gradually settle may be removed. Fragments of coal, peat, willow, &c., may be added, these carrying down the impurities as they settle. In order to avoid prolonged contact of the oil with the air, pressure filters (described in the section on Sugar) are preferred ; either the oil is placed at a higher altitude than the filter, or the pressure is applied by means of pumps, it being possible in this way to filter 1000 to 2000 kilos of oil in 24 hours. To purify with sulphuric acid (see later, Twitchell process), the latter is poured in a thin stream into the oil contained in a lead-lined vat and kept well stirred. After 7 to 8 hours, by which time small black clots of carbonised impurities have deposited, the oil is decanted into a second vat, washed two or three times with water at 40 to 60 (in some cases a small quantity of sodium carbonate is added to the second water), being stirred meanwhile or emul- sified by air from a Korting injector ; after being left to stand, it is either decanted or filtered. The water is sometimes intimately mixed with the oil to be washed by means of the so-called emulsor-centrifuge, (Fig. 267), consisting of two superposed metal plates with the concave parts inside and mounted on a hollow axle rotatable at 8000 to 10,000 revs, per minute, while through a central aperture commanded by two taps exactly adjust- able the oil and water are introduced in the desired proportions. The distance between the two plates can be altered so as to give a slit between their edges from 0-02 to 2 mm. .in width, the more or less completely emulsified mass being forced out through the slit by the plates themselves. If the oil does not separate from the water on standing, the emulsion may be destroyed by added powdered and calcined sodium sulphate or carbonate (which act as dehydrating agents) or by agitating the emulsion with animal black or one is raised and the other at its low position, excess of compressed water supplied by the pumps at any moment is directed to the latter accumulator, which is hence raised. ID this way, also, the final pressure of the hydraulic press can be utilised before discharging it, energy being thus saved that would otherwise be lost. By these means, a uniform and persistent pressure may be exerfed on several presses, but it is exerted, not gradually, but instantaneously, which may be disadvantageous in certain cases, unless indeed various accumu- lators at different pressures are employed. Accumulators with small pistons may be used for pressures up to 400 atmos. The circular iron rings composing the accumulator may be replaced by a single large cylinder filled with scrap iron or stones. The pressure of a hydraulic accumulator may be exerted in some degree gradually by connecting it with a compressed-air chamber (aromatic accumulator). As liquid for use in the accumulators, water, glycerine, or oil may be employed. OLIVE OIL 395 magnesium silicate (which separates the components) ; but the best results are obtained with centrifugal separators, like that used for milk (see p. 387), the water and impurities being forced to the periphery, where they adhere, while the oil is discharged by the central tube. The acid may also be mixed in the same way and continuous working may be attained by means of a battery of emulsors and another of centrifugal separators ; the latter serve well to purify the dregs of the oil and, in general, colloidal and soapy products of oils. When the emulsified or colloidal condition is due to the presence of gum or wax, it is preferable to initiate freezing of the glycerides, this breaking down the emulsion so that it can be filtered. When stable emulsions of oil and water are required, as is some- times the case, they can be obtained by pouring the oil into a mixture of water and the amide of a higher fatty acid or an acidyl derivative of an aromatic base, or both of these, together with the salt of a higher fatty acid (Kosters, 1906). To deodorise oils, they are passed through bone-black or, sometimes, elm-bark. Tn some cases, and more especially when very rancid, oils are purified by deacidifying them with a concentrated solution (8 to 10 Be. for cotton-seed oil and 36 to 38 Be. for olive oil) of caustic soda in amount slightly exceeding that calculated from the acid number ; this treatment, however, readily leads to the forma- tion of persistent emulsions and to loss of glycerides and also of fatty acids. These emul- sions, which are due to the presence of soaps, are broken down in the manner already de- scribed, first being heated to 50 to 60. If the acidity exceeds 30 per cent., the losses would be so high that deacidification is not advisable ; such oils (e.g. highly acid olive oil after refining with sulphuric acid) cannot be used as lubricants or for softening wool, but are used solely for soap, unless indeed the fatty acids are transformed into glycerides by treat- ment with glycerine as described on p. 373. Bleaching with hydrogen peroxide, dichromate or permanganate is carried out as with tallow (see p. 381), but if the oil is first deacidified, 100 grms. (instead of 1500 grms.) of dichromate per quintal are sufficient. If it is required to eliminate every trace of soap, the oil is heated with a boiling solution of 5 per cent, sulphuric acid. Vegetable oils are frequently decolorised nowadays with fuller's-earth (see p. 77). 1 OLIVE OIL is obtained by pressing the fresh olives of Olea europcea in the period from October to December (in Morocco, in August and September). The olive grows in abundance in Central and Southern Italy, on the shores of Lake Garda, on the Genoese Riviera, and in Southern France, Spain, Portugal, Dalmatia, Istria, Greece, Morocco, California, and Southern India. The composition of the fruit is given in the Table on p. 391. It is not advisable to extract the oil from stored or fermented olives, these giving the so-called huile tournante, which is rich in fatty acids and yields a persistent emulsion when shaken with soda solution, and a Turkey-red oil similar to the sulphoricinate (see p. 327) when treated with concentrated sulphuric acid. If the olives cannot be worked at once, fermentation is prevented by storing them in a cold, dry, and well-ventilated place. The fermentation (according to Tolomei) is due to an enzyme (olease) occurring with the oil, which, in the presence of air and light, it 1 Fuller's-earth has been long used in Northern Africa for clarifying olive oil; in Chicago it was thus employed as early as 1878, but its use was considerably extended subsequently to 1890. It consists of aluminium and magnesium hydrosilicates, and is found in granular or powdery deposits in Florida and also at Fraustadt, in Silesia. The decolorising action of this earth depends on its state of hydration, the maximum effect being obtained if it is first lightly roasted (at about 200), while if the roasting is carried too far so that all the water of hydration is lost, the decolorising power is entirely destroyed. The oil is shaken with 1 to 3 per cent, of the earth, and the mass heated for a short time at a temperature (60 to 100) varying with the nature of the oil and then passed to the filter-press, the first turbid portions of the filtrate being reflltered. FIG. 267. 396 ORGANIC CHEMISTRY decolorises ; if the olease is removed by washing the oil with water, the oil is not decolorised under the influence of light. The extraction of olive oil is not always effected by rational processes and plant, but usually the olives are first crushed by means of the ordinary edge-runners (see Fig. 210, p. 251). The pulp is next placed in suitable bags of tenacious vegetable fibre or wool surrounded by horsehair and then pressed, the type of press employed varying widely with the locality. The ring hydraulic press (see Fig. 261, p. 392) and other forms, still further improved, give excellent results. In some cases a moderate pressure is first employed, the result being oil of superfine quality (virgin oil). The residues are steeped in hot water and subjected to increased pressure. Repetition of this procedure, employing a still higher pres&ure, gives an industrial oil. The cake from the second pressing may, however, be agitated in a vat through which water flows ; part of the remaining oil is thus removed, this being collected in a second vat, where it undergoes protracted washing with water, yielding so-called washed oil. The Kuess-Funaro process (1902), which results in an improved yield and a readier extraction, consists in emulsifying each time with feebly alkaline aqueous solutions. 1 The residual cakes (known in Italy as same), after being dried, still contain 7 to 11 per cent, of oil, which is nowadays extracted in large works by means of carbon dkulphide, which gives the very green, so-called sulphocarbon oil, almost all of this being used in the manufacture of green soap for the textile industries. 2 Pure olive oil is yellowish or, in some cases, almost colourless or slightly green. The finer qualities taste but little ; freshly pressed Puglia oil has a rather bitter and unpleasant taste (due to camphene, eugenol, and other substances investigated by Canzoneri), which it gradually loses. The composition of olive oil varies with the district of origin and with the conditions of extraction, the solid glycerides fluctuating from 10 to 28 per cent, (more especially palmitin). The liquid glycerides, which occur to the extent of 70 to SO per cent., were formerly thought to consist of triolein alone, but the presence of linoleic acid (as much as 6 per cent., this explaining the high iodine number of the oil) has now been proved, and there appears also to be about 1-5 per cent, of a mixed glyceride and 0-2 to 1-5 per cent, of volatile acids, besides 0-7 to 1-6 per cent, of non-saponifiable substances (phytosterol and, according to Sani, an oil not yet defined). It contains a variable quantity of free fatty acids, and when impure readily becomes rancid. If the acid number excetels 16 (i.e. 8-1 per cent, of acids calculated as oleic acid), it cannot be used as machine oil as it attacks metals. 1 A new process of extracting olive oil proposed by Acapulco (1910-1911) and tested with favourable results in the experimental oil plant of the Portici Higher Agricultural School, is based on the different surface tensions of the two liquids (oil and water) which are present in the pulp of the olive and have to be separated, and hence on their different capillary behaviour towards the vegetable tissues constituting the pulp. The surface tension of the oil^is about one-half of that of the water, so that separation of the two liquids is easily attained by even slight diminution of the pressure below that of the atmosphere. The separation is also facilitated by rise of temperature and by the fact that the water present has a capillary constant higher than that of the oil, so that it remains more strongly adherent to the vegetable tissues. The essential part of the machinery of this process after the stones have been separated from the pulp consists of the so-called filtering extractors, formed of superposed metallic cylinders, inside which is a metal filtering cloth, an annular space communicating with the vacuum pump being left between the walls and the cloth. A starrer fitted with vanes continually moves the mass of pulp contained in the extractor and spreads it in thin layers on the filtering cloth so that the liquid portion is separated from the pulp. By steam-heating the extraction can be carried out at any temperature. But even in the cold the exhaustion of the pulp is more complete than that obtained by the older systems, while in the hot it surpasses that reached by pressing the " sanse " in the most powerful hydraulic presses. It is said that the Acapulco process is more economical than those previously used and that it lends itself to the production on a large scale of pure, slightly coloured oils of constant type. But, as yet, this process has not been subjected to decisive commercial tests. * To distinguish sulphocarbon oil, which has a lower iodine number (77 to 80), from that obtained by pressure, Halphen's test (1905) may be employed. To 50 c.c. of the oil heated to 100 are added 12 c.c. of'alcoholic caustic potash diluted with an equal volume of water, the mixture being heated for ten minutes at 110 and cooled to 100 ; 200 c.c. of hot water are then added and the liquid, after cooling, shaken with 200 c.c. of saturated sodium sulphate solution ; 20 c.c. of 30 per cent, copper sulphate are then added, and the liquid filtered. If the filtrate is not green, a little more of the copper sulphate solution is added and the liquid filtered again if necessary. 5 c.c. of silver nitrate solution (containing 1 vol. of 1 per cent, aqueous silver nitrate solution and 5 vols. of glacial acetic acid) are then added to the liquid, which is boiled, allowed to cool, supersaturated with ammonia and filtered, the filter being washed with dilute ammonia. If black silver sulphide remains on the filter, the presence of sulpho- carbon oil (or impure cruciferous oils colza, mustard, &c. which cannot be detected otherwise) is certain. Cusson (1909) has devised a simpler test : 200 grins, of the oil are vigorously shaken with 50 grms. of 90 per cent, alcohol and then distilled on a water-bath, the distillate being collected in a well-cooled flask containing a little alcoholic potash. Even traces of carbon disulphide thus yield potassium xanthate, which gives a yellow coloration or precipitate on addition of alcoholic cupric acetate solution. TESTING OF OLIVE OIL 397 Pure olive oil is used as a comestible and the very pure and more liquid qualities for oiling clocks, while the other qualities are employed in large quantities in the manufacture of soap, lubricants, burning oil, and Turkey-red oil. The purity of the oil is controlled by various tests referring to the constants given in the Table on p. 378, and by certain special tests. Olive oils of certain origins give abnormal constants, e.g. Algerian and Moroccan oils have an iodine number of 96 and are reddened by nitric acid ; pure Tunisian olive oil gives the reaction for sesame oil (Villavecchia and Fabris' test) but not the Belliez reaction (test for sesame oil with a saturated solution of resorcinol in benzene and nitric acid) ; the extraneous substances of Tunisian oil which give the Villavecchia and Fabris test can be removed by shaking the oil with hot water. Detection of added sesame oil is effected by Tortelli and Ruggeri's modification of Bau- douin's test on the fatty acids (see p. 384), or more rapidly on the oil itself by means of Villavecchia and Fabris' test, taking care to dilute 5 c.c. of the resulting red acid liquid with four times its volume of distilled water and to shake the mixture in a cylinder, and observing the lapse of time required for the disappearance of the red coloration. With any pure olive oil, if there is a coloration, this disappears within 5 minutes or, in exceptional cases, in 8 minutes, whilst if sesame oil (even only 3 per cent.) is present the colour will persist for 30 minutes (Zega and Todorovic, 1909). The presence of cotton-seed oil is indi- cated by the Halphen reaction (see p. 381) or by Tortelli and Ruggeri's modification of Becchi's reaction, which is carried out on the liquid fatty acids in the following manner : 20 c.c. of the suspected oil are hydrolysed with alcoholic potash in the ordinary way (see p. 379), the aqueous solution of the soap being neutralised with acetic acid and precipitated with lead acetate ; the lead salt, separated by filtration, is shaken with ether and the filtered ethereal solution decomposed in a separating funnel by dilute hydrochloric acid. The ethereal layer is filtered and the ether evaporated, and to 5 c.c. of the residue (liquid fatty acids) * are added 10 c.c. of 90 per cent, alcohol and 1 c.c. of 5 per cent, aqueous silver nitrate solution ; if a black precipitate is then formed on heating for some time on a water-bath at 60 to 70, the presence of cotton -seed oil is proved. In certain special cases the Becchi reaction alone is insufficient to indicate with certainty the presence of cotton -seed oil. Traces of mineral oils in vegetable oils are detected by the formation of a yellowish red solution on addition of a benzene solution of commercial picric acid (F. Schulz, 1908 ; see Note, p. 379). To detect fish oil in vegetable oil, 100 drops of the latter are treated with a mixture of 3 c.c. of chloroform and 3 c.c. of acetic acid, sufficient bromine being then added to produce a persistent brown coloration ; after ten minutes rest the vessel is introduced into boiling water, when the liquid will remain liquid if the vegetable oil is pure, whilst insoluble bromo -compounds will separate if fish oil is present. With boiled oil, the metals are first eliminated. Where the oil has been coloured yellow with auramine, this is detected by boiling 1 c.c. of the oil with 20 c.c. of 8 per cent, alcoholic potash and a little zinc powder in a reflux apparatus, 20 c.c. of pure benzene and 50 c.c. of water being added after cooling ; the benzene solution is evaporated and the residue taken up in glacial acetic acid, a blue coloration, becoming darker on heating, being formed if auramine is present. Sanse oil or sulphocarbon oil, extracted from the cake or marc by means of carbon disulphide, has a dark green colour, and the corresponding fatty acids have a rather low iodine number (as low as 75) and a somewhat higher melting-point than usual. The presence of arachis oil in olive oil is shown by the Tortelli and Ruggeri test, which has been modified by Fachini and Dorta (1910) as follows : 20 grms. of the oil are saponified with alcoholic potash, the alcohol being then expelled, the soap dissolved in water, the fatty acids liberated by hot dilute sulphuric acid, and the clear fused acids collected on a moist filter ; they are then washed with hot water and dissolved in 150 c.c. of pure, tepid acetone, water being subsequently added, drop by drop, until a turbidity is formed ; the liquid is finally rendered clear by the addition of a few drops of acetone at 40 to 45 and then left to crystallise. In presence of arachis oil, characteristic shining crystals separate at 15 ; after an hour these are collected on a filter, washed with 10 c.c. of dilute acetone (32 vols. water + 68 vols. acetone) and examined for arachic and lignoceric acids by the Tortelli 1 The liquid fatty acids can be separated, to a considerable extent if not quantitatively, from the solid oner by dissolving the mixtures in light petroleum or, better, in acetone and crystallising out almost all the solid fatty acids by cooling (to - 20) (Fachini and Dorta, 1910). According to Twitchell (U.S. Pat. 918,612, 1909) the liquid fatty acids are separated from the solid ones by fusion with 1 per cent, of aliphatic sulpho-acids, which render the liquid acids soluble even in water 398 ORGANIC CHEMISTRY and Ruggeri test : one-half is dissolved in 100 c.c. of 70 per cent, alcohol, warmed slightly and allowed to cool, separation of crystals indicating arachi die acid (m.pt. 75 to 76) with certainty. STATISTICS. The cultivation of the olive is widespread in Italy, where it extends to over a million hectares. The production and price vary considerably from year to year, sometimes causing serious agricultural crises. The production of the oil varies from 2 to 4 millions of hectolitres (250,000 to 350,000 being sulphocarbon oil). The mean annual production per hectare was 3-66 hectols. in 1879 (total, 3,400,000 hectols.) and 2-5 hectols. in 1899, the total being 2,515,000 hectols. The total production in Italy was 3,086,000 hectols. (3-04 per hectare) in 1890 ; 2,337,000 (2-16) in 1901 ; 1,846,000 (1-7) in 1902 ; 3,256,000 (2-99) in 1903 ; 3,412,300 in 1906; 2,559,000 (corresponding with 15,000,000 quintals of olives) in 1909 ; 1,384,580 (9,366,000 quintals of olives) in 1910, The exportation from Italy was 412,000 hectols. in 1898 ; 506,000 in 1899 ; 290,000 in 1900 ; 424,350 in 1901120,000-140,000 to France, 100,000 to Russia, 50,000 to North and 55,000-80,000 to South America, 60,000-80,000 to England, and 42,000-52,000 to Austria -Hungary. In 1908 the exports were 368,000 quintals (91,800 to the United States and 127,000 to the Argentine) ; in 1909, 184,500 ; and in 1910, 285,150 (96,000 to the United States, 73,500 to the Argentine, and 40,000 to France). The price varies according to the harvest in the different districts and to the requirements abroad ; in some years the producers sell at 4 per hectolitre and in others at 48s. to 56s. In 1908-1909, owing to the small crop in Puglio, caused largely by the drought, prices exceeded 7 10s. per quintal. In 1907 the products of the olive industry of Italy alone were valued at more than 8,000,000 (oil and residues) ; the number of presses was 52,000, these being distributed in 18,000 works, of which only 2400 employed steam. During the two or three months of the olive campaign, about 70,000 operatives are employed. Extraction of the oil by means of carbon disulphide is carried out in 60 works, consuming 780 h.p. (almost all steam- power) and employing 1230 workmen. The exportation of sulphocarbon and washed oil was about 60,000 quintals in 1900, 100,000 in 1904, 55,660 in 1909, and 131,400 in 1910. Italy imports olive oil every year, especially from Spain and Greece, the total amount being 39,000 quintals in 1908, 52,330 in 1909, and 58,450 in 1910. Portugal produces 250,000 to 350,000 hectols. per annum, and Spain (especially in Andalusia) about 2 or 3 millions of hectols., of lower quality than the Italian. In 1905 the Spanish production was 1,492,490 quintals, and in 1906, 1,336,650; the amount exported being 340,000 quintals in 1905, 190,000 in 1906, and about 150,000 in 1907. France, with about 150,000 hectares under olives, produces annually about 1,500,000 hectols. of oil. Greece produces 550,000 hectols. (in 1907) ; Asiatic Turkey (with Crete), 2,000,000 hectols., importing also 350,000-500,000 hectols., especially from Spain, Italy, Tunis, and Algeria, and exporting 200,000-250,000 hectols. of the finest quality. In 1906 Tunis produced 243,000 hectols., and in 1907 almost 400,000. The world's production is about 10,000,000 hectols. of the oil. Germany imports 10,000-13,000 tons, seven-tenths of it for industrial purposes. Austria imports about 6000 tons. England imported crude olive oil to the value of 300,000 and purified oil to the value of 386,190 in 1910. The imports into the United States were valued at 1 ,162,400 in 1910 and at 1,149,800 in 1911. CASTOR OIL is extracted from the seeds of Ricinus communes, a plant cultivated in India, Italy, Messina, California, Egypt, and Greece. The oval seeds are 10 to 15 mm. long, about 6 mm. broad, and rather flat, and are covered with a brownish or marbled, shining, brittle skin ; when peeled they contain 45 to 55 per cent, of oil. The oil was at one time extracted by pressing the ground seeds twice in the dry state and then pressing the residue after steeping in hot water. Nowadays, however, three consecutive pressings of the hot crushed seeds with increasing pressures are employed, modern hydraulic presses being used. This procedure yields first a fairly pure pale oil, then one less pure, and finally a more highly coloured oil for secondary industrial purposes. One hundred kilos of the seeds yield 9 kilos of^husks, 43 of residual cake, 20 of oil of the first, 10 of the second, and 8 of the third pressing. The oil is purified by heating with an equal volume of boiling water, which precipitates many protein and gummy substances ; it is decolorised by means of bone-black or by the ordinary processes given for tallow. The refined oil is almost colourless or faintly yellow, and has a high specific gravity, LINSEED OIL 399 considerable viscosity, and a peculiar, unpleasant taste and smell. It forms an excellent purgative, the less pure qualities being used in the manufacture of sulphoricinate (see p. 327) and of transparent soaps capable of retaining considerable quantities of water. Its soap differs from others in not rendering water opalescent. The residual cake, whether extracted with carbon disulphide or not, is injurious and cannot be used as cattle-food, but it is of value as manure, since it contains 4 to 5 per cent, of combined nitrogen and is sold at 8s. to 105. per quintal. Castor oil contains various glycerides but is free from tripalmitin. Triricinolein is solid, and there appear to be glycerides of a ricinoleic acid and of a ricinisoleic acid, also of a hydroxystearic acid (melting at 141 to 143) and a dihydroxystearic acid (which explains the characteristic high acetyl number of castor oil). The oil yields, besides ricinoleic acid, more or less highly polymerised compounds with less and less marked acid characters (e.g. ricinisoleic acid), these increasing in amount with the age of the oil. Castor oil is strongly dextro-rotatory (40-7 in a 20 mm. tube). Unlike other oils, it is soluble in all proportions in absolute alcohol and in glacial acetic acid ; at 15 it dissolves in 2 parts of 90 per cent, alcohol or 4 parts of 84 per cent, alcohol, but is insoluble in light petroleum or paraffin oil (both of which dissolve all other oils and fats). Hence, if a castor oil is insoluble in light petroleum and gives a clear solution with 5 vols. of 90 per cent, alcohol, it may be regarded as pure. The solubility relations are completely inverted if it is heated to 300 and 10 to 12 per cent, of it distilled ; there then remains a product termed floricin, which solidifies at -20, is insoluble in alcohol, dissolves in all proportions in mineral oil, and forms a stable emulsion with 5 parts of water. A similar product is also obtained by heating castor oil to 200 in presence of 1 per cent, of formaldehyde ; if heated with zinc chloride solution, it becomes dense. The potassium salt of the condensed product, with water and formaldehyde, gives a disinfectant solution producing the same effects as lysoform or ozoform. The constants of castor oil are given in the Table on p. 378. Italy produces a certain quantity of castor oil seeds, but the greater part is imported, this amounting to about 55,000 quintals in 1 892 and 80,000 (equal to about 27,000 quintals of oil) in 1904. In 1901 Italy exported 5420 quintals of castor oil ; in 1908, 3454 ; in 1909, 2292 ; and in 1910, 4766 quintals of the value of 18,110. In 1908 Germany imported 62,400 quintals of castor oil, and in 1909, 85,000. LINSEED OIL is a drying oil, as it contains much linoleic and linolenic acids (see pp. 303 and 304), and when spread out in a thin layer on a sheet of glass slowly forms a solid skin (varnish), this forming more rapidly with the boiled oil. Linseed oil is extracted from the seeds (containing 35 per cent, of oil) of Linum usitatis- simum, which are converted into flour by the ordinary edge-runner mills and pressed hot in hydraulic presses. Linseed is cultivated especially in the Baltic provinces of Russia, and also in Southern Russia, Eastern India, the United States, and the Argentine, and to a less extent in Egypt, Belgium, and Italy. Linseed oil extracted by means of solvents contains more unsaturated fatty acids and less volatile acids than the expressed oil. According to Fahrion (1903 and 1910), the fatty acids separated from linseed oil contain 17-5 per cent, of oleic acid, 30 per cent, of linolic acid, 38 per cent, of linolenic and iso- linolenic acids, 8 per cent, of palmitic and stearic acids, all combined with 4-2 per cent, of glycerine and 0-6 per cent, of non-saponifiable substances. The purity of the oil is indicated by means of the constants given in the Table on p. 378, especially by the iodine number and the refractive index, which, in the different qualities, varies from 1-484 to 1-488 at 15 (or from 81 to 85 Zeiss at 25 or 87 to 91 Zeiss at 15), whilst cotton-seed oil gives no more than 1-477 and maize oil no more than 1-4765 at 15, A good proportion of the oil is used in practice in the form of boiled linseed oil (see Note on next page), since on boiling it acquires drying properties especially necessary to the varnishes prepared with the oil. The drying power can be determined by Livache's method. On a watch-glass is spread 1 grm. of lead-powder (obtained by immersing a strip of zinc in the solution of a lead salt and washing the precipitate with water, alcohol and ether, and drying), on which 0-6 to 0-7 grm. of oil is allowed to fall slowly in drops, the whole being then weighed exactly and left at a moderate temperature in a well -lighted situation. After 18 hours the weight 400 ORGANIC CHEMISTRY begins to increase, the maximum increase (12 to 15 per cent.) being obtained within 2 or at most 3 days (it then diminishes slightly). Other drying oils give the following increases : ivalnut oil, 7-9 per cent. ; poppy-seed oil, 6-8 per cent. ; cotton-seed oil, 5-9 per cent. ; cod- liver oil, 7-4 per cent. ; the remaining oils increase in weight only after the fourth or fifth day to a maximum of 2-9 per cent, after seven days. The drying properties are determined best and most rapidly by spreading a given weight of the boiled linseed oil on a definite area of glass (1 mgrm. per sq. cm.) and leaving the latter in a horizontal position until the oil is no longer adhesive when pressed lightly with the finger (the temperature should always be noted). The drying power of an oil can be determined also from the ozone number (Molinari and Scansetti, 1910). In a 20 mm. tube, pure linseed oil gives a rotation of -0-3 in the Laurent saccharimeter at 15, whilst other resin oils and sesame oil are dextro-rotatory. Linseed oil is used mostly in the manufacture of lacs and varnishes, 1 mastics and lino- leum. The latter is obtained by oxidising (blowing) hot linseed oil, after addition of the dryer (see Note), for 18 to 20 hours with hot air until it thickens to linoxyn ; about 30 per cent, of colophony are then added, the whole being converted into a paste with cork- dust at a temperature exceeding 100. The mass swells and is compressed hot (140) on a strong textile previously varnished to protect it from moisture, the whole being repeatedly pressed between hot rollers. It is finally dried in suitable hot chambers, where it loses its smell and acquires elasticity and weight. It is coloured in the pasty condition with mineral colouring - matters . Linseed oil is used also for making soft, transparent soaps (see later). STATISTICS. Almost the whole of the linseed oil produced comes from India, Russia (about one-fourth), North America, and the Argentine. From 1895 to 1900 it amounted to about 1,500,000 tons, while in 1903 it exceeded 2,500,000 tons, the price varying from 10 to 12 per ton. North America produces about 200,000 tons of the oil ; the exports include very little oil but comprise 500,000 tons of cake out of a total of 700,000 produced. The imports of linseed oil to the United States were valued at 72,600 in 1911 and the 1 Oil Varnishes and Lacs arc liquids which, when spread out in a thin layer on an object, leave on drying a solid , shining skin insoluble in ether and water and almost impermeable. Varnishes and lacs have linseed oil as a basis, and are often mixed with mineral or organic colouring-matters. Oil varnishes are formed from linseed oil rendered drying by dissolving small quantities of certain minerals in the hot. Oil lacs are obtained by adding to the almost boiling oil varnish (free from gummy matter) the fused copal or other resin, and diluting with oil of turpentine at the moment of using : all these new components contribute to increase the fixation of oxygen. Crude linseed oil requires four to five days to dry in a thin layer, but the fixing of oxygen, that is, the drying, may be markedly accelerated by the presence of small quantities of dissolved metals which act as catalysts. At one time oil varnish (boileft linseed oil) was prepared by heating the oil to 220 to 300 for 2 to 3 hours in presence of minium, litharge, or manganese dioxide (dryers). This procedure yielded dark varnishes (boiled varnishes), and was accompanied by danger from flre, the heating being carried out in open iron vessels furnished with stirrers and heated directly over the fire. Nowadays the dryer (0-1 to 0-25 per cent. Mn or 0-5 per cent. Pb + 0-1 per cent. Mn is sufficient) is dissolved by heating at a far lower temperature (100 to 120 and best in a vacuum) for 4 or 5 hours (by indirect steam at 135 to 150), it being added (when the oil ceases frothing) as manganese borate or, better, manganese linoleate or resinate, and the mass stirred with com- pressed air ; in this way, the so-called cold varnishes are obtained. These are paler varnishes which dry in 6 to 8 hours, whilst the others require as long as 24 hours. It has been proposed to decolorise boiled linseed oil with ultra-violet rays. The drying is far more rapid in the hot than in the cold. Prolonged boiling of linseed oil without dryers increases not so much the drying properties as the consistency, certain components of the oil being polymerised and the iodine number consequently diminished ; these oils, thickened at 295 to 340", bear the names Dickiil, Standiil, and lithographers' varnish. The action of oxygen during the drying of varnishes seems to lead to the decomposition of the glycerides of the saturated acids and of oleic acid with subsequent complete oxidation of the glycerine and acids, the glyceride of hydroxylinolic acid (hydroxylinolein), insoluble in ether, being also formed as well as anhydrides and polymerised substances. In the manufacture of lacs, a difficult and important operation is the fusion of the copal- previously prepared in lumps in cylindrical or slightly conical, enamelled iron or aluminium vessels ; these are protected at the bottom by an iron or copper casing when heated by direct-fire heat and are provided with a cover and chimney to carry off the noxious vapours, which are carefully condensed or burnt. The temperature is closely watched by means of a thermometer immersed in the fused copal (300 to 360). It is nowadays regarded as preferable to heat with hot water under pressure (up to 300) circulating in coils situate in the lower part of the boiler. Complete, uniform fusion occupies 3 to 4 hours (with a loss in weight of 15 to 30 per cent.), the linseed oil containing the dryer and heated to about 100 being then mixed in ; if any turbidity appears, the mass is heated to 300. It is then allowed to cool to 150 to 200, the addition of the oil of turpentine which dissolves the lac and, if necessary, of the dryer, being then begun. The diluted lac is filtered under pressure and discharged into smaller vessels, in which it is allowed to cool completely. The addition of calcium salts of colophony renders the lac harder but more brittle. The copal is sometimes replaced by colophony and other resins, which are, however, readily saponifiable ; a mixture of Japanese wood oil with resin and a little lime gives a good lac. Lacs are improved by prolonged storage (at least a year). Linseed oil for making lacs should be free from gummy matters, which may be removed by filtration through magnesium aluminium hydrosilicate (see p. 395). The softer lacs contain more than 50 per cent, and the harder ones less than 50 per cent, of linseed oil. The United States imported 12,000 tons (488,000) of copal (kauri and dammar) in 1910 and 11,000 tons (410,000) in 1911. PALM OIL 401 exports at 37,200 ; walnut oil to the value of 239,200 was also imported in 1911. Germany treated 142,000 tons of linseed in 1891, about 250,000 in 1900, and more than-331,000 in 1903 ; consequently the importation of linseed oil, which amounted to 35,700 tons in 1890, fell to 3350 tons in 1905 and to 2059 in 1909. The importation of oil-cake (mostly linseed) into Germany is, roughly, about 500,000 tons (exports, 180,000 tons). France imports about 150,000 tons of linseed (1905-1906). Holland imports more than 200,000 tons of linseed and exports 82,000 ; it imports also about 200,000 tons of linseed cake and exports about 25,000 tons of oil. England imported about 310,000 tons of linseed in 1900 and almost 506,000 tons in 1904 ; the imports of pure linseed oil amounted to 19,936 tons in 1909 and to 37,242 tons (1,252,140) in 1910 ; 30,000 tons of the oil were exported in 1905, while in 1911 the exports were valued at 837,712. In 1905 Italy imported 1800 quintals of boiled linseed oil, and 3011 quintals (worth 9030) in 1910, besides 438,600 quintals of linseed in 1908 and 367,660 (worth 544,000) in 1910 ; the import duty is the same as for other vegetable oils, namely, 20s. 10^. (26 lire) per quintal for the boiled oil and 19s. 2d. (24 lire) for the crude. Italy also imported 26,432 quintals of varnish free from spirit, worth 150,920, in 1910. PALM OIL is extracted from the fruit of certain varieties of palm (Elais guineensis and Elais melanococca, which grow in Western and Central Africa and in America, and Astrocaryum acuale and Astrocaryum vulgare, growing in Guiana). The orange-brown fruit, of the size of walnuts, hangs in bunches. The pulp constitutes, according to the variety, 25 to 75 per cent, of the fruit, which contains a nut also yielding an oil (palm-nut or palm-kernel oil). The extraction of the oil in the districts where the palm is grown is carried out in an irrational manner, the fruit being sometimes heaped up until it putrefies and the oil then pressed out. In other cases the fruit is stored and compressed in excavations in clay soil, being left to putrefy until the oil separates at the surface. In other places the fruit is fermented for a month and then heated with water, so that the pulp becomes detached from the stone and can then be heated and pressed again with water until the fused oil comes to the top and can be decanted off. In these ways more than one-half of the oil is lost, and machinery is now being introduced for detaching and disintegrating the pulp and for the rational pressing of the latter. When freshly expressed it has a buttery consistency, an intense orange-yellow colour and a faint smell of violets ; the colour and odour persist in the soap prepared from it. It can be decolorised by heating it when exposed to the air and light, but this is effected best and most rapidly by fusing and heating it until it loses the water remaining from any preliminary heating with water for the removal of impurities ; this separates from the fused mass in 24 hours. After this it is introduced into a metal vat or cylinder provided with a cover and tube for carrying the gases to the chimney ; the fat is heated to 100 by means of an indirect steam coil and a vigorous and finely divided stream of air passed through the oil from a perforated tube. In a couple of hours' time decolorisation is com- plete ; at the same time the pleasant odour of the fat remains, although it is destroyed if the fat is decolorised by simple heating to 220. Chemical decolorisation is often employed, the oil (1000 kilos), already purified by treatment with water and by fusion, being heated in a boiler to 50, at which temperature 30 to 50 kilos of commercial hydrochloric acid and 8 to 10 kilos of potassium dichromate dissolved in 18 to 20 litres of boiling water are stirred in. After 15 to 20 minutes, 1 to 2 kilos of sulphuric acid are sometimes added, the stirring being continued until the oil becomes limpid ; stirring is then stopped and 70 to 80 kilos of boiling water sprayed on the oil to wash it. After standing overnight, the water is decanted off, the acid separated from below, and the oil washed once or twice by boiling with water. Even when fresh it contains 12 per cent, of free fatty acids, and as it becomes older it decomposes spontaneously with increasing ease, separation of fatty acids (up to 55 per cent.) and glycerine which can be extracted with water taking place. Besides free palmitic acid, the principal components are the glycerides of oleic and palmitic acids, up to 1 per cent, of stearic acid, a little linolic acid, and about 1 per cent, of heptadecylic acid, Cl7H3 4 2 . The colouring-matter of palm oil admits of various characteristic colour reactions : with sulphuric acid, a bluish green coloration is obtained, whilst mercurous nitrate colours it first canary -yellow, then pale green, and finally straw-yellow. II 26 402 Palm oil is used in large quantities in the manufacture of soap and candles, its value being related to the melting-point of its fatty acids. It is calculated that the palm oil placed on the market (that is, exclusive of the large amounts consumed where produced) amounts to 70,000 to 80,000 tons per annum. Germany imports about 14,000 tons of palm oil and exports 14,000 to 18,000 tons of palm-nut oil and coco-nut oil. Marseilles imports 18,000 to 20,000 tons of palm oil and Austria-Hungary 3000 to 5000. England imported 176,264 tons of crude palm oil in 1909 and 199,438 tons (3,056,600) in 1910, while the United States imported 42,000 tons in 1910 and 21,000 tons (645,000) in 1911, in addition to 5000 tons of palm-kernel oil. The price varies with the year from 40*. to 52s. per quintal. The best qualities of palm oil are from Lagos ; then come those of Old Calabar, Benin, and Acora ; while among the more impure varieties are those from Gabun, Liberia, and the Cameroons. PALM-NUT OIL (or Palm-kernel Oil) is obtained by crushing and then either pressing in hydraulic presses or extracting with solvents the stones contained in palm fruit ; freed from shell, the seed forms 9 to 25 per cent, of the weight of the fruit and contains 43 to 55 per cent, of fat, which is white or straw-coloured and free from fatty acids when fresh, although it turns rancid fairly easily in the air ; it melts at 26 to 30. It consists of about 1 5 to 25 per cent", of triolein, 33 per cent, of trigly cerides of stearic, palmitic, and myristic acids, and about 45 to 55 per cent, of triglycerides of lauric (in preponderance), capric, caprylic, and caproic acids. It bears a great resemblance to coco-nut oil, even in the property of its soaps of taking up large proportions of water as much as 600 per cent, (coco -nut soap up to 1200 per cent.) and of being somewhat soluble in solutions of salt. The total quantity of palm nuts placed on the market is about 1,125,000 tons. Germany now imports about 200,000 tons of palm nuts and copra (see Coco-nut), and 152,350 quintals of palm oil (in 1909) ; Austria, 30,000 tons ; France, about 7000 ; and England about 60,000 ; while Italy im- ported 81,920 quintals of palm oil of the value of 216,270 in 1910 (78,460 quintals in 1908). COCO-NUT OIL (or Coco-nut Butter) is obtained from the coco-nuts yielded twice a year by the palms Cocos nucifera and Cocos butyracea, which grow abundantly in Africa, Ceylon, Cochin China, and the Indies. The coco-nut is oval and about 20 to 25 cm. long and 12 to 16 cm. broad ; it is covered with a fibrous mass, used for making matting, cord, and baskets, and with a hard, woody shell, 8 to 12 mm. thick, which some time before maturation contains a sweetish, watery liquid (coco-nut milk), this subsequently disappearing and giving place to a soft edible pulp. The latter hardens in the air and is sold under the name of copra (60 to 70 per cent, of oil) for the extraction of oil. At the place of production this is carried out in a very primitive manner, but in European factories the dry pulp is ground, steeped in boiling water and pressed, first cold and then hot. The oil is nowadays decolorised with bone-black or absorbent earths (magnesium hydro- silicates), and in the white form thus obtained is used as a comestible (coco-nut butter ; see Margarine), after the free acids have been removed with highly concentrated solutions of caustic soda and after the odorous constituents have been expelled by means of super- heated steam. The best form for use as food is the softer, almost liquid butter obtained by the first pressing in the cold. Its digestibility is equal to that of margarine and butter. If it contains more than 2 per cent, of free fatty acids (expressed as oleic acid), it cannot be used for food and then goes to the soap factory as industrial coco-nut oil. Its composition is variable, and of the unsaturated acids it contains only oleic acid (about 10 per cent.), while glycerides of myristic and lauric acids are present in large quan- tities and those of caproic, caprylic, and capric acids to the extent of 2 to 3 per cent. The pure fat contains no free fatty acids, or at most traces. It has already been men- tioned that it gives a soap separable from solution only by very large quantities of salt ; it is, however, capable of absorbing as much as 10 to 12 times its own weight of water, and is hence highly valued by soap manufacturers. It is used alone for culinary purposes and for mixing with margarine and adulterating cacao butter. In. its analysis, attention is paid to the physical and chemical constants given in the Table on p 378. A large area of the earth's surface (about 1,400,000 hectares) is under coco-nut palms, which in a good year would yield 960,000 tons of coco-nut oil. In 1905 about 300,000 tons COTTON-SEED OIL, MAIZE OIL 403 of copra were placed on the market, and in 1906 only 200,000 tons, the average price being 19 per ton. England imports, on an average, 34,000 tons of copra ; France, 100,000 tons, of the value of 1,400,000 (the exports are 10, 700 tons) ; the figures for Germany are given above (see Palm-nut Oil). England imported 50,240 tons of crude coco-nut oil and 17,708 tons of the purified oil in 1909, and 53,968 tons of the crude and 50,021 tons (1,136,736) of purified oil in 1910. The United States imported 25,000 tons of the crude oil in 1910 and 43,000 tons (785,000) in 1911. Italy imported 344 quintals of coco-nuts in 1910 and also 20,225 quintals of coco-nut oil of the value of 66,340 (in 1908, 13,840 quintals). The Philippines exported 115,130 tons of copra in 1910, two-thirds of it to France. VEGETABLE TALLOW (Chinese Tallow) is obtained by pressing the fruit (separated more or less from the seeds) of Stillingia sebifera (tallow-tree), which grows in China, Indo- China, &c. Pressing of the seeds (3 per fruit) yields stillingia oil, which is to some extent drying (iodine number more than 135). The tallow, however, serves well for making soap and has an iodine number of about 30, but this varies somewhat owing to variation of the amount of stillingia oil present. The tallow melts at 35 to 44, and is sold in 40- to 50-kilo cakes wrapped in straw. COTTON-SEED OIL is obtained by pressing the shelled, washed seeds of the cotton plant (Gossypium herbaceum and barbadeuse, cultivated in North America, and G. religiosum, hirsutum, and arboreum, cultivated in Egypt, India, China, Siam, &c.). The crude oil is reddish brown (sulphuric acid produces a red coloration) and is decolorised by stirring with 6 to 10 per cent, of a caustic soda solution of 10 to 15 Be. and passing through it a vigorous current of air, first in the cold (40 to 50 minutes) and then when heated to 50 to 55 by indirect steam. It is then allowed to deposit, and is afterwards washed with 10 per cent, of salt water (at 10 Be.) to remove the last traces of soap, decanted" off, and passed through filter-presses to obtain it clear and of a fine straw-yellow colour. The fatty acids separated from the glycerides of cotton-seed cil contain about 26 per cent, of oleic acid, 47 per cent, of linolic acid (the oil is hence partly drying), and about 24 per cent, of saturated fatty acids (palmitic and up to 3 per cent, of a hydroxy-acid), besides a small proportion of an aldehydic substance (to which Becchi's reaction is due). It contains also 1-5 per cent, of a non-saponifiable sulphur compound and apparently a chloro-compound. Tests for the detection of cotton-seed oil in other oils have already been described (p. 397), and the analysis of the oil is carried out with reference to the constants given on p. 378. About two-thirds of all the cotton-seed oil is used directly or indirectly (as adulterant) as food ; the remainder (second and third qualities) serves, with palm oil and coco-nut oil, for making white soaps, although in some cases it gives rise, after some time, to yellowish spots. The world's production of cotton being about 3,300,000 tons, that of cotton-seed should be 6,600,000 tons, but in reality is only 5,000,000 tons (three-fifths in the United States), and the United States produce about 500,000 tons of cotton-seed oil (2,725,000 barrels in 1909 and 3,000,000 in 1910, one-fourth of this being exported) and 1,100,000 tons of cotton- seed cake. England produces about 70,000 tons of cotton-seed oil, imports 18,000 tons (1905) and exports about 18,000 tons. England imported 17,560 tons of cotton-seed oil in 1909 and 15,950 tons (562,672) in 1910, and also 690,000 tons (4,866,000) of cotton-seed in the latter year. In 1906, France imported more than 220,000 tons of the seed and 46,000 to 50,000 tons of the oil. The United States exported 85,000 tons (2,638,200) of cotton-seed 011 in 1910 and 155,000 tons (4,367,800) in 1911. Germany imported about 17,000 tons of the seed and about 55,000 tons of the oil in 1904. In 1906 Austria imported 20,500 tons of cotton-seed oil. Italy imported 31,328 quintals of the oil in 1907, 108,117 quintals in 1908, 306,250 quintals in 1909, and 35,801 quintals of the value of 117,430 in 1910. MAIZE OIL (in America, Corn Oil) is now prepared in large quantities in America and Italy from maize germs, which are separated during grinding. These germs contain 40 to 50 per cent, of oil, and after being pressed hot leave an excellent cake for cattle-food (10s. to 12*. per quintal). The dense oil has a fine golden yellow colour and a faint odour of maize, and serves well for soap-making and for adulterating edible oils and linseed oil. That obtained by extracting the dried grains from spirit manufacture (see p. 153) is reddish brown, and is used for burning and as a lubricant when mixed with olive and mineral 404 ORGANIC CHEMISTRY oils, but is not used alone as it tends to resinify. As a drying oil it has no great value. The fatty acids of the glycerides of maize oil are : stearic and palmitic (4 to 25 per cent.), oleic (about 40 per cent.), linolic and linolenic (about 45 per cent., so that the oil is partly a drying one), and small proportions of arachic, hypogaeic, caproic, caprylic, and capric acids ; the oil contains also about 1-2 per cent, of lecithin and 1-4 per cent, of non-saponifiable substances, mostly cholesterol or, more precisely, sitosterol, identical with that obtained from wheat and rye. If in North America (Illinois) alone the oil were extracted from the germs of all the maize produced (about 6,000,000 tons the world's total production being over 7,500,000 tons, 900,000 of this in Italy), more than 250,000 tons of the oil should be obtained. But only about 40,000 tons of maize oil are produced at the present time, about one-half of it being exported. SESAME OIL (Gingelly Oil, Teel Oil) is obtained from the seeds of Sesamum indicum (brown, oval, flat seeds, 4 mm. long, 2 mm. broad, and 1 mm. thick) and of Sesamum orientale (violet-brown or black), the latter giving as much as 50 per cent, of oil when pressed once in the cold and twice hot. The first oil expressed serves as a food for 250 millions of the inhabitants of India, where the area under sesame exceeds ten millions of acres (i.e. 40,000,000 hectares). The exportation of sesame seeds from India amounts to about 1,200,000 quintals annually, nearly all of this being directed to the Marseilles market, whence other countries are supplied. The Levant produces about one-tenth as much as India, and a little is produced in Africa, China, and Japan. In France the sesame oil industry is declining owing to the obstinate empiricism of the older manufacturers and to the almost prohibitive Customs duties of various countries, but more than 1000 truckloads of the oil are still exported per annum. Germany imported in 1890 only 140,000 quintals of the seeds, but in 1903 615,000 quintals, and in 1905 nearly 465,000. Austria-Hungary imports on an average 150,000 quintals yearly. Italy imported 174,722 quintals of sesame and arachis seeds in 1908, 309,000 in 1909, and 386,000, worth 617,400, in 1910. Sesame cake (dark or pale), so largely used as cattle-food, has the composition : water, 10 to 12 per cent. ; protein substances, 37 to 39 per cent. ; fat, 9 to 10-5 per cent. ; and ash, 9-5 per cent. Sesame oil has a golden -yellow colour, that from the Levant being the paler ; it consists of glycerides of stearic, palmitic, oleic, and linolic acids, 78 per cent, of the fatty acids being liquid with an iodine number of 140. The physical and chemical constants are given in the Table on p. 378, and the characteristic reactions for detecting it when mixed with other oils on p. 397. It is dextro-rotatory ( + 0-8 to + 2-4). The characteristic reactions, especially the colorimetric ones, are due to special components, such as sesamin ; a laevo -rotatory alcohol, sesamol, C 2G H 44 0, |H y O, which gives Baudouin's reaction (p. 384), and the methylene ether of hydroxyhydro- quinone, C 7 H 6 O 3 . Sesame oil is used in the manufacture of oleomargarine and soap and as burning oil. ARACHIS OIL (Earthnut Oil, Peanut Oil) is obtained from the seeds of Arachis hypo- gcea, cultivated in Brazil, the Congo, and India, and to some extent in Spain, France, and Italy. The shelled seeds give 30 to 35 per cent, of oil. In 1908 Italy imported 4735 quintals of arachis oil, in 1909 46,833 quintals, and in 1910 50,820 quintals, of the value of 182,960. The oil obtained by the first cold pressing is almost colourless, has a slight flavour of beans, and is largely used as a comestible and for adulterating olive oil, although it readily turns rancid. The second pressing in the cold gives burning oil, and the third, in the hot, oil for soap-making. The liquid components contain triolein and trilinolin ; the presence of hypogseic acid is uncertain ; the solid constituents are composed of triglycerides of ligno- ceric acid, and to a less extent of arachic acid (5 per cent, of the oil). In olive oil arachis 011 is detected by Renard's test, as modified by Tortelli and Ruggeri and by Fachini and Dorta (see p. 397). SOJA BEAN OIL (Chinese Bean Oil) is extracted from the beans of Soj'a liispida (or ftoja japonica or Phaseolus hispidus), which are cultivated in China and Japan (Formosa). The crushed beans are heated in jute bags over jets of steam and then pressed. A large part of the oil is used for soap-making. After purification by standing, the oil has a sp. gr. 0-9255 at 15 ; acidity, ; saponification number, 193-2 ; iodine number, 135 ; Hehner GRAPE- AND TOMATO-SEED OILS 405 number, 95-95 ; Reichert-Meissl number, 0-45 ; Maumene number, 86 to 87 ; index of refraction, 1-4750 at 20 ; solidification point, 8 to 16 ; melting-point of the fatty acids, 27 ; and solidification-point of the fatty acids, 22 (Oettinger and Buckta, 1911). The exportation of the oil from China amounts to 60,000 tons per annum. GRAPE-SEED OIL. The seeds of the grape contain 10 to 20 per cent, of oil (more in white and sweet grapes). They are separated from the skins by drying in the sun or in ovens and then beating. The sieved seeds are dried completely, ground, steeped in 10 per cent, of water, heated, and pressed ; the cake is broken up, treated with 20 to 25 per cent, of water, and pressed again, this treatment being repeated go that all the oil may be extracted. The oil can also be extracted by means of solvents (benzine or carbon disulphide). When dark-coloured (extracted with solvents), it can be readily decolorised with animal-black. It has not a very pleasant odour and is rather bitter (if expressed in the hot). . This oil consists of glycerides mainly of linolic acid, together with those of solid fatty acids (10 per cent.), and a little erucic, linolenic, and ricinoleic acids. It has the sp. gr. 0-9202 to 0-9350. It has slight drying properties and solidifies between 10 and 15; itsgaponificalion number is 178 to 180 ; iodine number, 94 to 96-5 ; Wollny number, 0-46 ; Maumene number, 52 to 54 ; and butyro-refractometer reading, 60 at 40. The acetyl number of the fatty acids varies from 43 to 144, according to the extent of oxidation ; it thus resembles castor oil to some extent, so that it is recommended for the manufacture of sulphoiicinate (see p. 327). The pure oil expressed in the cold is used as a food, and the other varieties for soap- making. But if purified with sulphuric acid it serves well as a lighting oil, not so much on account of its luminosity, which is rather low, but more especially because it gives a smokeless flame. After the removal of the fat, the cake contains 10 to 15 per cent, of water, 14 to 18 per cent, of protein substances, 8 to 10 per cent, of fat, and 6-5 to 7 per cent, of ash, and is used as cattle-food. In Italy the extraction of grape-seed oil is capable of considerable development. A few factories have already been erected in Southern Italy and in the North ; some of the works treat a certain amount of the seed. Seeds obtained from distilled vinasse are somewhat diminished in value. TOMATO-SEED OIL. In Italy 393,000 tons of tomatoes were produced in 1909 and 335,000 tons in 1910. In the province of Parma 84,000 tons are treated annually, and the residues (seeds, &c.) now yield 600 tons of oil (drying oil of the cotton-seed type). The refuse from tomato-ketchup factories (about 5 per cent, of the weight of the tomatoes) contains about 70 per cent, of aqueous liquid, 6 to 8 per cent, of dry skins, and 22 to 24 per cent, of dry seeds. One hundred kilos of tomatoes give 95 kilos of liquid juice, which is concentrated for preserve, and 1 per cent, of dry seeds containing 2 3 per cent, of oil, 18 per cent. (180 grms.) being extractable by pressure ; the remaining 820 grms. consists of cake (5-2 nitrogen, 12 per cent, fat, 22-7 per cent, cellulose, 21 per cent, non-nitrogenous extractives, 6-5 per cent, ash, 0-22 per cent, of dry skins, and 3-78 per cent, of aqueous liquid adhering to the moist residues). The oil expressed in the cold from sound seeds is straw-yellow, and with 20 per cent, of tallow gives a good washing soap. Analysis of the oil gives the following results (Fachini) : density at 15, 0-9215 ; refrac- tive index, 1-4765 ; acid number, 0-46 ; saponification number, 191-6 ; iodine number, 114 ; iodine number of the fatty acids, 122-7 ; iodine number of the liquid fatty acids, 142-2 j Hehner number, 93-8 ; acetyl number, 20-4. TREATMENT OF FATS FOR THE MANUFACTURE OF SOAP AND CANDLES Candles are mostly made from solid fatty acids (stearic and palmitic) obtained by decomposing fats and oils into glycerine and fatty acids and pressing from the latter the liquid fatty acids, which are used, either alone or together with the solid acids, for soap- making. Liquid oils and soft fats, which contain little stearic and palmitic acids, are 406 ORGANIC CHEMISTRY hence used not for candles but only for soap, but the stiffer fats are often treated in one and the same works for making candles and soap. The resolution of fats into acids and glycerine is carried out in very varied ways : by means of lime, sulphuric acid, superheated steam, or biological or catalytic methods. (1) Saponification with Lime and Separation of the Solid Fatty Acids. Theoretically 100 kilos of fat (see p. 377) require 9-5 kilos of lime for hydrolysis, but when this process was first used industrially by Milly in 1834 as much as 15 per cent, of lime was used, so that a very large amount of sulphuric acid was con- sumed in liberating the fatty acids from the calcium f oaps formed, while fatty acids were carried down by the enormous quantities of calcium sulphate formed and hence lost. On this account the process was not used, but Milly showed later (1855) that, by heating in an autoclave under pressure instead of in open pans, the amount of lime could be reduced to 2 to 3 per cent. that is, less than the theoretical quantity and yet practically complete saponi- fication effected (see p. 370). Indeed, after 1 hour 64 per cent, of the fat remained unsaponified ; after 2 hours, 24 per cent. ; after 4 hours, 15 per cent. ; after 6 hours, 9 per cent. ; after 9 hours, 2 per cent. ; and after 12 hours, 0-7 per cent. The saponification is now carried out in large vertical copper auto- claves (Fig. 268) (5 to 6 metres high, 1 to 1-2 metre in diameter, of sheet copper 15 to 20 mm. thick), into which are passed several quintals (up to 20) of the fused fat from the tank, A (Fig. 269), and then about one-third as much milk of lime, containing 2 to 3 per cent, of lime (calculated on the fat), from the vessel B, The heating is continued for 6 to 8 hours at a pressure of 8 to 10 atmos., steam free from air being passed in, first at low pressure from the generator, D, and then at high pressure (10 to 12 atmos.) by the tube, e (Fig. 270), reaching to the bottom of the autoclave and terminating in a perforated coil. The steam alone keeps the mass mixed without the special stirrers formerly used, if the precaution is taken of allowing a little steam to escape continually from a tap at the top. At the end of the operation the steam is shut off, and when the temperature has fallen to 125 to 130 (about 3-5 atmos. pressure) the internal pressure is utilised to discharge first of all the aqueous glycerine from below by opening the valve, c, connected with a tube reaching to the bottom of the autoclave. In a similar manner the fused and subdivided calcium FIG. 268. FIG. 269. soap mixed with free fatty acids is forced into the tank, E, where a further quantity of aqueous glycerine separates, or the calcium soap is passed directly to the lead-lined vessels, F, where it is decomposed by a sufficient quantity of sulphuric acid to neutralise all the lime added. 1 After shaking, the gypsum is deposited and can be separated, and the fatty 1 During recent years several factories have replaced lime by magnesia (calcined natural carbonate), which possesses various advantages : when it is used in the proportion of 1-5 to 2 per cent., a pressure of 4 to 5 atmo- spheres is sufficient to produce complete saponification, since the magnesium soap formed gradually emulsifies and almost dissolves in the remaining fat, which is thus easily resolved by the water and magnesia. Then, too, decomposition of the magnesium soap with sulphuric acid, instead of giving an insoluble and useless salt (calcium sulphate, which always retains a little fat), gives magnesium sulphate, which is soluble in water, readily separable HYDROLYSIS OF FATS 407 acids, which float, are washed several times with hot water and then, if the fatty acids are distilled as is done in certain factories where dark fats are treated they are forced by a pump, G, to the tank, H. The latter feeds a cast-iron or copper (this is considerably attacked) boiler, K, which is heated partly by almost direct-fire heat and partly by super- heated steam (at 180 to 230) passed into the interior from the superheater, J. The steam carries the fatty acids, which distil, into the tinned copper condensing coil, L ; these acids finally collect in a white condition, together with condensed water, in 8, while the non -condensed gases are evolved from the tube, M (see later : Decomposition with Sulphuric Acid). Where the fatty acids are not distilled, they are solidified by passing them into a number of superposed tin-plate pans (Fig. 271) fed by the tubes, D, from the fused fatty-acid tank, F. When all the pans are full, the tubes, D, are closed with wooden plugs, E, and in 24 hours many of the pans contain solid cakes, consisting of a mixture of solid stearic and palmitic acids and liquid oleic acid. In order to separate the latter, the cakes are wrapped in woollen or camel's-hair or goat's-hair cloths and are then placed between metal plates and pressed, first in the cold with a pressure gradually in- creasing to 200 to 260 atmos. A second pressing at 40, either in the same press or in a horizontal press, results in the almost com- plete separation of the oleic acid, which, however, retains in solution a little palmitic and stearic acids. The latter acids are separated by cooling the oleic acid and, after some time, filtering or decanting off the oleine (p. 298), which is then put on the market or used for soap- making. The solid white cakes of stearic and palmitic acids, freed from the dark edges, bear the commercial name of stearine and melt at 56 to 56-5. These are often melted again, washed with warm water, poured into pans to solidify, and then pressed hot in hydraulic presses so as to remove the final portions of oleic acids ; this product, known as double stearine, melts at 57-5 to 58. The solidification of the crude acids, after liberation by sulphuric acid, is now effected more rapidly and more perfectly by pass- ing the fused acids at g (Figs. 272 and 273) into a casing into which dips a large, rotating, double-walled cylinder. Between the walls flows a non -congealing solution like that from an ice machine (see vol. i, p. 231), and the layer of fatty acid solidi- fying at the surface is detached by means of a scraper, h, and falls into a cooled box, F, connected with the pump, P, and functioning as a filter-press. This process of the firm of Petit Fieres has now been improved by replacing the cylinder by a [ijft ? t T$\ i \ 7 highly cooled toothed wheel. In some cases, also, channelled cylinders are used, whilst in others the liquid fatty acids are FIG. 271. withdrawn from the cold pasty mass con- taining the mixture of liquid oleine and the stearine in small crystals, by immersing in the mass a rotating vertical cylinder formed of metallic gauze and covered with a well- stretched cloth ; inside the cylinder the pressure is reduced by means of a suction-pump, so that the liquid oleic acid is sucked in, while the stearic acid is gradually scraped from the surface of the cylinder and pressed in a hydraulic press. Messrs. Lanza Bros, of Turin, instead of separating the liquid from the solid fatty acids by means of hydraulic presses, suggest emulsifying and dissolving the liquid acids with solutions of sulpho -oleic acid, so that they separate at the surface, while crystals by simple decantation arid in some cases utilisable. For similar reasons, zinc oxide is now used in some of the Italian factories. Bottaro (1908) has suggested the use of sulphurous anhydride to decompose the calcium soap from the autoclave. ORGANIC CHEMISTRY of the solid fatty acids collect underneath (Ger. Pat. 191,238). The sulpho-oleic acid is prepared by shaking 100 parts of oleic acid with 50 parts of sulphuric acid of 66 Be. in the cold and then diluting with 4000 parts of water. The decomposition of fats by lime in an autoclave at not too high a pressure has the advantage of giving the fatty acids in a sufficiently clear condition to render distillation useless ; the resulting glycerine and stearine are also clear. 1 (2) Decomposition with Sulphuric Acid (proposed by Achard in 1777 and then by Fremy in 1836). This method is now used more especially for very dark fats, which should, however, be freed from impurities, dried by fusion at 120, and decanted after long standing. The fused fat is introduced into a double-walled, lead-lined, copper or iron boiler fitted with a hood for carrying off the sulphur dioxide which is always evolved. Accord- ing to the nature of the fat, it is heated with 5 to 10 per cent, of concentrated sulphuric acid at 120 for 1 to \\ hour, steam being passed through the jacket and the mass kept mixed by a current of air passing through it. The operation is finished when a test portion, placed on a dark plate, crystallises on cooling ; the mass is then passed into large wooden vats and heated with water until the emulsion first formed is re- solved into two layers, the gly- cerine below (this is separated FIG. 272. FIG. 273. and freed from sulphuric acid by means of lime) and the acids above. The latter is subsequently boiled several times with water until the excess of sulphuric acid is removed, the sulphuric ethers of oleic acid being decomposed with formation of solid hydroxystearic acid. The resulting fatty acids are dark in colour, since they retain in solution the impurities of the fat partially carbonised by the sulphuric acid ; to purify and whiten them, they are distilled with superheated steam, as described above (see also Fig. 269) ; the first and last portions which distil are the more coloured and these are redistilled. Hirzel (Ger. Pat. 172,224, 1906) has devised an arrangement for continuous distillation, all that is required being a boiler of moderate size into which the crude fatty acids are run in a constant stream ; the pure acids distil over, while the tar remaining at the bottom of the boiler is discharged. Redistillation of this tar gives a final residue of black stearine pitch, amounting to about 2 per cent, of the fatty acids distilled. In some works the fatty acids are distilled in a vacuum at a temperature not exceeding 240, higher temperatures than this giving a coloured product ; the acrolein and hydrocarbons given off are condensed. The fatty acids obtained by distillation are separated into liquid and solid by pressure in hydraulic presses, liquid distilled oleine and white, solid distilled stearine being thus obtained. This oleine always contains a little acrolein and hydrocarbons, as the crude fatty acids which are distilled invariably include a small proportion of non-saponified neutral fat. On the other hand, distillation results in the formation of an increased amount of solid fatty acids (about 15 to 18 per cent.), since sulphuric acid converts oleic 1 During recent years, industrial application has been made of the Krebitz process (Ger. Pat. 155,108, 1902), which is a simplification of the lime process with direct production of soda soap, and is attended by considerable saving in fuel, caustic soda, and plant. To the fused fat is added the necessary quantity of lime (10 to 12 per cent. CaO) mixed to a paste with three to four times its weight of water, the mass being well mixed, boiled for five minutes, covered, and allowed to stand overnight. By this means^saponification is complete and a calcium soap is obtained which can be readily ground up in a mill. When this is washed in a vat with a perforated bottom, the first portion of hot wash-water removes the major part of the glycerine as a solution of 10 to 20 per cent, concentration, while a second washing gives a more dilute glycerine solution which is used for the first washing of the calcium soap of a subsequent operation. When treated in the hot with sodium carbonate solutions, the calcium soap yields soda soap and calcium carbonate, which require skilled manipulation for their proper separa- tion. In this case also, fusion and treatment with hot water is employed for the complete removal of impurities. This process is not applicable to the manufacture of soft soaps. ENZYMIC HYDROLYSIS 409 acid partly into the corresponding sulphuric ether, which yields solid hydroxystearic acid when boiled with water : C 17 H 33 'C0 2 H + H 2 SO 4 .gQ jj + H 2 = H 2 S0 4 + Ci 7 H 34 <^Qj| During the distillation with superheated steam, the hydroxystearic acid is transformed almost entirely into iso-oleic acid (see p. 299). It must, however, be borne in mind that hydroxystearic acid is not Very good for making candles, as it accumulates in a fused state in the cup formed by the burning candle round the wick ; further, when melted with stearic acid it tends to separate in layers instead of giving a homogeneous mass. In order to obtain a greater proportion of solid fatty acids, some works combine these two systems of saponifying by means of lime and acid. The saponification is first carried out in autoclaves in the ordinary way, but not to completion, the acids and the remaining fat (4 to 5 per cent.) being then separated by means of sulphuric acid ; the fatty acids and fat are dried and completely saponified with 2 to 2-5 per cent, of concentrated sulphuric acid at a temperature of 110 to 120 maintained for an hour. The resulting fatty acids are not distilled but are simply washed with boiling water, being thus rendered rich in solid hydroxystearic acid ; this process also yields a much purer glycerine. L. Fournier (Fr. Pat. 262,263) has suggested a method of increasing the amount of solid fatty acids by effecting the sulphonation with concentrated sulphuric acid in a carbon disulphide solution of the fat, the reaction then proceeding immediately without heating. 1 (3) Hydrolysis by Hot Water under Pressure (proposed by Tilghmann in 1854) is but little used owing to the low yields obtained and the very high pressures required. The fat, emulsified with water, is circulated in coils arranged in a furnace so as to attain a temperature of 300 to 350. Direct distillation of fats with superheated steam and collection of the glycerine and fatty acids in the distillate always gives low yields. (4) The Biological or Enzymic Process has been applied industrially since 1902, as a result of the work of W. Connstein, E. Hoyer, and H. Wartenberg, and is based on the observations of Green and of Sigmund (1891) according to which, when oily seeds are pounded with water, fatty acids are gradually liberated by the action of lipolytic enzymes (see p. 112). It is found that the most active enzymes are those of castor oil seeds (in which they occur to the extent of 70 parts per 1000 of fat), especially after removal of 1 Transformation of Oleic Acid into Solid Fatty Acids. For some years (about 1877-1885), oleic acid was converted on an industrial scale in France and England (by the process of Olivier and Radisson) into solid palmitic add by utilising Varrentrapp's reaction, according to which this change is almost quantitative on fusion with solid caustic potash (see pp. 290 and 299) : C 1S H S1 O, + 2KOH = H 2 + CH,-CO 2 K. + C,,H 81 O Z K. But the greasiness and unpleasant odour of the candles obtained compared with those made from stearine, the necessity of distilling the resultant dark acid, and the difficulty of eliminating all the acetic acid, led to the abandonment of this process. Also de Wilde and Reychler's process for transforming oleine into stearine by heating in an autoclave at 260 to 280 with 1 per cent, of iodine or chlorine or bromine seems to have been given up in practice since 1890, the yield being less than 75 per cent, (the combined chlorine was eliminated by heating under 8 to 10 atmos. in presence of zinc dust or iron, and then decomposing the metallic soap). The industrial transformation of oleic acid into solid elaidic add by treatment with a little nitrous acid (see p. 299) does not give satisfactory practical results, first because elaidic acid is not a very good material for candle-making, and also because the reaction succeeds well only with fairly pure and fresh oleic acid and not with the commercial acid (partly polymerised). Max v. Schmidt treats 10 parts of oleic acid with 1 of zinc chloride at 180, then decomposes the zinc soap by boiling first with dilute HC1 and afterwards with water and finally distils the fatty acids, which can be separated into liquid and solid by means of hydraulic presses. By this process Beuedikt (1890) obtained 75-8 per cent, of stearolactone, C, 8 H3 4 O 2 (the internal anhydride of y -hydroxystearic add), 15-7 per cent, of iso-oleic acid, and 8-5 per cent, of other saturated acids. K. Hartl, jun. (Ger. Pat. 148,062, 1903), in order to avoid the browning produced by the action of sulphuric acid on the impurities of the oleic acid, does not treat the oleine directly with concentrated sulphuric acid (as had long been the custom ; see Shukoff, Ger. Pat. 150,798, 1902), but first distils the oleic acid in steam and after- wards treats it with sulphuric acid of 58 to 60 Be. (e.g. at a temperature of 60 to 80 and using 1 mol. of sulphuric acid per 1 mol. of oleic acid) ; the resulting fatty acids are then washed and decolorised by heating in open pans with 1 to 10 per cent, of zinc dust at 100, the zinc soap being finally decomposed by hot dilute hydrochloric acid. W. H. Burton (U.S. Pat. 772,129, 1904) uses a process similar to that of Fournier (see above), benzine or naphtha being employed as solvent and the sulphonic ethers being decomposed in solution by the direct action of steam. The general reaction of Sabatier and Senderens (see pp. 34 and 59) has also been applied practically (Ger. Pat. 141,029, 1902), a current of hydrogen being passed into the hot mixture of oleic acid and catalytic powdered nickel (reduced nickel) (see also E. Erdmann, Ger. Pat. 211,669, 1907) ; if the oleic acid is pure, it is transformed almost completely into stearic acid. A similar reduction, but with a lower yield, is obtained with the electric discharge (Ger. Pat. 167,107, 1904). A. Knorre (Ger. Pat. 172,690, 1903) treats an emulsion of oleie acid and formaldehyde with zinc dust. 410 the oil. But better results are now obtained by using aqueous emulsions rich in enzymes (extract of castor oil seeds), but much poorer in proteins (which are harmful) and con- taining 60 per cent, of water, 37 per cent, of castor oil, and 3 per cent, of proteins. When the seeds are used, a milky emulsion is obtained by crushing the seeds in presence of the necessary amount of water (50 to 60 per cent.) and is decanted off roughly from the skins and treated with 0-06 per cent, of acetic acid (calculated on the weight of fat to be decom- posed subsequently). Of the seeds or the enriched extract, 50 to 89 kilos are used per 1000 kilos of fat (the maximum for fats with the higher saponification number ; although tallow requires the maximum amount and a temperature of 40). To accelerate the decomposition, 0-15 to 0-20 per cent, (on the weight of fat) of manganese sulphate (acti- vator) dissolved in a little hot water is added, and if the fat contains much protein or gummy matter, it is well to clarify it by heating with 1 per cent, of sulphuric acid diluted with a little water ; the last traces of this acid are then removed by repeated and thorough washing with water, as they would be deleterious to the reaction. With liquid fats, the decomposition is carried out at 23 and with solid ones at 1 to 2 above the melting- point, provided however that this does not exceed 42, since at 44 the enzymes no longer act in the desired direction ; if necessary, fats with high melting-points are mixed with liquid oils. The practical working of the process is as follows : A leaden coil for indirect steam and a tube for the injection of air reach almost to the bottom of a lead-lined iron boiler with a conical base ; discharge cocks are fitted to the boiler at the bottom and at various heights. The fat and about 35 per cent, of water are heated to the desired temperature (see above), being kept stirred by means of a current of air. The castor-seed extract, mixed with 0-2 per cent, of manganese sulphate and 0-06 per cent, of acetic acid (on the weight of fat ; the reaction starts and proceeds well if the mass is faintly acid at first) is then added, the whole being mixed for about 15 minutes so as to give a homogeneous emulsion. The vessel is then tightly covered so that the temperature may be maintained, the mass being mixed from time to time to keep it emulsified. After 24 to 36 hours, when more than 90 per cent, of the fat is decomposed, the mass is mixed and heated to 80 to 85, 0-2 to 0-3 per cent, (of the weight of fat) of concentrated sulphuric acid (66 Be.) diluted with one-half its weight of water being then added. The whitish emulsion soon becomes dark owing to the separation of the fused fatty acids and when this occurs the heating and stirring are suspended and the mass left overnight. The various taps are then set in operation to separate the bottom layer of fairly concentrated glycerine, the intermediate emulsified layer (3 to 4 per cent, of the fatty acids, used for soap- making) and the clear fused fatty acids which are boiled with water to free them from sulphuric acid. Originally, when the seeds were used instead of the extract, the resulting glycerine was very dark, and it was necessary to decolorise it with bone-black (nowadays it is as good as that given by saponification with lime), while the intermediate emulsified layer formed as much as 22 per cent, of the total fatty acids (now only 2 to 4 per cent.). The aqueous glycerine (sweet water) of the enzymic process is first concentrated to 10 Be. in open pans, the sulphuric acid being separated by means of barium carbonate in the hot. The barium sulphate is removed by filter-pressing and the filtered liquid further con- centrated in a multiple-effect vacuum apparatus to 28 Be., a clear, brownish glycerine containing only 0-2 to 0-4 per cent, of ash being thus obtained. The biological process has spread rapidly during recent years, since the whole of the glycerine is readily recovered, while the fatty acids obtained are of far better quality than those prepared by decomposing the fat in autoclaves by means of lime, &c. The fatty acids from sulphocarbon olive oil retain, however, their characteristic green colour, and those from palm oil their orange colour. The fatty acids yielded by this process contain neither hydroxy-acids, as do those obtained under pressure, nor calcium soaps, and are hence more suitable for the manufacture of either candles or soap (see later, Soap). (5) Twitchell's Catalytic Process. The decomposition is here analogous to that with sulphuric acid (which also, strictly speaking, is catalytic), but with TwitchelVs reagent (benzenestearosulphonic acid) it takes place far more readily probably because this reagent dissolves in the fat more easily than does sulphuric acid. The fats are first purified by heating to 90 to 100 in a lead-lined covered vat (Fig. 274) with 1-5 to 2 per cent, of sulphuric acid at 60 Be., direct steam being passed in so that when the acid is discharged after standing overnight it has a specific gravity of 8 Be. (for cotton-seed or linseed oil, TWIT C HELL'S PROCESS 411 15 Be.). The purified fat is passed into another wooden vat, B, provided with a wooden cover, one half of which is removable ; it is here mixed with 20 per cent, of distilled or condensed water (from the tank G), the mixture being then boiled by direct steam and 0-5 to 0-15 per cent, of the Twitchell reagent added (the minimum with pure fats and the maximum with highly impure third-grade fats). The current of steam is continued so that a homogeneous emulsion is rapidly obtained, and after being heated in this way for 24 hours about 90 per cent, of the fatty acids are liberated and the glycerine separated. Xo more steam is then passed through the mass, but a slow jet is kept flowing into the space between the surface of the liquid and the cover to prevent the fatty acids from turning brown during the subsequent operations owing to contact with the air. In about an hour's time, the emulsion breaks up and the fatty acids float on the aqueous glycerine ; if the emulsion should not disappear, it is mixed gently for a few moments with 0-1 to 0-2 per cent, of sulphuric acid of 60 Be. and then left. The sweet water usually has a specific gravity of 5 Be. (15 per cent.) and forms 50 per cent, of the weight of the fat, and if this is not the case, the quantity of distilled water added initially and the dryness of the steam employed are varied when further quantities of fat are treated. The sweet water is neutralised with lime and concentrated (see p. 185). For soap-making the fatty acids may be used as they are, but as a rule the saponification is com- pleted by adding 10 percent, of pure water and heating for 12 to 24 hours with direct steam, any small amount of emulsion formed at the surface of the liquid by the steam being destroyed by the addition of a little sulphuric acid. In this way, 97 to 98 per cent, of the theoretical amount of fatty acids is obtained. Barium carbonate (1 part per 10 parts of Twitchell's reagent used, or more if sulphuric acid were added to destroy emulsion), mixed with a little water, is now added, and the whole heated for 15 to 20 minutes ; if the lower layer of water now has an acid reaction towards methyl orange, more barium car- bonate must be added. The current of steam, both in and above the liquid, is now stopped, since after this the fatty acids are no longer turned brown by the air. The sweet water drawn off after clarification is very dilute and is used in place of water in the treatment of further quantities of fat. After crystallising and pressing to separate the solid from the liquid acids (see above), the fatty acids are now ready for converting into soap and candles. In general they are less coloured as the amount of Twitchell's reagent used and the duration of its action are diminished. Good results are not obtained until after five or six operations, by which time the surface of the wooden vessels ceases to be attacked. Just as with the preceding process, the use of the Twitchell process has spread con- siderably in America and in Europe. 1 The Twitchell reagent (which costs about 1*. 2d. per kilo) and estimates for the plant may be obtained directly from Messrs. Joslin, Schmidt & Co., 3223 Spring Grove Avenue, Cincinnati, Ohio, or from their representatives in various countries. 1 The plant for a factory using the biological or catalytic process is considerably less expensive than for one employing autoclaves, while there is also a decided economy in the working expenses, as is shown by the follo.wing approximate figures, which show that these processes are of value, at any rate in countries where coal is dear. These data are from a large factory using the Twitchell process and treating about 70,000 kilos of fat per day 10.000 kilos at a time in each apparatus. The prices given are those current in Italy, and the cost is calculated for 100 kilos of fat treated ; the figures in brackets give the corresponding cost for the autoclave method : coal at 40 lire (32s.) per ton, 0-20 lira (0-82 lira) ; sulphuric acid, 0-09 lira (0-37) ; baryta or lime, 0-06 lira (0-11) ; labour, 0-03 lira (0-04) ; depreciation and repairs, 0-02 lira (0-26) ; Twitchell reagent, 0-80 lira. Hence the total cost of treating 100 kilos of fat will be at most 1-20 lira (lljrf.) with the Twitchell process and at least 1-60 lira (15-4rf.) with the ordinary autoclave process. In the case of small plants, the cost of working increases some- what with the Twitchell process, but there is always an advantage owing to the less initial outlay required. FIG. 274. 412 ORGANIC CHEMISTRY MANUFACTURE OF CANDLES x The prime materials for the manufacture of candles are the combustible fatty matter and the wick. A good candle should give a white light, should burn slowly, should not "gutter " or diffuse an unpleasant smell, should not be greasy to the touch, should be white and give a smokeless flame, and should not splutter, while the relation between the size of the wick and that of the candle must be properly chosen. The object of the wick is to feed the flame regularly with the melted material. It is usually made of filaments (15 to 20) of pure cotton or linen without knots. Animal fibres should be rejected, as they give an unpleasant smell and a fused carbonaceous mass which diminishes the luminosity. Wicks formed of filaments which are only twisted require frequent snuffing, since they do not bend on themselves and do not burn completely, whilst, if they are plaited or woven and twisted, as Cambac^res proposed, this incon- venience is overcome. For stearine candles obtained by fusion, the wick is of twisted cotton braid, while for more readily fusible materials (wax, tallow, &c.), more or less twisted wicks are used according as the candles are made by fusion or by compression. Nowadays wicks are made with suitable machines like those used for knitting, the^e effecting also the twisting of the filaments. Wicks which have not been pickled do not act well for candles, as they leave a carbonaceous residue which diminishes their capillary property. In 1830, Milly found that the combustion of the wick is facilitated by steeping it in a solution of boric or phosphoric acid, such treatment being, however, only of advantage with braided wicks. Many other substances have since been proposed for this purpose. Thus, in France the wicks are immersed for 3 hours in a solution of 1 kilo of boric acid in 50 litres of water, and are then pressed, centrifuged, and dried ; in some cases a trace of sulphuric acid is added to the bath. In Russia, the wick is left in a solution of sulphuric acid (50 grms. per litre), squeezed, dried in hot air, steeped in a bath containing 4-5 grms. of boric acid and 18 grms. of ammonium sulphate per litre of water, and then dried. Another solution giving good results is composed of 60 grms. of borax + 30 grms. KC1 + 30 grms. KN0 3 , + 30 grms. NH 3 + 3-5 litres of water. The borax renders the flame white. In general these products either induce a more ready oxidation (chlorates, nitrates) or melt the ash of the wick, which thus gradually falls by its own weight. In some cases the penetration of the solution into the wick is hastened by the addition of a little alcohol. If the candle is too large in comparison with the wick, the excess of stearine melts and forms a kind of cup with tall sides full of the fused stearine, which cannot be completely absorbed by the wick and so makes the flame smaller ; then, when the edges fall, the stearine overflows and produces guttering. If, on the other hand, the wick is too large, an insufficient quantity of wax is melted and no cup is formed to contain it, the candle guttering continually from the sides and the flame being less luminous. 1 The ancient Romans used for illuminating purposes a kind of torch steeped in wax or bitumen. Only after the second century of the Christian era was a distinction drawn between wax candles and those of tallow ; the use of the latter was regarded as a luxury, while wax candles were employed in churches. The Catholic religion used them exclusively for religious functions, and thus caused a great increase in the consumption, which diminished only after the spread of the Reformation. Very soon, however, the consumption of wax candles again increased very considerably owing to their extended use at the courts of kings and princes. Meanwhile the employment of tallow candles for domestic purposes was continually spreading, and in the eighteenth century several important factories were working in England ; but the candles produced were high in price and burned very quickly. Only after ChevreuTs work on the nature of fats in the early part of last century (after 1815) led to improvements in the saponiflcation and to the preparation of solid fatty acids was the rational manufacture of candles initiated. Chevreul himself, together with Gay-Lussac, patented in 1825 a process for preparing candles from stearic acid ; but the resulting industrial undertakings were soon abandoned, owing to the diffi- culties encountered in the saponiflcation and in the preparation of the wick. It was only when Cambaceres, in 1830, devised plaited and twisted wicks, and when Milly, in 1834, kitroduced saponification with lime and the subsequent decomposition of the calcium soap with sulphuric acid, that the manufacture was placed on a stable and remunerative basis. Milly's first factory for stearine candles was erected in Austria in 1837, and in 1840 one was started in Berlin and another in Paris. Important improvements were made in 1842 by saponifying the fats with sulphuric acid, and in 1854 by saponifying the fats and distilling the fatty acids with superheated water or steam (processes of Tilghmann, Berthelot, and Melsen). Almost immediately after this, however, the manufacture of paraffin candles was started, paraffin having been obtained in large quantities by Young (1850) by the dry distillation of bituminous coal (boghead, &c.), peat, shale, lignite, &c. ; this industry underwent further extension after paraffin had been extracted from petroleum and ozokerite (see p. 80). MANUFACTURE OF CANDLES 413 In 1904 a patent was filed for the manufacture of artificial silk candle-wicks, which seem to give good results. Formation of the Candles. The white blocks of stearic and palmitic acids from the presses are scraped at the surface and edges to remove adherent impurities. The purer residue is melted and shaken in a leaden vessel with sulphuric acid (3 Be.) to dissolve and separate the impurities (iron, hairs from the press bags, &c.) ; the sulphuric acid is then decanted off and the stearine washed repeatedly with boiling water to remove all trace of the mineral acid. In some cases the fused fatty acids are shaken with a little albumin, being then allowed to stand so that the coagulated albumin and the impurities may settle. In cooling, the stearine tends to crystallise, the resultant candles being then less homogenous and more brittle, At first arsenious ami acid was used to prevent crystallisation, but, now that this is prohibited, mm the stearine is kept continually shaken until it almost solidifies when it is introduced into the moulds, and the candles then rapidly solidified. It is often more convenient to add a little white wax or paraffin (2 to 10 per cent.), which also prevents crystallisation of the stearine. The quality and purity of the stearine are ascertained by the usual tests, the neutral fat being determined by Geitel's test (see p. 379), the paraffin, cerasin, cholesterol, and carnauba wax by the saponification number and by the non-saponifiable matter (see p. 379), and the amount of oleic acid by the melting-point (which is 56 to 56 -5 for pure stearine pressed once and 57-5 to 58 for doubly pressed stearine) and the solidification point, making use of de Schepper and Geitel's Table 1 obtained by mixing saponi- fication stearine, solidifying at 48, with oleine having a solidifying point of '5-4. Candles are made in three different ways: (1) by immersion; (2) by p IG> 275. fusion ; and (3) by pressure. The first of these methods is the oldest and is now almost entirely abandoned. It was employed originally for tallow candles, and is now sometimes used to mask the presence of inferior fat or stearine, the wicks suspended from frames being first immersed in the impure fused fat, while the outer layers are obtained by dipping into a purer fat or fatty acid. In China considerable use is still made of tallow candles of peculiar shape with a hole in the middle. Certain long tapers are obtained by pressure, the semi-fused wax or stearine and the wick being forced through a tube. But almost all candles are now made by fusion in highly perfected machines, which admit of a maximum output being rapidly obtained with a minimum of labour. The moulds, which are very smooth inside, have the shape of the candles with the pointed end below and the enlarged base at the top (Fig. 275) and are imperceptibly conical ; they are made of an alloy composed of 3 parts of tin and 1 part of lead. For the fusion of a large number of candles at a time (100 or more) a machine is used similar to that shown in Fig. 276. The moulds of all the candles pass through the closed metallic box, E D, to the bottom and cover of which they are screwed. Tepid or cold water can be passed 1 De Schepper and Geitel's Table of the solidifying points of mixtures of fatty acids : Tempera- Per cent. Tempera- Per cent. Tempera- Per cent. Tempera- Per cent. ture of of ture of of ture of of ture of of solidification stearine solidification stearine solidification stearine solidification stearine 5-4 16 7-7 27 21-7 38 50-5 6 0-3 17 8-8 28 23-3 39 54-5 7 0-8 18 9-8 29 25-2 40 58-9 8 1-2 19 11-1 30 27-2 41 63-6 9 1-7 20 12-1 31 29-2 42 68-5 10 2-5 21 13-2 32 31-5 43 73-5 11 3-2 22 14-5 33 33-8 44 78-9 12 3-8 23 15-7 34 36-6 45 83-5 13 4-7 24 17 35 39-5 46 89-0 14 5-6 25 18-5 36 43-0 47 94-1 15 6-6 26 20-0 37 46-9 48 100-0 414 ORGANIC CHEMISTRY at will through the box at / or H, so as to surround the moulds. The lower part of each mould contains a kind of small piston which has exactly the shape of the point of the candle and can be made to traverse the whole length of the mould, being joined to an iron tube, B, fixed to a frame capable of being raised and lowered by the rack and pinion, C. All the pistons can be raised at once so as to force all the solidified candles from the moulds. In order that the wick may be always in the middle of the candle, it is wound on bobbins, A, and passes through the iron tube which raises the piston to the upper part of the mould. The semi -fused, opalescent stearine, which is poured into the moulds kept by means of warm water (45 to 60) at a temperature slightly above the melting-point, is then cooled by passing cqld water round the moulds. When solidification is com- plete, the enlarged bases at the top of the candles FIG. 276. are cut off by a knife and the candles forced out and grasped by the rods, L. In rising, the candles unwind from the bobbins new wicks which are thus brought into the middle of the moulds ready for the next operation. When the second batch of candles is solidified in the moulds, the wicks of the first batch are cut so as to make way for the others to be removed from the moulds. When shorter candles are required, the pistons are raised in the moulds to the desired height and the stearine then run in. The candles thus obtained are bleached by arranging them ver- tically on trucks in metal gauze frames and leaving them for some days in the open air exposed to the action of the air, sunlight, and dew. After this, the candles are washed, polished, and sawn off to a uniform length in a machine of the Binet type (Fig. 277). The candles are first dipped in a bath, V, containing soapy water or a dilute solution of soda, and are then placed in the grooves of the wheel, M, the head being against the left-hand edge, while the bases are cut off by a small circular saw, TO ; the fragments drop on to the frame, X, and so into the box beneath. The candles fall into the grooves of the travelling endless plane, TM', and are rubbed and polished by a brush, B, moved excentrically from V ; when they reach M' they fall into the trough, E. The finished candles are stamped auto- FIG. 277. SOAP 415 matically with the trade mark and are then tied and wrapped up in packets of 12 or 24 (or I or 1 kilo) and placed in wooden boxes for transport. Some factories make lighter, perforated candles and some coloured candles or mixed candles containing wax or paraffin. To remove the semitransparency of paraffin candles and so make them resemble those of stearine, about 5 per cent, of stearine and 5 per cent, of paraffin oil are added. The same effect may be obtained with a small quantity of /7-naphthol (Ger. Pat. 165,503) or any other substance which dissolves the paraffin in the hot and deposits it in the cold in a finely divided state (e.g. solid fatty acids, amides, phenols, ketones, &c.). STATISTICS. The value of the stearine candles exported from England was 346,400 in 1892, 400,000 in 1900, 495,860 in 1909, and 485,220 in 1910, the output in 1907 being 10,000 tons, of the value of 1,640,000. France exported 25, 120 quintals of stearine candles, worth 196,000, in 1890, and 43,300 quintals, of the value of 180,000, in 1900. In 1909 Germany imported 2200 quintals of candles and exported 7880. The United States exported 1500 tons (58,200) in 1910 and 1450 tons (55,600) in 1911. In 1903 there were about 250 factories in Italy for candle-making only, the total horse-power of the engines being 185 and the number of operatives 1430 ; there were, in addition, 188 factories for both candles and soap, employing 2700 workpeople and using 830 h.p. In 1876 there were but 10 factories with 550 operatives. In 1875-1879 Italy imported on an average 6350 quintals of candles per annum and exported 650 quintals, whilst in 1900-1904 the imports averaged 551 and the exports 1420 quintals annually. In 1905 the imports were 869 quintals, worth 4172, and the exports 614 quintals, of the value of 2948 ; in 1910 the exports fell to 582 (2920) and the imports to 380 quintals (1900). MANUFACTURE OF SOAP 1 Theoretically soaps include all metallic salts of the higher fatty acids, but practically the name is given only to salts of oleic, stearic, and palmitic acids, and, in general, of the fatty acids contained in natural oils and fats. Importance attaches mainly to the sodium soaps and, to a less extent, to those of potassium and ammonium. It was at one time thought that soaps were composed largely of margaric acid, but it has been shown that this acid does not occur in natural fats, the confusion arising from the fact that a mixture of palmitic and stearic acids was obtained with a melting-point identical with that of synthetic margaric acid (see p. 290). Almost in its entirety soap is used for washing and for cleansing and removing grease from textile fibres, sweaty garments, and the greasy, dirty 1 History of Soap. Soap was not known to the ancient Hebrews and Phoenicians or to the Greeks of the time of Homer, who washed their garments with the ashes of plants and water, and by mechanical rubbing. Some races used the juices of certain plants, and later it was discovered that when ashes were heated with lime they gave rise to natron, which was much more effective than the ashes themselves. Yet the writers of the Bible, who are certainly not conscientious and exact historians, several times mention soap and quote the following supposed phrase of the prophet Jeremiah (who would have lived several centuries before the Christian era) : " Though thou wash thee with nitre [natron] and take thee much soap, yet thine iniquity is marked before me." Seneca and Pliny mention soap in their writings and attribute its discovery to the Gauls, who prepared it from the ashes of plants and goats' fat and used it as a hair-wash and for medicinal purposes (lead plaster). It is said that Galen (second century of the Christian era) proposed the use of soap for washing. In the excavations of Pompeii has been found a complete soap factory with utensils and saponified material. Marseilles did a large trade in soap as early as the ninth century, but in the eleventh century it had a serious rival for the premier position in Savona. In the fifteenth century the industry flourished at Venice, and in the seventeenth at Genoa, which, together with Savona, Marseilles, and Alicante, enjoyed a monopoly in soap-making. In England the industry began to develop after 1650, and in Germany it assumed considerable importance after Chevreul's investigations on fats (1810-1823). With the development of the soda industry and increase of the trade in palm oil and coco-nut oil, the conditions in Germany and, to some extent, in other countries favoured extension of soap-making. At the present time Marseilles, although partly surpassed by the large English factories, still preserves its early fame, which, however, the Italian factories have lost. But several times in the past the renown of Marseilles has been dimmed owing to the custom, even in the early days, of adulterating soap and of loading certain qualities of white soap with enormous quantities of water. This explains why. for several generations, the public preferred mottled soaps, which could not then be adulterated. It explains also the various laws promulgated in France against dishonest soap-makers, who in 1790 provoked a general protest of all the population and a petition to the deputies of the States General from all the laundresses of Marseilles to protest " against the adulteration of white soap and against the, malefactors who adulterate it to increase its weight." It does not appear that things have changed greatly after the lapse of 120 years, for, since the introduction of palm oil and coco-nut oil in 1850, the consumer has always paid for a considerable amount of water in place of soap. 416 ORGANIC CHEMISTRY epidermis of the human body, but it is sometimes employed as a subsidiary dressing in certain industrial operations, e.g. in the dyeing of silk and cotton, &c. The theory of the saponification of fats has already been discussed on p. 377, and we shall here consider the cleansing action of soaps. It is well known that the quantity of fat or grease that a soap is able to remove from a dirty garment is greater by far than corresponds with the amount of alkali liberated on dissolving the soap in water. Being formed from weak acids, soap in dilute aqueous solution is undoubtedly partly dissociated into caustic alkali and either acid soaps in the cold or fatty acids in the hot. This can readily be shown by the opalescence of the dilute aqueous solutions and by the violet colour imparted to phenolphthalein by a perfectly neutral (i.e. not yet dissociated) alcoholic solution or highly concentrated aqueous solution of soap, after pouring into a large quantity of water. If, then, part of the grease can be rendered soluble by the saponifying action of the alkali gradually liberated from the soap, another part is certainly carried away mechanically by the emulsifying action of the soap itself and of its fatty acids ; this action is accompanied by the abundant production of lather, which, together with the water, incorporates and removes all the grease with which it comes into contact. It is for this purpose the formation of lather and emulsification of the grease that rubbing is necessary in the washing of a garment with soapy water. A mere solution of caustic soda, even in excess, does not produce a detergent effect equal to that of soap. As regards the molecular condition of soap in its concentrated, non -dissociated solutions, it appears demonstrated that it there exists in a colloidal condition, since an increase in the concentration is not accompanied by rise in the boiling-point, which approximates to that of water, while the electrical conductivity is minimal. But, according to McBain and Taylor (1910), in highly concentrated solutions soap is apparently not a colloid, as it conducts the electric current. The solubility in water of almost all soaps is diminished rapidly to the point of complete separation by the addition of soluble salts which do not decompose the soap, e.g. NaCl, KC1, Na 2 S0 4 , NH 4 C1, Na 2 C0 3 , and even NaOH, &c., this action being due to a change in the density of the solution and in its degree of dissociation. This phenomenon is the basis of the salting- out or graining of soap during its manufacture, but it must be noted that if the fats or fatty acids used in the making of the soap contain hydroxy -acids, these are almost entirely lost, as they are not separated as insoluble soaps by salting out, and mostly pass into the spent ley. Hence account is now taken of the proportion of fatty hydroxy -acids (less soluble in benzine than ordinary fats or fatty acids) present in fatty materials. Sodium soaps are more stable than those of potassium or ammonium, since sodium salts partly displace potassium or ammonium from their soaps with formation of sodium soaps. Alkali soaps are precipitated by the soluble salts of the alkaline earths and heavy metals in the form of insoluble metallic soaps. Strong acids separate the weaker fatty acids from soaps. The alkaline soaps are usually soluble in alcohol and insoluble in ether, benzine, or benzene. Evaporation of the alcoholic solution yields a trans- parent soap. Saponification of fats is accompanied by increase in weight, each molecule of glyceride that decomposes fixing 3 molecules of alkali or water. A fat containing a mixture of glycerides with a mean molecular weight of 880, in reacting with 120 of NaOH (3 mols. or about 13-6 NaOH per 100 of fat), gives 92 of glycerine and 908 of water-free soap. So that theoretically 100 kilos of fat can produce about 10-5 kilos of glycerine and 102 of soap ; in practice about 1-5 to 2 kilos of glycerine are lost, while 140 to 160 kilos of soap, containing a considerable amount of water, are obtained. Potash soaps are softer than those of soda, and soaps of liquid fatty acids softer than those of solid fatty acids. STAGES OF SOAP- MAKING Soap may be made either from the fatty acids obtained from fats by the methods described above, or from the fats themselves. In the former case the saponification is carried out mainly by sodium carbonate, and is completed (since with the carbonate it proceeds only to the extent of about 90 per cent.) by caustic soda. But in the latter case concentrated solutions of caustic soda in the hot are employed ; the carbonate is, indeed, unable to resolve glycerides, and that amount of it which always occurs in the caustic alkali is lost during the subsequent operations of salting-out, &c. Mention has already been made (see p. 379) of the process of decomposing fats in an autoclave by means of ammonia and sodium chloride, which was studied by Leuchs (1859), Witelw (1876), Buisine (1883), and Polony (1882)', and improved by Garelli, Barbe, and de Paoli (Ger. Pat. 209,537, 1906). This process leads directly to the sodium soap with formation of ammonium chloride, from which the ammonia may be recovered in the usual way, and, according to the above patent, gradual decomposition of the ammonia by means of steam results in a considerable separation of the solid fatty acids from the liquid ones, the ammonia soaps of the former being the first to decompose. Such separa- tion can be effected also by cold water, which dissolves the ammonia soaps of the liquid fatty acids (oleates) almost exclusively. In the manufacture of soaps from fats or oils, various stages are to be distinguished : (1) mixing or pasting of the fat with the alkaline lye ; (2) mixing in the hot to form the soap and separate it partially from the excess of water ; (3) salting-out (or " graining " or " cutting the pan '') to render the soap insoluble and separate it from the lye, which thus collects under the layer of soap ; (4) boiling to saponify the last traces of fat, to eliminate the scum and the excess of water still remaining in the soap and to collect the latter into a perfectly homogeneous, curdy mass ; (5) the soap is often subjected to a finishing process, that is, a final treatment with dilute alkali hydroxide or carbonate solution, in order to separate the more thoroughly the residual impurities (aluminium or iron soaps) and so avoid a partial mottling, and to give to the soap, first, the quantity of water necessary to the particular type, and, secondly, a still more homogeneous appear- ance. 1 A well-finished soap contains 35 to 40 per cent, of water and only 0-20 to 0-36 per cent, of salt and free alkali together. When excess of free caustic soda remains in the soap, considerable efflorescence, due to formation of sodium carbonate by the carbon dioxide in the air, occurs at the surface during the subsequent drying. In order to avoid such a serious inconvenience, it is necessary to treat repeatedly with sodium carbonate solutions, because, even if a little of the latter is left in the soap, only a slight powder forms at the surface on drying and this can be readily eliminated. In some cases, a small proportion of a non-saponifiable fat (e.g. \vool fat) or even of a dense mineral oil is added to the soap, the caustic soda being thereby preserved from direct contact with the air. At one time the coppers used for soap -making were largely made of masonry, but nowadays they are almost universally of iron and are heated either by fire or by direct or indirect steam, as is shown in Figs. 278, 278A, 279, 279A. Small coppers hold 10 to 50 hectols. and large ones 100 to 400. For every 100 kilos of fat to be saponified, a copper-volume of 500 litres is taken. In most soap-works the mixing is done by wooden blades worked by hand, although coppers are made fitted with stirrers of various forms. The saponification of 100 kilos of fat or oil requires theoretically about 136 kilos of NaOH, but practically rather more than this amount is used. Tallow soap is made in the 1 Finishing is best effected when the soap contains a certain proportion of water, namely, 10 mols. of water (40'5 per cent.) per 1 mol. of sodium oleate, or 16 mols. (48-5 per cent.) per mol. of sodium stearatc. If the soap is more concentrated than this, it remains too viscous and opposes too great a resistance to the precipitation of the impurities and of the drops of saline and caustic solutions ; but if, in the finishing, the necessary quantity of water is restored (by adjusting the concentration of the lye), a small part of the soap dissolves, the mass becomes more liquid and, on standing, the impurities are able to fall to the bottom the more readily. Soaps which are too insoluble in the salt solution or caustic lye (colza, sesame 1 , linseed, poppy -seed, &c.) can be finished only when mixed with readily soluble soaps (coco-nut, castor, &c.). On the other hand, it is necessary to prevent the soap taking up too much water, for, if this happens, it pastes together and adheres to the sides of the boiler, does not transmit heat readily to the interior and hence boils with difficulty, is not easily finished and becomes uneven. Agitation of the mass and the consequent inclusion of a considerable amount of air are to be avoided, the finishing being thereby retarded. When the finishing is complete and the mass has been allowed to stand, a slight frothy layer is observed at the surface and then comes the thick layer of pure, homogeneous soap, well separated from the lye ; but above this is a small, irregular, and gelatinous layer composed of more soluble soaps (of hydroxy- acids) of calcium, magnesium, and iron, and of certain other impurities insoluble in the lye (colouring-matters, coagulated proteins, &c.), and it is this aiasu which forms the refuse. 11 27 418 ORGANIC CHEMISTRY following manner : The tallow is mixed and gently heated in the copper with about one- fourth of the necessary amount of caustic soda in the form of a solution of 10 Be. First of all an emulsion is formed and then saponification gradually proceeds, the mass beginning to become homogeneous and the volume increasing slightly. When a little is removed on the blade, it forms a jelly which does not separate the lye, and the soap-boiler judges FIG. 278. FIG. 278A. of the fixation of the alkali by observing when the caustic taste of the alkali disappears. Much of the fat remains unsaponified, so that a hot caustic soda solution of 12 to 14 Be. is gradually added until the stirred, boiling mass thickens, becomes clear and homogeneous, and falls from the spatula in transparent ribbons. At this stage, in order to judge if the alkali has been added in the proper proportion, a little of the soap is poured on to a glass plate ; if a solid white edge first forms round the drop of soap, while the interior of the FIG. 279. FIG. 279A. mass remains transparent until solidification is complete, the whole of the fat is saponified and there is no excess of alkali. But if the edge immediately turns greyish and the mass turbid, non -saponified fat is present and alkali lacking ; whereas, if the whole mass becomes covered with a whitish pellicle without previous formation of a solid edge, excess of alkali is present, this being corrected by adding a little fused tallow to the mass in the copper. Thus treated, the gluey paste, which has a slightly caustic taste, is boiled more strong] v until it loses sufficient water to form a homogeneous ropy paste on the mixing -blade, " OLE INE" SOAPS 419 At this stage the separation of the soap from the liquid is induced by the gradual addition of salt either in the solid state (4 to 8 per cent, of the weight of fat) or in con- centrated solution (20 to 22 Be.). The first addition of salt renders the mass more fluid, while successive additions cause separation of the soap, which finally floats on the lye, the latter being drawn off after some hours by means of either a tap or siphon. When hard water is used, a little sodium carbonate is always added to the salt. The residual lye should have, not a caustic but a brackish and somewhat sweet taste owing to the glycerine present, and its density should be at least 7 to 8 Be. (for soaps from coco-nut, palm-kernel, oxidised oils, &c., 16 to 24 Be.). If too little salt has been added, the lye will retain dissolved soap, and the separation of the latter will not be sharp, since between it and the lye will be formed a third layer consisting of an irregular, gela- tinous mass which increases the waste and diminishes the yield. With too much salt, the soapy mass separates rapidly and in large clots which retain the lye. But if the operation has been properly carried out, the soap adheres to the mixer in soft flocculent masses which, when squeezed between the fingers, are moderately stiff, do not exude liquid and give a hard and dry, not a sticky, flake. When treated repeatedly with salt solution, some soaps lose part of their combined alkali owing to the readiness with which they dissociate ; in such cases a little caustic soda is added to the salt. The soap is then subjected to the boiling process (in some cases this is preceded by a further heating with weak alkali of 4 to 6 Be. and a little salt, the subnatant lye being decanted after a time). This consists in covering the copper, boiling vigorously, and, if necessary, stirring to prevent the frothing mass from overflowing. By this means the small quantity of residual lye is concentrated and hence separates more easily, while the soap gradually becomes denser owing to the loss of nearly the whole of the water (only 15 to 35 per cent, being left). The bubbles at the surface gradually increase in size and then disappear completely, while large bubbles of steam form at the bottom of the copper, force their way noisily through the mass and produce large puffs at the surface. The heat (fire or steam) is then very soon stopped. A little of the soap pressed between the thumb and the palm of the hand then forms a dry, soft, waxy paste, but does not stick. The soap could next be moulded, but it is often subjected to a finishing process (see above), dilute caustic soda (3 to 6 Be., or hot water alone if the soap has been treated originally with excess of alkali, or very dilute sodium carbonate) being gradually added to the soap in the boiler, the mass being heated and gently stirred until it becomes more liquid, less granular and perfectly uniform. The copper is next covered and left for a day, the soap being then transferred to the moulds or cooling frames. To obtain white soap, an addition of 0-1 to 0-3 per cent, of sodium hydrosulphite is sometimes made to the mass before discharging. As a rule, soaps are made not from pure tallow, but from mixtures of various fats and oils, e.g. palm oil, olive oil, oleine, &c. ; in such cases the concentration of the caustic soda must be varied, olive oil soap, for example, requiring lye of 25 to 28 Be., which sometimes escapes salting-out. At one time Marseilles soaps were prepared from olive oil alone, very dilute lye being first used and then more and more concentrated ones. But nowadays cotton-seed, arachis, coco-nut, palm-kernel oils, &c., are generally added, the processes employed, whether for white or for Marseilles mottled soap, being those used for other soaps. SOAPS FROM FATTY ACIDS or OLEINE. Oleine, elaine, or commercial oleic acid forms a more or less dense liquid with a colour varying from pale yellow to dark brown. Less highly coloured is the oleine obtained by saponification of pure fats in autoclaves and separation from the stearine by pressing (oleine of saponification) or by enzymic or catalytic decomposition (catalytic oleine), whilst that obtained from impure fats (bone fat, &c.) or by means of sulphuric acid is generally darker and is separated after distillation of the fatty acids (distillation oleine). If an oleine contains more than 3 per cent, of non-saponifiable substances, it is certainly distillation oleine (1 to 9 per cent.), but a less content than this does not necessarily indicate oleine of saponificaticn since the modern methods of exact distillation yield oleines almost free from non-saponi- fiable matter. Oleine always contains small quantities of neutral fats and, more especially, of solid fatty acids (5 to 20 per cent, palmitic, stearic, &c.), but its iodine number should bs between 420 ORGANIC CHEMISTRY 75 and 85, and its acid number at least 179 (about 90 per cent, of fatty acids?, expressed as oleic acid). Oleine of saponification is now sold at a rather lower price than tallow, and distillation oleine at a still lower price (48-s. to 56s. per quintal). Besides for eoap-making it is Used for treating wool which is to be carded or combed. Pure oleic acid and its properties have already been considered on p. 298. The manufacture of soap from fatty acids (see p. 406 et seq.), although it gives no glycerine, is economical in various ways ; thus, it allows of a more rapid saponification with a diminished consumption of fuel and renders possible the use of sodium carbonate, which is cheaper than caustic soda. 100 kilos of oleine would require about 19 kilos of sodium carbonate (instead of 13-5 of caustic soda), but in practice only about 90 per cent, of this amount is used, the saponi- fication being completed with caustic soda in order to transform the small amount of neutral fat present in the commercial oleine. A hot solution (about 30 per cent.) of the whole of the sodium carbonate is prepared in a wide, shallow copper, the oleine being then added gradually in a thin stream, the mass being mixed and heated by a jet of direct steam so as to liberate the carbon dioxide and prevent the froth from overflowing ; the latter end is best attained by adding a "little salt to the soda solution at the beginning or by the passage of a current of air. The caustic soda solution (15 to 18 Be.) is then introduced and the whole heated, salted-out and boiled, as already described for tallow soap. Pure oleine soap is at first rather soft, but it gradually dries, hardens, and becomes of a paler yellowish brown colour than the fresh soap. When soap is made from oleine and fats together, the latter are first saponified and the oleine added subsequently. RESIN SOAPS are now made in large quantities and by almost all soap manu- facturers. Colophony (see Part III) contains acids which behave like the fatty acids and yield similar soaps, which lather well with water and, when mixed with ordinary fat soaps, diminish the price considerably, as colophony costs only 14*. to 28s. per quintal. The saponification of the resin is effected with a rather strong lye (to avoid excessive frothing). It is necessary to employ pure fats and pure resin (with the saponification number 160 to 180), and when saponification is complete, the soap must be well " finished " in order to avoid excess of alkali, which would cause efflorescence (also avoidable by the addition of a little sodium silicate at the end of the manufacture). The resin may be introduced as a powder directly into the fused fat, but it is more generally added after the fats have been saponified and the soap salted out and separated from the lye. The concentrated caustic soda (100 kilos at 20 Be. or 90 kilos at 25 Be. per 100 kilos of resin) is then added and the resin gradually disintegrated by heating and stirring. Boiling is continued until the froth almost disappears and the soapy mass separates well from the lye below and exhibits the proper consistency when pressed between the lingers. After the lye has been removed, the soap is finished with a little boiling water, then left for 12 to 24 hours, and finally solidified in the ordinary frames or moulds. Good resin soaps should not contain more than 40 per cent, of resin, but in some cases they show as much as 100 per cent, (compared with the fat), and it is a question whether resin soaps should be regarded as adulterated ; to this view the manufacturers object, for obvious reasons. Although attempts have been made at various congresses to fix limits (10, 20, or 30 per cent.) to the proportion of resin allowable, none of these are regarded. The case would be met by stamping the resin -content on every cake of soap, as there could then be no question of adulteration or fraud. Some soaps are not separated from the lye, or grained or finished, but are left mixed with the lye and the glycerine ; the fats employed must here be pure, since otherwise the impurities would colour the soap. Coco-nut oil and palm-kernel oil are more especially used, as they have the property of becoming incorporated or remaining dissolved in a large excess of alkali or salt and of forming hard soaps with even large proportions of water (200 to 300 per cent.). They are made by either the hot or the cold process, and are generally cheap soaps, as they can be resined and charged, not only with large quantities of water, but also with salt, silicate, talc, flour, &c. Solutions of salt or caustic soda ('20 Be.), even in excess, facilitate hardening, whilst potassium carbonate produces a certain softness and lustre. The silicate and salt are mixed with hot caustic soda and are MOTTLED SOAPS 421 added finally to the soap at 90 to 95. The method of procedure is that generally employed : the fat is added to part of the caustic or carbonate solution wilh which it is stirred and heated to boiling ; the rest of the alkali is then introduced and finally the salt or silicate solution in small portions ; the mass is mixed, left in the covered copper over- night, when it falls to a temperature of 75, then skimmed and cooled in the frames. 1 When these soaps are prepared in the cold, the palm-kernel or coco-nut oil is mixed with the theoretical quantity of concentrated caustic lye (for coco-nut oil, 50 per cent, of lye at 38 Be.), which saponifies these and other fats (tallow, lard, cotton-seed oil, arachis oil, resin, &c.), even in the cold, with spontaneous rise of temperature ; they are commonly loaded with silicate, talc, salt, &c. MOTTLED SOAPS. Until 30 to 40 years ago, mottled soap of the Marseilles type was made with olive oil, the mottling being produced by adding to the soap, either before or after graining, ferrous sulphate, ferric oxide, ultramarine, &c. (0-2 to 0-6 per cent, of the weight of fat), discharging into the cooling frames at a temperature of 75 to 80 and allowing to cool slowly (4 to 6 days). Mottling is satisfactory only when the soap does not contain more than 32 '5 to 34 per cent, of water, and hence constitutes a safeguard to the consumer showing that he is not being cheated with soap overcharged with water. Olive oil soap can be well mottled if it does not contain more than the above quantities of water and colouring-matter, and not more than 2 per cent, of salt, since it is only under these conditions that it acquires just that fluidity which, at the solidifying temperature, offers a resistance to the minute coloured particles (iron, aluminium, and manganese soaps, and metallic hydroxides) ; the latter gradually group themselves into veins, whilst the drops of lye and soluble salts fall to the bottom. If the quantity of water is raised, the equilibrium is displaced and the fluidity increased, so that the colouring-matters are deposited. But if other folid fats are used in conjunction with the olive oil, the required consistency can be obtained with as much as 50 per cent, of water. With coco-nut, palm-kernel, and palm oils, mottled soaps can be prepared containing 70 per cent, or even more of water, in addition to an increased amount of alkali. These soaps, however, should not contain more than 2 per cent, of sodium carbonate and less than 10 per cent, of dissolved salts ; otherwise the soap will effloresce on drying, provided that it is sufficiently stiff to permit of mottling. A type of mottled soap which is often prepared with a yield of 180 to 200 per cent. is that from almost equal quantities of sulphocarbon olive oil and coco-nut or palm oil. In this case the manufacture of the olive oil soap is carried out separately as far as the stage where it is separated from the lye, so as to remove the impurities ; it is then intro- duced into the pan where the coco-nut oil has been saponified in the hot with caustic soda of about 20 Be., together with some 13 per cent, of sodium carbonate dissolved in water. Unger (1869) found that, in order to prevent coco-nut or palm oil soap from efflorescing on drying, it should not contain more than 43 per cent, of sodium carbonate, calculated on the weight of coco-nut oil (i.e. 1 mol. of sodium carbonate per 4 mols. of pure coco -nut soap). After mixing, the two soaps are boiled and 4 to 5 per cent, (on the total fat) of sodium chloride solution of 24 Be. gradually added ; the heating is continued until the paste boils readily without adhering to the sides of the copper, and the steam evolved produces, at the surface of the soap, veinings and crevices in the form 1 High yields are given by the following mixtures Ss It Coco- Crude Palm Caustic Potassium 9-2 Yield nut oil palm- kernel oil oil Tallow Resin soda (26 B6.) carbonate 25-30 Be. 20-22 Be. SMC- o ~ kilos kilos About 250 % 90 [or 90] 10 60 65 40 300 % 100 60 100 65 300 % 50 40 20 15 60 65 65 (resined) 400 % 100 [or 100] 60 100 100 50 30 800 % 100 [or 100] 80 260 (20 Be.) 300 60 1000 % 100 ~ 150-160 (22 B6.) 800 422 of rosettes. The soap will then emit a hollow sound and will not form bubbles when struck with the stirrer, from which it falls in broad folds which become covered with a dry skin ; when placed on glass, it is quickly coated with a solid layer beneath which it remains fused, while between the fingers it does not pull out, but tends to solidify. It is important that it should not contain an excess of caustic soda (not more than 0-2 to 0-3 per cent. ; it is best neutral) as with finished soaps ; any excess may be eliminated by adding the calculated quantity of coco-nut oil or of hydrochloric acid, determined by titration. At this point the colouring-matter is well mixed in, the soap being then cooled to about 75 and poured into large solidifying frames (holding at least 10 quintals) so as to cause slow cooling (in winter these are wrapped round with cloths), and hence satisfactory mottling. These mottled soaps of high yield (up to 400 per cent.) bear the name of blue mottled or Eschweg soaps, and were largely used some years ago. Even now their consumption is considerable, as they have a higher detergent power than finished soaps owing to their richness in alkali carbonates ; they dry more rapidly than resin soaps and owing to their hardness they are preferred for laundry purposes, there being no waste even when the clothes are vigorously rubbed. The formation of mottling in soaps probably obeys the laws holding in the solidifica- tion of alloys (solid solutions) and the figures given on pp. 412 and 642, and in Plate III. of vol. i of this work ("Inorganic Chemistry ") represent well the impression produced by the mottling of soap. FIG. 280. FIG. 281. When almond-mottling is required, an iron rod 12 to 15 mm. in diameter is drawn vertically through the semi -solid soap in the solidifying frame, so as to make a kind of longitudinal cut ; similar cuts, parallel to the first, are then made throughout the whole mass at distances of 4 to 6 cm., and afterwards a similar series perpendicular to the others. When solidification is complete, the whole of the soap is traversed by dark markings in the shape of almonds arranged like the leaves on acacia twigs. Other mottlings are made either by machinery or by hand. For Eschweg soaps mixtures of various fats are used, e.g. 20 to 25 per cent, of tallow, 25 to 30 per cent, of bone fat, 10 to 15 per cent, of cotton-seed oil, 20 to 40 per cent, of palm-kernel oil, and 20 to 30 per cent, of coco-nut oil. The yield is usually 205 to 215 per cent., although additions of silicate (10 to 12 per cent.) are sometimes made. TRANSPARENT SOAPS were at one time obtained by dissolving ordinary soaps in alcohol, evaporating the latter and moulding the transparent residue. The amount of alcohol used was subsequently diminished by adding glycerine, and at the present time transparent or so-called glycerine soaps are made from mixtures of decolorised tallow with castor, linseed, and coco-nut oils, with addition of glycerine and also of 20 to 30 per cent, of saccharose or glucose, which enhances the transparency. To this mixture, melted . in the copper, is added caustic lye at 30 to 36 Be\, the whole being mixed until a homo- geneous emulsion is formed ; 2 to 5 per cent, of alcohol is then introduced, and the mass heated to 75, cooled to 50, and poured into the moulds. For some of thete soaps as much as 40 per cent, of pale resin is employed. SOFT SOAPS are usually potash soaps of linseed oil or oleine, while in summer cotton- seed, colza, sesame^ palm, or fish oil is also used. Some of these soaps are transparent (plain or variegated), others opaque and white or yellowish. For every 100 kilos of fat, about 160 kilos of caustic potash of 24 Be. are used, the yield being sometimes as much as 235 per cent. ; if caustic soda is partly COOLING, BARRING, ETC., OF SOAP 423 employed, a harder soap is obtained, but the yield is diminished. Also 10 to 15 per cent, of resin may be used or 10 to 15 per cent, of oil. In general these soaps contain carbonates. The boiling is carried out in the usual way, and is continued until frothing ceases, and a small portion placed on glass remains clear for some time without forming a skin and, on cooling, becomes turbid at the edges and exhibits slight veinings of lye. If this test portion remains clear but presents no such veinings, lack of alkali is indicated. FIG. 282. FiQ. 283. Many of the soft soaps now used contain white granules, produced by the addition of tallow or stoarine, which crystallises out throughout the mass of poap during the cooling, the latter occupying 4 to 8 weeks ; this change is known as figging and the yield of such soaps is often as high as 240 per cent. The mamifacture of soda soap from glycerides by means of lime and sodium carbonate (Krebitz process) has been described on p. 408. Cooling and Solidification. The soap from the copper is cooled in large chests or frames, formerly of wood but now of iron, as was suggested by Krull in 1876 (Fig. 280). The sides of these are fixed by means of bolts and nuts and hence fit perfectly and are readily taken apart. In some cases, the frames are mounted on three wheels so as to be transportable. To prevent any impurities depositing in one place and so producing mottling, the pasty soap in the frame is stirred with wooden crutches until it begins to solidify ; but, if slow solidification is required (for mottled soaps), the sides of the frame are covered with straw mat- tresses or wool, especially FIG. 284. in winter (Fig. 281). The frames vary in capacity from 100 to 6000 kilos and, according to the amount and quality of the soap, the cooling lasts one or several weeks. The walls of the frame are then removed and the large block cut into smaller prismatic blocks by means of thin steel wires worked by a toothed -wheel winch, which is applied to various points of the block (Fig. 282). The small blocks are discharged on to a truck carrying a platform which can be raised (Fig. 283) and are then transferred to the barring machine (Fig. 284), where each block is placed between A and B and forced by means of the plate A and the toothed wheel, R, against the frame, B, fitted with adjustable crossed steel wires. The long bars thus 424 ORGANIC CHEMISTRY obtained between B and C are then pressed against the vertical wires of the frame, C, and thus cut into cakes of the required size. There are many such machines of different types, some fitted with fixed and others with universal frames. During recent years a method has been devised of preparing cakes directly from the hot soap from the copper, without using the largo cutting machines (slabbers) ; in this way much time is saved, waste and scraps are diminished in amount and the subsequent seasoning shortened. The hot soap is rapidly cooled and compressed in the Klumpp apparatus (Fig. 285), being first transferred to the jacketed reservoir, L, where it is kept liquid by means of hot water in the jacket. The plate, c, consisting of a double -walled box surrounded by cold water, has a movable base, h, resting on the piston of a hydraulic pump, K. The box, c, is filled with liquid soap and the wheel, V, turned so as to press on to the surface of the soap the large plate, a, which is kept horizontal by the four rods, N, of the press, while inside it cold water circulates. When this plate is firmly fixed and the soap begins to solidify, a pressure of 50 atmos. is applied by means of the press, K, this pressure being increased to 250 atmos. when the soap is quite cold and solid. The ordinary cutting machines are then used to cut these slabs into marketable pieces, which lose little water even in the air. Seasoning or drying of the soap, to bring it to the degree of moistness required by the trade, is effected by keeping the cakes on frames in well-ventilated cham- bers for several weeks or even months. This slow drying is now generally replaced by drying in hot air. furnished cheaply by Perret furnaces, which burn waste coal or slack. The soap is spread out on gratings superposed on trucks, which are gradually introduced into a brickwork gallery ; hot air traverses the gallery, entering at the opposite end at 50 to 60 and being dis- charged at 35 to 45. The seasoning is FIG. 285. complete in 3 to 6 days, but if the tem- perature is too high at first, or the drying too rapid, the soap softens and becomes deformed and crushed. To give the cakes a smooth surface, and so render efflorescence and cracking more difficult, as they issue from the dryer they are subjected to the action of a slight steam-jet, which melts them superficially. STATISTICS. The largest and most up-to-date soap-factories of the present day are in the United States and England, next in importance being those of Germany and France. Italy imports more than 220,000 quintals of fat (1904) for soap and candle-making. In 1894 there were 300 factories and an appreciable exportation (33,000 quintals of common soap and 1000 of perfumed soap) which in 1903 reached 40,680 quintals of common soap, sent to England and the United States, and 1230 of perfumed soap to India and Egypt. In 1909 the exports were 28,450 quintals of common soap, worth 74,000, and 2150 of perfumed sorts, worth 19,340 ; in 1910, 40,000 quintals of common soaps, valued at 104,200, and 1915 of perfumed (17,240) were exported. The importation amounted to 16,369 quintals of ordinary and 964 of perfumed soap in 1903 ; to 19,000 quintals of common kinds from France and 1100 of perfumed from Germany, France, and England in 1904 ; to 23,230 quintals of common soap, valued at 48,400, 1120 of perfumed (15,680), and 5300 quintals of cart-grease and stiff fats, consisting largely of lime soaps, resins, and mineral oils and valued at 2544, in 1905 ; 40,000 quintals of ordinary soap, costing 96,000, 1617 quintals of perfumed soap (23,360), and 2500 quintals of cart -and engine -grease in 1910. The total production of soap in Italy in 1905 was estimated at about 1,500,000 quintals. France was at one time the greatest exporter of soap, 160,000 quintals of non-perfumed kinds being exported in 1890 and 268,000 (worth 440,000) in 1900 ; but at the present time England is far ahead. In 1898 France produced about 3,000,000 quintals of soap (one-fourth mottled), one-half of this in the Marseilles district ; the total value was 5,600,000. The port of Marseilles receives annually 5,000,000 quintals of oily fruits SOAP STATISTICS 425 and seeds, and a similar quantity of crude oils and fats for extraction and refining (including mineral oils), the total yearly production of the Marseilles oil, soap,- and .stearine industries being nearly 40,000,000. The province of Marseilles produced 500,000 quintals of mottled and 50,000 quintals of white soap in 1866 and 200,000 quintals of mottled and 1,400,000 of white soap in 1808. In 1894 England exported 290,000 quintals of soap and in 1897 about 370,000 ; in 1900 the exports of soap were valued at 920,000, and those of stearine candles at 400,000 ; in 1907 the soap exported amounted to 1,480,000 ; and in 1909 to 650,000 quintals (112,000 being imported in that year). The English Sunlight Company alone has a capital of 14,000,000. The production of soap in England in 1907 was as follows : soft soap, 49,900 tons ; toilet soap, 70,400 tons ; ordinary soap, 665,000 tons ; and various other soaps, 24,200 tons, the total value being estimated at 7,055,000. The exportation of ordinary soap alone amounted to 119,540 tons in 1909 and 134,560 tons (1,357,776) in 1910, in which year the imports were less than 20,000 tons. The total production of soap in England is about 4,000,000 quintals per annum. The United States produced soap to the value of 13,650,000 in 1904 and of 22,280,000 in 1909 ; the exports amounted to 789,000 in 1910 and 800,000 in 1911, and the imports to 165,000 in 1910 and 160,000 in 1911. In 1903 Germany imported 7,740,000 quintals of oils and fats, and in 1905 about 9,000,000 quintals, of the value of 13,000,000, the exports in 1905 being 2,432,000 quintals, worth 2,920,000. Also, in 1905, 306,320 quintals (145,000 in 1903) of oleine, valued at 300,000, were imported. In 1903 Germany exported 84,160 quintals of soap of all qualities, and in 1905 almost 99,000 quintals, worth 440,000, the imports in that year being 14,500 quintals, valued at 48,000 ; in 1905, the exports of soap were about 640,000. In 1906 Japan imported soap to the value of 36,000, and in 1907 52,000, half of the amount coming from Germany. The Argentine Republic possesses 200 soap factories, representing a total capital of about 240,000, and giving an annual production valued at about 640,000 ; considerable quantities of perfumed and medicinal soaps are imported. The value of soap l varies considerably with the quality, the degree of fineness, the 1 Analysis of Soap. As a rule, the commercial value of a soap is determined from tho quantity of combined fatty acids which it contains, and as the percentage of these varies with the degree of moistness, great care must be taken in sampling the soap. The cake is first weighed and the sample cut in such a way that the inner and outer portions are taken in the proper proportions ; the sample is then cut up fine, rapidly mixed and immedi- ately enclosed in a vessel with a ground stopper so that water may not be lost. The analysis consists of some or all of the following determinations : (1) Water. This estimation is not usually made, as it involves a long operation, while it is possible to cal- culate the proportion of water indirectly after all the other components have been determined. The direct estimation is made by weighing 5 to 10 grms. of the finely divided soap rapidly in a tared dish containing a small glass rod and filled to the extent of one-third with sand which has been previously calcined. The dish and its contents are heated first in an oven at 60 to 70", the fused soap being carefully mixed with the sand until a skin of soap no longer forms at the surface ; the temperature is then raised to 105 to 110, at which it is maintained until constant weight is reached. The total loss in weight represents the water. (2) Unsaponified Fat. The dry residue from the water estimation is introduced into a Soxhlet extractor (see p. 374) and extracted for a couple of hours on the water-bath with light petroleum in a tared flask ; the solvent is subsequently distilled off and the extracted fat dried at 110 until of constant weight. (3) Fatty Acids, Free Alkali, Glycerine, and Resin. The residual matter in the Soxhlet apparatus (or the dry soap itself) is extracted with neutralised absolute alcohol, which dissolves the soap, glycerine, and free caustic alkali ; the last of these is determined immediately by titrating the alcoholic solution with normal sulphuric acid in presence of phenolphthalein. The liquid is afterwards largely diluted with water, heated for a long time on the water-bath to remove all the alcohol, and treated with a measured volume, in excess, of normal sulphuric acid, the liquid being then heated in a beaker on a water-bath and on a sand-bath until the clear fatty acids (and the resin, if present) separate at the surface. After cooling, the solidified layer of acids is pierced with a rod and the liquid poured on to a tared filter in a stemless funnel, the fatty acids being then washed with hot water, and the whole brought on to the filter. The excess of free sulphuric acid in the whole of the wash-water is deter- mined by titration with normal caustic potash. This then gives the amount of sulphuric acid fixed by the alkali of the soap and hence also the combined alkali expressed as Na.jO. Evaporation of the liquid to dryness and extraction with absolute alcohol removes any glycerine present in the soap, this being weighed after evaporation of the alcohol. The fatty acids on the filter are treated with a couple of c.c. of a'lcohol to remove any moisture and then with sufficient light petroleum to dissolve all these acids ; the filtrate is evaporated in a tared dish, dried at 105 to constant weight and the residual fatty acids weighed. To determine any resin which may be present in the fatty acids, part of the latter is weighed, dissolved in 20 c.c. of alcohol and, after addition of phenol- phthalein, hydrolysed in the hot with a slight excess of alkali ; after cooling, the liquid is made up to 110 c.c. with ether, treated with powdered silver nitrate and allowed to deposit the precipitated silver stearate, palmitatc, and oleate. One-half of the filtered solution (containing soluble silver resinate) is treated with 20 c.c. of dilute 426 ORGANIC CHEMISTRY content of fatty acids, and the degree of purity. The ordinary soaps used in laundries and in the textile industries, which are made from sulphocarbon olive oil and contain 60 to 65 per cent, of fatty acids, cost 44s. to 48s. per quintal, according to the conditions of the market and the prices of prime materials (fats and oils). Soaps loaded with water and other substances may cost much less ; fine, perfumed soaps cost up to 4 to 8 per quintal. GG. POLYHYDRIC ALDEHYDIC OR KETONIC ALCOHOLS CARBOHYDRATES (Sugars, Starch, Cellulose) This group of substances might have been included in the preceding chapter, FF, where, in paragraphs D and E, certain very simple aldehydic and ketonic alcohols have been considered. But, partly owing to custom (since it has been the rule to include in the group of Carbohydrates only ketonic or aldehydic polyhydric alcohols with six [monosaccharides] or a multiple of six carbon atoms [polysaccharides] and containing hydrogen and oxygen in the proportion of 2 : 1, as in water), and partly because this group embraces all the sugars, which exhibit special characters very different from those of glycollic aldehyde (which should be the first member). So that even at the present time the carbohydrates are considered separately, although the brilliant researches of Emil Fischer, commenced in 1887, have extended this group to compounds with five, four, or three carbon atoms, on the one hand, and to monosaccharides with six, eight, or even nine carbon atoms on the other. These monosaccharides bear the name of Monoses (biases, trioses, tetroses, pentoses, hexoses, heptoses, octoses, nonoses, &c., according to the number of carbon atoms they contain), while the polysaccharides (formed by the con- densation of two or more monose molecules) are called generally polyoses and, in particular, hexabioses, hexatrioses, &c., according as they are formed by the condensation of two, three, &c., hexose molecules. A. MONOSES All the monoses are aldehydic or ketonic polyhydric alcohols containing H the characteristic grouping, C C , i.e. a hydroxyl group united with a OH 6 carbon atom adjacent to a carbonyl (CO) group. When the carbonyl exists H as an aldehydic group, C C-H, these monoses are called Aldoses, whilst OH 6 hydrochloric acid (1 : 2) and filtered, an aliquot part of the filtrate being evaporated in a tared capsule, dried at 100 and the residual resin weighed ; the weight of the resin is diminished by 0-00235 grm. for every 10 c.c. of ethereal solution of silver resinate, this being the amount of oleic acid removed by the ether. The true weight of the fatty acids, free from resin, can then be calculated. (4) Soda, Salt, Sulphates, Silicate, &c. The residue from the Soxhlet apparatus, after separation of the fat and soap, is treated two or three times with 50 to 60 c.c. of hot water and the solution filtered, made up to a definite volume and divided into four parts : one of these is titrated with normal sulphuric acid, using phenol- phthalein as indicator, to ascertain the sodium carbonate ; in a second portion, the sodium chloride is determined by titration with silver nitrate ; the third is precipitated with barium chloride and the weight of the barium sulphate and hence that of the sodium sulphate in the soap, determined. The fourth portion is treated with hydrochloric acid and the silica, thus separated from the silicate, weighed. (5) Ash and Mineral " Filling." The ash obtained by burning a definite weight of pure soap is about 40 per cent, greater than the total alkali (expressed as Na a O). If the proportion is much higher than is indicated by this relation, the excess represents mineral filling. CARBOHYDRATES 427 ? I when it exists as a ketonic group, C C C , they are termed ketoses, so OH 6 J that we have aldohexoses, ketohexoses, &C. 1 The monoses have the general properties of the aldehydes or ketones and hence form, on oxidation, the corresponding monobasic acids, e.g. pentonic, hexonic acids, &c. Since the aldoses contain a primary alcoholic group, X OH-CH 2 -[CH-OH] n -Cf , X H they can also be oxidised to dibasic acids containing the same number of carbon atoms, whilst when the ketoses are oxidised, the carbon atom chain is ruptured and acids with lower numbers of carbon atoms formed. On reduction, both the aldoses and the ketoses take up two atoms of hydrogen, forming the corresponding alcohols ; the hexoses give hexitols and the pentoses pentitols. Like all aldehydes, they reduce ammoniacal silver solutions in the hot, giving silver mirrors. When heated with alkali, they turn brown and then resinif y . They reduce alkaline copper solution (Fehling's solution) in the hot. When heated with excess of phenylhydrazine dissolved in acetic acid, they yield yellow, crystalline phenylosazones, insoluble in water. 2 In dealing with the hexoses later on, we shall see how the constitutions of the monoses in general are determined. Of the various monoses, containing from 2 to 9 carbon atoms, only certain of the hexoses are fermentable, that is, give alcohol and carbon dioxide under the action of ferments or enzymes (see pp. 112 and 122). Of the hexoses, some ferment readily, others with difficulty, and others again not at all, in dependence on their stereochemical configurations and possibly on the asym- metric constitution of the enzymes. d-Glucose, d-mannose, and d -fructose ferment easily, and d-galactose with difficulty, whilst 1-glucose and 1-mannose do not ferment. GENERAL METHODS OF FORMATION OF THE MONOSES : (a) From the polyoses by hydrolysis with dilute acids, water being added and several molecules of hexose obtained : Cia^On (saccharose) + H 2 = 2C 6 H 12 6 . (b) By oxidation of the corresponding alcohols by nitric acid : e.g. Arabitol, C 6 Hi 2 5 , 1 The two classes of sugars, aldoses and ketoses, are distinguished by means of Romijn's reaction with a solution of iodine and borax, which oxidises all the aldoses (galactose, glucose, mannose, arabinose, xylose, rhamnose maltose, lactose), while it either does not oxidise the Jceloses or oxidises them but slightly (sorbose, fructose ; saccharose and rafflnose are oxidised to a small extent). 8 They form first pJienylhydrazones (see p. 206) : H C OH H C OH c = io + H,!N-NH-C,H, = H 2 o + c = N-NH-C.H,; I I these phenylhydrazones then react with two other molecules of phenylhydrazine, giving ammonia, aniline, and phenylosazone : H c ;o :H NH,iNHC,H 6 C=N-NHC,H, = NH, + NHj-C.H, + HjO + | j aniline C=N-NHC,H, C = N-NH-C.H, I which is the characteristic group of the phenylosazones. The latter crystallise readily and in a pure state from a dilute pyridine solution. Reduction of the phenylosazones yields osamines, e.g. glucosamine, C,H n O,-NH,. 428 ORGANIC CHEMISTRY gives Arabinose, C 5 H 10 O 5 (pentose) ; xylitol (stereoisomeric with arabitol) gives xylose and Mannitol, C 6 Hj 4 O 6 , mannose. (c) Oxidation of glycerol gives dihydroxyaoetone, OH-CH 2 -CO-CH 2 'OH, which is a triose, its constitution being indicated by the fact that it forms a cyanohy- drin, OH-CH 2 -C(OH)(CN)-CH 2 -OH, the latter yielding trihydroxyisolutyric, acid, OH-CH 2 -C(OH)(COOH)-CH 2 -OH, and this, on reduction, isobutyric acid having a known constitution. (d) By treating the bromo -derivatives of the aldehydes with baryta water. Thus nionobromaldehyde gives Glycollic Aldehyde, ^ // Q CH 2 Br-C O 7 OH- CH 2 CH(OH) CH(OH) CH(OH) CH(OH) CH(OH) cf . Heptose By the same ketonic (lactonic) synthesis, the keptose can be converted into odose and nonose. 1 It was shown by Butlerow that formaldehyde and later by E. Fischer that glyceraldehyde can, under certain conditions and in the presence of bases (baryta), give rise to sugar (a-acrose). In 1905 H. and A. Euler found that under no conditions do other alkali hydroxides give an appreciable amount of sugar, whilst with dilute solutions of sodium carbonate or, better, with calcium carbonate or lead hydroxide at 100, first glycollic aldehyde and glyceric aldehyde are formed and finally a Tceto-arabinose, the phenylosazone of which melts at 159 to 161. The conditions for the production of hexoses from formaldehyde are not yet defined, but O. Loew stated that, with milk of lime, he obtained formose, which is a mixture containing i-fmctose (a-acrose). D. Berthelot and H. Gaudechon (1910) found that the action of ultra-violet rays on 10 per cent, solutions of various sugars at a temperature of 80 to 90 leads to the rapid formation of the following quantities of gas : CO CH 4 H 2 CO 2 Levulose (ketose) .... 83 8 9 15 Glucose (aldose) 12 12 76 22 Maltose (gives 2 mols. glucose) 12 11 77 21 Saccharose (gives glucose and levulose) .45 8 47 16 They found also that prolonged action of ultra-violet rays on a mixture of CO 2 and H, yields a small quantity of CO and of formaldehyde. These facts tend to confirm J. Stoklasa's observations (1906-1910) on the formation of hydrogen as final T ETHOSES AND PENTOSES 429 TETROSES, C 4 H 8 O 4 , and PENTOSES, C 6 H 10 O 5 Just as the hexoses can be converted into pentoses, the latter can give rise to TETROSES. For instance, d-, 1-, and i-erythrose are obtained by oxidising d-arabonic acid, d-arabinosoxime, and natural i-erythritol respec- tively with hydrogen peroxide : OH-CH 2 - [CH-OHVCOOH + = H 2 + C0 2 + OH-CH 2 - [CH-OH] 2 -CHO Tetrose The tetroses are also obtained by oxidising (with H 2 2 ) the calcium salts of pentonic acids in presence of ferric acetate, which acts as an oxidising catalyst. The pentoses (Arabinose, Xylose, &c.) occur abundantly as Pentapolyoses or Pentosans (Araban, Xylan) in many vegetable organisms (straw, wood, maize husks, &c.), from which they are obtained by simple boiling with dilute acids. 1 Pentoses do not ferment. product of the degradation of carbohydrates by the action of glycolytic enzymes, which have an important function in the assimilation of carbon dioxide in the chlorophyll cells, and also to render valid Stoklasa's hypothesis (1907) that the formaldehyde necessary to the formation of carbohydrates by the simple polymerisation assumed by I3acyer can result from the reaction, 2CO 2 + 2H a = 2 + 2H-CHO. Stoklasa and Zdobnicky (1910) have obtained inactive sugars and aldehyde by the action of ultra-violet rays on carbon dioxide and hydrogen in the nascent state in presence of caustic potash (with initial formation of potassium bicarbonate, which, in the nascent state and with nascent hydrogen, generates the sugar) and have disproved the view held by Fischer (1888-1889), Loew (1888-1889), Neuberg (1902), and Euler (1906) that, in the synthesis of sugars from formaldehyde, pentoses are formed ; the sugars they obtained are not asymmetric and are hence not fermented by ordinary alcoholic ferments. According to Stoklasa, the function of the chlorophyll in plants is to absorb the ultra-violet rays of sunlight. From the aqueous distillate of the leaves of various plants, F. Hartwig and T. Curtius (1910) have separated (by means of m-nitrobenzhydrazide, a : p-hexylenealdehyde, CH 3 -CH,-CH 2 -CH : CH-g( , the hydia- \H zone of which melts at 167. 1 By the term Pentosans are meant those polysaccharides which are related to the pentoses in the same way as are starch, inulin, &c., to the hexoses, and which give pentoses and also hexoses on hydrolysis. From starch they arc distinguished by their Isevo-rotation. From plants the pentosans are extracted by means of dilute alkali according to the method given by Tollens, Stone, and Schulze (1888-1901) : the finely divided vegetable matter is treated twice, for some hours at the ordinary temperature, with seven times its weight of 2 per cent. ammonia solution to eliminate in the soluble state part of the proteins, salts, &c., and to remove the more soluble part of the hcmicellulose (this would give little pentose on subsequent hydrolysis). After the dark ammoniacal liquid has been separated by filtration through cloth and by squeezing in a press, the solid residue is extracted with ten times its weight of 5 per cent, caustic soda solution, with which it is first macerated in the cold for ten to twelve hours, and then heated in a reflux apparatus on a water-bath for six hours. The mass is next filtered through cloth and the residue pressed and washed several times with water until the total volume of solution obtained is equal to that of the caustic soda solution used. This brown liquid is evaporated to some extent on a water-bath and is then treated in the cold with an equal volume of 90 per cent, alcohol. The voluminous, flocculent precipitate of gum (pentosans) thus obtained is collected on cloth, washed and purified by repeatedly dissolving in dilute acid and reprecipitating with alcohol, this procedure being continued until the gum leaves a minimal ash on incineration. To pass from the pentosans to the pentoses, the moist gum is hydrolysed (Conneler and Tollens, 1892 and 1903) by digestion for 12 hours with 25 parts of water and 2-5 parts of hydrochloric acid of sp. gr. 1-19, the mixture being finally heated on a water-bath until the furfural reaction (red coloration with aniline acetate paper) begins to make its appearance (about two hours). After filtration of the cold liquid and neutralisation with lead car- bonate (testing with Congo-red paper), a few drops of barium hydroxide are added and the liquid filtered to remove precipitated lead chloride and barium carbonate. The solution is concentrated on a water-bath und.er reduced pressure, mixed with a little alcohol, filtered and concentrated to a syrup. This is taken up with methyl alcohol and the solution filtered to remove mineral and other impurities. The alcohol is then evaporated and the residue seeded with a few crystals of xylose or arabinose and left in a desiccator until the whole mass crys- tallises (this sometimes requires several weeks). In order to separate the arabinose and xylose, which often occur together, Ruff and Ollendorff (1899) treat the mixed pentoses with eight times their weight of 75 per cent, alcohol and nearly their own weight of bcnzyl- phcnylhydrazine dissolved in a little absolute alcohol. After several weeks' rest with frequent shaking, there separates arabinose benzylphenylhydrazone, which, in the pure state melts at 174 and, when treated with excess of formaldehyde, liberates the arabinose ; the latter is soluble in water, whilst formaldehyde benzylphenylhydra- zone remains undissolved. The aqueous arabinose solution, after separation and concentration to a syrupy consistency, deposits pure urabinose in crystals. The corresponding hydrazone of xylose is soluble in 75 per cent, alcohol, and yields xylose when decomposed with formaldehyde in the manner described above. The xylose can also be separated, according to Bertrand and Tollens (1900), by treating the mixture of pentoses with 2 parts of water, 1 part of cadmium carbonate, and 0-5 part of bromine. The mixture is heated for a short time on the water-bath, then left for twelve hours, evaporated, taken up with water, filtered, again evaporated, and mixed with alcohol ; this pro- cedure yields crystals of cadmium bromoxylonale, C 5 H 9 O 6 BrCd. But before carrying out this separation, it is necessary to make sure that the mixture contains no galactose or glucose. These sugars can be detected by oxidising the mixture with nitric acid (sp. gr. 1-15) on the water-bath and evaporating the liquid to two-thirds of its volume. If the liquid remains turbid in the cold, the presence of mucic acid, derived from galactose, is indicated ; and if, after neutralising with potassium carbonate, acidifying with acetic acid and concentrating, potassium hydrogen saccharate separates, the presence of ylucose which gives saccharic acid on oxidation ig demonstrated. 430 ORGANIC CHEMISTRY Arabinose and xylose are aldoses, OH-CH 2 - [CH(OH)] 3 -CHO. By bromine water these two pentoses are oxidised with formation respectively of arabonic and xylonic acids, OH-CH 2 - [CH-OH] 3 -C0 2 H, which are stereoiso- ineric ; with more energetic oxidising agents, they give trihydroxyglutaric acid. On reduction they yield the corresponding alcohols, arabitol and xylitol (see p. 189), which are also stereoisomerides. By way of the corresponding cyanohydrins they can be converted into hexoses (via hexonic acids). All these reactions aid in establishing the constitution of these pentoses. As they contain asymmetric carbon atoms, these sugars are optically active, and they exhibit the phenomenon of muta-rotation ; thus, for freshly prepared solutions of xylose, the value of the specific rotation is [a] D = 75 to 80, while five minutes after the sugar is dissolved it has the stable rotation + 19. When pentoses are boiled with dilute sulphuric acid or with hydrochloric acid of sp. gr. 1-06 (12 per cent.), they yield furfural, C 4 H 3 O-CHO (aldehyde), which distils over and gives a characteristic and intense red coloration with aniline and hydrochloric acid, a phenylhydrazone with pheiiylhydrazine, and a slightly soluble condensation product with phloroglucinol. 1 Treatment of any pentose or hexose with caustic soda in presence of air or other oxidi- sing agent (e.g. HgO) yields no trace of saccharic acid, but gives formic acid and monobasic hydroxy-acids (e.g. glycollic, dl-glyceric, trihydroxybutyric, and various pentonic and hexonic acids) ; if air is excluded, aldotetroses, formaldehyde, a little 2 : 3-dienols, bioses, and glyceraldehyde are mainly formed. Recent work has shown that the furfural obtained on distillation of vegetable sub- stances with 12 per cent, hydrochloric acid is derived not merely from true pentosans, but also from oxycellulose, glycuronic acid, &c. ; niethylpentosans give methylfurfural. Hence Cross and Bevan suggest the name furfuroids for substances other than true pen- tosans which give furfural. On the other hand, it has been proposed by Tollens that the term pentosan be applied to the whole of the substances (furfuroids and true pentosans) which give furfural when distilled with 12 per cent, hydrochloric acid. Hydroxymethyl- furfural (see below) does not distil in presence of acids but undergoes resinification, and hence escapes the Tollens method of estimating furfural. Until comparatively recent times it was assumed that the pentosans were derived from the hexoses and poly hexoses, since it was known that 4:-hydroxymethylfurfuraldehyde, CHO-C : CH-CH : OCH 2 -OH is obtained on heating levulose, d-mannose, d-glucose, d-galactose, chitose, &c., in a sealed tube with 0-3 per cent, of oxalic acid, while 4,-bromomethylfurfural, CHO C : CH- CH : O CH 2 Br, 1 Quantitative Determination of Pentoses and Pentosans. Mint and Tollens (1902) distil in a flask similar to that shown in Fig. 17 (p. 11), about 5 gnns. of the substance with 100 c.c. of 12 per cent, hydrochloric acid, the heating being carried out in an oil-bath at 160. Thirty c.c. of liquid are distilled over every twelve to fifteen minutes, in which time 30 c.c. of fresh acid are added by means of a tapped funnel, this procedure being continued as long as the distillate reddens a strip of filter-paper moistened with an acetic acid solution of aniline. To the distillate is added an excess (double the amount of furfural expected) of pure phloroglucinol dissolved in 12 per cent, hydrochloric acid. The volume of the liquid is made up to 400 c.c. with the same acid in a graduated flask, which is well shaken and left for 12 hours, at the end of which time the precipitate is collected on a tared filter, washed with 150 c.c. of water, dried for four hours in an oven and weighed. The weight of furfural is obtained by dividing this weight by a variable factor, which has the following values for different amounts (in grms.) of the phloroglucinol compound : 0-20 (1-820) ; 0-22 (1-839) ; 0-24 (1-856) ; 0-26 (1-871) ; 0-28 (1-884) ; 0-30 (1-895) ; 0-32 (1-904) ; 0-34 (1-911) ; 0-36 (1-916) ; 0-38 (1-919) ; 0-40 (l-92ft) ; 0-45 (1-927) ; 0-50 (1-930) ; 0-60 or more (1-931). The xylan is calculated by multiplying the quantity of furfural by 1-64, the araban, by 2-02, while for mixed pentosans, the factor 1-84 is employed. Another method of procedure consists in precipitating the furfural with phenylhydrazine and estimating the nitrogen in the precipitate. Jolles (1906), however, neutralises almost completely (to methyl orange) the distillate containing the furfur- aldehyde, then adds 10 c.c. (moir, if necessary) of a decinormal sodium bisulphite solution, and after two hours titrates the excess of bisulphite with a deciuoimal iodine solution (1 c.c. of which corresponds with 0-0075 grui. HEXOSES 481 is obtained by heating levulose (or filter-paper, cotton, cellulose, straw, starch, dextrose, lactose, glycogen, &c.) under pressure with chloroform saturated at with hydrogen bromide. Further, when the oxime of levulose is heated with concentrated caustic potash solution, the uitrile is first formed and then hydrocyanic acid and d-arabinose : OH-CH 2 -[CH-OH] 4 -CH:NOH > H 2 + OH-CH 2 - [CH-OH] 4 -CN > HCN + OH-CH 2 -[CH-OH] 3 -CHO. Oxidation of d-gluconic acid with peroxides also gives d-arabinose. Ketohexoses in general, when heated with dilute acids (e.g. with 0-3 per cent, of oxalic acid under a pressure of 3 atmos.), are largely transformed into hydro xymethylfurfural, whilst the aldohexosea undergo this change only to a very slight extent ; if mineral acids are used, or oxalic acid in larger quantity, levulinic acid is obtained instead of hydroxy- methylfurfural. U. Nef 's recent work (1910) tends to show that, in plants, pentosans cannot be derived from the hexoses, but that they are formed rather from either aldotetroses and formalde- hyde or 2-carbon-atom sugars and glyceraldehyde. The hescoses, in their turn, would be formed, not from pentoses and formaldehyde, but rather from 2 mols. of glyceraldehyde or 3 mols. of a 2-carbon-atom sugar, or even from 1 mol. of a 2-carbon-atom sugar and 1 of an aldotetrose. XYLOSE is readily obtained by boiling with dilute sulphuric acid plants containing it, especially jute, bran, straw, or, better still, apricot stones or maize hueks. It bears ako the name of wood-sugar, and is yielded by the decomposition of gluconic acid. When pure, it crystallises and forms a phenylosazone melting at 160. rf-ARABINOSE is lee vo -rotatory, but is obtained from calcium d-gluconate and hydrogen peroxide and from d-glucose. In the pure state it forms prismatic crystals. i-ARABINOSE is the optically inactive racemic isomeride, and is found in the urine of persons suffering from pentosuria. Z-ARABINOSE is obtained by boiling vegetable gum with dilute sulphuric acid. It is dextro-rotatory, but is designated a laevo -compound because it is related chemically to 1-glucose. It forms sweet-tasting crystals melting at 160, and its phenylosazone melts at 157. Two other pentoses are ; RIBOSE, which, with nascent hydrogen, gives adonitol (a pentahydric alcohol, OH'CH 2 - [CH'OH] 3 'CH 2 'OH, and the only sugar -alcohol yet discovered in plants, the leaves of which are able to transform it into starch ; the sap of Adonis vernalis contains as much as 4 per cent, of adonitol) ; and d-lyxose, which is obtained from galactonic acid and melts at 101. Higher homologues are the Methylpentoses : FUCOSE, contained in algse ; CHINO- VOSE, ISORHAMNOSE, and RHAMNOSE (or Isodulcite), C 5 H ? 6 -CH 3 , which is obtained by boiling quercetin and other glucosides with dilute sulphuric acid. According to Rosenthaler (1909), Methylpentose in presence of pentoses can be recog- nised by heating the solution for a few minutes on a boiling water-bath with HC1 of sp. gr. 1-19 and observing the yellow liquid thus obtained in the spectroscope: methylfurfural, from methylpentose (even as little as 0-0005 grm.) gives absorption bands between the blue and green. The reaction is still more sensitive if a little acetone is added before heating, the liquid then being coloured red (by the methylfurfural) and giving a sharp absorption band in the yellow (D line) ; pentoses do not give this reaction if the liquid is heated. Other sensitive reactions are obtained with phloroglucinol, orcinol, resorcinol, pyrogallol, aniline acetate, &c. HEXOSES, C 6 H 12 6 ' These are of frequent natural occurrence and exist in various optically active stereoisomerides, since they contain four asymmetric carbon atoms, while they also form inactive racemic compounds. They are substances of sweet taste, and are extremely soluble in water, but in alcohol they dissolve but slightly and in ether not at all ; they crystallise with great difficulty and decompose when distilled. Their phenylhydrazones are soluble, and their phenylosa- zones insoluble in water. When boiled with hydrochloric acid they all give 432 ORGANIC CHEMISTRY (1) Levulinic Acid (CH 3 -CO-CH 2 -CH 2 -C0 2 H), the silver salt of which forms characteristic crystals, and (2) a brown amorphous mass of so-called humic substances. With methyl alcoholic ammonia, the hexoses form Osamines e.g. Glucosamine, C 6 H n 5 -NH 2 . They reduce Fehling's solution or ammoniacal silver solution in the hot, and with oxidising agents they yield hexonic acids and then lower acids down to oxalic. With lime they form additive compounds decomposable by carbonic acid ; with boiling milk of lime they turn brown and give Hexosaccharine (lactone of saccharic acid), C 6 H 10 5 . By the combined action of concentrated sul- phuric and nitric acids, they are converted into pentanitrates, while with alcohols and gaseous hydrogen chloride they form ethers (glucosides) . The aldohexoses give the fuchsine-sulphurous acid reaction (see p. 206), which is, however, not shown by the ketohexoses. The mode of formation of the phenylosazones is described on p. 427. With hydroxylamine they forjn oximes, e.g. d-Glucosoxime, which can be converted into the corresponding nitrile and then, by elimination of HCN, into the aldopentose (d-arabinose). The hexoses are formed in various organisms and can also be obtained by hydrolysing polyhexoses with dilute acids or enzymes. The optical activity of the hexoses indicated by the prefixes d-, 1-, and i- indicates the sign of that of the substances with which they are connected genetically, but the fact that the actual direction of the rotation does not always correspond with this prefix is a source of some confusion. It must also be noted that the rotatory powers of the hexoses and pentoses are lowered when the sugars are dissolved in a centinormal alkali solution at 37. Synthetically the hexoses can be obtained from formaldehyde (see Note, p. 428), as well as from the hexahydric alcohols by gentle oxidation and from the hexonic acids by reduction. E. Fischer has synthesised d-glucose com- pletely from glycerine, by way of (1) glyceraldehyde, (2) inactive fructose, which, with hydrogen, yields (3) inactive mannitol, oxidation of this giving (4) mannose and (5) racemic mannonic acid, the latter being resolved into its (6) active components by means of strychnine ; d-mamionic acid, in presence of pyridine and water in the hot, produces (7) d-gluconic acid and this, on reduction, d-glucose. The relations between hexoses and pentoses were indicated in the last Note (see p. 429). As was mentioned above, fermentation with yeast occurs only with d-glucose, d-fructose, d-galactose, d-mannose, and glycerose, no fermentation taking place with sorbose, the pentoses, 1-glucose, 1-fructose, 1-mannose, or d-mannoheptose. So that only the stereoisomerides of a certain group are fermentable. The structures of the hexoses are deduced partly from their general reactions and partly from the following facts : The chain of six carbon atoms in the hexoses is normal, since reduction with hydrogen yields a hexahydric alcohol, which is further reduced by heating with hydriodic acid to normal sec. hexyl iodide, CH 3 - CH 2 - CH 2 - CH 2 - CHI- CH 3 ; the constitution of the latter is shown by the fact that the corresponding secondary alcohol is oxidised to n-propylacetone,X)H 3 - CH 2 -CH 2 - CH 2 - CO CH 3 , this, on oxidation, giving finally butyric and acetic acids of known constitu- tion. The hexoses contain five hydroxyl groups, as they yield pentacetyl-deriva- tives when boiled with acetic anhydride and sodium : .acetate or zinc chloride. Their constitutional formula hence cannot be other than : GLUCOSE H H H H I I I I /H H 2 C C C C C C/ 1 1 J I I OH OH OH OH OH since, if two hydrox^yl groups were at any moment united with one carbon atom, a molecule of water would be eliminated immediately. Further, with hydrogen the hexoses form hexitols, which are not aldehydic but only alcoholic in character and do not give up H 2 O under any conditions, so that two hydroxyl groups are not combined with one carbon atom. Neither can it be supposed that three hydroxyl groups are united with the terminal carbon, thus : /OH C^-OH, because if this were so water would be readily separated and an X OH acid formed, in which case the aqueous solution should conduct the electric current and have a dissociation constant much greater than that of acetic acid ; but this is not found to be the case. Combination with bases does occur (with the hexabioses), but the com- pounds formed are additive compounds. Since then there are a number of different hexoses, all showing the same general behaviour, they must have the same constitution, the differences being due to differences in the spatial structure. Theoretically, 16 active stereoisomeric aldo-hexoses are possible, and 14 of them have been already prepared. The rotatory powers of the phenylosa- zones and phenylhydrazones may be of opposite signs to those of the corre- sponding hexoses. d-GLUCOSE (Grape Sugar, Dextrose, Starch Sugar), C 6 H 12 O 6 , is an aldose found in abundance in grapes and many other sweet fruits in company with d-fructose ; it also occurs in the urine of diabetic patients. It crystallises from water with 1H 2 0, which it loses at 120, and from alcohol in the anhydrous form, melting at 146. In aqueous solution it has the specific rotation + 53 at a temperature of 20, but it exhibits muta-rotation, the rotatory power being about double the above value in freshly prepared solutions which have not been boiled. Owing to its rotatory power, glucose can be estimated polarimetri- cally (see later, Sugar). When saccharose (a dextro-rotatory hexabiose) is heated with dilute acid, it is con- verted into a Isevo -rotatory mixture of equal proportions of glucose ( + ) and fructose or levulose ( ), which bears the name Invert Sugar, the change being known as inversion, since it is accompanied by alteration of the sign of the optical rotation. On oxidation, d-glucose gives d-Gluconic Acid, OH-CH 2 - [CH-OH] 4 -COOH, and then the dibasic Saccharic Acid, CO 2 H- [CH-OH] 4 -CO 2 H, which, like tartaric acid, gives a slightly soluble acid potassium salt ; the latter serves to characterise d-glucose, it being sufficient to oxidise with nitric acid and then precipitate the saccharic acid with saturated potassum acetate solution. When reduced, d-glucose yields d-sorbitol (hexahydric alcohol). The sugar forms a phenylosazone, melting at 204 to 205, and two phenylhydrazones, melting respectively at 115 and 144. When heated above 140, glucose is converted into caramel. In dilute solution it reduces Fehling's solution in the hot, and on this reaction is based the estimation of glucose. 1 1 Estimation of Glucose. In the chemical way the estimation is effected by means of Fehling's solution by the method described later in the section on Saccharose, about 10 grms. of solid glucose or 15 to 20 grms. of the syrupy product being dissolved in water, made up to 100 c.c. in a graduated flask and filtered through a dry, covered filter. Polarimetric estimation is not usually applicable owing to the presence of dextrin, some- times to the extent of 40 per cent., this increasing the rotation. The dextrin is determined by dissolving 5 grms. of the glucose in 400 c.c. of water, adding 40 c.c. of HC1 of sp. gr. 1-125, heating for two hours on a boiling water-bath, cooling, neutralising exactly with NaOH and making up to 500 c.c. The total dextrose (including that formed by hydrolysis of the dextrin) in this solution is now determined by means of Fehling's solution. The difference between the amounts of glucose found before and after the'action of acid, multiplied by 0-9, gives II 28 434 ORGANIC CHEMISTRY Barfoed has proposed the following reaction for detecting the presence of minimal quantities of glucose, (0-2 mgrm.) mixed with lactose, maltose, dextrin, and saccharose : to 5 c.c. of Barfoed's reagent (an acetic acid solution of normal cupric acetate) in a test- tube is added the dilute aqueous sugar solution (about 1 per cent.), the mixture being heated on a boiling water-bath for 3J minutes, allowed to cool for 10 minutes, and filtered. If the filter retains red cuprous oxide, the presence of dextrose is demonstrated. MANUFACTURE OF GLUCOSE. One hundred kilos of starch are mixed with 500 litres of water containing 5 kilos of concentrated sulphuric acid, and the mass heated to 40 to 50 and then introduced into a suitable autoclave or converter (conical or cylindrical, capable of withstanding 6 atmos.), coated internally with lead and externally with insu- lating material. A current of steam is then passed in and the temperature raised to 160. By allowing the steam to escape after this temperature has been reached, the empyreumatic oils (which are of disagreeable odour) are carried away ; the steam is condensed in cooled coils (the heat being used to heat water). The temperature of the mass is then maintained at 80 until a test portion gives no blue colour with iodine and no precipitate with lead acetate (or potassium silicate), these being indications of the saccharification of the dextrin and gummy matters ; a further sign of this is the non-formation of a precipitate with alcohol. The duration of the heating is S to 4 hours. The mass is then decanted into the neutralisation vats, which are furnished with stirrers, and finely divided calcium carbonate, suspended in a large quantity of water, gradually added in order to neutralise and precipitate the sulphuric acid. After thorough mixing of the mass, it is allowed to settle and the liquid then decanted into another vessel, where the calcium sulphate remaining in solution is precipitated by the addition of a little ammonium oxalate. The liquid is next filter-pressed, evaporated in a vacuum to 28 to 30 Be., decolorised in the hot by means of dry blood mixed with powdered wood charcoal or by passing through vertical filters filled with the charcoal similar to those used in sugar refineries. It is then concentrated in a vacuum (see Sugar Industry) either to 41 to 42 Be., to give solid compact glucose separating in the cooling vats (fitted with stirrers), or to about 65 Be., when ready formed crystals of glucose are added. The temperature is lowered to 18 to 20, and after 3 or 4 days the separated crystals centrifuged and so freed from the syrupy portion, which retains the dissolved dextrin and other impurities. To obtain granulated glucose the solution is concentrated only to 30 to 32 Be. ; after 8 to 10 days in the cold, a granular hydrated glucose separates. When a very dense liquid glucose (so dense that its specific gravity cannot be deter- mined with the ordinary hydrometers) is required, a little dextrin is left in the sugar so as to prevent crystallisation. The theoretical yield of pure glucose from 100 kilos of dry starch is 110 kilos. In some factories the starch is saccharified with a little nitric acid, which gives a less highly coloured syrup and is more rapid in its action. The nitric acid is then eliminated by means of sulphurous acid, which is oxidised at the expense of the nitric acid to sulphuric acid, this being readily precipitable with lime. The advantages of transforming starch into glucose by means of hydrofluoric acid consist in a rapid and complete hydrolysis, ready separation of the whole of the acid as barium fluoride, and the production of a glucose with a pure flavour. In 1901 Calmette found that, after heating crushed cereals with double the amount of the quantity of dextrin. The acidity should not exceed 2 c.c. of normal caustic soda per 100 grms. of syrup. The proportion of ash varies from 0-2 to 0-7 per cent. Solid commercial glucose contains 65 to 75 per cent, of glucose and the liquid 35 to 45 per cent. In pure solution, glucose can be estimated by means of the specific gravity : Density Degrees Per cent, of pure Density Degrees Per cent, of pure at 17-5 Be. glucose at 17-5 \ B6. glucose 1-0192 2-7 5 1-1310 16-4 35 1-0381 5-3 10 1-1494 18-8 40 1-0571 7-5 15 1-1680 20-6 45 1-0761 10-1 20 1-1863 22-7 50 1-0946 12-4 25 1-2040 24-4 55 1-1130 14-6 30 1-2218 26-1 60 FRUCTOSE 435 1 per cent, hydrochloric acid for 1 hour at 100, 1 hour at 110, and a third hour at 120, and then cooling, the mass may be converted completely into glucose by the action of Mucedince. USES. Large quantities of glucose are consumed for making sweet syrups, caramel, 1 fer- mented liquors, sweets and wine, preserving fruit, adulterating honey, dressing textiles, &c. In 1909-1910 Italy possessed 15 glucose factories, producing 676 quintals of the solid and 65,000 of the liquid sugar ; the total revenue from the tax of manufacture amounted to 51,000, solid glucose paying 32s. and the liquid 16s. per quintal. The Customs duty is 40,?. for liquid and 64s. for solid glucose per quintal. Importation is very f mall in amount and the exports only 300 quintals of liquid glucose. In the United States, where glucose is made from maize, the amount produced reached 800,000 quintals in 1907. In France the 16 factories working in 1908 to 1909 produced 200,000 quintals of glucose, about 48,000 being used in breweries while 72,000 were exported. England imported 46,500 tons (462,940) of glucose in 1909 and 62,500 tons (595,808) in 1910. The United States exported 90,000 tons, valued at 1,196,200, in 1911. The output in Germany, with about 26 factories, was as follows (quintals) : Solid Glucose Glucose Syrup Caramel Produced Exported Produced Exported Produced 1897-1898 72,000 348,000 1901-1902 99,400 28,874 492,700 76,800 48,000 1902-1903 96,170 12,026 545,300 30,620 1903-1904 75,050 6,113 469,461 . 1 3,000 35,630 1904-1905 53,000 2,890 324,340 10,432 34,690 1905-1906 91,900 582,750 43,000 1906-1907 88,300 477,506 . 44,244 1907-1908 81,836 466,340 48,461 1908-1909 87,623 506,600 44,180 The diminished production in 1904-1905 was due to a poor potato crop and an over- production of beet-sugar, the less an.cunt exported being caused partly by enhanced prices and partly by increased production in other countries. Glucose syrup with a specific gravity of 42 Be. is sold in Italy at 48s. to 52s. per quintal, whilst in Germany, where there is no manufacturing tax, it costs about 28s. ; the crystalline sugar costs rather more. d-FRUCTOSE (Levulose, Fruit-Sugar) occurs abundantly, together with glucose, in sweet fruits, and is also found in large quantities in honey (which contains natural invert sugar). The hydrolysis of inulin (a polyhexose found in dahlia tubers) yields d-fructose alone. The sugar is Icevo-rotatory and fermentable. It has the constitution of a ketose, OH CH 2 [CH OH] 3 CO CH 2 OH, hydrolysis of its cyanohydrin giving the heptonic acid, OH-CH 2 -[CH-OH] 3 -C(OH)-CH 2 -OH. COOH The phenylosazone of d-fructose is identical with that of d-glucose. Methylphenylhydrazine forms osazones only with ketoses and not with aldoses, with which, however, it forms colourless hydrazones, these being usually soluble and hence readily separable from the slightly soluble, intensely yellow osazones (see pp. 333 and 427). 1 Caramel (or sugar colouring) is prepared by fusing and heating glucose or saccharose at a temperature of 160 to 200 (not beyond this) in an iron vessel fitted with a stirrer. To glucose 1 to 3 per cent, of soda is also added to accelerate the operation and to neutralise the acid formed (saccharose also yields acid, being first partly inverted by the heating), and after the change is complete, 50 per cent, of hot water is added and the mass well mixed and filtered through charcoal. A brown, syrupy mass is thus obtained which is soluble in water or alcohol, giving a brown or yellow solution according to the dilution. That obtained from saccharose, which does not contain dextrin and dissolves completely in 80 per cent, alcohol, is used for colouring spirits, whilst that from glucose, which contains dextrin and is entirely soluble in 75 per cent, alcohol, is used to darken beer and vinegar. The presence of more than 5 per cent, of ash indicates that a caramel has been prepared from molasses ; good qualities contain only 1 per cent, of ash. In Germany caramel is exempt from taxation and costs about 32*. per quintal (for the production, see above). 436 ORGANIC CHEMISTRY When phenylosazones are heated gently with hydrochloric acid, they lose 2 mols. of phenylhydrazine with formation of osones which contain two carbonyl groups. Thus phenylglucosazone yields Glucosone, OH-CH 2 - [CH-OH] 3 -CO-CHO, and this when treated with nascent hydrogen (from zinc and acetic acid) takes up 2H at the terminal carbon atom, fructose being thus obtained from glucose. On the other hand, reduction of a ketose gives the corresponding hexahydric alcohol, which, on oxidation, yields the mono- basic hexonic acid ; the latter loses water, giving rise to the laetone, and this gives the a Idose on reduction. d-Fructose is lawo-rotatory ; [o] D = 92 at a temperature of 20. This sugar has been suggested for the use of diabetic and tuberculous patients and as a substitute for cane-sugar, since it is sweeter, and in syrups and honey it hinders the crystallisation of the other sugars. In view of these uses, attempts have been made to prepare fructose industrially. Honig in 1895 and Steiner in 1908 proposed its extraction from endive roots and dahlia tubers (these contain from 8 per cent, to 17 per cent, of inulin). The crushed tubers are treated in the hot (below 65) with a little milk of lime and with steam, and are then pressed. The juice, after defecation with clay, is allowed to crystallise in a rotating cooler, the mass of inulin crystals being centrifuged, redissolved in hot water, and converted into fructose by means of dilute acid (see Glucose) ; the 'fructose solution is concentrated in a vacuum. Steiner calculates that the sugar can be made by this process at a cost of 1 s: per kilo. A characteristic reaction for the detection of fructose in presence of other reducing sugars is obtained with the following solution: to a solution of 12 grms. of glycocoll in hot water are slowly added 6 grms. of pure cupric hydrate, the liquid being heated on a water-bath for about 15 minutes until complete solution takes place and then cooled to 60; after 50 grms. of potassium carbonate have been added, the volume is made up to 1 litre and the whole filtered. This reagent is reduced in the cold only by levulose (1 to 5 per cent, solution), the time required varying from 4 to 9 hours ; other sugars, including the pentoses, reduce it only at temperatures above 40. d-MANNOSE, C 6 H 12 6 , is an aldose stereoisomeric with glucose, and is fermentable ; it is obtained from mannitol, the corresponding alcohol, by oxidation. It melts at 195 to 200, and differs from other monoses in forming a phenylhydra/one only slightly soluble in water. With oxidising agents it forms first monobasic d-mannonic acid and then dibasic d-mannosaccharic acid, COOH-[CH-OH] 4 -COOH. A general method for converting one hexose into a stereoisomeric one, e.g. d-mannose into d-glucose, is as follows : the d-mannose is oxidised to d-mannonic acid and the latter dissolved in quinoline and the solution boiled ; in this way the acid undergoes partial transformation into the stereoisomeric d-gluconic acid, reduction of the lactone of which yields d-glucose. The reverse change of d-gluconic into d-mannonic acid is also produced to some extent by boiling with quinoline, so that d-glucose can be converted into d-mannose. These sugars (and acids) differ only in the space arrangement of the groups united with the asymmetric carbon atom in the a-position, OH'CH 2 -[CH'OH] 3 'CH(OH)'CHO, since the phenylosazone of d-mannose is identical with that of d-glucose, OH-CH 2 - [CH-OH] 3 -C -G II N-NHC 6 H 5 It is this a-carbon atom, adjacent to the aldehyde group, which is influenced when a hexonic acid is boiled with quinoline or pyridine. When glucose, fructose, or mannoseis treated with a very dilute alkali solution, a mixture of all three sugars results. The fructose seems to be an intermediate product, since the dextro-rotation of mannose gradually changes to a laevo-rotation, owing to formation of fructose, the amount of the Isevo-rotation subsequently diminishing as the fructose becomes converted into glucose. Z-MANNOSE and Z-GLUCOSE, C 6 H 12 O 6 (Aldoses), are obtained together from 1-arabi- nose by the cyanohydrin synthesis and reduction of the lactones of the resulting acids. Application of this synthesis to an aldehyde yields, in general, two optically active stereo- isomerides, since a new asymmetric carbon atom is created and the chances of formation of the two isomerides are equal. But the final mixture will be inactive only when the initial molecule is inactive, while, when this is optically active (as with arabinose), the mixture will be active, as the components will not have equal and opposite activities ; GLUCOSIDES 437 one of these will have a rotation greater than that of the original molecule by a certain amount and the other a rotation less by the same amount. d-GALACTOSE, C 6 H 12 O 6 (Aldose), is obtained by oxidising dulcitol, C 6 H 8 (OH) 6 , or by hydrolysing milk-sugar, in which case it is formed together with glucose. It melts at 168, is fermentable, and exhibits muta-rotation. It is an aldose, giving on oxidation first mono- basic d-galactonic acid and then dibasic mucic acid, COOH- [CH-OH] 4 -COOH, which is inactive. HEPTOSES, OCTOSES, and NONOSES have not been found in nature, but are prepared synthetically from mannose by means of the cyanohydrin synthesis. GLUCOSIDES These are of frequent occurrence in the vegetable kingdom and, when heated with acid or alkali or subjected to the action of certain enzymes, decompose into a glucose and an alcohol (or phenol, aldehyde, or nitrogen compound) ; they are hence ethereal derivatives of the monoses (e.g. Amygdalin, Salicin, Populin, Coniferin, &c.). Artificial glucosides have been prepared by E. Fischer by the interaction of an alcohol and a monose in presence of hydrochloric acid (which withdraws water). The glucosides are analogous in structure to the acetals R- C\ + 2CH 3 - OH = H 2 + R- COCHg (acetal), ^O X OCH 3 but only 1 molecule of alcohol takes part in the reaction : OH-CH 2 'CH(OH)-CH(OH)-CH(OH)-CH(OH)-CHO + CH 3 -OH = y |3 f the beet (hence the water-tank should have a capacity at least as great as 3 or 4 of the diffusors). Pressing and Drying the Pulp. The pulp (exhausted slices containing less than 0-5 per cent, of sugar) discharged from the diffusors is transported by a screw or endless band to an elevator which discharges it into the pulp-press (M, Fig. 305), where the water it contains (95 per cent.) is removed as completely as possible. Presses of various forms are used for this purpose. d s , V. When the juice reaches a s , the same operation is repeated, that is, a, (-, and d t are closed and c s , d t , and a, opened, so that the juice is forced from \ the preceding cylinders into VI through 5, d t , d,, and 6, the temperature being kept at 70 to 75. The juice from VI is not passed into VII, but part of it is first discharged (see later) into the juice measurer (and thence into a reservoir) by opening the main valve, M , and keeping d 7 shut ; when the amount in the measurer reaches a certain value, M is closed, A^ opened and diffusor VII filled from below in the usual manner. In all these cases the pressure is supplied by the water in the raised cistern. When VII is full, before the juice is passed into VIII, part of it is discharged into the measurer through the valve, M , as before ; _, ft these operations are repeated until the last diffusor is reached. When, however, IX is filled, it is advisable to discharge the water from diffusors I, II, and III, and to fill these with slices, so that, when the juice arrives at III, it is certain that the slices in IV (the first to be extracted) are exhausted ; IV is then emptied and recharged immediately with fresh slices. IV 4 V 5 VI 6 VII Fro. 304. The juice is then passed from III to IV, while V is discharged and recharged, and so on. The working thus assumes its normal course. In case of accident, the workman regulating the taps immediately shuts off t he steam and water, so as to prevent caramelisation of the juice and loss of sugar. The temperature is 35 in the diffusor following that which receives the fresh water, then rises to 60, and in the last diffusor (preceding that into which the water first passes) is 70" to 75. 454 ORGANIC CHEMISTRY That of the Kluscmann type consists of a vertical, revolving cone of perforated sheet' metal, C (Figs. 307, 308), fitted with oblique vanes and enclosed in a stationary cylinder, also perforated. The vanes, which are arranged helically alone the cone, compress the FIG. 305 FIG. 306. mass of pulp against the perforated cylinder and gradually move it downwards where the space becomes narrower, so that a considerable part of the water is squeezed out through the cone and cylinder, which are enclosed in a jacket, E ; all the water is carried off by the tubes F, G, and H, while the pressed pulp is discharged through the annular orifice, 7. An arrangement similar to this has also been combined with the pulp-elevator, which consists of an inclined screw, the pulp being thus raised and pressed at the same time. KLUSEMAKN PRESS 455 The Klusemaim press has been improved by Bergreen and others in order to diminish the amount of water left in the pulp. Each quintal of beet yields about 80 kilos of pressed pulp containing, on an average, 72 per cent, of water, 3 per cent, of ash, 1-8 per cent, of protein, 0-27 per cent, of fat, 6 per cent, of cellulose, and 17 per cent, of non -nitrogenous extractive matter. FIG. 309. The pressed pulp is loaded directly on the farmers' waggons to be used as fodder, about Wd. per quintal being paid for it ; but part of it (30 per cent, of the amount of beets they supply to the factory) is given to them free of cost. If the pulp cannot be sold immediately, it is stored in silos until sold. But if this is done, it readily undergoes putrefactive fermenta- tion, the gasogenic bacteria of which contaminate milk and cause inflation of cheese, so that in some countries, where fuel is not expensive, it is preferred to dry the pulp at once. It is known, too, that fresh pulp in silos loses as much as 40 per cent, of its solid matter, which is rendered soluble and volatile by bacteria, the sugar being converted almost completely into lactic acid. Of the various types of apparatus for drying the pulp, that of Biittner and Meyer (s$e Fig. 309), which was devised in 1887-1888 and rapidly came into use in Germany, 456 ORGANIC CHEMISTRY France, Belgium, and Austria, gives good results. In 1898 sixty German factories were employing pulp-driers on this plan. The moist pulp is raised by means of an elevator, p, and dropped at / into an upper chamber, B, composed of four semi-cylindrical channels containing mixers revolving in opposite senses, which stir and lift the pulp and at the same time transport it to the mixers of the similar chamber below ; thence it passes to a third chamber. A current of air at 400 from a furnace enters A at / and is moved in the same direction as the pulp by the aspirator, C, which then forces it into the dust chamber, D, and thence to the shaft. The pulp should issue at a temperature of 1 10 so that moisture may not condense on it, and the supply of pulp is regulated so that the final proportion of water present is 12 to 14 per cent. The composition of the dry pulp is as follows : 12 per cent, of water, 6-5 per cent, of ash, 8 per cent, of protein, 1-2 per cent, of fat, 18 per cent, of cellulose, and 55 per cent, of non-nitrogenous extractive substances (5 to 7 per cent, being sugar) ; it is sold in Italy at 6s. 6d. to 8s. per quintal. Recent tests made by Gorini (1911) show that the dry pulp is not sterile, and may hence be dangerous to milk during milking operations. THE STEFFEN PRO- CESS. Some years ago Carl Steffen patented (Ger. Pat. 149,593) a process of ex- tracting sugar from the beet without the use of diffusion, a process resembling that used by Achard 125 years ago (see Note on p. 448). The . beet slices (containing 75 to 80 per cent, of water) are pressed, giving a juice of 20 to 25 Brix. The remaining pulp is then heated to 85 with more dilute juice (15 to 17 Brix), which is thus enriched with sugar extracted from the pulp. The latter is compressed in a powerful press in the hot, the residual pulp being rich in sugar and hence of greater value for cattle-food. This process yields less molasses and more first-jet sugar, while it requires less expenditure of water, coal, and labour, and a less expensive plant, than when diffusers are used. For each quintal of beet there are 45 litres of water less to evaporate. The Steffen apparatus is shown diagramniatically in Fig. 310. The beets pass into an ordinary slicer, H, and the slices fall into G and then into a horizontal cylinder, M, containing the juice heated to 95 to 98 (600 litres of this juice and 100 kilos of cold slices give a mixture at 85). A horizontal screw, Z, transports the slices to T, where they meet a double- jacketed (the inner casing perforated) worm-conveyor, F, which raises them and presses them to some extent, so that the juice runs back into M . At the top of this conveyor they are discharged into a press of the type described on p. 455 (Figs. 307, 308). The expressed juice returns through the tube, V, to M, while the pulp falls into Y and is conveyed to the drying apparatus. In order to maintain the juice at a temperature of 85, part of it is continually forced by the pump, P, through the tube, X 2 , to the sieve, K 2 , then to the heater, C^, and through X 3 to the cylinder, M ; if necessary, steam is injected by means of the injector, C 2 . In order to dilute the juice in M so as to keep it always at 15 to 16 Brix, dilute sugar solution from the washing of the defecation mass in the filter-presses (see later) is introduced both directly into the cylinder, M, at R 2 an d into the inclined conveyor at J? a . The excess of juice flows continuously through the funnel, A, to the sieve, K l , which retains finely divided pulp, and then through the tube, Xj_, to the 1 defecation apparatus. This process admits of the rapid treatment of large masses of material, which is heated to 85 in 2 to 3 minutes and yields 70 to 80 per cent, of juice purer than diffusion juice and about 30 per cent, of pulp (containing 70 per cent, of water and 10 per cent, of sugar), which, after drying, contains 10 per cent, of water, 7-6 per cent, of proteins, 0-4 per cent. FIG. 310. DEFECATION OF THE JUICE 457 of fat, 10 per cent, of cellulose, 36 per cent, of non-nitrogenous extractive matters, 52 per cent, of sugar, and 4 per cent, of ash ; the expense of drying in Germany is about Q-5d. per 100 kilos of the dry pulp. The diminution of 2 to 2-5 per cent, in the yield of commercial sugar is compensated in various ways ; the dry pulp is worth about three times as much as diffusion pulp and is sold in Germany at lls. per quintal, in addition to which the diffusion process leads to various small absolute losses. It must be admitted that, after many trials and much discussion, during recent years, the most competent technical opinion varies with regard to the advantages claimed by the Steffen process. It can, however, be stated that only the most efficient diffusion plant can compete with the Steffen process, which up to the present has been found most advan- tageous in districts and in seasons in which prices for the dried saccharine pulp are more favourable than those of raw sugar. In 1910 a dozen factories in Germany alone produced 1,300,000 quintals of sugar by the Steffen process. In a new process devised by Claassen, all the water from the diffusion of the molasses and that resulting from the pressing of the exhausted pulp are used directly for the extrac- tion of the sliced beets in the diffusors. In this manner all the soluble substances of the beet are returned and utilised, so that an increased yield of sugar is obtained with a diminished consumption of water. This process requires, however, much supervision and care. The new process devised by Hyros and Rak employs more perfect machinery than the Steffen process, yet is identical with the latter in many points ; but the heating to 85 is carried out in three stages and the final pulp is not dried. This process has been little used; but, according to Herzfeld, could be combined advantageously with the Steffen process. Other processes, such as those of Bosse, Naudet, Garez, &c., are concerned mainly with the rapid heating of the slices below the slicing machine, pressure or diffusion then being employed. Juice Measurers. These are special automatic apparatus used to measure the juice extracted at intervals from the diffusors, each such quantity of juice being registered automatically on a strip of paper together with the time elapsing between one discharge and the next. This paper serves to control the working, while it also indicates any stoppages taking place. The underlying principle of such apparatus is the same as that on which alcohol meters (see p. 146) are based. The juice is then discharged through coarse filters to remove vegetable fibres, which are eventually rejected. This dilute juice (10 to 12 per cent, of sugar) has a reddish brown colour and is further subjected to a series of operations, to be described below. Defecation with Lime. In addition to sugar, the juice extracted from the beet contains proteins, pectic substances (colloidal substances of the carbohydrate group), and mineral salts. The pectic matters readily ferment, giving two gummy acids (Pectic Acid, CsoHnOso, and Pectosinic Acid, C 3 2H4 6 O 31 ), which convert the juice almost into a gelatinous mass and partially invert the sugar. When the fresh juice is treated with lime, if the latter is not in excess, "insoluble calcium pectates separate ; whilst if excess of lime is present, the juice is liable to lactic and butyric fermentation of the proteins with development of unpleasant odours. If the lime is added to the hot juice, no fermentation occurs and the whole of the organic impurities are preci- pitated ; but it is not possible to avoid a slight excess of lime, which forms insoluble tricalcium sucrate, so that the mass cannot be filtered immediately. Loss of sugar in this way is obviated by passing carbon dioxide through the turbid liquid, this readily decom- posing the sucrate with formation of calcium carbonate and liberation of the sugar. Excess of carbon dioxide must, however, be employed, since otherwise an insoluble double com- pound of sugar with calcium carbonate is formed. The operation of saturation with carbon dioxide must be controlled rigorously and continuously in the laboratory, since it is the principal source of loss. The treatment of the juice with lime is carried out at 85 in suitable vessels provided with stirrers. The lime is added in the quantity previously determined in the laboratory (2-5 to 3-5 per cent.), and may be as powder or in the form of milk of lime, the concen- tration of the latter being measured by means of automatic floating densimeters. Kowalski and Kosakowski have recently shown that if, as was long ago recommended, the juice is 458 ORGANIC CHEMISTRY FIG. 311. well agitated during defecation and heating, the total quantity of lime required may be reduced to as little as 1 -5 per cent. In France and Germany saturation with carbon dioxide is carried out in two phases, and in Austria in three phases. 1 The lime in the juice is estimated by means of soap solution (Pellet's method) in a way similar to that used to determine the hardness of water (vol. i, p. 215). In order to avoid the risk of redissolving the calcium carbonate (as bicarbonate), the saturation is first carried on for 20 to 40 minutes at a temperature of nearly 90 until a certain degree of alkalinity remains (0-11 to 0-13 percent.) ; the juice is then filtered, heated, saturated again for about 15 minutes until the alkalinity falls to 0-02 to 0-04, and finally filtered a second time. In Austria and Bohemia, however, a little lime (0-5 to 1 per cent, leaving an alkalinity of 0-05 to |oi 0-07) is added before the second saturation in the hot (95). The juice is then filtered and the third saturation carried out at 100 (10 minutes), the alkalinity being reduced to ,0-01 to 0-03. After a fresh filtration, the juice is thoroughly heated for a long time in another boiler, again filtered and despatched to the concentrators. In some fac- tories the third saturation is now made with sulphur dioxide, which has a greater purifying action than carbon dioxide and at the same time decolorises the solution. Liquid sulphur dioxide may be employed, but it is cheaper to produce the gas in furnaces (see vol. i, p. 244). In some works continuous saturation is practised, but the Austrian system seems to be the best, even though it leaves 0-06 per cent, of alkalinity. The iron saturation vessels (Figs. 311, 312) are pro- vided at the top with a large tube for the escape of the excess of gas. That used for the first saturation is often 7 metres high, but is filled with juice only to the height of 2 metres (30 to 50 hectols.), the remainder of the space gradually being filled with a dense froth ; that for the second saturation is 3 metres high, less foam being formed in this case (a large saturation chamber is shown in Fig. 313). If too much froth forms, it can be reduced by the addition of a little coco-nut oil. The juice is heated for the first saturation by means of a steam-coil, and the carbon dioxide is introduced at the bottom by a perforated tube, 6. A glass is inserted to permit of the operation being viewed, and a closed orifice, E, serves for the inspection and cleaning of the interior. The completion of saturation is shown by phenol - phthalein paper, which ceases to turn violet. Trained workmen also carry out titrations. A plant for saturation with sulphur dioxide is shown in Fig. 314. The air-pump, A, feeds the sulphur furnace, B, and the mixture of air and sulphurous acid then passes through the tube, C, into the saturator, D, the excess issuing by the tubes, E. 1 Thelime and carbon dioxide used in sugar-works are generally prepared in a vertical lime-furnace (see also vol. i, p. 489), the upper outlet of which communicates with one or two water-cisterns, into which the gas is drawn by an aspirator to be- washed and cooled before being conveyed to the saturators. Chalk of good quality (free from iron and containing little sulphate or silica) is used and is mixed with 9 to 10 per cent, of coke (anthracite should be avoided, in order to prevent the presence of odorous and tarry impurities in the gas). The gases contain about 30 per cent, of CO. 2 , and the size of the suction-pump is calculated on the basis that every quintal of lime produced corresponds with at least 300 cu. metres of gas. The treatment of 5000 quintals of beet per 24 hours requires about 300 quintals of chalk (occupying, in lumps, about 15 cu. metres), which give 170 quintals of quicklime with a consumption of about 85 quintals of coke (9-3 cu. metres in lumps). FIG. 312. FILTRATION OF THE JUICE 459 Continuous saturation processes with a counter- current of juice and carbon dioxide (Horsin-Deon, Raboux) are also used, but they do not seem to have any great advantage. Behm, Dammeyer, and Schalmeyer propose to purify the juice at 75 with a current of 40 to 50 amperes at 6 to 8 volts for 8 to 10 minutes, using zinc electrodes. This FIG. 313. treatment seems to result in the deposition of various organic impurities, but, although promising well, the process has not been adopted. Filtration of the Defecated, Saturated Juice. The precipitated calcium carbonate is separated by passing the juice through filter-presses?- which allow the clear sugar-juice FIG. 314. to pass through and retain the suspended impurities and the calcium carbonate in the form of cakes, which, after being washed, are readily extracted by unscrewing the press 1 Filter-presses are formed of a number of iron frames, alternately empty and fllled in and supported on two horizontal, parallel rods. An empty frame is shown at A (Figs. 315, 316) and a fllled-in one at B (Fig. 317). The latter is fllled in with sheet-iron grooved on both sides, the grooves ending below in two horizontal channels 460 ORGANIC CHEMISTRY and removing the frames ; they fall into conveyors or trucks underneath, and are often used as lime fertilisers. The first wash-water is added to the filtered juice, while the last is used to slake the lime for defecation. The pressed cake should contain less than 0-6 per cent, of sugar. The filtering surface of the filter-presses necessary after the first saturation is calcu- lated at 0-5 sq. metre per ton of beet worked in 24 hours ; after the second saturation communicating with a single tap, r (Fig. 317) ; the grooves of the two sides are covered with a perforated plate. On the empty frames are stretched cotton or linen cloths, which form two filtering surfaces of the same area as the frame. j[ The frames arc squeezed together and against the strengthened block, P, by the screw, V, so that FIG. 316. rani FIG. 315. FIG. 317. 246 hermetic joints arc formed at the edges of all the frames. Each frame is provided with bored projections, a and b, at the top and bottom. When the frames are joined up, the holes in the projections form two continuous channels. The turbid juice enters at a and thence passes through nf into all the empty frames, the air being forced out from these through the valve, d. When d is closed, the juice passes under pressure through the cloths on the two sides and the clear liquid flows down the grooves and is discharged at r into the tank, S. When the frames, A, are filled with calcium carbonate, the latter is washed with water to remove the sugar it retains. Since only the alternate grooved plates communicate with the tube, b, water intro- duced under pressure at b will pass through the cakes of calcium carbonate in the direc- tion of their thickness and into the grooved plates (not communicating with b) to be discharged at the taps r. In this way, each cake is brought into thorough contact with the washing water, which can be measured inS. In other filter-presses there are no empty plates (Figs. 318, 319), but each of these has \a central aperture over which the filter- cloth, with a hole exactly in the middle, is screwed with a ring from botli sides. The juice is introduced into the chambers between adjacent plates, and the wash-water passes under pressure into alternate (odd) plates from the tube, m, traversing the cakes, and collects in the other alternate (even) plates which communicate not with m but with k, the wash-water being thus discharged ; the air is initially discharged from the odd frames through i. Each press contains 20 to 50 plates, each 3 to 5 cm. thick, and with a length of side 60 to 100 cm. The juice to be filtered is pumped in under a pressure of 3 to 4 atmos. FlG. 319. CONCENTRATION OF THE JUICE 461 0-25 sq. metre suffices. The pressed cakes of chalk form 12 to 14 per cent, of the weight of the beets (i.e. four times the weight of quicklime used). The washing of these cakes requires 1 litre of water per kilo. After the second and third defecations, use is often made, not of niter-presses but of mechanical filters, which also serve for removing suspended matter and residues of the slices from the diffusor -juice. During the whole of its course from the diffusors and saturators, the juice is under pressure and should rise in temperature from 70 to 100 ; but since heat is lost in all the pipes, in order that monocalcium sucrate may not be deposited or the liquor become turbid, the use of heaters is necessary for the first and second saturation juices, &c. These heaters consist of a species of tubular boiler divided into three parts by two plates, p (Fig. 320) ; each of the two end parts is divided into 10 chambers communicating in pairs at the two ends alternately. Opposite chambers are connected by groups of long tubes, 4 to 5 cm. in diameter, through which the juice circulates ; steam enters at C, follows a sinuous path round the partition?, V, and finally issues at D. The juice enters, at A, chamber 1 of compartment I, and passes through the tubes to chamber 1 of compart - FIG. 320. ment II, then to chamber 2 of compartment II, through the tubes to chamber 2 of com- partment I, and so on, until it reaches chamber 10 of compartment I and hence leaves the heater at B. After the third saturation the juice passes into a final heater or boiler, where it is thoroughly boiled but not under pressure. The juice is moved by means of pumps, a separate one being used after each operation (for raw juice, first saturation juice, second saturation juice, &c.) ; double-action piston pumps or Girard pumps, with an efficiency of 80 to 85 per cent., are employed. IjWhen the tax is based on the volume and density of the defecated juice, before the latter goes to the evaporators it passes into tanks under the supervision of the Inland Revenue authorities, who measure the density at 85 to 90 and then reduce it to the normal temperature by means of tables. Thus, in Italy, up to 1903, tax was paid on 2000 grms. (before 1900, only on 1500 grms.) of sugar for every hectolitre of juice and every one-hundredth of a degree of density above 1. [CONCENTRATION OF THE JUICE. The defecated, saturated, and filtered juice is pale yellow and perfectly clear ; it contains 88 to 90 per cent, of water, 10 to 1 1 per cent, of sugar, and 0-8 to 1 per cent, of salts. The formation of crystallised sugar requires first considerable evaporation or concentration and then boiling. The suggestion was made in 1907 to concentrate juice by freezing and removing the ice (E. Monti, Ger. Pat. 194,235), but so far as is known this process has yielded no practical results. Evaporation is nowadays carried out excusively with indirect steam and in multiple- effect vacuum plant. The vacuum is obtained by means of pumps, combined with a baro- metric water -column, this method being introduced into the sugar industry by Rillieux (see modern triple- and sextuple-effect apparatus, vol. i, p. 442 et seq.) ; it admits of FIG. 321. ORGANIC CHEMISTRY considerable saving in fuel and avoids blackening or caramelisation of the juice, the boiling-point being lowered as the vacuum is increased. 1 The steam from the first body is not all utilised for the second body, part of it, and also part of that from the second body, serving to heat other plant (boilers, heaters, &c.). The first body is usually heated by the exhaust steam from the engines, which it leaves at a pressure of 1-5 to 2 atmos. The evaporation bodies are simply large wrought or cast iron (formerly copper) boilers surrounded by an insulating earth. These bodies are of various shapes and are placed sometimes vertically and some- times horizontally. They are usually divided into three com- partments by means of two partitions, held rigid by a number of brass tubes, 2 to 2-5 cm. in diameter, connecting the first and third compartments. In boilers with horizontal tubes (Figs. 321, 322, 323) the steam circulates in the tubes in a similar manner to the juice in the heater described above (Fig. 320), while the juice surrounds all the tubes. In vertical bodies (Fig. 324) the steam, entering at A and issuing at B, circulates in the chamber between the two partitions and heats the numerous connecting tubes. The saccharine solution is thus brought into a condition of vigorous ebullition and circulates rapidly between the lower and upper chambers, as indicated by the arrows in the figure. The level of the liquid, which can always be controlled by the^external glass tube, a, is kept just above jthe tubes ; in this way, less scum is formed, the free vapour space is increased, and danger of caramelisation is avoided. The boiling may be observed through the^window, r. In order to separate the drops of liquid carried away in the steam,' about two -thirds of the way up the JIG & boiler is placed a plate, P, with a large central aperture, G, above which is arranged a kind of metal umbrella, p, at a height adjustable by the levers, e, w, and h. This height is chosen so that the liquid condensing above P contains no sugar. But with horizontal evapo- rators the spray separators consist of large cylinders placed above the boiler (F, Fig. 321). The steam issuing from the boiler by the tubes, E, before passing to the exit pipe, H, traverses the finely perforated vertical plates, 8, which retain the drops of solution carried over by the steam, this effect being facilitated by the expansion and consequent slackening of the steam in F. The condensed liquid is returned to the bottom of the boiler by the tube, G. 1 The boiling-point of water for different degrees of vacuum is as follows (Regnault-Claassen) : with a vacuum of 50 mm., 98-1 ; 100 mm., 96-1 ; 150 mm., 94 ; 200 mm., 91-7 ; 300 mm., 86-5 ; 400 mm., 80-4 ; 500 mm., 72-5 ; 600 mm., 61-6 ; 650 mm., 53-6 ; 700 mm., 41-7 ; 720 mm., 34-2 ; 740 mm., 22-4 ; 750 mm., 11-8. It must, however, be remembered that saccharine solutions boil at higher temperatures than water. Thus, under the ordinary pressure, a solution containing 30 per cent, of sugar boils at 100-6 ; 60 per cent., 103-1 ; 80 per cent., 110-3 ; 85 per cent., 115. FIG. 322. EVAPORATORS 463 In exceptional cases, where a large amount of spray is persistently formed, this may be diminished by the addition to the boiler of a small quantity of coco -nut oil. Fig. 325 shows a triple-effect horizontal evaporator of the Wellner-Jelinek type, Fig. 326 a vertical triple effect, and Figs. 327 and 328 a vertical quadruple effect evaporator, , , ,. .' .. ' Fiu. 323. Fiu. 324. in elevation and plan. In the last of these, the steam passes from the body I through the tubes, a and a', to heat body II, the steam from which heats body III, and so on ; the steam from the last body, IV, proceeds through the tube, N, to the vacuum pump and the barometric water-column. FIG. 325. The evaporation in the separate bodies takes place under reduced pressure in the following manner. In a quadruple effect, steam at a temperature of 110 to 120 (from a boiler or from the exhaust of an engine) enters the tubes of body I, which contains a juice already concentrated to a considerable extent in the other evaporation bodies. Since the steam generated in body I proceeds to the heating tubes of II, where it condenses, a partial vacuum (e.g. of 150 mm.) is established in body I, in which boiling will hence take place at a temperature, say 94, below 100, and this will be the temperature of the steam which 464 ORGANIC CHEMISTRY heats the second body. But the steam given off by the rather more dilute solution of body II is larger in amount, so that a more marked vacuum (up to 350 mm.) is produced by the vigorous condensation of this steam- in the heating tubes of body III ; the sugar solution of body II will hence boil at about 84, and steam at this temperature is able to boil the more dilute liquid of body III, where the vacuum may be as high as 380 mm., PIG. 326, FIG. 327. PIG. 328. corresponding with a boiling-point of 74. The steam here produced goes to boil the defe- cated diffusion juice, which is introduced into body IV ; the latter is connected with the vacuum pump and with the barometric column, which produce a vacuum of about 630 to 640 mm., corresponding with a boiling-point of about 56. When the solution in body I has attained the desired concentration, it is discharged and replaced by that from body II, which in its turn is filled by that from III. The latter WATER-COOLERS 463 is then charged with fresh juice by means of the tube, (Fig. 32ft), which also serves for the passage of the juice from one vessel to the other. The steam is circulated between the various bodies by the tubes G and H, and every battery of heating tubes is connected with a condenser and separator, I. Fig. 329 shows the arrangement of the three barometric columns, M, P, and R, which produce the vacuum directly in the third body of a triple -effect evaporator (for small single- or double-effect plant one barometric column suffices). The pipe, K, conveys the steam from body 3 to the chamber, L, furnished with an iron barometer tube, M, at least 12 metres long, which dips into a well or water -tank, T. The condensation water collects in the tube, M, to a height corresponding with the vacuum formed in L, and hence in the body 3. But the majority of the steam condenses in the chamber, N, into the top of which the tube, O, introduces a fine cold-water spray which produces an abundant and rapid con- densation of steam and a considerable lowering of pressure, so that a large quantity of hot water passes into the vessel, U, from the barometer tube, P. A little steam condenses in the chamber, Q, communicating by the tube, 8, with the suction pump which main- tains the vacuum. The vacuum pump can also be connected, by means of three narrow tubes, with the three evaporation bodies, in which the vacuum can be regulated as desired. It is evident that in the three evaporation bodies, especially in P, the water must not be kept at too high a temperature, so that it may not evaporate in its turn and may help the con- densation of the steam. Certain sugar -works have recently made successful FIG. 329. use of the Kestner concentration system (see vol. i, p. 443), which gives an evaporative efficiency superior to that of the ordinary multiple -effect apparatus. Indeed, when there is a difference of, say, 7 between the temperature of the steam in the boiler (e.g. 135) and that of the heated juice (e.g. 128), an evaporation of 80 kilos of water per square metre of heating surface (i.e. 11-4 kilos per degree of temperature difference), is obtained with a reduction of the coal consumption to 5 kilos per 100 kilos of beet. In factories where there is not an abundance of water (that required by vacuum plant is ten to twelve times the quantity of juice to be concentrated), it is convenient to utilise the hot condensed water from the steam-engines (an engine of 350 to 400 h.p. requires about 1 cu. metre of water per minute for condensation) and that from the vacuum concen- tration batteries. This water is cooled in suitable atmospheric coolers, T (Fig. 330), so that it can be used in the barometric tubes and also for the washing and hydraulic trans- port of the beets. The tank, K, corresponds with that marked U in the preceding figure, A pump, A, forces this water to the top of the pile, T (see also vol. i, p. 454), whence it flows down over the faggots built up under a kind of hood, which produces a strongupwar draught of air and so evaporates and cools the water (e.g. from 50 to 60 down to 25 to 30). The latter collects underneath in the tank, r, and is then transferred by the pump, M, to the chamber, F, where the dissolved air is separated and passes out through the pipe, g (higher than O). The water rises in the tube, O, to the top of the barometric condenser, C, II 30 466 which is evacuated by the pump, B, and the tube, n ; the pipe, F or S, corresponds with the tube, K, of the preceding figure and communicates with the third evaporation body. Other more efficient arrangements are also used for the cooling of the hot water. Fig. 331 shows a system consisting of numbers of vertical rods arranged in layers crossing one another in a manner similar to those of the apparatus depicted in Fig. 244 (p. 283). The hot water, entering by the pipe, A, is distributed homogeneously by means of the tooth- edged channel, C, and collects in the vessel, B, underneath ; the air drawn upwards between the rods carries with it a cloud of steam. Another arrangement is shown in Fig. 332 ; here a wooden cap or cover fits over walls composed of sticks arranged in the form of Venetian blinds, while at the bottom a Korting injector produces a powerful jet of pulverised water in the shape of an inverted cone. The upward air-current evaporates the water while the latter ascends or while it flows down in a thin film on the boards (in this manner only 4 per cent, of the water is lost). Equally ingenious and simple is the cooling effected by forcing the hot water under pressure into a circular pipe fitted with a number of Korting pulverisers, catching the water in a large tank and, if necessary, passing it again through the pulverisers (Fig. 333) ; but by this procedure more than 10 per cent, of the water is lost. In those seasons of the year and on those days when the air is warm and dry, the temperature of the water can generally be reduced to that of the air ; but if the air is cold and not very dry, the temperature of the water remains 6 to 7 above that of the atmosphere. BOILING OF THE CONCENTRATED JUICE. The juice from the evaporators has a density of 28 to 30 Be. ( = 50 to 55 Brix) and an intense brown colour, and in order to induce crystallisation of the sugar it is necessary to concentrate it until not more than 15 per cent, of water remains (85 Brix). This concentration or boiling is carried out in simple vacuum boilers or vacuum pans, the juice being first filtered through mechanical filters, collected in tanks and drawn into the pans which are already evacuated. These pans resemble ordinary evaporators and are made of sheet-iron ; they may be either horizontal (like that shown in Figs. 311 and 312) or vertical. In the lower part of the pan is a dense coil of copper or brass pipes arranged either in a zigzag manner or in concentric circles, and through these passes the steam (Fig. 334) ; in some cases, however, the bottom of the pan is steam -jacketed (Fig. 335). The concentration or boiling is carried BOILING OF THE JUICE 467 out at as low a temperature as possible and the pan is fitted with a froth-separator (see Fig. 311), a tap for the removal of test-samples of the mass towards the end of the opera- tion, and a wide discharge pipe, K. The first thing to be done is to evacuate the pan by connecting it with the condenser and with the vacuum pump. Next the cock of the tube dipping into the concentrated juice tank is opened, the required quantity of juice being allowed to enter. Steam is then passed FIG. 331. FIG. 332. through the heating tubes. During the boiling, the level of the juice is not allowed to fall beneath the top of the heating tubes, since otherwise sugar would dry on these tubes and be decomposed ; so that fresh concentrated juice is introduced from time to time. At a certain stage of the concentration small crystals begin to form and gradually increase in size. The operator extracts samples and spreads them out on glass in order to, ascertain the size of the crystals and the density of the mass, and when he considers that sufficient of this massecuite consisting mainly of crystals with a certain amount of dark molasses FIG. 333. has been deposited on the tubes, the heating is stopped and the ordinary pressure estab- lished in the pan. The whole mass is then discharged from the outlet, K, into a large vessel furnished with stirrers, where it is gradually cooled and the crystallisation completed. The boiling and discharging of the massecuite occupy altogether about 10 hours. Fig. 336 shows a battery of Bock cylindrical crystallisers fitted with stirrers. Larger crystals are obtained by adding to the crystallising vessels a little unboiled juice, which lowers the sugar-content somewhat and retards the crystallisation. When no further crystallisation takes place, the mass is discharged, by means of a parachute at the bottom of the crystalliser, into the centrifuges, which readily separate the liquid molasses from the solid sugar. 468 ORGANIC CHEMISTRY This process of boiling is termed boiling to grain to distinguish it from the boiling to thread, now used only in refining. In the latter case the boiling is not continued until crystals form, the proper density of the boiled juice being ascertained by squeezing a drop between the finger and thumb and then sharply withdrawing the finger ; if a filament is Fret. 334. Fia. 335. thus formed, the boiling is not finished, but the breaking of the thread with formation of two projections indicates the end of the boiling. The syrup is then poured into moulds, which are kept lukewarm until the whole mass sets to an almost solid block composed of finer crystals than in the preceding case. FIG. 336. CENTRIFUGATION OF THE FIRST MASSECUITE. The centrifuges for the massecuite have drums of perforated steel with an inner coating of fine -meshed gauze. The diameter of the drum is about 80 to 100 cm., the height 40 to 45 cm., [and the speed of rotation 800 to 1000 per minute. The motive force is applied underneath, and the CENTRIFUGATION OF MASSECUITE 469 centrifuged sugar remaining in the drum is discharged either above (Fig. 337) or through a door which can be opened in the base of the drum (Fig. 338). The massecuite is passed directly from the crystallisers to the centrifuges, and, in order to effect more complete separation of the molasses adhering to the surface of the crystals, especially in the layer adjacent to the gauze, so-called covering or clearing is resorted to ; while the centrifuge is still in motion, the sugar is sprayed with finely divided cold or tepid water (Fig. 339), or even with a jet of steam applied inside or, better, to the outside of the basket, the molasses being thereby ren- dered more liquid. This pro- cedure naturally gives a whiter raw sugar (first product) but in diminished yield, a small part of the sugar being carried FIG. 337. away with the molasses by the water. This loss is diminished by using, in place of water or steam, sugar juices (syrups) gradually increasing in purity, so that the molasses and less pure syrups are removed and the sugar left covered with a solu- tion of pure sugar. In this way minute, moderately white crystals of sugar are obtained, and these are sometimes placed on the market without refining. But the public suspects them of being adul- terated and prefers quite white crystals or cubes. The molasses from the centrifugation of the first massecuite, after separation of the first-product sugar (first runnings), is further concentrated and boiled in syrup pans, which are similar to vertical evaporators and are worked under a vacuum, but are usually of single effect. FIG/ 338. The boiling is continued until the syrup gives a long thread (see above), the impurities present preventing boiling to grain. This second massecuite is then placed in large tanks in the molasses room, where it is kept for 25 to 30 days at a temperature of 35 to 40. The blocks of crystals which separate are broken up with suitable bladed machines, and are then delivered to the centrifuges by means of screws or piston pumps. The result- ing second-product sugar is rather yellow. The molasses which then separates is further concentrated and the third massecuite sent to the molasses room, but no more sugar separates, since the various potassium and other salts present prevent about five times their own weight of sugar from crystallising. This molasses is hence fold as it is for the preparation of cattle-foods or for the manufacture of spirit (see p. 140). In some 470 ORGANIC CHEMISTRY countries, however, it is treated by special processes for the extraction of the sugar still present. 1 Every 100 kilos of beet treated yield 1 to 3 kilos of molasses. The first- and second-product sugars from the centrifuges are sent to the stores, where they are sieved to break up the crusts, which retain, molasses. The two products are often mixed, put up in bags holding 100 kilos, and despatched to the refinery. SUGAR-REFINING. The raw sugar (first and second products, with a purity of 88 to 96 per cent.) is not usually placed on the market, but is purified in refineries, where it is dissolved in hot water, the purer and less coloured qualities of high rendcmcnt z being kept separate from the more impure grades of low rendement. The solution, with a density of 37 to 39 Be., is treated with a little lime, with 3 to 4 per cent, of animal charcoal and often with 2 per cent, of ox-blood, after which it is boiled, the frothy crust forming at the surface being continually broken. The suspended matter is then- removed by rapid me- chanical filters or by filter-presses. The residue (refinery black) is utilised as a manure, while the hot and still coloured solution is passed through a battery of f >ur or six tower filters, 8 to 9 metres i i height and 60 to 80 cm. in diameter, fi led with animal charcoal (Fig. 340 : A, tube for dense juice, B for dilute juice, C for water, D for steam) and previously heated with steam (D) to prevent the sugar separating and to obtain the maximum decolorising action of the charcoal, this being exerted in the hot. The animal charcoal or bone-blaek has a considerable affinity for colouring- matter and for lime, but only a slight one for sugar. But in course of time the pores of the charcoal become obstructed and its decolorising power diminished, so that after a few weeks it becomes necessary to revivify the charcoal. 3 The solution is passed through the filters in succession and, if necessary, this procedure is repeated. When the syrupy liquid is decolorised, it is con- centrated and boiled in ordinary single - eTect vacuum pans (of copper) until it shows the grain or short-thread test (see above). When the massecuite reaches this degree of concentration, it is poured into a jacketed copper vessel, in which it is kept at 85 to 90 to initiate the formation of large crystals. 1 In some works the second product is obtained much more rapidly by the Bock or the GrosSc process. In 'the first of these, the molasses is not left for 25 to 30 days in the molasses room but is crystallised in 4 to 5 days by continually shaking in large, jacketed drums heated to 90 to 95 and adding a considerable quantity (25 to 30 per cent.) of crystallised sugar. It is then allowed to cool slowly, but at certain times it is heated one or two degrees above the temperatures it shows at those times, so that the smaller crystals formed, and these only, are redissolved. When the mass has been cooled to 35, the crystalline blocks are crushed and centrifugcd, the amount required (25 to 30 per cent.) to induce the molasses (see above) to crystallise being previously removed. In the Grosse process, the mass is kept in motion by a vertical Archimedean screw rotating in the vacuum pan. With this procedure, crystallisation takes place in 48 hours and, after cooling to 40, the crystalline mass is disintegrated and centrifuged. Loblich, Zschene, Stenzel, and others have tried mixing the molases with fresh juice and defecating the mixture in the ordinary way, but this process does not seem to offer any great advantage. 1 The rendement expresses the, percentage of refined sugar obtainable from the raw sugar and is determined indirectly on the assumption that every 1 part of ash diminishes the refined sugar by 5 parts ; thus a raw sugar containing 96 per cent, of pure sugar and 0-4 per cent, of ash would give a rendement of 96 (0-4 x 5) = 94 per cent. The rendement 1 is regarded as low if it is less than 94 per cent. s Revivification of Animal Charcoal. The charcoal is first treated with hydrochloric acid to remove the calcium carbonate, and if more than 1-5 per cent, of calcium sulphate then remains, this is eliminated by means of hot soda solution. After washing, the wet charcoal is allowed to ferment (first alcoholic fermentation sets in FIG. 340. SUGAR-REFINING 471 It is then allowed to flow into conical copper moulds with their apices, closed by plugs, underneath. The mass, which has just begun to crystallise, is well stirred, and when it has assumed a certain consistency it is left at rest at a temperature of 35, so that all the molasses collects at the bottom and can be discharged by removing the plug. In order to remove the molasses completely, the sugar-loaves with their casings are introduced into the moulds of a Fesca centrifuge (Fig. 341), which holds sixteen of them, arranged alter- nately in two superposed series of eight. The point of the sugar-cone communicates with the aperture, b', of the drum of the centrifuge, and when the latter is charged it is fitted in the middle with a cylinder, hh' k, which rotates with the drum and is provided with channels, 8, communicating with all the cones, so that the covering solutions (see above) may be run in from the tank, r. These solutions consist of three or four pale syrups and three or four concen- trated solutions of pure sugar. In order to remove the last traces of yellow colour from the sugar and to blue it slightly, as is sometimes required, the final covering syrup is mixed with a minimal amount of ultra- marine (5 grins, per 100 quintals of sugar) or methyl or ethyl violet or, better still, according to a recent suggestion, indan- FIG. 341. threne. The white loaves thus obtained are then dried in suitable chambers or in revolving apparatus, at a temperature of 55. To obtain white sugar directly, the final massecuite is sometimes decolorised with 30 to 50 grins, of blankite per hectolitre (see Note, p. 444 ; blankite is pure, crystallised sodium hydrosulphite, the use of which is rapidly extending in sugar-works ; see vol. i, p. 465). The beet-sugar of commerce should always have a very faint alkaline reaction (towards phenolphthalein), since otherwise it undergoes partial inversion. Cane-sugar, however, has usually a slight acid reaction. Cube sugar was formerly obtained by sawing the large blocks, this entailing considerable loss. But at the present time suitable centrifuges (Adant type, Figs. 342 and 343) yield directly long rods of sugar of the requisite thickness, these being then sawn with a minimum of loss. A platform, F, carries eight vertical prisms, o, furnished with screws by which they are fixed to an upper annular disc. The latttr is slotted (c) to allow of massecuite being intro- duced into the chambers (a a) remaining between each prism and the next, and divided into a number FIG. 342. of tall narrow chambers by plates fixed in the grooves, 6. The platform is introduced into the cylinder, H, which fits tightly the periphery of the moulds, these being closed inside by a second cylinder. All the then acid fermentation and finally putrefaction), and is afterwards washed thoroughly with water, treated with steam, dried and gently ignited in long cast-iron tubes, C (Fig. 344), which are heated to about 400 by the gases from the furnace, A, access of air to the retorts being excluded. The cooled, free portions are then gradually discharged from the lower parts of the retorts (E) into covered metal waggons, so that the charcoal, which is still not quite cold, may not take fire in the air. The discharge of the putrid washing water from the fermented char- coal into rivers causes serious inconvenience, and nowadays this water is either passed on to the soil or subjected to biological purification (see vol. i, p. 222). The plant for decolorising with animal charcoal and the revivifying furnaces are very costly, a large amount of the charcoal at 20s. to 24s. per quintal being required. In 1908, Germany imported 51,666 quintals of animal charcoal and exported 35,019, while in 1900 the imports and exports were 39,839 and 32,018 quintals respectively. Italy imported 4756 quintals in 1908 ; 6789 in 1909 ; and 9863, costing 15,780, in 1910. Soxhlet avoids the carbon decolorising plant by using filter-presses the chambers of which are filled with a cake composed of wood-meal mixed with various indifferent materials (ground coke or pumice, &c.). By this means sugar solutions can be decolorised moderately well even in the cold. 472 ORGANIC CHEMISTRY chambers are filled with massecuite introduced through the slots, c, the whole being allowed to cool for 12 to 14 hours with occasional shaking. After complete crystallisa- tion, the whole platform is withdrawn by the crane/Cr, and placed in the centrifuge, D, FIG. 343. which makes about 700 revolutions per minute. The covering is effected at a reduced velocity with sugar solutions entering by the tube, C, from a reservoir at a height of 5 metres. After the sticks of sugar have been removed, the platform and moulds are washed with water and are then ready to receive a fresh quantity of massecuite. FIG. 344. N Pile or crushed sugar is obtained in a more simple manner by covering the crystalline sugar (from massecuite) in the centrifuge itself by means of water, steam, or pure Migar solution. Slight prolongation of the centrifugation yields a hard, compact mass, which is removed in large blocks and broken into small irregular pieces (pile sugar) by a special crusher having an indented drum (Fig. 345). UTILISATION OF MOLASSES 473 FIG. 345. Powdered sugar or farin is obtained by grinding lump sugar and any scraps between two smooth, horizontal rollers (d and d', Fig. 346) which are brought near to one another by springs and are furnished with scrapers, /, to detach the powdered sugar ; the latter is subsequently sieved. Powdered sugar can also be obtained by means of the Excelsior mill (see Fig. 162, p. 168), which yields as much'as 2000 kilos per hour of a sugar not too finely powdered. UTILISATION OF MOLASSES. The processes employed for the extraction of beet- sugar yield about 3 per cent, (of the weight of beets) of molasses, i.e. of dense, dark-coloured syrups, containing 40 to 50 per cent, of sugar. This does not crystallise owing to the presence in the molasses of 8 to 10 per cent, of mineral salts, which prevent about five times their weight of sugar from crystallising. So that, in general, it is difficult or almost impossible to extract sugar by direct crystallisation from syrups with a degree of purity less than 60 to 65 per cent. The percentage composition of molasses varies between the following limits : water, 19 to 28 (mean, 23) ; sugar, 45 to 54 (mean, 48) ; solids not sugar, 26 to 29 (mean, 28) ; ash, 6 to 8 (mean, 7 ; largely potassium salts) ; invert sugar, 1-25 to 1-85 (mean, 1-65). The degree of purity ranges from 62 to 67 per cent, (mean, 64 per cent.). The com- positions of various Italian molasses have been given in the Note on p. 140. The recovery of the sugar from molasses involves indirect processes which are not always convenient in practice, and when this is the case the molasees is employed for the manufacture of cattle-food or spirit (see p. 140). In spirit factories the molasses is diluted to 12 to 14 Be. (about 15 per cent, of sugar), when it can be fermented (see p. 140). 100 quintals of molasses yield 23 to 25 hectols. of alcohol (calculated as anhydrous spirit) and 1800 kilos of C0 2 . The potassium salts are extracted from the residual vinasse by the process described in vol. i, p. 435. 100 kilos of molasses give 35 kilos of concentrated vinasse (40 Be.), and by calcining this 10 kilos of vinasse charcoal are obtained. In some factories the vinasse is now treated for the recovery of the ammonia and fatty acids by the Effront process described on p. 155, without, however, losing the potas- sium salts. 1 In Italy, before the modification of the fiscal regulations which taxed the defecated saccharine 1 The molasses vinasse remaining after the distillation of the alcohol has a density of about 4 B6. and contains 6 to 7 per cent, of solids. When utilised, it is first concentrated to 40 Be 1 . (100 kilos of molasses give 35 kilos of this concentrated vinasse), when it contains 75 per cent, of solids with about 4 per cent, of nitrogen. About one-half of the solid substances are nitrogenous compounds. The solids contain 10 to 12 per cent, of betaine, 5 to 7 per cent, of glutamic acid, and 1 to 2 per cent, of leucine and isoleucine, besides varying quantities of amino- acids and nuclein bases ; the non-nitrogenous constituents consist of about 15 per cent, of fatty acids (formic, acetic, lactic, butyric, and homologous acids), and 15 to 20 per cent, of other organic compounds not com- pletely investigated. Effront thinks it possible, from 100 quintals of molasses, to obtain 75 kilos of ammonium sulphate and 95 to 120 kilos of fatty acidS, by the action of yeasts which decompose the amino-acids into ammonia and fatty acids, separable by distillation. But, according to P. Ehrlich, yeasts transform amino-acids into alcohol and,succinic acid, the formation of ammonia and fatty acids being clue not to yeasts but to butyric and other bacteria which always occur with yeasts, and decompose the amino-acids into ammonia, fatty acids, and various amines just as in ordinary putrefaction. Hence the effect of the Effront process could also be obtained by adding to the aqueous vinasse a little putrefied meat and allowing putrefaction to proceed. The manipulation of large masses of putrefied liquid would not, however, be very agreeable or hygienic. FIG. 346. 474 ORGANIC CHEMISTRY juices directly and left untaxed the sugar in the molasses, various factories applied certain of the chemical and physical methods used in other countries for the extraction of the sugar from molasses by means of osmosis, lime, strontia, baryta (formerly by means of alcohol), &c. When these methods (see later) are used, it is calculated that the final molasses does not exceed 0-5 to 1 per cent, of the weight of the original beets. In Italy the amount of molasses produced annually, including that from refineries, is 300,000 to 350,000 quintals, which is utilised almost entirely in spirit factories, 60 kilos of anhydrous alcohol being obtained from 100 kilos of sugar. Germany produces about 4,000,000 quintals, 2,200,000 being desaccharified by means of strontium, 1,250,000 uped as fodder, and 350,000 utilised by spirit factories. In 1908 England imported 84,128 tons for making spirit and cattle-food. In France, 346,000 quintals of molasses were returned to the agriculturist in 1907. (1) Osmosis Process. This was first proposed by Dubrunfaut in 1863, and is based on the osmotic properties of crystalloids, which pass through a membrane immersed in water (see vol. i, p. 102). But different crystalloids traverse the membrane at varying speeds, the sugar, for instance, far more slowly than salts. Hence, if the molasses is placed in a dialyser and surrounded with water, after a time the water will contain more salts than FIG. 347. sugar, while the molasses will be diluted with water but will contain relatively more sugar and less salts than at first. The apparatus now used for osmosis (Fig. 347) consists of a series of wooden frames 4 cm. in thickness and of the size of those used in filter-presses ; these are separated by sheets of parchment paper, the whole being pressed tightly together. The compartments thus formed are filled alternately with water and molasses. The upper part of the whole of the osmogen constitutes an open reservoir formed by the upper vertical projections of the frames. The molasses for feeding the alternate chambers is placed in this reservoir and is kept circulating in various ways. The water chambers are fed from the lower part and are discharged through a common upper tube as they become enriched with salts. The osmotic effects occur best in the hot, so that the molasses is introduced at 80 and the water at 90. The taps through which the liquids enter and leave the osmogen are regulated by automatic floats, which close or open the taps more or less so as to maintain a constant relation between the density of the exosmosed aqueous solution and that of the osmosed molasses. This relation is determined beforehand in the laboratory, and corresponds with the conditions least favourable to the loss of sugar with the osmosis water and most favour- able to the purity of the residual molasses. The exosmosed water generally has a density of 3 Brix (3 per cent, of sugar and salt together), and the osmosed molasses 35 to 40 Brix (measured at 75 C.) ; the latter is concentrated and boiled in ordinary syrup pans until it shows the string test. Crystallisa- tion is carried out in the molasses room at 40 to 45 or in the Grosse apparatus. The crystallised sugar is separated by centrifugation and the new molasses obtained again subjected to osmosis. This operation is repeated once or twice more in fact, until the SUGAR FROM MOLASSES 475 quantity of sugar extracted would be insufficient to pay the cost. In some cases the osmosis waters are concentrated and reosmosed. The final molasses and the final osmosis waters rich in salts and also in sugar serve for making spirit, shoe -polish, or potassium salts (see p. 155). They are also given to cattle, but must then be diluted with solid vegetable products as an excess of salts may exert harmful effects. (2) Lime Process. Steffen found that the addition of finely powdered, sieved quicklime in small portions to a solution of molasses of a suitable concentration (about 12 Brix, i.e. 1 per cent, of sugar, obtained from 1 quintal of molasses + 7 hectols. of water), and kept at a temperature below 15, results in the separation of insohible sucrate containing rather more lime than tricalcium sucrate, whilst the impurities remain dissolved in the aqueous molasses. FiG. 348. FIG. 349. The operation is carried out in a vessel (Figs. 348, 349) similar to the Grosse apparatus, the steam-pipes being used, however, for the circulation of cold water at about 12, so that after each addition of lime, when the temperature rises 7 to 8, it can be brought rapidly down below 15. The addition of lime is continued until all the sugar is precipi- tated (about 100 kilos of lime per 100 kilos of sugar), this being ascertained by reading the clear liquid in the saccharometer. The resultant sludgy mass is filter-pressed at a pressure not exceedingly atmosphere, the filtrate still containing about 0-5 per cent, of sugar, which can be separated as tricalcium sucrate by heating the liquid to 90 and filtering. The cakes of sucrate are washed several times in the filter-press and the fairly pure residue used to defecate fresh diffusion juice before saturation with carbon dioxide ; or the sucrate can be treated with any cold saccharine solution so as to form the soluble monosucrate, the precipitated excess of lime being removed by filtration_and the filtrate then saturated with carbon dioxide in the ordinary manner. (3) Strontia Process. When an excess of crystallised strontium hydroxide is added to a dilute sugar solution at a temperature of about 100 and the liquid boiled, a granular, 476 ORGANIC CHEMISTRY sandy precipitate of strontium disucrate is obtained, which is stable in the hot whilst in the cold it decomposes into sugar and strontium hydroxide. In a suitable boiler provided with steam-coils and stirrers, a 10 per cent, solution of strontium hydroxide is boiled, further quantities of the hydroxide being added until a 20 to 25 per cent, solution is obtained. The molasses is now added in amount equal to about one-third of that of the strontium solution, which is stirred rapidly and heated meanwhile. Strontium hydroxide is subsequently introduced in such amount that the mass has 12 to 13 per cent, of excess alkalinity. The total strontium hydroxide is related to the sugar in the molasses in about the proportion 2-5 : 1. The precipitated disucrate is filtered rapidly in the hot through bag-filters and washed with boiling 10 per cent, strontium hydroxide, the latter being jrecovered from the filtrate. The disucrate is then dissolved in a cold strontium hydroxide solution and the solution introduced into metallic vessels situate in an apartment kept below 10. In the course of three days one-half of the hydrate separates in a crystalline form, the saccharine solu- tion being then decanted and the residue centrifuged. The sugar solution is then saturated with carbon dioxide until it shows an alkalinity of 0-05, all the strontium being thus separated as carbonate. The very pure sugar solution obtained after filtration is concen- trated and boiled as usual, the crystallised sugar obtained being placed directly on the market without being refined. A somewhat different mode of procedure is that based on the formation of strontium monosucrate, but this does not yield the whole of the sugar as the above process does. In Germany the desaccharification of molasses is effected almost exclusively with strontia in large works specialising in such work. (4) Baryta Process. When solutions of molasses and of barium hydroxide are mixed in the hot in the proportion of 1 mol. of sugar to 1 mol. of the hydroxide, a heavy, sandy precipitate of barium monosucrate is formed which is stable to either hot or cold water ; this is collected as usual on filters and freed from impurities by washing with cold water. It is then saturated with carbon dioxide in order to liberate the sugar and, after dilution with other sugar juices, is filtered, concentrated, and crystallised. 1 YIELD AND COST OF PRODUCTION. Formerly a hectare of land yielded with difficulty 200 quintals of beet, but as the result of long-continued improvement of the methods of cultivation, manuring, selection of seed, &c., as much as 300 to 400 quintals are now obtained, and in certain special regions (e.g. Ferrarese) as much as 600 to 650. Italy contains 20,000,000 hectares of cultivated land (excluding forests), 5,000,000 being under corn, 1,600,000 under maize, and only 50,000 under beet, the yields being as follows : Hectares under beet Mean production per hectare Mean price per quintal Mean quantity of sugar per 100 kilos of beet 1905-1906 37,500 Quintals 253 Shillings 2-01 Kilos 12-24 1906-1907 37,954 271 2-02 13-56 1907-1908 41,000 307 2-15 13-98 1908-1909 51,000 280 13 1909-1910 36,000 1910-1911 50,000 For every quintal of beet worked, the loss is calculated to be 1-6 kilo of sugar in Italy 1 The barium carbonate filtered off is converted into the oxide and then into the hydroxide by heating in suitable high-temperature furnaces (see vol. i, p. 502). This barium process was used for some time in Italy, after it had been shown that no danger to health was to be feared from the use of a barium compound, since this is eliminated completely by carbon dioxide and the final traces by calcium sulphate. The barium hydroxide required is imported principally from America and Germany ; but by 1903, four factories had been erected in Italy for supplying all the baryta necessary to the sugar factories. One of these factories, at Calolzio, starts from barium sulphate ; another, at Milan, heats the barium carbonate from the sugar-works ; while the remaining two, at Foligno and Pont St. Martin respectively, treat barium carbonate in electric furnaces, making first barium carbide, which with water gives acetylene and barium hydroxide (Garelli's process). Such treatment of molasses in Italy was found feasible as long as the sugar extracted in this way remained free from taxation, that is, while the tax was levied solely on the defecated diffusion juice. But since 1904, the total quantity of sugar produced, including that extracted from molasses, has been liable to duty, and the molasses is consequently utilised in the distillery and in the manufacture of cattle-food. But recently some sugar factories have resorted to treatment of the molasses with barium sulphide, which is much cheaper than the hydroxide and is obtained directly from the sulphate in the electric furnace. SUGAR STATISTICS 477 and only 1 kilo in Germany. The cost of cultivating 1 hectare of beet, including manure, transport, &c., amounts to 12. Italian manufacturers calculate that in bad seasons the production of 100 kilos of refined sugar requires 10 quintals of beet, the cost of working these being Is. to Ss. (including 3s. for coal). Refining costs about 5s. 6d. (100 kilos of raw sugar give about 90 of refined). In Germany 100 kilos of beet gave not more than 8-4 of sugar in 1870, about 12-5 in 1890, and 15-8 (including that from the molasses) in 1909-1910. The mean production per hectare was 246 quintals of beet in 1871 and 300 in 1910. The consumption of coal in working 100 kilos of beet in Germany was 35 kilos in 1867, 24 kilos in 1877, 10 kilos in 1890, and 7 kilos (8 in Italy) in 1900. By the use of Kestner concentrators (see above) a further saving in coal has recently been effected. Every 100 kilos of beet treated yield 3 kilos of molasses (including 0-5 kilo from the refining process) containing 45 to 50 per cent, of sugar. Each quintal of beet gives 80 kilos of exhausted and pressed pulp or slices containing 70 per cent, of water. The cost of manufacturing 100 kilos of cane-sugar in Java varies from 12s. to 16s., and transport to England or the United States amounts to 2s. per quintal. STATISTICS. 1 The history of the development of the sugar industry in Europe and the importance this industry has assumed during the past quarter of a century have already been discussed on p. 448. Reference has also been made to the production of cane-sugar compared with that of beet-sugar. While in 1854 beet-sugar formed only 14 per cent, of the world's total production (1,423,000 tons), in 1866 the proportion was 30 per cent, (on 1 The Commercial, Customs, and Fiscal Conditions of the sugar industry in Italy and other countries. In some countries this great industry has been extended artificially owing to the direct and indirect help afforded by the State, and to the speculations of financiers. \Vith the excuse of protecting national industries, Govern- ments have levied heavy Customs duties, with the result that the public has paid dearly for its sugar, while manu- facturers have accumulated enormous profits and have been enabled to export sugar at less than cost price to other countries. At first the protective duty was from 24s. to 32s. per quintal, while in France it was raised to 64s. The form taken by the protection was then changed by the institution of export bounties, which allowed the sugar to be sold abroad at a low price, while large profits were made owing to the high prices at home and to the bounties. First Belgium and then France established a bounty of 8s. to 10s. for every quintal of sugar exported, France being thus subjected to an enormous burden amounting to over 2,000,000, without counting the rebate on the freight from the factory to the frontier. This enormous sum has been paid by the mass of the population, to the exclusive advantage of a few manufacturers (rule of the Meline Ministry). In Germany and Austria, where the export bounties were relatively low, the manufacturers formed sale syndicates (cartels), which operated in the following manner : the manufacturers pledged themselves to supply all the raw sugar to the refiners, who granted a bounty of 24s. per quintal to the manufacturer and sold the sugar to the home consumer at a very high price, there being no fear of competition, as they enjoyed a monopoly. The sufferers, as always, were the consumers. The home profits were so enormous that sugar could be sold abroad at less than cost price and competition thus vanquished. On the other hand, England, the greatest consumer of sugar, found its markets deluged with cheap Continental sugar, which competed seriously with that from its Colonies, which had also become considerable exporters. Under these conditions a more rational solution was found for the problem of sugar with reference to inter- national commerce. The initiation of such an undertaking could come only from England, who was able finally to impose her conditions on all countries sending sugar to her markets. The Brussels Convention, convoked on September 1, 1902, was subscribed to by England, Germany, Austria, France, Belgium, Holland, and Italy. The result was the abolition of export premiums and the reduction of the boundary duty to 5s. per quintal above the manufacturing tax, from September 1, 1903, onwards. Such duty was to be enjoyed only by those countries conforming to the Brussels Convention. Italy did thus conform in a modified way : the boundary duty remained as before, namely, 23s. for first quality and 16s. 6d. for second quality, while a pledge was given not to export sugar to other countries and to impose an exceptionally heavy Customs duty on countries not adhering to the Brussels Convention (especially on Russia and the Argentine Republic ; but Russia entered the Convention in January 1908, and pledged herself to export for six years not more than 200,000 tons per annum of bounty-fed sugar. After 1908 England held herself free to import premiumed sugar without imposing supertaxation). Spain and Sweden were treated like Italy by the Brussels Convention, to which then Luxemburg, Peru, and Switzerland conformed. In Spain there is now an overproduction crisis. This is the regime now in force in Europe. But in Italy the price of sugar fell, owing to overproduction and frenzied competition, to 92s. per quintal, so that in 1901-1903 almost all the sugar factories showed either minimal profits or considerable losses. Indeed, deducting the tax of 56s., there remains 36s. as the price of the sugar. And, according to the manufacturers, 10 quintals of beet, giving 1 of sugar, cost 16s., while the cost of production of crude sugar is 8s. (including 4s. for coal), that of refining about 6s. 4d. and that of transport Is. 8d : total, 32s. Thus only 4s. remains to provide interest on capital as well as depreciation. Hence, in 1904, all the sugar- makers combined to fpym a syndicate and raise prices, and early in 1905 an increase of 16s. (to 108s.) per quintal was enforced ; with a production of 1,000,000 quintals, this amounted to an annual burden on the consumer of 800,000. Adding to this the protective duty of 1,200,000, it will be seen that, for the luxury of a native sugar industry, the Italians pay an annual tax of 1,200,000 to 2,000,000, the sole gainers being some 30 factories with a capital of about 3,200,000 ; this in spite of the fact that Germany and Austria would supply sugar at 24s. to 26s. per quintal, so that, leaving aside the taxation of 2,800,000, sugar could be sold at IQd. instead of 15d. per kilo. The sugar manufacturers state that, owing to various causes, their capital yields on an average only 6 per cent. But it can only be regarded as a mistake to keep an industry alive under such artificial conditions, when the consumer evidently suffers considerable injury and the advantage to the agriculturist and the operative is doubtful and in any case mipinia.!. Attempts recently made to remedy this state of affairs have been unsuccessful. ORGANIC CHEMISTRY a total of 2,000,000 tons) ; in 1878, 44 per cent, (on 3,000,000 tons) ; in 1887, 47 per cent, (on more than 5,000,000 tons) ; in 1893, 55 per cent, (on about 6,000,000 tons) ; in 1899, 64*per cent, (on 7,500,000 tons) ; in 1901, 67 per cent, on almost 9,000,000 tons. In 1C09- 1910 cane-sugar again assumed first place, constituting 53-5 per cent, of the total world's production of nearly 15,000,000 tons. In Europe the total area under beet is about 2,000,000 "hectares, about 43,000 hectares being in Italy. Country and year No. of factories Production of beets (b) or sugar-cane (c) Output of raw sugar Remarks Europe Tons Tons Germany . . 1903 384 12,171,000 b 2,293,000 In 1907, 417 factories and refineries worked 14,187,000 tons of beet, obtaining 1,950,000 tons of raw sugar. In 1905-1906 the pro- duction was 2,400,000 tons, and in 1909-1910, 2,037,400 tons Austria-Hungary 1903 215 7,542.600 ft 1,290,000 355,000 tons in 1886 ; 665,000 . 8,507,000 in 1908 in 1894 ; 865,000 in 1900 ; 1,334,000 in 1906; 1,259,000 tons in 1909-1910 France . . 1903 296 6,315,300 6 1,080,000 Equal to 694,000 tons of refined ; 807,500 in 1909 in 1904, 540,000 tons Russia . . 1903 275 7,604,000 6 1,000,000 425,000 in 1887 ; 578, JOO in 1894 ; 8,800,000 in 1908 1,300,000 in 1906; 1,144,000 tons in 1909-1910 Belgium . . 1903 100 1,645,000 b 325,000 241,000 in 1909-1910 Holland . . 1903 29 1,023,000 b 204,000 175,000 in 1909-1910 Italy . . 1909 31 1,050,000 b 165,000 Or 150,000 refined Spain . .1904 32 580,000 6 64,300 69,000 in 1905; 103,340 (in 49 factories) in 1903 ; 86,000 in 1910 . 1904 27 250,000 c 24,500 After 1906 there was overproduc- tion. Cultivated between Gib- raltar and Almeria : 28,820 tons in 1905 ; 140,600 in 1908 ; 22,000 in 1910 Denmark . . 1903 7 415,000 b 65,000 58,500 in 1909 Roumania . 1904 5 182,700 b 23,500 Sweden . . 1909 21 864,400 b 122,0(10 Only 4000 in 1886 America United States . 1903 70 1,500,000 6 300,000 12,000 in 1893 ; 327,000 in 1904 ; 433,000 in 1907 ; 450,000 in 1909-1910 > c 170,000 ,, ,, maple 12,000 5400 in 1880 ; 10,000 in 1907 Cuba . . 1904 c 1,050,600 1,140,000 (home consumption, 26,000) in 1905-1906 ; 1,450,000 in 1907 ; 970,000 in 1907-1908 ; 1,520,000 in 1908 - 1909 ; 1,459,000 in 1910-1911 Trinidad . . 1906 c 54,000 45,600 in 1907 ; 41,600 in 1908 Other Antilles, Central America . 1906 c 410,000 South America (Dem- Argentine in 1909 produced 115,000 erara, Peru, Argen- and consumed 150,000 tons tine, Brazil) . 1902 c 440,000 Mexico . . 1909 c 160,000 143,000 (and 70,000 molasses) in 1908 ; exports 400,000 to 600,000 Asia Java . .1909 c 1,312,466 1,285,000 in 1907 Philippines . 1909 c 94,000 exported Production, 122,000 tons in 1907 and 138,000 in 1908 East Indies . 1904 c 1,300,000 Australia . 1902 470,000 Africa . 1902 (Egypt, Reunion, and .Mauritius) c 280,000 f 45.000(1904) Formosa (for Japan) 17 c < 71,000(1906) 1910 1206,000(1910) SUGAR STATISTICS 479 The world's production of sugar and that of the various countries is shown in the Table on the opposite page, which also gives the number of sugar factories.^ The total world's production of sugar is given by about 1400 factories and 300 refineries, and in 1904-1905 amounted to 11,684,000 tons of raw sugar, in 1905-1906 to 13,762,000 tons, in 1906-1907 to 14,420,000, in 1907-1908 to about 13,500,000, and in 1909-1910 to 16,200,000 tons, nine million tons of this being cane-sugar. In Japan a single refinery at Moji produces 2500 quintals of refined cane-sugar daily, and it is proposed to double the output. Two other refineries are found at Osaka and Tokyo respectively. These work mostly raw sugar from Java and export a considerable quantity of refined sugar to China and Corea. Formosa produced in 1907 about 70,000, in 1908 about 100,000, and in 1910 more than 200,000 tons of cane-sugar. In 1910 the areas under beet in the various countries of Europe were as follow (hectares) : Eussia, 675,000 ; Germany, 470,000 ; Austria-Hungary, 365,000 ; France, 235,000 ; Belgium, 67,500 ; Holland, 55,000 ; Italy, 32,000 in 1903, 38,000 in 1906, 51,000 in 1908, 36,000 in 1909, 50,000 in 1910 ; Sweden, 35,000 ; Denmark, 22,500 ; Spain, 18,000 ; Roumania, 13,000 ; Servia, 3300 ; Bulgaria, 1700 ; Switzerland, 950. To give an idea of the progress made by the beet-sugar industry during the last 50 years, the production of raw sugar in the two countries where this industry has developed most is given in the following table : In France In Germany Germany Yield of sugar per 100 kilos beet Annual consump- tion per head Tons Tons 1840 . 22,784 14,200 5-9 kilos 2-5 kilos 1850 . 62,165 53,300 7-3 3-1 1860 . 126,480 126,520 8-6 4-3 1870 . . . 282,136 186,000 8-6 4-7 1890 . 750,000 1,336,000 12-5 8-5 1903 . 1,080,000 1,921,000 14-4 13 1905 . 1,605,000 14-9 14-9 1906 . 730,000 2,400,000 14-7 17-0 1909 . ' . 807,500 2,037,400 16-3 19-5 The production in France varies, since the agriculturists require as much as 3*. per 100 kilos of beet. While the consumption was 40,000 tons in 1887 and 527,000 in 1902, it rose to 600,000 tons in 1908, owing to the modification of the fiscal conditions of 1903- 1904. The number of workpeople occupied for about two months was 38,000 in 1908, with an average wage of 3s. per day. The area under beet in France in 1907 was 210,000 hectares. Some of the large factories in France and Belgium have diffusion plants in the middle of the beet-growing districts, the sugar juices after treatment with lime being forced through pipes, often several kilometres long, to the factories, where they are further worked up. In 1906 England imported 1,583,000 tons of sugar, and in 1909 about 940,000 tons of refined and 815,000 tons of raw sugar. In 1910 the imports were 98,000 tons of raw sugar and 84,400 tons of refined sugar, the total value being 24,554,000 ; the exports were 31,000 tons. The United States imported 2,095,000 tons of raw sugar and 76,000 tons of refined sugar in 1910, and 2,049,000 tons (19,873,600) of raw and 160,000 tons of refined sugar in 1911. In Germany the beet-sugar industry has reached its greatest perfection and magnitude, and from 1880 to 1902 Germany was the largest exporter (as much as two-thirds of its own output). In 1909-1910, in spite of the diminution of exports resulting from the Brussels Convention, 1 Germany exported 423,000 tons of refined sugar and 310,000 tons of the 1 The Fiscal System in Germany from 1841 to 1866 was based on the quantity of beets, the object being to bring about improvements in the cultivation of the beet and hence increase in the sugar-content ; the tax corresponded with about 18s. per quintal, and was refunded to the manufacturer for all exported sugar. From 1870 to 1886 the tax was 1. "id. per quintal of beet, it being assumed that 12-5 kilos of beet were required to give 480 ORGANIC CHEMISTRY raw product, the home consumption 'being 1,260,000 tons. The exports were 740,000 tons in 1890, 883,000 in 1904, and 1,145,000 in 1906. In 1908-1909, 358 factories and 39 re- fineries were working in Germany. Certain German factories, employing 46 workmen, treat 4000 to 5000 quintals of beet, but in Italy many more employees are required. In 1909- 1910 Germany produced 10,600,000 tons of beet, but in 1910-1911 only 5,200,000 tons. In Austria large batteries of diffusors are used and a more complete exhaustion is obtained even at a lower temperature ; in general, indeed, the modern plants are more perfect than those in Germany. In 1908 Austria -Hungary exported 610,000 tons of refined and 195,000 of raw sugar. The following Table shows, for different countries : I, manufacturing tax in pence per kilo ; II, retail price in pence per kilo ; III, mean annual consumption in kilos per head in 1899 and 1909 ; IV, mean quintals of beet produced per hectare in 1908-1910 ; V, kilos of refined sugar obtained from 100 kilos of beet ; and VI, kilos of refined sugar from 1 hectare. I II III IV V VI 1899 1909 England . . 0-96 5-3 40 41-1 United States . 0-96 4-8 28-4 37-2 220 12-44 2706 Switzerland 0-67 4-8 25-7 30-2 Denmark 0-575 6-7 21-6 35-5 287 13-82 3950 Sweden and Norway . 2-88 7-7 /24-5\ 15 ' 7 J17-8J 266 14-26 3803 Germany . 1-92 6-2 13-7 19-7 284 16-35 4809 Holland 5-47 9-6 13 19-8 257 14-80 3803 France . . 2-6 7-2 12-8 16-9 265 13-03 3445 Belgium 1-92 6-7 10-5 15-1 281 14-37 4032 Austria-Hungary . 3-45 8-15 8-3 11-2 249 15-74 3909 Russia . . . ' 2-7 8-25 6 9-1 136 16-37 2230 Spain . . .-', 0-77 8-15 . 4-5 5-4 289 11-88 3439 Portugal . 6 6-2 Greece . . . 2-4 8-15 3 3-8 Roumania . . . r- 3-5 4-1 165 14-53 2392 Turkey . . . 5-47 9-6 3-5 5-7 Italy . . ,. - 6-7 14-4 2-8 3-9 299 11-27 3378 Servia f.. . ... . 3-17 7-7 3 3-5 The influence of the price of sugar on the consumption is shown not only by the above Table but also by the following significant facts : when the manufacturing tax was reduced by 40 per cent, in France in 1903-1904, the consumption increased by 61 per cent. ; in Germany a 33 per cent, reduction in the taxation produced an increase of about 60 per cent, in the consumption, and in Belgium 55 per cent, more sugar was consumed as a result of the lowering of the tax by 29 per cent. In Italy the sugar industry has developed only within the last fifteen years (see p. 477), as is shown by the Table on p. 481 (in 1906 the 39,500 hectares under beet gave a mean yield of 270 quintals of beet per hectare, the limits being 328 and 177). 1 kilo of sugar ; but even in 1870 1 kilo of sugar could be obtained from 11-9 kilos of beet, and in 1887 from 8-1 kilos. But since the exports increased enormously and the taxes refunded remained the same, the manu- facturers enjoyed indirectly a considerable export bounty, which diminished the Exchequer receipts from 3,000,000 to less than 760,000 (1888). A modification was hence made in the system of taxation, sugar produced and consumed at home paying a tax of 1 per quintal, while that exported was freed from tax and received a bounty of 2s. Gd. (raw) or 3s. 6<2. (refined) per 100 kilos (1896-1903). Further, the import duty was left at 2 per quintal, so that German producers were allowed to sell their sugar at high prices at home (even during the abundance of 1900-1901) and to employ part of their profits to lower the price of sugar sold abroad in competition with other countries. After the Brussels Convention, however, export bounties ceased and the import duty was reduced, to 5s. + 16s. (manufacturing tax in Germany). Under these new conditions, the exports diminished somewhat, but the home consumption increased owing to the lowered prices. The wholesale price in 1910 was 2 per quintal (that of sugar for export, without tax, being 19s.) ; the retail price was 14d. per kilo in 1875, Id. in 1902, and &d. in 1910. The German Government received 5.750,000 in sugar taxes in 1900-1901 and almost 8,000,000 in 1909-1910. ESTIMATION OF SUGAR 481 Year Output of raw sugar Imported raw sugar Total consump- tion of refined sugar Remarks Tons Tons Tons 1887 . ' . 184 140,000 125,500 100 tons of raw sugar taken as 1889 . 632 78,000 71,000 90 tons of refined 1894 . 2,090 75,000 70,000 1897 . 3,336 75,500 71,000 1901 . 73,800 37,100 99,880 1902 . 95,166 20,000 100,000 1903 . 128,000 5,200 120,000 Overproduction of 25,000 tons 1904 . ' , 79,000 2,100 73,000 1905 . 112,000 5,100 112,000 1906 . 136,000 1 5,000 (?) 138,245 Mean yield of refined sugar from the beets, 11-86 per cent., the mean content being 13-56 per cent. ; loss in working, 1-70 per cent. 1906-1907 . 106,400 23,738 (?) 1907-1908 . 136,000 4,903 145,000 The Italian Government received 60,000 in manufacturing tax and 2,680,000 in import duty in 1897, about 2,560,000 in tax and 320,000 in Customs duty in 1903, 3,920,000 in tax and 66,060 in Customs duty in 1909-1910. The production of beets in Italy was 1,256,660 tons in 1909 and 1,679,070 tons in 1910, the output of sugar being 161,600 tons in 1908-1909, 107,200 tons from 9,670,700 quintals of beet in 1909-1910, and 170,000 tons in 1910-1911. In 1910 Italy imported 5800 tons of first-grade and 655 tons of second-grade sugar. DETERMINATION OF SUGAR -CONTENT. Sugar is estimated in various ways. With an aqueous sugar solution, the content of saccharose can be determined by means of the specific gravity at 17-5, compared with water at 17-5, this being measured by hydrometers, pyknometers, &c. (see vol. i, p. 72). In the factory, use is generally made of a hydrometer (saccharometer), which, at 17-5, gives directly the percentage of saccharose present. These saccharometers were first proposed by Balling and were subsequently corrected by Brix, degrees Brix expressing the percentage of sugar. In France and Belgium, and sometimes also in Germany, saccharometers gauged at 15 and referred to water at 15 are used, and the Berlin Royal Commission for the control of standards prescribed the use of saccharometers giving the density of solutions at 20 referred to that of water at 4. The following Table gives the densities and degrees Brix (grammes of sugar per 100 grms. of solution) for the temperature 17-5, and also, for each 10, the values from the other two Tables, so that the intermediate values in these two Tables can be calculated roughly. The saccharometer is read with the precautions and in the manner indicated on p. 74 of vol. i and on p. 147 of this volume. The Table gives densities above 66 Brix, which cannot be determined by hydrometers, but which serve to calculate the degree of purity of impure saccharine solutions (molasses, &c. ; see later). It 482 MATEGCZEK AND SCHEIBLER'S TABLE, GIVING THE SPECIFIC GRAVITIES AND DEGREES BRIX OF SACCHARINE SOLUTIONS Sp. gr. . ]7-5 at 17^ OJ t "c g?H n Sp. gr. . 17-5 at i7-5- i* Ml ,v! P" Sp. gr. , 17-5 at lT.5 1 * 11 P Sp. gr. . 17-5 ^TTM l<-5 1 * &s P Sp. gr. 17-5 *wv 8 x I s 1-00388 i 1-08778 21 1-18460 41 1-29531 61 1-42258 81 1-00779 2 1-09257 22 1-18981 42 1-30177 62 1-42934 82 1-01173 3 1-09686 23 1-19505 43 1-30777 63 1-43614 83 1-01570 4 1-10145 24 1-20033 44 1-31381 64 1-44298 84 1-01970 5 1-10607 25 1-20565 45 1-31989 65 1-44986 85 1-02373 6 1-11072 26 1-21100 46 1-32601 66 1-45678 86 1-02779 7 1-11541 27 1-21639 47 1-33217 67 1-46374 87 1-03187 8 1-12013 28 1-22182 48 1-33836 68 1-47074 88 1-03599 9 1-12488 29 1-22728 49 1-34460 69 1-47778 89 1-04014 10 1-12967 30 1-23278 50 1-35088 70 1-48406 90 1-04027 r^ 1 Llo -I 10 1-12999 g] 30 rl5n 1-23330 - L15 J 50 [15T -s\ 15 J 70 1-48716 rjl Llo 1 90 [20i J 10 r20T 1-126984[ -J 30 l-229567[^-J 50 1-347174[?J 70 l-479976f^-l L 4 J 90 1-04431 11 1-13449 31 1-23832 51 1-35720 71 1-49199 91 1-04852 12 1-13934 32 1-24390 52 1-36355 72 1-49915 92 1-05276 13 1-14423 33 1-24951 53 1-36995 73 1-50635 93 1-05703 14 1-14915 34 1-25517 54 1-37639 74 1-51359 94 1-06133 15 1-15411 35 1-26086 55 1-38287 75 1-52087 95 1-06566 16 1-15917 36 1-26658 56 1-38939 76 1-52810 96 1-07002 17 1-16413 37 1-27235 57 1-39595 77 1-53550 97 1-07441 18 1-16920 38 1-27816 58 1-40254 78 1-54290 98 1-07884 19 1-17430 39 1-28400 59 1-40918 79 1-55040 99 1-08329 20 1-17943 40 1-28989 60 1-41586 80 1-55785 100 1-08354 [g] 20 rl5-i M7985 [^J 40 rl5i 1-29056 - Llo J 60 rl5*i 1-41628 [_] 80 1-56165 [If] 100 1-080959[ ---] 20 r20n M76447[ J 40 l-28S456r^-l L 4 - 60 r20*i 1-411715[ J 80 r20"~j 1551800[ J 100 If the degrees Brix are read with solutions at temperatures other than the normal, corrections must be made by means of the following Tables : STAMMER'S TABLE FOR REDUCING TO 17-5 DEGREES BRIX READ AT DIFFERENT TEMPERATURES DEGREES BRIX OF THE SOLUTIONS Temperature 5 10 15 20 25 30 35 40 50 60 70 75 13= } [These 0-18 0-19 0-21 0-22 0-24 0-26 0-27 0-28 0-29 0-33 0-35 0-39 corrections 15 to be subtracted "I 0-11 0-12 0-14 0-14 0-15 0-16 0-17 0-17 0-17 0-19 0-21 0-25 from the observed 1 ir J Brix degrees \ 0-02 0-03 0-03 0-03 0-04 0-04 0-04 0-04 0-04 0-05 0-05 0-06 18" \ 0-03 0-03 0-03 0-.03 0-03 0-03 0-03 0-03 0-03 0-03 0-03 0-02 19 These 0-08 0-08 0-09 0-09 0-10 0-10 0-10 0-10 0-10 0-10 0-08 0-06 corrections 21 to be added > 0-20 0-22 0-24 0-24 0-25 0-25 0-25 0-26 0-26 0-25 0-22 0-18 to the observed 23 Brix degrees 0-32 0-35 0-37 0-38 0-39 0-39 s O-39 0-40 0-42 0-39 0-36 0-33 25* 0-44 0-47 0-49 0-51 0-53 0-54 0-55 0-55 0-58 0-54 0-51 0-48 Example. If a sugar solution shows 40 Brix (i.e. 40 per cent, of sugar) at a tempera- ture of 23, 0-4 must be added to reduce the reading to the true Brix degrees at 17 ; so that 40 + 0-4 = 40-4 degrees Brix at 17. POLARIMETERS 483 SCHEIBLER'S TABLE SHOWING DEGREES BRIX AT 15 AND THE CORRESPONDING DEGREES AT OTHER TEMPERATURES (FROM 10 TO 25) Teinperaturc DEGREES BRIX OR PERCENTAGE OF SUGAR 10 5-15 10-19 15-22 20-24 25-27 30-29 35-30 40-31 50-33 60-35 70-36 75-36 12 . 5-10 10-12 15-14 20-15 25-17 30-18 35-18 40-19 50-20 60-21 70-21 75-21 14 5-04 10-04 15-05 20-05 25-06 30-06 35-06 40-07 50-07 60-07 70-07 75-07 15 5-00 10-00 15-00 20-00 25-00 30-00 35-00 40-00 50-00 60-00 70-00 75-00 ir 4-92 9-91 14-90 l?-89 24-88 29-87 34-37 39-87 49-86 59-86 69-86 74-86 19 . 4-83 9-80 14-78 19-77 24-75 29-74 34-73 39-73 49-72 59-71 69-71 74-71 21 . 4-72 9-69 14-66 19-64 24-62 29-60 34-59 39-59 49-57 59-57 69-57 74-57 23 4-61 9-57 14-53 19-50 24-48 29-46 34-45 39-44 49-42 59-42 69-42 74-42 25 . 4-49 9-44 14-40 19-36 24-34 29-32 34-30 39-29 49-27 59-27 69-27 74-28 Example. If a solution reads 19-36 Brix at a temperature of 25, this would corre- spond with 20 Brix at the normal temperature of 15. For intermediate values, either of temperature or of concentration, the corresponding results are easily obtained by inter- polation. Thus, 18 Brix at temperature 15 would give, at other temperatures, values higher than those corresponding with 15 Brix by three-fifths of the difference between the values in the 15 Brix and 20 Brix columns. So that a solution showing 18 Brix at the temperature 15 would show, at the temperature 17, 14-90 + (19-89 14-90) = 14-90 + 2-99 = 17-89 Brix. In the quantitative determination of sugar, use is commonly made of its action on polarised light (see pp. 26 and 330), this being measured in the polarimeter. The rotatory power of a sugar solution is proportional to the concentration and almost independent of the temperature. In these determinations it is necessary to use pure sugar solutions, decolorised by means of a little basic lead acetate, which precipitates the albuminoids, colouring-matters, and other impurities ; the filtered solution is examined in the polari- meter. If the saccharose is accompanied by another optically active sugar for instance, glucose (dextro-rotatory) allowance must be made for the rotation of the latter. In such a case the diminution in rotation produced by inversion of the saccharose with dilute acid would give the amount of this sugar. POLARIMETERS AND SACCHARIMETERS. 1 One of the best-known polarimeters is the Laurent shadow instrument (Fig. 351 ), which contains, in place of the compensator 1 It has been mentioned already (see p. 26) that crystals of Iceland spar and quartz have the property of decomposing a ray of light into two polarised rays, the ordinary and the extraordinary. If a prism of Iceland spar with length greater than the breadth, with its acute angle of 68, is cut diagonally and lengthwise so as to divide it into two rectangular triangular prisms (Fig. 350), and these be cemented together again with Canada balsam, the result is a Nicol prism. When a ray of light, Im, enters the nicol, of the two refracted rays (mo, mp), the ordinary one, nw, is totally reflected by the layer of Canada balsam and is thrown out of the crystal (or), whilst the extraordinary ray, mp, passes through the prism (pqs) and emerges polarised. This ray is able subsequently to traverse a second nicol only when the principal section of this analysing nicol is parallel to that of the first polarising nicol. If, on the other hand, the two principal sections are perpendicular, the ray undergoes total reflection and will not pass through the second nicol ; in intermediate positions, varying quantities of light are allowed to pass. If a layer of water is placed between the perpen- dicular nicols, still no light will pass through the analyser. But if a sugar solution is interposed, the light passes with a greater or less intensity through the analyser, which must be rotated through a certain angle (proportional to the quantity of sugar) to produce total disappearance of the light. In order to determine exactly when the luminous ray is extinguished (even in this case a kind of half-shadow is always observed), Solcil attempted to divide the luminous field into two halves with complementary colours. Indeed, if a ray of polarised light is passed through a quarts plate placed FIG. 350. 484 ORGANIC CHEMISTRY and double -polarisation quartz plate, a special semicircular quartz plate, D, half a wave- length in thickness and occupying one-half of the field. The polariser, B, is rotated by means of the rod, X, and the rotation which restores the two halves of the field to the same luminosity is indicated on a graduated circle, C, provided with a vernier, read by means of the lens, N, and illuminated by the mirror, M. The source of monochromatic light is a double Bunsen flame coloured with sodium chloride, the light being collected by the lens, B, and the observation made through the eye-piece, 0. The scale of the apparatus is regulated by the screw, Z, so that it reads zero when the two halves of the field are equally illuminated. If a tube containing a liquid, interposed between the two nicols, causes the right-hand half of the field to darken, the liquid is dextro-rotatory, while darkening of the left-hand half indicates a laevo -rotatory com- pound. From the rotation read on the scale, the specific rota- tion can be calculated by the formulae given on p. 27. The practical examination FIG. 351. of sugars is made with polari- meters furnished with special scales and known as saccharimeters; the Laurent polari- meter has a saccharimetric graduation as well as that showing circular degrees. In the French saccharimeters (Soleil and Laurent) the 100 division corresponds with a normal aqueous solution of pure saccharose (obtained by precipitation of a very concen- trated aqueous solution with alcohol and drying at 60 to 70) containing 16-350 grms. in 100 c.c. at 17-5, the reading being made in a tube 20 cm. long (the same reading is given by a quartz plate 1 mm. in thickness). In the German instruments (Ventzke-Scheibler, Schmidt and Haensch) the 100 reading is obtained with a length of 20 cm. of a saccharose solution of sp. gr. 1-1, which contains 26-048 grms. per 100 c.c. at a temperature of between the two nicols, one half of this plate being dextro- and the other Isevo-rotatory, and the junction of the two lying exactly on the axis of the light, the two halves of the field will appear illuminated with com- plementary colours. If the plate is 3-75 mm. in thickness and the analyser is rotated through 24-5, the two halves of the field are almost completely extinguished and assume a pale red coloration, similar in the two halves. But if a sugar solution is interposed, the two halves assume different colours, extinction being restored by rotation of the analysing nicol. Later Soleil suggested compensating the rotation of the sugar solution by introducing, 123 FIG. 352. to a greater or less extent, between the nicols, a conical quartz plate w compensator, moved by a rack indicating on a scale the thickness of the plate and hence the equivalent rotation. The more modern saccharimeters of the Soleil-Ventzke type have two compensators, each formed of two quartz wedges (MN and HK, Fig. 352) of opposite rotations, and are fitted also with the Lippich polariser formed of three nicols (P), which give a field divided into three zones ; when these zones are not equally illuminated, the two lateral ones show a colour different from that of the middle one. The analyser is enclosed in a metal box to protect it from dust. The two compensators with their scales are regulated by two screws, V and V. When the two scales indicate zero, the three zones should be equally illuminated. POLARIMETRY 485 17-5 . 1 So that a reading of one division corresponds with. 0-26048 grm. of saccharose per 100 c.c., or 1 grm. of sugar per 100 c.c. gives a reading of 3-839 divisions. The source of light for modern saccharimeters is an incandescent gas-burner enclosed in a blackened metal chimney fitted with a ground-glass window, or an incandescent electric lamp of at least 32 candle-power with a ground-glass globe and also enclosed in a black case. In order that the apparatus may not become heated, the lamp should be placed at a distance of about 15 cm., and to render the luminous fields more distinct the light is passed first through a glass cell with parallel walls filled with 6 per cent, bichromate solu- tion in a layer 15 mm. thick ; in this way the more refractive rays are absorbed and a uniform yellow light obtained. The normal tube of the saccharimeter contains a layer of liquid exactly 20 cm. long, but for very dilute and slightly rotating solutions tubes of 30, 40, and 50 cm. are used, whilst for solutions which are not quite colourless tubes of 10 or 5 cm. may be employed ; in all cases the readings are referred to the normal length of 20 cm. Some tubes are provided with an aperture for the introduction of a thermometer, so that the temperature of the solution may be read in the instrument. The saccharimeter scale extends from to 100 divisions on the positive side and to 30 on the negative side. The integral divisions are given by the zero of the vernier, N (Fig. 353), and the decimal parts by that division of the vernier scale which coincides exactly with a division on the scale ; in Fig. 353 the reading is + 2-6 divisions. 2 The specific rotatory power of saccharose varies little with the concentration (up to 30 per cent.) and with the temperature (between 15 and 25), but it is best to work near to 20, FIG. 353. when [ a ] 2 D = + 65-5 . 3 1 That is, in 100 Mohr c.c., 1 Mohr c.c. being the volume of 1 grm. of water at 17-5 weighed in air with brass weights. The true c.c. is the volume of 1 grm. of water at 4 weighed in vacua ; calculation on the basis of the coefficient of expansion of water shows that 100 Mohr c.c. are equal to 100-234 true c.c., so that 100 true c.c. of the normal saccharose solution at 17-5 would cojtain 25-987 grms. of saccharose. The International Commission for uniform methods of sugar analysis proposed in 1900 the fixing of the 100 point of the saccharimeter by a length of 20 cm. of a solution obtained by dissolving 26 grms. of pure saccharose in water to a volume of 100 true c.c. at 20 and polarising at 20 (100 true c.c. of water at 20 weigh 99-7174 grms. in air and 99-8294 grms. in vacua). 2 With double compensation saccharimeters (furnished with two scales, a working scale and a control, V and V, Fig. 352) the procedure is as follows : When the tube with the sugar solution is introduced between the nicols the control scale is placed at zero, the working scale being then moved by the screw until the field is uniformly illuminated and its position read. The sugar solution is next removed and the control scale moved until the field is again uniform, the reading of this scale being nearly equal to the first reading of the working scale. The tube of solution is now again introduced and the position of the working scale, near to the zero-point, read after its adjustment to give uniform luminosity. Finally the tube is again removed and the control scale moved until the field is uniform and its position read. The final result is obtained by subtracting the mean of the second pair of readings from the mean of the first pair. Thus, if the readings were + 78-6, + 78-4, + 0-2, and - 0-3, the result would be 78-5 0-05 = + 78-45. 3 Invert sugar, on the contrary, has a rotatory power varying markedly with the concentration and tempera- ture. A solution of saccharose containing the normal weight (26-048 grms.) contains, after inversion, 27-419 grms. of invert sugar, and if this is contained in 100 c.c. it gives a deviation of 32-66 in a 20 cm. tube at 20. The variation per degree of temperature is 0-5, so that at this reading would be 42-66 and, in general, at any temperature, t, it would be 42-66 + 0-5 t. If no account is taken of variations due to the concentration, 1 division Ventzke corresponds with 0-8395 grm. of invert sugar in 100 c.c. (Mohr), the solution being read in a 20 cm. tube at 20" ; or 1 grm. of invert sugar dissolved in 100 Mohr c.c. gives a reading of 1-191 division. The specific rotatory power of invert sugar for different concentrations (from 1 to 35 per cent.) is given by the formula : [a] 20 = _ 19-657 0-03611c, c indicating the weight of invert sugar in 100 c.c. For concentrations near 15 per cent, the value 20-2 may be taken for the specific rotation of invert sugar, 1 circular degree then corresponding with 2-475 grm. of invert sugar in 100 true c.c. and 1 grm. of invert sugar in 100 true c.c. giving a rotation in circular degrees of 0-404. Glucose has a specific rotation, [a] 2 D + 52-8", which is constant after muta-rotation has ceased (see p. 27), i.e. if the observation is made after the solution has been either left for 24 hours or boiled for 15 minutes. The concentration and the temperature have virtually no influence on the rotatory power and, with a 20 cm. tube, 1" corresponds with 0-947 grm. in 100 true c.c. and hence 1 grm. of glucose in 100 true c.c. will give a rotation of 1-056. One division on the Ventzke saccharimeter corresponds with 0-3448 (corrected for the dispersion), and the normal weight is 32-71 grms. of glucose in 100 Mohr c.c. ; thus 1 Ventzke division, with a 20 cm. tube and a temperature of 20 corresponds with 0-3271 grm. of glucose per 100 Mohr c.c., or 1 grm. of glucose per 100 Mohr c.c. gives a rotation of 3-057 Ventzke divisions. For fructose (levulose) the data are uncertain owing to the difficulty of obtaining pure crystals, and the rotatory power varies with the concentration (for solutions of about 10 per cent, strength, [a] 2 ^ = 93) and with the temperature (an increase of 1 of temperature diminishes the specific rotatory power by 0-67). One division on the Ventzke saccharimeter corresponds with 0-1838 grm. of levulose in 100 Mohr c.c., or 1 grm. of levulose gives a rotation of 5-439 Ventzke divisions in a 20 cm. tube at the temperature 20. For C, ? H 8? O U + H,O, after the disappearance of the muta-rotation, the specific rotation, which is 486 ORGANIC CHEMISTRY CHEMICAL DETERMINATION OF SUGARS. With the exception of saccharose and raffinose, the sugars (glucose, levulose, &c.) reduce Fehling's solution (an alkaline solution of copper sulphate containing salts of organic hydroxy-acids ; see pp. 212 and 335) in the hot, with separation of a corresponding amount of cuprous oxide. Fehling's solution is obtained by mixing, just before using, equal volumes of the two following solutions : (a) 69-278 grms. of pure crystallised copper sulphate (CuS0 4 + 5H O), air- dried until constant in weight, dissolved in water to 1 litre ; (b) 346 gims. of Rcchelle tali (sodium potassium tartrate) and 100 grms. of pure solid sodium hydroxide dissolved in water to 1 litre. Since saccharose does not reduce Fehling's solution, it must be first inverted. For this purpose, 9-5 grms. of the sugar are dissolved in 700 c.c. of N/5-hydro- chloric acid and the solution heated for 30 minutes in a water-bath at 75, neutralised with caustic soda, and made up to 1 litre. This solution, which contains 10 grms. of invert sugar, is then ready for testing. The Fehling test may be either volumetric or gravimetric, the concentration of the sugar being reduced to about 1 per cent, (by a preliminary trial) and the details of the procedure being followed exactly. Volumetric method : 40 c.c. of water and 10 c.c. of Fehling's solution (5 c.c. of each of the component solutions) are brought to boiling in an Erlenmeyer flask, a measured quantity (4 to 5 c.c.) of the sugar solution run in from a burette, and the liquid again heated and kept boiling for a definite time (2 minutes for glucose or invert sugar, 4 minutes for maltose, and 6 for lactose) ; the flame is then removed, a few drops of the liquid filtered, and the filtrate acidified with a little acetic acid and tested with a drop of potassium ferrocyanide solution. If a red coloration is produced, the test is repeated with a larger quantity of sugar solution, whilst if no red coloration appears, a less quantity of the sugar is tried. This procedure is continued until in the last two tests, representing excess and deficiency of the sugar solution, the difference between the two volumes is not more than 0-1 c.c. ; the mean of these two volumes is employed in calculating the sugar-content of the solution. 100 c.c. of undiluted Fehling's solution, under the above conditions, correspond with 0-4945 grm. of glucose, 0-533 of levulose, 0-515 of invert sugar, 0-740 of maltose, and 0-676 of lactose (hydrated). The gravimetric estimation is carried out as follows (Allihn's method) : To 60 c.c. of Fehling's solution, diluted with 60 c.c. of boiled distilled water and heated to boiling, are added 25 c.c. of the sugar solution of about 1 per cent, concentration, the liquid being thsn again heated and kept boiling for a definite time (2 minutes for glucose, levulose, and invert sugar, 4 for maltose, and 6 for lactose). The solution is then filtered at once, with the aid of a filter-pump, through a dried and weighed Soxhlet tube containing a layer of asbestos, the cuprous oxide being repeatedly washed with a total quantity of 300 to 400 c.c. of boiling water, then with two or three portions of alcohol, and finally with ether. The tube is then dried in an oven, and the cuprous oxide subsequently reduced to metallic copper by passing a current of dry hydrogen through the tube and gently heating the oxide with a small flame ; the hydrogen is kept passing until the tube is quite cold, when the weight is taken. From the weight of copper thus obtained, the but slightly influenced by the concentration, is [a] 2 D = + 52-53; this diminishes by 0-075 for every degree rise in temperature. One degree of rotation corresponds with 0-9519 grm. of lactose in 100 true c.c., so that 1 grm. of lactose gives a rotation of 1-051. For the Ventzke saccharimeter, the normal weight is 32-95 grms. per 100 Mohr c.c. (32-83 grms. in 100 true c.c.), so that 1 Ventzke division in a 20 cm. tube corresponds with 0-3295 grm. of lactose in 100 Mohr c.c. and 1 grm. of lactose in 100 Mohr c.c. gives a rotation of 3-035 Ventzke divisions. Maltose has a specific rotation (after muta-rotation has been destroyed ; see Glucose) varying with the tem- perature and concentration according to the equation : [a]' = 140-375 0-01837 c 0-095 t, where t indicates the temperature and c the percentage by weight of anhydrous maltose. For medium concentrations, W^ = + 138-2 and 1 corresponds with 0-3618 grm. of maltose per 100 true c.c. with a 20 cm. tube and 1 grm. of maltose in 100 c.c. gives a rotation of 2-764. For the Ventzke saccharimeter, the normal weight is 12-55 grms. of maltose in 100 true c.c. or 12-58 grms. in 100 Mohr c.c., and for a 20 cm. tube at the temperature 20, 1 Ventzke division corresponds with 0-1255 grm. of maltose in 100 true c.c., while 1 grm. of maltose in 100 true (Mohr) c.c. gives a rotation of 7-968 (7-949) Ventzke divisions. N Raffinose, C^HajO,, + 5H 2 O, has the specific rotation, [a] 2 ^ = + 104-5, which is almost independent of the temperature and concentration ; for anhydrous raffinose the value is + 123-15. One degree of rotation corresponds with 0-4785 grm. of hydrated raffinose in 100 true c.c. with a 20 cm. tube and 1 grm. of raffinose in 100 true c.c. gives a rotation of 2-09. For the Ventzke saccharimeter the normal weight is 16-576 grms. per 100 Mohr c.c. or 16-537 grm. per 100 true c.c., so that 1 Ventzke division corresponds with 0-16576 grm. of raffinose in 100 Mohr c.c. and 1 grm. of raffinose in 100 Mobr c.c. gives a rotation of 6-033 Ventzke divisions in a 20 cm. tube, QUOTIENT OF PURITY, ETC. 487 corresponding weight of sugar is read off from the following Table, all the numbers representing milligrams : hi ft ft 8 Glucose fa > CUD H" I tf> Maltose Lactose n 1 s Glucose Oi OJ || Maltose Lactose S Pi ft 6 Glucose la a Maltose Lactose 30 16 _ 25-3 155 79-1 81-6 135-9 112-6 280 145-5 151-9 247-8 208-3 35 18-5 29-6 160 81-7 84-3 140-4 116-4 285 148-3 154-9 252-2 212-3 40 20-9 33-9 165 84-3 87-0 144-9 120-2 290 151-0 157-8 256-5 216-3 45 23-4 33-3 170 86-9 89-7 149-4 123-9 295 153-8 160-8 261-1 220-3 50 25-9 42-6 175 89-5 92-4 153-8 127-8 300 156-5 163-8 265-5 224-4 55 28-4 47-0 180 92-1 95-2 153-3 131-6 305 159-3 166-8 269-9 228-3 60 30-8 51-3 185 94-7 97-8 162-7 135-4 310 162-0 169-7 232-2 65 33-3 55-7 190 97-3 100-6 167-2 139-3 315 164-3 172-7 236-1 70 35-3 60-1 195 100-0 103-4 171-6 143-1 320 167-5 175-6 240-0 75 38-3 64-5 200 102-6 106-3 176-1 146-9 325 170-3 178-6 243-9 80 40-8 68-9 205 105-3 109-1 180-5 150-7 330 173-1 181-6 247-7 85 43-4 73-2 210 107-9 111-9 185-0 154-5 335 175-9 184-7 251-6 90 45-9 46-9 77-7 215 110-6 114-7 189-5 158-2 ; 340 178-7 187-8 255-7 95 48-4 49-5 82-1 220 113-2 117-5 193-9 161-9 345 181-5 190-8 259-8 100 50-9 52-1 86-6 71-6 225 115-9 120-4 198-4 165-7 350 184-3 193-8 263-9 105 53-5 54-8 91-0 75-3 230 118-5 123-2 202-9 169-4 355 187-2 196-8 268-0 110 56-0 57-5 95-5 79-0 235 121-2 126-0 207-4 173-1 360 190-0 199-8 272-1 115 58-6 60-1 99-9 82-7 240 123-9 128-9 211-8 176-9 365 192-9 203-0 276-2 120 61-1 62-8 104-4 86-4 245 126-6 131-8 216-3 180-8 370 195-7 206-1 280-5 125 63-7 65-5 108-9 90-1 250 129-2 134-6 220-8 184-8 ; 375 198-6 209-2 284-8 130 66-2 68-1 113-4 93-8 255 131-9 137-5 225-3 188-7 380 201-4 212-4 289-1 135 63-8 70-8 17-9 96-6 260 134-6 140-4 229-8 192-5 385 204-3 215-5 293-4 140 71-3 73-5 22-4 101-3 265 137-3 143-2 234-3 196-4 390 207-1 218-7 297-7 145 73-9 76-1 26-9 105-1 270 140-0 146-1 238-8 200-3 395 210-0 221-8 302-0 150 76-5 73-9 31-4 108-8 275 142-8 149-0 243-3 204-3 400 212-9 224-9 306-3 NON-SUGAR, APPARENT DENSITY, TRUE DENSITY, AND QUOTIENT OF PURITY. Sugars and their solutions are distinguished, commercially and industrially, by their content of saccharose, water, and solids not sugar (e.g. salts and various organic substances). The Brix saccharometer is graduated with pure sugar solutions, and hence gives results which are increasingly inaccurate as the degree of impurity of the sugar solutions increases. Apparent density is that shown by the Brix hydrometer, while the real density corresponds with the true content of sugar determined by direct analysis (by the polarimeter or, after inversion, by Fehling's solution). The difference between the real and apparent densities, expressed in degrees Brix, indicates the non-sugar in Brix degrees, while the ratio between the real and apparent densities, in degrees Brix, is termed the quotient of purity and, when multiplied by 100, shows the percentage of sugar present independently of the water. In the analysis of a mixture of various sugars a number of optical and chemical tests must be made in order to deduce, directly or indirectly, the quantities of the separate components (see Villa vecchia, Chim. Anal. Tecnol., vol. ii, pp. 223 et seq.). 1 1 If only saccharose and another sugar are present, p grms. of the mixture are dissolved in water to 100 c.c. and the polarisation, P, read ; if a t is the rotation of 1 grm. of saccharose per 100 c.c. and a 2 that of 1 grm. of the other sugar, the quantities x and y of saccharose and the other sugar respectively are given by the formulae (I) x = ~ atP (II) y = ' P ~ since x + y = p (III) and ap + a^y = P (IV). The values of a lt a 2 , and P must be given their proper algebraic signs (+ or ). A. In the special case of a mixture of saccharose and glucose, the components x and y may be determined in various ways : (1) The glucose (y) may be estimated by means of Fehling's solution ; formula IV then gives x = (V). ai Since saccharose reduces Fehling's solution to a very slight extent, small proportions of glucose are best deter- mined by means of Soldaini's reagent, which consists of 150 grms. of potassium bicarbonate, 104-4 grms. of normal potassium carbonate, and 100 c.c. of tho copper sulphate solution used for Fehling's solution, made up to a litre with water. (2) The solution of the mixture is polarised (P), the saccharose being inverted and the polarisation again read (P,). If a 3 is the ro tation of 1 grm. of invert sugar (= 1-191), then, since 1 grm. of saccharose gives P P, 1-053 grm. of invert sugar, we have 1-053 a,x + a$ PI (VI) and hence x = - (VII) and g.P. - 1-053 3 P a 2 (a! l-053a 3 ) 3-839P, + 1-254P 5-093 X 3-057 (VIII) a t having the value 3-839 and o, 3-057, it follows that x = a l l-053a 3 P _ p ' (IX) and (X), which are the quantities of the two sugars in p grms. of the mixture. The 488 ORGANICCHEMISTRY The total ash of a sugar is determined by weighing 3 grms. into a tared platinum dish, moistening it with a few drops of concentrated sulphuric acid, carbonising over a Bunsen flame and incinerating in a muffle at a low red heat (about 700) so that the ash does not fuse. From the sulphated ash, one-tenth of its weight is deducted in order to correct for the increase due to the formation of sulphates. By means of tables the quantity of soluble ash can also be deduced. The water is determined by heating 5 to 10 grms. of the sugar in a flat glass dish covered with a clock-glass at 105 to 110 first for 2 hours and subsequently to constant weight. Subtraction from 100 of the water and the sugar gives the percentage of total non-sugar, while further subtraction of the ash gives the organic non-sugar. The alkalinity of the sugar is determined by titrating an aqueous solution of 20 grms. of the product with decinormal sulphuric acid in presence of phenolphthalein ; the result is calculated as grammes of CaO per 100 grms. of sugar. lOOz WOy Percentages will therefore be - and - respectively. For a mixture of saccharose and levulose, a 2 = 5-439, 3-839P, 1-254P so that y = - JvT^TTi - > f r saccharose and invert sugar, 2 = 1-191 and the denominator becomes * 3-839P, -f 1-254P 6-066 instead of 27-701 ; for mixtures of saccharose and maltose, a, = 7-949 and y = - 7~alq > for saccharose and lactose hydrate, a 2 = 3-035. (3) The glucose is first determined by means of Fehling's solution ; in another portion of the solution the .saccharose is inverted and the reducing sugars again estimated with Fehling's solution ; the difference between these two estimations gives the invert sugar and this, multiplied by 0-95, the saccharose. B. With a mixture of saccharose and rafflnose, the polarisation is determined before (P) and after (PJ inversion ; ! and a 2 being the known rotations of 1 grm. of each of the two sugars and a 3 and a 4 those of 1 grm. of the respective inverted products, it follows that : a& + a^j = P t (XIII) and 1-053 a a x + 1-036 a^y = P! (XIV) ; substitution in these of the values c^ = 3-839, a t = 7-11, 1-053 a t = -- 1-254 and 1-036 a 4 = 3-643 gives 0-5124P P, 1-254P + 3-839P t and y - ^: - . For the determination of the rafflnose by means of methyl- "' phenylhydrazine, in presence of saccharose and invert sugar, see Rafflnose, p. 442. C. When two reducing sugars, but neither saccharose nor rafflnose is present, it is sufficient to measure the polarisation and apply formulae I to IV. But if a non-saccharine substance is also present, it is necessary to determine also the number (F) of c.c. of Fehling's solution reduced by a weight, p, of the substance ; if 6j and 6 a are the volumes (c.c.) of Fehling's solution reduced by 1 grm. of each of the two sugars dissolved in 100 c.c., then : a t x + a& = P (XV) and bjX + b 2 y = F (XVI) and hence x = ' ~ "l and y = a ' ' ~ ', . With a mixture of glucose and levulose, j = 3-057, a., = 5-439, 6j = 202-4, and 6, = 186, so that x - - - 1669 3-057.F - 202-4P and y = - i-^r - . For mixtures of glucose and maltose, a 2 has the value 7-940 and 6 2 135 ; these last lOO9 two numbers hold also for mixtures of invert sugar and maltose, but then a t 1-191 and 6 t = 194 ; for mixtures of glucose and lactose, a^ 3-057, a 2 = 3-035, b^ = 202-4, and 6 2 = 148, while for invert sugar and lactose, a. 2 and & 2 have the values just given, but a t = 1-191 and bi = 194. D. With a mixture of saccharose (x), glucose (y), and levulose (z), if a weight, p, is dissolved to 100 c.c., and !, a 2 , a 3 , a t represent the respective rotations of 1 grm. of each of these sugars and of invert sugar in 100 c.c., 6 2 and 6 3 the number of c.c. of Fehling's solution reduced by 1 grm. of each of the reducing sugars, P and P,, the polarisations before and after inversion, and F the number of c.c. of Fehling's solution reduced by weight p of the substance, then (XVII) P = OjX + a s y + a 3 z ; P t = 1-053 a t x + 1 : 75). The origin of cellulose in plants cannot be regarded as a condensation of starch ; the latter is probably converted into glucose, which gives cellulose on condensation. The preparation of nitrocellulose (pyroxyline, guncotton, collodion-cotton) has already been described in the chapter on Explosives (p. 232). Cellulose Formate (Blumer, Ger. Pat. 179,590) has also been prepared. At 210 cotton begins to decompose with evolution of carbon monoxide and dioxide, 1 Numerous attempts have been made to convert wood industrially into saccharine substances and so prepare alcohol (see p. 142), but it is only recently (1910 and 1911) that Flechsig, Ost, and Wilkening showed that cellulose can be transformed completely into fermentable glucose by dissolving it in concentrated sulphuric acid, diluting until the solution contains only 1 to 2 per cent, of acid, and then heating at 110 to 120 (but not to 125, as was done by Simonsen, since a part of the glucose is thereby destroyed). 2 Cellulose Acetate forms a horny, amorphous mass soluble in chloroform, tetrachloroethane, aniline, pure acetic acid, and boiling nitrobenzene. The less highly acetylated products are soluble in alcohol, giving a solution which, together with camphor, serves for the preparation of cellite films for cinematographs ; these films are considerably less inflammable than those of celluloid. The triacetyl-compound is used for making artificial silk (see later), and is prepared by treating hydrocellulose in the cold with acetic anhydride, a few drops of concentrated sulphuric acid, and a little glacial acetic acid or phenolsulphonic acid (see also the following patents : Gcr. Pats. 118,358, 153,350, 159,524, 163,316, 175,379, 185,837, 203,178, 203,642, 206,950, 224,330 ; Fr. Pats. 316,500, 319,848, 324,862, 345,764, 368,738, 368,766, 371,357, 371,447, 385,179, 385,180; U.S. Pats. 733,729, 826,229, 838,350 ; Eng. Pat. 9998, 1905). More or less successful attempts have also been made to acetylate cellulose in the hot with acetyl chloride and metallic acetates, the reaction being facilitated by the addition of a small quantity of pyridine or quinoline and, in some cases, of a solvent of cellulose acetate (e.g. acetone, nitrobenzene, naphthalene, A-c.) ; see Ger. Pats. 85,329, 86,368, 105,347, 139,669, and U.S. Pat. 709,922, according to which phenol- or naphthol-sulphonic acids are added. ^ The following method of manufacture (from Fr. Pat. 347,906) admits of the direct acetylation of cotton textiles and may be taken as an example : 10 kilos of defatted cotton, containing 10 to 20 per cent, of moisture, are heated with 40 kilos of acetic anhydride (containing 0-25 per cent, of concentrated sulphuric acid) and 150 kilos of benzene, at 70 to 75, in a reflux apparatus until a small portion of the cotton dissolves completely in chloro- form ; the whole mass is then pressed and dried. Cross, Bevan, and Briggs (1907) obtain cellulose acetates easily and cheaply, without preparing hydrocellulose ; cellulose is treated directly with a mixture of 100 parts of glacial acetic acid, 30 of zinc chloride, and 100 of acetic anhydride, the whole being heated for 36 hours at 45. HYDROCELLULOSE, MALTOL, ETC. 505 methyl alcohol, acetic acid, acetone, hydrocarbons, &c. (see Distillation of Wood, p. 272). By the dry distillation of pure cellulose (Swedish filter-paper) under ordinary pressure, E. Erdmann and C. Schaefer (1910) obtained about 5 per cent, of tar, 42 per cent, of acid liquors, and a residue of carbon, together with gas containing 66 per cent. CO, 19 per cent. CH 4 , 11-5 per cent. H 2 , &c. ; from the acid liquors, acetone, formaldehyde, furfural, methoxyfurfural, maltol (C 6 H 6 3 , which, according to Peratoner and Tamburello, has CH- 0- OCH 3 , the constitution || ), and y-valerolactone were separated. With lapse of CH-CO-OOH f time or under the action of bacteria, &c., cellulose undergoes various changes (see Peat, Lignite, Coal, vol. i ; and Methane, p. 32 of this volume). The action of sulphuric acid on cellulose varies somewhat with the concentration of the acid, the duration of the reaction, and the temperature. The concentrated acid has a gelatinising action and dissolves part of the cellulose, which is reprecipitable by water or ammonia. If the action is protracted, the very friable Hydrocellulose, C 12 H 2 2Oii [(C 6 H 10 5 ) 2 , H 2 O], is formed, but, in general, hydrocelluloses of diminishing molecular weight and increasing friability (e.g. cellobiose ; see above) are successively formed. The hydrocellulose formed in the preparation of artificial silk is only slightly friable, and has probably the formula (C 6 H 10 5 ) 6 , H 2 0. Since also these hydrocelluloses exhibit rather different behaviour towards dyes, it has been suggested that the name hydrocellulose be given to that resulting from advanced hydrolysis by non-oxidising acids ; the increase of weight during this change, owing to the addition of hydrolytic water, is 3-5 to 5 per cent., this being lost at above 125, whilst the hygroscopic moisture is expelled at 104. This hydrocellulose reduces Fehling's solution (Ost ; Cross and Bevan, 1909). On the other hand the name cellulose hydrate or hydracellulose is given to that obtained by gentle alka- line hydrolysis, which produces an augmentation in weight of 8 to 10 per cent. ; here, too, this hydrolytic water is given up at temperatures above 125. Hydracelluloee does not reduce Fehling's solution. Schwalbe (1907) measured the reducing power of hydrocellulose towards Fehling's solution. 1 Cross and Bevan proposed for cellulose the formula : H OH H OH o:c \ X C C' X H H Oil H OH some polymeride of this, such as H OH H OH H OH H OH H OH c a c ---- c, H C (T C ' &c H OH H OH H OH H OH On the basis of the formation of the trinitrate and triacetate, Green (1894) suggested for cellulose a formula CH(OH) CH CH(OHK (or some multiple of it) containing 3 OH, namely : | ^>^ /' an< * * or hydrocellulose the CH(OH) CH - CH 2 - / CH(OH) CH - CH(OH) 2 formula | ^>O ; these constitutions explain the formation of furfural by the decomposition CH(OH) CH - CH 2 - OH of cellulose and also the formation, under the action of oxidising and bleaching agents, of oxycellulose containing ketonic groups which react with phenylhydrazine, reduce Fehling's solution, and admit of direct dyeing by basic dyes (e.g. methylene blue). Two oxycelluloses are, however, distinguished : the one very similar to hydrocellu- lose and insoluble in boiling dilute alkali, and the other possessed of considerable reducing power and soluble in alkali. The hardening of cellulose in the formation of wood is due to its partial transformation into LIGNIN, which is not yet well defined chemically but certainly contains methoxy-groi'ps, which explain the formation of methyl alcohol and acetic acid when wood is distilled. According to Green, lignin is formed by dehydration of cellulose CH:C - CH-OH and would be a polymeride of | ^>O ~^ > ^ > -^ u * Klason is of the opinion that lignin is a kind of glucoside CH:C - CH-OH 506 ORGANIC CHEMISTRY When sheets of pure, unsized paper are immersed for a few minutes in sulphuric acid of 50 to 60 Be. and then washed immediately in a plentiful supply of water, they are converted into parchment paper (artificial parchment), amyloids being formed at the surface. These artificial parchments are distinguished from the natural ones by the presence of nitrogen in the latter, and from paraffined paper by the extraction of the paraffin from these by ether. Parchment paper is rendered softer and more transparent by immersion in glycerine or glucose solution. If cellulose pulp is well ground and beaten in the Hollander until it forms an almost gelatinous pulp, a translucent paper can be obtained which is similar to artificial parchment and, under the name of pergamin, is largely used as a wrap- ping for foods and fatty materials ; this may easily be distinguished from vegetable parchment, which is composed of cellulose hydrate (amyloids) and is hence coloured blue by a solution of iodine in potassium iodide, whilst pergamin gives no such coloration. With concentrated zinc chloride solution, cellulose gives compounds similar to those it forms with sulphuric acid : papers thus prepared and then superposed and compressed form the so-called vulcanised paper ; this is very hard, impermeable to water, and a bad conductor of electricity, and is used for making plaques, tubes, and noiseless gearing. When cellulose is treated for a long time with energetic oxidising agents, it is converted into oxycellulose (C^gH^O] 6 ) v , which lowers the resistance of the tissues and, unlike cellu- lose, reduces Fehling's solution and fixes, although feebly, basic dyes and alizarine without a mordant. Hydrocellulose reduces Fehling's solution slightly and is not coloured by basic dyes. When cellulose (spun or woven cotton) is treated in the cold with concentrated caustic soda solution (25 to 35 Be.), it swells and becomes semi-transparent owing to the forma- tion of sodiocellulose, and treatment of this with a large amount of water converts it into hydrocellulose (see above), the original appearance of the cellulose being retained. But in the hot sodiocellulose cannot be obtained (see Part III, Textile Fibres and Mercerised Cotton), prolongation of the action then resulting in decomposition into oxalic acid. Hygroscopic water held by cellulose is eliminated by heating at 100 to 105 ; the water of hydration in hydrocellulose is determined by heating in toluene or petroleum or at 130. The hydration occurring during mercerisation increases the weight of the cotton by 8 to 10 per cent. Mercerised Cotton is distinguished from ordinary cotton by Knecht's test : the material is dyed in the hot with 5 c.c. of benzopurpurin 4B solution (0-1 grm. in 100 c.c. of water), and when it has taken the colour well, about 2 c.c. of concentrated hydrochloric acid are added drop by drop to the bath ; the non -mercerised cotton then turns bluish black, while the mercerised remains red. If oxycellulose (which is formed even by the action of calcium hypochlorite) is present, the material is dyed with Congo red and the acid then added, the ordinary cotton and the oxycellulose assuming the bluish black colour, while the mercerised cotton remains red ; but if the material is thoroughly washed, the pure cotton becomes red, the oxycellulose retaining its bluish black and the mercerised cotton its red colour. PAPER INDUSTRY As prime material in the paper industry, use has been and is still made of all the cellu- losic fibres obtained from most widely differing plants, 1 linen and cotton rags, straw, wood, hemp, &c. with two aromatic nuclei containing mcthoxy- and hydroxy-groups, also lateral groups, -CH : CH and CH 2 -OH, besides the fundamental cellulose grouping ; it is probably represented by the formula (C^H^Oj,));. Dry wood contains 26 to 30 per cent, of lignin. Schultze, Tollens, and Konig hold the view that the hard part of wood is formed of cellulose, together with small proportions of pentosans and of lignin. The formation of wood in plants has been recently attributed by Wislicenus to the 'colloidal character of the plant fluids which, in the initial phase, transport into the tissues the cellulosc-hydrogcl as a superficial, chemically indifferent sub- stance ; in a second phase, the latter is lignifled by absorption and surface gelatinisation of the colloidal meta- bolic substances contained in the sap. Lignocellulose is hydrolysed and dissolved by zinc chloride solution and by ammoniacal copper oxide solution, dilute acids and alkalis also exerting a hydrolysing action. Lignin gives a number of colour reactions, e.g. with aniline sulphate (yellow), with phloroglucinol and hydrochloric acid (red) ; with potassium ferricyanide it forms potassium ferrocyanide, and with fhchsine decolorised by sulphur dioxide it gives a red colour ; it fixes various aniline dyes (e.g. methylene blue, eosin, &c.) directly. Wood is regarded by Cross and Bevan as an ester of lignocellulose, derived from cellulose (polyhydric alcohol) and lignic acid (lignin). When pure cellulose is subjected to dry distillation, it does not yield methyl alcohol, which is, however, formed from wood ; the alcohol must hence be derived from the lignin. But acetic acid is formed from both lignin and cellulose. 1 History of the Paper Industry. The origin of paper dates back to the second century B.C., when the first traces of it were evident in China. In early times races marked their records and writings on stone, wood, PAPER 507 It is not possible here to review all the wonderful mechanical improvements which rendered paper-making one of the most interesting and important industries of the nine- teenth century. From the arrival of the wood in the factory to the despatch of the rolls or reams of paper, all the operations are carried out mechanically by means of perfected machinery, which is not only more rapid in its action but more accurate than hand labour. A description cannot be given here of all the varied and ingenious dressings employed to obtain different kinds of paper, or of the mineral loading of kaolin, barium sulphate, gypsum, &c., with which some papers are so impregnated that the mineral substances exceed the vegetable matter, to the delight of the tradesman who sells gypsum for cheese or sausages. What will be attempted here will be simply a brief description of the various treatments to which the raw material is subjected to convert it into paper. Paper factories require a plentiful supply of pure water, which must not contain iron and should be filtered if turbid. .The rags, gathered in places of all sorts and in all conditions, are acquired from the rag- merchants, who separate those of wool and silk, which go to wool factories, &c., and often sort the remaining linen and cotton rags into light and dark sorts. It is calculated that Italy produces about 600,000 quintals of rags, only some 35,000 of which are made into paper, while in 1905 20,700 quintals of vegetable rags (at 9*. 6d. per quintal) were imported, together with 30,000 quintals of animal rags (at 48s.) and and parchment. In the seventh and eighth centuries the Japanese and other neighbouring peoples learnt how to prepare paper from the bark of various trees, this industry then becoming known to the Arabs, but only much later in Europe. In 1190 paper made its appearance in Germany, in 1250 in France, in 1275 in Italy, and in 1430 in Switzerland. In the East, besides bark, cotton and linen rags were also employed for paper -making. In Italy the first important factory furnished with grinders and pistons for the preparation of the raw material was erected at Fabriano in 1320. With the subsequent discovery of printing, the paper industry underwent an unforeseen and marked development, and grew to enormous proportions in the nineteenth century. About the middle of the eighteenth century, the pistons and grindstones in use up to that time for treating the raw materials were gradually replaced by the so-called Hollanders, which led to an increase in the output and an improvement in the quality of the product. The demand for paper increased largely at the end of the eighteenth century, the form being improved and the price lowered. Mechanics and chemistry came to the aid of the paper manufacturer, and as early as the beginning of the nineteenth century the paste of cotton or linen fibres, mixed in large tanks, was transformed into a thin sheet of paper by means of a revolving, perforated drum, through which the water escaped. It was about 1825 that rudimentary continuous machines were first employed, these supplying an uninterrupted strip of paper a metre in width at a rate of 10 metres per minute. The imposing and complex, but very accurate, continuous machines of the present day give paper as much as 4 metres wide at 150 metres per minute. Great advances were also made in the chemical treatment of the raw materials. In the first quarter of the nineteenth century, the putrefaction to which the rags were subjected so that they might be more easily disintegrated was replaced by heating with soda and lime in open boilers and, later on, in closed boilers under steam pressure. Then came bleaching of the fibres with gaseous chlorine and subsequently with chloride of lime. The yellow cellulose obtained from straw can also be bleached in this way, and since 1830 has been used in large quantities for the commoner papers and for mixing with rags. Sizing of paper by means of resin soap, although suggested in 1800, only later came into general use. With the rapidly increasing consumption of paper, there came a time of dearth of raw materials ; cotton and linen rags were no longer obtainable in sufficient quantities, and straw could not be used alone. It hence became necessary to look for other sources of cellulose, and it is to Keller that we owe the happy solution of this pressing problem. In 1843 he succeeded in utilising wood-cellulose by means of machines which, rotating rapidly against logs of wood kept wet, gradually converted the wood into an aqueous pulp made up of the separate fibres ; these machines were improved later by Volter, and the first factories of mechanical wood-pulp were erected. Thig inexhaustible material can be purified by boiling it with caustic soda in digesters under pressure and bleaching the resultant brown mass with chloride of lime ; this procedure gives chemical ivood-pulp, which to-day forms the basis of almost all kinds of paper, from the finest to the commonest. In 1884 Dahl effected considerable economy in the manufacture of wood pulp by replacing the expensive caustic soda to a large extent by sodium sulphate ; calcination of the evaporated residue of the exhausted lye yields mainly caustic soda, sodium carbonate, sulphide, thiosulphate, &c., and a solution of this product acts on wood, giving a whiter and more resistant product. But although this process was applicable with advantage to straw cellulose, which gives good results only when treated with alkali or sulphate (the consumption of straw is limited nowadays by its increasingly high price), it was not convenient for dealing with the enormous quantities of wood necessary to meet the growing demands for paper. As early as 1865, Tilgman in America had attempted the chemical purification of mechanical wood-pulp by digestion with acid sulphites, and in 1874 Ekman's large factory at Bergvik was working regularly with magnesium bisulphite. Meanwhile, Professor Mitscherlich of Monaco (1872) had suggested the improvement of this process by using calcium bisulphite in large digesters under pressure. From that time and especially after the improvements introduced by Keller, the use of bisulphite spread gradually in Germany and other European countries and received a fresh impetus on the lapse of Mit- scherlich's patents. At the present time, with rare exceptions these including the treatment of straw, which contains silicates not attacked by bisulphite almost all wood-pulp is transformed into cellulose by the bisulphite process. This process not only effects economy in the digestion of the wood-pulp, but results in an increased yield of a whiter and more resistant product. With improvements in the chemical methods and especially by the use of energetic bleaching processes (chlorine, chloride of lime, electrolytic alkali, hypochlorite, &c.), it became possible to utilise the wood of many different trees from the fir to the poplar so that there is now no danger that raw material for paper-making may some day fail. In Canada alone there are still forests large enough to supply the whole world with paper for 800 years, even with a much larger annual consumption than at present. 508 ORGANIC CHEMISTRY 11,200 of mixed rags (at 8s. lOd.) ; in 1910 the corresponding amounts were 15,300, 21,000, and 21,400 quintals respectively. The exports in 1905 were 15,000 (at 28s.) 12,000, and 210 quintals respectively ; in 1906 2631 quintals were exported, in 1908 only 508, and in 1910 less than 200 quintals. Also 65,000 quintals of macerated paper (recovered waste paper) are utilised every year in Italy. FIG. 382. FIG. 383. In 1909 17,777 tons of linen and cotton rags, of the value of 180,000, were imported into England. The rags arrive at the paper factory in large bundles, some light and others dark. Preference is given to linen rags, since these give longer and tougher fibres and are used also to improve those of cotton. The first operation to which the rags should be subjected is disinfection, either by heat (great care being then taken to avoid fires) or by gaseous disinfectants (e.g. by introducing the bales into large iron cylinders, which are then evacuated and filled with formaldehyde vapour). In many factories, however, this disinfection is omitted, the health of the sorters being thus jeopardised. Sorting is carried out by workpeople who spread the loose rags on tables and separate carefully those which are more or less white and those which are coloured to vary- ing degrees ; the larger pieces are then cut by special cutters (Fig. 382), having a number of hori- zontal knives fixed to the peri- phery of a cylinder, the seams, buttons, hooks, &c., being pre- viously removed. The different qualities then pass to suitable machines to be cleaned and brushed. Fig. 383 shows a simple form of duster, in which the rags are beaten vigorously by pegs or rapidly revolving hori- zontal wooden cylinders and carried to the opposite end of ^ e macmne > while the dust is removed by an air-draught to be deposited in chambers or in large bag-filters of various types (Fichter, Beeth, &c.). After this the rags are washed a little with water in vessels similar to hollanders (see p. 513) without knives but with a vaned wheel and a gauze drum for renewing the water. They are next removed to revolving spherical boilers, where the residual dirt is eliminated and any dye, fat, resin, starch, gum, or other impurity destroyed. This is effected by boiling, sometimes with soda or caustic soda, but more commonly with lime (2 to 5 per cent, on the weight of the rags) and water. These boilers (Fig. 384) hold as much as 2000 FIG 384 MECHANICAL PULP 509 kilos of rags and make about two revolutions per minute, while steam is passed in through a tube traversing the axis until a pressure of 2 to 3 atmos. is reached. The boilers are coated with insulating material, and the boiling lasts for 6 to 12 hours, according to the nature of the material. When the boiling is finished, the steam under pressure is released into the adjacent boiler, in which the operation is just starting, and the rags removed, rinsed well in water, and reduced to a fine pulp in machines similar to hollanders (see later) with cast-iron or reinforced concrete tanks, the knives of the drum not being set too close to those of the fixed plate. About 20 horse-power is required by the hollanders for a charge of 200 kilos of rags. The loss in weight in all the operations up to the present stage varies, according to the quality of the material, from 12 per cent, to 40 per cent. In hollanders or similar vessels holding up to 800 kilos of rags, the bleaching is carried out with a clear solution of chloride of lime, of which 2 to 10 kilos are required per 100 kilos of rags ; a little sulphuric acid (100 to 200 grms. per 10 kilos of chloride of lime) is finally added to liberate all the chlorine from the bleaching agent. In some factories fresh electrolytic solutions of sodium hypochlorite (see vol. i, p. 457) are used. The bleaching must not be too prolonged, and the pulp is afterwards washed in large quantities of water until all smell of chlorine has disappeared and potassium iodide starch paper is no longer turned blue or blue litmus paper reddened ; as a pre- caution, 30 to 50 grms. of sodium thiosulphate (antichlor) and soda are added to each vessel. The bleached mass or half-stuff, as it is called, is freed from water and allowed to drain for some days in brickwork chambers with pave- ments of absorbent grooved bricks. From these it is taken in the moist state as required for mixing with bleached wood-pulp. The mixture is beaten in true hol- landers, the knives being set more or less close according as more or less fine refined pulp is required. WOOD-PULP (Mechanical Pulp). The treatment to which the woody parts of the various plants suitable for paper-making [fir, pine, larch, poplar (Populus nigra or, better, Populus canadensis), beech, birch, esparto (of which Algeria exports half a million quintals annually), straw, hemp, broom, &c.], varies somewhat, as the cellulose and the surrounding lignin are present in different proportions and in different states of aggregation. 1 Logs containing few knots are cut into the required lengths (40 cm.), which, after the knots have been removed by a boring machine, are barked in another machine. The logs are then defibred by being pressed against a stone mill, which revolves rapidly and removes the fibres tangentially. This mill is about 1 J metre in diameter and 35 to 40 cm. thick, and it revolves either horizontally or vertically (at 150 to 180 turns per minute). To the latter type belongs the vertical grinder devised by Vbith and subsequently improved in various ways (Fig. 385). The three chambers corresponding with the three toothed rods, B, contain 1 In the disintegrated wood, the proportion of cellulose is determined by digesting several times with sodium bisulphite solution and then treating repeatedly with chlorine at 0, by which means almost all the constituents except the cellulose are dissolved. For the determination of the crude cellulose in plants, Weender's older method, modified by Uermeberg and Stohmann, has been largely replaced by that of Gabriel (or Lange and Konig) : 2 grms. of the finely divided substance are heated in a beaker with 60 c.c. of alkaline glycerine (33 grms. of caustic soda dissolved in a litre of glycerine) at 180, the mass being then cooled to 340 and poured into a basin con- taining 200 c.c of boiling water, with which it is mixed and allowed to settle. The supernatant liquid is drawn off thiough a siphon covered with cloth at the end dipping into the liquid, and the deposit boiled with 200 c.c. of water which is siphoned off as before. The boiling is repeated with 200 c.c. of water containing 5 c.c. of con- centrated hydrochloric ^cid, and the residue finally brought on to a tared filter, washed with water, alcohol, and ether successively, dried and weighed as crude cellulose. To determine the pure cellulose, almost free from pentosans, ash, &c., KSnig's method is used : 3 grms. of FIG. 385. 510 ORGANIC CHEMISTRY the logs cut to the proper length, and, while the grinder revolves, these are pressed against it by the corresponding covers which are forced down by the toothed rods ; the latter connect with gearing worked by a chain, D, the velocity of which is proportioned to that of the grinder. The pressure is nowadays exerted hydraulically ; Fig. 386 shows a series of such vertical grinders in which hydraulic pressure is employed. Horizontal grinders (Fig. FIG. 386. 387, vertical section ; Fig. 388, general view) with hydraulic pressure are now widely used, as they admit of a larger number of logs being ground at the same time. While in opera- tion, the grinder is continually sprayed with water to prevent heating and to remove the woody fibres as they are liberated. According to the pressure of the logs on the grinder and to the speed of the latter FIG. 387. FIG. 388. a more or less fine pulp is obtained with a smaller or larger content of splinters, dust, and other irregular and unusable portions ; these are removed by means of sloping sieves, the finely divided, air-dried material are treated with 200 c.c. of glycerine (sp. gr. 1-230) containing 4 grms of concentrated sulphuric acid in a dish which is heated in an oven at 137 for exactly one hour, the liquid being then allowed to cool to 80 to 100, mixed with 200 to 250 c.c. of hot water, boMed and filtered hot through an asbestos filter with the help of a pump. The filter is then washed with 300 to 400 c.c. of hot water, then with boiling alcohol, and finally with a hot mixture of alcohol and ether. The filter and its contents are next introduced into a platinum crucible, which is dried at 105 to 110 and weighed. The crude cellulose is then ashed by heating to redness, the loss in weight thus produced representing the crude cellulose free from ash. If, in a second estima- tion, the cellulose is not dried and ashed, but is repeatedly treated for several hours with strong hydrogen per- oxide and ammonia, and finally washed, dried, weighed, ashed, and again weighed, the proportion of pure, white cellulose is obtained. The difference between the crude and the pure cellulose represents the lignin. CHEMICAL PULP 511 B and C (Fig. 389), on to which the channel, A, conveys the water to carry away the crude wood-pulp, while powerful water -jets carry the splinters (b), the good fibre (c), and the dust (E) to various collecting channels. Cylindrical or superposed sieves are also used. When the wood-pulp is to be used immediately for making paper, it is mixed with the necessary quantities of rag-pulp and dressing and worked up as described below. But generally the wood-pulp is placed on the market, in which case the water is removed and the pulp converted into sheets by sucking it on to drums of metal gauze or travelling planes, through which the water is drawn by suction ; the continuous layer of pulp is cut into lengths and is best dispatched in the wet state (with 40 to 60 per cent, of water). But sometimes the FIG. 389. sheets are dried on hot drums, although this renders difficult the subsequent treatment necessary to transform them into pulp in the hollanders. Wood-pulp is yellowish or rather brown, and still contains all the encrusting substance (lignin) ; it cannot be used as it is for paper, the action of light altering its colour imme- diately. It cannot be bleached with chloride of lime or alkaline reagents, which intensify its yellow colour ; but good results are obtained with sulphur dioxide, which does not, indeed, remove the yellow tint but prevents the browning or reddening which gradually sets in. Barked and cleaned logs yield about one -half their weight of dry wood-pulp (containing 12 to 15 per cent, of moisture). CHEMICAL WOOD-PULP. This is obtained by removing the encrusting matter from the wood by means of various chemical agents. It was Payen who first, in 1840, attempted this purifica- tion with nitric acid, and who afterwards tried caustic alkalis, sulphurous acid, &c. The prepara- tion of the cellulose in the chemical way can be effected by (a) the soda process or (b) the bisulphite process. (a) The logs freed from bark and knots are converted into sticks 1 cm. thick, which are heated for some hours with caustic soda of 12 Be. under a pressure of 6 to 8 atmos. (160 to 170) in large digesters, 100 to 200 cu. metres in capacity. Various types of digester are in use, Fig. 390 showing the vertical type devised by Sinclair. This consists of an iron cylinder, A, 5 to 6 metres in height, with conical extremities, a charging orifice, C, a wide horizontal discharge tube, C lt a tube, b, by which the caustic soda is introduced, and an inner perforated jacket, which is filled to the extent of four-fifths with the sticks. The FIG. 390. reservoir, G, contains a supply of caustic soda solution, and circulation in the digester can be effected with the help of a Korting injector, the cocks of the tubes, h lt and h, being opened ; the latter conveys the alkali on to the sticks, while that collected between the perforated jacket and the inner wall of the digester ascends through h v The hot gases from the hearth, K, heat the digester and pass through E to the chimney. At the end of the operation the highly coloured alkali is discharged from the tap, V, and can be used for several successive treatments, being reinforced each time with a little sodium carbonate 512 ORGANIC CHEMISTRY FIG. 391. The soda is eventually recovered from this liquor by evaporating in a vacuum, calcining the residue, extracting the sodium carbonate thus formed with water, boiling with milk of lime, and decanting the resultant caustic soda solution (see vol. i, p. 441). But for this recovery of the soda, this process would be inapplicable. A method which is more econo- mical and more generally used consists in reinforcing the alkali liquor first used with sodium sulphate, instead of the carbonate, for subsequent operations ; the liquor is then ultimately evaporated in a vacuum and calcined, the sodium sulphate, in presence of carbonised organic matter, being converted partly into caustic soda and partly into sodium sulphide (which exerts on wood the same action as caustic soda), just as in the preparation of soda by the Leblanc process (see vol. i, p. 468). Extraction of the calcined mass with water yields a liquor containing sodium sulphate, sulphide, and carbonate, and is ready to act on fresh quantities of wood in the digester. Cellulose thus prepared is termed sulphate pulp. The concentration of the alkaline liquor is accompanied by the production of pungent and disagreeable odours, which are a source of annoyance to the neighbourhood, so that hi certain countries (e.g. Scandinavia) such concentration is prohibited. It has been suggested to destroy these odours (due to mercaptan) by nitrous vapours. Use is also made of horizontal autoclaves arranged in series like sugar diffusors (see p. 451), while ordinary vertical iron digesters, as shown in Fig. 391, are largely employed. The digesters can be heated with indirect steam for 24 to 48 hours, or, more economically and rapidly, by direct steam (10 to 15 hours) to 140 to 150 (12 to 15 atmos.), but the yield is then rather lower and the mass slightly more attacked. The residual cellulose is washed, in the ^digesters themselves or in hollanders, with water and steam and is then mixed with the quantity of rag half -stuff necessary for the kind of paper required, the whole being then worked in the'hollander into the refined pulp (see later). (b) Calcium Bisulphite (Mitscherlich) or Mag- nesium Bisulphite (Ekman) Process. This process is the one most largely used at the present time, as it gives a cellulose of better quality than the preceding method. The wood is heated under pressure (115 to 130 or 2-5 to 4 atmos.) in large autoclaves lined inside with cement or brickwork with a solution of calcium bisulphite, Ca(S0 3 H) 2 , or magnesium bisul- phite, which dissolves the encrusting matter but does not act on the cellulose 1 ; the liquid is circulated inside the boiler by means of an injector or by leaving a small upper tap slightly open. The bisulphite solution of 4 to 5 Be. (about 30 grms. of SO 2 per litre, approximately one -third being combined with lime) is prepared in very tall wooden towers (that of Harpf being as much as 55 metres high), usually lined with lead and filled with limestone or dolomite (Fig. 392). A current of sulphur dioxide ascends from the bottom to the top of the tower, while the trough, 6 ls supplied by the reservoir, S, at the top, yields a fine spray of water ; the bisulphite solution is collected at the bottom. 1 Lignin is dissolved with remarkable ease by calcium bisulphite, giving a stable soluble compound, the sulphur dioxide in which is neither detectable by iodine, nor capable of being set free by sulphuric acid, nor able to exert reducing action. Sulphurous acid alone does not act so well as the bisulphite, the lime being necessary for the formation of these sulphonic salts and for the neutralisation of the sulphuric acid always formed. FIG. 392. Harpf 's tower has ten gratings (I to X), connected by steps not shown in the figure ; each of these can be charged and attended to independently of the others by means of the door, k. The first six gratings are cleaned every four weeks, but the others far less often. The sulphur dioxide issues from pyrites furnaces into the iron tube, c, and passes down the earthenware pipe b, B B being for convenience of cleaning. The calcium or magnesium bisulphite solution deposits its suspended matter in L and is then discharged into storage tanks. When the whole of the tower is to be washed, the plug, P, of the cistern is raised. To ascertain the completion of the action of the bisulphite on the wood in the digesters, a sample of the liquid is removed now and then and treated in a graduated tube with ammonia ; when the calcium sulphite occupies one-sixteenth of the volume of the sample the heating is stopped, and when this fraction is reduced to one thirty-second the operation is finished and the coloured liquor can be discharged. The wood is sometimes treated with steam before being introduced into the bisulphite boiler. The whole operation, in- cluding charging and discharging, preliminary treatment of the wood and action of the bi- sulphite, lasts 50 to 60 hours. The spent bisulphite liquor is highly coloured and charged with salts, gummy matters, tannin, glucose, pentoses, acetic acid, nitrogenous .compounds, &c., and it is usually forbidden to turn it into watercourses or bottomless wells ; so that it is often purified by precipitation of the sulphite with lime, the calcium sulphite being then reconverted into the bisulphite by sulphur dioxide. Attempts have also been made, but with little success, to evaporate the residual liquor and so obtain adhesive gummy substances utilisable in the preparation of coal briquettes. In a factory with two boilers, each of 120 cu. metres capacity (12 to 15 metres high, 3-5 to 4 metres in diameter, and about 2 cm. thick), each of these is charged with about 200 quintals of wood and 85 cu. metres of bisulphite solution. With a monthly output of 1000 quintals of cellulose, the daily production of spent liqxior is 30 cu. metres, the organic residue amounting to 8 per cent, and the ash to 2 per cent. The rational disposal of these spent liquors is always a serious problem, which still awaits solution ; the attempts made to prepare alcohol from them are mentioned in the note on p. 142. The gases emitted can be deodorised by means of nitrous vapours, which attack the mercaptans. The yield of cellulose varies with the quality of the wood, but is about 50 to 55 per cent. (c) Electric Process. This was proposed by Kellner, and consists in passing through closed receptacles containing the wood a solution of sodium chloride at 126, through which an electric current passes ; the chlorine, hypochlorous acid, and caustic soda act together in the nascent state, dissolving the encrusting substances of the wood and libera- ting the cellulose. This process has not yet been much used. MECHANICAL REFINING OF THE CELLULOSE AND MECHANICAL WOOD- PULP. The mass of wood, more or less finely divided, extracted from the digesters is coarsely defibred in suitable disintegrating machines, and the cellulose and the mechanical pulp, either together or separately, according to the kind of paper required, are introduced into the so-called Hollanders, where they are completely defibred and converted into a very fine pulp ; bleaching with calcium hypochlorite and the subsequent washing are also carried out in the hollanders, as is the addition of dressing, colour, size, resin, alum, &c., necessary for the desired paper. The hollander beating machine consists of a large, oblong wooden or, better, cement vessel (A, Figs. 393 and 394), in the middle of which is a vertical, longitudinal partition, B, which does not extend to the ends of the vessel. In one part of the vessel is a large revolving drum, D, furnished at its periphery with a number of cutters which circulate the 'water containing the cellulose or mechanical pulp. The bottom of this part of the vessel is in the form of a ridge (PR, Fig. 394), and at a point, F, on one of the slopes are fitted cutters ; the drum can be moved up or down by means of the lever, HG, and the distance between its cutters and those at F thus adjusted as required. The movement of the water produced by the rotation of the drum causes almost the whole of the cellulose and pulp to pass between the fixed and revolving cutters, and after some time the woody fibres swim sepa- rately in the water. As the process goes on, the knives are gradually brought closer together until the desired degree of fineness is attained. The mass passes up the plane, P, down the plane, R, round the partition, B, again up the plane, P, and so on. n 33 514 ORGANIC CHEMISTRY The washing water can be changed by immersing in the free half of the vessel a fine gauze drum from which the water can be aspirated by means of a pump. This drum is then raised by the chain and pulley, R (Fig. 393), and fresh water introduced into*the vessel. To avoid spurting from the drum, D, it is fitted with a cover, T. In the base of the vessel and in front of the inclined plane is a recess for catching pieces of iron or stone accidentally present in the wood-pulp, the cutters thus being protected from damage. Fig. 193 on p. 237 shows a battery o^hollanders, which are also used for guncotton. )f li FIG. 393. SIZING AND FORMATION OF THE PAPER. The refined pulp in the hollander, containing the different raw materials (rags, wood-pulp, cellulose, &c.) in the requisite proportions, is blued and sized before being transferred to the continuous machines. The blueing is effected by adding, a short time before the end of the be'ating, 500 to 1000 grms. of ultramarine, Prussian blue, or aniline blue ; a little later the size is added, which renders the paper impervious to water and prevents ink from running on it ; if blotting-paper or filter-paper is required, the sizing is omitted. Sizing can be carried out on the finished FIG. 394. paper, but it is usually preferred to add the dressing directly to the finished pulp while this is still suspended in water, since in this way all the fibres become coated with the size without losing the power of adhering, one to the other, to form a homogeneous, felted mass of paper. Animal size was at one time used, but, owing to its ready putrefaction or alteration even while it is being applied, it has been ..almost entirely replaced by resin (colophony) previously rendered soluble (resin soap) by means of caustic soda. With water this soap forms very fine, homogeneous and persistent emulsions, the efficacy of which may be increased by the addition of starch paste (in amount sometimes equal to that of the resin) or of casein dissolved in dilute soda solution. The total dressing added amounts to 2 to 5 per cent, of the dry paper. FORMATION OF THE PAPER 515 In order to precipitate the resin in a fine state of division on the fibres, a solution of aluminium sulphate (or of potash alum) is added to the homogeneous mixture of pulp and resin soap ; as was shown by Wurster, this effects the precipitation of the resin, starch (or casein), and a very small amount of aluminium resinate. Nowadays one-half of the aluminium sulphate is sometimes replaced by the cheaper magnesium sulphate. The so-called loaded papers are obtained by adding, in addition, a considerable quantity (some- times 50 per cent.) of kaolin, barium sulphate, talc, or cal- cium sulphate. The colouring-matters (mineral dyes, lakes, or sub- stantive aniline dyes) are also added directly to the finished pulp, organic dyes being the more commonly used. The lakes are produced by mixing basic dyes with the pulp and then precipitating with tannin solutions ; for direct dyeing, substantive dyes (see later, ""*?"" Colouring - Matters) are em- PIG. 395. ployed. Powdered lakes ob- tained by precipitating either acid aniline dyes with aluminium hydroxide or basic dyes with tannin or tartar emetic may also be used. After all these additions have been made, separation of any of the components from the homogeneous pulp is prevented by conveying the latter into two vats, where it is kept in motion by stirrers, the resultant milk being more or less dense according to the thickness of paper required. Before going to the continuous machine to be converted into paper, the pulp is passed through a purifier (Pig. 395) which removes any clots of fibre still present. This purifier consists of two or three slightly inclined, oscillating plates, perforated with very fine slots ; when the pulp is fed regularly on to these plates, the fine fibres pass through while the lumps are discharged into channels provided for the purpose. The homogeneous pulp collected under the vibrating plates is conveyed to the continuous machine at an almost absolutely regular speed, and on this depends the uniformity in the thickness of the resultant paper ; the pulp regulator or feeder should hence be constructed with great care. If this homogeneous pulp is placed on a very fine sieve, the water passes through, leaving a thin layer of interlaced, adhering fibres which can be removed in the form of a wet sheet. The preparation of the PIG. 396. paper in the continuous machine takes place in a similar manner. The pulp is distributed uniformly on a very fine endless copper gauze after a good proportion of its water has been removed by draining and suction. A cloth then passes the wet sheet to a pair of rolls, which compress it and give it more consistency ; other rolls heated to 130 gradually dry the paper, while others, again, press it and give it a little polish. When it leaves the endless gauze, the paper is sufficiently consistent to be conveyed to the supercalendar (Fig. 396), where it is pressed and polished between several pairs of rolls. Other machines wind it into rolls, cut it, rule it, &c. A large modern continuous machine may cost several thousands of pounds. A general view of such a machine is shown in Fig. 397 ; the two vats of pulp are seen at a, while b represents the circular feeder carrying buckets, c the drum sieve which collects the pulp and passes it as a wet sheet to the metal gauze, d, this transferring it to the cloth at / and 516 ORGANIC CHEMISTRY passing back round the rollers, e, underneath to take up fresh pulp ; g shows the drying rolls and h where the cloth returns, the continuous length of paper being drawn off at i to the winding apparatus. It is not possible here to consider the different kinds of paper now manufactured, or the different pulps required, or the special modern machines devised to meet all the require- ments of the trade, but a few words may be devoted to the testing of paper, 1 the pulp used being recognisable under the microscope by the magnitude and form of the fibres (see Figs. 398 et seq.). As will be shown in the chapter on Textile Fibres, the fibres of paper are corroded and somewhat distorted and resemble the original fibres only in certain characters. The fibres of the white fir are shown in Fig. 398 at A and in transverse section at B ; they are brown and are characterised by the pores arranged in concentric circles. Fig. 399 shows at B altered cotton fibres and at L those of linen. Fig. 400 gives an idea of the microscopical appearance of mechanical wood-pulp of the conifers (fir, pine, &c.) with medullary rays, while Fig. 401 shows chemical pulp from the conifers ; in the latter case, the concentric circular pores are less marked and the fibres more homogeneous. Fig. 402 shows straw cellulose with the very thin parenchymatous cells, a, rounded at the ends, and the superficial toothed cells of the epidermis, o, mixed with the bulk of ordinary elongated and striated fibres. Esparto fibres resemble those of straw to some extent but FIG. 397. are lacking in thin and terminal cells, while the toothed edges are different in nature and are found in smaller cells than in straw ; esparto contains certain isolated fibres having 1 Testing of Paper. The presence of mineral loading is detected by determining, in a platinum crucible, the ash of 1 to 2 grms. of the paper, cut up and dried at 100 to 105 ; non-loaded paper contains 0'4 to 2'5 per cent, of ash. To detect the presence of mechanical wood-pulp, the paper is immersed in an aqueous solution of aniline sulphate, which imparts a golden-yellow colour to the crude wood fibre ; or use may be made of aqueous phloroglucinol faintly acidified with hydrochloric acid, this dyeing the crude wood fibre (mechanical pulp) red. The impermeability or solidity of the sizing is determined by Leonardi's method ; on to the paper, stretched and inclined at 60, a solution containing 1 per cent, of ferric chloride, 1 per cent, of gum arabic, and 0-2 per cent, of phenol is allowed to fall drop by drop so as to form a number of moist strips which are then allowed to dry ; similar strips, crossing the first and perpendicular to them, are next made with a solution containing 1 per cent. of tannin and 0-2 per cent, of phenol ; the formation of a black stain of tannate of iron at the point of intersection indicates bad sizing, absence of stain shows perfect sizing, and stains more or less grey denote more or less good sizing. Resin sizing is recognised by pouring a few drops of ether on to the paper and allowing them to evaporate ; the formation of transparent rings indicates the probable presence of resin. Or a few grms. of the paper may be boiled with absolute alcohol containing a few drops of pure acetic acid, the solution being afterwards poured into distilled water ; if the latter becomes turbid, the presence of resin is certain. To detect animal sizing, a few grms. of the paper are boiled with a very small quantity of distilled water, the liquid being filtered, highly concentrated and treated with a solution of tannin ; if size is present, whitish grey flocks are formed, which, when observed under the microscope in contact with a dilute solution of iodine in potassium iodide, are seen to be coloured brown, while if starch is present this is coloured blue ; the test for starch may be made directly on the paper itself. The presence of free mineral acid is ascertained by boiling the paper in a little distilled water and noting if the solution turns Congo-red paper blue or black. For the microscopical examination (see Figs. 398-402), the fibres are4iberated as follows : 3 to 5 sq. cm. of the paper are boiled and vigorously shaken for two minutes with 3 to 4 per cent, caustic soda solution, the pulp thus formed being poured on to a very fine metal sieve and washed well with tepid water. The fibres are then tested microchemically with solutions containing (1) 6 parts of iodine, potassium iodide, 10 parts of glycerol, and 90 of water, and (2) 100 part? zinc chloride, 10-5 of potassium iodide, 0-5 of iodine, and 75 of water, the clear liquid being, in this case, decanted from the precipitate formed ; linen, hemp, and cotton are coloured pale to dark brown by solution (1), the thin fibres remaining almost colourless, while with solution (2) a more or less intense wine-red coloration is obtained. An alcoholic solution of phloroglucinol containing hydrochloric acid does not colour pure cellulose but reddena PAPER STATISTICS 517 the form of teeth or elongated pears. Spain exported more than 90,000 tons of esparto in 1872 and about 46,000 in 1900. Algeria now exports' 80,000 tons, Tunis 30,000, Tripoli 75,000, and Morocco 4000. Algeria contains 5,000,000 hectares under esparto. England imports about 200,000 tons of esparto per annum. STATISTICS. Books and reviews often contain contradictory and fantastic statistics concerning the output of paper. According to the most trustworthy data, the world's production of paper and pasteboard in 1906 amounted to about 8,000,000 tons, and that of cellulose in 1908 was estimated at 1,600,000 tons of the value of 16,000,000. In 1904 the United States produced 2,000,000 tons of mechanical pulp and 4,000,000 tons of paper and pasteboard, worth 32,000,000 ; in 1860 the output was 200,000 tons, of the value of 5,200,000. The wood converted into mechanical pulp represented a value of FIG. 401. 5,600,000 in 1908, and more than 6,800,000 (from 253 factories) in 1909. The value of the exports was 1,480,000 in 1904 but is rapidly increasing, and already exceeds 4,000,000 ; in 1904 there were 1200 paper and mechanical pulp factories, with a total capital of about 40,000,000, one-half of this representing machinery. In 1911 the United States imported 263,000 tons of mechanical wood-pulp, 213.000 tons of cellulose (1,296,000), and 86,000 tons of bleached pulp (737,800). impure cellulose, the presence of wood-pulp (i.e. impure cellulose) in paper being hence detectable in this manner. Further, aniline sulphate or naphthylamine hydrochloride colours impure cellulose yellow, but does not alter pure cellulose. The bursting strain of paper, called also the degree of elasticity, is determined in the directions of the length and breadth by means of suitable dynamometric apparatus, the elongation which occurs before rupture being expressed as a percentage of the length (this varies from 1-5 to 4 per cent, for different papers). The breaking length expresses the length of a uniform strip of paper which would tear under its own weight if suspended from one end : if a strip 10 cm. wide of paper of which 1 sq. metre weighs 70 grms. breaks under a load of 3500 grms. S500 the breaking length is X 1000 = 5000. The resistance to folding is determined roughly by crushing and rubbing an irregular ball of the paper between the hands ; when different papers are compared in this way, that with the least number of creases is the best. 518 ORGANIC CHEMISTRY Germany in 1899 contained 900 paper and pasteboard factories, 72 wood-cellulose factories, 30 straw-cellulose factories, and 600 mechanical wood-pulp factories, using a total of 125,000 hydraulic horse-power and 75,000 steam horse-power, employing 65,000 operatives, and producing 270,000 tons of cellulose (550,000 tons in 1909), 300,000 tons of mechanical pulp, and 800,000 ton s of paper and pasteboard. In 1 884 the output of paper and paste-board was 200,000, and in 1904 more than 1,200,000 tons of the value of 12,600,000. The imports in 1904 were 24,000 tons of paper and pasteboard, the same quantity of mechanical pulp, and about 47,000 tons (32,550 tons in 1909) of cellulose ; the exports were 64,000 tons of cellulose in 1904 (147,088 tons in 1909), 6000 tons of mechanical pulp, and 250,000 tons of paper and pasteboard. In 1908 Germany imported 833,480 tons of wood for paper, and in 1909 about 1,653,000 tons, exporting about 40,000 tons. The price of wood- cellulose in Germany was 42s. per quintal in 1852, and is to-day less than 16s. In Norway the first manufactory of mechanical pulp was erected in 1870 and the first of cellulose in 1880, and in 1905 the paper industry occupied 8000 workpeople, the output being 100,000 tons of paper. In recent years this industry has advanced considerably, 27 factories now possessing a total of 60 continuous machines and the production of paper in 1910 being 150,000 tons (of the .value of 1,520,000), nine-tenths of this being for export. In 1891 Sweden possessed 40 paper factories, occupying 3000 workpeople and producing 36,000 tons of paper, worth 480,000. In 1906 the output of paper was 209,000 tons (2,400,000), and in 1907, 225,000 tons (2,560,000). Sweden produced mechanical wood- pulp to the value of 3,080,000 in 1906 and 3,760,000 in 1907, part of it being exported. France produced about 75,000 tons of paper in 1860, about 60,000 operatives being employed in the industry in 1901 ; in 1904 the output was almost 450,000 tons. In Russia the consumption of paper is continually increasing, but the amount produced is almost stationary : 163,800 tons in 1897, 177,000 in 1900, and 205,000 (7,600,000) in 1906 ; so that Russia imports a considerable quantity of paper, even from Japan and China, but more especially from Finland, whose exports of paper to Russia have increased sixfold during the last ten years. Finland exported nearly 43,000 tons of mechanical pulp and about 13,000 tons of cellu- lose in 1906, largely to Russia. Austria produced about 350,000 tons of paper in 1904. England in 1860 produced about 100,000 tons of paper and now produces rather less than Germany ; in 1909 the imports included, besides rags (q.v.), 197,501 tons of esparto, &c., of the value of 720,000, and 749,740 tons of mechanical wood-pulp and cellulose, worth 3,480,000. The imports of raw materials for paper-making into England were valued at 4,741,230 and the exports at 820,730 in 1911 ; the imports of paper of different kinds amounted to 6,574,500 and the exports to 3,311,867 in 1911. In 1906 Spain produced more than 35,000 tons of paper. The paper industry in Italy has increased very considerably in recent years, the produc- tion being 60,000 tons in 1876 and almost 70,000 in 1886, the importation of cellulose being as follows : 1800 tons in 1886 ; 13,600 in 1896 ; 24,300 in 1901 ; 42,000 in 1905 ; 46,700 in 1907 ; 54,000 in 1908 ; and 63,100 tons, of the value of 706,400, in 1910. The production of cellulose in Italy is very small, there being but three factories. In 1896 28 mechanical pulp factories produced 10,000 tons of the pulp, 4200 tons of which were imported in the same year and 8741 tons (62,936) in 1910. The total production of paper and cardboard in Italy in 1907 was about 200,000 tons ; 30,000 tons of this was used for daily papers weighing 45 grms. per sq. metre (24s. to 28s. par quintal). The Customs duty on paper in Italy varies from about 6s. to 36s. per quintal for different qualities. The following numbers represent the mean annual consumption of paper in kilos per inhabitant for various countries, these being regarded as a-jough indication of progress : x About 75,000 new books are published per annum in the whole world, these requiring 25,000 tons of mechani- cal pulp alone. In addition, about 30,000 periodicals are published with a total circulation of nearly 11,000,000,000 copies and for these 1000 tons of mechanical pulp are consumed per day. Of the total output of paper, 32 per cent, is for ordinary printing, 10 per cent, consists of fine paper and writing paper, 10 per cent, of brown paper and cardboard, 6-3 per cent, of fine cellulose and rag paper for fine printing; 5 per cent, of straw paper and card, 3 per cent, of paper for placards, &c., 3 per cent, of wall-paper, 0-6 per cent. CONSUMPTIONOF PAPER 519 United States, 19-3 ; England, 17-2 ; Germany, 14 ; France, 11-5 ; Austria, 9-5 ; Italy, 7-5 ; Spain, 2-5 ; Russia, 2-3 ; Servia, 0-6 ; China, 0-6 ; India, 0-13. of drawing paper, 0-5 per cent, of silk paper, cigarette paper, and paper for making flowers ; 0-4 per cent, of blotting- and filter-paper, <$rc. Although the consumption of paper has increased to an extent that would have been incredible a few years ago, yet the day is far distant when a scarcity of raw material will be experienced. Canada alone, with its 322,000,000 hectares of forest land can supply the whole world for several centuries. Of other reserves of forest the most important are those of the United States, 200,000,000 hectares ; Russia, 184,000,000 ; Queensland, 86,000,000 ; Siberia, 38,000,000 ; British India and Burmah, 26.000,000 ; Finland, Sweden, and Japan (excluding Formosa and Hokkaido), 20,000,000 each ; Germany, 17,000,000 ; Austria and France, 10,000,000 each ; Hungary, Croatia, and Slavonia, 9,000,000 ; New Zealand, 8,000,000 ; Asiatic Turkey, 7,000,000 ; Norway, 6,000,000 ; Hokkaido (Japan), 6,000,000 ; Italy, 4,500,000, &c. In Burmah and elsewhere there are immense tracts of bamboo, which will one day be utilised for the manufacture of paper. It cannot, however, be denied that an immense amount of wood is used for building purposes, and in Italy, for instance, many of the forests have been destroyed, so that the imports of wood, which in that country amounted to 840,000 in 1871, increased to 2,000,000 in 1900, to 2,840,000 in 1905, and to still greater extents (mostly from Austria-Hungary and America) in recent years (see vol. i, p. 204). PART III. CYCLIC COMPOUNDS THE aliphatic series contains various groups of closed-chain compounds (e.g. lactones, uric acid derivatives, anhydrides of dibasic acids), which are readily opened by simple reactions giving ordinary open-chain compounds of the fatty series. Numerous substances are, however, known containing a closed-chain nucleus which is composed of 3, 4, 5, or more commonly 6, carbon atoms united in a special manner and is resistant to the most energetic reagents. These compounds form the important group of aromatic compounds. Other groups of cyclic substances are also known with nuclei composed, not of carbon atoms alone, but of several elements, e.g. pyridine, C 5 H 5 N, in which the nucleus contains 5 carbon atoms and 1 nitrogen atom ; pyrrole, C 4 H 5 N, with C 4 and N in the nucleus ; furan, C 4 H 4 0, with a C 4 nucleus ; thiophene, C 4 H 4 S, with a C 4 S nucleus ; pyrazole, C 3 H 4 N 2 , with the nucleus C 3 N 2 , &c. These compounds are called heterocyclic. There are also many substances derived from more complex nuclei formed by the condensation of two or more of the nuclei mentioned above, e.g. naph- thalene, C 10 H 8 , in which are condensed two benzene nuclei held together by two carbon atoms common to the two nuclei, and quinoline, with a nucleus analogous to that of naphthalene but composed of one benzene and one pyridine nucleus. AA. ISOCYCLIC COMPOUNDS These contain 1 or several homogeneous carbon atom rings, and can be sub- divided, according to the type of linking, into (1) Polymethylene Compounds, which contain singly linked carbon atoms and are less resistant to chemical reagents than (2) Benzene Derivatives, where the carbon atoms are linked very differently (see later}. Compounds of the first group approach those of the aliphatic group in their chemical properties and are hence intermediate to methane and benzene derivatives. I. CYCLOPARAFFINS AND CYCLO-OLEFINES OR POLYMETHYLENE COMPOUNDS xCH 2 TRIMETHYLENE (Cyclopropane), CH 2 \ | , is obtained by the action of sodium ^CU 2 on ay-dibromopropane, CH 2 Br CH 2 CH 2 Br, the bromine being eliminated and the chain closed. It is a gas which liquefies at a pressure of 5 to 6 atmos. and combines very slowly with bromine or hydriodic acid giving open-chain compounds, so that it is easily dis- tinguished from propylene CH 2 : CH>CH 3 . Its heat of combustion is much greater than that of propylene, into which it is partially converted at 400. Its derivatives are obtained from ethylene bromide by means of the ethyl malonate synthesis (see p. 308). CH 2 \ xCO2-H Trimethylenedicarboxylic Acid, | /C\ was obtained by Perkin by the CH/ X CO 2 H interaction of ethylene bromide and ethyl sodiomalonate. 530 AROMATIC COMPOUNDS 521 TETRAMETHYLENE (Cyclobutane) is not known in the free state, but derivatives of it are obtainable by syntheses similar to those used for trimethylene compounds. vx.ri 2 v*il 2 \ PENTAMETHYLENE (Cydopentane), \ */>CH 2 , is a liquid boiling at 50; vs.tl 2 ' v'.rio its derivatives are prepared by the ethyl malonate synthesis. According to Baeyer's tension hypothesis (see p. 88 and Fig. 247, p. 306), it is easy to understand why pentamethylene is the most stable of the preceding compounds, a ring of five carbon atoms being the only one which can be formed without tension of the linkings. Indeed, while trimethylene combines with Br or HI with rupture of the ring, penta- methylene does not unite with bromine and resists the action of nitric or sulphuric acid like a saturated hydrocarbon, the properties of saturated open- and closed-chain compounds hence differing but little. KETOPENTAMETHYLENE (Cyclopentanone), C 5 H 8 O, is obtained by the dry distilla- tion of calcium adipate : CH 2 .CH 2 .CO(\ CH 2 .CH 2X \Ca = CaC0 3 + | )CO ; CH 2 -CH 2 .CO(K CH 2 .CH/ by reduction and subsequent treatment with HI it gives pentamethylene, whilst oxidising agents convert it into glutaric acid, these reactions proving its constitution. Ketohexa- methylene is obtained similarly by distilling Calcium Pimelate, C 7 H 10 4 Ca, and higher homologues by distilling the corresponding calcium salts of higher dibasic acids ; Calcium Suberate, CgH^O^Ca, for example, yields Ketoheptamethylene (suberone). The yield diminishes with increase of the number of carbon atoms. CH : CH. CYCLOPENTADIENE, | yCH 2 , is a liquid boiling at 41, and is found in CH : CH/ the first distillate of crude benzene and also in illuminating gas ; it combines with iodine and with hydrogen sulphide. The presence of two double linkings in the nucleus is deduced from the fixation of four atoms of halogen. The two hydrogen atoms of the CH 2 readily react, e.g. with acetone, giving intensely red hydrocarbons : CH : CH\ CHg\ CH : CHx >CH 2 + )CO = H 2 +| / CH.-CH/ CH 3 < CH:CH X CH 3 this compound is known as dimethylfulvene, fulvene being an isomeride of benzene of the CH : CH^ structure | /C : CH 2 . CH : CH/ II. BENZENE DERIVATIVES OR AROMATIC COMPOUNDS It was observed by several chemists about the middle of last century that a whole series of compounds, mostly aromatic in nature, besides exhibiting certain common physical and chemical characters, showed on analysis pro- portions of hydrogen very low in comparison with those of carbon and also very low compared with those of hydrogen in saturated or unsaturated com- pounds of the methane series, e.g. C n H 2n -j- 2 , C n H 2 n, CH 2W -2' & c - In general the hydrocarbons of these substances correspond with the fundamental formula, C n H 2M _ 6 , and the various transformations of the aromatic substances often yield Benzene, C 6 H 6 , from which they can again be prepared. If the constitutional formula of benzene were an open -chain one, it would be necessary to assume the presence of double or triple linkings between carbon and carbon which would lead to ready addition of bromine and to ready oxidation. But these reactions do not occur, and the great stability of the com- pounds of this group, and of benzene in particular, can be explained only by the existence of a stable nucleus of carbon atoms, probably joined in the form of a closed ring. It was found later that benzene forms only one monosubstituted product (nitrobenzene, bromobenzene, &c.), and that all the hydrogen atoms of benzene exist under similar 522 ORGANIC CHEMISTRY conditions ; three isomeric disubstituted products (e.g. dinitro- or dibromo -benzene) are, however, known. With the empirical formula C 6 H 6 correspond the three rational formulae : (a) C 4 (CH 3 ) 2 , (/3) C 3 (CH 2 ) 3 , and (y) (CH) 6 . Formulae (a) and (/3) would give only two isomeric disubsti- tuted products, whilst in the case of (y), if the six CH groups were joined in the form not of an open chain but of a closed ring, the six hydrogen atoms would be under the same conditions, and the formation of a single monosubstituted product and of three iscmeric disubstituted products would be explained. It was Kekule who, in 1865, first advanced the ingenious hypothesis that the funda- mental compound of aromatic substances is benzene, the constitutional formula of which must be represented as a closed, hexagonal chain of carbon atoms united alternately by single and double linkings, the fourth valency of each carbon atom being united to a hydrogen atom. Such an arrangement is figured in the scheme H C HC/iy= HC CH G H or, if the six carbon atoms are represented by tetrahedra (see p. 18 et seg.), in the diagram shown in Fig. 403. The carbon atoms combined with the substituents in the three disub- stituted derivatives would then be : (a) 1 and 2 (ortho- derivatives), (b) 1 and 3 (meto-derivatives), and (c) 1 and 4 (para-derivatives) ; the 1 : 5- and 1 : 6- com- pounds would be identical with the 1 : 3- and 1 : 2- compounds respectively. For the sake of shortness, the terms ortho-, meta-, and para- are contracted to o-, m- and p-, these being prefixed to the names of the compounds. FIG. 403. The constitutional formula given for ben- zene by Kekule and also those of Claus (1867), Baeyer (1868), Korner (1869), and Ladenburg (1870) would seem to indicate the possible existence of 2 ortho-substituted derivatives, since the 1 and 2 carbon atoms are joined by a double linking and numbers 1 and 6 by a single linking. Hence Claus and Korner pro- posed the hexagonal formula with the fourth valencies of the carbon atoms joined diagonally (para -linking) (Fig. 404, A), while Ladenburg preferred the prismatic formula (Fig. 404, B l} B 2 , and B 3 ), and Armstrong and Baeyer the centric formula, with the fourth valencies in a latent (or potential) state and directed towards the centre (Fig. 404, C) ; see also Fig. 405. In order to obtain a better interpretation of the formation of the disubstituted iso- merides of benzene, Kekule (1872) developed his theory further on the assumption that the linkings between the carbon atoms are to be regarded as vibrations, so that carbon atoms 2 and 6 of the Kekule formula are in identical conditions. These oscillations would explain why benzene does not unite readily with halogens or ozone (see p. 88 ; also Ann. Soc. Chim., Milan, 1907, p. 116, and Berichte der dewt. chem. GeselL, 1908, p. 2782) or give Baeyer's permanganate reaction (see p. 88), thus behaving almost like a saturated compound. But even Kekule's oscillatory formula does not explain completely the optical and thermal behaviour of the aromatic compounds or the interesting results obtained by Baeyer on the hydrogenated derivatives of benzene subsequently to 1886. Indeed, when two or four hydrogen atoms are added to benzene so as to form dihydro- or tetrahydro- benzene, the latter are found to be quite different from true aromatic compounds and to STRUCTURE OF BENZENE 523 resemble olefine compounds ; it must, then, be assumed that where the hydrogen has not been added, true double linkings are formed capable of combining with halogens or ozone and of giving Baeyer's permanganate reaction. Baeyer's centric formula would harmonise with this behaviour, since each of the valencies directed towards the centre is kept in equilibrium with all the others, stability being thus conferred on the molecule ; if, then, two or four of the central valencies are used in the addition of hydrogen or other groups, the remaining central valencies become true, olefinic, double linkings. There are, however, aromatic compounds, especially those with several condensed benzene nuclei, with which Baeyer's centric formula alone cannot be assumed. In 1899 FIG. 404. Thiele attempted to harmonise all the chemical and physical phenomena observed with benzene and its derivatives on the assumption that when two carbon atoms are united by a double linking the two affinities are not completely utilised, parts of the unsatisfied valencies (partial valencies) remaining. These are regarded as bringing about addi- tion processes, and are represented by dotted lines, e.g. C = C, C = C C = C, &c. But when, as in the ^ latter formula, a conjugated system of double bonds is present, the addition of hydrogen, halogens, &c., occurs only at the two extreme carbon atoms, the partial valencies of the two middle atoms forming a new inactive double bond, C = C C = C ; after the addition at the extreme carbon atoms, the central inactive bond becomes C - C= C - C active again, the constitution then being, H H In Kekule's benzene formula, we may assume the existence of three conjugated double H (?\ HC CH bonds with three inactive bonds, thus, || |) ; it would then be clear why ben- HC CH FIG. 405. zene,, being without partial valencies, would not readily form additive products, and why, when even a single inactive double bond is broken down, true active olefinic double linkings would appear (see Theory of Double Linking, Note on p. 88). A plausible explanation of the constitution of benzene is also arrived at by means of the ideas of motochemistry, according to which double or single linkings are represented by double or single vibrations or blows per unit of time (E. Molinari, Gazzetta Chimica Italiana, 1893, vol. ii, p. 4.1, and Journal fur praktische Chemie, 1893, p. 113). ISOMERISM IN BENZENE DERIVATIVES It has been seen already that when one of the hydrogen atoms of benzene is replaced by a halogen or an organic residue, the same monosubstituted compound is always obtained, no matter at what point of the molecule the substitution occurs. If two substituent groups, either similar or different, are introduced, three disubstituted derivatives are obtainable. If the benzene 524 ORGANIC CHEMISTRY molecule is represented simply by a hexagon, each angle of which indicates a carbon atom united with a hydrogen atom, replacement of the latter by another atom or group (x, y, z, &c.) may be shown by placing the symbol of the substituent at the" angle of the hexagon. With disubstituted com- pounds, if one group is assumed to occupy the position 1, the other may go to either 2 or 6 (or tho -position), 3 or 5 (meta), or 4 (para). Benzene Monosubstitutcd Ortho-compounds Meta-derivatives Para-derivative benzene 1 : 2- and 1:6- 1:3- and 1 : 5- (identical) (identical) With the trisubstituted derivatives, three isomerides are possible when the three substituents are similar (1:2:3- or vicinal, identical with 1:6:5-; the symmetrical, 1:3:5-, identical with 2:4:6-; and finally, the unsym- metrical, 1:2:4-, identical with 1:5:4-): Vicinal (i>) Symmetrical (s-) Unsymmetrical (as-) When one of the three substituents is different from the remaining two, six isomerides are possible : 'A /* Vicinal Unsymmetrical Symmetrical With four similar substituent groups, it will readily be seen that three isomerides are possible. The number of isomerides may be further increased in cases where one or more of the substituents form lateral chains capable of isomerism, e.g. saturated hydrocarbon or unsaturated alcohol or acid groups ; in these compounds, further replacement of hydrogen may occur either in the benzene nucleus or in the side-chain, fresh cases of isomerism being thus possible. It was Korner (1869-1874) who first showed how it is possible to deter- mine experimentally the 'positions of the various substituent groups in the benzene nucleus ; examples will be given later. GENERAL CHARACTERS OF BENZENE DERIVATIVES While the saturated hydrocarbons of the aliphatic series offer considerable resistance to oxidising agents and to concentrated sulphuric or nitric acid, those of the aromatic series readily give nitro -derivatives with nitric acid, and sulphonic derivatives, having an acid character, with sulphuric acid : C 6 H 6 + HN0 3 = H 2 + C 6 H 5 -N0 2 (nitrobenzene) ; C 6 H 6 + H 2 S0 4 = H 2 + C 6 H 5 -S0 3 H (benzenesulphonic acid). In the latter, the sulphur is united directly to a carbon atom of the benzene nucleus, this being confirmed by the fact that benzenesulphonic acid is also obtained by the action of oxidising agents on thiophenol, C 6 H 5 -SH, in which the sulphur is known to be joined to carbon. PREPARATION OF BENZENE, ETC. 525 Oxidation of aromatic hydrocarbons containing side-chains leads to the replacement of the latter by carboxyl groups, C0 2 H, the benzene nucleus remaining unchanged ; in this way the various aromatic acids are obtained : )CH 3 + 30 = H 2 + >C0 2 H ; Toluene Benzoic acid CH 2 'CH 3 C0 2 H + 90 = 3H 2 + CH 8 Ethyltoluene Isophthalic acid The halogen substitution derivatives, which are readily obtained by the direct action of the halogens, have less reactive properties than the halogen compounds of the aliphatic series and are more resistant to substitution. The hydroxyl-derivatives (e.g. phenol, C 6 H 5 -OH) are more decidedly acid in character than the alcohols of the fatty series, the phenyl group, C 6 H 5 , for example, being more negative than the ethyl group ; their resistance to oxidising agents is similar to that of the tertiary alcohols, to which they are analogous in constitution, the grouped- OH being present in both cases. The amino-derivatives, which are readily obtainable by reducing the nitro- derivatives (CgHg-NOa + 6H = 2H 2 + C 6 H 5 -NH 2 , aniline) with inter- mediate formation of azo-compounds (q.v.), are easily converted by the action of nitrous acid into diazo-compounds ; the latter are formed only seldom and with difficulty in the case of aliphatic compounds. In their last investigations Korner and Contardi (1908) show how, with the substitution products of benzene, the formation of one isomeride rather than another sometimes depends on minimal differences in the physical conditions under which the reactions take place. Thus, in the nitration of aniline or of halogenated derivatives, a very slight difference in the concen- tration (even in the second decimal place of the specific gravity) is sufficient to alter the yield very considerably or even to give entirely different products. FORMATION OF BENZENE AND ITS DERIVATIVES When vapours of aliphatic compounds are passed through red-hot tubes, the products formed contain aromatic compounds. At a red heat acetylene gives benzene (the reverse reaction is also possible) : 3C 2 H 2 = C 6 H 6 . When allylene, C 3 H 4 , is distilled with dilute sulphuric acid, mesitylene, C 6 H 3 (CH 3 ) 3 (1:3: 5), is obtained, while under similar conditions crotonylene, C 4 H 6 , forms hexamethylbenzene, C 6 (CH 3 ) 6 . In presence of concentrated sulphuric acid, several aliphatic ketones undergo condensation to aromatic hydrocarbons ; thus, acetone forms 1:3:5- trimethylbenzene, 3C 3 H 6 = 3H 2 + C 6 H 3 (CH 3 ) 3 . Acetoacetaldehyde, CH 3 -CO-CH 2 -CHO, when liberated from its sodium derivative, is transformed immediately into triacetylbenzene, C 6 H 3 (COCH 3 ) 3 . Various aromatic compounds can also be obtained by the action of sodium on ethyl bromoacetoacetate or ethyl succinate, by heating ethyl sodiomalonate and by certain other syntheses. From the tar obtained by distilling coal, wood, or lignite, many aromatic compounds can be separated : 5 to 10 per cent, of naphthalene, 1 to 1-5 per cent, of benzene and toluene, besides quinoline, anthracene, &c. Benzoic and salicylic acids, bitter almond oil, &c., occur naturally in the vegetable kingdom. 526 ORGANICCHEMISTRY A. AROMATIC HYDROCARBONS Those with saturated side-chains are colourless, . refractive liquids of characteristic odour, insoluble in water, but extremely soluble in ether or absolute alcohol ; they are lighter than water (0-830 to 0-806). General Methods of Preparation. (1) Alkyl chlorides and aromatic hydro carbons in presence of aluminium chloride give mono- and poly-sub- stituted hydrocarbons, which can be separated by fractional distillation : C 6 H 6 + CH 3 C1 - HC1 + C 6 H 5 -CH 3 (Friedel and Craft's synthesis) ; inter- mediate aluminium compounds are first formed. Ferric chloride, zinc chloride, or zinc turnings act in the same way as aluminium chloride. The latter salt also brings about the decomposition of the higher hydrocarbons into more simple ones. (2) In presence of sodium, monobromo-substitution derivatives of aromatic hydrocarbons and alkyl bromide or iodide give higher aromatic hydrocarbons (Fittig's synthesis, analogous to that of Wurtz for the aliphatic series) : C 6 H 5 Br + C 2 H 5 I + Na 2 == NaBr + Nal + C 6 H 5 - C 2 H 5 . (3) Distillation of calcium salts with soda lime (analogous to the synthesis of aliphatic hydrocarbons) : (C 6 H 5 C0 2 ) 2 Ca + Ca(OH) 2 = 2CaC0 3 + 2C 6 H 6 . Calcium benzoatc (4) Aromatic sulphonic derivatives give the hydrocarbons when heatsd with sulphuric or hydrochloric acid, best in presence of steam : C 6 H 5 -S0 3 H + H 2 = H 2 S0 4 + C 6 H 6 . On this reaction is based the methcd used for separating aromatic hydrocarbons from those of the aliphatic series, the former with concentrated sulphuric acid giving soluble and the latter (paraffins) insoluble sulphonic acids. (5) When an aromatic hydrocarbon is dissolved in an alcohol in presence of zinc chloride at about 300, water separates and a higher hvdrocarbon is formed : C 6 H 6 + C 5 H n OH = H 2 + C 6 H 5 -C 5 H n . DISTILLATION OF TAR The cheapest and most abundant hydrocarbons used as raw material for the prepara- tion of large numbers of important aromatic compounds (from artificial perfumes to aniline dyes) are obtained by the distillation of tar. While at one time this product con- stituted an unpleasant and inconvenient residue of the illuminating gas industry (see pp. 36-38, 52, and 81), it is now so much in demand by large manufacturers of chemical products that it is sometimes very scarce, and attention has been turned to the utilisation of the tar produced in metallurgical coke factories, this having been formerly discarded. Westphalian coal gives, on an average, 2-5 per cent, of tar, that of Saahr as much as 4 per cent., and that of Silesia even more than 4 per cent. The first attempt to utilise tar dates back to 1834 when, in a works at Manchester, it was distilled out of contact with air in primitive retorts, the liquid products being collected and the residual pitch employed for making black varnish. Bethell subsequently patented a process for obtaining from tar creosote oil for the impregnation and preservation of wood. Still later the more volatile products of the distillation of tar were used both as an illuminant and as a cleaning liquid. Nitrobenzene was then prepared from it to replace essence of mirbane. But it became possible to develop an industry for the regular utilisation of tar only after the wonderful discovery by Perkin (1856), who prepared synthetically the first arti- ficial coal-tar dye, thus laying the foundation of one of the most important industries for which the nineteenth century is famous, AROMATIC HYDROCARBONS 527 COMMONEST AROMATIC HYDROCARBONS WITH A SINGLE BENZENE NUCLEUS Name national formula Position of substituents Melting- point Boiling- point Specific gravity C 6 H, Benzene + 5-4 +80-4 0-874 (^) C,H 8 Toluene or methyl- benzene C.H 6 -CH 3 liquid 110 0-869 (16) C 8 H 10 o-Xylenc=o-dimcthyl- benzane C 6 H 4 (CH 3 ) 2 1:2 - 28 142 0-893 (0) m-Xylene = m-dimethyl- benzsne 1:3 - 53 139 0-881 (0) p-Xylene = p-dimethyl- benzene M 1:4 + 13 138 0-880 (0) Ethylbenzene C 6 TI d -C 2 H 5 liquid 136 0-883 (0) C 9 H 12 Hemiinellithcne = tri- methylbcnzene (v) C 6 H 3 (CII 3 ), 1:2:3 ,, 175 Pseudocumene = tri- methylbenzsne (ns) 1:2:4 169-5 0-895 (0) Mesitylene = trimethyl- benzene (s) 1:3:5 tj 165 0-865 (14) n-Propylbenzene '. C,H 5 -C 3 H 7 ,, 159 0-867 (14) Isopropylbenzene = cuniene M ,, 153 0-866 (16) C 10 H 14 Prehnitene = tetra- methylbenzene C 6 H 2 (CH 3 ) 4 1:2:3:4 4 204 Isodurene = tetra- methylbenzene (as) M 1:2:3:5 liquid 195 Durene = tetramethyl- benzene (s) - 1:2:4:5 + 80 192 m-Cymene methyl- isopropylbenzene . C 8 H 4 -CH 3 (C 3 n,) 1:3 liquid 175 0-862 (20) Cymeue = methyliso- propylbenzcne ,, 1 :4 175 0-856 (20) n-Butylbenzene . C 6 H 6 -C 4 H 9 180 0-864 (15) sec. Butylbenzene Jy 175 0-867 (15) Isobutylbenzene n 171 0-371 (15) tert. Butylbenzene .. 167 0-871 (15) C n H 18 Pentamethylbenzene C 8 H(CH 3 ) 5 1:2:3:4:5 + 51-5 231 0-847 (104 n-Amylbenzene . C 9 H 6 -C 5 Hu . liquid 202 0-860 (22) Isoamylbenzene . ,, ,, 194 0-885 (18) C 12 H 18 Hexamethylbenz.iue C 6 (CH 3 ) 8 1:2:3:4:5:6 + 166 265 Ci 3 H 20 n-Heptylbenzene O IT P IT liquid 109 (10 mm.) C 14 H 22 n-Octylbenzene . C 6 H 8 -C 8 H 17 - 7 263 0-852 (14) Ci 6 H 28 Pentaethylbenzene C 8 H(C 2 H 5 ) 5 1:2:3:4:5 liquid 277 0-896 (20) C 18 H 30 Hexaethylbenzene C6(C 8 H 6 ) 8 1:2:3:4:5:6 + 129 298 0-830. (130) C 22 H 38 Cetylbenzene CeHj-CmH.,, + 27 230 (15 mm.) 0-857 (27) C 24 H 42 Octadecylbenzene C 8 H S -C 18 H 37 + 36 249 (15 mm.) Hexapropylbenzene C 6 (C 3 H 7 ) 6 1:2:3:4:5:6 + 118 MM C 25 H 44 Trimethylcetylbenzene C fl H 2 (CH 3 ) 3 (C 18 H 33 ) 1:3:5:2 + 40 258 (15 mm.) 0-845 (40) Numerous industries then arose for the more complete and more rational utilisation of tar for employing to the best advantage the various products of its fractional distillation. Since that time a continuous series of mechanical improvements in the plant and chemical ones in the processes have been introduced. Improvements in the coke furnaces to admit of the collection of the whole of the products of distillation and of the rational recovery of the heat have been dealt with in vol. i (p. 366). After separation from the ammoniacal liquors of gas manufacture (by ceritrifugation), tar forms a dense, almost viscous, blackish (since it contains 10 to 30 per cent, of suspended carbon particles) liquid of sp. gr. 1-1 to 1-3. It contains many varied acid, basic, and indifferent products ; the first can be extracted by agitating with aqueous alkali solution, the second with acids, while the neutral compounds, consisting principally of aromatic hydrocarbons, form the residue. The composition of tar varies, however, with the nature of the coal, the type of furnace, and the temperature of distillation. It seems that tar contains at least 300 different substances, of which 150 have been established either directly or indirectly and 90 have been isolated with certainty and studied, although only four have wide application in the pure state : benzene, phenol, toluene, and naphthalene. Only to a small extent is tar used as it is : for varnishes, coal briquettes, bitmnenised 528 ORGANIC CHEMISTRY paper, lampblack, 1 treating roads to render them less dusty, &c. But for such purposes the residue from the distillation of tar can also be used. A little tar is used in preparing the basic lining of Bessemer converters for the manufac- ture of steel. Nowadays, however, tar is mostly subjected to distillation for the extraction of the following products : (1) Indifferent substances, in which benzene hydrocarbons predominate (benzene, toluene, xylene, tri- and tetra -methyl benzene, and, to a still greater extent, naphthalene, anthracene, &c.), those of the methane series being small in amount (these occur abundantly in the distillation products of lignite-tar, see pp. 81 and 82). Small quan- tities of nitrogen compounds occur, such as acetonitrile, benzonitrile, carbazole and pyrrole derivatives, and also traces of carbon disulphide, thiophene, cumarone, &c. ; (2) Acid substances, among which phenol (carbolic acid), cresol, xylenol, and the naphthols abound. (3) Basic substances, which are found in small amount and contain small proportions of pyridine and quinoline compounds and a trace of aniline. Wood-tar is of less value than coal-tar ; its most important constituents are those soluble in alkali, these consisting of methyl ethers of polyhydric phenols (pyrocatechol, pyrogallol and homologues, forming creosote oil), which are used for making guaiacol. Wood-tar is distilled in a vacuum, the gases which do not condense being utilised for power or heating purposes, as they have a calorific value of 6000 to 9000 cals. per cubic metre. The distillation products of coal-tar are approximately as follows : benzene and its homologues, 1-5 to 2-5 per cent. ; phenol and its homologues, 0-5 to 2 ; pyridine and quinoline bases, 0-2 to 0-3 ; naphthalene, 4 to 6 ; heavy oils, 20 to 26 ; anthracene and phenanthrene, 0-5 to 2 ; pitch, 55 to 62 (38 being asphalte soluble in benzene and 24 carbon or insoluble) ; ammoniacal liquor, 4 ; gas and loss, 1-25. Another source of aromatic products is the distillation and heating of lignite-tar and 1 Lampblack is prepared by the incomplete combustion of tar, colophony, vegetable oils, the pitch or heavy oils from tar, &c. The liquid or fused substance of the receivers, a, is passed through pipes to the long pans, A (Fig. 406), in which it is heated while a carefully regulated minimal air-current is passed over the surface of the Section A JC-T*. m FIG. 406. liquid so as to burn the vapours incompletely and separate the greater part of the carbon in a free and finely divided state. This is carried away by the air into the first arched chamber, B, where it is partly deposited, then into the second arched chamber, c, and finally into D (before the chimney, 0), in which the final traces of lampblack are deposited on a thin cloth in front of the mouth of the shaft. This operation is continued for five days, the sixth day (Sunday) being occupied in cooling down and the seventh in restarting. A very fine lamp- black is obtained by burning paraffin oil in a kind of lamp with a wide thin jet and allowing the flame to impinge on an iron cylinder inside which water circulates ; the cylinder thus cools the flame and the lampblack deposited on it is removed from time to time by an automatic scraper. With more or less intense cooling the lampblack has a lower or higher specific gravity. 100 kilos of tar yield 25 kilos of lampblack, while 100 kilos of resin residue give 20 kilos. Lampblack contains, in addition to free carbon, tarry impurities and oily distillation products. Attempts have been made, apparently without success, to prepare lampblack by exploding acetylene with a measured proportion of air in closed vessels. The Frank process seems to be more advantageous ; in this, acetylene is burned with a certain proportion of carbon monoxide or dioxide : C 2 H 2 + CO = H 2 + 30. Swedish lampblack costs 16s. to 20s. per quintal, that from resinous wood 40s. to 52s., and that from lamps 8 to 20. It is used for making black varnishes, printers' ink, boot polish, &c. Boot polish is made by mixing ;ampblack with wax, molasses, turpentine, and sometimes also sulphuric acid or a little chestnut tannin extract to preserve the skin or leather. Italy imported and exported the following quantities (quintals) : 1905 1907 1908 S 1909 1910 Lampblack .Imports . . . t exports 1070 50 1986 20 1684 56 1910 26 2440 worth 4617 97 186 Boot polish {sports . . . 1 exports . . . 2335 1786 3500 1960 4140 1420 4240 1540 6900 10980 2518 4028 DISTILLATION OF TAR 529 FIG. 407. petroleum residues (see Cracking Process, &c., p. 74). By dropping the tar into very hot retorts and evacuating the latter, oils for various industrial purposes and also gas are obtained. The distillation is carried out in large wrought-iron vessels J holding several hundred quintals of tar. The old type of boiler is shown in Fig. 407, but preference is now given to horizontal stills, which are sometimes multitubular, like locomotive boilers, in order to obtain more homogeneous and more rapid heat- ing. It will be seen in the figure that direct -fire heat is used (at b) ; the mass is mixed at intervals by means of a stirrer or of a steam-jet introduced at x and subdivided on the arched base of the still by a number of pipes, z. The tar enters at r, and at the end of the operation the pitch is discharged through a much wider orifice than that marked a. A ther- mometer or pyrometer is inserted at v, while t serves as exit for the vapours, which are condensed in a coil surrounded by cold water in the case of the first pro- ducts and by hot water in that of the last products ; these are collected in order of density in a number of small receivers, from which they are passed to large J store-tanks. The stills are arranged in batteries under light roofs open at the sides so that the damage in case of fire or explosion may be minimised, the further precaution being taken of placing the fire hearths outside in the open. When the products formed at 270 are distilled over, the yield is increased and the pitch rendered more liquid, and so prevented from charring, by introducing a current of superheated steam, this remov- ing various substances (anthracene oil) which would otherwise remain in the pitch. The latter is then discharged, while hot, into old, disused steam boilers so as to avoid contact with the air, which might ignite the mass ; when almost cold but still fluid, it is run into shallow vessels or pits dug in the earth and allowed to solidify. With a still holding 300 to 400 quintals, each distillation (including charging and discharg- ing) lasts about 4 days. Distillation in a vacuum saves time (less than 30 hours being required for 200 quintals) and lessens repairs. In order to avoid decomposition of the pro- ducts distilling 'at high temperatures and to make the distillation continuous and thus increase the output and economise fuel, Wernecke (Ger. Pat. 201,372, 1907) has proposed the use of a conical, stepped still. A, fitted with a number of superposed peripheral channels, E, inside (Fig. 408). The cover, B, is fitted with a vapour outlet, b, and a pipe, a, for the continuous intro- duction of the tar (which first passes through a heater, where the water and light oils are distilled). The latter enters the uppermost channel, E, and overflows into the lower channels, gradually diminishing in volume owing to the distillation of various products ; the more or less liquid pitch is discharged at d. The vapours of the medium oils pass through the upper orifice, b, to refrigerators, but those of the heavy oils from the lower channels are collected by the perforated pipe, DF, which is provided with a cap, G, and is surrounded by metal gauze, and carries them through c to refrigerators. The still is heated by the 1 The rapid wear of the iron vessels and coils is due especially to HO, NH 3 , H 2 S, HCN", &c., formed by the dissociation at high temperatures of chlorides (e.g. ammonium chloride, dissociating at 360), sulphides, cyanides &c., and perhaps also by certain electrolytic processes. The base of the still is often 18 to 20 mm. in thickness. Cast-iron coils last better than those of wrought-iron, and are composed of superposed straight tubes connected at alternate ends by semicircular pipes of cast-iron. II 34 FIG. 408. 530 ORGANIC CHEMISTRY gases from the hearth, r, which circulate in the flues, e. The distillation is only interrupted once in 4 to 8 weeks to allow of the removal of the coke deposited on the inner surface of the still. Although the total capacity of the channels is only 600 kilos of tar, the daily output is equal to that of a still of the old type holding 2500 to 3000 kilos. Such a still also serves well for the distillation of lubricating oils from petroleum residues. In order to prevent the tar from frothing, it is necessary first to free it from water as completely as possible. This is done by maintaining the mass in open vessels for some hours at 40 to 50 or by adding quicklime or gypsum as a dehydrating agent. 1 The products which distil below 110 at ordinary pressure (sp. gr. 0-900 to 0-920) are somewhat similar to the ammoniacal liquor of gasworks, and consist of a more or less coloured liquid on which floats an oil containing a little benzene and toluene. The second portion which is collected is that distilling between 110 and 210, this forming the so-called light tar-oils (sp. gr. 0-935 to 0-995). From 210 to 240 the phenols or medium oils or creosote oils are collected. The next fraction consists of heavy oils (up to 270 ; sp. gr. 1 to 1-040), and the final one, the anthra- cene oil, passes over up to 270 (sp. gr. 1-050 to 1-095) and forms a buttery mass composed of oils and crystalline substances. According to whether the pitch (residue) is required to be more or less liquid or solid, the distillation is suspended after the third or fourth fraction has been collected ; to render the pitch shiny, it is mixed in the still with the heavy oil remaining after the crystallisation of the anthracene. This oil is also used with other lubricants for making oil -gas (see p. 57), lampblack, &c. Anthracene was very dear until a few years ago, and to obtain the maximum yield the heating was continued under reduced pressure after the distillation in superheated steam. Tar from metallurgical coke manufacture gives distillation products differing in their proportions from those of coal-tar, and even with the latter the tar from the hydraulic main (p. 43) is richer in pitch and .poorer in light oils than the tar from the condensers and separators. The tar from coke factories contains more light oils and less water than that from gasworks, and distils more regularly and more rapidly. 2 STATISTICS AND PRICES. The price of tar varies considerably with the locality, quality, season, demand, &c., the limits being about Is. Id. and 3s. 2d. per quintal. In Germany the price reached the maximum of 4s. Qd. per quintal in 1885 and fell to Is. 8d. in 1898 ; it now varies from 2s. 4d. to 2s. 5d. In 1900 the world's production of the various distillation products of tar was as follows : 50,000 tons of naphthalene, 24,000 of benzene, and 6000 of toluene. One-half of this output comes from coke factories and about one- third from Germany. In 1904 Germany produced 277,000 tons of tar in metallurgical coke factories (in 1908 about 632,400 tons) and 225,000 tons in lighting gasworks (in 1908 about 1 The separation of water is necessary in order that the distillation may be regular; if -water, is present, dis- tillation is very slow and is accompanied by bumping and frothing. When a sufficient separation cannot be obtained by decantation from the tepid tar after standing, special methods are used. According to Ger. Pat. 161,528 the tar in the still is heated first at the surface and subsequently in lower and lower layers to the bottom. Oppenheimer and Kant remove the water by means of gypsum or cement (Eng. Pat. 12,696, 1903). By centrifuging the tar in a non-perforated drum (as for starch, see p. 498), the proportion of water can be reduced to 1 to 2 per cent. In large distilleries the water is now eliminated by passing the tar from elevated tanks to the cooling coils in which the vapours of the tar are condensed ; the tar heated in this way to 50 to 60 enters a small rectifying column fitted to a large retort (150 to 200 hcctols.) almost full of tar already heated to 200 and freed from water. An overflow pipe to the retort delivers tar almost without water, while from the top of the column issue steam and a considerable proportion of the light oils, which are condensed in cooling coils. The estimation of water in tar is not easy, since when the tar is heated in a dish it readily froths and overflows. H. Beck and llispler (1909 and 1904) allow 200 grins, of the tar to fall drop by drop from a separating funnel on to about 500 grms. of water-free heavy tar-oil'contained in a flask of about 2 litres ; each drop of tar, as it falls, is instantly evaporated, and the water distilling over is condensed in the refrigerator connected with the flask and collected, together with a little tar-oil, in a graduated cylinder ; the temperature is finally raised to 300. The cylinder is kept at a moderate temperature, so that the water separates from the oil ; its volume is then read; If much naphthalene also distils over, it is difficult to read the volume of the water ; in this case, the whole of the distillate is poured on to a small filter-paper steeped in benzene, so that only the tar-oil filters. The filter- paper is subsequently pierced and the water allowed to pass into a graduated cylinder. E. Ott, on the other hand, heats 400 grms. of tar in a copper retort, the heating being carried out from the top by means of an annular gas-pipe with oriflees in its lower side. 1 In a large German coke-tar distillery, where retorts holding 350 ^quintals were used, the mean yields of several years were as follows : Ammoniacal liquor, 4-27 per cent. ; light" oils, 4-06 ; medium oils, 10-38 ; heavy oils, 6-11 ; anthracene oil, 13-71 ; pitch, 60-49 ; loss, 0-93. The tar distilled contained on an average 24 per cent, of matter (carbon) insoluble in benzene, and the mean cost of distilling 1000 kilos of tar was as follows : Labour, 7-7d. ; coal (at Is. 7d. per quintal), 14-4n! ; steam, CH(CH 3 ) 2 , also gives benzoic acid on oxidation, and is formed from benzene with either isopropyl iodide or normal propyl iodide (in the latter case aluminium chloride is necessary to cause molecular rearrange- ment) ; it is obtained also on distilling cuminic acid, C 6 H 4 (C 3 H 7 )-C0 2 H, or by the inter- action of benzal chloride, C 6 H 5 -CHCI 2 , and zinc methyl. TETRAMETHYLBENZENES, C 6 H 2 (CH 3 ) 4 . The best known of these are the fol- lowing : (a) Durene (1:2:4:5) or s-tetramethyl benzene, which is found, together with isodurene, in tar ; it is a solid, has a smell resembling that of camphor, and is prepared synthetically from toluene and methyl chloride. (b) and (c) Isodurenes, two isomerides being known (1:2:3:4 and 1:2:3:5) (see Table, p. 527). (d) p-Methylisopropylbenzene or cymene, CH 3 <^ ^> CH(CH 3 ) 2 , is a liquid of pleasant odour, b.pt. 185. It occurs naturally in cumin oil (from Cuminum cyminum) and in various essential esters, and it can be prepared by heating camphor with phosphoric anhydride or by the interaction of oil of turpentine and iodine. On oxidation it yields various acids. (e) m-Isocymene is found in resin oil. Hexamethylbenzene (mellithene), C 6 (CH 3 ) 6 , m.pt. 164, is a stable compound and can be neither nitrated nor sulphonated, owing to the absence of hydrogen atoms from the nucleus. When oxidised with potassium permanganate, it gives Mellitic Acid, C 6 (C0 2 H) 6 . HYDROCARBONS WITH UNSATURATED SIDE-CHAINS As far as the nucleus is concerned, these compounds behave like true benzene deriva- tives, whilst by means of the unsaturated side-chain they give all the reactions of unsaturated methane derivatives. STYRENE, C 6 H 5 -CH : CH 2 , occurs in storax and is formed on heating cinnamic acid, which loses CO 2 : C 6 H 5 CH : CH CO 2 H = C0 2 + C 6 H 5 -CH:CH 2 . It is a liquid of pleasant odour boiling at 146, and tends to polymerise to Metastyrene. Styrene combines with bromine, iodine, hydrogen, &c., in the same way as olefines do. When it is treated with nitric acid, a nitro -group is introduced into the side- chain, giving Nitrostyrene, C 6 H 5 -CH : CH*N0 2 , the constitution of this being shown by its formation from benzal- dehyde and nitromethane : C 6 H 5 -CHO + CH 3 -N0 2 = C 6 H 5 -CH : CH-N0 2 + H 2 0. Styrene serves for the synthesis of anthracene (q.v.). PHENYLACETYLENE, C 6 H 5 -C : CH, is a liquid of pleasant odour boiling at 142, and is prepared by converting acetophenone, C 6 H 5 'COCH 3 , by means of PC1 6 , into the dichloro -derivative, C 6 H 5 - CC1 2 -CH 3 , and eliminating 2HC1 from the latter by the action of potassium hydroxide. It is also obtained by the cautious distillation of Phenylpropiolic Acid, C 6 H 6 -C i OC0 2 H. Like acetylene, it forms metallic compounds ; treatment with concentrated sulphuric acid results in the addition of H 2 0, subsequent dilution with water giving acetophenone. 536 ORGANIC CHEMISTRY B. HALOGEN SUBSTITUTION PRODUCTS OF BENZENE Halogens act on benzene and its homologues, replacing one or more atoms of hydrogen and forming colourless liquids (sometimes crystalline substances) which are heavier than water, distil unchanged, and dissolve in alcohol and in ether. In aromatic hydrocarbons, a halogen in the benzene nucleus is held much more firmly than one in a side-chain and cannot be replaced by hydroxyl by the action of silver hydroxide or by the ammo-group by treatment with ammonia ; only by sodium or sodium alkoxide at about 200 can the halogen be eliminated. . The chlorine in the nucleus of chloro toluene is united as firmly as in chloro- benzene, whilst the chlorine in benzyl chloride is readily replaceable, just as is the case with that in methane derivatives. To ascertain whether the halogen is present in the nucleus or in the side-chain, the oxidation products are studied; thus, chlorotoluene gives chlorobenzoic acid, C 6 H 4 C1-C0 2 H, whilst benzyl chloride yields benzoic acid. * For distinguishing isomeric halogen derivatives, the same methods are used as for the xylenes, &c. In order to be able to name aromatic derivatives the more readily, the following names are given to the more common of the different groups or aromatic residues (known as aryl radicals and denoted generally by Ar) : OH, phenolic ; C0 2 H, carboxyl ; 0-CH 3 , methoxy ; C 6 H 5 , phenyl ; -CH 2 - C 6 H 5 , benzyl ; -CO-C 6 H 5 , benzoyl ; -CN, nitrile ; -S0 3 H, sulpho C ~ or sulphonic; C-C 6 H 5 , benzenyl; C 6 H 4 >0, phthalyl ; CH-C 6 H 5 , benzylidene or benzal ; C 6 H 4 -C 6 H 4 , diphenylene. General Methods of Formation. (1) In direct sunlight, chlorine and bromine act on benzene, giving additive products, e.g. C 6 H 6 C1 6 and C 6 H 6 Br 6 , but in diffused light (best in presence of traces of iodine, aluminium chloride, anti- mony trichloride, &c.), substitution products are formed. With homologues of benzene, if the reaction is carried out in the cold and in the dark (or in diffused light) or in presence of iodine (which acts catalytically), the halogen only enters the benzene ring (even in the hot, if iodine is present), whilst in the hot or in direct sunlight, the substitution takes place principally in the side-chain. (2) By heating halogenated acids with lime : C 6 H 4 C1-C0 2 H = C 6 H 5 C1 + C0 2 . (3) By withdrawing oxygen from oxygenated compounds (phenols, aromatic alcohols, ketones, acids, aldehydes) by means of PC1 5 ; e.g. C 6 H 5 -OH + PC1 5 = POC1 3 + HC1 + C 6 H 5 C1. (4) By boiling with cuprous chloride or potassium iodide the diazo- compounds obtained from the corresponding nitro- or amino-compounds : C 6 H 5 N : NCI = C 6 H 5 C1 + N 2 ; C 6 H 5 N : NCI + KI = KC1 + N 2 + C 6 H 5 I. (5) lodo-derivatives may be obtained by the action of iodine, iodic acid being added to oxidise the hydriodic acid which is formed. They are, how- ever, usually obtained by process (4). (6) lodobenzene, C 6 H 5 I, unites with two atoms of chlorine, forming iodoso- benzene chloride, C 6 H 5 IC1 2 , digestion of which with alkali yields iodosobenzene, C 6 H 5 1 : 0, the latter, when heated or oxidised (with chloride of lime) giving iodylbenzene, 2C 6 H 5 IO = C 6 H 5 I + C 6 H 5 I0 2 or C 6 H 5 IO + = C 6 H 5 I0 2 (an explosive, crystalline compound). Chlorination or bromination of toluene yields the para- and ortho-derivatives in equal quantities ; the me ta -derivative is obtained indirectly (from diazO- compounds). HALOGENATED BENZENE COMPOUNDS 537 PRINCIPAL HALOGEN DERIVATIVES OF BENZENE Empirical formula Name Melting- point Boiling- point Specific gravity Chloro-derivatives C 6 H 5 C1 Monochloro benzene 42 + 132 1-1 28 at C 6 H 4 C1 2 . o-Dichloro benzene (1:2) . . 179 m- (1:3). 172 P- (1:4) ... + 53 172 C6H 3 CI 3 * . v-Trichloro benzene (1:2:3) 16 218 as- (1:2:4) 63 213 s- (1:3:5) 54 208 C 6 H 2 C1 4 . v-Tetrachloro benzene (1 : 2 : 3 : 4) 46 254 as- (1:2:3: 5) 50 246 (1:2:4:5) 137 244 C 6 HC1 5 Pentachlorobenzene . . 86 276 CgClg . , Hexachlorobenzene . . . 226 326 Bromo-derivatives C 6 H 5 Br Monobronio benzene . . - 31 + 155 1-51 7 at C 6 H 4 Br 2 o-Dibromo benzene (1:2) 1 224 2-003 at m- (1:3) . + 1 220 1-955 at 20 P- (1:4) . 87 219 1-841 at 89 C 6 H 3 Br 3 . r-Tribromobenzene (1:2:3) 87 as- (1:2:4) 44 275 s- (1:3:5) 120 278 C 6 H 2 Br 4 . v-Tetrabro mo benzene (1:2:3:4) as- (1:2:3: 5) 98 329 (1:2:4:5) 175 C 6 Br 6 . Hexabromobenzene . . above315 C 6 H 4 Br-CH 3 o-Bro mo toluene (1:2) -26 181 1-422 at 20 m- (1:3) . . - 39-8 184 1-410 at 20 P- (1 : 4) + 28 185 1-392 at 20 C 6 H 5 -CH 2 Br Benzyl bromide . * liquid 198 1-438 at 22 lodo -derivatives C 6 H 5 I lodobenzene . ..- - 30 188 CeH^ o-Di-iodobenzene (1:2) -, + 27 286 m- (1:3) .. 40 285 P- (1:4) . 129 285 BENZYL CHLORIDE, C 6 H 5 -CH 2 C1, is a colourless liquid with a pungent odour, melting at 49 and boiling at 178 ; its specific gravity at 15 is 1-113. It was first prepared by Cannizzaro in 1853, and is obtained by chlorinating boiling toluene. With potassium acetate this chloride gives the acetyl-derivative, with potassium hydrosulphide a mercaptan, and with ammonia amino-bases. On protracted boiling with water it is transformed into benzyl alcohol, while boiling with lead nitrate converts it into benzalde- hyde ; when heated with finely divided copper, it loses chlorine and condenses to dibenzyl, C 6 H 5 CH 2 CH 2 C 6 H 5 . It is used for the preparation of oil of bitter almonds and for numerous aromatic syn- theses, its chlorine atom being readily replaceable. The commercial product costs 3s. 5d. per kilo and the chemically pure 5s. Id. Benzyl Bromide, when treated with potassium iodide, gives Benzyl Iodide. These products are also formed from benzyl alcohol, C 6 H 5 'CH 2 -OH, and halogen hydracids ; they may be converted back into the alcohol by boiling with water or potassium carbonate solution. 538 ORGANIC CHEMISTRY BENZAL CHLORIDE, C 6 H 5 -CHC1 2 , and Benzotrichloride, C 6 H 5 -CC1 3 , are obtained either by protracted chlorination of boiling toluene or by the action of PC1 5 or benzal- dehyds or benzoic acid. Benzal chloride boils at 204 and has the sp. gr. 1-295 at 16, while the trichloride melts at - 22, boils at 213, and has the sp. gr. 1-380 at 14. Mixed halogen derivatives are known, as also is Hexachlorohexahydrobenzene, C 6 H 6 C1 6 . Numerous halogenated derivatives of unsaturated aromatic hydrocarbons have likewise been prepared, e.g. o-Bromostyrene, C 6 H 5 -CBr : CH 2 , and /3-Bromostyrene, C 6 H 5 -CH:CHBr. C. SULPHONIC ACIDS These are formed directly from the aromatic hydrocarbons by the action of concentrated or fuming sulphuric acid or of chlorosulphonic acid, C1-S0 3 H. Improved yields are obtained in presence of mercury or ferrous sulphate, which exerts a catalytic action. They are crystalline substances, readily soluble in water and even hygro- scopic, and are separated from the excess of sulphuric acid either by means of their calcium or barium salts, which are soluble, or by saturation of the aqueous solution with sodium chloride and subsequent cooling ; in the latter case, the sodium sulphonate separates, this being decomposed with the calcu- lated quantity of a mineral acid and the free sulphonic acid extracted with ether. When treated with superheated steam or with hydrochloric acid, they lose the sulphonic group, the aromatic hydrocarbon being thus regenerated. With PC1 5 they form the acid chlorides, e.g. C 6 H 5 - S0 2 C1, which, with ammonium carbonate, yield the sulphamides, C 6 H 5 -S0 2 -NH 2 (see later). On energetic reduction, thiophenol (phenyl hydrosulphide), C 6 H 5 -SH, is formed. BENZENESULPHONIC ACID, C 6 H 5 -S0 3 H, is obtained by the direct action of con- centrated sulphuric acid on benzene : C 6 H 6 + H 2 S0 4 = H 2 O + C 6 H 5 -S0 3 H. Its barium and lead salts being soluble, it can be readily separated from the excess of sulphuric acid. It is very stable and is not decomposed on boiling with alkali or acid (as is ethylsulphonic acid), but if heated with hydrochloric acid at 150 or with superheated steam in presence of concentrated phosphoric acid, it takes up water, giving benzene : C 6 H 5 S0 3 H + H 2 O = C 6 H 6 + H 2 S0 4 . When distilled with potassium cyanide, it forms benzonitrile, C 6 H B S0 3 K + KCN = K 2 S0 3 + C 6 H 6 -CN. When fused with alkali it forms phenol, C 6 H 5 -S0 3 K + KOH = K 2 S0 3 + C 6 H 5 -OH, while with PC1 5 it yields Benzene Sulphochloride, C 6 H 5 -S0 3 H + PC1 5 = POC1 3 + HC1 + C 6 H 5 -S0 2 C1 (decomposable by water). With ammonia, ammonium carbonate, or primary or secondary amines, benzene sulphochloride gives more or less substituted Benzenesulphonamides, e.g. C 6 H 5 -S0 2 'NH 2 , C 6 H 5 -S0 2 -NHR, C 6 H 6 -S0 2 -NR 2 , which crystallise well. As the tertiary amines do not give this reaction, they can be separated from other amines. Owing to the highly acid character of the SO 2 group, the amino-group does not form salts, but its hydrogen can be replaced by metals, e.g. by dissolving in sodium hydroxide solution. Sulphur trioxide concerts benzene into Sulphobenzide (sulphone), (C 6 H 5 ) 2 SO 2 . Nitration of benzenesulphonic acid yields mainly m-nitrobenzenesulphonic acid, but small quantities of the ortho- and para -derivatives are also formed. Reduction of p-nitrobenzenesulphonic acid yields Sulphanilic Acid (p-aminobenzene- sulphonic acid), NH 2 'C 6 H 4 -S0 3 H (discovered by Gerhardt in 1845), which is also obtained on heating aniline with fuming sulphuric acid or on heating aniline sulphate at 200. This acid and also the corresponding meta-acid are used in the manufacture of artificial dye- stuffs, and both of them can be diazotised (see later). Sulphonic compounds and their salts are of importance in the dye industry as they give dyes soluble in water and readily applied to the dyeing of textile fabrics. Polysulphonic acids of benzene and its homologues are also known, some PHENOLS 530 of them serving for the separation of isomeric aromatic hydrocarbons (see Toluene). D. PHENOLS Phenols contain hydroxyl groups in place of one or more hydrogen atoms of the benzene nucleus. They have a characteristic odour (phenol, thymol), and certain of them are partially soluble in water, while all of them are soluble in alcohol and in ether ; they distil unchanged and have a more or less marked antiseptic action. Their properties resemble, to some extent, those of tertiary alcohols and those of weak acids. Thus, ethers are formed by the action of alkyl halogen compounds on the sodium derivatives of the phenols, anisole, C 6 H 5 - OCH 3 , and phenyl sulphate, CgHg-O-SOgH, being obtained in this way ; the latter compound is readily hydrolysed. They are, however, stable towards oxidising agents, nitric acid forming substitution products. The hydroxyl group is with some difficulty replaced by chlorine by the action of PC1 5 . They act as weak acids, but with alkalis form stable salts, which are soluble in water, are decomposed even by carbonic acid, - and show only slight electrical conductivity. Halogens and nitric acid replace the benzene hydrogen of phenols more easily than that of benzene itself or its homologues, so that even in dilute solution phenol can be precipitated quantitatively as tribromophenol by the action of bromine water. If the hydroxyl group is joined to a side-chain and not to the benzene nucleus directly, the compound is an aromatic alcohol and not a phenol. Oxidation of homologues of phenol yields hydroxy-acids, the side-chain being oxidised while the phenolic groups remain intact. When distilled with zinc dust, phenols give the corresponding aromatic hydrocarbons. In aqueous neutral solution, phenols give a violet, green, or other coloration with ferric chloride, calcium hypochlorite, or, in some cases, iodine. In general, they exert a reducing action. With nitrous acid, phenols form isonitroso-derivatives (oximes), and, in presence of concentrated sulphuric acid, intensely coloured solutions are formed which are turned blue by potash (Liebermanris reaction). The sodium or potassium derivatives of the phenols (phenoxides), with carbonic acid (or (CC1 4 + KOH) give aromatic hydroxy-acids : ~C 6 H 5 -OH + C0 2 = OH-C 6 H 4 -C0 2 H. With chloroform and sodium hydroxide, they yield the corresponding alde- hydes. They react with diazo-compounds and various other compounds forming colouring-matters (see later). The action of zinc chloride (or calcium chloride) and ammonia on phenols results in replacement of the OH by NH 2 . (a) MONOHYDRIC PHENOLS These are found alone or together with polyhydric phenols, and partly in the form of ethers (e.g. guaiacol, OH-C 6 H 4 - OCH 3 , cresol, &c.) in the tar obtained by the dry distillation of wood or coal. They are separated from the tar-oils by means of caustic soda, which renders them soluble, and, after separation, are set free by mineral acid and subjected to fractional distillation. They are also obtained industrially by fusing salts of sulphonic acids with alkali (in iron vessels ; in the laboratory silver vessels are used) : C 6 H 5 -S0 3 K + 2KOH = C 6 H 5 -OK + K 2 S0 4 + H 2 0. If the nucleus contains chlorine atoms, these are also substituted by hydroxyl groups by this reaction. 540 O5 CO 1O * CO "^ CO CO O OOOO O5 : ! 0) ,J5 a ^ 0) -i-J 2 >> = x il ^^ Cl- O J M 2 - s s M* HH ss H w o u o ^ % i_? HH c W d 1 W ,r-COM WWW W CARBOLICACID 541 Phenols are formed by boiling diazo-compounds (see later) with water in dilute sulphuric acid solution : C 8 H 5 -N 2 C1 + H 2 = N 2 + HC1 + C 6 H 5 - OH. Also, when benzene is oxidised with H 2 2 or with oxygen in presence of aluminium chloride, phenols are obtained. Chlorine atoms or amino-groups joined directly to the nucleus can be replaced by hydroxyl-groups by the action of sodium hydroxide, but only when the nucleus contains also strongly negative groups, e.g. N0 2 . PHENOL (Carbolic Acid), C 6 H 5 - OH, was first discovered byRunge in tar and occurs, to a small extent, in combination in urine. It is separated from tar-oils (see p. 530) by treatment with caustic soda solution (sp. gr. 1-09) and agitation by a current of air ; steam is passed through the decanted alkaline solution of phenol, this removing the naphthalene, &c. The phenol is then liberated by H 2 S0 4 or CO 2 (e.g. flue gases) and washed several times with water, crude carbolic acid (containing 40 per cent, of phenol, the rest creosote, &c.) of sp. gr. 1-05 to 1-06 being thus obtained. 1 This is purified by repeated distillation between 175 and 185 or, better, rectification until it crystallises at the ordinary temperature and no longer turns red in the air. To free it from final traces of cresol, it is diluted with 12 to 15 per cent, of water and the hydrate crystallised at 8 to 10 (cresol hydrate crystallises at 20), centrifuged and distilled until a strength of 99 per cent, is attained ; repetition of the operation and of the distilla- tion (in earthenware vessels) gives chemically pure phenol. Minimal quantities of water prevent crystallisation at the ordinary temperature. The low price of benzene renders practicable the industrial synthesis of phenol ; by means of fuming sulphuric acid the siilphonic acid is formed, this being then fused with one-half of its weight of caustic soda: C 6 H 5 -SO 3 Na + 2NaOH = H 2 O + Na 2 S0 3 + C 6 H 5 - ONa. The addition of acid then liberates pure synthetic phenol, which has very little smell and is suitable for the manufacture of picric and salicylic acids, &c. Pure phenol crystallises in long, colourless needles melting at 42-5 and boiling unchanged at 183 ; it has a specific gravity of 1-084, dissolves in 15 parts of water at 16, and is readily soluble in alcohol or ether. It has a characteristic odour, is poisonous, and on account of its great antiseptic power is largely used as a disinfectant in medicine and surgery 2 ; in many cases it is, however, replaced by other antiseptics (corrosive sublimate, cresols, &c.) which have not the unpleasant odour of phenol. The maximum antiseptic action of phenol is exerted in aqueous solution and in presence of acid, owing to its partial dissociation into the ions CeHgO 1 and H' ; according to Pfliigge, when dissolved in pure alcohol or in oil, it has no antiseptic action, since it is then not dissociated. It dissolves in caustic alkali solutions (forming phenoxides, e.g. C 6 H 5 -ONa), but not in those of alkali carbonates. With formaldehyde it forms resinous condensation products (artificial sealing-wax : Resit, 1909). A pine splinter, moistened with hydrochloric acid, is coloured bluish green by phenol. 1 Testing- of Carbolic Acid. Commercial pure phenol melts at 39, other pure forms melting at 30" to 35 and boiling at 183 to 186. When pure, phenol should dissolve completely to a clear solution in 15 parts of water and should leave no residue on evaporation. Phenol which does not crystallise at the ordinary temperature contains at least 10 per cent, of higher phenol liquors. The exact quantitative estimation of pure phenol (not containing cresols, which behave like phenol) is effected by transforming it into tribromobenzene by Koppe- schaar's method. There is no characteristic reaction for distinguishing the phenols from cresols, but the latter are the less soluble in water. An approximate method, which is used in practice, and is suggested also in the German Pharmacopeia, for determining the phenol-content of crude carbolic acid is as follows : 10 vols. of the product are shaken for a long time with 90 vols. of sodium hydroxide solution (sp. gr. 1-079) in a graduated cylinder and then left to stand until two layers separate ; the volume of the undissolved non-phenol is then read off and, after this has been removed, the residue is acidified with HC1 and NaCl added to separate the whole of the phenol, the volume of which is subsequently measured. * The action of Antiseptics or Disinfectants (see also p. 127) depends on the chemical character of the antiseptic substance and partly on the quantity and nature of the substance to be disinfected. The poisonous action of disinfectants is the result of a chemical action between the proteins of the plasma of the living cells, this having varying affinities towards different antiseptics ; the concentration of the latter, the duration of the action, CH 2 , which contains no double linking and corresponds with triketohexamethylene ; in accord with this structure, it reacts with 3 mols. of hydroxylamine, giving a trioxime. On the other hand, it behaves also as a trihydroxybenzene or trihydric phenol, giving a triacetyl-derivative with acetyl chloride, so that it is able to exist in two tautomeric forms. This explains why, when it is treated with alcoholic potash or with an alkyl iodide, the alkyl groups unite with carbon and not with oxygen (as they would with a triphenol), giving, e.g. hexaniethylphloroglucinol. Pure phloroglucinol costs about 16 per kilo. 546 O R G A N I C C H E M I S T R Y (d) POLYHYDRIC PHENOLS From dinitroresorcinol is obtained a Tetrahydroxybenzene, C 6 H 2 (OH) 4 (1:2:4:5). which boils at 220, while chloranilic acid (see later) is formed by the oxidation of the dichloro- derivative. HEXAHYDROXYBENZENE, C 6 (OH) 6 , is obtained as potassium derivative, C 6 6 K 6 , in the manufacture of potassium by reduction of its carbonate : K 2 C0 3 + C 2 = SCO + K 2 and 6K + 6CO = C 6 6 K 6 . These reactions represent a further example of the synthesis of organic substances from inorganic matter. Hexahydroxybenzene is a white, crystalline substance which oxidises readily in the air and yields benzene when distilled with zinc dust. Of the additive products formed by polyhydric phenols with hydrogen, quercitol and inusitol may be mentioned. QUERCITOL (Pentahydroxycyclohexane or Acorn Sugar) / OH-CH/ >CH 2 , \CH(OH).CH(OH)/ s found in acorns and is similar to mannitol ; it has a sweet taste and forms monoclinic prisms melting at 234, its specific rotation being [a]p 6 = + 24-16. When heated to 240 in a vacuum or fused with alkali it loses water yielding various aromatic derivatives (hydroquinone, quinone, and pyrogallol) ; on reduction with HI, it gives benzene, phenol, pyrogallol, quinone, and hexane. When oxidised with nitric acid it forms mucic and trihydroxyglutaric acids, while with permanganate it yields malonic acid, the presence of the methylene group, CH 2 , being thus confirmed. It forms a pentacetyl-derivative, an explosive pentanitrate, and a pentachlorohydrin, C 6 H 7 C1 5 , melting at 102 ; the formation of these compounds demonstrates the presence of five hydro xyl groups. INOSITOL (Hexahydroxycyclohexane or Muscle Sugar), C 6 H 6 (OH) 6 , is similar to quercitol but contains a CH OH group in place of the CH 2 . It has the appearance and, to some extent, the sweet taste of the sugars, with which it was for long confused. That it is a cyclohexane derivative is shown by the formation of phenol, benzene, and triiodophenol on reduction with HI, and that of quinone and some of its derivatives on treatment with PC1 5 . The presence of six hydroxyl groups is proved by the formation of a hexa-qcetate (m.pt. 212) when it is treated with acetic anhydride and zinc chloride, and of a hexanitrate, C 6 H 6 (NO 3 ) 6 (m.pt. 120), under the action of concentrated sulphuric and nitric acids ; the hexanitrate is highly explosive and reduces Fehling's solution. Four optical isomerides are known: (1) inactive; (2) dextro-rotatory, [a] D +68-4, crystallising with 2H 2 and melting at 247 j (3) laevo- rotatory, [a] D 65, m.pt. 247 ; (4) racemic, melting at 250. Baeyers stereochemical conceptions indicate eight possible isomerides, according to the arrangement of the OH and H above or below the plane of the hexagon. Inositol, especially the inactive form, occurs in beans, lentils, peas, the muscles of the heart, the brain, &c. The inactive modification crystallises from water with 2H 2 at temperatures below 50 and in an anhydrous form, m.p. 225, at higher temperatures ; it boils unchanged in a vacuum at 319 and is not fermented by yeasts. It does not combine with phenyl- hydrazine or reduce Fehling's solution, but it reduces ammoniacal silver nitrate solution ; it forms a basic lead salt, (C 6 H u O 6 ) 2 Pb, PbO. It does not yield quercitol when reduced, so that the hydroxyl groups are symmetrically distributed. The rnonomethyl ether of i-inositol, or bornesitol, is found in Borneo rubber, and the dimethyl ether, or dambonitol, C 6 H 6 (OH) 4 (OCH 3 ) 2 , in Gabon rubber. The monomethyl ether of d-inositol, or pinitol, which occurs in many plants and plant-juices, melts at 186, sublimes at 200, and has a rotation of + 67-5. The monomethyl ether of 1-inositol, or quebrachitol, melts at 186, boils at 200 in vaciw, and with HI forms 1-inositol ; it occurs in quebracho bark. E. QUINONES These may be regarded as derivatives of phenols obtained by elimination of hydroxyl groups, with consequent displacement and partial elimination of the double linkings of the benzene nucleus. They are usually yellow and Q U I N O N E S 547 of pungent odour and possess oxidising properties ; they are volatile in steam, with partial decomposition. Oxidation of meta- and ortho-diphenols does not yield quinones. BENZOQUINONE or simply Quinone, C 6 H 4 O 2 , can be obtained by oxidising either p-aminophenol or sulphanilic acid (1:4 NH 2 -C 6 H 4 -S0 3 H), or p-phenolsulphonic acid, or hydroquinone, or aniline (on a large scale) with chromic acid. On sublimation it forms fine yellow crystals which melt at 116, giving a characteristic irritating odour. It is soluble in alcohol or ether and slightly so in cold water. It fixes hydrogen, which transforms it into hydroquinone, while the halogens give addition or substitution products according to the conditions. With HC1 it forms monochloro- hydroquinone, C 6 H 4 2 + HC1 = C 6 H 3 C1(OH) 2 . With amines and with phenols it forms dyes which crystallise well but are only slightly soluble. With hydroquinone it forms a condensation product, Quinhydrone, C 6 H 4 O 2 - C 6 H 4 (OH) 2 , which consists of green prisms with a metallic lustre, and may be regarded as an intermediate product in the oxidation of hydroquinone or in the reduction of quinone 0:C/ \CH-0. 0-CH< Constitution. That quinone contains two carbonyl groups is deduced from the fact that with hydroxylamine it yields quinone monoxime and quinonedioxime : CO C : NOH C : NOH /\ CH I HCll I CH CO CO C : NOH Quinonemonoxime (nitrosophenol) It contains two double linkages, since in benzene solution it absorbs four atoms of bromine, while ozone is also fixed quantitatively (see pp. 88 and 299). OH CO /\ y\ The transformation of hydroquinone, |X| > into quinone, || || , is an evident OH CO example of the convertibility of the centric form of benzene into that with two double linkages. Tetrachloroquinone (chloranil), C 6 C1 4 2 , prepared by oxidising trichlorophenol with dichromate and sulphuric acid, serves for the manufacture of coal-tar dyes ; the com^ mercial product costs 20s. per kilo, and the pure 80*. Toluquinone, C 6 H 3 O 2 -CH 3 , xyloquinone, thymoquinone, &c., are known, as also are quinoneimides (e.g. C 6 H 4 0-NH) and quinonediimides [e.g. C 6 H 4 (NH) 2 ], F. NITRO-DERIVATIVES OF AROMATIC HYDROCARBONS These are readily obtained by treating the hydrocarbons with concentrated nitric acid, best in presence of concentrated sulphuric acid, which fixes the water as it is formed : C 6 H 6 + HN0 3 = H 2 + C 6 H 5 -N0 2 . With the hydrocarbons homologous with benzene, nitration is still more easy, but not more than ' three nitro-groups can be introduced directly ; . tetranitro-derivatives are prepared indirectly. Aromatic nitre-compounds cannot be obtained by the action of silver nitrite on chlorobenzenes, as is the case with those of the fatty series ; but this method serves for the intro- duction of nitro-groups into side-chains. The nitro-compounds are liquid or solid and usually more or less yellow, although some are red ; they are heavier than water and dissolve readily in 548 ORGANIC CHEMISTRY >, i BD IO CM ' O CO COCOCMCMCOCOCO 1 1 o IO rH 49 1 oo o 33 rH SC 4;, S .5 oooo ooo ooo 00 O5 CM O5 OO O -* I GO CO O 1 O rH O O5 I-H CO CO 1 IO TH <* 1 CMCOCOCM CM CM CM CMCMCM o 10 1 1 CM O 1 CM I' 1 ooooo. .ooooo r c!oo CO IT*- O CM CM t^ O CO rH O O5 CM '^ CM C i-H O t~~ CM *O i-H rH IO l^ CM ~ GO C + rH rH rH . 9 ooooo 3 CM ^ CO G5 1O 5 00 <* 00 O 00 I-H CM rH O O .H O O rH i-H 3 GO CO K CO CO CT 1 I- *& CM 3 rH 1 ( C o 5 OS 4 o 1 i 1 CM CO -* ^ ^ C! rH rH rH " " C~:j CM T 1 1 I-H *' ?. | <- 4) S! . x^ fl 4) 4) O -^ 3 4) 4) 4) 3 4) ^^ 4) O 5 2 '3 ft <*> c S g| s C3 Q^ O S 33 -0 S 3 2 N 4) 4) -*-* O S O ^ SUf a ^ -a o o rS 3 .-S'lS " 5 1 1 jz; O O -H -g S S T3 O 4J -4J H ^ 'S - a ^ 5 S .- 2 "I -a" o "S 43 9 fl 4) , . . . C! ^3 "^ rj 'S S3 - ^S a 4) jD 4) 4) 4) D 0-S3 S33 ^ ^^4) initro-m-xjdene (dimet tromesitylene (trimeth nitromesitylene (trime initrohemimellithene ( initropseudocumene initromesitylene troprehnitene (tetrame trocymene (methylisoj nitroprehnitene (tetrai nitroisodurene mtroaurene initro-tert. butyltoluei t^oSOHa!cSoS& M 3 H 9 6 1 NITROBENZENE 549 alcohol, ether, o acetic acid, but are mostly insoluble in water. They distil unchanged and are volatile in steam. The nitro-group is united very firmly to the nucleus, especially in mono- nitrobenzene, and is not directly replaceable. It can be reduced to the amino- group by means of nascent hydrogen in acid solution ; reduction in alkaline solution results in the formation of azoxy-, azo-, and hydrazo-compounds, whilst in neutral solution or with hydrogen sulphide, the nitro-group becomes a hydroxyl- amino-group. On electrolytic reduction, nitro-derivatives yield amino-phenoK Polynitrobenzenes are easily obtained by the action of fuming nitric acid in the hot ; the me ta -derivative is formed first and this, by further nitration with nitric and fuming sulphuric acids at 140, gives symm. trinitrobenzene. The polynitro-compounds react more readily than mononitro-derivatives ; when the former are oxidised, a phenolic group is formed, while the nitro- groups remain intact. With para- and ortho-dinitrobenzenes, sodium alkoxide replaces one nitro-group quantitatively, whilst with m-dinitrobenzene no reaction occurs : C 6 H 4 (N0 2 ) 2 + C 2 H 5 -ONa = NaN0 2 + N0 2 -C 6 H 4 -OC 2 H 5 . By boiling o-dinitrobenzene with caustic soda, o-nitrophenol is formed : _ N0 2 _OH NaOH - NaN0 2 + < while boiling with alcoholic ammonia yields o-nitralinine : N0 2 NH 2 A N0 2 + NH 3 = HNO 2 + C >N0 2 . NITROBENZENE, C 6 H 5 -N0 2 , is an almost colourless, faintly yellow, refractive liquid which has the sp. gr. T209 at 15 and, after solidification, melts at 3 and boils at 208. Owing to its pleasant bitter-almond smell, it is used in perfumery under the name of artificial essence of mirbane, but its vapour is somewhat poisonous. It is insoluble in water, but it mixes in all pro- portions with alcohol, ether, or benzene. It is of considerable industrial importance, as it forms the raw material for the manufacture of aniline, benzidine, quinoline, azobenzene, various explosives, &c. On a large scale it is prepared in wrought- or cast-iron vessels, employing precautions and methods similar to those used in making nitroglycerine (see p. 225). The nitro -sulphuric mixture, consisting of 120 kilos of HN0 3 (42 Be.) and 180 kilos of H 2 S0 4 (66 Be.), is poured gradually (in 8 hours) into 100 kilos of benzene. The mass is kept mixed by means of a stirrer, and during the first 5 to 6 hours is maintained at 25 by means of cold water circulating outride the apparatus. In the final phase of the reaction the temperature is raised by external steam to 70 to 90, the heating being then stopped, while the stirring is continued for a further 6 hours. The mass is then forced by a suitable elevator into a tank with a conical base. The acid mixture gradually settles to the bottom, while the nitrobenzene floats ; the former is then drawn off through taps (see Nitroglycerine), and the nitrobenzene, after repeated washing with water, distilled in a current of steam from a vessel with a jacketed bottom heated with steam at 2 to 3 atmos. pressure. A second distillation yields moderately pure nitrobenzene. According to Ger. Pat. 221,787 of 1907, nitro- benzene can also be obtained by running benzene into a mixture of sulphuric acid and sodium nitrate at 70 to 90. It is sold at 72s. to 104s. per quintal. The imports of nitrobenzene into Italy are as follow : 11 quintals in 1900, 18 in 1907, 138 in 1908, and 182, of the value of 874, in 1910. 550 DINITROBENZENES. By the action of fuming nitric acid or of a suitable nitro- sulphuric mixture on benzene, m-dinitrobenzene is formed along with small proportions of the ortho- and para-compounds. The me ta -derivative crystallises from alcohol in colourless needles, m.pt. 90, and is insoluble in water, but readily soluble in alcohol or ether. The ortho- and para-isomerides are obtained indirectly from the corresponding dinitroanilines (see Aniline) by elimination of the amino-group ; both form colourless crystals. On reduction, m-dinitrobenzene gives first m-nitraniline and then m-phenylene- diamine. The crude product costs 6 per quintal, and the chemically pure 8s. per kilo. Dinitrobenzene dust is somewhat poisonous. TRINITROBENZENE (symm. ), C 6 H 3 (N0 2 ) 3 . Attempts are now being made to utilise this compound as an explosive. It is obtained by oxidising trinitrotoluene (see later) with sulphuric and chromic acids (Ger. Pat. 127,325) and decomposing the resultant trinitro- benzoic acid by heat, NITROTOLUENES MONONITROTOLUENES, NO 2 -C 6 H 4 -CH 3 (ortho-, m.pt. -10, b.pt. 218 ; meta-, m.pt. + 16, b.pt. 230 ; para-, m.pt. + 54, b.pt. 236). Large proportions of the ortho- and para -compounds and a small proportion of the meta-compound are obtained when toluene is nitrated with fuming nitric acid, the relative amounts varying with the condi- tions of the reaction. Thus, if highly concentrated nitric acid (sp. gr. 1-53) is employed and the mass is not cooled, 65 per cent, of p-nitrotoluene is obtained ; but if such a weak acid is used that it scarcely reacts, and the reacting mass is cooled, 67 per cent, of o-nitro- toluene is formed. When toluene is nitrated with nitric and sulphuric acids, 60 to 66 per cent, of the ortho -compound is obtained (100 kilos of toluene are added, in the course of 12 hours, to a mixture of 100 kilos of nitric acid of 44 Be. with 150 kilos of sulphuric acid of 66 Be., the mass being stirred and cooled and the decanted nitro -products washed with water and alkali) ; the unaltered toluene is distilled off in steam and the nitrotoluene eventually distilled by means of superheated steam. The ortho- and para -compounds are separated by fractional distillation ; 40 per cent, of ortho -compound distils at 222 to 223, then a little meta-, and above 230 the p-nitrotoluene. The separation of the ortho- and para-isomerides may also be effected by cooling to 6 (Ger. Pat. 158,219, Fr. Pat. 350,200), when almost colourless p-nitrotoluene (melting at 54 and boiling at 236 when pure) crystallises out ; o-nitrotoluene is a yellow liquid, which solidifies at 10-5, boils at 218, and has the sp. gr. 1-168 at 15. The crude mixture of these two isomerides costs 96s. to 1125. per quintal and is used, either as it is or after separation, for the manufacture of toluidine, tolidine, and fuchsine. m-Nitrotoluene is formed in small quantity in the direct nitration of toluene (see also Dinitro toluene), but in the pure state is obtained only indirectly from m-nitro-p-toluidinc N0 2 2 , by Griess's reaction (see Aniline). Only with difficulty is it nitrated further to dinitrotoluene, thus differing from the other mononitro toluenes. It has no important practical use, and when impure costs 4s. per kilo, and when pure 32s. It forms crystals melting at 16, boils at 230-5, and has the sp. gr. 1-168 at 22. DINITROTOLUENES, C 6 H 3 (CH 3 )(NO 2 ) 2 , exist in six isomeric forms, which are prepared and named in various ways. Denoting the methyl group by M (always in position 1) and the nitro -group by N, the isomerides have the following configurations : 5 3 4 M M N m-dinitro toluene ordinary dinitrotoluene o : p-dinitrotoluene a-dinitro toluene 2 : 4-dinitrotoluene m.pt. 70-5 N p-dinitrotoluene e -p-dinitrotoluene 2 : 5-dinitrotoluene m.pt. 48 X N o : m-dinitrotoluene 1:2: 3-dinitrotoluene m.pt. 63 551 N N X o : o-dinitrotoluene m : m-dinitrotoluene m : p-dinitrotoluene j3-dinitrotoluene S-dinitrotoluene y-dinitrotoluene 2 : 6-dinitrotoluene 3 : o-dinitrotoluene 3 : 4-dinitrotoluene m.pt. 61 m.pt. 92 m.pt. 60 Of the various names, the last given in each case is the simplest and clearest. When toluene is nitrated directly with a suitable nitro-sulphuric mixture (richer in nitric acid and poorer in water than for mononitrotoluene) and the mass is finally heated almost to boiling, the main product is ordinary solid dinitrotoluene (2 : 4), a little trinitro- toluene and 2 : 5-dinitrotoluene being also formed. About 35 per cent, of the crude mass always consists of a liquid product which is separated by centrifugation and was thought to be another isomeride, but Glaus, Becker, Nolting, and Witt have shown it to be a mixture of 2 : 4- and 2 : 6-dinitrotoluenes and 40 per cent, of mononitrotoluenes (equal parts of p- and m- and a little o-) ; the mononitrotoluenes can be removed by distillation in a vigorous current of superheated steam. This orange-red mixture of liquid products gela- tinises collodion -cotton well and serves for the preparation of incongealable dynamites and powders or dynamites with ammonium nitrate as basis. 2 : 4-Dinitrotoluene is prepared as described above and is the one in most common indus- trial use, while it serves also for making ordinary (2:4:6) trinitrotoluene. It is purified by crystallisation from alcohol or carbon disulphide and forms monoclinic crystals melting at 70-5 ; it is insoluble in water, slightly soluble in cold alcohol or ether, still less so in carbon disulphide (2-2 per cent.), and readily soluble in benzene. It dissolves in alkali, giving a red solution, from which acids precipitate a reddish brown substance. Fuming nitric acid oxidises it slowly and in the hot gives the corresponding o : p-dinitrolenzoic acid, C 6 H 3 (CO 2 H)(N0 2 )2- With hot, concentrated nitro-sulphuric mixture, it forms ordi- nary trinitrotoluene (see below). Ammonium sulphide reduces it in the cold to o-nitro-p- toluidine (m.pt. 105), while in the hot, p-niiro-o-toluidine (m.pt. 78) is also formed. By zinc and hydrochloric acid it is reduced to tolylenediamine. 2 : 6-Dinitrotoluene is obtained along with the 2 : 4-isomeride and accumulates in the mother-liquors, when mononitrotoluene (ortho) is nitrated further. It is prepared in the pure state by eliminating the amino-group from dinitro-p-toluidine (m.pt. 168). It forms shining needles, m.pt. 61, dissolves to some extent in alcohol, and with ammonium sulphide gives o-nitro-o-toluidine. 2 : 3-Dinitrotoluene is obtained by heating o : m-dinitro-p-toluic acid with dilute hydro- chloric acid for 6 hours at 265 and distilling in a current of steam, the crystals formed being pressed or centrifuged ; it separates from light petroleum solution in yellow crystals, m.pt. 63. 2 : 5-Dinitrotoluene is obtained together with the 2 : 4-derivative when toluene or nitrotoluene is run into fuming nitric acid ; it crystallises from alcohol in yellow needles, m.pt. 48. Alcoholic ammonium sulphide reduces it to o-nitro-m-toluidine. 3 : 5-Dinitrotoluene is formed by eliminating the amino-group by diazotisation (see Aniline) from dinitro-o-toluidine (m.pt. 208) or from m : m-dinitro-p-toluidine (m.pt. 168). From water, in which it is sparingly soluble, it crystallises in needles, m.pt. 92. It is soluble slightly in light petroleum, more so in cold alcohol or in carbon disulphide, and readily in chloroform, ether, or benzene. It distils easily in a current of steam, and with benzene forms the crystalline double compound, C 6 H 3 (CH 3 )(N0 2 ) 2 + C 6 H 6 . 3 : 4-Dinitrotoluene is obtained by protracted agitation of m -nitrotoluene with concen- trated nitric acid (sp. gr. 1-54). From carbon disulphide (which dissolves 2-19 per cent.), it crystallises in long needles melting at 60. TRINITROTOLUENES. The following six isomerides are possible, only the first three being known : O R G A N I C C H E M I S T R Y M M M N N a-triuitrotoluene 2:4: 6-trinitrotoluenc melts at 82 and solidifies at 80 5 M X N 2:3: 4-trinitrotoUiene X /3-triiiitiotolu3ne 2:3: 6-trinitrotoluene m.pt. 112 M X 3:4: 5-tiinitrotolucue X N p-trinitrotoluene 2:4: 5-trinitrotolaene m.pt. 104 N X X 3 : 5-trinitiotolucne Rudeloff (1907) is of opinion that, together with a -trinitrotoluene, two other isomerides, melting at 73 and 78, are formed, but these are probably more or less impure a-compounds. a-TRINITROTOLUENE (ordinary or 2 : 4 : 6-Trinitrotoluene) is formed on heating toluene for several days or, better, 2 : 4-dinitrotoluene for some hours, with a highly con- centrated nitro -sulphuric mixture, the operation being begun at a low temperature and with constant mixing, and the temperature being raised gradually to 100. After removal of the acids by decantation, the mass is washed with boiling water and purified by crystal- lisation from alcohol or from concentrated sulphuric acid, in which it dissolves in the hot (V. Vender, Fr. Pat. 405,812 of 1909). It forms pale yellow crystals which darken under the influence of light ; it melts at 82 and solidifies at 80-5. At a higher temperature it undergoes partial sublimation, and when heated rapidly to 240 it sometimes explodes. It is very slightly soluble in water (0-164 per cent, at 100 and 0-021 per cent, at 15) ; the mixture of nitric and sulphuric acids containing 15 per cent, of water dissolves from 2 per cent, to 5 per cent, according to the proportion of nitric acid present ; 99 per cent, sulphuric acid dissolves it to the extent of 66 per cent, at 100 and of 10 to 12 per cent, at 20. Alcohol dissolves 2 to 3 per cent, of it in the cold and 25 per cent, in the hot ; it is readily soluble in ether, acetone, or benzene ; cold carbon disulphide dissolves only 0-39 per cent. When non -compressed the crystals have the density 0-8 to 1, but if they are fused and allowed to solidify under ordinary pressure the density is 1-54 to 1-57 ; while if the solidifi- cation takes place under a pressure of 3 to 4 atmos. (Bichel, 1906) or with rapid cooling (Nobel Dynamite Co., Hamburg, 1907), the value 1-61 to 1-62 is attained. When the crystals are compressed in a hydraulic press to 200 to 600 kilos (or to 3000 kilos) per square centimetre, they assume a density of 1-59 (or 1-68). When aniline is poured into an alcoholic solution of trinitrotoluene, a double compound, C 6 H 2 (CH 3 )(N0 2 )3 + C 6 H 5 'NH 2 , separates in red acicular crystals melting at 84. If heated at 180 with ten times its weight of fuming nitric acid, trinitrotoluene is converted into s-trinitrobenzene. While picric acid (which is now partly replaced by trinitrotoluene as an explosive) readily forms with metals picrates dangerous to handle, trinitrotoluene does not react with metals and can be manipulated safely even in the hot, since it burns slowly without exploding ; it is not hygroscopic and does not form a bitter and poisonous powder like picric acid. It is highly stable to shock, and when compressed is exploded with a mercury fulminate cap ; but when fused and then solidified it is exploded only by a detonator of moderately compressed, crystalline trinitrotoluene, which in its turn is exploded by a fulminate cap. The velocity of detonation in a charge 50 mm. in diameter and with a density of 1-55 is 7500 metres (picric acid, 8000 metres). The theoretical decomposition is expressed by : 2C 6 H 2 (CH 3 )(NO 2 ) 3 = 12CO + 2CH 4 + H 2 + 3N 2 , 1 kilo giving 778 litres of gases, which are incompletely burnt owing to lack of oxygen. The use of trinitrotoluene as an explosive was suggested prior to 1890, and attempts PHENYLNITROMETHANE 553 were made to compensate the deficiency of oxygen by addition of ammonium nitrate. But it has been largely used, mainly as a result of Bichel's investigations, only since 1904, and in the crystalline state it now forms a very important military explosive. In the com-' pressed or solidified state it is used for charging projectiles, grenades, &c. (it does not serve for propelling projectiles, owing to its shattering power and to the abundance of fumes it forms on explosion). Different firms produce it under various names (trotyl, trolite, trilite, trinol, tritole). Germany manufactures 12,000 to 15,000 quintals per annum, and Italy 4000 to 5000 ; the price varies from 2s. Gd. to 4. per kilo, according as it is crystallised, granulated, fused, or compressed. For some time a plastic product called plastrotyl (Bichel, 1906) was prepared from trinitrotoluene, resin, collodion-cotton, and crude liquid dinitrotoluene, but this is no longer manufactured. /3-TRINITROTOLUENE or 2 : 3 : 6-Trinitrotoluene is formed in small proportion with a large proportion of the y-isomeride (see below) when m-nitro toluene is boiled for a day with nitric and sulphuric acids. It separates from carbon disulphide or alcohol in colourless crystals melting at 112, and is readily soluble in ether, acetone, or benzene. With alcoholic ammonia in the hot it gives y-dinitrotoluidine (m.pt. 94). y-TRINITROTOLUENE or 2 : 4 : 5-Trinitrotoluene is formed with the /3-isomeride (see above), from which it can be separated in virtue of its slight solubility in alcohol or carbon disulphide. It forms yellowish, shining crystals, melting at 104. When heated with alcoholic ammonia it forms /3-Dinitrotoluidine, while with aniline in the cold it gives Phenyldinitrotoluidine, melting at 193. Also with aniline its hot alcoholic solution gives orange crystals of y-Dinitrotolylphenylamine, m.pt. 142. CHLORO- and BROMO-NITROBENZENES. The para-derivatives melt at higher temperatures than the meta- and these at higher temperatures than the ortho -com pounds. This rule often holds with aromatic compounds. Nitration of chlorobenzene yields much para- and little or^o-derivative ; the meta- compound is prepared indirectly from m-nitraniline by transforming the amino -group and replacing it by halogen. TRINITROTERT.BUTYLXYLENE has an odour of musk and is used as a perfume. PHENYLNITROMETHANE, C 6 H 5 -CH 2 -NO 2 , contains the nitro-group in the side- chain, as is shown by its method of preparation : C 6 H 5 .CH 2 C1 + AgN0 2 = AgCl + C 6 H 6 - CH 2 . N0 2 . Benzyl chloride It is obtained also by heating toluene with nitric acid (sp. gr. 1-12) under pressure. This compound exists in two isomeric (or tautomeric) forms, one being known as a pseudo- acid : (1) C 6 H 5 .CH 2 .N0 2 and (2) C 6 H 6 . CH : NO OH (pseudo-acid) ; the former does not react with ferric chloride, while the latter gives a coloration. Modification (1) is a liquid, and its aqueous solution gives, with sodium alkoxide, the sodium salt of the pseudo- acid ; when the acid is liberated by means of a mineral acid it forms a crystalline product, which has the same composition as the original compound and gradually changes into this, becoming liquid. The presence of a hydroxyl group in the pseudo-acid is demonstrated by the formation of the characteristic dibenzhydroxamic (or dibenzoylhydroxamic) acid by treatment with benzoyl chloride : C 6 H 5 -CH : NO-ONa + C 6 H 5 .CO-C1 = NaCl + C 6 H 5 .CH : NO O CO C 6 H 5 C 6 H 5 .CO-NH.O.CO.C 6 H 5 Dibenzhydroxamic acid That these isonitro -compounds contain hydroxyl is shown also by the fact that they react in the cold with phenyl isocyanate, while the nitro -compounds do not. _N0 2 Similar behaviour is shown by m-Nitrophenylnitromethane, ^ \ ; the ~CH 2 .N0 2 passage from the yellow pseudo-acid to the colourless nitro-compound is clearly shown by the change both in colour and in electrical conductivity, which is very high for the pseudo-acid (as for acids in general) and almost zero for the normal nitro-compound, into which it is gradually converted. 554 ORGANIC CHEMISTRY These nitro -derivatives of the side chain can hence yield metallic derivatives of the pseudo-acids ; treatment of these derivatives with acid yields the normal form, and the latter in presence of alkali is only slowly neutralised, this being characteristic of the pseudo- acids. In benzene solution the true acids combine rapidly with ammonia, forming insoluble ammonium salts, while pseudo -acids combine only slowly or not at all with ammonia. G. AMINO-DERIVATIVES OF AROMATIC HYDROCARBONS When the hydrogen atoms of benzene are replaced by amino-groups or the hydrogen of ammonia or of a primary aliphatic amine by phenyl -groups, the resulting products are mono-, dl-, or tri-amines in the first case and secondary and tertiary amines in the second. Some of the aromatic amines are similar to but weaker than the aliphatic bases, the phenyl group being somewhat negative in character compared with the positive alkyl groups. Aromatic amines form salts with acids and double salts with platinum chloride. In contact with the vapours of volatile inorganic acids they form white fumes in the air in the same way as ammonia ; they distil undecomposed. The diamines are more highly basic than the monamines. Isomerides of the amines are formed when the amino group enters side chains. 1. PRIMARY MONAMINES Primary, secondary, and tertiary aromatic monamines are distinguished by the same reactions as are used for aliphatic amines (by nitrous acid, &c. ; see p. 201). Formation, (a) Mono-, di-amines. &c., are usually obtained by reducing the nitro-derivatives with tin or stannous chloride and hydrochloric acid, or with iron and acetic acid, or with ammonium sulphide, &c. : -C 6 H 5 -N0 2 + 6H = 2H 2 + C 6 H 5 -NH 2 . The reduction may also be effected electrolytically (see later, p. 566). (b) By heating phenols (or, better, nitrophenols or naphthols) with ammo- niacal zinc chloride at 300, primary amines are readily obtained with small proportions of secondary amines : C 6 H 5 -OH + NH 3 = H.,0 + C 6 H 5 -NH 2 . (c) By heating secondary and tertiary bases (substituted amines) with concentrated hydrochloric acid at 180,. C 6 H 5 -N(CH 3 ) 2 + 2HC1 = C 6 H 5 -NH 2 . + 2CH 3 C1 ; at higher temperatures the alkyl chloride reacts with the nucleus, giving homologous amines higher than the original one : C 6 H 5 -NH 2 + CH 3 C1 = C 6 H 4 (CH 3 )-NH 2 , HC1. In the same way, trimethylphenylammonium iodide yields mesidine hydriodide, C 6 H 2 (CH 3 ) 3 -NH 2 , HI (the methyl groups of the nucleus never assume the meto-position). Properties. The primary monamines are liquid or solid and turn brown in the air. With acids they form crystalline salts soluble in water, but with carbonic acid they do not give salts, so that they may be liberated from their salts by means of sodium carbonate. With platinum chloride they form double salts (platinichlorides), e.g. (C 6 H 5 -NH 2 , HC1) 2 , PtCl 4 , which are only slightly soluble and serve for the separation of these bases. With methyl iodide they form secondary, tertiary, and quaternary com- pounds : C 6 H 5 -NH-CH 3 , HI - - C 6 H 5 -N(CH 3 } 2 , HI C 6 H 5 -N(CH 3 ) 3 I ; the base can easily be separated from the acid by caustic soda. Benzaldehyde reacts with aniline, forming benzylideneaniline : C 6 H 5 -CHO + C 6 H 5 -NH 2 = H 2 + C 6 H 5 -CH : N-C 6 H 5 , while acetaldehyde gives ethyli- denediphenyldiamine : 2C,H 6 -NH 2 + CH 3 -CHO = H 2 OT 6 AROMATIC AMINES 555 >> a O O O O 0000 OO O 60 1 cj c3 c3 c3c3 c3 c3 ci cj I|||c3c3 r; i i III 1 I ^OiOO f i CO 00 1 cstOiOi oit-- i i t^ oo t-- r~Tt< ec O O^ C^ ' O O2 C^ O^ O Ci O O^ P. j> ^OO Oi OOOO OO O I a? ooooooooooooooooooooooo as ooooo >C O 1C 00 CO 1C GO Illl ll 1 ^ i ia ,,. 6 6 a a a M o H &< a g rifrf W 00 w w 0* 556 ORGANIC CHEMISTRY The action of the organic acids on amines gives acianilides, which are decomposable by alkali : C 6 H 5 -NH 2 + CH,,-COOH = H 2 + C 6 H 5 -NH-C 2 H n O. Acotanilide When heated with chloroform and alcoholic potash, the primary amines form isonitriles (carbylamines), which have most unpleasant odours. With carbon disulphide they give thioureas. which with P 2 5 give mustard oils of the aromatic series. With nitrous acid (or nitrites) in acid solution, amines yield diazo- or diazoamino-compounds, these giving phenols when boiled with water. Where the amino-group is in the side-chain, no diazo-derivative is formed. Aniline, see later. 2. SECONDARY MONAMINES These are basic in character, not when they are purely aromatic compounds, but only when they contain also aliphatic radicals. These mixed derivatives are obtained from primary amines by treatment with methyl iodide and, if the acetylated primary base is employed, the simultaneous formation of tertiary base is avoided : C 6 H 5 -NH-COCH 3 + CH 3 I = HI + C 6 H 5 -N(CH 3 ) (COCH 3 ) ; the acetyl group may be removed by subsequent hydrolysis. The secondary bases may be separated from the tertiary by means of nitrous acid (potassium nitrite), with which the former yield nitrosamines : C 6 H 5 -NH-CH 3 + NO- OH = H 2 + C 6 H 5 -N(NO)-CH 3 , which are neutral compounds, insoluble in water. When these nitrosamines are heated with hydrochloric acid (alcoholic), the NO group passes into the benzene nucleus : C 6 H 5 -N(NO)-CH 3 gives C 6 H 4 (NO)-NH-CH 3 . Pure aromatic secondary monamines are obtained by heating the primary bases with the corresponding hydrochlorides : C 6 H 5 -NH 2 + C 6 H 5 -NH 2 , HC1 = (C 6 H 5 ) 2 NH + NH 4 C1. 3. TERTIARY MONAMINES These are formed by alkylating primary or secondary bases. Triphenylamine is obtained from bromobenzene by the action of dipotassio- aniline : 2C 6 H 5 Br + C 6 H 5 -NK 2 = 2KBr + (C 6 H 5 ) 3 N. The purely aromatic tertiary monamines are not basic in character, and hence do not form salts. They do not give isonitriles with chloroform, or mustard oils with CS 2 . With alkyl iodides they form quaternary compounds. When they are treated with nitrous acid, the NO group enters the benzene nucleus, this reaction distinguishing these bases from the tertiary bases of the fatty series. 4, QUATERNARY BASES These are analogous to the corresponding aliphatic compounds. Trimethyl- phenylammonium Hydroxide, C 6 H 5 -N(CH 3 ) 3 OH, for example, is strongly alkaline, colourless, and bitter, and is decomposed on heating. 5. DIAMINES, TRIAMINES, TETRAMINES, ETC. These are obtained by reducing the corresponding nitroamino- or polynitro- derivatives ; thus Tetraminobenzene is formed from dinitro-m-diaminobenzene. The polyamines give various reactions with nitroso -compounds of tertiary amines, with certain azo-dyes, &c. The diamines and polyamines are solid substances, which distil unde- ANILINE 557 composed and are soluble in hot water. They are colourless, but turn brown in the air with a rapidity increasing with the number of ammo-groups ; they give characteristic colorations with ferric chloride. The ORTHODIAMINES form Anhydro-bases or Benziminazoles, e.g. / NH \ C 6 H 4 \ >OCH 3 . Further, aldehydes react with the hydrochlorides |X N of diamines, forming Anhydro-bases or Aldehydo-bases. Glyoxals yield Quinoxaline, &c., while nitrous acid gives Azimino- / NH \ compounds, e.g. Aziminobenzene (aminoazophenylene), C 6 H 4 \ 2^' \N^ m-DIAMINES give, with nitrous acid, yellowish brown colouring-matters (Bismarck brown : sensitive reaction). With diazobenzene chloride they yield azo-dyes (chrysoidin). When oxidised together with p-diamines, they give a blue colour which becomes red on boiling. p-DIAMINES, when oxidised with Mn0 2 + H 2 S0 4 , yield quinoiie, C 6 H 4 2 , and a bomologue with a peculiar odour ; some of them give colouring-matters when treated with solutions of hydrogen sulphide and ferric chloride. ANILINE (Aminobenzene, Phenylamine), C 6 H 5 -NH a , was discovered in 1826 by Unverdorbeii among the products of the dry distillation of indigo and was called crystalline, since with acids it readily formed crystalline masses. It was- then found also by Runge in 1834 in coal-tar, and he named it kyanol or blue oil, since with hypochlorite it gave a blue coloration and its salts a violet coloration. In 1841 Fritsche obtained it by distilling indigo with potash, and he termed it, after the native name of the plant, " anil," aniline. In 1842 Zinin gave the name benzidam to the product obtained by reducing nitrobenzene with ammo- nium sulphide. The identity of these various substances and their true con- stitution was proved by Hofmann in 1843. Industrially it is prepared by treating nitrobenzene with nascent hydrogen produced by the action of hydrochloric acid on iron filings or, better, turnings, as was proposed in 1864 by Bechamp, who first used acetic acid in place of hydrochloric : C 6 H B .N0 2 + 6HC1 + 3Fe = C 6 H 5 .NH 2 + 3FeCl 2 + 2H 2 O. The quantity of HC1 consumed is, however, only one-fortieth of the theoretical amount, so that after a certain point the reduction is perhaps continued by the action of the iron on water in presence of ferrous chloride: 2Fe + C 6 H 5 -N0 2 + 4H 2 = 2Fe(OH) 3 + C 6 H 5 .NH 2 . The apparatus for manufacturing aniline consists of a cast-iron cylinder (the lower half is furnished with a discharge tap and is replaceable, as it corrodes rapidly) provided with a cover, through which pass a vertical stirrer worked by toothed wheels and a direct-steam coil. The cover is also fitted with a reflux condenser and a hopper with a wooden plug for the introduction of the iron turnings. A tube fixed laterally to the lower part of the reflux condenser carries off the aniline distilling with the steam to a condensing coil on one side. The operation is carried out as follows : 300 litres of water, 180 kilos of iron turnings, and 60 kilos of concentrated hydrochloric acid are kept stirred in the cylinder while 750 kilos of nitrobenzene are introduced. The reaction is started by a jet of direct steam, and is afterwards maintained by gradually adding moist iron turnings up to a total quantity of 650 kilos ; these additions are made over a period of 6 to 7 hours and are arranged so that the mass is kept hot, but the reaction is allowed to calm down before fresh iron is introduced. If the reaction becomes violent, benzene and ammonia are formed instead of aniline. A further quantity of 100 to 150 kilos of iron turnings is added, nitrobenzene evaporating with the water is condensed in the reflux condenser. At the end of the operation the vessel contains aniline, aniline hydrochloridc, ferric oxide and o- and p-toluidines, together with a little unaltered nitrobenzene and some impurities sucll as azobenzene, &c. Thick milk of lime is then added until the reaction is strongly 558 ORGANIC CHEMISTRY alkaline, and the mass distilled with superheated direct steam. The condensed distillate separates into two layers, the lower one of aniline and the upper one of water containing 2 to 3 per cent, of aniline in suspension or solution ; this lower layer is used in the reduction of subsequent quantities of nitrobenzene. The decanted aniline is purified by distillation from an iron still. The decomposition of the aniline hydrochloride by milk of lime takes place according to the equation : 2C 6 H 6 .NH 2 , HC1 + Ca(OH) 2 = CaCl 2 + 2H 2 O + 2C 6 H 6 .NH 2 . A purer product is obtained by rectification of the aniline in a vacuum apparatus. It has also been proposed (Ger. Pat. 184,809) to reduce nitrobenzene by means of sodium bisulphite in the hot. At one time the nitrobenzene employed was obtained from crude 90 per cent, benzene containing toluene, the resultant product being a mixture of aniline and toluidine, which served well for the preparation of certain dyes. But nowadays it is often regarded as preferable to start from pure benzene and pure toluene separately and to mix the aniline and toluidine subsequently in the required proportions. Aniline can also be obtained by other processes which have not yet been applied on a large scale, e.g. by passing a mixture of nitrobenzene vapour with excess of hydrogen (or water-gas) over reduced copper turnings, heated to 300 to 400 ; the copper acts as a catalyst and remains unchanged (Ger. Pat. 139,457). Some importance is now being assumed by the electrolytic process, according to which nitro -derivatives can be converted into amino -derivatives in presence of metallic salts (e.g. copper salts), which also separate at the cathode (see p. 566). Aniline is a liquid which boils at 183 to 184, has the sp. gr. 1-024 at 16, and solidifies at 8 (or 20 if impure). It is colourless and refractive, but becomes brown in the air at a rate increasing with the proportion of impurities present. It is soluble in alcohol, ether, benzene, fatty oils and, to a slight extent (1 : 30) in water, and it dissolves sulphur (in the hot), phosphorus, camphor, indigo, a little water (in the hot), &c. ; it is readily oxidisable. It distils easily and completely in steam, and its vapour is somewhat poisonous * and combustible. As a base it is weaker than ammonia in the cold but stronger in the hot, but its aqueous solution does not react with litmus or turmeric paper. Although it is a weak base, it precipitates salts of zinc, aluminium, and iron, and in the hot it displaces ammonia from various salts. With formaldehyde it gives a characteristic (for aniline and for the aldehyde) condensation product, (C 6 H 5 - N : CH 2 ) 3 , melting at 40. With chloride of lime a solution of aniline becomes intensely blue if pure or violet if impure (sensitive reaction), the colour rapidly changing to brown ; if the aniline solution is very dilute this coloration does not appear, but a red colour will then form on further addition of a few drops of ammonium hydrosulphide, minimal traces (1 : 250,000) of aniline being thus detectable. Aniline or one of its salts forms p -amino - benzenesulphonic acid with concentrated sulphuric acid, but in presence of a drop of potassium dichromate solution a fine blue colour is produced which disappears very rapidly ; in dilute solution a green and then a black colour (aniline black) is formed. Different methods of oxidising aniline give varied products : azobenzene, nitroso- and nitro-benzene, /3-phenylhydroxylaniine, p-aminophenol, quinone, p-aminophenylamine, violaniline (with arsenic acid). Oxidation of a mixture of aniline and toluidine yields fuchsine, while a mixture of aniline and p-diamine gives safranine. 2 Chlorine transforms dry aniline into 1 Aniline acts on the nervous system, and even when its action is slight the edges of the lips are turned bluish and an effect similar to drunkenness is produced, but the face becomes pale and the appetite fails ; in such cases Epsom salts are administered as purgative, alcoholic liquors being harmful. Clothes soaked in aniline may produce serious poisoning, the lips becoming dark blue or even black, and giddiness so acute as to cause collapse When this happens, recourse should be had to excitants or ablution or to- small doses of ether administered internally. Benzene and nitrobenzene vapours are also injurious to health. For the making of aniline black and other dyes, the following qualities of aniline are placed on the market aniline oil for Hue, which is almost pure aniline, b.pt. 182 to 186, sp. gr. 1-034 to 1-036 ; aniline oil for red, con sisting of about 1 part of aniline and 2 parts of o- and p-toluidines and boiling at 190 to 198 ; aniline oil for stfranine~'sj>. gr. 1-032 to 1-034, containing 35 to 50 per cent, of aniline and 50 to 65 per cent, of o-toluidine Aniline oil is tested commercially by measuring the fractions distilling at different temperatures from 100 grins- ANILINE DERIVATIVES 559 a tarry substance, while in presence of water trichJoraniline and trichlorophenol are formed. The action of calcium hypochlorite on a solution of aniline in chloroform yields azobenzene. In 1909 Germany imported 639 quintals of aniline and exported 78,835 quintals (70,452 in 1908). According to the official statistics, Italy imported the following quantities of aniline oil and salt : 426 quintals in 1908 ; 577 in 1909 ; 3695, of the value of 20,680, in 1910. But these figures are obviously inaccurate, and private information shows that the Italian consumption (nearly all imported) must exceed 8000 quintals per annum, one-half of this being aniline salt (hydrochloride) and the other oil. In 1911 England exported aniline and toluidine oils to the value of 39,814. Chemically pure aniline costs 2*. to 4s. per kilo ; aniline oil for blue and for black costs 104s. to 112s. per quintal, and that for red 120s. to 152s. ; lower prices than these some- times prevail. Aniline salt costs about 10s. less per quintal than the oil. There is no import duty on the oil or salt in Italy. Some of the more important salts and derivatives of aniline and its homologues are as follow : ANILINE HYDROCHLORIDE (Aniline Salt), C 6 H 5 -NH 2 , HC1, is obtained pure and dry in white crystals by passing a current of dry hydrogen chloride into an ethereal solu- tion of aniline. It melts at 198 and partly sublimes, and boils unchanged at 245 ; it dissolves readily in water or alcohol, but is insoluble in ether. For price, see above. It is prepared industrially by neutralising aniline at 100 with concentrated hydro- chloric acid, free from chlorine. After standing for some days, crystalline aniline salt separates out, this being centrifuged and dried at 50 ; the mother- liquors are then evapo- rated and crystallised. In the air the white scales assume a reddish or greenish tint. In presence of HC1 its aqueous solution imparts a yellow colour to pine-wood or elder-pith. ANILINE SULPHATE, (C 6 H 5 .NH 2 ) 2 , H 2 SO 4 , is only slightly soluble in water. Various other salts of organic and inorganic acids are known. ANILINE PLATINICHLORIDE, (C 6 H 6 -NH 2 , HC1) 2 , PtCl 4 , forms yellow leaflets dissolving readily in water and, to a less extent, in alcohol. METHYLANILINE, C 6 H 5 -NH-CH 3 , is obtained by heating aniline hydrochloride with methyl alcohol (free from acetone) at 200 in an enamelled iron autoclave. It is a colourless liquid, sp. gr. 0-976 at 15, b.pt. 191, with an odour resembling, but stronger than, that of aniline. With chloride of lime it gives first a violet and then a brown colora- tion. The corresponding nitrosamine, C 6 H 6 -NO-CH 3 , is obtained by methylating phenyl- nitrosamine or by treating methylaniline with nitrous acid. It forms a yellow oil which distils unchanged only in a current of steam and gives Liebermann's reaction, characteristic of the nitrosamines and of various nitroso -derivatives ; this reaction consists in the forma- tion of a dark blue coloration when the nitroso -compound is heated with phenol and sulphuric acid and the liquid then diluted with water and neutralised with potash. DIMETHYLANILINE, C 6 H 5 -N(CH 3 )2, is a mixed tertiary amine and is obtained by heating aniline hydrochloride with methyl alcohol, methyl chloride being formed as an intermediate product and reacting with the aniline. If, however, dimethylaniline hydro - chloride is heated with gaseous hydrogen chloride at 180, methyl chloride and aniline are formed. When dimethylaniline is heated to a high temperature, the alkyl groups pass into the nucleus. The hydrogen in the para-position of these dialkylamines is readily replaceable by different groups ; thus, the action of nitrous acid yields p-nitrosodimethyl- aniline, which forms green crystals and gives a yellow hydrochloride. Permanganate converts the NO group into NO 2 , giving nitrodimethylaniline (m.pt. 162), while boiling with caustic soda results in the elimination of dimethylamine and the formation of nitroso- phenol, NO-C 6 H 4 -OH. It gives a straw-yellow coloration with chloride of lime and reacts with aldehydes and various other compounds. DIPHENYLAMINE, C 6 H 5 .NH-C 6 H 5 , is obtained by heating aniline with its hydro- chloride : C 6 H 5 .NH 2 , HO + C 6 H 5 .NH 2 = NH 4 C1 + C 6 H 5 .NH.C 6 H 5 . of the oil in a suitable distilling flask fitted with a thermometer graduated in fifths of a degree from 150 to 225 the heating being carried out on a sand-bath. The best qualities of aniline oil give 95 to 98 per cent of distillate between 182 and 185. It is also advisable to make small dyeing tests with aniline black in order to ascertain which of the different aniline oils and salts on the market gives the finest and most intense black (tee later, Dyeing Proce'ieg). 560 It melts at 54 and boils at 310, and forms a very sensitive reagent for the detection of traces of nitric acid, with which, in presence of concentrated sulphuric acid, it gives an intense blue coloration (also given with nitrous acid and various oxidising agents ; see Detection of Nitrates in Water, voL i, p. 214). Various nitro- and nitroso-derivatives are known, as well as triphenylamine, N(C 6 H 5 ) 3 , which crystallises in large plates melting at 127 and distils unchanged. BENZYLANILINE (Benzylphenylamine), C 6 H 5 -CH 2 .NH.C 6 H 5 , is obtained either by heating benzyl chloride (1 mol.) with aniline (2 mols.) or by reducing thiobenzanilide, u 6 H 6 .CS-NH-C 6 H 5 . It forms crystals melting at 33 and boils at 310. ANILIDES are derivatives of aniline in which one or both of the hydrogen atoms of the amino-group of aniline are replaced by one or two inorganic or organic acid residues ; in the latter case, compounds of considerable interest are formed. ACETANILIDE (Antifebrin), C 6 H 5 -NH-COCH 3 , is obtained by boiling a mixture of aniline and glacial acetic acid for a couple of days in an earthenware vessel fitted with a reflux condenser : C 6 H 6 .NH 2 + CHg-CO-OH = H 2 + C 6 H 5 .NH.COCH 3 . It is purified by repeatedly crystallising or distilling, best in vacua. It melts at 113, boils at 295, and dissolves in 174 parts of cold or 18 parts of boiling water or in 3^ parts of alcohol ; it is readily soluble in ether or chloroform. The hydrogen atom united to nitrogen can be replaced by metals (Na, K, &c.). It causes considerable lowering of the temperature of animal organisms, and is hence used as an antipyretic. It costs about 2s. 6d. per kilo. Di- and Tri-acetanilides have analogous properties, and Methylacetanilide, C 6 H 5 -N(CH 3 )'COCH 3 , is used under the name of exalgin as a specific against headache. PHENYLSULPHAMINIC ACID, C 6 H 5 -NH-SO 3 H, is obtained by the action of sulphur trioxide on the amine, and is very unstable except in the form of salts. CHLORACETANILIDE, C 6 H 4 Q.NH-COCH 3 , exists in three isomeric forms: the or tho -compound, melting at 88 ; the meta-, at 72-5 ; and the para-, at 172. The chloro- and bromo -derivatives of acetanilide and other anilides are obtained by the action of chlorine or bromine on the anilide or by the interaction of acctyl chloride and the substi- tuted anilines. Another series of isomerides is that in which the substitution is in the acid group, e.g. Phenylchloracetamide, C 6 H 5 NH CO CH 2 C1 (m.pt. 134), which is obtained froni chloracetyl chloride and aniline. Phenyldichlor- (m.pt. 118) and phenyltrichlor- acetamide (m.pt. 82) are also known. NITRACET ANILIDE, NO 2 -C 6 H 4 -NH-COCH 3 . The three isomerides are obtained by the action of acetyl chloride on the corresponding nitranilines ; the o-compound melts at 92 (yellowish crystals), the m- at 142, and the p- at 207. PHENYLACET ANILIDE (Diphenylacetamide), (C 6 H 5 ) 2 N-CO-CH 3 , is obtained by treating a benzene solution of diphenylamine with acetyl chloride ; it melts at 99-5. BENZANILIDE (Phenylbenzamide), C G H 5 .NH.c6c 6 H 5 , is prepared from benzoyl chloride and aniline and melts at 162. It is very stable, but is decomposed by fusion with alkali. It is insoluble in water, but dissolves in alcohol. PHENYLGLYCOCOLL (Phenylaminoacetic or Anilidoacetic Acid), C 6 H 5 .NH-CH 2 - CO 2 H, is obtained by protracted heating of chloroacetic acid (1 mol.) and aniline (2 mols.) with water. It forms crystals melting at 127, gives characteristic mercury and copper salts, ,CH 2 and when heated at 150 gives up water and yields the anhydride C 6 H 5 N\ | , melting at X CO 263. HOMOLOGUES OF ANILINE, POLYAMINES, AND THEIR DERIVATIVES (see Table, p. 555) ORTHO- and PARA-TOLUIDINES, CH 3 .C 6 H 4 -NHo, are obtained by reducing the corresponding nitro -com pounds. Since the three isomerides are formed simultaneously in the nitration of toluene, reduction yields a mixture of the three toluidines (m-toluidine in small amount). In order to separate them, the mixture is poured into a solution of oxalic acid containing hydrochloric acid and the liquid heated to boiling ; the p-toluidine oxalate, which is only slightly soluble in water and insoluble in ether, is then separated, the filtrate containing the soluble hydrochlorides of tho other toluidines. Also Wiilfing PHENYLENEDIAMINES 561 has shown that only amines which have the para-position free can be converted (by HC1 + NaNO 2 ) into the corresponding aminoazo-derivatives, the unaltered p-toluidine being then separable by distillation in steam. p-Toluidine can also be separated by cooling, since it freezes first. The toluidines are distinguished from aniline by the different solubilities of the nitrates, hydrochlorides, and acetyl-derivatives. p-Toluidine, like the meta-com- pound, costs double as much as the ortho-isomeride. o-Toluidine, which is also found in coal-tar, is a liquid (sp. gr. 1-09) boiling at 199 and turning brown in the air. p-Toluidine is a solid melting at 43, and boils at 198 ; it is sparingly soluble in cold water, but dis- solves readily in alcohol, ether, or benzene. The toluidines are used in the manufacture of dyes. m-TOLUIDINE is obtained indirectly by nitrating acetylated p-toluidine, the compound CH 3 / \NH-COCH 3 being thus formed ; the acetyl-group is then elimi- ~N0 2 nated by boiling with hydrochloric acid and the amino-group by diazotisation. Reduction of the resultant m-nitrotoluene yields m-toluidine, which is a colourless oil (sp. gr. 0-998 at 25) boiling at 197. The crude product costs 4s. per kilo and the pure ten times as much. XYLIDINES. Six isomerides are known (see Table, p. 555), and all are formed together by nitrating crude xylene and reducing the resulting nitro -compounds ; the most im- portant is m-xylidine. Various methods of separating the different xylidines are known, almost all of them being patented and based on the varying solubilities of the acetates and hydrochlorides of p- and m-xylidines. The separate isomerides are prepared pure from the corresponding pure nitro-compounds. BENZYL AMINE, C 6 H 5 -CH 2 -NH 2 , is isomeric with the toluidines and behaves like the amines of the aliphatic series. It is obtained together with di- and tri-benzylaminc by heating benzyl chloride with ammonia. It is a colourless liquid of ammoniacal odour and boils at 185 ; it has an alkaline reaction and is a more energetic base than aniline, the amino-group being further removed from the benzene nucleus, which has a somewhat negative (acid) influence. PHENYLENEDIAMINES, C 6 H 4 (NH 2 ) 2 , are obtained by reducing the corresponding dinitrobenzenes or nitroanilines with iron and hydrochloric acid. m-Phenylenediamine is also obtained by electrolysing m-nitroaniline in aqueous saline solution in presence of a cathode of copper or of powdered copper (Ger. Pat. 131,404). It forms acicular crystals melting at 63, boils at 287, and readily undergoes change in the air ; its hydrochloride is, however, stable. It is used in the manufacture of dyes and also as a reagent for detecting traces of nitrous acid, with which it forms a brownish yellow coloration (Bismarck brown), p-Phenylenediamine is obtained by the reduction of aminoazo benzene (dissolved in aniline) with hydrogen sulphide, or, more easily, by heating p-dichlorobenzene or p-chloraniline with ammonia in presence of a copper salt (Ger. Pat. 204,408). It melts at 147, boils at 267, and forms crystals which are soluble in water and blacken a little in the air ; when pure it costs 40s. per kilo, the commercial product being sold at about 14s. As well as for making dyes, it is frequently employed for dyeing hair by oxidising it with hydrogen peroxide, but its use for this purpose should be .prohibited owing to its poisonous properties (see below, p-Tolylenediamine). Its asymmetric dimethyl-derivative, NH 2 .C 6 H 4 -N(CH 3 ) 2 , is used in presence of ferric chloride to detect traces of hydrogen sulphide (methylene blue being formed). o-Phenylenediamine is of no practical importance. Commercial m-phenylenediamine costs about 6s. and its hydrochloride 7s. per kilo, the pure products costing about six times as much. TOLYLENEDIAMINES, C 6 H 3 (CH 3 )(NH 2 ) 2 . The most common of these is the o : p- compound, i.e. the one with the amino-groups in the 2 and 4 positions and the methyl group in the position 1. It is obtained by reducing the corresponding din itro toluene (see p. 550) with iron and hydrochloric acid and is used for making dyes and, together with sodium sulphite, for dyeing hair, as it does not seem to be injurious to health, as p-phenylene- diamine is. It costs about 16s. per kilo. NITROANILINES. Concentrated nitric acid acts very energetically on aniline, and in order that the nitro-groups may be introduced into the benzene nucleus without the amino-group being attacked, either the ammo group is acetylated or the nitration is carried out in presence of a large proportion of 562 ORGANIC CHEMISTRY concentrated sulphuric acid. In the former case, ortho- and, to a still greater extent, para-Nitroacetanilide, N0 2 -C 6 H 4 -NH-C 2 H 3 0, are obtained, the acetyl group being then removed by hydrolysis with HC1 or KOH ; in the second case a mixture of m- and p-nitroanilines, together with a little of the ortho- compound, are obtained. The ortho- and meta -derivatives distil unchanged in steam. Boiling with alkali results in the elimination of the amino-groups and the formation of nitrophenols. PICRAMIDE, (NO 2 ) 3 C 6 H 2 -NH 2 , is a yellow substance melting at 188 ; on hydrolysis it gives picric acid. H. NITROPHENOLS, AMINOPHENOLS NITROPHENOLS. The ortho- and para -compounds are obtained mixed by treating phenol with dilute nitric acid, a larger proportion of the para- derivative being formed in the cold and of the ortho- in the hot. The latter is volatile in steam, and can hence be readily separated from the former. m-Nitroaniline gives m-nitrophenol only by passing through the diazo- compound, but o- and p-nitroanilines give the corresponding nitrophenols when simply fused with potash. Nitrophenols are more markedly acid than the phenols and decompose the alkali carbonates, forming Nitrophenoxides. _N0 2 PICRIC ACID (Trinitrophenol), N0 2 ^ /OH, was discovered in 1771 ~N0 2 by Amato di Welter, but was first used as a dye and much later as an explosive. It is formed by the action of concentrated nitric acid on various substances, such as silk, wool, indigo, &c., and by the oxidation of s-trini- trobenzene with potassium ferricyanide. Further nitro-groups cannot be introduced directly into picric acid. It is prepared industrially as follows : equal weights of sulphuric acid (66 Be.) and pure phenol are heated at 120 in a cast-iron vessel and continually stirred until a small portion of the mass dissolves in water without separation of phenol. The phenolsulphonic acid thus obtained is poured into two parts of cold water and the solution introduced gradually into earthenware jars containing 65 per cent, nitric acid (sp. gr. 1-400) in the proportion of 3-5 parts per 1 part of phenol. The jars are surrounded by a water-bath and are covered over so that the nitrous fumes, which are at first freely evolved, may be drawn off. Towards the end of the reaction the water-bath is heated to boiling. The stages of the process are represented by the following equations : (1) C 6 H 5 -OH + H 2 SO, t = H 2 + OH-C 6 H 4 -S0 3 H ; (2) OH-C 6 H 4 -SO 3 H + 3HNO 3 = 2H 2 + H 2 SO 4 + OH-C 6 H 2 (N0 2 ) 3 . When the mass is cool it solidifies, and it is then centrifuged and washed with a little water ; by this means the picric acid crystals can be efficiently separated from the mother- liquor. The acid can also be prepared by the following process, the details of which are kept secret by the various manufacturers : To pure crystallised phenol (m.pt. 40), fused in a number of pear-shaped retorts by means of indirect steam, is added a mixture of nitric and sulphuric acids in proportions varying in different works. When the reaction is finished, the clots of picric acid formed are fused and allowed to fall into a trough con- taining cold water, with which they are kept stirred, the water being repeatedly renewed until washing is complete. The crystallised picric acid is centrifuged, again melted and run into cold water, the size of the yellow scales separating out increasing with the tem- perature of the fused acid ; the crystals are then centrifuged, spread out on tables, and dried in a current of air at 40 to 60. PICRIC ACID 5C3 A suggestion has been made to prepare picric acid in the cold, as follows (Fr. Pat. 345,441) : 1 part of crude phenol is stirred into a mixture of 10 parts of nitric acid (sp. gr. 1-4) with 3 parts of denatured alcohol, the mass being poured into hot water at the end of the reaction ; the yield is good, but part of the alcohol is oxidised and lost. When phenol is dear, aniline is sometimes used, being converted into the sulphonic acid, diazotised, and treated with the theoretical quantity of nitric acid (Ger. Pat. 125,096). Properties. Picric acid forms yellowish, very bitter, and somewhat poisonous leaflets, which melt at 122'5 and have the sp. gr. 1-7635 or, in the fused state, 1-62. It burns without exploding, but if it is heated in a closed vessel, or if its vapour is superheated, it may explode with great violence. . In the open, mercury fulminate is not able to explode it, a detonator of dry guncotton (or lead picrate) with a mercury fulminate cap being necessary. When it is exploded in a closed vessel, its shattering effect is double that of dynamite. One hundred parts of water dissolve O626 part of picric acid at 5, 1-161 part at 15, 1'225 part at 20, or 3' 89 parts at 77. It is readily soluble in alcohol, and benzene dissolves 8 to 10 per cent, of it at the ordinary tempera- ture. In aqueous solution it is dissociated to some extent and shows a marked acid action. The yellow colour of its aqueous solution is due to the anion ; in light petroleum it gives a colourless solution, and is hence noil -ionised. It is non-volatile in steam. Its hydroxyl-group is highly reactive, owing to the presence of the three nitro -groups. The potassium and ammonium salts are exploded by percussion, whilst the free acid requires a detonator. With many aromatic hydrocarbons it forms well-crystallised, molecular compounds which serve for the identification and separation of the hydro- carbons ; picric acid is eliminated from these compounds by ammonia. With potassium cyanide it gives a characteristic and sensitive coloration (isopurpuric acid). With nitron acetate it gives a precipitate of nitron picronitrate, C2oH 16 N 4 , C 6 H 3 O(N0 2 ) 3 , which is insoluble in extremely dilute aqueous solutions acidified with sulphuric acid, and can be filtered off, washed with water, dried at 110, and weighed. N=C v NITRON has the structure N-C 6 H 6 J>N-C 6 H 5 , and in presence of acetic acid C 6 H 5 -N CH precipitates N0 3 ions from very dilute solutions even when nitrites are also present. The decomposition of picric acid on explosion has not been thoroughly investigated, but is represented approximately by the equation : C 6 H 2 (OH)(N0 2 ) 3 = 6CO + H 2 + 3N + H ; the acid is hence too poor in oxygen to give the maximum effect, the carbon monoxide and hydrogen not being oxidised. Uses. Picric acid is employed in the preparation of certain organic compounds and was at one time used for dyeing silk and wool yellow, but the colour is not very stable. It is now mostly used as an explosive, either as acid or in the form of ammonium or potas- sium salt, these exploding at 310 or on percussion (see Explosives, pp. 215 et seq.). Melinite, a very powerful explosive suggested by Turpin for filling grenades, is merely picric acid which has been fused in a tinned vessel ; it is poured into the empty grenade, the interior of which is also tinned. From ammonium picrate and ammonium salts of trinitrocresol, sometimes with addition of potassium nitrate, powerful and stable explosives are obtained, these bearing various names (lyddite, ecrasite, &c.). In 1905 Germany produced 10,350 quintals of picric acid (at 9 per quintal) for export alone. AMINOPHENOLS, NH 2 -C fi H 4 -OH, are crystalline, colourless substances, which turn brown and resinify in the air. They are formed by reducing 564 ORGANIC CHEMISTRY nitrophenols and form salts only with acids. p-Aminophenol, melting at 183, . is obtained by electrolytic reduction of nitrobenzene in acid solution ; it is stable in a solution of sodium sulphite and is used thus as a photographic developer under the name rodinal. Methyl-p-aminophenol, or metol, also serves as a developer. Aromatic photographic developers (see vol. i, p. 800) should contain several hydroxyl- or ammo-groups, or at least one group of each kind ; if the hydrogen of the hydroxyl- and amino -groups is partly replaced, the compounds lose their developing properties, unless some of these groups remain unchanged. AMINO ANISOLES (Anisidines), NH 2 -C 6 H 4 -OCH 3 , and Phenetidines, NH 2 .C 6 H 4 . OC 2 H 5 , are used in making azo-dyes and are similar to aniline. Glacial acetic acid yields, for example, PHENACETIN (Acetyl-p-phenetidine), CH 3 .(X).NH.C 6 H 4 .OC 2 H 5 , Phenetole being C 6 H 5 -OC 2 H 5 . Phenacetin is used as an antipyretic and antineuralgic and forms colourless and tasteless white crystals, m.pt. 135, which are soluble in alcohol and slightly so in water. It costs about 6s. per kilo. DIAMINOPHENOL (1 : 2 : 4) is obtained from the dinitrophenol and forms the photo- graphic developer, amidol (see above). DIHYDROXYDIAMINOARSENOBENZENE is the product prepared by Ehrlich and Bertheim as hydrochloride and placed on the market in 1910 under the name salvarsan or 606. It is a straw-yellow powder, dissolving in water to an acid solution, and it contains 34 per cent, of arsenic. It also bears the name Hata, since it was Dr. Hata, of the Ehrlich Institute, who first injected it into animals and found it to be highly efficacious in cases of syphilis in rabbits, who were able to withstand a certain dose of the preparation. It was applied to man by Alt in the case of a syphilitic paralytic, and was subsequently largely used with success by Iversen. Salvarsan is a specific remedy for syphilis, the spirochetes being killed in 24 to 48 hours and the syphilitic symptoms disappearing rapidly even where treatment with mercury or iodine is without effect. The cure seems, however, to be very painful, relapse and secondary effects sometimes occurring. The firm of Meister, Lucius und Briining (Hochst, near Frankfort), who make salvarsan, sold a million pounds' worth of it in 1911. THIOPHENOL (Phenyl Hydrosulphide), C 6 H 5 -SH, is obtained by heating phenol with phosphorus pentasulphide or by reducing benzenesulphonic chloride, C 6 H 5 -S0 2 C1. It is a liquid of very unpleasant odour and exhibits the characters of the mercaptans. It readily forms salts, that of mercury, (C 6 H 5 S) 2 Hg, for example, crystallising in needles. When oxidised in ammoniacal solution, thiophenol yields Phenyl Bisulphide, (C 6 H 5 ) 2 S 2 , melting at 60. Phenyl Sulphide, (C 6 H 5 ) 2 S, is obtained from thiophenol and diazobenzene chloride, and has an alliaceous odour. AMINOTHIOPHENOLS, NH 2 -C 6 H 4 -SH. The ortho-compound readily forms con- /Nv densation products of the type C 6 H 4 <' /CH, or of greater complexity, such as primu- NgK line (a yellow dye diazotised on the fibre), which is obtained by heating p-toluidine with sulphur and then sulphonating. When heated with sodium sulphide and sulphur, p-amino- phenol yields Vidal black, which colours cotton in an alkaline and reducing bath of sodium sulphide. The black thus obtained is brilliant and stable, like most of these sulphur dyes. PHENOLSULPHONIC ACID, OH-C 6 H 4 -SO 3 H, is obtained from phenol and con- centrated sulphuric acid or, better, from benzenesulphonic acid. The ortho- and para- compounds are preferably form 3d, and the former is transformed into the latter on heating. The meta -derivative is prepared indirectly. The ortho -compound is used as an antiseptic under the name sozolic acid or aseptol. AZO-DERIVATIVES I. AZO-, DIAZO-, AND DIAZOAMINO-COMPOUNDS AND HYDRAZINES 1. AZO-DERIVATIVES These are intermediate reduction products of nitro -compounds and contain a characteristic group of two nitrogen atoms, each of which is united to an aromatic group. In acid solution hydrogen reduces nitro-derivatives directly to aromatic amines, but in alkaline solution two benzene nuclei condense and become joined by two nitrogen atoms. In this way the following compounds can be obtained from nitrobenzene : (1) Azoxybenzene, C 6 H 5 -N N-C 6 H 5 ; (2) Azo- \0/ benzene, C 6 H,,-N : N-C 6 H 5 ; (3) Hydrazobenzene, C C H 5 -NH-NH-C C H 5 . Reduction of nitrobenzene with zinc dust in neutral solution yields Phenyl- hydroxylamine, C G H 5 -NH-OH. When aliphatic amines are oxidised, the alkyl groups are detached in the form of acids and ammonia is generated, but the aromatic amines yield important intermediate compounds, e.g. azoxy-derivatives. AZOBENZENE (Benzeneazobenzene), C 6 H 5 -N : N-C 6 H 5 , is obtained by reducing nitrobenzene with a solution of stannous chloride in excess of potassium hydroxide or by distilling azoxybenzene with iron filings. It forms orange-red crystals melting at 68 and boils at 295 without decomposition ; it is insoluble in water and is volatile in steam. On reduction in acid solution it yields benzidine : NH/ Higher homologues, such as Azotoluene, are also known. AZOXYBENZENE is formed by oxidising aniline with potassium permanganate in alkaline solution or, better, by boiling nitrobenzene with alcoholic potash. It forms pale yellow crystals melting at 36. When heated with concentrated sulphuric acid, it is con- verted into HYDROXYAZOBENZENE : C 6 H 5 N - N C 6 H 5 > C 6 H G N : N C 6 H 4 . OH. \0/ Hydroxyazo-compounds are formed also by the action of diazo-compounds on phenols (especially resorcinol and the naphthols) in presence of alkali : C 6 H 5 .N 2 .C1 + C 6 H 5 .OK = C 6 H 5 -N : N-C 6 H 4 .OH + KC1. These compounds form yellow, red, or brown crystals, readily soluble in alcohol but insoluble in water. They are azo-dyes (tropceolins). AMINOAZOBENZENES are obtained by the following methods, which introduce the amino-group into the para-position. Aminoazobenzene itself is formed by nitrating azo- benzene and reducing the mononitroazo benzene thus obtained ; or by transposition of the diazoamino -compounds (see p. 569), and hence indirectly from diazobenzene and a primary or secondary amine ; or by coupling diazo-compounds with tertiary amines, in which case the aminic hydrogen of the aminoazo -compounds is substituted. If the aminic group cannot enter the para-position, owing to this being occupied, the reaction becomes more difficult and o -aminoazo -derivatives are formed. The interaction of diazo-compounds with m-diamines yields diaminoazobenzenes, which are yellow, red, or brown dyes and are termed Chrysoidines, C 6 H 5 .N 2 .C1 + C 6 H 4 (NH 2 ) 2 = HC1 + C 6 H 5 -N : N.C 6 H ? (NH 2 ) 2 (chrysoidine). The amino-group of p-aminoazobenzenes can also be diazotised, giving diazo-compounds, which again react with amines to form a group of substances called bisazo -compounds or tetrazo-compounds, e.g. C 6 H 5 -N: N-C C H 4 -N : N-C 6 H d -NHo ; trisazo -compounds are also known. These substances are used for Biebrich scarlet, croceine, &c. 566 ORGANIC CHEMISTRY HYDRAZOBENZENE, C 6 H 5 -NH-NH-C 6 H 5 , is obtained by reducing azobenzene or nitrobenzene with zinc dust and alcoholic potash, and forms colourless crystals melting at 126. With energetic reducing agents it gives aniline, while oxidising agents (FeCl 3 or atmospheric oxygen) convert it into azobenzene. Under the action of a strong acid it undergoes transformation, even in the cold, into Jenzidine (diaminodiphenyl) : NH-NH < > -H- NH 2 < \ / >NH 2 (benzidine) which forms a sulphate only slightly soluble in cold water. The formation of benzidine in this way shows that it contains the amino-groups in the para -positions, and this is con- firmed by the fact that this transformation does not occur with a hydrazobenzene in which the par a -hydrogen is replaced by another group. Electrolytic Reduction of Nitroderivatives. This has been studied more especially by Gattermann, Haber and Elbs, who found that, in the electrolytic conversion of nitro- benzene to aniline in acid solution, various intermediate products are formed, the primary ones being : C 6 H- 5 N0 2 - -> C 6 H 5 -NO * C 6 H 5 -NH-OH > C 6 H 5 -NH 2 : Nitrobenzene Nitrosobenzene Phenylhydroxylamine Aniline while in alkaline alcoholic solution two secondary reactions occur, the nitrosobenzene first formed reacting with the phenylhydroxylamine formed later, giving azoxybenzene : C 6 H 5 -NH-OH + C 6 H 5 'NO = H + C 6 H 5 -N - N-C 6 H 5 , \0/ this being subsequently reduced to hydrazobenzene, which reacts with the excess of nitro- benzene, forming azobenzene and azoxybenzene. The reduction of hydrazobenzene to aniline requires a tension at the cathode much greater than suffices for the formation of nitrosobenzene and phenylhydroxylamine ; with 1-47 volts, only traces of aniline are formed. 2. DIAZO-DERIVATIVES In the diazo-compounds of the aromatic series (discovered by P. Griess in 1860) the characteristic group, N 2 , is united to only one aromatic radical (aryl, Ar) and to an acid residue (X). This group therefore forms two series of compounds. (1) Diazonium salts, in which one atom of "nitrogen is pentavalent as in ammonium salts. Hantzsch showed their structure to be : Ar-N : N. (2) True diazo-compounds with two trivalent nitrogen atoms, Ar-N : N-X ; these exist in two stereoisomeric forms (see p. 22), the somewhat unstable syn-diazo-compound,s, Ar-N, and the stable anti-diazo-compounds, Ar-N. The II II X-N N-X two groups Ar and X are far apart in the cmfo'-compounds, so that they cannot easily react, these compounds hence being the more stable. The cyanide of antidiazo-p-chlorobenzene, C1-C 6 H 4 -N, is not decomposed by powdered copper N-CN and, on the other hand, cannot have the constitution of a diazonium salt, C1-C 6 H 4 -N : N, which, like ammonium salts, should be colourless (whereas CN DIAZO-COMPOUNDS 567 the cyanide is yellow) and should have an alkaline reaction and conduct the electric current in aqueous solution ; neither of these properties is shown by this cyanide, although they are found with the analogous diazoanisole cyanide, CH 3 0-C 6 H 6 -N N. CN The antidiazotates behave partly like acids and the corresponding pseudo- acids. Indeed, antidiazo -hydrate gives the reaction for hydroxyl and forms a conducting aqueous solution ; it is unstable and is converted by acids into the nitrosamine (pseudo-acid), which no longer gives the reactions for hydroxyl, does not conduct, has a neutral reaction, and in dry ethereal solution does not form the ammonium salt with ammonia (as, for example, Phenylnitro- methane does). By alkali the nitrosamine is immediately reconverted into the antidiazotate : Ar-N Ar-N-H N-OH N:O Antidiazotate Nitrosamine Preparation. The gradual addition of sodium nitrite (1 grm.-mol.) solution to a solution of the salt of the amine (1 grm.-mol.) cooled with ice results in the formation of the soluble diazonium salt : C 6 H 5 -NH 2 , HN0 3 + NO-OH = 2H 2 + C 6 H 5 -N 2 -N0 3 . Aniline nitrate Phenyldiazonium nitrate C 6 H 5 -NH 2 , HC1 + NO-OH = 2H 2 + C 6 H 5 -N 2 -C1. Aniline hydrochloride Phenyldiazonium chloride These diazonium salts are highly explosive when dry, so that they are always used in aqueous solution, when they are completely harmless. In these compounds the group C 6 H 5 -N 2 - behaves like the ammonium cation and with strong mineral acids gives neutral salts, while the salts formed with carbonic acid have alkaline reactions, since, like the alkaline carbonates (see vol. i, pp. 91 and 436), they readily undergo hydrolytic dissociation, These salts have extremely high conductivities, and hence are dissociated like potassium and ammonium chlorides, and like these, too, they form diazo- nium platinichloride, (C 6 H 5 -N 2 -Cl) 2 PtCl 4 . The hydroxide, C 6 H 5 -N 2 -OH (from the chloride + AgOH), is known, although it has not yet been isolated ; it is soluble, colourless, and strongly alkaline. All these reactions indicate the existence of a pentavalent nitrogen atom in the group N 2 . Two constitutional C 6 H 5 -N : N formulae are hence possible : C 6 H 5 - N j NX and | ; various reactions indicate the latter to be the more probable (see above). There are various ways of eliminating the nitrogen from diazo -compounds in the free state, union taking place between the benzene nucleus and the other group joined to the N 2 complex : (a) By heating the aqueous solution of a diazonium salt a phenol is formed : C G H 5 -N 2 -C1 + H = C 6 H 5 -OH + N 2 + HC1. (6) When a diazonium salt is heated with alcohol the benzene nucleus unites with the alkoxy-group : C 6 H 5 -N 2 -HS0 4 + C 2 H 5 -OH = C 6 H 5 -0-C 2 H 5 + N 2 + H 2 S0 4 ; under certain conditions, however, the alcohol is oxidised and aldehyde libe- rated along with the nitrogen : N0 2 'C 6 H 4 -N,-C1 + C 2 H 5 -OH = C 6 H 6 -N0 2 + N 2 + HC1 + CH 3 -CHO. p-NitrodiRRobenzcne chloride AceUldebyde ORGANIC CHEMISTRY (c) When a diazonium chloride is treated with cuprous chloride dissolved in concentrated hydrochloric acid (Sandmeyer), the chlorine (or other halogen) is introduced into the nucleus : C 6 H 5 -N 2 C1 = C 6 H 5 -C1 + N 2 . The same result is produced by finely divided copper, which, however, acts catalytically (Gattermann). (d) The cyanogen group is introduced into the nucleus by the action of potassium salt in presence of a copper compound : C 6 H 5 -N 2 C1 + KCN = KC1 + N 2 + C 6 H 5 -CN. Benzonitrile This is a general reaction for obtaining (by subsequent hydrolysis) aromatic acids. (e) Dry diazobenzene chloride, when treated with benzene in presence of aluminium chloride, gives diphenyl : C 6 H 5 N 2 C1 + C 6 H 6 = N 2 + HC1 + C 6 H 5 .C 6 H 5 . With tertiary amines, diazonium salts condense in the para -position, giving aminoazo-derivatives : C 6 H 5 N 2 C1 + C 6 H 6 N(CH 3 ) 2 '= HC1 + C 6 H 5 -N : N-C 6 H 4 (CH 3 ) 2 . Diazonium salts also form hydro xyazobenzenes (see p. 565). DIAZOBENZENE CHLORIDE (Phenyldiazonium Chloride), C 6 H 5 -N 2 .C1, forms colourless needles soluble in water and is obtained by the action of moist AgCl on the corresponding bromide ; the bromide is obtained in nacreous scales by the interaction of ethereal solutions of bromine and diazoaminobenzene (tribromoaniline remains in the solution). DIAZOBENZENE NITRATE (Phenyldiazonium Nitrate), C 6 H 6 .N 2 .N0 3 , is the salt which is most widely used, and is obtained by passing nitroso-nitric fumes into a cold ethereal solution of diazoaminobenzene or into an aqueous paste of aniline nitrate until this is dissolved ; to the filtered liquid are added the triple volume of alcohol and then ether until the nitrate separates in colourless needles. It is readily soluble in water but insoluble in ether, benzene, chloroform, &c. It has a strong acid reaction and is easily exploded by shock. DIAZOBENZENE SULPHATE (Phenyldiazonium Sulphate), C 6 H 5 -N 2 HS0 4 , is best obtained by treating a concentrated solution of crude diazobenzene nitrate with moderately concentrated sulphuric acid, precipitating several times with excess of alcohol and with ether, and allowing to crystallise in a desiccator. It forms crystals which are readily soluble in water and deflagrate at 100. DIAZOBENZENE PERBROMIDE, C 6 H 5 -N 2 -Br3, is prepared by the action of hydro- bromic acid and bromine water on diazobenzene salts. xN 2 -OH DIAZOBENZENESULPHONIC ACID, C 6 tt/ , is known as anhydride, X S0 8 H /N 2 C 6 H 4 C | , and is obtained by adding a mixture of sodium sulphanilate and sodium X S0 3 nitrite to dilute sulphuric acid. It forms white needles readily soluble in water, and is used to prepare azo-dyes. With KOH, phenyldiazonium hydroxide forms a potassium compound, C 6 H 5 'N 2 -OK, and hence behaves as an acid besides as a base. But as it cannot be assumed that these two functions are exhibited to such marked extents by one and the same substance, Hantzsch supposes that, in aqueous solution, it forms a mixture of phenyldiazonium hydroxide, C 6 H 6 .N.OH, HI and syn-diazobenzene hydroxide, C 6 H 5 -N : s N-OH, so that the general N reactions mentioned above would be explained thus : C 6 H 5 -N : N + H-OH = HC1 + C G H 5 .N : N-OH > C tJ H 5 .OH + N : N. I Cl DIAZOAMINO-DERIVATIVES 569 None of the reactions referred to above can be explained well without assuming the passage of diazonium salts with pentavalent nitrogen into true diazo -compounds with trivalent nitrogen ( N == N ) (see above). 3. DIAZOAMINO-DERIVATIVES These contain the aminodiazo-group, N = N NH , and are yellow, crystalline substances which do not combine with acids. They are obtained by adding to diazo-salts (freshly formed in solution) primary or secondary amines, e.g. aniline hydrochloride ; the separation of the yellow crystalline mass is hastened by addition of concentrated sodium acetate solution : C 6 H 5 -N 2 -C1 + C 6 H 5 -NH 2 = HCl + C 6 H 5 -N 2 -NHC 6 H 5 . To 2 mols. of aniline and 3 mols. of hydrochloric acid, kept cool with ice, is slowly added 1 mol. of sodium nitrite, the liquid being then precipitated with concentrated sodium acetate solution. They are also formed directly from free aniline and nitrous acid, in whicl 1 case diazobenzene hydroxide must be regarded as an intermediate product : (a) C 6 H 5 -NH 2 + HN0 2 = H 2 + C 6 H 5 -N 2 - OH ; (6) C 6 H 5 -N 2 -OH + C.H 5 -NH a = H 2 + C 6 H 5 -N 2 -NHC 6 H 5 . With nitrous acid in acid solution, diazoamino-compounds are converted into diazonium salts, the remaining aminic residue, NHC 6 H 5 . being diazo- tised : C 6 H 5 -N : N-NHC 6 H 5 + HN0 2 + 2HC1 == 2H 2 + 2C 6 H 6 -N 2 C1. When heated with aniline hydrochloride, diazoaminobenzene solution yields aminoazobenzene, which is a colouring- matter from which others are derived. In this transformation, which is common to all diazoamino-compounds, the aniline hydrochloride acts catalytically and takes no part in the reaction ; the amino-group is carried to the para-position with respect to the diazo- group, or, if this is occupied, to the ortho -position : C 6 H 5 - N : N- NHC 6 H 5 -> C 6 H 5 - N : N- C 6 H 4 - NH 2 . Diazoaminobenzene Amiuoazobcnzene It has been shown by H. Goldschmidt that the velocity constant of this transformation increases with the amount of the catalyst (aniline hydrochloride), and the catalytic powers of the different amine salts are propprtional to their degrees of dissociation in water. 4. HYDRAZINES These compounds are obtained by reducing diazonium salts with a hydro- chloric acid solution of stannous chloride : C 6 H 5 N 2 C1 + 4H = C 6 H 5 -NH-NH 2 , HCl. Phenylhydrazine hydrochloride The use of sodium sulphite in place of stannous chloride gives first the diazosulphonate, which, when treated with zinc dust and acetic acid and subsequently boiled with hydrochloric acid, gives phenylhydrazine hydro- chloride ; this salt separates out, being only slightly soluble in water and less so in acid. The three stages of the reaction are as follow : (a) C 6 H 5 N 2 C1 + Na 2 SO 3 = C 6 H 5 -N 2 -S0 3 Na + NaCl. Diazobenzcne chloride Sodium diazobenzenesulphonate (6) C 6 H 5 -N 2 -S0 3 Na + 2H = C 6 H 5 -NH-NH-S0 3 Na. Sodium phenylhyd rayanesulphonate (c) C 6 H 5 -NH-NH-S0 3 Na + H 2 = NaHS0 4 + C 6 H 5 -NH-NH 2 . Phenylhydrazine 570 ORGANIC CHEMISTRY PHENYLHYDRAZINE, C 6 H 5 -NH-NH 2 , is the most important member of this group and has a basic character, forming well-crystallised salts. It is a colourless, oily liquid which turns brown in the air ; it dissolves only slightly in water, melts at 17-5, and boils at 241 with slight decomposition. With energetic reducing agents it forms aniline and ammonia, and with oxidising agents (e.g. chloride of lime) it can form diazonium compounds, but usually nitrogen is eliminated with formation of water and benzene. It gives characteristic reactions with lactones, sugars, aldehydes, and ketones (see pp. 206 and 210). The constitution of phenylhydrazine is proved by the fact that nitrosomethylaniline, C 6 H 5 -N(CH 3 )-NO (obtained from the secondary amine, methylaniline, C 6 H 5 -NH-CH 3 , by the action of nitrous acid), on reduction, yields as.phenylmethylhydrazine, C 6 H 5 .N(CH 3 ).NH2, which can also be obtained from phenylhydrazine by the action of metallic sodium (this replaces the iminic hydrogen) and subsequently of methyl iodide : C 6 H 5 .NH.NH 2 > C 6 H 5 .N.NH 2 > C 6 H 5 .N-NH 2 . I I Na CH 3 Replacement of the aminic hydrogen by an acid residue yields hydrazides (a and^3), which give a reddish violet coloration with sulphuric acid and potassium dichromate. The hydrazides are insoluble in water and may hence be used for the precipitation of soluble acids. DIPHENYLHYDRAZINE, (C 6 H 5 ) 2 N.NH 2 , is obtained by reducing Diphenylnitro samine, (C 6 H 5 ) 2 N-NO, in alkaline solution with zinc dust and acetic acid. It is a base boiling at 34 almost without decomposition, and oxidising readily in the air ; its salts are unstable. It is insoluble in water and hence reduces Fehling's solution only slightly, even in the hot. With concentrated sulphuric acid it gives a blue coloration. The action of oxidising agents distinguishes it from the isomeric hydrazobenzene ; the latter gives azobenzene, whilst diphenylhydrazine yields in the cold Tetraphenyltetrazone, (C 6 H 5 ) 2 N-N: N-N(C 6 H 5 ) 2 , and in the hot diphenylamine and violet colouring- matters with abundant evolution of nitrogen. With nitrous acid, hydrazobenzene forms nitroso- derivatives, whilst diphenylhydrazine, like other secondary hydrazines, gives diphenyl- nitrosamine and nitrous oxide. BENZYLPHENYLHYDRAZINE, C 6 H 5 -CH 2 -N(C 6 H 5 ).NH 2 , is obtained from phenyl- hydrazine and benzyl chloride. Benzylhydrazine, C<5H 5 .CH 2 .NH-NH 2 , boiling at 135 (in vacuo), is also known. p-NITROPHENYLHYDRAZINE, obtained from p-nitraniline, forms yellow crystals and is a useful reagent for aldehydes and ketones. /3-PHENYLHYDROXYL AMINE, C 6 H 5 .NH.OH, is obtained by the gentle oxidation of aniline or the cautious reduction of nitrobenzene with zinc dust and water, and forms colourless crystals melting at 81. With oxygen it gives p-aminophenol, with oxygen azoxybenzene and with dichromate nitrosobenzene. The a-isomeride, NH 2 -OC 6 H 6 , is of little importance. L. AROMATIC ALCOHOLS, ALDEHYDES, AND KETONES In these compounds the primary alcohol group, the aldehyde group, or the ketonic group forms a side-chain to the benzene nucleus and shows all the general properties of these groups. Di- and trihydric alcohols are also known, e.g. Phthalic Alcohol (ortho) ; Xylylene Alcohol (para), C 6 H 4 (CH 2 - OH) 2 ; Phenylglycerol, C 6 H 5 -CH(OH)-CH(OH)-CH 2 - OH. BENZYL ALCOHOL, C 6 H 5 -CH 2 -OH (discovered by Cannizzaro in 1853), is isomeric with the cresols, CH 3 -C 6 H 4 -OH, and is obtained by th interaction of benzyl chloride and potassium acetate and subsequent hydrolysis of the acetyl-derivative thus obtained, or, better, by the action of aqueous potassium hydroxide on benzaldehyde : 2C 6 H 5 .CHO + KOH = C 6 H 5 .CO 2 K + C 6 H 5 .CH 2 .OH. The alcohol readily gives benzyl chloride when treated with PC1 6 . On oxidation BENZALDEHYDE 571 it gives first benzaldehyde and then benzoic acid, its constitution being thus proved. It forms simple and mixed ethers and esters. It differs from aliphatic alcohols by resinifying with sulphuric acid. It has the characters of a true alcohol and is hence insoluble in alkali (unlike the phenols). It is slightly soluble in alcohol and boils at 206. Various higher homologues are known : Tolylene Alcohols, CH 3 C 6 H 4 CH 2 OH ; Cumyl Alcohol (p-), C 3 H 7 .C 6 H 4 .CH 2 .OH, &c. Styryl Alcohol, C 6 H 5 CH : CH CH 2 OH, containing an unsaturated side-chain, is found as ester (styracin) in storax ; it forms acicular crystals with an odour of hyacinth. With alcoholic potash aromatic aldehydes are partly oxidised and partly reduced, benzaldehyde, for instance, being converted into potassium benzoate and benzyl alcohol : 2C 6 H 5 - CHO + KOH = C 6 H 5 - C0 2 K + C 6 H 5 - CH 2 - OH. With dimethylaniline or phenol these aldehydes give derivatives of Tri- phenylmethane : xC 6 H 4 -OH C 6 H 5 - CHO + 2C 6 H 5 - OH = H 2 + C 6 H 5 - CH ... P- 1 4 210 CH 3 O-C;H 4 -CO 2 H Anisic . 1 4 184 CH 3 .C,H 3 (OH)-C0 2 H. o-Hydroxytoluic 1:2-3 164 If m- ,, 1:3-4 177 P- . 1:4-3 151 * /3-m- 1:3-2 168 OH-C,,H 4 -CH 2 -CH 2 -CO,H . Hydro-p-coumaric 1 : 4 118 C,H 5 -CH(OH)-CO 2 H . Mandelic 118 C,H 6 .CH(CH a -OH).C0 2 H . Tropic . 117 C,H S -CO-CO 2 H . Benzoylformic 65 C.H 6 -CO-CH 2 -CO 2 H . Benzoylacetic . 103 C e H 3 (OH) 2 -C0 2 H Protocatechuic 1:3-4 199 C 6 H 3 (CH 3 )(OH) 2 .C0 2 H Orsellinic (1 -methyl 3 : 5-dihydroxybenzene -2-carboxylic ..... 1:3:5-2 176 C e H 2 (OH) 3 .C0 2 H . .' Gallic (3:4: 5-trihydroxybenzenecarboxylic 3:4:5-1 221 fuged, and purified by sublimation. Jessnitzer (Ger. Pat. 236,489 of 1910) proposes to oxidise with calcium hypochlorite instead of with nitric acid. Benzonitrile, C 6 H 5 -CN, found in the middle tar oils, gives pure benzoic acid when hydrolysed (Ger. Pat. 109,122). According to Ger. Pat. 136,410, benzoic and phthalic acids are readily obtainable by heating naphthol or other naphthalene derivative with fused or dissolved alkali in presence of metallic oxides (Mn0 2 , CuO, Fe 2 3 ) ; the benzoic acid is separated from the mixture by distillation. Benzoic acid is used in medicine, in making certain aniline blues, in the seasoning of tobacco, in printing textiles, and for preserving foodstuffs, 1 although it has not been shown to be harmless when used in this way ; experi- ments made in the United States in 1910 showed that doses of 1 grm. per 1 Of the various methods for detecting benzoic acid in foods, the following may be mentioned. According to Joncscu (1909) the presence of benzoic acid in milk may be shown by converting it into salicylic acid by means of 3 per cent, hydrogen peroxide diluted ten times, and then testing for salicylic acid with ferric chloride solution ( sp. gr. 1-28) diluted ten times (as in the examination of beer, see p. 179). In the case of butter, this is acidified with sulphuric acid and distilled with steam, the distillate being tested as above (see also Salicylic Acid). II 37 578 ORGANIC CHEMISTRY day of benzoic acid or sodium benzoate have no injurious effect. It costs about 3s. to 4:8. per kilo. BENZOIC ANHYDRIDE, (C 6 H 5 -CO) 2 0, is obtained by heating an alkali benzoate with benzoyl chloride : C 6 H 5 -C0 2 Na + C 6 H 5 -CO-C1= NaCl + (C 6 H 5 -CO) 2 O, or, according to Ger. Pat. 146,690, by heating nearly 2 parts of sodium chlorosulphonate, Cl'SO 3 Na, with 3 parts of sodium benzoate; by changing these proportions, benzoyl chloride (see below) may be obtained. In the cold it is not decomposed by water, but on boiling it gives benzoic acid. It costs 16s. to 20s. per kilo according to its purity. BENZOYL CHLORIDE, C 6 H 5 -CO-C1, is formed by the action of PC1 6 or POC1 3 on benzoic acid, and is obtained industrially either by the action of chlorine on benzaldehyde or from sodium chlorosulphonate (see above, Benzoic Anhydride). It is a colourless liquid which boils at 194, and has a very pungent odour. Water decomposes it very slowly in the cold (distinction from acetyl chloride) giving hydrochloric and benzoic acids. It reacts readily with many compounds in alkaline solution, introducing into them the benzoyl group (Schotten and Baumann's method). For instance, a mixture of benzoyl chloride with a little potassium hydroxide acts in the cold on aniline, forming Benzanilide,- C 6 H 5 NH CO C 6 H 5 (white compound, melting at 158, and boiling unaltered). With hydroxylamine it gives Benzhydroxamic Acid, C 6 H 6 -CO-NH-OH, which gives a violet coloration with ferric chloride. Benzoyl chloride is used in the preparation of benzaldehyde and of various dyes ; it costs about 5s. 6d. per kilo, or, in the highly purified state, 1 6s. ETHYL BENZOATE, C 6 H 5 -CO 2 C 2 H 5 , has an odour of mint, and is obtained by heating benzoic acid with alcohol in presence of sulphuric acid. BENZAMIDE,C 6 H 5 -CO-NH 2 , is obtained by the action of ammonia (or ammonium carbonate) on benzoyl chloride, or by the interaction of sulphuric acid and benzonitrile. It forms nacreous crystals melting at 130, and is soluble in boiling water. It forms metallic derivatives more easily than acetamide. BENZHYDRAZIDE, C 6 H 5 -CO-NH-NH 2 , is obtained from hydrazine hydrate and benzoic ester ; with nitrous acid, it gives / N BENZAZIDE (Benzoylazoimide), C 6 H 5 -CO-N/ || , which is readily hydrolysed, \N giving hydrazoic and benzoic acids. HIPPURIC ACID, C 6 H 5 -CO-NH-CH 2 -CO 2 H, is obtained by heating benzoic acid with glycocoll. It occurs in the urine after ingestion of benzoic acid or toluene, and is found in considerable quantities in the urine of horses and other herbivorous animals. It forms rhombic crystals melting at 187, and is soluble in hot water. CHLOROBENZOIC ACIDS, C 6 H 4 C1-CO 2 H. The halogen enters preferably the meta- position and nitric acid (in presence of concentrated sulphuric acid) gives mainly m-Nitro- benzoic Acid, NO 2 -C 6 H 4 -C0 2 H, which, on reduction, yields Azobenzoic Acids and Aminobenzoic Acids, NH 2 C 6 H 4 C0 2 H. The latter, like glycine, exhibit the functions N: N of both acids and bases ; with nitrous acid, they form Diazobenzoic Acids, C 6 H 4 <^ ~ 1 ^>. ANTHRANILIC ACID (o-Aminobenzoic Acid) is formed in the synthesis of indigo and also from phthalimide. It is prepared by boiling the potassium derivative of phthalylhydroxylamine with aqueous sodium carbonate : OH + H 2 = C0 2 + OX<^H. \ Numerous patents have been taken out for its manufacture (Ger. Pats. 130,302, 136,788, 138,188, 145,604, 146,716, &c.). It melts at 145, and is largely used in the manufacture of dyes, drugs, and perfumes. The pure product costs 64s. and the crude 20s. per kilo. JOO Anthranilic acid forms an internal anhydride, Anthranil, C 6 H 4 / | . NNH TOLUIC, XYLIC, AND CINNAMIC ACIDS 579 Of the dibasic Sulphobenzoic Acids, C 6 H 4 (S0 3 H)(C02H), the ortho-isomeride is of interest, since its imino-derivative forms SACCHARIN (o-Benzoicsulphimide), SO which is a white powder slightly soluble in water, and is 500 times as sweet as sugar. Saccharin is prepared as follows : toluene is heated at 100 with concentrated sulphuric acid and the mixture of o- and p-toluenemonosulphonic acids thus obtained converted into calcium salt and thence into sodium salt. This is dried and distilled in presence of phosphorus trichloride and chlorine and the o- and p-toluenesulphonic chlorides frozen and centrifuged to separate the crystalline para- compound from the liquid ortho-compound. With ammonia, the latter gives o-toluene- sulphamide, which is oxidised by permanganate to the potassium salt of o-benzene- sulphaminic acid and treatment of this with an acid results in the separation of crystals of saccharin. It has no nutritive value, but is harmless in the amounts usually introduced into foods ; in large doses, it is antiseptic, antifermentative, and diuretic. It melts at 224. "In Italy its consumption is prohibited, for fiscal reasons. It costs 40s. to 48s. per kilo. With alkali carbonates it forms soluble saccharin (crystallose). Saccharin was discovered in 1878, and in 1896 there were three factories in Germany producing 33,528 kilos ; in 1897 four factories made 34,682 kilos ; in 1898 five factories made 78,363 kilos and in 1899 six factories 130,287 kilos ; in 1901 189,734 kilos were made, and in 1902 174,777. Its sale is now prohibited in Germany, but six firms in Switzerland produce annually about 80,000 kilos which they despatch as contraband to various countries. In 1908 a fine of nearly 25,000 was inflicted at Domodossola for the smuggling of 623 kilos of saccharin. A pharmacist in Hungary was found in 1908 to have sold saccharin illegally to the value of 20,000. In Russia 56,332 kilos were imported in 1899 when its consumption was allowed, but after its importation was prohibited it fell to 831 kilos in 1906, although a considerable amount is introduced without the knowledge of the Customs authorities. An International Convention at Brussels in 1909 passed a resolution that all countries should prohibit the use of saccharin in foods and beverages and placed severe restrictions on its sale. TOLUIC ACIDS, CH 3 -C 6 H 4 vCOOH. The three isomerid.es are obtained by oxidising the corresponding xylenes with dilute nitric acid (see Table, p. 577). p-Toluic acid is formed also by the oxidation of turpentine. Phenylacetic Acid(a-Toluic acid), C 6 H 5 -CH 2 -C0 2 H, is isomeric with the toluic acids, but it gives benzoic acid on oxidation, whereas the toluic acids give pJithalic acids. XYLIC ACIDS, C 6 H 3 (CH 3 ) 2 -CO 2 H ; various isomerides are known (see Table, p. 577). CUMINIC ACID (p-Isopropylbenzoic Acid), C 3 H 7 -C 6 H 4 -CO 2 H, is formed in animal organisms by the oxidation of cymene, and is obtained by oxidation of Roman chamomile oil with permanganate. It melts at 117 and yields cumene when distilled with lime. CINNAMIC ACID, C 6 H 5 -CH : CH-C0 2 H, is found in storax and in certain balsams (Tolu, Peru, &c.), and remains as sodium salt when these are distilled with caustic soda. It is prepared according to Perkin's synthesis (p. 291) by heating benzaldehyde with sodium acetate in presence of a dehydrating agent (acetic anhydride) ; or by heating benzylidene chloride (benzal chloride) with sodium acetate in an autoclave at 200 ; or by the malonic synthesis from benzaldehyde and ammonia : C 6 H 5 - CHO + CH 2 (COOH) 2 = H 2 + C0 2 + C 6 H 5 - CH : CH- COOH. Cinnamic acid melts at 133 and boils at about 300. It readily forms addi- tive products owing to the double linking in the side-chain, and on this account also reduces permanganate in presence of sodium carbonate (Baeyer's reaction, p. 88). 580 ORGANICCHEMISTRY According to theory, the presence of the double linking should result in the existence of two stereoisomerides : C 6 H 5 -OH C 6 H 5 -OH II and || H-OC0 2 H C0 2 H-OH But, in addition to these, two others, Allocinnamic and Isocinnamic Acids, are known and are apparently polymorphous modifications of the maleic form, although this question studied by Liebermann, Michael, and Erlenmeyer, jun. has not yet been completely decided. Cinnamic acid costs 16s. per kilo, and is used in medicine and in the synthesis of various perfumes. PHENYLPROPIOLIC ACID, C 6 H 5 -C : . C -CO 2 H, is obtained by heating the dibromide of ethyl cinnamate with alcoholic potash; C 6 H 5 -CHBr-CHBr-C0 2 C 2 H 5 + H 2 = 2HBr + C 6 H 5 -C OC0 2 H + C 2 H 5 -OH. It forms shining needles which melt at 137 and readily sublime. Its sodum salt is used. n 1 to 3 per cent, solution as an inhalation in cases of tuberculosis, and costs 4 per kilo o-Nitrophenylpropiolic Acid, obtained in a similar manner from ethyl o-nitrocinnamate, s used in the synthesis of indigo. The basicity of these acids is given by the number of carboxyl groups, and the phenomena of isomerism are similar to those of the dihalogenated deriva- tives. The carboxyl groups may be united directly to the benzene nucleus or to side-chains, and by means of them esters, amides, acid chlorides, &c., can be formed. PHTHALIC ACID (Phenylene-o-dicarboxylic Acid), C 6 H 40, which melts at 128 and boils at 277, but sublimes considerably below this temperature ; the anhydride has a characteristic s odour and gives phthalic acid when boiled with water. With PC1 5 , phthalic acid gives Phthalyl Chloride, C 6 H 4 < ( ^ ) 2 >0, which givesPhthalide, f (CH 6 ) 2 on reduction, and Phthalophenone,C 6 H 4 >O, with benzene ( + A1C1 3 ). POLYBASIC AROMATIC ACIDS 581 When heated with phenols and sulphuric acid, phthalic anhydride forms phthaleins, e.g. C(C 6 H 4 -OH) 2 O + 2C 6 H 5 -OH = H 2 + C 6 H 4 < >0 (phenolphthalein). Phenolphthalein is a yellow powder and, being a phenol, dissolves in alkali, the solution having a violet-red colour (it forms an excellent indicator, see vol. i, p. 97). When heated with resorcinol in presence of zinc chloride at 210, phthalic anhydride yields Fluorescein (resorcinolphthalein), 6 H 3 (OH)>' which, even in very dilute alkaline solution, shows an intense greenish yellow fluorescence while by transmitted light the solution appears reddish (see Triphenylmethane Dyes). Tetrabromofluorescein, or eosin, gives alkaline solutions showing a marked reddish green-yellow fluorescence, and is used for dyeing silk red, producing a beautiful fluorescent effect ; the colour is, however, not very stable, especially towards light. CO With dry ammonia in the hot, phthalic anhydride gives Phthalimide, CgH^^pp.^NH, which is of importance since the iminic hydrogen can be replaced by metals and the latter under the action of alkyl halides, by alkyl groups. The compounds thus obtained, when heated with acid or alkali, yield phthalic acid and a primary amine free from secondary or tertiary amine (important general synthesis of primary amines, discovered by Gabriel) : r*o r*o C ti H 4 NK + C 2 H 5 Br = KBr + C 6 H 4 <^>NC 2 H 5 , and C 6 H 4 <>NC 2 H 5 + 2H 2 = C 6 H 4 <^H + (y^.j^ Phthalic acid is used in the synthesis of indigo and of dyes of the pyronine group, and is usually placed on the market as the anhydride (although called acid) at a price of 6 per quintal (65 per cent, strength) ; chemically pure, it costs 4s. per kilo. ISOPHTHALIC ACID, C 6 H 4 (C0 2 H) 2 (1 : 3), is obtained by oxidation of colophony with nitric acid, or, in general, by the oxidation of meta-derivatives of benzene. The barium salt is soluble in water. TEREPHTHALIC ACID, C 6 H 4 (CO 2 H) 2 (1 : 4), is formed by oxidising oil of turpentine or chamomile oil, or by oxidising p-toluic acid with permanganate. It is almost insoluble in water and alcohol and sublimes unchanged. It gives a sparingly soluble barium salt, but does not form an anhydride. POLYBASIC ACIDS. The tri-, tetra-, penta-, and hexa -car boxy lie acids are known, but are of little practical importance. The Benzenetricarboxylic Acids are : TRIMESIC ACID (1:3:5) derived from mesitylene ; TRIMELLITIC ACID (1:2:4) obtained from colophony ; HEMIMELLITIC ACID (1:2: 3). The Benzenetetracarboxylic Acids are : PYROMELLITIC ACID (1:2:4:5), melting at 264 ; PREHNITIC ACID (1:2:3:4), melting at 237 and forming an anhydride Mellophanic Acid (1:3:4:5), which melts, and is converted into anhydride, at 280. MELLITIC ACID (Benzenehexacarboxylic Acid), C 6 (COOH) 6 , is obtained from mellite, which is a kind of mineral found in deposits of lignite, and consists of yellow, quadratic octahedra of aluminium mellitate, C 6 (COO) 6 A1 2 + 18H 2 O. Mellitic acid may also be obtained by oxidising wood charcoal with alkaline perman- ganate. It forms needles insoluble in water and alcohol and, when heated, loses 2H 2 O and 1 4 CO CO 2CO 2 , forming Pyromellitic Anhydride, 0<< C Q>>C 6 H 2 < C Q>0, which gives Pyromellitic 2 5 Acid, C 6 H 2 (CO 2 H) 4 , with water. Mellitic acid cannot form substitution products, since all the benzene hydrogens are already substituted, but on reduction with sodium amalgam it readily yields Hydromellitic Acid, C 6 H 6 (COOH) 6 , which gives benzene when distilled with lime. 582 ORGANIC CHEMISTRY (c) HYDROXY-ACIDS AND PHENOLIC ACIDS These are formed by the methods given on p. 575 or by oxidising homologues of phenol or fusing them with alkali. The basicity is given by the number of carboxyl and phenolic groups, both of these leading to salt-formation, but the basicity towards sodium carbonate is determined by the carboxyl groups alone. When both the carboxyl and hydroxyl groups are etherified, only the former can be subsequently hydrolysed. SALICYLIC ACID (o-Hydroxybenzoic Acid), OH C 6 H 4 - COOH, is the most important of the hydroxy -acids. It is derived from salicin (glucoside of willow bark), which, when hydrolysed, first gives glucose and Saligenin : I TT H _ P TT -*' I (\ XT O 1 2 U - ^e^l^QJJ . QJJ ' ^6 1 12 U 6' Salicin . Saligenin Glucose the saligenin giving salicylic acid on oxidation. The acid is found as methyl ester in the essence of Gaultheria pracumbens. It is prepared industrially by heating sodium phenoxide with carbon dioxide in an autoclave at 140, according to Kolbe's process ; from the resulting sodium salicylate the acid is liberated by treatment with a mineral acid. In Marasse's method a mixture of phenol and potassium carbonate is heated in presence of C0 2 at 140 to 160. It forms white crystals melting at 156, subliming at 200, and distilling in superheated steam at 170. It is readily soluble in alcohol or in ether, and 1 part dissolves in 444 parts of water at 15 and in 13 parts of hot water. When / heated with POC1 3 it gives the Internal Anhydride, CgH^ I , which forms X CO a white powder softening at 110 and melting at 261. With bromine water it gives a precipitate, C 6 H 2 Br 3 -OBr, and with ferric chloride it gives a violet coloration even in alcoholic solution (phenol is coloured only in aqueous solution). With lime-water in the hot it forms a basic salt, and can thus be separated from its isomerides, which do not give this reaction. It is used as an antiseptic for preserving foodstuffs, 1 and in the manufacture of dyes and perfumes. Its sodium salt is largely used as a medicine. When heated to 200 it loses C0 2 , giving Phenyl Salicylate (salol) : 20H-C 6 H 4 -CO,H = C0 2 + H 2 O + OH-C 6 H 4 -C0 2 C 6 H 5 , which is used as an antiseptic for the intestines. Salicylic acid costs 12 14s. per quintal. In 1905 Germany exported 5018 quintals of the acid and its sodium salt. Acetylsalicylic Acid is used in medicine under the name aspirin. m and p-HYDROXYBENZOIC ACIDS give insoluble basic barium salts and yield no coloration with ferric chloride ; the m-acid is more stable to heat than the o- or p-acid. Anisic Acid, CH 3 'C 6 H 4 -CO 2 H, resembles the monobasic acids more than the phenols and is obtained from p-hydroxybenzoic acid, methyl alcohol, potassium hydroxide, and methyl iodide, the dimethyl ether obtained being then partially hydrolysed. Methyl Salicylate, OH'C 6 H 4 'C0 2 CH 3 , forms 90 per cent, of oil of Gaultheria, and is prepared artificially by the interaction of salicylic acid (2 parts) and methyl alcohol (2 parts) 1 The examination of foods for the presence of salicylic acid is carried out in the same way as with beer (p. 179). But baked starchy substances (bread, CH 2 , is PIPERONYLIC ACID, which is also obtained by oxidising piperinic acid. GALLIC ACID (3:4: 5 - ? rihydroxybenzenecarboxylic Acid), C 6 H 2 (OH) 3 -CO 2 H, occurs naturally as glucosid.es in various plants and in tea, gall-nuts, &c. It is formed by the action of mould on solutions of tannin or by boiling the latter with dilute acid or caustic soda. It reduces gold and silver salts and becomes oxidised and turns brown in the air. With ferric chloride it gives a black coloration, and, on this account, it is used in making ink l ; its reducing properties are utilised in photography. When pure it forms colourless needles (+ H 2 0) which decompose at 200 into carbon dioxide and pyrogallol. It is only slightly soluble in ether or cold water but dissolves readily in alcohol or hot water. Chemically pure gallic acid costs about 5s. per kilo. There are a number of hydroxy '-acids with hydro xyl and carboxyl groups in the side-chains : mention may be made of : (1) COUMARIC ACID (o-Hydroxycinnamic Acid), OH-C 6 H 4 -CH : CH-CO 2 H, which does not give an anhydride owing to its fumaroid structure (see Fumaric Acid), while the maleic stereoisomeride, Coumarinic Acid, is known only as salts, since in the free state it immediately forms Coumarin, CO C 6 H 4 \ | ; the latter may also be obtained by heating salicylic acid X CH : CH with sodium acetate (Perkin synthesis ; see Aldehydes). (2) MANDELIC ACID, C 6 H 5 -CH(OH)-CO 2 H ; of the various stereo- isomerides, that occurring naturally is laevo-rotatory, while that obtained synthetically (from benzaldehyde and hydrocyanic acid, with subsequent hydrolysis) is the racemic form. In solutions of the latter, certain Schizomycetes destroy the d- and leave the 1-isomeride, whilst Penicillium i INK is made by adding to aqueous gallic acid or tannin ferrous sulphate solution slightly acidified with acetic or hydrochloric acid in order to prevent oxidation and the formation of a black precipitate. To this brownish solution is added a solution of indigo-carmine or logwood to render the writing visible. When the ink is exposed on the paper to the air, it becomes black and insoluble, owing to the evaporation or neutralisation of the acid by the sizing of the paper (albumen, &c.), and the consequent ready oxidation by atmospheric oxygen, which changes the original blue colour to a deep black. To make the ink adhere without spreading, a little gum is added, and to preserve it, a little phenol [1 Ltre of this normal ink may be obtained from 23-4 grms. of tannin, 7-7 grms. of gallic acid, 10 grms. of gum, 2-5 grms. of hydrochloric acid (as gas) or 7-5 grms. of the concentrated acid, 30 grms. of ferrous sulphate, 1 grm. of phenol, and the rest water ; the liquid is left at rest for four days and then decanted from the deposit and coloured with indigo-carmine or logwood extract]. A logwood ink may be obtained as follows : 20 grms. of dry logwood extract or 30 grms. of the paste (hsematei'n) are dissolved in 800 c.c. of water and to the hot solution are added 15 grms. of soda crystals (7 grms. of Solvay soda), and then, drop by drop, and with shaking, 100 c.c. of a solution containing 1 grm. of normal potassium chromate ; this process gives a fine blue-black tint, and the ink, which does not attack steel pens, and dries easily can be preserved by a trace of phenol. Coloured inks are aqueous, gummy solutions of aniline dyes. Copying inks are similar to ordinary writing inks, but are more concentrated, and contain also glyerine, sugar, dextrin, calcium chloride, &c., by which the writing is kept moist for some time. 584 ORGANIC CHEMISTRY glaucum destroys the 1- and leaves the d-compound. Also, if the cinchonine salt of the racemic form is prepared, the d-salt crystallises out first. The Dihydroxycinnamic Acids include : CAFFEIC ACID (see Chapter on Gluco- sides), FERULIC ACID and UMBELLIC ACID (p-hydroxy-o-coumaric acid, which is readily transformed into its anhydride, umbelliferone) ; a similar acid is PIPERINIC ACID, 3 -CH : CH-CH : CH-C0 2 H, which is formed in the decomposition of piperine. The derivatives of the Trihydroxycinnamic Acids are dealt with in the Chapter on Glucosides (cesculin and daphnin from horse chestnuts and Daphne mezereum, &c., respectively). Mention may be made here of ^ESCULETIN (a Dihydroxycoumarin), ,O CO C 6 H 2 (OH) 2 ^ | , and of the isomeric DAPHNETIN, which have also been obtained ' \CH:CH synthetically. TANNIN (Gallotannic or Tannic Acid), C 14 H 10 O 9 , was studied originally by Berzelius, Pelouze, and Liebig. According" to Hlasiwetz (1867) and to U. Schiff (1873), tannin is probably a partial and mixed anhydride of gallic acid, 2 mols. of which are condensed with loss of 1 mol. of water from a carboxyl and a hydroxyl group and formation of a Digallic Acid (or ether of 3-gallolylgallic acid) : OH/\C0 2 H 2 = H 2 + OHl J OH OH OH OH According to the investigations of Nierensteln (1908) on the acetyl-derivatives and hydrolysis, commercial tannin would seem to be a mixture of digallic acid and Leucotannin (or ether of 3-hydroxygallolylgallic acid) : j CH(OH) /NcOjjH OH OH There appear, however, to be various more or less highly polymerised tannins with widely varying molecular weights. Some uncertainty still prevails as to the true molecular magnitude of tannin. Paterno (1907), from a study of the colloidal solutions, arrived at molecular weights varying from 430 to 470 (i.e. C 21 ....), while Walden (1898), by the ebullioscopic method, obtained numbers between 760 (about C 35 . . . .) and 1560 (about C 70 ....), which are sharply distinguished from that of digallic acid (332). P. Biginelli (1911), on the basis of the property shown by tannin of forming additive products with water, > alcohol and ether [e.g. C 41 H 32 O 26 , C 4 H 10 O (ether), which is stable even in a vacuum and is analogous to the oily compound, C 41 H 32 O 25 , 6C 4 H 10 O, 7H 2 0, previously obtained by Pelouze, and to others of Biginelli's compounds, namely, C 41 H 32 O 25 , 6C 4 H 10 0; C 41 H 32 O 26 , 6C 2 H 5 -OH, and C 41 H 32 26 , 5H 2 0], and also on the loss of C0 2 and H 2 with formation of Hexahydroxybenzophenone, C 13 H 10 O 7 , when tannin is heated in aqueous solution with lead dioxide (the C0 2 liberated was estimated), holds that tannin has the formula C 41 H 32 25 , and that it is probably a glucoside. It was, indeed, observed by Liebig and also by Hlasiwetz that when tannin is boiled with dilute sulphuric acid it decomposes into gallic acid and dextrin or gum (reacting with 6H 2 0) ; but Etti (1884) and Lowe found that tannin purified with ethyl acetate does not yield saccharine substances (dextrin, &c.). Tannin is widespread in nature and occurs in abundance in sumac (Rhus coriaria), gall-nuts and oak-galls, which are pathological excrescences caused by incision of the oak branches by insects. To extract the tannin, the gall-nuts are ground to a coarse powder, which is treated in a battery of diffusors similar to those used for extracting beet-sugar (see p. 451). The crude aqueous solution of tannin thus obtained is filtered through a battery of filters and extracted, TANNIN 585 in a closed copper vessel fitted with a stirrer, with crude ether (aqueous or not free from alcohol). After the liquid has been left at rest in vats for 8 to 10 days, the dense lower layer containing the tannin is decanted and freed from ether by distillation. The evaporation of the water present is effected in heated, rapidly rotating drums, or on zinc plates placed in desiccators. The dry mass is then subjected to short and gentle treatment with steam a very soft, pale, ethereal tannin being thus obtained. Tannin solutions are also concentrated under reduced pressure in multiple-effect apparatus (see Sugar, p. 461). Aqueous or Alcoholic Tannin, which is extracted by water or alcohol without being purified by means of ether, is less pure. Pure tannin forms a pale yellow light powder or sometimes crystals. It is darkened in colour by light, turns brown in the air, and dissolves in its own weight of water, double its weight of alcohol or eight times its weight of glycerol or ethyl acetate. It is almost insoluble in ether, benzene, chloroform, petroleum ether or carbon disulphide. With iron salts it forms a bluish black precipitate and with albumin or starch, a gelatinous precipitate. In aqueous solution it is dextro-rotatory ( + 15 to -f 20). According to the degree of purity, it costs from 10 to 14 per quintal, and it is used mainly, in conjunction with antimony salts, as a mordant in the dyeing of cotton with basic dyes. It is employed also in making ink and, along with gelatine, in clarifying beer and wine, forming with the gelatine a gummy precipitate which gradually settles and carries down with it the suspended matter of the liquid. In 1905 Germany exported 7040 quintals of pure tannin of the value 80,000, while in 1909 the exports were 8135 and the imports 772 quintals. In 1908 Turkey produced 70,000 tons of valonia (Quercus cegilops), the harvesting of which employs 70,000 workpeople. The United States consumed in 1909 different tanning materials to the value of 4,200,000. Imports into England amounted to : 1910 1911 Tanning extracts . , . 749,410 .. 739,329 Tanning barks . . .. 225,642 .. 243,128 Myrobolams -. . . . 225,168 .. 138,844 Sumac . . . . . 105,620 . . 103,981 Valonia 169,948 .. 121,227 Gall-nuts. . . ... 37,329 Powdered barks or woods are used, either before or after extraction, in tanning hides. These tannin extracts [from oak bark (containing 10 to 20 per cent, of tannin), mimosa (30 per cent.), leaves and twigs of sumac (15 to 30 per cent.), valonia (20 to 45 per cent.), Asiatic gall-nuts (55 to 75 per cent.), European gall-nuts (25 to 30 per cent.), divi-divi (40 per cent.), myrobolams (30 per cent.), quebracho wood (22 per cent.), horse-chestnut bark (2 to 3 per cent.), catechu or cutch (40 or 50 per cent.), &c.], are now rationally prepared on an enormous scale by extracting the finely divided material with hot water in batteries of diffusors. The dilute solutions (1-5 to 3 Be.) are filtered and then concentrated in a triple-effect vacuum evaporator (see p. 461) to the density 25-30 Be. For some years, however, certain extracts have been clarified or partially decolorised with alkali sulphite, bisulphite, or hydrosulphite (patented by Lepetit, Dollfus and Gansser, 1896) before concentration. The bisulphite renders the extracts much more soluble, as it converts part of the tannin substances into soluble sulphonic compounds, while in the resinous extract of quebracho it also causes decomposition of a glucoside present, giving the product the property of imparting a yellow colour to skins with an aniline mordant. Decoloration is, however, due more especially to the hydrosulphite either added directly (Lepetit's patent) or produced by reduction of the bisulphite added to the extract (1) by zinc or aluminium dust (Eng. Pat. 11,502 of 1902) ; (2) by treating the crude extract with aluminium sulphate and sodium bisulphate and then heating under pressure at 120-130 (U.S. Pat. 740,283) ; (3) by treating the extract with a mixture of formaldehyde-bisulphite and forinaldehyde-sulphoxylate (Fr. Pat. 362,780) ; or (4) according to the recent patent of L. Dufour (Genoa), by reducing the sulphite with thiosulphate, and then with formaldehyde. Use has also been made of the waste sulphite liquors from the manufacture of cellulose 586 ORGANIC CHEMISTRY (Ger. Pat. 132,224 and 152,236 ; U.S. Pat. 909,343, January 1909), of aluminium amalgam (Ger. Pat. 220,021), and of chromous salts (chloride, sulphate, acetate, &c.) An interesting method of clarifying quebracho extract and rendering it soluble even in the cold is that of A. Redlich, L. Pollak, and C. Jurenka (Ger. Pat. 212,876 of 1908) : The paste deposited from the crude, cooled extract is shaken for six to seven hours with 1 part per thousand of soda at 50 to 100, 50 litres of the red solution thus obtained being mixed with 1000 litres of the crude extract previously decanted and the whole left to stand. A flocculent deposit is thus obtained and a pale solution of pure extract which is decanted off and can be concentrated ; the flocculent precipitate can be dissolved again in dilute soda and used to clarify further quantities of crude extract. Any excess of red, alkaline solution may be employed for clarifying extracts of sumac, &c. The price of tanning extracts is roughly proportional to their content of tannin or tannin substances, 1 which may vary from 20 per cent, to 50 per cent., but for a given content of tannin, extracts rich in red or orange colouring-matters have the greater value ; these matters are estimated in special colorimeters or in the spectroscope. A. Gansser has recently (1909) suggested the replacement of the direct test on hide by one on strips of animalised cotton (the latter being immersed in a bath of gelatine and then in one of formaldehyde) ; the resultant colour on the textile is similar to that obtained on hides. In 1905 Germany imported 58,000 quintals of sumac (40,550), 139,054 quintals of quebracho extract (257,250), and 126,315 tons of quebracho wood (600,000), 145,000 quintals of quebracho extract (254,800) being exported. The chestnut extract produced in Corsica amounted to 22,032 tons in 1906, to 18,275 tons in 1907 (the diminution being due to strikes), and to 25,000 tons in 1909. The United States consumed about 70,000 tons of solid quebracho extract in 1908. For the manufacture of tannin extracts (e.g. from chestnut wood) to pay, at least 300 quintals of wood must be treated per day ; the plant costs over 8000. TANNING OF HIDES. The hides of oxen, horses, sheep, &c., even when freed from hair and flesh (i.e. in the form of corium), do not keep and readily putrefy during drying or in presence of moisture. When dressed (this was carried out as early as 2000 B.C.), and, more especially, when tanned, the hides are more tenacious and resistant, do not putrefy, and do not gelatinise with boiling water, since the fibres on which the tanning material is fixed (to the extent of 30 per cent, or even more) do not agglutinate during drying, and hence remain fibrous and do not become compact and horny. The corium or derma, i.e. the fibrous substance of the skin, is converted by tanning into leather. 2 Rational ANALYSIS OF TANNING MATERIALS. A solution is prepared containing not more than 0-6 to 0-8 grra. of dry residue per litre ; for this purpose 9 to 10 grins, of solid extract or 15 to 20 grms. of liquid extract are dissolved in a litre of tepid water. Of the various analytical methods, the least inexact is that of Procter, which was accepted by the International Congress of Leather-Trades Chemists at Turin, 1904. The amount of total soluble substances is determined, the difference between this and the non-tannins (not fixed by powdered hide) giving the tannins. The total soluble substances are determined by evaporating 100 c.c. of the clear, filtered solution to dryness, and drying the residue at 100 to 105 until of constant weight. Non-tannins. Powdered hide of the best quality is employed. With this is filled a glass bell or funnel (Fig. 410), holding about 30 c.c. and 3-5 cm. high and 3 cm. wide; the funnel is fitted with a rubber stopper, through which passes a capillary glass tube (2 mm. diameter) bent in the form of a syphon. The short limb of the tube pene- trates 1 cm. below the stopper, and its end is surrounded with cotton- or glass-wool to retain the hide. The funnel holds about 7 grms. (not less than 5) of slightly compressed powdered hide, and the mouth of the funnel is closed by well-washed muslin tied tightly on. The funnel is arranged almost on the bottom of a 200 c.c. beaker, con- taining a little of the filtered tannin solution, and is left for an hour so that the hide powder may become moistened uniformly. The beaker is then filled with the tannin solution and suction applied to the long limb of the syphon (about 20 cm. longer than the short limb) so that about 90 to 100 c.c. flows out in 1-5 to 2 hours. The first 30 c.c. or so of the filtrate is discarded until, indeed, a small portion fails to give a turbidity with the liquid obtained by treating 2 grms. of the hide powder with 60 c.c. of distilled water and filtering. Of the clear liquid free from tannin substances, 50 c.c. are evaporated in a platinum dish and the residue dried at 100 to 105, until of constant weight. This weight is multiplied by 2 and subtracted from the total soluble substances (see above). " THEORY OF TANNING. In the first half of last century, Davy, Seguin, Dumas, and Berzelius regarded the absorption of tannin by hides as a chemical reaction. In 1858 Knapp defined leather as an animal skin the fibres of which do not adhere during drying owing to the pores separating the fibres being filled with the tannin ; tanning would hence be a simple physical phenomenon. Similar views were expressed by Reiner (1872), Heinzerling (1882), Schroder and Passler (1892). Th. Koruer (1898-1903) also regarded it as a physical process, since neither the tanning material nor the fibres constituting the hides are clcctrolytically dissociated, and therefore cannot combine to form a kind of salt. Herzog Adler, and Wislicenus (1904) also supported the physical theory. FIG. 410. THEORY O'F TANNING 587 tanning was introduced only when the anatomical structure of the skin became exactly known and the effects of tanning materials on the different parts of the hide were studied. Various methods of tanning are in use : (a) Mineral Tanning or tawing, by means of alum and sodium chloride ; (6) Oil Tanning or chamoising, with fatty materials ; (c) Ordinary Tanning with tannin substances ; (d) Chrome Tanning, using chromium salts (tanning with formaldehyde, proposed by Trillat and Payne ; with quinone by Meunier and Seyewetz ; with naphthols by Weinschenck ; with rare earths by Garelli ; with fatty acids by Knapp, or with the corresponding ammonium soaps by Garelli and Corridi, 1909.) The preparation of the skins for tanning (swelling, unhairing, &c.) is carried out as described below under ordinary tanning. (1) Mineral Tanning or tawing is frequently used for light lamb, sheep, and goat skins, which, after unhairing (see later), are passed into the limes and are then, just as in ordinary tanning, swelled in an acid bath, which also removes all the lime. They are then placed in the tanning vat containing alum or sodium chloride solution, without impregnating them with fatty substances. For twenty hides, about 1500 grm. of alum and 500 grm. of sodium chloride are dissolved in 50 litres of tepid water. The hides are well saturated with this bath and are heaped up still wet for two or three days, after which they are pressed, washed, and allowed to dry in the air. The finishing of the tanned hides is carried out as described later. As it has been established that the hide is capable of absorbing at its surface like a colloidal solution, Stiassny (1908) holds that tanning consists simply of a physical absorption, since tannin reacts with scarcely any of the known hydrolytic products of hides. Just as colouring-matters are fixed by carbon, silica, and alumina without there being any special groups to effect combination, so also in tanning all the known phenomena support the physical absorption hypothesis. According to Stiassny, every tanning process consists in the absorption of a dissolved colloidal substance by the gel of the hide and in simultaneous secondary transformations (polymerisations, oxidations, &c.), to which the absorbed matter is subjected by the catalytic action of the hide, and which render the absorbed tannin insoluble and the process irreversible. This is more a physico-chemical than a physical theory. Konnstcin (Vienna) also regards the phenomena as a physical one, owing to the absence of stoicheiomctric relations. On the other hand, Miintz (1870) and Schreiner (1890) hold that tanning must be due to a chemical phenomenon, since the same hide always absorbs the same maximum amount of a given tanning material, but Schroder and Piisslcr advance the objection that below the limit of maximum absorption, the quantity fixed varies with the concentration of the bath, there being no stoichciometric relations characteristic of chemical combination. Suida, Gelmo, and Fahriou (1903-1908) revert to the chemical theory, and assert that, as tanning is preceded by treatment with acid or mordant, slight dissociation or hydrolysis may occur (as in the case in the dyeing of wool) Further, hide powder fixes substantive dyes better than wool itself, and that the combination docs not exhibit stoicheiometric proportions is explained by the fact that the hide consists of compact fibres and not of separate molecules as in solution, so that the tanning liquor penetrates only slowly into the interior of the mass, and is gradually impoverished and exhausted. Fahrion (1908-1910) points out that in tanning with formaldehyde there can be no question of colloidal material (as with tannin), and with regard to the elimination of alum or tannin from leather by the mere action of water, this is due to pseudo-tanning, i.e. to the formation of labile, readily hydrolysable compounds, the tannin of which becomes distributed between the hide and the water. With reference to the non-stoicheiometric relations, he observes that the fixation of more tannin from concentrated than from dilute solutions is in accord with the law of mass action for reversible chemical reactions. According to lleidcnhain, Zacharias, and Fahrion (1908), both the dyeing and the tanning process occur in two phases, the absorption and penetration of the tanning substance and the subsequent chemical combination of this substance with the hide. Garelli (1907-1910), from the results of his tanning experiments with rare earths (ceria, thoria, zirconia), supports this theory, and holds that all substances which in aqueous solution can undergo hydrolysis forming basic hydroxides or salts (like chromium, iron, and aluminium salts) arc capable of tanning hidi's (i.e. the hide hydrolyses and decomposes the salts, which thus deposit hydrates or basic salts on the fibres of the corium or derma, the fibres and the salts combining to form leather). Thus, Garelli effected tanning with the rare earths, i.e. with compounds of the trivalent (cerium, lanthanum, and didymium) or tetravalent elements (cerium, thorium, and zirconium ; Zacharias had used stannic salts in 1903), and the tanning, as when alum is used, is facilitated by sodium chloride (this was not used with eerie salts, which would generate chlorine). The most effective tannings are those in which an oxidation plays a part (the metals pass from the higher to the lower valency) and those with alum, which cannot give salts of lower valency but are not very stable, and do not resist even the prolonged action of cold water (pseudo-tanning). Chromium salts are reduced to oxides by the skin and fixed, while oils and fats must be oxidised (to hydroxy -acids), as otherwise the tanning is not complete. It. Lepetit (Ann. d. Soc. chim. di Milano, 1907, p. 83) asserts that in the tanning of sole and upper leather it is not sufficient to effect separation and stabilisation of the fibres, but that it is necessary to produce swelling and tilling of the interstices between the fibres with phlobaphenes. These arc colloidal substances dissolved or suspended in the tannin extracts and consisting partly of internal anhydrides of soluble tannins (see p. 582) and partly of condensation products of formaldehyde with polyphenols and phenolcarboxylic acids derived- from the tanning vegetable organisms. Indeed, according to Nierenstein, the products of the reaction between formalin and poly phenols exhibit tanning properties, and at the present time glove leather is successfully tanned by formaldehyde (Trillat and Payne). Also Weinschenck (1907-1908) stated that a- and /3-naphthols in presence of formaldehyde are able to tan hides, but this is denied by Stiassny and llicevuto (1908). In tanning with quinone derivatives (suggested by Meunier and Seyewetz) leather is formed, owing to the hydroquinone derived from the quinone reacting with the amino-groups of the proteins. With formaldehyde, there is probably production, by aldol condensation, of complex colloidal polymerides of formaldehyde (especially in presence of alkali carbonate), these reacting with aminic complexes in the same way as formaldehyde and the aldols react with aniline (see p. 558). Thuau (1909) found that if the hides are previously treated with formaldehyde .subsequent chrome tanning is hastened. 588 ORGANIC CHEMISTRY Mineral tanning is usually a rapid process, and the alum combines with the corium and preserves it, but the leather is not so lasting as that prepared with tannin and can still be gelatinised by prolonged boiling with water. Chromium salts (the alum and chloride) are often used nowadays in place of alum and sodium chloride. (2) Chrome Tanning has assumed considerable importance of recent years (since 1895), as it is rapid and furnishes boot leather highly resistant to wet ; it is often used also for girths, &c. (see later, Rapid Tanning). This method of tanning can be carried out in two baths or in a single bath. In the first case, the prepared hides are steeped in a cold bath containing 40 grms. of dichromate and 40 grms. of hydrochloric acid for every four litres of water and for every kilo of hide. Only one-half of the reagents are added to the bath at first, the hides being stirred with a reel for 2 to 3 hours, after which the remainder of the reagents is added "and the stirring continued for 1 to 2 hours. The hides are then removed from the bath, allowed to drain, stretched, and placed in the reducing bath containing 105 grms. of sodium thiosulphate per 4 litres of tepid water and per kilo of hide. Here they are stirred for half an hour, after which 50 grms. of hydrochloric acid are gradually added (in the course of 1 to 2 hours) per kilo of hide ; the hides are finally rinsed with water and the tanning is at an end. When, as is now more commonly the case, a single bath is used, this contains a solution of either basic chromium oxychloride, Cr 2 (OH) 5 Cl, and common salt, or chromo-base, which is a basic sulphate prepared by the firm of Lepetit, Dollfuss, and Gansser ; the procedure is as in the preceding case. The use of chromium lactate has been recommended, since lactic acid reduces chromium salts, even in the cold. (3) Oil Tanning or chamoising. This is used to obtain very soft leather for gloves, clothing, &c. Deer, stag, lamb, kid skins, &c., are smeared or rubbed with various fats (fish oil, wool fat, paraffin, egg-yolk, alum, carbolic acid, sodium chloride, &c.), the absorp- tion of which is effected by repeated working of the skins, followed by drying in tepid chambers ; the skins are thus rendered impermeable, while they can be washed many times without losing their tanning. The superficial fat is finally removed by washing in soda solution, the emulsion thus formed, known as degras (see p. 389) being used for currying ordinary hides. Heavier hides (cow, horse, ox, buffalo) intended for saddlery are subjected to mineral tanning (without being treated with lime) and afterwards to a kind of oil tanning which imparts to the leather considerable resistance to tension. (4) Ordinary Tanning (vegetable tanning). The fresh hides as they come from the slaughterers are termed green hides, and in this condition an ox-hide will weigh from 30 to 40 kilos, its weight being reduced to one-half by tanning. Many hides are imported from South America in the dried and salted or smoked state. Ox-hides give the heaviest leather for boot-soles, while for lighter soles cow-hide is used ; the uppers are made preferably from calf-skin. Saddles are made from horse-hide, pig-skin, and seal-skin, while sheep-skin is used for bookbinding leather and goat-skin for morocco leather. Deer-skin, goat-skin, &c., are tanned with oil to obtain chamois or buff leather (see above). The hides are first softened by soaking for 2 days or longer (according as they are green or dry) in water, which removes blood and other adherent impurities. They are then placed on a " beam " (Fig. 411) and scraped on the flesh side with a curved knife (Fig. 412), which is drawn across them horizontally. They are then soaked for 24 hours, scraped again, washed in water for a few hours, thrown on the beam and allowed to drain. This operation is hastened if the softened hides are subjected to fulling in a revolving vessel (Fig. 413) or in a vat containing cold water in which they are worked with wooden mallets. In order to remove the hair fixed in the epidermis (not in the corium), the epidermis must be attacked and almost destroyed, this being effected in various ways (by putrefaction, lime, or sulphides). Putrefaction ("sweating") is carried out by salting the flesh side of the hides or sprinkling them with crude acetic acid, bending the hides in two longi- tudinally with the hair outside and stacking them in tanks or in a warm chamber (30 to 50) ; fermentation soon sets in, accompanied by heating and evolution of ammonia, the hides being then unhaired on the beam with a suitable knife. In order to avoid the possibility of excessive heating, the hides are sometimes placed in cement troughs fitted with perforated, wooden, false bottoms, water being sprayed on to the hides at the top, so that the temperature is kept down to 10 to 12 ; after 8 to 12 days the hides can be TANNING PRACTICE 589 readily unhaired. The more delicate skins of small animals are treated with sulphides, being smeared with rusma, which consists of a mixture of 1 part of arsenic sulphide (orpiment) with 2 to 3 parts of slaked lime ; calcium hydrosulphide is also used and gives better results. In recent years, sodium sulphide has also been used for heavy hides, unhair- ing being easily carried out by scraping the hides (after washing) with a knife against the set of the hair, the operation being facilitated, if necessary, by sprinkling a little sand or ashes on the hide ; the hair serves for the manufacture of felt, but that treated with sulphide is converted into fertiliser. When unhaired, the hides are well washed in water and beaten on a large beam with the hair side uppermost ; if necessary, the removal of the flesh is then completed by means of a knife, the useful part of the hide, i.e. the corium, then remaining. The hides have by this time lost about 12 per cent, in weight, and those which have been limed are next kept for two or three days in several successive infusions of barley flour or bran (" bran drench ") in active acid fermentation ; to these are added sulphurous or sulphuric acid, lactic acid (or better, according to Boekringer, Ger. Pat. 234,584 of 1909, a solution of lactic anhydride in ammonium lactate), or acetic acid, the calcium soaps on the hides being thus decomposed ; the acids separate at the surface and the soluble calcium salts FIG. 411. FIG. 412. FIG. 413. are eliminated by washing (at one time, mixtures of dog and bird dung with water were used, the action of these being due to enzymes and amine hydrochlorides). After a few days the hides swell up to double their original size and become yellowish and transparent. Excessive swelling is prevented by the addition of a little tanning material to the infusion. All these preparatory operations are required to make the material to be tanned more permeable and more uniform in its behaviour towards the tanning agents, which are fixed to the extent of about 30 per cent, (calculated on the dry corium). 1 The tanning can now be carried out by the following methods : (a) Infusion tanning. This process, which is used for lighter hides, consists in passing the hides into tanning baths of gradually increasing strength, so that the tanning may be gradual and penetrative. The total time required is 6 to 9 weeks, and between each bath and the succeeding one the hides are drained, pressed, and fulled in order to facilitate the absorption of the tannin. (b) Tanning in layers was once largely used but is now employed more particularly for sole leather. Fifty or sixty hides are placed, alternately with layers of powdered or crushed tanning material (bark, wood, &c.), in a cement or wooden vessel, the empty spaces being then filled with the tanning material and the whole covered with water. The vessel is then closed with an air-tight cover and left for about 2 months, the hides being then transferred to a second similar vessel containing rather less tanning material, where they are left for 3 to .4 months, and finally to a third vessel containing still less tanning material (4 to 5 months). 1 F. Carini (Ann. d. Soc. chim. di Milano, 1903, p. 23, and 1904, p. 144) proposes to use the hydrostatic balance in order to obtain the weight of the dry hide from that of the wet hide, without drying. The hides can thus be followed through all the operations, from their entry in a more or less moist state. The quantity of tanning material fixed can also be determined at any moment in this way. 590 ORGANIC CHEMISTRY FIG. 414. If the hides are very heavy and resistant, they are passed to a fourth and sometimes to a fifth bath or pit, the whole operation then occupying about two years and the consump- tion of bark being about five times the weight of the dry hides. The completion of the tanning is ascertained by cutting the hide and observing that the section is uniform and without horny or fleshy layers, and that the grain does not crack when the hide is carefully bent. (c) Rapid tanning, which gives a greater output of leather, has been attempted in many different ways : By immersing and com- pressing the hides in relatively concentrated tanning baths pre- pared from active, modern ex- tracts, and containing a certain amount of acid to prevent wrinkling of the hides, the tan- ning liquor being circulated by means of pumps without moving the hides ; or the skins are placed in revolving barrels or drums, the lower half dipping into tanning liquor so that the hides are pressed at intervals. The diffusion process is also applied by placing the tanning bath in bags composed of various hides sewn together. Tanning in a vacuum has likewise been used in order to effect better penetration of the tanning material, considerable pressure being exerted automatically on the hides at regular intervals, and the operation being facilitated by gentle heat, &c. By these rapid processes (see also Use of Quinone, Ger. Pat. 206957, 1907) tanning can be completed in 6 to 8 weeks, this including the preliminary preparation of the hides. The actual tanning may, indeed, be limited to 30 hours if revolving barrels are used with hot, highly concentrated tanning baths (8 to 10 Be). When such a rapid process is used it is, however, indispensable to eliminate all traces of lime beforehand by immersion in formic acid solution. Other very rapid methods which are largely used are chrome tanning (see above) and formaldehyde tanning as proposed by Payne. The tanned hides are then subjected to finishing, which varies considerably with the nature of the hide and the kind of leather required. For sole leather, the hides from the layers are first dried in the shade and are then beaten or hammered by means of a suitable machine (Fig. 414) ; in this way the leather is rendered more compact, so that the wear of the sole is diminished. To make the leather of uniform thickness and to eliminate lumps, scurf, wrinkles, &c., the hide is scraped or shaved on the under side (i.e. not the hair side) with a sharp curved knife or, better, with W IG 415 a machine carrying steel blades on a moving band, which can be brought more or less near to the skin, the latter being stretched on a movable trolley (Fig. 415) so as to facilitate this tedious and troublesome operation. Regular graining is attained by stretching the hide on a bench, one side being fixed to the edge of the bench and the hide then bent over on itself, while a block of wood having a concave base with pointed grooves (Fig. 416) is moved backwards and forwards over the fold, which is gradually displaced until the whole surface of the hide is covered. Artificial grain is nowadays imparted by pressing the hides between special fluted cylinders. Those hides which are required to show, not graining, but a smooth surface are first rendered perfectly uniform at the surface by rubbing both sides with pumice by hand or HYDROGENATED BENZENE COMPOUNDS 591 Fio. 416. more conveniently by a kind of spindle-shaped grindstone covered with emery (Fig. 417), against which the surface of the hide is gently pressed. In some cases this operation is completed by polishing the bloom side with a concave piece of wood, similar to that of Fig. 41 6, but with a smooth surface lined with cork. The polishing is finished on the bloom side with a heavy, very smooth roller, moved horizontally by a rod connected with an eccentric. All these finishing operations are carried out mechanically by machines which are continually being improved and which cannot be described here. Finally, many leathers with which a certain degree of softness is required, are greased with fish oil or a mixture of this with tallow or other fats (wool fat) or degras (see p. 389). In this operation, which amounts to a second tanning (chamoising, see above), the tanned and still moist hides are well smeared with the fat and exposed to the air until the whole of it is absorbed. Leather for boot uppers is coloured black on the flesh side by rubbing with concentrated solutions of iron acetate and sulphate, treating with oil, wax, soap, lampblack, &c., and then polishing with smooth wood until a shining surface is obtained. For special purposes hides and leathers are coloured with basic or mordant aniline colours, the hides being first prepared by immersion for 12 hours in cold water in which is dissolved the white of an egg for each hide. The dyeing is carried out at a temperature of 30. Certain leathers are varnished with ordinary resin var- nishes. In order to supply the great demand for large hides for the hoods, &c., of carriages, ox-hides and cow- hides are nowadays divided, the more resistant part being kept for the hoods, and the flesh side for the seats, &c. The use of pure water is indispensable in all tanning operations, since water which is too hard and rich in lime readily produces white efflorescence on the hides. The presence of iron in the water results in the formation of dark patches, while suspended organic matter is always harmful ; waters containing these substances must hence be thoroughly purified before use (see vol. i, pp. 218 and 665). In order to avoid the formation of the white efflorescence due to the combination of lime with the fatty matters of the tanning materials it has been proposed to replace the fats by mineral oils, which do not give calcium salts, or to wash the hides well with dilute lactic or formic acid which form soluble calcium salts. The suggestion has also been made that the hides be dressed, not with fats, but with the anhydrides or lactones of fatty acids, as these form calcium salts more slowly (the purgatol recently placed on the market consists mainly of anhydrides or lactones). England's exports and imports of hides are as follow : FIG. 417. Raw hides Tanned hides, leather 1910 1911 1910 1911 Imports 12,882,326 11,104,326 11,824,741 12,227,606 Exports 1,757,762 1,685,583 4,686,485 4,880,932 O. HYDROGENATED BENZENE COMPOUNDS Considerable interest attaches to the numerous hydrophthalic acids studied by Baeyer in their various constitutional and stereo -isomerides (cis- and trans- isomerides ; see p. 21). They behave largely like unsaturated aliphatic compounds (see p. 520), as they no longer possess the stability of the true benzene nucleus. The position of the true double linkings in these compounds is determined by the addition of bromine or by subsequent elimination of the latter by reduction, with or without substitution of hydrogen, according as the two bromine atoms are in para- or ortho-positions. Simple boiling with alkali often effects displacement 592 ORGANIC CHEMISTRY of a double bond (as with oleic acid ; seep. 293), so that it is possible to pass from one isomeride to another. The di-, tetra-, and hexa-hydrophthalic and terephthalic acids can be dehydrogenated in stages by heating with bromine at 200 ; many of them form anhydrides. From the results of his investigations on the hydrophthalic acids Baeyer drew important conclusions concerning the constitution of the benzene nucleus. Many important hydrogenated benzene derivatives occur naturally, among them the naphthenes, found in abundance in Russian petroleum (see p. 63), which contain hexamethylene groupings (see Polymethylenes, p. 520). Syn- thetically they may be obtained, for example, from calcium pimelate : 222 2 >CO (ketohexamethylene) . Also, by condensing 2 mols. of ethyl succinate with sodium and then hydrolysing the product and heating at 200, p-diketohexamethylene is obtained. Hydrogenation of benzene and its homologues, by passing their vapours, mixed with hydrogen, over heated finely divided nickel, yields hexamethylene l and its homologues, hexahydrophenol (b.pt. 160-5), and p-diketohexamethylene (m.pt. 78). The latter gives the corresponding alcohol, quinitol (p-dihydroxy- hexamethylene), which forms various cis- and trans -isomerides. Inositol, C 6 H 12 6 , the hexahydric alcohol derived from hexamethylene, is isomeric with the hexoses, but with HI or PC1 5 yields true benzene derivatives. Various naphthenic acids are obtained by oxidation of the naphthenes of petroleum (see p. 63), and are distinguished from open-chain acids by forming soluble magnesium and calcium salts ; by this means they can be detected when used in the manufacture of soaps. Still more interesting are the terpenes and the camphors, which are found in various plants and form the principal constituents of many ethereal oils and essences and of many resins^ QUINIC ACID (TetrahydroxyhexahydrobenzoicAcid), CO 2 H-C 6 H 7 (OH) 4 , is optically active, but only an inactive modification is known. It is obtained from the roots of coffee, cinchona, &c., and forms white crystals. TERPENES These are regarded chemically as hydrogenated derivatives of cymene (dihydrocymene) and its homologues, and have the generic formula Ci H 16 . They are not soluble in water, but can be readily isolated from the natural products owing to their volatility in steam. The chemical constitutions of the principal terpenes have been established mainly by O. Wallach's investigations over a period of more than twenty years. By their syntheses, their halogenated additive compounds, their behaviour towards oxidising agents and their molecular refraction (see p. 26), it has been shown that they contain two double linkings and a closed ring of six carbon atoms. There is, however, a group of more complex terpenes (pinene, camphene, fenchene, &c.) which have only one double bond. In order to define the position of the double linkages (A), Baeyer numbered the fundamental carbon atoms of the cymene as in the first figure of the following scheme, which shows the 1 HEXAMETHYLENE (hexahydrobenzene, cydohexane, or naphthene) is found in Caucasian petroleums and is obtained synthetically from iodohexamethylene or 1 : 3-dibromopropane. It is a colourless liquid smelling like petroleum, and it boils at 80 and resists the action of permanganate. By hydriodie acid at high temperatures it is converted into methylpentamethylene. constitution of five terpadienes out of fourteen possible theoretically without counting enantiomorphs. 5 8| v/ c_ c c 10 8 9 H 2 H CH, H C 3 H 7 I H H x/ t A H C 3 H 7 II HCH 3 V H Aa H C 3 H 7 III FIG. 418. H 2 Ho CH, H/\H H. , H 2 f~ H CH 3 -C-CH 3 CH 3 -C=CH 2 IV V 10 H 16 (N0 2 )(NO), or Nitrosochlorides, C 10 H 16 (NO)C1, which are To indicate the position of the double linking in the side-chain, instead of giving only the lower number of the two carbon atoms united to the double linking, as in the case of the nucleus (e.g. Ill = A 3 > 5 -terpadiene or limonene ; I = A 1 > 4 -terpadiene), the numbers of both the carbon atoms united to the double linking are given, the higher number being bracketed (e.g. IV = AWUerpadiene ; V = A^O-terpadiene). In the official nomenclature the name terpane is given to Hexahydrocymene, C 10 H 20 , Tetrahydrocymene, C 10 H 18 , being called terpene and the Dihydrocymenes, G 10 H 16 , terpadienes. As separated from plants or fruits, the terpenes are generally mixtures, and when obtained from conifers are termed oil of turpentine. Essence of lemon gives citrene ; thyme, thymene ; cumin, carvene ; orange, hesperidine, &c. Although their boiling-points differ little (160 to 180), they form tetra- bromo-derivatives and dihydrochlorides with widely different melting-points, these compounds hence serving for their separation. Properties. Owing to the presence of double linkings, which act as in aliphatic compounds, the terpenes can combine with four bromine atoms or two mols. of HC1 (the halogen being readily replaced by hydroxyl, with formation of camphor) and also react with nitrous acid or nitrosyl chloride, forming solid Nitrosites, C also solid and sometimes blue. They oxidise easily and with mild oxidising agents give benzene derivatives, whilst on energetic oxidation they resinify ; they polymerise readily, and by acids, for instance, are converted into more stable isomerides. In alcoholic solution they give characteristic colorations with concentrated sulphuric acid. They are usually optically active. They often accompany the natural perfumes of fruits and flowers, which, now that they have been subjected to thorough chemical study, can be obtained purer and of increased value. 1 1 PERFUME INDUSTRY. A considerable number of the natural perfumes have been prepared from the very earliest times, but with the perfected methods of extraction now available they are obtained in higher yields and in a more highly refined condition. The most abundant supplies of raw material have always been, and are still, obtained from eastern countries, where whole provinces are often devoted to the cultivation of flowers. The most delicate perfumes are those obtained from flowers which contain, along with the odorous principle, other substances which refine the aroma and render it softer. The name artificial perfumes was at one time given to mixtures, in proportions carefully chosen, of the fundamental natural essences, a great variety of perfumes being thus obtained ; this, however, required a very highly developed sense of smell in the operator. The discovery of artificial perfumes did not diminisli the consumption of the natural products since these became cheaper and thus appealed to a large public. The consumption of perfumes fluctuates with the fortunes of a nation. The early Eastern races and then the ancient Egyptians introduced perfumes into religious ceremonies, their secular use being often forbidden. Gradually, however, they became used for domestic purposes, together with many diifercnt pomades and, in some cases, dyes. Egyptian pomades were held in high esteem by Cleopatra. With the ancient Greeks, the use of perfumes and cosmetics assumed considerable importance and often degenerated into abuse, and Socrates states that if even a slave is anointed with a good perfume he will exhale the same odour as his master. Perfumery flourished under the Romans and declined with the Empire, being re-established in Italy only a II 38 594 ORGANIC CHEMISTRY CINENE (A J 8 ( 9 )-Terpadiene or Dipentene ; Inactive Limonene), C 10 H 16 , is found together with cineol in oleum cince and also in Laurus camphora and in Russian and Swedish turpentine oils. It is formed by isomeric change when camphene, active limonene, pinene, &c., are subjected to protracted heating at 260 to 270, and is obtained, together with isoprene, when rubber is distilled, 2 mols of the isoprene, CH 2 : CH-C(CH 3 ) : CH 2 , under- going condensation. It has a pleasant odour of lemons, and boils at 176. Nitrosodipentene (inactive the time of the Renaissance. It then passed into France, where it became a true national industry, culminating at the time of the perfumed Court of Louis XV. Until about the middle of last century, France enjoyed almost a monopoly in this industry, but when science pervaded this branch of human activity, the clever French rule-of-thumb manufacturers did not grasp quickly enough the benefit to be derived from a rational development of their industry, of which England and Russia, and more especially, during the past quarter of a century, Germany have taken advantage. At Grasse and Cannes, in the south of France, however, the natural perfume industry is still of importance, certain factories dealing with as much as 3000 kilos of violets (40 to 50 millions of flowers) at a time. As has been already mentioned, the prime materials come mainly from Eastern Europe, and at the present time also from the Far East. But the cultivation of plants for perfumes is still largely carried on in the South of France and in Sicily. In annuals the essential oil is formed in the green organs, and the majority of it is found ill the flowers before ertilisation. The extraction of perfumes from Dowers and leaves is carried out in various ways : (1) By distillation with direct or indirect steam or in vacua, the distillates of different densities being separated ; this method is used for lavender, rosemary, thyme, orange blossom, and roses, which are unaltered at steam heat. (2) By infusion for 12 to 48 hours at 60 to 65 with pure fats (olive oil, &c.), the flowers being renewed four to six times until the fat is highly perfumed ; the extracted flowers are pressed to free them from fat, and the perfumed fat run into enamelled iron vessels as a concentrated pomade ; in this way are treated cassia, violets, jonquils, and sometimes orange blossom and roses, when mixed with other flowers. (3) By absorption in the cold of the more delicate perfumes of jessamine, heliotrope, and tuberoses ; in vessels with glass walls smeared with fat or covered with cloth soaked in oil, the petals are pressed and rubbed, being renewed every day ; after some days or at the end of the season the perfumed fats are shaken for a long time with alcohol, which extracts all the perfume. To obtain colourless products, Piver passes a slow current of air through the flowers and then on to the fatty surface. (4) By dissolution. The use of this method is spreading, as it gives highly concentrated, very delicate perfumes. The flowers are immersed in petroleum ether, carbon disulphide, &c., the perfume being extracted by a current of steam from the solvent, which is afterwards recovered. (5) By pressure with hand or hydraulic presses, this method being employed with orange-peel, bergamot, iris rhizomes, &c. The yields obtained per 1000 kilos of leaves or flowers are about as follow : 1 kilo of orange oil or neroli, from the flowers (value 24 to 28), or 3 kilos of petit grain (from the leaves) ; 1 kilo of essence of basil (6 to 8 per kilo) ; 1200 grms. of essence of citronella (88s. per kilo) ; 9 to 15 kilos of eucalyptus oil (from the leaves) ; 120 grms. of essence of jessamine (from fresh flowers) ; 1 kilo of geranium oil (from flowers and leaves) ; 10 kilos of oil of lavender ; 6 kilos of marjoram oil ; 2 kilos of mint oil; 3 kilos of myrtle oil ; 2 to 10 kilos of rosemary oil ; and 200 to 500 grms. of rose oil. The exports from Sicily and Calabria and the imports to Italy of essences were as follow : Exports from Sicily and Calabria 1906 1907 1908 1909 1910 Orange oil . . kilos 136,739 162,274 173,265 242,762 143,825 (97,800) Bergamot oil . . ,, 63,510 87,538 74,842 73,803 64,788 (82,930) Lemon oil . . ,, 440,500 - 469,385 476,842 364,647 425,076 (154,025) Various (mint, mandarin, &c.) oils . . . kilos 28,500 31,800 13,500 (14,800) Imports into Italy Clovu oil . . . kilos 2,446 2,165 1,538 1,969 1,878 (1,502) Mint oil . . . 6,628 6,484 4,391 5,334 8,931 (14,290) Rose oil ... ,, 493 341 101 158 109 (3,920) Various oils . '. ,, 67,797 77,510 81,830 90,661 99,228 (79,380) As much as 100 quintals of flowers for perfumes (at 28s. per quintal) are dispatched per day from San Remo ii the spring and summer. In the neighbourhood of Giasse, Cannes, and Nice the production in 1902 was 2,500,000 kilos of orange blossom, 3,000,000 kilos of rose leaves, 200,000 kilos of jessamine, 150,000 kilos of violets, 150,000 kilos of tuberoses, &c., all these being extracted on the spot. In Germany, although the climate does not seem very favourable, the cultivation of certain flowers for perfumes is largely carried on in some districts. The perfumery factories have hundreds of hectares of land under flowers not only for commercial purposes, but also for analytical and research work. One hectare yields 10,000 to 15,000 kilos of rose leaves. At one time the firm of Schiiumel (Leipzig) treated as much as 600,000 kilos of fresh rose leaves per day, 300 kilos of rose oil being extracted ; this was repeated two or three times in a month (June). A kilo of the oil is sometimes obtained from 2000 kilos of the leaves. Rose cultivation is, however, carried on most extensively in Turkey and Bulgaria, where preference is given to the red rose (Rosa damasccena), which gives on an average 1 kilo of oil per 4000 kilos of leaves, although white roses (Rosa alba), giving 1 kilo of oil per 5000 kilos of fresh petals, are also largely grown. The product from the latter variety is less fine, but it gives an oil crystallising at 18 to 20 and is used to mask oils of lower quality ; the market value of the oil is judged more particularly from the freezing-point, which should be between 17 and 19 for good qualities. Adulteration with alcohol or spermaceti is easily discovered, but it is more difficult to detect additions of geranium oil or palmarosa oil. In 1887 Turkey produced 2400 kilos of pure rose oil (attar of roses), whilst in 1904 and 1906 the output reached 3600 kilos. The annual production varies very considerably, as the plants suffer greatly in dry seasons, especially water IB scarce in the month of May preceding the harvest; in 1907 indeed, the output was only 2000 kilos. PERFUMES 595 carvoxime) melts at 93. With HC1, cinene gives two stereoisorneiic dipentcne dihydro- chlorides (1 : 4-dichloroterpanes), melting at 50 and 25. The tetrabromide melts at 125. CARVENE (rf-Limonene, Hesperidine, Citrene), Ci H 16 , forms the greater part of orange-peel oil and also occurs abundantly in cumin oil, anethum oil, &c. ; lemon oil is a mixture of pinene and limonene. It is a liquid boiling at 175 and is optically active although readily convertible into inactive dipentene. It forms a dextro-rotatory tetrabromide melting at 104. In Bulgaria roses are still more largely grown, and here, too, the production varies widely. The exports of pure oil were as follows : 3190 kilos (71,280) in 1897 ; 3900 kilos in 1902 ; 6200 in 1903 ; 5000 in 1904 ; less than 4500 in 1905. In 1907 the exports were valued at 168,000, aud in 1908 at 184,000 ; in 1909 6053 kilos were exported. At one time two-thirds of the oil went to France, but now only one-third goes to the French factories the rest being sent to England and Germany. In 1910 England imported natural ethereal oils to the value of 320,218, artificial ethereal oils to the value of 34,369, and alcoholic perfumes to the value of 90,176. The United States imported perfumes and other toilet preparations to the value of 303,600 in 1911. The price of attar of roses varies from 32 to 80 per kilo, and was formerly higher than this. In 1904 H. von Soden patented a process for obtaining more refined and delicate perfumes from flowers. He first obtained a petroleum ether extract which was then evaporated and the residue taken up in alcohol, the latter being distilled off and the residue distilled in steam. It must, however, be pointed out that with this process, 1 kilo of the finest rose oil would now cost 1520 and 1 kilo of oil of violets almost 4000. From what has been already stated, it will be recognised that considerable interest attaches to the study of the composition and constitution of these essences and to their artificial production by synthetical methods. In former times, various artificial perfumes have been obtained empirically, as was also the case with the first coal- tar dye, yet it has required systematic chemical investigation to open up new fields in this direction. During the last thirty years, the consumption of perfumes has increased from 480,000 to 2,400,000, owing to the diminished prices of the natural and artificial products. The first artificial perfume was nitrobenzene or artificial myrbane oil, which was discovered by Mitscherlich in 1834, placed on the market by Colles and manufactured on a large scale by nitrating benzene from tar by Mansfield in 1847. In about 1840, Piria oxidised salicin (a glucoside found in willow bark) and thus obtained salicyl- aldehyde, which is the pleasant smelling essence of Spircea ulmaria (meadow-sweet). A few years later in 1844 Cahours succeeded in isolating the active principle of gaultheria or wintergreen oil, consisting of methyl salicylate, which can be obtained synthetically by heating salicylic acid with methyl alcohol (wood spirit) and sulphuric acid. Many of the natural perfumes contain aldehydes, and in 1853 Bertagnini showed how they could be separated pure by first combining them with bisulphite. Benzaldehyde was synthesised by Cahours in 1868, and coumarin, the essence of Asperula odorata, by Perkiu in 1875. In 1876 Haarmaun and Tiemann ascertained the constitution of vanillin, later preparing it from coniferiu or, better still, from eugenol extracted from clove oil. In 1888 Baur prepared artificial musk. In 1893 Tiemann and Kriigcr succeeded in effecting the synthesis of violet oil, previously obtained at enormous expense from the natural flowers and costing more than 600 per kilo. They also separated irone, the odorous principle of iris root, and determined its chemical constitution. Immediately afterwards they prepared synthetically an isomeride of irone, ionone (see later) to which the delicate odour of the violet is due. These investigators heated citral, which occurs in abundance in lemons, with acetone, acetic anhydride, acetic acid, and sodium acetate, obtaining first pseudo-ionone, which has an unpleasant smell, and, when treated with mineral acid, yields ionone. These processes were patented by Tiemann and disposed of by him to the most important perfume manufacturers for 40,000. The study of the chemical constitution of the components of perfumes reveals a certain relation between the aroma and the presence of definite atomic groupings (osmophores) and attempts were made to establish a perfume theory on a similar basis to the colour theory of aniline dyes, the characteristic groups of which are termed chromo- phores. It has not yet been found possible to formulate a theory as rigorous as that for the colouring-matters, and all that has been fixed is that aldehydes, ketones, mixed ethers, &c., often enter into the constitution of perfumes, and that the introduction of certain alcoholic residues into the molecules may intensify or modify the aroma. The action of perfumes on the olfactory nerves is not thoroughly understood, although it is regarded by some as due to vibrations of the ether similar to those by which light and heat are transmitted, these vibrations originating from the oxidation of the substance in the air. This hypothesis seems to be supported by the fact that many odorous substances emit no smell when worked and distilled in an inert gas instead of in air. It is now, however, generally assumed that the smell is propagated by small particles or molecules, which become detached and, in the state of gas, come into contact with and excite the papillae of the nasal mucous membrane. The fact that certain substances have little smell in the pure or concentrated state and acquire their-maximum smell only when con- siderably diluted, is well explained by modern views on solutions, dissociation in dilute solutions giving rise to the corresponding ions, which become detached and excite the olfactory sense. That minimal traces of these substances transmit perfume is shown by the retention of this property by garments which have been washed five or six times see Experiment described in vol. i, p. 3). A series of tests, controlled by the olfactometer, showed that most men - who have by no means a very delicate sense of smell in comparison with other animals perceived the odour of 1 part of prussic acid in 100,000 of water, 7 per cent, of the individuals examined detecting it in a dilution of 1 in 2,000,000. Of the women tested, however, not one was able to detect prussic acid in a dilution as small as 1 in 20,000. These results support the view that male animals are very sensitive to the odour of the females, which serves to excite their sexual passions. Some individuals, termed anosmic, are quite without sense of smell. The following data give an idea of the influence exercised by the artificial products on the prices of perfumes in general : vanillin cost 120 per kilo in 1878, 35 in 1890, and 3 in 1892, while for heliotropine the price was 100 per kilo in 1881, 15 in 1890, and about 30*. in 1902. That the consumption of the natural products has not been diminished but has increased is shown by the importation of vanilla to France, which amounted to 29,000 kilos in the period 1857-1866, and to 137,000 kilos in 1887-1896. For the year 1897 it was calculated that the total imports into and exports from Germany of ethereal oils and perfumes amounted to 920,000. The Italian imports of alcoholic perfumes in 1904 were valued at 18,640 and those of non-alcoholic at 18,560, while the exports were valued at 3440 and 16,000 respectively. For fruit essences see p. 371. 596 ORGANIC CHEMISTRY Z-LIMONENE, C 10 H 16 , the constitution of which is shown on p. 593 (V), can be obtained from d-carvone, and occurs, together with 1-pinene, in pine oil. Its tetrabromide melts at 104. SYLVESTRENE, C 10 H 16 , is possibly derived from m-cymene and forms a dextro- rotatory component of turpentine. It boils at 176 and gives an intense blue coloration with concentrated sulphuric acid and acetic anhydride. TERPINOLENE (A 1>4(8) -Terpadiene), C 10 H 16 , has the constitution shown at IV on p. 593. It is obtained by the elimination of water from terpineol and melts at 185. TERPINENE, Ci H 16 , boiling at 179 to 180, is obtained in the transformation of various terpenes. Its nitrosite forms monoclinic crystals melting at 155. DIHYDROCYMENE, C 10 H 16 , obtained synthetically from ethyl succinylsuccinate, boils at 174. PHELLANDRENE, Ci H 16 , is known in both the laevo- and dextro-rotatory forms, these having the same chemical and physical properties (excepting the optical rotation) and boiling at 172. The former (1-) is found in Australian eucalyptus oil and the latter in Anethum foeniculum and in water-fennel oil (Phellandrium aquaticum). MENTHENE, C 10 H 18 , boils at 167. MENTHANE (Hexahydrocymene), C 10 H 20 , boiling at 170, does not occur naturally, but is obtained by hydrogenating cymene in presence of nickel. COMPLEX TERPENES Like the preceding, these are composed of a monocyclic system, but with two rings ; they have only one double linking, and hence combine with two atoms of hydrogen or halogen. They can be converted readily into cymene and its derivatives. The following four diagrams show how a trimethylene ring or bridge is formed in Carane (not known in the free state, 'although the corresponding saturated, synthetic ketone, Carone, is known), a tetramethylene ring in pinane and pinene, and a pentamethylene ring in camphane : CH, CH, CH 2 CH, C CH S CH 2 CH, CH Camphane PINENE (Terebenthene, Laurene, Menthene, &c.), QioHie (constitution, see above), forms one of the principal components of oil of turpentine, occurs also in sage and juniper oil, and, mixed with sylvestrene and dipentene, forms Russian and Swedish turpentine oil. When incisions are made at suitable seasons in certain varieties of pine, fir, and larch, a kind of balsam is exuded in the form of a juice which gradually changes to a soft resin, more or less clear according to the quality. This is known as ordinary turpentine or American, French, Venetian, according to the particular tree and to the locality of origin. When turpentine is distilled with steam, the liquid essence or oil of turpentine (turps) is collected separately, the residue, which is solid in the cold, being Colophony. 1 The direct extraction 1 COLOPHONY (rosin) is hard and brittle, its sp. gr. being 1-050 to 1-085 at 15 and its fracture shining and conchoidal. According to the quality, its colour varies from yellow to Drown, but it gives a whitish powder. At 70 it becomes soft and it forms a kind of emulsion with hot water. It always melts below 135 and it is readily soluble in alcohol (1 in 10), ether, benzene, petroleum ether, and carbon disulphide. It burns with a smoky Uame and, when subjected to dry distillation out of contact with the air, yields resin oil. It contains abietic acid, Ci 9 H 28 O 2 , which has two double Unkings, melts at 165, and is soluble in hot alcohol. From gallipot rosin (Pinus maritima) pimaric acid, C 20 H 30 O 2 , m.pt. 148, has been obtained. Colophony has the rotatory power 69-6, and the acid number 145 to 185. One cubic metre of flr contains about 10 kilos of turpentine, which yields as much as 7 kilos of colophony, COLOPHONY 597 of the turpentine from resinous woods by means of suitable solvents (hot wood-tar mixed with pine oil ; U.S. Pat. 852,236) has been suggested. Oil of turpentine is rectified by heating with steam in presence of 0-5 per cent, of quicklime. As the oil always resinifies to some extent when exposed to the air, it is often desirable to redistil it before use. The strong and less agreeable odour of Russian and Greek turpentine oils is removed or lessened by shaking with a solution of permanganate, dichromate, or persulphate. Fresh oil of turpentine is clear, colourless, and highly mobile ; it has the sp. gr. 0-855 to 0-876 and boils at 156 to 161. It absorbs and combines with considerable quantities of ozone and oxygen part of the latter being converted into ozone and the oil at the same time resinifying. It dissolves sulphur, phosphorus, rubber, and resins, and is hence used for varnishes, lacs, oil paints, &C. 1 Permanganate in acid solution transforms it partly into Pinonic Acid, C 10 H 16 3 , while with dilute nitric acid it gives Terephthalic and Terebinic Acids, C 7 H 10 O 4 . It reacts violently with iodine in the hot, forming cymene. The relation between resins and aromatic compounds is established by the fact that when the former are distilled with zinc dust they form aromatic hydrocarbons, while if fused with potash they give di- and tri-hydroxy- benzenes. Resin substitutes or artificial resins are now prepared by heating phenols with formaldehyde in presence of hydroxy-acids (e.g. tartaric acid) or mineral acids (Blumer, Eng. Pat. 12,880 of 1902, and Fr. Pat. 361,539 of 1905 ; also Baekeland, 1909). According to the preponderance of laevo- or dextro-pinene, turpentine oil is laevo-rotatory (Venetian, German, and French) or dextro-rotatory (Australian). Pinene contains only one double linking, and hence unites with only 1 mol. of HC1, giving Pinene Hydrochloride, C 10 H 17 C1, which melts at 125, and has the smell of camphor ( Artificial Camphor). When treated with alcoholic potash, this hydrochloride is converted into CAMPHENE, C 10 H 16 , m.pt. 50, which is known in three optical modifications and IB while 1 cu. metre of pine gives 22 kilos of turpentine, this leaving 16-6 kilos of colophony ; the larch gives an intermediate yield. Colophony is used in large quantities for mixing with soaps (see Resin Soaps, p. 420), for sizing paper, for making varnishes, mastics, &c. In the United States 35 per cent, of the total output is used in soap-making. Large quantities of it are incorporated with artificial wax (cerasin), which is thus cheapened ; to deodorise the resin, it is finely ground, macerated with dilute sulphuric acid for five or six days and then suspended in hot water and subjected to a jet of steam for some time. After this treatment it melts and mixes well with the cerasin. Colophony is also used for making sealing-wax by mixing with shellac, turpentine, and a larger or smaller number of mineral substances (chalk, burnt gypsum, magnesia, zinc oxide, baryta, kaolin, &c.), according to the quality required ; the fused mass is coloured with cinnabar (for the finer red qualities), minium, ferric oxide, or red ochre. The best qualities contain only 40 per cent, of mineral matter and are mainly shellac, while the inferior kinds contain as much as 70 per cent, of mineral matter, the residue being principally colophony. Sealing-wax is coloured black by lampblack or boneblack, green by Prussian blue, yellow by chrome yellow, or blue by ultramarine ; when fused, colophony may be coloured also with algol or indanthrene dyes (?..). Italy imported 18 quintals of sealing-wax in 1898 and 61 in 1910, the corresponding exports amounting to 21 and 86 quintals respectively. In 1905 Germany imported 41,042 quintals of shellac and sealing-wax, of the value of 779,800, and exported 9575 quintals (196,280). In 1909 200,000 cases of sealing-wax were dispatched from Calcutta to England, Germany, and the United States. The importation of colophony into Italy amounted in 1896-1899 to an average of 122,700 quintals ; in 1905 to 125,000 quintals (72,450) ; and in 1910 to 149,000 quintals (119,200), mainly from North America. Its price varies from 12. to 28s. per quintal. In 1910 England imported 75,000 tons of.rosin (colophony), of the value of 880,582, and 8700 tons of shellac and sealing-wax, of the value of 627,629. The United States exported 2,269,000 barrels (2,474,800) of rosin in 1910 and 2,415,000 barrels (3,241,600) in 1911, and imported 12,000 tons (638,200) of sealing-wax in 1910 and 8000 tons (478,600) in 1911. 1 OIL OF TURPENTINE. Most common are the French, English, Russian, German, and American varieties, 40,000 tons of the last-named being landed at Hamburg, London, and Antwerp in 1897. In 1908 the outpiit in the United States was 1,700,000 quintals (2,800,000), one-half of this being produced in Florida. In 1902 Germany imported 63,600 barrels, in 1906 about 68,000 barrels, and in 1908 77,000 barrels (68,000 from America and 9000 from France) : in 1909 the imports were 318,884 quintals (1,200,000), and the exports 12,457 quintals, 100,000 tons of turpentine resins and balsams (880,000) being also imported, and 21,000 tons exported. In 1909 England imported 222,000 quintals of oil of turpentine, and in 1910 23,612 tons (1,001,216), whilst the United States exported 14,252,000 gallons (1,925,400) in 1910 and 18,198,000 gallons (2,187,400) in 1911. Italy's imports of oil of turpentine were as follow : 30,963 quintals in 1906 ; 30,088 in 1907 ; 33,316 in 1908 ; 26,932 in 1909, and 27,941 (111,760) in 1910. In the United States (on the Savannah market) the output was calculated at 675,000 barrels in 1907-1908, 725,000 in 1908-1909, and 580,000 in 1909-1910. In 1904 there were 1287 turpentine distilleries in the United States, with a total capital of 1,400,000, and in 1909 1585 distilleries with a capital of 2,480,000 and an output valued at 5,200,000. The price varies from 56s. to 76*. per quintal and reached a minimum in 1908. The smell of European turpentine oil has been improved by treatment with oxidising agents, such as perman- ganate, persulphates, or chromic acid, or, better still, with hydrogen peroxide sodium peroxide, barium peroxide, or oxides of nitrogen. By suitable application of Halphen's reagent (p. 381) or mercuric acetate, C. Grimaldi (1910) was able to detect adulteration with pine oil or resin oil. 598 ORGANIC CHEMISTRY transformed by oxidising agents into camphor and by ozone into the ozonide (Harries, 1910), these reactions establishing its constitution. FENCHENE is similar to camphene but is an optically inactive liquid, boiling at 158 to 160 ; it resists the action of nitric acid, but not that of permanganate. CAMPHANE, C 10 H 18 , forms white volatile crystals melting at 154 and boiling at 160, and is obtained by reducing d- or 1-bornyl iodide. It is optically inactive, and is the saturated hydrocarbon of the camphor nucleus. HOMOLOGUES OF TERPENES. The most interesting lower homologue is Hemi- terpene or Isoprene, C 5 H 8 (see p. 90), which gives various terpenic polymerisation products, such as (C 6 H 8 ) 3 (Clovene, Cedrene, Caryophyllene, &c.), C 20 H 32 (Colophene), C^H^ (Rubber), 1 &c. 1 RUBBER (caoutchouc) is obtained from the milky juice exuding when incisions are made in the stems of certain plants. The latter are mainly tropical trees (Apocynece. Moracece, Euphorbiacece, &c. ; Siphonia elastica, more especially in Brazil, and Urceola elnstica in Eastern India). According to Henri (1906-1908), the faintly alkaline latex contains the rubber, ready formed, in the form of minute emulsified drops (50 .millions per c.c.), which are in continual movement, and of which this author was able to obtain a cinematographic representation ; a coagulum is produced by acids, by salts of divalent metals (Ca, Mg, Ba, &c.), and less rapidly by salts of trlvalent metals, &c., but not by alkalis. The conditions of coagulation, which are not identical with different varieties of latex, are in general related to the quality of the rubber yielded. The best quality (Para rubber) is obtained by drying superposed thin layers of fresh latex in a mould by means of hot gases until about a hundred layers, each about 0-5 mm. in thickness, are obtained. The commoner qualities are set by the heat of the sun, with addition of acid, water, formalin, or a trace of mercuric chloride ; the electrolytic separation of rubber has also been suggested (Ger. Pat 218,927 of 1908), but, according to Pahl (Ger. Pat. 237,789 of 1910) hydrofluoric acid or carbon dioxide gives the best results. The fundamental component of rubber, the hydrocarbon, C 10 H 1( , is mixed with varying quantities of a resin soluble in alcohol or in acetone ; Para rubber contains less than 2 per cent, of resin, but the inferior qualities as much as 8 to 10 per cent. The removal of mineral substances and organic detritus is effected by manipulating, softening, and cutting the rubber in cold water, first in a kind of hollander (see Paper), then in a mechanical pulping machine and between rolls ; the water is then removed and the rubber dried at 40 to 50 The consumption of rubber on a large scale began fifty or sixty years ago, after it had been found possible to render it unattackable by ordinary reagents and solvents and to keep it elastic even when exposed to the air and to heat. This is attained by vulcanising (suggested in 1839 by Goodyear and by Hancock), which consists in mixing sulphur with the rubber and heating in an oven or in a steam apparatus at 110 to 140. Vulcanisation can also be carried out in the cold, by immersing the rubber in a mixture of sulphur chloride and carbon disulphide. The weighting materials or fillers, e.g. metallic oxides, kaolin, barytes, &c. (10 to 15 per cent.), are added with the sulphur. In place of sulphur chloride, which may give rise to a little hydrochloric acid, Bloch (Ger. Pat. 219,525 of 1908) has suggested hydrogen disulphide, H 2 S 2 , or trisulphide, H 2 S 3 , dissolved in acetone or carbon disulphide. According to C. O. Weber and Henriques (1894), in vulcanisation in the cold, the excess of S 2 C1 2 may form compounds with the rubber varying in composition from (C, Hi,) 24 , S 2 C] 2 with 4-3 per cent. S, to (C 10 H I6 ) 2 ,, (S 2 C1 2 ) 21 with 23-6 per cent. 8. E. Stern (1909) holds that the quantity of sulphur fixed is variable, while Hinrichsen (1910) maintains that the amount of S 2 C1 2 combined is constant. Ostwald (1910) explains vulcanisation as an adsorption phenomenon of the colloidal rubber, and assumes that the sulphur forms a series of reaction products, the first and last members of which cannot be isolated, and that the process is partly reversible. By protracted vulcanisation of rubber with 60 to 75 per cent, of sulphur or sulphide and mixing in minera substances (gypsum, chalk, inorganic colours, &c.), ebonite is obtained. (Guttapercha is similar to rubber bu contains oxygen.) Rubber has a brown or black colour and is insoluble in water when not vulcanised, and more or less soluble n chloroform, ether, petroleum ether, benzene, or carbon disulphide. A 7 ulcanised rubber is almost insoluble in these substances, but dichlorethylene, C 2 H 2 C1 2 , forms an excellent and non-inflammable solvent. With age, rubber (tubing, &c.) becomes hard and brittle, and cracks. According to Wo. Ostwald (Ger. Pat 221,310 of 1908), it lasts longer if quinoline, aniline, dimethylaniline, A-c., is used in its preparation. Rubber is recovered from vulcanised waste by subjection of the latter, after removal of the impurities and comminution, to the action of steam under a pressure of 6 atmos. By this means a large part of the sulphur seems to be converted into sulphuric acid, which can be readily removed with water or soda. The residue is com- pressed into strips, but it is always an inferior product. There are now numerous patents, some of them fanciful for devulcanising rubber by means of alkaline solutions, phenols, naphthalene, aniline (Ger. Pat. 99,689), &c. In 1907 Tissier obtained good results by macerating used, finely divided rubber with double its weight of terpineol in a closed vessel at 120 to 150, then diluting with 4 parts of benzene and decanting the solution from the impurities. The benzene is recovered by direct distillation and the terpineol by distillation in steam. In general, devulcanisation is based on depolymerisation of the vulcanised rubber where the sulphur is not united chemically with the rubber ; it is almost impossible to eliminate the sulphur which is combined chemically. Old rubber, well devulcanised and then again vulcanised, seems to give a more resistant product but of lower quality. The United States import 10,000 tons of used rubber for " reclaiming." The Mitchell process is often, used in America for obtaining rubber from old articles (boots, rubbered textiles, &c.), which are treated with sulphuric acid of 20 to 25 Be 1 ., this destroying the textile fibres but not the rubber. The chemical constitution of the hydrocarbon of rubber, C 10 H,,, was determined by Harries (1905) by means of its ozonide, C, H,,O 6 , which decomposes into levulinic aldehyde, so that the hydrocarbon must be regarded as OH 3 -C CH 2 CH 2 CH derived from an eight-carbon-atom ring (a ring never yet found in natural products); || || CH- CH 2 -CH 2 -C-CH 8 . In 1909 Harries obtained true artificial rubber by polymerising isoprenc in presence of glacial acetic acid in sealed tubes at 100 : (2C 5 H 8 )j. = (Ci H 16 ) r , but the process is too expensive to be used industrially. The firm of Bayer (Elberfeld) also obtained artificial rubber from isoprene and from Erythrene C 4 H, (see p. 90 ; also Ger. Pat. 235,423 and 235,686 of 1909 and Fr. Pat. 425,582 of 1911), by prolonged heating in presence of benzene, &c., but this product is also very expensive ; in consequence of this method of formation, the formula attributed RUBBER 599 Other hydrocarbons related to the terpenes are : ionene and irene, two isomerides of the formula, C^H^, the ketones of which, C^H^O, are irone and ionone, i.e. the aromatic principle of iris root, having a marked violet smell. IONONE (Artificial Essence of Violets) was prepared synthetically by Tiemann and Kriiger in 1883 by shaking equal proportions of citral and acetone with barium hydroxide solution, extracting with ether and expelling the latter by evaporation. The fraction of the residue boiling at 138 to 155 is Pseudoionone, which is transformed into the isomeric ionone by the action of dilute acid (Ger. Pat. 75,120). According to Ger. Pat. 113,672, the condensation may be effected by water in an autoclave at 170, while in presence of sodamide it takes place at the ordinary temperature (Ger. Pat. 147,839). See also Ger. Pat. 138,939. by Harries to the hydrocarbon is now contested. Various patents have recently been taken out for the preparation of isoprene, dimethylbutadiene, erythrene, &c., as prime materials for artificial rubber (Eng. Pats. 29,566 and 29,277 of 1909). The Badische Anilin und Sodafabrik (Ludwigshafen) obtained rubber by heating isoprene and dimethyl- butadiene (Fr. Pats. 417,170 and 417,768 and Eng. Pat. 14,281 of 1910) in presence of alkali, which has a poly- merising action. Harries (1911) showed that various isomeric artificial rubbers exist, with the generic formula fioTI,,, C 8 H 12 . Contrary to Weber's statement, Hinrichsen (1909) showed that the latex of rubber trees does not contain diterprncs, which polymerise to form rubber, but that the latter exists ready formed in the latex. The world's production of rubber was 52,190 tons in 1899 ; 59,750 in 1901 ; 68,500 in 1904 ; 73,680 in 1905 ; 75,300 in 1907, and about 80,000 tons (30,000,000) in 1910-1911. The French East African possessions give 7000 tons ; the French Congo 3000 and the Belgian Congo 6000. The total consumption in 1904 was distributed as follows : United States, 26,470 tons ; Germany, 12,800 (about 15,600 in 1909) ; England, 10,000 ; France, 4130 ; Austria-Hungary, 1320 ; Holland, 1218 ; Belgium, 748 (increasing rapidly) ; Italy, 548. One-half of the world's output of rubber comes from Brazil, which produced 28,600 tons in 1902 and nearly 34,500 tons in 1908. Mexico exports rubber to the value of 1,280,000. The exports from Ceylon were 450 tons in 1908 ; 750 in 1909, and 1700 in 1910, while Abyssinia exported 9 tons in 1908 and 79 tons (15,280) in 1909. A considerable part of the total output of rubber comes from Africa (Senegal, Madagascar, the Congo, the Cameroons, &c.), the exports being 16,000 tons in 1900 and 23,500 in 1906. The East Indies produce about 2000 tons per annum, but the culti- vation of rubber is increasing rapidly. Brazil exported about 27,000 tons of rubber, at 60 per ton, in 1902 ; 31,600 tons in 1905 ; and 35,000 tons, valued at 1,720,000, in 1906. In Malacca the English have rapidly extended the rubber plantations, the exports being valued at ,17,000 in 1900, ana at 720,000 (1580 tong) in 1908, 1017 tons being exported in 1907. The number of rubber-producing trees was 27,500,000 in 1907, and 37,500,000 in- 1908. The most suitable climate and soil for rubber are found at Malacca. The areas under nibber in different countries in 1911 were as follow': Malacca, 170,000 hectares ; Ceylon and Southern India, 110,000 ; Borneo, 35,000 ; Mexico, Brazil, and Africa, together, 45,000 ; German colonies, 20,000 The mean yield of rubber is calculated at 42 kilos per hectare in Ceylon and 38 in Samoa. The Russo- American Ilubber Company of St. Petersburg produced rubber articles to the value of 4,800,000 in 1907, 5,360,000 in 1908, and 5,800,000 in 1909. In Germany there were 339 factories (with 12,500 employees) for making rubber and gutta-percha articles in 1895, and 100 factories, with 35,000 employees, a capital of 5,600,000, and an annual output valued at 10,000,000, in 1908. ID 1904 Germany imported 17,407 tons of rubber of the value of 5,400,000, and exported 4569 tons, worth 1,400,000. In 1908 Germany imported more than 13,000 tons of raw rubber, 1500 tons of it from the German African colonies, especially the Cameroons ; in 1910 the imports were about 10,000 tons. In Italy the firm of Pirelli and Co., founded in 1872, has two factories for the working of rubber, with a total capital of 600,000, an annual turnover of 640,000 (including submarine cables), and about 4000 operatives. A less important factory is that of the Italian company for the manufacture of rubber at Milan. Eaw rubber imported into Italy pays no duty, but the manufactured product pays from 2 to 3 per quintal. The imports of raw rubber and gutta-percha amounted to 6688 quintals in 1904, 7669 in 1905, 1500 in 1908, and 18,800 (980,000) in 1910; the exports, in the form of string, ribbon, and tubing, being 639 quintals in 1904, 1183 in 1905, and 109 in 1910. The total imports were 5715 quintals in 1904, 8061 in 1905, 2400 in 1908, and 2300 (112,000) in 1910 ; and the tota exports, 3580 quintals in 1904, 4884 in 1905, 625 in 1908, and 1150 (92,000) in 1910. Belgium occupies 1900 workpeople in the rubber industry and France 9000 (in 1901). The amount of rubber imported into England was 35,000 tons (14,138,200) in 1909 and 43,500 tons(26,096,790 together with 4800 tons (1,136,500) of gutta-percha, in 1910. The imports of rubber to the United States were 40,500 tons (19,600,000) in 1910 and 37,000 tons (14,880,000) in 1911, in addition to 16,000 tons (601,000) of waste rubber in 1910 and 8600 tons (306,000) in 1911. Gutta-percha resembles rubber, and is obtained mostly in Singapore and Borneo from a large tree, Isonandra percha. After purification and manipulation in hot water, it sets in the cold to a hard mass, soluble slightly in ether and alcohol and more readily in hot benzene, carbon disulphide, or chloroform. It is used as an electrica insulator and for making various articles. The price of raw rubber has increased rapidly from 320*. per quintal in 1850 to 448s. in 1890 and 664s. in 1907 (at Hamburg). The finer qualities of Para rubber cost 720s. in 1902, 1280s. in 1905, and 1280s. to 1600s. in 1909 Gutta-percha costs 360s. per quintal. It is estimated that the cost of native labour in the Congo district is not more than 120s. per quintal. RUBBER SUBSTITUTES. Many of these have been prepared, but the only one of much practical import- ance is the so-called factis, of which two types are on the market : white and brown or black. The latter is made by boiling rape oil or linseed oil in an open vessel for two hours, cooling, and passing a current of air through it for thirty-six hours. It is then vulcanised by adding 2 per cent, of flowers of sulphur, heating for two hours at 140 adding a further 1 per cent, of sulphur, and raising the temperature to 150, when it begins to rise. White factis is obtained by treating the oil with 20 to 25 per cent, of sulphur chloride (free from dichloride) ; the energy of the reaction may be modified by adding the sulphur chloride dissolved in carbon disulphide. The mass is obtained in sheets or blocks by pouring it immediately on to cold metal plates or moulds. These substitutes are almost as elastic as rubber and are used to adulterate rubber, their price being 72s. to 96s. per quintal ; they are insoluble In water or acid, but dissolve slightly in dilute alkali. They are distinguished from rubber by being saponiflable with alcoholic potash. 600 ORGANIC CHEMISTRY The constitution of synthetic ionone is : CH 3 CH 3 v C . CH 2 CH-CH:CH-CO-CH 3 I CIi2 C CH 3 \/ CH Ionone (100 per cent.) costs 152 per kilo, and fi-ionone, 60 ; the 20 per cent, solutions are sold at one-fifth of these prices. CAMPHORS While the terpenes are liquids, -the camphors are generally solid. They contain alcoholic or ketonic oxygen, and the principal ones with a single ring are : Menthone, C 10 H ]8 O, and Terpinol with the same formula, while Menthol H, n 9 . Among the camphors and Carvomenthol are C 10 H 20 0, andTerpin C 10 j.j. 20 w 2 . with complex rings are true Camphor, Fenchone, and Carone. C 10 H 16 0, and Borneol, C 10 H 18 0. The camphors poorer in hydrogen and oxygen contain double linkings, form additive products, and are readily oxidised, while the others behave like saturated compounds. When reduced with sodium, the ketonic camphors yield the alcoholic camphors, which are converted into the former on oxidation. It is possible to pass from the camphors to the terpenes by way of the chlorides, and reduction of the alcoholic camphors often gives the terpene hydrocarbons. Thus, the Terpane (hexahydrocymene) can be obtained by reducing the Terpanol (menthol, C 10 H 20 0), which contains a hydroxyl or secondary alcoholic group, this being transformed by oxidation into the ketonic group with formation of Terpanone (menthone), 1 so that the hydroxyl should be in the ortho -position with respect to the CH 3 and C 3 H 7 groups, as is shown below in the constitutional formulae. 214 On the other hand, since Carvacrol, C 6 H 3 (OH)(CH 3 )(C 3 H 7 ) (isomeric with carvone or carvol), of known constitution, gives on reduction a terpanol (carvomenthol with the hydroxyl in the position 2) different from that of menthol, the hydroxyl of the latter must be in position 3 : H CH< H CH, H CH< H CH, H 2 H(OH) H, H C 3 H 7 Menthol (terpanol), C 10 H 20 O H, Menthone (terpanone), C 10 H 18 O H(OH) H, H C 3 H 7 Carvomenthol Ci H 20 O H C,H 7 Carvomenthone Ci.H I8 MENTHOL (3-Terpanol),C 10 H 19 - OH, occurs in abundance in oil of peppermint, from which it can be obtained crystalline by cooling. It melts at 42, boils at 213, and has the strong odour of peppermint. The position of the OH is established by the fact that, 1 Ciamician and Silber (1910) showed that, in alcoholic solution and nnder the action of light menthone i hydrolysed with formation of decoic acid, and nn aldehyde isomeric with citronellal (p. 210) CAMPHORS 601 with bromine in chloroform solution, menthone (which is the corresponding ketone, boiling at 207, and having a strong smell of peppermint) gives dibromomenthone, and elimination of 2HBr from the latter gives thymol having the known constitution, C 3 H Y ^>CH 3 ; o!T~ the CH 3 and OH are here undoubtedly in the meta -position, since elimination of the C 3 H 7 by means of P 2 O 5 yields m-cresol. When heated with copper sulphate, menthol yields cymene. Four isomerides of menthol are possible theoretically. It is used as an anaesthetic and as a disinfectant. PULEGONE (A 4 ( 8 )-Terpen-3-one), C 10 H 16 O, predominates in oil of pennyroyal (Mentha pulegiiim). It is a ketone boiling at 222, and on reduction gives menthol, so that the carbonyl group is in position 3. CARVONE (Carvol or Terpadien-2-one), C 10 H 14 O, is a ketone giving Carvoxime, C 10 H 14 : NOH, which exists in optical isomerides and is identical with nitrosolimonene. It forms the principal component of cumin oil, boils at 228, and is converted into Carvacrol, C 10 H 13 - OH, when heated with potash or phosphoric acid : CH 3 CH 3 I ' I C C /\ /\ CH CO CH C-OH CH 2 CH 2 CH CH \/ ' \/ CH C I I CH 3 C = CH 2 CH 3 CH CH 3 Carvone Carvaorol TERPENOL (A 4(s) -Terpen-l-ol), C 10 H 18 O, melts at 70, and, like tetramethylethylene, forms a solid blue nitrosochloride, the double linking being in the 4(8)-position, between two tertiary carbon atoms. TERPINEOL (A^Terpen-S-ol), C 10 H 18 O, melts at 35, boils at 218, and is known in the form of various optically active isomerides. It has a pleasing odour of lily of the valley, lilac, and cyclamen, and occurs in ethereal oils. With sulphuric acid it forms terpin hydrate, which is also converted back into terpineol by sulphuric acid. TERPIN (1 : 8-Terpandiol), Ci H 18 (OH) 2 . Terpin hydrate, CxoHaoO^HaO, is slowly formed from oil of turpentine, C 10 H 16 , in contact with dilute nitric acid and alcohol. This crystalline hydrate melts at 117 and then loses 1 mol. of H 2 0, anhydrous terpin distilling over at 258. This is optically inactive and is not obtainable in active modifications, so that the presence of asymmetric carbon atoms is excluded. The hydrate is also obtainable from geraniol by the prolonged action of 5 per cent, sulphuric acid, 2H 2 being added at the double linkings : CH 3 CH 3 CH 3 CH 3 \/ i I C CH 3 C OH CH 3 C OH II I I CH CH 2 CH / / /\ CH 2 CH 2 -OH nw CH 2 CH 2 -OH CH 2 CH I " I . Jl'Un I | TT n + TT TT > | n 2^ CH 2 CH CH 2 CH 2 CH 2 CH 2 Y Y . Y I /\ /\ CH, CH 3 OH CH 3 OH Geraniol Terpin hydrate Terpin 602 ORGANIC CHEMISTRY Nitric oxide oxidises terpin, giving Terebic Acid, which has the known constitution C0 2 H (CH 3 ) 2 : C CH CH 2 ' ! I' O CO so that the position 8 must be occupied by a hydroxyl ; the other hydroxyl can only be in position 1, since otherwise an asymmetric carbon atom would be obtained. CINEOL, CioHigO, has the constitution of terpin less H 2 O, which is eliminated from the two hydroxyls, an atom of oxygen thus remaining united to the two carbon atoms 1 and 8. Cineol melts at 1, boils at 176 and occurs in abundance in eucalyptus oil and in oil of wormseed. FENCHONE, C 10 H 16 O. The dextro-form occurs in fennel oil and the la?vo in thuja oil. It is a ketone similar to camphor and can be converted into Fenchene. CAMPHOR (ordinary camphor, laurel camphor, or. Japan Camphor). C 10 H 16 O, is the constituent which separates in the solid form from the essential oil of Laurus camphora, a tree which is cultivated in China, Japan, and Formosa , and grows well in Southern Europe (Italy). The wood (thirty to forty years old) is chopped up and boiled with water until the camphor floats at the surface ; on cooling, the crude camphor sets to a solid mass, which can readily be separated. In some cases the camphor is distilled directly from the wood in a current of steam. The yield is about 1 kilo per quintal of wood. The crude product is refinsd by mixing with quick- lime and charcoal and subliming at a gentle heat. It is obtained thus as a white, crystalline, and not very hard mass, which has a characteristic odour, and partially sublimes at the ordinary temperature. It melts at 178, boils at 207, and has the sp. gr. 0-922-0-995 (the finer Borneo camphor has sp. gr. 1-10). In alcoholic solution it is more or less dextro- rotatory, according to its origin, but matricaria camphor (from the leaves of feverfew, Matricaria parthenium) is laevo-rotatory. With iodine in the hot it forms carvacrol (we above), while oxidation with nitric acid gives Camphoric Acid, C 8 H 14 (C0 2 H) 2 , which exists in two active and two inactive forms. Further oxidation yields Camphoronic Acid. C 9 H 14 6 , which gives trimethylsuccinic acid on dry distillation. When distilled with P 2 5 , camphor loses H 2 and forms cymene. On reduction with nascent hydrogen, ordinary camphor gives Borneol (Borneo camphor), C 10 H 17 -OH, which melts at 208, boils at 212, and when oxidised gives ordinary camphor, which it strongly resembles. Between 1860 and 1893 various constitutional formulae for camphor were proposed by Kekule, Armstrong, Bredt (1884), and G. Oddo (1891), the last of whom gave a formula which explained well all the reactions and properties observed up to that time. More and more acceptable constitutions were given by Widmann (1891), Collie (1892), Bouveault (1892), &c., and finally by Bredt (1893). The constitution of camphor now seems to be definitely established as the result of various syntheses, especially that from ethyl oxalate and ethyl /3/3-dimethylglutarate, two compounds which are obtainable synthetically from their elements. The various stages in this synthesis are as follow, R indicating the alkyl group : COOR H-CH-C0 2 R CO CH C0 2 R CO CH- C0 2 R + CH.-C-CH, CH. CH., "C 'CHr COOR H-CH-C0 2 R CO CH C0 2 R CO C(CH 3 )-C0 2 R Ethyl Ethy /3/3-dimethyl- Ethyl diketoapo- Ethyl diketo- oxalate glutarate camphorate camphorate CH , C CH t CAMPHOR 603 \ CH 2 CH - C0 2 R CH 2 CH CH 2v CH 2 CH-CH 2 -ON q C CH, >0 CH, C CH 3 r * J *~>i.i. 3 CH 2 C(CH 3 )-C0 2 R CH 2 C(CH 3 ).Or CH 2 C(CH 3 ) CO 2 H Ethyl campliorate Campholidc Homocamphoric nitrile CH 2 -CH CH 2 .C0 2 H CH 2 CH CH 2 CH , C CH : CH, C CH CH 2 C(CH 3 ) C0 2 H CH 2 C(CH 3 ) CO Homocamphoric acid Camphor a This constitutional formula proposed for a-camphor by Bredt, although still contested, is the one generally accepted by chemists, since it corresponds best with most of the reactions of camphor. In 1911, Bredt and Hilbing prepared (3-camphor, containing the CO group in the /3-position, from bornylene- carboxylic acid ; it melts at 182 and boils at 213-4. Camphor forms strongly rotating energetic sulphonic acids, e.g. CH-S0 3 H C 8 H 14 <^ | , which are able to resoive many racemic compounds into ^CO their active components. Since many terpenes give camphor on oxidation, many attempts have been made to prepare artificial camphor from oil of turpentine. 1 The latter contains pinene, C 10 H 16 , which is readily convertible into borneol, C 10 H ]7 -OH, or isoborneol, this giving the inactive racemic compound corresponding with natural camphor on oxidation. According to Ger. Pat. 134,553, when anhydrous turpentine is heated for a long time at 120 to 130 with dry oxalic acid, a mixture of camphor with pinyl formate and oxalate is obtained ; after washing with water, the latter are hydrolysed with alkali and the resultant borneol converted into camphor by oxidation with dichromate and sulphuric acid. At Monville, near Rouen, a factory was erected in 1906 to manufacture artificial camphor by the process described in Fr. Pat. 349,896 (of Behal, Magnier, and Tissier, and similar to U.S. Pat. 779,377) : A mixture of oil of turpentine and salicylic acid is heated and, after elimination of the excess of the reagents, the isoborneol ether is hydrolysed to a mixture of borneol and isoborneol. Another factory, near Calais, utilises Schering's method (Fr. Pat. 341,513), already in use on a large scale in Berlin, and also applied in a factory established in 1909 in Finland. According to Fr. Pat. 349,852, pinene hydrochloride is first prepared and then heated under pressure with lead acetate in acetic acid solution, thus giving camphene, which with permanganate forms camphor ; or treatment of the pinene hydrochloride with a formate gives the formic ester of borneol, which can be readily hydrolysed. The final oxidation to obtain camphor is carried out in various ways : by oxidising the borneol, in benzene or petroleum ether solution, with aqueous alkaline permanganate (Ger. Pat. 157,590), or by means of ozone, air, or chlorine water (see Eng. Pat. 28,036 of 1907 and Ger. Pats. 166,722 1 It has been pointed out that a difficulty in the way of the further development of the present, artificial camphor industry may be the excessive price of oil of turpentine, this having risen from 56s. per quintal in 1900 to 96s. in 1906 ; these conditions might easily be aggravated by the formation of a trust. Further, the demand for camphoi may diminish in the future, since substitutes are continually being found capable of replacing it in celluloid, which up to the present has consumed about two-thirds of the total camphor produced. The fact that natural camphor almost entirely monopolised by the Japanese Government can be sold, without loss, at 144s. per quintal constitutes a menace to the future of artificial camphor, which could never be sold at that price and depends on a raw material the price of which cannot be regulated. In addition to the 70 per cent, absorbed in the manufacture of celluloid, natural camphor is used for explosive powder and guncotton (2 per cent.), for pharmaceutical preparations (13 per cent.) and for various other purposes (15 per cent.) 604 ORGANIC CHEMISTRY and 154,107), or by oxidising isoborneol in aqueous acid solution with permanganate (Ger. Pat. 197,161 of 1906). Camphor was obtained by A. Hesse by means of the Grignard reaction, and it is also formed by fusing borneol with finely divided nickel (1911). Natural camphor may be distinguished from the artificial product by mixing it intimately with an equal weight of chloral hydrate : the former gives a syrupy mass, but the latter does not liquefy. Camphor is used in pharmacy, for fireworks and nightlights and, in large quantities, in the manu- facture of celluloid x and for rendering explosives insensitive to shock. The price of camphor varies somewhat, and during the Russo-Japanese War rose considerably ; it is usually about 24 per quintal. The cost price of artificial camphor seems to be about 4s. per kilo. The production of camphor in Japan and Formosa (State monopoly) amounted to about 3,500,000 kilos (three-fourths in Formosa) in 1906, and to 4,300,000 kilos, 2,000,000 kilos being exported (two-thirds to Havre, London, and Hamburg, and one-third to America), in 1907. After the war with Russia, Japan, with her monopoly of the production of camphor, tried to raise the price. From about 2s. 5d. per kilo in 1903 it became 4s. 5d. in 1906 and 1907. At the same time large plantations were laid out in Formosa, 1,300,000 trees being planted in 1907, about 1,400,000 in 1908, and more than 5,000,000 in 1909. The rise in price caused increased production of artificial camphor in Europe, and, owing to this competition, the price fell again to 2s. per kilo at Japanese ports in 1911. The exporta- tion from Japan, which had fallen to 1,500,000 kilos in 1908, rose to 2,430,000 kilos in 1909, but steps are now being taken to regulate the output so that the price may not be lower than the actual cost of production, this being about Is. Qd. in Formosa and 2s. Wd. for camphor produced in Japan. In 1906 Japan prepared 1,600,000 kilos and. in 1907 more than 3,000,000 kilos of camphor oil. An association was formed in the United States in 1908 to sterilise the camphor trees of Florida and Texas, the system of cultivation being thus improved. The exports of camphor from China were as follow : 120 quintals in 1902 ; 660 in 1903 ; 725 in 1904; 2450 (60,000) in 1905 ; 6000 (220,000) in 1906 ; 11,600 (280,000) in 1907 the total output being 16,000 quintals ; 4820 (108,000) in 1908, the total output being 8000 quintals ; 100,000 worth in 1909 and still less in 1910. Trees have been used up in China without new plantations, which require 40 years before giving good trees, being established. Continuance of this procedure will result in the complete destruction of camphor trees in China in seven or eight years. England imported nearly 387 tons of camphor in 1899 ; France about 546 tons and Germany more than 1000 tons (320,000) in 1903. The average price of camphor in Hamburg was 141s. in 1881-1885, 181s. in 1886-1890, 235s. in 1891-1895, 268s. in 1896, and 200s. in 1897. It now tends to increase owing to the Japanese monopoly. The United States imported in 1910 1700 tons (226,000) of crude natural camphor, and in 1911 1150 tons (165,000), besides 160 tons (23,000) of purified and artificial camphor. The exports of celluloid and articles made therefrom amounted to 286,800 in 1910 and 420,200 in 1911. The world's consumption of camphor is about 5000 to 6000 tons, almost all placed on the market by Japan, which has collected Chinese camphor since the monopoly there came to an end in 1904. Camphor is not produced in Italy, where, however, according to the excellent -mono- 1 Celluloid is obtained by mixing nitrocellulose and camphor in the following manner : well-stabilised powdered, and partially dried collodion-cotton (with 10 to 11 per cent. N. ; see p. 239) is soaked in alcohol in a covered centrifuge, then gelatinised with alcohol and one-third or one-fourth of its weight of camphor, coloured, if necessary, homogenised between rolls and then formed into dense, compact blocks by pressing while hot. It is then ready to be cut, sawn, compressed, polished, &c., its marked plasticity when hot being utilised in working it. It is a homogeneous, transparent, colourless, or yellowish substance without teste and of sp. gr. 1-37. If sufficiently dry it is odourless, but, when rubbed or heated, it develops a slight smell of camphor. It is a very bad conductor of heat and electricity, and its elasticity is about equal to that of ivory. Celluloid is used for making toys, balls, combs, walking-stick handles, tortoiseshell objects (substitutes for tortoiseshell, amber, ebonite, &c.), films, &c. It has the disadvantage of burning rapidly and energetically (without exploding) when brought into contact with an ignited or incandescent body. If the collodion -cotton used is well stabilised, celluloid will withstand a temperature of 125 or even higher. It can be charged with mineral substances to render it less inflammable and heavier. CONDENSED BENZENE NUCLEI 605 graph presented by Giglioli at the International Congress of Applied Chemistry at Rome (1906), it could be produced advantageously on a large scale. Italian requirements are met by the importation of 20 to 25 tons per annum of refined camphor from Germany at a price of 280s. per quintal prior to 1898, 400s. after 1902, and 480s. after 1908 ; the import duty is 20s. per quintal. Italy imported 1000 quintals of celluloid in 1906, 2407 in 1909, and 4267 (102,410) in 1910. The value of the imports of celluloid into Japan were : 48,000 in 1905, 80,000 in 1906, 30,000 in 1907, and 60,000 in 1908. The import duty is 4 per quintal for celluloid in strips or in the crude state, and 40 per cent, for manufactured articles. Two large celluloid factories erected in Japan in 1908 produce annually 500,000 kilos, 300,000 kilos for exportation, and the rest for home consumption, this being previously supplied by importation from Germany (five-sixths) and England (one-sixth). In three large centres (Xeckaren, Troisdorf, and Eilenburg), Germany produces annually about 5,500,000 kilos of celluloid of the value of 1,360,000. P. CONDENSED BENZENE NUCLEI DIPHENYL AND ITS DERIVATIVES DIPHENYL, C 6 H 5 -C 6 H 5 , or <^ \ is formed by treating an ethereal solution of bromobenzene with sodium (Fittig), by the trans- formation of hydrazobenzene, or by diazotising benzidine and decomposing the resultant product. It can also be obtained by passing benzene vapour through a red-hot tube. It forms colourless crystals melting at 71 and boiling at 254, and is soluble in alcohol and in ether. On oxidation with chromic acid, it gives benzoic acid, its constitution being thus confirmed. Of monosubstituted products of diphenyl, three isomerides are possible, corresponding with the o-, m-, and p- positions with respect to the carbon joined to the second nucleus. Disubstituted derivatives exist in numerous isomeric forms, as the substitution may occur in only one nucleus or in both ; in general, however, the substituents enter preferably the para-positions. BENZIDINE (p : p-Diaminodiphenyl), NH 2 -C 6 H 4 -C 6 H 4 'NH 2 . Nitration of diphenyl yields p : p-dinitrodiphenyl, which, when reduced with zinc dust in alkaline solution, gives benzidine. The latter may also be obtained by electrolysis of nitrobenzene ; see also Ger. Pat. 122,046, according to which azobenzene is electrolysed in hydrochloric acid solution in presence of stannous chloride. When pure, benzidine forms colourless scales melting at 122 and then subliming. It dissolves slightly in cold water, but readily in hot water, ether, or alcohol. It is a diacid base and gives a sulphate, C ]2 H 8 (NH 2 ) 2 ,H 2 S0 4 , almost insoluble in water. It is largely used in making substantive dyestuffs (such as Congo red and chrysamine, which dye cotton without mordants), being first diazotised and then combined with naphthylamine or naphthalenesulphonic acids. Crude benzidine costs about 5s. per kilo and the pure product 48s. The crude sulphate in paste (63 per cent.) costs 2s. per kilo and the pure 36s. A higher homologue of benzidine is o-Tolidine, C 12 H 6 (CH 3 ) 2 (NH 2 ) 2 , which melts at 128, and the diazo-compound of which combines with naphthionic acid to form a red substantive dyestuff, benzopurpurin 45. The dimethoxy- compound, (O'CH 3 ) 2 , of tolidine is dianisidine, which with a-naphthol-a- sulphonic acid forms benzoazurin G (substantive blue). CeH 4 \ CARBAZOLE, | NH, is found in coal-tar, and can be obtained 606 ORGANIC CHEMIST RjY synthetically by distilling o-aminodiphenyl over red-hot lime or by gently heating diphenylamine vapour. The unions of the nitrogen with the two phenyl groups are in the diortho- positions, so that carbazole may be regarded as a pyrrole derivative (see later). It forms colourless scales melting at 238 and readily subliming, and it dissolves in concentrated sulphuric acid, giving a yellow coloration. From diphenyl can be derived : four isomeric Dihydroxydiphenyls, C 12 H 8 (OH) 2 ; the Diphenylsulphonic Acids ; Diphenyl Oxide, (.C 6 H 4 ) 2 ; Hexa- hydroxydiphenyl, C 12 H 8 (OH) 6 (the mother-substance of caerulignone) ; and the Diphenylcarboxylic Acids (the di-p-acid is a white powder, insoluble in water, alcohol or ether ; the di-o-acid is Diphenic Acid, CO 2 H C 6 H 4 C 6 H 4 CO 2 H, m.pt. 229) which give diphenyl when heated with lime. 2. DIPHENYLMETHANE AND ITS DERIVATIVES These compounds may be obtained by condensing either 2 mols. of benzene (or its homologues) with one of methylene chloride, or 1 mol. of benzyl chloride (or benzoyl chloride) with one of benzene (or its homologues or derivatives) in presence of aluminium chloride : 2C 6 H 6 + CH 2 C1 2 - 2HCI+ C 6 H 5 -CH 2 -C 6 H d CgHg -j- CgH 5 'CH 2 'Cl = HC1 -f- GgHij' GH 2 - C 6 H 5 . Condensation of 2 mols. of benzene with aldehydes (Baeyer) or 1 mol. of an aromatic aldehyde with one of benzene (V. Meyer) under the influence of concentrated sulphuric acid (ketones, phenols, tertiary anilines, &c., also act similarly) : 2C 6 H 6 + CHg-CHO = H a O + CH 3 -CH(C 6 H 5 ) 2 ^6^6 + C 6 H 5 -CH 2 -OH = H 2 + C 6 H 5 -CH 2 -C 6 H 5 . DIPHENYLMETHANE, C 6 H 5 -CH 2 -C 6 H 5 , forms white crystals melting at 26 and boiling at 262, has a smell of oranges and is soluble in alcohol or in ether. It is obtained synthetically (see above). With water at 150, its bromo- derivative, CHBr(C 6 H 5 ) 2 , is converted into Benzhydrol (diphenylcarbinol), (C 6 H 5 ) 2 CH-OH, which is also obtained on reducing benzophenone. p-Diaminodiphenylmethane, CH 2 (C 6 H 4 -NH 2 ) 2 , and Tetramethyldiamino- benzhydrol, OH-CH[C 6 H 4 -N(CH 3 ),] 2 , are used in the preparation of dyestuffs. BENZOPHENONE (Diphenylketone), C 6 H 5 -CO-C 6 H 5 (see p. 572). o-DIHYDROXYBENZOPHENONE, [C 6 H 4 (OH)] 2 CO, by the elimination of a molecule CO of water from the two hydroxyls, gives Xanthone, C 6 H 4 <^ ~ ^>C 6 H 4 . p-Dihydroxy- benzophenone is obtained from anisaldehyde, so that the hydroxyl groups must be in the para -posit ions. Trihydroxybenzophenone is formed by the condensation of benzoic acid with pyrogallol in presence of zinc chloride. It is used in dyeing under the name alizarin yellow C (see Dyestuffs). Other higher derivatives of diphenylmethane are a s s follow : as-DIPHENYLETHANE (see later symm.dibenzyl) is liquid and is formed from paraldehyde and benzene (see above). Benzilic Acid (diphenylgly collie acid), (C 6 H 6 ) 2 C(OH)-CO 2 H, is a solid and is obtained by the action of KOH on benzil ; by reduction with HI it gives Diphenylacetic Acid, (C 6 H 5 )2CH'CO 2 H. Tolylphenylmethane, C 6 H 5 -CH 2 -C 8 H 4 'CH 3 , exists in several isomeric forms. TRIPHENYLMETHANE 607 Tolyl Phenyl Ketones, C 6 H 6 -COC 6 H 4 -CH 3 . The stereoisomeric oximes of these ketones were employed by Hantzsch in developing the stereochemistry of nitrogen (see pp. 20, 210, and 306). Benzoylbenzoic Acids, C 6 H 5 -CO-C 6 H 4 -COoH : the ortho-acid gives anthraquinone when heated at 180 with P 2 5 . C 6 H 4 -. FLUORENE (Diphenylenemethane), | /CH 2 , is found in coal-tar, and is C 6 H 4 / formed on heating diphenylmethane vapour. It melts at 113, boils at 295, and forms scales showing a violet fluorescence. 3. TRIPHENYLMETHANE AND ITS DERIVATIVES These are prepared synthetically by processes analogous to those used for diphenyl methane, but under such conditions as to lead to the condensation of three benzene nuclei in the methane molecule. The action of chloroform on benzene in presence of A1C1 3 gives TRIPHENYLMETHANE (m.pt. 93; b.pt. 359) : CHC1 3 + 3C 6 H 6 = 3HC1 + CH(C 6 H 5 ) 3 . The condensation of benzaldehyde and dimethylaniline yields Tetramethyl- diaminotriphenylmethane, C 6 H 5 -CH[C 6 H 4 -N(CH 3 ) 2 ] 2 , which is a leuco-base (see Dyestuffs) of malachite green ; phenols, &c., condense similarly. When this colourless leuco-base is oxidised with Pb0 2 and HC1, it gives Tetramethyl- diaminotriphenylcarbinol, C 6 H 5 C(OH) [C 6 H 4 N(CH 3 ) 2 ] 2 which is also a colour- less base and forms colourless salts. When, however, these salts are heated in solution, they lose water and form an intense green colour ing -matter, the double salt of this with zinc chloride or oxalate being known as malachite green : /C 6 H 4 N(CH 3 ) 2 ,HC1 /C 6 H 4 N(CH 3 ) 2 ,HC1 C 6 H 5 -C< -C 6 H 4 -N(CH 3 ) 2 ,HC1 H 2 + C 6 H 5 -C/ / \_ N(CH } / or C 6 H 5 -C<^ ; C 6 H 4 -NC1(CH 3 ) 2 on reduction, the colouring-matter (+ 2H) gives the leuco-base again. PARAROSANILINE is obtained by oxidising 1 grm.-mol. of p-toluidine and 2 grm.-mols. of aniline with arsenic acid or nitrobenzene. The methyl of the toluidine furnishes the carbon atom for the methane nucleus : CH 3 -C 6 H 4 -NH 2 + 2C 6 H 5 -NH 2 + 30 = 2H 2 + OH-C^-C 6 H 4 -NH 2 (base). X C 6 H 4 -NH 2 With acids, this base gives a red colouring- matter which is precipitated by alkali. When reduced with zinc and hydrochloric acid it yields paraleucaniline, HC(C 6 H 4 -NH 2 ) 3 , in colourless crystals which give the coloured base again on oxidation. Elimination of the amino-groups by diazotisation leads to triphenylmethane, while nitration of the latter, followed by reduction, gives paraleucaniline, which yields triaminotriphenylcarbinol on oxidation. When treated with acids the latter loses H 2 0, giving the colouring-matter : .(C 6 H 4 -NH 2 ) 2 X (C 6 H 4 -NH 2 ) 2 OH-Cf = H 2 + CClf X C 6 H NH 2 ,HC1 X C 6 H 4 NH a 608 ORGANIC CHEMISTRY ROSANILINE is formed by oxidising a mixture of o- and p-toluidines and aniline with arsenious anhydride, mercuric nitrate, or nitrobenzene, the carbon of the methane nucleus being furnished in this case also by the p-toluidine : CH CTT C 6 H/ 3 (p) + C 6 H/ 3 (o) + C 6 H 5 -NH 2 + 30 = X NH 2 X NH 2 2H 2 + OH-O V X (C 6 H 4 -NH 2 ) 2 Rosaniline hydrochloride (with 1 HC1) or fuchsine forms crystals with a green metallic lustre, while the aqueous solution is red owing to the presence of the monovalent cation, C 19 H 18 N 3 , the salt being almost completely ionised. All fuchsine salts, at the same dilution, give the same absorption spectrum, as they contain the same cation. If 3HC1 are combined, the salts become yellow (yellow trivalent cation) ; indeed, with excess of HC1 fuchsine is almost decolorised, although in dilute solution the red cation is again formed by dissociation. 1 Replacement of the hydrogen atoms of the amino-groups by alkyl groups gives various colouring-matters, the intensity of the violet colour increasing with the number of methyl groups. Pentamethylpararosaniline is the methyl violet of commerce. ROSOLIC ACID and AURIN are the phenolic compounds corresponding with rosaniline and pararosaniline, from the diazo -compounds of which they are obtained by boiling with water : f!H ff! H -OH^ V^iJ. q Sf\\J Ci. A \J J--L /O ^C F~ ^ X C H Rosolic acid Aurin They are colouring-matters of an acid character and of but little importance and they form dark red prisms with a greenish, metallic reflection. PHTHALOPHENONE, C^-C 6 H 5 , may be regarded as a deriva- tive of phthalic acid (see p. 580) or of tripheny] methane. It is the anhydride of triphenylcarbinol-o-carboxylic acid, C(OH)(C 6 H 5 ) 2 (C 6 H 4 -C0 2 H), and is obtained by heating phthalyl chloride with benzene in presence of aluminium chloride. It forms scales melting at 115 and dissolves in alkali giving a salt of the acid, the latter not being obtainable in the free state. Its phenolic derivatives are the phthaleins (see p. 581). HEXAPHENYLETHANE, (C 6 H 5 ) 3 C.C(C 6 H 5 ) 3 , is of some interest theoreti- cally, as its molecule was at first regarded as C(C 6 H 5 ) 3 (Triphenylmethyl) and was 1 It is commonly thought that in the hydrochloride the chlorine is joined to the amiuo-group and not to the carbon of the methane, since, as Tortelli showed (1895), all the chlorine is precipitable by silver nitrate ; the com- pound is hence a salt and not an ether. It cannot, however, be denied that there are compounds, such as triphenyl- methyl chloride, (C 6 H 6 ) 3 C Cl, which behave similarly, being hydrolysable by water and then completely precipitable by silver nitrate. Then, too, methyl iodide is hydrolysed by water alone to the extent of 0-6 per cent, in forty-three hours, whilst in the presence of silver nitrate 96 per cent, of the iodide is hydrolysed in the same time. It is hence more accurate to state that, alter hydrolysis, these ethereal compounds behave like salts. Kosenstiehl maintains that every double decomposition between salts (especially organic) is preceded by hydrolysis, and those salts and ethereal compounds which hydrolyse slowly he calls bradolytes, and those which hydrolyse rapidly, stenolytes. DIBENZYL,ETC. 609 looked upon as the first example of an organic compound containing trivalent carbon. But cryoscopic examination shows that it has the doubled molecular weight, and hence indicates the constitution (C 6 H 5 ) 3 C-C(C 6 H 5 ) 3 . It was prepared by Gomberg by the action of zinc on triphenylchloromethane, and is a solid, stable substance which, in solution, has a yellow colour and becomes unstable owing to its great power of reacting ; with the oxygen of the air it forms a peroxide, (C 6 H 5 ) 3 -OO 'OC^CeH^g. On account of the facility with which it forms additive products, hexaphenylethane is regarded by some as having in solution the constitution : H H (C 6 H 5 ) 2 : C : C I I H H An analogous compound is Pentaphenylethane, (C 6 H 5 ) 3 C CH(C 6 H 5 ) 2 , stable at the ordinary temperature but not in the hot. 4. DIBENZYL AND ITS DERIVATIVES The constitution of these compounds is shown by their methods of synthesis and by the fact that they all yield benzoic acid on oxidation. DIBENZYL (symm. Diphenylethane), C 6 H 5 -CH 2 -CH 2 -C 6 H 5 , is obtained from benzyl chloride and sodium : 2C 6 H 5 -CH 2 -C1 + Na 2 == 2NaCl + C 6 H 5 -CH 2 -CH 2 -C 6 H 5 ; it melts at 52. STILBENE (symm. Diphenylethylene), C 6 H 5 -CH : CH-C 6 H 5 , melts at 125, and is obtained from benzal chloride (benzylidene chloride) and sodium. Owing to its double linking, it can unite with two atoms of Br, which can be eliminated as HBr by treatment with alcoholic potash, the resulting product being TOLANE (Diphenylacetylene), CfiHg-C : C-C 6 H 5 , melting at 60, and behaving like an acetylene derivative. p-DIAMINOSTILBENE, NH 2 -C 6 H 4 -CH : CH-C 6 H 4 -NH 2 , is used, especially in the form of the corresponding sulphonic acids, for the preparation of various substantive dyestuffs. BENZOIN, C 6 H 5 -CH(OH)-CO-C 6 H 5 , is formed by oxidising HYDROBENZOIN, C 6 H 5 -CH(OH)-CH(OH)-C 6 H 5 , which is obtained by treating benzaldehyde with sodium amalgam. Benzoin exists in two stereoisomeric modifications, melting at 138 and 119. It reduces Fehling's solution even in the cold (giving benzil) and forms a phenylosazone, since it contains, like the sugars, the group CO-CH(OH). BENZIL, C 6 H 5 -CO-CO-C 6 H 5 , is a yellow diketone and forms three benzildioximes (see pp. 22, 210, and 572) : C 6 H 5 -C --- OC 6 H 5 C 6 H 5 -C --- OC 6 H 5 C 6 H 5 -C -- OC 6 H 6 II II II II II II N-OH N-OH OH-N N-OH N-OH OH-N Amphi-benzildioxime Anti-benziUioxime Syn-benzildioxime When heated with alcoholic potash, benzil combines with H 2 O, giving benzylic acid : CH-CO-CO-C 6 H 5 + H-OH = (C 6 H DESOXYBENZOIN, C 6 H 5 -CH 2 -CO-C 6 H 5 , is obtained from phenylacetyl chloride, C 6 H 5 -CH 2 -CO-C1, and benzene in presence of aluminium chloride, and also from benzoin and benzil. It melts at 55 and gives dibenzil when reduced with hydriodic acid. ii 39 610 ORGANIC CHEMISTRY HEXABENZYLETHANE, (C 6 H 6 -CH 2 ) 3 C-C(CH 2 -C 6 H 6 )3, was prepared by F. Schmerda (1909) by heating tribenzylcarbinol with hydriodic acid in a sealed tube at 200, the product being shaken with bisulphite, extracted with ether and the latter distilled off. It forms a yellowish crystalline mass which is recrystallised from acetone and glacial acetic acid ; it melts at 80 to 81. From the mother-liquor dibenzyl is obtained. 5. NAPHTHALENE AND ITS DERIVATIVES NAPHTHALENE, C 10 H 8 , occurs in abundance in crude illuminating gas and in coal- tar. When the latter is distilled (see p. 526 et seq.), the naphthalene is obtained from the portions distilling between 170 and 230 and by redistilling the residues of the oils from which the carbolic acid has been extracted with caustic soda, care being taken to surround the condenser coils with hot water to prevent stoppages. The first separation of the naphthalene from the crude oils yielded at various stages of the distillation is effected by cooling in large tanks, crystallised naphthalene separating out. The oily impurities of the crystals are removed in a hydraulic press with heated plates. Attempts have been made to centrifuge the crude naphthalene, but even when this is steamed in the centrifuge, the resi- dual product is always very impure and unsuit- able for distillation or sublimation. In conse- quence of this, use has been made of hydraulic presses with horizontal rods and vertical plates FIG. 419. heated by steam, but these give insufficient pressure and too much waste, and require too much time and attention. The best results are given by presses with vertical columns and ring plates (similar to the presses described on p. 392), which work continuously and readily attain a pressure of 102 kilos per square centimetre with a diminished consumption of steam. Nowadays hydraulic presses with perforated steel bells are used similar to those xised for oily seeds and in 10 hours each of these can effect 30 compressions of 100 kilos ; when several presses are worked, hydraulic accumulators (see p. 393) are used. If well pressed, naphthalene has the mean solidifying point 78-6 and 95-5 per cent, of it distils between 216-5 and 218-5. Attempts have been made to purify naphthalene with a solution of resin soap, but such a method is too expensive (a centrifuged naphthalene containing 7 per cent, of oil gives, with 5 per cent, of colophony and the corresponding quantity of caustic alkali solution, 85 per cent, of pure naphthalene with the solidifying point 78-8). The compressed naphthalene is purified further in metal vessels with conical bases and fitted with stirrers (sometimes with air- jets). In these the molten naphthalene is agitated for 15 minutes with 5 per cent, of sulphuric acid of 50 Be (already used once) to dry the mass somewhat and free it from pyridine compounds ; after removal of this acid, the mass is shaken successively with 5 to 6 per cent, of sulphuric acid of 60 Be. for 30 minutes, 4 per cent, of hot water, 4 per cent, of caustic soda solution of 19 Be. (already used once), and, finally, 2 per cent, of hot water. After settling and removal of the water as far as is possible by decantation, the naphthalene is distilled in large stills holding 100 to 150 quintals and furnished with a rectifying column 2 to 3 metres high. Water distils over first and then pure naphthalene, which is collected in metal boxes, allowed to crystallise in moulds and granulated by means of a crusher ; the solidifying point is then 70-7, while 97-5 per cent, distils between 216-6 and 218. A purer product, in the form of large, shining scales, can be obtained by sublimation (instead of distillation) in an open vessel, a (Fig. 419), having an area of 2 to 3 cu. metres and covered with an inclined wooden plane leading to a large wooden chamber, 20 to 25 cu. metres in capacity. The naphthalene is heated by a pressure steam-coil NAPHTHALENE Gil and sublimes and condenses in the large chamber, forming on the walls a thick layer of shining, white scales of pure naphthalene. In order to avoid loss and to obtain continuous working, the naphthalene is introduced into long cylindrical boilers, bricked in like steam boilers and connected with a large wooden chamber (350 cU. metres), which has a base fitted with conical outlets leading to sacks for catching the naphthalene as it becomes detached from the walls (these are knocked from time to time). In this way 70 kilos of pure naphthalene are obtained per 12 hours for each 100 cu. metres of capacity. Pure naphthalene forms shining scales melting at 79-6 and boiling at 218. It is insoluble in water, but dissolves readily in boiling alcohol or in ether j it volatilises even at the ordinary temperature and distils readily in steam. Naphthalene is used in large quantities in the preparation of various dyestuffs (eosin, indigo, Martius' yellow, tropseolin, Biebrich scarlet, croceine scarlet, &c.), phthalic acid, lampblack, varnishes, and cart-grease, and is employed also as an antiseptic and as a preventative of moth in clothes. For some time it has been mixed with camphor in order to render celluloid less inflammable and less explosive. Crude naphthalene costs 11s. to 12s. per quintal, while the pure white scales are sold at 16s., pure in tapers at 17s. 6d., and chemically pure at 80s. per quintal. In 1910 Italy imported 371 quintals of naphthalene, of the value of 386 (in 1907, 42 quintals), and exported 2214 quintals (2200) ; the production in Italy was 8600 tons (71,880) in 1908. In 1909 Germany imported 77,445 and exported 63,544 quintals. England exported 3650 tons of naphthalene, of the value of 28,110, in 1911. Constitution of Naphthalene. The following structural formula is attributed to naphthalene : rl rl H C C I II H C C \ C H C H a H and to indicate the positions occupied by groups replacing the hydrogens in derivatives the carbon atoms are numbered or lettered with Greek letters, thus : That the two nuclei are united by means of two carbon atoms in the ortho-position is shown by the fact that oxidation of naphthalene in such a way as to destroy one of the nuclei results in the formation of phthalic acid, which is known to contain two carboxyl groups in adjacent positions. Further, since when phenylisocrotonic acid is heated, a naphthalene derivative, namely, a-naphthol, results, it is clear that the second nucleus is formed by the elimination of a molecule of water with closure of the chain of the four carbon atoms of the side-chain of the original acid and two ortho-carbon atoms in the benzene nucleus : CH CH CH CH I HC^/NCH H 2 + . TJp llfvtr ptr Trri J I \_^A > V.CL xVyJTLo XX V-'.j CH CO -OH \/c\/ H CH^C-OH 612 ORGANIC CHEMISTRY That there are two condensed benzene nuclei is also deduced from the fact that oxidation of a-nitronaphthalene gives nitrophthalic acid, the benzene nucleus containing the nitro- group being preserved and the other destroyed. If, however, the nitro-group is first reduced to an amino-group, oxidation results in the destruction of the nucleus containing the amino-group and in the preservation of the other, phthalic acid, which undoubtedly contains a benzene nucleus, being formed. That the linkings between carbon and carbon are different in the two nuclei is shown by the addition of four hydrogen atoms to one of the nuclei, which probably has true double linkings, while the other nucleus would seem to have a true benzenic character with centric linkings (Bamberger) ; further, the addition of ozone proves with certainty the presence of olefinic double linkings (E. Molinari, 1907) : CH CH C HC HC \ CH CH C CH CH With nitric acid, naphthalene gives a nitro-derivative, and with sulphuric acid, various sulphonic acids. The hydroxy-derivatives resemble phenols and the amino-derivatives are capable of diazotisation. Hydronaphthalene is readily formed by the addition of nascent hydrogen and behaves likes hydrophenol, i.e. like an unsaturated hydrocarbon of the aliphatic series. The isomerides of the substitution products of naphthalene are more numerous than in the case of benzene. Thus, there are two isoineric mono- substituted derivatives, the a-compound with the substituent in the 1-, 4-, 5-, or 8-position, and the /3-compound with the substituent in the 2-, 3-, 6-, or 7-position. The isomeric disubstituted compounds with two similar substituents are ten in number, while with two different substituting groups fourteen isomerides are possible, and, in some cases, all known. Compounds with substituents in the 1- and 8- or the 4- and 5-positions are known as aa- or pen-compounds, e.g. Perinaphthalenedicarboxylic Acid, >COOH , which readily forms an anhydride owing to the proximity >COOH of the hydroxyls. The number of isomerides being so large, it is sometimes difficult to deter- mine the constitution of any derivative. To this end the oxidation products are often studied, the formation of phthalic acid indicating that all the substituents are in the one benzene nucleus destroyed by the oxidation, while the formation of a substituted phthalic acid indicates the opposite to be the case. NITRON APHTH ALENES, Ci H 7 NO 2 . Of the two isomerides, the /3 is of no industrial importance. a -NITRON APHTH ALENE is obtained on the large scale by nitrating naphthalene (10 parts) with a mixture of 8 parts of nitric acid (sp. gr. 1-49) and 10 parts of sulphuric acid (sp. gr. 1-84), the temperature being kept at 70 for six hours. The supernatant acid is decanted off in the hot and the fused nitronaphthalene washed several times with hot water and then granulated by pouring slowly into cold water with vigorous agitation. It forms a yellow crystalline mass which melts at 59, boils at 304, and in the molten condition has the sp. gr. 1-223 ; it is insoluble in water %ut dissolves readily in benzene, ether, carbon disulphide, or hot alcohol. The crude commercial product costs 64s. to 76s. per quintal and the pure crystals 96s. It is used in the manufacture of dyestuffs, especially for making n-naphthylamine and thence various azo-derivatives or naphthol derivatives. It is also employed in the NAPHTHALENE DERIVATIVES 613 treatment of oils. When it is stored alone, there is no greater danger than in the storage of oil. When used in mixtures not excessively hot, it presents no special danger. With reducing agents, ti-nitronaphthalene gives u-NAPHTHYLAMINE, Ci H 7 -NH 2 , and this, by way of the diazo-compound, yields u-naphthol, identical with that obtained from phenylisocrotonic acid. ct-Naphthylamine can also be obtained from ci-naphthol by means of ammonia and calcium chloride, or ammonia and ammonium sulphite. Indus- trially it is obtained by the various processes mentioned above for aniline : usually 60 parts of dry u-nitronaphthalene are added gradually to a hot mixture of 80 parts of iron turnings, 4 parts of hydrochloric acid, and a little water, the whole being mixed and kept at 70 for 6 to 8 hours ; slaked lime (about 5 parts) is next added until the reaction becomes alkaline, the naphthylamine being distilled from retorts and condensed at 60, and subsequently purified by rectification. It consists of pleasant -smelling, white crystals, melting at 50, readily subliming and boiling at 300. With oxidising agents it gives red or blue colorations. The isomerif; /3-NAPHTHYLAMINE is obtained by heating 10 parts of /3-naphthol with 4 parts of caustic soda and 4 parts of ammonium chloride in an autoclave at 160 for 60 to 70 hours ; the unchanged naphthol is removed by means of sodium hydroxide solution and the /3-naphthylamine extracted from the residue by hydrochloric acid. The preparation with ammonium sulphite (see above) gives a better yield owing to the formation of the sulphuric ether of /3-naphthol, which reacts more readily with ammonia. It forms shining, odourless scales which melt at 112, boil at 294, and are not coloured by oxidising agents. The separation of a- from /3-naphthylamine is effected by solvents, such as xylene, chlorobenzene, &c., which dissolve both isomerides in the hot and deposit almost all of the a -compound in the cold. The commercial a-derivative costs about 8 per quintal and the /3-compound three times as much. Both are used for making azo-dyestuffs. The NAPHTHALENESULPHONIC ACIDS are obtained from naphthalene and con- centrated sulphuric acid. They form deliquescent crystals and when fused with KOH give the naphthols ; the a-sulphonic acid in presence of sulphuric acid at 160 is converted into the /3-acid. a- and /3-NAPHTHOLS, Ci H 7 -OH, are found in coal-tar, and may be prepared from the sulphonic acids or amines (see above). They form shining scales with a phenolic odour, and dissolve slightly in hot water and more readily in alcohol or ether. a-Naphthol melts at 95 and boils at 282 ; /3-naphthol melts at 122 and boils at 288. Their hydroxyl groups are more readily substituted than those of the phenols. With ferric chloride, aqueous a-naphthol gives a violet precipitate, while /3-naphthol gives a green coloration OTT and precipitates Dinaphthol, C^oHe-cCV, QTT The two naphthols give ethers, e.g. Neroline, C^H^-O^IIg, which has a fruity odour, Betol or Naphthosalol (the salicylic ester of /3-naphthol), C 10 H 7 -0-CO-C 6 H 4 -OH, melts at 95, and is used in medicine under the name of salol. NAPHTHIONIC ACID (l-Naphthylamine-4-sulphonic Acid), C 10 H 6 (NH 2 )(S0 3 H), or SO a H , is formed by sulphonating a-naphthylamine, and is used for preparing NH 2 Congo red and other dyes. The solutions of its salts have an intense reddish blue fluorescence. Of the a- and /3-naphthylaminesulphonic acids, 13 isomerides are known. Eikonogen, used as a photographic developer, is the sodium salt of o 1 -Amino-/3 - naphthol-/3 3 -sulphonic Acid. ((-NAPHTHAQUINONE, , is obtained in yellow crystals melting at 125 by oxidising naphthalene with chromic acid in boiling acetic acid solution. From its constitution those of other substitution products of naphthalene can be 614 ORGANIC CHEMISTRY deduced, since when the substituent groups are in the para-position, oxidation always leads ultimately to a-naphthaquinone. It is volatile in steam. /3-NAPHTHAQUINONE, C 10 H 6 O 2 or , is formed by the oxidation of 1 : 2-aminonaphthol, and crystallises in reddish yellow leaflets blackening at 115 to 120. The following compounds are also known : Oxy- and Dioxynaphthaquinones (naphtlta- zarinblack) ; a- and /3-Methylnaphthalenes, C 10 H 7 -CH 3 ; Naphthoic Acids, C 10 H 7 -CO H ; HydroxynaphthoicAcids,C 10 H 6 (OH)(C0 2 H);NaphthalicAcid,C 10 H f) (CO 2 H)o;Dinaphtnyl, C 10 H 7 -C 10 H 7 ; Acenaphthene, C 10 H 6 , in which the unions with the ethylene group are in the a x - and appositions (found in tar, colourless, melting at 85, boiling at 277, and giving naphthalic acid on oxidation). ADDITION PRODUCTS OF NAPHTHALENE Naphthalene gives additive products more readily than benzene does, those containing four atoms of chlorine or hydrogen being well known. It has been shown that this addition occurs in only one of the nuclei, and similar behaviour is shown on oxidation. Chlorine reacts with naphthalene at the ordinary temperature and forms Naphthalene Tetrachloride, H HC1 H H \ HC1 HC1 which forms colouiless crystals melting at 181 and gives phthalic H HC1 acid on oxidation, and Dichloronaphthalene, C 10 H 6 C1 2 , when treated with alcoholic potash. When ^(3-naphthylamine is reduced (Na + amyl alcohol), four hydrogen atoms are added to the nucleus containing the amino-group, giving Tetrahydronaphthylamine, H H 2 , which behaves exactly like an aliphatic amine and does not form diazo-compounds ; it is oxidised by permanganate, giving o-Carboxyhydrocinnamic Acid, /C-H 2 * C.H 2 * COjjH C 6 H 4 <^ . o-Naphthylamine also gives a tetrahydro-derivative, which \C0 2 H behaves, however, as an aromatic amine and can be diazotised ; on oxidation it gives Adipic CH 2 CH 2 COOH Acid, I . which shows that the four hydrogen atoms are added to the benzene CH 2 COOH CH 2 H 2 nucleus which does not contain the amino-group : 11 H. H H 2 NH 2 INDENE, C 9 H 8 , may be regarded as formed by the condensation of a benzene group ANTHRACENE 615 CH /\ HC C CH with a pentamethylene group : ) || || . It is a yellow oil boiling at 1 80, and is HC C CH CH CH 2 found in coal-tar and in crude pseudocumene ; it has an odour of naphthalene and gives phthalic acid on oxidation and Indrene, C 9 H 10> on reduction. 6. ANTHRACENE GROUP a. CH ^ ANTHRACENE, C 14 H 10 , or K . , is found in coal-tar to the extent of 0-25 to 0-45 per cent. The crude anthracene oil which passes over at a high temperature (above 270) in the distillation of tar is subjected to a further rectification which yields a 50 per cent, anthracene. This is purified by distillation from iron retorts with potassium carbonate, C^H^x which holds back the large amount of Carbazole, | /NH, as the non- volatile C 6 H 4 C 6 H 4\ potassium compound, | ^>NK. The distillate then contains only anthracene C 6 H 4 and phenanthrene, the latter being removed by dissolving it in carbon disul- phide or a mixture of this solvent with concentrated sulphuric acid (Ger. Pat. 164,508 and Fr. Pat. 349,337). The residual anthracene is purified by crystalli- sation from crude benzene (see Treatment of Tar described on p. 526 et seq.). and by sublimation with superheated steam. The proposal has also been made to purify crude anthracene (containing, say, 46 per cent, of anthracene and 13 per cent, of carbazole) with hot naphtha and sulphuric acid, which convert all the basic substances into salts and dissolve them, the anthracene being afterwards separated by decantation. Evaporation of the naphtha gives anthracene of about 84 per cent, strength and this gives a product of 95 per cent, purity on crystallisation from benzene. It forms shining, colourless scales with a blue fluorescence, and melts at 216-5 and boils at 351 ; it dissolves slightly in ether or alcohol, but is readily soluble in hot benzene. Sunlight gradually converts it into the polymeric para- Anthracene (Cj 4 H 10 ) 2 . With picric acid it forms a molecular condensation product, C 14 H 10 ,C 6 H 2 (N0 2 ) 3 OH, melting at 138. By reducing agents, anthra- CH cene is transformed into Hydroanthracene, C 6 H 4 << riTT 2 >.C 6 H 4 , which melts i ^^2 at 107 and is readily soluble in alcohol. It is used for the manufacture of anthraquinone and alizarin. Crude anthracene oil (green grease) is sold at 11s. to 12s. Qd. per quintal, crude 20 per cent, anthracene at Is. 6d. per kilo and the purified product at 6s. to 8s. per kilo. Its constitution is deduced from its various syntheses. Anschiitz obtained it from tetrabromoethane and benzene in presence of A1C1 3 : CHBr 2 / CH \ 2C 6 H 6 + | = 4HBr + C 6 H 4 ^ | ))C 6 H 4 . CHBr 2 ^CH/ 616 ORGANIC CHEMISTRY It is formed also when o-tolyl phenyl ketone ia heated with zinc dust : H 2 + C 6 H 4 < S CH' >C 6 H 4 ; this synthesis establishes the ortho-position of the connections between the two nuclei and also the presence of the CH-CH group. Confirmatory evidence is obtained from the following synthesis : 4Na = 4NaBr Br H CH 2 Br o-Bromobenzyl bromide which, on oxidation, loses 2H and gives anthracene. Phthalic anhydride, when heated with benzene and A1C1 3 , gives o-benzoylbenzoic acid, from which PC1 6 eliminates water with formation of anthraquinone, the latter giving anthracene when reduced with zinc dust in the hot : ,CO-C fi H fi S\) anthracene. Centric linkings do not seem to be present in the nuclei of anthracene, which readily combine with ozone (E. Molinari, 1907), this property being characteristic of olefine double linkings (see p. 88). SUBSTITUTION PRODUCTS OF ANTHRACENE The possible isomerides are here very numerous, but only few of them have yet been prepared. Three monosubstituted isomerides are possible, as is seen from the constitutional formula (see above). The constitution of the isomerides is ascertained from a study of the oxidation products and of the methods of synthesis. When the substituents are in the y x or y 2 position, oxidation gives anthraquinone. 1 CO ANTHRAQUINONE, C 14 H 8 2 , or , is obtained very easily by oxidising anthracene with dichromate and dilute sulphuric acid in 1 Of the many Derivatives of Anthracene, the following may be mentioned : anthraccnecarboxylic aciiis (, ft, and y) ; chlorobromoanthracenes, which contain the halogens in the y-positions, as they form anthraquinone on oxidation ; niiro- and dinitro-anthracenes (y) ; p-anthramine, C 14 H 8 -NH 2 , obtained from /3-anthrol and NH, ; anthrols (a and CH CH _CH 2 ; anthrone, C,H 4 C,H 4 ; the anthrahydroquinones, C CH(OH) chrysazol, rufol, and flavol) ; anthracenegtdphonic and disulphonic acids ; unthraquinonesulpJionic acids ; hydroxy- anthraquinones, C 14 H,O 2 -OH; quinizarin (a, ; a a -dihydroxyanthraquinone) ; purpuroxanthin (a l : ^,-dihydroxy- anthraquinone) ; C 6 H4(CO 2 )-C 6 H(OH) S ( : ft : a,) is purpurin (the isomeric flavopurpurin, anthrapurpurin, anthragallol, &c., are also known) ; OH C 6 H 3 CO 2 C 6 H 3 OH (anthraflavinic and isoanthraflavinic acidt, with which correspond anthrarufin, chrysazin, &c.) ; tetrahydroxyanthraquinones (rufiopin, anthrachrysone, quinalizarin) ; hexahydroxyanthraquinones (rufigallic acid, &c.) ; methyl- and dimethyl-anthracenes, C 14 H,-CH S and C 14 H 8 (CH 3 )., ; CH 2 s phenylanthracene, C 14 H,-C,H, ; alkylanthrahydrides, C,H 4 \ ^>C,H 4 ; phenylanthranol (phthalidinf), CHR C(C.H ) C(C.H.)(OH) f -H 4 <^ | ^>C,H 4 ; phenylhydroxyanthranol (phthalideine), C,H 4 <^ ^>C 8 H 4 ; anthraceneearb- C(OH) ~CO~ _CH,_ oxylic acids (a, /3, y), C 14 H,-CO a H ; alkylhydroanthmnoU, C,H 4 <^ ">C,H,, &c. CR(OH) ALIZARIN 617 the hot, or, better, with nitric acid, which does not give nitro-derivatives. It can also be obtained from phthalic anhydride and benzene in presence of A1C1 3 (see above) or by electrolysing anthracene in 20 per cent, sulphuric acid in presence of cerium, chromium, or manganese salts (Ger. Pat. 152,063, and Perkin, 1904). It can be purified by crystallisation from nitrobenzene or aniline, which dissolve it in the hot but not in the cold. It gives two isomeric monosubstituted derivatives. It forms yellowish needles melting at 274 and boiling at about 360, and it dissolves in concentrated sulphuric acid, but is precipitated unchanged on dilution. It is very stable, is not easily oxidised and has the character of a diketone rather than of a quinone. It is not readily reduced, is only slightly volatile and has no pungent odour. That the two lateral benzene nuclei have centric linkings and not olefinic double bonds is shown by the fact that, unlike anthracene (see above), anthraquinone does not fix ozone. When fused with potash, it gives benzoic acid and, when heated with zinc dust and NaOH, Hydroxyanthranol, 6 H 4 < ^Q^>C 6 H 4 , which has a blood-red colour in alkaline solution and is oxidised to anthraquinone in the air. Reduction of anthraquinone with Sn and HC1 gives Anthranol, C(OH) C 6 H 4 ^ [ ^>C 6 H 4 , which is a weak phenol. X C(OH) X More energetic reduction, such as distillation over zinc dust, yields anthra- cene. The Bohn-Schmidt reaction permits of the introduction of sulphonic or nitro-groups into the non-substituted or the substituted nucleus of anthra- quinone derivatives, according as the reaction occurs in presence or in absence of boric acid, a- or /3-Nitroderivatives can also be obtained, at will, by means of the same reaction (Ger. Pat. 163,042 of 1905), which is facilitated by the presence of mercury salts. Commercial anthraquinone costs about 6s. per kilo, and the sublimed chemically pure product 28s. The most important derivative of anthraquinone is the 1 : 2-dihydroxy- compound or alizarin. CO OH ALIZARIN (Dihydroxyanthraquinone), C 14 H 8 O 4 , or CO was at one time obtained exclusively from madder roots (Rubia tinctorum), from which Ruberythric Acid (a glucoside of the formula C 26 H 28 O 14 ) is extracted ; this is separated into glucose and alizarin by boiling with dilute sulphuric acid. It is a very beautiful red colouring -matter and was known to the ancients. Since 1870, 1 following Grabe and Liebermann's synthesis (1869), it has been prepared only artificially in the following manner : anthracene is converted by oxidation with H 2 S0 4 and Na 2 Cr 2 7 into crude anthracene. This is then heated at 100 with concentrated sulphuric acid, which leaves the anthra- quinone unaltered, while it converts the impurities into sulphonic acids soluble in water. The anthraquinone is then filtered and washed and heated at 160 with fuming sulphuric acid (containing 50 per cent, of free S0 3 ), which converts it largely into the monosulphonic acid. The latter is dissolved in water and filtered to separate it from unaltered anthra- quinone ; neutralisation of the solution with caustic soda results in the deposition of the sodium salt, which is only slightly soluble in cold water. One hundred parts of this salt are mixed with 25 parts of caustic soda and 12 to 14 parts of potassium chlorate, which facilitates 1 In 1868 Prance produced and exported madder to the value of 1,720,000 and 1,240,000 respectively. The exportation fell to 800,000 in 1871 and to 160,000 in 1876, the production then ceasing entirely. 618 ORGANIC CHEMISTRY the reaction ; the mixture is dissolved in the smallest possible amount of water and the liquid heated at 180 for 2 days in an autoclave fitted with a stirrer. The sulphonic group is thus replaced by hydroxyl (or ONa), and at the same time a second OH group is formed by the action of the chlorate : co pn C 6 H 4 C 6 H 3 -S0 3 Na + 3NaOH + O = Na 2 S0 3 + 2H 2 + C 6 H 4 <^>C 6 H 2 (ONa) 2 . Sodium anthraquinone- monosulphonate The fused mass is run into water and acidified with sulphuric acid, the colouring- matter (alizarin) being thus liberated. According to Fr. Pat. 333,144, if fuming sulphuric acid acts on anthraquinone in presence of mercury, there is no partial formation of the m-sulphonic compound, the sulpho-group entering exclusively the ortho-position to the ketonic group. Alizarin may also be prepared (Ger. Pat. 186,526) without sulphonation by treating, say, 300 kilos of a mixture of NaOH and KOH with 30 kilos of NaC10 3 (or Na 2 2 , BaO 2 , Pb0 2 , &c.) dissolved in 100 litres of water, 100 kilos of anthraquinone being then added and the liquid heated at 200 in an oil-bath until the oxidising agent disappears. After this, the mass is poured into water through which air is then passed ; the alizarin is precipitated with milk of lime, the precipitate being filtered off and decomposed with HC1 and the alizarin purified from anthraquinone residues by means of caustic soda. This method yields a purer product than other processes. Alizarin has been prepared recently by passing an electric current through a mixture of anthraquinone and fused potash. Alizarin sublimes in fine, orange-red needles, melts at 289, and is almost insoluble in water and slightly soluble in alcohol ; owing to its phenolic groups it dissolves in alkali and also forms a diacetyl-derivative. When distilled with zinc dust it forms anthracene. With metallic oxides it forms insoluble lakes of various colours, and on this is based its use in dyeing. With ferric oxide it gives a bluish black colour and with lime a blue lake ; the lakes of tin and aluminium are red (Turkey red). The constitution of alizarin is shown also by its synthesis from phthalic anhydride and catechol at 150 in presence of sulphuric acid : CeH^QQ^O + C 6 H 4 S (Thiophene). H 2 OC0 2 H HC=CH X /3 l a 1 Thiophene is a colourless and almost odourless, refractive liquid, boiling at 84, and having the sp. gr. 1-062 at 23. The presence of the double linkings is confirmed by the quantitative addition of ozone. Pure thiophene, prepared synthetically, costs 18 per kilo. CH : C(CH 3 K Dimethylthiophene (thioxene), \ /S, is obtained by the interaction of the CH : C(CH 3 )/ enolic form of acetonylacetone and phosphorus pentasulphide, and 1 : 4-diketones in general yield higher homologues of thiophene, which, when oxidised, give carboxyl groups in place of the side-chains. Thiophene compounds, such as halogen and nitro-derivatives, sulphonic acids, &c., behave very similarly to those of benzene. With isatin and concentrated sulphuric acid, thiophene gives a blue coloration (indo- phenin, C^H^NCS). 3. PYRROLE, C4H 5 N, is found in small quantity in tar and in larger quantity in Dippel animal oil (bone oil), especially in the fraction distilling at about 130, which is freed from pyridine bases by saponifying with soda and washing with dilate sulphuric 621 acid. It is purified by converting into the potassium derivative, C 4 H 4 NK (by the action of potassium), which is washed with ether, in which it is insoluble, and then treated with water, the pyrrole being thus liberated. After fractional distillation, it is obtained as a light, colourless oil, boiling at 131, and possessing a faint odour of chloroform. It readily turns brown and polymerises under the action of light. With isatin and sulphuric acid it gives the blue indophenin reaction (see above). A reaction characteristic of the pyrroles is the red coloration they give with a pine shaving moistened with HC1. The hydrogen of the iminic group is replaceable by metals, acetyl, and alkyl groups. Pyrrole now forms the basis of a number of important compounds, which are obtained by various syntheses investigated by Ciamician and his collaborators during the past quarter of a century. The constitutional formula of pyrrole is as follows : /3'HC CH/3 Us 2|| - u'HC CHrt i N H this being deduced from a number of reactions and syntheses, e.g. the formation of pyrrole by the action of ammonia on y-diketones or on succinic aldehyde, with intermediate formation of diammonaldehyde : CH 2 -CHO CH:CHv + NH 3 = 2H 2 + | )sNH. CH 2 -CHO CH-.CH./ This pyridine nucleus occurs frequently in nature, in combination with other groups in alkaloids (nicotine, &c.), in the colouring-matter of the blood and of chlorophyll, &c. CH 2 -CH:N-OH When boiled with hydroxylamine, pyrrole gives Succindialdoxime, | , CH 2 -CH:N-OH CH 2 -CHO which, with nitrous acid, gives succinic aldehyde, I CH 2 CHO CH 2 -COv Pyrrole is formed by the distillation of succinimide, | /NH, with sodium CH 2 -CCK or zinc dust, while the oxidation of pyrrole with chromic acid gives maleimide, CH-C(X II >H. CH-CCK Pyrrole is changed by acids ; with HC1 in the hot, it polymerises and condenses to a red mass (pyrrole red). It has a faint acid character, but gives a hydrochloride, (C 4 H 5 N) 3 , HC1, only in ethereal solution. With the halogens it gives not additive products but only, like benzene, substituted derivatives. Tetraiodopyrrole (iodol) is obtained from pyrrole by the action of an alcoholic, alkaline solution of iodine ; it is an efficient antiseptic and is used instead of iodoform, being without the unpleasant odour of the latter. It melts at 190, and is colourless when freshly prepared, but it gradually turns brown and deposits iodine. With nitric and sulphuric acids, pyrrole resinifies ; the nitro-derivatives, which contain the isonitro-group, NOOH, are prepared indirectly (e.g. with alkyl nitrate). Pyrrole is analogous in many of its properties with the substituted phenols and anilines ; thus, a methyl- or acetyl-group united to the nitrogen (N-derivatives) is displaced, on heating, to a carbon atom (C-derivatives) : 622 ORGANIC CHEMISTRY HC CH HC CH HC CH HC CH II II II II II II II II HC CH HC C-CH 3 ; HC CH HC C-CO-CH 3 . \/ \X \/ \/ N NH N NH I I CH 3 CO-CH 3 Potassium pyrrolate, C 4 H 4 NK, and CO 2 give pyrrolecar boxy lie acid, C 4 H 3 (C0 2 H)-NH (m.pt. 102) ; this loses C0 2 and gives pyrrole again when heated, while it loses water and N CO forms a dimolecular anhydride, Pyrocoll, , when treated with CO N acetic anhydride. Like the substituted phenols, the C-alkylpyrroles give pyrrolecarboxylic acids by simple fusion with potash. In analogy with the formation of nitrosophenols from phenols, pyrrole, with ethyl nitrite in presence of sodium alkoxide, forms Nitrosopyrrole, which exists in tautomeric modifications : HC C : NOH HC CH II II II HC CH and HC C : NOH. \/ V N N By means of chloroform and sodium alkoxide, another atom of carbon is introduced into the nucleus, a pyridine derivative being formed. Hydrogenated derivatives of pyrrole are formed more easily than those of benzene, and, like the latter, do not show purely aromatic properties. When pyrrole is reduced by means of zinc and hot acetic or cold hydrochloric acid, it yields Dihydropyrrole (or pyrroline, m.pt. 91), which, with HI and P, gives Tetrahydropyrrole (or pyrrolidine, b.pt. 87), H 2 H 2 H H 2 ; the latter, together with N-methylpyrroline, are the simplest cyclic alkaloids NH known and are found in tobacco. Pyrrolidine is found in carrot seeds and a C-methyl- pyrroline in pepper. When proteins are decomposed by means of trypsin or hydrochloric acid, the amino- acids formed are accompanied by laevo-rotatory a-Pyrrolidinecarboxylic Acid. Among the products formed by the degradation of egg albumin by baryta is a'-pyrrolidone-o -carboxylic Acid, OC CH-CO 2 H, which is also known as pyroglutamic acid ; it melts at 183, has NH a neutral reaction and, when heated, loses C0 2 and H 2 0, forming pyrrole. PYRAZOLE, C 3 H 4 N 2 , is a heterocyclic compound with two nitrogen atoms in the ortho-positions. It can, indeed, be obtained by the condensation of 1 mol. of diazomethane with 1 mol. of acetylene : CH N CH = N |||. + CH/H | ")NH (Pyrazole). CH X N CH=CH X s It is very stable, melts at 70, is a feeble base, and has a neutral reaction in water. The a'/3'-dihydro-compound is known as Pyrazoline, C 3 H 6 N 2 , and the a'-keto-derivative CH : Nv pf this, | /NH, as Pyrazolone. Condensation of methylphenylhydrazine, PYRIDINE 623 CH 3 NH NH C 6 H 5 , with ethyl acetoacetate yields Dimethylphenylpyrazolone, CH 3 -C-N(CH 3 K /N'C 6 H 6 , which bears the name antipyrine and is used medicinally H-C OK owing to its marked antipyretic action on the animal organism ; it melts at 113, dissolves in water and in alcohol, and gives a greenish blue coloration with nitrous acid and a red coloration with ferric chloride. NC=Hx THIAZOLE, C 3 H 3 NS, or | /S, may be regarded as thiophens with one CH CH=CH X group replaced by N. It shows analogies with the pyridine bases. Just as benzene may be obtained from aniline, thiazole may be obtained from aminothiazole (see below). AMINOTHI AZOLE, C 3 H 2 NS NH 2 , is obtained by the action of monochloracetaldehy de on pseudo-thiourea : CH 2 -C1 HN V CH + >C NH 2 = HC1 + H 2 + || CHO HS X CH and is a base analogous to aniline. N=CH\ IMIDAZOLE or Glyoxaline, C 3 H 4 N 2 or | NH, melting at 92, is a strong base with a fishy odour, and is isomeric with pyrazole (see above) ; it is obtained by the action of ammonia on glyoxal in presence of a little formaldehyde. Alloxan (see p. 366) may be regarded as a derivative of imidazole. LYSIDINE, Methyldihydroimidazole or Ethenylethylenediamine, C 3 H 3 (CH 3 )N 2 H 2 , is administered as a solvent for uric acid. OXAZOLE, C 3 H 3 NO, or | /O, is also termed Furazole, owing to its analogy HC=CH/ with furfuran (see above). Its phenyl derivatives are known, as also are those of Isooxazole, HC=N, OSOTRIAZOLE, | /NH, is faintly acid and also faintly basic in character. HC=W It melts at 22, boils at 204, and is soluble in water. TRIAZOLE (or Pyrrodiazole), | /NH, melts at 121, and is extremely soluble HC = W in water. HC=Nx TETRAZOLE, /NH, is a weak acid which forms explosive salts ; it melts N=N/ at 155 and is soluble in water. HC=Nx AZOXAZOLE, | O, is also termed Furazan. 4. PYRIDINE AND ITS DERIVATIVES Pyridine is a heterocyclic nucleus containing five carbon atoms and one nitrogen. It resembles benzene in its behaviour, but it is more stable or more indifferent towards sulphuric, nitric, and chromic acids, permanganate, &c. Oxidation of the homologues with side-chains gives pyridinecarboxylic acids, and the latter, when distilled with lime, give pyridine. Its hydro-derivatives are readily formed in a similar manner to hydro- benzenes, 624 ORGANIC CHEMISTRY Halogen derivatives are obtained more easily by the action of PC1 5 or SbCl 5 at a high temperature than by the action of the halogens themselves. Oxidising agents attack only the side-chains and not the pyridine nucleus. With sulphuric acid, a pyridinesulphonic acid is obtained, and this gives a hydroxyl-derivative of pyridine on fusion with potash, or a nitrile when treated with KCN. There is hence a marked analogy to benzene, although direct nitration of pyridine is not possible unless phenolic or aminic groups are present. Pyridine and its derivatives are decidedly basic in character (tertiary bases) and form soluble salts with hydrochloric or sulphuric acid and insoluble ones with chromic acid ; the double salts with platinum and gold chlorides are slightly soluble. Like tertiary bases, they combine with methyl iodide to form quaternary bases. From the complex alkaloidal groupings, pyridine compounds are often obtained either by distillation with caustic potash or merely by energetic oxidation . Coal-tar and Dippel animal oil contain various pyridine compounds which are separated by conversion into salts. General methods of formation, (a) The oxidation of quinoline (see later) yields first quinolinic acid (pyridinedicarboxylic acid), C 5 H 3 N(CO 2 H) 2 , which then loses C0 2 , giving pyridine. /3-Methylpyridine is obtained by distilling acraldehyde- ammonia ; this explains the presence of pyridine products in Dippel oil, acrolei'n and ammonia being formed in the dry distillation of non- defatted bones. An important synthesis is the general one of Hantzsch by which Ethyl Dihydrocollidinedicarboxylate, for example, is obtained by heating aldehyde- ammonia with ethyl acetoacetate ; other pyridine compounds are obtained from different aldehyde-ammonias and /3--ketonic acids : 2CH 3 -CO-CH 2 -C0 2 C 2 H 5 + CH 3 -CHO + NH 3 = C 5 N(CH 3 ) 3 H 2 (C0 2 C 2 H 5 ) 2 + 3H 2 0. From the ester thus formed the hydrogen of the NH and CH is eliminated by means of nitrous acid, and the resulting collidinedicarboxylic acid, when treated with potash and distilled with lime, loses the two carboxyls and gives collidine (trimethylpyridine) ; oxidation of the latter gives pyridinecarboxylic acid and elimination of carboxyl from this in the ordinary way forms pyridine. When ethylidene chloride is heated with alcoholic ammonia, it yields Aldehydine, C 8 H U -N. The constitution of pyridine corresponds with that of benzene, in which one methinic group, CH, has been replaced by a nitrogen atom. Korner in 1869 proposed the following constitutional formula, which still agrees well with all the general properties of the pyridine compounds : CH HC HC CH i N N When pyridine is reduced with alcohol and sodium, it fixes six atoms of hydrogen, giving Piperidine or hexahydropyridine, the constitution of which is shown by its synthesis when pentamethylenediamine hydrochloride is rapidly heated : /CH 2 -CH 2 -NH 2 /CH 2 -CH. 2 \ CH/ NH 3 + CH 2 < ')NH. When piperidine is heated with sulphuric acid it gives pyridine, and the latter, when strongly heated with hydriodic acid, gives normal pentane. The constitution of pyridine is confirmed by the fact that the isomeric substitution products correspond exactly in number with those derivable theoretically from Korner's formula. There are, indeed, three monosubstituted isomerides (a, ft, and y), and six disubstituted isomerides : aa', aft, aft', /3y, and ftft'. The position of a substituent group is determined by converting it into a carboxyl group with formation of the corresponding acid of known constitution (see later). Thus, picolinic acid has the carboxyl in the a-position, nicotinic acid in the ft-, and isonicotinic acid in the y-position. PYRIDINE, C 5 H 5 N, is a coiourless liquid, boiling at 115 and having the sp. gr. 1-0033 at 0. It dissolves in water in all proportions and has a slight alkaline reaction (not sensitive to phenolphthalein, slightly to litmus, and more so to methyl orange). It has an unpleasant odour and is hence used to denature alcohol (see p. 152). It forms a slightly soluble f errocyanide, by means of which it can be purified. It forms pyridineammonium iodides, e.g. C 5 H 6 N,CH 3 I, which with KOH in the hot gives Dihydromethylpyridine, C 5 H 4 H 2 -NCH 3 , with a characteristic pungent odour. Metallic sodium polymerises pyridine, forming Dipyridine, C 10 H 10 N 2 (b.pt. 290) and y-Dipyridyl, C 10 H 8 N 2 or NC 5 H 4 -C 6 H 4 N (m.pt. 114). With sulphuric acid it gives /3-Pyridinesulphonic Acid, NC 5 H 4 'S0 3 H. Pyridine is administered in cases of asthma and has been suggested as a means of purifying synthetic indigo. Mixed pyridine bases for denaturing cost about Is. 2d. per kilo and pure pyridine 8s. Of the homologues of pyridine, the following may be mentioned : PICOLINE (Methylpyridine), NC 5 H 4 'CH 3 , exists as three isomeric liquids similar to pyridine and of disagreeable odour ; their boiling-points are : a, 129 ; ft, 142 ; y, 144. Besides by general synthetical methods (see above), /3-picoline is formed by heating strych- nine with lime, a -Methylpyridine condenses with aldehydes by means of the methyl group, giving alkines : NC 6 H 4 -CH 3 + CH 3 -CHO = NC 5 H 4 -CH 2 -CH(OH)-CH 3 . This a-picolylalkine gives up a molecule of water yielding a pyridine derivative with an un- saturated side-chain, e.g., a-allylpyridine, NC 6 H 4 -CH : CH-CH 3 . These reactions proceed in one stage if zinc chloride is present with the aldehyde. LUTIDINES (Dimethylpyridines), NC 5 H 3 (CH 3 ) 2 ; three isomerides are known, with the boiling-points : aa', 143 ; ftft', 170 ; ay, 157. COLLIDINES (Trimethylpyridines), NC 6 H 2 (CH 3 ) 3 , are isomeric with propylpyridine. a-Allylpyridine (see above) fixes hydrogen (alcohol and sodium), giving the alkaloid CONIINE (inactive racemic), which is a-Propylpiperidine ; fractional crystallisation of the tartrate separates the laevo- from the dextro-form, the latter being identical with natural coniine (the poison of hemlock), boiling at 1 67. The asymmetric carbon atom causing the activity is the a- one united with the propyl group. PYRIDONES or HYDROXYPYRIDINES, NC 6 H 4 -OH. The three isomerides are known, their boiling-points being : a, 107 ; ft, 124, and y, 148. They are obtained by heating the corresponding hydroxypyridinecarboxylic acids with lime. They are phenolic in character and give red or yellow colorations with ferric chloride. a-Hydroxy- pyridine forms two series of derivatives corresponding with the two tautomeric formulae : /C(OHK / C \ C 2 H 2 \ /C 2 H<{ and C 2 rI 2 \ x-C 2 H 2 , the former giving, for instance, a Methoxypyridine and the latter a Methylpyridone. PYRIDINEMONOCARBOXYLIC ACIDS, NC S H 4 -C0 2 H. The three isomerides are as follow : a or Picolinic Acid, m.pt. 135 ; ft or Nicotinic or Nicotic Acid, m.pt. 231 ; y or isonicotinic acid, m.pt. 309, They are formed by oxidation of pyridine derivatives with a side-chain or by elimination n 40 626 ORGANIC CHEMISTRY of one carboxyl from the pyridinedicarboxylic acids, that nearer to the nitrogen being the more easily eliminated. Nicotinic acid is obtained on oxidation of nicotine. When boiled with sodium amalgam in a highly alkaline solution, these acids lose nitrogen as NH 3 and give saturated, open-chain, dibasic hydroxy-acids. When the carboxyl is in the a-position (with the dicarboxylic acids also), an orange coloration is given with FeS0 4 . As they are both acid and basic in character, they exhibit analogies with glycocoll (see p. 355). The PYRIDINEDICARBOXYLIC ACIDS, NC 5 H 3 (C02H) 2 , have the following melting- points : cm' or Dipicolinic Acid, 226 ; /3/3' or Dinicotinic Acid, 323 ; a/3 or Quinolinic Acid, 190 ; Isocinchomeronic Acid, 236 ; ay or Lutidinic Acid, 235 ; /3y or Cinchomeronic Acid, 249. Quinohnic acid is formed by the oxidation of quinoline, its constitution being thus established ; and since in the hot it loses C0 2 from the a-position, giving nicotinic acid, the constitution of the latter is also fixed. Pyridinetricarboxylic Acids (obtained by oxidising cinchonine or quinine), as well as Pentacarboxylic Acids and Hydroxypyridinecarboxylic Acids, are also known. HYDROPYRIDINES. The Dihydropyridines are mentioned above. The tetra- hydropyridines and their derivatives are known also as piperideines, while the hexahydro- pyridines and their derivatives included in the term piperidines embrace pipecoline, NC 5 H 10 -CH 3 ; lupetidine, NC 6 H 9 (CH 3 ) 2 ; copellidine, NC 5 H 8 (CH 3 ) 3 , &c. PIPERIDINE, NC 6 H n , is obtained by heating Pipeline or piperylpiperidine, C 6 H 10 N-C 12 H 9 3 (m.pt. 129), which is the alkaloid contained in pepper, and is formed by the condensation of 1 mol. of piperic or piper inic acid, C 12 H 10 4 , or CH 2 C 6 H 3 -CH : CH-CH : CH-C0 2 H with 1 mol. of piperidine. For the constitution and syntheses of the latter, see p. 624. Piperidine boils at 106, has an odour of pepper, is strongly basic, and is soluble in water or alcohol. With H 2 2 it gives aminovaleraldehyde. Piperidine, being a secondary base, forms with 2CH 3 I an ammonium iodide derivative which, when distilled with silver oxide, gives an unsaturated open-chain, tertiary base ; in its turn the latter, with CH 3 I, Ag 2 0, and distillation, loses trimethylamine and forms Piperilene, CH 2 : CH-CH 2 -CH : CH 2 . To the group of heterocyclic compounds belong the following, which are of little importance : N CH /\ /\ /\ /\ HC OH HG CH HC CH CH 2 CH 2 HC CH HC CH N N CH 2 CH CO N CH NH Pyrone or pyrocomane Pyrazine or aldine Pyrimidine Morpholine, (m.pt. 32) J the iso- (m.pt. 47) is basic or m-diazine a base, b.pt. meric <*-pyrone is and with H gives (m.pt. 22) 129 coumalin piperazine, C 4 H 10 N 2 From these compounds can be derived coumalinic or comanic acid, C 5 H 3 2 -C0 2 H (also formed from malic acid); Meconic Acid, CgHO^OH) (C0 2 H) 2 , which can be obtained from opium and gives pyromeconic acid by elimination of C0 2 ; Chelidonic Acid, C 5 H 2 O 2 (CO 2 H) 2 , which is found in celandine, loses CO 2 giving comanic acid and pyrone. ^ ALKALOIDS These are found in various plants and have medicinal and often poisonous properties ; some of them, such as caffeine, theobromine, &c., were described on p. 368, and the principal ones having basic characters (vegetable bases) will be considered here. ALKALOIDS 627 They are almost all laevo-rotatory and have an alkaline reaction and a bitter taste. They are soluble in alcohol and to a less extent in ether, and are usually insoluble in water and in alkali ; in acids they dissolve with formation of crystallisable salts. Nearly all alkaloids are precipitated from their solutions by tannin, phosphomolybdic acid, potassium mercury iodide, HgI 2 ,KI, or aromatic nitro- derivatives (e.g. picric acid, &c.), &C. 1 From plants they are extracted with acid solutions and are then liberated with alkali and either distilled in steam or, if they are non-volatile, filtered. When converted into salts by means of strong acids, their specific rotatory power is not greatly influenced, since these acids are almost completely dis- sociated in aqueous solution ; with weak acids, however, the salts are only slightly dissociated and hence the rotatory power is different, being due to very different ions. A. Pictet (1906) regards the alkaloids not as assimilation products of the organism, but rather as nitrogenous decomposition products of proteins, nucleins, chlorophyll, &c., which have condensed with other substances present in the plants. It is supposed that alkaloids containing the pyrrole group have their origin in protein or chlorophyll, in which such group is certainly present, while those with a pyridine grouping have a similar origin, the trans- formation of the pyrrole into the pyridine nucleus being possible even in the laboratory ; the pyridine group itself does not appear to exist in the proteins, chlorophyll, &c. 2 1 Separation and Tests of Alkaloids. A mixture of these is separated as follows : I. From the neutral or acid aqueous solution, ether extracts : digitalin, picrotoxin, and colchicine, and from a solution of these the first and last are precipitated by tannin. II. From the alkaline aqueous solution, ether extracts : coniine, nicotine, brucine, delphinine, narcotine, vera.* trine, atropine, strychnine, aconitine, quinine, codeine, and physostigmine. III. From the alkaline aqueous solution, cliloroform extracts : cinchonine, caffeine, curarine, morphine, solanine, and theobromine. The separate alkaloids can be distinguished by the following colorimctric tests, arranged by Hager. The colours are represented shortly (as with the colouring-matters ; see later) as follow : = orange ; B = blue ; Br = brown ; D = decolorised or colourless ; Y yellow ; Or = grey ; Bl = black ; R = red ; r = rose ; Gn = green ; V = violet ; + = intense ; = weak. The reagents most commonly used are : (1) Erdmann's reagent : to 20 drops of a solution containing 10 drops of HNO 3 (sp. gr. 1-153) and 20 c.c. of water are added 40 c.c. of concentrated H a SO 4 . One cubic centimetre of this liquid is poured ou to 1 to 2 grms. of the dry aklaloid and the changes observed after 15 to 30 minutes. (2) Friihde's reagent : 0-5 grm. sodium molybdate in 100 c.c. cone. HjSO 4 . (3) Mandelin's reagent : 1 grm. ammonium vanadate in 200 grms. H 2 SO (monohydrate), (4) Marquis's reagent : a solution of formalin in sulphuric acid. (5) Lafou's reagent : sulphuric acid solution of ammonium seleuite. See Table on p. 628. Synthesis of Alkaloids and Medicine. Even during the most remote ages human beings sought remedies for their ailments in the principles contained in various plants and animals. Galen (A.D. 131-200) studied various medicines more rationally than had been previously done by Hippocrates (400 B.C.). Numerous medicines proposed by Galen were used as sovereign remedies for some centuries, until indeed Paracelsus (1493-1541) gave a new direction to medicine by contesting the theory of Galen and of Avicenna and by founding iatrochemistry, which had such a large following in the Middle Ages, and which ultimately degene- rated into the most fantastic sorcery (see History of Chemistry, vol. i, p. 14). Modern chemistry alone could yield medicine any real support, by rigorous control of the physiological and chemical actions of all the natural and artificial drugs. In the past the curative properties of various substances were discovered by pure chance ; this was the case, for instance, with antifebrin (acetanilide), which was administered to a patient in mistake for naphthalene. But nowadays a rational procedure is followed, use being made either of analogy in chemical constitution between the substance under consideration and others of known action or of systematic physiological tests, first on animals and afterwards on human beings. Until the beginning of the nineteenth century, the energies of chemists were directed to the discovery of the active and essential principles of those parts of plants successfully applied in medicine. When these were isolated in the pure state, attempts were made to establish their chemical structures and, in some cases, to effect their manufacture synthetically. A early as 1805 Serturner discovered and isolated morphine, the active principle of opium, and in 1821 Pelletier and Caventou discovered the alkaloids of cinchona bark, which were studied in 1850 by Strecker with the object of ascertaining their chemical constitution. The synthesis of these alkaloids was by no means an easy task, but in cases where they themselves have not been obtained by laboratory reactions, simple derivatives have been prepared, and these often exhibit similar therapeutic properties. Thus synthesis has given codeine (or methylmorphine) and dionine (ethylmorphine), which in many cases are excellent substitutes for morphine, as they are scarcely if at all poisonous. Derivatives of cocaine, such as eucaine (a derivative of y-methoxypiperidine ; Ger. Pats. 90,235 and 97,672), and of quinine, such as euquinine (the carbethoxy-derivative of quinine, without the bitter taste of the mother-substance), have also been prepared. Chemical investigation not only gives new products but leads to improved manufacture and consequent cheapen- 628 ORGANIC CHEMISTRY Alkaloid H 2 S0 4 cone. HN0 3 sp. gr. 1-4 Erd- mann's reagent Frohde's reagent Mando- lin's reagent Marquis's reagent Lafou's reagent 2 per cent, aqueous furfural Aconitine Y-Br; F Y-Br ; in Y-Br, after 24 the hot then D hours Br- R-Br R; after 4 hours D Atropine DOT Br substance D D Br, solu- tion D Brucine r +.R,then Band R, then r then F F ; after 24 hours D Quinine D D - D Dor Gn Quinidine - D D - D - D Cinchonine D D D D Digitalin Br then R - Br R-Br, then + Othen R ; after +R; in 15 hours 30 minute +R Bl-Br ; after 24 hours * 6'n-F Caffeine D D D D Codeine D, after 8 r then F D then B G'nthcn B Gn in hot F Gn days B or after 24 B hours F Cocaine D D D D Colchicine + F FthenB Y Fand and F -Gn-Y Coniine D Dthen D Y -FthenD Morphine D, in hot then F .Rtiien V, then R Gn R R then F Br . Gn, Br, and Gn after 24 hours F Narcotine - Y then F then D Y,0 Gr, then R Gn-Br, in and Br, Y, r the hot R after 24 hours r Narceine Br then F F Fthen Br then F 0, in the F B, Br - Br andD hotB Nicotine D Fthen D Fthenr R-V and then D Papaverine V then B V then B F then B YD Physostigmine FthenGn ' Pipeline R then and -Y-Br Fthen Br-Gn R GnBr Y-Gn with alkali -f BrBl toGnB R Solanine 0, and Dthen -Y JSthen Br R-Br R then F after 20 -B Br and F hours Br Strychnine D and with Y D ; with D V bichro- MnO 2 F mate F then R Thebaine R then Y R then O R then O R R - R then and Z D Theobromine D D D Z> Veratrine O then R Y O then R F then R R RBr F, in hot B Adrenaline Y Br then Y,Br RBr Gn -Y GnR Berberine Gn + RBr Gn Gn-Br B Fthen Fin hot, BrGn VBr Gn Hydrastine Gn Gr R Br Picrotoxin O YGn -R Digitoxin Br BrV r BrV V ANESTHETICS 629 The consumption of alkaloids is growing in all countries. The imports into Italy (exclusive of quinine) amounted to 14,200 kilos in 1908 and 17,320 kilos, of the value of 138,545, in 1910. CONIINE, C 8 H 17 N, is found in hemlock (Conium maculatum). For its constitution and syntheses, see above. NICOTINE, C 10 H 14 N 2 , is a strong diacid base which, in combination with malic and citric acids, forms the poisonous alkaloid of tobacco. It is an oil boiling at 247 and possess- ing a very strong odour ; it is soluble in water, alcohol, or ether, and turns brown in the ing of the old ones. Thus, quinine, which twenty years ago cost 40 per kilo, Is now sold in a highly pure state for 32s. Vast works now turn out enormous quantities of synthetic drugs, although these are administered in closes of centigrammes ; thus, antipyrine, discovered by Knorr, was consumed to the extent of hundreds of thousands of kilos in the first few years during which influenza made its appearance. Modern industrial conditions have rendered possible the development of serotherapy (see p. 115). and great results are now promised by organotherapy or ototherapy. This is based on the fairly general phenomenon that in the different organs of a healthy individual substances are continually produced capable of guarding them against different affections. This principle, introduced vaguely and confusedly by Brown-Se'quard in France in 1891, was in 1895 brought forward with triumph by Baumann, who found that in many persons goitre is due to deficient secretion of iodo-products by the thyroid glands (see vol. i, p. 151), and, having extracted the active iodine principle, thyroidin, from the thyroid of healthy sheep, that this constitutes a rapid and effective cure for goitre. For the treatment of other diseased organs, ovarin, cerebrin, nucleiti, &c., were prepared from the corresponding organs of healthy animals. Coal-tar derivatives have been employed for the synthesis, not only of artificial alkaloids, antipyretics, and antiseptics, but also of an important group of anaesthetic or hypnotic substances which have been of great service to medicine and especially to surgery in rendering painless the most complicated operations. At first, substances such as ether and chloroform were employed which produced general ancesthesia of the organism ; but the use of these, especially of chloroform, was attended by much inconvenience and often by death of the patient. Sulphuric ether was recognised as an anaesthetic by Faraday as early as 1818, but it was used for the first time by the American doctor, 0. W. Long, in 1842. The anaesthetic is carried by the blood into contact with the nerve-centres which perceive pain, producing a poisoning and a paralysis which last for some time, but at the same time those centres which govern the action of the heart and of respiration are also affected, thus causing the dangers and disturbances accompanying general anaesthesia. The nervous currents start from the periphery, from the points where the surgical operation begins, and are transmitted to the brain, which transforms them into the sensation of pain, and it is precisely by the influence of the anaesthetic on the cerebral centres that pain is avoided. But anaesthesia ceases to be dangerous when the paralysis is effected on the peripheral nerve-centres at the beginning of the nervous currents, without, however, reaching the brain. In this way the idea of local ancesthesia was arrived at, this being much more rational and much less dangerous, since by its means only the single organ or region of the body to be operated on is rendered insensible. To chloroform, ether, &c., were added, in 1885, cocaine, which paralyses only the sensitive peripheral nerves and does not influence the motor nerves. It can now be indicated which specific atomic groupings in the molecules of anaesthetics or hypnotics confer on these their special properties. Hypnotics include those of (1) the chloral hydrate group, to which belong also chloralamide (chloralformamide) /-1TTT /I TT and paraldehyde ; (2) the tert. amyl alcohol class, rH*^> c< CoH *' cliaracter i sed bv the presence of a hydroxyl and of a carbon atom united to three alkyl groups, the action of these compounds increasing with the molecular weight ; (3) the intermediate dormiol [tert. amylchloral, CCl a 'CH(OH)(OC 5 H,,)] class ; (4) the urethane deri- vatives, including hedonal (methylpropylcarbinol urethane, NHj'CO-O-CH(CH s )(C s H 7 ) ; (5) a group of compounds containing a single carbon atom united to two alkyl groups and to two sulphonic residues, e.g. trional, ("*TT SO "C* TT Tr 3 ^> c< Cso 1 '-C H ( met fyl sl 'lP Jlona l or diethylsulphonemethylethylmethane) ; (6) a group studied by B. Fischer and consisting of urea derivatives, e.g. NH s -CO-NH-CO-CH(CjH 5 ), (diethylacelylurea)ot, better, diethylmalonylurea, CO<^ NJ T] CO ^>C<^Q ! V S (diethylbarbituric acid), which bears the name of veronal (m.pt. 191" ; it was prepared by E. Fischer and J. Mering, patented by Messrs. Merck in 1903 and then made by Messrs. Fr. Bayer, of Elberfeld) and serves to replace chloroform, being free from the dangerous consequences of the latter (provided that it is not administered to patients with weak kidneys). Change of the alkyl groups in veronal is accompanied by change in the properties ; thus, dimethylbarbituric acid has no hypnotic properties, dipropylbarbituric acid is more effective than veronal, while dibenzylbarbituric acid is without action, possibly owing to its slight solubility. According to H. Meyer and Overton, all substances capable of dissolving fats are more or less anaesthetic, and according to Nicloux (1909) the substance of the nervous system contains an abundance of lipoids, i.e. of compounds soluble in the same solvents as fats and hence capable of fixing the anaesthetics (they may contain nitrogen and also phosphorus). Thus the quantity of anaesthetic fixed by the organism and hence effective is directly related to the quantity of lipoids present in the various parts of the body. It is also interesting that structural isomerism produces marked change in the physiological action, tropacocaine, for instance, being j>u anaesthetic, while benzoyltropine acts as a mydriatic. Of the numerous other anaesthetics, orthoform (methyl ester of m-amino-p-hydroxybenzoic acid), alipme holocaine, may be mentioned. But in order that local anaesthesia may be efficacious and lasting, it is necessary to prevent the anaesthetic inoculated at a certain place from being carried away (resorbed) by the blood, and this was at first attained by causing the venous blood at that place to stagnate by preventing circulation. The same end was reached later by intense local cooling produced by the rapid evaporation of ethyl or methyl chloride. For internal surgical operations (e.g. in the thorax, &c.), adrenaline, C,Ha(OH) 2 -CH(OH)-CH 2 'NH>CH,, is of the greatest use, as it produces considerable contraction of the blood-vessels without driving all the blood from them, although it prevents fresh blood from arriving ; the anaesthetic can thus be kept as long aa is desired in the inoculated region. The substitution of cocaine by stovaine (less poisonous) leads to partial spinal ancesthesia orjmedullary anaesthesia, which now permits the most difficult surgical operations on the abdominal organs and even renders possible painless childbirth. 630 ORGANIC CHEMISTRY air. When oxidised by permanganate it forms nicotinic acid, and as further it contains also a pyrrolidine group, its constitution is represented as follows : /CH^CH, CH/ \C CH< ^N CH^ S N(CH 3 ) CH 2 Synthetically it is obtained from /3-aminopyridine which is converted into its mucic cid salt, and then passes through the following stages : N< N CH CH .CH-CH S NH-CH /3-Pyridylpyrrole a : /3-Pyridylpyrrole ,CH CH Nicotine X N(CH 3 ) Nicotyrine CH Practically it is prepared from ordinary tobacco extract, by diluting, rendering strongly alkaline with NaOH, and extracting with ether. From the ethereal solution, the alkaloid is extracted by shaking with dilute sulphuric acid and decanting off the acid solution. The latter is again made strongly alkaline and shaken with ether, and the ethereal solution dehydrated by means of solid NaOH. The ether is then distilled off and the remaining nicotine distilled in a stream of hydrogen. It is a very powerful poison and is used medicinally to counteract nervous irregularity of the heart and is employed in agriculture, as tobacco extract, to kill insects. 1 Impure 75 per cent, nicotine costs 148s. per kilo, and the pure product 184s. 1 Tobacco is a herbaceous plant, originally an annual but now sometimes a biennial, of the order Solonacese (Nicotiana tabacum), which includes about fifty species and sub-species of American origin, e.g. the Virginia tobacco plant (Nicotiana tabacum, see Fig. 420), the Maryland large-leaved tobacco (N. latissima, N. rustica, N. suffruticosa, C<( C 2 H/ X CH 2 -N(CH 3 ) 2 C 2 H/ X CH 2 -N(CH 3 ) 2 Stovaine Alipine prepared by Messrs. Bayer in 1905. Both of these are less poisonous than cocaine, but have not its property of contracting the blood-vessels. They are therefore mixed with adrenaline, which shows this property in a marked degree and also diminishes the toxicity of certain alkaloids, especially of cocaine. NARCOTINE, C 22 H 23 7 N, exists to the extent of 6 per cent, in opium, melts at 126, and is a slightly poisonous, weak, tertiary base containing three methoxyl groups. When hydrolysed, narcotine gives Meconic Anhydride, C^H^O^, and Cotarnine, C 12 H 13 3 N, which is a derivative of isoquinoline (see later), and with bromine gives dibromopyridine. STRYCHNINE, C 21 H 22 O 2 N 2 , is present, with Brucine, C 23 H 26 O 4 N 2 , and Curarine, in the seeds of Strychnos nux vomica. They are very powerful poisons, which, even in small doses, cause death, accompanied by tetanic muscular contorsions ; curarine is used as an antidote to the other two alkaloids. Strychnine melts at 265, and is a mono-acid tertiary base slightly soluble in water ; it gives indole and quinoline when fused with potash and /3-picoline on distillation with lime. QUININE, C 20 H 24 O 2 N 2 . The bark of various species of cinchona has yielded, up to the present, twenty-four alkaloids, the most important being quinine and Cinchonine, 634 ORGANIC CHEMISTRY C 19 H 22 O 2 N 2 , both of these possessing in different degrees febrifugic properties. The other alkaloids include Hydroquinine, C 20 H 26 O 2 N 2 ; Cinchonidine, C 19 H 22 ON 2 ; Hydro- cinchonidine, CjgH^ON-j ; Quinidine, C 2 oH 24 O 2 N 2 , &c. Quinine is laevo- rotatory, slightly soluble in water and odourless and has an intensely bitter taste ; it melts at 177, or, when crystallised with 3H 2 O, at 57. It is a di-acid base, containing two tertiary nitrogen atoms capable of salt-formation with two equivalents of acid, then often giving aqueous solutions showing blue fluorescence characteristic of quinine. It contains a hydroxyl and a methoxyl group, and its constitutional formula, although not completely established, must consist of two cyclic systems, NC 10 H 15 (OH) NC 9 H 5 -OCH 3 , the first being somewhat analogous to tropine (see above) and the second representing 5-methoxyquinoline, which can be obtained by fusing quinine with potash. After pro- tracted investigation, W. Konigs (1906-1907) arrived at the following probable structures for cinchonine and quinine : N- *^X1 2 (\J1LJ\J ( 1 CH C C /\/\ H 2 C HC C CH \ ^Hj ^ J i 2 H 2 CH-CH : CH 1 II 1 CH HC C CH \/\7 CH N Cinchonine (OH)C CH,, CH 2 CH 2 H 2 C I CH-CH : CH 2 \l/ CH CH 2 CH C /\/\ CH 3 0-C C CH I II I HC C CH \/\/ CH N Quinine Rabe (1906-1907), however, proposed for cinchonine the formula : CH 2 CH CH CH : CH 2 */ CH 2 dr 2 I ' I > C(OH) N CH 2 which is in harmony with the Beckmann oxime reaction. Oxidation of quinine gives, among other products, Quinic Acid, C 9 H 5 N(OCH 3 )-CO 2 H. To combat fever, especially malarial fever, use is made of the normal sulphate of qiiinine, (C 20 H 24 2 N 2 ) 2 ,H 2 SO 4 ,8H 2 (from alcohol it crystallises with 2H 2 0), or of quinine hydro- chloride, C 20 H 24 2 N 2 ,HC1,2H 2 O, which is far more readily soluble in water. Quinine bisulphate or acid sulphate contains 1 mol. s of quinine per 1 mol. of sulphuric acid. Quinine is extracted from the finely ground bark by mixing it with lime and extracting with hot mineral oils (paraffin oil, &c.) of high boiling-point. From this solution the alkaloid is obtained by shaking with dilute sulphuric acid, neutralisation of the acid solution with sodium carbonate in the hot resulting in the crystallisation of most of the quinine QUINOLINE 635 as sulphate from the cold solution, the other alkaloids remaining dissolved. From the sulphate the quinine is liberated by means of ammonia. The purification of quinine is not easy and is sometimes effected by precipitating it from solution as tartrate by addition of Rochelle salt. Statistics. Quinine bisulphate costs about 28s. per kilo ; the sulphate 32*. ; and the hydrochloride 40s. Of the world's output of cinchona bark, 90 per cent, conies from Java, which in 1900 exported 60,000 quintals, and in each year from 1905-1909 more than 80,000 quintals of the bark, giving 6 to 6-5 per cent, of quinine sulphate. In 1898 Germany imported 3537 tons of cinchona bark, worth about 128,000 ; ii? 1905 the imports amounted to 2594 tons, of the value of 168,000, and in the same y eat Germany exported 1404 quintals of quinine and its salts, of the value of 224,000, and 461 quintals of other alkaloids, of the value of 424,000 ; in 1907 the exports were 1700 quintals and in 1908 about 1500 quintals at 22s. per kilo. England imported 1080 tons (35,759) of cinchona bark in 1909, 1123 tons (39,520) in 1910, and 1020 tons (37,169) in 191], while the imports and exports of quinint, salts were as follow : Imports Exports 1909 . . . . 82,556 55,065 1910 . \- . . . 90,771 56,866 1911 ... . . . 98,056 75,080 The United States imported 1500 tons (52,200) of cinchona and similar barks in 1910 and 1550 tons (55,000) in 1911 ; also quinine salts and alkaloids to the value of 76,400 in 1910 and 95,400 in 1911. In 1904 Italy imported 1627 quintals (in 1908, 1384) of cinchona bark of the value of 13,650. In 1878, at the time of the Fabrica Lombarda of quinine in Milan, Italy consumed 10,000 kilos of quinine (5000 furnished by the Fabrica Lombarda, which also sent 20,000 kilos, at 28 to 32 per kilo, to Russia). After 1902, in consequence of the valuable studies of Ross, Grassi, and Celli on malaria (a disease which is transmitted by the Anopheles mosquito and against which a couple of small doses of quinine per week render one immune), a Government monopoly was instituted to distribute quinine cheaply or gratuitously in the malarial centres. The beneficial results obtained are shown by the following figures : in 1902-1903 the consumption of quinine distributed in this way was 2242 kilos ; in 1903-1904 7234 kilos ; in 1904-1905 14,071 kilos ; in 1905-1906 18,712 kilos, and in 1906-1907 21,723 kilos. The mortality from malaria, which was 21,000 in 1887 and 15,865 in 1900, fell to 9908 in 1902, to 8513 in 1903, to 8501 in 1904, to 7838 in 1905, and to 4690 in 1906. In addition to these advantages, the Italian Government made in 1906 a profit of more than 1400 from its commerce in quinine. There is now scarcely any quinine made in Italy, but the imports amount to 30,000 to 40,000 kilos, 20,000 to 30,000 kilos being converted into pastilles and sold practically at cost price to combat malaria. 5. QUINOLINE AND ITS DERIVATIVES Quinoline and pyridine are related in the same way as naphthalene and benzene. CH CH QUINOLINE, C 9 H 7 N, i.e. Hc^ ' isa CH highly refractive, colourless liquid of peculiar odour and is found in bone tar and also in coal-tar, but is now prepared in the pure state by Skraup's synthesis. It is slightly soluble in water, has the sp. gr. 1-1081 at 0, boils at 236 and functions as a tertiary base (the nitrogen not being combined with nitrogen). With acids it forms salts, e.g. the bichromate (C 9 H 7 N) 2 H 2 Cro0 7 . 636 ORGANIC CHEMISTRY Its constitution is deduced from the following syntheses : (1) By the interaction of Allylaniline and Pb0 2 at a red heat : H CH = 2H0 H NH N (2) Skraup obtained it by heating aniline with glycerol, sulphuric acid, and nitrobenzene ; in this way acrolei'n is formed, which then gives Acroleih- aniline, C 6 H 6 -N : CH-CH : CH 2 . The nitrobenzene acts purely as an oxi- dising agent and can be replaced by As 2 3 . (3) o-Nitrocinnamaldehyde on reduction gives o-aminocinnamaldehyde, which loses 1 mol. H 2 and yields quinoline. the fact that the latter is an ortho- derivative of benzene being thus proved : H H 2 = When quinoline is oxidised, the benzene nucleus is attacked first, with ^COOH formation of a dibasic Quinolinic Acid, COOH , which gives pyridine, N , when distilled with lime. Hence, as was suggested Jong ago byKorner, N quinoline contains a benzene and also a pyridine nucleus. It is analogous to naphthalene, one u-CH group being replaced by a nitrogen atom. .That the linkings in quinoline are, at least in part, olefinic double bonds is shown by the behaviour of this compound to ozone. Quinoline forms many isomeric derivatives, seven monosubstitutecl, twenty-one disubstituted, and still more trisubstituted compounds being possible. The positions of the replaceable hydrogen atoms are indicated by numbers or by the letters , /3, and y for the pyridine nucleus and o, m, p, a (ortho-, meta-, para-, ara-) for the benzene nucleus. The constitution of quinoline derivatives can be determined by means of the general synthesis of Skraup, variously substituted anilines with the sub- stituents in the benzene nucleus being used ; or often by oxidation, which usually attacks the benzene nucleus and not the pyridine nucleus, so that it is easily ascertained whether the substituent is in the one or the other nucleus. The sulpho-acids (or sulphonic acids) of quinoline, when fused with KOH, give hydroxy- quinolines, and these, on being heated with KCN, form cyanoquinolines, which are converted by hydrolysis into the corresponding quinolinecarboxylic acids those containing the carboxyl in the benzene nucleus are called quinolinebenzocarboxylic acids. Oxidation of cinchonine gives cinchonic acid, C 9 H 6 N-CO 2 H (m.pt. 254), which is quinoline-y-carboxylic acid, and from this is derived quinic acid (see above], C 9 H 5 N(OCH 3 )-C0 2 H (p ; y), con- sisting of yellow prisms melting at 280. When acridine is oxidised it yields quinoline- - : /3-dicarboxylic acid or acridic acid. QUINOLINE DERIVATIVES 63? Carbostyril is 2 -Hydroxy quinoline, , and has the character of the phenols, OH dissolving in alkali and being reprecipitated by C0 2 , &c. When quinoline is reduced with nascent hydrogen, this unites with the nitrogenated H 2 nucleus, forming Tetrahydroquinoline, C 9 H n N, or , which behaves as a \/\/ 2 NH secondary aromatic amine (^>NH). If the reduction is pushed further, the hydrogen is added also to the benzene nucleus, forming decahydroquinoline, C 9 H 17 N, which behaves like an aromatic amine. Quinaldine or a-Methylquinoline, CjoHgN, is found in coal-tar and boils at 246 ; with phthalic anhydride it gives a fine colouring- matter, Quinoline Yellow, C 10 H 7 N(CO) 2 C 6 H 4 . When quinoline is heated with metallic sodium it gives diquinolyl, CgHeN-CgHgN, analogous to dipyridyl and diphenyl. Polymerisation of quinoline yields diquinoline, (C 9 H 7 N) 2 , crystallising in yellow needles. METHOXYQUINOLINE, C 9 Ht;N-OCH 3 , corresponding with anisole, resembles quinoline; among its derivatives are the antipyretic, Thalline, C 9 H 10 N-OCH 3 , and Analgen (o-ethoxy-a-benzoylaminoquinoline). ISOQUINOLINE, C 9 H 7 N or \ I /\ I / , is a colourless liquid boiling at 237, melting at 21 and forming a slightly soluble sulphate. It is obtained from tar and also synthetically by heating the ammonium salt of homophthalic acid : xCH 2 -COONH 4 CH 2 -CO C 6 H 4 <^ , = 2H 2 + NH 3 + C 6 H 4 / | , X COONH 4 \CO NH riTi rrCO, forms reddish yellow prisms soluble in alcohol and in hot Water, and may be regarded as the lactam (see p. 355) of Isatinic Acid, NH 2 C 6 H 4 CO COOH. It is obtained from o-nitrobenzoylformic acid (see later, Indole), by oxidising indigo with NK nitric acid, &c. It dissolves in KOH, giving first a violet colour (C 6 H 4 <^ ^>CO), while TU-TT in the hot it yields potassium isatinate, CeH^-c^,,, 2 . Oxidation of isatin with chromic OU ' V^UjjIY xNH-CO acid gives rise to Isatic Acid (anhydride of anthranilcarboxylic acid), C 6 H 4 ^ XX) -O / co \ From PseUdoisatin, C 6 H 4 (' ^C-OH (which would be a lacttm) is derived the methyl ether or Methylpseudoisatin, C 6 H 4 (' V>OCH 3 (red powder). Methylisatin, \NCO, is formed by reducing isatin with zinc and W il' -- HC1 and readily gives isatin again on oxidation. It is the internal anhydride of o-amino- mandelic acid, and exhibits both basic and acid properties. It crystallises in colourless prisms, melting at 180. OXINDOLE, C 6 H 4 <CO, acts both as an acid and as a base, and hence dissolves in alkali and in HC1. It is the lactam of o-aminophenylacetic acid, and can, indeed, be obtained by reducing o-nitrophenylacetic acid. It forms colourless needles, m.pt. 120, and forms dioxyindole on oxidation. NH- INDOXYL, C 6 H 4 <^ ^ CH ' is isomeric wit h tne preceding compound, and is x C(OH) r formed by fusing indigo with KOH or by the elimination of CO 2 from indoxylic acid or indophore'. It occurs in the urine of herbivorous animals in the form of Potassium Indoxylsulphate, NH C 8 H 6 N'0'SO 3 K(?'nrfica of the urine). Derivatives of Pseudoindoxyl, C 6 H 4 <_~ ^>CH 2 , are also known. NH - X SKATOLE, C 6 H 4 <' ^**, is formed during the putrefaction of protein or by fusing the latter with KOH, and is hence found in the faeces ; it is found in the African Viverra civetta. It forms white scales, m.pt. 95, with an intense fecal odour. / NH \ INDOLE, C 6 H 4 (' ^C**, is of importance owing to its intimate connection with \CH^ indigo. By treating o-nitrobenzoyl chloride with AgCN, the nitrile is obtained and this, on hydrolysis, gives o-nitrobenzoylformic acid : COC1(1) CO-CN CO-COOH. LeH *ndigo fell to 115,000 hectares ; in 1909-1910 there was a slight increase to 117,450 hectares. 640 ORGANIC CHEMISTRY argentea, and others of less importance). They are herbaceous shrubs 50 to 100 cm. in height, covered with silky hairs, with pinnate leaves and many small leaves. From the results of tests made at Calcutta it would seem that Indigofera leptostachya, cultivated in Java but indigenous to Natal, is better in every respect than Indigofera tinctoria, while it lasts four to five years. Still better results seem to be given by Indigofera erecta. In order that the indigo may be extracted from the cut plants, it is necessary that the glucoside they contain (indicari) consisting of a compound of glucose with indigotin (the leuco-base of indigo) be decomposed by fermentation in large vessels with water. After 10 to 14 hours the glucose is fermented, while the indigo, owing to the presence of ammonia, forms a yellowish solution. The liquid is transferred to deep vats, where it is subjected to " beating " for 2 to 3 hours with wooden paddles or wheels, or to " blowing " by means of a current of air. The oxidation thus effected causes the separation of the indigo in flocks, which are removed by decantation after 3 to 4 hours : C(OH) NH NH C 6 H 4 Indigotin /C(OHk 2C 6 H 4 ^ )CH > \ NH / Indoxyl r*n pn /V>W v s\j\J \ C\P . fur \ri TI 6"4\ /V : ^\ /^G^i Indigo blue The 5 per cent, indigo paste separated by decanta- tion is passed through sieves to remove fragments of the plants and is then boiled by means of steam for 15 minutes in order to sterilise the mass which would otherwise undergo change and to eliminate FIG. 422. part of the brown matter and to effect better sepa- ration of the particles of indigo. These then deposit more easily and are collected on a large cloth filter, the first liquid passing through being returned to the filter until it comes through faint red ; the 8 to 12 per cent, paste thus obtained is pressed in primitive presses. The large cakes thus formed contain about 80 per cent, of water and are cut into small cubes, which are arranged on grids, dried in the air for two or three months and placed on the market in boxes holding 50 to 140 kilos under the name of cakes. During the drying, these cakes evolve ammonia and become covered with mould, which is finally removed with brushes. The yield of indigo is about 0-2 per cent, on the weight of the green plant or 2 per cent, on that of the dry plant. To combat the competition of artificial indigo, various improvements have been intro- duced during recent years into the methods of cultivation, manuring, and extraction ; attention may be directed to the rational fermentation with suitable enzymes (oxydases) proposed by Calmette and others (Fr. Pata. 300,826 and 302,169). The indigo-content of the cakes varies considerably, some of those on the market containing only 20 per cent, and others as much as 90 per cent. It hence becomes necessary to determine the value of any sample on the basis of the proportion of pure indigo ascer- tained by exact analysis. 1 According to Fr. Pat. 323,036 an increased yield and an 1 Analysis of Commercial Indigo. Commercial indigo from Bengal contains, on an average, 60 per cent, of indigotin ; that of Madras, 30 to 50 per cent. ; that of Java, 72 to 82 per cent. ; that of Guatemala, about 40 per cent. ; that of Martinique, 60 to 70 per cent. ; and that of Cambay, China, and Tonkin, 8 to 15 per cent. Indtgotin can be estimated as follows : 1 grm. of well-dried indigo is mixed (in a bottle with a ground stopper) fithlO grms. of garnets or glass beads and 20 c.c. of sulphuric acid mixture (composed of 3 parts of concentrated sulphuric acid and 1 part of oleum containing 20 per cent, of free S0 8 ). The mass is thoroughly mixed and is terwards shaken occasionally over a period of 12 hours or so, until solution is complete, the whole being then poured carefully into cold water and the bottle thoroughly rinsed out. The aqueous solution is boiled for 10 minutes ana altered, the filter being washed with hot Water until the washings become colourless and the filtrate then mods INDIGO 641 improved product are obtained by macerating the fresh plants in presence of tannin materials which leave only the indigo undissolved. The cakes of indigo are blackish blue in colour and give a fracture showing a bronzy reflection. Natural indigo always contains, besides indigotin, other substances and colour- ing-matters (such as indigo gum, indigo brown and red, &c.) which affect the tint, sometimes favourably. A good Bengal indigo gave, on analysis, 62 per cent, of indigo blue, 7-3 per cent, of indigo red, 4-7 per cent, of indigo brown, 1-5 per cent, of indigo gum, 6 per cent, of water, and 19 per cent, of mineral matter. Pure or refined indigo is obtained in various ways, e.g. the crude indigo is treated with a mixture of concentrated acetic and sulphuric acids, the indigo alone passing into solution as sulphate, which is decomposed after filtration by excess of water, this precipitating pure indigo or indigotin. In order to avoid dilution with water and loss of acid, it has been proposed to separate the sulphuric acid directly by addition of calcined sodium sulphate, which transforms it into bisulphate ; the acetic acid is then distilled off and the bisulphate removed together with a little water. According to Ger. Pat. 134,139 pure indigo is extracted from the crude product by means of hot, crude pyridine. To purify artificial indigo, it is heated, according to Ger. Pat. 179,351, at 200 to 270, at which temperature it does not sublime or decompose, while the indigo red and other impurities are destroyed, leaving an indigo highly valued for its fine bronzing. Of some interest is colloidal indigo, which behaves like dissolved indigo, and has been recently prepared by Mohlau by heating, out of contact with the air, a suspension of indigo in an aqueous solution of alkali and sodium hydrosulphite, the liquid being treated, after cooling, with protalbinic acid (obtained by Mohlau by the alkah'ne hydrolysis of protein and subsequent dialysis ; this acid has the power of precipitating various metals in a colloidal state from their salts). Addition of hydrogen peroxide to the filtered liquid gives indigo blue in the colloidal condition, which is retained even after evaporation. Properties. Pure indigo forms a dark blue powder which, when rubbed, gives a metallic, coppery reflection. It sublimes at about 170, giving red vapour and forming copper-red, shining prisms. It is insoluble in water, alcohol, ether, alkali, or acid, and dissolves only slightly, even in the hot, in amyl alcohol, chloroform, phenol, carbon disulphide, pure acetic acid, nitro- benzene, aniline or melted paraffin. It has neither odour nor taste and is indeed an almost completely indifferent substance ; this explains why, although materials have been dyed from time immemorial in the Far East, in Europe no process for dyeing textile fibres was discovered for so many centuries until the sixteenth. The portion soluble in hot aniline colours this blue but colours fused paraffin purple-red ; from these solutions, rhombic crystals showing marked dichroism separate on cooling. From hot oil of turpentine indigo crystallises in blue plates. Concentrated sulphuric acid converts it in the hot into a monosulphonic derivative, soluble in water but insoluble in salt solutions. With fuming sulphuric acid it forms the disulphonic compound, which gives more soluble salts, the sodium salt being sold as a paste under the name of indigo-carmine, this dyeing wool like an acid aniline dye. When dry distilled, indigo gives aniline and other aromatic compounds. up to a litre. Fifty cubic centimetres of this solution are mixed with 900 c.c. of distilled water, and the liquid titrated with 0-05 per cent, potassium permanganate solution until the blue colour becomes golden yellow without green reflection. In order to accustom the eye to this end-point, which is not sharp, it is advisable to make a com- parative test with pure indigo of known strength ; 1 c.c. of the permanganate solution corresponds with about 0-00125 grm. of indigotin. In order to prepare pure 100 per cent, indigo for purposes of comparison, 10 grins, of pure, powdered artificial indigo (98 per cent.) marked B.A.S.P. or M.L.B.) are treated in a beaker with 120 grms. of caustic soda solution (sp. gr. 1-21), 330 grms. of concentrated sodium hydrosulphite solution and 100 grms. of water (or, if 50 grms. of 20 per cent, indigo paste are taken, only 60 grms. of water are added), the mixture being heated on a water-bath at 40 to 50 with occasional shaking and the air being gradually expelled from the beaker by means of a current of coal-gas. When solution is complete, the liquid is rapidly filtered and a current of air passed into the yellow or greenish filtrate. The precipitated indigo is collected on a hardened filter and washed first with hot water, then with hot dilute hydrochloric acid (30 c.c. of the concentrated acid diluted to a litre), next with water again, and repeatedly with alcohol and with alcohol and ether. When dried at 101 to 110 until of constant weight, the product represents pure 100'per cent, indigo.. n 41 ORGANIC CHEMISTRY Energetic oxidising agents (nitric or chromic acid or permanganate) decolorise it more or less rapidly, converting it into isatin. Chlorine, bromine, and iodine give halogenated derivatives of isatin. The white indigotin, which is the leuco-base of indigo blue, is obtained from the latter in a soluble form, by the action of alkaline reducing agents (sodium amalgam, ferrous sulphate, hypophosphorous or hydrosulphurous acid, glucose, gallic acid, &c.) or enzymes. When heated with acid, the greenish yellow alkaline solution deposits indigotin white, which is readily converted into the blue form by the oxygen of the air. Indigo may be regarded as a substantive dye which colours both animal and vegetable fibres without a mordant. It is first reduced in the vats by means of enzymes in presence of sugar, urine, zinc, arsenic, or reducing salts (sulphites, hydrosulphites), thus becoming decolorised, soluble in alkali and capable of impregnating textile fibres, on which it becomes firmly fixed when rendered insoluble by the action of atmospheric oxygen. In 1890 the German Government permitted alizarin blue to be used for dyeing part of the cloth for military uniforms, these having been previously coloured exclusively with indigo. The first efforts to ascertain the chemical nature of indigo were those of Erdmann and of Laurent, who simultaneously (in 1841) obtained isatin by oxidising indigo with nitric acid. In 1848 Fritzsche obtained aniline by distilling indigo with caustic potash ; Baeyer and Knop, in 1865, reduced indigo to dioxyindole, oxindole, and indole, the last of these being prepared synthetically by Baeyer and Emmerling in 1869 from o-nitro- cinnamic acid. In 1870 Engler and Emmerling effected the first complete synthesis of indigo by heating o-nitroacetophenone with lime and zinc dust, and in 1874 Nencki pre- pared indigo by oxidising indole with ozone. In an interesting series of studies extending from 1870 to 1878 Baeyer and his pupils established the constitution of, and synthesised, oxindole, transforming it into isatin, and the latter, in various ways, into indigo. The new complete synthesis effected by Baeyer in 1880-1882 firmly established the structure of the indigo molecule. Of the new syntheses of indigo following that of Baeyer which, in spite of costly attempts, could not be rendered capable of industrial application the most important from a practical point of view is that of Heumann (1890), in which fusion of phenylglycine- o-carboxylic acid with alkali is succeeded by oxidation. The starting-point and the various intermediate products of Baeyer 's 1880 synthesis of indigo are as follow : P TT JUH 2 -CO 2 H CH 2 _^C(: NOH) Mi4< N Q > CeH^^jj^-U. > C 6 H 4 <_ NH _ o-Nitrophenylacetic acid Oxindole Isatoxime Amino-oxindole Isatin Isatin chloride CO CO C 6 H 4 < NH >C : C< Indigo Baeyer's other synthesis, which was tried on an industrial scale by the Badische Anilin- und Soda-Fabrik of Ludwigshafen in 1882, and gave a yield of 60 per cent., started from benzaldehyde, the product of the interaction of benzylidene chloride and sodium acetate being nitrated (and subsequently esterified) and a mixture of 70 per cent, of o-nitrocinnamic acid and 30 per cent, of p-nitrocinnamic acid thus obtained. After removal of the latter, the former is converted into the dibromide, which, with alcoholic potash, loses 2HBr and forms o-nitrophenylpropiolic acid, this giving indigo when heated with alkali and glucose : CH : CH-C0 2 H C : C-C0 2 H CO ^o^^NOg Le 4< -N0 2 C 6 H 4 < N O-Wtrocinnamie acid o-Jf itrophenylpropiojic acicl Indigo SYNTHESES OF INDIGO 643 Owing to the high price of o-nitrophenylpropiolic acid, this artificial indigo is used only for printing textiles. In 1882, by means of a new and theoretically elegant synthesis, Baeyer and Drewsen succeeded in raising the yield to 70 per cent. ; o-nitrobenzaldehyde and acetone were condensed in presence of caustic soda, indigo being formed as follows : f 2CH 3 -CO-CH 3 = 2C 6 H 4 <^ (OH) ' CH2 ' CO ' CH3 = o-Xitrobenxaldehyde Acetone 2H 2 + 2CH 3 -C0 2 H Indigo In printing, the synthesis takes place directly on the textile, the acetone being rendered soluble by conversion into the bisulphite compound (Kalle's salt). The industrial prepara- tion of o-nitrobenzaldehyde presented, however, a serious disadvantage, the direct nitration of benzaldehyde yielding a considerable proportion of the unusable m-nitrobenzaldehyde ; while, starting from benzil, the p-nitro-compound is obtained. A happy solution of this difficulty was found in the preparation of o-nitro toluene directly from toluene (only 40 per cent, of p-nitrotoluene is formed), oxidation with manganese dioxide and su'phuric acid then giving a good yield of o-nitrobenzaldehyde. To the general application of thn process were opposed a number of difficulties. In order that the artificial indigo might displace the natural product, the annual consumption of which was about 5,000,000 to 6,000,000 of kilos (100 per cent.), it was necessary that there should be on the market a sufficient quantity of raw material (toluene) at a reasonable price. It was found that, even although the use of modern metallurgical coke furnaces (see vol. i, p. 366, and this vol., p. 530) increased the quantity of crude benzene (in 1900 the total output in Europe amounted to 30,000 tons), yet, since the latter contains only one-sixth of its weight of toluene and since 4 kilos of toluene are required to furnish 1 kilo of artificial indigo, the use of all the toluene extractable from the benzene on the market would give only 1,000,000 kilos of indigo, i.e. one-fifth or one-sixth of the whole consumption. Increase of the production of crude benzene for the purpose of obtaining more toluene would lead to over-production of unusable benzene, and hence to increase in the price of toluene and hence in that of artificial indigo, which would be unable to compete with the natural product. After much further investigation and many unsuccessful trials, the industrial prepara- tion of artificial indigo has, however, become an accomplished fact. Having acquired Baeyer's patents for a sum approaching 20,000 without deriving any practical benefit from them, the Badische Anilin- und Soda-Fabrik of Ludwigshafen did not hesitate to purchase later the patents of K. Heumann, who was the first to discover, in 1890, that indigo is obtained on fusion of phenylglycocoll with caustic potash, but that a better yield is obtained if the phenylglycocoll is replaced by phenylglycine-o-carboxylic acid, C 6 H 4 (CO 2 H)(NH-CH 2 -CO 2 H). The economical preparation of this acid necessitated investigations and trials extending over more than seven years, and the synthesis became of industrial value only when it was found possible to employ naphthalene as the initial substance. Quite 50,000 tons of naphthalene are produced annually in the distillation of tar, and up to that time only about 15,000 tons of this had been utilised, the rest being left in the heavy tar-oils or used for making lamp-black (p. 528). The complete synthesis takes place in the following stages : Naphthalene CO. \\r -CO/ Phthalimide NH C0 2 H + NH 2 Anthranilic acid HOI -!- C0 2 H , ,-NH-CH 2 'C0 2 H PhenylglycJne-0-ca.rboxylie acid 644 /N Vrvym / \-rvn~m. / \-r,n . .m./\ C(OH) V., NH ' C(OH) CH NH ' NH/ \NH Indoxylic acid Indoxyl Indigo The oxidation of naphthalene to phthalic anhydride by means of chromic acid is too expensive, but the same end was attained by the use of fuming sulphuric acid rich in sulphur trioxide, after it had become possible to prepare this cheaply by the catalytic method (see vol. i). The action of the acid was moderated with mercury bisulphate, while the sulphur dioxide was recovered by the catalytic process (in 1901 the Badische Company recovered in this way, for the manufacture of phthalic anhydride alone, about 40,000 tons of sulphur dioxide). Phthalimide is then easily obtained by the action of ammonia, while the monochloro- acetic acid can be prepared cheaply and in large quantity by using the liquid chlorine (1,000,000 kilos in 1900) resulting from the electrolytic manufacture of caustic soda or potash and glacial acetic acid (about 20,000 quintals obtained per annum from the distilla- tion of 100,000 cu. metres of wood). The reaction between anthranilic acid and mono- chloroacetic acid proceeds readily, but the formation of indoxylic acid was found to be much more difficult, the conditions required for the fusion of the phenylglycinecarboxylic acid being inconvenient ; this obstacle was, however, finally overcome. The ultimate oxidation of the indoxyl is effected by means of a current of air. The indigo separates in small crystals, and in order to obtain it in a finely divided state, it is converted into sulphate and this decomposed with water. After being washed, the paste thus formed is identical with natural indigo and is, indeed, of greater value owing to its higher purity and to its constancy of composition. Process of the Farbwerke vormals Meister, .Lucius und Bruning (of Hdchst). This consists in the action of sodamide (obtained by treating gaseous ammonia with sodium) on phenylglycocoll, subsequently heating in an autoclave at 250 : NH 2 Na + C 6 H 5 -NH-CH 2 -C0 2 Na = NH 3 + Na 2 O + C 6 H 4 <>CH 2 (indoxyl), 2 mols. of the indoxyl then condensing in presence of oxygen : + 2 = 2H 2 + This process was originally patented by the Deutsche Gold- und Silber-Scheide Anstalt (Frankfort), from whom it was purchased. A yield as high as 65 per cent, has been obtained, but sodium at 28d. per kilo is too expensive to make the process practicable. Sandmeyer's synthesis (patented by Messrs. Geigy of Basle ; Eng. Pat. 15,497 of 1899). Aniline is treated with carbon disulphide in presence of alcoholic potash, diphenylthio- urea being obtained: CS 2 + KOH + 2C 6 H 5 -NH 2 = KHS + CS(NH-C 6 H 5 ) 2 + H 2 0. The action of lead cyanide on diphenylthiourea gives Hydrocyanocarbodiphenyl- C 6 H 5 - N x imide, /C.CN, which with ammonium sulphide yields the Thioamide, C 6 H 5 -NH/ C 6 H 5 - N ^ NH 2 yC C/' , and this with sulphuric acid forms a-Isatinanilide, CH-NH/ ^S C-NH-C 6 H 5 , CO/ 4 \ reduction of the latter by means of ammonium sulphide then giving indigo. All the materials used in this synthesis are cheap, but the indigo produced was not able to compete for long with that of the Badische Company and of Messrs. Meister, Lucius und Bruning, who continually lowered the price in order to suppress natural indigo and made use of the two improved Heumann processes starting from phenylglycocoll and phenylglycinecarboxylic acid. The struggle, lasting for more than twenty years, between the producers of natural indigo and the scientific men connected with the various industrial undertakings has now ended in uncontested victory for the latter. The figures already given showing the areas under indigo at different times (see p. 639) justify the conviction that in a few years time Indigofera tinctoria will be of interest only historically, just as is the case with madder, now supplanted by artificial alizarin. With its lower price, its more ready applicability in dyeing, and the considerable use now made of its halogenated derivatives, the consumption of indigo will certainly increase. In 1908, owing to the slight difficulty of reducing indigo, even when finely powdered, several firms placed on the market the leuco-product itself (indigo white), this being obtained by reduction with iron and alkali, or, better, with hydmsnlp}u.te(Grandmougin), &c. The following figures will give a clearer idea of the commercial and industrial importance of indigo, both natural and artificial. Statistics. The production in India was 50,000 quintals in 1892 and 75,000 quintals (containing 56 to 70 per cent, of indigotin), of the value of 3,200,000, in 1896, while in 1909 it was only 12,000 and in 1910 9000 quintals (240,000). Of Indian indigo 60 per cent. is sold at Calcutta, which supplies Europe and America, 30 per cent, at Madras to Egypt and the East, and 10 per cent, at Bombay and Karachi. In 1882 the Indian Government abolished the export duty on indigo. Until 1865 almost all the indigo was sent to London, which was the centre of the European trade. In 1905-1906 exportation from India had fallen to 15,000 quintals (400,000), the cultivation of indigo being replaced by that of rubber (28,000 quintals), turmeric (25,000 quintals), hemp, cotton, tanning plants, &c During recent years the cultivation of natural indigo has increased in the districts more suitable to it and diminished in those less fitted. The amount of indigo produced in British India in 1911 was 6 per cent, in advance of that of the preceding year, although the area under cultivation was 2 per cent. less. England imported : f 1909 500 tons of the value of 139,335 Natural indigo J 1910 167 . 43,054 [1911 245 . 67,430 I 1909 1670 . 117,100 Artificial indigo -! 1910 1450 . 101,249 [ 1911 1215 . 85,143 The United States imported 3100 tons of natural and artificial indigo of the value of 229,800 in 1910 and 3400 tons of the value of 224,600 in 1911. In 1854 the Philippines exported 194,727 kilos of indigo paste (17,445^ and liquid indigo (tintarron) (5470), while in 1866 the amounts were 251,574 kilos of indigo paste (96,950) and 959,206 kilos of liquid indigo (28,180). The industry was still flourishing in 1875-1881, when the producers began to adulterate with sand and other substances ; prices were thus ruined and fell from 12 per quintal to 4, the cultivation being to some extent abandoned. With careful cultivation, as much as 4 quintals of good indigo can be obtained per hectare. By 1905 the exportation had diminished to a total of 250,000 ki'los of pasty and liquid indigo. The output in Java amounted to 547,000 kilos in 1904 and to 500,000 in 1905, but in 1908 the exports were only 105,000 and in 1909 100,000 kilos. In 1895 the consumption of indigo in different countries was as follows : England, 13,000 quintals ; United States, 11,500 ; Germany, 10,000 ; France, 7100 ; Belgium, 1500 ; Austro-Hungary, 5500. In 1911 the world's consumption of indigo (calculated for 100 per cent.) was estimated to be about 60,000 quintals, but is possibly higher than this, the amounts used in China and some other countries not being known exactly. In 1900 the Badische Anilin- und Soda-Fabrik produced 10,000 quintals of artificial indigo, which corresponds with the output from 104,000 hectares. Italy imported the following quantities of natural and artificial indigo : 1903 1906 1907 1908 1909 1910 Natural, quintals -. ; 5564 1419 972 944 910 474 (13,270) Artificial, quintals ;' " . 2956 3028 3474 4243 .5164 (72,300) The quantities of artificial indigo (20 per cent.) exported from Germany in 1900 (and in 1905), in quintals, were as follows : to England 1668 (15,612) ; to France, 1000 (1350) ; 646 ORGANIC CHEMISTRY to Austria-Hungary, 3773 (11,407); to Russia, 950 (3160); to Italy, 1078 (3200, worth 76,800; besides 2160 quintals of natural indigo of the value of 52,000); to Belgium, 385 (2346) ; to Switzerland, 595 (819) ; to the United States, 4926 (25,357) ; China, 1 189 (26,000) ; and Japan, 174 (7000). In 1907 the total production of artificial indigo was about 43,200 quintals (of 100 per cent.), i.e. four-fifths of the world's consumption. In the same year Germany exported artificial indigo to the value of 2,000,000 (in 1910 2,160,000) and imported natural indigo worth 60,000 ; in 1908 the exports were 154,560 quintals, and in 1910 about 161 quintals. The price of natural indigo reached its maximum of 22s. per kilo in 1870, at which time aniline dyes came into competition with it. The price of artificial indigo (calculated to 100 per cent.) in 1897 was 15s. to 16*. per kilo, a corresponding amount of the natural product costing 16*. to 18s. In 1900 natural indigo cost 12*., while in 1905 artificial indigo was sold at one-half the price of the natural dye, i.e. at about 1*. Id. per kilo of 20 per cent, strength. The first artificial indigo plant of the Badische Anilin- und Soda-Fabrik in 1897 cost 480,000, and in 1900 two competitors, namely, Messrs. Meister, Lucius und Briining and Messrs. Geigy, made their appearance, the considerable fall in price thus produced resulting in Messrs. Geigy's abandonment of the manufacture and of the fusion of the indigo interests of the two remaining firms with a capital of 1,200,000. In 1910 the manufacture of artificial indigo was started by the Rahtjen Company of Hamburg which is a company with a capital of 280,000 and makes use of Rahtjen's improved Sandmeyer process and by the firm of Heyden (Radebeuf), which employs the phenyl- glycine method. The Society of Chemical Industry in Basle also began making artificial indigo in 1911-1912. R. COLOURING-MATTERS Only a certain proportion of the innumerable coloured substances are capable of being fixed on vegetable or animal fibres, imparting to them a more or less stable coloration, and only those able to fulfil this function, directly or indirectly, belong to the true Colouring-Matters. Coloured substances are those which absorb constituents of white light of certain definite wave-lengths, emitting the rest. Generally speaking, only the luminous waves visible to the eye have yet been closely studied, and it is probable that new laws, possibly more important than those already known, will be discovered when the infra-red and ultra- violet rays absorbed or reflected by coloured substances are considered. Hartley has indeed shown that the apparently colourless substance, benzene, is, strictly speaking, coloured, as it absorbs certain ultra-violet rays invisible to the eye, and that in the benzene series the luminous vibrations are gradually rendered slower and so made visible as the molecular weight is increased by substituent groups. Dichroic Substances allow certain rays to traverse them and reflect certain others, so that they appear to be of one colour by transmitted, and of another by reflected, light ; such are, for example, fluorescent substances. Certain alkaline fluorides, such as those of the alkali metals, allow infra-red and ultra- violet radiations to pass through them, while various nitrates, nitric acid, the hydrocarbons, the aldehydes, &c., although they do not retain any of the constituents of white Mght and hence appear colourless, yet do absorb waves of many wave-lengths. Light itself is to the human organism only a sensation due to absorption of a portion of the radiations by the crystalline lens of the eye. Between coloured and non-coloured substances there is often complete or nearly complete identity in chemical composition, so that the colour depends, not on the composition, but only on the constitution or atomic structure of the molecule. It is now universally admitted that the colour of substances is closely CHROMOGENS 647' dependent on the presence in the molecule of certain well-defined atomic groupings or nuclei. The various organic substances of the benzene series which form coloured or colouring matters always contain these groups (chromophores, see below), some of which are univalent and quite simple, e.g. N0 2 , X N=N , X CO . But of more importance are the divalent groups formed of a benzene nucleus of the constitution >>C<< pTT where X 2 may be 0, NH, NR, CR 2 , while the other two valencies in the para-position may be satisfied by 0, NH, N, R 2 , Cl. As early as 1867 Graebe and Liebermann arrived at the conclusion that the colouring-matters capable of fixing hydrogen with decoloration and forma- tion of the so-called leuco-bases (see p. 607), are transformed into coloured substances on oxidation. 1 In 1876 N. 0. Witt defined the nature of these simple groups, which are always contained in the more complex benzene groups characteristic of the colouring-matters, terming the former chromophores and the latter chromogens? 1 Some of these leuco-products regenerate the original colouring-matter simply by oxidation, while others do not. For instance, reduction of nitro-groups gives, as final products, amino-derivatives, which yield nitro-groups again on oxidation. The complete reduction of azo-compounds yields amino-groups, but there may also be intermediate, less highly reduced products (hydrazo-compounds), which are themselves new leuco-derivatives. 2 Examples of chromogens are : >=N-C,H 4 -NH 2 ; Indamine / -N-C 8 H 4 -NH 2 ; =C=(C,H 4 -NH,) S Indophenol Rosaniline Rosolic acid CO CO Anthraquinone NH Thiodiphenylamine NH, OH N Thionine In these chromogens is seen the analogy between the chromophores in the different molecules, characterised by divalent or polyvalent atoms or atomic groups (=NH,=N' , =C=O, S, O , ^>CO) united to the ring in a closed chain, the whole forming the true chromophore, which, joined to the rest of the molecule, gives the chromogen. The passage from simple to more complex chromophores is often accompanied by change from a yellow colour to a more intense yellow or to red or blue. Resorufln ORGANIC CHEMISTRY The latter are colourless or slightly coloured but approximate to the colouring- matters in chemical composition. Jt is, indeed, sufficient to introduce into the chromogen, in place of hydrogen, a salt-forming basic or acid radical (OH, SO 3 H, C0 2 H, NH 2 ), to produce the colouring-matter. Thus, nitrobenzene, C 6 H 5 'N0 2 , is a chromogen which becomes a true colouring-matter in nitro- phenol, C 6 H 4 (OH)'N0 2 , and in nitraniline (phenol itself, C 6 H 5 'OH, is colourless). The intensity of the coloration increases with the number of these acid or basic groups ; thus, Aminoazobenzene, C^H^Na-NH^, is pale yellow, Diaminoazobenzene, C 12 H 8 N 2 (NH 2 ) 2 , is orange, and Triaminoazobenzene, C 12 H 7 N 2 (NH 2 ) 3 , is brown. Such regularities often occur with artificial colouring-matters, so that the colour of a new compound of a certain constitution can be foretold before the compound is prepared. 1 In the light of the above definition it would be difficult to understand how colouring-matters could be formed of hydrocarbons alone, since these contain none of the characteristic chromophores just mentioned. The few hydrocarbon colouring-matters were for some time regarded as exceptions, but it was found later that they contain a characteristic complex chromophore, different from those previously known and with molecular weight higher than a certain limit. The following two chromophores, for example, are well defined : | \~> and X X / Further, what are usually the more energetic chromophores cease to be so when they occur in molecules which are small or poor in carbon. To this is due the very small number of colouring-matters in the aliphatic series. Thirteen chromophores of well-defined constitution are now known, while concerning others there is still doubt owing to the pseudoisomerism (tautomerism) they exhibit. 2 1 With fuchsines (rosanilines and p-rosanilines) the colour becomes more intense and more violet with increase in the alkyl groups replacing the aminic hydrogen. The faintly acid, ph<*iolic colouring-matters which are fixed by mordants give highly resistant colours if they contain at least two OH groups, or OH and COOH, in the ortho- position, and better still if these are also in ortho-positions with respect to the chromophores. In the colouring- matters of the nitrophenol group, the colour passes from greenish yellow to orange-yellow as the distance between the OH and NOj groups increases. Fast colours on mordants are given especially by those colouring-matters containing hydroxyl-groups in the ortho-position with respect to one another and to the chromophore (alizarin, &c.). Of the triphenylmethane colouring-matters, those which have a sulphonic group (SO 8 H) in the ortho- position with respect to the central carbon atom are stable to alkali and to soap (Suais and Sandmeyer). 8 According to Hantzsch (1906) all the true nitro-hydrocarbons of the aromatic or aliphatic series and also all polynitro-compounds are colourless when quite pure, so that the NO 2 group by itself is never a chromophore. Only certain nitrophenols are coloured when their phenolic hydrogen is free and hence mobile (forming tautomeric com- pounds) and for the same reason all salts of nitrophenols are coloured. By the discovery of the quinonic (aci-) ethers of nitrophenols besides the true ethers, it was shown that many colourless or almost colourless hydrogenated compounds capable of forming highly hydrogenated salts, are pseudo-acids, so that the coloured salts are derived from a hydrogenated compound differing from the original ; if it were possible to obtain these free, they also would be coloured. Nitrophenols are certainly true tautomeric hydrogenated compounds which give two series of structurally isomeric ethers, such as are given also by nitrous, sulphurous, hydrocyanic, and cyanic acids. The true nitro- phenolic ethers are colourless, while the aci-ethers (tautomeric) are coloured an intense red ; the former correspond with the general formula, C t H 4 : O *"+' (derived from the colourless true nitrophenol, C,H, -OH ), and the hitter with C.H^ [derivatives of aci-nitrophenol (quinonic), C,H 4 ^ ]. It is ^NO'OC ; ,H t ,, + t ^\NO'OH hence possible to tell, from the mere colour, to which of the two groups a given nitro-compound belongs. When true nitrophenols (even in the solid state) are slightly coloured, it is assumed that a minimal quantity of Ihe aci-nitrophenol is dissolved in a large quantity of true nitrophenol (solid solution). Also, the fact that the colour of the substance is sometimes not intensified by increase in the number of nitro-groups is explainable, not on the old view of tht theory of chromophores, but only by the new theory of transposition (tautomerism) : ,on ^ ,o C.FT./ C,H 4 ^ \NO'OH SALT-FORMING GROUPS 640 In practical dyeing it is of interest to know, not only that a substance has colouring properties but to what chemical conditions or groups these properties are due. Especially is this the case with animal and vegetable textile fibres (see Theory of Dyeing, p. 708). Chromophores are generally of basic (electropositive) or acid character (electro-negative, e.g. the quinonoid group, &c.) and when they form coloured substances do not retain their colouring properties ; the latter are, however, manifested if the basic or acid character is reinforced or even inverted by means of salt-forming groups. The acid groups (S0 3 H, C0 2 H, &c.) have, however, a slight influence on the colour. Thus, azobenzene, C 6 H 5 'N : N'C 6 H 6 , although a coloured chromogen (containing the chromophore -N : N-) does not colour textile fibres since it is neutral, while its sulphonic derivative is a feeble colouring- matter. The basic groups (especially NH 2 and, in some cases, OH, &c.), on the other hand, exert considerable influence on the colour, and Witt calls them auxochrome groups to distinguish them from the acid groups, which he terms salt-forming groups. 1 The tendency to tautomeric transposition may, indeed, be increased or diminished by the entry of new groups. Thus, in solutions of nitrophenols and their salts, the coloration is not as it would be according to the modern theory of indicators (see vol. i, p. 97) due to ionisation, but rather to the formation of coloured tautomeric compounds (aci-nitrophenolic ethers) in agreement with the old chemical theory of indicators. It is thus proved that the formation of coloured salts and coloured ions derived from colourless hydrogenated compounds is of a purely chemical nature. It is caused first of all by intramolecular transposition, from which, by the action of a positive metal (salt), there results a negative quinonic atomic grouping (chromophore), the appearance of coloured ions being a secondary reaction. Hence the actions of chromophore and of auxochrome cannot be held to be distinct but are exerted together, both of them (nitro- and phenol-group) causing the appearance of colour at the expense of their mutual transformation, which generates a quinonic grouping containing neither nitro- notfphenolic group. These views may be extended to other groups of organic substances since, in general, colourless acids unchange- able in constitution (i.e. not giving tautomeric forms) give only colourless ions and yield colourless salts with colour- less metallic oxides, and colourless ethers and esters with colourless organic radicals (alkyl and acyl). If -oloured ions and salts are derived from a colourless alkyl compound, it may be stated with certainty that intramolecular change occurs. According to E. Fischer and 0. Fischer (1900) many colouring-matters derive their properties from the presence in the molecule of quinonoid groups, although A. v. Baeyer (1902-1905) and Hantzsch (1905) showed that the true quinone group does not always cause coloration (i.e. is not the chromophore), and Kostanecki and Haller pointed out that, in addition to the two carbonyl groups of the quinone, two ethylene double linkings must be present, CO CO HOf ;CH H,C' > ; in fact, diketohexamethylene, CH, , which has not these double bonds, is completely CH, HtH "CH H,C CO CO colourless. As a metaquinone with two ethylene double linkings cannot exist such meta-compounds are incapable of producing colouring-matters. CH CH cm ' n -a t ^-HLj / \ / ' ' i NH ' Rosaniline, HN = C\ >C =- C( l * (see corresponding base, p. 608), which is coloured \ / XC.H^NH, CH CH CH=CH contains as chromophore the group HN = C/ \r>^-' and as auxochfomes two ammo-groups. When the \ /<- CH=CH salt with a single molecule of HC1 is obtained and the substance is dyed red, proof is given that the salt is formed with the imino-group of the chromophore, since a red coloration is formed on the fibres. On the other hand, salts of rosaniline with two or three molecules of HC1, which form salts also with the auxochrome amino-groups, are yellow but do not dye textile fibres yellow. It can hence be affirmed that the auxochromes do not unite with the fibres and hence have no action as salt-forming groups but only contribute to increase the basic character of the colouring-matter or even to increase the intensity of the colour ; this is clearly shown with safranine (see later and also above, Aminoazobenzene). In general, the union of an acid chromophore with a basic auxochrome gives colouring-matters of slight intensity ; for instance, the nitroanilines are feeble and the nitrophenols more intense colouring-matters. The replacement of the hydrogen of the auxochrome OH by a metal increases the power of the auxochrome, while an alkyl or aromatic radical lowers it and an acid radical often annuls it. Substitution of the hydrogen of the auxochrome NHj by alkyl radicals raises the colouring power, while two aromatic radicals sometimes lower it considerably, exceptions to this being shown by sulphonal and picryl, C,Hj(NO,)j, which cause the NH, group to assume an acid character. The hydrazinic and hydroxylaminic groups also behave as auxochromes ; thus phenyl- hydrazine is slightly yellow while aniline is colourless, and nitrophenylhydrazine is more highly coloured than nitroaniline. Anthraquiuone (faintly acid chromogen) gives an intensely coloured derivative with hydrazine ORGANIC CHEMISTRY The acid or basic character of a colouring-matter decides also its behaviour towards different textile fibres vegetable and animal and in general acid colour ing -matters contain the group S0 3 H or COOH, the feebly acid ones the group OH, and the basic ones the groups NH 2 , NHR, NR 2 , either as chromophore or as auxochrome. If the auxochrome of a colouring-matter is weak and the chromophore strong, or vice versa, the colouring-matter is generally feeble. For dyeing purposes the colouring-matters are placed on the market in a state soluble in water, the auxochrome groups being converted where possible into salts (e.g. S0 3 Na, &c.). When wool (which is both basic and acid in cha- racter) is dyed with acid colours, since the basic properties of the wool are usually not sufficiently strong to displace the metal (Na) of the acid colour, an energetic acid (acetic or sulphuric) is added to the hot aqueous dyeing bath, this liberating the acid residue of the colouring-matter, which can then combine with the basic group of the wool to form a coloured stable insoluble salt in the fibre itself. Thus wool is dyed directly both by acid and by basic colours (with the latter it is not necessary to render the bath acid). Cotton, on the other hand, is not usually dyed by acid dyes but only by basic ones, and then only when the fibres are previously mordanted with tannin materials and metallic salts. During the past twenty years, however, numerous neutral or substantive dyestuffs have been discovered, capable of dyeing cotton directly in a neutral or faintly alkaline, but not acid, bath, previous mordanting being unnecessary. Many of these colouring-matters have a common benzidine group (see p. 605), others contain a basic group (primulin) and others again a phenolic group (curcumin). Colouring-matters sometimes acquire this property by mere accumulation of chromophores in a single molecule (Rupe, 1901 ). The nature of the metal present in these colouring-matters alters to some extent the properties and the affinity towards cotton, but this is always related to the capillary constant of the aqueous solution. The precipitation of the unaltered colouring- matter on the fibres is facilitated by increasing the osmotic pressure of the bath by the addition of considerable quantities of salts (NaCl or Na 2 S0 4 ). As a rule phenolic compounds form weak colouring-matters, but they have the property of giving intensely coloured lakes with metals (phenoxides), the metallic atom united to the phenolic oxygen functioning as an energetic auxochrome. These colouring-matters having no affinity for textile fibres, the latter are previously charged with metallic oxides (mordants). Lakes of different colours are formed with different metals (Hummel hence called such colouring-matters polygenetic), but for practical purposes it is indispensable that they should be resistant to atmospheric agents and to ordinary physico- chemical treatment. 1 The best among these substances are those containing in the ortho-position either two phenolic groups (OH) or one OH and one COOH, and of such those are best in which these groups are in the ortho- position with respect to the chromophore (Liebermann and Kostanecki, groups. The hydroxylamine derivatives are few in number and have been but little studied. H. Kaufmann (1911) has shown that two auxochromes reinforce one another when they are in the para-position and to a less extent or not at all when they are in the ortho- or meta-position. This rule is confirmed, not only by the greater intensity of the colour, but also by the increased luminescence or fluorescence assumed by these substances when they are exposed to ultra-violet rays (see vol. i, p. 121) ; in solution, only compounds of the para- series give direct fluorescence. By the law of distribution it is proved that the maximum and sometimes the only effect of auxo- chromes in the para-position is exerted when the chromophore and auxochrome are in the same benzene nucleus. 1 In addition to what has been already stated with reference to the application of lakes, it may be said that they are derived from acid or basic colouiing-matters, coloured pigments or colouring-matters of the anthra- quinone group. The soluble acid colouring-matters are precipitated by salts of calcium, barium, strontium, alu- minium (chlorides), magnesium (sulphate), &c. Solutions of basic dyes are precipitated by tannin, Turkey red oil, resin, or, more commonly, sodium phosphate or sodium arsenate. Anthraquinone dyes (alizarin, coerulein, &c.) form lakes with greater difficulty, and it is necessary to observe rigorously the proper temperature conditions. In pieparing lakes, great importance attaches also to the substance on which the precipitated lake is deposited or with which it is mixed (aluminium hydroxide, barytes, zinc or lead white, ferric oxide, fresh aluminium silicate, &c.), and of these, the ones more easily decomposable by dilute acids retain the colour best. Lake-formation is hence not a simple absorption phenomenon but also a chemical phenomenon. DYES IN RELATION TO FIBRES 1887-1893) and in which the auxochrome is formed from iron, aluminium, or chromium. Not all colouring-matters which give insoluble lakes can be fixed on fibres mordanted with metallic oxides, and this perhaps depends on the fact that only certain coloured lakes are capable of combining with the fibre, the constitution of the colouring-matter (see Alizarin) being here also of considerable importance. When basic or neutral colouring-matters are sulphonated with concentrated H 2 S0 4 , acid colouring -matters (Simpson and Nicholson, 1862) are often obtained. In the form of soluble salts of the alkali metals, these can be fixed directly, in an acid bath, on animal fibres with the same colour as the colouring-matter, the animal fibre forming a kind of new salt ; indeed the fibre assumes the colour of the original salt of the colouring-matter and never that of its free coloured acid liberated in the bath by means of acetic or sulphuric acid (see above, Process of Dyeing). These acid colours are fixed also by cotton, provided the latter is first rendered basic either by nitrating and then reducing, or by oxidising (oxycellulose), or by hydrating (with NaOH : mercerisation), or by treating with NH 3 under pressure in presence of ZnCl 2 (Vignon). Basic colouring -matters which owe their basicity to the chromophore and more especially to the auxochrome NH 2 , form salts with acids and are used in practice in the form of hydrochlorides, sulphates, &c., from . hot acidified aqueous solutions of which wool and silk fix the coloured base. These basic dyes also form insoluble salts with tannin, &c., so that they are capable of dyeing cotton - which has no affinity for basic dyes if this is previously mordanted by prolonged immersion in cold solutions of tannin extracts (sumac, &c.), followed by fixation of the tannin in another bath containing an antimony, aluminium or iron salt, or gelatine. In the subsequent dyeing-bath, the dye is fixed rapidly, even in the cold (the fixation is more regular, i.e. slower, in presence of a little alum). The full (intense), bright colours thus obtained on cotton resist the different reagents well but are destroyed during washing by the rubbing. In 1901 C. Favre suggested the use of resorcinol and formaldehyde as mordants in place of tannin. 1 Many colouring-matters exert a poisonous 1 Behaviour of Colouring-Matters towards different Fibres and Mordants, according to Noelting. If a skein of wool, silk, or cotton is immersed for some hours in a solution of a basic ferric salt, the fibres assume a brown colour, having fixed a certain amount of ferric oxide or basic salt. The same holds generally for all salts of oxides corresponding with the formula R 2 O,. The salts of protoxides (RO), e.g. those of copper, iron, manganese, nickel, cobalt, &c., especially the tartrates or in presence of tartar, are fixed by wool or silk, but not at all or but slightly by vegetable fibres. Not only metallic salts, but also certain organic substances (tannin materials) and salts of hydroxyoleic and hydroxystearic (sulpho-oleates) acids, can be fixed by fibres. A large number of colouring-matters are fixed directly on animal fibres in a neutral or acid bath, more rarely in an alkaline bath. To this group belong the nitro-derivatives of the phenols and amines : the azo, basic, and acid dyes ; basic, acid, or sulphonated derivatives of triphenylmethane ; certain phthaleins (fluorescein and eosin) ; the aminophenazines, safranines, thioindamines, phenoxazine derivatives (gallocyanine and Meldola's blue), phenylacridine complexes (phosphine), quinoline complexes (cyanine, quinoline red, quinophthalone), hydrazides, osazones (tartrazine), ketonimides (auramine) and, among the natural colours, indigo-carmine, berberine, safHower, saffron, archil, and catechu. Almost all of these dyes are fixed in minimal quantity or not at all on vegetable fibres. Those which are fixed by the latter are less numerous and include : a first group of substances which aie fixed only with difficulty (better with tannin), e.g. certain aminoazo-compounds, phenylene brown, chrysoidine, methylene blue, Victoria blue, safranine ; a second group fixed stably and directly and consisting of numerous azo-derivatives of beoeidine, tolidine, diaminostilbene, p-phenylenediamine, naphthylenediamine, diamino- azobenzene, diaminoazoxybenzene and its homologues, diaminodiphenylamine, canarine (oxidation product of thiocyanates), and the sulphur dyes of Croissant and Bretonnierc ; a third group which do not dye wool, cotton, or silk directly but give bright fast colours if these fibres (especially with wool) are previously mordanted with salts of iron, aluminium, or chromium : such are certain phthaleins (gallein), derivatives of anthraquinone (alizarin, purpurin, alizarin orange, anthragallol), anthraquinoline (alizarin blue), phenoxyanthranol (ccerulein), and almost all the natural colouring-matters (logwood, cochineal, quercitron, cudbear, sandalwood, &c.). Noelting gave the name substantive dyes to those which dye animal and vegetable fibres directly and that of adjective dyes to those which dye the fibres only after mordanting. Certain dyestuffs are fixed directly by wool and silk and only indirectly by cotton, i.e. when the latter has been mordanted. Such are gallocyanine and various carboxylic acids of azo-compounds. In dyeing with aniline black, the fibre fixes both the aniline salt and also the oxidising agent, the latter oxidising the aniline on the fibre with formation of an insoluble aniline black. Dyes which are not fixed directly by cotton, dye it only after mordanting with tannin or sulpholeic acids if they are basic in character, or after mordanting with metallic oxides, with or with- out sulpholeates, if they are acid. Further, various substantive colouring-matters have the property of fixing others on them ; for instance, chrys. amine and canarine, which are yellow, fix basic colouring-matters, such as fuchsine forming an orange, malachit 652 ORGANIC CHEMISTRY action oil micro-organisms, as they unite with the protoplasm, and even, in dilute solution, cause death (Th. Bokorny, 1906). In recent years attempts have been made, but without practical success, to utilise the colouring-matters produced by certain chromogenic bacteria, e.g. B. prodigiosus) . MANUFACTURE OF COLOURING-MATTERS Since 1856-1860, when Perkin in England made mauveine and Renard and Frank in France made fuchsine on an industrial scale, scientific progress in colouring-matters has advanced pari passu with the industrial development. In the history of the artificial colouring- matters, side by side with the names of the scientific men, such as Perkin, Williams, A. W. Hofmann, Graebe, Liebermann, Baeyer, Witt, Nietzki, Noelting, Caro, &c., who laid the first stones in this marvellous chemical edifice, are those, not less worthy, of the brilliant and daring industrial workers who, by uninterrupted energy and the application of ingenious processes, carried these theoretical discoveries into the larger field of industry and commerce. 1 green forming a yellowish green, and methylene blue forming a blue colour. All the benzidine colours have the same property, to .which Noelting gives the name secondary dyeing, a term applicable also to all dyeing with mordants . Direct dyeing would then be primary dyeing. In some cases a third colouring-matter can be superposed ; for instance, the violet lake of alizarin and iron combines with methyl violet giving a brilliant triple lake. The red lake of alizarin, alumina, and lime, which is not very bright and rather opaque, is rendered brilliant and more fast by the fixation of a sulpholeate, which forms a quadruple lake ; finally, this can still fix tin from a soapy solution of tin salt, a new lake with five components being formed. If a tissue removed from a solution of a basic iron salt, instead of being washed immediately (in which case it becomes yellowish), is treated directly with alkali or soap (or with a solution of a salt the acid residue of which forms an insoluble compound with oxide of iron), it becomes much more intensely coloured and the quantity of iron fixed by the fibres is considerably increased. Oxide of iron can be accumulated on the fibre, not only, as just mentioned, from an alkaline bath, but also by impregnating the fibre itself with ferrous salts of volatile acids, e.g. the acetate, and then exposing it in the moist state to the air. The ferrous salt is thus converted into basic ferric salt, this in warm, moist air losing part of its acid and undergoing change into an insoluble, highly basic salt, which is not removed from the fibre even by repeated washing. In order to help the action of the air and render a larger quantity of basic salt insoluble, the fibre may be passed into a bath of cow-dung or lime and potassium silicate, phosphate, or arsenate. Aluminium salts are similarly rendered insoluble by formation of a basic salt. The basic chromium salt is fixed by a subsequent bath of sodium carbonate or, better still, by impregnating the tissue with a solution of chromium sesquioxide in caustic soda and exposing it,to the air, the caustic alkali being thus converted into carbonate, which precipitates the sesquioxide of chromium ; instead of exposure to the air, the action of steam may be employed. Chromous oxide is precipi- tated by simple washing of the impregnated tissue with a tin salt. Sulphoricinate is fixed by solutions of aluminium salts and tannin by solutions of tartar emetic or ferric or aluminium salts. The action of a chromate bath or catechu is twofold ; first, the catechu undergoes oxidation with considerable darkening, and then combination takes place between the oxidation product and the chromium sesquioxide resulting from the reduction of the chromate. 1 At first France was at the head of the aniline dye industry, with numerous pioneers, such as Verguin, Renard Brothers, Frank, Poirrier, Guinon Mamas and Bonnet, Coupler, Girard and de Laire, Baubigny, Persoz, Bardy, Lauth, Kopp, Rosenstiel, Roussin, R-NC1 : N( + R'-OH, phenol) > HC1+R-N : N-R'-OH. The diazo-group enters in the para-position to H, OH, or NH 2 , or if this is occupied, in the ortho-position. Azo- colouring-matters are so numerous and so varied in constitution and behaviour that they may be divided into several sub-groups. The MONOAZO-COMPOUNDS may be sulphonated (aminoazo-derivatives give basic colouring- matters and the hydroxyazo-derivatives without carboxyl, acid colouring- matters) or not sulphonated (the aminoazo- compounds give basic and acid colouring- matters and the hydroxyazo- compounds basic and adjective colouring-matters). POLY- AZO-COMPOUNDS yield substantive and adjective dyestuffs (i.e. without benzidine nuclei and then form acid, basic, and mordant colouring- matters). Finally theic is the sub-group, the members of which are generated directly on the cotton fibre. (a) Aminoazo-derivatives. These are obtained in the usual way, in the cold and in alkaline solution, from diazo-compounds (amino- or not) and amines. Among these are fast yellow, acid yellow, tropceolin, the oranges, Indian yellow (nitro- derivative of phenylaminoazobenzenesulphonic acid), orange IV or tropceolin 00 (sodium salt of the non-nitrated product, S0 3 H-C 6 H 4 -N : N-C 6 H 4 -NH-C 6 H 5 ) and vesuvine or Bismarck brown, which is the hydrochloride of triaminoazobenzene, NH 3 -C 6 H 4 -N : N-C 6 H 3 (NH 2 ) 2 , mixed with C 6 H 4 [-N : N-C 6 H 3 (NH 2 ) 2 ] 2 . Indoin is a basic blue obtained by coupling diazotiscd safraninc with /3-naphthol. On textiles they are not very fast to light, the less fast being those which do not contain the sulphonic group. In printing textiles these colours are corroded by the stannous chloride. (b) Hydroxyazo-derivatives (or azoxy-compounds), e.g. hydroxyazobenzcne, C 6 H 6 'N : N'C 6 H4'OH. Tropceolin is a dihydroxyazobenzenesulphonic acid. N:N- Of greatest importance are the derivatives of a- and /5-naphthols, (a) N: N- /\/\OH OH and (/3), the compounds with the auxochrome in the ortho : /3-position with respect to the chromophore (-N : N-) being colouring-matters of greater fastness to acid and alkali than the ortho- : u-compounds. But if another azo group be introduced into the OH latter, it will occupy the /3-(ortho)-position, N : N- , the fast brown dyestuffs N : N- being obtained. Those most used are the sulphonic derivatives, obtained from various naphtholsulphonic acids. s Of the numerous colouring-matters of this group, the most important are : orange II tropceolin OOO N. II or N. I, croceine orange, orange G, &c., Ponceau (various), Bordeaux S, N-C, H 6 .S0 3 H (4), amaranth, rocelline, croceine, azorubin S I j > &c. BENZIDINE 657 (c) Azo- Colouring-Matters derived from Carboxylic Acids are obtained from carb- oxylic diazo-compounds and phenols or amines. These compounds (especially the o-hydroxycarboxylic acids, such as salicylic acid) have an affinity for metallic mordants, particularly for chromium oxide. The hydroxyl and carboxyl groups are in the ortho-positions. Among the nitrobenzeneazosulphonic acids are alizarin yellow, the diamond yellows, &c., which, on cotton and wool, give colours very resistant to light and to fulling. The hydroxy- azo-acids include various tropseolins (V, B, 0, OOO, &c.), chrysoin, cochineal scarlet, ponceau, palatine scarlet, &c. (d) Azo- Colouring-Matters derived from Dihydroxynaphthalenes. Several of these compounds are fixed by mordants when they have two hydroxyl groups in the ortho- (1 :) 2 OH OH peri- (1:8) position, as in anthraquinone (see Alizarin) and . But S0 3 H these compounds are used practically, not on mordants, but for the dyeing of wool, as they give very regular results (such are the azofuchsines), while the peridihydroxynaphthalenes are used on mordants and form the so-called chromotrop colouring- matters, which dye unmordanted wool in an acid bath, giving a fine red turned violet by addition of alumina mordants or blue- black with chrome mordants. POLYAZO- COLOURING-MATTERS (di- and tetra-azo) contain the chromophore X : N- several times and vary according as the chromophores are in the same benzene nucleus or in different nuclei and as the auxochromes are or are not in the same nuclei as the chromophores. Here are found benzidine derivatives in which the two chromophores are in two different nuclei, joined by a single linking. Among the sulphonic derivatives are, for example, Biebrich scarlet, and the croceines, while among the polyazo-compounds are also naphthol black, naphthylamine black D, diamond black (which is obtained from aminosalicylic acid and is fixed by mordants), &c. BENZIDINE, NH 2 <^ /NH 2 , when treated with nitrous acid, gives a tetrazo-derivative which yields yellow, red, blue, or violet colours on combination with amines or phenols. With naphthionic acid, tetrazodiphenyl gives Congo red, which was the first substantive dyestuff obtained and was patented by C. Bb'ttiger in 1884 : the free sulphonic acid is blue while the salts are red and are ixed directly on cotton, but have the disadvantage of becoming blue or black in contact with even weak acids. The Benzopurpurins (see p. 605) are obtained in a similar manner. These benzidine derivatives cease to form substantive colouring-matters if the meta- positions (with respect to the NH 2 ) are occupied. Substantive or direct colours, when fixed on cotton, function as weak mordants for basic dyestuffs. The different firms making colouring -matters place on the market a large number of substantive dyes under various names. For instance, Messrs. Casella have a long and important series of diamine colours (diamine yellow, green, red, black, blue, &c.), while Meister, Lucius und Briining call their substantive colouring- matters dianil colours. The Bayer Company have the most numerous and important series of substantive dyes, which they term benzidine or benzo dyestuffs (e.g. benzo azurines, benzo browns, benzo reds, &c.). The Actien-Gesellschaft fur Anilin-Fabrikation, Berlin, call these dyes Columbia, Zambesi, &c. II 42 658 ORGANIC CHEMISTRY o The Derivatives (e.g. sulphonic) of azoxystilbene, C 6 H 4 <^ /C 6 H 4 , have the >0 = (X H H special property of dyeing cotton directly in an acid bath. The firm of Meister, Lucius und Briining, in 1896, placed on the market a class of strongly basic colouring-matters (Janos dyes), which colour cotton directly without previous mordanting in an acid bath and also dye with the same colour the wool and cotton of a mixed fabric when the latter is boiled in a bath acidified with sulphuric acid. These dyes change their tint temporarily if brought into contact with hot objects (hot iron). Of very great importance is the group of azo-dyes produced directly on the fibre by processes of diazotisation and combination, these bearing the name of Ingrain Colours. Cotton fabrics or yarns are impregnated in the cold with a base (aniline, p-nitraniline, aminoazobenzene, benzidine, safranine, &c.) or they may be first dyed with one of the substantive tetrazo-dyes containing free auxochrome amino-groups (e.g. diamine black, primuline yellow, benzo brown, blue, or black, &c.). They are then transferred for 15 minutes to a wooden vessel containing a cold diazotising solution, this consisting, per 100 kilos of cotton, of 2000 litres of water, 2 to 4 kilos of sodium nitrite, and 6 to 10 kilos of hydrochloric acid at 20 Be. ; this diazotisation is carried out in dimly lighted rooms, since sunlight readily decomposes the diazo-compounds formed. After removal from this bath, the cotton is allowed to drain for a short time and is then placed in a developing bath (coupling bath) containing 2000 litres of water, 0-5 kilos of sodium carbonate and 0-5 to 1 kilo of /3-naphthol previously dissolved in 415 to 430 grms. of caustic soda at 40 Be. The cotton is manipulated rapidly and in a few minutes intense development of the colour takes place. When substantive dyestuffs are thus further diazotised on the fibre, they exhibit increased fastness to scouring, and this is still more the case if the fabric is subse- quently treated with a bath of potassium or sodium bichromate at 90 to 95 for 20 minutes ; a final copper sulphate bath at 50 for 25 minutes gives greater fastness to light ; but both opper and chromium compounds diminish the brightness of the colour to some extent, and on this account the firm of Geigy suggests the use of a final bath of formalin. Instead of /3-naphthol, a-naphthol, resorcinol, phenylenediamine, benzonitrole (diazotised p-nitrani- line), &c., may be used. By this method of diazotising and developing on the fibre the original tint of the basic substance is intensified, certain yellows become orange or scarlet (p-nitraniline gives with /3-naphthol a fine scarlet similar to Turkey red, while with a-naph- thol it yields a violet-red), certain reds become brown or even blue, the blues become intense blacks, &c. Different developers give different colours or shades. The coupling of a phenol with a diazo-compound is prevented by the presence of a reducer which destroys the latter ; as reducing agent stannous chloride was at one time used, but use is now made of sodium or zinc hydrosulphite, which permits of the printing of textiles in white designs on a coloured ground. III. HYDRAZONE AND PYRAZOLONE COLOURING-MATTERS. Hydrazones .are obtained by the action of phenylhydrazine, C 6 H 5 -NH-NH 2 , on compounds con- taining ketonic groups (see p. 210). Thus, for example, the condensation of phenyl- CO hydrazine with a-naphthaquinone, gives a hydrazone of the constitution CO C 6 H B NH N : CVoHe : 0. The same compound is obtained by the interaction of a-naphthol .and diazobenzene, so that its constitution might be that indicated by the equation : C 6 H 6 -N 2 -C1 + CioHT-OH = HC1 + C 6 H 5 -N : N-QoHe-OH, one hydrogen atom being mobile and oscillating between nitrogen and oxygen. The hydrazones may hence be regarded as azo-compounds and can be prepared from diazo- derivatives and phenols. This is true for aromatic compounds (which can be diazotised), QUINONE DYE STUFFS 659 but not for those of the aliphatic series, which are only exceptionally diazotised ; in the latter case, the hydrazones must be obtained by means of phenylhydrazine. The colouring-matters of the hydrazone group have not as yet been practically applied, as they are too weak. It was formerly thought that tartrazin was a hydrazone, but Anschiitz showed it to be a pyrazolone. In general the Tartrazins are obtained by con- densing, in hot acid solution, the aromatic hydrazines (sulphonated) with dihydroxy- tartaric acid, CO 2 H'C(OH) 2 -C(OH) 2 'CO 2 H, which probably reacts with phenylhydrazine CO 2 H-C : N-NH-C 6 H 4 -SO 3 H as a true di-ketone, C0 2 H CO CO CO 2 H, giving ; C0 2 H-C : N-NH-C 6 H 4 -S0 3 H a molecule of water is then lost from a carboxyl- and an imino-group, /N(C 6 H 4 -S0 3 H)-CO < 1 C(CO 2 H) C : N NH C 6 H 4 SO 3 H. The sodium salt is used as a fast yellow for wool, in an acid bath. Some tartrazin nitrates are fixed also by mordants. In an acid bath tartrazin dyes wool a bright and fairly fast yellow. IV. COLOURING-MATTERS DERIVED FROM QUINONES AND QUINON- OXIMES. All these colouring-matters give very fast tints on fibres mordanted with metallic oxides with which they form lakes. If the hydroxyl groups present are not in the ortho-position with respect to one another and to the chromophore C0<^ , the lakes formed have no affinity for the fibres. 1 The most important colours of this group are formed by introducing into the chromo- phores, naphthalene groups ; e.g. Naphthazarin, which is a dihydroxynaphthaquinone, The quinonoximes contain the group : N-OH in place of the ketonic oxygen ; they have properties similar to the hydroxyquinones, and here too the affinity for metallic mordants is most marked in the derivatives of the orthoquinones. A few colouring- matters derived from the oxime 0=\ j>=NOH, are known, e.g. fast green for cotton, naphthol green, &c. Among these quinone derivatives are almost all the alizarin (see p. 617) and anthracene (see p. 615) colouring-matters, purpurin, &c., in all gradations from yellows to reds, blues, blacks, greens, &c. For hundreds of years Alizarin was the sole representative of a group of excellent colours, and was only obtained naturally mixed with purpurin, from which it was separated with difficulty. Nowadays, not only is alizarin prepared artificially, but there are quite fifty other colouring-matters of this group, fast to light and chemical and atmospheric reagents. And while nature yields colours such as madder and indigo in an impure condition (as these are secondary products of vegetable life) and not directly applicable for dyeing, the artificial products are highly pure, much brighter in colour and more easily utilisable as dyes. Alizarin and anthracene dyes, which are the prototypes of mordant colouring -matters, are used in large quantities for the fast dyeing of wool for clothing and military uniforms. As a rule the wool is mordanted first, by boiling for an hour with an aqueous solution containing 2 to 3 per cent, of potassium dichromate and 1 per cent, of sulphuric acid and 1 Mordant colouring-matters are generally obtained with the following groups in the ortho-position : OH and NO (or CO and NOH), 2NOH, 2OH. Also, according to Noelting (1909), in the anthraquinone series intense mordant dyes are obtained also with OH and NH 2 in the ortho-position (less important and less intense are those with OH and NHj in the para-position). 660 ORGANIC CHEMISTRY amounting to 15 to 20 times the weight of the wool. After mordanting, the wool is rinsed well in water and dyed in a solution of the dyestuff faintly acidified with acetic acid ; this bath is heated very gradually to boiling, the latter being maintained for 1 to 2 hours to obtain the maximum intensity and fastness. If fresh addition of the colouring-matter is necessary in order to obtain the desired shade, it is best first to lower the temperature of the bath to 40 to 50with cold water in order to prevent non-uniformity of tint. V. DIPHENYL- AND TRIPHENYL - METHANE COLOURING - MATTERS, P TT r 1 TT CHo^ 6 ;; 5 and C 6 H 5 CH<^ 6 - [T 5 . It has been shown on p. 647 that in these colouring - ^C 6 .H. 5 "^e-H-s matters the chromophore consists of the benzene group with two double linkings in the para-position, R=Cy /** The mode of formation and the general properties of diphenyl- and triphenyl- methane derivatives were described on pp. 606, 607. In this group are found Auramine (basic) and Pyronine (also basic) which dye wool in an acid bath and cotton mordanted with tannin. The rosaniline group embraces all the basic colouring- matters derived from triphenyl- methane, e.g. malachite green, methyl violet, for my I violet, fuchsine, &c., while with sulphonic and other groups, acid dyes are obtained, such as patent blue (carmine blue), acid fuchsine, &c. There are also azo- derivatives of triphenylmethane, such as Rosamine, which dyes silk violet-red with a yellow fluorescence, and has the formula : C 6 H 3 [N(CH 3 ) 2 ] C fi H s -C/ ^C 6 H 3 [N(CH 3 ) 2 C1] The Rosolic Acid group, O= >-r H formed b y fusion of phenol with oxalic acid in presence of concentrated H 2 S0 4 , also furnishes numerous colouring- matters, e.g. aurine, coralline, pittacal, chrome violet. Benzoazurin is formed from 1 mol. of phenylchloroform with 2 mols. of phenol : / =\ r\ TT . Off : < > : C0 + 2C 6 H 5 -OH = H 2 + C 6 H 4 < >0. NXK -co Phenolphthalein Phthaleins (see p. 581) with the hydroxyls in the para-positions are of some importance ; if resorcinol, C 6 H 4 (OH) 2 , is used in place of phenol, Fluorescein is obtained : C 6 H 4 CO O while if dimethylaminophenol is taken instead of resorcinol, or if fluorescein chloride is heated with a secondary amine, NHR 2 , fine red colouring-matters, Rhodamines, which are basic in character, result : O R 2 : NCI = NR 2 QUINONIMIDE GROUP 661 If previously brominated phthalic anhydride is used, the Eosins are obtained : C 6 H 4 -C | CO-^O C 6 HBr 2 (ONa) these give beautiful fluorescent red colours on silk but are not very fast to light (see p. 581). VI. COLOURING-MATTERS OF THE QUINONIMIDE GROUP. To this belong the derivatives of indophenol and indamine. Of the hypothetical quinonimides, HN=: >=0 and NH= >=XH, various derivatives and condensation products are known, e.g. Indamine, H 2 N/ X >-N=< / ' "\=NH; Indophenol, OH< The Thiazones, e.g. thiodiphenylamine, 0= >=N < >NH,. with indamines form Thiazimes (e.g. Lauih's violet or thionine, meihylene blue, methylene green, &c., which are basic dyes). The Oxamines and Osazones have an oxygen atom in place of the sulphur of thiazones, /\/\/\ , and undergo various condensations : Capri blue, naphthol blue, Nile blue, &c., which are also basic. The Cyanamines are related to Nile blue ; Resorufin is an osazone, namely, hydroxy- 0= /\/\/\OH diphenosazone, ; Gallocyanine,Ci 5 H 12 5 N 2 ,is obtained by heating nitrosodimethylaniline with gallic acid in alcoholic solution. They dye chrome-mordanted wool a very fast violet, and are used in printing linen, which is treated with sodium bisulphite and chromium acetate and subsequently steamed. The Azines were formerly called Safranines ; the simplest type is Phenazine, C 6 H 4 <^ | j>C 6 H 4 . The eurodines are used for dyeing cotton mordanted with tannin. The Safranines contain four nitrogen atoms and three aromatic nuclei : ,Nv NH 2 Cl and are strongly basic and give red colours on cotton mordanted with tannin. Indulins are obtained by heating aniline hydrochloride with aminoazobenzene. The following constitution has recently been established for one of the indulins : 662 ORGANIC CHEMISTRY N-C 6 H 5 C 6 H 5 -HN- -NH-CH S N-C 6 H 5 The Quinoxalines contain the nucleus represented by the formula : .N. C H ; the Fluorindines can be SHT/ H X C H H VII. ANILINE BLACK. The oxidation in various ways of aniline salts in acid solution gives aniline black, which is of considerable importance in the dyeing of cotton. Among the different oxidising agents, a special place is occupied by vanadium salts (suggested by Witz in 1877), which bring about the oxidation of large quantities of aniline (transferring oxygen by catalytic action)^; 1 part of vanadium, in presence of a sufficiency of potassium chlorate, oxidises as much as 270,000 parts of aniline hydrochloride. In point of efficiency, vanadium is followed by caesium and then copper, the action of iron being much less. Aniline black has a feebly acid character and is insoluble in almost all solvents. It dissolves with difficulty in aniline and forms with it a violet and then a brown colour ; phenol dissolves it more easily, giving a green coloration. With fuming H 2 S0 4 , it yields soluble, coloured sulpho-compounds. Acetic anhydride gives a faintly coloured acetyl- derivative, and potassium dichromate a violet-black product. When treated with per- manganate and then with oxalic acid, aniline black is partially decolorised. Energetic reducing agents (Sn + HC1) decompose it completely. 1 1 The chemical constitution of aniline black has been the subject of much discussion. Assuming that the first intermediate product of the oxidation of aniline is aniline black (Jv ictzki), it cannot be true, as is often thought, that the transformation of aniline into quinone by oxidation takes place through the intermediate stages of phenyl- hydroxylamine and p-aminophenol, since these do not yield aniline black on oxidation, phenylhydroxylamine giving a nitrosobenzene and not a quinone ; nor can aminodiphenylamine (Never, 1907) be formed, since this, on oxidation, gives emeraldine, a compound never obtained in the oxidation of aniline. It has now been found possible to convert aniline black to the extent of 95 per cent, into quinone by oxidising with lead peroxide (chromic acid giving less than 80 per cent.), so that the indaminic formula (proposed by Bucherer. 1907) can no longer be attributed to aniline black, since, according to this, it would give only 50 per cent, of quinone. This result led B. Willstatter and S. Dorogi (1909) to suggest for aniline black the formula (C 6 H 4 N : C 6 HYNH) 4 , i.e. C 48 H a ,N s> which is confirmed by the fact that the oxidation requires 1J atom of oxygen per molecule of aniline with a yield of 97 per cent. Further, the determination of the molecular weight by hydrolysis of aniline black with dilute sulphuric acid at 200 indicates clearly the separation of one-eighth of the nitrogen as ammonia : C 6 H 4 N : C 6 H 4 : NH + H 2 = C 6 H 4 N : C e H 4 : O + NH 3 . All these results point to the trebly quinonoid formula of aniline black as the most probable : This aniline black is obtained by oxidising aniline in the cold with rather less than the theoretical quantity of dichromate, chlorate, or persulphate. Further oxidation with H 2 O 2 , for example, results in the elimination of 2H and the formation of a quadruply quinonoid aniline black, C 48 H 34 N 8 , the base of which is very dark blue-black while the salts are dark green. It absorbs only 2JHC1 whilst the trebly quinonoid black absorbs 4HC1 ; all of the latter are displaced by ammonia, which, however, in the former case, leaves 1HC1 (4-5 per cent, of 01 in the nucleus). In practice the quadruply quinonoid black is obtained with excess of a slow oxidising agent acting in' the cold, e.^.with chlorate and copper sulphate or with chlorate and vanadium. On hydrolysis, the quadruply quinonoid black also loses one-eighth of its nitrogen as ammonia, forming the more complete black, C 18 H S aON 7 , which is not turned green by SO 2 . Oxidation of the corresponding product of hydrolysis of the trebly quinonoid black gives the same quadruply quinonoid black, C 48 H 3 ,ON 7 . The practical preparation of aniline black in a single bath leads to the quadruply quinonoid black that turns green, and further oxidation of this in the hot yields the black QUINOLINE, THIAZOLE, ETC. 663 In practice aniline black is produced directly on the fibre and the use of this very stable colouring-matter is due especially to the studies and initiative of Prud'homme, C. Koecklir, Paraf, &c. After many improvements, the production of aniline black (termed also oxidation black or fine black) directly on cotton fibre is now carried out as follows (the quantities given are for 50 kilos of cotton). The three following solutions are prepared separately : I. 5-5 kilos of aniline oil (see p. 558) + 6-25 kilos of commercial HC1 + 50 litres of water ; II. 3-5 kilos of sodium (or potassium) chlorate + 50 litres of water (1-5 kilo of starch is sometimes added) ; III. 3 kilos of potassium ferrocyanide in 20 litres of water. When cool, the solutions are mixed (1 grm. of vanadium chloride is sometimes introduced) and the yarn or fabric immersed until it is well soaked. It is then gently pressed and passed slowly over rollers through the oxidation chamber (see illustration given later) so that at least an hour elapses before it emerges at the opposite end. The temperature of the chamber should not exceed 50 and the humidity 25. The fabric assumes a coarse greenish colour, which is changed to a fine black when it is transferred to a Jigger (see later) containing 2 kilos of potassium dichromate, 250 grms. of sulphuric acid and 100 to 120 litres of water at the temperature 50. The black thus obtained, when thoroughly washed, is turned green only to a slight extent in the light. VIII. QUINOLINE AND ACRIDINE COLOURING -MATTERS. Among the quinoline dyestuffs are quinoline yellow (water- or alcohol-soluble), quinoline red, ,CH : CH cyanine, &c. ; all of them contain one or more of the chromophores, C 6 H 4 \ | , \N : CH or its homologues. Acridine derivatives possibly contain a quinonoid chromophore of the formula H 3 ( : NHk C 6 H 4 They are obtained by condensing m-diamines with formaldehyde, heating the resulting tetraminodiphenylmethane with acid to remove ammonia, and finally oxidising with ferric chloride. To this group belong acridine orange and yellow, phosphine, benzoflavin, &c. = C-N X IX. THIAZOLE COLOURS. These contain the group ^C with the chromophore C=X and are formed by heating p-toluidine with sulphur, the resulting Primulin being probably of the constitution X*v / N v CH 3 -C 6 H 3 < )C-C 6 H 3 < }C-C 6 H 4 -NH 2 : x s/ \s/ it may be easily sulphonated, dyes cotton directly and may be diazotised and developed on the fibre (see p. 658). The methyl derivative is Thioflavin. These colouring-matters are not very fast against light. X. COLOURING-MATTERS OF THE OXYKETONES, XANTHONE, FLAVONE, COUMARIN. This group embraces many valuable mordant colouring-matters : alizarin yellow, anthracene yellow, alizarin black (see Alizarin Colouring-Matters, p. 659), flavopurpurin, alizarin green, alizarin blue, alizarin cyanine, anthracene blue, &c. The characteristic group of the xanthones is , and that of the flavones CO CR CH Indian yellow is a hydroxy- derivative of xanthone. XI. INDIGO, INDIGOIDS, AND OTHER NATURAL COLOURING-MATTERS. In addition to what has been stated with reference to indigo (see p. 639 et seq.), it may be said that there are a number of derivatives of artificial indigo which are reduced with hydro- which does not turn green, the terminal imino-group being hydrolysed. This latter black is obtained also by the two- (or more) bath process or by steaming. Oxidation of aminodiphenylamine instead of aniline gives first the reddish blue imine (C !4 ..,.), emeraldine, which then polymeiises, forming the black (trebly quinonoid). 664 ORGANIC CHEMISTRY sulphite and alkali and give very fast colours which are superior to indigo and resist even concentrated solutions of chloride of lime. 1 The Indanthrene Colours, which were at first very expensive, are now obtainable at more reasonable prices and give medium and dark shades. They are so resistant to various reagents that they are used as pigments in place of ultramarine, &c. ; they are used also for blueing sugar and other foodstuffs, as they are fast to light and non-poisonous. 1 Bohn has given the name vat dyestuffs to those insoluble pigments the molecule of which contains at least one ketonic group capable of being reduced (e.g. by hydrosulphites), taking up hydrogen and thus becoming soluble in an alkaline liquid and flxable by animal and vegetable fibres. These vat dyestuffs may be divided into two classes : indigoids and indanthrene derivatives. The first class comprises two series : symmetrical (indigo, c = c , NX s Symmetrical with sulphur chromogen (4) NH, CO >NH Indigo and its halogenated and other substitution deriva- tives : chloro-, bromo-, alkyl-, and naphthol-indigo. The substitution takes place in the benzene nucleus ; many polybromo-derivatives are formed. The colours range from reddish blue to greenish blue. The antique purple recently studied by Friedlander is 6 : 6'-di- bromoindigo. 5-Bromoindigo (pure indigo R), 5 : 5'-di- bromoindigo (pure indigo 2B), 5:7: 5'-tribromoindigo (Ciba blue B), 5:7:5': 7'-tetrabromoindigo (Ciba blue 2B or indigo 4B) have been prepared. CO N Besides the chromophore also the chromophore =C< >C = of indigo, these have Belonging to this family are : Ciba grey G (monobromo-derivatives), Ciba violet R, B, 3 R (these are polybromo-derivatives of Ciba violet A). 'The first term is Friedlander's thloindiyo (or thioindigo red B) ; Ciba bordeaux B (5 : 5'-dibronTothioindigo) and numerous derivatives in which the 5- and o'-positions are occupied by alkoxy- and thioalkyl-groups have been prepared, among these being red and brown colours and the various colours of the helindone series of Meister Lucius und Briining. Indirubin is not a colouring-matter, since on reduction it forms indigo. But use is made of tetrabromoindirubin (Ciba heliotrope B) : X NH< /CO v C.HjBr Indirubin with asymmetric nitrogen chromogen '\ >C=C< >NH CO - C = NX" CO \ \ _ Thioindigo scarlet B I The dibromo-derivative forms thioindigo scarlet G (or Ciba red G) : Q C*O fi TT / NP n/ \"\rw L H *N X U - \ . X^ 11 >Is known by the name of thioindigo scarlet 2 G (Ciba scarlet G). This is a new family which has given the first yellow colours of the indigoid group (Ciba indigo yellow 3 OH; OH it dissolves in alkali, giving an orange-yellow coloration and yields phloroglucinol and protocatechuic acid when fused with alkali. It is sulphonated by concentrated sulphuric acid, forming a direct dye for wool. It dyes more especially animal fibres (wool) either previously mordanted or with an alum or chrome mordant added to the dye-bath. Similar behaviour is shown by flavin, which is a more concentrated preparation of quercitron and contains quercetrin and quercetin. Natural INDIAN YELLOW is still extracted in Bengal from the evaporated residue of the urine of cows fed on mango leaves. It contains a hydroxyl derivative of xanthone, namely, Euxanthone, as glycoronic ester (euxanthinic acid, C^H^O.^), which is decomposed by hot hydrochloric acid into Euxanthine, C 13 H 8 4 or HQ/ N CO O OH (obtained synthetically by condensing hydroquinonecarboxylic acid with /3-resorcylic acid). Natural Indian yellow functions as a mordant dyestuff, but is now scarcely used for textiles as it is not very stable to light. CHLOROPHYLL 669 BRAZIL WOOD or Red Wood is obtained from the trunk of Ccesalpina brasiliensis and other varieties. The colourless glucoside it contains gives, on fermentation or when treated with acids, glucose and Brazilin, C l6 R 1 ^O 5 or C 6 H 3 (OH) 2 -C 4 H 4 0-C 6 H 5 O 2 , which is coloured carmine by alkali and decolorised by acids or reducing agents ; it gives intensely coloured lakes and oxidises in the air, forming Brazilein,C 16 H 12 5 , while with concentrated nitric acid it gives trinitroresorcinol and, when fused with alkali, resorcinol. It is a red mordant (alum or chrome) colouring- matter, but is only slightly fast to light. Brazilin seems to have a constitution analogous to that of haematoxylin (see p. 666) with a hydroxyl group less in the first benzene nucleus, brazilein being apparently the corresponding quinonoid derivative similar to haematein (see above). SANDAL WOOD is the wood of Pterocarpus santalinus, which grows in Madagascar,, tropical Asia, and Ceylon. Santalin or Santalic Acid, C^Hj 6 6 , which forms the colouring- matter of this wood, occurs in abundance in other plants (in barwood or Baphia nitida of Sierra Leone and in camwood or kambewood from West Africa). Santalin gives resorcinol, acetic acid, &c., when fused with alkali, but its constitution is not yet established. It is a mordant colouring-matter, like logwood, and was once used with alizarin to dye cotton red. CATECHU (or Cutch) and GAMBIER are extracted from various plants of India, Bengal, Malay, &c. (palm, mimosa, Rubiaceae, Acacia catechu, Areca catechu, Uncaria gambier, &c.). They contain tannin and colourless catechol, partly combined to a brown colouring- matter. When fused with alkali, they give phloroglucinol, pyrocatechol, and protocatechuic acid. With various mordants they give stable browns or olives, which do not, however, withstand chlorine or alkali. On cotton they give reddish or yellowish brown colours which become fast to light after treatment with a-lkali dichromate at 60 to 70 (khaki used for uniforms in the British, German, and Italian armies). Nowadays a much faster khaki is obtained by impregnating the white fabric in a cold concentrated bath of pyrolignite of iron, chromium acetate, and a very small proportion of manganese chloride, drying it thoroughly, immersing it in a boiling bath of caustic soda (11 Be.) and a little sulphoricinate, and oxidising in a hot-air chamber or by means of dichromate solution. With a less concentrated soda bath or one not boiling, the metallic oxides would be precipitated superficially on the fibre, and the dry fabric would be dustv and would wear out sewing needles. This khaki is very fast against light, scouring, and chlorine, but does not resist perspira- tion (test with a mixture at 1 Be. of hydrochloric, formic, and acetic acids for 5 hours). Fastness to perspiration is given by boiling the dyed fabric for 2 hours in a silicate bath at 6 to 7 Be. Statistics. See above. CHLOROPHYLL is not a colouring- matter for textiles but is the green pigment which occurs in many plants (those which assimilate C0 2 ) and brings about the transforma- tion of the carbon dioxide into starch in the leaves under the action of sunlight especially of certain rays of the spectrum and apparently also with the help of an enzyme (Will- statter and Stoll, 1911) known as Chlorophyllase. With starch, wax, &c., it forms the characteristic chlorophyll granules of green leaves. It is soluble in oil, alcohol, ether, or chloroform, its solutions showing blood-red fluorescence and readily undergoing change. Its constitution is still uncertain, and it does not appear to contain combined iron as was formerly thought. Following the indications of the botanists Borodin (1882) and Monte verde (1893), Willstatter and Benz (1908) obtained a pure chlorophyll 1 (2 grms. from 1 kilo of dried leaves) in dark, bluish black crystals with a metallic lustre, which are insoluble in petroleum ether but soluble in alcohol or ether, giving a bluish fluorescence. The green solution of this product, which exhibits the same spectrum as the chlorophyll of fresh leaves, is turned brown by alkali, but again becomes green. Its formula is probably C 55 H 72 O 6 N 4 Mg, and the magnesium present (3 per cent.) is perhaps the cause of the catalytic action effecting the transformation of CO 2 into starch ; it does not contain phosphorus, as many, including 1 As chlorophyll readily undergoes change, it is extracted in the cold with methyl alcohol from the carefully dried, powdered leaves (Willstatter), previously washed with petroleum ether. In order to separate it from other colouring impurities, its alcoholic extract is suitably diluted and extracted with ether (benzene or carbon disulphide), many of the impurities remaining dissolved in the alcohol ; or the alcoholic extract may be shaken with a large amount of water, which dissolves the chlorophyll in the colloidal state, the decanted aqueous solution being treated with salt and extracted with petroleum ether containing a little alcohol. From this solution the- chlorophyll is deposited pure if the whole of the alcohol is eliminated by washing, 670 ORGANICCHEMJSTRY Stoklasa, have thought. Acids remove all the magnesium, the residue being Phoeophytin, which is similar to chlorophyll, is ethereal in character, and forms various products (e.g. phytol, phytochlorin, and phytorodin) when hydrolysed with alkali. Phytol forms one- third by weight of the chlorophyll of plants and is a primary, unsaturated, monohydric alcohol, C 20 H 40 0. Plants produce also an amorphous chlorophyll which, unlike the other, gives phytol on hydrolysis. It is thought that it is analogous in chemical composition to the colouring-matter of the blood (see later), since both yield pyrrole when distilled with zinc dust. Willstatter and Isler (1911) showed that chlorophyll contains two colouring- matters : (a) bluish green and (b) yellowish green, thus confirming the hypotheses of Stokes (1867 and 1873) and of Tswett (1906) ; the two colours are separated by more or less dilute alcohol. Chlorophyll is used in practice to colour oils, soaps, fats, preserved vegetables, &c. ; it costs 8s. per kilo or, for the highly purified product, 80s. per kilo. XII. SULPHUR COLOURING-MATTERS. These colouring- matters, which have been discovered since 1893, are very fast on cotton, which they dye directly without a mordant, but in alkaline and reducing solution (sodium sulphide and sometimes, a little glucose) which prevents any unevenness which might be produced in the colouring owing to contact with the air. The sulphur colouring-matters do not dye wool or silk in presence of sodium silicate (or of blood or diastofor), so that two colours can be obtained on wool and cotton fabrics, the wool being dyed first with an acid dyestuff and the cotton subse quently with a sulphur colouring-matter in a bath of sodium sulphide and silicate (or blood or diastofor). They are obtained by melting together sulphur or sodium sulphide and various other colouring-matters or other organic compounds. Cachou de Laval has been known since 1873 but has been used but little It was obtained by Croissant and Bretonniere by fusing sawdust, bran, or the like with sodium sulphide. In 1893 the discovery of Vidal black turned the attention of manufacturers to this interesting group of colouring-matters, which now include almost all tints except red, and are obtained by fusing with sulphur or sodium sulphide, derivatives of benzene, naphthalene, diphenylamine, anthraquinone, &c. These colouring- matters are placed on the market by various firms under different names, although their compositions are practically the same : the firm of Cassella calls them immedial colours ; the Bayer Company, katigenic colours ; the Badische Company, kriogenic colours ; the Berlin Aktien-Gesellschaft, sulphur colouring-matters, &c. The constitution of these colours has not been firmly established, but during recent years a little light has been thrown on them. According to Sandmeyer (1901) they are derivatives of Piazthiol, N, /S, the compound soluble in sodium sulphide having the constitution W \ R Nv xNa /S = S\ ; but nowadays other interpretations have been given. N/ X Na I R When diphenylamine-derivatives are fused with Xa 2 S, black colouring matters are preferably formed, with aminohydroxydiphenytamine derivatives and the corresponding N-alkyl and N-aryl compounds blue colours are obtained, while in presence of stable metasubstituted compounds, brown or yellow colouring-matters are formed. In general the reaction takes place with preliminary formation of aromatic mercaptans or polymercaptans (in the ortho-position with respect to N or to 0), which give further condensation products, e.g. black derivatives of thiodiphenylamine (of thiazine), / \ / ' x ^" ,s/ \/ and yellow or brown colouring-matters derived from thiazole (see above), TESTING OF DYESTUFFS 671 They form insoluble condensed products (disulphides) with the oxygen of the air, these being rendered soluble again by alkaline reducing agents (sodium sulphide, hydrosulphites, &c.). The fixation and development of the colour in the cotton fibres consist simply in the oxidation of the mercaptan to disulphide. The black or blue sulphur colouring-matters are quinonimino-derivatives of the thiazine group. These colouring-matters are now used in large quantities, the production of sulphur black alone in 1909 being estimated at nearly 5,000,000 kilos. It has been proposed (1909) to render them faster to washing by treatment with formaldehyde or by immersion in a nickel sulphate bath. TESTING OF COLOURING-MATTERS Of the thousands of colouring-matters sold by different firms under most varied and fanciful names, the majority represent, not chemical individuals, but intimate mixtures of several colours which give directly the tints desired. The colouring- matters obtained at the end of the manufacture by precipitation or separation from their solutions by means of salt (just as with soap) are not sold in the pure state, but are diluted with 50 per cent, or 75 per cent, of finely ground sodium chloride or sulphate. A mixture may be distinguished from a chemical individual by the following simple test : a few milligrams are blown in a cloud from a watch-glass and are caught on a moist filter-paper spread on a sheet of glass at a short distance from the watch-glass. If the filter-paper were not too moist, it shows on drying isolated, swollen points of colour, the uniformity or non-uniformity of which is readily seen. A variation of this test consists in sprinkling a little of the powder on to the surface of concentrated sulphuric acid contained in a flat porcelain capsule. The use of the spectroscope has been suggested for differentiating between various groups of colouring- matters, the positions of the absorption bands being observed when white light is passed through an aqueous or alcoholic solution of the colouring-matter of definite concentration contained in a glass vessel with parallel glass walls. The spectroscope is now, however, scarcely ever used, owing to the uncertainty of the results obtained ; but it is useful in detecting the colouring-matter of the blood (see later Haemoglobin). The qualitative analysis of colouring- matters for the detection of the principal groups may be carried out according to the method of A. G. Rota L or to those of Weingartner and Green. The latter, which are largely used, are briefly as follow : I. COLOURING-MATTERS SOLUBLE IN WATER. (A) If the aqueous solution gives a precipitate with a solution containing 10 per cent, of tannin and 10 per cent, of sodium acetate, the presence of basic colour ing -matters is denoted : (1) If the solution of the colouring-matter is reduced with zinc dust and dilute hydro- chloric acid, a few drops of the decolorised solution are placed on a piece of filter-paper : (la) The reappearance of the original colour of the substance when the paper is waved 1 Rota's method, extended by Buzzi (1911), for analysing colouring-matters consists of four series of tests : A. This is based on the usually quinonoid character of these matters and hence on their behaviour towards acid reducing agents, preferably stannous chloride ; the alkaline reducing agents do not serve well, as with all colouring-matters they give leuco-derivatives which are not very characteristic. The behaviour with SnCl a + HC1 permits of the division of all colouring-matters into the following four groups : I. Those which are decomposed may contain the following chromogens (p. 647) : OH = N-OH >NH X=( , &c. \ / II. Those which are reduced to colourless leuco-compounds, which can be reoxidlsed, contain the chromogen (P. 647) : N N N H,N NH,-C1 (orO) 672 in the air indicates azines, oxazines, thiazines, and acridines, i.e. according to the colour, pyronine, safranine, rosinduline, phosphine, benzoflavin, indulin, &c. (16) If the original colour appears but weakly or not at all, but is formed immediately on moistening with a drop of 1 per cent, chromic acid solution, the colouring-matter belongs to the rhodamines or to the triphenylmethane group ; (Ic) The non-appearance of the original colour under any conditions indicates auramine, thioflavin, chrysoidin, Janos colours, Bismarck brown. III. Colouring-matters which are neither reduced nor decomposed, but have a basic character and arc partly decolorised or precipitated by caustic soda, contain the chromogens (p. 647) : H,N /\ NH 2 -C1 (or = O) \/\/\/ C H,N NH, IV. Those which are neither reduced nor decomposed, andjhave a phenolic character (feebly acid) and are increased in colour and solubility by caustic soda, contain the chromogens : T OH \_/ =0i OH Groups III and IV always contain the chromophore OH OH and to these belong the acridines, the thiazoles, theauramines, the rosanilines, the pyronines, the rosamines, the phthaleins, the rhodamines, the hydroxyacetones, the hydroxyanthraquinones, the coumarins, flavone, flavonal, &c. B. To distinguish between the different chromogens of the separate groups, other special reactions are used, for instance : The acridines, with concentrated sulphuric acid, give a fluorescence resembling that of petroleum. The 0o-dyestuffs, with concentrated nitric acid, regenerate the respective diazo-salts. The hydroxyacetonic, hydroxyquinonic, &c., colouring-matters are precipitated as lakes by stannous chloride and subsequent treatment with sodium acetate. The transformation of azo- colouring-matters and their derivatives into thiazole (polychromin). The conversion, by special reagents, of one colouring-matter into another, e.g. gallein into ccarulein. C. After the restriction of the colouring-matter to one of the four groups, and after the various tests for defining more exactly the character of the chromophore have been carried out, the process of identification is continued by means of systematic dyeing tests which vary with the auxochromes and salt-forming groups (see p. 649), imparting to the colouring-matter a basic, acid, phenolic, substantive, or a mixed character, such as basic phenolic, acid phenolic, substantive basic, substantive phenolic. The group with azo-chromophores contains, for example, Bismarck brown, which is basic (see p. 656) ; metanil SO,H yellow, _ N = N NH ; which is acid ; alizarin yellow R, stantive ; chromotrop 2R, )OH, which is phenolic ; Congo red (see p. 657), which is sub- CO 2 H SO,H / NH N=\ / = o - which is acid phenolic ; carbazole yellow, \ / NH > NH N=: ) NH N=r OH :=O SO 3 H which is substantive-phenolic in character. CO 2 Na DYEING TESTS B. Non-precipitation of the solution by tannin, &e. (.t?r above) denotes the presence of arid colouring-matters : (2) The solution of the colour is reduced as in (1) or with Zn -h NH 3 and a drop placed on a strip of paper : (2a) The reappearance of the original colour on shaking the paper in the air indicates sulphonic or mordant dyes of the groups of azines, oxazines, thiaziiies, soluble indulin, nigrosins or azocarmine, thiocarmine, indigo-carmine, gallocyanine, Mikado orange. (26) If the coloration reappears only after treatment with chromic acid or ammonia vapour, the original aqueous solution is acidified with sulphuric acid and shaken with ether ; coloration of the ether and complete or almost complete decolorisation of the solu- tion indicates phthaleins or auramines, while non-coloration of the ether shows triphcnyl- methane dyes. (2r) Xon- coloration of the paper even when heated in a flame or treated with ammonia vapour points to azo-, nitro-, nitroso-, or hydrazine-colours, which, when burnt in powder directly on a platinum foil, give coloured vapours (e.g. naphthol yellow S, picric acid, Victoria yellow). (2d) If on reduction the solution is not decolorised but becomes reddish brown and in In the group with hydroxyazinc chromophorcs are, for instance, Meldola's blue, K , which is basic; gallocyanine (see p. 661), N CO 2 H C1(CH 3 ) 2 N O which is basic-phenolic in character. So, also, the thiazinc group (see p. 061) contains methylene blue, which is basic, and thiomnnine, which is acid. The dyeing tests arc made in hot neutral and acid baths, in each of which tour samples arc immersed, nann-ly, cotton, cotton mordanted witli tannin, wool, and wool mordanted with dichromate (for the mordanting, see pp. 651 and 706). The more or less intense colours assumed by the samples give indications concerning the character of the colouring-matter (see p. 650), and confirmation of this is obtained by various tests on the dyed fabric : () The colour is substantive if, when the dyed sample of natural wool is heated in faintly alkaline water, the colour passes to the white cotton placed in (he same bath ; (b) The colour is acid if the change indicated in it is not observed, and if, when the bath is acidified, the wool takes up the colour it gave up to the alkaline bath ; (c) The colour is basic if in bath () t be colour passes from the wool to a sample of white cotton mordanted with tannin ; (rf) The colour is phenolic if the tint on mordanted wool varies with the nature of the mordant. Tests may also be made on the solution of the colouring-matter ; thus, if it is precipitated by tannin or picric acid, the colour is basic ; if ether extracts the colouring-matter in an acid medium, the colour is phenolic, whereas if ether extracts the coloured base in an alkaline medium, the colour is basic. If it is established that the colouring-matter, containing a given chromophore, is basic in nature, all acid, sub- stantive, phenolic, A-c., colouring-matters with the same chromophore are excluded. 1). For the further individualisation of the colouring-matter, useful information is given by the following reactions characteristic of the substituent radicals. The NH 2 group is recognised by diazotising and then coupling (see p. 658), by which means a new azo- colouring- matter is formed, or by boiling the diazotised product with water, the formation of the OH group being shown by the increased solubility in NaOH compared with that of the original colour. The more or less basic groups are indicated by the greater or less sensitiveness of the solutioii to mineral acids : The N(CH 3 ) 2 group is sensitive, as seen in -methyl violet and methyl orange ; The NH 2 group is less sensitive, as in fuchsine and acid yellow ; The group NH- is less sensitive still, as in aniline blue and metanil yellow. Different colorations with different concentrations of acid indicate several salt-forming groups. To complete the characterisation of a colouring-matter, the latter must be tested for halogens and nitro-groups. Thus, to distinguish alizarin yellow 11 (see above) from diamond yellow O, CO 2 H( >OH, the CO 2 H nitro -group is tested for by reduction and diazotisation, its presence indicating alizarin yellow. Other colouring- matters are differentiated by testing for chlorine and bromine. The azo-dyestufls are characterised also by the formation of the corresponding diazonium nitrates when treated with concentrated nitric acid : )OH )OH SO, II then by testing for diazo-eompound with /3-naphthol and ascertaining the solubility of the nitro-derivative, the position of the sulphonic group in the molecule may be determined. The Tables given on pp. 674-679 afford considerable help in the rapid characterisation of colouring-matters. n 43 674 ORGANIC CHEMISTRY f da Q >, i- 48 M So Si- a a is a > 3 - -a aa o e S - T , (B O Q 1 "S :> -I |* 5. tT+ I * |+ 5 + o '- 5 10 per cent. NaOH ^ c ^ Q a 15 J, s ? > ^a++K"og c -g GPS ^ "o'*' f :: ! ||_|_ ^ 8 4 ' ''' i ** i 8 fapH )N FIBRES HNO, (sp. gr. 1-40) on F _S B C B * i ! S ^ ^ 5? 1 I SH i !H S v ^ o CO P 3 fe = C a K. b> fi 9 ^ s ^ 1 33 ->>a != H P3 PH H i H O te lo o 1-1 > ' a > 5 > - || llo""l >n5. + a c RECOGNITIO] o O ^ g II ^ ^ ^^ ^ 0-5 M f MfQi 1 * * S 1 n* D S tt ^"a'^Hrli^O 1 " B a pj PR H W PH pc, " 03 to S ll_l_io"-)-_^ PH ! " + II ^ Hi 11 11 g , C/3 o S ffl . ...... ............ O t4 UH O >*i INAME OF COLOU MATTER ^ s 1 c o " S . | - ; 'i ; 1 I Is 2 SI 10 ^S-f.s.s^ ""ScJ ^aa^S rt ' "". "oooj'aH ^'S* ''"iM^f'S.S^Qi ^<5^ BL &L ^ "^ao^.o^-^o^^ "o "2 e^cj-cS '2'E'S3ci'o'o^-H'?~ a | | =1 |S | 1 I 4 fl | 1 I 1 I I 1 | WOOL-DYES 675 '*'* o | PH M SH H . ? S 9o *3 la"5, + s 'r-jl'i +-+ o ^ * - I * ? a 31+ I S I Q ! ^ II ^ h e a a o o o MM II * , pq pq ^2 O II n ^ o ^ _0 O M ^ 3 ^""CM Oe'S c *^| 'ri^ftc- o-t^'" ao ^oocs ob~ == N C M Emm fr * HO S 2 -c ^ i i |g be v 076 ORGANIC C 11 E M I S T R Y + -g "lStl=o +0)1 -? 1-1 ~. ' J - 7-1 ,- ! c . o S o 11 !!> wt u PH PH II PH ^ PH fQ ' ? J8 i ^M rt ' pq 5 7J ^ T C PH EH *f PH 3- o o 5 th"" 1 'r* M xUrt^aSa^^P -,,. . + p^S ^05^* Sort' x 1 t>r ' " a e K ' ^-S^-C fe O O |w A " ca O -o:^ < pj^i - 05 r " as " 11 " II 1 PS N f- h |i l ^|Jtl flpll "tii PH II 1 H 5 t !!; 5wa!i 1 PH EH II PH 10 per cent. H 2 S0 4 S'i ' + + ! >- 1 + ' ii PH J?^-^ ^#c- :>f ^ - -aO i ^ "5 "5 + g 3 5 1 II S334- ' I ,0,0 _S S PH M . t- M^HgBMB Q ,>^ ^X ^ !*5i1ft?Hl**ii?, i M ffi ,fq J * I * *{H| .S^ PH M PL, EH p> OO^ ++H 5 + PH ^ S e rf 1 B ^ && pc< II II PH PE^ i : '3 S !*> 3 O 11 .?. ,..".:.. * g S H H a .? oj i-i * P^ S s K _ | 's 1 ' ^^'SSO^^STSI^O g g "E S-s "- g .S.H .s g 5 g-g-r ggSg^llc^ 1 ?^^ g 1 1 ' ' ' ' C D . ...--. "S .1 = a 1*^ if -p33-= ^ ? s s H 1,1 J J "3 ^ 1 "* s ii 2. 2 BROWN AND BLACK DYESTUFFS 677 B _, S IH 33 A w w 7 II fa f*< S PQ 4- I mord dark, a E II II 4- M , i ii a II fa II i a 3 2 Q _ i sg i 55 ii as ill a S>fl ORGANIC CHEMISTRY _ M p p c - pq p? i I IV. " s s "oS "3 J5 w jj pq *j ti pq fe'-^l, ffi _|>H " > *? 1 ltil | I 1 5|ees= RED AND BROWN COTTON-DYES 679 i Q ^ Q O Q J2 . | ii A M ^ a | . o 1^0 ft SP S c 2"g ^ fe 8 "" ""> SH * 1 I | pq "3 "3 o ^;O , P | s -c 1 1 "38 S I ^ "^ N y c ^ >. d" "a 5 3 .S b O ^ B d" M d d a " q 3 3 >H H rt 3 PH o > > P3 ^ V 7 g ^ " " o * " > n 1o " t, a aJ e B d B a hitiiil li'ilfc*'**" 1 j riS d B a a + oj c3 M * pq -M ll+l 1 B 3 || ^ .^ 2 . S.B ^ 6HfN ^ II ii ii [4 II 5 -3 II S fc- B S 1 1 If Oo M >H 1 M II II *O ^ 5 " K* 1 M | t, O t. ti -" B II O c8 o i' ii ^ ^PHBO^PH^+^O^ | |-M p; +B O >*^ss 1 II * 1* PH II 5 *" II * 6n ** \ S PM = II h p!, || || _ 1-4 , Q h * >? LH n * * ,-, a .- fi OH P B a i* oioi.JI ?i7l "a ill B B fi . || d 3 OH B |H II M d II + O - 5 ! w * ?^ **.<* B-'V* B * { * i I B _ o "jj 1 5 a ii rt ! + B ! B - 1 II B B 03 fa -*-' c&4 -*-* ^ l " 1 ii^ 1 c ESI *o3 "5 B B *gj ^ 13 ^ oo W II Pt a ^ 'i '5 I' " Pn " N II + II ^T II __| " C .j- h &H II &H 6H a) O ' &H 52 II Jl> o " ll .3 d ^ d d d 3 3 U B d a d d "T 1 V co'cajv^^'toiB'"'" i'Sa 3 3 In "S d d d d d + 33 3 ^ 330 3 3333 M II "3 "3 "3 a a PH ^ ^ 3 " II i* d s 3 .2 c" o .2 ^ >- t, ^ n C M -G h . 8 -3 S3 " I' " 'I || B B >> t, > ^ * PH" O 3 * a 1 ^ 1 || || I' PH || II SS -|_ j o | > "Jnll -S S II I' "*^*^r^ ^"^fc4 > bi !C:ctn "^^"^^^ ^'>ll II *-" * ft^ "5* * B * U, K, B ^ ^ * O^ ~ ' "B^BTQir^PHp^^^^aa O (J 1 Tj 1 ^T o ^ fc- pq B 1-5^ ^l rt + + >+ rt ^ >r fc<1tf V"" " * ^> a S W 4- 1 B M | ^ 2 || ^ t-^ tn" r*^ Lrt" ' 4" II II II II II M ^ PH * . t* 1 &H * II CM 1 i ||6Htoft|ai^fi( ^ t, -0 B a * ^ S E - "^ . 2 3 o "2 ' If jj "a, 15 B 2 3 - o O 3 -a 1- "''?'.-- II 3 o ii B . "o ^ II o ' ' * U ' ' a ' B (J . G ^ C tA o 8 (A 5). TEXTILE FIBRES Before a description is given of the processes and plant used in dyeing textile Hbres, the physico-chemical properties of these may be outlined. WOOL. Only sheep and certain goats furnish true wool used in the great textile industries. The wool fibre is readily distinguished from the hairs of other animals by its softness and fineness and by its waviness and curling, which can be seen with the naked eye. Also under the micro- scope a marked difference from all other hairs is easily discernible (Figs. 424 and 425). The whole filament seems to be composed of closely superposed scales, which are more or less large according to the quality of the wool. It is the saw-like or serrated structure of these scales which explains why wool readily forms a felt when rubbed, the filaments becoming more or less firmly attached one to the other. 1 The quality of a wool is closely dependent on the breed of sheep producing it and only partially on the climate, food, and age. The yield of wool is greatest from the second to the sixth year. The finer wools, furnished generally by the merino breed, 2 are long, slender, soft, and very wavy and form the so-called combing wool for the best woollens. Shorter wools cannot be combed but only carded (Hiksian, AWvm), although nowadays nearly all could be combed with the improved machinery available, and a large part of the carded FIG. 424. II III IV FIG. 425. ' The whole of the wool covering the sheep forms the fleece, which is kept entire even after shearing (this is done in May) owing to the scaly structure of the filaments. Wool obtained by shearing twice a yenr is called liistose, while that from slaughtered sheep is termed skin ivool and frequently contains dead hairs, which have little affinity for colouring-matters and arc often impure owing to the use of lime, arsenic, &c., as preservatives. If the sheep is washed in the tank before shearing, the wool is known as washed, the other being called in grease or unwashed. The fleece (weighing 2-5 to 3 kilos) contains different parts of different qualities and these the sorters separate by cutting. In one and the same fleece the finest wool is that of the shoulders, then conies that of the neck, stomach, flanks, and back, the poorest qualities being those of the head and legs. Certain African sheep (Morocco), and, to some extent, the Lincoln, Leicester, and Wellington breeds give long, coarse, and only slightly curved fibres, which are used for special fabrics and for mattresses. * Merinos are indigenous to the plains of Estremadura and Andalusia (Spain), where they were jealously guarded for some centuries, exportation being prevented. In the nineteenth century the Spaniards themselves introduced them into the Argentine, where three principal types were developed : Rambouillet, Negrette, and Lincoln, and a similar result followed the concessions made to France and Sweden. The English introduced them, with great success, into Australia and Cape Colony. Ths Electoral breed originated in merinos which were imported in ] 760 into the Electorate of Hesse, and spread into Silc.-sia, Saxony, Wiirtcmberg, Hanover, Moravia, and Hungary ; it now furnishes a large proportion of the raw material of German and Austiian wool factories. In England the C'/ieviot breed has assumed considerable importance and yields a long, yellowish wool, 682 wool is obtained from shoddy. 1 The lengths of wool fibres vary from 4 cm. to 30 cm. and the diameter from 0-014 mm. to 0-06 mm. The finer wools (merinos, Fig. 424) have as many as 13 waves per centimetre, while the more ordinary ones have only 3 (Fig. 424 B, natural size). The number of sheep in different countries in 1906 was as follows (in millions) : Australia, 72-8 ; Argentine, 74-4 ; Russia, 61-5 ; United States, 50-6 ; England, 29-2 ; New Zealand, 20-1; Uruguay, 17-9; France, 17-8; British India, 17-6; Spain, 13-3; Cape Colony, 11-8; Algeria, 9-1; Hungary, 8-1; Germany, 7-9; Italy, 11-2 (in 1908); Bulgaria, 6-9; Roumania, 5-7; Mexico, 3-4; Servia, 3-1; Austria, 2-6 ; Canada, 1-8; Sweden, 1-1 ; Norway, 1 ; Denmark, 0-9 ; Holland, 0-7 ; Natal, 0-6 ; Belgium, 0-2 ; Switzer- land, 0-2. The world's production of wool in 1903 was about 1,300,000 tons, namely, 450,000 in Europe, 140,000 in North America, 240,000 in South America, 2500 in Central America and the West Indies, 130,000 in Asia, 240,000 in Australia, 60,000 in Africa, and 23,COO in Oceania. For the separate countries the figures (tons) were as follow : United States, 130,000; England and Ireland, 60,000; France, 48,000; Spain, 4800; British South Africa, 47,000 ; Uruguay, 45,000 ; British India, 40,000 ; European Turkey and Balkan Peninsula, 30,000 ; Austria-Hungary, 29,000 ; Russia in Asia, 27,000 ; Germany , 23,000 ; Central Asia, 22,000 ; China, 16,000 ; Asiatic Turkey, 15,000 ; Algeria and Tunis, 14,000 ; Italy, 10,000 ; Venezuela, 7000 ; Portugal, 6000 ; Norway and Sweden, 3500 ; Chili, 3500 ; Mexico, 2300 ; Egypt, 1400 ; and Brazil, 700. The Argentine Republic exported less than 18,000 tons in 1860, nearly 66,000 in 1S70, about 98,000 in 1880, about 120,000 in 1890, and more than 200,000 (and 34,000 tons of skins) in 1895, while in 1905 the exports were about 191,000 tons of wool and 27,000 tons of skins, besides 120,000 live sheep and 3,325,000 frozen carcases. The wool exported from Morocco in 1908 was valued at 240,000, that from Algeria in 1906 at 720,000 (640,000 to France), and that from Tunis in 1906 at 100,000. The great market for wool in Europe is at Antwerp, and the price is fixed by auction, account being taken of the yields of the various wools (Conditioning, see later) after washing, some of them losing 40 to 70 per cent, of their weight owing to the removal of dirt, grease, &c. ; the normal or natural moisture, after washing and drying, is taken as 18-25 per cent. The price of raw wool varies somewhat from year to year and even in the same season not so flue as merino. Crossbreeds, obtained by crossing Argentines with Cheviots, arc also largely bred in England. The Russian breeds are derived from pure and Saxon merinos. The commonest varieties are the fitiugsk, aidarsk, rescetiovesk, and romanovsk (this is used for furs). In France the wool of the Burgundy and of the Berry uhighly valued. Italian wools, which were once famous, are now of little importance, and only Apulia, the Tuscan marshes, and the Roman province furnish a small part of the wool consumed in Italy. Good wool is also obtained from certain breeds of goats, such as those of Cashmir, which flourish in the Himalayas, nearly 5000 metres above sea-level. They furnish a very fine wool mixed, however, with much white or grey hair ; it is exported to France and Russia. The Thibetan goat, acclimatised also in France and in Bengal, likewise yields a valuable wool. The Angora goat of Asia Minor gives milk and a long wool (mohair) valued for its lustre, even after dyeing. The vicuna of the Peruvian, Chilian, and Mexican mountains gives a fine wool, used in certain cloths, which are now made partly from rabbit fur (the name vicuna or vigogne yarn is also applied to fabrics of wool and cotton which are quite distinct from vicuna wool). Alpaca is greyish, and is fuinished by a kind of tall, long-necked sheep (llama) indigenous to Peru. Camel-hair, which is worked like wool, has coarse fibres, and in its natural colour is woven into certain very strong textiles used, for instance, for the seats and curtains of railway carriages. 1 Shoddy is obtained by disintegrating woollen rags (previously sorted with respect to colour, and separated from those mixed with cotton) by means of an opener or devil, formed of a drum furnished with a number of steel points and rotating rapidly inside a second, fixed drum also provided with points ; from this the rags issue in short, flocculent fibres, which are carded and then spun. This industry, started in England in 1845 and since then extended to other countries, allows of the utilisation of- all woollen waste (fabrics and yarn) ; England alone imports from all parts of the world about 15,000 tons of woollen rags per annum. The coloured rags may often be partially decolorised by boiling them with 2 to 3 per cent, potassium dichromate and a little sulphuric acid. Admixed cotton (sewing and other) may be eliminated from the rags by so-called carbonisation, which consists in immersing the rags in sulphuric acid (4 to 5 Be\), centrifuging and heating them in ovens, the temperatme of which is raised to 120 to 140. In the course of an hour the cellulose of the cotton is transformed into brittle hydrocellulose and partly carbonised, so that it can be easily removed^by subsequent rubbing or by washing with water, this also carrying away the acid from the wool, which is not affected by such treatment. In some cases, hydrochloric acid vapour or aluminium chloride solution is used instead of sulphuric acid. The carbonised wool shows increased affinity for acid colouring-matters. Also woollen fabrics which contain bits or fibres of cotton in such quantity that it is impracticable to pick them out by hand, are carbonised with sulphuric acid or aluminium chloride before dyeing and are thoroughly washed after removal from the oven. Decolorised shoddy mixed with new wool can he recognised under the microscope owing to its different colour, which often recalls the original tint. Italy produces annually 100,000 to 120,000 quintals of shoddy. WOOL 683 from about Is. 2d. to 2s. per kilo. Australian wool is worth more than that from the Argentine. Unwashed wool (Australian weighs about 160 kilos per bale), after sorting, is washed with soap and soda at 45 to 50 in vessels (Leviathans) provided with loose forks for mixing and, when rinsed, is dried in revolving drums by means of hot air. The washed (or salted, such as Italian or Cape wool, weighing about 110 kilos per bale) wool is then carded or combed. In some districts the washing is preceded by treatment with benzene to remove the grease (see p. 393). The great European market for combed wool, not yet spun but wound into balls of 4 to 5 kilos (tups), is in France, at Roubaix (and also at Tourcoing and Lille), where prices are fixed at auction, although there is a considerable trade in combed wool at Bradford and to a less extent at Leipzig. These wools are classified, according to their fineness, as A, B, ... F, the first being the finest and the last the commoner sorts ; very fine wools are marked AA or AAA. Before being spun the washed wool is subjected to the operation of blending, i.e. the various qualities of wool (fine, ordinary, long, short, waste, shoddy, &c.) being mixed so as to obtain yarn of the desired count and fabric corresponding with the price and quality. To facilitate spinning and avoid felting, the wool is slightly oiled (with olive oil, commercial oleine, soap emulsion, &c., but not with non-saponifiable substances, such as mineral oils or resins, which would be difficult to remove from the fabric by washing, and would lead to irregular dyeing). In passing through the combs or cards, the various fibres are perfectly mixed and rendered parallel. The coarse strands (tops) are gradually converted into finer but not twisted strands, which are wound on bobbins (prepared) and are then, by means of ingenious, self-acting machines of enormous capacity, spun to the desired fineness to give, when twisted, yarn of the required count. 1 During spinning, the air of the room must be kept moistened with water vapour (see vol. i, p. 291) to prevent the parallel fibres from diverging and giving a non-uniform yarn. Satisfactory weaving also requires a certain degree of moisture. Italy's imports and exports of wool (raw, carded, combed, spun, woven, &c.) from 1905 to 1910 were as follow: 1905 1906 1907 1908 1909 1910 Imports, quintals Exports, ,, 184,936 59,164 202,834 47,996 228,626 38,862 257,808 28,348 265,643 39,351 278,432 (5,736,600) 41,697 (1,043,500) Woollen yarns of counts above 10 (international) pay 52s. per quintal on entry into Italy, while combed wool fabrics weighing less than 200 grms. per sq. metre pay 10 per quintal. The quantity of wool consumed (production + importation exportation) in Italy was 20,160 tons in 1886 and about 28,000 in 1905. The imports of raw wool into Japan were : 1700 tons in 1894, 9622 tons (1,028,000) in 1904, and 10,240 tons (1,880,000) in 1907. Chemical Properties of Wool. Pure wool consists of C, H, O, N, and S, the last varying somewhat in amount and being partly removed by repeated washing in boiling water. It is hence improbable that wool consists of a single chemical compound (it was at one time thought to be keratin, containing 4 to 5 per cent. S, but there appear to be other substances also). In 1888 Richard showed that the compounds forming wool contain NH 2 and NH groups. In a solution of alkali or a salt, wool fixes chemically part of the alkali or salt. Concentrated alkali dissolves wool, forming amino-acids, the most important being lanugic acid, which was isolated by Knecht and Appleyard and exhibits the same behaviour towards colouring-matters as does wool. 1 The Count of Yarn, either cotton or wool, is given by the number of kilometres weighing 1 kilo (international count) or half a kilo (French count). In Great Britain, the count represents the number of hanks of 840 yards (1 yard = 0-914 metre) per 1 Ib (453 grms.) ; hence English count No. 1 is equal to French count No. 0-847 and to international count No. 1-694. Division of the international count by 1-66 gives the English count, multiplication of the French count by 2 gives the international count, while division of the English count by 1-18 gives the French count. A thread spun from two yarns of count 60 has the count 30. its weight per unit length being doubled. Fine wools are spun so as to give a count of CO to 80 or even of 120, while the commoner qualities give counts of 'SO or even less. For till; the International Congress at Paris in 1900 accepted the Italian count, which expresses the weight in denari (one denaro = 0-05 grm.) of a length of -150 metres, tlic finer yarns thus having the lower counts. Silk is often spun to a count of 12 to 20 denari, and artificial silk to 60 to 120 denari. 684 ORGANIC CHEMISTRY It is probable, therefore, that wool contains at least one carboxyl group. The affinity of wool for acid colouring-matters (often sulphonic acids) is explained by the presence of amino-groups and that for basic dyes by the presence of the carboxyl group. Certain highly basic colouring -matters (such as methyl green) do not, however, colour wool, the acid character of which is too weak, while they colour silk, which is more markedly acid. The fixation of metallic oxides (of Cr, Fe, Cu, Al, &c.) in the mordanting of wool is due to the formation of salts with the carboxyl group. The salt-forming property of wool can be easily demonstrated by immersing it in a hot colourless solution of rosanilinc (base), which colours it red just as though it were dyed with red rosaniline hydro - chloride. Knecht, Witt, and Nilsen have shown that the action of chlorine on wool is to intensify its acid character, so that it fixes basic dyes the more readily ; at the same time it loses partially its capa- city to felt. Bolley found that wool de- composes potassium bitartrale in boiling solution, generating the neutral tartrate and lixing tartaric acid. In 1898 Kertesz utilised industrially, for the simultaneous production of two colours on wool, the property this shows of fixing acid colour- ing-matters more intensely at points where it has been care- fully treated with caustic soda, the latter neutralising the carb- oxyl group and thus rendering the basic character more pro- nounced. Wool loses much of its affinity for acid colours when treated with phosphotungstic acid, but recovers it when subjected to the action of ammonium bicarbonate (Scrida, 1909). Of practical importance is the behaviour of wool (or cotton) waste containing ordinary oils or fats (not wool-fat), as it readily ignites owing to energetic oxidation and causes fires (see Pyrophoric Substances, vol. i, p. 174). An aqueous extract of pure wool gives a precipitate with either tannin or basic lead acetate, while true glue or gelatine yields no precipitate with the latter reagent. Pure wool contains 14 per cent. N. COTTON is the white down surrounding the black cotton-seed and is contained in capsules (each weighing about 30 grms., 10 grms. being cotton) which, to the number of 300 to 400, form the fruit of Gossypium a shrub 2 to 4 metres in height (see Fig. 426). When the fruit is ripe (in America in August), the capsule opens and throws out a white tuft of cotton, which is fixed to the seeds. After harvesting, the cotton is freed from seeds by means of cotton-gins and compressed hydraulically into bales holding 180 to 200 kilos. Cotton, is produced most abundantly in North America and, to s a less extent, in South America (Brazil, Peru, Colombia, &c.), and the Antilles (Haiti, Cuba, &c.). Its cultivation is also of importance in the East Indies, Syria, Macedonia, &c. Egyptian cotton (makb) is valued on account of its lustre and length of fibre. Cotton is also grown in Australia. Attempts have recently been made to cultivate it in the Italian colony of Eritrea, but without great success. FIG. 426. MERCERISED COTTON 685 The best qualities have fibres 30 to 40 mm. in length and the lower qualities 10 to 14 mm. The fibres are 0-015 to 0-020 mm. in thickness and under the microscope have the appearance of flattened ribbons with a twist here and there (Fig. 427, the upper part of which shows the transverse sections). When treated with ammoniacal copper oxide solution, cotton swells very considerably, forming superposed capsules sepa- rated by constrictions (Fig. 428). By cold concen- trated caustic soda solution (30 to 35 Be.) the flat fibre is converted into a cylindrical one almost circular in section (Fig. 427 / ; see Mercerisation) with a thin central channel. If immersion in the soda is prolonged for two or three minutes, during which the skein or fabric is kept stretched, and the soda is subsequently washed away while the tension is maintained, the skein will not contract and the fibres present a lustrous appearance (mercerised cotton) and are stronger and heavier than in their original state (soda-cellulose and then hydrocellulose are formed). 1 The chemical characters of cotton are those of cellulose described on p. 503, purified cotton being pure cellulose. For its behaviour towards different dyes, see p. 651, and also later. The world's production of cotton is about 3,500,000 tons per annum, about three-fifths of this quantity being given by the United States, which exports about 1,700,000 tons, nearly one-half to England, about one- quarter to Germany, and the remaining quarter to 1 History and Properties of Mercerised Cotton. In 1844 J. Mercer, chemist in a Lancashire calico-printing works, having Flo. 427. filtered a concentrated caustic soda solution through a cotton filter, (Magnified 300 timos) noticed that the cloth had contracted somewhat and had become thicker and transparent. Before filtration the liquid had the sp. gr. 1-300, but after filtration only 1-265. On studying the phenomenon more closely, Mercer found he could reproduce it at will with yarn immersed in caustic soda solution of 20 to 30 Be 1 ., while he estab- lished with certainty that, under such treatment, the cotton fibre shortens by 20 to 25 per cent., thickens and becomes stronger (by about 50 per cent.) and of increased affinity for colouring-matters. He showed, too, that the phenomenon is more rapid and more intense at low temperatures, while at the boiling-point no contraction occurs. Similar changes are produced by treating cotton with sulphuric acid of 50 to 55 B6. or with zinc chloride solution. In October 1850 Mercer was granted an English patent (13,296) for increasing, by this treatment, the resistance and compactness of cotton and its affinity for dyes. In 1884 P. and C. Depoully patented a process for the partial morcerisation of fabrics by which parts of the fabric were brought FlG. 428. into contact with an alkali solution ; these parts contracted and caused (Magnified 200 times) the other parts to curl, beautiful crape effects being thus obtained. In 1896 the textile world was astounded to see on the market samples of fine cotton of the most brilliant colours and the lustre and feel of silk. This product was prepared by the great dyeing firm Thomas and Prevost of Crefeld, according to their German Patent, No. 85,564 of March 24, 1895, which reads: ". . . . improvement in the mercerisation of vegetable fibres with alkaline or acid solutions, by subjecting the tightly stretched yarn or fabric to the. action of alkali (caustic soda of 15 to 32 Be.), or of acid (sulphuric acid of 49-5 to 55-5 Be.), the stretching being maintained until washing is complete when it is relieved spontaneously and the shortening of the yarn or fabric thus prevented." The specification does not refer to the lustre assumed by the yarn, but this is mentioned in a later addition. These Thomas and Prevost patents were, however, annulled a couple of years later in all countries, since various competitors found that an identical process had been patented (No. 4452) in England in 1890 by H. A. Lowe but had not been renewed within a year because Lowe could not find an English manufacturer disposed to make prac- tical use of it. Large quantities of mercerised cotton are now freely produced in all countries. The shortening of the fibre and its increase in resistance produced by concentrated alkali solution may be under- stood if the changes occurring in the fibre itself are followed under the microscope. While the fibre of ordinary cotton is seen to be a flattened empty tube with an occasional twist, that treated with caustic soda without stretching is shortened and swollen and forms an oval, almost round tube with thickened walls, but still with an internal channel ; outside it shows creases and a rough surface. But by mercerisation under tension, the fibre becomes like a straight, round tube, smooth and without visible creases outside and almost entirely filled up inside. These :hanges explain the silky lustre and also the increased strength, the fibre becoming more compact. Buatroek'i experiments showed that mercerisation occurs very rapidly : with caustic soda of 30" Be. the shortening of the fibre after one minute is 23 per cent, and after 33 minutes 29 per ceut., which is the maximum attainable. 686 ORGANIC CHEMISTRY the rest of Europe) especially France, Austria, and Italy). The cotton exported from the United States in 1901 represented a value of 62,000,000. J Mexico produces 45,000 tons ; Egypt, about 250,000 ; British India, 450,000 ; Japan, 30,000 ; and in the South of Italy there are about 12,000 hectares under cotton, about 5000 tons being produced. In 1899 Italy imported about 130,000 tons of cotton, mostly from the United States. The import duty in Italy is 29d. per quintal, yarns paying from 14s. to 48s. according to the count and fabrics 48*. to 104s. ; on exported yarns and fabrics the Government grants a rebate of 3s. per quintal for yarns and 3s. 6d. for fabrics per quintal (freed from dressing), the weight of the cotton being increased by 8 per cent, to allow for its natural moisture. The conversion of cotton from flock to yarn is effected by carding or combing in a similar manner to shoddy (see above). Very fine counts (150) are spun in some countries, but in Italy, where at one time 30 was the finest, 60 and 90 are the usual ones, although 130 is sometimes obtained. ... The immense importance of the cotton industry is shown by the Table on page 687. which refers to the year 1905 (in the previous year the production was 13,635,000 bales). In one of the cotton mills of the United States 1 34 workpeople are sufficient to overlook 2000 Northrop looms, a clever workman attending as many as 20 looms, while with the less expert the number never falls below 12 ; these looms make 165 strokes per minute with good warp and weft. In Italy the consumption of cotton yarn and fabric amounted in 1905 to 813,000 quintals, or 2-5 kilos per head of the population. As a result of the Turko-Italian War Italy has lost the cotton trade in the Near East, which it had previously captured in com- petition with England and Germany. Japan in 1903 with 4933 looms produced 69,800,000 metres of cotton fabric, the exports being valued at 720,000 ; in 1905 7128 looms turned out 104,500,000 metres, exports W. Vieweg (1908) determines the degree of mercerisation by a method based on the fact that, in 13 to 24 per cent. NaOH solution, cotton fixes an amount of NaOH corresponding with (C H 10 O D )jNaOH, while in a 40 per cent, solution it fixes double this amount, (C 6 H, O 5 ) 2 2NaOH. This soda-cellulose loses its soda when washed, and the recovered cellulose has the property of taking up more or less caustic soda in a 2 per cent. NaOH solution, non-mercerised cotton fixing 1 per cent., and mercerised 1 to 3 per cent, of NaOH according to the degree of previous mercerisation. In practice this degree of mercerisation is ascertained as follows : 3 grms. of the dry mercerised cotton are shaken for an hour with 200 c.c. of exactly 2 per cent. NaOH solution in a separating funnel, 50 c.c. of the solution being then titrated with seminormal acid and the amount of NaOH absorbed by the cotton calculated. A qualitative test for detecting mercerised cotton mixed with ordinary cotton and oxycellulose was given on p. 506. To ascertain if a fabric is mercerised H. David (1907) places a drop of concentrated soda on the fabric, which is then washed and dyed with a substantive dye ; a more intense colouring on the place touched by the soda indicates that the original fabric was not mercerised. When cotton is mercerised with tension its strength increases by 35 per cent., and when mercerised without tension by as much as 68 per cent. The elasticity is greater in cotton mercerised without tension (27 per cent.), while with cotton mercerised under tension it is unchanged (20 per cent.). The lustre of mercerised cotton is not altered by washing or dyeing. In order to obtain satisfactory results and a good lustre by mercerising, it is best to use long-fibred cotton ; the shorter the fibre the greater must be the tension. It is also necessary to boil the cotton thoroughly and wash it completely before placing it in the caustic soda bath, as otherwise, besides obtaining less lustre, there is great danger of irregular dyeing. The dyeing is carried out in the usual way with basic dyes, being preceded by mordanting, or, better, with substantive dyes in baths containing a little Turkey-red oil or soap, the temperature being kept low at the start to avoid non-uniformity. Old caustic soda baths, which become largely converted into sodium carbonate and so diminish in activity, can be used for soap-making. To impart a silky feel to mercerised cotton, the latter is well washed, immersed for a few minutes in a calcium acetate bath at 0-5 B6., pressed, introduced into a bath of Marseilles soap (1 grm. per litre), again pressed, placed in an acetic or tartaric acid bath (10 grms. per litre) and finally pressed and dried without washing. Mercerised may be distinguished from unmercerised cotton by immersion in a solution of 5 parts KI + 20 water + 2 iodine + 30 ZnClj in 12 water. All cotton is thus coloured blue, but thorough washing with water decolorises only that which has not been mercerised (H. Lange, 1903). 1 The increase in the production, consumption, and exportation of cotton in the United States is shown by the following figures [in 1874 the production (home consumption + exportation) was 3.500,000 bales of 500 lb.] : Home consumption Exports Imports bales bales bales 1903 . . v . . 3,980,567 6,290,245 1904 . ... 4,523,208 9,119,614 1905 ... . 4,877,465 6,975,494 1906 ... . 1,974,199 8,825.237 202,733 1907 . . . 4,493,028 7,779,508 140,8fi9 1909 . . . 10,300,000 1910 . . . 12,000,000 In the United States in 1907 cotton-seeds gave also 660,000 tons of oil and 1,490,000 tons of pressed oil-cake for cattle-food. These products were partly exported oil to the value of 3,400,000 and cake to the value of 2,320,000. FLAX 687 Consumption Country umber of Spindles Looms Workpeople in bales of mills 200 to 225 kilos England . 2207 50,964,874 704,357 550,000 3,640,000 United States, North 573 14,810,164 340,682 197,137 2,167,700 South 659 8,050,879 174,324 120,000 2,203,406 Russia . , 227 6,554,577 154,577 350,000 1,177,000 Poland . 56 1,268,547 12,000 35,000 325,000 Germany 870 8,832,016 211,818 350,000 1,761,369 France . 420 6,150,00(1 206,000 90,000 840,000 Austria ... ISO 3,280,330 110,000 100,000 650,000 Hungary 3 103,400 Switzerland 68 1,711,300 17,385 19,000 100,000 Italy . 760 2,435,000 110,000 139,000 560,000 Spain . , 257 2,614,500 68,289 300,000 Portugal 15 160,000 Syria 35 372,000 10,000 80,000 Norway 9 87,832 2,293 2,635 12,000 Denmark 3 6.0,000 18,000 Holland. . 23 376,234 20,100 17,000 67,000 Belgium 43 1,222,138 24,000 15,000 100,000 Roumania 40,000 Turkey . . . 5 80,000 ':'!* 23,000 Greece . 970,000 2,100 15,000 Asia Minor 4 60,000 18,000 India 191 5,119,121 45,337 184,779 1,744,766 China 15 620,000 2,200 Japan . . 64 1,332,000 68,261 900,000 Brazil . 142 450,000 23,000 20,000 250,000 Canada . 22 773,538 18,267 10,000 99,000 Mexico 114 628,096 20,287 26,000 140,000 Total 6224 119,127,146 2,117,016 2,295,120 17,511,241 being 1,360,000, and in 1907 9260 looms made 125,000,000 metres, the exportation amounting to 1,880,000. FLAX (Linum usitatissimum) is a herbaceous annual, growing usually in temperate regions, and reaching a height of 60 to 80 cm. (Fig. 429). It bears clusters of blue flowers which give capsules (Fig. 430, 2) containing flattened lenticular seeds (Fig. 430, 1). It was cultivated first in Egypt, then in Greece, and later in Italy and various other parts of Euro[ j (Belgium, Holland, Russia, &c.) ; in Italy the cultivation has diminished very considerably, although it is still followed in some parts and is carried on in the south of Sicily for obtaining the seeds. There are two ordinary varieties which are grown for both fibre and seed : autumn or winter flax, which has a coarse fibre and is sown in October and harvested at the end of spring, the ground being left free for another crop ; and that sown in March, which is pulled in the summer when the seeds begin to brown but are not quite ripe. Flax plants are pulled by hand and arranged in sheaves to dry and to mature the seeds. After removal of the latter by threshing, the plants are made into large bundles, which are left for 15 to 20 days in stagnant water, where the action of micro-organisms (Amylobacter, butyric bacteria) results in the dissolution of those parts of the tissues which unite the long fibres to the cortex and to the pith. The bundles are then opened and dried in the field. Instead of being retted in this way, flax is in some countries heated in large autoclaves for half an hour at 125 with water from a preceding operation and then for an hour with steam at a pressure of 5 atmos. The dried flax is freed from the friable cortex by bruising between sticks, the operation being completed by blows from scutching knives (the waste forms the tow). The flax is then combed and placed on the market in large, 688 ORGANIC CHEMISTRY twisted tresses of 200 to 300 grms. at 144s. per quintal or 80s. to 96s. for short fibre. In Italy, a hectare of winter flax yields about 300 kilos of fibre and 900 kilos of seed, March flax giving 200 and 700 kilos respectively ; in Ireland, Belgium, and Germany double as much fibre is obtained. The world's production is about 6,000,000 quintals, more than one-half of which is produced in Russia (where 1,000,000 hectares are under flax and two-thirds of the output is exported), while Germany produces about 450,000 quintals //,_, (importing 600,000 and exporting 250,000), Austria'-Hungary 400,000, France 400,000, Belgium 250,000. North America about 200,000, Italy (from 52,000 hectares) less than 150,000 (with 100,000 spindles for flax and hemp), and England about 120,000 quintals. England, however, imports 700,000 quintals of flax (two- thirds from Russia and one-third from Holland and Belgium) to supply its 1,500,000 spindles (three-fourths in Ireland for fine yarn and one- fourth in Scotland). The cultivation of flax is falling in all countries except Russia. Thus, France had at one time 120,000 hectares under flax but now has only 20,000 (in spite of Government awards of 100,000 annually to encourage its growing), about 800,000 quintals being im- ported (four-fifths from Russia) to supply its 700,000 spindles, 20,000 hand looms, and 22,000 power looms. Italy has not more than 50,000 hectares under flax, and for the manufacture of fine fabrics imports annually about 40,000 quintals of fine or semi-fine flax and about 1400 quintals of undressed flax. The flax fibre has a diameter of 0-02 mm. and is readily distinguishable under the microscope from other vegetable fibres (Fig. 431 : 1, spiral stria- tion ; 2, extremity of the fibre and polygonal section ; 3, bruised places). The fibre is spun into yarn in the same way as with cotton, but special machines are used for the recombing and repreparing of coarse fibres, which are drawn out in the moist state to a finer thread, and, at a certain stage, twisted. The tow from these operations is worked up by carding (see Shoddy). Flax can be spun by hand to a count of 300, but by machinery only to 200 ; certain qualities of flax can be hand-spun, for very fine work, to a count of 1400, such yarn costing as much as 80 per kilo. HEMP (Cannabis saliva) belongs to the order Cannabineae and bears male and female flowers on different plants (dioecious). When growing wild it branches (Fig. 432), but when cultivated for industrial purposes it grows to a height of 2 metres or more without branching and has a finer and closer tuft in the case of the female plants (Fig. 433). Of the different varieties of hemp (jute, Manila, New Zealand, and ordinary), the most important is the ordinary. It is sown very close in heavy, deeply worked soil, and is gathered in August, the plants being dried in bundles on the ground. The treatment is similar to that of flax, but with a more protracted maceration. The residue from the breaking is used to some extent in paper-making ; the hemp, more or less combed, is twisted into tresses like flax and made up into bales of 150 kilos. Hemp fibres have a Fro. 431. iiiflrd 200 times) J IT T E 689 diameter of 0-04-0-05 mm. and are easily distinguished microscopically rom other fibres (Fig. 435 : 1, displaced fibres ; 2, a-d, form of the tip of the fibre ; 3, section of a bundle of fibres ; 4, striation : the crossed transverse lines are not always seen, the parallel FIG. 432. FIG. 433. FIG. 434. longitudinal striations being more common). The long stems are cut into three lengths of about 70 cm. and are combed first by hand and then by a machine with long, coarse points, the waste forming the first and second tow, which can be subsequently carded. A third combing is carried out with finer and closer teeth, the coarse and then the finer ribbon being passed through machines similar to but coarser than those used for cotton and wool (preparing), and finally twisted for coarse twine yarn, for canvas yarn (count of 7 to 10), &c. Two twines twisted together give a string, several strings combined and twisted form a rope, and several ropes a cable. As well as for string, rope, &c., hemp is largely used for making coarse-, strong cloth for bags, waggon covers, sails, &c. In order to render hemp fabrics more compact and durable, they are sometimes mercerised. The output of hemp in Europe is less than 4,000,000 quintals, 1,000,000 coming from Russia, 960,000 from Italy, 750,000 from Austria-Hungary, 600,000 from France, 20,000 from Belgium, and 10,000 from Holland. Italy exports nearly 40,000 quintals of string and TIG. 435. rope, 35,000 of rough hemp and flax, and 1200 (Magnified 200 times) of twisted hemp and flax. JUTE (Corchorus capsularis of the order Tiliacese) has been grown on an enormous scale in India and Bengal from time immemorial and is now replacing indigo. Even in 1851 India exported 282,350 quintals, and in 1858 the exports of jute sacks were valued at almost 240,000. These figures are now nearly doubled, owing to the development n 44 690 ORGANIC CHEMIST R Y of the large works in Calcutta. In Europe its cultivation was commenced subseqxiently to 1830. It is grown also in. South America and in the United States. Jute requires a moist, hot climate and soil. It is sown in spring, and the plants, 15 to 20 cm. apart, mature in four months and attain a height of 3 to 4 metres. The shape of the leaves, stem, seeds, &c., is shown in Pig. 436. It is treated in a similar manner to hemp, and the bales, weighing 180 kilos, are tightly pressed for transport. The principal European, centre of the jute trade and industry is at Dundee. The jute fibre is brownish yellow, and is bleached in a faintly alkaline chloride of lime bath (5 Be.) at 25 to 30, then rinsed, immersed in a 0-5 per cent, sulphuric acid bath for 15 minutes, and finally thoroughly washed. Raw jute fibres are easily distinguished from other fibres under the microscope (see Fig. 437 : 1, irregular lumen of the fibre dotted at the top ; 2-, fibre with broken lumen ; 3, tip of fibres ; 4 and 5, sections of fibre with thin or thick walls) and show more or less lustre according to their fine- ness. Jute competes directly Avith hemp since it serves for making the same articles (sacks, packing cloth, carpets, tents, furniture coverings, &c.), but when made of jute these cannot be washed. In 1901 Italy imported 249,000 quintals of raw jute to be manufactured and 4000 of jute yarn (some again exported), and exported about 15,000 quintals of jute tissue (jute, flax, and hemp fabrics are highly protected in Italy, the duty ranging from 8s. to 16s. or even more for the finer counts and for tissues). The consumption of jute in different countries is as follows: England, 1.280,000 bales (of 180 kilos) ; India, 1,200,000 ; United States, 540,000 ; Ger- many, 450,000 ; France, 260,000 ; Austria-Hungary, 170,000 ; Italy, 120,000 ; Belgium, 100,000, &c. Raw jute in bales costs 28s. to 36s. per quintal. SILK. The Chinese seem to have known the silk- worm as early as 2600 years B.C. Although they understood the preparation of silk materials, they did not at once trade with other races, but maintained great secrecy on the rearing of silkworms and strictly prohibited the exportation of the eggs. According to tradition it was only in 150 B.C. that silkworms arrived in Japan, where they were imported secretly by the daughter of a Chinese emperor, and whence they spread later throughout the rest of Asia. They were apparently imported into Italy in the sixth century by three monks who hid them in their staves, although the manufacture of imported silk was begun in Italy three centuries earlier. From that time up to the present Italy has maintained the first place among the countries of Europe for tin- rearing of silkworms and the production of silk. 1 \ 1 Silk is produced by one of the Lepidoptera, Bombyx mori, a larva which after birth (when it weighs about 0-5 mgrm.), feeds on mulberry leaves (Moms alba) and atta.ns the height of its development (with a weight of 3 to 5 grins.) in five weeks, passing through four moults or sleeps during which it casts its skin. It finally passes to brushwood arranged above, where it constructs a cocoon with the silky exudation secreted by two long glands tilled with fibroin and leading along the body beside the intestinal canal to two very fine apertures in the mouth. The two contiguous and parallel threads thus formed are immediately stuck together by a liquid (sericin) exuded by two other channels near the first pair, the result being an apparently single thread, which is Fic.436. f - FIG. 437. (Magnified 200 times) S I L K W O R M C U L T U R E 601 In 144;} Florence contained 84 large silk factories and in 1580 Milan began to acquire the ascendancy, but fell back later, to advance again in the middle of the nineteenth either white or some shade of yellow (the double thread is shown in Fig. 438). In three days the silkworm is trans- formed into a chrysalis from which the butterfly originates (in 10 to 14 days) if the temperature ia sufficiently high (15 to 30). The butterfly emits from its month an alkaline liquid with which it moistens one end of the cocoon and then perforates it and issues to proceed to the coupling necessary for the preservation of the species. Immediately aff pnvards the female di'ixisits numerous fertile eggs (qraine), and both it and also the male die, ( heir short life-cycle being at an end (Fig. 439). One kilo of cocoons gives three ounces of eggs. Part of tho eggs (or of the butterflies) are selected under the microscope and are kept in a. cool place until the following spring, when they are hatched by incubating for a couple of weeks in an oven, the young worms being distributed to the rearing-houses. In 1904 Italy exported 1521 kilos of eggs, of the value of 20,240 ; in 1908 9228 kilos ; in 1909 2885 kilos, and in 1910 3330 kilos, worth 40,000. The imports of graine (from France) were 4178 kilos in 1906, 18,928 in 1908, 13,629 in 1909, and 5612, worth 44,880, in 1910. By moans of extreme cleanliness, disinfection of the brushwood and microscopic tests of the eggs, the numerous diseases which cause havoc among silkworms at all stages (calcino, flacherie, 8, while Lyons possessed 10,000 looms as early as 1685, 40,000 in 1834, and 05,000 in 1852 (present con- ditions are indicated later). Raw silk consists of 60 to 70 per cent, of Fibroin (the fundamental constituent of pure silk) and 25 to 35 per cent, of Sericin, which is the with hot soap solution (80 to 85) containing a little sodium carbonate, and are then well rinsed in tepid water. 1 If the wares are to remain white, they are sulphured (see Note) or treated with hydrogen peroxide solution, the characteristic rustle (scroop or crackle) of silk being imparted by immersion in a 1 to 2 per cent, sulphuric or acetic acid bath, centrifugation and drying without rinsing. Dyeing is in general carried out in soap baths, using one-third or one-fourth of the soap solution remaining after the boiling of the raw silk, acidifying it with sulphuric acid, boiling and agitating. The silk is immersed in this emulsion for a time and then removed, the bath being diluted with water and the colouring -matter (acid or basic) ; the dyeing is begun at 35 to 40, the temperature being gradually raised almost to the boiling-point. Acid colouring-matters are fixed by silk also from a hot acidified aqueous solution, but the tints are not so lasting. The dyed silk is rinsed in water and transferred to the acid bath to obtain the crackle, which becomes more pronounced as the acidity and temperature of the bath are raised, but the acid remaining in the dry fibre slowly attacks it, with injury to its tenacity and elasticity. Nowadays silk is usually weighted, i.e. impregnated with various substances (organic and inorganic), in order to increase its weight (by 30 to 40 per cent, and sometimes, with black silk, even by 300 per cent, or more). Silk possesses, indeed, the property of absorbing from solution large quantities of tannin ; this can be fixed by means of salts, and fresh tannin can then be absorbed, and so on. Successive amounts of insoluble metallic salts (tin salts, phosphates, silicates, &c.) may also be precipitated on silk. To weight white silk, the boiled silk is soaked for an hour in a stannic chloride bath of 25 to 30 Be. [at one time pink salt, SnCl 4 ,2NH 4 Cl (see vol. i, p. 609) was largely used, but at the present time, crystallised tin salt, SnCl 4 ,5H 2 O, is mostly employed], manipulated for 30 to 40 minutes in a hot disodium phosphate bath (4 to 5 Be.), washed slightly with water, introduced into a sodium silicate bath (3 to 4 Be. ) and again washed. Treatment with this series of baths (stannic chloride, phosphate, and silicate) is repeated several times, according to the degree of weighting desired ; five such repetitions give a weighting of 100 to 120 per cent, (the weight being doubled). 2 Weighted silk can be dyed, and in the preparation 1 It is generally necessary to ascertain, before dyeing, what will be the loss in weight of the silk during ungum- ming or stripping. White Italian silk loses on an average 21-5 per cent. ; Japanese, 20 per cent. ; Canton and Chinese, 24 per cent. ; raw yellow Italian, 24 per cent. ; and cliappe, 4 per cent. The loss, which includes also any weighting of the yarn with vaseline, soap, oils, glycerine, &c., is determined as follows : 50 grms. of the silk are manipulated in a solution of 15 grms. of seasoned Marseilles soap of good quality in a litre of hot water, which is allowed to boil gently for half an hour, and are then removed, pressed or centrifuged, boiled for a further period of 30 minutes in a soap bath similar to the first, and washed thoroughly with water until the latter remains clear ; after being centrifuged, the silk is dried in an oven until of constant weight. The loss of weight on stripping is referred to 100 grms. of dry silk, so that allowance should be made for the normal humidity (11 per cent.) of silk. 2 The phenomenon of iveighting is explained, according to Sisley (1911), by regarding silk as a colloid (see vol. i, p. 102), which absorbs hydrogels (e.g. stannic) of various salts of polybasic acids. But many substances which give precipitates and insoluble salts do not serve for weighting, since they are not firmly retained by the silk fibre and are therefore eliminated during washing and dyeing and are not dyed. The weightings which have given the best results in practice are : (1) tin hydroxide (used as early as 1869 in a Lyons dyeworks) ; (2) tin phosphate ; (3) tin silicophosphate ; (4) tin and aluminium silicophosphates. Sisley (1896) showed, and Franckel and Fasal (1897) and Sevcrini (1906) confirmed, that weighting is due purely to a physical and not to a chemical phenomenon, since the weighting bath undergoes no chemical change and no alteration in concentration. Further, when silk soaked in stannic chloride is washed with water, the precipitated stannic hydroxide which is formed in abundance as a result of hydrolysis is not fixed by the silk and is derived from the chloride on the surface of the thread, that absorbed inside the fibre remaining as a kind of colloidal solution of stannic hydroxide in hydrochloric acid ; the acid diffuses into the fibre, which retains it, whilst the stannic hydroxide is fixed as a gel and does not influence the feel and lustre of the silk. The absorption of stannic chloride is avoided if the silk is previously treated with tannin. In 12 hours silk which has absorbed II per cent, of tannin fixes from a stannic chloride bath of 30 B6., only 1-25 per cent, of SnO 2 , while silk without tannin fixes about 12 per cent, of SnO 2 from the same bath ; these different silks also take up varying quantities of colouring-matters. When washed, the stannic hydroxide formed on the fibre is Sn(OH) 4 or ,Sn 34,590 48,050 50,020 India, exported from Cal- 1 cutta and Bombay . *> (mo 4,380 261 2,930 2,560 325 World's total, quintals . 94,380 1 Hi. 000 1.V2.950 170,530 190,920 209,130 242.000 raw silk is calculated to lose '24 per cent, on stripping, the weighting will be 0-95 : 0-85 = 76 : x (76 is the percentage of silk remaining after stripping) and x = 68 hence the dyed dry silk contains 76 parts of dry stripped silk (or 100 of raw silk) and 08 of weighting, total 144. The silk was hence weighted 44 per cent. Gianoli and Colomba (190.7) showed, however, that in some eases when metustannic acid is formed on the fibre, e.g. by the fixation of 1 in salts with sodium carbonate, the whole of the weighting is not eliminated by hydrofluoric acid, even when this is followed by a bath of I1C1. A more certain result is then obtained by the old method (see abuce) or by using Urst soda and then potassium hydrogen oxalate. P. Hermann (1909) proposes to modify the alternate treatment with hydrochloric aeid and caustie potash (Kistenpart, 1908) of black on tin salt and eatechu, by replacing thu caustic potash with a solution of normal caustic potash and concentrated glycerine (28 B6.) in equal parts, the latter preserving the silk, readily dissolving Prussian blue (by treatment for an hour in the cold or 10 minutes at 80*0, but leaving the oxide and tannate of iron unchanged. 696 ORGANIC CHEMISTRY The world's production was 18 millions of kilos in 1903 ; 20-5 in 1904 ; 18-5 in 1905 ; 21 in 1906 ; 22 in 1907 ; 24 in 1908; and, in spite of diminished European production, 24-2 in 1909 (15-7 from the Far East, 3-1 from the Levant, and 5-4 from Europe). In China the exportation of real silk tends to diminish, but that of wild silk (or tussah) increases ; this is produced by Anterea mylitta and is readily recognised under the micro- scope (Fig. 441 ). China exported 1 .2(10,000 kilos in 1900 ; 1 ,325,000 in 1903, and 2,000,000 in 1904. To the quantity of raw silk produced in Italy from home-grown cocoons must be added that obtained from cocoons imported from abroad, viz. 3000 quintals in 1893 ; 7320 in 1898 ; 11,000 in 1903, and 13,000 in 1906. The mean annual importation from 1901 to 1905 of cocoons (calculated dry) was 37,736 quintals (46,000 in 1906) with a mean yield of 1 kilo of silk per 4 kilos of dry cocoons (at 7*. to Qs. M. per kilo) or per 11-5 kilos of fresh cocoons. To the 60,000 quintals of raw silk yarn produced in Italy must be added 24,000 quintals of silk simply treated and imported from the Far East to be spun and twisted. But only about 10,000 quintals are woven in Italy, the rest being exported (50,000 quintals of raw silk and 39,000 of twisted). It is estimated that in 1903 there were more than 61,000 basins (against 54,000 in 1891) ; about 95,000 operatives for treating the silk (100 basins require 10 to 12 quintals of fuel per day); more than 1,667,000 spindles (against 1,500,000 in 1891) with 54,000 workpeople ; and 20,000 looms (of which one-half are power-looms running at 100 to 160 picks per minute, and the remainder hand-looms at 50 to 60 per minute) with 30,000 workpeople (against 10,000 looms in 1891), almost all of these being in the province of Como and neighbouring districts. Eighty per cent, of the workpeople are women. The Italian weaving industry is capable of considerable extension, its produce being valued at only 3,200,000, while Switzerland 1 (with 35,000 looms) produces silk fabrics to the value of 5,600,000, France 2 (with 140,000 looms) 19,600,000 ; England about 13,600,000 (im- FIG. 441. porting 8,800,000) with 87,000 looms, and about the same for Germany. If Italy were to weave the 8,000,000 worth of yarn which it exports, the value would be increased to 16,000,000 (a kilo of fabric costs about double as much as a kilo of yarn) while 200,000 more workpeople would be employed. Mention has been made of the silk waste industry in Italy on p. 692. During the past twenty-five years the silk-weaving industry has become of considerable 1 Switzerland has two very important centres at Zurich and Basle, where the output of silk goods is continually increasing, although the production of cocoons is gradually diminishing. In the canton of Tieino, where the silk- worm is reared, the cocoons produced have diminished from 187,500 kilos in 1872 to 58,000 in 1904, while there lias been a "Corresponding increase in the importation of raw silk from China, Japan, and Italy. This importation rose from 514,400 kilos in 1893 to 637,000 (worth 960,000) in 1902, but about one-third of this, after being twisted in the Swiss factories, is exported to Germany, Russia, and Italy. In the canton of Zurich alone in 1900 there were at work about 21,000 hand-looms and 13,330 power-looms for silk and mixed silk fabrics. The Swiss exports of pure silk tissues in 1893 were 966,700 kilos (2,506,100), those of mixed tissues being valued at 580,000. In 1903 the exports of silk fabrics were 1,760,300 kilos, worth 3,780,000, while the total imports in the same year were 149,000 kilos (330,800) of silk fabrics and also mixed fabrics to the value of 112,000. One-half of the exports goes to England. The silk ribbon and embroidery industry of Switzerland is steadily advancing. Germany is a large importer of raw silk (about 3,000,000 kilos, largely Italian), and, besides supplying home demands, exports considerable quantities of manufactured goods (see Table later). Russia consumes about 1,500,000 kilos of raw silk annually. a None the less interesting is the condition of affairs in France, although the production of fresh cocoons is only 8,000,000 kilos (1905). The imports of raw silk are calculated to be about 9,000,000 kilos, and the silk industry (almost entirely concentrated in the city of Lyons) occupies one of the foremost positions among French industries. The province of Lyons contains more than 25,000 power-looms for silk-weaving, in addition to a larger number of hand-looms. In order to reduce the importation of raw silk and increase that of cocoons, and so encourage the direct spinning of the latter, the French Government in 1892 offered a premium of 16 for every new four-threaded basin established, but the results did not come up to expectations. While in 1893 the production of silk goods was valued at 15,150,000, in 1902 it reached 17,800,000. The French exportation of silk wares of all kinds amounted in 1896 to 4,220\000 kilos, worth about 10,000,000, while in 1904 it rose to 5,700,000 kilos, of the value of 13,200,000 (including about 1,200,000 worth despatched by parcel post). The value of the products \yoven in Lyons in 1904 was 16,360,000. in 1905 15,640,000, and in 1906 17,040,000. In the department of Saint-Ktienne the output of silk ribbon in 1906 was valued at 3,760,000, one-third of it for export. The French home consumption of silk wares is about 4,000,000 kilos, this large amount helping considerably to maintain the silk industry in an active condition. ITALIAN SILK INDUSTRY 697 importance in the United States, where raw silk is almost free from customs duty, while the manufactured products (yarn and fabric) are very heavily taxed. These conditions have led to the rapid development of American spinning and weaving. 1 The importation of raw silk into the United States shows continuous and rapid increase, the annual averages being: 1881-1885,15,300 quintals; 1886-1890,23,100; 1891-1895,31,300; 1896-1900, 43,500 ; 1901-1905, 65,300 quintals, which is about one-third of the world's production (excluding the local consumption of the Far East). The Italian silk industry has passed through various crises, not on account of excessive production since working on stock is not usual with silk articles and the demand is often greater than the supply but owing to various circumstances, not the least among which are the tariffs raised against Italy as retaliation for the protection of many Italian industries by the tariff of July 1887. The most acute crises of the Italian silk industry were those of 1893 and 1903, which were the cause of numerous financial disasters, and that of 1907-1908, the effects of which are still felt, and which resulted from the great American crisis and is now being aggravated by French and Japanese competition. The quin- quennial average price of raw Italian silk fell gradually from 62-1*. per kilo in 1876-1880 to 38 -Is. in 1901-1905, mainly owing to increase in the world's production (see Table, p. 695). In 1906 and 1907 a rise in price of raw silk occurred ; thus, that of organsine sublime (count i) was 40s. per kilo at the end of 1905, and rose to 49s. Qd. towards the end of 1906 and to 60s. Qd. in August 1907, after which a fall took place owing to the American crisis. Silk-twisting in Italy in 1910 employed 800,000 spindles (four-fifths in Lombardy and the remainder in Piedmont), which produced 4,500,000 kilos of organsine and tram, about one-half from imported raw silk. The silk-waste which was produced in Italy in 1910 (and was exported to the extent of two-fifths while the remainder was worked up in Italy) amounted altogether to 5,300,000 kilos of the value of 500,000. Silk, carded and combed in Italy, amounts to about 1,500,000 kilos and the chappe yarn to almost 900,000 kilos, of which 200,000 kilos are consumed in Italy and the rest exported. Six thousand workpeople are employed in the treatment of waste, the ten estab- lishments in this trade containing about 80,000 spindles in 1912. The exportation of fabrics from Italy was 288,428 kilos in 1892 ; 443,371 in 1895 ; 1,011,567 in 1900 ; 1,254,416 in 1905; and 1,304,750 in 1908. The countries with large outputs of cocoons are not always large consumers of silk wares, while in general large consumers are not producers. Italy has a total internal consumption of 6500 to 7500 quintals of silk articles, and the relation between home consumption and exportation for the principal countries in 1899 was as follows : France Germany . Austria . Italy . . As regards the quantity of raw silk passing through their conditioning establishments, the two principal silk markets in the world are Lyons and Milan, which together receive 1 The protective duty on manufactured wares was 50 per cent, ad valorem in 1883, while it rose to 75 per cent, in 1897, and later to 90 per cent. In 1882 there were only 8000 power-looms (including 2500 for ribbon) and 3100 hand-looms for silk in the United States, while in 1901 the number of power-looms was 52,000 (7000 for ribbon) and that of hand-looms was reduced to 800. In the same period the number of spindles for twisting and spinning increased from 450,000 to 1,900,000. The output of silk gloves was 2000 dozens in 1887 and more than 180,000 dozens (200,000) in 1901. The production of silk articles increased sixtyfold during the latter half of the nineteenth century. The output in America is, however, not equal to tin- consumption, the proportion bet\ per cent, for silk fabrics. 85 per cent, for ribbons, and 5:5 per cent, for velvet. In 1901 the IT silk wares to the value of 5,760,000, now diminished to 3.200,000-43 pel cent, from l-'ra Japan, 17 per cent, from Germany, and 16 percent, from Switzerland. The American (!o\ times, by offering prizes, attempted to initiate the cultivation of mnlberiies and the reaiing < poor success, probably because skilled agricultural labour is lacking and is not easy to turn Home consumption Exports Home consumption Exports Per cent. Per cent. Per cent. Per cent. 61 39 Switzerland 5 95 60 40 United States 95-100 0-5 88-5 12-5 China . . about 50 about 50 20 80 Japan 50 50 cl States import c to 20 Be. ur one of 5 per cent, caustic soda filaments so slender that 225,000 metres do not weigh 1 kilo (Pauly or Fremery and Urban silk). 1 (3) That obtained by decomposing, with ammonium sulphate, cellulose thiocarbonate (sodio-cellulose xanthate), suitably matured (i.e. polymerised) and converted into slender fibres by being forced through a platinum plate furnished with eighteen very small orifices so as to give simultaneously eighteen filaments, 1,000,000 metres of one of these weighing less than 1 kilo (viscose silk of Cross, Bevan, Beadle, Stearn, and Tophar, 1892-1900). 2 (4) The silk prepared from cellulose acetate by Cross and Bevan seems to be free from I he defects mentioned above and to be superior to all other artificial silks in its resistance, which is equal to that of natural silk. The manufacture of this was started a few years ago by Count Donnersmark, using acetic anhydride and chloroform, but it is too costly to compete with other silks, and is dyed only in dilute alcoholic solutions (Ger. Pat. 152,432). Excellent solvents for cellulose acetate have been found in tetrachloroethane and formic acid (Ger. Pat. 237,718 of 1907). (5) Millar and Hummel's Vandura silk, obtained from gelatine solution and now from casein, is not used praetieally. ((>) K. Hofmann (Ger. Pat. 227,198 of 1909) obtains artificial silk and also hair and films, by dissolving cellulose at 220 in a mixture of concentrated phosphoric and acetic acids, and then precipitating with water or salt or alkali solution. The raw material for the de Chardonnet and Fremery silks is cotton waste, which should answer the same requirements as that used in making collodion and guncotton. For viscose the raw material is wood-cellulose, such as is used in paper-making. Although the chemical processes arc different, the final product for all three types of silk is more or less oxidised hydrocellulose ; Chardonnet-Lehner silk consists of hydrated oxy- cellulose. dangerous to the wearer and to warehouses in which it was stored, owing to its inflammability. Attempts to render the silk harmless by the addition of various substances proved futile, and the problem was sqlved subsequently to 1893 by the elimination of the nitro-groups combined with the cellulose and the regeneration of the latter without alteration of its lustre. When treated in this way it burns 'quite like other cotton or silk fibre. But this operation is accompanied by a new disadvantage which has not yet beeii completely removed. On denitration, the artificial silk loses part of its resistance, and when it is wetted the resistance and elasticity diminish further by two-thirds. But in spite of this defect the product is marketable, its production being started before the Paris Exhibition of 1900 and subsequently prosecuted on a large scale. 1 The first patent for this process was that of Despeissis in 1890, but this was not renewed in a year's time. The process was improved and rendered practicable by Pauly, Bronnert, Fremery, and Urban, and the manufac- ture was undertaken by the Vereinigten-Glanzstoff Fabriken of Elberfeld. Well defatted cotton waste is lixiviated with sodium carbonate and hydroxide in an autoclave for 3 to 4 hours, rinsed, bleached with cold hypochlorite solution, well washed and centrifuged. The mass is then treated with concentrated caustic soda to mercerise it and form sodio-cellulose, which is more soluble than cellulose ; or concentrated ammonia is added to a mixture of the centrifuged cotton witli caustic soda and copper sulphate solutions until the whole dissolves (6 to 7 kilos of cotton per 100 litres of solution). When cellulose is to be dissolved in cupro-ammoniacal solution the latter is prepared beforehand in large tanks (in cellars) containing scrap copper and concentrated ammonia solution kept in circulation by a pwnp which also injects air until each litre of solution contains about 15 gnus, of dissolved copper. In this liquid, stirred now and then, cellulose dissolves in 6 to 8 days, the solubility increasing as the amount of copper present increases and as the temperature is lowered (between and 4). As soon as the cellulose lias dissolved and the mass become dense and stringy it must be filtered under pressure, since if this is delayed i no or three days the cellulose begins to undergo depolymerisation (especially in a warm place), and the mass loses its viscosity, with the result that the silk obtained is of poor quality, irregular and weak. Spinning follows closely on filtration. Tlie threads from the capillary glass tubes were at one time coagulated by passing them into sulphuric acid of about 20 Be., but there is then danger of weakening of the fibre owing to excessive hydration, which is facilitated by the rise of temperature caused by the neutralisation of the ammonia. On this account it is now preferred to produce coagulation by means of 5 per cent, caustic soda, this giving a softer and more lustrous silk from which a very weak sulphuric acid bath readily eliminates the traces of copper hydrate precipitated by the soda. According to Ger. Pat. 221,041 (1908) coagulation with alkaline sulphite or bisulphite solution appears advantageous. 1 According to F. Todtenhaupt (1909), to obtain perfect solution and transformation into xanthate, the carbon disulphide is diluted with an indifferent liquid, e.g. benzene, ligroin, CC1, &c., the sodio-cellulose being thus the more completely and easily penetrated. Before spinning, the pulp is matured by heating at 70 to 90 or by leaving for a longer time at 15 to 18. In making viscose silk it was for some years found difficult to iiit on the exact maturation-point (polymerisation of the cellulose) and to avoid excessive adhesive properties. But this difficulty has now been overcome and about 1,500,000 kilos of viscose silk are produced annually. It must, however, be remembered that of the single thread (lava) of the silkworm, 6,000,000 to 7,000,000 metres are required to weigh 1 kilo. 700 ORGANIC CHEMISTRY Of the innumerable now patents for the manufacture of artificial silk, mention may be made of the following : Luiniere Bros, of Lyons dissolve the nitrocellulose in amyl acetate so as to avoid clotting (Ger. Pat. IBS, 173 of 1905). It has been suggested to add resin, oil, oleic acid, &c., to the solvent to economise the latter and to give a more homogeneous solution, hut such additions retard the subsequent denitration. Also the addition of acid leads to the formation of oxycellulose during the denitration, with diminution in the strength of the silk. To obtain Chardonnet silk, collodion-cotton is prepared in the way described in the section on Explosives (pp. 232 et seq.), and after elimination of the acid by thorough washing, the cotton is pressed hydraulically or centrifuged to reduce the moisture-content to 25 to 30 per cent. In this condition it is dissolved in 5 to 10 times its weight of a mixture of 3 parts of ether and 2 of alcohol, with which it is shaken for a couple of hours in revolving iron drums ; de Chardonnet first prepared solutions of collodion with dried nitrocellulose, but Lehner of Frankfort found that moist nitrocellulose also dissolves in alcohol and ether, avoiding the danger of drying and also giving a more homogeneous fibre. If a little mineral acid is added to the collodion solution, the mass becomes much more fluid and requires less pressure for spinning [according to Eng. Pat. 10,932 of 1910, acetylene tetra- chloride (see p. 102) is an excellent solvent for nitrocellulose]. The dense collodion solution is passed under a pressure of 40 atmos. through a cotton-wool filter, then left for a couple of days for the air-bubbles to escape, and finally forced first through cotton-wool and then through capillary glass tubes having a bore of 0-2 to 0-08 mm., under a pressure of 60 to 80 atmos. The slender threads issuing from the capillary tubes under pressure and in a closed-in machine, through which a current of air passes to carry off the alcohol and ether vapour, 1 are united in a number varying from 6 to 20, and under a water-jet are wound on glass spools in a coagulated condition, but still somewhat adhesive owing to the moisture left in the nitrocellulose. After a short time on these spools the fibre solidifies completely and can be manipulated without danger of the filaments adhering. It is then combined, twisted, and reeled in the same way as silk. The artificial silk fibre thus obtained is as lustrous and strong as natural silk, even in the moist state. But it has a rather horny feel, is completely impermeable to water, and hence cannot be dyed in vat, while it exhibits also the serious disadvantage of ready inflammability, which came to be avoided by elimination of the nitro-groups (1893). Ferrous chloride, formaldehyde, thiocarbonate, &c., were tried as denitrating agents, but the best results were obtained with hydrosulphides of ammonium, calcium (0-4 to 0-5 per cent, solution), and magnesium, and with dilute sodium sulphide solution acting for 3 to 4 hours in the cold. The denitration must be carried out with great care since if as little as 0-1 per cent, of nitrogen remains, irregular striations are obtained on dyeing. In practice all but 0-05 per cent, of N can be eliminated, it being impossible to push the denitration as far as the disappearance of the diphenylamine reaction without considerable attack of the fibre. The regularity of the dyeing is also largely influenced by the manner in which the white silk is first dried. Too rapid drying at a temperatiire above 70 and with too dry air, gives rise at many points to oxycellulose which alternates with hydro- cellulose and hence gives rise to non-uniformity of tint. After denitration; the silk contains only minimal traces of nitro-groups, but these are in sufficient amount to allow of the distinction of Chardonnet silk from other artificial and from natural silks, by means of the diphenylamine reaction. 'Artificial silks may also be differentiated from natural silk by microscopic examination (see Figs. 442, 443). Denitrated silk is less resistant and, as with other qualities of artificial silk, the resis- ARTIFICIAL SILK DYEING 701 banco is considerably less in (he moist state ; under such conditions it can still compete with heavily weighted natural silks. 1 In general a libro of artificial silk can be distinguished from one of natural silk owing ( o t he small resistance of the former to tension when in the moist condition. In bleaching artificial silk, the latter is passed first into a weak sulphoricinate bath at 1<> and then into a weak bath of sodium hypochlorite and acetic acid or of calcium hypo- chlorite (0-15 per thousand) or of permanganate (see Cotton). It is then dried, the loss in weight (water and denitration) being nearly 50 per cent. When the artificial silk factories supply a homogeneous product, dyeing is usually accomplished without difficulty on skeins of yarn, just as with cotton and silk. The methods of dyeing arc those used for cotton or, more exactly, for mercerised cotton, which is also cellulose. The dyeing can bo carried out without special mordants if substantive dyesfuH's (diamme, beir/.o, congo, &c.) are used in a bath of sodium sulphate and a little sodium carbonate at the temperature of 50 to 60, various precautions being taken in the manipulation, With basic dyes, a tannin or tartar emetic mordant is used, just as with cotton, the d KJ<;. 442. (Pauly) d, sign of crossed fibres; s7, striation ; b, air- bubbles; q, fine transverse striations; B, sections of fibres 1 According to Hassack the strengths are as follow : Elasticity Xatural silks boiled and lustred . . . .20 ,. ,, red, slightly weighted . . .20 ., ., blue-black, 100 per cent, weighting . 20 black, 140 per cent, weighting . . 20 500 ,, . . 20 Cellulose acetate silk ...... 17 White Chardonnet silk . . . . I o U'hner (Frankfort) silk . . . . . f I'auly (Elbcrfeld) silk 14 Viscose silk ....... 14 Cotton thread 14 FIG. 443. (Chardonnet)a, air- bubbles ; B, sections of fibres Tenacity in kilos per sq. mm. Dry 37-5 20-0 12-1 7-9 2-2 10-2 14-1 17-1 19-1 21-5 11-5 Moist 35 15-6 8-0 6-3 5-8 1-7 4-3 3-2 18-6 Th< ,-/tiisticit./j is the elongation exhibited by 100 cm. of the fibre before breaking. The tenacity or resistance of natural silk is 3 to 13 grms. for the single thread (bava). Echallier (Lyons) has recently increased the resistance of viscose in the moist state by treating it in a bath containing 15 per cent, of formaldehyde, 5 per cent, of alum and 5 per cent, of lactic acid. A further disadvantage of artificial silk is its high specific gravity, the same weight of yarn of the same size giving a larger quantity of fabric in the case of the natural silk than with the artificial. Hut while, with the first artificial silks, the excess of specific gravity was 15 to 20 per cent., the difference is now reduced to 7 to 8 per cent., and further progress in this direction is not improbable. Natural silk has the sp. gr. 1-36 and cellulose acetate silk 1-251, while other artificial silks show values exceeding 1-5. Marked advances have been made also in the count of the thread. Until a few years ago. only yarn of 120 denari (75,000 metres per kilo) could be made, but nowadays counts of 80 denari (112,000 metres per kilo) are regularly spun, and in some cases, with Lehnei silk, 40 denari (225,000 metres per kilo) has been reached. These are still far from the fineness of natural silk (10 to 20 denari) but represent an appreciable step forward. The machinery used in spinning artificial silk has now undergone further improvements which permit of the product ion at once of bundles of threads, these being subjected during their development to rapid rotation so that the completely twisted yarn is obtained in a single operation. There are also machines which give two bundles of threads twisted in opposite directions and at the same time wind the two bundles one on the other so as to produce finished organsine of two threads. ORGANIC CHEMISTRY dyeing being commenced in the cold and terminated at a gentle heat in presence of 2 to 3 per cent, of acetic acid. Certain basic colours dye Chardonnet silk even without mordanting. The new sulphur colours are also used. These different processes give all colours, from the pale and more delicate ones to black, in all shades. One merit of artificial silk is that it cannot be weighted so heavily or so easily as natural silk. Only when black can it be relatively heavily weighted. Cellulose acetate silk is not readily dyed by aqueous solutions of colouring-matters, but as it easily fixes phenols even from dilute solution, a tine pa rani tramline red can be obtained by passing the silk into a hot 0-5 per cent. /3-naphthol bath and then into a 1-5 per cent, p-nitraniline hydrochloride bath containing sodium acetate. The most valuable property of artificial silk is its great lustre, which exceeds that of natural silk and permits of its use for a large number of different articles. Beautiful new effects are obtained by using it as weft in figured textiles with warp of natural silk, a new opening being thus provided for the latter. It is also used with advantage as weft in silk ribbons. For some years it has held almost undisputed sway in the lace industry, and a single Italian manufacturer of this material consumes annually 15,000 to 20,000 kilos of artificial silk. Fringe and cord for ornamenting garments, lace, embroidery, &c., are now largely made from artificial silk. Special articles which cannot be obtained with natural silk are made from the artificial product. There is now a large consumption of artificial hair prepared from artificial silk by fusing together several thin fibres so as to form a single large compact filament which, unlike large fibres obtained directly bv spinning, is flexible and resistant. This artificial white hair, which can be dyed various colours, is in great demand as a substitute for horsehair, which is difficult to bleach and also rather expensive owing to the increased demand for horses for military purposes. This hair is used for various ornaments but mostly for making wigs for ladies and artificial bristles. Another interesting application of artificial silk is in the manufacture of incandescent gas-mantles according to Plaissetty's patent : such mantles are more resistant to shock, even after burning, and can be used in trains. Largely used also is a new product obtained from viscose, namely, a kind of ebonite, which serves well for the manufacture of artistically worked and coloured umbrella handles, knife handles, &c., and resists the actipn of the acids and alkalis with which it is likely to come into contact. Casein products, 1 which have also been suggested for these purposes, cannot compete with viscose ebonite, which exhibits marked advantages over bone and horn in the manu- facture of brushes, as it can be more easily worked and more easily pierced to allow of the fixing of the bristles. As hair-ornaments for ladies, great use is made of artificial silk in thin sheets or ribbons showing brilliant colours and sparkle. Artificial silk is also used in large quantity for making materials for tapestry, upholstery, neckties, hat-linings, &c., with which no re- sistance to the action of water is required. With zinc salts viscose smeared on paper or . fabric shows fine silky effects and fine results are also obtained with bronze powder made into a paste with viscose and spread on different cloths. Important new outlets would offer themselves for artificial silk if the resistance to the action of water could be improved. It seems to be a question of saturating the hydroxyl groups of hydrocellulose so as to render the latter stable towards water, and the most promising attempt yet made is that with cellulose acetate, which gives a silk highly resistant but as yet too expensive, since acetic anhydride is used in its manufacture, while the cellulose acetate must be dissolved in chloroform to be spun. In America this new product is used as an electrical insulator (its dielectric constant is 4 and that of viscose 7, compared with 5-6 for porcelain). The last patents of the Badische Anilin- und Soda-Fabrik in 1904 and the more recent ones of Friedrich Bayer and Co., indicate that a speedy solution of this important problem may be hoped for. 1 According to a Dutch patent of 1911 (No. 431,052), part of the casein suited to the manufacture can be sepa- rated by precipitating the unsuitable casein (which gives brittle products) from skim-milk by means of sodium pyrophosphate solution (3 grms. of the salt per litre of milk). From the decanted liquid, the soluble part of the casein is then precipitated by means of dilute acid. This precipitate is pressed, dissolved in a little dilute ammonia, filtered, reprecipitated with acid, again pressed, rendered plastic with a little ammonia, and spun ; the thread is- rendered insoluble by means of Dilute formaldehyde solution. TESTS FOR FIBRES 70.3 In Kurope there are 30 large artificial silk works, with a capital of more than 2,400,000, among these being the original Chardonnet factory at Besancon, which produces 2000 kilos of silk per day, and the equally important ones at Frankfort and Tubize (Belgium), and that of the Glanzstoff Company at Elberfeld. In France there are 7 factories working profitably ; in Germany, 8 ; Belgium, 3 ; England, 2 ; Spain, 1 ; Austria-Hungary, 4 ; America, 3 ; Russia, 3 ; Japan, 1. The world's production of artificial silk was about 2,500,000 kilos in 1905 and more tlian 6.000.000 kilos in 1011, about 2,500,000 being nitrocellulose silk, an equal amount ammoniacal copper oxide silk, and nearly 1,500,000 kilos viscose silk. The output in France was about 1,700,000 kilos in 1908, the exports being 63,700 kilos in 1908, 78.500 in 1909, and 161,700 in 1910. Italy consumes large quantities of artificial silk. The three large Italian factories (Padua, Pa via, and Turin) arc working under adverse conditions owing to the excessive cost of patents and the keen foreign competition. As is shown by the following figures, the importation of artificial silk into Italy is continuously increasing: 190.> 1906 1907 1908 1909 1910 kilos kilos kilos kilos kilos kilos White . Dyed . 275 2,338 12,900 8,080 25,500 10.920 38,250 2,762 68,822 1,080 210,000 1,560 (126,030) (1,060) The exports wen,- White . Dyed . 572 5,238 18,890 1,187 82,472 5,299 83,942 5,422 (50,365) (3,685) The exportation is mainly from the works at Padua, which is under contract to send its whole output to a German factory. In order to withstand competition better, the artificial silk factory at Pa via began in 1911 to use the ammoniacal copper oxide process instead of that with nitrocellulose which had been previously employed. Artificial silk, which was sold at 28s. to 32,9. per kilo in 1903 and 1904, could be bought at 20s. in 1905, while the price fell to 16s. in 1906, 13s. 6d. in 1908, and 12s. in 1910, the poorer qualities being sold at 6s. to 8-9. per kilo. When artificial silk can be placed on the market at 8s. to 10s. per kilo and this would appear likely at no distant date, owing to the early lapse of the principal patents a new era. of activity for this industry will begin, as it will become possible to displace not only tussah silk but also all the heavily loaded silks of the fabrics commonly used more especially as train silk. The most authoritative information indicates that the cost of manufacture of Char- donnet silk should not now exceed 9s. 6rf. per ki!o, allowing for the recovery of most of the alcohol and ether (this seems to be effected successfully by passing the air containing them through higher alcohols such as butyl or amyl, or by washing the fresh fibre on spools with water ; see also Note, p. 700). while that of the Glanzstoff (Elberfeld) artificial silk should ultimately fall to 6s. to 7s. per kilo, and that of viscose silk to 5s. to 6s. These figures explain the almost fantastic profits realised by the principal factories, which have sold concessions and rapidly redeemed the cost of their plant and are now enabled to pay dividends of 50 per cent., 100 per cent., or even more. CHEMICAL TESTS FOR RECOGNISING DIFFERENT TEXTILE FIBRES The commonest test for distinguishing animal from vegetable fibres consists in burning a thread ; the former burn slowly, giving an odour of burnt nails and forming a round granule of carbon at the point of the thread where combustion ceases, while vegetable lihres burn more rapidly, are converted into ash and give but little smell, which 'recalls that of burnt paper. Other reactions are as follow : Boiling 10 per cent, caustic potash : Hemp, jute, flax, cotton, and artificial silk are insoluble and are not coloured (excepting jute, which becomes yellow) ; wool, silk, and artificial gelatine silk dissolve after a few minutes. Cold cone, sulphuric acid (after 2 hours) : Hemp, flax, jute, cotton, unweighted silk, and artificial silk are soluble or almost so, hemp being coloured brownish yellow, jute 704 ORGANIC CHEMISTRY brownish black, and mercerised cotton yellowish, while the rest remain colourless. Wool and weighted silk do not dissolve. Boiling zinc chloride (60 Be.) : Flax, hemp, jute, and cotton are insoluble, jute alone being coloured a faint brown. Wool, silk, and artificial silk are soluble. Schweitzer's reagent (see p. 503), after 2 hours in the cold, dissolves more or less completely (better if freshly prepared), hemp, flax, jute., cotton, unweighted silk (in less than an hour) and artificial silk. Wool is insoluble. Milton's reagent (solution of mercury in an equal weight of nitric acid of sp. gr. 1-41, first cold, then heated gently, diluted with double the volume of water and decanted after standing) : Cotton, flax, hemp, and Chardonnet-Lehner artificial silk are not coloured ; jute is turned yellow, wool and pure silk violet-red, and weighted silk and tussah silk ochre-red. Cone, aqueous fuchsine (just decolorised with NaOH) : Wool and silk arc coloured red, while cotton and flax remain uncoloured. Silver nitrate solution : Wool is coloured violet to black, while cotton and flax are not coloured. lodo-zinc chloride solution (1 part iodine + SKI + 30 fused ZnCL + 14 water) in the cold : Flax, hemp, cotton, and artificial silk are coloured violet-brown (mercerised cotton almost black) : jute, wool, and tussah silk are turned yellowish and with time become colourless ; true silk is not coloured. Lowe's reagent (shake 10 grms. copper sulphate, 100 c.c. of water, and 5 grins, of pure glycerine and add caustic potash in quantity scarcely sufficient to redissolve the precipitate formed) in the cold dissolves only natural silk and is used for the quantitative separation of natural from artificial silk. Diphenylamine sulphate(l grm. in 100 c.c. cone. H 2 SO 4 ) in the cold : Hemp, flax, jute, and tussah silk are dissolved, giving more or less intense brown colorations (flax dissolves less easily and is less coloured) ; cotton and wool dissolve with yellow coloration ; silk dissolves, giving a colourless or faintly brown solution ; artificial silk assumes an intense, characteristic blue colour. Molisch's reagent (obtained by dissolving 15 grms. of n-naphthol in 100 c.c. alcohol) : the fibre, dyed or otherwise, is first purified by boiling with 2 per cent, sodium carbonate solution and washing thoroughly with water. One centigramme of the fibre is treated with 1 c.c. of water, 2 drops of Molisch's reagent, and 1 c.c. of cone. HaSOj ; all the vege- table fibres, including artificial silk, dissolve with a violet-blue coloration ; wool is insoluble and is coloured reddish ; silk is dissolved, giving a reddish (or, if weighted, an intense red) solution ; tussah silk dissolves, yielding a yellowish solution. Iodine solution (1 grm. KI, 100 c.c. H 2 O, and excess of iodine) : 0-1 grm. of the white fibre, purified as above with sodium carbonate, is treated with a few drops of iodine solution, the excess being removed by means of filter-paper ; hemp, flax, cotton, and artificial silk are coloured blackish brown (flax more intensely than hemp and unmercerised cotton reddish brown) ; wool and silk become orange-yellow and jute reddish yellow. It is often of importance for trade or fiscal purposes to determine quantitatively sub- stances extraneous to textile fibres in order to ascertain their commercial weight. This is determined by means of the so-called conditioning. In conditioning, which is now carried out officially, the moisture is estimated by drying in an oven with automatic regulation, and thus determining very exactly the amount of dry fibre (absolute weight) remaining after silk has been heated at 120 or wool and cotton at 105 to 110. To obtain the commercial weight the absolute weight is increased by the normal moisture which the hygroscopic fibre absorbs from the air, this being fixed at 12 per cent, for flax and hemp, 13-75 per cent, for /e,s8-5 per cent, for cotton, 18-25 per cent, for combed wool, 17 per cent, for spun and carded wool, and 11 per cent, for xilk (120) ; also the amount of dressing in the fibre must be deducted. It is, however, to be noted that usually wool has only 11 per cent., silk 8-5 per cent., and cotton 7-5 per cent, of moisture when in ordinary surroundings. Dressing : 5 grms. of the fabric are well washed with water, wrung out, boiled for DYEING AND PRINTING TESTS 15 minutes in 150 c.c. of 0-1 per cent, sodium carbonate solution, washed in water and rubbed all the fibres being grasped heated to boiling with 150 c.c. of 1 per cent. HC1 and kept on the steam-bath for 15 minutes, again washed and rubbed, boiled for 15 minutes with distilled water, washed with cold water, pressed in a towel, washed two or three times with alcohol and two or three times with ether, dried in the air and then in an oven to constant weight. The loss in weight, after allowing for the moisture (see preceding determination) repre- sents the dressing and colouring-matter ; the latter is almost always a negligible quantity, but in the case of black may be taken at about 0-3 per cent, of the weight of the pure fibre. Mixed Cotton and Wool Fabric. After the moisture and dressing have been determined, the cotton may be estimated and the wool deduced by difference or vice versa. The cotton is determined by boiling 3 grms. of yarn or fabric with 100 c.c. of 10 per cent, caustic potash solution, the wool quickly dissolving ; the residue is well washed with water, boiled for 15 minutes with distilled water, squeezed, washed with alcohol and with ether, and finally heated at 100 to 105 until of constant weight, representing the dry cotton. In reducing this to percentage, account is taken of the moisture and of the dressing. If, however, the wool is to be determined directly and the cotton by difference, 3 grms. of the fabric are boiled for 15 minutes with 0-1 per cent, sodium carbonate solution, rinsed in water, well wrung out in a towel and left for two hours in cold sulphuric acid of 58? Be. ; it is then washed in a large amount of water care being taken that the remaining wool does not become heated boiled for 15 minutes in distilled water, squeezed, washed with alcohol and with ether, and dried at 100 to 105 until of constant weight, which represents the dry wool. Mixed Cotton and Silk Fabric. After the moisture and dressing have been determined (see above), the same piece of dried fabric is immersed for a minute in a boiling solution of zinc chloride (60 Be.) and washed first with water slightly acidified with HNO 3 and then with pure water until the wash water gives no zinc precipitate with ammonium sulphide, the remaining cotton being washed with alcohol and with ether and dried at 100 to 105 until of constant weight ; the silk is calculated by difference. In the case of tussah silk, the action of the zinc chloride is prolonged somewhat. In order that no loss may occur with a heavily weighted silk, the dressing is eliminated by means of sodium carbonate alone, treatment with hydrochloric acid being omitted. Mixed Wool and Silk Fabric. The silk is dissolved in zinc chloride and the residual wool weighed, the silk being determined by difference (see above). Natural and Artificial Silk Fabric. The natural silk is dissolved in Lowe's reagent (see above). Cotton and Linen Fabric. As a rule the different fibres can be separated by hand, but when this is not possible the cotton (after the moisture and dressing have been deter- mined on the same piece of fabric) is dissolved by immersing the tissue for 1 to 2 minutes in concentrated sulphuric acid ; the fibre is washed well with water being rubbed mean- while then with water and ammonia, and again with water, the linen remaining being dried and weighed. The cotton is obtained by difference. Different Artificial Silks. Those from nitrocellulose (de Chardonnet, Lehner, &c.) contain traces of nitro-derivatives and with diphenylamine and sulphuric acid give a blue reaction, which is not shown by other silks. P. Maschner (1910) distinguishes different silks by treatment with concentrated H 2 SO 4 ; that from nitrocellulose colours the liquid a faint yellow only after 40 to 60 minutes ; amrnoniacal copper oxide silk is coloured yellow or brownish yellow immediately, while the liquid becomes brownish yellow after 40 to 60 minutes ; viscose is at once coloured carmine-red, the liquid turning brown after 40 to 60 minutes. The fibres dissolve after about 20 minutes and then carbonise. DYEING AND PRINTING TESTS ON TEXTILE FIBRES Of some importance are the tests which admit of the classification of colouring-matters according to their basic, acid, neutral, or mordant character. To this end, dyeing or printing tests are made on a small scale with wool and cotton (see also p. 671 et seq.). Tests made with colorimeters, which compare the. intensities of coloration of solutions in tubes of equal lengths or vessels of equal thickness, are of*little practical value. Hence to ascertain II 45 Fio. 444. 706 the dyeing power of any commercial product, the latter is compared with a standard colouring-matter by weighing out equal quantities (0-1 to 1 grm. per litre of water) of the two, and dyeing equal weights of wool, cotton, or silk fabric with definite volumes of the more or less diluted solutions. The quantity of dye used is always referred to the weight of the fabric, independently of the dilution of the bath ; this is especially the case with wool (0-1 per cent, of the dye for pale colours and 2 to 4 per cent, for dark colours). The dyeing tests are made on 1 to 2 grms. of wool or cotton yarn or tissue in glass or porcelain beakers of 150 to 250 c.c. capacity, these being heated in a bath of concentrated sodium sulphate solution or of glycerine giving a temperature of 101 to 102 in the dye-bath (see Fig. 444). If the bath retains much colour after the dyeing, a second portion of the textile is dyed without adding fresh dye. If the cotton is raw it must first be boiled for an hour in a 0-5 per cent, caustic soda solution, and then thoroughly rinsed with water. If light colours are used, the cotton is also bleached in calcium hypochlorite solution (less than 1 Be.) at 25 to 35 for an hour, washed Avith water, immersed in a 1 per cent, sodium bisulphite bath (antichlor), and well rinsed in water. Wool, if impure, is heated at 60 for 10 minutes with a solution containing 0-5 per cent, of soap and 0-1 per cent, of sodium carbonate, and then well rinsed with water. Also silk, if not already discharged, is washed with hot soap sohition. The comparative dyeing tests should be made on equal quantities of textile fibre wetted uniformly before introduction into the dyeing bath. Silk is dyed like wool, but the bath is made less acid and the temperature rather lower. W ool is dj^ed in an aqueous bath containing 10 to 15 per cent, of sodium sulphate and 5 per cent, of sulphuric acid (or 6 to 7 per cent, of sodium bisulphate the German Weinsteinpreparat in place of the sulphuric acid) calculated on the weight of fibre ; the bath is stirred continually with a glass rod and heated gently to boiling, being kept slowly boiling for 20 to 30 minutes ; the wool is then rinsed and dried either in the air or in a water-oven. The above procedure is followed more especially for acid dyes ; with basic dyes, one-quarter of the amount of sulphuric acid is sufficient. When wool is dyed with acid dyes, it is not merely necessary to add to the dye-bath the quantity of sulphuric acid required to liberate the acid residue of the dye so that this can be fixed on the wool, but in order that the latter may be dyed intensely and well, 20 to 30 times the theoretical amount of sulphuric acid must be added (E. Knecht, 1888). With mordant dyes, the wool is mordanted with 3 per cent, of potassium dichromate and 2-5 per cent, of cream of tartar (on the weight of wool) and about 100 times the weight of water, heating gradually to boiling and maintaining this for nearly an hour, the water evaporated being gradually replaced ; the wool is then rinsed and dyed in the dye- bath, which contains a little acetic acid(l per cent, on the fibre), and is mixed continuously and brought slowly to the boil, boiling being maintained for about an hour. Knecht and Hibbert (1903-1905) determine the concentration of the colouring-matters in the different solutions by reduction with standard titanium trichloride solution ; crystal violet, for example, fixes 2H, giving the colourless leuco-derivative. Cotton is dyed with substantive dyes in more concentrated baths (50 of water to 1 of cotton) containing 30 to 50 per cent, of sodium chloride or sulphate and 1 to 2 per cent. of sodium carbonate (on the fibre) ; this is heated slowly and kept boiling for 30 to 40 minutes ; in general the bath is not exhausted and can be used for a second portion of cotton. In the case of sulphur colours, 20 to 30 per cent, of sodium sulphide are added to the bath and in some cases 2 to 3 per cent, of glucose, and during the dyeing the cotton is kept immersed and out of contact with the air. Wlfen basic colour ing -matters are used the cotton is previously mordanted with 2 to 4 per cent, of tannin dissolved in water, being left in contact with this solution for 6 to 7 hours (overnight) at 50 to 60 (the tannin is fixed more slowly in the cold) ; the cotton is then wrung, immersed for 10 minutes in a bath containing 2 per cent, of tartar emetic (antimony potassium tartrate) at 40, rinsed with water and dyed in the -tepid (30 to 40) dye-bath for 20 to 30 minutes. FASTNESS TESTS 707 Dyeing on a large scale is carried on under the same conditions, but the calculations are made on a longer time, and great precautions are taken in the moving of the fibre and in raising the temperature, so as to obtain uniformity. For dark colours, the tannin is fixed with ferric nitrate instead of with tartar emetic. Industrial dyeing apparatus is shown more in detail later (p. 717). PRINTING TESTS. The object of printing is to colour the fabric or yarn in a definite pattern or with different colours, part of the fibre being possibly left unaltered. In the first rudimentary printing processes, the fabric was printed with resin or a kind of cement, the uncovered parts being dyed as usual and the preserving substance subsequently removed. It is now usually regarded as preferable to stamp, i.e. to print, on the fabric or yarn the colour mixed with thickening (gum, dextrin, gum tragacanth, &c.) by means of metal rolls on which the desired pattern is engraved. The engraved roll is coated with the pasty colour by rotating against a rubber or cloth roller (furnisher), one-half of which dips in a vessel containing the thickened colour ; a knife (doctor) is arranged so as to scrape the excess of colour from the metal roll, and the yarn or fabric then passes over the latter under pressure. In order to fix the colour and prevent it from spreading, the fibre is subjected for 30 to 60 minutes to the action of steam at about 105 (see p. 731). By this means the colour is fixed without immersing the printed fibre. The latter is subsequently washed with an abundance of cold water (or with tepid soap and water), which removes all excess of colour and thickening agent. In other cases similar effects are obtained by dyeing uniformly in the ordinary way and then printing on the dyed fabric reagents which decolorise (corrode) the dye at the points of contact. Sometimes other colours are introduced with the corroding agent, so that the white parts are dyed a lighter or darker shade or a different colour from the foundation. A kilo of thickened colour for printing wool black the wool having been previously subjected to slight chlorination to make it take up the colouring-matter better (by immersion in a cold calcium hypochlorite bath at 0-5 Be. and then in very dilute HC1, washing, and drying) may be obtained as follows: 750 c.c. of water, 100 grms. of gum, and 100 grms. of British gum (dextrin) are heated in a jacketed vessel by means of indirect steam and kept well mixed, 60 grms. of anthracite black E G and 10 grms. of milling yellow O (and, in some cases, 8 grms. of anthracene acid brown R) being added. When the paste is boiled uniform, it is allowed to cool, and before it is used a solution containing 80 c.c. of water, 120 c.c. of acetic acid (6 Be.), and 40 grms. of sodium chlorate is well mixed in. For printing cotton textiles, colours are used which form insoluble lakes with tannin or metallic oxides ; such are basic and mordant colouring-matters (alizarin, &c.). The former are dissolved in acetic acid and tannin (or a solution of 50 parts of tannin, 50 of water, and 5 of tartaric acid) and the latter (alizarin, &c.) in chromium (or iron, aluminium, &c.) acetate, dextrin, gum, &c., being added in either case. Fabrics treated with tannin, after being steamed at the ordinary pressure and before being washed, are passed into a bath containing 5 to 10 grms. of tartar emetic per litre at 60. FASTNESS TESTS. The fastness of a colour is only relative and must be considered with reference to the purposes for which the dyed fibre is required ; for example, it would be superfluous to require fastness against light in dyed fibres or fabrics to be used for stockings, linings, &c. The dyed specimen is mixed with similar undyed fibre and subjected to the following tests, as required. Mordanted colours answer all these tests fairly well, but in other cases more or less of the colour is given up. Fastness against Water. The sample is immersed in 50 times its weight of cold water for 12 hours or for 1 hour in water at 60 to 70 (and is left to cool in the bath) and is then dried in the oven. Note is taken of the colour assumed by the water and by the white fibre, especially where the latter comes into contract with the dyed fibre. Fastness against Soap, Alkali, and Washing. The skein of white and dyed fibre is immersed in 50 times its weight of an aqueous solution containing 10 grms. of Marseilles soap and 10 grms. of soda per litre. The bath is heated at 60 for 30 minutes and allowed to cool, the skein being then removed, well rinsed, and dried. The changes in colour of the bath and the white and dyed fibres are observed. Fastness against Milling. This test is carried out with a soap and soda solution, of double the above concentration, at 40, the skein being continually rubbed between the hands for 30 minutes, and then well washed and dried in the overy Colours fast to- milling Y08 ORGANIC CHEMISTRY should not soil the white portion of the skein and should give up only a minimal amount of colour to the bath. Fastness against Bleach. If the colour is on wool or silk it is immersed in a 2 per cent, sodium bisulphite bath acidified at the moment of using with a few drops of hydrochloric acid, and, after 30 minutes, washed and dried. When the colour is on cotton, the test is made with a calcium hypochlorite bath at 0-5 Be. for half an hour. Fastness against Scouring. Indigo, Turkey-red, and all basic dyes on cotton mordanted with tannin, even when dry, give up a little colour to a white handkerchief with which they are scoured. Other dyes should not soil the white. Fastness against Acid. The test is carried out for an hour with 1 per cent, sulphuric acid at 60 to 70. Fastness against Perspiration. In some cases this test is made with a 1 per cent, acetic acid solution for 30 minutes at 60, the skein being dried at 60 under slight pressure, without rinsing and after thorough rubbing. In others, an alkaline test is made as in testing fastness against washing but the unrinsed skein is subsequently scrubbed and dried at 60 under slight pressure. Fastness against Ironing. The dyed tissue or yarn is ironed with a very hot iron (1 30 to 140), note being taken whether, after cooling and exposure to the air for 15 minutes, the fabric resumes its original colour. Many colours are changed by ironing hot, but return to their initial state in the cold. Fastness against Steaming. The yarn is placed in a glass tube, through which steam at 110 is passed for two or three minutes. Fastness against Light. One half of a skein of yarn or of a strip of fabric is tightly enclosed between two pieces of card, while the other half is left free ; the whole is then hung in the open air exposed to the sun and weather. For pale colours, an exposure of at least two days, and for dark colours, one of at least four days, is necessary in summer, while in winter or in cloudy or rainy weather (the skein must be sheltered from rain), at least double or even treble these exposures are necessary. The covered and uncovered portions are subsequently compared. Fastness of the Dressing against Rain. A few drops of water are sprinkled on the fabric, especially finer woollen ones, and after exposure to the air it is noted whether the drops have left faint spots; In some cases the fabric is scratched with the thumb-nail ; a paler streak should not result. This test is not applied to cotton fabrics strongly dressed, since the nail will sometimes detach the dressing itself. THEORY OF DYEING. The phenomenon of dyeing was at one time thought to be due to the porosity and capillarity of fibres which were thus enabled to absorb, and become impregnated with, dyes. The possibility of chemical combination between the dye and the fibre was regarded as excluded, it being asserted that in such case the fibre would undergo marked change. The different colouring powers of substances were explained as due to different molecular magnitudes. Even at the beginning of last century, in Chreveul's time, these ideas prevailed, and only in the case of mordant dyeing was any chemical fixation of the dyestuff assumed. Later on, Bergman, J. Persoz, &c., arrived at a purely chemical conception of the phenomenon of dyeing. But when in 1885 substantive cotton dyestuffs of almost neutral character made their appearance, the chemical theory, which was based mainly on the basic or acidic nature of the dyestuffs, was in some degree shaken. Many then accepted a new theory in harmony with the osmotic phenomena of solutions, the more readily because no definite and constant relation between the amount of fibre and that of dyestuff combined had been established. The chemical theory was, and is still, however, upheld by many authorities on the subject, more particularly by Noelting, by Knecht, and by Vignon, who have pointed out that alloys form well charac- terised compounds which exhibit no definite chemical relations between the components and may be regarded as true-solid solutions of one substance in excess of the other. Further, they were able to show that silk and wool, in combining with colouring-matters, set free the acid united with the base of the dyestuff, this acid being found in the dye-bath. Also, with certain acid dyestuffs (e.g. naphthol yellow), Knecht and Appleyard found a constant relation between fibre and dyestuff. Jacquemin asserts that if there were no question of chemical combination, the dry dyed tissue should have the colour of the dry colouring-matter, whereas it has the same THEORYOFDYEING 709 colour as the dissolved colouring-matter. Nietzki finds that with certain highly basic colours (e.g. methyl green), wool cannot of itself displace the mineral acid of the colouring base, the addition of ammonia being necessary ; while, with the same colouring-matters the more markedly acidic silk is dyed without any addition. An interesting fact, which supports the chemical theory, is that the base of rosaniline is colourless and becomes red (fuchsine) only when converted into a salt with HC1 ; a similar change is produced if wool is immersed in a colourless rosaniline (base) bath, the wool being dyed red owing to the formation of a salt. If the dyeing is effected directly by rosaniline hydrochloride, the bath ultimately contains the hydrochloric acid which is displaced by the acid of the wool fibre (Jacquemin and Knecht, 1888). Moreover Richard (1888), Vignon (1890), and Nietzki (1890) showed that silk and also wool are active both towards acids and towards bases, so that in chemical characters they are comparable with the amino-acids. The fibre may even be replaced by albumin, which is dyed by the same dyestuffs as wool, &c. According to W. Suida (1907) the dyeing of wool is accompanied by liberation of the base of the dyestutt' which combines (or forms salts) with the textile fibre, the latter function- ing as a polybasic acid in virtue of its guanidyl and irnidazole groups. Also Vignon showed that when wool and silk are dyed with basic or acid colouring-matters heat is developed, so that the dyeing may be regarded as a true, exothermic chemical reaction. According to Vignon cotton is not dyed directly by basic or acid dyestuffs (which are usually salts) since it has not the reactive force to decompose them ; but if it is previously oxidised or animated, it fixes these dyestuffs partially with development of heat. Further, the difference in fastness against light of the same colouring-matter (e.g. methylene blue) fixed on cotton (with tannin) and on wool or silk would appear to favour the chemical hypothesis of the phenomenon of dyeing. In 1889 O. N". Witt advanced a new theory, which explains also the dyeing of cotton with substantive and mordant dyes. According to Witt, dyeing consists merely of a solution of the colouring-matter in the fibre, analogous to that of solution of coloured metallic oxides in glass. So that the colouring-matter passes from a liquid solvent (dye- bath) to a solid one the fibre itself just as occurs with alloys or in the extraction with ether of a substance dissolved in another solvent in which it is less soluble than in ether assuming that the two solvents are mutually insoluble. Dyeing on mordants is similarly explained as due to the solvent properties of the fibres for the metallic salts, these then fixing the colouring-: Batter from the dye-bath. The dyeing of cotton with substantive dyestuffs is regarded as the result of the marked solvent power of cotton (cellulose) for these dyes. In support of his theory, Witt cites the fact that silk dyed with fuchsine gives up its colour to alcohol, which is a better solvent for fuchsine than is silk, while if the alcohol is then diluted with water, the colour is again fixed by the silk. To this observation Knecht (1902) made the reply that, with substantive colouring- matters, lanuginic and sericinic acids form insoluble lakes, i.e. true compounds, while with fuchsine they form lakes soluble in alcohol ; it is therefore to be supposed that the fuchsine extracted by Witt with alcohol is in reality the soluble lake formed by the fuchsine with the components of the fibre. Rosenstiehl (1894), Reisse (1896), and Gillet (1898), after various quantitative dyeing tests, decided in favour of the chemical hypothesis. In 1894-1895 Georgievics advanced a number of arguments in favour of a purely mechanical theory of dyeing (his predecessors of a century earlier being Hellot and Le Pileur d'Apligny, and those of more recent times Walter Crum, Spohn, and Hwass). Com- paring the latter with occlusion of gases by solids or with the mechanical fixation of dyes on sand or on powdered charcoal, &c., he maintained that colouring-matters fixed on fibres have the same properties as those not so fixed, and that there can hence be no question of a chemical reaction (but see above, Knecht's experiment), since some dyestuffs fixed on fibres can be separated by mere sublimation, while in other cases (with methylene blue and indigo carmine) the coefficient of distribution of the colouring- matter in the fibre and in the solution is constant. According to Krafft (1899), dyeing generally consists in a deposition, on or in the fibre, of adhesive and resistant colouring salts in the colloidal state. Biltz (1905) has succeeded in producing true dyeing phenomena by replacing the textile fibre (cotton) by aluminium hydroxide or other hydroxides which behave as hydrogels 710 ORGANIC CHEMISTRY (see vol. i, p. 102) towards the colouring-matter, which is regarded as a colloid (ben/o- purpurin and sulphur dyes). Freundlich and Losev (1907) have shown that carbon not only fixes colouring -matters but decomposes basic colouring-matters, fixing the coloured base in the colloidal state and leaving the acid in solution, in the same way as happens with wool or silk. Knecht has recently (1909) found that the amount of colouring-matter fixed by charcoal is related to the quantity of nitrogenous matter remaining in the charcoal even after ignition, so that here a true chemical reaction occurs ; this investigator has ako shown that colouring-matters cannot be regarded as colloids, since 'they are electrolytes and diffuse through membranes. In 1909 Dreaper and Davis demonstrated that basic colouring-matters are fixed in constant quantity on calcined sand, and in increased quantity if the dye solution contains sodium chloride. Rosenstiehl assumes that the phenomenon of dyeing is explainable by the cohesive force between the colouring-matter and the textile fibre, this force varying with the liquid or gaseous medium in which the dyeing takes place and depending on or being produced by the osmotic pressure of this medium. According to Miiller and Slassarski (1909) dyeing may be regarded as a phenomenon of adsorption of the colouring- matter by the colloid, i.e. the textile fibre. There is hence not chemical combination, but fixation, under definite conditions (of moisture and tem- perature). Mercerised cotton fixes colouring-matter on account of its more marked colloidal character. The process of fixation or adsorption may also be reversible and all the pheno- mena of direct dyeing depend on the relative coefficient of adsorption of the colloid (fibre) for the colouring-matter. Freundlich and Losev and Pelet-Jolivet attribute dyeing to adsorption because the fixation of the colouring-matter from solution by any textile fibre or x obeys the formula, = K-C (where denotes the ratio between the quantity of colour m n v m absorbed and the weight of the textile fibre, K and are constants, and C indicates the n final concentration of the colouring-matter), which also regulates the adsorption of gases by solid substances and that of various dissolved substances by animal charcoal. It cannot, however, be denied that certain limited chemical processes also correspond with this formula, and that many phenomena accompanying dyeing are most simply explained chemically. Indeed, W. .1. Miiller and Slassarski (1910), by means of experiments on the dyeirg of "artificial silk, show that the absorbed colour varies in quantity with the chemical pro- perties of the cellulose (raw, oxycellulose, hydrocellulose). Every hypothesis is supported by some experimental fact, and it would seem that, according to the nature of the fibre, of the colouring-matter, and of the dyeing process, the phenomenon is explainable either on purely physical or on purely chemical grounds, but more generally on both. O. Weber (1891, 1899) and Gnehm (1898) explain the various phenomena of dyeing in the following way : (1) Dyeing on mordanted cotton is due to the formation of lakes between the colouring-matter and the mordant precipitated mechanically on the cotton. (2) Azo- colouring-matters formed directly on the fibre (see p. 658) or pigments held by it, ultramarine, cinnabar, ochre, Guinea green, &c.) are merely precipitates deposited mechani- cally in the pores of the fibre. (3) The direct dyeing of cotton with substantive dyes consists in solution of the colouring salt in the cell juice, and the marked fastness against washing of these colours on cotton is due to their slow diffusion with the juice (Miiller- Jacobs and Weber). (4) Dyeing of tannin-mordanted cotton with basic or indigo colours is a true mechanical occlusion. (5) Direct dyeing of wool and silk and other animal fibres with basic or acid colouring-matters is due partly to mechanical absorption, and partly to chemical combination, of the colouring-matter by the fibre. (6) The dyeing of mordanted animal fibres is explained by the formation of insoluble lakes, partly by the mordant fixed chemically by the fibre, and partly by that fixed mechanically within the fibre, but is never caused by combination of the unchanged fibre with the colouring-matter. As regards the mordanting of wool, it has been shown that when this is boiled with metallic salts, it fixes not only the basic part but also, the acid part of the salt (only of unstable salts, e.g. sulphate of Al, Cr, Cu, or Fe, and not sodium sulphate or chloride); the latter part is eliminated to some extent by water, but the basic part is fixed more stably. TEXTILE MACHINERY FIG. 445. MACHINERY USED IN DYEING AND FINISHING TEXTILE FIBRES The limits of this treatise do not allow of the inclusion of a complete description of all the machinery used in works where textile fibres are dyed and finished. We shall hence confine ourselves to illustrating some of the principal washing, dyeing, and dressing machines. WASHING AND PREPARATION. At the dye-house, textile fibres arrive either raw (cotton and wool in flock) or combed (wool in skeins or tops) or spun in skeins or on bobbins (wool, cotton, silk), or more commonly woven in "pieces 30 to 100 metres long and 60 to 140 cm. wide (woollen, cotton, silk, or mixed fabrics). Wool is sometimes supplied free from its natural fat (see p. 681) but, whether as fabric or as yarn, contains the fat or dressing used in weaving or spinning. Cotton is still in the raw state, and, in order that the colouring-matter may be fixed well, it is subjected to energetic boiling under slight pressure with water and with soda. With either flock or skein cotton, this treat- ment is carried out in large, closed, iron or copper boilers (Fig. 445), provided with pumps or steam-injectors for circulating the liquid, the textile material not being moved as it might be damaged. As a rule the boiler is either evacuated or freed from air by a current of steam, since air damages the fibre owing to formation of oxy- cellulose, and also gives dark lye ; along with the caustic soda, vigorously frothing soap (from castor oil, for example) is introduced. The washing of cotton goods to rid them of the starch with which the weft was charged for weaving purposes was at one time carried out by heating them with milk of lime, but better results are obtained by heating with dilute caustic soda solution in an autoclave under steam-pressure. Nowadays the goods are often passed through a lukewarm bath of diamalt or diastofor (malt extracts rich in diastase) and left in heaps overnight, the- starch being thus transformed into soluble dextrin and maltose. The latter products are removed by thorough rinsing in water : the material passes between the two rollers A and B (Fig. 446) into the water, round the roller C, up between A and B, down again and so on until it reaches the middle, where it is removed, together with a similar piece intro- duced at the other end of the machine ; the pieces of material are tied end to end and passed through this washer in a continuous length ; an abundant supply of water enters the vessel at D and is drawn off through another pipe. When washed the goods are soured with a solution of sulphuric acid (0-5 Be\), either cold or tepid (with the latter the action is very rapid, even with more dilute acid) ; the pieces may be tied together in cords and passed through this solution (*ee Fig. 446). Bleaching is then effected in a clear chloride of lime bath (0-5 to 0-75 Be) : this occupies some hours in the cold, or, if the liquid is lukewarm, the material may be passed continuously through it as before. Then follows rinsing and treatment with antichlor (sodium bisulphite). Skeins of cotton yarn may also be bleached with chloride of lime in an apparatus jvith Fia.' r 446. 712 ORGANIC CHEMISTRY automatic circulation of the liquid, as is shown in Fig. 445, while the rinsing may be effected in rotating machines (Fig. 447), where each skein rotates on a reel and all the reels rotate horizontally in a circulation vessel, a water-spray being used meanwhile. According to Pick and Erban cotton may be bleached in the cold, without preliminary boiling with alkali, by means of sodium hypochlorite solution mixed with sulphoricinate ; in this way, the resistance of the fibre is retained better, while time is saved (Ger. Pat. 176,609 of 1 906). Cotton or cotton and wool FIG. 447. FIG. 448. fabrics may be bleached by passing them repeatedly into a sodium permanganate bath (0-6 to 0-7 per cent, of the permanganate on the weight of ftbre) until the bath is almost decolorised and the fibre turned brown, then into a sodium sulphite or sodium nitrite bath (0-6 to 0-7 per cent, on the fibre) and finally into sulphuric acid (4 per cent, on the fibre). The Washing of skeins of wool yarn in a tepid bath (50 to 60) is carried out by passing the skein for a minute between two rolls (Fig. 448), then twisting the skein and again FIG. 449. FIG. 450. squeezing it. Subsequent thorough washing with water in the vessel shown in Fig. 447, for example renders the skein of wool ready for dyeing. In all these operations and in those which follow, woollen yarns are treated with greater care than cotton ones, it being necessary to manipulate, press, and rub them as little as. possible and only very slowly in order to avoid felting. Bleaching of washed woollen yarns or fabrics (wrung out uniformly by means of centri- fuges : see p. 468) by sulphuring is effected by stretching them out on rods in tightly closed chambers in which sulphur has been previously burnt in a cup situate in an angle heated by a furnace outside. Here the wool is left overnight, and in the morning the windows BLEACHING, WASHING 713 are opened and the wool dried and deodorised in the air. The amount of sulphur burnt is 2 to 3 per cent, on the weight of the wool, or less if the chamber is a small one and deficiency of air is maintained in order to avoid sub- limation of the sulphur and its deposition as a yellow powder in the wool. Bleaching with Hydrogen Peroxide is carried out in the cold or at a gentle heat, and for woollen yarn, paraffined wooden vessels, or, better, cement troughs, are used. Woollen or silk fabrics are wound into a vessel similar to that used for dyeing (see later), or, better, on a jigger (see later). The bath is prepared by diluting commercial 10 to 12 vol. H 2 O 2 with 8 to 10 times its volume of water, and rendering it slightly alkaline with ammonia (see vol. i, p. 235). After use the bath is preserved by acidification with sulphuric acid. More economical bleaching is obtained with sodium peroxide, which, however, must be used with great caution (see vol. i, p. 440) ; FIG. 451. better results are obtained with sodium perborate (see vol. i, p. 480) in a bath containing, say, 200 litres of water, 600 grms. of sulphuric acid of 66 Be., and 1-8 kilo of sodium silicate at 40 Be. FIG. 452. Washing of Woollen Fabrics is carried out in various ways. A number of the pieces, the two ends of each being tied together, are wound round in a trough fitted with a pair of pressure rollers, A and B (Figs. 449, 450), and containing hot soap and soda solution. Beneath the rolls is a wooden channel, G, to collect the expressed liquid, which for some 4T(D A \ P & q a o 6 : ? < r * a a Q D Fio.^453. time is allowed to run back through r, but when dirty is run off outside. Thorough rinsing with water is carried out in the same vessel. It must be noted that almost all washing and dyeing machinery is fitted with arrangements for obtaining different velocities of the moving parts, with pipes for water and steam, &c. Very heavy woollen fabrics are more easily washed at their full width in vessels (Fig. 451 ) similar to the preceding. But the lighter ones are most conveniently dealt with by joining 714 ORGANIC CHEMISTRY the pieces end to end so as to form a single piece, which is treated in the machine shown in Fig. 452, and, in diagrammatic section, in Fig. 453. This is furnished with three pairs of rolls, A, B, and C, which press the pieces in their passage from one vessel to the next, while a slow current of water enters at R and takes a zigzag course through the succeeding vessels ; a little soap and soda solution is gradually added in vessels 1, 2, and 3, which are heated by steam-pipes, while the dirty water is discharged continuously from S. FIG. 454. FIG. 455. For making certain articles, woollens must be subjected to Milling, which transforms them into more or less close cloth. When the pieces are rolled up, moistened with soap solution, and then continually compressed and rubbed together, the wool is felted and cloth formed in the course of a few hours. The milling machine in which this is effected is shown in Figs. 454 and 455. The material is caught between the three wooden rollers A, B, and C, which compress them and force them into the wooden channel R 8, where the pressure of the plate R may be increased by the spring A ; the expressed liquid collects in the channel E and is at first returned but later discharged. If any knots were formed they would stick at P and raise a spring, T, thus stopping the driving- belt. With certain heavy fabrics already soaked with oleine, milling is carried out with addition of a little soda solution, which saponifies the oleic acid. In some cases dilute sulphuric acid is used, but 1 idler results are apparently obtained with 1 per cent, lactic acid solution, the wool then retaining greater lustre and elas- ticity (G. Ita, Ger. Pat. 236,153 of 1910). Some fabrics which are required to take bright designs and a very smooth and shiny surface (satin, &c.) are freed from the down always accompanying textile fibres especially after washing, &c. by passing them, quite taut, quickly over a row of gas-jets (or over a sheet of heated copper or a strip of metal heated electrically), which burn the hair on the face and sometimes on the reverse of the fabric too (see Fig. 456, where the gas-jets run horizontally from A and B). The removal of cotton fibres or bits of vegetable matter (which would become more noticeable after dyeing) from woollens may be effected by hand, but is more commonly attained by Carbonisation. In this the fabric is impregnated uniformly with sulphuric Fit;. 456. 715 acid of about 4 Be. (or aluminium chloride solution), centrifuged and heated at 125 to 135 being passed at width either over a series of tinned sheet-iron or copper rollers (similar FIG. 457. to those used for drying woven goods after dyeing) through which steam at 2 to 3 atmos. is passed (see Fig. 481, p. 724) or else slowly through a large oven heated with hot air or with branched pipes fed with steam under pressure (see Fig. 457). In this way all the FIG. 458. vegetable fibres are incinerated or carbonised and are eliminated in the subsequent souring, which occupies an hour and is effected by means of a large quantity of water in the washing vessels already described (Figs. 449, 450). I As has been mentioned, woollen fabrics exhibit a tendency to felt and shrink, and these actions may become very pronounced during dyeing, when the material is kept moving 716 ORGANIC CHEMISTRY in boiling baths for two or three hours. In order to avoid these changes, which likewise often spoil the design, the fabric is subjected to Fixing, which consists in heating it in a stretched condition in vigorously boiling water, i.e. at a temperature rather higher than any it will experience in subsequent operations ; scalding of the fibres in this way causes FIG. 459. partial loss of their elasticity and power of contraction, and the fabric shrinks less during dyeing. Light fabrics are fixed in the so-called revolver machine (Fig. 458), in which the material is wound in compact rolls on reels dipping into a vessel of water kept briskly boiling ; each reel may have six rolls and one reel is arranged in each of two adjacent vessels. The axis of each reel revolves during the winding, and when the first reel has FIG. 460. received the first six rolls, the first roll begins to unwind to form another on the second reel, so that the part of the fabric which was peripheral on the first roll becomes central in the roll of the second reel. This procedure prevents any subsequent irregularity of colouring owing to the more ready and more intense fixation of the dye on the parts subjected to the most prolonged action of the boiling water. Each roll may contain from 100 to 300 metres of fabric, which is fixed in about an hour. Certain heavy woollens with a satin surface (and also mixed wool and cotton goods unions or cotton goods with a satin foundation) are fixed, and at the same time furnished DYEING 717 FIG. 401. with a lustre which persists even after dyeing, by so-called crabbing. The machine in which this is carried out consists essentially of two or three pairs of superposed heavy rolls of solid iron (Figs. 459, 460). One-half of the lower roll of each pair dips into a long narrow vessel of water kept boiling by direct steam. The stretched, smooth cloth is wound in compact rolls on the lower roll, and is then allowed to revolve for 30 to 40 minutes in the boiling water, being pressed by the upper roll, which revolves freely and can be weighted by means of levers. The fabric then passes to the lower roller of the adjacent vessel and so on. DYEING. Cotton and wool in flock were at one time dyed by immersing them in cloth or net in open wooden vessels containing the hot dye-bath. Use was afterwards made of mechanical apparatus similar to that shown in Fig. 445, where the material remains stationary on a false bottom, below which the liquid is drawn off and pumped to the top. It was, however, often found that the liquid did not pass uniformly through the whole of the fibre but went more easily through that which was least compressed and which contained channels, thus producing irregular dyeing. Almost universal use is now made of mechanical apparatus similar to the above, but with the fibre highly compressed (-see Fig. 461). In this case the pump, which must be more powerful, causes complete penetration of the liquid, and much better results are obtained. Skeins of yarn can also be dyed in this apparatus when they are well com- pressed. After the discharge of the dye-bath (kept, if required, for a subsequent operation), the dyed fibre may be washed in the same vessel. To dye combed wool (tops) wound on to bobbins by suitable machines (Fig. 462), very general use is made of Obermaier mechanical apparatus of the revolver type, in which the bobbins are arranged in as many horizontal, cylindrical cases fitting into a vertical cylinder closed at the top and communicating below with the pipe of a pump, which it fits exactly (Fig. 463) ; the mode of action is shown clearly by the figure. A more simple appa- ratus which carries larger charges and is largely used also for yarn on bobbins with crossed thread, is that of Halle shown in Fig. 464, where may be seen the false bottom supporting the bobbins, the pump for circu- lating the dye solution and the perforated cover pressed down by vertical screws. In these mechanical apparatus it , is always possible to reverse the sense in which the liquid circulates, homogeneous dyeing being thus more easily obtained. With skeins of spun fibre, various methods of dyeing are in use : in the old method, still largely used, the skeins are threaded on smaoth round sticks so that one-half of the skein is immersed in the dye-bath, the skeins being turned or inverted on the stick from time to time by hand (see Fig. 465). The form of the wooden vessel is now simpler, as is seen from Figs. 466 and 467, which show the perforated false bottom below which are the direct or indirect steam-pipes for heating the bath, and the perforated wall, P, outside of which the colour is gradually added so that it may not come into immediate contact with the neighbouring skeins. FIG. 462. 718 ORGANIC CHEMISTRY A mechanical apparatus for dyeing skeins is shown in Fig. 468. The skeins are threaded on^rods which are rotated by toothed wheels, while the whole frame can be raised from or lowered into the bath by a toothed rack. Still better is the Klauder-Weldon revolving apparatus shown in Figs. 469 and 470 : on a large bronze wheel, one-half of which dips FIG. 463. into a trough while the other half is covered, are fixed axial and peripheral rods, which keep the skeins taut. The wheel revolves slowly in the dye-bath, and the pegs, b, at the ends of the peripheral rods knock against an iron striker inside the trough, so that the rods revolve slightly each time ; hence the skeins threaded on them are moved a few centi- metres. Two workmen suffice for the charging and discharging of 100 to 200 kilos of wool '. Fia. 464. or cotton, while during the dyeing one man can look after three or four of these machines, adding the necessary colour now and then by means of tjie copper funnel A. The steam for heating the bath reaches the bottom of the trough by the tube d. At e is an automatic indicator which shows when any particular peripheral rod does not turn owing to the skein being caught. The rapidity of revolution may be altered, but, as a rule, the movement is slow in order that the wool may not be felted. In recent years a happy solution has been found to the problem of dyeing cotton or woollen yarn while still wound on the tubes of the spinning machine as spools or cops, DYEING OF SPOOLS AND SKEINS 719 thus avoiding the winding into skeins and preserving the fibre better. At first the per- forated tubes of the bobbins were inserted in drums which rotated in the bath and from the interior of which the air or liquid was pumped, the bath being hence circulated from the interior to the exterior of every bobbin and vice versa (Figs. 471, 472). There are various other arrange- ments, but recently a good reception has been every- where accorded to an apparatus devised by De Keuke- laeres of Brussels. This compresses the skeins or bobbins in a square iron or copper case on to a perforated false bottom, while, before the case is covered with a per- forated metal plate, the yarn is covered with sea-sand, which is forced into all the pores of the mass not occupied by fibre by means of a water-jet. The cover is then fitted and screwed tight, and the bath circulated through the mass of yarn by means of a pump capable of develop- ing considerable pressure ; the liquid may circulate from bottom to top and vice versa and, finding no channels open, is obliged to traverse the fibre uniformly. When the dyeing is finished, it suffices to place the bobbins in a perforated basket and to shake this in a vessel of water to separate the whole, of the sand, which collects at the bottom of the vessel and can be used again. FIG. 466. For dyeing skeins of cotton with Turkey red, which is the fastest red for cotton, the latter must be prepared and mordanted. It is not bleached with chlorine but is boiled for a long time with a caustic soda solution (0-75 Be.) under pressure (2 atmos.) for 4 to 5 hours. When washed, the skeins of cotton are passed repeatedly into a bath of neutralised ammonium sulphoricinate (20 kilos of 50 per cent, strength per 100 litres of water ; see p. 327) ; this operation is readily done with a suitable machine (Fig. 473), which is fitted with ingenious contrivances for pressing, wringing, un- twisting, and immersing the skeins in the sulphoricinate bath repeatedly and auto- matically. When thoroughly soaked, the p IG ^QJ skeins are dried at 50 to 60, then steamed under an excess pressure of 0-5 atmos. in an autoclave for an hour, and afterwards passed into the mordanting bath, consisting of a basic aluminium sulphate solution (7 Be.) at 45 (with an iron mordant, a violet colour is obtained instead of red ; with one of tin an orange colour, and with one of chromium a reddish brown colour ; but these mordants are rarely used in practice) ; they are subsequently dried at 45. 720 ORGANIC CHEMISTRY Use is often made next of a tepid bath consisting either of a little chalk suspended in water or of sodium arsenate, to remove any sulphoricinate not stably fixed, and hence to give subsequently a brighter colour. After this preparation, the skeins are passed into the dye-bath (10 to 15 per cent, of alizarin paste, calculated on the weight of cotton) contained in wooden vats and heated by tinned copper steam-coils ; the temperature is first kept at FIG. 468. 25 for an hour and is then raised in 30 minutes to 65 to 70, the goods being manipulated for an hour. The dyed skeins are dried and are often introduced, without washing, into a second sulphoricinate bath, being then steamed for an hour in an autoclave at 1 atmos. ; the colour is not very bright but is made so by immersing the material for half an hour in a 0-5 per cent, soap solution heated under slight pressure (0-5 to 0-25 atmos.). Thorough FIG. 469. washing with water is followed by drying at a gentle heat. Although Turkey red is removed to a small extent if the material is scoured with a white fabric, yet it is the fastest red against washing and light now prepared on cotton. Kornfeld (1910) regards the fastness of Turkey red as due, not to the constitution of alizarin, but rather to the formation of a highly resistant double salt of aluminium oleate and the calcium salt of alizarin, and still more to the polymerisation of the fatty acid molecules under the action of steam. According to a patent by Kornfeld, Turkey red dyeing may be carried out in the usual mechanical apparatus with circulation of the bath, the alizarin being rendered soluble by means of sucrate of lime. DYEING OF COTTONS AND WOOLLENS 721 Cotton Fabrics are sometimes dyed in ropes with vessels similar to those used for wool (see later), but more usually in the so-called jigger (Fig. 474), which is a rather shallow wooden trough provided with two outside rollers worked alternately by gearing so as to wind or unwind the pieces (3-4) ; the latter are sewn end to end and are kept quite taut, and pass below two small rollers close to the bottom of the trough. The dye solution in the bath may be heated at will by direct or indirect steam.' FIG. 470. The jigger is often used also for dyeing unions, i.e. fabrics composed of cotton warp and wool weft, since these do not cockle or wrinkle, as all-wool goods would do, when passed under tension from one roll to another. Woollens are usually dyed in wooden vessels provided with one or two reels which raise the goods in ropes from the front part of the vessel and drop them into the bath, the inclined wall at the back forcing them in folds on to the bottom of the vessel itself (Figs. 475, 476). FIG. 471. FIG. 472. In some cases the velocity of rotation of the reels can be varied at will, being accelerated at the moment when the colour is introduced into the perforated compartment which admits of its gradual passage into the whole of the bath. The perforated steam-pipe also passes into the bottom of this compartment and is so arranged that the steam does not strike against the pieces, as this would result in irregular dyeing. The velocity of the reel must not be too high (20 to 50 cm. per second), as otherwise the wool would felt and the bath cool too rapidly. When the pieces are introduced into the vessel, one end is thrown over the reel and then stitched with twine to the other end (see Fig. 476). In some cases the materials (e.g. cashmeres) are twisted, by the movement in the trough, into very thin cords, into which penetration of the colouring-matter is difficult and irregular ; in order ii 46 722 ORGANIC CHEMISTRY to avoid these disadvantages, such fabrics arc first folded in two lengthwise and the selvedges then stitched together. During the dyeing operation, the dyer cuts off small samples of the fabric from time to time, washes them, dries them in a warm towel and compares them with a specimen, so that fresh addition of colour may be made where necessary. Such fresh colour is dissolved apart in a wooden bucket in a few litres of the hot dye-bath, the solution being always FIG. 473. passed through a very fine hair-sieve to remove granules of undissolved dye, which would spot the material ; the steam-cock is closed while the new dye is being gradually added. The dyeing of woollen fabrics is commenced with a bath of tepid water (40 to 50) with the addition of 10 to 15 per cent, of crystallised sodium sulphate and 2-3 percent, of concentrated H. 2 SO 4 (or 5 to 6 per cent, of sodium bisulphate) (these proportions referring to the weight of the fibre). The colouring-matter (a few grammes for pale colours and as FIG. 474. much as 5 kilos of black per 100 kilos of material) is added in several portions at the begin- ning of the operation, the goods being slowly moved meanwhile. In the course of an hour the bath is brought to boiling and this may last one or two hours before the dyeing is complete. Finally the steam-tap is shut and the goods discharged into a vessel of cold water. After being rinsed and folded roughly by hand they are left to drip on beams for some time, a further part of their water being removed by two or three minutes' centrifuging (see p. 468). The goods are then ready to be dried in the apparatus described later. When very delicate wool or wool and silk fabrics (with gathers and embroidery) are t-o be dyed, they are sometimes wound concentrically on hooks fitted to a frame such as that shown in Fig. 477. In this case the frame is only moved now and then, so that the fabric may not be injured. DRYING MACHINES Textile Fibres in Flock are dried in a series of superposed chambers with perforated bases on which the moist, centrifuged fibre is spread (Fig. 478, 1). At II is seen a counter- poised elevator on which is placed the charged chamber ready to be introduced into its position in the series in place of one containing fibre already dried. The air used for the drying is forced in by the fan A, and is heated in the tubular steam heater B. The lower chambers are dried first, and when these are discharged, the remaining ones are lowered FIG. 475. FIG. 476. automatically and fresh ones introduced at the top. Yarn on bobbins or spindles can also be dried in these chambers. Skeins of yarn may be dried by threading them, after centrifuging, on rods and fixing these horizontally in frames in a chamber heated by branched steam-pipes on its base ; the moist air issues from vent-holes fitted to the ceiling. In some cases the yarn is dried in hot chambers, the skeins being stretched over revolving reels furnished with central steam-pipes, as is shown in Fig. 479. Good results are also obtained with the continuous drying machine, in which the skeins are placed on rods, &c., carried by chains moving in a drying chamber (Fig. 480) supplied FIG. 477. FIG. 478. at A with hot, dry air. The dry yarn issues continuously at Z, while the moist air finds an outlet at B. Fabrics as they come from the centrifuge are usually dried by passing them, well stretched, over a battery of seven or nine copper drums F (Fig. 481 ). These are all moved regularly by gearing, the rate being regulated by means of the large disc B, which is actuated at a point more or less distant from its centre by the friction roller C ; the latter is turned by the pulley A, .joined by belting to the general system of power transmission. The dried fabrics are then examined throughout their entire length and breadth before a well-lighted window in order to ascertain if there are any defects in dyeing or otherwise, so that these may be remedied before dressing. 724 ORGANIC CHEMISTRY Dressing of Fabrics is effected by impregnating them with solution of gum, bone glue, starch, &c. The fabric is passed beneath a roller dipping into a vessel containing the solution, and is then pressed by a second roller superposed to the first in a kind of foulard like that shown in Fig. 482 ; the vessel may have the section shown in Fig. 483. The gummed fabrics are subjected to mechanical treatment varying according to the type required. Dressing increases the strength and weight of the tissue, which is next dried and at the same time pulled out both lengthwise and breadthwise in order FIG. 479. as nearly as possible to the dimensions they possessed before dyeing. This is effected by means of the so-called tentering frame, into which the tissue passes, fixed laterally by the selvedges on two chains carrying clips or needle-points ; the distance between the two chains is gradually increased to the desired width, which is shown on a graduated iron FIG. 481. bar, A (Fig. 484). Fig. 485 shows a complete frame with s the gumming machine B and two operatives fixing the selvedges on the points of the chains. The widened cloth is dried throughout its whole length by a current of hot air blown into a long chamber beneath, and finally by a' heated drum, C. These frames are 8 to 12 metres long, but are sometimes constructed on several stories in order to save length. Fig. 486 gives a better view of the frame in outline_:_the gummed, centrifuged, and folded cloth lies ready on the two benches, CALENDARS 725 B ; the air is heated at T and the fan V forces the hot air into the long chamber, R ; the cloth enters at jB and issues at G. Milled fabrics and certain others which are required to present a hairy surface are passed to the so-called raising gigs (Fig. 487), consisting of one or more large drums carrying numbers of metallic points or strings of the spiny capsular heads of Dipsacus fullonum (10 to 20 cm. in length, Fig. 488) on spindles. The drums or spindles revolve so that the \ FIG. 482. FIG. 483. points just touch the stretched surface of the cloth and draw from it fairly long hairs, which are then rendered uniform by passing the dry cloth to the cutting and brushing machines furnished with cylindrical brushes and with drums fitted with cutting edges arranged helically (see Fig. 489) ; the first brush, A, raises the hair, the cutter, B, cuts or crops it off uniformly, and the second brush, C, sets it regularly all in the same direction. A similar operation is carried out with velvets, which are, however, woven specially, and often in two superposed pieces attached by a large number of fibres, which are then cut exactly in two so as to give two separate pieces each with a hairy face. FIG. 484. When the fabrics are required to have a very smooth, shiny surface, they~are passed after gumming to the so-called calenders. A common type of the latter for wool and unions, which require but little pressure, is that shown in Fig. 490 : the cloth is seized by the selvedges by two discs fitted with bands, A (called a palmer), which enlarge the cloth to the required size and then pass it on to a continuous felt, C, which transfers it in a well- stretched and compressed condition on to a copper drum, B, heated by steam under slight pressure. For cotton or cotton and silk fabrics, use is made of calenders with several superposed and heated cylinders to which pressure may be imparted by means of suitable levers (Fig. 491 ) in such a way as to exert a kind of friction on the cloth passing from one cylinder to the other. When a very high finish is required on certain satin fabrics of cotton, they are passed between two massive steel cylinders which are under very high pressure 726 ORGANIC CHEMISTRY (hydraulic) and one of which is fluted with very fine striations (as many as 10 to 25 per millimetre, as suggested by Schreiner) ; these leave their stable imprint on the fabric like so many minute, shining cylinders like silk fibres, which reflect light under any angle ; .;. r . FIG. 485. this finish is known as silk finish (or Schreiner finish). Similar calenders are used for obtaining special watered effects (moire). On woollen fabrics calenders generally produce a so-called false finish like that of a bright sheet of metal. This is not regarded as desirable by the merchants, and, further, FIG. 486. such a finish will show rain- drops, even after drying. In order to avoid this inconvenience and the better to fix the material in both directions, so that it will not shrink when worn, it is subjected to so-called steaming, i.e. to the action of steam under a pressure of 2 to 3 atmos. (some colours will not withstand this operation). The fabric is well stretched Fio. 487. FIG. 488. and wound, together with a cloth, round a perforated cylinder ; the roll of two or three pieces thus obtained is wrapped in cloth fastened by strings, the cylinder being then fixed vertically on a steam-cock (Fig. 492). The steam, under pressure, is obliged to traverse the whole of the roll of fabric, and when it issues in a dense cloud (after a few minutes) the operation is at an end ; the roll is then removed, but is allowed to cool without unrolling, PRESSING 727 since in that way it acquires a better and more resistant lustre. The latter is also found to be improved by carrying out the steaming in a vacuum, the rolls G H (Fig. 493) being introduced into a kind of horizontal jacketed autoclave, X, previously heated by passing steam through FIG. 489. the jacket ; when the cover L has been tightly closed, the autoclave is evacuated by passing steam into it and condensing the steam by a water-spray in the cylindrical chamber W, which communicates with the autoclave by means of the tap R. After this the steam is passed through the roll of fabric, either from the inside to the outside or vice versa, by fixing the roll in a suitable manne^-to the steam-cock. FIG. 490. Of the various other operations comprised in the finishing of fabrics, only that of pressing between hot card need be referred to ; this gives lustre to cloths which are not subjected to steaming and in general imparts a very soft, pleasant feel, more particularly to the finer woollens. In this operation, which is the last of importance, the best effect is obtained when 10 to 15 per cent, of moisture is present, so that fabrics which are too dry are treated first with a slight steam- jet, being meanwhile wrapped on drums in large rolls ; after some 728 ORGANIC CHEMISTRY hours these rolls are unwound and the fabric arranged in regular folds, between each adjacent pair of which is inserted a piece of hot, smooth card. The whole is then left FIG. 491. under pressure in a hydraulic press (Fig. 494) for 10 to 12 hours. In order to obtain uniform heating while the pressure is being exerted, presses are now used with double pillars in which steam circulates (Fig. 495) ; also the pillars are sometimes heated electrically. FIG 493. <\ For the folding or rolling of fabrics, and also for measuring, simple and rapid machines have been devised. For the Mercerisation of cotton yarn in hanks (see p. 685) a machine such as that shown in Fig. 496 is used. The uniformly moist skeins, as they come from the centrifuge, are stretched in a thin layer between the two cylinders, A and B, the distance between which FIG. 494. Fia. 496. FIG. 495. 730 ORGANIC CHEMISTRY can be increased so that the skeins are considerably stretched. Then, when the rollers are revolving, a lever is operated to raise the iron vessel, C, containing cold caustic soda solution of 25 to 30 Be., one-half of each cylinder dipping into the soda. At the end of a few minutes the imbibition is complete, the soda solution is drawn off into a tank provided with a pump, while a copious supply of water is sprayed on to the skeins, which are pressed by the roller R. When washing is complete, the tension is relieved and the skein removed. There are also other machines for mercerising fabrics, these being kept stretched by contrivances similar to those used in the tentering frame (see Figs. 484 to 486), while the caustic soda is removed from the fabrics by means of suction pumps. The fabric is then washed with a little hot water so as to give a moderately strong solution of caustic soda, which may be used to dissolve solid caustic soda or may with advantage be con- centrated in multiple-effect evaporators (see vol. i, p. 442). The caustic soda is removed completely from the fabric by thorough washing in cold-water, then in a slightly acid bath and finally in water. The Printing of textiles, as indicated on p. 707, is carried out by pressing, with a FIG. 497. FIG. 498. rubber roller, A (Fig. 497), the fabric or yarn against a copper cylinder, B, on which the design is engraved. The copper cylinder receives the pasty colour from a roller, /, dipping into the vessel, C, containing the thickened colour solution, a blade, D, then scraping away the excess of colour so that only the hollows of the design remained filled. Between the rubber cylinder and the fabric, T, to be printed runs a continuous band, E, which is kept taut by the con- trivance V. The arrangement used, with the adjacent drying chamber, o, is shown in Fig. 498 : the vessel of colouring-matter is at cd, and the fabric is unwound from g together with the accompanying cloth h, and the continuous pressure cloth i ; the dyed and dry fabric is collected in folds at I, while the cloth h is rewound at r, and i returns constantly to the printing cylinder. Wh>n several colours are to be printed on one and the same fabric, a number of rolls and colour vessels are required, ^ IG 499 as is shown diagrammatically in Fig. 499. Fig. 500 shows a complex machine for the printing of textiles in twelve colours at once ; highly skilled workmen are required to regulate its working with accuracy. COLOUR-PRINTING MACHINES 7B1 A simple arrangement for printing yarn in skeins by hand is shown in Fig. 501. The skeins are kept taut between the rods A and B and the printing rollers, which are not very clear in the figure, are below A. The printed skeins are hung on rods fitted to a framework, this being introduced into an autoclave to be treated with steam under pressure (Fig. 502). FIG. 500. Printing colours are boiled with the thickening agents in suitable double-bottomed boilers, heated by means of steam and furnished with stirrers. Fig. 503 shows a battery of such colour-pans. Fra. 501. FIG. 502. When mention was made of aniline black (p. 662), it was stated that the complete develop- ment of this colour is obtained in an oxidation chamber (Fig. 504). In the case of yarn, the method of continuous drying illustrated in Fig. 480 (p. 724) gives good results. But with fabrics use is generally made of a chamber with revolving rollers, where the fabric FIG. 503. FIG. 504. FIG. 505. PROTEINS 733 traverses slowly a very long path and isaues completely black ; a hood is arranged to carry of? acid vapours. Of great importance in this operation is the regulation of the tem- perature, of the draught and of the velocity with which the fabric passes through the chamber. Unexpected stoppages are dangerous, as they may lead to corrosion of the fabric or alteration of the colour. To polish and soften silk, the skeins are stretched, twisted, and rubbed repeatedly on a smooth rod fixed in the wall. But nowadays this is done by machines (Fig. 505), which act automatically and give a large output. T. PROTEINS OR ALBUMINOIDS These are fundamental products in the formation and constitution of animal and vegetable organisms. The protoplasm of vegetable and animal cells, which is the origin of the metabolic processes and hence of the life of the organism, consists of protein substances, which are also indispensable components of foodstuffs. From a physiological point of view they are therefore of the utmost -sig- nificance, but their chemical nature is very complex and is still little understood, although the investigations of Emil Fischer and a number of able collaborators during the past ten years have to some extent pierced the veil surrounding this most important group of organic compounds, which had been previously studied, as regards some of their more superficial characters, by Ritthausen, Hoppe-Seyler, Hammarsten, Neumeister, Pfliiger, Hedin, Kuster, Nencki and Sieber, &c. The numerous substances comprised in this group are all composed of C, H, 0, N, and S, with, in a few cases, P ; their percentage compositions vary between the following limits : C, 50 to 55 ; H, 6-9 to 7-3 ; 0, 19 to 24 ; N, 15 to 19; S, 0-3 to 2-4. The molecular magnitudes of these substances cannot be established with certainty, since it is not easy to isolate single individuals, only very few of them crystallise, none are transformable into vapour, and in no case are true solutions obtainable capable of cryoscopic or ebullioscopic measurement ; their solutions are colloidal. Direct or indirect attempts to determine their molecular weights have given numbers varying from 10,000 to 30,000. Both the sulphur and the nitrogen occur in two groupings, being partly removed by hot potash and partly more stably combined. Absolute alcohol coagulates proteins and precipitates them to some degree unchanged from their aqueous solutions. They are also precipitated unaltered by solutions of sodium chloride, magnesium sulphate or ammonium sulphate of different concentrations, which are characteristic of the various proteins. Proteins are coagulated and precipitated from their aqueous solutions by small quantities of mineral acids (nitric acid may be in excess). They have a feeble acid character and form salts as insoluble precipitates with metallic salts, e.g. ferric chloride, acidified mercuric chloride, copper sulphate, &c., and they dissolve small amounts of freshly precipitated ferric hydroxide. From these metallic precipitates proteins are liberated in a changed form. Less pronounced is their basic character (like the amino-acids, they behave as both acids and bases at the same time), although egg-albumin is completely precipitated by weak acids, such as tannin, phosphotungstic acid, and picric acid. Aqueous solutions of the proteins^are coagulated on heating to different characteristic temperatures, and the coagulated proteins dissolve only in an excess of acid or alkali in the hot, their constitution being modified thereby and H 2 S and NH 3 sometimes evolved : with alkalis they form albuminates and with acids, Acid-Albumins (syntonins, see p. 737), both insoluble in water and reprecipitable by neutralisation. By the protracted action of these two 734 ORGANIC CHEMISTRY reagents (Hydrolysis, seebelow) or by the action of pancreatic juice, which contains Tryptase (seep. 112), they yield various amino- or diamino-acids : glycocoll, alanine, phenylalanine, aspartic acid, glutaminic acid, leucine (in abundance), pyrrolidinecarboxylic acids, tyrosine, serine, triaminotrihydroxydodecanoic acid, /3-indoleacetic acid, arginine, lysine, ornithine, tryptophane, cystine (sulphur compound), &c., all of them optically active with the exception of glycocoll. When a piece of boiled egg-albumin is heated at 37 with gastric juice, it rapidly dissolves with formation of Peptones and Albumoses. The peptones, passing into the intestines, undergo further hydrolysis, and as final products yield amino-acids. The complete hydrolysis of the albumin may be effected more rapidly by means of a concentrated acid (e.g. HC1), which gives amino-acids and also ammonia. By putrefaction various other substances are formed : Ptomaines, such as cadaverine (see p. 214), putrescine or tetramethylene- diamine, &c. ; glucosamine, methylamine, ammonia, /3-indoleacetic acid, phenyl- acetic acid, carbonic acid, hydrogen sulphide, formic to caproic acids, partly of normal structure and partly optically active (valeric and caproic), &c. ; indole, skatole, phenol, cresol, mercaptan," methane, &c., all of these being oxidation or reduction products of the original compounds obtained. The action of pathogenic bacteria on proteins yields poisonous substances, the Toxalbumins, which are similar in composition to the proteins and lose their toxicity when their aqueous solutions are heated. The following reactions are characteristic of the proteins : Protein solutions give a violet coloration (like biuret) with alkali and a few drops of .2 per cent, copper sulphate solution (biuret reaction). With nitric acid in the hot and even in excess a yellow precipitate is formed (xanihoprotein reaction). With Millon's reagent (see p. 704) a red coagulum is formed on boiling. The degradation or hydrolysis of proteins, when it is complete and takes account of all the more or less complex groups composing the protein molecule, will permit of an attempt, with probability of success, to synthesise these substances completely, Such more or less gradual decompositions are attained by protracted heating (for different times with different proteins and in some cases for 200 hours) in an autoclave, or by means of soda or baryta (Schiitzen- berger), or, better, 25 per cent, solutions of hydrochloric or sulphuric acid. But even under these conditions some of the intermediate compounds cannot be detected, the hydrolysis being in many cases too rapid. Hugounenq and Morel (International Congress of Applied Chemistry, London, 1909) have obtained a somewhat more gradual hydrolysis by using 15 to 25 per cent, hydrofluoric acid solutions and heating for 100 to 150 hours. The separation of the numerous amino-acids resulting from the hydrolysis of the proteins constitutes a difficult problem, which has recently been solved by E. Fischer for the amino-acids and by Kossel for the diamino- acids. Fischer subjects the esters of the amino-acids to fractional distillation in vacuo and thus determines their separate amounts. It is thought that the amino-acids occur in the proteins in a condensed form similar to Glycylglycine, NH 2 -CH 2 -CO -NH -CH 2 -C0 2 H. Indeed Fischer was able to synthesise the so-called Polypeptides, which contain such groups and in many respects resemble the natural peptones derived from proteins (see later) ; the esters of the amino-acids readily give up alcohol and undergo ketonic condensation to polyanhydrides, and tjiese, under the influence of alkali, take up a molecule of water, giving the peptides : 2NH 2 -CH 2 -C0 2 C 2 H 5 = 2C 2 H 5 -OH + NH<^' C ^>NH (and this + H 2 0) Ethylglycocoll Double Anhydride or Dlketopiperazine ALBUMINS 735 NH 2 CH 2 CO NH CH 2 C0 2 H. Dipeptidc or Glycylgylcine By chlorinating the carboxyl of the dipeptide with PC1 5 in acetyl chloride solution, a second molecule of ethylglycocoll may be caused to react with formation of a tripeptide, and so on, higher polypeptides similar to the natural ones being ultimately obtained, X---CO-C1 + NH 2 -CH 2 -C0 2 -C 2 H 5 = HC1 + X- -CO-NH-CH 2 -C0 2 -C 2 H 5 ; these polypeptides are completely hydrolysed by hot concentrated HC1, are digested by tryptase, withstand cold alkali, are soluble in water and insoluble in alcohol, and give the reactions of the proteins (see below). These syntheses, which represent the first small step towards the synthesis of the proteins, give an idea of the enormous difficulties to be overcome before the natural proteins can be reconstructed. Indeed, since the dipeptides have molecular weights of about 100, while with the proteins the molecular weight certainly exceeds 10,000, at least 100 of these groups must be present. Also, as several of the amino-acids contain one or more asymmetric carbon atoms, stereoisomerism is possible, and so likewise is tautomerism, e.g. -HN-CO N:C(OH) The investigations of Fischer have resulted in the synthetical preparation of more than a hundred of the simpler polypeptides, the highest of which is an octadecapeptide ; but on ascending the series the complications and difficulties increase disproportionately. This problem could occupy a whole generation of chemists, and its solution would be a glorious triumph for the twentieth century, as it would banish for ever the Malthusian threat that one day humanity will be starved owing to the disproportion between the population and the productive capacity of the earth. Indeed, while it is not possible to replace the proteins in human nutriment by fats or carbohydrates these alone leading to rapid decay of the organism and to death proteins of themselves are able to supply all the needs of the organism. So that the insufficient production of proteins in nature at some future time would of a certainty be accompanied by famine, unless a method of synthesising proteins by chemical means had previously been discovered. Berthelot imagined that one day the air would supply the oxygen and nitrogen, and water the hydrogen for this synthesis. And it is not for us to deny that the dream of yesterday may become the reality of to-morrow, if chemistry learns how to imitate the simplicity and economy of the natural synthetical processes best exemplified in plants, which from carbon dioxide, water, and nitrates are able to effect continuous production of carbohydrates, fats, and proteins. Our laboratory synthetical methods are still too cumbersome, too indirect, and generally too costly. Only when the action of catalysts and light and the laws of colloids have been more closely studied can any hope be entertained of a more rapid progress in the synthesis of such complex organic substances. The numerous different proteins are usually classified in the following groups and sub- groups : I. NATURAL PROTEINS (1) ALBUMINS (of eggs or Egg-albumin, of blood serum or serum-albumin, of milk or lactalbumin, of muscles, of plants, &c.). These are the most common and also the best known of the proteins, since they can be isolated as definite, crystalline, chemical individuals. They are soluble in water, dilute acid or alkali, or neutral solutions of NaCl, MgSO 4 , or (NH 4 ) 2 S0 4 (the globulins being 736 ORGANIC CHEMISTRY insoluble), but in acid solution these salts precipitate the albumins. In the hot they are coagulated. The products of the putrefaction of albumin contain also p-Hydroxyphenylacetic Acid, OH C 6 H 4 CH 2 CO 2 H, which occurs likewise in urine (acicular crystals coloured greenish by ferric chloride). There exists nowadays a considerable trade in dry albumin obtained from the egg and from blood. In various countries, egg-yolks l are preserved in salt and employed in different industries (for tanning, making lecithin, culinary purposes, &c.), and the fresh white sepa- rated is diluted with a little water, beaten until it forms a froth, allowed to stand until the latter is destroyed, filtered through woollen bags, and evaporated in a stream of air at 30 to 40 in large shallow pans ; after 40 to 60 hours there remains a thin, yellowish, transparent pellicle, which is completely soluble in water and keeps without developing any unpleasant odour. From fresh blood (from the butcher's) pure albumin is separated with greater difficulty. The blood is first allowed to undergo spontaneous coagulation, the blood globules and other impurities thus collecting in a compact mass so as to allow of the ready decantation of the faintly coloured liquid serum containing the albumin ; or, after coagulation, the blood may be introduced immediately into a centrifugal separator (see p. 395). The centrifugcd or decanted liquid is beaten (without dilution), filtered, decolorised with charcoal, and dried as above. In many cases decolorisation is difficult, and the albumin has to be precipitated with lead acetate ; the decanted precipitate is washed and suspended in water, which is then saturated with carbon dioxide, the lead carbonate being allowed to settle. The clear albumin solution is treated with a little hydrogen sulphide, which removes traces of lead, and filtered, and the pure solution evaporated as with egg-albumin. According to Ger. Pat. 143,042, the serum-albumin is coagulated by means of salt, dissolved in ammonia and treated at the boiling-point with hydrogen peroxide, the excess of ammonia being subsequently driven off. The method described in Eng. Pat. 10,227 (1905) consists in treating the serum successively with hydrosulphite, acetic acid, and sodium acetate, the liquid being then neutralised with ammonia and evaporated as usual. Albumin is used in various industries : for photographic papers, in textile printing, in printing titles in gold-leaf on books, as a clarifying agent in wine-making (see p. 156), &c. Egg-albumin costs, according to its degree of purity, 24 to 28 per quintal. Blackish blood-albumin is sold at 48s. to 60s. per quintal, the dark at 88s., the pale at 5 to 8, and the pale powdered at 128s. to 208s. (2) GLOBULINS (of plants or Phytoglobulins, Serum-globulin, Lactoglobulin, &c.) are insoluble in water but soluble in dilute acid or alkali. At 30 they are precipitated 1 The eggs produced by different breeds of hens are of varying size, and weight, (from 45 to 65 grms. ; duck, goose, and turkey eggs weigh from twice to four times as much) and are composed of about 60 per cent, of white, 30 per cent, of yolk, and 10 per cent, of shell (mainly calcium carbonate) ; the white contains 86 per cent, of water and 13 per cent, of albumin, and the yolk about 51-5 per cent, of water, 28-5 per cent, of fats, 15-8 per cent, of proteins (principally vitellin), 2 per cent, of salts, 0-45 per cent, of cholesterol, 1-2 per cent, of phosphoglyceric acid, and 0-4 per cent, of extractive substances. As regards its nutritive value, an egg weighing 60 grms. is equivalent to 50 grms. of meat, while its heat value is about 80 calories. Continuous evaporation of water takes place thiough the shell of the egg, and the volume of the contents diminishes, leaving a free air-space varying in size in different eggs which may be observed by looking through the egg at a candle flame in a dark chamber. Fresh eggs are also distinguishable from stale ones by the specific gravity : fresh eggs sink in a salt solution of sp. gr. 1-078, those 2 to 3 weeks old in one of sp. gr. 1-060, those 3 to 5 weeks old in one of T050, and rotten eggs in one of sp. gr. 1-015. It has also been observed that fresh eggs float horizontally on a denser liquid, those 4 to 6 days old at an angle of 20, those 8 to 10 days old at an angle of about 45, and those 15 to 20 days old at an angle of 60. The preservation of eggs is of considerable importance, since in summer eggs are abundant and cheap, while in winter they are scarce and cost double as much. A common means of preservation formerly employed consisted in immersing the eggs in water saturated with lime (which partially filled up the pores of the shell with calcium carbonate), but in this way they acquire an unpleasant taste ; an improvement is effected by adding 5 per cent, of sodium chloride to the lime water. Others preserve them in pounded salt or in salt and bran, pointed end down, while others again smear them with wax, vaseline, and oil or tallow. Large quantities of eggs are now preserved for some months (May to November) by placing them in thin layers on wooden lattices in cold chambers, which are kept at a temperature of 1 to 2 and a humidity of 70 to 80, and afe well ventilated, preferably by means of an apparatus producing ozonised air. In certain cases good results are obtained by preserving the eggs in 10 per cent, sodium silicate solution, although such eggs often burst during subsequent boiling. A mere coating of the silicate or of collodion is of little avail. For transport eggs are arranged in layers, with alternate layers of old straw, in wooden boxes. It is estimated that Italy produces from 5,000,000,000 to 6,000,000,000 eggs (about 60,000,000 being hens' eggs) and the exportation, which in 1905 exceeded 320,000 quintals, worth about 1,800,000, fell to 228,500 quintals in 1907, a large part of the English market (to which France sends more than 1,000,000,000 eggs) and also of the German market (captured by Ilussia and Denmark) being lost. PROTEIDS. 737 unchanged, completely by solutions of ammonium or magnesium sulphate and partly by sodium chloride solution. Their solutions are coagulated by heat. (3) NUCLEO-ALBUMINS (Vitellin, Casein, &c.) are acid in character and decompose carbonates ; they are slightly soluble in water, but dissolve with formation of salts ,in caustic soda or ammonia and are then coagulated neither by heat nor by alcohol. They contain phosphorus (0-85 per cent, in casein) but are distinct from the nucleo-proteins, which give xanthine bases among their decomposition products. Casein is found in milk (see p. 385) and is coagulated by rennet or by dilute acids at 50 ; it is soluble in borax or potassium carbonate and is rendered insoluble by formaldehyde. Converted into salts in various ways, it is placed on the market as a concentrated and readily digestible food (plasmon, nutrose, tropon, &c.) ; it is mixed with mineral colouring- matters to make var- nishes. The hydrolysis of casein yields various amino-acids and complex tribasic acids (Skraup). Vegetable caseins are also known. To obtain pure caseinin the laboratory, diluted skim-milk to which 0-5 per cent, of acetic acid has been added is heated to 55 to 60 and the precipitated casein collected on cloth, washed well with water, redissolved in very dilute ammonia, decanted or filtered to remove the undissolved fat and nuclein and then reprecipitated with acetic acid as at first. It is again collected on cloth, washed with alcohol and then with ether, and dried in a vacuum. Prepared in this way, it is free from fat, leaves less than 0-5 per cent, of ash and contains 1 5-5 to 18 per cent, of nitrogen. From ordinary casein a modification known as paracasein, containing 14-8 to 15 per cent, of nitrogen, may apparently be separated. Commercial casein (see p. 385) contains less than 3 per cent, of ash and less than 0-1 per cent, of fat, and costs 64s. to 80s. per quintal. Riegel (Ger. Pat. 117,979 of 1900) precipitates it in a highly pure state from milk by means of ethylsulphuric acid. Casein is detected on textiles or paper by Adamkiewicz's reaction, a drop of a mixture of glyoxylic and sulphuric acids being placed on the surface, which is then gently heated over a flame : in presence of casein, the drop of liquid assumes a transitory violet-red colour. (4) PROTEINS WHICH COAGULATE (Fibrinogen, Myosin, &c.) are distinguished by exhibiting a first coagulation under the influence of certain enzymes and a further coagulation by heat or absolute alcohol. (5) HISTONES (Globin, Nucleo-histone, &c.) contain sulphur and are markedly basic in character ; they are precipitated by alkalis, and in acid solution give insoluble compounds with the albumins. Nucleo -histories are obtained from the leucocytes of the thymus gland and from the testes of certain fish. The protein part of the haemoglobin molecule of the red blood corpuscles consists of a histone, globin. The histones have certain pro- perties in common with the peptones and albumoses. (6) PROT AMINES (Salmin, Clupein, Sturin, &c.) do not contain sulphur but contain up to 25 per cent, of nitrogen and are composed mainly of diamino-acids (arginine) ; they are obtained from the spermatazoa of many fishes (salmon, herring, sturgeon, &c.). They and the histones are the least complex proteins. They are still more basic in character than the histones and readily form platinichlorides, sulphates, and picrates, which are all crystalline. They are precipitated by dilute alkalis. II. MODIFIED PROTEINS (1) ALBUMOSES and PEPTONES are derived from true proteins by various trans- formations. The albumoses are soluble and cannot be coagulated, but are precipitable by ammonium sulphate and other salts. The peptones are regarded as the last decomposition products of the proteins which give protein reactions (the biuret reaction) ; on decomposition they give amino-acids without intermediate products. (2) SALTS OF PROTEINS (Syntonins or Acid-albumins, Albuminates) are markedly acid in character. III. CONJUGATED PROTEINS (PROTEIDS) These represent combinations of proteins with other complex substances, and are coagulable by alcohol. (1) HAEMOGLOBIN is the colouring-matter of red blood corpuscles and seems to be composed of a protein combined with a colouring-matter containing iron, as it can be decomposed into albumin and Haematin, Pe(C 1G H 3 2O 2 N 2 )<,, the latter being a brown II 47 738 ORGANIC CHEMISTRY substance containing 8 per cent, of iron. The haemoglobin of venous blood is of considerable importance in respiration, as it combines very readily with atmospheric oxygen (when the blood traverses the lungs) forming Oxyhaemoglobin, which is found in arterial blood and carries the oxygen to the tissues, afterwards returning to the veins. With acetic acid and sodium chloride it gives hccmatin hydrochloride (hcemin) in characteristic, microscopic crystals in the form of reddish brown needles. Blood-spots (even old ones) may be detected by Teichmann's test : to a solution of the spot in a little glacial acetic acid are added a trace of sodium chloride and then a small quantity of pure concentrated acetic acid, the liquid being heated to boiling on a watch-glass and one or two drops of the hot solution placed on a microscope slide and allowed to evaporate slowly in the cold ; a drop of water is added, a cover-glass applied, and the slide observed under the microscope. The brown hsemin crystals resemble barley-corns, but are sometimes rhombohedral and generally crossed in groups (Fig. 506) ; viewed in polarised light between crossed hicols, they appear luminous and golden on a dark ground. They are insoluble in water or cold acetic acid, but dissolve in alkali. Blood-stains may also be identified by means of the catalytic action of the haemoglobin, which colours alcoholic guaiacol tincture or alkaline phenolphthalein previously decolorised by zinc-dust or, better, the leuco-base of malachite green ( F - Michel > 191 1). 1 T. Gigli (1910) states that a very sensitive reaction is given by a fresh mixture of 3 drops of benzidine (5 per cent, solution in acetic acid) and 2 drops of 3 per cent, hydrogen _ TT . , peroxide solution ; a blue coloration is given immediately FIG. 506. Hsemin crystals at & i different magnifications. b y a trace of blood. Bardach and Silberstem (1910) pro- pose the use of guaiacum resin and sodium perborate. Oxyhaemoglobin has a composition differing little from that of the proteins, but it contains 0-4 per cent, of iron combined in the ferric state, as with haemin and haematin, whilst the reduction product of the latter, -i.e. haemoglobin, is a ferrous compound (W. Kiister, 1910). In a vacuum (or under the action of ammonium sulphide) it loses oxygen, giving haemoglobin. Haemoglobin forms a red crystalline powder soluble in water and reprecipi table in the crystalline state by alcohol. Both haemoglobin and Oxyhaemoglobin give characteristic absorption spectra. Haemoglobin and also its ash exert a catalytic action in certain combustion phenomena ; e.g. sugar moistened with a little human blood burns with great energy. When a current of carbonic oxide is passed into a solution of red Oxyhaemoglobin (defibrinated blood) the oxygen is x displaced and the liquid assumes a violet-red colour, carboxyhcemoglobin which can be obtained in bluish crystals being formed. An aqueous solution of this compound (blood poisoned with carbonic oxide) gives two characteristic absorption bands between the D and E lines of the spectrum, and these bands do not unite or disappear as happens in the case of Oxyhaemoglobin when a few drops of ammonium sulphide are added to the solution. Haemoglobin itself gives a single absorption band between the D and E lines. (2) NUCLEOPROTEINS or Nucleins have a pronounced acid character and are insoluble in water and acids, but soluble in alkali. They represent compounds of proteins 1 Blood-spots may also be detected by means of hydrogen peroxide : it is sometimes sufficient to press a piece of moistened filter-paper on the dry blood-spot and then to immerse it in hydrogen peroxide solution, to obtain a copious evolution of oxygen. To ascertain from what animal the blood comes, and in general to discover if it is human blood, Uhlonhuth's test (1909), based on the formation of different antitoxins in different animals (see p.115) serves. Tristovitch and Bordet (1899) showed, indeed, that if an extraneous serum (e.g. human) is injected in several doses into the blood of an animal (e.g. a guinea-pig), the serum of this animal (antiserum) ultimately acquires the property of precipitating (or rendering turbid in the case of dilute serum or dilute blood) the blood of the animal which furnished the injected serum (e.g. man). If even a very dilute solution of blood (obtained, for instance, by extracting a dried blood-spot with a little water) is cleared by filtration and treated separately with different clear antiserums to ascertain with which of them a turbidity is produced, it can be stated with certainty Wiat the blood-spot was derived from the animal whose serum, when injected into the guinea-pig, produced the antiserum rendering the blood solution turbid. The test must be applied very carefully and with parallel control experiments ; it does not distinguish between the bloods of similar animals, e.g. hens and pigeons, sheep and goats, apes and men. The difference between various species becomes more evident when dilute solutions or, better, dilute blood and a little concentrated anti- serum are employed. All these phenomena, studied by Uhlenhuth, and subsequently by others, are based on the precipitation of the albuminoid substances of the different serums (precipitins), and they allow of the determination of the character of blood-spots sixty years old. Clear solutions and sterilised vessels are always used for the test. ALBUMINOIDS with a Nucleic Acid, which is phosphoric acid neutralised partially by basic organic groups, such as xanthine, guanine, &c. The nucleins contain 5-7 per cent. P, 41 per cent. C, and 31 per cent. O, and are hence sharply distinguished from true proteins although they give the same colour reactions. They form the fundamental constituents of cell nuclei. (3) GLUCOPROTEINS are acid in character and are formed of a protein combined with a sugar derivative. They are insohible in water and with a little lime-water give neutral, frothy, and ropy solutions which are not coagulated by heat or by nitric acid. When hydrolysed with alkali or acid they yield sugar, peptones, and Syntonins. These compounds, which are poor in nitrogen (11-7 to 12-3 per cent.), include the Mucins. IV. ALBUMINOIDS These constitute the fundamental parts of the cartilaginous tissues and epidermis of animals and comprise : (1 ) ELASTIN, which forms the elastic part of the tendons and ligaments, is insoluble in dilute acid or alkali, but with the latter loses the whole of its sulphur. (2) KERATIN is the principal constituent of the nails, horns, feathers, epidermis, hair, &c. It is insoluble in water, but when heated under pressure, best in presence of alkali, it dissolves with partial decomposition. It contains 4-5 per cent, of sulphur, which is eliminated to some extent by boiling water. With nitric acid it gives the yellow xanthoprotein reaction (see above, Blood-spots on skin treated with nitric acid). (3) The COLLAGENS are abundant in bones, hair, tendons, and cartilage. They combine with water at the boiling-point and dissolve, forming ordinary glue or gelatine, which is precipitated by tannin or by mercuric chloride acidified with HC1 but not by mineral acids. They contain stably combined sulphur. They consist, to the extent of 85 per cent., of amino-acids (Skraup, Biehler and Bottcher, 1909-1910), and, like the protamines, are true proteins containing methoxy- and azomethyl-groups. Unlike casein, they give little glutamic acid on hydrolysis. On hydrolysing them with caustic baryta, E. Fischer and R. Boehner (1910) obtained Proline [a-pyrrolidinecarboxylic acid) as primary product ; a -Amino-o-hydroxy valeric Acid, which is obtained from gelatine, does not give proline with baryta. By digesting gelatine with trypsin, Levene (1910) obtained mainly Prolylglycocoll Anhydride. The absorptive power of the collagens for carbon disulphide, which in presence of alkali leads to thiohydration, allows of their differentiation from agglutinating substances (Sadikow, 1910) ; the agglutination of gelatine is not only a disgregation of the collagen molecule, but also a condensation of the side-chains. Gelatine which has undergone prolonged exposure to light loses some of its absorptive power for water owing to the formation of formaldehyde, which hardens the glue (Meisling, 1909). On hydrolytic decomposition, the collagens give glycocoll (while the albumins give tyrosine), leucine, glutamic acid, and asparagine. 1 Very dilute solutions of glue give, with boiling ammonium molybdate solutions, a characteristic precipitate and coloured solution, which may be applied to quantitative estimations (E. Schmidt, 1910). 1 Manufacture of Glue and Gelatine. The prime materials are bones and hide waste, generally untanned and preserved with lime. From bones the fat is first extracted (see p. 392 and also vol. i, p. 508), and the crushed bones then heated for a couple of hours in a large autoclave with water and steam under pressure, so as to convert the ossein into soluble gelatine ; this treatment is repeated two or three times, the final more dilute solutions being used for a subsequent operation. In some cases, however, the bones and hence also the glue are freed from calcium phosphate by treatment with four times their weight of 6 to 7 per cent, hydrochloric acid (sp. gr. 1-05) until complete softening occurs ; the calcium phosphate is precipitated from the solution by means of lime and calcium carbonate, while the ossein, placed in a double-bottomed vessel heated by steam, is rapidly converted into a solution of glue. According to Ger. Pat. 144,398, the calcium phosphate maybe dissolved by aqueous SO 2 under pressure (only the treatment under pressure is patented). The solution obtained by either of these methods, with a concentration of 17 to 18 on the glue-densimeter in summer and 12 to 14 in winter, can be partly decolorised, while still hot, by a current of sulphur dioxide ; it is then introduced into zinc moulds surrounded by cold water to solidify. The solidification of these solutions (and even more dilute ones) is now hastened by refrigeration. The solid blocks of glue are then cut into suitable sizes and dried on wide-meshed nets arranged on trolleys, which are placed in chambers through which air at 25 to 30 is circulated by means of fans. If the air is above this temperature the glue will melt, while if it is" too dry the cakes are deformed. On this account and also because it would readily putrefy, glue is not made in summer. Dry bone-glue contains 15 to 20 per cent, of water and costs 52. to 68. per quintal. Skin-glue (leather glue) is prepared from hide-waste and also other waste (nerves, cartilage, Ac.) by defatting with carbon disulphide and softening or swelling in water, which likewise removes impurities. It is then macerated for three weeks in a series of vessels containing milk of lime, which is frequently renewed to eliminate any remaining fat, blood, &c. It is then thoroughly washed in water and the last traces of lime (which would make the glue turbid) removed by means of dilute hydrochloric acid, or, better, of sulphur dioxide or phosphoric acid. The 740 ORGANIC CHEMISTRY V. VARIOUS PROTEINS Spongin enters into the formation of sponges ; its hydrolytic products approximate more to those of the collagens than to those of the albumins, but they are more resistant to the action of soda and baryta than collagens. Cornein constitutes coral and gives leucine on hydrolysis. Fibroin and Sericin are obtained from silk (see p. 692) ; fibroin dissolves in energetic alkalis with elimination of ammonia and formation of Sericoin, and when completely hydrolysed it yields tyrosine and glycocoll but not leucine. The Enzymes (see p. Ill) belong to the group of complex albumins. GLUCOSIDES AND OTHER SUBSTANCES OF UNCERTAIN OR UNKNOWN COMPOSITION Glucosides have been denned and the synthesis of artificial glucosides described on pp. 432 and 437. They are compounds of aromatic or aliphatic compounds with carbo- hydrates. In vegetable organisms these glucosides form, according to Pfeffer, difficultly dialysable substances which serve the plants as reserve material, gradually becoming utilisable as they are decomposed by the various enzymes occurring in other cells. This was well shown by T. Weevers (1903 and 1908) for Salicin, which is decomposed (by emulsin) into glucose and saligenin (hydroxy benzyl alcohol), the latter being probably further transformed into a final product known as Catechol. The latter is a phenol found throughout the whole plant (e.g. Salix purpurea), and its quantity is inversely proportional to that of the salicin present ; it is possible that it reacts with fresh quantities of glucose regenerating salicin. While the sugars are gradually utilised in the growth of the plant, the aromatic group (which serves as a reserve of carbon for bacteria but not for enzymes) is used in the continuous reconstruction of the glucoside. So that plants are able to prepare reserve materials in different ways : when the carbohydrates are not utilised, they are transformed into insoluble starch, or into glycogen, or into glucosides. AMYGDALIN, already mentioned on p. 113, has a composition corresponding with C 20 H270 U N and forms colourless crystals which are soluble in water and melt at 200. waste prepared in this way is treated with hot water and steam in wooden vessels with false bottoms and the first solutions, showing densities of 16 to 20 on the glue-densimeter, are solidified in moulds as above. The two or three succeeding extracts, which are more dilute, are concentrated to 20 to 22 in a single or multiple-effect vacuum apparatus (see p. 461), surmounted by a column with perforated discs to break up the froth, and are then allowed to set. Good results are now obtained with Kestner concentrators (see vol. i, p. 443). The waste used gives about one-third of its weight of dry glue, which is sold at 96. to 128*. per quintal. The finer qualities, filtered, decolorised, and prepared from pure, fresh, raw materials, bear the name of gelatine and cost 6 to 12 per quintal. t In order to utilise tanned hides in the manufacture of glue it is necessary to untan them by successive treatment with dilute alkali solution, water, and lime ; if chrome tanned, they are treated first with dilute sulphuric acid, then with an abundant supply of water and finally with lime. In either case, the remaining traces of lime arc removed by means of dilute HC1, the latter being eliminated by treatment with alkali and washing with water (Eng. Pat. 22,738 of 1902). Fish-glue is obtained from the well-purified swimming-bladders of various species of Acipenser, especially of Acipenser sturio (sturgeon), by treatment with acid, lime, steam, water, CH : CH C0 2 H. PHLORETIN, C 15 H 14 O 5 , and its glucoside, PHLORIDZIN, C 21 H 24 O 10 , are found in plants, and in cases of glycosuria in animals. IRIDIN, C 24 H 26 O 13 , is found in the roots of the Florentine iris and yields Irigeninand glucose on hydrolysis. ARBUTIN, C 12 H ]6 O 7 , occurs in the leaves of the bear-berry and gives glucose and hydroquinone on hydrolysis. Methylarbutin gives glucose and methylhydroquinone. CONIFERIN, C 16 H 22 O 8 + 2H 2 O, is hydrolysed to glucose and Coniferyl Alcohol, which gives vanillin on oxidation. It is found in the sap of Coniferce. SINIGRIN (Myronic Acid), C 10 H 17 O 9 NS 2 ; hydrolysis of its potassium salt, which occurs in black mustard seed, gives glucose, potassium bisulphate, and allyl mustard oil. SANTONIN, C 15 H 18 O 3 ; its constitution has been studied more especially by Canniz- zaro and his pupils. It is a naphthalene derivative and is found in worm-seed (santonica). ALOIN, C 17 H 18 7 , an anthracene derivative, occurs in aloes and is a strong pur- gative. LECITHIN (composition, see p. 374) is a characteristic component of egg-yolk and of brain and nerve matter and is a crystalline waxy substance, which dissolves in alcohol or ether and with water forms an opalescent liquid. When hydrolysed it yields glycero- phosphoric, oleic, and palmitic acids, together with choline., and it may therefore be regarded as a glyceride (see pp. 183, 372). Considerable use has been made of it (and also of bromo- and iodo-lecithin) in recent years as a medicine. Lecithin is extracted on the large scale from egg-yolk, and new processes are described in Fr. Pats. 371,391 and 406,634 of 1908. Pure lecithin costs up to 8 per kilo. CEREBRIN, C 17 H :!3 O 3 N, occurs in the nerves. IODOTHYRIN (see vol. i, p. 151) is the iodine compound of the thyroid gland. Bile Compounds include TAUROCHOLIC ACID, C 26 H 45 O 7 NS, and GLYCOCHOLIC ACID,C 2 6H 43 O 6 N, as sodium salts. When decomposed by alkali, both acids yield Cholic Acid, OH-C 2 iH 32 (CH 2 -OH) 2 (CO 2 H), glycine and taurine. Bile also contains colouring, matters such as BILIVERDIN, BILIFUCHSIN, and BILIRUBIN, CtHi,O 4 N,. ORGANIC CHEMISTRY CANTHARIDIN, C 10 H 12 O 4 , occurring in cantharides, causes blistering of the skin, and sublimes in thin scales. CHITIN forms the skeletal matter of crustaceans. It is insoluble in alkali (unlike keratin) and when hydrolysed by acid gives a glucosamine. Fusion with potash at 184 yields acetic acid and Chitosan, which also forms the glucosamine with acid. CHOLESTEROL, C 27 H 46 0, occurs in many plants and animals (that of plants is called Phytosterol), generally together with fats and oils ; certain physical differences but virtually no differences in chemical behaviour are observable in products of different origin. Its constitution has not been definitely established, but, owing more especially to the investi- gations of A. Windaus, many of its component groups have been ascertained. A doubt whether the complex contained one or two double linkings formerly existed, but the addition of ozone (Molinari and Fenaroli, 1908) shows the presence of two such linkings in both phytosterols and other cholesterols. It forms shining scales melting at 147, and in constitution it resembles the terpenes more than the substances of any other group, but in all probability it does not contain benzene groups. Minimal quantities of cholesterol may be detected by Tschugajew's reaction, which consists in the formation of a more or less intense red coloration when a small quantity of a substance containing, cholesterol is poured into fused anhydrous tri- chloroacetic acid. In alcoholic solution, cholesterol and phytosterol (but not their ethers) form an insoluble compound with Digitonin ; this reaction serves for the estimation of these substances and for their separation from other animal and vegetable organic compounds, such as hydrocarbons, &c. INDEX ABRIN, 115 Abrus praecatorius, 115 Accumulators, Hydraulic, 393 Acenaphthene, 614 Acer saccharinum nigrum, 443 Acetaldehyde, 208, 567 Estimation, 209 Acetals, 205, 437 Acetamide, 198, 362 Acetamidine, 357 Acetamido -chloride, 356 Acetanilide, 556, 560 Acetates, 284-287 Acetic anhydride, 320 Acetifiers, 283 Acetimino-chloride, 356 Acetiminothiomethyl hydriodide, 357 Acetins, 214, 223 Acetoacetaldehyde, 334 Acetobromamide, 352 Acetochlorhexoses, 438 Acetometers, 284 Acetonamines, 210 Acetone, 107, 211 Acetonealcohol, 333 Acetonitrile, 198, 199 Acetonylacetone, 333, 334 Acetophenone, 572 Acetophenoneacetone, 572 Acetoxime, 210 Acetyl chloride, 319 iodide, 319 number, 188, 189 Acetylacetone, 334 Acetylcarbinol, 333 Acetylene, 92 hydrocarbons, 90 Acetylethylamine, 351 Acetylglycocoll, 355 Acetylhydrazides, 358 Acetylides, 91, 301 Acetyl-p-phenetidine, 564 Acetylthiourea, 365 Acetylourea, 364 Achroodextrin, 116 Acianilides, 556 Acichlorides, 317 Acid, Abietic, 173, 596 Acetaldehydedisulphonic, 213 Acetic, 270 Acetoacetic, 332 Acetonediacetic, 345 Acetonedicarboxylic, 344 Acetonetricarboxylic, 351 Acetonic, 326 Aceturic, 323, 355 Acetylenecarboxylic, 301 Acetylenedicarboxylic, 315 Acetylsalicylic, 582 Aconitic, 315, 345 Acridic, 636 Acrylic, 294 Adipic, 297, 302, 614 Alkylphosphonic, 202 743 Acid, Allanturic, 366 Allocinnamic, 580 Allocrotonic, 295 Allophanic, 364 Alloxanic, 366 7-Allylbutyric, 297 Allylsuccinic, 312 Aminoacetic, 317, 322, 355 o-Aminobenzoic, 578 Aminoethylsulphonic, 214 a-Aminoglutaric, 356 a-Amino-/3-hydroxypropionic, 355 a-Amino-5-hydroxyvaleric, 739 a-Aminopropionic, 354 Aminosuccinic, 355 a-Amino-/3-thiolactic, 332 Amylacetylenecarboxylic, 300 Amylmalonic, 308 Angelic, 296 Anilidoacetic, 560 Anisic, 577, 582 Anthraflavinic, 616 Anthranilic, 578, 643 Arabonic, 328, 430 Arachidic, 265, 398 Aspartic, 355 Atropic, 577 Azelaic, 305, 311 Azulmic, 359 Barbituric, 366, 368 Behenic, 265 Behenolic, 300, 302 Benzenecarboxylic, 575, 577 Benzenehexacarboxylic, 575, 581 Benzenehexamethanoic, 575 Benzenemethanoic, 575 Benzenestearosulphonic, 410 Benzenesulphonic, 524, 538 Benzhydroxamic, 578 Benzilic, 606 Benzoic, 525, 575, 576 Benzoylacetic, 577 Benzoylformic, 577 Bornylenecarboxylic, 603 Brassidic, 300 Brassylic, 300, 305 Bromosuccinic, 313 Butylacetylenecarboxylic, 300 Butylfumaric, 312 Butylmaleic, 312 Butylmalonic, 308 Butylsuccinic, 310 Butyric, 288 Cacodylic, 202 Caffeic, 584, 741 Caffetannic, 369 Camphoric, 602 Camphoronic, 315, 602 Capric, 289 Caproic, 289 Caprylic, 289 Carbamic, 363 Carbaminic, 363 Carbolic, 541 744 INDEX Acid, o-Carboxyhydrocinnamic, 614 Carminic, 668 Cerotic, 290, 373 Cetylmalonic, 308 Chelidonic, 626 Chloroacetic, 317 Chlorobenzoic, 578 a- (/3- 7-) Chlorobutyric, 318 Chlorocarbonic, 363 a- (^3-) Chloropropionic, 318 Cholic, 741 Cinchomeronic, 626, 637 Cinchonic, 636 Cinnamic, 575, 576, 579 Citraconic, 20, 312, 314 Citramalic, 335 Citric, 345 Citronellic, 297 Citrylideneacetic, 304 Comanic, 626 Coumalinic, 626 Coumaric, 577, 583 Coumarinic, 583 Crotonic, 20, 295 Cuminic, 577 Cyanic, 359 Cyanoacetic, 318 Cyanuric, 359, 362 Cyclogeranic, 303 Decamethylenedicarboxylic, 305 Decoic, 289 Dehydroundecenoic, 300 Desoxalic, 351 Diacetosuccinic, 345 Diacetylenedicarboxylic, 315 Diacetylglutaric, 345 Dialkylphosphonic, 202 Diallylacetic, 303 Dialuric, 366 a6-Diaminovaleric, 328 Diaterebinic, 335 Diazobenzenesulphonic, 568 Dibasic quinolinic, 636 Dibenzhydroxamic, 553 ^7-Dibromobutyric, 295 /3/3-Dibromopro picnic, 318 Dicetylmalonic, 308 Dichloracetic, 318 aa- (a/3-)Dichloropropionic, 318 Diethylmaleic, 312 Diethylmalonic, 308 Digallic, 584 Diglycollic, 322 a/3-Dihydroxybutyric, 295 Dihydroxymalonic, 344 ajS-Dihydroxypropionic, 328 Dihydroxystearic, 299, 328 Dihydroxytartaric, 344 Diisoamylmalonic, 308 Diisobutylmalonic, 308 Dimethylacetic, 288 a/3-Dimethylacrylic, 296 Dimethylarsenic, 202 Dimethylfumaric, 312, 314 aa- (ay-, 77-) Dimethylitaconic, 312 Dimethylmaleic, 312, 314 Dimethylmalonic, 308 Dimcthyloxaminic, 200 Dimethylparabanic, 367 Dimethylpseudouric, 368 Dimethylsuccinic, 310 Dimethyltrihydroxycinnamic, 632 Dinicotinic, 626 Dioctylmalonic, 308 Acid, Diphenic, 606, 619 Diphenylacetic, 606 Diphenylcarboxylic, 572 Dipicolinic, 626 Dipropylmalonic, 308 Dithiocarbamic, 365 Dithiocarbonic, 364 Dithiocarbonylic, 365 Dodecamethylenedicarboxylic, 305 Durenecarboxylic, 577 Durylic, 577 Elseostearic, 304 Elaidjc, 298 - Erucic, 300 Erythric, 328 Ethanetricarboxylic, 345 Ethanthiolic, 351 Ethanthiolthiolic, 351 Ethylacetylenecarboxylic, 300 Ethylcarbonic, 363 Ethyleneaminosulphonic, 356 Ethylenelactic, 325 Ethylenesuccinic, 310 Ethylfumaric, 312 Ethylhydroxamic, 358 Ethylideneacetic, 295 Ethylidenelactic, 323, 325 Ethylidenepropionic, 296 Ethylidenesuccinic, 311 Ethylisopropylmalonic, 308 a- (7-) Ethylitaconic, 312 Ethylmaleic, 312 Ethylmalonic, 308 Ethylmethylacetic, 289 Ethylnitric, 198 Ethylsulphonic, 197 Ethylsulphuric, 89, 197 Ethylsulphurous, 197 Euxanthinic, 668 Ferulic, 584, 741 Flaveanhydric, 359 Formic, 268 Formothiohydroxamic, 360 Formylacetic, 326, 329 Fulminic, 360 Fumaric, 21, 312, 313 Galactonic, 328, 437 Gallic, 577, 583 Gallolylgallic, 584 Gallotannic, 584 Geranic, 303 Glucoheptonic, 329 Gluconic, 328, 433, 438 Glutaconic, 312, 314 Glutamic, 356 Glutaric, 305, 311 Glyceric, 185, 328 Glycerophosphoric, 214 Glycocholic, 741 Glycolglycollic, 322 Glycollic, 322 Glycolsulphuric, 213 Glycoluric, 364 Glycuronic, 329 Glyoxylic, 329 Gulonic, 328 Hemcllitic, 577 Hemirngllitic, 581 Heptoic, 289 Heptylacetylenecarboxylic, 300 Heptylsuccinic, 310 Hexahydrotetrahydroxybenzoic, 59 Hexahydroxystearic, 304 Hexantetroloic, 328 INDEX 745 Acid, Hexylacetylenecarboxylic, 300 Hexylsuccinic, 310 Hippuric, 355, 576, 578 Homocamphoric, 603 Hydantoic, 364 Hydracrylic, 325 Hydratropic, 577 Hydrazoic, 366 Hydrochelidonic, 345 Hydrocinnamic, 577 Hydrocyanic, 358 Hydromellitic, 581 Hydromucic, 312 Hydro muconic, 315 Hydroparacoumaric, 577 Hydro xyacetic, 317, 322 /3-Hydroxyacrylic, 326, 329 Hydroxybenzoic, 577, 582 a- (/3-) Hydroxybutyrio, 326 a-Hydroxycaproic, 326 o -Hydro xycinnamic, 583 Hydroxycitric, 351 Hydro xyethylsulphonic, 213 Hydroxygallolylgallic, 584 a- (/3-) Hydroxyglutaric, 335 a-Hydroxyisobutyric, 326 a-Hydroxyisovalcric, 326 Hydroxymalonic, 334 Hydroxymethylsulphonic, 213 a-Hydroxymyristic, 326 Hydroxyoleic, 326 a-Hydroxypalmitic, 326 /3-Hydroxypelargonic, 327 p-Hydroxyphenylacetic, 583, 736 a-Hydroxypropionic, 323 /3-Hydroxypropionic, 325 a-Hydroxystearic, 326 Hydroxysuccinic, 335 Hydroxytoluic, 577 a-Hydroxyvaleric, 326 Hypogseic, 291 lehthyolsulphonic, 84 Idonic, 328 Iminodithiocarbamic, 365 Iminodithiocarbonic, 365 Iminothiocarbonic, 365 Indoxylic, 638, 644 /3-Iodopropionic, 318 Isatic, 638 Isatinic, 638 Isethionic, 213 Isoamylmalonic, 308 Isoanthraflavinic, 616 Lsobutylaticonic, 314 Isobutylfumaric, 312 Isobutylmaleic, 312 Isobutylmalonic, 308 Isobutyric, 288 Isocinchomeronic, 626 Isocinnamic, 580 Isocrotonic, 21, 295 Isocyanic, 359 Isodurenecarboxylic, 577 Isodurylic, 577 Isoerucic, 300 Isolinolenic, 304 Isonicotinic, 625 Iso-oleic, 299 Isophthalic, 525, 581 Isopropylacetylenecarboxylic, 300 Isopropylfumaric, 312 7-Isopropylitaconic, 312 Isopropylmaleic, 312 Isopropylmalonic, 308 Acid, Isopurpuric, 563 Isosaccharinic, 328 Isosuccinic, 311 Isovaleric, 289 Itaconic, 312, 313 Itamalic, 335 Jecorinic, 304 /3-Ketobutyric, 332 Lactic, 321, 323 Lactobionic, 438 Lanugic, 683 Laurie, 265, 289, 302 Leucinic, 326 Levulinic, 326, 333, 431 Lignic, 506 Lignoceric, 265, 397 Linolenic, 304 Linolic, 303 Lutidinic, 626 Lyxonic, 328 Malamic, 353 Maleic, 21, 312, 313 Malic, 335, 353 Malonic, 3081 368 Maltobionic, 438 Mandelic, 577, 583 d -Harmonic, 328, 436 d.-Mannosaccharic, 436 Margaric, 290, 415 Meconic, 633 Melissic, 290 Mellitic, 535, 575, 581 Mellophanic, 581 Mesaconic, 21, 312, 314 Mesitylenecarboxylic, 577 Mesitylenic, 577 Mesotartaric, 336 Mesoxalic, 334, 344 Metacrylic, 296 Metasaccharinic, 328 Methionic, 192, 213 Methylacetylenecarboxylic, 300 a-Methylacrylic, 296 /3-Methylacrylic, 295 0-Methyladipic, 311 Methylbutylmalonic, 308 l-Methylcyclohexylidenc-4-acetic, 19 Methylenedisulphonic, 213 7-Methylene-y-methylpyrotartaric, 312 Methylenesuccinic, 313 Methylethylglycollic, 326 Methylethylitaconic, 312 Methylethylmaleic, 312 Methylethylmalonic, 308 Methylfumaric, 314 Methylisobutylmalonic, 308 Methylisopropylmaleic, 312 Methylisopropylmalonic, 308 a- (7-) Methylitaconic, 312 Methylmaleic, 314 Methylmalonic, 308 Methylmethyleneacetic, 296 Methylpropiolic, 301 Methylpropylmaleic, 312 Methylpropylmalonic, 308 Monochloroacetic, 317, 321 Monothiocarbamic, 364 Monothiocarbonic, 364 Monothiocarbonylamic, 365 Monothiocarbonylic, 365 Mucic, 344, 437 Muconic, 315 Myristic, 289 Myronic, 741 746 INDEX Acid, Naphthalic, 614 Naphthionic, 613 l-Naphthylamine-4-sulphonic, 613 Nicotinic, 625 m-Nitro benzole, 578 o-Nitrocinnamic, 642 Nitrohydroxylaminic, 206 o-Nitrophenylacetic, 642 o-Nitrophenylpropiolic, 642 Nonoic, 289, 300 Nonylacetylenecarboxylk', 300 Nucleic, 739 (Enanthic, 289 Oleic, 298 Aa/3-Oleic, 299 Orsellinic, 577 Oxalacetic, 344 Oxalic, 306 Oxaluric, 366 Oxamic, 352 Palmitic, 289 Parabanic, 366 Paralactic, 325 Paratartaric, 336 Pectic, 457 Pectosinic, 457 Pelargonic, 289 Pentadecoic, 265 Pentamethylbenzoic, 577 Pentylmalonic, 308 Perinaphthalenedicarboxylic, 612 Perthiocyanic, 360 (8-Phenanthrenecarboxylic, 619 Phenylacetic, 575, 577, 579 Phenylaminoacetic, 560 a-Phenyl-o-nitrocinnamic, 618 Phenylene-o-dicarboxylic, 580 Phenylglycine-o-carboxylic, 643 Phenylpropiolic, 535, 575, 577, 580 Phenylsulphaminic, 560 Phenylsulphuric, 542 Phthalic, 580 Picolinic, 625 Picric, 245, 562, 655 Pimaric, 596 Pimelic, 305, 521 Pinonic, 597 Piperic (piperinic), 584, 626 Piperonylic, 583 Pivalic, 289 Prehnitinecarboxylic, 577 Prehnitic, 581 Prehnitylic, 577 as. Propanetricarboxylic, 345 s. Propanetricarboxylic, 315, 345 Propargylic, 301 Propinoic, 301 Propiolic, 301 Propionic, 288 Propylacetylenecarboxylic, 300 Propylfumaric, 312 Propylitaconic, 312 Propylmaleic, 312 Propylmalonic, 308 Propylsuccinic, 310 Protalbinic, 641 Protocatechuic, 577, 583 Pseudouric, 368 Purpuric, 367 /3-Pyridinesulphonic, 625 Pyrocinchonic, 312, 314 Pyrogallic, 545 Pyroligneous, 276 Pyromeconic .626 Acid, Pyromellitic, 581 Pyromucic, 620 Pyrotartaric, 311 Pyroterebic, 297 Pyrroglutamic, 622 a-Pyrrolidinecarboxylic, 622, 739 a'-Pyrrolidone-a-carboxylic, 622 Pyruvic, 321, 331 Quinic, 592, 636 Quinolinecarboxylic, 636 Quinoline-a : |3-dicarboxylic, 636 Quinolinesulphonic, 636 Quinolinic, 626 Racemic, 20, 336 Rhamnohexonic, 29 Rhodanic, 360 Rhodinic, 297 Ribonic, 328 Ricinelaidinic, 326 Ricinoleic, 326 Ricinoleinsulphonic, 3 7 Roccellic, 305 Rosolic, 608, 647, 660 Rubeanhydric, 359 Ruberythric, 617 Rufigallic, 616 . Saccharic, 344 d-Saccharic, 433 Saccharinic, 328 Salicylic, 179, 577, 582 Santalic, 669 Sarcolactic, 325 Sativic, 304 Sebacic, 298, 305 Sinapic, 632 Sorbic, 303 Sozolic, 564 Stearic, 290 Stearolic, 302 Suberic, 305, 521 Succinamic, 353 Succinic, 305, 310 Sulphanilic, 538 Sulpho-oleic, 407 Talonic, 328 Tannic, 584 Tariric, 302 Tartaric, 20, 335 Tartronic, 185, 334 Taurocholic, 214, 356, 741 Telfairic, 304 Teraconic, 312 Teracrylic, 297 Terebic, 297, 602 Terebinic, 335, 597 Terephthalic, 581, 597 Terpenylic, 297 Tetrabromostearic, 304 Tetracetylenedicarboxylic, 315 Tetrahydroxystearic, 304 Tetrolic, 295, 301 Thioacetic, 351 Thiocyanic, 360 Thiocyanuric, 360 Tiglic, 296 Toluic, 577 Tricarballylic, 303, 315, 345 Trichloroacetic, 318 Trihydroxybenzoic, 583 Trihydroxyglutaric, 343, 430 Trihydroxyisobutyric, 428 Trimellitic, 581 Trimeeic, 329, 581 Trimethylacetic, 289 INDEX 747 Acid, Trimethylenedicarboxylic, 520 aa/3-Trimethyltricarballylic, 315 Trithiocarbonic, 364, 365 Tropic, 577, 631 Umbellic, 584 Undecoic, 289 Undecolic, 302 Uric, 368 Valeric, 288 Vanillic, 583 Veratric, 583 Vinylacetic, 294 /3-Vinylacrylic, 303 Violuric, 368 Xanthic, 365 Xanthonic, 365 Xylic, 577 Xylonic, 328, 430 Acid-albumins, 733, 737 Acidol, 355 Acids, Affinity constants, 266 Heat of neutralisation of organic, 2.~> Alkylsulphonic, 195 Alkylsulphuric, 197 Aminobenzoic, 578 Anthracenecarboxylic, 616 Anthracenesulphonic, 616 Anthraquinonesulphonic, 616 Aromatic, 575 et seq. Azobenzoic, 578 Benzenedicarboxylic, 575 Benzenetetracarboxylic, 581 Benzenetricarboxylic, 575 Benzoylbenzoic, 607 Diazobenzoic, 578 Dibasic, 196 Dihydroxybenzoic, 583 Dihydroxycinnamic, 584 Dihydroxystearic, 326 Diolefinedicarboxylic, 315 Diphenylcarboxylic, 606 Diphenylsulphonic, 606 Heptonic, 329 Hexabromostearic, 304 Hexahydroxystearic, 304 Hexonic, 328, 427 Homoaspartic, 356 Hydrophthalic, 591 Hydroxamic, 206, 358 Hydroximic, 358 Hydroxyolefinecarboxylic, 326 Hydroxypyridinecarboxylic, 626 a-Ketonic, 331 j8-Ketonic, 330 7-Ketonic, 331 Ketonic dibasic, 344 Lactic, 19, 323 Monobasic, 196 Monobasic aldehydic, 329 Monobasic ketonic, 330 Naphthalenesulphonic, 613 Naphthenic, 592 Naphthoic, 614 Naphtholsulphonic, 656 Olefinecarboxylic, 291 Olefinedicarboxylic, 312 Phenolic, 576, 582 Phenolsulphonic, 543, 564 Phthalic, 575, 580 Polybasic aromatic, 580 Polybasic fatty, 304 Pyridinecarboxylic, 625, C2G Pyrotartaric, 311 Quinolinebenzocarboxylic, 636 Acids, Quinolinecarboxylic, 636 Saturated dibasic, 304 Saturated monobasic fatty, 264 Succinic, 310 Sulphobenzoic, 579 Sulphonic, 538 Tartaric, 18, 20, 335 Tetrabasic, 316 Toluic, 579 Tribasic, 196, 315 Unsaturated dibasic, 312 Unsaturated monobasic fatty, 291 Unsaturated monobasic, of the series CH 2 - 4 2 , 300 with two double bonds, 302 with three double bonds, 304 with triple linking, 300 Xylic, 579 Aconitine, 628 Aconitum napellus, 345 Acridine, 647 Acridines, 672 Acrolein, 209 Acroleinammonia, 209 Acroleinaniline, 636 ' Acrose, 329, 428 Acrylic aldehyde, 209 Activators,. 410 Adenine, 368, 369 Adonitol, 431 Adrenaline, 629 Aesculetin, 584, 741 Aesculin,. 584, 741 Affinity constants, 267 Agglutinins, 115 Agro cotto, 347, 348 Alanine, 325, 354 Albumin, Living, 114 Albuminates, 737 Albuminoids, 733, 739 Albumins, 735 Albumoses, 734, 737 Alcohol, Absolute, 109, 144 Acetoisopropyl, 333 Acetone, 333 Allyl, 182 Amyl, 105, 145, 181 Anisic, 573 Benzyl, 570 Butyl, 104, 105, 180 Caproyl, 181 Capryl, 181 Ceryl, 105, 181 Cetyl, 181 Coniferyl, 741 Cumyl, 571 Decyl, 105 Denatured, 145, 149, 152 Dodecyl, 105 Ethyl, 108 Fluorene, 572 Furfuryl, 620 Glycide, 214 Heptyl, 105, 181 Hexadecyl, 105, 181 Hexyl, 181 Hydroxy benzyl, 573 Isobutyl, 105, 181 Isohexyl, 181 Isopropyl, 105, 181 Melissyl, 181 Methyl, 106-108, 142 Monochloroethyl, 21 Myricyl, 105, 181 748 INDEX Alcohol, Nonyl, 105 Octodecyl, 105 Octyl, 105, 181 Oenanthyl, 181 of crystallisation, 105 Pentadecyl, 105 Phthalic, 570 Propargyl, 182 Propyl, 105, 181 Styryl, 571 Tetradecyl, 105 Tolylene, 571 Tridecyl, 105 Undecyl, 105 Vanillic, 573 Vinyl, 182 Xylylene, 570 Alcohol, Amylo process, 129 Effront process, 141 Fiscal regulations, 149 from fruit, 141 from lees, 143 from molasses, 140 from vinasse, 143 from wine, 143 from wood, 142 Industrial preparation, 116 meters, 146 Rectification, 138 Solid, 109 Statistics, 149 Syntheses, 108 Tests, 144, 146 Windisch's Table, 148 Yield, 128 Alcohols, 102 Aldehydic, 329 Aromatic ketonic, 573 Constitution, 103 Derivatives of monohydric, 190 of polyhydric, 213 Dihydric, 182 Higher monohydric, 180 Ketonic, 330, 333 Nomenclature, 104 Polyhydric, 182, 188 aldehydic, or ketonic, 426 Primary, 103, 104 Saturated monohydric, 103, 105 Secondary, 103, 104 Tertiary, 103, 104 Tetrahydric, 188 Trihydric, 183 Unsaturated, 182 Alcoholene, 153 Alcoholism, 150 Alcoholometer, Gay Lussac, 146 Tralles, 146 Alcoholometry, 146 Aldehyde -ammonias, 205 Aldehydes, 96, 103, 204 Aromatic, 570 Determination by Strache's method, 212 Schiff's reagent, 206 with unsaturatcd radicals, 209 Aldehydine, 624 Aldehydo-catalase, 112 Aldims, 574 Aldine, 626 Aldines, 572 Aldohexoses, 427, 431 Aldoketenes, 213 Aldol, 329 Aldols, 205 Aldoses, 426 Aldoximes, 206 Alembics, 133 Alga;, 60 Alipine, 99, 629, 633 Alizarin, 617, 654, 659 astrol, 677 cyanine, 663 irisol, 677 saphirol, 677 Alkaloids, 626 Synthesis, 627 Table, 628 Tests, 627 Alkines, 625 Alkoxides, 103 Alkoxy-groups, Estimation, 542 Alkylanthrahydrides, 616 Alkylenes, 87 Alkylhydrazines, 202 Alkylhydroanthranols, 616 Alkylhydroxylamines, 202 Alkyls, 29 Alkyl halides, 94, 95 Estimation, 100 Alkylisoureas, 364 Allantoin, 367 Allene, 90, 314 Alloisomerism, 21, 22 Alloxan, 366, 623 Alloxantin, 367 Allyl bromide, 102 chloride, 102 iodide, 102 isothiocyanate, 361 mustard oil, 361 thiocyanate, 361 Allylaniline, 636 Allylene, 90, 314 Almonds, composition of sweet, 391 Aloin, 741 Amaranth, 656 Amidases, 155 Amides, 211, 351 of carbonic acid, 363 of hydroxy-acids, 353 Amidines, 199, 357 Amido-chlorides, 356 Amidol, 564 Amidoximes, 358 Amimides, 357 Amines, 200 Aromatic, 555 Amino-acids, 351 Derivatives, 354 Aminoanisoles, 564 Aminazobenzene, 569 Aminoazobenzenes, 565 Aminoazo-derivatives, 568 Aminobenzene, 557 Amino-derivatives of aromatic hydrocarbons, 554 Aminoguanidine, 366 Amino-oxindole. 642 Aminophenols, 562, 563 Aminothiazole, 623 Aminothiophenols, 564 Ammelide, 362 Ammc line, 362 Ammonium carbamate, 363 cyanate, 359 ichthyolsulphonate, 84 thiocyanate, 361 Amygdalin, 113, 740 INDEX 749 Amyl acetate, 371 Amylacetylene, 334 Amylase, 111, 112 Amylbenzene, 527 Amylene, 9J hydrate, 181 Amylodextrin, 116 Amyloids, 504 Amylomyces Rouxii, 130, 131 Amylo process, 129 Anaesthesia, 94, 98, 029 Anaesthesin, 633 Anaesthesiophore, 633 Anaesthetics, 98, 629 Mild local, 633 Analgen, 637 Analysis, Elementary, 7 Qualitative, 6 Quantitative, 7 Anethole, 543, 574 Anhydrides, 319, 320 Internal, 319 Mixed, 319 Anhydro-bases, 557 Anilides, 560 Aniline, 557, 566 hydrochloride, 559, 567 nitrate, 567 platinichloride, 559 salt, 559 sulphate, 559 Anisaldehyde, 574 Anisidines, 564 Anisole, 542 Annatto, 388 Anterea mylitta, 696 Anthracene, 615 derivatives, 616 Anthrachrysone, 616 Anthraflavone, 665 Anthragallol, 616 Anthrahydroquinones, 616 Anthramine, 616 Anthranil, 578 Anthranol, 616, 617 Anthrapurpurine, 616 Anthraquinones, 616, 647 Anthrarufin, 616 Anthrols, 616 Anthrone, 616 Antialdoximes, 210 Anti-bodies, 115 Antichlor, 706 Antidiazo-p-chlorbenzene, 566 Antidiazotates, 567 Antifebrin, 560 Antiketoximes, 211 Antilactase, 115 Antimorphine, 115 Antipepsin, 115 Antipyrine, 623 Antique purple, 664 Antirennet, 115 Antiricin, 115 Antiseptics, 127, 541 Antiserum, 738 Antitoxins, 115 Apigenin, 638 Araban, 429 Arabinose, 428, 429, 431 benzylphenylhydrazone, 429 Arabitol, 189, 427, 430 Arachis nuts, 391 Arbutin, 741 Archil, 544, 667 Arginine, 328 Aristol, 543 Aromatic compounds, 28, 521 Arrack, 160 Artificial hair, 702 Artificial parthenogenesis, 115 Ascomycetes, 111 Aseptol, 564 Asparagine, 19, 358 Aspartamide, 356 Aspergillus oryzse, 130 Asphalte, 59, 83 mastic, 83 Aspirin, 582 Astatki, 67, 74 Asymmetric syntheses, 114 Asymmetry, Absolute, 22 Relative, 22 Atole, 160 Atropa belladonna, 631 Atropine, 628, 631 Attenuation of fermented liquids, 12 \ 13 174 Auramine, 660 Aurin, 608, 660 Auxochromes, 649 Axite, 245 Azides, 358 Aziminobenzene, 557 Azimino-compounds, 557 Azines, 661 Azobenzene, 565 Azo carmine, Acid, 676 Azo-derivatives, 565 Azodicarbonamide, 366 Azoflavine, 675 Azofuchsines, 657 Azolitmin, 668 Azorubin S, 656 Azotoluene, 565 Azoxazole, 623 Azoxybenzene, 565 BACILLUS aceticus, 122, 280 acidi laevolactici, 325 acidificans longissimus, 128 butylicus, 181 Delbriickii, 126 ethaceticus, 328 saprogenes vini, 339 Bacteria, Acetic, 122, 280 Butyric, 122 Chromogenic, 110 Lactic, 122 Pathogenic, 110 Reproduction, 110 Saprophytic, 110 Zymogenic, 110 Bagasse, 444 Balling's Table, 167 Ballistite, 244 Baphia nitida, 669 Barley, 161 Barwood, 669 Bases, Aldehydo-, 557 Aminic, 200 Ammonium, 200 Arsonium, 202 Iminic, 200 Nitrilic, 200 Primary, 200 Quaternary, 200, 556 Secondary, 200 750 INDEX Bases, Tertiary, 200 Vegetable, 626 Beckmann rearrangement, 211, 573 Beer, 161 Alcohol-free, 178 Analysis, 178 Attenuation, 178 Cask pitching, 176 Composition, 178 Detection of antiseptics, 179 Fermentation, 171 Mashing, 168 Pasteurisation, 177 Racking, 176 Statistics, 179 Beet, 445 Cultivation, 446 Production, 478 Sugar-content, 446 Treatment, 448 Beet -pulp press, 453 Benzal chloride, 538 Benzalacetone, 572 Benzalacetophenone, 572 Benzalazine, 573 Benzaldehyde, 571, 595 homologues, 571 phenylhydrazone, 573 Benzaldoxime, 572 Benzamide, 578 Benzanilide, 560 Benzanthrone, 665 Benzene, 528, 533 Artificial, 533 Chloro -derivatives, 537 Derivatives, 521 Formation, 525 Formulae, 522 from Naphtha, 75 Haloid derivatives, 537 Homologues, 534 Hydrogenated derivatives, 591 Isomeric derivatives, 523 Tests, 534 Benzene sulphochloride, 538 Benzenesulphamides, 538 Benzhydrazide, 578 Benzhydrol, 572, 606 Tetramethyldiamino-derivative, 606 Benzidam, 557 Benzidine, 566, 605, 650, 657 Benzil, 609 Benziminazoles, 557 Benzine, Crude, 73 from petroleum, 66 Benzoazurin, 605, 660, 678 blue, 678 blue-black, 678 browns, 679 flavin, 663 orange, 678 purpurin, 605 Benzoin, 609 Benzonitrile, 568 Benzophenone, 572, 606 oxime, 673 Benzopinacone, 572 Benzoquinone, 547 Benzotrichloride, 538 Benzoyl, 14 chloride, 578 Benzoylacetone, 572 Benzoylazoimide, 578 Benzoylcarbinol, 574 Benzoylsalicin, 741 Benzoyltropine, 629 Benzyl bromide, 537 chloride, 537, 553 iodide, 537 Benzylamine, 561 Benzylaniline, 560 Benzyldioximes, 609 Benzylglucoside, 437 Benzylhydrazine, 570 Benzylphenylamine, 560 Benzylphenylhydrazine, 570 Berberine, 633 Bergamot, 347 Betaine, 323, 355 hydrochloride, 355 Betol, 613 Bile compounds, 741 Bilifuchsin, 741 Bilineurine, 214 Bilirubin, 741 Biliverdin, 741 Biogen theory, 114 Bisazo -compounds, 565 Bismuth tribromophenoxide, 101 Bisulphite aldehyde compounds, 205 Bitumen, 83 Biuret, 364 Bixa orellana, 388 Black, Acid azo, 674 Alizarin, 663 ' Aniline, 651, 662, 731 Anthracene, 674 Bone, 470 Columbia, 677 Diamine, 677 Diamond, 657 Fine, 663 Immedial, 677 Naphthazarin, 614 Naphthol, 657 Naphthylamine, 657 Oxidation, 663 Pluto, 677 Sulphur, 671 Vidal, 564, 677 Zambesi, 677 Blankite, 445 Blastomycetes, 111 Bleaching of textiles, 712 Blood, 114 stains, Identification, 738 Blue, Algol, 665 Alizarin, 663 Alkali, 676 Anthracene, 663 Capri, 661 Carmine, 660 Ciba, 664 Diaminogen, 678 Immedial, 678 Indanthrene, 665 Indrone, 665 Janos, 677 Lanacyl, 677 Meldola's, 673 Methylene, 661, 673 Naphthol, 661 Nile, 661 Oxamine, 678 Sulphur, 678 Victoria, 677 Wool, 677 Boghead coal, 84 INDEX 751 Bombyx mori, 690 Boot-polish, 475, 528 Bordeaux, Ciba, 664 Indanthrene, 665 S, 656 Borneo!, 600, 602 Bornesitol, 546 Boudineuse, 231. 24.S Bradolytes, 608 Brandy, see. Cognac Brazilein, 669 Brazilin, 669 British gum, 501 Bromobenzenes, 537 4-Bromomethylfurfural, 430 Bromonitrobenzenes, 553 Bromostyrenes, 538 Bronze, Diamine, 679 Brown, Alizarin, 677 Anthracene, 677 Bismarck, 557, 658. 672. 677 Cibanone, 665 Diamine, 679 Indanthrene, 665 .lanos, 679 Pluto, 679 Pyrogen, 679 Sulphur, 679 Thiazine, 679 Brucine, 628, 633 . Bulgarian ferment, 440 Butandiene, 90 Butandiine, 94 Butandione, 333 Butanes, 91 Butanolone, 333 Butanols, 180 Butanone, 212 Butantetrol, 189 Butenes, 80 Butter, 385 Analysis, 387 Artificial, 387 Cacao, 369 Coco -nut, 402 Renovated, 388 Buttirol, 385 Butyl iodides, 97 Butylbenzene, 527 Butylenes, 80 Butyroflavine, 383 Butyrolactone, 322 Butyrometer, Gerber, 386 Butyrorefractometer, Zeiss, 375 Butyryl chlorides, 319 Byssus, 698 CACAO butter, 369 Cachou de Laval, 670 Immedial, 679 Sulphur, (>79 Cacodyl, 20, 202 chloride, 202 oxide, 202 Cadaverine, 214 Cadmium bromoxylonate, 429 Caffeine, 368, 369, 626 Calcium acetate, 285 benzoate, 526 butyrate, 288 carbide, 92 citrate, 346, 350 cyanamidc, 362 dilactate, 325 Calcium formate, 270 lactate, 325 oxalate, 307 tartrate, 336 Calendars, 725 Calorimeter, Junker's gas, 54 Campeachy, 666 Camphane, 598 Camphene, 597 Campholide, 603 Camphor, 600, 602 Artificial, 597, 603 Camphors, 600 Camwood, 669 Candles, 412 Paraffin, 415 Statistics, 415 Stearine, 414 Tallow, 413 Cannel coal, 84 Cantharidin, 742 Capillarimeter, 149 Caps, 255 Caramel, 433, 435 Carane, 596 Carbamide, 363 Carbamidyl chloride, 363 Carbazide, 364 Carbazole, 605, 615 Carbenes, 83 Carbinol, 104, 108 Carbocyclic compounds, 28 Carbodiimide, 18, 362 Carbodiphenylimide, 362 Carbodynamite, 231 Carbohydrates, 426 Carbohydrazide, 364 Carbolineum, 532 Carbon, Asymmetric, 18, 27 chains, 16 Estimation, 7 sulphochloride, 365 tetrachloride, 101 Valency, 14 Carbon oxychloride, see Phosgene Carbonisation of textiles, 682, 714 Carbonite, 231 Carbonites, 246 Carbostyril, 637 Carboxyhsemoglobin, 738 Carboxyl, 103 Carbyl sulphate, 213 Carbylamines, 199, 556 Carone, 596, 600 Carotin, 692 Cart-grease, 82, 424 Carvacrol, 543, 600 Carvacrylamine, 555 Carvene, 593, 595 Carvol, 601 Carvomenthol, 600 Carvomenthone, 600 Carvone, 601 Caryophyllene, 598 Casein, 385, 702, 737 Vegetable, 737 Castor oil, 326, 393 Catalases, 112 Catalysts, Inorganic, 113 Organic, 113 Catechol, 543 Catechu, 585, 669 Cedrene, 598 Cellase, 112 752 Cellite films, 504 Cellobiose, 504 Cellose, 504 Celluloid, 604, 611 Cellulose, 503 acetate 504 formate, 504 hydrate, 505 Wood, 511 Centrifuges, 385, 468 Cerasin, 61, 81, 85, 86, 597 Cereals, Starch-content, 117, 118 Cerebrin, 741 Cerotene, 90 Cerotin, 181 Ceryl cerotate, 181, 372 Cevadine, 632 Cetyl palmitate, 372 Cetylbenzene, 527 Chalkone, 572 Chamberland flasks, 123 Chamoising, 588 Champy drums, 253 Chappe (silk), 692 Charcoal, Animal, 470 Wood, 106 Chartreuse, 160 Cheddite, 255 Cheese, 385 Filled, 385 Margarine, 382 Chica, 160 Chinovose, 431 Chitin, 742 - Chitosan, 742 Chlamydomucor oryzae, 130 Chloracetanilide, 560 Chloral, 209 hydrate, 209 Chloralamide, 352 Chloramides, 199 Chloranhydrides, 317 Chloranil, 547 Chlorhydrins, 87, 214 Chlorobenzenes, 537 Chlorocruorin, 114 Chloroethane, 97 Chloroform, 98 Pictet, 99 Tests, 100 Chloromethane, 96 Chloronitrobenzenes, 553 Chlorophyll, 111, 669 Chlorophyllase, 669 Chloropicrin, 198 et-Chloropropylene, 102 Chocolate, 368 Cholesterol, 742 Cholestrophane, 367 Choline, 214 Chromogens, 647 Chromone, 637 Chromophores, 595, 647 ft seq. Chromotrop, 676 Chronograph, le Boulenge's, 262 Chrysamine R, 678 Chrysazin, 616 Chrysazol, 616 Chrysene, 619 Chrysin, 638 Chrysoidin, 557 Chrysoidines, 565 Chrysoin, 656 Chrysophenin, 678 INDEX Cider, 159 Cinchona alkaloids, 633 Cinchonidine, 634 Cinchonine, 628, 634 Cinene, 594 Cineol, 602 Cinnamaldehyde, 572 Citral, 182, 209 Citrene, 593, 595 Citromyces citricus, 346 Pfefferianus and Glaber, 34 '. Citronellal, 210, 297 Citronellol, 182 Citrus bergamica, 347 industry, 347 limetta, 347 limonium, 347 Classification of organic substances, 28 Clovene, 598 Clupein, 737 Coagulation, Enzymic, 112 Coal, Cannel, 84 for gas, 36, 37 Coal-gas, 36 Coal-tar, 83 Cocaine, 99, 633 Cocci, 110 Coccus cacti, 668 Cochineal, 668 Cocoa, 368 Coco-nut, Composition. 391 oil (or butter), 378, 402 Cocoons, 690 Production, 695 Waste, 692 Codamine, 632 Codeine, 627, 628, 632 Ccerulein, 650, 651 Ccerulignone, 606 Coffee, 369 Cognac, 144, 160 Colchicine-, 628 Collagens, 739 Collidines, 625 Collodion, 700 cotton, 232, 239, 700 Colophene, 598 Colophony, 420, 596 Colorimeters, 705 Colouring-matters, 646 Aci-aminoanthraquinone, 665 Acid, 650, 655, 673 Acridine, 663 Adjective, 651 Aminoazo, 656 Azo, 655, 673 Basic, 650, 655, 673 Benzidine, 657 Benzo, 657 Chromotrop, 657 Classification of, 654 Coumarin, 663 Diamine, 657 Dianil, 657 Diphenylmethane, 660 Fastness of, 707 Flavone, 663 Hydrazone, 658 Hydroxyazo, 656 Immedial, 670 Indanthrene, 664, 665 Ingrain, 658 Insoluble, 655 Jan os, 658 INDEX 753 Colouring-matters, Kathigenic, 670 Kriogenic, 670 Manufacture of, 652 Monoazo, 656 Mordant, 655, 659 Natural, 651 Neutral, 650 Nitro, 655 Oxyketone, 663 Phenolic, 673 Polyazo, 656 Pyrazolone, 658 Quinoline, 663 Quinonimide, 661, 665 Quinone, 659 Statistics, 653 Substantive, 650, 651, 655, 673 Sulphur, 670 Tests, 671, 680 Theory of, 646, 708 Thiazole, 663 Triphenylmethane, 660 Vat, 664 Water-soluble, 671 Xanthone, 663 Concentrators, Multiple effect, 463, 464 Condensation, Aldehyde, 205 Aldol, 205 Condenser, Liebig's, 2 Conditioning of textiles, 704 Conductivity, Electrical, 27 Conidia, 111 Coniferin, 574, 741 Coniine, 19, 90, 625, 629 Conylene, 90 Coolers, Wort, 170 Copellidine, 626 Copper aceto-arsenite, 287 Copra, 402 Cops, 718 Coralline, 660 Corchorus capsularis, 689 Cordite, 244 Corium, 586 Cornein, 740 Cotarnine, 633 Cotton, 684 Bleaching, 711 Mercerisation, 506, 685 Production, 685 seed, 391 Coumalin, 626 Coumarin, 576, 583, 595 Coumarins, 663 Count of yarn, 683 Crabbing, 717 Cracking of oils, 74 Cream of tartar, 336, 337 Creatine, 366 Creatinine, 366 Creoline, 543 Creosol, 544 Creosote, 83 oils, 531 Cresols, 543 Croceine, 656 Crotonaldehyde, 209 Crotonylene, 90 Crushers, 220, 261 Cryptopine, 632 Crystalline, 557 Crystallisation, 2 Crystallose, 579 Crystals, Hemihedral, 18 II Crystals, Liquid, 116 Mixed, 23 Cudbear, 668 Cumene, 535 Cumidine, 535 Cuminaldehyde, 572 Cuminol, 572 Cunerol, 385 Curagao, 160 Curarine, 627, 633 Cutch, 585, 669 Cyamelide, 359 Cyanamide, 18, 362 Cyanamines, 661 Cyanates, 359 Cyanides, Alkyl, 198 Cyano-acids, 317 Cyanogen, 358 chloride, 359 compounds, 358 of coal-gas, 47 sulphide, 360 trichloride, 359 Cyanohydrins, 199 Cyanole, 676 Cyanoquinolines, 636 Cyanurtriamide, 362 Cyclic compounds, 87, 520 Cycloheptanone, 297 Cyclohexane, 63 Cyclo-olefines, 28, 520 Cycloparaffins, 28, 520 Cyclopentadiene, 521 Cyclopentane, 63 Cyclopentanone, 521 Cyclopropane, 87 Cymene, 209, 535 Cymogen, 35 Cynarase, 115 Cysteine, 332, 356 Cystine, 332, 356 Cytase, 112 DAMBONITOL, 546 Daphnetin, 584, 741 Daphnin, 584 Datura stramonium, 631 Decahydroquinoline, 637 Decane, 31 Degragene, 389 Degras, 389, 588 Artificial, 389 Degree of dissociation, 267 fermentation, 129 (apparent), 129 (real), 129 mercerisation, 686 rancidity, 375 viscosity, 79 Degrees Brix, 129, 481 Dehusker, 121 Delphinin, 627 Denaturants, 152 Denatured alcohol, 152 Densimeter, Brix, 447, 481 Legal, 174 Dephlegmators, 67, 133, 139 Derricks, 58, 65 Desiohthyol, 84 Desmo bacteria, 110 Desmotropy, 17 Desoxybenzoin, 609 Detonation, 215 Detonators, 265 754 Developers, Photographic, 564 Dextrase, 123 Dextrin, 501 in glucose, 433 Dextrinase, 112, 171 Dextrose, 433 Diacetamide, 352 Diacetanilide, 560 Diacethydrazide, 358 Diacetyl, 332, 333 Diacetylene, 94 Diacetylglycol, 182 Dialdehydes, 329 Diallyl, 90 Diamalt, 500, 711 Diamines, Aromatic, 556 Diamino-acids, 356 Diaminoazobenzenes, 565 Diaminobenzenes, 556 p : p-Diaminodiphenyl, 605 p-Diaminodiphenylmethane, 606 Diaminogen, 678 Diaminophenol, 564 p-Diaminostilbene, 609 Dianisidine, 605 Dianthraquinonedihydroazine, 665 Diastase, 111, 112, 116 Diastofor, 500, 695, 711 m-Diazine, 626 Diazoaminobenzene, 569 Diazoamino-ompounds, 556, 565, 569 Diazoanisole cyanide, 567 Diazobenzene bromide, 568 chloride, 568 nitrate, 568 perbromide, 568 sulphate, 568 Diazo black, 677 blue, 678 brown, 679 compounds, 202, 556, 565, 566 Diazoguanidine, 366 Diazomethane, 202 Diazonium platinichloride, 567 salts, 566 Diazotisation, 658 Dibenzyl, 609 derivatives, 609 Dibromopyranthrene, 665 Dicetyl, 31 Dichlorhydrin, 214 Dichlorethane, 98 Dichloromethane, 98 Dichloronaphthalene, 614 Dickol, 400 Dicrucin, 372 Dicyanodiamide, 362 Dieline, 102 Diethylacetylurea, 629 Diethylamine, 201 Diethylcarbinol, 105, 203 Diethylcyanamide, 362 Diethylenediamine, 214 Diethylmalonylurea, 629 Diethylsulphone, 196 Diffusors, 451, 452 Digits lin, 628, 741 Digi^onin, 741, 742 Digitoxin, 741 Diglycerol, 184 Diglycollamides, 353 Diglycollimide, 353 Dihydrazides, 358 Dihydric phenols, 543 INDEX Dihydroanthracene, 616 Dihydrocymene, 593, 596 Dihydropyridines, 626 Dihydropyrrole, 62 Dihydroxyacetonase, 123 Dihydroxyacetone, 123, 333 Dihydroxyanthraquinone, 617 o-Dihydroxybenzophenone, 606 Dihydroxycoumarin, 584, 741 Dihydroxydiaminoarsenobenzene, 564 Dihydroxydiethylamine, 213 Dihydroxydiphen3'ls, 606 p-Dihydroxyhexamethylene, 592 Dihydroxynaphthaquinone, 659 Dihydroxytoluene, 544 Diketobutane, 333 p-Diketohexamethylene, 592 Diketohexane, 334 Diketonamines, 210 Diketones, 333 Diketopiperazine, 734 Dimethylacetamide, 351 Dimethylacetol, 333 Dimethylamine, 201 Dimethylaniline, 559 Dimethylanthracene, 616 Dimethylarsenic acid, 202 chloride, 202 Dimethylarsine, 202 Dimethylbenzenes, 534 Dimethylbutadiene, 599 Dimethylcarbinol, 180 Dimethylethylcarbinol, 105, 181 Dimethylfulvene, 521 Dimethylmethane, 35 Dimethyloxamide, 200 Dimethylphenylpyrazolone, 623 Dimethylpyridines, 625 Dimethylthiophene, 620 Dimorphism, 24 Dinaphthol, 613 Dinaphthyl, 614 Dinitroacetylglycerine, 223 Dinitroanthracenes, 616 Dinitrobenzenes, 550 Dinitrodurene, 548 Dinitroformylglycerine, 223 Dinitroglycerine, 222 Dinitroisodurene, 548 Dinitromesitylene, 548 Dinitromonochlorhydrin, 223 Dinitroprehnitene, 548 7-Dinitrotolylphenylamine, 553 /3-Dinitrotoluidine, 553 Dinitroxylenes, 548 Diolefines, 90 Dionine, 627 Dioxyindole, 638 Dioxynaphthaquinones, 614 Dipentene dihydrochloricles, 595 Dipeptides, 734 Diphenyl, 568, 605 derivatives, 605 Diphenylacetamide, 560 Diphenylacetylene, 609 Diphenylamine, 559 Diphenylcarbinol, 606 Diphenyleneketone, 573 Diphenylenemethane, 607 as. Diphenylethane, 606, 609 Diphenylethylene, 609 Diphenylhydrazine, 570 Diphenyl ketone, 572 Diphenylmethane, 604 INDEX 755 Diphenylnitrosamine, 570 Diplococci, 110 Dipropargyl, 94 Dipsacus fullonum, 725 Dipyridine, 625 7-Dipyridyl, 625 Diquinoline, 637 Diquinolyl, 637 Disinfectants, 541 Distillation, Fractional, 2, 66 of fermented liquids, 132 Theory, 3 Vacuum, 4 Wood, 272 Distillery residues, Utilisation, 153 Disulphides, 195 Disulphoxides, 195 Dithioglycol chloride, 213 Dithiourethane, 365 Diureides, 367 Docosane, 31 Dodecane, 31 Dormiol, 99, 629 Dotriacontane, 31 Dryers for oils, 400 Drying ovens for explosives, 239, 253 textiles, 723 Drying power of oils, 399 Dulcitol, 190 Durene, 535 Durra, 117, 151, 154 Dyeing, Theory, 708 Dyestuffs, see Colouring-matters Dyewoods, 666 Dynamites, 229 Analysis, 259 Gelatine, 241 Gelatinised, 242 Gum, 241 Manufacture, 230 Non-congealing, 551 Properties, 231 Safety, 231 with active bases, 231 with inert bases, 230 EBONITE, 598, 702 Ebullioscope, 147 Ecgonine, 633 Echinochrom, 114 Ecrasite, 563 Effusiometer, Bunsen's, 55 Eggs, 736 Preservation, 736 Ehrlich's side-chain theory, 114 Eicosane, 31 Eikonogen, 613 Elastin, 739 Emeraldine, 662 Emulsin, 112, 113 Emulsor, Kuhlmann, 226 Emulsor-centrifuge, 394 Enantiomorphism, 19, 24 Enantiotropy, 108 Encaustic, 377 Engenhos, 444 Enzymes, 22, 112, 740 Equilibrated action, 113 Glycolytic, 429 Synthetic action, 113 Eosin, 581 Eosins, 661 Epichlorhydrin, 214 Equilibrium in saponification, 370 Erica B, 679 Eriocyanine, 677 Erythrene, 90 Erythrin, 189 Erythritol, 90, 189 Erythrodextrin, 116 Erythrolein, 668 Erythrolitmin, 668 Erythroxylon coca, 633 Ester ificat ion, Laws, 370 Esters, 103 Ethanal, 208 Ethanamide, 352 Ethanamidine, 357 Ethandial, 329 Ethandiol, 183 Ethane, 17, 23, 34 Polychloro-derivatives, 101 Ethanol, 108 Ethene, 29, 89 Ethenol, 182 Ethenylethylenediamine, 623 Ether, 192 Petroleum, 35, 66 Ethers, 190 Ethine, 29, 92 Ethyl, 29 acetate, 331, 371 acetoacetate, 331, 332, 372 benzoate, 578 bromide, 95 bromopropionate, 309 butyrate, 372 carbonate, 363 chloride, 95, 97 chloroacetoacetate, 333 chlorocarbonate, 363 chloroformate, 363 cyanurate, 359 diacetylsuccinate, 333 diazoacetate, 323, 356 dichloroacetoacetate, 333 dihydrocollidinedicarboxylate, 624 diketoapocamphorate, 602 diketocamphorate, 602 /3 /3-dimethylglutarate, 602 dimethylmalonate, 372 fluoride, 95, 97 formate, 331, 371 hydroxycrotonate, 330 iodide, 95, 97 isocyanate, 359 isocyanurate, 359 malonate, 308 methyl ketone, 212 mustard oil, 361 nitrate, 197 nitrite, 197 oxalate, 200 peroxide, 194 hydrate, 194 phosphate, 197 sodioacetoacetate, 331, 332 sodiomalonate, 309 sodiomethylmalonate, 309 sulphate, 197 sulphite, 197 thioacetate, 351 thiocyanate, 361 Ethylacetamide, 351 Ethylacetamido-chloride, 356 Ethylacetimino-chloride, 356 Ethylacetylene, 90 Ethylamine, 199, 201 756 Ethylamine ethyldithiocarbamate, 365 hydrochloride, 351 Ethylbenzene, 527, 534 Ethylcarbinol, 180 Ethylcyanamide, 362 Ethylene, 87, 89 bromide, 98, 182 chloride, 98 cyanide, 213 iodide, 98 monothiohydrate, 213 oxide, 213 Polychloro -derivatives, 102 Ethylenecyanohydrin, 213 Ethylenediamine, 213 Ethylenic compounds, 98 Ethylglycocoll, 734 Ethylhydrazine, 206 Ethylideneacetone, 333 Ethylidene chloride, 98 Ethylidenecyanohydrin, 199, 213, 221 Ethylidenic compounds, 98 Ethylmagnesium bromide, 203 Ethylmercaptan, 196 Ethylsulphone, 196 Ethyltoluene, 525 Ethylurethane, 363 Etiline, 102 Eucaine, 99, 627 Eugenol, 544 Euquinine, 627 Eurodines, 661 Euxanthine, 668 Euxanthone, 668 Exalgin, 560 Excelsior mill, 168, 250 Exhausters, 48 Exploders, 258 Explosion, 215 by influence, 221 Determination of, 220 Heat of, 216 Pressure of gases, 218 Velocity of combustion, 219 projectiles, 262 reaction, 219 wave, 219 Volume of gases, 217 Explosive, Favier's, 220 Explasives, Abel's test for, 260 Analysis of, 259 Charging density of, 218 Classification of, 222 Destruction of waste, 258 Non-congealing, 223 Progressive, 219 Safety, 33, 246 Sensitiveness of, 261 Shattering, 219 Sprengel's, 245 Stabilisation of, 228 " Statistics of, 263 Storage of, 258 Theory of, 215 Uses of, 263 Extractor, Merz universal, 393 Pallenberg, 393 Soxhlet, 374 Wegelin and Hubner, 393 FACTIS, 599 Fat, Bone, 378, 388 Crude, 390 Goose, 378 INDEX Fat, Hog's, 388 Horse, 378 Ox, 378 Wool, 389 Fats, 372 Acid number of, 375 Animal, 379 Chemical and physical constants of, 378 Dropping-point of, 6, 375 Estimation of, 374 Industrial treatment of, 405 et seq. Rancidity of, 375 Saponification of, 377 Fehling's solution, 212, 486 Felt, 681 Fenchene, 598 Fenchone, 602 Fermentation, Alcoholic, 110, 121, 171 Lactic, 324 Ferrugine, 694 Fibres, see Textile fibres Fibrinogen, 114, 737 Fibroin, 692, 740 Filite, 244 Films for cinematographs, 504 Filter-presses, 459 Filters, Charcoal, 470 Mechanical, 461 Firedamp, 33, 246 Fishery statistics, 60 Flavanthrene, 665 Flavin, 638, 668 Flavol, 616 Flavone, 637, 663 Flavopurpurin, 616, 663 Flax, 687 Autumn, 687 March, 687 Fleece, 681 Floricin, 399 Floss (silk), 691 Flour, Wheat, 497 Fluoranthrene, 619 Fluorene, 607 Fluorescein, 581, 660 Fodder, Molassic, 140 Nutritive value of, 153 Forcite, 242 Forests, 519 Formaldehyde, 206 Formalin (Formol), Analysis of, 207 Formamide, 352 Formates, 270 Formhydrazide, 358 Formins, 214 Formose, 428 Formula, Constitutional, 14 Empirical, 12 Fleischmann's, 386 Structural, 14 Formulae, Rational, 17 Unitary, 14 Formyl chloride, 317 Formyloxime chloride, 358 Fractionator, 136 Fructose, 435 Specific Rotation of, 485 Fruit essences, Artificial, 289, 371 Fuchsine, 608, 660 Fucose, 431 Fulgurite, 231, 245 Fuller's-earth, 77, 395 Fulminate of mercury, 255 Analysis of, 256 INDEX 757 Fulvene, 521 Fumaria officinalis, 313 Furan, 619, 620 Furazan, 623 Furazole, 623 Furfuraldehyde (Furfural), 145, 430, 620 Furnace, Combustion, 7 Gas, 41 Fusel oil, 90, 122, 144 Fuses, 255 Bickford, 257 Electric, 257 GALACTOSE, 430, 437 Galalith, 386 Galazin, 160 Galbanum, 544 Gallocyanine, 661, 673 Gambier, 669 Gas, Air, 53 Blue, 52 Illuminating, 36 et seq. Marsh, 32 Oil, 57, 82 Producer, 53 Riche, 53 Suction, 53 Water, 52, 82 Gases, Permanent, 33 Gasogen, 42 Gasolene, 35, 66 Gasometers, 48 Suspended, 49 Telescopic, 48 Gaultheria procumbens, 106 Gelatine, 739 Blasting, 241 dynamites, 241, 242 Geranial, 209 Geranine, 679 Geraniol, 182, 209, 601 Gin, 160 Globin, 737 Globulins, 736 Glucoproteins, 739 Glucosamine, 427, 432 Glucose, 433, 437 Detection of, 434 Estimation of, 433, 486 Granulated, 434 Hydrated, 434 Manufacture of, 434 Glucosides, 432, 437, 740 Glucosone, 436 Glucosoxime, 432 Glue, 739 Analysis of, 740 Manufacture of, 739 Glutarimide, 354 Gluten, 497 Glyceraldehyde, 329 Glycerides, 183, 370, 372 Synthesis of, 373 Glycerol (Glycerine), 35, 122, 183 Qualities of, 188 Refractive index, 184 Statistics, 188 Tests for, 188 Glycerose, 329 Glycine (Glycocoll), 317, 322, 355 Glycocyamidine, 366 Glycocyamine, 366 Glycogen, 114, 503 Glycol, 183 Glycol acetates, 213 chlorohydrin, 213 dinitrate, 213 Ethyl ethers of, 213 mercaptan, 213 Glycollamide, 353 Glycollic aldehyde, 329, 428 Glycollide, 322 Glycols, 88, 182 Higher, 183 Propylene, 183 Glycosine, 329 Glycylglycine, 734 Glyoxal, 329 Glyoxaline, 329, 623 Glyoxiline, 231 Gnoscopine, 633 Gommeline, 501 " Grains," 169 Grape -must, 155 Greek fire, 248 Green naphtha, 84 oil, 84 starch, 495 Green, Algol, 665 Alizarin, 663 Brilliant, 675 Diamine, 678 Diamond, 675 Fast, for cotton, 659 Indanthrene, 665 Italian, 678 Janos, 675 Malachite, 607, 660 Methylene, 661 Naphthol, 659 Pyrogen, 678 Schweinfurth's, 287 Sulphur, 678 Grisounite, 247 Guaiacol, 544 Guanamines, 365 Guanidine, 365 Amino-derivative of, 366 nitrate, 366 Nitro -derivative of, 366 Guanine, 368, 369 Gum, 502 arabic, 502 Artificial, 501 British, 501 dynamites, 241 Starch, 501 tragacanth, 502 Guncotton, 232 Compression of, 238 Manufacture of, 234 Properties of, 233 Pulping of, 237 Stabilisation of, 237 Thomson and Nathan's process for, 236 Uses of, 238 Guttapercha, 599 H^IMATEIN, 666 Haematin, 737 hydrochloride, 738 Haematoxylin, 666 Hsemin, 738 Hsemocyanin, 114 Haemoerythrin, 114 Haemoglobin, 113. 114, 737 Hair, Artificial, 702 Half-stuff, 509 758 INDEX Halides, Acid, 316, 317 Halogens, Detection of, 7 Estimation of, 11 Hansena fermentation vessels, 175 Hardened glass, 80 Heat of combustion, 25 explosion, 216 formation, 25 of explosives, 216 neutralisation, 25 Heaters for sugar-juice, 452, 461 Hedonal, 99, 629 Helicin, 741 Heliotrope, Ciba, 664 Heliotropin, 574 Hemicellulose, 503 Hemimellithene, 527 Hemiterpene, 598 Hemp, 688 seeds, 391 Heneicosane, 31 Hentriacontane, 31, 36 Heptachloropropane, 101 Heptacosane, 31, 36 Heptadecane, 31 Heptaldehyde, 209 Heptane, 31, 36 Heptoses, 437 Heptylbenzene, 106 Heracleum giganteum, 106, 181 spondylium, 181 Hesperidene, 595 Hesperidin, 741 Heterocyclic compounds, 28, 520, 619 Hexabenzylethane, 610 Hexabioses, 438 Hexabromobenzene, 537 Hexacetylmannitol, 189 Hexachlorobenzene, 537 Hexachlorohexahydrobenzene, 538 Hexacontane, 31, 36 Hexacosane, 31 Hexadecane, 31 Hexadione, 334 Hexaethylbenzene, 527 Hexahydrobenzene, 592 Hexahydrocymene, 593, 596 Hexahydrophenol, 592 Hexahydropyridine, 626 Hexahydroxybenzene, 546 Hexahydroxybenzophenone, 584 Hexahydroxycyclohexane, 546 Hexahydroxydiphenyl, 606 Hexamethyl benzene, 91, 527, 535 Hexamethylene, 87, 592 Hexamethylenetetramine, 157, 205, 207 Hexandiine, 94 Hexanes, 31, 35 Hexanhexol, 189 Hexanol, 181 Hexaphenylethane, 608 Hexapropylbenzene, 527 Hexine, 90 Hexitols, 427, 433 Hexosaccharinc, 432 Hexoses, 431 Constitution of, 432 Hides, Dyeing of, 591 Finishing of, 590 Graining of, 590 Tanning of, 586 Histones, 737 Hollanders, 508, 513 Holocaine, 99, 629 Homoasparagincs, 356 Homocamphoric nitrile, 603 Homologues of aniline, 560 benzaldehyde, 571 phenol, 543 succinic acid, 310, 311 terpenes, 598 Homology, 23 Homophthalimide, 637 Homopyrocatechol, 544 Honey, 435 Hops, 162 Decoction of, 170 ' Humic substances, 432 Humulus lupulus, 162 Hydantoin, 364 Hydracellulose, 505 Hydramine, 213 Hydrastine, 632 Hydrastinine, 632 Hydraulic accumulators, 393 gas main, 43 press, 251, 392 Hydrazides, 358, 570 Hydrazines, 565, 569 Hydrazobenzene, 566 Hydrazodicarbonamide, 366 Hydrazones, 206, 427 Hydroanthracene, 615 Hydroanthranols, 616 Hydrobenzamide, 571 Hydrobenzoin, 609 Hydrocarbons, 28, 30 Aromatic, 526 of petroleum, 62 of the CH 2 2 series, 90 of the CH 2 4 and CH 2 - 6 series, 94 Saturated, 28, 30, 31 Unsaturated, 28, 87, 96 with triple linkings, 90 with unsaturated side-chains, 535 Hydrocellulose, 503, 504 Hydrocinchonidine, 634 Hydrocotarnine, 633 Hydrocyanocarbodiphenylimide, 644 Hydrogen, Estimation of, 7 Nascent, 32 Typical alcoholic, 106 Hydrolysis, 104, 438 Enzymic, 112 Hydronaphthalene, 612 Hydropyridines, 626 Hydroquinine, 634 Hydroquinone, 544 Hydroxy-acids, Aromatic, 582 Higher, 326 polybasic, 351 Polyvalent dibasic, 334 monobasic, 328 tribasic, 345 Saturated monobasic, 326 Unsaturated monobasic, 320 Hydroxy-alcohols, 573 Hydroxy-aldehydes, Aromatic, 573 Hydro xyanthranol, 617 Hydroxyanthraquinones, 616 Hydroxyazo-compounds, 565 Hydroxybenzaldehydes, 574 jS-Hydroxybutyraldehyde, 205 Hydroxyethylamine, 213 Hydroxyethyltrimethylammonium hydroxide, 214 Hydroxyhydroquinone, 545 Hydroxylamine, 167 INDEX 759 Hydroxylamine derivatives of acids, 358 Hydroxylinolein, 400 Hydroxymethyleneacetone, 331, 334 Hydroxymethyleneketones, 334 Hydroxymethylfurfural, 430 Hydroxynitriles, 199 Hydroxypyridines, 625 2-Hydroxyquinoline, 637 Hydroxytoluenes, 543 Hyoscyamine, 632 Hyphomycetes, 111, 130 Hypnone, 572 Hypnotics, 98, 629 Hypoxanthine, 368 ICHTHYOFORM, 84 Ichthyol, 84 Ichthyolsulphonates, 84 Iditol, 190 Illuminating gas, 36 Analysis of, 53 Calorific value of, 38, 54, 55 Composition, of , 38 History of, 36 Lighting power of, 55 Meters, 50 Physical and chemical testing of, 53 Price of, 51 Properties of, '. 8 Purification of, 43, 46 Separation of naphthalene from, 44 Statistics of, 52 Yield of, 51 Imidazole, 623 Imides, 353 Iminocarbamide, 365 Iminocarbamideazide, 366 Iminochlorides, 356 Iminoethers, 352, 353 Iminothioethers, 357 Iminourea, 365 Impermeable fabrics, 286 Indamine, 647, 661 Indanthrene, 471, 665 Indazin, 676 Indazole, 639 Indene, 614 Index of refraction, 26, 375 Indican, 640 of urine, 638 Indigo, 639, 663 Analysis of, 640 blue, 640 carmine, 641 Colloidal, 641 extract, 676 Properties of, 641 Statistics of, 645 artificial, 645 Syntheses of, 642 Indigofera erecta, 640 leptostachya, 640 tinctoria, 639 Indigoids, 663 Indigolignoids, 665 Indigotin, 640 Indirubin, 664 Indoin, 656 Indole, 638 Indolignone, 665 Indonaphthalene, 665 Indophenin, 620, 621 Indoxyl, 638, 640, 644 Indrene, 615 Indulins, 661 Infusorial earth, 223, 229 Injectors, Korting, 48 Ink, 583 Inositol, 546, 592 Inulin, 435, 436 Inversion, 433, 441 Invert sugar, 433, 485 Invertase (Invertin), 111, 112, 441 lodobenzene, 536, 537 lodobutane, 97 lodoform, 100 Tests for, 101 lodol, 621 lodopropane, 97 lodosobenzene, 536 chloride, 536 lodourethane, 363 lodylbenzene, 536 lonene, 599 Ionic concentration, 441 lonone, 595, 599 Irene, 599 Iridin, 741 Irigenin, 741 Irone, 599 Isatin, 638, 642 chloride, 642 a-Isatinanilide, 644 Isatis tinctoria, 639 Isatoxime, 642 Isoamyl valerate, 372 Isoamylbenzene, 527 Isobutane, 35 Isobutyl iodide, 97 Isobutylbenzene, 527 Isobutylcarbinol, 105, 122 Isocyanates, 359 Isocyanides, 199 Isocyclic compounds, 28 Isocymene, 535 Isodulcitol, 431, 668 Isodurenes, 527, 535 Isoduridine, 555 Isoeugenol, 544, 57 Isolactose, 438 Isology, 23 Isomaltose, 113, 438 Isomelamine, 362 Isomerides, 16 Boiling-points of, 24 Melting-points of, 24 Metameric, 17 Optical, 18 Racemic, 19 Isomerism, 14, 16 Cis- and trans-, 21 Space, 18 Isonitriles, 199, 358 Isonitrosoketones, 211, 333 Iso-oxazole, 623 Isopentane, 35 Isoprene, 90, 598 Isopropyl iodide, 96, 97 Isopropylacetylene, 90 Isopropylbenzaldehyde, 572 Isopropylbenzene, 527, 535 Isoquinoline, 637 Isorhamnose, 431 Isovaleryl chloride, 319 Isoviolanthrene, 665 Isuret, 358 Ivory, Artificial, 290 760 INDEX JIGGERS, 721 Jute, 689 KEFIR, 115, 160 Keratin, 683, 739 Kerosene, 63 Keto -aldehydes, 330, 334 Keto-arabinose, 428 Ketoheptamethylene, 521 Ketohexamethylene, 521, 592 Ketohexoses, 427, 431 Ketoketenes, 213 Ketones, 96, 103, 203, 204, 210 Aromatic, 572 Strache's estimation of, 212 Ketonimides, 203 Ketopentamethylene, 521 Ketoses, 427, 435 Ketoximes, 210, 572 Beckmann's transposition of, 210, 573 Khaki, 669 Kieselguhr, 223, 229 Kneading machine, Werner-Pfleiden r, 384 Koji, 130 Koumis, 160 Kunerol, 385 Kwan, 58 Kyanol, 557 LACCASE, 112 Lacs, 400 Lactalbumin, 735 Lactams, 355 Lactases, 112 Lactic acid bacillus, 323 Lactides, 321 Lactite, 386 Lactoglobulin, 736 Lactone, Bromobutyric, 295 Isocaproic, 297 Lactones, 295, 322, 428 Lactose, 438, 486 Testing of, 440 Lactyl chloride, 325 Lager beer, 170, 172 Lakes, 650 Lamp, Carcel, 56 Hefner, Alteneck, 56 Lampblack, 528 Lanite, 244 Lard, 388 Laudamine, 632 Laudanidine, 632 Laudanosine, 632 Laurene, 596 Lautopine, 633 Law of Dalton, 5 esterification, 370 Hess-Berthelot, 25 refraction, 26 Leather, 586 Lecanora tartarea, 668 Lecithin, 374 Lecithins, 215 Lees, Wine, 143, 337, 341 Lemons, Cultivation of, 347 Treatment of, 347 Leucine, 19, 355, 734 Leu co -bases, 607, 647 Leucocytes, 115 Leucotannin, 584 Levulose, 435, 485 Life, Origin of, 114 Light Polarised, 26, 330 Light, Sources of, 57 Standards of, 55 Lignin, 505, 512 Estimation of, 509 Ligroin, 66 Limonene, 593, 595, 596 Linoleum, 400 Linseed, Composition of, 391 oil, 399 Linum usitatissimum, 399, 687 Lipase, 112, 409 Lipoids, 629 Lippich polariser, 484 Liqueurs, 159 Liquids, Specific gravity of, (i Lithoclastitc, 231 Lithographers' varnish, 400 Litmus, 668 Lupetidine, 626 Lupulin, 170, 173 Luteolin, 638 Lutidines, 625 Lyddite, 245, 503 Lysidine, 623 Lysi e, 328, 356 Lysins, 115 Lysoform, 208 Lysol, 543 Lyxose, 431 MACLURIN, 668 Madder, 617 Magnesia, Effervescent, 346 Maize, 119, 162 Composition of, 391 oil, 403 Malamide, 353 Malaria, 635 Malt, 112, 116 Cleaning of, 168 Dias^atic power of, 107 Evaluation of, 167 Green, 164 Grinding of, 168 Kilning of, 166 Mashing of, 168 Maltase, 111, 112, 117, 438 Malting, 164 Maltodextrinase, 112, 171 Maltol, 505, 582 Maltose, 113, 167, 438, 486 Mammoth pump, 226 Manna, 189 Mannide, 190 Mannitan, 190 Mannitol, 189, 428 Hexacetyl-derivative of, 189 Mannose, 436 Mannotetrose, 489 Mannotriose, 190 Margarine, 382, 383 cheese, 382 Statistics of, 384 Margol, 383 Marsala, 159 Mashing apparatus, 169 Massecuite, 468 Masut, 67, 74 Meconidine, 632 Melam, 362 Melamine, 362 Melene, 90 Melibiase, 112 Melibiose, 438, 442 INDEX 761 Melinite, 245, 563 Melissyl palmitate, 372 Mellite, 581 Mellithene, 535 Menthane, 596 Menthene, 596 Menthol, 600 Menthone, 600 Mercaptan, 196 Mercaptans, 195 Mercaptide, Mercuric, 196 Sodium, 196 Mercaptides, 195 Mercaptol, 210 Mercerisation, 506, 686, 728 Mercury fulminate, 255 Mesidine, 555 Mesitol, 540 Mesityl oxide, 212 Mesitylene, 91, 535 Metacymene, 527 Metadiamines, 557 Metaformaldehyde, 207 Metaldehyde, 208 Metalepsy, 15 Metamerism, 17, 192 Metastyrene, 535 Meters, Alcohol, 146 Automatic gas, 51 Dry gas, 51 Gas, 50 Methanal, 206 Methanamide, 352 Methanamidoxime, 358 Methane, 23, 32 Derivatives of, 30 Industrial uses of, 34 Preparation of, 34 Properties of, 33 Methanol, 106 Methanthiol, 196 Methene, 89 Methenylamidoxime, 358 Methoxymethane, 192 Methoxypyridine, 625 Methyl, 29 acetate, 371 chloride, 96 cyanide, 198, 199 iodide, 97 isothiocyanate, 361 mustard oil, 361 nonyl ketone, 332 sulphide, 196 Methylacetanilide, 560 Methylacetylurea, 352 Methylal, 209 Methylamine, 201 hydrochloride, 97, 201 sulphate, 201 Methylaniline, 559 Methylanthracene, 616 Methylarbutin, 741 Methylbenzene, 534 Methylbutanol, 181 Methylcyanamide, 362 Methyldihydroimidazole, 623 Methylene, 89 bromide, 95, 98 chloride, 95, 98 iodide, 95, 98 Methylethylacetylene, 90 Methylethylcarbinol, 181 Methylglyoxal, 334 Methylheptanone, 303 p-Methylisopropylbenzene, 535 Methylisopropylcarbinol, 105 Methylnaphthalenes, 614 Methylpentoses, 431 Methylpropane, 29, 35 Methylpropanol, 181 Methylpseudoisatin, 638 Methylpyridine, 625 Methylpyridone, 625 a-Methylquinoline, 637 Methylsulphonal, 629 Methyluracyl, 367 Methylurethane, 358 Metol, 564 , Microbes, 110 Micrococci, 110 Micron, 110 Milk, 112, 385 Analysis of, 386 C jco-nut, 402 Fermented, 160 Skim, 385 Milling, 714 Moellon, 389 Molasses, 469, 473 Beet-sugar, 140 Lactose, 440 Recovery of sugar from, 474-476 Utilisation of, 473 Monoacetin, 214 Monoacylhydrazides, 358 Monoazo -compounds, 656 Monochlorhydrin, 214 Mononitroglycerine, 222 Mononitrotoluenes, 550 Monosaccharides, 426 Monoses, 426 Formation of, 427 Mordanting, 651, 706, 710 Mordants, 650 Morin, 638 Morphine, 115, 628, 632 Morpholine, 626 Morphotropy, 24 Morus tinctoria, 668 Motochemistry, 523 Moulds, 111 Mucins, 739 Mucors, 111, 130, 131, 346 Murexide, 367, 368 Muscarine, 214 Mustard, Black, 361 seed, 391 Muta-rotation, 27, 430, 485 Mycoderma aceti, 280, 281 vini, 281 Myosin, 737 Myristin, 289 NAPHTHA, 58 Naphthalene, 531, 610, 643 derivatives, 610 from coal-gas, 44 tetrachloride, 614 a-Naphthaquinone, 613 /3-Naphthaquinone, 614 Naphthazarin, 659 Naphthenes, 63, 592 ' Naphthindigo, 664 Naphthols, 613 a- (and -) Naphthylamine, 613 Narceine, 628, 632 Narcotine, 628, 632, 633 762 INDEX Natron, 415 Neroline, 613 Neurine, 214 Nic"ol prism, 483 Nicotine, 628, 629 Nicotyrine, 630 Nitracetanilides, 560, 562 Nitriles, 198, 575 Constitution of, 199 Nitroacetins, 224 Nitroanilines, 561 Nitroanthracenes, 616 Nitrobenzaldehyde, 572, 643 Nitrobenzene, 549, 566 Nitrocellulose, 232 constitution of, 232 Nitrochlorhydrin, 223 Nitrocymene, 548 Nitro-derivatives, Aromatic, 197, 547, 548 Electrolytic reduction of, 566 Nitrodimethylaniline, 559 Nitroethane, 198 Nitroform, 198 Nitroformins, 224 Nitrogen, Detection of, 6 Estimation by Dumas' method, 10 Kjeldahl's method, 10 Will and Varrentrapp's method, 11 Stereoisomerism of, 22 Nitroglycerine, 223 Filtration of, 228 Manufacture of, 225 Stabilisation of, 228 Uses of, 229 Nitroguanidine, 366 Nitrohexane, 198 Nitromesitylene, 548 Nitron, 563 Nitronaphthalenes, 612 .Nitrophenols, 562 Nitrophenoxides, 562 p-Nitrophenylhydrazine, 570 Nitroprehnitene, 548 sec. Nitropropane, 198 Nitro.amines, 201, 556, 567 Nitrosites, 593 Nitrosochlorides, 593 p-Nitrosodimethylaniline, 559 Nitrosodipentene, 594 Nitrosophenol, 559 Nitrosopyrroles, 622 Nitrotoluenes, 550 Nitrourea, 364 Nitrourethane, 363 Nitroxylenes, 548 Nomenclature, Official, 28 Nonane, 31 Nonodecane, 31 Nonoses, 437 Non-sugar, 447, 473 Nonyl aldehyde, 209 Nuclei, Condensed benzene, 605 Nucleins, 738 Nucleo-albumins, 737 Nucleo-histone, 737 Nucleoproteins, 738 Number, Acetyl, 188, 189 Acetyl acid, 189 Acetyl saponification, 189 Arid, 86, 375 Butter, 387 Ester, 377 Hehner, 373 Iodine, 375 Number, Kottstorf, 379 Maumene, 376 Polenske, 387 Reichert-Meissl-Wollny, 373, 387 Saponification, 379 Nutrose, 386, 737 Nux vornica, 633 OCTADECAPEPTIDE, 735 Octadecylbenzene, 527 Octadiene, 90 Octane, 31 Octocosane, 31 Octodecane, 31 Octoses, 437 Octylbenzene, 527 (Enanthaldehydo, 209 (Enoxydase, 112 Oil, Acetone, 212 Allyl mustard, 361 Almond, 378 Aniline, 558, 559 Anise, 543 Anthracene, 530, 532 Arachis, 378, 397, 404 Bitter almond, 571 Boiled linseed, 399 Bone, 620 Camphor, 604 Castor, 326, 378, 398 Clove, 544 Coco-nut, 378, 402 Colz.i, 378 Cotton-seed, 378, 403 Chinese bean, 404 Cod-liver, 378, 389, 400 Dippel animal, 620 Ethyl mustard, 361 Fish, 389 for gas, 82 Gelatinised vaseline, 80 Gingdly, 404 Grape seed, 405 Hempseed, 378 Linseed, 378, 399 Maize, 378, 403 Methyl mustard, 361 Oleo, 384 Olive, 373, 378, 395 Palm, 378, 401 Palm-nut (Palm-kernel), 378, 401, 402 Paraffin, 66, 81, 82 Poppy-seed, 378, 400 Propyl mustard, 361 Resin, 78, 596 Sanse, 397 Sesame, 378, 404 Soja-bean, 378, 404 Solar, 63, 82 Sperm, 389 Stillingia, 403 Sulphocarbon, 396 Teel, 404 Tomato-seed, 405 Turkey-red, 327, 395 Turpentine, 593, 596, 597 Walnut, 400 W r ashed olive, 396 Whale, 378, 389 Wool, 390 Oil-cake, 391, 405 Oil-gas, 57 Oils, Animr.l, 378, 379 INDEX 763 Oils, Bleaching of, 395 Blown, 375 Creosote, 531 Deodorisation of, 395 Drying, 374 Engine, 78 for gas, 82 Flash-point of, 79 Heavy, 66 Mineral lubricating, 63, 74 Mustard, 361 Oxidised, 375 Refining of, 394 Spindle, 78 Thickened, 400 Vegetable, 378, 390 Viscosity of, 72, 79 Olease, 395 Olefines, 28, 87 Constitution of, 89 Nomenclature of, 87 Preparation of, 88 Table of, 87 Oleine, 298, 407 Catalytic, 419 Distilled, 408, 419 Saponification, 419 Transformation into stearme, 41 Wool, 390 Oleomargarine, 382 Olive, Composition of, 391 oil, 395 Opium, 632 . Estimation of morphine in,