'Illllli 1 nun 1 INTERNATIONAL CHEMICAL SERIES H. P. TALBOT, PH.D., CONSULTING EDITOR TECHNICAL METHODS OF ANALYSIS PUBLISHERS OF BOOKS F O B^ Coal Age * Electric Railway Journal Electrical World * Engineering. News -Record American Machinist v The Contractor Engineering 8 Mining Journal ^ Power Metallurgical 6 Chemical Engineering Electrical Merchandising TECHNICAL METHODS OF ANALYSIS AS EMPLOYED IN THE LABORATORIES OF ARTHUR D. LITTLE, INC., CAMBRIDGE, MASS. EDITED BY ROGER CASTLE GRIFFIN Director of Analytical Department FIRST EDITION SECOND IMPRESSION McGRAW-HILL BOOK COMPANY, INC. NEW YORK: 370 SEVENTH AVENUE LONDON: 6 & 8 BOUVERIE ST., E. C. 4 1921 0-7 COPYRIGHT, 1921, BY THE McGRAW-HILL BOOK COMPANY. INC. THIS BOOK IS AFFECTIONATELY DEDICATED TO THE MEMORY OF MY FATHER tll (Stiffin 4G0150 PREFACE THIS book contains a representative selection of analytical methods which have been adopted as standard procedures in a large commercial laboratory engaged in technical analysis. With very few exceptions the methods here given have been used many times in this laboratory and have been proved to give satis- factory results in the hands of different analysts. In certain cases it has been thought best to include, for the sake of complete- ness, certain standard procedures which have not been thoroughly tested in our laboratory but which have, nevertheless, received official recognition from an authoritative body of chemists. Such a case is the Roese-Gottlieb method on page 418. Endeavor has been made to give such directions in each pro- cedure that any one reasonably familiar with analytical technique can readily follow them. A certain amount of cross referencing has been necessary on account of economy of space but, in order to avoid the annoyance caused by carrying this to extremes, the directions, when they are not too long, have been repeated in most cases. No attempt has been made to give experimental data to show the accuracy of the method nor to enter into the theory of the procedure. On the other hand in many cases it has seemed advisable to give brief descriptions of the properties which a given material should normally possess and sufficient other infor- mation to enable the analyst to translate his results into practical language. Certain specifications and requirements of an authori- tative nature have also been included. It is obviously impossible in a book of this size to enter all the fields of analytical chemistry. Certain classes of work, such as drugs, alkaloids, and medicines, have been omitted, as they are of interest mainly to specialists in these lines. This is also true of the rare elements and, to a certain extent, of gas analysis. The analysis of mineral rocks, glasses, and vitreous materials would vii viii PREFACE require a book in itself. The present book is not intended for the specialist, confining himself to comparatively narrow limits. It aims rather to include methods which are typical of the procedures employed in the usual commercial analyses. Certain special methods have been included, when not too long, in the belief that they are not otherwise readily obtainable and that they illustrate procedures which may be applicable to other problems. The analyst whose particular problem is not covered by the methods here given is referred to the bibliography in the Appendix. Some of the methods herein described are original. By far the greater number, however, have been obtained from other sources. In many cases they have been adopted without change; others have been modified in the light of experience gained from their use. Since the collection of methods has been built up gradually over a period of years it is not possible in every case to give proper credit for the original source. The methods of the Association of Official Agricultural Chemists have been found particularly valuable, although it has sometimes seemed more convenient to change their arrangement. The various publica- tions of the United States Bureaus at Washington have also been drawn upon freely. The tables in the Appendix have been confined strictly to those concerned with quantitative chemical analysis. The values in these tables have all been independently calculated from the 1920 International Atomic Weights. Every effort has been made to have them correct. If any errors should be discovered the editor will be grateful to those who bring them to his attention. Refer- ences to other tables are given in the text; and tables of properties, specific gravity tables, etc., may be found in the handbooks referred to in the bibliography in the Appendix. This bibliography aims to include at least one authoritative book in each of the fields of analytical chemistry. Several cuts have been included as aids to the description of procedures. Acknowledgment is made for certain of these as follows: Fig. 9, American Society for Testing Materials, Standard, Serial D-21-16; Fig. 10, U. S. Bur. Mines Tech. Paper 166, Petroleum Technology 39; Figs. 16, 17, 18, 19 and 20, Paper 25, No. 15, 19-23 (1919); Figs. 21, 22 and 23, Bureau of Chem., Circular 107; Figs. 27 and 28, Am. Soc. Test. Mat., Standard PREFACE ix C-9-17; Fig. 29, courtesy of the Fairbanks Co. The table of constants of oils, fats and waxes on pages 232-239 was prepared from data obtained largely from Lewkowitsch: " Chemical Technology and Analysis of Oils, Fats and .Waxes." The editor wishes to express his appreciation to Dr. H. P. Talbot for valuable criticisms and suggestions and to Dr. R. S. Williams for reading part of the manuscript. Thanks are also especially due to Dr. C. J. West, who prepared the bibliography and whose aid in arranging the manuscript and reading the proof is greatly appreciated. Mr. H. C. Parish and Mrs. Helen B. Colson also rendered valued assistance in checking the numerical tables. ROGER C. GRIFFIN. CONTENTS CHAPTER I REAGENTS Liquid Reagents 1 Standard Volumetric Solutions 6 Indicators 12 Care of Platinum 14 Recovery of Platinum Residues 16 CHAPTER II GENERAL INORGANIC ANALYSES Sulfur 17 Fuming Sulfuric Acid (Oleum) , 17 Alkalies 19 Ammonium Hydroxide 23 Table Salt 24 Sodium Nitrite 28 Sodium Sulfide 28 Sodium Silicate (Water Glass) 29 Potassium or Sodium Bichromate 30 Cyanides of Potassium and Sodium 31 Acetate of Lime 32 Antimony Sulfide 34 Determination of Small Amounts of Arsenic 35 Determination of Potassium in Fertilizers, etc 41 Lead Arsenate 46 Bordeaux Mixture 50 Bordeaux Mixture with Paris Green and Lead Arsenate 52 Paris Green 54 Lime Sulfur Solution 59 Corrosive Sublimate in Medicated Gauze 62 Asbestos Magnesia Pipe Covering 62 xi xii CONTENTS CHAPTER III GENERAL ORGANIC ANALYSES PAGE Nitrogen 64 Methyl Alcohol 71 Grain Alcohol or Cologne Spirits 74 Formaldehyde Solution 79 Formic Acid 81 Acetic Anhydride 82 Glycerol 85 Dextrin or British Gum 91 Albumin 93 Tannic Acid 95 Indigo 97 Nicotine in Tobacco and Tobacco Extract 100 Nicotine Solution 103 CHAPTER IV ANALYSIS OF METALS Sampling Iron and Steel * 106 Carbon Steel 107 Alloy Steel 116 Pig and Cast Iron 129 Tin in Tin Ores. 134 Zinc (Spelter) 137 Zinc Dust 141 Brass and Bronze 142 Nickel Silver 152 White Metals 154 Mercury in Zinc Amalgam 163 Testing of Galvanizing or Sheradizing on Iron and Steel 163 Tinning Test for Tinned Iron and Steel 166 CHAPTER V ANALYSIS OF FUELS Coal Sampling 167 Coal, Proximate Analysis and Heating Value 172 Coal, Ultimate Analysis 180 Phosphorus in Coal and Coke 184 Coal Ash and Refuse 185 Gasoline 185 Heating Value and Sulfur Content of Liquid Fuels 190 CONTENTS xiii CHAPTER VI ANALYSIS OF PAINTS AND PAINT MATERIALS PAGE Turpentine 193 Turpentine, Electric Railway Specifications 196 Lipseed Oil 197 Tung Oil 198 Mixed Paints and Pigments in Oil 200 Green Graphite Pole Paint 207 Red Lead and Orange Mineral 210 White Lead (Basic Carbonate) 211 Chrome Yellow 213 Oil Varnishes 217 Japan Driers 221 Shellac and Shellac Varnishes 223 Black Air Drying Insulating Varnish and Black Baking Insulating Varnish 227 CHAPTER VII ANALYSIS OF OILS, FATS, WAXES AND SOAPS Animal and Vegetable Oils and Fats 230 Lubricating Oils 254 Unsaponifiable Matter in Oils '. 261 Castor Oil 263 Sulfonated Oils (Turkey Red Oils) 263 Lard Oil. 265 Olive Oil 266 Tallow 268 Greases 270 Degras (Wool Grease) 274 Beeswax 275 Paraffin Wax 278 Soap > 279 CHAPTER VIII ANALYSIS OF WOOD, PAPER AND PAPER-MAKING CHEMICALS Cellulose in Wood 289 Wood Pulp Sampling and Testing 290 Sulfate Cook Liquor 294 Sulfite Acid ; . 301 Alum 303 Aniline Dyes 307 Blanc Fixe . . .308 xiv CONTENTS PAGE Bleach (Bleaching Powder) 312 Bleach Consumption of Pulp 313 Casein 315 Clay for Paper Filler 318 Crown Filler , 319 Glue 320 Lime 324 Limestone 327 Rosin 327 Rosin Size and Rosin Size Milk 329 Satin White 332 Talc for Paper Filler 334 Ultramarine 336 Fibers in Paper 337 Standard Papers for Fiber Analysis 339 Chemical Analysis of Paper 339 Physical Testing of Paper 345 Sizing in Paper 355 Tarnishing Test for Paper 362 Cotton Cellulose (Cotton Linters) for Nitration 363 Wood Distillate Products. . . 368 CHAPTER IX ANALYSIS OF TEXTILES AND TEXTILE FIBERS Structural Analysis of Textile Fabrics 372 Fibers in Cloth and Yarns 377 Chemical Tests on Ropes and Twines 382 Differentiation of Rope and Cordage Fibers 383 Asbestos Cotton Twine . 385 CHAPTER X ANALYSIS OF FOODSTUFFS Ammonia in Eggs , 387 Coloring Matter in Foods 389 Crude Fiber 393 Sulfur Dioxide in Foodstuffs 395 Reducing Sugars and Sucrose ...... 396 Saccharine Products 408 Honey 420 Maple Products 424 Butter and Butter Substitutes, , . 427 CONTENTS XV PAGE Cocoa, Chocolate and Cacao Products 430 Feed Stuffs and Mixed Grains 438 Milk and Cream 443 Condensed or Evaporated Milk 449 Cheese 451 Vinegar 454 Lemon and Orange Extracts 458 Orange Oil and Lemon Oil 464 Oil of Peppermint 464 Peppermint, Spearmint, and Wintergreen Extracts 467 Vanilla Extract.. . 468 CHAPTER XI MISCELLANEOUS ANALYSES Leather 474 Chromium in Chrome Salts and Leathers 477 Sumac Extract 479 20% Para Rubber Compound 480 Case Hardening Compounds 484 Cutting Compounds 486 Metal Polishes 488 Soldering Paste 490 Sanitary Analysis of Water and Sewage 492 Industrial Water 514 Boiler Scale 522 Fertilizers 524 Carbolineum and Similar Wood Preserving Oils 530 Gypsy Moth Cresote 536 Coal-Tar Roofing Pitch 536 Bituminous and Asphaltic Road Binders 537 Crude Coal-Tar and Water-Gas Tar 548 Spent Oxide 554 Mortar and Concrete 559 Sampling and Physical Testing of Portland Cement 561 Chemical Analysis of Portland Cement 572 Mechanical Testing of Sand and Gravel 576 TABLES: International Atomic Weights 584 Molecular and Atomic Group Weights 586 Analytical Factors 608 Volumetric Solutions 615 Bibliography 627 Index . 635 TECHNICAL METHODS OF ANALYSIS CHAPTER I REAGENTS LIQUID REAGENTS General. In making up reagents always use chemicals of highest purity, unless otherwise instructed. In the following direc- tions water is understood to be distilled water. All solutions should be filtered unless perfectly clear or unless directions state to the contrary. Formulas here given are generally for 1 liter of reagent. The ordinary green glass acid bottle contains about 2.5 liters. Acetic Acid, Glacial Contains 99.5% HC 2 H 3 2 . The sp. gr. is about 1.058. Acetic Acid, Dilute (sp. gr. 1.044). Mix 400 cc. of glacial acetic acid with 1 liter of water. This solution contains about 30% of HC 2 H 3 2 . Alcoholic Potash (Half -normal). Dissolve 29 grams of pure KOH (preferably purified by alcohol) in 1 liter of 95% alcohol. NOTE. The alcohol before use should be tested with a little NaOH and if it gives a decidedly yellow solution, showing an excessive aldehyde content, treat it as follows: Dissolve about 1.5 grams of AgNOa in 3 cc. of water and add to 1 liter of the alcohol. Shake thoroughly. Dissolve 3 grams of NaOH in 15 cc. of warm alcohol, cool and add to the main solution. Shake thor- oughly, let settle, siphon off the clear liquid and distill, adding a few pieces of pumice to prevent bumping. The alcohol and KOH mixture should be allowed to stand until all the KOH has dissolved and then filtered rapidly, or siphoned off, to remove the insoluble carbonate. 2. if.: TECHNICAL METHODS OF ANALYSIS ,^^ \ilumina -Or earn. Prepare a cold saturated solution of alum ' (KAl' 1 snilfate) i in water. Add NH 4 OH with constant stirring until the solution is alkaline to litmus; let settle and wash by decantation with water until the wash water gives only a slight test for sulfate with BaCk. Pour off the excess of water and store the residual cream in a stoppered bottle. Ammonium Acetate. To 1250 cc. of glacial acetic acid add cautiously 1000 cc. of cone. NH 4 OH (sp. gr. 0.90), a little at a time, with constant stirring, and cooling if necessary. The solu- tion contains about 70% of ammonium acetate, NIL^HaC^. Ammonium Carbonate. Dissolve 250 grams of ammonium carbonate crystals in 1 liter of water and add 100 cc. of cone. NILiOH. This solution contains about 22% of the carbonate, which is generally assigned the formula (NH^COa-NHtCC^NH^. Ammonium Chloride (10%). Dissolve 100 grams of NHjCl in water in a liter volumetric flask and dilute to the mark. Ammonium Hydroxide, Cone. (sp. gr. 0.90). The solution contains 28-29% of NHs by weight. Ammonium Hydroxide, Dilute (sp. gr. 0.96). Mix 400 cc. of cone. NE^OH with 700 cc. of water. The solution contains about 10% of NHa by weight. Ammonium Molybdate. Solution No. 1. Weigh 113 grams of 100% c. P. molybdic acid (MoOs) or an equivalent amount of weaker acid. Add 300 cc. of water, 175 cc. of cone. NIL^OH, and, after mixing, add slowly with stirring 75 cc. of cone. HNOs. Solution No. 2. To 1200 cc. of water add 500 cc. of cone. HN0 3 . When both solutions are perfectly cold, pour No. 1 into No. 2 (do not pour No. 2 into No. 1). Add 5 cc. of (NH^HPO* reagent, shake well, let stand until clear and filter. This reagent contains about 8% of ammonium molybdate, (NH Ammonium Nitrate (20%}. Dissolve 200 grams of in water and make up to 1 liter. Ammonium Oxalate (4%)- Dissolve 40 grams of 2O4-2H2O in water and make up to 1 liter. NOTE. The percentage strength of this solution refers to the crystallized salt and not to the actual content of (NH 4 )2C 2 O4. This note also applies to other reagents, here given, which are made from chemicals containing water of crystallization. REAGENTS 3 Ammonium Phosphate (10%}. Dissolve 100 grams of (NH4)2HPO4 in water and dilute to 1 liter. Ammonium Sulfate (25%}. Dissolve 250 grams of (NH4) 2 S0 4 in water and dilute to 1 liter. Ammonium Sulfide (Colorless). Pass H 2 S gas into 750 cc. of cone. NHUOH until saturated. Then add 500 cc. more of cone. NILtOH and 1000 cc. of water. Ammonium Polysulfide (Yellow). Make up a bottle of color- less (NH4)2S as above, add to it 50-75 grams of powdered sulfur, and shake. Barium Chloride (10%). Dissolve 100 grams of BaCl 2 -2H 2 O in water and dilute to 1 liter. Barium Hydroxide (5%). Dissolve 50 grams of Ba(OH) 2 -8H 2 O in water and dilute to 1 liter. Calcium Chloride (10%). Dissolve 100 grams of CaCl 2 -6H 2 O in water and dilute to 1 liter. Calcium Hydroxide (Lime Water). Make a saturated solution of Ca"(OH) 2 and keep tightly stoppered. Decant or filter before use. Fehling's Copper Solution (Soxhlet Modification). Dissolve 69.28 grams of CuSC>4-5H 2 in water, dilute to 1 liter and filter through prepared asbestos. Fehling's Alkaline Tartrate Solution (Soxhlet Modification). Dissolve 346 grams of Rochelle salts (NaK tartrate) and 100 grams of NaOH in water and dilute to 1 liter; let stand for two days and filter through prepared asbestos. NOTE. The two above solutions are also used in the Munson and Walker Method. Fehling's Copper Solution (Allihn's Modification). Same as Soxhlet modification. Fehling's Alkaline Tartrate Solution (Allihn's Modification). Dissolve 346 grams of Rochelle salts and 250 grams of KOH in water and dilute to 1 liter. Ferric Chloride (10%). Dissolve 100 grams of FeCl3-6H 2 O in water and dilute to 1 liter. Hydrochloric Acid, Cone. (sp. gr. 1.18-1.19). This reagent contains 35.5-37.5% of HC1 by weight. Hydrochloric Acid } Dilute (sp. gr. 1.12). Mix 500 cc. of cone. 4 TECHNICAL METHODS OF ANALYSIS HC1 with 400 cc. of water. The solution contains about 20% of HCL Lead Acetate, Basic. Boil for 0.5 hour 430 grams of normal lead acetate, 130 grams of litharge (PbO) and 1 liter of water. Let cool and settle. Dilute the supernatant liquid to sp. gr. 1.25 (room temperature) with freshly boiled water. NOTE. U. S. P. lead subacetate solution may also be used in place of the above solution. '- *' Lead Acetate, Normal (10%). Dissolve 100 grams of Pb(C2H 3 02)2-3H 2 O in water and dilute to 1 liter. Magnesium Ammonium Chloride (Magnesia Mixture). Dis- solve 90 grams of MgCl2-6H20 (or 45 grams anhydrous MgCb) and 240 grams of NILiCl in 1 liter of water, and add 50 cc. of cone. NHiOH. Ten cc. of this solution will precipitate about 0.25 gram of H3PO 4 or 0.18 gram of P 2 O 5 . Magnesia Wash Solution (for washing magnesium ammonium phosphate precipitate). Dissolve 100 grams of NELiNOs in water, add 335 cc. of cone. NELiOH, and dilute to 1 liter. Mercuric Chloride (5%}. Dissolve 50 grams of HgCl2 in water and dilute to 1 liter. Nitric Acid, Cone. (sp. gr. 1.42). This reagent contains 69-70% of HNO 3 . Nitric Acid, Dilute (sp. gr. 1.20). Dilute 400 cc. of cone. HNO 3 with 600 cc. of water. This reagent contains about 32% of HNO 3 . Nitric Acid, Red Fuming (sp. gr. 1.80). This is cone. HNO 3 saturated with nitrogen peroxide. (Not to be made up in the laboratory.) Potassium Bichromate (4%). Dissolve 40 grams of K^C^O? in water and dilute to 1 liter. Potassium Ferricyanide (1 %) .Dissolve 10 grams of K 3 Fe(CN)e in water and dilute to 1 liter. Potassium Ferrocyanide (1.5%}. Dissolve 15 grams of K4Fe(CN) 6 -3H 2 in water and dilute to 1 liter. Potassium Sulfocyanate (Thiocyanate) (1%). Dissolve 10 grams of KSCN in water and dilute to 1 liter. Silver Nitrate (2.5%}. Dissolve 25 grams of crystallized REAGENTS 5 AgNOs in water and dilute to 1 liter. The solution should be kept in dark-Colored glass-stoppered bottles. Sodium Carbonate (15%}. Dissolve 150 grams of anhydrous Na2COs in water and dilute to 1 liter. It is best not to keep it in a glass-stoppered bottle. (Use a clean rubber stopper.) Sodium Hydroxide (10%). Dissolve 100 grams of NaOH (electrolytic) in water and dilute to 1 liter. It should be kept in a bottle stoppered with a clean rubber stopper and not exposed to the air any more than necessary. Sodium Phosphate (10%). Dissolve 100 grams of Na 2 HPO4'12H 2 O in water and dilute to 1 liter. Sulfuric Acid, Cone. (sp. gr. 1.84, or 66 Be\). This reagent contains about 94% of H2SO4. Sulfuric Acid, Dilute. Into 800 cc. of water pour cautiously 200 cc. of cone. H 2 SO4, with constant stirring. This solution contains about 30% of H 2 SC>4. Wijs Solution for Iodine Number. This solution may be pre- pared in two ways: (1) Dissolve separately 7.9 grams of iodine trichloride and 8.7 grams of sublimed iodine in glacial acetic acid by warming gently on the water bath, carefully covered to prevent absorption of water. Then pour both solutions into a liter volumetric flask, rinsing the containers into the flask with glacial acetic acid. Dilute to the mark with glacial acetic acid at 20 C. (2) Dissolve 13 grams of resublimed iodine in 1000 cc. of pure glacial acetic acid. Titrate 25 cc. with 0.1 Nthiosulfate. Remove another 25 cc. to a small flask. Pass washed and dried chlorine gas into -the main volume until the original thiosulfate titration is just doubled. Then add the small amount of original solution to neutralize any possible free chlorine. Preserve in amber-colored glass-stoppered bottles sealed with paraffin until ready for use. NOTE. Moisture spoils Wijs Solution. In making up by the second method, the chlorine must be passed through a washing bottle containing water and then through two bottles containing cone. H 2 SO 4 . TECHNICAL METHODS OF ANALYSIS STANDARD VOLUMETRIC SOLUTIONS General. The solutions which are to be kept in stock for general laboratory use are the following: 0.5 N Hydrochloric Acid 0.1 N Hydrochloric Acid 0.1 N Potassium or Sodium Hydroxide 0.1 N Oxalic Acid 0.1 N Potassium Permanganate 0.1 N Sodium Thiosulfate 0.1 N Iodine 0.1 N Potassium Bichromate 0.1 N Silver Nitrate 0.1 N Sulfocyanate 0.1 N Arsenious Acid In making up standard solutions all weighings must be madf with standardized weights and all volumetric apparatus (pipettes, burettes, and flasks) must have been calibrated carefully at 20 C. In titrating standard solutions the burettes should be allowed to drain at least three minutes before cheeking the reading and proper calibration corrections must be made. It is not necessary to have the solutions precisely 0.1 or 0.5 normal, provided the exact strength is known. The strength of the solution is expressed in terms of a " factor." This " factor " is the ratio between a given number of cc. of the solution in question and the number of cc. qf a theoretically correct solution. In other words, if 50 cc. of a given solution of NaOH are neutralized by 45 cc. of exactly 0.1 N HC1, then the NaOH solution has a fac- tor of 0.900 which is obtained by dividing 45 by 50. In using factor solutions the number of cubic centimeters of the solution in question used in titrating, multiplied by its factor, gives the corresponding number of cc. of a solution of correct strength. All standard solutions for the laboratory are to be made up at 20 C. or made up at a known temperature and the " factors " corrected to 20 C. (A table of corrections is given on page 13.) The " factors " of these solutions should in no case be greater than 1.005 or less than 0.995. The factor of each solution should REAGENTS 7 be determined at intervals not exceeding one month and the figures entered in a record book kept for that purpose and also on the label of the bottle. Each solution should be titrated by two chem- ists independently and their results must agree satisfactorily and be accepted before the solution is released for use. 0.5 N Hydrochloric Acid. (18.235 grams absolute HC1 per liter or about 43.0 cc. of cone. HC1, sp.gr. 1.20.) STANDARDIZATION. Make up the desired amount of solution, mix thoroughly and standardize against pure Na2C03. The sodium carbonate is best prepared by heating sodium bicarbonate * of the highest purity by one of the following two methods : I. Half fill a platinum dish with pure powdered NaHCOs, place it in an air bath already heated to about 200 C. and raise the temperature to 270-280 (never more than 300). Let remain at this temperature 0.5 hour, then cool in a desiccator and, before quite cold, transfer to a warm, dry, stoppered tube or bottle, out of which it may be weighed rapidly when wanted. For each standardization of 0.5 N acid weigh out accurately about 1.1 grams of the resulting Na2COs. II. Accurately weigh a platinum crucible and place in it about 1.75 grams of pure NaHCOs. Place in an asbestos disc with a hole cut in it which will admit the crucible to about half its depth. Cover the crucible and heat at a temperature which will just give a very dull red on the bottom. Continue the heating for at least 0.5 hour, cool in a desiccator and weigh accurately the resulting Na 2 CO 3 . After the Na2COs is prepared (it should be anhydrous and free from lumps), dissolve the accurately weighed portion in about 100 cc. of water, add 2 drops of methyl orange and titrate with the acid to the point where the color changes from yellow to pinkish orange. For very accurate work the end point should be matched against the color of 100 cc. of distilled water saturated with CO2 and containing 2 drops of methyl orange solution. CALCULATION. 1 gram of pure Na2COs is equivalent to 37.736 cc. 0.5 N acid. * The sodium bicarbonate used for this purpose should be used, for no other. Before use it must be carefully tested and its purity ascertained. The bottle should then be labeled "for standardizing purposes only." TECHNICAL METHODS OF ANALYSIS If A = weight of Na 2 CO 3 taken, B = cc. of HC1 (titration), and F = the factor of the solution, 37.732 A then F = B The solution should be so made up that F is greater than 1. Add the required amount of distilled water to make the solution exactly 0.5 N, mix thoroughly and re-standardize the fresh solu- tion as above. EXAMPLE. If the factor is 1.042, then for each liter of the solution there should be added 42 cc. of water. 0.1 N Hydrochloric Acid. (3.647 grams hydrogen chloride per liter or about 8.45 cc. of cone. HC1.) STANDARDIZATION. Follow the same method as for 0.5 N HC1; dissolve about 1 gram of Na2COs (accurately weighed) in 500 cc. of water in an accurate volumetric flask, pipette out 100 cc. of this diluted solution with an accurate pipette and titrate ' with the 0.1 N HC1. 0.1 N Caustic. (5.611 grams KOH or 4.001 grams NaOH per liter. Weigh out about 5.8 grams of stick KOH or 4.2 grams of stick NaOH for each liter of solution.) STANDARDIZATION. Pipette out -50 cc. of the solution and titrate against 0.1 N HC1, using two drops of methyl orange indi- cator. Titrate a second 50 cc. portion with phenolphthalein indicator. The factor for each indicator should be written on the bottle. The factor is obtained by multiplying the factor of the 0.1 N HC1 by the number of cc. of the latter used in titration and dividing by 50 the figure thus obtained. 0.1 N Oxalic Acid. (6.303 grams H 2 C2O4-2H 2 O per liter.) STANDARDIZATION. Pipette out 50 cc. of the solution and titrate with 0.1 N caustic and phenolphthalein. Multiply the num- ber of cc. of 0.1 N caustic used in the titration by its phenol- phthalein factor and divide by 50 to obtain the factor of the solution. NOTE. This solution may also be standardized against 0.1 N KMnO 4 solution. It should be kept in a dark-colored bottle away from light. 0.1 N Potassium Permanganate. (3.161 grams KMn04 per liter.) ' NOTE. The KMnO 4 should be dissolved in a small amount of distilled REAGENTS 9 and filtered through glass wool or a Gooch crucible with an asbestos mat before diluting to proper volume. This solution should be kept in a dark-colored bottle away from light. STANDARDIZATION. Weigh out accurately 6.700 grams of pure, freshly dried sodium oxalate. Dissolve in 250-300 cc. of hot distilled water. Transfer to a liter volumetric flask and make up to volume at 20 C. This solution will be exactly 0.1 N. Pipette out 50 cc. of the above solution into an Erlenmeyer flask. Add 5 cc. of cone. H2SO4 and heat to boiling. Titrate immediately with the KMnC>4 solution, adding the latter drop by drop at first. The first appearance of a faint but permanent pink color shall be taken as the end point. To obtain the factor of the KMnCX solution, divide 50 cc. by the number of cc. of KMn04 required for the titration. 1st Optional Method. Pipette out 50 cc. of 0.1 N oxalic acid solution and add 50 cc. of distilled water and 5 cc. of cone. H^SCX; heat to boiling and titrate with the KMnO4 solution until a perma- nent pink color forms. Multiply the factor of the 0.1 N oxalic acid solution by 50 and divide by the number of cc. of KMnO4 used, to obtain the factor of the latter. 2d Optional Method. Weigh out 1.9607 grams of pure ferrous ammonium sulfate, Fe(NH4) 2(804)2-61120. Dissolve in 100 cc. of distilled water, add 5 cc. of cone. H2SO4, cool and titrate imme- diately with the KMnO4 solution until a permanent pink color is formed. The factor is obtained by dividing by 50 the number of cc. of KMnO4 solution used. 0.1 N Sodium Thiosulf ate. (24.820 grams Na 2 S 2 O 3 -5H 2 O per liter.) STANDARDIZATION. Standardize the solution against 0.1 N K 2 Cr20? solution as follows: Weigh out accurately 4.9033 grams of c. P., freshly dried K^C^O?, dissolve in about 200 cc. of distilled water and make up to volume in a liter graduated flask at 20 C. This will make a 0.1 N solution. Place in a 350 cc. glass-stoppered bottle 100 cc. of distilled water and 30 cc. of 10 per cent KI solution. Then add from an accurate pipette 50 cc. of the above bichromate solution followed by about 7 cc. of cone. HC1. Shake and let stand for three minutes. Cool under the tap so that when the stopper is removed any adhering liquid will be sucked in. Wash the stopper carefully, and titrate 10 TECHNICAL METHODS OF ANALYSIS the contents of the bottle with thiosulfate solution. When the yellow color of iodine has almost disappeared, add about 1 cc. of starch solution and continue the titration until the deep blue color of the solution changes to sea-green. By conducting the titration carefully this change should be brought about by a single drop of thiosulfate solution. Divide the number of cc. of bichromate solution taken (i.e. 50 cc.), by the number of cc. of thiosulfate required for the titra- tion. The quotient will be the factor of the 0.1 N thiosulfate. Optional Method. Standardize the solution against 0.1 N KMn04 solution as follows: Place in a 350 cc. glass-stoppered flask or bottle 150 cc. of distilled water containing 5 cc. of cone. H2S04; cool thoroughly and then add 25 cc. of 10% KI solution. Then pipette into this mixture 50 cc. of 0.1 N KMnC>4 solution. Stopper the flask and let stand ten minutes; then titrate the iodine set free with 0.1 N Na2S2Os solution, using starch indicator, but not adding the latter until the iodine color has nearly disappeared. The final disappearance of the blue color is the end point. To obtain the factor, multiply the factor of the 0.1 N KMnC>4 solution by 50 and divide this result by the number of cc. of 0.1 N thio- sulfate used in the titration. 0.1 N Iodine. (12.692 grams sublimed iodine per liter.) NOTE. This solution should be kept in a dark-colored bottle away from light. Weigh out 12.7 grams of iodine for each liter of solution. Also weigh out (for each liter) 20-25 grams of pure KI and dissolve in as little water as possible. Then add the iodine and after it has dis- solved make up to the proper volume. STANDARDIZATION. Pipette out 50 cc. of 0.1 N thiosulfate solution and titrate with the iodine solution, using starch indi- cator, until a permanent blue color forms. In this case the starch may be added directly at the beginning of the titration. The factor is obtained by multiplying the factor of the 0.1 N thiosulfate solution by 50 and dividing the result by the number of cc. of iodine solution required for titration. 0.1 N Potassium Bichromate. (4.9033 grams K^C^O? per liter.) NOTE. This salt can be obtained in very pure condition and when made up accurately should give a 0.1 N solution. REAGENTS 11 STANDARDIZATION. Pipette 50 cc. of the solution into a 350 cc. glass-stoppered flask or bottle; add 150 cc. of water and 5 cc. of cone. H2SO4 and after cooling thoroughly add 25 cc. of 10% KI solution. Stopper the flask and let stand ten minutes, then titrate the iodine set free with 0.1 N thiosulfate solution, using starch indicator but not adding it until the yellow color of the iodine has nearly disappeared. The end point is denoted by the change in color of the solution from deep blue to light green. The factor is obtained by multiplying the number of cc. of thiosulfate solution used by its factor and dividing the result by 50. 0.1 N Silver Nitrate. (16.989 grams AgNO 3 per liter.) .STANDARDIZATION. Pipette out 25 cc. of the solution, dilute to about 250 cc., add a slight excess of dil. HC1 and let stand until clear. Filter the AgCl on a weighed Gooch crucible ; wash with an extremely dilute solution of HC1, and finally once with cold dis- tilled water. Dry at 110 C. Place the Gooch crucible in a large platinum crucible and ignite gently until the edges of the precipi- tate just begin to fuse. Cool and weigh the AgCl. The factor of the solution is obtained by dividing the weight of AgCl found by 0.3584. 0,1 N Sulfocyanate. (9.717 grams KSCN or 7.611 grams NH 4 SCN per liter.) STANDARDIZATION. Pipette 50 cc. of 0.1 N AgNOs solution into a white porcelain dish and add 100 cc. of water, 5 cc. of dil. HNOs and 5 cc. of ferric nitrate solution. The latter should be approx- imately a 10% solution and free from chlorides (see page 12). Titrate with 0.1 N sulfocyanate solution until a permanent red coloration of the liquid appears. (This is best seen by artificial light.) The factor is obtained by multiplying the factor of the 0.1 N AgNOs solution by 50 and dividing the product by the number of cc. of sulfocyanate solution required in the titration. 0.1 N Arsenious Acid. (4.948 grams As20s per liter.) Dis- solve 4.96 grams of the purest sublimed As2Os powder in about 250 cc. of distilled water in which has been dissolved about 20 grams of pure Na2CC>3. The mixture needs warming and shaking for some time in order to complete the solution. When the solu- tion is clear, cool and make up to 1 liter at 20 C. STANDARDIZATION. Pipette out 50 cc. into a beaker, dilute with 100 cc. of distilled water and titrate with 0.1 N iodine, using 12 TECHNICAL METHODS OF ANALYSIS starch indicator. The starch should not be added, however, till near the end of the titration. The factor is obtained by multi- plying the number of cc. of 0.1 N iodine solution used by its factor and dividing the product by 50. INDICATORS The following indicator solutions should be kept in stock: Methyl Orange Methyl Red Phenolphthalein Starch Potassium or Sodium Chromate Ferric Nitrate These are to be made up as follows: Methyl Orange. Dissolve 1 gram in distilled water and dilute to 1 liter. Methyl Red. Dissolve 1 gram in 100 cc. of 95% alcohol. Phenolphthalein. Dissolve 5 grams in 500 cc. of 50% alcohol. Since this solution will be slightly acid, it must be neutralized by adding 0.01 N alkali cautiously till a faint pink color appears, then just removing the color with a drop or two of 0.01 N acid. Starch. Triturate 5 grams of arrowroot starch with a little cold water and then add, with constant stirring, 1000 cc. of boiling water. Set aside to cool and then decant, or better filter. Add 2 cc. of Oil of Cassia or of CHCls as a preservative. NOTE. So-called "soluble starch" must not be used as an indicator. Potassium or Sodium Chromate. Dissolve 25 grams of K2Cr04 or 21 grams of Na2Cr04 in a small amount of distilled water. Add a drop or two of AgNOs solution to remove any chloride (sufficient AgNO 3 must be added to form a brick-red precipitate), filter and dilute to 250 cc. Ferric Nitrate. Dissolve 10 grams of Fe(NO 3 )3-9H 2 O in distilled water, add a few drops of dil. HNOs and make up to about 100 cc. A portion of this solution should be tested with AgNOa solution to make sure that it contains no chloride. NOTE. Instead of ferric nitrate, the solution may be made up from ferric alum, Fe 2 (NH 4 )2(SO4)4-24H 2 O. To this solution should be added a little cone. REAGENTS 13 8 rHOqCO^>OCOOOO5OrH OOOOOOOOrHrHC^ "~O "O 'J rH(MCOTfrllOCOl>.OOO5O CO CO CO CO CO CO *T^ CO t^S rH 0-Q 00001OT^CO(NOO5OOCO ^G, ^i C^ ^^ C^ C^ CO CO CO CO rH OOO'OOOOOOrH oSflj SdQ Ol O5 O rH b- ss^^^^.. ^ x * V to ^^ iO ^2 ^O O^ ^J^ O^ CO fH C^ S^H T-H C^ Ol C^ CO CO '^ ^O C^ OOOOOOOOOi f 1 .OCOcO OOOrHrHrH(NC^(M O O O s> O O O O CO CO rH rH O O 00 rH iO 00 rH rH rH 000 00 CO 00 o o 5. This should be calculated from the amount of P 2 Os found. To determine the correct amount, proceed as follows: Weigh out 1 gram of ferrous ammonium sulfate, dissolve in a 26 TECHNICAL METHODS OF ANALYSIS little water, boil with a few drops of cone. HNO 3 , make up to 100 cc. and use the proper aliquot as calculated from the following approximate factors: P 2 O 5 Xl.l2 = Fe 2 C>3 X 5 = Ferrous ammonium sulfate. Ferrous ammonium sulfate X 0.2036 = Fe 2 03. After adding the iron solution, add a slight excess of NILiOH and boil until the odor is nearly gone. This precipitates all the P 2 O 5 as FePO 4 and the excess of Fe as Fe(OH) 3 . Filter, wash, ignite, blast and weigh as Fe 2 03-f-Al 2 03+P 2 05. From this weight subtract the amount of P 2 0s previously determined and the weight of Fe 2 03 added. The remainder will be the Fe 2 O 3 H-Al 2 3 in the salt. Total Lime. In the filtrate from the Fe 2 0s all the Ca is present as CaCl 2 . Heat to boiling, add a slight excess of ammonium oxalate and let stand until clear (several hours if possible). Filter, ignite (finally with a blast lamp) in a platinum crucible and weigh as CaO. NOTE. The CaC 2 O 4 may also be titrated directly with O.I N KMnO 4 in the usual way (see page 326). Magnesia. Concentrate the filtrate from the . CaC 2 O4 until crystals begin to form. Then add a little water to dissolve the crystals, and finally add a large excess of ammonium or sodium phosphate and cone. NE^OH. Let stand overnight. Filter on a weighed Gooch crucible and wash with dilute NKUOH; ignite, first gently and then strongly, until white or light gray, and weigh as Mg 2 P 2 0?. Calculate to MgO. CALCULATION. Mg 2 P 2 O 7 X 0.3621 = MgO. Sulfur Trioxide. To 100 cc. of the original solution (equivalent to 10 grams) add 5 cc. of dil. HC1 and heat to boiling. Then add 5 cc. of 10% BaCl 2 solution diluted to 100 cc. This should be added boiling hot, drop by drop. Let stand overnight. Filter while hot and wash thoroughly with hot water. Ignite in a platinum crucible and weigh as BaSC>4. Calculate to 863. CALCULATION. BaSO 4 X 0.3430 = SO 3 . Barium. If 80s was found, Ba cannot be present in the solu- tion but might be present in the insoluble portion. If SOs was GENERAL INORGANIC ANALYSES 27 not found, test for Ba by adding 5 cc. of dil. H^SO* to 100 cc. of the original solution. Heat to boiling and let stand several hours, preferably overnight. Filter hot, wash with hot water, ignite in a platinum crucible and weigh as BaSO4. Calculate to BaCk. CALCULATION. BaSO 4 X 0.8923 = BaCl 2 . Salt. Dilute 100 cc. of the original solution to 500 cc. Thoroughly mix and pipette out 25 cc. of this solution, equivalent to 0.5 gram of the sample. Dilute to about 250 cc., add 5 cc. of dil. HNOa and precipitate with an excess of AgNOa solution in a large glass-stoppered Erlenmeyer flask. Shake violently, let stand at least one hour, filter on a weighed Gooch crucible, wash with water containing a few drops of AgNOa solution and finally once with boiling distilled water. Dry at 105 C. Place the Gooch crucible in a large platinum crucible, and heat gently until the edges of the AgCl just begin to fuse. Cool in a desiccator and weigh. Calculate to NaCl. CALCULATION. AgCl X 0.4078 = NaCl. Potash. Determination of potash is seldom necessary. If desired, proceed according to directions on page 41. Calculations. Calculate SOa to CaSO4. If there is an excess of SOa, calculate this to MgSC>4, and if still an excess, to Na2SO4. If CaO is in excess of SOa, calculate the excess to CaCOa (if the salt solution is turbid and shows presence of carbonates) or to CaO (if the salt solution is alkaline) or to CaCU (if the solu- tion is clear and neutral). All ?2Os must first be calculated to Caa(PO4)2, and any excess over CaO to Mga(P04)2, and any further excess to Na2HP04. Any excess of MgO over SOa or P20s calculate to MgCOa, MgO or MgCl2 (under the same conditions as for CaO) . Report Fe2Oa and A^Oa as such. If any BaCb, CaCU or MgCl2 is present, subtract the equivalent amount of AgCl from the total AgCl before calculating the latter to NaCl. If any potash is present, calculate it to KC1 and subtract the equivalent amount from the NaCl. NOTE. If the amount of iron is desired, it is best determined colorimet- rically as in water analysis. (See page 513.) 28 TECHNICAL METHODS OF ANALYSIS SODIUM NITRITE Determination of NaNC>2. Weigh out on a balanced watch glass exactly 4 grams of the sample and dissolve in water. Filter, if the solution contains much suspended matter. Dilute to 1 liter in a volumetric flask. Into a clean Erlenmeyer flask of 300-400 cc. capacity pipette 50 cc. of 0.1 N KMnO4 solution and dilute with 150 cc. of water. Pipette into this 25 cc. of the nitrite solution to be analyzed, equiva- lent to 0.1 gram of the original sample. Heat just to boiling. Add 20 cc. of dil. H2S04 and let stand ten minutes. Cool under the tap. Add 30 cc. of 10% KI solution. Upon adding the H2SO4 there will be a heavy precipitate of Mn02, but the KI solution will dissolve this and should give a perfectly clear, brown- ish red solution. Titrate this with 0.1 N thiosulfate, adding about' 5 cc. of starch solution when the color begins to get pale. Run a " blank " with 50 cc. of the 0.1 N KMn0 4 solution, going through all the operations except the addition of the nitrite solution, and subtract the blank from the titration of the sample. Calculate the difference to NaNC>2. CALCULATION. 1 cc. 0.1 N Na 2 S 2 O 3 = 0.003451 gram NaNO 2 . NOTE. After adding the H 2 SO 4 to the hot solution, the liquid should be distinctly pink or magenta after allowing the precipitate to settle. If it is not, too little KMnO 4 has been used, and the analysis must be repeated with a larger excess. SODIUM SULFIDE General. Sodium sulfide occurs commercially in two forms (1) crystals, Na2S-9H2O, containing theoretically 13.35% S and (2) fused sodium sulfide, Na2$, containing theoretically 41.07% S. As a matter of fact, the crystals often contain more than the theoretical amount of S, due to loss of moisture ; and the fused material is seldom completely dehydrated and usually contains about 25% S or 62% Na 2 S. Sulfide-sulfur. Dissolve 10 grams of the material in water and make up to 500 cc. Pipette out 25 cc. and add to this 50 cc. of 0.1 N iodine (more if necessary). Titrate back the excess of GENERAL INORGANIC ANALYSES 29 iodine with 0.1 N thiosulfate. The iodine precipitates sulfur according to the reaction Calculate the percentages of S and of Na^S. CALCULATIONS. 1 cc. 0.1 N iodine = 0.001603 gram S. = 0.003903 gram Na 2 S. NOTE. This determination is the only determination generally necessary. If it is desired to determine iron, dissolve 5 grams, acidify with HC1, boil off H 2 S, add a few drops of HNO 3 and continue boiling. Filter, precipitate iron in the filtrate with NH 4 OH, dissolve in hot 5% H 2 SO4, pass through a Jones reductor (see page 138) and titrate with standard permanganate (or determine colorimetrically in the usual manner). SODIUM SILICATE (WATER GLASS) General. Anhydrous water glass is generally given the formula , containing 20.45% of Na 2 O. According to the method of manufacture, however, its actual composition varies consider- ably from this and in commerce it is furnished as a thick solution, generally of a gravity either 40, 50, or 60 Baume. When exposed to the air, it sets first to a stiff jelly and then to a hard mass. Specific Gravity. If the solution is thin enough, determine the sp. gr. with a hydrometer. Otherwise fill a graduated flask to the mark with the solution and compare its weight with an equal vol- ume of water at the same temperature. Moisture. Moisture is not completely given off at 105 C. To get accurate results the material must be ignited. Weigh accurately about 10 grams in a beaker. Dilute with water, transfer to a 500 cc. volumetric flask and make up to the mark. Mix thoroughly and pipette 50 cc. into a weighed platinum dish. Dry in the oven at 105 C. and finally ignite cautiously to avoid spattering. Subtract the weight of residue from the weight of the aliquot taken and calculate the per cent loss on ignition. Report as moisture. Silica. Place another aliquot of 50 cc. of the solution in a platinum dish, dilute to about 100 cc., add slowly 5 cc. of cone. HC1 and evaporate to dryness. Heat at 135 C. for two hours, add a little cone. HC1 and then 5 cc. of water. Warm a few 30 TECHNICAL METHODS OF ANALYSIS moments, filter and wash with hot water. Evaporate the filtrate to dryness and heat as before for 0.5 hour. Take up with cone. HC1 and water and filter. Combine both silica residues in a weighed platinum crucible, dry in the oven, then ignite intensely with a blast lamp, cool in desiccator, and weigh. . Report the result as total SiO2. NOTE. From the above figure the amount of sodium silicate can be approximately calculated as follows: SiC>2 X 1.257 = Na 2 Si4Og. Sodium Oxide. The Na20 alkalinity can be determined by titrating with 0.1 N acid and phenolphthalein. Take an aliquot of 50 cc., corresponding to 1 gram, dilute to 500 cc. with water free from CO2 and titrate until the pink color of the phenolphthalein just disappears. Calculate to Na20. CALCULATION. 1 cc. 0.1 N acid = 0.003100 gram Na 2 O. (2) The total Na 2 O, which will include the sodium of impurities, such as sodium chloride and sulfate, is determined as follows: Combine the filtrates from the SiC>2 determination, add a slight excess of NELiOH, and boil. Then add a few cc. of ammonium carbonate solution and digest for some time. If any precipitate forms, filter it off. Make the filtrate slightly acid with H^SCU and evaporate to dryness in a weighed platinum dish. Ignite until no more white fumes are given off. Saturate with water, add a few drops of ammonium carbonate solution, evaporate to dry- ness, again ignite and weigh as Na2SC>4. Calculate the total Na2O. CALCULATION. Na 2 S0 4 X 0.4364 = Na 2 O. Sodium Sulfate and Chloride. If it is desired to determine these impurities, proceed in the usual way, using a very dilute solu- tion. Make the solution acid in the cold, adding dil. acid a little at a time. If cone, acid or heat is applied, the silicic acid will precipitate. In determining sulfate, the solution may be heated after it has been made acid, if it is sufficiently dilute. POTASSIUM OR SODIUM BICHROMATE Potassium Bichromate. Dissolve 8-9 grams of the sample, accurately weighed, in distilled water and dilute to 1 liter. Pipette 25 cc. of this solution into a wide-mouth, glass-stoppered bottle; add 15 cc. of a 10% KI solution and then 7 CQ, of cone. HC1. GENERAL INORGANIC ANALYSES 31 Run in from a burette 0.1 N sodium thiosulfate until the brown color is nearly gone. Then add a few drops of starch solution and complete the titration carefully until the color just changes from dark blue to light green. CALCULATION. 1 cc. 0.1 N Na 2 S 2 3 = 0.004903 gram K 2 Cr 2 7 . Sodium Bichromate. The same procedure may be used for the analysis of sodium bichromate, using the following factors for calculation : 1 cc. 0.1 N Na 2 S 2 3 = 0.004367 gram Na 2 Cr 2 O 7 . = 0.004967 gram Na 2 Cr 2 O 7 -2H 2 O. CYANIDES OF POTASSIUM AND SODIUM General. There is very little pure potassium cyanide on the market to-day. Most material so labeled contains varying amounts of sodium cyanide. As an insecticide or poison, however, it is chiefly bought and sold on the cyanogen content. The following procedures are the official methods of the Asso- ciation of Official Agricultural Chemists: CAUTION. Cyanides are extremely dangerous poisons. They should not be handled with the hands: and if necessary to grind them, a mask should be worn, to prevent inhaling the fine particles. On no account add a strong acid to a solution of a cyanide. Cyanogen. Weigh about 10 grams of the sample in a weigh- ing bottle, dissolve in water, and make up to volume in a liter graduated flask. Pipette a 100 cc. aliquot and titrate it with 0.1 N AgN0 3 , drop by drop, stirring constantly, until 1 drop produces a permanent turbidity. In calculating the results, 1 equivalent of silver is equal to 2 equivalents of cyanogen, according to the fol- lowing equation: 2NaCN+ AgNO 3 = NaCNAgCN+NaNO 3 . Reserve the titrated solution for the determination of chlorine. CALCULATIONS. 1 cc. of 0.1 N AgNO 3 = 0.005203 gram CN. = 0.009802 gram NaCN. = 0.01302 gram KCN. 32 TECHNICAL METHODS OF ANALYSIS Chlorine. After completion of the titration for cyanogen, as directed above, add a few cc. of 10% K2CrO4 solution as indicator and continue the titration with 0.1 N AgNOs to the appearance of the red-brown color of Ag2Cr04. The first titration with AgNOs represents the cyanogen present according to the above equation. The second titration represents the cyanogen and chlorine according to the following equation : NaCNAgCN+NaCl+2AgNO 3 = 2NaNO 3 +2AgCN+ AgCl. Therefore, subtract the first reading of the burette from the final reading and calculate the difference to chlorine. CALCULATION. 1 cc. of 0.1 N AgN0 3 = 0.003546 gram Cl. = 0.005846 gram NaCl. = 0.007456 gram KC1. REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 72. ACETATE OF LIME General. Commercial gray acetate of lime is the crude product from which acetic acid is obtained. In the distillation of wood, a mixture of acetic acid, wood alcohol, tar, etc., is obtained. The acetic acid and alcohol are separated by distillation from the tar. This mixture is neutralized with lime and the alcohol distilled off, leaving acetate of lime. The only determinations which are usually necessary are the amounts of mois'ture and of true acetate of lime. A good grade of acetate should contain not more than 5% of moisture and should show at least 80% of acetate of lime when analyzed by the fol- lowing procedure. Moisture. Dry 5-10 grams at 100 C. for two hours. This is generally sufficient to drive off all the moisture, but to make certain of this, after cooling in a desiccator and weighing, return it to the oven for another half hour and again weigh. Report the total loss as moisture. Acetate of Lime. Determine the amount of acetic acid by distillation and calculate it to calcium acetate. The apparatus is shown in Fig. 1. Weigh accurately 2 grams of the sample into GENERAL INORGANIC ANALYSES 33 the long-necked, 250 cc. Kjeldahl flask A. Add 25 cc. of water and connect the flask by a 2-hole cork with the water reservoir (separatory funnel) E and the condenser F. Add to the flask 15 cc. of cone, phosphoric acid solution, sp. gr. 1.7. Boil the con- tents of flask A over a Rose burner until the contents are about 20 cc. The acetic acid and steam will be condensed and caught in the receiver G, to which has been added 30 cc. of 0.5 N NaOH solution. FIG. 1. Apparatus for Analysis of Acetate of Lime, Regulate the stop cock M so that the water will enter drop by drop into the flask A, keeping the contents at about 20 cc. Con- tinue the distillation for two hours; remove the receiver G, replac- ing it with a fresh receiver, and titrate with 0.5 N NaOH and phe- nolphthalein. At the end of fifteen minutes, titrate the amount of acid in the second receiver, if any. Continue the distillation until no more acetic acid comes over. From the total amount of 34 TECHNICAL METHODS OF ANALYSIS 0.5 N NaOH neutralized, calculate the amounts of acetic acid and of calcium acetate. CALCULATION. 1 cc. 0.5 N caustic = 0.03002 gram HC 2 H 3 O 2 . = 0.03954 gram Ca(C 2 H 3 O2)2. NOTE. Make a careful "blank" with all reagents and correct for any acid obtained. This is very important and must not be neglected. REFERENCE. The above method was furnished by Still well and Gladding, New York, in 1909 and has given good results in this laboratory. ANTIMONY SULFIDE General. The impure antimony pentasulphide used in the rubber trade is known as " Golden Sulfide of Antimony." It can be made by boiling the crude black antimonious sulfide (Sb 2 Ss) with hydrated lime, soda ash, charcoal and sulfur. The mixture is then filtered and concentrated to crystallization, giving Schlippe's salt, NasSbS^ This, when treated with H 2 SO 4 , pro- duces the orange pentasulfide, which is allowed to settle, washed by decantation and dried. It generally contains a considerable amount of calcium sulfate, free S, and antimony oxide (Sb 2 0s), but for rubber work should be entirely free from acids and chlorides. The free sulfur generally ranges from 5 to as much as 30%; the color varies from a full orange-tan to almost purple-scarlet. The behavior of different samples with boiling CS 2 varies widely; in some cases reduction to Sb 2 Ss takes place. The same is true of the behavior of different samples on drying at 110 C. Free Sulfur. Weigh accurately 2 grams into a beaker and dis- solve the Sb 2 Ss with cone. NH 4 OH. Filter the residue on a filter paper, which has been dried and weighed in a weighing bottle, and wash with dilute NILtOH until the filtrate shows no trace of Sb 2 Ss on acidulation with HC1. Dry this filter paper and contents for five hours (or to constant weight) at a temperature not over 60 C. The loss in weight is the Sb 2 Ss, plus any moisture which may have been present. (For accurate work the moisture should be deter- mined on a separate sample by drying at a temperature not above 60 C.) Carefully fold the filter paper, place in an extraction thimble in a Soxhlet extractor, and extract for twelve hours with CS 2 . Distill off most of the CS 2 into the top of the Soxhlet, cool, and GENERAL INORGANIC ANALYSES 35 remove the flask. Evaporate the rest of the C&2 spontaneously, dry at not over 100 C. and weigh the free sulfur. NOTES. (1) Before disconnecting the flask it must be allowed to cool thoroughly, as hot CS 2 will ignite spontaneously. (2) In case the determination of free sulphur alone is desired and the analysis is urgent it may be extracted directly with chloroform or with acetone without the previous treatment with ammonia. Antimony Oxide (Antimonious Acid). Transfer the residue from the C$2 extraction to a beaker. Evaporate off all 82 and then dissolve in 40 cc. of HC1. (If the material does not com- pletely dissolve in HC1, add one-third its volume of HNOs and evaporate to dryness. Then take up the residue with 40 cc. of cone. HC1.) To the HC1 solution add 1 gram of potassium chlorate, boil until all Cl is driven off and evaporate to about 25 cc. Cool and add enough water to dissolve any salts which may have crystallized. Add 1-2 grams of KI crystals and titrate with 0.1 N thiosulfate, using starch indicator. Do not add the starch until the titration is almost finished. Carry out the titration in an Erlenmeyer flask, preferably glass-stoppered, as otherwise iodine escapes when the KI is added. The method depends upon the reduction of SbCU to SbCla by the KI and titration of the liber- ated iodine. CALCULATION. 1 cc. 0.1 N thiosulf ate = 0.00721 gram Sb 2 O3. = 0.00601 gramSb. Calcium Sulfate. If calcium sulfate is present, it is usually customary to confirm it qualitatively and report the amount " by difference," adding together the percentages of moisture, free S, Sb2$5 and Sb2Oa and subtracting from 100%. REFERENCES. Weber, C. O.: "The Chemistry of India Rubber," page 186; Heil and Esch: "Manufacture of Rubber Goods." DETERMINATION OF SMALL AMOUNTS OF ARSENIC General. The three principal classes of materials (other than arsenic salts) in which arsenic often has to be determined are : 1. Foods, drugs, and chemicals. 2. Wall papers and textiles. 3. Arsenic bronzes. 36 TECHNICAL METHODS OF ANALYSIS The principal methods employed for the first two classes are the Marsh method and various modifications of the Gutzeit method. MARSH METHOD The apparatus consists of a generating flask (an ordinary 8-ounce wide-mouth bottle with a 2-hole stopper is suitable) with a funnel tube, a U-tube containing cotton moistened with 10% lead acetate solution (to remove [28), a CaCb drying tube, and a hard glass tube of 8 mm. bore drawn down near the end to a uniform constriction about 4 cm. long and 1 mm. inside diameter and also at the very end to a narrow exit tube. Instead of hard glass, transparent silica makes very satisfactory tubes. The tube is supported over a 3-burner furnace, the part in contact with the flame being wrapped with wire gauze. Introduce into the generating flask 20-30 grams of arsenic-free zinc (either stick or mossy) and a perforated platinum disc to produce an electric couple. Insert the stopper and add through the funnel tube sufficient 20% H^SCU to start the reaction and drive out all air. When any danger of explosion is over,* heat the tube to bright redness. After running the current long enough to prove the absence of arsenic in the reagents, add slowly through the funnel tube a solution of the material in 20% H 280,4, or the solution obtained by one of the procedures described below, containing about 20% of H2SO4, keeping a steady evolution of gas. When the flow slackens, add 30% H 2 SO 4 and later 40% H2S04 until all the As has been expelled, which usually takes from two to three hours. If no As mirror forms in the constriction tube in one hour, further test may be abandoned. * Test for this as follows : Invert a small test-tube over the capillary exit of the Marsh tube, in a nearly vertical position, so that the test-tube will be gradually filled with the generated gas. Hold in this position for about one minute. Then place the thumb over the open end of the test-tube, reverse the position of the latter, bring the open end of the tube near a gas flame and remove the thumb. If the tube is filled with an explosive mixture, the con- tents will ignite with a peculiar noise resembling the yelp of a dog. Repeat this at intervals until the tube can be filled with a gas which is non-explosive and ignites quietly. As hydrogen is lighter than air the test-tube must be held inverted while filling and then reversed when brought to the flame. GENERAL INORGANIC ANALYSES 3Y If the amount of As is sufficient, cut off the constriction from the tube and weigh it, or weigh the whole tube. Then dissolve the As in a solution of sodium hypochlorite (Sb is insoluble). Wash with water and then .with alcohol, dry, cool and weigh. The loss is metallic arsenic. If the amount of As is very small, compare the mirror with a series of standard mirrors prepared in the same apparatus using quantities of a standard solution containing from 0.005 to 0.05 mg; of As20s. To prepare the standard solution, dissolve 1 gram of pure AS20s in arsenic-free NaOH solution. Acidify with H2SO4, make up to 1 liter and dilute 10 cc. of this stock solution to 1 liter. Of the latter solution 1 cc. = 0.01 mg. of SANGER-BLACK-GUTZEIT METHOD (MODIFIED) Reagents. (a) Cone, nitric and sulfuric acids, arsenic-free (sp. gr. 1.42 and 1.84, respectively). (b) Sulfuric acid (1:2). (c) Zinc, arsenic-free Stick zinc broken into pieces approxi- mately 1 cm. in length. (d) Lead acetate paper Heavy filter paper soaked in 20% lead acetate solution, dried and cut into pieces about 4.5 by 16 cm. (e) Lead acetate cotton Absorbent cotton soaked in 5% lead acetate solution. (/) Mercuric bromide paper Cut heavy, close-textured draft- ing paper (similar to Whatman's cold pressed) into strips exactly 2.5 mm. wide and about 12 cm. long. Soak for an hour in a 5% solution of HgBr2 in 95% alcohol, squeeze out the excess of solution and dry on glass rods. Cut off the ends of the strips before using. (d) 20% potassium iodide solution. (h) Stannous chloride solution 40 grams of stannous chloride crystals made up to 100 cc. with cone. HC1. (i) Standard arsenic solution Dissolve 1 gram of AsoOa in 25 cc. of 20% NaOH solution, neutralize with dil. H 2 SO 4 , add 10 cc. of cone. H 2 SO4 and dilute to 1 liter with recently boiled water. One cc. of this solution contains 1 mg. of arsenious oxide (As2Os). Dilute 20 cc. of this solution to 1 liter. Fifty cc. of the latter 38 TECHNICAL METHODS OF ANALYSIS -Mercuric Bromide Paper Lead Acetate Cotton solution when diluted to 1 liter give a dilute standard solution containing 0.001 mg. of arsenious oxide (As2O,3) per cc., which is used to prepare the standard stains. The dilute solutions must be freshly prepared immediately before use. Apparatus. Use a 2-ounce wide- mouth bottle as a generator. Fit this by means of a perforated rubber stopper with a glass tube, 1 cm. in diameter and 6 cm. long, containing a piece of the lead acetate paper rolled into a cylinder. Connect this tube by means of a per- forated rubber stopper with a similar tube filled with the lead acetate cotton, squeezed to remove excess of the solu- tion. The cotton in all tubes used should be uniformly moist to obtain compara- tive stains. Connect the second tube by means of a perforated rubber stopper with a narrow glass tube, 3 mm. internal diameter and 12 cm. long, containing a strip of the mercuric bromide paper. (See Fig. 2.) Rubber stoppers used for connections must be free from any white coating. Preparation of Solution. Weigh 5-50 grams of the finely divided and well- mixed sample into a porcelain casserole, the amount selected depending upon the character of the material and the ease with which it is oxidized. With dry, highly nitrogenous substances employ 5 grams; pulped vegetables, 25 grams; liquids with low solid content, like beer or vinegar, 50 grams. Add 10-15 cc. of HNO 3 , cover the casserole by setting a watch glass inside the rim, convex side upward, heat until vigorous action is over, cool and add 10 cc. of cone. H2SO4. Heat on a wire gauze over a flame until the mixture turns dark brown or black, then Lead Acetate Paper Solution -Stick Zrac- FIG. 2. Sanger-Black- Gutzeit Apparatus for Arsenic Determination. GENERAL INORGANIC ANALYSES 39 add more HNOs in 5 cc. portions, heating between each addition until the liquid remains colorless or yellow when evaporated until SOs fumes are evolved. To remove completely all nitric or nitrous acid, evaporate to about 5 cc., cool, dilute with 10-15 cc. of water and again evaporate until white SOs fumes are evolved. Cool, dilute with water, again cool, and make up with water to a definite volume (usually 25-100 cc., depending upon the amount of sample taken and its arsenic content). Determination. Introduce 20 cc. of the solution or, if the amount of arsenic is large, an aliquot containing not more than 0.03 mg. of As20s prepared as directed under Preparation of Solution, into the generator of the apparatus as described above and add 20 cc. of dil. EkSCU. If the total volume is less than 40 cc., dilute to that volume with water, and add 4 cc. of 20% KI solu- tion. Heat to about 90 C., add 3 drops of SnCl2 solution and heat for ten minutes. Cool the generator and its contents in a pan containing water and ice; when cold, add about 15 grams of the stick zinc and connect the entire apparatus described above. Keep the bottles in ice water for fifteen minutes, then remove from the bath and allow the evolution of gas to proceed for an hour longer. Remove the sensitized paper and compare the stain with similar ones produced under like conditions with known amounts of arsenic, using portions of the standard arsenic solution, containing 0.001, 0.002, 0.005, 0.010, 0.015, 0.025 and 0.030 mg. of As2Os, adding such quantities of water and H^SCU that the same volume and acid strength are maintained as above. NOTE. It is very necessary in making comparisons that the same apparatus be used and the same proportions of the same reagents. METHODS OF PREPARING SAMPLES FOR TEST (1) Foods, Drugs, and Chemicals. (a) Syrups, baking powders and other materials soluble in water or acid do not need preliminary treatment. (b) Beer is treated as follows: Measure 100 cc. (freed from C02 by agitation) in a 7-inch porcelain evaporating dish; add 20 cc. of pure cone. HNOs and 3 cc. of pure cone. H2SO4 and heat cautiously until vigorous chemical action sets in. Turn the flarne low, or remove it altogether, and stir vigorously until the frothing ceases, 40 TECHNICAL METHODS OF ANALYSIS after which the liquid may be boiled freely. Transfer to a large casserole and continue boiling until nearly all the HNOs is driven off. Then, holding the casserole by the handle, continue the heating until the mass chars and fumes of SOs are given off, giving the casserole a rotary motion to prevent spattering. The residue should be reduced to a dry, black pulverulent char soon after the SOs fumes begin to come off freely. If still liquid, stir in pieces of filter paper while still heating until the residue is dry, avoiding an excess of paper. Cool, add 50 cc. of water and remove the masses of char from the sides of the dish with a stirring rod. Heat to boiling and filter. Use the filtrate for the Marsh apparatus, adding it gradually. (c) Meat, vegetables and the like may be treated as follows, varying the proportion of acids to suit the conditions: Heat at 150-160 C. in a porcelain dish 100 grams of the finely divided material with 23 cc. of pure cone. HNOs, stirring occasionally. When the mixture assumes a deep orange color, remove from the heat, add 3 cc. of pure cone. H2S04 and stir while nitrous fumes are given off. Heat to 180 C. and add, while still hot, drop by drop with stirring, 8 cc. of HN0 3 . Then heat at 200 C. until S0 3 fumes come off and a dry charred mass remains. Pulverize the mass, extract with hot water, filter, evaporate to small volume, take up in cold 20% H2SO4 and treat by the Marsh or Gutzeit method. ALTERNATIVE METHOD. Digest at room temperature for some hours 5-20 grams of the material in a casserole with about an equal bulk of HNO 3 . Add 20 cc. cone. H 2 SO 4 and digest further at a gentle heat until the mixture begins to char. Add about 2 cc. of HNO 3 and heat until SO 3 fumes appear, repeating the addition of acid and heating until oxidation appears to be practically complete. Remove all HNOs by dilution and evaporation to the fuming stage. Then dilute with 4 volumes of water. At this point about twice the bulk of saturated SC>2 solution may be added and the evaporation repeated, thus reducing to the arsenious condition, but this is not usually necessary. (2) Wall Paper and Textiles. Take a piece of paper 3.25X4 inches (equivalent to 0.01 sq. yd.) or a piece of cloth 12X10.8 inches (equivalent to 0.1 sq. yd.). Cut in small pieces and place in a porcelain dish or Kjeldahl flask. Add about 50 cc. of a mixture of 25 cc. cone. H2SO4 and 1 cc. cone. HNOa. Heat on a low flame until completely charred, and then continue heating GENERAL INORGANIC ANALYSES 41 until strong fumes of SOs appear. Cool, dilute and filter into a liter flask. Cool and dilute to the mark. Treat 250 cc. of this solution by the Gutzeit method, using 2 grams of arsenic-free zinc and letting the reaction run until the Zn is all dissolved. If a strong reaction is obtained, repeat, using a smaller aliquot. A very faint yellow color on the paper indicates the following amounts of metallic arsenic per square yard : Arsenic (Grains per Square Yard) Solution Taken cc. Paper Cloth 500 0.007 0.0007 400 0.009 0.0009 300 0.012 0.0012 200 0.018 0.0018 100 0.036 0.0036 10 0.36 0.036 1 3.6 0.36 NOTES. (1) The Massachusetts law allows not over 0.1 grain of metallic arsenic per square yard of wall paper and not over 0.01 grain in dress goods. (2) For court cases use the Marsh method and compare with standards. (3) Determination of Arsenic in Alloys. (See page 149.) REFERENCES. Leach: "Food Inspection and Analysis," 1913 Edition, pages 74 and 728; Am. Chem. J. 11, 250; Proc. Am. Acad. Arts Sci. 26, 24; J. Soc. Chem. Ind. 26, 1115 (1907); J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 171. DETERMINATION OF POTASSIUM IN FERTILIZERS, SOILS, PLANT ASHES, ETC., AND IN POTASH SALTS (A) LINDO-GLADDING METHOD (1) Preparation of Reagents. (a) Ammonium Chloride Solution. Dissolve 100 grams of NKiCl in 500 cc. of water, add 5 to 10 grams of pulverized K^PtCle, and shake at intervals for 6-8 hours. Let the mixture settle overnight and filter. The residue may be used for the preparation of a fresh supply. (6) Platinum Solution. The platinum chloride solution used contains 1 gram of metallic platinum (2.1 grams of H^PtCle) per 10 cc. 42 TECHNICAL METHODS OF ANALYSIS (c) 80% Alcohol Sp. gr. 0.8645 at 15 C.* (2) Methods of Making Solution. (a) Mixed Fertilizers. Place 2.5 grams of the sample upon a 12.5 cm. filter paper and wash with successive small portions of boiling water till the filtrate is about 200 cc. Add to the latter 2 cc. of cone. HC1 and heat to boiling. Transfer to a 250 cc. graduated flask and add to the hot solution a slight excess of NELiOH and sufficient (NH4)2C2O4 solution to precipitate all the lime present. Cool, dilute to 250 cc., mix, and pass through a dry filter. (6) Potash Salts, Muriate and Sulfate of Potash, Sulfate of Potash and Magnesia, and Kainit. Dissolve 2.5 grams and dilute to 250 cc. without the addition of NH 4 OH and (NH4) 2 C 2 O4. (c) Organic Compounds. When it is desired to determine the total amount of K^O, in organic substances, such as cottonseed meal, tobacco stems, etc., saturate 10 grams with cone. H2S04 and ignite in a muffle at a low red heat to destroy organic matter. Add a little cone. HC1, warm slightly in order to loosen the mass from the dish, and proceed as directed above for Mixed Ferti- lizers (a). (3) Determination. (a) Mixed Fertilizers. Evaporate 50 cc. of the solution made according to (2), corresponding to 0.5 gram of the sample, nearly to dryness, add 1 cc. of dil. H2SO4 (1:1), evaporate to dry- ness, and ignite to perfect whiteness. All K 2 is in the form of non-volatile K2SO4 and a full red heat must be maintained until the residue is perfectly white. Dissolve the residue in hot water, using at least 20 cc. for each 0.1 gram of K^O. Add a few drops of HC1 and an excess of the platinum solution. Evaporate on a water bath to a -thick paste in a porcelain dish and treat the residue with 80% alcohol, avoiding exposure to NH%. Filter and wash the precipitate thoroughly with 80% alcohol both by decantation and on the filter, continuing the washing after the fil- trate is colorless. Wash finally with 10 cc. of NEUCl solution (1, a) to remove impurities from the precipitate, and repeat this washing 5 or 6 times. Wash again thoroughly with 80% alcohol * Denatured alcohol, made up according to formula 30 (U. S. I. R. Reg. No. 30, revised) and diluted with water to make 80% by volume may also be used. GENERAL INORGANIC ANALYSES 43 and dry the precipitate for thirty minutes at 100 C. Weigh as K 2 PtCl 6 and calculate to K 2 O. NOTE. The precipitate should be completely soluble in water. (b) Muriate of Potash. Acidify 50 cc. of the solution prepared according to (2, 6) with a few drops of HC1, add 10 cc. of platinum solution and evaporate to a thick paste. Treat the residue as under (3, a). (c) Sulfate of Potash, Sulfate of Potash and Magnesia, and Kainit. Acidify 50 cc. of the solution, prepared according to (2, b) with a few drops of HC1 and add 15 cc. of platinum solution. Evaporate the mixture and proceed as directed under (3, a), except that 25 cc. portions of NHiCl solution should be used. (d) Water-soluble Potash in Wood Ashes and Cotton Hull Ashes. Use the above method, making the solution according to (2, a) and pay special attention to the note under (3, a). (4) Factors. For the conversion of K^PtCle to KC1, use the factor 0.3067; to K 2 SO 4 , 0.3584; and to K 2 O, 0.1938. (B) OPTIONAL METHOD NOTE. Method (A) is preferable in the presence of soluble sulphates. Reagents. The same as for the Lindo-Gladding method (A). Preparation of Solution. Prepare the solution as directed under the Lindo-Gladding method (A, 2) omitting in all cases the addition of ammonium hydroxide and oxalate. Determination. Dilute 25 cc. of the solution (50 cc. if less than 10% of K 2 O is present) to 150 cc. Heat to boiling and add, drop by drop, and with constant stirring, a slight excess of BaCl 2 solution. Without filtering, add in the same manner Ba(OH) 2 solution in slight excess. Filter while hot and wash until the precipitate is free from Cl. Add to the filtrate 1 cc. of cone. NHiOH and then a saturated solution of ammonium carbonate until the excess of Ba is precipitated. Heat and add, in fine pow- der, 0.5 gram of pure oxalic acid or 0.75 gram of (NH4) 2 C 2 04. Filter, wash free from Cl, evaporate the filtrate to dry ness in a platinum dish and ignite carefully over the free flame below red heat until all volatile matter is driven off. Digest the residue with hot water, filter through a small filter and dilute the filtrate, 44 TECHNICAL METHODS OF ANALYSIS if necessary, so that for each 0.1 gram of K^O there will be at least 20 cc. of liquid. Acidify with a few drops of HC1 and add platinum solution in excess. Evaporate in a porcelain dish on the water bath to a thick paste and treat the residue with 80% alcohol, both by decantation and after collecting on a weighed Gooch. Dry for thirty minutes at 100 C. and weigh. If there is an appearance of foreign matter in the precipitate, it should be washed as described above under Method (A) with several portions of 10 cc. each of the NEUCl solution. (C) J. LAWRENCE SMITH METHOD For Rocks and Silicious Materials Mix 0.5 gram of the sample, finely ground, with 0.5 gram of pure dry NHUCl, by gentle trituration in an agate mortar, then add 4 grams of dry powdered CaCOs and mix intimately.* Place the mixture in a large platinum crucible, rinsing the mortar with a little of the CaCOs powder. Place the crucible in a hole cut in a sheet of asbestos, the hole of such a size that not more than two- thirds of the crucible will be below the asbestos. Heat very gently over a small Bunsen burner until fumes of NELt salts no longer appear. Then heat with a higher flame until the lower part of the crucible is brought to a red heat. Not more than three-quarters of the crucible should be red and it should be kept well covered during the fusion. Keep this temperature constant for forty to sixty minutes. The temperature desired is that which suffices to keep in a state of fusion the CaCU formed by the reaction of NHiCl with CaCOs. The mass, however, does not become liquid, since the fused CaCl2 is absorbed by the large quantity of CaC0 3 present. The silicate itself should not fuse, since this would render impos- sible the disintegration of the mass at the end of the operation. Moreover, too high a temperature causes a volatilization of alkali chlorides. Certain silicates, e.g., those which contain much ferrous iron, may fuse when heated with the above mixture, even if no higher temperature is employed than is necessary to effect decom- position. If this occurs, it is better to repeat the ignition with a new portion, using 8-10 parts of CaCOs. * For soils double the above amounts. GENERAL INORGANIC ANALYSES 45 The mass contracts in volume during the ignition, and is usually easily detached from the crucible. Boil it for 0.5 hour in a covered porcelain dish with 50-75 cc. of water, replacing water lost by evaporation. Decant the solution from the residue upon a filter, boil the residue a few minutes with water, and decant again. If the residue is now all in a finely disintegrated state, it may be brought upon the filter and washed. But if, as is often the case, a portion remains coherent or in a coarsely granular state, it must be reduced to a fine state of division by trituration with a porcelain or agate pestle in the dish, and boiling with water again. By a few repetitions of the trituration, boiling and decant- ing, allowing the fine suspended portion to pass upon the filter each time, the whole can usually be transferred to the filter in properly disintegrated condition in the course of an hour. Next wash until a few drops of the washings acidified with HNOs give but a slight turbidity with AgNOs. The filtrate now contains the alkalies of the silicate as chlorides together with cal- cium chloride and hydroxide. It is not advisable to concentrate this filtrate in glass, since it might dissolve an appreciable quan- tity of sodium. Precipitate, therefore, the Ca at once with ammo- nium carbonate; let the precipitate settle, and concentrate the supernatant solution in a porcelain (or platinum) dish, decanting it into the latter, portionwise if necessary, finally rinsing the pre- cipitate into the porcelain dish. When the whole is thus reduced to about 30 cc., add a little more ammonium carbonate and NH4OH, heat and filter into a platinum (or porcelain) dish. Evap- orate to dryness on a water bath, expel NILtCl by gentle ignition, and dissolve the residual alkali chlorides in 3-5 cc. of water. A little black or dark-brown flocculent matter usually remains undissolved, and the solution may still contain traces of Ca. Add 2 or 3 drops of ammonium carbonate and NEUOH, warm gently, and filter through a very small filter into an unweighed but weighable platinum dish. Evaporate to dryness on a water bath, heat at dull red to incipient fusion of the alkali chlorides and, after cooling, weigh. Dissolve the mixed chlorides in water and filter through a small filter into a porcelain dish. Ignite the filter in the platinum vessel previously used and weigh. Sub- tract this weight from the first weight to obtain the weight of the NaCl+KCl. Determine the K^O in the filtrate in the porcelain 46 TECHNICAL METHODS OF ANALYSIS dish by precipitating with platinum chloride as previously described, adding sufficient platinum chloride to combine with the total weight of alkali chlorides calculated as Nad, i.e., an amount of metallic Pt 1.67 times the weight of alkali chlorides found. NOTES. (1) Smith's method is the most convenient of all methods for extracting alkalies from silicates, and is universally applicable, except perhaps in the presence of boric acid. When carried out as here described, results are sufficiently accurate in most cases. If, however, the silicate is rich in alkalies, a loss amounting to 0.1 or 0.2% of the mineral is possible. If great accuracy is desired in such cases, a repetition of the whole process may be applied to the residue left by treatment of the ignited mass with water. It need hardly be mentioned that unless care be taken to use reagents perfectly free from soda and to avoid action of the solution on glass, an amount of soda may be intro- duced from these sources equal to 0.1 or 0.2%. (2) The above methods are the official methods of the Assoc. Official Agr. Chemists. The factors are based on 1920 atomic weights. REFERENCES. Journal Assoc. Official Agr. Chemists, Methods of Analy- sis (1916), pages 12, 24, 26; Fresenius: ' 'Quantitative Chemical Analysis," page 426. LEAD ARSENATE General. Lead arsenate, furnished on the market as an insecticide, generally consists either of Pb 3 (AsO 4 )2 or PbHAsO4 or a mixture of the two. The former is prepared from lead acetate and sodium arsenate according to the reaction: 3Pb(C 2 H30 2 )2 3H 2 O+2Na 2 HAsO 4 7H 2 O = Pb 3 (AsO4) 2 +4NaC 2 H3O 2 -3H2O+2HC 2 H3O 2 +llH 2 O. The second form is made from lead nitrate and sodium arsenate, probably according to the reaction: Pb(NO 3 ) 2 +Na 2 HAs0 4 7H 2 O = PbHAs0 4 +2NaNO 3 +7H 2 0. Pb 3 (AsO 4 ) 2 contains theoretically 74.44% of PbO and 25.56% of As 2 0s. On the other hand, PbHAsO 4 contains theoretically 64.29% of PbO, 33.11% of As 2 O 5 and 2.60% of water of con- stitution. In technical analysis it is often customary to report the sum of the total PbO+As 2 05 as " Lead Arsenate." To give the seller full benefit, however, this sum should be divided by 0.9740 and the result reported as lead arsenate, thus allowing for the maximum amount of water of constitution. GENERAL INORGANIC ANALYSES 47 Total Volatile Matter (Moisture, etc.). In case the sample is in the form of a paste, as it usually is, it should be very thor- oughly and rapidly mixed and about 50 grams dried to constant weight in a flat glass Petri dish at 105 C. and the total loss in weight determined. Save a portion of the original paste t for determination of free acetic acid and free ammonia, if these deter- minations are desired. Grind the dried sample to a fine powder. Mix well, transfer a small portion to a sample bottle and again dry for one to two hours at 105-110 C. Use this anhydrous material for the determination of total PbO and total As2Os. Total Lead Oxide. Dissolve 2 grams of the dry powder in 50 cc. of HNOs (1 : 4) on the' steam bath. The sample should dissolve without residue. (In case there is an appreciable residue, it may be filtered out, ignited and weighed and reported as Insol- uble Matter.) Transfer to a 250 cc. volumetric flask, cool to room tempera- ture and make up to the mark. Pipette 50 cc. of this solution into a 400 cc. beaker; dilute to at least 300 cc.; heat nearly to boiling; add NH^OH to incipient precipitation and then dilute HNOs (1 : 10) to redissolve the precipitate, avoiding more than 1-2 cc. excess. Add slowly from a pipette a hot solution of K2Cr04 or K2Cr207 until an excess is indicated by the yellow color of the solution. Boil for two or three minutes. The PbCr04 will settle out, leaving a clear solution, in about fifteen minutes. Filter on a weighed Gooch crucible, washing thoroughly by decantation with hot water. Dry to constant weight at 140- 150 C. and calculate to PbO. CALCULATION. PbCrO 4 X 0.6906 = PbO. NOTES. (1) The PbCKX may contain a small amount of lead arsenate which causes slightly high results. This error rarely amounts to more than 0.1-0.2%. (2) Instead of drying the PbCrO 4 at 140-150 C. it is permissible to dry at 105 C., then place the Gooch in a larger platinum crucible and ignite at dull redness. The flame must not come in direct contact with the PbCrO 4 . Total Arsenic Oxide. Transfer 100 cc. of the HN0 3 solution prepared above to a porcelain or platinum dish, add 6 cc. of cone. H2S04, evaporate to a syrupy consistency on the water bath and then on the hot plate to copious white fumes. Cool, wash 48 TECHNICAL METHODS OF ANALYSIS into a 100 cc. volumetric flask with water, and make up to the mark. Filter through a dry filter and use a 50 cc. aliquot for analysis. Transfer this to a 400 cc. Erlenmeyer flask, add about 200 cc. of water, 4 cc. of cone. H2S04 and 1 gram of KI crystals. Boil until the solution is colorless or only faint yellow. The volume must not be allowed to become less than 50 cc. Cool the solution under running water, dilute to about 300 cc. and add a little starch solution. If this colors the solution blue, add a drop or two of 0.1 N Na2S2Os until the color is just discharged, then add a drop of methyl orange indicator and powdered Na2CO3, cau- tiously at first to avoid loss by foaming. When the solution, after mixing, becomes yellow, add just enough dil. H^SCX to produce a pink color; and then make alkaline again with an excess of NaHCOs powder. Add a considerable excess of bicarbonate. Finally titrate the solution with 0.05 N or 0.1 N iodine solution to the appearance of a blue color throughout the solution. CALCULATION. 1 cc. 0.1 N iodine = 0.005748 gram As2C>5. NOTE. A generous excess of bicarbonate insures a sharp end point in titrating. Lead Arsenate. Add together the total PbO and divide the sum by 0.9740 and report the result as lead arsenate. Total Water-soluble Matter. Weigh to 0.01 gram about 4 grams of paste (or 2 grams if the sample is a dry powder), place in a tightly stoppered flask or bottle with 250 cc. of freshly boiled and cooled distilled water for each gram of paste and keep at 32 C. for twenty-four hours, shaking well every hour of the working day (8 times in all), filtering at the end of twenty-four hours. It is generally most satisfactory to filter through a Gooch crucible, rinsing out the filtering flask once or twice with the first portions of filtrate and discarding them. Continue the filtration until about 350 cc. of filtrate have been obtained. Fill with this a graduated 250 cc. flask, first rinsing the latter several times with small por- tions of the liquid. Evaporate the entire 250 cc. in a weighed platinum dish to dryness on the steam bath. Dry to constant weight at 100 C. and weigh the total soluble matter. The weight in centigrams gives the direct percentage. NOTE. In most cases it is sufficiently accurate to let stand for two or three days at room temperature (shaking frequently), instead of twenty-four GENERAL INORGANIC ANALYSES 49 hours at 32 C. The individual for whom the analysis is made often specifies the time for digesting for water-soluble impurities. Water-soluble PbO. Dissolve the residue of total soluble matter, above determined, in a small amount of water, add a few cc. of cone. HNOs and evaporate nearly to dryness on the steam bath; wash this into a small beaker with a little water, add 1 cc. of cone. H2S04, evaporate to a syrup on the steam bath and then to white fumes on the hot plate. Cool, add 10 cc. of water; and swirl the beaker gently to throw any PbSCU into the center. If any is present, let stand one hour, or for very accurate work, overnight. Then filter on a tiny filter, and wash with 5% H^SCU solution. Dry the filter paper and ignite in a weighed porcelain crucible. Treat the residue with a little HNOs, which is after- wards evaporated off, and then with a drop or two of H2S04. Ignite gently, cool and weigh as PbSC>4. Calculate to PbO. CALCULATION. PbS0 4 X 0.7360 = PbO. Water-soluble AS2O5. Water-soluble As205 is determined in the filtrate from the water-soluble PbO in exactly the same man- ner as for total As2Os above, using, however, 0.01 N instead of 0.1 N iodine for titration. It is important that the solution shall be perfectly clear and the titration carefully made. Make correc- tions for the iodine necessary to produce the same color using the same chemicals and volumes of solution. (See note (2) at end of method.) Free Acetic Acid. Weigh out 2-10 grams of the original moist sample and transfer to a 250 cc. flask with 20-40 cc. of water. Connect with a condenser and heat to boiling. Pass steam through the solution, regulating the flame under the flask so that the volume will remain nearly constant, and collect about 200 cc. of distillate. Titrate the distillate with 0.1 N NaOH and phenol- phthalein, and calculate to acetic acid. CALCULATION. 1 cc. 0.1 N caustic = 0.0060 gram HC2HsO2. Free Ammonia. Wash 2-10 grams of the original sample into a 200 cc. distilling flask and dilute to about 150 cc. Concen- trate by distillation to about 25 cc. and collect the distillate in 0.1 N-HC1. Titrate the excess of HC1 with standard alkali, using methyl red or cochineal indicator, and calculate the NHs from the amount of acid consumed. CALCULATION. 1 cc. 0.1 N HC1 = 0.0017 gram NH 3 . 50 TECHNICAL METHODS OF ANALYSIS Calculation of Results. From the total loss on drying sub- tract any free acetic acid or free ammonia which the sample may contain, and report the difference as Moisture. In case free acetic acid and free ammonia are not determined, it is customary to report the total loss on drying as Moisture, loss at 105 C. The percentages of total PbO, total As2O5, and water-soluble impurities should be calculated to the absolute dry basis. These results should then be calculated to the original paste by multiply- ing each by 100% minus the per cent of total moisture at 105 C. Subtract the per cent of soluble As20s and of soluble PbO from the total soluble matter and report the difference as " Soluble impurities." NOTES. (1) The procedures for Moisture, Total PbO, Total As 2 O 5 , and Water-soluble As 2 O 5 are essentially the methods of the Association of Official Agricultural Chemists as published in its Journal, Methods of Analysis (1916), page 67, slight changes being made for convenience in conforming to our usage. (2) Lead arsenate as now made seldom contains an appreciable amount of water-soluble PbO, consequently the water-soluble As 2 O 5 can be determined directly on the filtered solution without going through the steps of removing lead. BORDEAUX MIXTURE General. Standard Bordeaux Mixture is made up in the pro- portions of 6 pounds of CuS04 and 4 pounds of lime to 50 gallons of water; but a more common mixture is 5 : 5 : 50, and for peach trees the general mixture is 3 : 9 : 50. The following procedures are essentially the official methods of the Association of Official Agricultural Chemists. NOTE. Bordeaux Mixture is often mixed with other insecticides. For analysis of such mixtures see page 52. Moisture. (A) Powder. Dry 2 grams to constant weight at 105-110 C. and express the loss in weight as moisture. (B) Paste. Beat about 100 grams in the oven at 90-100 C. un- til dry enough to powder readily and determine the loss in weight. Powder this partly dried sample and determine the remaining moisture in 2 grams as for powder above. Determine GENERAL INORGANIC ANALYSES 51 directed later, both in the original paste and in this partially dried sample. Calculate the total moisture by the following formula: (100-a) (6+c) in which M=per cent total moisture in original paste; a = per cent loss in weight of first drying; b = per cent loss in weight of second drying ; c = per cent CO2 in paste after first drying; and d = per cent total CCb in original paste. Carbon Dioxide. (A) Apparatus. This consists of a 200 cc. Erlenmeyer flask closed with a 2-hole stopper; one of these holes is fitted with a dropping funnel, the stem of which extends almost to the bottom of the flask; the outlet of a condenser, which is inclined upward at an angle of 30 from the horizontal, passes downward through the other hole. The upper end of the con- denser is connected with a CaC^ tube which in turn is connected with a double U-tube filled in the middle with pumice fragments, previously saturated with CuSO* solution and subsequently dehydrated, and with CaCl2 at either end. Then follow two weighed U-tubes for absorbing the CO2, the first filled with porous soda-lime, and the second, one-third with soda-lime and two- thirds with CaCU, the latter reagent being placed at the exit end of the train. A Geissler bulb, partly filled with H2SO4, is attached to the last U-tube to show the rate of gas flow. An aspirator is connected with the Geissler bulb to draw air through the apparatus. An absorption tower filled with soda-lime is con- nected with the mouth of the dropping funnel to remove CC>2 from the air entering the apparatus. (B) Determination. Weigh 2 grams of the dry powder, or 10 grams of paste, into the Erlenmeyer flask, add about 20 cc. of water, attach the flask to the apparatus, omitting the 2 weighed U-tubes, and draw CO2-free air through the apparatus until the original air is displaced. Then attach the weighed U- tubes in the position as described above, close the stop-cock of the dropping funnel, fill half full with dil. HC1 (1:1), reconnect with the soda-lime tower, and let the acid flow into the Erlen- meyer flask, slowly if there is much CO2, rapidly if there is little. 52 TECHNICAL METHODS OF ANALYSIS When effervescence diminishes, place a low Bunsen flame under the flask and start the flow of water through the condenser, a slow current of air being allowed to flow through the apparatus at the same time. Maintain a steady but quiet ebullition, and a slow air current through the apparatus. Boil for a few minutes after water has begun to condense in the condenser, then remove the flame and continue the aspiration of air at the rate of about 2 bubbles per second until the apparatus is cool. Disconnect the tared absorption tubes, cool in the balance case and weigh. The increase in weight is due to C02. Copper. Dissolve 2 grams of the dry powdered sample in about 75 cc. of water and 20 cc. of cone. HNOs in a 250 cc. beaker. Dilute to 200 cc., and electrolyze as described in the method for Brass and Bronze, page 145. Calculate per cent of Cu in the sample. Test the solution to make sure all Cu has been removed. REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916", page 69. BORDEAUX MIXTURE WITH PARIS GREEN AND LEAD ARSENATE General. Bordeaux Mixture is often mixed with other insecti- cides and sold under various names. " Pyrox " is generally a mixture of copper and lead arsenates, or Paris Green, with Bor- deaux Mixture. The following procedures are those of the Association of Official Agricultural Chemists. BORDEAUX MIXTURE WITH PARIS GREEN Moisture and Carbon Dioxide (Official). These are deter- mined as in Bordeaux Mixture. Copper. Method I (Tentative). Dissolve 2 grams of the dry powdered sample in a few cc. of cone. HNOs, add 25 cc. of a 3% solution of H2O2 and warm for 5-10 minutes. Make slightly alkaline with NHiOH and then slightly acid again with dil. HNOs. Transfer to a weighed 150 cc. platinum dish, add 15-20 cc. of H2O2, dilute to 100 cc. and electrolyze, using a rotating spiral anode and a current not exceeding 2 amperes. After electrolysis has proceeded for about twenty minutes, add to the electrolyte 0.5 gram of ferric sulfate dissolved in a few cc. of water together with a drop or two of HNOs. After all Cu is deposited, wash the GENERAL INORGANIC ANALYSES 53 deposit with water by siphoning, then rinse with alcohol, dry for a few minutes in the oven, weigh and calculate the per cent of Cu. (Do not pass the current for more than five to ten minutes after all the copper has been deposited without adding more ferric sulfate solution.) Method II. (Tentative). Treat 1 gram of the dry, powdered sample with 20 cc. of water and 5-6 cc. of cone. HNOs, heat to boiling, cool, and add a slight excess of cone. NB^OH. Wash the solution and precipitate into a 150 cc. weighed platinum dish, and electrolyze, using a rotating anode and a current of about 4 amperes and 3-4 volts for about ninety minutes (or until all Cu is deposited). Wash the deposit by siphoning until it is clean, being careful not to use too much wash water. Dissolve the Cu in 5 cc. of cone. HNOs, dilute to 100 cc. and electrolyze as before, except that all the Cu will be deposited in thirty minutes. Wash the deposit with water by siphoning, then rinse with alcohol, dry for a minute or so in an oven, weigh and calculate the per cent of Cu. Total Arsenic (Official) . Proceed as directed for Total Arsenic in Paris Green (page 55) using an amount of the dry, powdered sample equal to the As2Os equivalent of 500 cc. of the standard iodine solution. The number of cc. of the standard iodine solution used, divided by 2, represents directly the per cent of total arsenic in the sample, expressed as As203. Total Arsenious Oxide (Tentative). Proceed as directed for Total Arsenious Oxide in Paris Green (page 56) using an amount of the dry, powdered sample equal to the As2Os equivalent of 200 cc. of the standard iodine solution. Before titrating, all Cu must be in solution. The corrected number of cc. of the standard iodine solution used, divided by 2, represents directly the total As2Os in the sample. Water-soluble Arsenious Oxide (Tentative). Proceed as described under Water-soluble Arsenious Oxide in Paris Green (page 57), using 2 grams of the sample, and slightly acidify with HC1 the aliquot employed before adding the excess of NaHCOs. 54 TECHNICAL METHODS OF ANALYSIS BORDEAUX MIXTURE WITH LEAD ARSENATE Moisture and Carbon Dioxide (Official). Proceed as above described. Copper (Tentative). Proceed as above described, using Method II. Lead Oxide (Tentative). Dissolve the PbO 2 (which will contain a little arsenic) from the anodes used in the copper elec- trolysis, under Method II, by means of dil. HNOs and a little H2O2, and add to this solution the washings from both electrolyses of Cu. Add NHiCl to dissolve any PbSO4 which may have pre- cipitated out, and make the solution up to 1 liter. Concentrate a 500 cc. aliquot of this solution to about 300 cc. (all H2O2 must be expelled from the solution) ; transfer to a 400 cc. beaker and precipitate the Pb as PbCrC>4 as directed on page 47 for the determination of Total Lead Oxide in Lead Arsenate. Total Arsenic (Official). Proceed as directed under Total Arsenic in Paris Green, page 55, using an amount of the dry, powdered sample equal to the As2Os equivalent of 500 cc. of the standard iodine solution. The number of cc. of the standard iodine solution used, divided by 2, represents directly the per cent of total arsenic in the sample expressed as As2Os. Water-soluble Arsenic Oxide (Tentative). Proceed as directed under Water-soluble Arsenic Oxide in Lead Arsenate, page 49. REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), pages 71, 72. PARIS GREEN General. Paris Green is the aceto-arsenite of copper. The theoretical composition is: As2Oa 58.55%, CuO 31.39%, acetic acid 11.84%. The commercial material, however, always con- tains more or less Na2SO4, but in good samples the amount should not be much over 1%. It is also sometimes adulterated with CaSO 4 . The following procedures are the methods of the Association of Official Agricultural Chemists: Preparation of Sample (Tentative). Mix thoroughly before analysis. Make water-soluble arsenic determinations on the sample as received without further pulverizing or drying. GENERAL INORGANIC ANALYSES 55 Moisture (Tentative). Dry 2 grams at 105-110 C. for five hours and report the loss as moisture. Total Arsenic (Official). (Arsenic present as arsenate is titrated as As2Os.) (A) REAGENTS. (a) Starch Indicator. Mix about 0.5 gram of finely powdered potato starch with cold water to a thin paste; pour into about 100 cc. of boiling water. (b) Standard Arsenious Oxide Solution. Dissolve 4 grams of pure As2O3 in a beaker by boiling with about 300-400 cc. of water containing 20 cc. of cone. H2SO4; cool, transfer to a liter graduated flask and dilute to the mark. (c) Standard Iodine Solution. Prepare an approximately 0.1 N solution as follows: Mix intimately 12.70 grams of pure iodine with twice its weight of pure KI. Dissolve in a small amount of water, filter and dilute the filtrate to 1 liter in a gra4- uated flask. Standardize against (6) above as follows: Pipette 50 cc. of the As2Os solution into an Erlenmeyer flask, dilute to about 400 cc., neutralize with NaHCOs, add 4-5 grams in excess, and add the standard iodine solution from a burette, shaking the flask continuously, until the yellow color disappears slowly from the solution, then add 5 cc. of the starch indicator and continue adding the iodine solution, drop by drop, until a permanent blue color is obtained. Calculate the value of the standard iodine solution in terms of As2Os and of As2O5. Occasionally restandard- ize the iodine against freshly prepared As2Os solution. (B) APPARATUS. The apparatus is shown in Fig. 3. The distillation flask rests on a metal gauze which fits over a circular hole in a heavy sheet of asbestos board. The first two Erlenmeyer flasks are of 500 and 1000 cc. capacity and contain about 40 and 100 cc. of water, respectively. Both of these flasks should be placed in a pan and kept surrounded with cracked ice and water. The third flask, containing a small amount of water, is used as a trap. (C) DETERMINATION. Weigh an amount of the sample equal to the As2Os equivalent of 250 cc. of the standard iodine solution, and wash into the distillation flask by means of 100 cc. of cone. HC1. Add 5 grams of cuprous chloride and distill. When the volume in the distillation flask is reduced to about 40 co., add 50 cc. of cone. HC1 by means of the dropping funnel and 56 TECHNICAL METHODS OF ANALYSIS continue distillation until 200 cc. of the acid distillate have passed over. Then wash down the condenser and all connecting tubes carefully, transfer these washings and contents of the three Erlen- meyer flasks to a liter graduated flask and dilute to the mark. Mix thoroughly, pipette 400 cc. into an Erlenmeyer flask and nearly neutralize with a saturated solution of NaOH or KOH, using a few drops of phenolphthalein indicator and keeping the solution well cooled. Continue as directed under Reagents (c), beginning with "neutralize with NaHCOs." The number of cc. of iodine used in FIG. 3. Apparatus for Determining Arsenic in Paris Green. this titration represents directly the total per cent of arsenic in the sample, expressed as As2O 3 . NOTE. In case the regular 0.1 N iodine solution of the laboratory is used, the amount can be calculated from 1 cc. 0.1 N iodine = 0.004948 gram Total Arsenious Oxide. (The following methods determine As, and Sb if present, as the -ous oxides, As20a and Sb2Oa, respec- tively. Ferrous and cuprous salts vitiate the results.) METHOD I. C. C. Hedges Method, Modified (Tentative). (a) Reagents. The reagents and solutions are the same as those described above under Total Arsenic. (6) Determination. Weigh an amount of the sample equal to the As2Os equivalent of 100 cc. of the standard iodine solution, wash into an Erlenmeyer flask with 10-15 cc. of dil. HC1 (1:1) followed by about 100 cc. of water, and heat on the steam bath to complete solution, at a temperature not exceeding 60 C. Cool, neutralize with NaHCO 3 , add 4-5 grams in excess, and then suf- GENERAL INORGANIC ANALYSES 57 ficient 25% NIrLtCl solution to dissolve the precipitated copper. Dilute somewhat and titrate as directed under (C) above. A correc- tion must be applied for the amount of iodine solution necessary to produce a blue color with starch in the presence of copper (using an equivalent weight of copper sulfate). The corrected number of cc. of the standard iodine solution used represents directly the per cent of As2Os in the sample. ' METHOD II. C. M. Smith Method, Modified (Tentative). Proceed as above using dil. H2&O4 (1:4) instead of dil. HC1. The solution in this case may be heated to boiling. Sodium Acetate-soluble Arsenious Oxide (Tentative). (A) REAGENTS. (a) Sodium Acetate Solution. Prepare a solu- tion containing 12.5 grams of NaC2Hs02 3H2O in each 25 cc. The other reagents are described under Total Arsenic. (B) DETERMINATION. Place 1 gram of the sample in a 100 cc. flask and boil for five minutes with 25 cc. of the sodium acetate solution. Dilute to the mark, shake, and pass through a dry filter paper. Titrate an aliquot of this filtrate as directed under (C) above. Calculate the amount of As2Os present and express the result as per cent of sodium acetate-soluble As20s. Water-soluble Arsenious Oxide (Tentative). (A) REAGENTS. Same as for Total Arsenic. (B) DETERMINATION. To 1 gram of the sample in a liter Florence flask add 1 liter of recently boiled water which has been cooled to exactly 32 C. Stopper the flask and place in a water bath kept at 32 C. by means of a thermostat. Digest for twenty- four hours, shaking hourly for eight hours during this period.* Filter through a dry filter and titrate 250 cc. of the filtrate as directed under (C) above. Correct for the amount of the standard iodine necessary to produce the same color, using the same reagents and volume. Calculate the amount of As20s present and express the result as per cent of water-soluble As2Os. Total CuO. (A) ELECTROLYTIC METHOD (OFFICIAL). Treat 2 grams of the sample in a beaker with 100 cc. of water and about 2 grams of NaOH and boil thoroughly until all Cu is precipitated as Cu2O. Filter, wash well with hot water, dissolve the precipitate in hot dil. HNOs, cool, transfer to a 250 cc. graduated flask and dilute to the mark. Use 50-100 cc. of this solution for the elec- * See note on page 48 under Total Water-soluble Matter in Lead Arsenate. 58 TECHNICAL METHODS OF ANALYSIS trolytic determination of Cu either with a rotating cathode and stationary anode in a beaker, or with a rotating spiral anode, using a weighed 150 cc. platinum dish for the cathode, and a current of about 3 amperes. After all the Cu is deposited, wash the deposit with water by siphoning, then rinse with alcohol, dry for a few minutes in an oven, weigh and calculate to per cent of CuO. CALCULATION. CuX 1.2517 - CuO. (B) THIOSULFATE METHOD (OFFICIAL). Determine Cu in another aliquot of the HNOs solution of Cu2O prepared as de- scribed above, by titrating with 0.1 N thiosulfate solution as fol- lows : After washing the precipitated Cu2O filtered on Gooch cru- cible, cover the Gooch with a watch glass and dissolve the oxide with 5 cc. of warm HNOs (1 : 1) poured under the watch glass with a pipette. Catch the filtrate in a 250 cc. flask, wash the watch glass and Gooch free of Cu, using about 50 cc. of water. Boil to expel red fumes, add 5 cc. of bromine water, boil off the Br, remove from the heat and add a slight excess of cone. NH 4 OH (about 7 cc. are required). Again boil until the excess of ammonia is expelled, as shown by a change of color of the liquid and partial precipita- tion. Then add a slight excess of 80% acetic acid (3 or 4 cc.) and boil for one minute. Cool to room temperature and add 10 cc. of 30% KI solution. Titrate at once with thiosulfate solution until the brown tinge has become weak, then add sufficient starch indicator (see reagents under Total Arsenic) to produce a marked blue coloration. Continue the titration cautiously until the color due to free iodine has entirely vanished. The blue color changes toward the end to a faint lilac. If at this point the thiosulfate be added drop by drop and a little time allowed for complete reaction after each addition, there is no difficulty in determining the end point within a single drop. One cc. of the thiosulfate solution will be found to correspond to about 0.0064 gram of Cu. Calculate to per cent CuO. NOTE. The thiosulfate solution should be standardized against pure copper foil. Weigh out accurately about 0.2 gram; dissolve in a 250 cc. flask by warming with 5 cc. of a mixture of equal parts of cone. HNOs and water. Dilute to 50 cc., boil to expel red fumes, add 5 cc. of bromine water and proceed as above. REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 63. GENERAL INORGANIC ANALYSES 59 LIME SULFUR SOLUTION General. Lime sulfur solutions are made by boiling, either by direct heat or by live steam, sulfur and freshly slaked lime in water. The resulting solution contains considerable amounts of poly- sulfides and thiosulfate and very small amounts of sulfate and sul- fite. In commercial practice, 1 part of lime and 2 parts of sulfur are used. Lime comparatively free from magnesia must be employed. The sample for analysis should be kept tightly stoppered and not exposed to air, as it will very easily precipitate free sulfur. The determinations below marked " Official " are those of the Association of Official Agricultural Chemists. Specific Gravity. Determine the sp. gr. with a Westphal balance at 15.5 C. (60 F.) and calculate to degrees Baume; or if there is sufficient sample at hand, determine the gravity directly with a Baume hydrometer. Total Sulfur (Official). Measure and accurately weigh from a stoppered weighing bottle about 10 cc. of the clear sample. Trans- fer to a 250 cc. graduated flask, partly filled with recently boiled and cooled distilled water, and dilute to the mark with the same water. For the determination of total sulfur and for other deter- minations (unless otherwise directed) use 10 cc. aliquots of this solution. Transfer a 10 cc. aliquot to a 400 cc. beaker, add about 3 grams of Na2O2, cover immediately with a watch glass and warm on the steam bath, with frequent shaking, until all S is oxidized to sulfate, adding more peroxide if necessary. Dilute, acidify with HC1, evaporate to dryness, treat with water acidified with HC1, boil, and filter to remove silica, if present. Dilute the fil- trate to 300 cc., add 50 cc. of cone. HC1, heat to boiling and pre- cipitate with excess of 10% BaCl2 solution slowly and stirring constantly. (The rate is best regulated by attaching a suitable capillary tip to a burette containing the BaCl2 solution.) Evapo- rate to dryness on the steam bath, take up with hot water, filter through a quantitative filter paper, wash until free from Cl and ignite to constant weight over a Bunsen burner. Weigh as BaSO4 and calculate to sulfur. Previous to use, test all reagents for sulfur and, if present, make corrections accordingly. CALCULATION. BaSO 4 X 0.1373 = Sulfur. 60 TECHNICAL METHODS OF ANALYSIS Sulfur as Polysulfides. Pipette 10 cc. of the solution de- scribed above into a small beaker and add about 30 cc. of water. Then run into the solution from a burette 0.1 N HC1, drop by drop, with constant stirring, until the yellow tint has practically disap- peared. Add 2 drops of methyl orange and continue the addition of acid until the first permanent pink color appears. (After stand- ing for some time this tint will gradually disappear.) The solu- tion will be milky white from finely divided S present, but it is not difficult to ascertain the exact end point of the reaction. Let the solution stand a few moments to permit S to settle and collect together. Filter on a weighed Gooch crucible, wash thor- oughly, dry at about 40 C. and weigh directly as free S. This represents S combined in the form of polysulfides. NOTES. (1) Precipitation of S by means of weak acid does not decom- pose thiosulfate in solution. (2) An optional method of determining sulfur is as follows: Filter on a small filter paper instead of on a Gooch crucible, and after thoroughly washing, gently boil the paper and contents in 50 cc. of 10% KOH solution until all sulfur is dissolved. After cooling add 50 cc. of a 3% solution of H 2 O 2 , free from sulfates. Heat on the steam bath for exactly thirty minutes and then acidify with HC1 and precipitate with BaCl 2 in the usual manner. Finally weigh as BaSO 4 . Run a "blank" on the KOH solution and subtract any sulfur found. Sulfur as Sulfide (Official). Dilute 10 cc. of the solution pre- pared as for Total Sulfur to about 100 cc. and add ammoniacal ZnCU solution (see notes) until the sulfide is all precipitated, which will be shown by adding a drop of the clear solution to a few drops of nickel sulf ate solution ; if any sulfide remains, this will cause a black precipitate. Filter at once and thoroughly wash the pre- cipitate with cold water. Transfer the filter paper and precipitate to a beaker; cover with water, disintegrate with a glass rod and add about 3 grams of Na2C>2, keeping the beaker well covered with a watch glass. Warm on a steam bath with frequent shaking until all S is oxidized to sulfate, adding more Na 2 O 2 if necessary. Make slightly acid with HC1, filter to remove shreds of filter paper, wash thoroughly with hot water and determine sulfur in the filtrate exactly as under Total Sulfur above. This gives both monosulfide and polysulfide sulfur. Subtract the polysulfide sulfur as previously determined and report the difference as (mono) sulfide sulfur. GENERAL INORGANIC ANALYSES 61 NOTES. (1) Ammoniacal zinc chloride solution Dissolve 50 grams of pure ZnCl 2 in water. Add sufficient NH 4 OH to redissolve the precipitate first formed; then add 50 grams of NH^Cl and dilute to 1 liter. (2) Blank determinations of the amount of S in the reagents used should be made and corrections applied. (3) The amount of sulfur as polysulfide is almost always quite close to 3.5 times the amount of sulfur as sulfide. This shows that the polysulfide in solution is not one compound only, but probably a mixture of CaS 4 and CaS 6 . Sulfur as Thiosulfate (Official). Dilute 50 cc. of the solution prepared as for Total Sulfur to about 100 cc. in a 250 cc. grad- uated flask. Add ammoniacal ZnCU solution until in slight excess and make up to the mark. Shake thoroughly and filter through a dry filter. To 200 cc. of the filtrate add methyl orange and exactly neutralize with 0.1 N HC1. Titrate this neutral solution with 0.1 N iodine, using a few drops of starch paste as indicator. From the amount of iodine solution required calculate the sulfur present as thiosulfate. CALCULATION. 1 cc. 0.1 N iodine = 0.0064 12 gram sulfur. = 0.01581 gram Na 2 S 2 3 . Sulfur as Sulfate (Official). To the solution from the deter- mination of thiosulfate add 2 or 3 drops of HC1. Precipitate cold with 10% BaCl2 solution and let stand in the cold overnight. Filter, ignite and weigh as BaSOi. From this weight calculate the sulfur and report as sulfate sulfur. NOTE. In case any sulfite is present, the determination of thiosulfate will be too high. As calcium sulfite is nearly insoluble, however, it will not be present in more than traces and the error from this cause will be negligible. It is unusual for the combined sulfate and sulfite to amount to more than a small fraction of 1%. Total Lime (Official). To 25 cc. of the solution prepared as under Total Sulfur add 10 cc. of cone. HC1; evaporate to dry- ness on the steam bath, treat with water and a little HC1, warm until all the CaCl2 is dissolved, and filter from S and any 8162 that may be present. Oxidize the filtrate by boiling with a little cone. HNOs; make ammoniacal, filter from Fe and Al if present; heat to boiling and precipitate the Ca with (NH4)2C2O4 solution. Filter, wash and ignite over a blast lamp to constant weight. Weigh as CaO. (The calcium oxalate may be titrated with KMnO4 in the usual way, if desired, instead of igniting it.) REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 76. 62 TECHNICAL METHODS OF ANALYSIS CORROSIVE SUBLIMATE IN MEDICATED GAUZE Weigh 30 grams of gauze, or the whole sample if less than this, and place in a 200 cc. separatory funnel; pack firmly, and add 200 cc. of warm dil. HC1 (15 cc. of cone. HC1 per liter). Let the acid drain slowly into a 1000 cc. beaker, about 2 or 3 drops per second. When completely drained, add 100 cc. more of the warm dil. HC1, and continue the draining operation until complete. Repeat 5 times, so that the gauze is washed with a total of 800 cc. of acid. Pass H2S through the solution for one hour and let stand over- night. Filter through a tared filter or Gooch crucible. Wash 6 times with water, and then 3 times with 95% alcohol. Stopper the bottom of the funnel with a small cork; add CS2 sufficient to cover the precipitate of HgS completely, and let stand for 0.5 hour. Remove the stopper and let drain. Wash once with C$2, and 3 times with alcohol. Dry and weigh. In case a Gooch crucible is usedj set the crucible in a beaker containing CS 2 sufficient to cover the precipitate. Let stand 0.5 hour, etc. Calculate parts of HgCl2 per thousand parts of gauze. ' WX 1.167X100 CALCULATION. ^ = parts HgCb per thousand, o where W weight of HgS, and S = weight of gauze. NOTE. This is the method prescribed by the U. S. Government for testing medicated surgical dressings. ASBESTOS MAGNESIA PIPE COVERING General. This material consists of long-fibered asbestos mixed with magnesium carbonate. The specifications on which it is gen- erally purchased are as follows : Long-fibered asbestos not less than 10%. Magnesium carbonate not less than 85%. The magnesium carbonate is calculated to the empirical for- mula Mg(OH) 2 -4MgCO 3 -5H 2 O. Since the asbestos and magnesium carbonate do not adhere very strongly to each other it is necessary to use considerable care GENERAL INORGANIC ANALYSES 63 in sampling this material. A considerable number of representa- tive lumps should be taken for the analysis and these lumps should not be shaken nor squeezed on account of danger of loss of carbonate. If it is attempted to grind up the whole sample and quarter it down, some magnesium carbonate is very likely to be lost. Asbestos. Weigh out 5-10 grams into a beaker and treat with excess of hot dil. acetic acid (1:4) until there is no further effer- vescence. Filter off the asbestos and wash thoroughly with hot water. Dry the residue of asbestos at 105 C., ignite in a platinum crucible and weigh directly as asbestos. Magnesium Carbonate. Weigh out 5-10 grams into a beaker and treat with excess of hot dil. HC1 until no further effervescence occurs. Filter into a 500 cc. or 1000 cc. graduated flask and wash thoroughly with hot water. Cool, and dilute to the mark. Pipette an aliquot of this solution, representing about 0.5 gram of the original substance, into a beaker, dilute to about 100 cc., add a slight excess of NHiOH and boil. Then add, without filter- ing, 5 cc. of (NH4)2C204 solution and again boil. Let settle till clear, and then filter. If there is a considerable amount of calcium or of iron and alumina, dissolve the precipitate in HC1 and repre- cipitate with NILtOH and (NH4)2C2O4 as before. Make the final filtrate slightly acid and concentrate until crystallization begins. Then cool and dilute until the crystals go into solution. Add a strong excess of NFLt or Na phosphate and stir thoroughly. Then add about one-third the volume of cone. NELtOH. Let stand several hours, filter through a Gooch crucible, ignite slowly at first and then strongly to constant weight and weigh as Mg2P2Oj. Calculate to magnesium carbonate. CALCULATION. Mg 2 P 2 O 7 X 0.8724 = Mg(OH) 2 4MgC0 3 5H 2 O. NOTES. (1) Pure asbestos is a magnesium silicate. The natural product, however, also contains more or less Fe, Al, and SiO 2 . If in the determination of asbestos, therefore, HC1 were used instead of acetic acid more of these con- stituents would dissolve and the per cent of insoluble matter would generally be considerably less. (2) If the amount of sample is limited, the filtrate from the acetic acid treatment may be made up to volume and used for the magnesia determina- tion. In this case add 25 cc. of 10% NH 4 C1 solution before adding the phos- phate. CHAPTER III GENERAL ORGANIC ANALYSES NITROGEN General. The principle of the determination of nitrogen in organic materials and in fertilizers is its conversion into NHs and a determination of the amount of NHs so formed. The method to be employed depends upon whether or not nitrates are present. In every case a " blank " should be carried out to correct for the presence of small amounts of nitrogen in the reagents employed. In running the blank, employ the same amount of each reagent as is used in the determination. In this laboratory we have found the Gunning methods prefer- able to the Kjeldahl for most substances. (See general notes on page 66.) METHODS TO BE USED IN THE ABSENCE OF NITRATES Kjeldahl Method.* Place 0.7-3.5 grams of the substance to be analyzed (depending upon the N content) in a 500 cc. pear- shaped digestion flask. Add approximately 0.7 gram of mercuric oxide, f or its equivalent of metallic Hg, and 20-30 cc. of cone. H2S04. From 0.1 to 0.3 gram of crystallized CuS04 may also be used in addition to the Hg or in place of it. Place the flask in an inclined position and heat gently below the boiling point of the acid for five to fifteen minutes, i.e., until frothing has ceased. (A small piece of paraffin may be added to prevent extreme foaming.) Then raise the heat until the acid boils briskly and digest for four hours after the mixture is colorless, or nearly so. Remove the * For ordinary work 0.5 N acid is recommended. For work in deter- mining very small amounts of nitrogen 0.1 N acid is recommended. t If mercuric oxide is used, it should be prepared in the wet way but not from mercuric nitrate. 64 GENERAL ORGANIC ANALYSES 65 flask from the flame, hold it upright, and while still hot drop in KMn(>4 (finely pulverized) in small quantities until, after shaking, the liquid remains green or purple. After cooling, dilute with about 200 cc. of distilled water, add a few pieces of granulated zinc or pumice stone and 25 cc. of K 2 S solution (40 grams of commercial K2S in 1 liter of water) wijbh shaking. Next add 50 cc. of a saturated solution of NaOH (free from N compounds) or sufficient to make the solution strongly alkaline, pouring it down the side of the flask so that it does not mix at once with the acid solution (50 cc. are usually enough). Connect the flask with the condenser before heating, mix the contents well by swirling and then distill until all the NHs has passed over, collecting the distillate in an excess of standard acid. As a general rule, 50 cc. of 0.1 N HC1 or 10 cc. of 0.5 N will be suf- ficient. It should have cochineal, methyl orange, or methyl red indi- cator (see general note 3) added to it before the distillation starts; and if the color changes before it is completed, the determination should be repeated, using either less of the original material or more of the standard acid. The first 150 cc. of the distillate will generally contain all the NHs. The distillation usually requires forty to ninety minutes. Wash down the condenser with distilled water and titrate the excess of standard acid with 0.1 N alkali. In most cases the use of KMnCX is quite unnecessary but it is believed that in exceptional cases it is required for complete oxidation and in view of the uncertainty it is always used. The K 2 S removes all the Hg from the solution and thus prevents the formation of mercuro-ammonium compounds, which are not completely decomposed by the caustic. The addition of Zn gives rise to an evolution of hydrogen and prevents violent bumping. Gunning Method. The apparatus used is the same as that employed in the Kjeldahl method. Place the substance to be analyzed in a 500 cc. digestion flask, using 0.7-3.5 grams according to the proportion of N. Add 10 grams of powdered K 2 SO 4 or 7.5 grams of dry powdered Na 2 SO 4 (use Baker's c. P. special, free from nitrogen) and 25 cc. of cone. [2804. Add also about 0.2 gram of crystallized CuS0 4 or 0.1 gram of copper wire. Conduct the digestion exactly as in the Kjeldahl .process, starting with a temperature below boiling point and increasing the heat gradually until frothing ceases. Digest 66 TECHNICAL METHODS OF ANALYSIS for four hours after the mixture is colorless, or nearly so. Do not add either KMnC>4 or K2S. Cool, dilute, add an excess of NaOH, distill and titrate as in the Kjeldahl method. In neutralizing, it is advisable to add a few drops of methyl orange or litmus indicator by which one can tell when an excess of NaOH has been added. ' NOTES. (1) It is well for convenience to use the same amount of H 2 SO 4 each time for digestion and to determine how much of the strong NaOH is necessary to neutralize this amount of acid, marking the amount on the bottle containing the NaOH. (2) If copper wire of the same size is always used, it can easily be deter- mined how long a piece will weigh 0.1 g. and then cut up a number of pieces of the proper length. The use of NaaSO4 and copper wire has not yet been made an official method of the A. O. A. C. but sufficient work has been done to show that it gives accurate results. Kjeldahl-Gunning-Arnold Method. Place 0.7-3.5 grams of the material (according to the N content) in the digestion flask. Add 15-18 grams of powdered K 2 S0 4 (or 10-12 grams Na 2 SO 4 ), 1 gram of CuSO4, 1 gram of HgO (or its equivalent of metallic Hg), and 25 cc. of cone. H2S04. Heat gently until frothing ceases, then boil briskly and continue digestion for at least two hours after oxidation is complete; cool, dilute with about 200 cc. of water, add 50 cc. of K^S solution, make strongly alkaline with NaOH solution and complete the distillation as under the Kjeldahl method. GENERAL NOTES. (1) A blank determination should be run, using all reagents in the same amount as in the regular determination. (2) We have found the Gunning methods the most convenient for use but either method gives accurate results, and the Kjeldahl is quicker. (3) It is generally recommended that cochineal or methyl red be used in titrating back the excess of acid, but accurate results can be obtained by using methyl orange as follows: To a container of similar size and shape to that containing the distillate, add an amount of distilled water equal to the volume of the distillate and add the same number of drops of indicator (two or three are sufficient) as were added to the standard acid. Then titrate until the color matches the color of the distilled water containing the methyl orange. (The titration of the blank on the reagents should, of course, be carried to the same end point.) (4) With certain materials it will be necessary to add considerably more H 2 SO 4 than the amounts given above. In this case, care must be taken that the mixture in the flask is alkaline before the distillation is started. (5) Only a moderate excess of the NaOH solution should be added since considerable excess often causes frothing. GENERAL ORGANIC ANALYSES 67 METHODS TO BE USED IN THE PRESENCE OF NITRATES The results obtained by these methods give total nitrogen and include the nitrogen of the nitrates. Modified Kjeldahl Method. Place 0.7-3.5 grams of the substance in a Kjeldahl digestion flask: (1) Add 30 cc. of cone. H2SO4 containing 1 gram of salicylic acid. Shake until thor- oughly mixed, let stand for at least thirty minutes and then add 5 grams of crystallized sodium thiosulfate. Or, (2) add to the substance 30 cc. of cone. H2S04 containing 2 grams of salicylic acid, let stand at least thirty minutes and then add gradually 2 grams of zinc dust, shaking the contents of the flask at the same time. The zinc dust should be an impalpable powder. Gran- ulated Zn or Zn borings will not answer. Place the flask on the stand for holding digestion flasks and heat over a low flame until all danger from frothing has passed. Then raise the heat until the acid boils briskly and continue boiling until white fumes no longer escape from the flask. This requires about five to ten minutes. Then add approximately 0.7 gram of HgO or its equivalent in metallic Hg. Continue the boiling until the liquid in the flask is colorless or nearly so. In case the contents of the flask are likely to become solid before this point is reached, add 10 cc. more of cone. H2SO4. Complete the oxida- tion with a little KMnC>4 in the usual way and proceed with the distillation as described in the Kjeldahl method. A blank should be run on all the reagents employed. Modified Gunning Method. In a 500 cc. digestion flask place 0.7-3.5 grams of the substance to be analyzed, add 30 cc. of cone. H2SO4 containing 1 gram of salicylic acid dissolved in it, shake until thoroughly mixed and let stand for at least thirty minutes with frequent shaking. Add 5 grams of sodium thio- sulfate and heat the solution for five minutes. Cool, add 10 grams of powdered K 2 SO 4 (Baker's c. P. special) and heat very gently until foaming ceases and then strongly until nearly colorless. Dilute, neutralize and distill as in the Gunning method. Absolute or Cuprous Oxide Method. By this method the nitrogen is all set free as such and measured in an azotometer. It is not ordinarily employed in commercial analysis but is described 68 TECHNICAL METHODS OF ANALYSIS in the Journal of the Association of Official Agricultural Chemists, Methods of Analysis (1916), page 8. AMMONIACAL NITROGEN MgO Method. Place 0.7-3.5 grams of the material, according to the NHs content, in a distillation flask with about 200 cc. of water and 5 grams or more of MgO, free from carbonates. Then connect the flask with a condenser and distill 100 cc. of the liquid into a measured quantity of standard acid and titrate the excess of acid with standard alkali. NITRIC AND AMMONIACAL NITROGEN Ulsch-Street Method. Place 1 gram of the sample in a 500 cc. flask, add about 30 cc. of water and 2-3 grams of reduced iron and, after standing sufficiently long to insure solution of the sol- uble nitrates and ammonium salts, add 10 cc. of a mixture of cone. H2SO4 with an equal volume of water. Shake thoroughly, place a long-stemmed funnel in the neck of the flask to prevent mechanical loss and let stand for a short time until the violence of the reaction has moderated. Heat the solution slowly; then boil for five min- utes and cool. Add about 100 cc. of water, a little paraffin, and 7-10 grams of MgO, free or nearly free from carbonates. Connect with a condenser such as is used in the Kjeldahl method and boil the mixture for forty minutes nearly to dryness, collecting the NHs in a measured quantity of standard acid, and titrate the excess with standard alkali in the usual way. The nitrogen ob- tained represents the nitrates plus the ammonium salts contained in the sample. In the analysis of nitrate salts proceed as above, except that 25 cc. of the nitrate solution (equivalent to 0.25 gram of the sample) are employed with 5 grams of reduced iron. After boiling, add 75 cc. of water and an excess of NaOH solution and complete the determination as above. Zinc-Iron Method. Dissolve 10 grams of the sample in water and dilute to 500 cc. Place 25 cc. of this solution, corre- sponding to 0.5 gram of the substance, in a 400 cc. distillation flask. Add 120 cc. of water, 5 grams of well-washed and dried zinc dust, GENERAL ORGANIC ANALYSES 69 and 5 grams of reduced iron. To the solution add 80 cc. of sat- urated NaOH solution, connect the flask with the condenser and conduct the distillation simultaneously with the reduction, col- lecting the NHa in excess of standard acid. Continue the dis- tillation until 100 cc. have been distilled and titrate the excess acid with standard alkali. NITROGEN IN NITRATES Ferrous Sulfate-Zinc-Soda Method (Tentative). Place 0.5 gram of the nitrate salt in a 600-700 cc. flask, add 200 cc. of water, 5 grams of powdered zinc, 1-2 grams of ferrous sulfate and 50 cc. of 30% NaOH solution. Connect with the distilling apparatus and distill, collecting the distillate in the usual way in 0.1 N acid and titrating the excess with standard alkali. METHODS FOR "AVAILABLE ORGANIC NITROGEN" Organic Nitrogen Soluble in Neutral Permanganate. As a preliminary test for the determination of water-insoluble organic N, place 1 gram of the material on an 11 cm. filter paper and wash with water at room temperature until the filtrate measures 250 cc. Dry and determine the N in the residue by the Kjeldahl or Gun- ning method, making a correction for the N of the filter if necessary. Place a quantity of the material equivalent to 0.050 gram of water-insoluble organic N as determined above on a moistened 11 cm. filter paper and wash with water at room temperature until the filtrate measures 250 cc. Transfer the insoluble residue with 25 cc. of tepid water to a 300 cc. Griffin low-form beaker. Add 1 gram of Na 2 CO 3 ; mix and add 100 cc. of 2% KMnO 4 solu- tion; cover with a watch glass and immerse for thirty minutes in a steam or hot-water bath, keeping the level of the liquid in the beaker below that of the water in the bath. Stir twice at inter- vals of ten minutes. At the end of the digestion, remove from the bath, add immediately 100 cc. of cold water and filter through a heavy 15 cm. folded filter. Wash with small quantities of cold water until the filtrate measures about 400 cc. Determine N in the residue and filter paper by the Kjeldahl or Gunning method, 70 TECHNICAL METHODS OF ANALYSIS correcting for N contained in the paper. The N thus obtained is the inactive water-soluble organic N. Subtract this result from the total water-insoluble organic N to obtain the percentage of organic N soluble in neutral permanganate. Organic Nitrogen Soluble in Alkaline Permanganate.* (A) PKEPARATION OF SAMPLE. /. Mixed Fertilizers. Place an amount equivalent to 0.050 gram of water-insoluble organic N determined as above on a filter paper and wash with water at room temperature until the filtrate measures 250 cc. II. Raw Materials. Place an amount of material equivalent to 0.050 gram of water-insoluble organic N, determined as above, in a small mortar. Add about 2 grams of powdered rock phos- phate, mix thoroughly, transfer to a filter paper and wash with water at room temperature until the filtrate measures 250 cc. When much fat or oil is present, it is well to wash with ether before extracting with water. (B) DETERMINATION. Dry the residue from the water extract above at a temperature not exceeding 80 C. and transfer from the filter to a 500-600 cc. Kjeldahl distilling flask. Add 20 cc. of water, 15-20 small glass beads or fragments of pumice stone, a piece of paraffin the size of a pea and 100 cc. of alkaline permanganate solution (25 grams of pure KMnCX and 150 grams of NaOH separately dissolved in water, the solutions cooled, mixed and made to a volume of 1 liter). Connect with an upright condenser, to the lower. end of which a receiver containing standard acid has been attached. Digest slowly for at least thirty minutes below the distillation point with a very low flame, using coarse wire gauze and asbestos paper between the flask and flame. Gradually raise the temperature and after any danger from frothing has passed, distill until 95 cc. of the distillate are obtained, and titrate as usual. When a tendency to froth is noticed, lengthen the digestion period and no trouble will be experienced when the distillation is begun. During the digestion gently rotate the flask occasionally, particu- larly if the material shows a tendency to adhere to the sides. The N thus obtained is the active water-insoluble organic N. CALCULATIONS. The following factors will be found useful in calculating: * This is not applicable to fertilizers containing cottonseed meal or castor pomace. GENERAL ORGANIC ANALYSES 71 1 cc. of 0.1 N acid = 0.001401 gram N. = 0.001703 gram NH 3 . = 0.01011 gram KNO 3 . = 0.002604 gram (NH 4 ) 2 0. = 0.008937 gram casein (NX 6.38). = 0.007844 gram glue (NX 5.6). = 0.008755 gram protein (NX 6.25). = 0.007985 gram protein (NX5.7).* REFERENCES. The above methods are all official methods of the Associa- tion of Official Agricultural Chemists unless otherwise indicated, except the Zinc-Ferrous Sulfate-Soda method for Nitrates, which is tentative. See Journal of Association of Official Agricultural Chemists, Methods of Analysis (1916), pages 5-12. See also Journal of American Leather Chemists' Asso- ciation 11, 454 (1916). METHYL ALCOHOL General. This method applies to the analysis of methyl alcohol to be used for denaturing ethyl alcohol. The methyl alcohol submitted must be partially purified wood alcohol obtained by the destructive distillation of wood. It must conform to the following analytical requirements: Color. It shall not be darker .than the color produced by a freshly prepared solution of 2 cc. of 0.1 N iodine diluted to 1000 cc. with distilled water. Specific Gravity. It must have a sp. gr. of not more than 0.830 at 60 F. (15.56 C.), corresponding to 91 of Tralles' Scale. Boiling Point. 100 cc. slowly heated in a flask under conditions as described below must give a distillate of not less than 90 cc. at a temperature not exceeding 75 C. at normal barometric pres- sure (760 mm.). 100 cc. of wood spirit are run into a short-necked copper flask of about 180-200 cc. capacity and the flask placed on an asbestos plate having a circular opening of 30 mm. diameter. In the neck of this flask is fitted a fractionating tube 12 mm. wide and 170 mm. long, with a bulb just 1 cm. below the side tube, which is con- nected with a Liebig's condenser having a water jacket not less than 400 mm. long. In the upper opening of the fractionating * In wheat products. 72 TECHNICAL METHODS OF ANALYSIS tube is placed a standardized thermometer, so adjusted that its mercury bulb comes in the center of the bulb. The distillation is conducted in such a manner that 5 cc. pass over in one minute. The distillate is run into a graduated cylinder, and when the tem- perature of 75 C. has been reached at the normal barometric pressure of 760 mm. at least 90 cc. shall have been collected. Should the barometer vary from 760 mm. during the distilla- tion, 1 C. shall be allowed for every variation of 30 mm. For example, at 770 mm. 90 cc. should have distilled at 75.3 C. and at 750 mm. 90 cc. should have distilled at 74.7 C. Miscibility with Water. It must give a clear or only slightly opalescent solution when mixed with twice its volume of water. Acetone Content. It must contain not more than 20 nor less than 10 grams per 100 cc. of acetone and other substances esti- mated as acetone when tested by the following (Messinger) method : With a standardized pipette measure 10 cc. of the sample into a 500 cc. glass-stoppered flask and make up to the mark with dis- tilled water. Pipette out 5 cc. of this solution (with standardized pipette) and treat with 10 cc. of 2 N NaOH solution. Then add with shaking 50 cc. of 0.1 N iodine solution and make the mixture acid with dil. H2S04 three minutes after the addition of the iodine. Titrate back the excess of iodine with 0.1 N sodium thiosulfate solution, using a few drops of starch solution as indicator. From 10.3 to 20.7 cc. of 0.1 N iodine solution should be used by the spirit. IMPORTANT. The solution must be kept at a temperature between 15 and 20 C. CALCULATION. Let X = grams of acetone in 100 cc. of spirit, Y = number of cc. of 0. 1 N iodine solution required, and N = volume of spirit taken for titration ; 7X0.096672 then, X = - .* Blank Correction. -T. D. 2268 (Dec. 4, 1915) recommends that a blank correction be made as follows : * This is the figure given in Regulations No. 30, Revised, United States Internal Revenue. Using 1920 atomic weights the figure becomes 0.096772. GENERAL ORGANIC ANALYSES 73 Weigh accurately from a weighing bottle about 16 grams of Kahlbaum's c. P. acetone into a standardized 100 cc. graduated flask partially filled with Kahlbaum's c. P. methyl alcohol. Make up accurately to the mark with more of the methyl alcohol. Determine the amount of acetone in this solution, following the method above described. If less acetone is found than was added, add the difference as the blank correction to the acetone found in the sample of denaturant. If more acetone is found in the blank subtract the difference from that found in the denaturant. Sep- arate blanks should be run with each sample of denaturant, taking care that all the conditions are kept the same. Esters. It should contain not more than 5 grams of esters per 100 cc. of spirit, calculated as methyl acetate and determined as follows : Five cc. of wood spirit are run into a flask and 10 cc. of N NaOH free from carbonates are added, the flask connected with a return condenser and boiled for two hours. Instead of digesting at boiling temperature the flask may be allowed to stand over- night at room temperature and then heated on a steam bath for thirty minutes with an ordinary tube condenser. The liquid after digestion is cooled and the excess NaOH titrated with N H2SO4 and phenolphthalein. CALCULATION. Methyl acetate (grams per 100 cc. of spirit) _ 0.074 Xcc. of N NaOH consumed X 100 cc. spirit taken = 1.48 Xcc. of N NaOH consumed. Bromine Absorption. It must contain a sufficient quantity of impurities derived from the wood so that not more than 25 cc. nor less than 15 cc. shall be required to decolorize a standard solution containing 0.5 gram of bromine, as follows: The standard Br solution is made by dissolving 12.406 grams of KBr and 3.481 grams of KBrOs (which is of tested purity and has been dried for two hours at 100 C.) in a liter of water. 50 cc. of the standard solution, containing 0.5 gram of Br, are placed in a glass-stoppered flask having a capacity of about 200 cc. This is acidified with 10 cc. of H2SO4 (1 : 4), the whole shaken and allowed to stand a few minutes. The wood alcohol is then 74 TECHNICAL METHODS OF ANALYSIS allowed to flow slowly into the mixture, drop by drop, from a burette until the color is entirely discharged. The rate of flow through the burette shall not exceed 5 cc. per minute. The tem- perature of the mixture should be 20 C. In addition to the above requirements the methyl alcohol must be of such a character as to render the ethyl alcohol with which it is mixed unfit for use as a beverage. REFERENCES. Regulations No. 30 Revised, United States Internal Revenue, July 15, 1907. See also, T. D. 2779. GRAIN ALCOHOL OR COLOGNE SPIRITS General. The sample should be clear and colorless and have the characteristic odor of ethyl alcohol. Medicinal alcohol should pass all the qualitative tests of the U. S. Pharmacopoeia. Specific Gravity. Determine the sp. gr. at 60 F. with a West- phal balance or with a pycnometer. The temperature must be exactly 60 F. (15.56 -C.). Ethyl Alcohol. Calculate the per cent of ethyl alcohol both by volume and by weight from the sp. gr. The tables for these cal- culations wilL be found in Leach's Food Inspection and Analysis (3d edition), pages 661-374 (4th edition), pages 690-703; also in Van Nostrand's Chemical Annual. Proof. To obtain the proof multiply by 2 the per cent of ethyl alcohol by volume. Non-volatile Residue. Evaporate 100 cc. in a weighed plat- inum dish nearly to dryness on the water bath, then transfer to a water oven and dry at the temperature of boiling water for 2.5 hours. Cool in desiccator and weigh. To obtain the per- centage of non-volatile residue divide the weight obtained by the sp. gr. of the sample. Acidity, Calculated as Acetic Acid. Titrate 100 cc. of the sample with 0.1 N alkali and phenolphthalein, until a permanent pink color is formed. CALCULATION. 1 cc. 0.1 N alkali = 0.0060 gram acetic acid. Esters, Calculated as Ethyl Acetate. Dilute 250 cc. of the sample with 30 cc. of water and distill slowly into a graduated 250 cc. flask until nearly filled to the mark. Make up to volume GENERAL ORGANIC ANALYSES 75 with water. Shake and mix thoroughly and use aliquot portions of this solution (solution A) for the determination of esters, alde- hydes and furfural. Pipette 50 cc. of solution A into a 300 cc. Erlenmeyer flask and exactly neutralize with 0.1 N alkali and phenolphthalein. Then add 25-50 cc. excess of the 0.1 N alkali, accurately measured. Either boil for one hour with a reflux condenser or let stand over- night in the stoppered flask and heat with a reflux condenser for one-half hour below the boiling point. Cool, and titrate with 0.1 N acid and phenolphthalein. Multiply the number of cc. of 0.1 N alkali consumed in the saponification by 0.0088. The result is the weight in grams of the esters calculated as ethyl acetate. Divide this weight by the sp. gr. of the sample and mul- tiply by 2 to obtain the percentage. Aldehydes (Qualitative AgNO 3 Test.) Make up a solution of the following: 3 grams c. P. AgNOs, 3 grams c. P. NaOH, 20 cc. cone. NH4OH. Dissolve the AgNOs in a little water in a 100 cc. flask, add the NELtOH and then the NaOH and then make up to 100 cc. Dilute 10 cc. of the sample with 10 cc. of water and place in a glass-stop- pered bottle, adding 1 cc. of the above alkaline AgNOs solution. Let stand for one hour in the dark and filter immediately. Test the filtrate for Ag by adding an excess of HNOs and then a few drops of HC1. If this produces a precipitate of AgCl (showing unreduced Ag salts), the alcohol contains less than the maximum allowable amount of aldehyde. NOTE. This test is copied from Government specification, January 20, 1907. Aldehydes (Quantitative). SOLUTIONS REQUIRED (a) Alco- hol Free from Aldehydes. Place 1500 cc. of ordinary 95% alcohol in a 2 liter distilling flask, add about 25 grams of NaOH (or KOH) and distill down to about 100 cc. Add to the distillate 2.5 grams of meta-phenylenediamine hydrochloride, stopper and let stand for several days (or place on the steam bath in a large flask and reflux for several hours). Then distill slowly, rejecting the first 100 cc. and the last 200 cc. (6) Sulfite-Fuchsin Solution. Dissolve 0.50 gram of pure fuchsin in 500 cc. of water, then add 5 grams of SO2 dissolved 76 TECHNICAL METHODS OF ANALYSIS in water.* Make up to a liter and let stand until colorless. Pre- pare the solution in small quantities as it retains its strength for only a few days. (c) Standard Acetaldehyde Solution. Grind about 5 grams of aldehyde ammonia in a mortar with ether and decant the ether. Repeat this operation several times, then dry the purified salt in a current of air and finally in a vacuum desiccator over H 2 SO 4 . Weigh out 1.386 grams of this purified aldehyde ammonia and dis- solve in 50 cc. of 95% alcohol. To this add 22.7 cc. of N alcoholic H2S04 (a solution of 95% alcohol containing 49.04 grams of H2SO4 per liter). Then make up to 100 cc. with 95 per cent alcohol and add 0.8 cc. extra to compensate for the volume of (NEL^SCU precipitated. Let this stand overnight and filter. This solu- tion contains 1 gram of acetaldehyde in 100 cc. and will retain its strength. The standard found most convenient for use is 2 cc. of this strong aldehyde solution diluted to 100 cc. with 50% alcohol (by volume). One cc. of this solution = 0.0002 gram of acetaldehyde. This solution should be made up fresh, as it loses its strength in a day or two. PROCEDURE. Determine the aldehyde in solution A as follows: Dilute 10 cc. to 50 cc. with aldehyde-free alcohol (50% by vol- ume). Add 25 cc. of the fuchsin solution and let stand fifteen minutes at 15 C. The solutions and the reagents should all be at 15 C. before they are mixed. Compare standards of known strength (prepared from the standard acetaldehyde solution) in the same way and match them colorimetrically with the sample. Calculate the percentage of aldehydes in the original sample (10 cc. of Solution A = 10 cc. of original sample). Furfural. Dilute 20 cc. of solution A to 50 cc. with furfural- free alcohol (50% by volume). To this add 2 cc. of colorless aniline and 0.5 cc. of dil. HC1 (5 : 4) and keep for fifteen minutes in a water bath at about 15 C. Prepare standards of known strength from the standard furfural solution in the same way and match up the colors. Calculate the weight of furfural in the original sample. * Saturate a liter of distilled water with SO 2 gas. Titrate an aliquot with 0.1 N iodine and calculate the amount of SO 2 in 1 cc. of the solution. Then measure out a sufficient number of cc. to contain 5 grams of GENERAL ORGANIC ANALYSES 77 Standard Furfural Solutions. Dissolve 1 gram of freshly redistilled furfural in 100 cc. of 95% alcohol. This strong solution will keep. Make standards by diluting 1 cc. of this solution to 100 cc. with 50% alcohol (by volume). One cc. of this weak solution = 0.0001 gram furfural. Fusel Oil. Dilute 50 cc. of the original sample with 50 cc. of distilled water in a 300 cc. Erlenmeyer flask. Add 20 cc. of 0.5 N NaOH and saponify the mixture by boiling for one hour under a reflux condenser. (The same result will be obtained by letting the mixture stand at room temperature overnight.) Connect the flask to a condenser and distill over 90 cc., then add 25 cc. of water to the flask and continue the distillation until a total of 115 cc. has been collected. Nearly saturate the distillate with finely ground salt (NaCl) and add a saturated solution of salt until the sp. gr. is 1.10. Extract this salt solution four times with CCU,* using 40, 30, 20, and 10 cc., respectively. To the CCU extract con- tained in the separatory funnel add 10 cc. of KOH solution (1 : 1). Cool the mixture in ice water to approximately C. Similarly cool 100 cc. of a solution of KMnO4 (20 grams per liter) accurately measured in a flask. To the contents of the separatory funnel add the bulk of the KMnC>4 solution, but without rinsing, retaining the residue to be added at a later stage. Remove the mixture from the bath and shake vigorously for five minutes. Set aside for thirty minutes with occasional shaking, letting the mixture warm up to room temperature (20-25 C.). Accurately measure into a liter Erlenmeyer flask 100 cc. of a solution of H202 about 2% stronger than .the permanganate solu- tion, acidulate with 100 cc. of an approximately 25% H2SO4 solution and slowly add the contents of the separatory funnel, with constant shaking, keeping the acid solution constantly in excess. Rinse the separatory funnel and the flask containing the residue of KMnC>4 with water, and add to the peroxide solu- tion. Finally titrate the excess of H202 with a standard KMnC>4 solution (10 grams to the liter). Run a " blank " determination using the same amounts of the stronger KMnO4, KOH, H 2 O2, and H 2 S04 solutions, and * Purify a liter of ordinary CC1 4 by boiling for several hours under a reflux condenser with 40 cc. of cone. H 2 SO 4 and 5 grams K 2 Cr 2 O 7 in 40 cc. of water. Distill off the CC1 4 and then redistill over BaCO 3 . 78 TECHNICAL METHODS OF ANALYSIS titrate the residual EbCb with the standard KMnCU, as before. Subtract the " blank " titration from the first tit ration and calculate the weight of amyl alcohol corresponding to the differ- ence. Standardize the weaker KMn04 solution carefully against oxalic acid of normal strength. If the solution contains exactly 10 grams of KMnO4 per liter as determined by the titration, then each cc. = 0.696 gram of amyl alcohol. Methyl Alcohol (Wood Spirit). (A) RICHE AND BARDY METHOD.* This method for the detection of methyl alcohol in commercial spirit of wine depends on the formation of methylaniline violet. The procedure is described on page 462. (B) OXIDATION METHOD. Dilute 20 cc. of the sample to 100 cc. and place in a 200 cc. distilling flask. Add from 5-8 grams of chromic acid (CrOa) to oxidize the methyl alcohol to formaldehyde. Distill, collect 10 cc. of the distillate, and test for formaldehyde by one of the following methods: (1) Hehner's Method. To the distillate obtained above add 100 cc. of fresh milk and mix. Fill a test tube about one-third full of the mixture and pour cone. H2S04 carefully down the side, not allowing the two to mix. A violet or blue color at the junction of the two liquids indicates formaldehyde. This test is sensitive to about 1 part in 10,000 of the liquid tested. (2) Leach's Method. Mix the distillate above described with an equal volume of pure milk in a porcelain casserole and add about 10 cc. of cone. HC1 containing about 1 cc. of 10% FeCla solution to each 500 cc. of acid: Heat to 80-90 C. directly over the gas flame, giving the casserole a rotary motion to break up clots. A violet color indicates formaldehyde. (3) Morphine-Sulfate Test. Dissolve 0.5 gram of morphine- sulfate in 500 cc. of cone. H^SCU. Place about 5 cc. of this solu- tion in a test tube and add 1-2 cc. of the distillate above referred to. A violet color on standing shows formaldehyde. NOTE. A blank test should always be carried out simultaneously with pure ethyl alcohol and with ethyl alcohol to which is added a little methyl alcohol. * Tentative method of Assoc. Official Agri'. Chemists, see its Journal, Methods of Analysis (1916), page 246. GENERAL ORGANIC ANALYSES 79 Nitrates. Neutralize 50 cc. of the sample with 0.1 N alkali (phenolphthalein) and evaporate nearly to dryness. Take up with a little distilled water and add 1 cc. of phenoldisulfonic acid; then make alkaline with NILiOH. A bright orange yellow color indicates the presence of nitrates. A " blank " should be run at the same time, evaporating 50 cc. of distilled water to make sure that no nitrate fumes have been picked up during the evap- oration. Sulfur Compounds. Place 100 cc. of the sample in a platinum dish and make slightly alkaline with 0.1 N NaOH (phenolphtha- lein). Add 5 cc. of H2O2, evaporate to dryness and ignite over an alcohol lamp. Dissolve the residue in 50 cc. of water, add 5 cc. of dil. HC1, bring to boiling and add 5 cc. of 10% BaC^ solution. If any precipitate forms, filter it out, ignite and weigh in the usual manner. Calculate to sulfur. CALCULATION. BaS0 4 X 0.1373 = 8. NOTE. A "blank" should be run at the same time, using the same amounts of each reagent as was used in the determination. If the blank shows any sulfur, its amount should be determined and subtracted from the weight previously found. REFERENCES. U. S. Dept. of Agriculture, Bureau of Chem., Cir. No. 74, p. 5; U. S. Dept. of Agriculture, Bureau of Chem., Bull. 107, revised, p. 185; Leach: ''Food Inspection and Analysis," 3d edition, pp. 745-750; U. S. Pharmacopoeia, 9th edition, p. 35. FORMALDEHYDE SOLUTION General. Formaldehyde is a gaseous substance of the formula HCHO. It is used in aqueous solution as a disinfectant, insecti- cide and deodorizer. The U. S. P. solution should contain not less than 37% by weight of HCHO. The solution also generally contains more or less methyl alcohol to prevent polymerization. The following procedures for the determination of formalde- hyde are recognized as official methods by the Association of Official Agricultural Chemists. Hydrogen Peroxide Method. (A) REAGENTS. (a) Normal 04. (b) Normal NaOH (1 cc. = 0.03002 gram HCHO). (c) Hydrogen Peroxide: An approximately 3% solution. If 80 TECHNICAL METHODS OF ANALYSIS the solution is acid, neutralize with (fr), using litmus solution as indicator. (d) Litmus Solution: A solution of purified litmus. (B) DETERMINATION. Measure 50 cc. of N NaOH into a 500 cc. Erlenmeyer flask and add 50 cc. of the H 2 2 . Then add 3 cc. of the formaldehyde solution under examination, letting the point of the pipette reach nearly to the liquid in the flask. Place a funnel in the neck of the flask and heat on the steam bath for five minutes, shaking occasionally. Remove from the steam bath, wash the funnel with water, cool the flask to about room tempera- ture, and titrate with N acid, using litmus solution as indicator. It is necessary to cool the flask before titration with the acid to get a sharp end point. Calculate per cent of formaldehyde. Cyanide Method. (A) REAGENTS. (a) 0.1 N AgN0 3 . (6) 0.1 N NH 4 SCN. (c) KCN solution: Dissolve 3.1 grams of KCN in 500 cc. of water. (d) 50% HNO S . (B) DETERMINATION. Treat 15 cc. of 0.1 N AgNOs with 6 drops of 50% HNOs in a 50 cc. graduated flask. Add 10 cc. of KCN solution. Dilute to volume; shake well; filter through a dry filter, and titrate 25 cc. of the filtrate with 0.1 N NttiSCN, using 5 cc. of a saturated solution of ferric alum as indicator, and con- tinue the titration until the first appearance of a permanent light- brown color. Acidify another 15 cc. portion of 0.1 N AgNOs with 6 drops of 50% HN0 3 and treat with 10 cc. of the KCN solution to which has been added a measured quantity (the weight of which must be calculated from the sp. gr.) of the formaldehyde solution, containing not over 2.5 grams of a 1% solution or the equivalent. Make up to 50 cc., filter and titrate a 25 cc. aliquot with 0.1 N NILtSCN for the excess of Ag as before. The differ- ence between the number of cc. of 0.1 N sulfocyanate used in these 2 titrations, multiplied by 2, gives the number of cc. of 0.1 N sul- focyanate corresponding to the KCN consumed by the formal- dehyde. Calculate the percentage of formaldehyde present. CALCULATION. 1 cc. 0.1 N sulfocyanate = 0.003002 gram HCHO. REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 75. GENERAL ORGANIC ANALYSES 81 FORMIC ACID General. Formic acid occurs in solutions of varying strengths, such as 30, 50, 75 and 90%. The weaker solutions are generally made by diluting the stronger. There are also often present HC1 and H 2 S04, the amount depending upon the extent to which the material has been purified. Specific Gravity at 15.5 C. This may be determined by the Westphal balance; if great accuracy is desired, a pycnometer should be used. Sulfuric Acid. Weigh approximately 20 grams from a weighing bottle into a beaker. Dilute with 250 cc. of distilled water; add 5 cc. of cone. HC1; heat to boiling, and add an excess of hot 10% BaCl2 solution. Continue the boiling for at least fifteen minutes. Let stand several hours. Filter out the BaSCU, wash, ignite, and weigh in the usual manner. Calculate to [2804. ' CALCULATION. BaSO 4 X 0.4202 = H 2 S0 4 . Hydrochloric Acid. Weigh 10-20 grams of the liquid into a 300 cc. Erlenmeyer flask. (This flask should previously have been cleaned with sulfuric acid bichromate solution, to remove all the grease and prevent sticking of the precipitate.) Dilute with 150 cc. of distilled water; add 5 cc. cone. HNOs and then AgNOs solution in excess. Shake the liquid in the flask until the precipitate settles clearly. Filter through a weighed Gooch crucible, washing with cold distilled water. Dry the precipitate in the crucible at 100-110 C., then place the Gooch crucible in a large platinum crucible and ignite gently until the edges of the precipitate just begin to fuse. Cool in a desiccator and weigh the AgCl. Calculate to HC1. CALCULATION. AgCl X 0.2544 = HC1. Formic Acid. Weigh 8-10 grams accurately from a weighing bottle into a 500 cc. graduated flask about one-half full of dis- tilled water. Make up to the mark and mix thoroughly. Pipette 50 cc. of this solution into a 300 cc. Erlenmeyer flask. Make the solution alkaline with Na 2 C03 solution. Warm and add a measured excess of standard 0.1 N KMn0 4 . It is very necessary that an excess of the permanganate be added. This oxidizes the formic acid to CO 2 and water and throws down a heavy precip- itate of manganese dioxide. Then acidify with 10 cc. of dil. 82 TECHNICAL METHODS OF ANALYSIS H2S04. Run in from a burette a measured volume of 0.1 N oxalic acid * until all the precipitate has dissolved and all the permanganate color has disappeared. Finally titrate the excess of H2C2O4 with the 0.1 N KMnCU. The difference between the total 0.1 N KMn04 used and the amount which was equivalent to the H 2 C2O 4 gives the number of cc. of 0.1 N KMnO4 required to oxidize the formic acid. CALCULATION. 1 cc. 0.1 N KMnO 4 = 0.002301 gram formic acid. NOTE. If other acids are known to be absent, the amount of formic acid can be determined by titrating directly with 0.1 N NaOH and phenol- phthalein. 1 cc. 0.1 N NaOH =0.004602 gram formic acid. REFERENCE. Am. Chem. J. 17, 539 (1895). ACETIC ANHYDRIDE General. Acetic anhydride is a very corrosive substance and its vapors are extremely irritating to the eyes and lungs. Care should therefore be exercised in working with it, and any spilled upon the hands should be immediately washed off. It is a con- densation of 2 molecules of acetic acid with elimination of water. The reaction, however, is reversible, and there is always present some acetic acid, due to action of water } thus: (CH 3 CO) 2 0+H 2 ^ 2CH 3 COOH. Determination of Anhydride Content. When the material is pure and contains only acetic anhydride and acetic acid, Method No. 1 (direct titration), described below, gives satisfactory results. In case of impure samples, however, Method No. 2 (aniline method) should be employed. The latter method is the one employed by the U. S. Government and should always be used unless otherwise expressly directed. METHOD No. 1. DIRECT TITRATION. Weigh accurately about 25 grams from a weighing bottle into a 2-liter volumetric flask, partly filled with CO 2 -free water; let stand overnight, * It is not absolutely necessary to have the oxalic acid of a definite strength. In case, however, it is not 0.1 N, measure off a volume of the oxalic acid equal to that used in the determination and titrate it against the 0.1 N KMnOi. GENERAL ORGANIC ANALYSES 83 make up to volume with CO2-free water and titrate 50 cc. aliquots with 0.1 N NaOH and phenolphthalein. Calculate the titration to acetic acid. 1 cc. 0.1 N NaOH = 0.006004 gram acetic acid. Since 100% of acetic anhydride is equivalent to 117.651% of acetic acid, then percentage of acetic anhydride = X 100 117 . ool lOU = 5.665 (A -100), where A is the per cent of acetic acid found by titration. NOTE. As any error in the acetic acid result is increased nearly sixfold in the final results, titration should be carried out with great care and the aver- age of several aliquots should be taken. Corrections for burette calibrations must be made and also corrections for temperature effects on the standard solution. The standard NaOH should be as free from carbonate as possible. METHOD No. 2. ANILINE METHOD. Shake the sample thoroughly and rinse a burette twice with portions of it. Clean and dry two 50 cc. weighing bottles, and two of 5 cc. size. (Do not use alcohol for drying the weighing bottles or any of the apparatus used in weighing out the anhydride.) Into each of the 5 cc. weighing bottles, which have been care- fully weighed, run 2 cc. of anhydride from the burette. Imme- diately stopper the bottles and weigh again to obtain the weight of anhydride. Then drop very carefully each weighing bottle with its contents into a separate 300 cc. Erlenmeyer flask containing 50 cc. of N NaOH and 50 cc. of distilled water. Loosen the stopper of the weighing bottle slightly before dropping it into the flask, so that it can be freed from the weighing bottle and the contents mixed with the NaOH solution. Let the mixture stand thirty to forty- five minutes at room temperature with occasional shaking. Then titrate with N or 0.5 N acid and phenolphthalein. Add 2 or 3 cc. of acid in excess and let the solution stand about fifteen minutes longer with occasional shaking. Finally titrate back the excess of acid with 0.1 N NaOH to the first permanent pink color. From this titration calculate the number of cc. of N NaOH represent- ing the total acetic acid from 100 grams of the sample. Call this A. 84 TECHNICAL METHODS OF ANALYSIS Into each of the 50 cc. weighing bottles run about 20 cc. of perfectly dry and recently distilled aniline. (The aniline should be distilled over solid NaOH, discarding the first portion of distillate.) Then add 2 cc. of anhydride from the burette. This operation requires considerable care. Do not add the anhydride too rapidly, and distribute it through the aniline as thoroughly as possible by keeping the aniline swirling slightly in the weighing bottle during the addition. When the anhydride has been added, stopper the bottle and set aside to come to room temperature; then weigh to obtain the weight of anhydride added. At the end of about one hour from the time of addition of anhydride to aniline, transfer the mixture to a 500 cc. volumetric flask and make up to the mark with a solution of equal parts of neutral alcohol and distilled water. Titrate 50 cc. of this solu- tion with 0.1 N NaOH. From this titration calculate the number of cc. of N NaOH corresponding to the residual acetic acid from 100 grams of the sample. Call this B. Subtract B from A to obtain the number of cc. of N NaOH corresponding to one-half the anhydride in 100 grams of sample. This value multiplied by 0.10207* gives grams of acetic anhydride per 100 grams of sample, i.e., the percentage. Mineral Acids. Shake up a portion of the sample with cold water and test the solution in the usual way for HC1 with AgNOs and for H2&O4 with BaC^. The determination may be made quantitative by starting with a weighed amount of the anhydride sample. NOTE. If mineral acids are found present, corrections must be made for the amount of NaOH they will neutralize in making the anhydride titration. Boiling Point. The boiling point of pure acetic anhydride is approximately 138 C. and a satisfactory sample should all distill between 130 and 140 C. REFERENCE. Worden: "Technology of Cellulose Esters," Vol. VIII, pages 2910-2917. * It will be noted that the factor is the molecular weight of acetic anhydride divided by 100. The method may be applied to other anhydrides, e.g., butyric anhydride (Mol. Wt. = 158.15), in which case the factor will be 0.01 of the -molecular weight of the anhydride in question. GENERAL ORGANIC ANALYSES 85 GLYCEROL General. There are two recognized methods for the deter- mination of glycerol (1) the acetin method, which depends upon the conversion of glycerol into triacetin with sodium acetate and acetic anhydride and a quantitative saponification of the tri- acetin thus formed; (2) the bichromate method, which is based on the fact that K^C^O? in the presence of H^SCX completely oxidizes glycerol to CO2 and H^O. The acetin method is the one agreed upon at a conference of delegates from the British, French, German, and American com- mittees on glycerol analysis as giving results nearer to the truth than the bichromate method on crude glycerines in general. It is the method to be employed whenever possible, but for the application of this method the solution must not contain over 60% of water. In general, therefore, use the acetin method for crude or refined glycerines and the bichromate method for soap lyes. ACETIN METHOD Reagents Required. (A) Best Acetic Anhydride. This should be carefully selected. A good sample must not require more than 0.1 cc. of N NaOH for saponification of the impurities when a blank is run on 7.5 cc. Only a slight color should develop during diges- tion of the blank. (See under " Blank Test " below.) The anhydride may be tested for strength by the aniline method described under analysis of Acetic Anhydride (page 83). (B) Pure Fused Sodium Acetate. Refuse the ordinary salt in a platinum, silica, or nickel dish, avoiding charring, powder quickly and keep in a stoppered bottle or desiccator. It is very important that the sodium acetate be anhydrous. (C) NaOH (approximately N and Free from Carbonate). Dis- solve pure NaOH in its own weight of water (free from CO2). Let settle till clear or filter through asbestos. Dilute the clear solution with water free from CO2 to the strength required. (D) N NaOH (Free from Carbonate). Prepare this solution as above but carefully standardize it. Some NaOH solutions show a marked diminution in strength after boiling; such solutions should be rejected. 86 TECHNICAL METHODS OF ANALYSIS (E) N Acid. Standardize this against the N NaOH. (F) Phenolphthalein Solution. Dissolve the powder in alcohol sufficient to make a 0.5% solution and neutralize with N NaOH to a very slight pink color. Procedure. In a narrow-mouthed flask (preferably round- bottomed), capacity about 120 cc., which has been thoroughly cleaned and dried, weigh accurately and as rapidly as possible 1.25-1.50 grams of the glycerine. A Grethun or Lunge pipette will be found convenient. Add about 3 grams of anhydrous sodium acetate and 7.5 cc. of acetic anhydride, and connect the flask with an upright Liebig condenser. For convenience the inner tube of this condenser should not be over 50 cm. long and 9-10 mm. inside diameter. The flask is connected to the condenser by either a ground glass joint (preferably) or a rubber stopper. If a rubber stopper is used it should have had a preliminary treatment with hot acetic anhydride vapor. Heat the contents and keep just boiling for one hour, taking precautions to prevent the salts from drying on the sides of the flask. Let the flask cool somewhat, and through the condenser tube add 50 cc. of distilled water, free from CO2, at a temperature of about 80 C., taking care that the flask is not loosened from the condenser. The object of cooling is to avoid any sudden rush of vapors from the flask on adding water, and to avoid breaking the flask. Time is saved by adding the water before the contents of the flask solidify, but the contents may be allowed to solidify and the test proceeded with the next day without detriment, bearing in mind that the anhydride in excess is much more effectively hydrolyzed in hot than in cold water. The contents of the flask may be warmed to, but must not exceed, 80 C. until solution is complete (a few dark flocks may remain in suspension, repre- senting organic impurities in the crude glycerine). By giving the flask a rotary motion, solution is more quickly effected. Cool the flask and contents without removing the condenser. When quite cold, wash down the inside of the condenser tube, detach the flask, wash off the stopper or ground glass connection into the flask, and filter the contents through an acid-washed filter into a Pyrex glass flask of about 1 liter capacity. Wash thoroughly with cold distilled water free from CC>2. Add 2 cc. of phenol- GENERAL ORGANIC ANALYSES 87 phthalein solution (F), then run in NaOH solution (C) or (D) until a faint pinkish yellow color appears throughout the solution. This neutralization must be done most carefully; the alkali should be run down the sides of the flask, the contents of which are kept rapidly swirling with occasional agitation or change of motion until the solution is nearly neutralized, as indicated by the slower disappearance of the color developed locally by the alkali running into the mixture. When this point is reached the sides of the flask are washed down with C02-free water and the alkali subsequently added, drop by drop, mixing after each drop until the desired tint is obtained. Now run in from a burette 50 cc. or a calculated excess of N NaOH (D) and note carefully the exact amount. Boil gently for fifteen minutes, the flask being fitted with a glass tube acting as a partial condenser. Cool as quickly as possible and titrate the excess of NaOH with N acid (E) until the pinkish yellow or chosen end-point color just remains.* A further addition of the indicator at this point will cause an increase of the pink color; this must be neglected, and the first end point taken. From the N NaOH consumed calculate the percentage of glycerol (including acetylizable impurities) after making the cor- rection for the blank test described below. 1 cc. N NaOH = 0.03069 gram glycerol. The coefficient of expansion for normal solutions is 0.00033 per cc. for each degree Centigrade. A correction should be made on this account if necessary. Blank Test. As the acetic anhydride and sodium acetate may contain impurities which affect the result, it is necessary to make a blank test, using the same quantities of acetic anhydride, sodium acetate and water as in the analysis. It is not necessary to filter the solution of the melt in this case, but sufficient time must be allowed for the hydrolysis of the anhydride before pro- ceeding with the neutralization. After neutralization it is not necessary to add more than 10 cc. of the N alkali (D), as this represents the excess usually present after the saponification of *A precipitate at this point is an indication of the presence of iron or alumina, and high results will be obtained unless a correction is made as described below. 88 TECHNICAL METHODS OF ANALYSIS the average soap lye crude. In determining the acid equivalent to the N NaOH, however, the entire amount taken in the analysis, 50 cc., should be titrated after dilution with 300 cc. of water free from CC>2 and without boiling. Determination of the Glycerol Value of the Acetylizable Im- purities. Certain crude glycerines may contain a considerable amount of acetylizable impurities other than glycerol. To deter- mine the amount of these impurities, dissolve the total residue at 160 C. (see below) in 1 or 2 cc. of water, wash into the acetylizing flask and evaporate to dryness. Then add anhydrous sodium acetate and acetic anhydride in the usual amounts and proceed as described iri the regular analysis. True Glycerol Content. After correcting for the blank cal- culate the result obtained in the preceding paragraph to glycerol and subtract the amount from the total amount of glycerol obtained in the analysis. Total Residue at 160 C. If the glycerine is acid, make it slightly alkaline with Na2COs to prevent the loss of organic acids. To avoid the formation of polyglycerols this alkalinity must not exceed 0.2% Na2O. If, therefore, the glycerine is too strongly alkaline, sufficient N HC1 must be added to bring the alkalinity down to 0.2%. Place 10 grams of the sample in a 100 cc. flask, dilute with water and add the calculated quantity of N HC1 or Na2CO3 to give the required degree of alkalinity. Dilute to 100 cc., mix thoroughly and pipette 10 cc. into a weighed Petri or similar dish, 2.5 inches in diameter and 0.5 inch deep, with a flat bottom. In the case of crude glycerines abnormally high in organic residue, a smaller amount should be taken so that the amount of the organic residue does not materially exceed 30-40 milligrams. Place the dish on a water bath until most of the water is evap- orated, then place in an oven and evaporate the glycerine, or most of it at least, at a temperature of 130-140 C. When only a slight vapor is seen to come off, take off the dish and let it cool. Add 0.5-1 cc. of water and bring the residue wholly or nearly into solution with a rotary motion. Place the dish on the water bath or on top of the oven until the excess water has evap- orated and the residue is in such a condition that it will not spurt if the oven is raised to 160 C. In the meantime set the oven at GENERAL ORGANIC ANALYSES 89 exactly 160 C. From this point the time of heating must be strictly observed. Place the dish in the oven and maintain at exactly 160 C. for one hour. Remove the dish, cool, treat the residue with water, and evaporate the water as before. Then place the dish again in the oven and heat a second time for exactly one hour. Place the dish in a desiccator and let cool over H2SO4. Weigh the cooled dish. Again moisten with water and heat at 160 C. for one hour. Repeat the operation until a constant loss of 1-1.5 mg. per hour is obtained. In the case of acid glycerine correct for the N Na 2 CC>3 added by subtracting 0.03 gram for each cc. added. In the case of alkaline glycerine correct for the amount of N HC1 added by calculating the increase in weight due to the conversion of the NaOH and Na 2 C03 to NaCl. From the corrected weight calcu- late the percentage of total residue at 160 C. This residue is taken for the determination of the non-volatile acetylizable impuri- ties. (See above.) BICHROMATE METHOD Reagents Required. (A) Pure K 2 Cr 2 7j powdered and dried at 110-120 C. in air free from dust or organic vapors. This is taken as the standard. (B) Dilute Bichromate Solution. Dissolve 7.4564 grams of the above bichromate in distilled water and make up the solution to 1 liter at 15.5 C. (C) Ferrous Ammonium Sulfate. It is never safe to assume this salt to be constant in composition, and the solution must be standardized against the bichromate as follows: Dissolve 3.7282 grams of bichromate (A) in 50 cc. of water. Add 50 cc. of 50% H 2 S0 4 (by volume), and to the cold undiluted solution add from a weighing bottle a moderate excess of the ferrous ammonium sulfate, and titrate back with the dilute bichromate (B). Calculate the value of the ferrous salt in terms of bichromate. (D) Silver Carbonate. Prepare this as required for each test. Make up a 0.5% Ag 2 SO 4 solution and precipitate the Ag 2 CO 3 from 140 cc. of this solution with about 4.9 cc. of N Na 2 CO 3 solution (a little less than the calculated quantity of N Na 2 CO 3 should 90 TECHNICAL METHODS OF ANALYSIS be used as an excess prevents rapid settling). Let settle, pour off the liquid and wash once by decantation. (E) Subacetate of Lead. Boil a 10% solution of pure lead ace- tate with an excess of litharge (PbO) for one hour, keeping the volume constant, and filter while hot. Disregard any precipitate which subsequently forms. Preserve out of contact with CCb. (F) Potassium Ferricyanide. Use a freshly prepared, very dilute (about 0.1%) solution of this salt. Procedure. Weigh out 20 grams of the glycerine, make up with water to 250 cc. in a volumetric flask and pipette out 25 cc. into a 100 cc. graduated flask. To this add the Ag 2 CO3, let stand with occasional agitation for about ten minutes, and add a slight excess (about 5 cc. in most cases) of basic lead acetate (E). Let stand a few minutes, dilute with distilled water to 100 cc. and then add 0.15 cc. to compensate for the volume of the precipitate. Mix thoroughly, filter through an air-dry filter into a test tube, rejecting the first 10 cc. and return the filtrate, if not clear and bright. Test a portion of the filtrate with a little basic lead acetate, which should produce no further precipitate. In the great majority of cases 5 cc. are ample, but occasionally a crude will be found requiring more, and in this case another aliquot of 25 cc. of the dilute glycerine should be taken and purified with 6 cc. of basic lead acetate. Care must be taken to avoid a marked excess of basic acetate. When the filtrate is coming through perfectly clear, collect sufficient in an Erlenmeyer flask so that 25 cc. may be pipetted into a flask or beaker which has been previously cleaned with K 2 Cr 2 7 and H 2 SO 4 . To this add 12 drops of H 2 SO 4 (1 : 4) to precipitate the small excess of lead as PbSO 4 . Then add 3.7282 grams of the powdered K 2 Cr 2 07. Rinse down the bichromate with 25 cc. of water and let stand with occasional shaking until all the bichromate is dissolved. (No reduction will take place in the cold.) Add 50 cc. of 50% (by volume) H 2 S04, immerse the vessel in boiling water for two hours and keep protected from dust and organic vapors, such as alcohol, until the titration is completed. Add from a weighing bottle a slight excess of the ferrous ammo- nium sulfate, making spot tests on a porcelain plate with ferri- cyanide indicator until an excess is shown. Then titrate back GENERAL ORGANIC ANALYSES 91 with the dilute bichromate solution. Calculate the percentage of glycerol from the amount of bichromate reduced. CALCULATION. 1 gram K2Cr2O7 = 0.13411 gram glycerol. NOTES. (1) The percentage of glycerol obtained above includes any oxidizable impurities present after the purification. A correction for the non- volatile impurities may be made by running a bichromate test on the residue at 160 C. (2) It is important that the concentration of acid in the oxidation mixture and the time of oxidation should be strictly adhered to. (3) Before the bichromate is added to the glycerine solution it is essential that the slight excess of lead be precipitated with H 2 SO 4 , as stipulated. (4) For crudes practically free from chlorides the quantity of Ag 2 COa may be reduced to one-fifth and the basic lead acetate to 0.5 cc. (5) It is sometimes advisable to add a little K 2 SO 4 to insure a clear filtrate. (6) Neither the acetin nor bichromate method is correct in theory or in practice on crudes containing trimethyleneglycol or polyglycerols but the acetin method gives nearer the truth. REFERENCE. J. Ind. Eng. Chem. 3, 679 (1911). Approved Report of the Sub-committee on Glycerine Analysis. DEXTRIN OR BRITISH GUM General. Dextrin corresponds to the formula (CeHioOs^ and is generally considered an intermediate product between starch and dextrose. The commercial product is made by heating dry starch to 200-250 C. or by moistening the starch with acid, drying at 50 C. and then heating to 140-170 C. The product is an indefinite mixture of several dextrins with unchanged starch and may also contain more or less dextrose. The dextrins are soluble in cold water and form a thick viscous syrup which has strong adhesive properties and is much used as a substitute for gum arabic in the preparation of mucilage and for thickening color in calico printing, etc. Moisture. Dry 5 grams to constant weight at 100 C. in a weighed platinum dish. Ash. Ignite carefully the residue from the moisture deter- mination to a white or grayish white ash. Cool in a desiccator and weigh. Insoluble in Cold Water (Starch) .Stir into about 250 cc. of water 25 grams of the sample. Wash into a 500 cc. graduated flask, Shake occasionally for several hours, dilute to volume and 92 TECHNICAL METHODS OF ANALYSIS let stand overnight. Pipette out 50 cc. of the clear supernatant liquor, evaporate in a weighed dish on the water bath and dry to constant weight at 100 C. This gives the weight of the soluble solids in 2.5 grams. Divide by 2.5 and multiply by 100 to obtain the percentage of soluble solids. Add to this the percentage of moisture and subtract the sum from 100%. The difference is insoluble matter (unconverted starch). Dextrose. (Munson and Walker Method.) Mix 10 grams of the sample with water, stir thoroughly and wash into a 250 cc. graduated flask. Add 5 cc. of lead subacetate solution, dilute to volume, mix thoroughly, let settle and filter through a dry filter. Do not wash. Add a few crystals of anhydrous K2C2O4 to remove the lead and again filter through a dry filter. Pipette out 50 cc. of this solution and determine reducing sugars according to the Munson and Walker method described on page 403. Calculate to dextrose, using the third column of figures in the Munson and Walker tables. Dextrin. Pipette 100 cc. of the solution of the material pre- pared for the dextrose determination (after the lead has been removed and the solution filtered), into a 250 cc. flask, add 20 cc. of HC1 (1:1) and 100 cc. of water. Heat in a boiling water bath for 2.5 hours. Cool and nearly neutralize to litmus paper with NaOH. Dilute to 500 cc. in a graduated flask, pipette out 50 cc. (equivalent to 0.4 gram of the original) and determine the reducing sugars by the Munson and Walker method as described above under " Dextrose." Calculate the total reducing sugars as dextrose. Subtract from this figure the percentage of dextrose, previously determined, and multiply the difference by 0.9 to obtain the per cent of dextrin. NOTE. By this method the amount of soluble starch, if any, is included with the dextrin, but as their properties are, for textile purposes, quite sim- ilar, the additional work involved in the separation is not worth while. Viscosity. Determine the viscosity by means of a Dudley pipette at 25 C. The pipette should be standardized with a sugar solution made by dissolving 120 grams of pure cane sugar in 100 cc. of distilled water at 25 C. This solution should give a viscosity of about 100 seconds. In determining the viscosity, the pipette should be surrounded by a jacket containing water at the same GENERAL ORGANIC ANALYSES. 93 temperature. Fill the pipette slightly above the mark with the solution, then set it exactly on the upper mark. Release the solution with one hand and start the stop watch with the other. Determine the number of seconds required by the solution to run from the top mark to the bottom mark. Several determinations should be made and the average reported. After standardizing the pipette with sugar solution determine the viscosity of a solution of the sample in the same way. Mix 40 grams of the sample with about 150 cc. of distilled water, heat to boiling for exactly one minute, stirring thoroughly; transfer to a 200 cc. flask, cool and make up to volume at 25 C. Use this solution for the vis- cosity determination. NOTE. The viscosity of the solution depends largely on the time of heating. The material should be stirred up with a little warm water and then trans- ferred rapidly to a beaker containing the rest of the water which should be near the boiling point. It should be rapidly heated to boiling and boiled for exactly one minute. ALBUMIN General. Commercial albumin is chiefly from two sources, eggs and blood-serum. The latter is cheaper than egg albumin and has a better thickening power. It is largely used for fixing dyes and pigments in calico printing in all but the finest colors. The blood-serum albumin varies from a dirty yellow color to the black color of " dried blood." Egg albumin is generally transparent and of a light yellow color. It should be free from blisters which indicate partial coagulation. On treatment with cold water, commercial albumin of good quality should dissolve almost com- pletely. In making the test, the albumin should always be added to the water and not vice versa. (If it is desired to keep the solu- tion, add about 1% of arsenious oxide.) Qualitative Tests. Commercial albumin is frequently adul- terated with dextrin, gums, sugar, flour, etc. For qualitative examination grind to a powder and add 5 grams (accurately weighed) slowly to about 50 cc. of water until the soluble matter is dissolved. Pure and high-grade samples should leave no residue. Add a few drops of acetic acid and filter through silk or fine muslin into a 500 cc. volumetric flask. The insoluble residue may con- 94 TECHNICAL METHODS OF ANALYSIS sist of coagulated albumin, casein, starch, or membranous matter. Treat it with very dilute NaOH solution and then exactly neu- tralize the nitrate with acetic acid. If casein is present it will be dissolved by the NaOH and reprecipitated on neutraliza- tion.. Make the original filtrate up to 500 cc. Pour about 100 cc. into a beaker and heat to boiling. This should coagulate the albumin. Filter and treat the filtrate with a little acetic acid and potassium ferrocyanide to make sure that no proteins remain in the solution. If there is a precipitate, filter it out. (A precip- itate generally indicates casein, although it must be remembered that any zinc would be thrown out as white ferrocyanide.) Cool the filtrate and add a little concentrated tannic acid solution. This will precipitate any gelatin or glue. Filter, and concen- trate the filtrate to a small bulk. Cool and treat with a con- siderable excess of alcohol. Any precipitate indicates the presence of gums or dextrin. Filter, boil off all the alcohol, heat with dilute HC1 and test the solution in the usual way with Fehling's solution after neutralizing excess of HC1. Any reduction indicates the presence of sugar but does not necessarily prove its presence. Sugar may also be extracted by treating the original solid sample with alcohol. Soluble Coagulable Albumin. Pipette 100 cc. of the original water solution (equivalent to 1 gram) into a beaker. Add about 1 gram of sodium acetate and heat to boiling. Filter* the floc- culent precipitate on a tared filter using a platinum cone and suc- tion. Wash with hot water, dry at 100 C. and weigh. Moisture. Dry 2 grams of the material in a weighed porce- lain dish to constant weight at 100 C. Report the loss as moisture. Ash. Instead of determining the actual residue on ignition ia the usual way, it is better on account of the fusible nature of the ash to proceed as follows: Treat the residue from the moisture determination with a few drops of cone. HNO 3 and 2 or 3 drops of H 2 S0 4 . On heating gently, the albumin dissolves to a clear yellow liquid which may be evaporated to dryness without trouble. Ignite the dry residue and weigh. Report the result as " sulfate ash." Allen gives the following results obtained on various samples: GENERAL ORGANIC ANALYSES 95 Designation. Sulfate Ash, Per Cent. Itjiiti albumin ... No. 1 7 4 No 2 7 Blood-serum, albumin. Refined 9.1 Blood-serum albumin Prime 8.5 Blood-serum albumin No. 1 9.2 Blood-serum albumin No 2 8 9 Blood-serum albumin No. 3 9.7 Blood-serum albumin Black 6.2 The addition of sugar or dextrin to the albumin would lower the amount of ash. If it is abnormally high it may also be exam- ined for zinc or other inorganic materials. REFERENCE. Allen: "Commercial Organic Analysis" (2d ed., 1898), Vol. IV, page 44. TANNIC ACID General. Many processes have been devised for estimating tannic acid in the substances known as " Tannins." Most of these methods, however, have had in view the valuation of tannins for tanning leather, but it does not necessarily follow that they are equally serviceable for valuing tannins when used in dyeing. Proctor's modification of LoewenthaFs method has been very generally adopted both for tanning and dyeing purposes. Never- theless, if the tannin is to be used in the leather industries, the Loewenthal method of analysis should not be used. In such cases use the Official Method of the American Leather Chemists' Association. The determinations most generally required on tannins or tannic acid for dyeing purposes are as follows: Moisture. Weigh 1 gram if the material is solid, 5 grams if liquid, in a platinum crucible or dish and dry at 100 C. to con- stant weight. Report loss in weight as moisture. Ash. Ignite the above, carefully at first, and finally to the full heat of the Bunsen burner; cool in a desiccator and weigh. Total Astringency (Loewenthal-Proctor Method.). Dissolve 1 gram in water and dilute to 1 liter. Pipette 10 cc. of this solu- 96 TECHNICAL METHODS OF ANALYSIS tion into a large porcelain dish containing 750 cc. of water. Add 25 cc. of indigo carmine solution and titrate with standard KMnO4 exactly as under Sumac analysis (page 479). Astringent Non-tannins. To 50 cc. of the above solution in a strong 8-ounce bottle, add 25 cc. of 2% gelatin solution, 25 cc. of saturated NaCl acid solution and 10 grams of china clay. Shake thoroughly for about five minutes and filter through a dry filter. This removes the tannins. Test a portion of the filtrate with more of the gelatin solution to see if precipitation is complete. (If not complete, make up a gelatin solution stronger than 2% and repeat the process, using 25 cc. of the latter solution.) Titrate 20 cc. of this filtrate, equivalent to 10 cc. of the original solution, and calculate the percentage of astringent non-tannins. Tannic Acids or Tannins. The difference between the total astringency, calculated as tannin, and the astringent non-tannins gives the percentage of tannins. NOTES. (1) The solutions referred to above are made as follows: (a) Indigo Carmine Solution. Dissolve 5 grams of pure indigo carmine in water, add 50 grams of cone. H 2 SO 4 and dilute to 1 liter. If indigo carmine is not available, sulfonate 1 gram of pure indigo with 60 cc. of cone. H 2 SO 4 for six hours at 70-80 C. and dilute with H 2 O to 1 liter. (6) KMn04 Solution (approx. 0.01 N). This should be accurately stand- ardized against oxalic acid or sodium oxalate. (c) Gelatin Solution. Soak 20 grams of Nelson's gelatin in water for two or three hours. Then dissolve on the steam bath with the addition of more water and dilute to 1 liter. (d) Saturated NaCl Add Solution. Make up a 5 per cent solution of H 2 SO 4 and saturate with ordinary salt. (2) In calculating the tannin titrations use the factor 1 cc. 0.1 N KMnO 4 = 0.004157 gram tannin. (3) It is often customary in analyzing mixed tannins or extracts containing tannin to calculate the tannin in terms of oxalic acid. 1 cc. 0.1 N KMnO 4 = 0.006303 gram H 2 C 2 O 4 -2H 2 O. (4) The above method is especially intended for pure tannic acid and tannins for use in dyeing. REFERENCE. Knecht-Rawson-Loewenthal : " Manual of Dyeing," Vol. II, 2d Edition, page 802. GENERAL ORGANIC ANALYSES 97 INDIGO POWDER OR PASTE General. The methods employed for testing indigo may be broadly classified into three groups: (1) Conversion into sulfonic acid. (a) Indigotin estimated by oxidation. (6) Indigotin estimated by reduction. (2) Indigotin reduced in an alkaline solution, the indigo tin re-oxidized, separated, purified, and weighed, or dissolved in acid and titrated as in (1). (3) Extraction by volatile solvents. There are two kinds of indigo met with, natural and synthetic. At the present time the synthetic has practically replaced the natural. (The European War, however, has stimulated again the production of natural indigo.) The natural indigo usually comes in the form of lumps and contains besides indigo blue, or indigotin, other substances both of organic and inorganic nature which owe their presence to the process of manufacture. Their amount varies according to the quantity of indigo blue that is contained in the sample and varies widely in indigoes of different origin. The mineral impurities consist of sand, silicates, and CaCOs. The organic impurities consist of (1) indigo gluten or gum, (2) indigo brown, and (3) indigo red or indirubin. Synthetic indigo is practically pure indigotin and is usually put out in the form of a paste containing about 20% indigotin and 80% water. Moisture. If the sample is in the form of lumps, grind to a fine powder and dry 1 gram at 100-105 C. to constant weight. If the sample is in the form of a paste, mix thoroughly and weigh out as rapidly as possible 10-15 grams and dry to constant weight at 100-105 C. Report loss in weight as moisture. Indigotin. (By reduction with hydrosulfite.) This method gives the most exact figures for the indigotin content and is not influenced by the impurities contained in the indigo. It also is the quickest method and the danger of errors is reduced to a 98 TECHNICAL METHODS OF ANALYSIS minimum owing to the change in color being very easy to recog- nize. SULFONATION. Weigh accurately 1 gram of the very finely powdered dry indigo and heat for five hours with 6 cc. of cone. H2S04 at 105-120 F. with frequent stirring, then pour into water and make up to 1 liter. In the case of natural indigo there is usually an insoluble residue which, however, contains no indigotin. The temperature should not be allowed to exceed 120 F. as other- wise much darker solutions are obtained, which render the titration more difficult. PKEPARATION OF SOLUTIONS. Any freshly prepared solution of sodium hydrosulfite may be used as a titrating solution, pro- vided that it does not contain more than 1% alkali. It is advisable to use a hydrosulfite solution of such strength that 25-30 cc. will decolorize 0.1 gram of indigo or 100 cc. of an 0.1% solution of indigo. To prepare the hydrosulfite solution mix 400 cc. of sodium bisulfite liquor of sp. gr. 1.36-1.38 (72-76 Tw.) with 950 cc. of H2O, and then add 35 grams of zinc powder which has been pre- viously worked to a paste with 50 cc. of H^O. The zinc paste should all be added within fifteen minutes in small portions, stirring or shaking gently. After the mixture has stood for one hour, draw off the clear liquid into lime water prepared by slaking 45 grams of good quicklime with 200 cc. hot water. Stir the mixture for some time and then let it stand quietly for about twelve hours. Then draw off the clear hydrosulfite solution and add for every liter 5 cc. of NaOH solution of sp. gr. 1.38 (76 Tw.). The solution should show a distinctly alkaline reaction. If it does not, add a little more NaOH solution. Standardization of Hydrosulfite Solution. Weigh out accu- rately 1 gram of pure standard indigo powder of known indigotin content and sulfonate as above described under " sulfonation." Make this up to 1 liter with distilled water. Pipette 100 cc. of this solution into a flask and run in from a burette the hydro- sulfite solution until all the blue color has disappeared. A stream of CO2 or illuminating gas must be run into the titrating flask to expel all air and have a reducing atmosphere before the titration is begun. The tip of the burette should be drawn out to a point about 8-10 cm. long. Have this tip below the surface of the liquid until the blue color has nearly disappeared. Then raise GENERAL ORGANIC ANALYSES 99 the tip and run in, drop by drop, the hydrosulfite solution until all the color has disappeared. This must be done as rapidly as possible to prevent oxidation. From the titration calculate the strength of the hydrosulfite solution in terms of indigotin. Then dilute so that it requires 25-30 cc. to decolorize 100 cc. of the standard indigotin solu- tion. Retitrate this corrected solution and determine the exact strength in terms of indigotin. Before making the final titra- tion, place the hydrosulfite solu- tion in a 2-liter bottle and pour in sufficient benzol to form a layer 1-2 cm. deep to exclude the air. Also connect up the bottle so that a stream of illu- minating gas can be passed into it. (See Fig. 4.) TITRATION OF SULFONATED SAMPLE. Pipette into the titrat- ing flask an aliquot of the solu- tion prepared as described under " sulfonation " which will cor- respond to about 0.1 gram of indigotin, and titrate with the hydrosulfite solution exactly as in the standardization. This should require at least 25 cc. of hydro- sulfite solution. If less is required, repeat using a larger aliquot. Always run the titration in duplicate. NOTE. The titration must be made as rapidly as possible and at the same time great care must be used not to overrun the end point. If the titration is carried out too slowly, there is danger of the indigo solution oxidizing back to indigo blue and the results will be too high. With a little experience there should not be much difficulty in determining the end point, especially with synthetic indigoes, but with the impure natural indigoes there is apt to be some trouble in determining the exact end point and it is best, especially if one is not used to the titration, to run several titrations and report the average. Mineral Matter (Ash). Weigh accurately 1 gram of the dried powder into a platinum dish and ignite gently until all organic FIG. 4. Apparatus for Titrating Indigotin. 100 TECHNICAL METHODS OF ANALYSIS matter is burned off. As calcium carbonate is generally present care should be taken not to ignite sufficiently to drive off the CC^ For more accurate work blast the ash, determine quantitatively the CaO, calculate it to CaCOs and correct the mineral matter accordingly. REFERENCE. Badische Anilin- u. Soda-Fabrik: "Indigo Pure," page 26. NICOTINE IN TOBACCO AND TOBACCO EXTRACT General. The samples should be analyzed as received as any attempt at artificial drying is likely to cause loss of nicotine. (If the tobacco is too moist to be ground it may be dried at a tem- perature not exceeding 60 C.) The material should be ground so as to pass through a No. 20 sieve or finer. Kissling Method.* (A) SOLUTIONS REQUIRED. (a) Alcoholic NaOH. Dissolve 6 grams of NaOH in 40 cc. of water and 60 cc. of 90% alcohol. (6) 0.4% Sodium Hydroxide. Dissolve 4 grams of NaOH in 1 liter of water. (c) Sulfuric Acid. Approximately 0.1 N solution, carefully standardized. (d) Phenacetolin Solution. Make a 0.5% solution in alcohol. (e) Cochineal Solution. Digest with frequent agitation 3 grams of pulverized cochineal in a mixture of 50 cc. of 95% alcohol and 200 cc. of water for one or two days at room temperature and then filter. (B) DETERMINATION. Weigh into a small beaker 5-6 grams of tobacco extract or 20 grams of finely powdered tobacco, previously dried at 60 C., if necessary. Add 10 cc. of the alcoholic NaOH solution and follow, in the case of tobacco extract, with enough pure powdered CaCOs to form a moist but not lumpy mass. Mix the whole thoroughly. Transfer to a Soxhlet extractor and ex- haust for about five hours with ether. Evaporate the ether at a low temperature, and take up. the residue with 50 cc. of the 0.4% NaOH solution. Transfer this residue by means of water to a 500 cc. Kjeldahl flask, and distill with steam, using a condenser through which water is flowing rapidly. Use a 3-bend outflow * Official method of the Assoc. of Official Agr. Chemists; see its Journal, II, Methods of Analysis (1916), page 73. GENERAL ORGANIC ANALYSES iVj ; ', tube and a few pieces of pumice and a small piece of prevent bumping and frothing. Continue distillation till all the nicotine has passed over, the distillate usually varying from 400 to 500 cc. When completed, only about 15 cc. of the liquid should remain in the distillation flask. Titrate the distillate with 0.1 N H 2 SO4, using phenacetolin or cochineal as indicator. One molecule of H 2 S04 is equivalent to 2 molecules of nicotine. CALCULATION. 1 cc. 0.1 N acid = 0.01622 gram nicotine (Ci H 14 N 2 ). Silicotungstate Method * (Bertrand and Javillier). Distill in a current of steam as in the Kissling method and precipitate the nicotine from the distillate as nicotine-silicotungstate, a rose-white salt with the following formula: 2Ci Hi 4 N 2 - 2H 2 OSiO 2 12WO 3 5H 2 O. This salt on being ignited leaves a residue of Si0 2 and WOs, from which is calculated the weight of nicotine originally present. (A) REAGENTS. (a) Silicotungstic Add Solution. Prepare a 12% solution of the silicotungstic acid having the following for- mula : 4H 2 OSi0 2 12W0 3 22H 2 O. (b) Sodium or Potassium Hydroxide Solution (1 : 2). (c) Dilute Hydrochloric Acid (1 : 4). (B) DETERMINATION. Weigh such an amount of the prep- aration as will contain preferably between 0.1 and 1.0 gram of nicotine (if the sample contains very little nicotine, about 0.1%, do not increase the amount to the point where it interferes with the distillation) ; wash with water into a 500 cc. round-bottomed dis- tillation flask; add a little paraffin to prevent frothing, a few small pieces of pumice and a slight excess of the NaOH or KOH solution, using phenolphthalein as an indicator. Distill rapidly in a cur- rent of steam through a well-cooled condenser, connected by an adaptor with a suitable flask containing 10 cc. of the dil. HC1. When distillation is well under way, heat the distillation flask to reduce the volume of the liquid as far as practicable without bump- ing or undue separation of insoluble matter. Distill until a few cc. of the distillate show no cloud or opalescence when treated * Official method of the Assoc. of Official Agr. Chemists; see its Journal, II, Methods of Analysis (1916), page 74. 102, > t ; TECHNICAL METHODS OF ANALYSIS w?th ^ drop, of silicotungstic acid and a drop of the dil. HC1. Confirm the alkalinity of the residue in the distillation flask with phenolphthalein. Make up the distillate, which may amount to 1000-1500 cc. to a convenient volume (the solution may be concentrated on the steam bath without loss of nicotine), mix well and pass through a large dry filter if not clear. Test a por- tion with methyl orange to assure its acidity. Pipette an aliquot, containing about 0.1 gram of nicotine, into a beaker (if the sam- ples contain very small amounts of nicotine, an aliquot containing as little as 0.01 gram of nicotine may be used), add to each 100 cc. of liquid 3 cc. of the dil. HC1, or more if the necessity is indicated by the test with methyl orange, and add 1 cc. of silicotungstic acid solution for each 0.01 gram of nicotine supposed to be present. Stir thoroughly and let stand overnight. Before filtering, stir the precipitate to see that it settles quickly and is in crystalline form; then filter on an ashless filter paper, and wash with cold dil. HC1 (1 : 1000). Transfer the paper and precipitate to a weighed platinum crucible, dry carefully, and ignite until all carbon is destroyed. Finally heat over a Teclu or Meker burner for not more than ten minutes. The weight of the residue multiplied by 0.114 gives the weight of nicotine present in the aliquot. Ash. Incinerate 2 grams of the sample in a weighed platinum dish. Moisten with water, dry and again ignite. Repeat until no more particles of carbon remain. Cool in a desiccator and weigh. Dissolve the ash in hot water and filter. Save the filtrat<~. Ignite and weigh the water-insoluble ash. Cool the filtrate and titrate with 0.1 N acid and methyl orange. Report the alkalinity of the water-soluble ash as the number of cc. of 0.1 N acid required to neutralize the water-soluble ash from 1 gram of the sample. Dissolve the water-insoluble ash in 25 cc. of 0.5 N HC1 and titrate the excess with 0.1 N alkali and methyl orange. Deduct the number of cc. of 0.1 N alkali from 125 and 'divide by 2. This will give the alkalinity of the water-insoluble ash in the same terms as above. (In some cases it may be necessary to use double the amount of 0.5 N acid. In that case subtract the cc. of 0.1 N alkali from 250 and divide by 2.) After titration of the water-insoluble ash, add cone. HC1 and boil. Filter the bulk of the solution through an ashless filter and add cone. HC1 to the residue. Warm, dilute, and filter through GENERAL ORGANIC ANALYSES 103 the same filter. Wash with distilled water, ignite, and weigh as " hydrochloric acid-insoluble ash." NOTES. (1) Cigarette tobacco contains about 1.0-3.3% of nicotine (av. about 1.7%), cigars about 1.5%, chewing tobacco about 1.1%. (2) Cigarette papers. Nearly all cigarette papers have chemical fillers, presumably to improve their burning qualities and color. These generally consist of carbonates and oxides of Al, Ca and Mg. The best papers are made from pure linen sized with starch. A little KNOs is also sometimes added. REFERENCE. Azor Thurston: "Analysis of Cigarettes, Cigars and Tobacco." Bulletin No. 2, Agr. Commission of Ohio, Dairy and Food Div., Bureau of Drugs, Nov., 1914. NICOTINE SOLUTION General. This method is for the analysis of nicotine solutions for use in preparing specially denatured alcohol according to Formula No. 4. The tobacco denaturant must conform to the following analytical requirements: Determination of Nicotine. Measure 20 cc. of the solution into a special 250 cc. dephlegmating flask (Fig. 5) ;* add 10 cc. of 0.1 N KOH; make the liquid up to 50 cc. and distill in a current of steam until the distillate is no longer alkaline (about 500 cc.). Wash the distillate into a 700 cc. flask and fill two other flasks with the same amount of distilled water. Add 8 drops of resa- zurin solution f to each flask as an indicator, and add 1 drop of 0.1 N acid to one blank and 1 drop of 0.1 N alkali to the other. This should give in the case of the acid solution a pink coloration and in the case of the alkali solution a blue coloration by trans- mitted light. Then titrate the nicotine distillate with 0.1 N H2SO4 and make comparisons with the 2 blanks until the nicotine distillate shows an end point of all red and no blue on transmitted light. This end point is difficult to detect without some practice and it should be approached with care. The best way to observe the final end point is to place all three flasks in a line on white * United States Internal Revenue Regulations No. 30 specifies a Kjeldahl flask fitted with a suitable bulb tube, but we have found the special flask more convenient and equally accurate. f Resazurin Solution: Dissolve 0.2 gram of the crystals in distilled water, add 40 cc. of 0.1 N NH 3 solution and dilute to 1 liter. 104 TECHNICAL METHODS OF ANALYSIS paper and have a sheet of white paper back of them. A com- parison of color can be made much more easily in this way. Not less than 23.2 cc. of 0.1 N H2S04 should be required for the neu- tralization. (This is equivalent to 1.88% of nicotine, which is the minimum amount that is demanded in the tobacco denaturant.) CALCULATION. 16218 Cc. of 0.1 N H2S04 required X =per cent of nicotine. Fio. 5. Flask for Nicotine Determination. Test of Coloring Matter. The original formula for nicotine denaturant required that it be colored with a mixture of blue and yellow dyes, thus giving a green solution. On August 11, 1916, due to the scarcity of dyestuffs, the Treasury Department ruled that the yellow dye may be omitted and the denaturant colored only with methylene blue. In case the denaturant is of a green color, proceed as follows for the test of coloring matter: Take 1 cc. of the denaturant and make up to 100 cc. with water, acidulating with a few drops of H2SO4. Immerse in this solution a piece of white cotton cloth and boil the solution. Continue the process, adding more cloth and more water if necessary, until GENERAL ORGANIC ANALYSES 105 all the blue color in the solution is fixed on the cloth. Then add a piece of white woolen cloth and boil the bath as before until all the yellow color is fixed upon the cloth. Both the cotton and woolen cloths should show decided colors the cotton blue and the woolen yellow. If the denaturant is blue, no dyeing test is required. Intensity of Color. If the denaturant is green, it must contain sufficient coloring matter so that when observed in an eighth-inch cell of Levibond's tintometer it will show a color of an intensity not less than No. 24 yellow combined with No. 3 blue. If the denaturant is blue, proceed as follows: Dilute 1 cc. of the denaturant material with 100 cc. of water and compare 50 cc. of this solution in a 50 cc. Nessler tube with 50 cc. of a solution containing 5 grams of c. P. copper sulfate (OuSC^-SH^O) in 100 cc. of water. The color of the diluted nicotine solution should be at least as intense as that of the copper solution. NOTES. (1) The original regulations called for the use of rosolic acid as indicator. Resazurin, however, gives a better end point and hi a letter from the Commissioner of Internal Revenue at Washington, dated September 4, 1913, the use of the latter was sanctioned. (2) Resazurin crystals are made by Theodore Schuchardt, Corlits, Ger- many, and may be obtained from Eimer & Amend, New York, N. Y. REFERENCES. Regulations No. 30 Revised, United States Internal Rev- enueJuly 15, 1907. Treasury Decision (1223), September 5, 1907. CHAPTER IV ANALYSIS OF METALS SAMPLING IRON AND STEEL General. Select at random a sufficient number of bars, rods, rails, I-beams, or car axles, etc., as the case may be, to represent the shipment and carefully mark them. From each of these pieces procure a sample composed of mixed borings weighing between 1 and 2 ounces. Location of Borings. In order to eliminate as far as possible the effect of possible segregation, take the borings at different points of the piece, selection being made with reference to the rapidity with which the different portions may have cooled. This is preferably done on a cross-section of the piece. For example in sampling a steel rail, make one boring close to the outside of the head, one at the center of the head, another at the middle of the web, and one near the center of the foot. Where access to a cross-section is impossible, take borings from the outside which will represent as nearly as possible the desired locations. Cleaning the Surface. When borirgs are to be made from a longitudinal surface, remove all rust and scale by means of an emery wheel, emery cloth, or other abrasive; in no case, however, should any considerable portion of the first borings be discarded. When drilling into a cross section, discard enough of the borings to insure absence of rust or dirt from the portion taken. Size of Borings. In order that a perfect rr ixing of the samples from different parts of the piece may be pcssitle, it is desirable to have borings in the form of small thin chips, rather than long, spiral turnings. This may be accomplished by slightly dulling the cutting edge of a half-inch twist drill or by using a half-inch flat drill. Freedom from Contamination. It is absolutely essential that the utmost care be taken to prevent contamination of the sample 106 ANALYSIS OF METALS 107 with grease, oil, or dirt of any kind. Brush the piece of iron or steel to be sampled free from any loose material which may adhere to it; collect the borings on a clean white sheet of paper. Use no oil, grease, or other lubricating substance on the drill. Place the combined borings from each piece in clean, wide- mouthed, glass-stoppered bottles; stopper tightly and carefully mark each bottle for identification. CARBON STEEL General. This method covers the analysis of plain carbon steels, i.e., those containing only C, Mn, P, Si and S. In those cases where an alternative method is given, the latter is to be used only when, on account of lack of chemicals or appa- ratus, the first method cannot be employed. Preparation of Sample. The points to be observed in sampling steel for analysis are as follows: (1) Samples must be. drillings or chips cut by some machine tool without the application of water, oil, or other lubricant, and free from scale, grease, dirt or other foreign substance. If sam- ples are taken by drilling, the diameter of the drill should be not less than 0.5 nor more than 0.75 inch. (2) Samples must be uniformly fine and well mixed before analysis. (3) If the composition of the surface of a sample piece has been altered by case hardening or decarbonization, do not include in the sample for analysis drillings which represent this outer surface. (4) Heat-treated samples should be annealed before sampling. (5) Round or flat samples up to and including 1 inch in thick- ness must be drilled through the entire thickness of metal; or, if the sample is taken on a milling machine, it must be taken by machining off the entire cross section. (6) From material having a cross section larger than 1 inch samples should be taken at any point midway between the outside and the center by drilling parallel to the axis. If this is not prac- ticable, the piece may be drilled on its side, but do not collect the drillings until they represent the above-named portion of the sample. 108 TECHNICAL METHODS OF ANALYSIS (7) Whenever a sample is received in the form of drillings, they should always be examined for oil and other foreign matter. Any oil must be completely removed with ether and the drillings then thoroughly dried. If foreign particles are present, go over the sample with a magnet and use only that portion for analysis which is picked up by it. Total Carbon. Determine carbon by direct combustion in the electric furnace. The general arrangement of the combustion apparatus is as follows (see Fig. 6) : h g FIG. 6. Apparatus for Determining Carbon in Steel. (a) Oxygen tank. (6) Safety jar for H 2 SO 4 . (c) CaCl 2 jar. (d) Soda-lime jar. (e) Raskins' multiple tube electric furnace. (/) U-tube for granular Zn. (g) U-tube for CaCl 2 . (h) Soda-lime absorption tube for CO 2 . (f) Bottle for Ba(OH) 2 solution (exhaustion and gas speed indicator). The combustion apparatus should be tested for leaks from time to time (see Blair: " Chemical Analysis of Iron," page 138, Fig. 63). Before weighing the absorption tubes in the morning ANALYSIS OF METALS 109 connect them to the combustion apparatus and pass oxygen through for ten minutes, weigh and again pass oxygen through for thirty minutes and weigh. This should be continued until the weight is constant. For steels containing 0.30-1.50% carbon, take 2 grams of fine drillings not over 0.25 mm. thick. For lower percentages of C, take 3-5 grams of drillings 'bet ween 20- and 60-mesh size. Trans- fer the sample to an alundum or clay boat, protected by a layer of ignited A^Os or alundum. Place the drillings in as compact a mass as possible. If curly drillings are scattered along the entire length of the boat instead of being put in a deep, compact body, borings that are a little thick will frequently be found to contain still unburned metal. Drillings lying in close contact heat each other to incandescence during burning with oxygen. Introduce the boat with the charge into the center of the combustion tube, maintained at a temperature of 1000 C. Admit oxygen at the rate of 25 bubbles per ten seconds or 250 cc. every ten minutes. As soon as the steel begins to burn, there is at first a rapid evolution of gas, which quickly ceases. When oxida- tion of the charge is completed, oxygen begins to flow at normal speed again. Allow thirty minutes for oxidation of the sample and sweeping of all C02 from the combustion tube into the absorp- tion apparatus. Weigh the absorption tube and from the weight of CO2 calcu- late the per cent of C in the sample. Wt. of C0 2 CALCULATION. X 27.28 = per cent C. Wt. of sample NOTES. (1) In case of very heavy drillings, those that will not pass a 20- mesh sieve, the analysis should be made by solution. Dissolve 2-5 grams of steel, using 75 cc. per gram of a solution of copper potassium chloride. Follow the procedure described for the determination of Total Carbon in Pig and Cast Iron on page 130. This method, however, is not reliable for alloy steels and should be used on plain steels only when drillings cannot be burned 3ompletely by direct combustion. (2) After refilling the CaCl 2 U-tube (g) the contents should be saturated with CO 2 , as the fresh drier may absorb CO 2 and thereby cause lower results it first. (3) Soda-lime should be hydroxides of Na and Ca. The most satisfactory soda-lime for carbon combustions in steel is 12-mesh size and contains 15% )f H 2 0. (4) Soda-lime absorption tubes should be filled three-quarters full with 110 TECHNICAL METHODS OF ANALYSIS soda-lime and the remainder with CaCl 2 . The CaCl 2 will absorb any moisture which might be liberated by the soda-lime during absorption- of CO 2 . (5) Soda-lime tubes will absorb from 0.6-0.7 gram of CO 2 before it is necessary to renew the chemicals. Saturation will be indicated by precipita- tion of BaCO 3 in the Ba(OH) 2 solution (i). (6) The same grade of CaCl 2 and of soda-lime should be used in the train preceding and following the combustion furnace. Graphitic Carbon. Weigh 10 grams of steel in a 200 cc. beaker and add 150 cc. of HNOs (1 : 3). Heat the beaker on the steam bath until the steel is all dissolved. Filter on a loose-bottom Gooch crucible through asbestos and wash with water, HC1 (1:1), H 2 0, NH 4 OH (1%), H 2 0, HC1 (1 : 1), H 2 0, and again H 2 O. Dry for one hour at 100 C. Transfer the asbestos with the carbon to the combustion boat and determine the carbon as under total carbon. Combined Carbon. This is the difference between the total and the graphitic carbon. Manganese (Bismuthate Method). Titration with sodium arsenite (Method I, below) is sufficiently accurate for ordinary pur- poses. In case of disputes, however, or where extreme accuracy is desired, the titration should be made with ferrous ammonium sulfate and KMnCU (Method II). The procedure up to the point of titration is the same in either case, as follows: Dissolve exactly 1 gram of drillings in 50 cc. of HNO 3 (1 : 3) in a 200 cc. Erlenmeyer flask, cool and add about 0.5 gram of sodium bis- muthate. Boil until the pink color has disappeared and dissolve any precipitate of MnO 2 by adding a few drops of a saturated solution of FeS04 or of Na 2 S 2 Os; then heat until all nitrous oxide fumes have been driven off, cool to 60 F., add 1 gram of Na bismuthate and shake the flask vigorously for a few minutes. Add 50 cc. of 3% HNOs* and filter the solution into a suction flask through an extra porous alundum thimble. The thimble should not be filled so full that any of the solution comes in con- tact with the rubber connection. Wash with 50-100 cc. of the same acid and finally with water. Titrate the filtrate by one of the following methods: * This 3% HNO 3 should be prepared by adding 60 cc. of cone. HNO 3 to 1940 cc. of H 2 O; then add 4 or 5 grams of bismuthate, shake and let stand overnight before using. This destroys lower oxides of nitrogen which have a tendency to cause low results. ANALYSIS OF METALS 111 METHOD I. Titrate the filtrate in the flask with standard sodium arsenite solution until 2 drops produce no further change in color. Until the end point is reached, the arsenite solution will produce a clear spot where it strikes the liquid. Sodium Arsenite Solution. For each- liter dissolve 2.5 grams of anhydrous Na2COs in distilled water, add 1 gram of c. P. As20s to the hot Na2COs solution and boil until the As2Os is dissolved. Cool and dilute to 1 liter. STANDARDIZATION. Measure into a 300 cc. Erlenmeyer flask exactly 20 cc. of 0.05 N KMnO4 solution. Dilute to 150 cc., add 15 cc. of cone, water-white HNOs and titrate with arsenite solution in the same manner as above. Make the standardization in triplicate, and if agreement is satisfactory, take the average. Cal- culate the Mn factor of the solution. 0.0110 CALCULATION. Mn factor = . cc. of arsenite used METHOD II. Add to the filtrate in the flask an excess of stand- ard ferrous ammonium sulfate solution, then titrate back the excess with 0.05 N KMnQt solution to the appearance of a faint permanent pink color. CALCULATION. Run a blank, exactly as in the regular deter- mination, to obtain the equivalent of the KMnCX solution in terms of the ferrous ammonium sulfate solution. Subtract the number of cc. of KMnCX solution used in the back titration from the KMnQi equivalent of the ferrous ammo- nium sulfate added. This gives the volume of KMnO* required by the Mn in the sample, which, multiplied by the Mn factor of the KMnOi, gives the amount of Mn in the sample. Standard Solutions. (a) 0.05 N KMnCU: Dissolve 1.58 grams of pure KMnO4 in water and dilute to 1 liter. (b) Ferrous Ammonium Sulfate: Dissolve 20 grams of Fe(NH4) 2 (SO4)2-6H 2 in water, add 50 cc. of cone. H 2 S0 4 and dilute to 1 liter. STANDARDIZATION. Standardize the KMnCU solution against standard sodium oxalate of the U. S. Bureau of Standards as follows: In a 300 cc. Erlenmeyer flask dissolve 0.150 gram of sodium oxalate in 125 cc. of hot water (80-90 C.) and add 10 cc. of H 2 SO 4 (1:1). Titrate at once with 0.05 N KMn0 4 , stirring 112 TECHNICAL METHODS OF ANALYSIS the liquid vigorously and continuously. The KMn(>4 must not be added more rapidly than 10-15 cc. per minute, and the last 0.5-1 cc. must be added dropwise, with particular care to allow each drop to be fully decolorized before the next is added. The solution should not be below 60 C. when the end point is reached. 0.0246 CALCULATION. Mn f actor = T ^ -. cc. KMnO* used If the Fe factor of the KMn04 has already been determined against Na2C2Oi as above, the Mn factor may be calculated from the Fe factor by multiplying the latter by 0.1967. Phosphorus. Dissolve 2 grams of the sample in a 500 cc. Erlenmeyer flask in 60 cc. of HN0 3 (1 : 3). To the boiling solution, free from red fumes, add 5 cc. of saturated KMnO4 solution. Con- tinue boiling until the pink color disappears. If no precipitate of brown oxide of manganese remains, add a little more KMnOi. When a permanent brown precipitate does remain, cool the flask and add a crystal of tartaric acid. Boil this solution until clear, remove the flask from the hot plate and after two or three minutes add 15 cc. of cone. NILiOH. Bring back with a little HNO 3 if a precipitate of Fe(OH) 3 remains. Warm to 80 C., add 60 cc. of molybdate solution and shake five minutes. Let stand 0.5 hour, or until the precipitate settles. Filter at once through an 11 cm. filter paper, wash the yellow precipitate five times with 2% HN0 3 , then with 1% KN0 3 solution (NaN0 3 must not be used) until free from add (approximately 15 times). Place the filter and contents in the original flask, which has been thoroughly rinsed with water, add approximately 50 cc. of cold distilled water and a measured excess of standard NaOH from a burette, 5 cc. at a time, in sufficient amount to completely dissolve the yellow precipitate. Cork the flask and agitate vio- lently until the filter paper is disintegrated. Add 3 drops of 1% phenolphthalein solution with a medicine dropper and titrate with standard HNO 3 to the disappearance of the pink color. CALCULATION. Run a blank titration the same way as in the regular determination to obtain the value of the NaOH solution in terms of the HNO 3 solution. Subtract the number of cc. of HN0 3 used from the number corresponding to the volume of NaOH added. The difference is the cc. of HNO 3 equivalent to the P in the sample. This multiplied by the value of the HNOs ANALYSIS OF METALS 113 in terms of P gives the weight of P in the sample. This weight divided by the weight of steel taken X 100 = per cent P. SOLUTIONS. (A) Molybdate Solution: (1) 151 grams of MoO 3 (85%); 600 cc. of water; 150 cc. of NH 4 OH (cone). (2) 1000 cc. of water; 675 cc. of HNOs (cone.); 800 cc. of solution (1). To prepare solution (1) pour all of the MoOs into the water. Shake to get in suspension and> before it settles, quickly pour the NH4OH into the bottle. Shake until the powder is dissolved. To prepare solution (2) mix the acid and water and cool thoroughly. Then by means of a funnel pour in rapidly all of solution (1). Shake and add a few crystals of sodium or ammonium phos- phate. Shake and let stand overnight before using. Use only the clear supernatant liquid. (B) Standard Nitric Acid: Dilute 10 cc. of cone. HNO 3 to 1 liter. (The HNOs must be water white.) (C) Standard Sodium Hydroxide: Dissolve 8 grams of NaOH in 400 cc. of water, add sufficient Ba(OH) 2 solution to precipitate all carbonates, filter at once and make up to 1 liter. (D) 1 % KNOz Solution (for washing) : 10 grams per liter. (E) 2% Nitric Add (for washing): 20 cc. of cone. HNOs per liter. STANDARDIZATION. First, determine the strength of the HN0 3 in terms of the NaOH. Second, determine the strength of the HNOs in terms of phos- phorus by titrating 0.5 gram of pure yellow ammonium phos- phomolybdate by the above method. (The phosphomolybdate should be dried for one hour at 100 C. before using.) _, , 0.00825 CALCULATION P f actor = TTXT/^ j- cc. HN0 3 used The standard ammonium phosphomolybdate may be prepared as follows: Acidify a dilute solution of Na2HPO^ with HNOs,. add an excess of MoOs solution, filter, wash the precipitate thoroughly with hot water, dry at 150 C. and keep in glass-stoppered vial. Determine the content of phosphorus by the pyrophosphate 114 TECHNICAL METHODS OF ANALYSIS method.* Pure ammonium phosphomolybdate contains 1.65% of phosphorus. NOTES. See under the Alternative Method below. Phosphorus (Alternative Method). Instead of titrating the yellow precipitate, filter on a 9 cm. filter paper which has been previously dried and weighed in a glass-stoppered weighing bottle. Wash the precipitate at least 5 times with 2% HNOa to remove all Fe salts. Absorb with a blotter the excess of moisture from the paper and precipitate and dry in the oven at 100 C. for one hour. Weigh the paper and contents in the weighing bottle. Using 2 grams of steel, the weight of precipitate in milligrams X 0.00825X100= per cent phosphorus. NOTES. (1) In neutralizing solutions to which NH 4 OH has been added, HNO 3 should be added until the solution is straw color, otherwise a small amount of iron may separate causing high results. (2) It is essential that the yellow precipitate be washed with solutions of exactly the strength specified and until free from acid. (3) When running a number of samples, the flask containing the filter and yellow precipitate must be kept tightly corked, and alkali should not be added until the remainder of the operation can be promptly carried through. (4) When the alternative method of weighing the yellow precipitate is used, care must be observed that the filter paper is previously dried and weighed in a weighing bottle and that the final weight of filter and contents is made in exactly the same manner as the first weight of the filter. Silicon. Weigh 4 grams of steel into a 12 cm. casserole or evaporating dish and add 60 cc. of silicon mixture (see below). When effervescence ceases, rinse the sides of the dish with water and cover with a watch glass. Place on the hot plate and boil slowly but continuously to dryness. Bake to dehydrate SiO2, and cool. Take up with 75 cc. of HC1 (1 : 3) and bring to a boil. Keep at the boiling point until the solution is clear. Filter while hot through an 11 cm. filter, washing alternately with hot HC1 (1:1) and hot water until free from Fe, then wash the paper free from acid. Ignite in a platinum crucible and weigh as SiC^. Calculate to Si. (If the precipitate is red, volatilize with HF. and weigh again. The loss in weight is SiCb. Calculate to Si.) CALCULATION. SiO 2 X 0.4693 = Si. * Dissolve a weighed amount of the yellow precipitate in NH 4 OH and precipitate with excess of magnesia mixture. Filter, ignite and weigh as Mg2P 2 O 7 . (See page 527.) ANALYSIS OF METALS 115 Silicon Mixture. Into 1500 cc. of water pour 500 cc. of cone. HNOs, and then 150 cc. of cone. H 2 SO4- CAUTION. Mix in the above order, adding the H 2 SO 4 very slowly, and let cool before shaking. Sulfur (Evolution Method). Weigh 5 grams of steel into a 500 cc. Florence flask provided with a double perforated stopper carrying a 2-bulb safety tube, extending below the surface of the liquid, and an exit tube for gas, the latter connected to a delivery tube extending to the bottom of a tall glass tumbler. Place 30 cc. of CdCl2 solution in the tumbler and fill two-thirds with water. Add 100 cc. of HC1 (1 : 1) to the drillings in the flask. Place the flask over an Argand burner, heat gently at first, and, when the drillings are dissolved, boil until the steam drives the last trace of [28 from the evolution flask. Disconnect the flask, add 2 cc. of starch indicator * and 30 cc. of cone. HC1 to the solution in the tumbler, titrating at once with standard iodine solution to the appearance of a permanent blue. From the factor of the iodine solution calculate the amount of S. SOLUTIONS. (a) Cadmium Chloride Solution: Dissolve 30 grams of CdCl 2 in 300 cc. of H 2 O and add this to 800 cc. of H 2 O and 1200 cc. of NH 4 OH. (6) Iodine Solution: Dissolve 4 grams of iodine and 8 grams of KI in a little H 2 O and make up to 1 liter. STANDARDIZATION. The iodine solution should be standardized against 10 cc. of 0.1 N sodium thiosulfate solution, very accurately pipetted out, and the factor calculated so that it will give per cent S on a 5-gram sample. CALCULATION. 1 cc. of 0.1 N Na 2 S 2 O 3 = 0.001603 gram S. Then S factor = X 10 X cc. of iodine used 5 0.003206 cc. of iodine used NOTES. (1) Care must be taken to prevent drillings from sticking to the neck of the wet flask. (2) When all H 2 S has been driven over, the tube extending into the tumbler will become hot from condensed steam, and further heating should be avoided. * See page 12. 116 TECHNICAL METHODS OF ANALYSIS (3) In cases of extreme accuracy the aqua regia method should be used. (See Blair, 7th Edition, page 66.) (4) In case of very coarse drillings, which dissolve slowly, considerable acid will be carried over and neutralize the NHs in the CdCl2 solution. In such cases care must be taken to keep the solution up to its original strength in NHa. If such samples are encountered it is preferable to use the Elliott Method as described on page 132. REFERENCES. Blair: "The Chemical Analysis of Iron"; Lord and Demorest: "Metallurgical Analysis." ALLOY STEEL General. This method covers the analysis of alloy carbon steels, i.e., those containing, in addition to C, Mn, P, Si, and S, one or more of the following: Cu, Ni, Cr, V, W, Mo and Ti. In those cases where an alternative method is given, the latter is to be used only when, on account of lack of chemicals or appa- ratus, the first method cannot be employed. Sampling. The points to be observed in sampling alloy steels for analysis are the same as those mentioned under Carbon Steels on page 107. Total Carbon. Use the procedure for Carbon Steels, page 108, omitting Note 1. Manganese (Bismuthate Method). Dissolve 1 gram of steel in 60 cc. of HC1 (1:1). When solution is complete add HNOs, drop by drop, to oxidize the Fe. Add an equal volume of EbO and boil ten minutes. Filter and wash any WOs with dilute HC1 (1 : 7). Replace HC1 in the filtrate with HNO 3 by evaporating nearly but not quite to dryness, removing the last traces with AgNO 3 . Then add 75 cc. of HNO 3 (1 : 3). Cool and proceed according to the Bismuthate method for Manganese in Carbon Steel, page 110, titrating with standard sodium arsenite solution as there described. Manganese (Alternative Method). In case the Cr content of the steel is over 2%, proceed as follows : Dissolve 1 gram of steel in 60 cc. of HC1 (1:1). Proceed as in the bismuthate method just described until the last traces of chlorides have been replaced by HNOs by evaporation. Add 100 cc. of colorless cone. HNOa, set on the hot plate and heat to incipient boiling. Then drop in powdered NaClOa or KClOa, a little at a time, adding each portion ANALYSIS OF METALS 117 when the effervescence produced by the preceding portion has ceased. By the time 2-2.5 grams have been added the Mn02 will have separated as a fine brown powder. Add 0.5 gram more of chlorate and boil gently ten minutes. If any SiC>2 is present in the solution after three or four minutes' boiling, add a few drops of pure HF. Then add 1 gram more of the chlorate and 25 cc. of cone. HNOs and boil ten minutes longer. Remove from the plate and cool by setting the beaker in water. When the MnC>2 has settled, filter without dilution through a Gooch crucible with a removable bottom and asbestos mat. Finally transfer the MnC>2 to the filter and wash the beaker and filter mat with color- less * cone. HNOa 3 or 4 times, or until the filtrate is colorless. This can be done without using more than 1520 cc., by adding only a little each time and letting each portion run through before adding the next. Finally wash with a little cold water. After washing the Mn(>2 with cold water till free from acid (letting each successive portion of water run entirely through before adding the next, so as not to use in all more than 20 cc.), wash the asbestos and precipitate back into the beaker, which always has some MnC>2 adhering to it. To the asbestos and MnC>2 in the beaker add standard ferrous ammonium sulfate solution (see note) from a burette, 5 cc. at a time, until after stirring and warming the MnC>2 is completely dissolved. Break up with a glass rod all lumps of asbestos and precipitate. Add a little water and run in standard KMn04 solution (see note) till the pink color remains permanent for two or three minutes. Read the burette and deduct the amount of KMn04 from its equivalent of ferrous ammonium sulfate solu- tion. The difference is the amount of standard KMnOi solution equivalent to the Mn present in the precipitate. Multiply this difference by the Mn factor of the KMnC>4 solution and divide by the weight of sample taken. This result multiplied by 100 gives the per cent of Mn in the sample. NOTE. The standard ferrous ammonium sulfate and standard KMnO 4 solution, described under determination of Manganese in the method for Carbon Steels (page 111), should be used for this titration. The Mn factor * If the HNOs is colored by lower oxides of N (from standing and the action of light), it can be purified by blowing a strong current of air through until it becomes colorless. 118 TECHNICAL METHODS OF ANALYSIS is not the same, however, as when the bismuthate method is used. In the bismuthate method the following reaction takes place: In the above reaction 1 Mn is equivalent to 5 Fe and the Mn factor may be calculated from the Fe factor of the KMnO 4 by the following equation : Mn factor = Fe factor X ^77. 279 . 20 = Fe f actor X 0.1967. When the MnO 2 is titrated in the method above described, however, the following reaction takes place : In this reaction 1 Mn is equivalent to 2 Fe and the Mn factor may be cal culated from the Fe factor of the KMnO 4 by the following equation: 54 93 Mn factor = Fe factor X 111.68 = Fe factor XO. 4919. Phosphorus. Dissolve 2-5 grams of steel in HNOs (1 : 3) in a porcelain dish, evaporate to dryness and ignite to dull red; cool, dissolve in HC1 (1:1), filter and convert to nitrate by evaporating nearly to dryness with HNOs several times. Proceed as under " Phosphorus " in Carbon Steel, page 112. Silicon. In alloy steels containing no tungsten determine Si as in Carbon Steels, page 114. NOTES. (1) Silicious residues are likely to be contaminated with Ti and Al and should be evaporated with HF and H 2 SO 4 . (2) When W is present, the WO 3 and SiO 2 are weighed together and sep- arated as described later under Tungsten (page 126). Sulfur. Dissolve 5 grams of steel in 50 cc. of cone. HC1 and 50 cc. of cone. HNOs. Evaporate to about 50 cc. and cautiously add 2 grams of Na2COs. Transfer the solution to a 5-inch por- celain dish and evaporate to dryness. Dissolve the residue in 100 cc. of HC1 and evaporate to 10 cc., then add 100 cc. of H^O and 10 cc. of HC1. Filter, heat to boiling, and precipitate the SOs with 10 cc. of a hot saturated solution of BaCl2, added drop by drop, and let stand overnight. Filter while hot, wash with a little very dilute HC1 and finally with cold water. Dry, ignite ANALYSIS OF METALS 119 and weigh as BaS04. If this ignited precipitate is reddish in color, it shows that Y^Os has been carried down with the BaSO4. In this case fuse with Na2COs, dissolve in water, filter, acidulate the filtrate with HC1, boil and proceed as before. Calculate the BaSC>4 toS. CALCULATION. BaSO 4 X 0.1373 = S. Copper. Dissolve 5 grams of sample in a mixture of 150 cc. of water and 12 cc. of cone. H 2 SO 4 . Dilute to about 500 cc. with hot water, heat to boiling and add 3 grams of sodium thiosulfate dis- solved in 10 cc. of hot water. Boil a few minutes, let the precip- itate settle, filter and wash with hot water. Dry the precipitate (which besides the CuS may contain graphite, silica, etc.), transfer to a small beaker, burn the filter and add the ash to the main portion. Digest the whole with aqua regia, dilute with hot water, filter, wash, add a few drops of H^SCU, evaporate to fumes of SOa, cool, dissolve in water, add 15 cc. of cone. HNOs, dilute to about 250 cc. and determine the Cu by electrolysis. (See page 145.) NOTES. (1) In the case of a steel insoluble in acid or containing tungsten, treat as described for Manganese (Bismuthate method) and filter off the WO 3 . To this filtrate add 12 cc. of cone. H^SCX, evaporate to fumes of SOs, dilute to 500 cc. with hot water and proceed as described above. (2) Instead of determining the Cu by electrolysis, it may be determined as CuO. Dilute the sulfate solution, obtained by any of the methods men- tioned above, with water to about 50 cc., add an excess of NH 4 OH, filter, wash with ammoniacal water, and pass H^S through the cold solution. Filter, and wash with H 2 S water. Dissolve the sulfide in aqua regia in a small por- celain dish, evaporate nearly to dryness, dilute with hot water, heat to boiling and add a slight excess of a dilute solution of NaOH or KOH. Filter on a small ashless filter, wash with hot water, dry, transfer the precipitate to a platinum crucible; burn the filter and add its ash to the precipitate; moisten the whole with HNO 3 , and heat, very gently at first, increasing the heat slowly to redness. Cool, and weigh as CuO. Calculate to Cu. CALCULATION. CuO X0.7989 = Cu. Nickel. Dissolve 1 gram of steel drillings in a 150 cc. beaker with 20 cc. of HC1 (1 : 1). When action ceases, add 10 cc. of HN0 3 (1:1). Boil until red fumes have been driven off, add 100 cc. of citric acid solution, dilute to 300 cc., and add with a pipette exactly 5 cc. of standard AgNOs solution. Now add just suf- ficient NILtOH to destroy the cloudiness, then 2 cc. of KI solu- tion, and titrate with standard KCN solution to the disappear- ance of turbidity. The end point is easy for an experienced operator 120 TECHNICAL METHODS OF ANALYSIS to detect, and is not reached as long as a drop, when striking the solution, produces a spot clearer than the liquid around it. As soon as the end point is reached all turbidity will have disappeared. The cyanide first reacts with the nickel, then attacks the iodide. If it is thought that the end point is passed, add a measured amount of AgNOs until a turbidity just appears. It is best to have another beaker containing a solution to be titrated, and to which no AgNOs has been added, placed beside the one being titrated so as to have a clear solution of the same color to compare with. If the citric acid was dirty, the solutions will be cloudy and should be filtered before AgNO 3 is added. SOLUTIONS. (1) Standard Silver Nitrate: Dissolve 2.885 grams of AgNOs in water and dilute to X liter. (2) Potassium Iodide: Dissolve 50 grams of KI in 250 cc. of water. (3) Citric Acid Solution: 380 grams c. P. (NH 4 ) 2 SO 4 ; 270 cc. cone. NH 4 OH; 1430 cc. H 2 O; 240 grams c. p. citric acid. NOTE. In case (NH 4 ) 2 SO 4 is not available the solution may be made as follows : (a) Pour 166 cc. of cone. H 2 SO 4 into 250 cc. of H 2 O. (6) Pour 712 cc. of cone. NH 4 OH into 880 cc. of H 2 O. Then pour (a) into (b) very carefully, as considerable heat is produced; and finally dissolve 240 grams of citric acid in the mixture. (4) Standard Cyanide: Dissolve 8.850 grams of pure KCN (or 9.030 grams of 98% KCN) and 10 grams of KOH in distilled H 2 and dilute to 2000 cc. STANDARDIZATION. The KCN solution is most conveniently standardized against a U. S. Bureau of Standards 3.50% nickel steel, the same procedure being followed as above, each cc. of the above solution being equivalent to approximately 0.10% nickel. An additional titration must be made in order to obtain the relation between the AgNOs and the KCN solution. This is accomplished in the solution after titration for standardization by adding exactly 10 cc. of standard AgNOs to the solution and running in standard KCN to the disappearance of turbidity. One-half of the amount of KCN here required is to be subtracted from the ANALYSIS OF METALS 121 amount used in each determination. Example: Suppose that 10 cc. of AgNOs required 5.2 cc. of KCN to destroy the turbidity. Suppose further that it took 5.6 cc. of KCN in the actual deter- mination. Then 5.6 2.6* = 3.0 cc. KCN actually required by the sample to destroy turbidity. Hence (3.0 cc.XNi Factor) v X100= per cent Ni. (wt. of sample) Assuming the standard steel contains exactly 3.50% Ni, then 0.0350 Ni f actor = (cc. KCN required) (cc. KCN equiv. to 5 cc. AgNOs) NOTES. (1) The presence of sulfates is necessary to obtain a sharp end reaction. Agl is soluble in a large excess of NH 4 OH, so care should be taken to have the solution only slightly alkaline with NH 4 OH. But it must be alkaline. (2) If the titrated solutions are allowed to remain in open beakers for some time a white film forms on the surface, but no account is to be taken of it. 3. When 2% or more Cr is present, proceed exactly as described above except to add 200 cc. of the citric acid solution instead of 100 cc. (4) The AgNO 3 solution used should not be stronger than that indicated above, for when a strong silver solution is used, the Agl, instead of forming a turbid solution, settles out as a curdy precipitate, which does not readily react with the cyanide. (5) Such elements as V, Cr, W, Mo or Mn do not interfere, even when present in large amounts in the sample. Copper, however, is titrated with KCN, but is usually present in negligible quantities. If its presence is sus- pected, nickel should be determined by the dimethylglyoxime method (below). Nickel (Alternative Method). Dissolve 1 gram of the sample in 25 cc. of HC1 (1:1). When solution is complete, add HN0 3 , drop by drop, to oxidize Fe. Add an equal volume of water and boil ten minutes. Filter into a 600 cc. beaker and wash with dil. HC1. Boil, and add 10 grams of citric acid dissolved in 50 cc. of water. Dilute to approximately 300 cc., render slightly alkaline with NttiOH, then add a slight excess of acetic acid and heat to boiling. Now add about 20 cc. of dimethylglyoxime solution, or 5 times as much dimethylglyoxime as there is Ni present. Then add NEUOH until the solution smells slightly of NHa or reacts alkaline. While still hot, filter on a weighed Gooch crucible, * One-half of 5.2 cc. 122 TECHNICAL METHODS OF ANALYSIS wash well, dry at 110-120 C. for forty-five minutes and weigh. Calculate to Ni. CALCULATION. Nickel glyoximeX 0.2031 = Ni. Chromium. Dissolve 1 gram of steel in 20 cc. of H^SCX (1 : 3). When solution is complete, add HNOs drop by drop, until Fe is oxidized. (5 cc. of HNOs (1 ' 3) is generally sufficient.) Boil to remove nitrous fumes and dilute to 150 cc. with hot water. Add to the boiling solution (a pipette is convenient) a few drops at a time of strong KMnO4 solution, boiling between each addition, until a permanent precipitate of Mn02 is produced which does not disappear after twenty minutes' boiling. An excess over the amount necessary to do this should be avoided as it will render reduction of excess KMnO4 to MnC>2 very difficult and will neces- sitate prolonged boiling. Remove from the heat and let settle. If reduction of excess KMnC>4 to MnO2 is complete, the precipitate will settle quickly leaving a clear yellow supernatant liquid. If reduction is not complete, the precipitate will not settle quickly and will leave a pink or reddish supernatant liquid and must be boiled again. Filter the cold solution, preferably into a 500 cc. Erlenmeyer flask; add 5 cc. of syrupy HsPCX, and titrate, adding a known excess of ferrous ammonium sulfate, and determining the excess with standard KMnC>4. Subtract the KMnO4 required for the excess titration from the ferrous ammonium sulfate orig- inally added (in terms of the KMnO4) and calculate the difference to Cr. The HaPCU decolorizes ferric salts, giving a better end point, and should also be added to the Fe(NH4)2(S04)2 solution when obtaining its relation to the KMnC>4 solution. SOLUTIONS. (1) Ferrous Ammonium Sulfate: Dissolve 50 grams of Fe(NH 4 )2 (864)2 -6H 2 O in 1900 cc. of distilled water and add 100 cc. of cone. H 2 SC>4. (2) Permanganate: Dissolve 1.58 grams of KMnO* in water and dilute to 1 liter. STANDARDIZATION. The Cr factor of the KMnCU solution is ob- tained by standardizing against 0.7 gram of pure Fe(NH4)2(SO4)2- 6H 2 O dissolved in 15 cc. of cone. H 2 SO 4 +150 cc. of H 2 O to ob- tain the Fe factor and calculating from this the Cr factor (also the vanadium factor if desired). ANALYSIS OF METALS 123 CALCULATION. Since 0.7 gram of Fe(NH4)2 (864)2 -GH^O is equivalent to exactly 0.1 gram of Fe, therefore the amount of Fe equivalent to 1 cc. of the KMnCU = . cc. KMnO4 used Since 1 Cr is equivalent to 3 Fe in this reaction, we have : 52 Cr f actor = Fe f actor X JT-T = Fef actor X 0.3 104. Vanadium (Qualitative Test). Dissolve 0.5 gram of sample in 10 cc. of H2S04 (1 : 3), heating until action ceases. Add 5 cc. of cone. HNOs and boil until no more red fumes are evolved. If tung- sten is present, filter. Pour half of the solution into each of two 6-inch test tubes. Then add to one tube 5 cc. of water and to the other 5 cc. of 3% HfoCfe solution. If V is present, the tube to which the peroxide was added will be distinctly redder than the other, even if there be only a few* hundredths of a per cent of V in the sample. If Ti, but no V, is present, the color will be a clear yellow. If a red color is produced, V is present and pos- sibly Ti also. Add 0.05 N FeSC>4 solution, 1 cc. at a time, shaking after each addition, until the red color gradually fades. If Ti is present, the red will change to a clear yellow. If none is present, the red will gradually fade out without changing to yellow. Vanadium (Quantitative). Dissolve 5 grams of drillings in 30 cc. of dil. HC1 (1 : 1) in a small covered beaker. When dis- solved, add a few drops of HF, warm, and add gradually 1 or 2 cc. of HNO 3 to oxidize the Fe. Evaporate to about 10 cc. Cool and pour into a 150 cc. separatory funnel with a very short stem. Wash out the beaker with warm HC1 (1:1) transferring all the solution to the funnel. To keep the volume small, use successive small portions of 5-6 cc. of acid in washing. The total volume of solution in the funnel should not exceed 40 cc. Now cool the funnel and contents and cautiously add 45-50 cc. of ether, pre- viously saturated with HC1. Stopper and shake vigorously for five minutes, keeping cool by holding under running water. Set the funnel in a rack and let stand until the ether separates and the line between the two layers is sharp. Then remove the stopper 124 TECHNICAL METHODS OF ANALYSIS carefully and draw off the aqueous solution down to the stop-cock, being careful to empty the stem. Add 5-6 cc. of HC1 (1 : 1) to the ether in the funnel and shake again. Let separate as before, and draw off the acid, adding this to the first extract. It is desir- able to use as little acid as possible in washing, as it takes up some Fe from the ether and increases that to be subsequently removed from the aqueous solution. During the whole operation the ether and the funnel should feel cool to the hand; if allowed to become too warm, vapor pres- sure may blow the stopper out. The aqueous solution contains some dissolved ether, a little Feds (which should not exceed 2% of that in the original solution) and the whole of the chlorides of V, Ni, Co, Mn, Al, and Cu. Evaporate this solution to dryness several times with a large excess of HC1 to reduce vanadic acid to divanadyl chloride, according to the equation: V 2 O 5 +6HC1 = 2VOC1 2 +3H 2 O+C1 2 . Add 5 cc. of cone. H2S04 and evaporate until all HC1 is expelled and fumes of 80s appear. By this operation divanadyl sulfate is formed: 2VOC1 2 +2H 2 S04==V 2 2 (S04) 2 +4HC1. Let cool and dilute to 200-300 cc. Warm to about 60 C. and titrate with standard KMnC>4 solution. The reaction is: 5V 2 O 2 (S04) 2 +2KMn04+8H 2 S04 = 5V 2 O 2 (S0 4 )3+K 2 S04+2MnS04+8H 2 0. Comparing this with the reaction for titrating ferrous iron : 10FeS04+2KMn04+8H 2 SO 4 = 5Fe 2 ($04)3+K 2 S0 4 +2MnS04+8H 2 O we have 10V : 510 = lOFe : 558 or, V = 0.9133 Fe. Hence, the value of the permanganate solution in terms of Fe, multiplied by 0.9133, gives its value in terms of V. NOTES. (1) Small amounts of ferric salts do not interfere with the reac- tion, but when Cr is present to the extent of 2% or over, the color of the ANALYSIS OF METALS 125 chrome salt masks the end reaction. The color of the divanadyl sulfate is pale blue, while that of the chromium sulfate is green. It is, therefore, impos- sible to tell from the color of the solution under the conditions whether the sample contains V or not, the discoloration of the permanganate solution being the only indication of its presence. (2) Run a blank on the reagents along with the sample and deduct from the titration before calculating to V. Chromium and Vanadium. Dissolve 1 gram of steel in 50 cc. of water and 10 cc. of cone. H^SO* in a 500 cc. beaker and proceed as described previously under Chromium until the solution is ready to titrate. Then add a standard ferrous ammonium sul- fate solution until the solution loses all brown tints and assumes a practically colorless shade in plain vanadium steels or a green color in chrome steels. Then add a few more cc. to make sure of an excess. Now add standard KMnCX solution, slowly and with vigorous agitation of the solution, until a very faint pink is ob- tained that persists after thirty seconds' stirring. Should even as much as 5 or 6% of Cr be present, a practiced operator can easily detect the pink tints through the chrome green. If it is desired to know the chromium content, the amount of ferrous ammonium sulfate and KMnCX solutions used should be recorded and the Cr content calculated as described previously under this element. Next, just destroy the pink tinge with ferrous ammonium sul- fate solution. The solution is now ready for the vanadium titra- tion. Add 2 cc. of a 2% solution of potassium ferricyanide (always use the same amount) and drop in slowly, and with stirring, standard ferrous ammonium sulfate until a drop produces a green color free from yellow tints. If much Cr is present, add ferrous ammonium sulfate until the green chromium color begins to darken. It is well to add from time to time during the titration a drop of the indicator and note if a green coloration is produced at the point where the drop mixes with the solution. The Fe value of the ferrous ammonium sulfate multiplied by 0.9133 gives the amount of V present. A " blank " must be run on a steel free from V but otherwise of similar composition. Also, it is best to run at the same time a V determination on a steel containing a known amount of V. . NOTES. (1) The presence of Cr increases the blank. If Cu is present, it will precipitate out when ferricyanide is added. In such case, add ferri- 126 TECHNICAL METHODS OF ANALYSIS cyanide before the MnO2 is filtered off, and filter off the copper ferricyanide with the MnO 2 . If a further precipitate is produced when ferricyanide is added for titration, run another determination using double the amount of ferri- cyanide to precipitate the Cu. Nickel, if present, will also slowly precipitate with ferricyanide. Mo does not interfere. If the sample is a W steel, when it is dissolved it should be digested until the precipitated WO 3 is a bright yellow. Then enough permanganate should be added to cause the precipitate to be colored chocolate by the MnO 2 formed. (2) It is absolutely essential in V and Cr titrations, when coming back with permanganate, to add the latter until 3 drops give a faint pink which remains visible after thirty seconds' stirring. The ferrous ammonium sul- fate should be added until 3 drops produce a distinct darkening of the green but not a blue. A better way for an analyst unfamiliar with the reaction is to add an excess of permanganate and titrate the excess with standard arsenite. Tungsten. Weigh accurately 1 gram of drillings into a 250 cc. beaker, add 10 cc. of aqua regia (75 cc. of cone. HC1+25 cc. of cone. HNOs), and, after violent action ceases, evaporate to dryness. Dissolve the residue by boiling with 20 cc. of HC1 (1 : 1) and a few drops of HNOs. Now dilute with 100 cc. of H^O and boil until the WOs precipitate is bright yellow. Filter, wash well with hot HC1 (1:1) but no water, to remove all Fe. Ignite in a plat- inum crucible until all paper is burned off, but do not heat the WOs hotter than a dull red. Cool in a desiccator and weigh. Add 3 drops of EbSCX and 5 cc. of HF and evaporate off the acids under a good hood, finally driving off the H^SCX by heating the crucible near the top. This evaporation must be carried out with extreme care to avoid loss by spattering. Heat to dull red and weigh. The loss in weight is Si02. Calculate to Si. CALCULATION. SiO 2 X 0.4693 = Si. The residue of WOs always contains some Fe20s. To correct for this, fuse it with 5 grams of Na2COs and a few small crystals of NaNOs until all WOs is dissolved. Dissolve the cold cake with hot water and filter off the residue of Fe2Os. Wash well with hot water, burn off the paper and weigh the residue. Subtract the weight so obtained from the weight of the impure WOs after the SiO2 was driven off. The difference is the weight of the pure WOs. Calculate to tungsten. CALCULATION. WO 3 X 0.7931 = W. NOTE. If, when filtering the WO 3 , some of the precipitate adheres to the sides of the beaker, wash the acid out of the beaker, then remove with a few cc. of NH 4 OH, rinsing into the platinum crucible with a little water. Evaporate to dryness and then place the filter paper in this crucible to ignite. ANALYSIS OF METALS 127 Molybdenum (Qualitative Test). Dissolve 0.5 gram of drill- ings in 25 cc. of HC1 (1 : 1). Add 2 grams of KC1O 3 and heat until the residue is bright yellow, if W is present. Filter and add to the filtrate 10 grams of KOH dissolved in 10 cc. of water. Boil for several minutes. Filter and pour the solution into a large test-tube. Add HG1 until crystals of KC1 begin to form, then add a few grains of granulated tin and heat just to boiling but no more. Cool and add 0.5 gram of KCNS. If Mo is present, a red color develops, the depth of color depending upon the amount present. The precipitation with KOH is to separate out the Fe, which would also give a red color with KCNS. The solution must not be heated too long with the tin present, otherwise the delicacy of the test will be impaired. The test will indicate the presence of as little as 0.2% of Mo or less. Molybdenum (Quantitative). Place 2 grams of drillings in a 450 cc. beaker and cover with 50 cc. of cone. HC1. Heat to boiling and add cone. HNOs, a few drops at a time. Con- tinue to heat the solution and add HNOs, a few drops at a time, until the sample is in solution and the Fe is oxidized. Very little more HNOa should be added than necessary to oxidize the Fe. A black film of carbonaceous matter will remain. Evaporate to the beginning of pastiness, add 50 cc. of hot water and 10 cc. of HC1 and boil a few minutes. The WOs separates, if present. Filter and wash with hot water acidulated with HC1. To the filtrate add a solution of NaOH, shaking the flask well during the addition, until most of the free acid is neutralized, but not until a darkening in color takes place. Transfer the solution to a sep- aratory funnel. Open the stop-cock of the funnel so that the solution runs out in drops, and let the drops fall into a 500 cc. graduated flask con- taining 150 cc. of a 6% solution of NaOH heated nearly to boiling. Shake the flask vigorously while the stream of drops is running in. This is important, as otherwise some Mo will be carried down with the Fe(OH)s. Finally wash the funnel. The Fe and most of the Cr are precipitated as hydroxides. A little Cr goes into the fil- trate as chromate. Fill the 500 cc. flask up to the mark, mix well and filter through a large paper into a 250 cc. graduated flask, rejecting the first few cc. of filtrate. Transfer 250 cc. of the filtrate to a 500 cc. 128 TECHNICAL METHODS OF ANALYSIS beaker and add HC1 until the solution is just acid to methyl orange, then add 4 cc. of cone. HC1 in excess. Add a few drops of sulfurous acid to reduce the small amount of CrOs usually present, and boil. Add 40 cc. of NH* acetate, made by adding 30% acetic acid to cone. NILiOH until the latter is neutralized. Now add 40 cc. of a 1% Pb acetate solution, stir well, filter through a close filter paper and wash well with hot water. Ignite in a porcelain crucible and weigh as lead molybdate. Calculate to Mo. CALCULATION. PbMoO 4 X 0.2614 = Mo. NOTES. (1) When the sample is dissolved and oxidized with HNO 3 , a little molybdic acid may precipitate, and would appear with the WO 3 if any were present. (2) The treatment with NaOH forms Na 2 MoO 4 which is soluble, while the Fe separates as Fe(OH) 3 . If the acid solution is not added slowly and with vigorous stirring the Fe(OH) 3 will carry down with it some molybdic acid. Ammonium acetate is added to reduce the acidity, according to the disso- ciation theory. (3) If much W is present, the WO 3 should be dissolved in NaOH solution, diluted to 50 cc., HC1 added until the solution is acid and then 20 cc. in excess, and the solution evaporated to 10 cc., diluted to 50 cc. and boiled; the WO 3 filtered off and the filtrate added to the main filtrate. Titanium (Qualitative Test). See Vanadium (Qualitative). Titanium (Quantitative). Add to 1 gram of sample in a 500 cc. flask 40 cc. of H2SO4 (1 : 3). Heat to boiling until no more action takes place. Disregard any residue. Dilute to 250 cc. and add NHtOH until a slight precipitate forms, then add a few grams of Na2S20s and a few drops of H^SC^, or enough to make the solution clear. Heat until all Fe is reduced, as shown by testing a drop with KCNS; then add a solution containing 25 cc. of water, 15 cc. of cone. NBLtOH, and 10 grams of KCN. Heat to boiling for several minutes. Prepare a filter by shaking two 9 cm. filter papers in a flask with water until well macerated and then pouring the pulp through a funnel containing a platinum cone. Place this funnel in a suction flask and pour the solution through, using suction. The solution should filter quickly to prevent oxidation of Fe. Wash well with water. Burn the paper and contents in a platinum crucible until all paper is consumed; then add 4 grams of KHSO4, which has been previously fused to remove water; fuse, and maintain at bright red heat for several minutes. Cool, then add water enough to ANALYSIS OF METALS 129 half fill the crucible and 5 cc. of H 2 S04, and heat until the cake is all dissolved. Cool, transfer to a Nessler tube or other color com- paritor, dilute to 100 cc. and add 3 cc. of H 2 2 (ordinary 3% solution). If Ti is present, a yellow color will immediately appear. To the other tube add 100 cc. of 5% H 2 S0 4 , 3 cc. of H 2 2 and standard titanium sulfate solution a little at a time, shaking after each addition, until the color matches the color of the solution of the sample. The Ti in the sample will then be the same as the amount added to the comparison tube. NOTES. (1) H 2 O 2 containing acetanilide or any other preservative should not be used. The pure 3% H 2 O 2 solution is satisfactory. (2) Standard Titanium Solution: Ignite pure TiO 2 at dull red heat to con- stant weight. Weigh 0.5 gram of the anhydrous powder in a platinum crucible with 5 grams of pure KHSC>4. Melt cautiously and keep at a low red heat for 5-10 minutes until the TiO 2 all dissolves and the liquid becomes clear. Partially cool the crucible and add 5 cc. of cone. H 2 SO 4 , then heat again till the mass liquefies. Cool, and put the crucible and contents into 200-300 cc. of water containing 5% of H 2 SO 4 . When the fusion is dis- solved, wash and remove the crucible. The TiO 2 should dissolve to a clear solution. If any residue remains, filter it out, wash, ignite and weigh, and deduct it from the TiO 2 taken, using the difference in calculating the strength of the solution. Finally, dilute the solution with 5% H 2 S(>4 till 1 cc. contains 0.001 gram of titanium. TiO 2 X0.6005 = Ti. REFERENCES. Lord and Demorest: "Metallurgical Analysis," Fourth Edition; Blair: "Chemical Analysis of Iron"; Johnson: " Chemical Analysis of Special Steels." PIG AND CAST IRON General. This method applies to all classes and grades of pig, charcoal and cast irons. Where an alternative procedure is given, the latter is to be used only when, on account of lack of chemicals or apparatus, the first method cannot be employed. Sampling. The points to be observed in sampling iron for analysis are as follows: (1) The sample should preferably be drillings taken without the application of water, oil or other lubricant, and should be free from sand, dirt or other foreign matter. A magnet must not be used in separating the iron from foreign matter on account of the danger of losing graphitic carbon which is non-magnetic. 130 TECHNICAL METHODS OF ANALYSIS (2) All samples should be ground uniformly in a hardened steel mortar. (3) Drillings in the case of cast iron should represent the entire cross section of material less than 1 inch thick. On material over 1 inch thick, the sample should be taken midway between the surface and the center. (4) Drillings in the case of pig iron should be taken midway between the surface and the center. Total Carbon. Determine total carbon by the solution-com- bustion method. For the general arrangement of the combustion apparatus, see page 108. Dissolve 1 gram of sample, using 75 cc. of a solution of copper potassium chloride. Stir constantly with a mechanical stirrer until the sample is dissolved (see Note 1) or let stand overnight. (Complete solution is shown by the disappearance of all Cu from the bottom of the beaker.) Filter on a previously ignited asbestos mat in a loose bottom Gooch crucible. Wash 5 times with HC1 (1 : 10), then 5 times with water. Do not let air draw through the mat and residue during filtering and subsequent washing. Transfer the mat and contents to the combustion boat, being care- ful not to lose any of the carbon, and dry at 100 C. Introduce the boat with the residue into the center of the com- bustion tube, which is maintained at 850-900 C. for this deter- mination (also that of graphitic carbon). Admit oxygen at the approximate rate of 26 bubbles per ten seconds, or 250 cc. every ten minutes. As soon as the carbon begins to burn, there is at first a rapid evolution of gas, which quickly ceases. When the oxidation of the residue is completed, the oxygen begins to flow at normal speed again. Allow thirty minutes for oxidation of the carbon residue and sweeping of all CO2 from the combustion tube into the absorption apparatus. Weigh the CC>2 formed and cal- culate to C. CALCULATION. CO 2 X 0.2728 = C. Copper Potassium Chloride Solution: Dissolve 900 grams of 2KCl-CuCl 2 -2H 2 O in 2700 cc. of water and add 215 cc. of cone. HC1. Filter the solution through glass wool and asbestos. NOTES. (1) We have found the rotating electrode a very convenient ptirrer for getting the sample into solution. Of course no current should be passed when it is used for this purpose ANALYSIS OF METALS 131 (2) This method cannot be used if the material is an alloy iron. In this case a portion of the sample passing a 20- but not a 30-mesh sieve must be mixed with 2 grams of litharge and the carbon determined by direct com- bustion at 1000 C. A "blank" must be carefully determined upon the litharge at the same temperature before using, and this blank deducted from the weight of CO 2 found. (3) See also notes (2) to (6) inclusive, under Total Carbon in Carbon Steel, page 109. Graphitic Carbon. Weigh 1 gram of sample into a 200 cc. beaker and add 60 cc. of HNO 3 (1 : 3). Heat the beaker on the steam bath until the iron is all dissolved. Filter on a loose bottom Gooch crucible through asbestos and wash with water; HC1 (1 : 1); H 2 O; NHjOH (1%); H 2 O; HC1 (1:1); H 2 O; and H 2 O, in the order indicated. Transfer the asbestos and contents to the combustion boat, heat until dry in the oven at 100 C. and determine carbon in the regular way. Combined Carbon. Combined C equals total C minus gra- phitic C. In special cases the combined C may be determined colori- metrically as follows: Dissolve 0.2 gram of sample in 6 cc. of HNOs (1 : 3) in a small test tube, filter, washing with the least possible amount of water, and compare the color with a standard pig iron run in the same manner. The standard iron must not vary more than 0.1% in carbon from the unknown. Manganese (Bismuthate Method). Dissolve 1 gram of drill- ings in 25 cc. of HNOs (1 : 3) in a small beaker. When dissolved, filter into a 200 cc. Erlenmeyer flask, wash with 30 cc. of HNO 3 (1 : 3), cool the filtrate, add about 0.5 gram of Na bismuthate, boil until the pink color has disappeared and dissolve any pre- cipitate of MnO 2 by .adding a few drops of a saturated solution of FeSO 4 or of sodium thiosulfate; then heat until all nitrous fumes have been driven off, cool to 60 F., add about 0.5 gram of Na bismuthate and shake the flask vigorously for a few minutes. Add 50 cc. of 3% HNOs* and filter the solution into a suction flask through an extra-porous alundum thimble, taking care not to fill the thimble so full that any of the solution comes in contact * This acid is prepared by adding 60 cc. of cone. HNO 3 to 1940 cc. of H 2 O and then adding 3-4 grams of Na bismuthate and shaking. It should be allowed to stand overnight before using. 132 TECHNICAL METHODS OF ANALYSIS with the rubber connection. Wash with 50-100 cc. of the same acid and finally with water. From here proceed as directed on page 111, using for the standardization an iron of known man- ganese content instead of a steel. Phosphorus. Dissolve 1 gram of the sample in a 250 cc. beaker with 100 cc. of HN0 3 (1 : 3). Add 5 cc. of saturated KMnO 4 solu- tion and boil until the excess KMnC>4 is decomposed and the solu- tion contains precipitated Mn02. Add a few crystals of tartaric acid and boil until the solution clears, then a few minutes longer, to drive off all nitrous fumes. Filter into a 500 cc. Erlenmeyer flask and wash the precipitate thoroughly with hot water. Add 15 cc. of cone. NELtOH and proceed as described under Phosphorus in Carbon Steel on page 112. Silicon. Weigh 1 gram of the iron into a 12 cm. casserole or evaporating dish and add 60 cc. of silicon mixture (see Note 2). When effervescence ceases, rinse the sides of the dish with water and cover with a watch glass. Place on the hot plate and boil slowly but continuously until a white crust of FeSO4 forms and SOs fumes appear. Remove the dish from the hot plate and cool. Dilute immediately with 75 cc. of HC1 (1 : 3) and bring to a boil; keep at the boiling point until the solution is clear. Filter while hot through an 11 cm. filter, washing with hot HC1 (1 : 1) and hot water alternately until SiC>2 and paper are free from iron. Then wash the paper free from acid. Ignite strongly and weigh as SiC>2. Calculate to Si. CALCULATION. SiO 2 X 0.4693 = Si. NOTES. (1) If the precipitate is red from undissolved Fe, volatilize with HF. Ignite and weigh again. The loss in weight is SiO 2 . Calculate to Si. (2) Silicon Mixture. To 1500 cc. of water add 500 cc. cone. HNO 3 and then 150 cc. of cone. I^SCX with constant stirring. Sulfur (Evolution Method). Proceed as on page 115. NOTE. In the case of very coarse drillings which dissolve slowly, consid- erable acid may be carried over and neutralize the NH 3 in the CdCl 2 solu- tion. In such cases care must be taken to keep the solution up to original strength in NH 3 . If such samples are encountered, it is preferable to use the Elliott Method as described below. Sulfur (Elliott Evolution Method). Thoroughly mix 5 grams of the sample with 0.25 gram of pure, finely powdered anhydrous ANALYSIS OF METALS 133 potassium ferrocyanide, and wrap in an 11 cm. filter paper. Place in a small porcelain crucible, cover and anneal at 750-850 C. for twenty minutes in a closed muffle. Cool slowly outside the muffle. The material should be covered practically completely by the charred paper, if the temperature has not been above 850 C. or if it has not been in the muffle too long. After cooling, crush the contents of the crucible slightly in a glass mortar and finally transfer to a 500 cc. Florence flask. Connect the evolution flask with a condensing tube (lX8-in. test tube) containing about 2 inches of water and standing in a conical beaker filled with cold water. A tube from this dips into a second test tube containing 60 cc. of CdCb solution, which again is connected with another test tube containing more (See Fig. 7.) FIG. 7. Apparatus for Sulfur in Iron. (Elliott Evolution Method.) Add 50 cc. of cone. HC1 to the evolution flask and heat until the sample is in solution and all H^S has been driven over. Discon- nect the apparatus and wash the CdS into a tumbler. Add 3 cc. of starch solution and 15 cc. of cone. HC1 and titrate at once with standard iodine solution to the appearance of a permanent blue. SOLUTIONS. (a) Cadmium Chloride Solution. Dissolve 20 grams of CdCU in water with the aid of a few drops of HC1. Then add NHiOH until the precipitate completely redissolves. Make slightly acid with acetic acid and then add 20 cc. excess. Dilute to 2 liters. (6) Standard Iodine Solution. The same as used in the pre- vious evolution method (page 115). (c) Starch Solution. The same as already described in the previous evolution method (see page 12). STANDARDIZATION. Standardize exactly as described under Sulfur in Carbon Steel (page 115). 134 TECHNICAL METHODS OF ANALYSIS NOTE. Copper, titanium and vanadium are sometimes called for in this class of material, and they may be detected and quantitatively determined as described in Blair, 7th edition, pages 184-185. REFERENCES. Blair: " Chemical Analysis of Iron." Lord and Demorest : " Metallurgical Analysis." TIN IN TIN ORES General. Tin seldom exists in nature in the metallic state. It occurs both in veins and in alluvial deposits. Cassiterite or tin-stone is its most important ore. This is a tin oxide (SnO2) and generally occurs in alluvial deposits. The color of the oxide may be black, brown, reddish yellow, red and brownish white. The streak is white to brownish. When pure the ore contains 78.77% Sn. Its sp. gr. is 6.8-7.1. Another ore is stannite (sp. gr. 4.3^4.52), a compound of Sn, S, Fe, Cu, and sometimes Zn. The impurities most frequently associated with the oxide are pyrite, arsenopyrite, wolframite (tungstate of Fe and Mn), chalcopyrite, titaniferous iron, columbite, iron oxide, tourmaline, and sometimes blende and galena. To determine whether the mineral is SnC>2, fuse a finely ground portion in a porcelain crucible with 3 or 4 times its weight of NaCN and dissolve the mass in water. A tin globule, or globules, will be found if the material is SnC>2. The sample may also be mixed with Na2COs and char- coal and fused on charcoal in a reducing flame. When in veins the gangue generally consists of granite, slate, syenite, quartz or feldspar, and often carries garnets and zircons. Fluorspar is also frequently present, and by some is considered a good indicator of tin-stone. Portions of the deposit are often very rich, but average ore, whether from veins or placers, carries only 1-5% SnO2- On this account samples can very rarely be assayed by fire, or even analyzed in the wet way directly. Owing to its high specific gravity, however, we can resort to washing and concentration, thus separating it from gangue and some other impurities. Wolframite, unfortunately, has a sp. gr. of 7.2-7.5, slightly higher than SnO2. This necessitates subsequent purification when this mineral is present. If the tin-stone carries much iron oxide, this must be removed with acids, otherwise the resulting tin will ANALYSIS OF METALS 135 not collect in a button, but will contain Fe and be a porous and magnetic mass. The following are the steps in the assay: 1. Concentration. 2. Roasting concentrates. 3. Panning concentrates and boiling in aqua regia. 4. Panning concentrates again. 5. Assaying final concentrates. If the concentrates obtained from the first panning are very pure, some of the later steps may be omitted. Concentration. Weigh 500-1000 grams of ore, crushed at least as fine as 40-mesh size. If crushed too fine, the SnC>2 will slime badly, but it must be fine enough to liberate SnC>2 from the gangue. Carefully pan the ore again and again until no more concentrates can be obtained. This is done by placing the ore in a shallow pan, running a little water into the pan and shaking. Then carefully pour off the lighter constituents of ore or tailings, allowing the heavy metallic compounds to settle to the bottom. Do not pan too closely; if a little gangue is left with the concentrates it does no harm. Discard the waste matter or tailings. Roasting. Dry the concentrates (consisting of SnO2, pyrite, and whatever heavy material happens to be in the ore, together with a small amount of gangue) and then place in a clay or iron roasting dish. Place this in a muffle, the bottom of which is hardly red, and heat slowly. When the odor of 862 can no longer be detected, remove the dish, cool, and stir a little fine charcoal into the ore. This reduces sulfates, arsenates and antimoniates to lower forms and enables Sb and As to be set free, and is espe- cially necessary when As is present. Roast again, and repeat until a dead roast is obtained. This point is reached when no more fumes of SO2 or As2Os are given off. Everything in the concentrates should now be in the form of oxides. They can now be panned to remove oxides of iron and silica and then treated with acid, or treated with acid directly and then panned. Treatment with Acid. Boil the concentrates in aqua regia, in which SnC>2 is insoluble. This practically removes everything except some SiO2, TiC>2 and compounds insoluble in aqua regia. If much SiC>2 is present, pan again. In some cases it is well to grind finely the tailings from this concentration and pan again. 136 TECHNICAL METHODS OF ANALYSIS Dry the total concentrates, weigh and grind to pass an 80-mesh sieve. Assaying. The concentrates are now ready for assaying, which may be done by various methods. The following cyanide method of Levol has been found to give excellent results. On clean ores or concentrates it is very accurate, but in ores con- taining much foreign matter, the assay is rendered much more difficult and the time of the fusion must be increased. Mix 5-10 grams of concentrates with 4 times as much NaCN. (Poison!) Have a good layer of NaCN in the bottom of the crucible, next put in the mixture of concentrates and NaCN and then place a layer of NaCN on top. Use Battersea A or Denver fire clay crucible of similar size. Heat very s owly at first in a Fletcher gas muffle furnace and just fuse to reduce SnO2 to Sn and then keep just fused for twenty to thirty minutes. Increase the temperature ten to fifteen minutes longer, remove from the fire, tap gently and transfer to some place where the fumes will not be carried into the laboratory. The purer the SnO2, the shorter is the necessary period of fusion. With some ores the fusion may be completed in ten minutes with good results, but ordinarily the above time will be required for a satisfactory fusion. Never let the fuson boil, which causes low results. Let the crucible become perfectly cold, being careful not to dis- turb during cooling. Break the crucible and free the button from the mass. If decomposition of the ore is complete, a clean bright tin button should be obtained. If fusion is incomplete, some ore will still remain undecomposed and the entire fusion must be repeated. Free the button entirely from cyanide, dry and weigh. Calculate the per cent of total metal in the original ore. On very high-grade ores, containing 50% or better of Sn, it is often possible to fuse the ore directly with NaCN, omitting con- centration, roasting and acid treatment. If these steps are omit- ted it may be necessary, in order to obtain a clean button, to add a little Na2COs to the fusion mixture in order to take care of silica. A trial fusion should be made first, however, with cyanide alone and then if a satisfactory button is not obtained, another fusion should be made by adding a little Na2COs. The button may contain, besides metallic Sn, small amounts of ANALYSIS OF METALS 137 Sb and other metals. In order to obtain the true tin content of the ore it is necessary to determine the amount of tin in the button. This may be done by the volumetric method described under Solder on page 155. The percentage of tin obtained should, of course, be calculated back to the original ore. ZINC (SPELTER) General. Spelter ordinarily used for brass and similar alloys is usually classified, according 'to the amounts of Pb and other impurities present, into four grades: A, " High Grade "; B, " Inter- mediate "; C, " Brass Special "; and D, " Selected." A fifth and still lower grade, " Prime Western," is principally used for gal- vanizing. These grades are covered by the specifications of the American Society for Testing Materials in its Triennial Standards for 1918, pages 437-438. The methods of sampling and analysis described below are those generally accepted in the United States as standard in all the larger laboratories of both producers and consumers of zinc and zinc products. The methods of analysis are those originally proposed by Elliott and Storer and Price. Alternative methods are only to be used when apparatus or chemicals required in the preferred method are not available. Sampling. Select 10 slabs at random from a carload and saw each slab completely across from the middle of one long side to the middle of the other and use the sawdust for analysis. Or, drill three 9 mm. (f inch) holes along one diagonal of each slab, boring completely through and taking care to make fine drillings. The holes should be drilled as nearly as possible at the middle and half- way between either end and the middle of such diagonals. Go over the drillings or sawings with a powerful magnet to take out any iron which may have come from the drill or saw, and mix the sample thoroughly. The drill or saw must be thoroughly cleaned before use, and no lubricant shall be used in either drilling or saw- ing. Lead (Electrolytic Method). Place 8.643 * grams of sample in * An empirical factor weight 8.643 is used instead of the theoretical one (8.662) according to the American Society for Testing Materials: Triennial Standards, 1918 and E. F. Smith's " Electro Analysis," Fourth Edition, page 102. 138 TECHNICAL METHODS OF ANALYSIS a 400 cc. beaker and add sufficient water to cover. Then add gradually and cautiously 30 cc. of cone. HNOs. When action is complete, boil the solution for a few minutes to expel nitrous fumes. Wash the watch glass and the sides of the beaker and transfer the solution to a 250 cc. beaker. Dilute to 200 cc. and electrolyze, using a rotating platinum cathode and a stationary platinum gauze anode, with a current of 5 amperes. The time required for elec- trolysis is from thirty to forty-five minutes, depending upon the amount of Pb present. Test the solution for complete depo- sition of Pb by washing the watch glasses and sides of the beaker until the depth of the solution is increased about 0.5 inch. Then continue the current for fifteen minutes, and if the newly exposed surface is still bright, the Pb is completely deposited. Wash the anode three or four times with distilled water, once with alcohol and then dry in the oven, or on the hot plate, at 210 C. for 0.5 hour and weigh as PbCb. The weight of PbCb in milligrams, divided by 100, will give the percentage of Pb. The PbC>2 deposit can be readily removed by covering the anode with dil. HNOs and inserting a piece of Cu. Lead (Alternative " Lead Acid " Method) .Place in a 350 cc. beaker 25, 15, 10 or 5 grams of drillings or sawings, according to whether the spelter is of Grade A, B, C, or D, respectively, and add, according to the size of sample taken, 300, 180, 120 or 60 cc. of " lead acid."* After all but about 1 gram of Zn is dissolved, filter on a close filter and wash out the beaker twice with " lead acid " from a wash bottle. Wash the undissolved matter from the filter into the original beaker with water and dissolve with a small amount of hot HNO 3 (1 : 1). Add 40 cc. of " lead acid/' and evaporate on the hot plate until strong fumes of SOs escape. When cool, add 35 cc. of water (which is the quantity of water evaporated * " Lead acid " is a solution of one volume of H 2 SO 4 in seven volumes of water, saturated with PbSO 4 . It is prepared as follows: Pour 300 cc. of cone. H 2 SO 4 into 1800 cc. of water; dissolve 1 gram of lead acetate in 300 cc. of water and add to the hot solution with stirring. Let the solution settle for several days and siphon off through a thick asbestos filter for use. When " lead acid " is used, it is unnecessary to consider the solubility of PbSO 4 , since the solution is always brought back to the same volume as the volume of " lead acid " originally added; consequently when the PbSO 4 is filtered, no more lead remains in the filtrate than was originally added in the " lead acid." ANALYSIS OF METALS 139 from the " lead acid ")> an d heat to boiling. Add the first filtrate (containing the greater part of the zinc, and possibly a small amount of PbSO4), stir well, and let stand for at least five hours, preferably overnight. Filter on a Gooch crucible, wash with " lead acid," then with a mixture of equal parts of alcohol and water, and finally with alcohol alone. Set the Gooch crucible inside a porcelain crucible in order to avoid reduction of PbSO4 by the flame gases and mechanical disintegration of the asbestos mat. Ignite for five minutes at the full heat of a Tirrill burner. Cool and weigh as PbSCX. Calculate to Pb. CALCULATION. PbSO 4 X 0.6833 = Pb. Iron. Place 25 grams of the sample in a tall 700 cc. beaker and dissolve cautiously in 125 cc. of cone. HNOs. Boil, dilute to about 300 cc., add 10 grams of NUiCl and then NILjOH until precipitated Zn(OH)2 has redissolved. Boil, let settle and filter on an 11 cm. S. 4 solution, containing approximately 0.2 gram per liter. 1 cc. of KMnC>4 solution will equal about 0.000334 gram of Fe. Run a " blank " with the same amounts of acid and water and correct accordingly. Standardization. Standardize the KMnC^ against sodium oxalate obtained from the Bureau of Standards, using the following procedure: Weigh in duplicate 0.0200 gram portions of Na2C2C>4 into a 200 cc. Erlenmeyer flask, dissolve in 50-75 cc. of hot water (80-90 C.), and add 10 cc. of H 2 S0 4 (1:1). Titrate at once with the KMnC>4 solution to be standardized, stirring vigorously and continuously. About 49-50 cc. of KMnC^ solution will probably be required. The KMnC>4 must not be added more rapidly than 10-15 cc. per minute and the last 0.5-1 cc. must be added with particular care to allow each drop to be fully decolor- ized before the next is added. The solution should not be below * For manipulation of the Jones reductor see page 148. If, before passing the solution through the reductor, a large amount of PbSO4 is present, it is well to filter it off so as to prevent clogging the reductor. 140 TECHNICAL METHODS OF ANALYSIS 60 C. at the time the end point is reached. The use of a small thermometer as a stirring rod will be found convenient in this titration. 0.0167 CALCULATION. Fe factor = . cc. KMn04 used Cadmium. Place 25 grams of drillings in a tall 500 cc. beaker; add 250 cc. of water and 55 cc. of cone. HC1 and stir. When action has almost ceased, add more acid with stirring, using about 2 cc. at a time, and letting stand after each addition of acid, until finally all but about 2 grams of Zn has been dissolved. About 60 cc. of acid in all are usually required. It is best to allow the first 55 cc. of acid to act overnight. Filter, first transferring one of the undissolved pieces of zinc to the filter paper, and wash twice with water. Discard the filtrate. Wash the undissolved matter on the filter paper into a 500 cc. beaker; cover, and dissolve in HNOs. Transfer to a casserole, add 20 cc. of H2S04 (1:1) and evaporate until fumes of 80s appear. Take up with about 100 cc. of water, boil, cool, and let stand until any PbSO^ present settles completely. Filter off the PbSC>4, wash with water, retain the filtrate and discard the PbSO 4 . Dilute the filtrate to 400 cc., add about 10 grams of NELiCl, and pass in H^S for one hour. It is occasionally necessary to start precipitation of the CdS by the addition of a drop or two of NELiOH to the dilute solution. Let stand until the precipitate has settled, and then filter off the impure CdS in a loose-bottomed Gooch crucible. Remove the CdS by carefully punching out the bottom into a tall 200 cc. beaker. Wipe off any CdS remaining on the sides of the crucible, using a little asbestos pulp; add 60 cc. of H 2 SO4 (1:5) and boil for 0.5 hour. In the case of spelters carrying large amounts of Cd, it may be necessary to add more acid. The dilute acid dis- solves CdS and ZnS, but not PbS. Filter and dilute to 300 cc. ; add about 5 grams of NKiCl and precipitate with H^S again, in order to get rid of traces of Zn. In case a large amount of Cd is present, a third precipitation may be necessary. After the final precipitation, let settle, filter and transfer to a weighed platinum dish; cover, and dissolve in HC1 (1 : 3). Also dissolve the sulfide remaining on the filter paper in hot HC1 (1 : 3) and add it to solution in the platinum dish. Add ANALYSIS OF METALS 141 a little H2S04 and evaporate the solution on the hot plate until copious SOs fumes escape. Dilute with water, add a few cc. of cone. HNOs to oxidize any filter paper shreds, and again evap- orate the solution until fumes of SOs come off freely. Remove the excess of H^SCX by heating the dish cautiously, and finally heat to between 500 and 600 C., or to dull redness, and weigh as CdSO 4 . Calculate to Cd. CALCULATION. CdSO 4 X 0.5392 = Cd. Cadmium (Alternative Method). Proceed as above until the CdS has been dissolved in HC1. Oxidize with HNOs and filter from any sulfur. Transfer the solution to a 200 cc. beaker, add a drop or two of phenolphthalein and then pure NaOH or KOH solution until a permanent red color is obtained. Add a strong solution of pure KCN, with constant stirring, until the precipitate of Cd(OH)2 is completely redissolved. Avoid using excess of KCN. Dilute the solution to 200 cc. and electrolyze with a current of 5 amperes, using a rotating platinum anode or cathode. (The Cd deposits on the cathode.) The time required is one to two hours. The solution should always be tested for Cd as follows: Raise the liquid in the beaker * and then note after twenty minutes the newly exposed surface of the electrode. If it is still bright the Cd is completely deposited. Finally wash the electrode with distilled water and then with alcohol. Dry at 100 C., coo' and weigh. The increase is metallic Cd. REFERENCES. Report of Sub-committee of Division of Industrial Chem- ists and Chemical Engineers. J. Ind. Eng. Chem. 7, 547 (June, 1915). ZINC DUST General. Commercial zinc dust is bought and sold upon its actual reducing power, which is measured by the amount of metallic Zn present or, as it is termed in the trade, " available zinc." The determination of the amount of available Zn, by means of K2Cr2O7, and H2SO4, depends upon the following reaction : Standardization of Solution. Prepare a solution of pure containing about 40 grams per liter. Pipette exactly * As under " Lead (Electrolytic Method) " above. 142 TECHNICAL METHODS OF ANALYSIS 10 cc. of this solution, with certified pipette, into an Erlenmeyer flask. Add about 150 ce. of water, 3 cc. of cone. H 2 SO 4 and 15 cc. of 10% KI solution. Titrate the liberated iodine with 0.1 N thio- sulfate, adding a little starch indicator when the end point has been nearly approached. The exact end point is indicated by a change in color from dark brownish green to clear light green. This change is very sharp. Calculate the exact strength of the bichromate solution from the factor: 1 cc. 0.1 N thiosulfate = 0.004903 gram K 2 Cr 2 O 7 . NOTE. It is possible to obtain K 2 Cr 2 O 7 sufficiently pure so that the solution does not need to be standardized. Procedure. Weigh accurately 1 gram of zinc dust into a 350 cc. beaker, add 50 cc. of standard K 2 Cr 2 O7 solution and dilute with about 200 cc. of distilled water. Then add, drop by drop, with constant stirring, 20 cc. of dilute H2SO4 at the rate of 1 cc. per minute. It is very important that the liquid should be con- stantly stirred and no bubbles of hydrogen should escape. Finally, add 5 cc. of cone. H 2 S04 and let stand until there is no further action. There will generally be more or less metallic Pb which will not be attacked. Heat just to boiling and then cool to room temperature. Transfer the iquid to a 500 cc. flask and dilute exactly to the mark. Mix thoroughly and pipette 200 cc. into a beaker; add 15 cc. of 10% KI solution and titrate with 0.1 N thiosulfate as previously described. Calculate the amount of K 2 Cr 2 C>7 equivalent to the thiosulfate titration and multiply by 2.5. Subtract this from the total amount of K 2 Cr 2 0? in the 50 cc. originally taken and from the difference calculate the amount of metallic Zn. CALCULATION. K 2 Cr 2 O 7 X 0.6666 = Zn. REFERENCE. Classen : " Ausgewahlte Methoden der Analytischen Chemie," Vol. 1, page 353. BRASS AND BRONZE General. This method covers the analysis of alloys of the brass or bronze type containing two or more of the following elements: Cu, Pb, Sn, Zn, P, Fe, Ni, Mn, As, Sb, and Al (the latter only in small amounts). It does not include white metals ANALYSIS OF METALS 143 and aluminum alloys, containing considerable Al. The sample for analysis should be in the shape of very fine drillings free from oil, iron, dirt, or other foreign matter. If the composition of the alloy is unknown, a qualitative analysis should be made, using a 5-gram sample. In those cases where an alternative method is given, the latter is to be used only when, on account of lack of chemicals or appa- ratus, the first method cannot be employed. Phosphorus. Dissolve 1 gram of the sample in 15 cc. of fum- ing HNOs and evaporate on the hot plate until most of the free acid is expelled. Add 10 cc. of cone. HC1 and evaporate to dry- ness. Dissolve the residue in 10 cc. of HC1 and 50 cc. of water, heat to boiling and precipitate the Pb, Sn, and Cu with granulated c. P. metallic Zn, 20-mesh. Use an excess of the Zn and let the reaction continue until no Pb, Sn, or Cu remains in the solution. Just before filtering add 5-10 cc. of cone. HC1 to the solution. Filter immediately on a rapid filter, in the cone of which has been placed 1-2 grams of c. P. granulated 20-mesh zinc. Wash well with hot water and neutralize this solution (approximately 150 cc. in volume) with NH^OH, adding the latter only until a permanent, heavy, curdy, white precipitate forms. Bring back again with cone. HNOs, adding approximately 5 cc. in excess. Heat the solution to 70 C. and precipitate the phosphorus by adding 60 cc. of ammonium molybdate solution. Shake for five minutes and let the solution stand at least 0.5 hour, or until the yellow precip- itate has completely settled. Filter through an 11 cm. filter paper, wash the precipitate five times with 2% HN0 3 , then with 1% KNOs solution until free from acid (approximately fifteen times) . Place the filter and contents in the original flask, which has been thoroughly rinsed with water, and add approximately 50 cc. of cool distilled water and a measured amount of standard NaOH from a burette, 5 cc. at a time (sufficient to completely dissolve the yellow precipitate). Cork the flask and agitate violently until the filter paper is disintegrated; add 3 drops of 1% phenolphtha- lein solution with a medicine dropper, and titrate with standard HNOs to the disappearance of the pink color. The same standard acid and alkali may be used as employed for determining Phos- phorus in Steel, as described on page 113. 144 TECHNICAL METHODS OF ANALYSIS CALCULATION. Subtract the number of cc. of HNO 3 used from the number corresponding to the volume of NaOH added. The difference is the cc. of HNO 3 equivalent to the phosphorus in the sample, and this, multiplied by the value of the HNO 3 in terms of phosphorus, gives the weight of phosphorus in the sample. This weight multiplied by 100 gives the per cent of phosphorus. Tin. (a) If the alloy contains over 1.5% of Sn: Weigh 1 gram into a 250 cc. beaker, add 15 cc. of dil. HN0 3 (2:1), cover imme- diately with a watch glass and, when violent action ceases, boil until no more red fumes are given off. Place the beaker on a water bath and digest for one-half hour. Dilute with 50 cc. of water and filter at once, if the alloy contains phosphorus, washing very thoroughly with 2% HN0 3 . Test the last drops of the filtrate with a little potassium ferro- cyanide or ammonium sulfide solution; neither solution should give any precipitate. Place the filter paper and contents in a weighed porcelain crucible, smoke off the paper and ignite for fif- teen minutes in the full heat of a Tirrill burner. Cool in a desic- cator and weigh. This represents Sn02-fSb204+P2O 5 , together with traces of Fe, Cu and Pb. (Save these mixed oxides.) The phosphorus and antimony are separately determined, as below, and subtracted from the above weight. The remainder is SnO2. Calculate to Sn. CALCULATION. SnO 2 X0.7877 = Sn. (b) If the alloy contains less than 1 .5% of Sn: Dissolve 5 grams in 40 cc. of HNO 3 (2:1) and treat as above described. In this case the filtrate should be made up to volume and an aliquot equivalent to 1 gram taken for the subsequent determinations of Cu, Pb and Zn. NOTE. In the case of brass or bronze containing less than 10% tin and less than 0.7% phosphorus, no further purification of the mixed oxides is necessary; but if above these limits, the mixed oxides must be purified as follows : Fuse the ignited precipitate with a mixture of finely powdered sulfur and Na 2 CO 3 in the proportion of 1 part of the precipitate to 3 parts each of sulfur and of Na 2 CO 3 in a covered porcelain crucible until the odor of SO 2 has disappeared. Cool and dissolve the fusion in hot water; add an excess of NaaSOs to convert any polysulfides to monosulfides, filter and wash the precipitate thoroughly. This precipitate contains Cu, Pb, and Fe. Dissolve in dil. HNO 3 (1 : 3). Determine the Cu and Pb electrolytically and the Fe with NH 4 OH in the regular way. Calculate the weights thus found to the oxides and subtract from the original precipitate. ANALYSIS OF METALS 145 CALCULATIONS. P X 2.2886 = P 2 5 . SbXl.2662 = Sb 2 O 4 . PbO 2 X0.9331 = PbO. FeX 1.4298 = Fe 2 3 . CuX 1.2517 = CuO. Copper and Lead. (a) When lead is 10% or less: Dilute the filtrate from the Sn determination to approximately 200 cc. and add cone. HNOs until the solution contains approximately 10%. Then add 1 cc. of cone. H 2 SC>4. Electrolyze the solution in a beaker, starting with 2 amperes current and gradually working up to 3, as the blue color of the solution disappears. Use either a rotating anode or a rotating cathode and continue the electrolysis for forty-five minutes. At the end of this time stop the rotation of the electrode but do not turn off the electrolysis current. Lower the beaker and shut off the current just before the electrodes come out of the solution. Quickly wash the electrodes with a stream of distilled water from a wash bottle. Remove and immerse imme- diately in methyl alcohol. Then burn off the alcohol in the air keeping the electrode in constant motion. Cool in a desiccator and weigh. The cathode contains the Cu in the metallic state and the anode contains the Pb as Pb0 2 . Test the solution for complete removal of Cu and Pb by further electrolysis with fresh electrodes and weigh any further deposit which may form. CALCULATION. PbO 2 X 0.8643 = Pb.* NOTES. (1) In case no Pb is present, the addition of 4 cc. of cone. H 2 SO 4 before electrolyzing aids materially in getting a good copper desposit. (2) Extreme care must be exercised that the electrodes are completely dry before weighing. The alcohol should be changed frequently, as it soon absorbs water and drying becomes difficult. (3) In case there is only a trace of Pb present, a separate 5-gram sample should be run for this element alone. (b) In the case of alloys containing more than 10% of Pb the PbO 2 is likely to flake off the anode and electrolysis is not a suitable method. Proceed therefore as follows: To the filtrate from the Sn determination, add 10 cc. of cone. H 2 S04. Evaporate the solution until fumes of 80s come off * Empirical factor (page 137). 146 TECHNICAL METHODS OF ANALYSIS freely. Then cool thoroughly and add 150 cc. of water; boil and let stand until perfectly cool. Filter the PbSC^ on a weighed Gooch crucible, and wash with a 5% solution of H2SO4, and then with a mixture of equal parts of alcohol and water until the washings are free from acid. The alcohol should not be run into the main filtrate as it will interfere with the subsequent elec- trolysis for Cu. It is used merely to remove the acid of the first washing liquid. Dry in the oven to remove alcohol and moisture, then set the Gooch crucible inside of a platinum crucible and ignite to a dull red heat. Cool in a desiccator and weigh as PbSO.*. Cal- culate to Pb. CALCULATION. PbSO 4 X 0.6833 = Pb. NOTE. In case this procedure is followed, the Cu is determined by elec- trolysis of the filtrate as above described. Zinc. To the filtrate * from the Cu and Pb determinations add 5 cc. (not more) of cone. H2SO4 and evaporate to the appear- ance of 80s fumes. Cool, rinse into a suitable beaker, and dilute to approximately 150 cc. Add 50 cc. of 30% NaOH solution and electrolyze the solution, using a current of 3 amperes and 2.5 volts, and a cathode which has been plated with Cu. The cathode used previously in the determination of Cu is very suitable. The Zn generally deposits out in fifteen minutes. After weighing the deposit, however, dissolve off the Zn in HC1 (1:1), wash, dry and weigh ; then electrolyze for ten minutes more in the same solution to insure the complete removal of all Zn. The manipulation for this determination is exactly as in the Cu determination and the Zn is weighed as metallic zinc. NOTES. (1) The amount of zinc weighed on the electrode should not be over 0.1 gram. If, therefore, the alloy contains more than 10% of Zn, dilute the filtrate from the Cu and Pb determinations to a suitable volume in a volu- metric flask and take an aliquot that will yield not more than 0.1 gram of Zn. (2) If Pb has been determined as PbSO 4 and 10 ccy of H 2 SO 4 used, add 25 cc. more of the 30% NaOH solution to take care of the extra acid. (3) Blanks should be run frequently on the NaOH to make sure that it contains no Zn. J. T. Baker's c. p. NaOH (electrolytic) has been found satis- factory. *0r to an aliquot. (See Note 1.) ANALYSIS OF METALS 147 Zinc (Alternative Method). In the nitrate from the electrol- ysis of Cu and Pb precipitate the Fe and Al as hydroxides by the addition of an excess of NHiOH. Filter and determine Zn in the nitrate as follows: Add cone. HC1 to the above solution until it is faintly acid, then add 2 cc. in excess. Now add an excess 'of a 10% solution of NEU or Na phosphate and heat to boiling. Add NELiOH rapidly until all the precipitate redissolves. Bring back to acidity slowly with acetic acid and add 1 cc. excess. Stir briskly, care being exercised that the policeman does not touch the sides of the beaker, until the precipitate becomes crystalline. Set the beaker aside for several hours, filter on a weighed Gooch crucible, wash with hot water, ignite and weigh as Zn2P2O7. Calculate to Zn. CALCULATION. Zn 2 P 2 7 X 0.4289 = Zn. Manganese. Dissolve 10 grams in 70 cc. of HNOs (1 : 1), adding only a small amount of acid at a time to avoid loss by foam- ing. Evaporate to approximatly half volume. Add 30-40 cc. of water, filter off any Sn0 2 and wash the precipitate thoroughly with hot water. In case both Mn and Fe are desired, make up the solution to volume, retaining one-half for the determination of Fe and the remainder for the Mn. determination. Most sam- ples of so-called manganese bronze contain a, very small amount of Mn, in which case take an aliquot representing 5 grams. In no case, however, should the aliquot taken for analysis contain over 0.015 gram of Mn. Remove the Cu and Pb by electrolysis and evaporate to 50-75 cc.; cool to tap water temperature and add approximately 0.5 gram of sodium bismuthate. Let stand for several minutes with occasional shaking. Place an alundum filtering tube in a Gooch crucible holder, making the connection with rubber. Filter the solution through the alundum with suc- tion. The alundum tube should fit well down into the glass cru- cible holder and the solution should at no time come within 0.5 inch of the place where the rubber connection is made on the out- side of the tube. Titrate the filtered solution at once with stand- ard sodium arsenite solution. NOTE. The arsenite solution is made to contain approximately 1 gram As 2 O 3 per liter. The solution described under Manganese in Steel on page 111, is satisfactory. 148 TECHNICAL METHODS OF ANALYSIS Manganese (Alternative Method). In case the estimations of Sn and Mn only are desired, evaporate the filtrate from the Sn determination to small bulk. Add 50 cc. of cone. HNOs (which must be water-white) and bring to a boil. Add 2-3 grams of KClOs, a few crystals at a time, and boil until the Mn(>2 is com- pletely precipitated and all nitrous fumes are driven off. Filter through a Gooch crucible and wash free from add with hot water, allowing all HNOs to run through the filter before starting to wash with water. Transfer the asbestos and precipitate to the original flask and add a measured excess of standard ferrous ammonium sulfate solution, 5 cc. at a time. Agitate until the brown MnO2 is dissolved. It may be necessary to break up some of the lumps with a stirring rod before solution is possible. Now titrate the excess of ferrous iron to a pink color with standard KMn04. (The same standard ferrous ammonium sulfate and KMn(>4 solutions may be used as are used for the determination of Chro- mium in Steel, page 122.) Deduct from the amount of ferrous ammonium sulfate used (expressed in terms of the KMn04 solu- tion) the amount of KMnCU required for the back titration. The difference is the amount of standard KMnCU solution consumed by the Mn present in the precipitate. Calculate the Mn factor of the KMnC>4 solution by multiplying its Fe factor by 0.4919. Iron. Use a 5-gram aliquot from the Mn determination, or, in case this element is not determined, dissolve 5-10 grams of the finely divided sample in HNOs, boil until nitrous fumes are ex- pelled and dilute with 100 cc. of water. Filter off any meta- stannic acid and wash thoroughly with hot water. Add a small amount of NILiCl to the filtrate and an excess of NH4OH suf- ficient to redissolve any white Zn(OH)2 formed. Heat to boiling, filter and wash the precipitate with hot water until the washings are free from Cu. Dissolve in hot HC1 (1:1) and repeat the process. After the second precipitation, dissolve the Fe(OH)s by pouring hot H2SO4 (1:4) through the filter paper, washing the filter thoroughly with the 1 : 4 acid and hot water. Dilute until the solution contains 5% of H2S04. Meanwhile prepare the Jones reductor* as follows: * The Jones reductor consists of a glass tube about 30 cm. long and 18 mm. inside diameter with a glass stopcock at the bottom. In filling it, first place a platinum spiral or a few glass beads in the bottom and then a plug of glass ANALYSIS OF METALS 149 Connect the suction bottle with a vacuum pump, fill the reductor while the stop-cock is closed (or nearly so) with warm 5% H2S04 and then open the stop-cock so that the acid runs through slowly. Continue to pour acid in until 200 cc. have passed through, then close the cock while a small quantity of liquid is still left in the funnel. Remove the filtrate and again pass through 100 cc. of warm 5% acid. Test this with 0.1 N KMnO4 solution. A single drop should color it permanently; if it does not, repeat the wash- ing. Be sure that no air enters the reductor. (If it is impos- sible to obtain an acid solution which does not color with 1 drop of KMnC>4 solution, the entire filtrate may be titrated and the determination used as a negative " blank " in the determination.) Pour the acid iron solution while hot (but not boiling) through the reductor at a rate not exceeding 50 cc. per minute. Wash out the beaker with 5% H^SCU and follow the iron solution with- out interruption with 175 cc. of warm acid, and finally with 75 cc. of distilled water, leaving the funnel partially filled. Remove the filter bottle and cool the solution under the water tap. Add 10 cc. of dil. H 2 S0 4 and titrate to a faint pink with 0.05 N KMn0 4 solution directly in the filter bottle. Calculate the titration to Fe. CALCULATION. 1 cc. 0.05 N KMn0 4 = 0.00279 gram Fe. NOTE. If the alloy contains tin, the iron obtained from the purification of the SnO 2 precipitate must be added to that found in the main solution. Arsenic. Weigh 3-10 grams of the sample (according to the amount of As present) into a Kjeldahl flask; add a solution of FeCls (made by dissolving 20 grams Fe20s in 150 cc. cone. HC1) ; distill slowly, collecting the distillate in a liter Erlenmeyer flask. When the distillation is complete (after two-thirds of the solution has distilled over) neutralize the distillate with stick NaOH or strong NaOH solution; acidify again with HC1 and bring back until faintly alkaline with NaHCOs solution (the solution must be wool followed by a thin layer of asbestos such as is used for Gooch crucibles. Finally fill the tube with amalgamated zinc to within about 5 cm. of the top and cover it with a little glass wool serving as a filter. The amalgamated zinc may be prepared as follows: Dissolve 5 grams of mercury in 25 cc. of HNO 3 (1 '. 1), dilute to 250 cc. and transfer to a heavy liter flask. Add to the solution 50 grams of granulated zinc (20-30-mesh size), shake the mixture thoroughly for 1-2 minutes and then pour off the solution and wash the zinc free from acid with distilled water. 150 TECHNICAL METHODS OF ANALYSIS kept cool during this process) ; add a few drops of starch solution and titrate with 0.01 N iodine solution; calculate to As. A " blank " distillation should always be made, using the same amount of reagents but omitting the sample. The titration obtained on the " blank " should be subtracted from that required by the sample. The iodine solution must always be freshly standardized against pure As 2 03 and the factor thus obtained. Standard 0.01 N Iodine Solution. Dissolve 1.27 grams of resublimed iodine and 2 grams of KI in 200 cc. of water in a cas- serole. When the solution is complete, transfer to a graduated flask and dilute to 1 liter. (Or, dilute 100 cc. of the laboratory stock solution of 0.1 N iodine to 1 liter.) Standardize the solu- tion against c. P. As 2 C>3 as follows: Weigh out 0.1000 gram of As 2 3 into a small beaker, dissolve in NaOH solution (2-3 grams of NaOH and 10 cc. of H 2 O). Dilute to 250 cc. in a volumetric flask with CO 2 -free distilled water. Acidify 50 cc. of this stock solution with 5 cc. of cone. HC1 and then make slightly alkaline with NaHCOs. Titrate with the iodine, using starch as an indicator. Calculate the arsenic factor of the solution. 0.00757 CALCULATION. As factor = ,. cc. Iodine required Antimony. Dissolve 5-10 grams of the sample in a 500 cc. Erlenmeyer flask in 30 cc. of cone. HNOs. Add sufficient pure tin so that the ratio Sn : Sb is at least 3:1, otherwise the Sb will not be completely precipitated. Evaporate the solution to 10 cc. and then dilute with hot water to 250 cc. Boil for fifteen to twenty minutes and then let the precipitate settle with the flask at an angle of 45. The precipitate should settle out well. If it does, carefully decant as much of the solution as possible, but do not lose any of the precipitate. If the precipitate does not settle out well, decant through a Gooch crucible and then return the precipitate and asbestos to the flask. If much Cu is present, it is advisable to add more hot water and let settle; then decant again. The object is to get rid of most of the Cu. Add 15 cc. of cone. H 2 SO4 and 4-5 grams of K 2 S(>4 to the flask and evaporate to white fumes. Do not drive off all free H 2 S04 so that the melt becomes hard on cooling. Then add 0.5 ANALYSIS -OF METALS 151 gram of powdered tartaric acid. Heat strongly until the solution becomes light colored, or until all of the carbon is destroyed. If necessary, add a little HN0 3 and heat to the appearance of 80s fumes. This treatment leaves the Sb in the proper condition for titration. Cool, add 50 cc. of water and 10 cc. of cone. HC1, and heat to solution of all that is soluble. Cool very thoroughly. Add 110 cc. more water and 10 cc. more acid. Cool again and then titrate at once with 0.1 N KMnO 4 . The end point is distinct but it fades after a short time. From the known Sb factor of the KMn04 solution, calculate the per cent of Sb. NOTE. The KMnO 4 solution should be standardized against c. p. anti- mony and the value of the solution in terms of Sb calculated, then the personal error in the end point will be the same as in the actual analysis. The same KMnO 4 solution may be used as is described under Antimony in White Metals on page 155. The method of standardization is also there described. Nickel. Whenever a bronze is found to contain over 15% of Pb, nickel is very apt to be present. For the determination, add 5 cc. of cone. H2SO4 to the solution from which Cu and Pb have been removed, and evaporate to SOs fumes. Dilute with 50 cc. of water and heat until all soluble salts are dissolved. Quickly pour this solution into 50 cc. of 30% NaOH solution. Heat to boiling, whereupon any Ni should be precipitated as the apple green Ni(OH)2. Dilute the solution to 150 cc., let the precipitate settle and filter through a weighed Gooch crucible. Wash thor- oughly with hot water. Transfer the filtrate to the original beaker and dissolve all the Ni(OH)2 through the Gooch crucible with the least possible amount of warm dil. H 2 SO4. Wash the Gooch crucible thoroughly and reprecipitate the Ni by pouring the solu- tion into 10 cc. more of 30% NaOH solution diluted to about 30 cc. Heat to boiling. Let the precipitate settle and filter through the same Gooch crucib'e. Wash with hot water, ignite at a white heat and weigh as NiO. Calculate to Ni. CALCULATION. NiO X 0.7858 = Ni. NOTE. If zinc is desired, and has not already been determined, add the second filtrate to the first, evaporate if necessary, and transfer to a 250 cc. beaker for electrolysis. Electrolyze as previously described under Zinc. Aluminum. This element is best determined after removal of zinc by electrolysis. Acidify the solution, which is now alkaline 152 TECHNICAL METHODS OF ANALYSIS with NaOH, with cone. HC1 and precipitate Fe and Al by adding a slight excess of NH 4 OH and bringing to a boil. Let settle, filter and wash thoroughly with hot water. Ignite strongly in a platinum crucible and weigh as Fe203+Al203. Fuse with a considerable excess of anhydrous KHS04, take up the fusion with H 2 O and add sufficient cone. H2S04 to give an acid concentration of 5%. Warm the solution, pass through a Jones reductor and titrate the Fe with 0.05 N KMnCU. Calculate to Fe20s and subtract from the weight of combined oxides to obtain the AbOs. Calculate to Al. CALCULATION. A1 2 O 3 X 0.5303 = Al. NOTE. Inasmuch as the NaOH which has been used for the zinc electrol- ysis may contain an appreciable amount of Fe or Al, a blank determination should be made upon an equal amount of NaOH solution and any Fe and Al so obtained should be subtracted. REFERENCE. Price and Meade: " Technical Analysis of Brass and the Non-ferrous Alloys." NICKEL SILVER General. Nickel Silver (German Silver) is an alloy of Cu, Ni and Zn, containing occasionally small amounts of Pb, Fe, Mn and other minor impurities. The method of analysis here described has been devised to take advantage of the well-known dimethylglyoxime precipitation of Ni and has given excellent satisfaction in this laboratory. Repeated comparisons with other methods have demonstrated that it gives accurate results and requires much less time and manipulation. Copper. Weigh 1 gram of the alloy into a beaker and dissolve in 15 cc. of HNOs (1 : 1). When solution is complete and all nitrous fumes have been removed by boiling, dilute to a suitable volume for electrolysis, adding sufficient cone. HNOs to give an acid concentration of approximately 10% if rotating electrodes, or 3% if the stationary type is employed. When using rotating electrodes and a 10% acid concentration, a current density increas- ing from 2 amperes at the beginning to 3 amperes near the end of the electrolysis, works very satisfactorily. In the case of sta- tionary electrodes and a 3% acid concentration, a current of 0.5 ampere is preferable. With rotating electrodes it is possible to ANALYSIS OF METALS 153 remove all copper in one hour. The metallic Cu is weighed as such on the electrode in the usual manner. (See page 145.) Lead. Small amounts of Pb may be present as an impurity in the Zn used in the alloy. This will be entirely deposited on the anode as Pb(>2 and may be weighed as such, calculating to metallic Pb. CALCULATION. Pb0 2 X 0.8643 = Pb.* Nickel. After the removal of the Cu and any Pb present transfer the solution to a 500 cc. volumetric flask and dilute to the mark. Pipette 100 cc., representing 0.2 gram, into a 600 cc. beaker. Make slightly but distinctly alkaline with NH4OH, dilute to approximately 400 cc., heat to incipient boiling and filter off any Fe(OH)3 precipitate. In the nitrate, precipitate nickel as nickel glyoxime with 75 cc. of saturated alcoholic solution of dimethylglyoxime. Heat to slightly below boiling for one hour, filter on a Gooch crucible previously ignited and weighed, and wash with hot water. Always test the first few cc. of the filtrate with a few drops of dimethylglyoxime reagent to make certain that no Ni remains unprecipitated. If it is found that some Ni still remains in the fil- trate, return the latter to the beaker, add a few cc. more of the reagent, heat just below boiling for fifteen to twenty minutes more and again filter. If all Ni is now precipitated, filter, wash, dry for one hour at 105-120 C., cool and weigh. Calculate to metallic Ni and multiply the result by 5 to correct for aliquoting. CALCULATION. Nickel glyoxime X 0.203 1 =Ni. Zinc. Transfer the filtrate from the Ni precipitate to a beaker and evaporate until the odor of NHs can no longer be detected. Cool the solution, transfer to a 500 cc. Erlenmeyer flask to avoid loss by spattering, add 5-6 cc. of cone. H2SO4 and evaporate the solution to the appearance of dense white fumes. Cool thoroughly and take up with water. Add 50 cc. of 30% NaOH solution and electrolyze on a copper coated electrode according to the method on page 146. Weigh as metallic Zn and multiply the weight by 5 to correct for aliquoting. Impurities. The determination of Pb has been already described. Fe, Mn and other impurities may be determined as in Brass and Bronze (see page 142). NOTE. This method was adapted bv H. C. Parish of this laboratory. * This is an empirical factor ^ee under Spelter, page 137). 154 TECHNICAL METHODS OF ANALYSIS WHITE METALS General. This method applies to the analysis of white alloys containing Sb, Sn, Cu, Zn, Pb, Fe, Ni, Mn, Al, and Mg. These are most conveniently divided into (1) Solders, (2) Babbitts, (3) Type Metals, and (4) Aluminum Alloys. In those cases where two methods are given for the same determination, the first method is preferable and the alternative method should be used only when made necessary by lack of chem- icals or apparatus for the first method. Sampling. A representative sample is best obtained by the use of a hacksaw. The sawdust from the sample is especially desirable because the accuracy of some of the methods depends upon the fineness of the sample. In case a hacksaw cannot be used, obtain fine drillings with a small drill. A magnet should always be passed through the sample before starting the analysis. (1) SOLDER General. Solders usually consist of Sn and Pb with a small amount of Sb as impurity. Tin (Gravimetric Method). Weigh accurately 0.5 gram of sample into a 250 cc. beaker, add 15 cc. of HNOs (2:1), cover immediately with a watch glass, and, when violent action ceases, boil until no more red fumes are given off. Place the beaker on the water bath and digest thirty minutes. Dilute with 30 cc. of hot water. Let the precipitate settle and filter, washing with 2% HNOs. Place the filter paper and contents in a weighed por- celain crucible. Smoke off the paper and ignite fifteen minutes in the full heat of a Tirrill burner. Cool in a desiccator and weigh. This represents SnC>2+ 80204, together with a small amount of PbO. Fuse the ignited precipitate with a mixture of finely powdered sulfur and Na2CO3, in the proportions of 1 part of precipitate to 3 parts each of S and Na2COs, in a covered por- celain crucible until the odor of SO2 has disappeared. Cool and dissolve the fusion in hot water; add an excess of Na2SOs to con- vert any polysulfides to monosulfides; filter and wash the pre- cipitate thoroughly. This precipitate contains the Pb. Dissolve it in dil. HNOs (1 : 3) and determine the Pb electrolytically in the usual way, ANALYSIS OF METALS 155 weighing as PbO2 on the anode. Calculate the weight thus found to PbO and subtract from the original weight of the pre- cipitate. (Also calculate it to Pb and add to the main Pb deter- mination as later described.) Also calculate the Sb (determined as described later) to Sb2O4 and subtract this from the original precipitate. The remainder is SnO2. Calculate this to Sn.- CALCULATIONS. Sb X 1.2662 = Sb 2 O 4 . Pb0 2 X0.9331 = PbO. PbO 2 X 0.8643 = Pb.* SnO 2 X 0.7877 = Sn. Tin (Alternative Method). Weigh accurately 0.5 gram of sample into a 300 cc. Erlenmeyer flask fitted with a 1-hole stopper carrying a 1 mm. capillary tube running into a 6-inch test tube as shown in Fig. 8. Add to the flask 25 cc. of 10% Na 2 CO 3 solution, 50 cc. of hot water, 50 cc. of cone. HC1 and 1 drop of 5% SbCls solution. Fill the test tube one-third full with 10% Na 2 CO3 solution and insert the capillary tube into it. Place on the water bath and heat until all except the Cu and Sb is dissolved, which requires about fifteen minutes. Remove the flask from the water bath, fill the test tube with 10% Na 2 C0 3 solution and cool under running water. When FlG< 8 - "" A PP aratus , j j , ,1" t *. i for Determination of cool add to the flask 5 cc. of arrowroot starch Tin - n Wnite M e t a i s solution, then 25 cc. of 10% Na 2 CO3 solution and titrate with 0.1 N iodine, covering the flask with a rubber washer which fits over the tip of the burette. The starch and Na 2 CC>3 solution are best added from a pipette by inserting its tip between the stopper and the neck of the flask, allowing as little air as possible to enter the flask as it causes low results. NOTE. Standardize the iodine solution against 0.2500 gram of c. p. tin, following exactly the details of this method. Divide 0.2500 by the number of cc. of I solution used for titration. The quotient is the factor of the solution in terms of tin. Antimony. Weigh accurately 1 gram of sample into a 500 cc. Pyrex Erlenmeyer flask. Add 15 cc. of cone. H 2 SC>4 and 4-5 * Empirical factor. 156 TECHNICAL METHODS OF ANALYSIS grams of K2SO4. Heat until the residue is perfectly white, but do not drive off all the free H 2 SO 4 , which makes the melt hard on cooling. Cool thoroughly, add 50 cc. of 10% tartaric acid solu- tion and 10 cc. of cone. HC1. Heat until all soluble matter is dissolved, cool very thoroughly, add 110 cc. of water and 10 cc. more of cone. HC1, cool again and titrate at once with 0. 1 N KMnO4. The end point is distinct but fades quite rapidly. From the known strength of the KMnCX solution calculate the per cent of Sb. STANDARDIZATION. Standardize the KMnCU solution against c. P. antimony and calculate the value of the solution in terms of Sb; then the personal error in the end point will be the same as in the actual analysis. Use the following procedure: Weigh out exactly 0.1500 gram of c. P. antimony and an equal quantity of c. p. tin. Place in a 500 cc. Erlenmeyer flask, add 15 cc. of cone. H2SO4 and 4-5 grams of K^SCX. Heat until the metals are in solution or entirely decomposed and all separated S is driven off. Do not drive off enough SOs to cause the melt to harden on cooling. Cool, add 50 cc. of 10% tartaric acid solution and 10 cc. of cone. HC1, and heat until a clear solution is obtained. Cool the solution to tap water temperature; then add 110 cc. more of water and 10 cc. more of cone. HC1. Cool again and titrate at once to a pink color with KMnC>4 solution. 0.1500 CALCULATION. Sb factor = ^ r - -. cc. of KMn(J4 used Lead. To the filtrate from the gravimetric Sn determination add 10 cc. of cone. H2SO4, and evaporate until copious fumes of SOs are evolved. Cool thoroughly and take up with 100 cc. of water; boil and let stand until perfectly cold (preferably over- night). Filter the PbSC>4 on a weighed ignited Gooch crucible and wash with 5% H2SO4, and then with 50% ethyl alcohol until the washings are free from acid. Do not run alcohol into the main filtrate as it will interfere with subsequent electrolysis for Cu. It is used merely for removing the acid from the PbSO 4 . Place the Gooch crucible inside of a platinum crucible and ignite. Cool in a desiccator and weigh as PbSC>4. Calculate to Pb. CALCULATION. PbSO 4 X 0.6833 = Pb. NOTE. Any Pb found in purifying the tin-antimony oxides, as previously described, must be added to the amount here found. ANALYSIS OF METALS 157 Copper. Dilute the filtrate from the Pb determination to approximately 200 cc. and electrolyze for copper in a 250 cc. beaker, as described under Brass and Bronze, page 145. Zinc. To the solution from which Cu has been removed, add 75 cc. of 30% NaOH solution and electrolyze as described under Brass and Bronze, page 146. (2) BABBITT METALS General. Babbitts may be divided into two general classes: (A) Those in which Sn predominates. (B) Those containing a high percentage of Pb. (A) High-Tin Babbitts Tin. Weigh out accurately 0.5 gram of sample, transfer to a 250 cc. beaker and determine Sn and Sb as oxides, as described above under Solder. Antimony. Using a 1 gram sample, determine the Sb by the method previously described for Sb in Solder. Copper and Lead. Dilute the filtrate from the Sn determina- tion to about 200 cc. and add cone. HNOs until the solution con- tains approximately 10%. Electrolyze as described under Copper in Brass and Bronze, page 145. The Pb will be deposited at the same time on the anode as Pb02. Calculate its weight to metallic Pb. CALCULATION. PbO 2 X 0.8643 = Pb. * NOTE. To the Cu and Pb thus determined must be added the Cu and Pb recovered as impurities from the SnC>2 by the procedure previously described in the method for Solder. At the same time that the Pb from the purification of the SnO 2 is deposited on the anode, the Cu is deposited on the cathode. Zinc. To the liquid from which the Cu and Pb have been removed add 2-3 cc. of cone. H2S04, and evaporate to copious fumes of SOs. Cool, add 50 cc. of 30% NaOH solution, and elec- trolyze for Zn as described under Brass and Bronze, page 146. Copper and Lead (Alternative Method). Treat 1 gram of finely divided sample in a 250 cc. covered beaker with 10 cc. of aqua regia.f Add a little KClOa and heat; then add a little tar- * Empirical factor. 1 Aqua regia. 1 part cone. HNO 3 , 3 parts cone. HC1. 158 TECHNICAL METHODS OF ANALYSIS taric acid and dilute with water. Make slightly alkaline with NaOH. If a precipitate forms, make acid again with HC1, add more tartaric acid and then make alkaline with NaOH. When a clear solution is obtained, add 25 cc. of saturated Na 2 S solution. Digest on the steam bath for thirty minutes, stirring frequently. Let the precipitate of PbS and CuS settle; filter on an asbestos mat in a loose bottom Gooch crucible and wash with 2% Na2S solu- tion. Transfer the asbestos mat and crucible to a small beaker and dissolve the sulfides in 20 cc. of cone. HNOs. Dilute to 200 cc. and determine the Pb and Cu by simultaneous electrolysis, (B) High-Lead Babbitts Tin. Dissolve exactly 1 gram of sample in a 250 cc. beaker with 15 cc. of HNOs (2 : 1). If the material is low in Sn, add sufficient pure Sn, accurately weighed, so that the ratio Sn : Sb is at least 3:1, otherwise the Sb will not be completely precip- itated. Complete the determination of Sn as previously described under Solder, deducting the equivalent of pure Sn added. Antimony. Determine Sb in the same manner as described under Solder, page 155. Lead. Determine Pb as described under Solder, page 156. Copper. If Cu is present, electrolyze the solution as pre- viously described for Cu in Solder, page 157. Zinc. If Zn is present, treat the liquid from the Pb deter- mination and electrolyze it as described under Zinc in High-Tin Babbitts, above, (3) TYPE METALS (LINOTYPE, STEREOTYPE, MONOTYPE, ETC.) Tin. Heat 1 gram of the sample, accurately weighed, in a 300 cc. Erlenmeyer flask, with 50 cc. of cone. HC1 on the steam bath until all action ceases. While the solution is still hot add 2 or 3 small crystals of KClOa at a time and shake until a clear solution is obtained. Avoid an excess of KClOs. Now add 50 cc. of cone. HC1, 100 cc. of hot H 2 0, 2 grams of steel drillings (see note 1) and 25 cc. of 10% Na2COs solution, in the order given. ANALYSIS OF METALS 159 Quickly insert the stopper carrying the bent capillary tube and test tube, as shown on page 155. The test tube must be one-third full of 10% Na2COs solution. Place on the steam bath and heat until the steel is entirely dissolved (see note 2). Fill the test tube with 10% Na2COs solution and cool the flask to tap water tempera- ture. Add to the flask 5 cc. of arrowroot starch, then 25 cc. oT 10% Na2CC>3 solution, from pipettes, and complete the deter- mination as described in the alternative method for Tin in Solder (page 155), being careful to exclude air as completely as possible. NOTES. (1) Plain carbon steel drillings, containing 0.35-0.60% carbon, give the best results. They must not be too fine. (2) All yellow color due to chlorine should be destroyed. If not, the analy- sis must be repeated, as too much KC1O 3 has been added and the results will be low. (3) Some trouble may be experienced at first in seeing the end point on account of suspended carbon. This can best be seen by looking down through the bottom of the flask where light shows through the solution. As soon as the blue end point is reached, the solution becomes opaque at this point; and as soon as the carbon settles, the deep blue color can be seen distinctly. (4) A blank determination should be made by each operator, testing 0.1500 gram of pure Sb (but no Sn) and proceeding exactly as described above. This blank titration should be subtracted from the total titration. (5) The iodine solution used should be standardized against pure Sn exactly as described under the determination of Sn in Solder, page 155. Tin (Alternative Method). Weigh accurately 1 gram of sam- ple, together with sufficient c. P. tin so that the ratio Sn : Sb is at least 3:1, into a 250 cc. beaker. Determine the Sn gravimet- rically as above described under Solder, page 154. It will be necessary to purify the precipitate as there described. NOTE. The amount of c. P. tin added should be calculated to SnO 2 and, together with the impurities and Sb 2 O 4 , deducted from the total weight of the precipitate. CALCULATION. Sn X 1 .2696 = SnC>2 . Antimony. Determine Sb on 0.5 gram of the sample exactly as described under Solder, page 155. Lead. Treat 0.5 gram of the finely divided sample in a 250 cc. Erlenmeyer flask with 4-7 grams of tartaric acid (depending upon the amount of Sn and Sb present), 15 cc. of water and 4 cc. of cone. HNOs, and heat on the steam bath until a clear solution is obtained. Add cautiously and with constant stirring 4 cc. of 160 TECHNICAL METHODS OF ANALYSIS cone. H2S04, dilute with 50 cc. of distilled water and let cool until the precipitate settles completely (at least one hour). All Pb will be precipitated as PbSO4. Filter on a Gooch crucible, wash with 5% H2S04 and then with 50% ethyl alcohol until the wash- ings are free from acid. Set the crucible inside a platinum crucible, ignite and weigh as PbS04, as described under Lead in Solder, page 156, (4) ALUMINUM ALLOYS General. In accordance with common practice with high Al alloys, the amounts of all other elements present are determined and the Al taken by difference. This is on account of the diffi- culties that are usually encountered in determining Al where it is present in such large amounts. All the determinations should be made in duplicate. It is always advisable to make a quali- tative analysis before attempting quantitative work. Silicon. Dissolve 1 gram, accurately weighed, of well-mixed drillings in 35 cc. of acid mixture (see below), using a porcelain dish covered with a watch glass. When the drillings are com- pletely dissolved, evaporate the solution to dryness and bake on the hot plate for at least one-half hour. This insures complete dehydration of SiC>2. Take up the residue with HC1 (1 : 4), filter, and wash with hot water. Ignite and weigh as Si0 2 . Cal- culate to Si by multiplying by 0.4693. Acid Mixture. Pour 150 cc. of cone. H2S04 into 450 cc. of water. Cool and add 100 cc. of cone. HNOa and then 300 cc. of cone. HC1. Aluminum. If for any reason it is desired to make a direct determination of aluminum, dilute the filtrate from the Si deter- mination to 500 cc. in a volumetric flask, and, after thoroughly mixing, take a 100 cc. aliquot for analysis. Dilute the solution to approximately 300 cc., add a pinch of tannic acid and a slight excess of NttiOH. Boil until the odor of NH 3 is nearly gone. Filter and wash with hot water. Dissolve the precipitate in hot HC1 (1 : 1), add a pinch of tannic acid, reprecipitate exactly as pre- viously, and wash with hot water. Dry the paper and residue, ignite with a blast lamp in a platinum crucible to constant weight ANALYSIS OF METALS 161 and weigh as Al203+Fe203. Subtract the Fe20s present (as later determined) and calculate the remainder to Al. CALCULATION. A1 2 3 X 0.5303 = Al. Manganese. Dissolve 1 gram of finely divided sample in 50 cc. of HNOs (1 : 3) in a 200 cc. Erlenmeyer flask. Since Al dissolves very slowly in HNOs, this may require some time. When entirely in solution, cool to about 15 C., add 0.5 gram of sodium bismuth- ate and agitate for a few minutes. Add 50 cc. of 3% HNOs, and filter through an extra-porous alundum thimble. Wash with 50-100 cc. of the same acid and finally with water. Do not let the solution come in contact with rubber connections. Titrate the filtrate in the flask with standard sodium arsenite solution (see note 1) as described under Manganese in Steel, page 111. NOTES. (I) The total amount of Mn titrated in this way should not be over 0.0125 gram. If, therefore, the material contains over 1.25% of Mn, use a 0.5 gram sample. (2) Arsenite Solution. The arsenite solution which is used for titrating Mn in Steel will be suitable (see page 111). (3) In case the sample is not fine, it may be necessary to use a small amount of HC1 to effect solution, and subsequently drive this off by evaporating to a small volume with successive portions of cone. HNO 3 , before proceeding with the method above described. Tin. Weigh accurately a 1 gram sample, together with approx- imately 5 grams of NaOH, into a beaker and add just enough water to cover the sample. When violent action has ceased, dilute to about 100 cc. and boil for four or five minutes. Cool, filter, and wash the residue with hot water until free from alkali. The Al and practically all the Zn are dissolved, while other metals remain unaffected. Save the filtrate for subsequent determina- tion of Zn. Wash the bulk of the residue from the NaOH treatment into a beaker. Ash the filter and add it to the beaker. Add 20 cc. of cone. HNOs and evaporate to approximately 10 cc. Dilute to 50 cc. with water. Filter, wash, ignite in a porcelain dish, and weigh as SnC>2. Calculate to Sn. CALCULATION. Sn0 2 X 0.7877 = Sn. Copper and Lead. Bring the volume of the filtrate to 200 cc. and add 10 cc. of cone. HNOs. Electrolyze for Cu and Pb as described under Brass and Bronze, page 145. 162 TECHNICAL METHODS OF ANALYSIS Iron. Make the solution from the Cu and Pb determination alkaline with NHiOH, and boil. Filter off the precipitate of Fe(OH)3 and wash with hot water. Dissolve this precipitate from the filter by pouring hot dil. H2SO upon it, catching the solution in a beaker. Dilute this solution until the H^SC^ present is approximately 5%. Heat nearly to boiling and pass through a Jones reductor. Cool and titrate with 0.1 N KMn0 4 solution. CALCULATION. 1. cc. 0.1 N KMn0 4 = 0.005584 gram Fe. = 0.007984 gram Fe 2 3 . NOTE. A blank should be run on the reductor (using the same amounts of acid and of solution) and subtracted from the titration of the sample. (See page 148.) Nickel. Heat to boiling the above filtrate from the Fe(OH)s precipitate and add a 1% solution of dimethylglyoxime, (CHs)2C2(NOH)2, in ethyl alcohol, until the amount of reagent is about 5 times that of the Ni supposed to be present. A large excess does no harm but it is unnecessary and the reagent is very expen- sive. Let the solution stand on the water bath for thirty minutes and filter while still hot through a weighed porcelain Gooch crucible. Wash with hot water and dry to constant weight at 110-120 C. Weigh as nickel glyoxime, CgHuN^Ni, and calculate to Ni. CALCULATION. C 8 Hi 4 N40 4 NiX 0.2031 = Ni. Zinc. Dilute the filtrate from the original NaOH treatment in the tin determination to 200 cc. and determine the Zn by elec- trolysis as described in the method for Brass and Bronze, page 146. Discard the solution after the Zn is removed. Some Zn may remain insoluble in NaOH and should be recovered as follows: Boil the filtrate from the Ni determination until the free NHs is expelled. Acidify with dil. H2SO4, adding the equivalent of 2-3 cc. of cone. H2SO4 in excess. Transfer to a 500 cc. Pyrex Erlenmeyer flask and evaporate to 80s fumes. If the solution is dark colored, add a few cc. of cone. HNOs and again evaporate to SOs fumes. Take up with water, cool, add 50 cc. of 30% NaOH solution, dilute to 200 cc. in a 250 cc. beaker and electrolyze for Zn as previously described. Add the Zn obtained here to that obtained by electrolysis of the filtrate from the original NaOH treatment. Magnesium. After any Zn present has been removed by elec- trolysis, acidify with cone. HC1, adding about 10 cc. in excess. ANALYSIS OF METALS 163 Dilute to about 300 cc., cool to tap water temperature and add 30 cc. of a saturated solution of microcosmic salt, NEUNaHPO* -4H20. Add NEUOH drop by drop, stirring vigorously to make the pre- cipitate crystalline. Then add an excess of cone. NHiOH, approx- imately 10% of the volume of the whole solution. Let stand over- night, filter through a weighed Gooch crucible, and wash with cold water containing 5% of cone. NHiOH and 5% of NILiNOs. Ignite until completely white. Weigh as Mg2?2O7 and calculate to Mg. CALCULATION. Mg 2 P2O 7 X 0.2184 = Mg. REFERENCES. Price and Meade: "Technical Analysis of Brass and Non-Ferrous Alloys;" E. Blough: "Analysis of Aluminum and Its Com- mercial Alloys." MERCURY IN ZINC AMALGAM General. So-called Battery Zincs are often made of zinc amal- gam containing 1-3% of mercury. The following method is a convenient one for the determination of the amount of Hg and gives accurate results. Determination. Dissolve 4-5 grams of finely divided alloy in about 75 cc. of HC1 (1:1) and boil two to three hours. This dis- solves Zn, whereas any Hg and Pb which are present will be separated as metal. Decant off the solution, wash the residue several times with hot water and transfer to a porcelain crucible. Place in the oven and dry at 100 C. Cool in a desiccator and weigh. Then ignite at a red heat and again cool in a desiccator and weigh. The loss, represented by the difference in weight, is mercury. NOTE. The ignition should be conducted under a hood and care taken not to breathe the mercury fumes, which are poisonous. REFERENCE. This method has been used for some time at the laboratory of the N. Y., N. H. & H. Railroad. TESTING OF GALVANIZING OR SHERADIZING ON IRON AND STEEL General. There are two methods of applying a protective zinc coating to iron and steel. Galvanizing deposits a layer of metallic zinc on the surface and gives a bright, smooth, shiny surface generally showing the char- acteristic crystalline structure of Zn. 164 TECHNICAL METHODS OF ANALYSIS Sheradizing forms a dull gray (nearly slate colored) coating generally more or less rough. This coating is not metallic Zn as in the case of galvanizing but is a mixture of metallic Zn and a Zn-Fe alloy. The Preece test, described herewith, was originally designed for galvanized articles and particular care must be employed, in applying it to sheradized articles, to brush the specimen vigor- ously with a wire brush after each dip and not to be deceived by an apparent plating of copper which can be removed by further dips and scrubbing. Cleaning. The samples shall be cleaned before testing, first with CCU, benzine or turpentine, and cotton waste (not with a brush),* and then thoroughly rinsed in clean water and wiped dry with clean cotton waste. The samples must be clean and dry before each immersion in the test solution. Solution. The standard solution of copper sulfate shall con- sist of commercial copper sulfate crystals dissolved in cold water, in the proportion of about 36 parts by weight of crystals to 100 parts of water. The solution shall be neutralized by the addition of an excess of chemically pure CuO. The presence of an excess of CuO will be shown by the sediment of this reagent at the bot- tom of the containing vessel. The neutralized solution shall be filtered before use by passing through filter paper. The filtered solution shall have a sp. gr. of 1.186 at 65 F.f (reading the scale at the level of the solution) at the beginning of each test. In case the filtered solution is high in sp. gr., clean water shall be added to reduce the sp. gr. to 1.186 at 65 F. In case the filtered solution is low in sp. gr., a filtered solution of a higher sp. gr. shall be added to make the sp. gr. 1.186 at 65 F. As soon as the stronger solution is taken from the vessel con- taining the unfiltered neutralized stock solution, additional crystals and water must be added to the stock solution. An excess of CuO shall always be kept in the unfiltered stock solution. Quantity of Solution. Wire samples shall be tested in a glass jar of at least 2 inches inside diameter. The jar without the wire samples shall be filled with standard solution to a depth of at least * Except for sheradized articles. t This is equivalent to a reading of 22.7 on a Baiiine* hydrometer at 65 F. ANALYSIS OF METALS 165 4 inches. Hardware samples shall be tested in a glass or earthen- ware jar containing at least 0.5 pint of standard solution for each hardware sample. Solutions shall not be used for more than one series of 4 immersions. Samples. Not more than 7 wires shall be simultaneously immersed, and not more than 1 sample of galvanized material other than wire shall be immersed, in the specific quantity of solution. The samples shall not be grouped or twisted together, but shall be well separated so as to permit the action of the. solution to be uniform upon all immersed portions of the samples. Test. The clean and dry samples shall be immersed in the required quantity of standard solution in accordance with the following cycle of immersions. The temperature of the solution shall be maintained between 62-68 F. at all times during the following test. First. Immerse for one minute, wash and wipe dry.* Second Immerse for one minute, wash and wipe dry. Third. Immerse for one minute, wash and wipe dry. Fourth. Immerse for one minute, wash and wipe dry. After each immersion the samples shall be immediately washed in clean water having a temperature between 62-68 F., and wiped dry with cotton waste.* In testing wire of No. 13 B. W. G. and smaller sizes, the time of the fourth dip shall be reduced to one-half minute. Interpretation. If the samples as tested above show bright, firmly adhering metallic copper deposits, they are considered as having failed to pass the test. In the case of sheradized articles there will be a copper deposit formed on each dip, due to the Fe-Zn alloy. This deposit, however, is not firmly adhering nor of the typical bright color of the Cu plate which indicates failure. It can be removed by vigorous scrubbing with a wire brush under running water and care must be taken not to reject sheradized articles unless they show a firmly adhering Cu deposit which is bright colored and cannot be removed by wire brushing or by further dipping and brushing. * In the case of sheradized materials the specimens must be quickly removed to running water after each dip and brushed vigorously with a wire brush while in the running water. 166 TECHNICAL METHODS OF ANALYSIS Copper deposits on zinc or within 1 inch of the cut end may be disregarded. If one wire in a group of 7 wires immersed together shows a Cu deposit, or if there is reasonable doubt as to the presence of the Cu deposit, two check tests should be made upon the 7 wires and the report based on the majority of the sets of tests. In case the article has a cut screw the thread shall stand one one-minute immersion; the rest of the article shall withstand the specified 4 immersions. The threads of nuts are not required to stand the galvanizing test. REFERENCE. Amer. Tel. & Tel. Co. Specification 3110, Feb. 3, 1908. TINNING TEST FOR TINNED IRON AND STEEL Method of Am. Elect. Railway Engineering Assoc. (a) Preparation of Samples. Samples of the wire or other material to be tested shall be thoroughly cleaned with alcohol. '(6) Tinning Test. Immerse in HC1 (sp. gr. 1.088 at 15.5 C.) for exactly one minute. Rinse in pure water and immerse in an aqueous solution of sodium sulfide (sp. gr. 1.142, or 28 Baume) for exactly thirty seconds. Again wash in pure water and repeat the operation three times. At the end of the fourth immersion in Na2$, the wire shall show no sign of blackening. (c) Sodium Sulfide Solution. The Na2$ solution shall contain an excess of sulfur and shall have sufficient strength to thoroughly blacken a piece of untinned copper wire in five seconds. Each new solution made up shall be tested for strength with a piece of untinned copper wire. Method of American Tel. and Tel. Co. Immerse the samples of wire or other material to be tested in a current of pure H^S gas, saturated with water vapor, at a temperature of not less than 24 C. nor more than 26 C., for four hours. At the end of this time the samples should show no signs of blackening. CHAPTER V ANALYSIS OF FUELS COAL SAMPLING General. The main object in taking a sample of coal is to secure a small portion of the coal which represents as nearly as possible the entire shipment. It is extremely important that the sample be properly taken, as in many cases it is entirely impos- sible to obtain another sample. Wherever possible collect the sample while the coal is being loaded into or unloaded from cars, boats, trucks or other conveyor. When coal is being crushed as received it is often advantageous to take the sample as it comes from the crusher. Samples taken only from the surface of piles or bins are generally unreliable. Gross Sample. In collecting the sample use a shovel or other tool which will take equal portions. For slack coal or small sizes of anthracite each shovelful may be as small as 5-10 Ibs., but for lump coal or run-of-mine the amount of each shovelful should be 10-30 Ibs. The total sample obtained in this way is called the gross sample. Wherever possible the gross sample should not be less than 1000 Ibs., except in the case of slack coal and small sizes of anthracite (not greater than f inch), where 500 Ibs. are sufficient. If the coal contains an unusual amount of slate or other impurities and if the pieces of such impurities are very large, collect a gross sample of 1500 Ibs. or more. The gross sample should contain the same proportions of lump, fine, and impurities as in the coal sampled. The gross sample should preferably be collected in a large receptacle with cover attached. A gross sample should be taken for each 1000 tons or less delivered, unless otherwise specified. When necessary to sample from loaded car or in bins or piles, the shovelfuls should be taken by a systematic plan in sufficient 167 168 TECHNICAL METHODS OF ANALYSIS First stage in the preparation of I.OOO-pound sample Crush 1,000-pound sample on hard, clean surface to I" size I.OOO-pound sample crushed to 1" and coned Second stage. Mix by forming long pile. A spreading out first shovelful. B long pile completed .._.. Crush 500-pound sample 500 pounds Crushed to&" and coned Mix by forming long pile. Third stage. Crush 250-pound sample (fig. 10. 4) to !^" 250-pounds crush'ed to j" n ^ c n *& Mix by forming new cone Fourth stage. Crush 125-pound sample (fig. 16: A, A) Mix by rolling on blanket on blanket to%" size Form coie after mixing Fifth stag*. Crush 60-pound sample (fig. 22: A, A) Mix by rolling-on blanket to !4" size Form cone after mixing Sixth stage. sg? EfEir_L_irr" ^fe 29 Crush 30-pound sample (fig. 26: A, A) Mix by rolling on blanket or 4-mesh size Form cone after mixing FIG. 9. Method of Preparing a ANALYSIS OF FUELS 169 ..5 Halving by alternate shovel method. Shovelfuls 1, 3, 5, etc., reserved as 5, A; 2, 4, 6, etc.. rejected as 5. B Long pile divided into two parts; A reserve; B reject Shovelful _ by alternate shovel method, uls 1,3, 5, etc.. reserved asKX.4; 2, 4, 6, etc., rejected as 10, B NOTE SELECT A HARD, CLEAN SURFACE, FREE OF CRACKS AND PROTECTED FROM RAIN, SNOW, WIND, AND BEATING SUN. DO NOT LET CINDERS, SAND, CHIRPINGS FROM FLOOR, OR ANY OTHER FOREIGN MATTER GET INTO THE SAMPLE. PROTECT SAMPLE FROM LOSS OR GAIN IN MOISTURE Long pile divided into two parts; A reserve; B reject Retain opposite quarters A, A Reject quarters B, B Retain opposite quarters A, A. Reject quarters B, B Quarter after flattening cone Simple divided into quarters Retain opposite quarters A, A. Reject quarters B, B Quarter after flattening i Sample divided into quarters Fill two 5-pound sample containers from A, A, 0/W for laboratory, one for reserve Sample of Coal by Hand. 170 TECHNICAL METHODS OF ANALYSIS number to make the proper gross sample and from as nearly all parts of the pile as possible, discarding the outer surface and securing as nearly as possible the same amount from the top, mid- dle and bottom of the coal. Laboratory Sample. Having secured the gross sample (pro- tected from the weather to avoid loss or gain in moisture), reduce it down to a smaller sample as follows (see Fig. 9) : Break down large lumps of coal and impurities on a clean, hard, dry floor with a suitable maul or sledge until, as judged by the eye, there are no pieces larger than specified. (See table.) Then thoroughly mix by shoveling over and over and form in a conical pile. Reduce this down by the alternate shovel method as follows : Take a shovelful from the conical pile and spread it out in a straight line having a width equal to the width of the shovel and a length of 5-10 feet. Then spread the next shovelful in the opposite direction over the top of the first. Continue this, occasionally flattening the pile, until all the coal has been formed into one long narrow pile. Starting at one end, on the side, take one shovelful from the bottom and set it aside. Then advance along the side of the pile a distance equal to the width of 4 the shovel, take a second shovelful and set it aside in a second pile. Again advance in the same direction one shovel width, take a third shovelful and add to the first. Take the fourth shovelful in the same manner and add it to the second. Proceed in this way, putting even shovelfuls in one pile and odd in another. Con- tinue to advance in the same direction around the original pile, so that its size will be gradually reduced in a uniform manner. Finally there will be two piles containing approximately the same amount of coal. Discard one of these. Spread out the other and crush down to the size indicated in the table. Then form in a conical pile, followed by a long narrow pile, and again reduce by the alternate shovel method. After the gross sample has been reduced by the above method to approximately 250 Ibs., further reduce it in quantity by the quartering method. Before each quartering crush the sample to the prescribed fineness. For quartering, form into a conical pile, and flatten out the cone. Divide with a shovel or board into 4 equal segments and discard 2 opposite quarters. Mix the 2 remaining quarters, crush down to proper size, mix, reform into ANALYSIS OF FUELS 171 a conical pile and quarter as before. Continue the process until no lumps are greater than 3^ inch (or 4-mesh screen size) and the final sample amounts to about 1-2 quarts. TABLE SHOWING REQUIRED FINENESS FOR GIVEN WEIGHT OF SAMPLE Weight of Sample Largest Size Allowable 1,000 Ibs. or over 1 inch 500 Ibs ! inch 250 Ibs \ inch 125 Ibs f inch 60 Ibs i inch 30 Ibs. or less r 5 inch (4-mesh) In reducing the gross sample, when the quantity reaches less than 125 Ibs. it should be placed on a mixing canvas about 6X8 feet and mixed by raising first one end of the canvas and then the other, so as to roll the coal back and forth. The sample should be worked down as rapidly as possible to avoid loss of moisture and immediately placed in an air-tight can or container. The outside of the container should be plainly marked and the corresponding description placed inside. Data. The following data should accompany the sample: Coal delivered to Sampled by Date Amount taken for original sample Amount of coal sample represents Sampled from barge, car or pile Car (initial and number) Barge or vessel Trade name of coal Grade (f , slack, nut, run-of-mine, etc) Remarks (appearance of coal, lumps, slate, sulphur balls, weather conditions, etc.) Sold by Mined by County State Mine REFERENCE. This is essentially the Standard Method of the American Society for Testing Materials, adopted 1916, Serial D-21-16. 172 TECHNICAL METHODS OF ANALYSIS COAL Proximate Analysis and Heating Value Preparation of Laboratory Sample. (A) When Coal Appears Dry. If the sample is coarser than 4-mesh (^ inch) and larger in amount than 10 Ibs., quickly crush it with jaw crusher to pass a 4- mesh sieve and reduce by a riffle to between 5-10 Ibs. Then grind the whole sample to 20-mesh size. If moisture is of special importance, remove about 60 grams immediately with a spoon from various parts of the sample and place at once in a dry rubber- stoppered bottle. This sample is to be used for the determination of total moisture. Thoroughly mix the main portion of the sample, reducing on a small riffle to about 120 grams, and pulverize to 60-mesh. If desired, this 60-mesh sample may be further reduced by a small riffle to 2-3 ounces. The sample will become partly dry on grinding, hence if moist- ure is important, compute the analysis of the 60-mesh sample to the " bone dry " basis by dividing each result by 1 minus the moisture content expressed as a decimal. Compute the analysis of the coal " as received " from the bone dry analysis by multi- plying each figure by 1 minus the total moisture found in the larger 20-mesh sample, also expressed as a decimal. (B) When Coal Appears Wet. Spread the whole sample on weighed pans; weigh quickly and air-dry in a special drying oven at 10-15 C. above room temperature. Continue drying until the loss in weight is not more than 0.1% per hour. Then finish sampling as above under dry coal. When moisture is important correct the moisture found in the 20-mesh air-dried sample to total moisture " as received " as follows: 100 per cent air-drying loss X (per cent H 2 in 20-mesh) + J-UU (per cent air-drying loss) = (total moisture as received) . Compute the analysis to the bone-dry basis and the " as received " basis as under dry coal above, using for the " as re- ceived " computation the total moisture as found by the above formula in place of the moisture found in the 20-mesh coal. NOTES. (1) In the general run of coal analysis the actual moisture is not important, as most purchases are based on bone-dry figures. In such ANALYSIS OF FUELS 173 it is not necessary to take a special 20-mesh sample, but wet coal should always be air-dried before grinding. In the latter case the formula for cor- recting to total moisture " as received " becomes 190 per cent air drying loss r X(per cent H 2 O in 60-mesh sample) + (per 100 cent air-drying loss). (2) Freshly ground or wet coal loses moisture rapidly, hence sampling operations between opening the container and taking the 20-mesh total moisture sample must be conducted as quickly as possible with very little exposure to air. (3) The accuracy of the method of preparing laboratory samples should be checked frequently by re-sampling rejected portions and preparing a duplicate sample. Ash determinations on the two samples should not differ more than the following limits: No carbonates present . 4% Considerable carbonate and pyrite present 0.7% Ash over 12% (containing considerable carbonate and pyrite) 1-0% Moisture. Weigh 1 gram of the 60-mesh sample quickly and accurately in a 15 cc. weighed platinum crucible with tight-fitting capsule cover, previously heated and cooled in a desiccator over H2S04. Place in the special moisture oven with cover removed and dry for exactly one hour at 104-108 C. Remove and place the crucible with tight-fitting cover in a desiccator over cone. H2SO4. When cool, immediately weigh the covered cru- cible. Report the loss of weight as moisture. NOTES. (1) The moisture determination is made in a specially designed double-walled oven. The space between the walls is filled with a glycerin-water mixture of such strength that the boiling point is between 104 C. and 108 C. Concentration of the solution is prevented by means of a reflux condenser fitted into the top of the oven. Air is pre-dried by passing it through cone. H 2 SO 4 at such a rate that the air is renewed in the oven 2-4 times a minute. The pre-dried air is then led through .a coil of block tin tubing which passes through the heated solution of glycerin and water. Air is thus pre-heated before enter- ing the oven. It escapes by means of a small orifice in the door of the oven. (2) Permissible tolerances: Same analyst Different analysts Moisture under 5% 0.2% 0.3% Moisture over 5% 0.3% 0.5% Volatile Matter. Place the crucible with tight-fitting capsule cover containing the residue from the moisture determination on a nichrome wire triangle and heat for exactly seven minutes over a 174 TECHNICAL METHODS OF ANALYSIS Tirrill burner burning artificial gas. The temperature at the bottom of the crucible must be regulated by means of a thermo- couple so that it is maintained at 950 C. (20 C.). Burner and crucible are protected by a wind shield. Cool in air and weigh without disturbing the cover. The further loss of weight thus obtained is reported as volatile matter. Permissible tolerances: Same analyst Different analysts Bituminous coals 0.5% 1-0% Lignites 1.0% 2.0% Fixed Carbon. Place the crucible containing the residue from the Volatile Matter determination (coke) in an artificial gas- fired muffle which is maintained at a temperature of 900-950 C. Continue heating until no particles of unburned coal appear upon stirring with a platinum stirring wire. Cool in air and weigh with cover as soon as cold. Report the further loss in weight thus obtained as fixed carbon. Ash. The residue from the last burning in the muffle is the ash. NOTES. (1) The result thus obtained is " unconnected ash " and is the ash percentage always reported unless otherwise specified. (2) High ash coals require longer heating than low ash coals. (3) Permissible tolerances: Same analyst Different analysts No carbonates present 0.2% 0.3% Carbonates present 0.3% 0.5% More than 12% ash, containing carbon- ate and pyrite 0.5% 1.0% Sulfur. Weigh 1.3734 grams of 60-mesh sample and thor- oughly mix on glazed paper with 3 grams of Eschka mixture. Transfer the uniform mixture to a 30 cc. Coors glazed porcelain crucible and cover with about 1 gram of Eschka mixture. Place the crucible in the cold gas-fired muffle and gradually raise the temperature to 870-925 C. (cherry-red heat) in about one hour. Maintain the maximum temperature for about one- half hour and then let the crucible cool in the muffle. When burning is complete, all trace of black coal will have disappeared and only a light, reddish gray mass remains. Make sure such is the case. ANALYSIS OF FUELS 175 When cool, empty the contents into a 150 cc. beaker and digest with 75-100 cc. of hot distilled water on the hot plate for thirty to forty-five minutes with occasional stirring. Then filter through a rapid 11 cm. filter paper. Wash the insoluble matter by decant ation. After several washings in this manner, transfer the insoluble matter to the filter and wash 5 times with hot dis- tilled water, keeping the mixture well agitated. Treat the fil- trate, amounting to about 250 cc., with 10 cc. of saturated bro- mine water, make slightly acid with cone. HC1 (5 cc.) and boil to expel liberated Br. Make sure the solution is acid by testing with litmus. Add slowly from a pipette, with constant stirring, 10 cc. of a 10% solution of BaCl2-2H20. Continue boiling for fifteen minutes and let stand overnight just below boiling point. The solution is now clear and the BaSQi precipitate is granular. Filter through an ashless 11 cm. filter paper and wash with hot distilled water until a drop of AgNOs solution shows no precip- itate in the filtrate. Place the wet filter containing the pre- cipitate of BaSC>4 in a weighed alundum (No. or 00) or platinum crucible, allowing free access of air by folding the paper over the precipitate loosely to prevent spattering. Place in a cold muffle and smoke the paper off gradually. At no time let it burn with a flame. After the paper is practically consumed, raise the tem- perature to approximately 900 C. and heat to constant weight. Cool and weigh the crucible and precipitate. Calculate to sulfur. CALCULATION. Grams of BaSO4XlO = per cent sulfur. NOTES. (1) Always run a blank determination with each analysis, using the same amounts of all reagents that were employed in the regular determina- tion. Deduct the sulfur found in the blank from the amount found in the sample. (2) Examine the residue of Eschka mixture for sulfur after digesting by dissolving it in HC1 and treating with bromine water and BaCl 2 . When an appreciable amount of sulfur is found add it to the main precipitate. (3) Determinations of -ash in coal or coke must not be made in the same muffle at the same time with sulfur determinations. (4) Reagents and Solutions. Eschka Mixture: Mix 2 parts by weight of light calcined MgO and 1 part by weight of anhydrous Na 2 CO 3 . Both mate- rials should be free as possible from sulfur. Grind the materials together and after thoroughly mixing pass through an ordinary flour sieve. Keep the mixture in glass-stoppered bottle. Saturated Bromine Water: Add excess of bromine to 1000 cc. of distilled water and mix, 176 TECHNICAL METHODS OF ANALYSIS Barium Chloride Solution: Dissolve 100 grams of BaCl 2 -2H 2 O in distilled water, and dilute to 1000 cc. (5) Permissible Tolerances: Same analyst Different analysts Sulfur under 2% 0.05% 0.10% Sulfur over 2% 0.10% 0.20% Heating Value (B.T.U.). The calorimetrrc determination for British Thermal Units is made' in an Emerson bomb calorimeter with a gold lining. Weigh accurately 1 gram of the 60-mesh sample into the fuel tray lined with recently ignited asbestos. Bituminous coal may be used either in the form of a briquette or as a powder. In the cases of anthracite and coke, the powder is always used. Set the tray in place with the iron wire connecting the terminals touching the coal. The iron wire used should be about No. 34 B. & S. gauge and the heat due to it must be subtracted from the final result. 3-| inches of No. 34 iron wire have a heating value of 25 B.T.U. Place 1-2 cc. of water in the bottom of the bomb to saturate with moisture the oxygen used for combustion. Screw the lid down tightly against the lead gasket. Then force oxygen into the bomb very slowly until the pressure within registers 18-20 atmospheres. At this point close the needle-point valve just tight enough to prevent leakage of gas. Place the bomb in the bucket containing exactly 1795 grams of water at a temperature about 4 C. lower than room temperature. In the cases of anthracite screenings and fuels whose heating value is low, the temperature of the water should be about 3.5 C. lower than room temperature. The initial temperature of the water in the bucket should be so adjusted that the final temperature after combustion will be approximately 0.5 C. above room tempera- ture, under' which conditions the total correction for heat gained from or lost to the surroundings will be small when the rise of temperature is 3.5-3.7 C. Connect the current terminals, adjust the stirring apparatus, cover the calorimeter jackets, attach the stirring device and insert the thermometer. Before taking any readings, allow the stirrer to mix the water thoroughly for two or three minutes. Thermometers used for all temperature observations in cal- ANALYSIS OF FUELS 177 orimetric work should be graduated to 0.01 C. and calibrated by the U. S. Bureau of Standards. Correct all readings according to the calibration certificate from the Bureau of Standards. Before each reading tap the thermometer slightly to avoid errors caused by lag of the mercury meniscus. Read all temperature observa- tions to 0.001 C. with the special reading lens. The actual determination is divided into three five-minute periods: (1) the preliminary period; (2) the combustion period; and (3) the final period. Take readings at intervals of one minute during the preliminary period. Immediately at the end of the fifth minute turn on the current of approximately 12 volts for about one-half second, which ignites the coal. Continue readings at intervals of one-half minute during the combustion period. Then take readings at intervals of one minute during the final period. After the last reading of the final period take the bomb out of the bucket and reduce the pressure in the bomb to atmospheric pressure by opening the needle-point valve. Remove the lid and very carefully inspect the interior of the bomb for traces of un- burned coal and iron wire. If any trace of unburned coal is evident, the determination is worthless. A correction for any unburned wire must be made. Wash out the bomb and fuel tray thoroughly with distilled water and titrate the washings with methyl orange and standard Na2COs solution of such strength that 1 cc. is equivalent to 0.004896 gram of HN0 3 , the heat of formation of which is 2 B.T.U. Make a. correction of 23 B.T.U. for each per cent of sulfur present. This corrects for the difference between the heats of formation of HNOs and of 862 to SOs, and also for the Fe present in the coal as pyrites being burned to Fe2Os. SOLUTIONS. (1) Titrating Solution. Dissolve 4.122 grams of c. P. anhydrous Na2COs in 1000 cc. of distilled water. 1 cc. of this solution is equivalent to 2 B.T.U. per pound. (2) Indicator. Dissolve 1 gram of methyl orange in 1000 cc. of distilled water. CALCULATION. The difference between the final and the initial corrected temperature observations of the preliminary period, divided by 7, gives the rate of change during the preliminary period. 178 TECHNICAL METHODS OF ANALYSIS The difference between the final and initial corrected tem- perature observations of the combustion period gives the cor- rected observed rise in temperature during the combustion period. The difference between the initial and final corrected tempera- ture observations of the final period, multiplied by f, gives the rate of change during the final period. When the final reading is greater than the initial reading, the difference is a minus quantity; and when the final reading is less, the difference is plus. Add the algebraic sum of the rates of change of the preliminary and final periods to the observed rise of temperature of the com- bustion period, and multiply the corrected rise in temperature by the water equivalent factor of the apparatus. This product is the iota 7 heat developed, expressed in B.T.U. This must be corrected for Fe wire, sulfur, and " titer " (acid formed) as below. Multiply the " titer " reading (number of cc. of Na2COs solu- tion) by 2, since each cc. is equivalent to 2 B.T.U. Add this product to the heating value of the Fe wire used, and deduct the sum from the total heat developed. Multiply the per cent of sulfur in the sample by 23 and deduct this product also from the total heat developed. The result after all these corrections is the heating value of the sample in B.T.U. per pound. Permissible Tolerances : Same analyst Different analysts 12,000 B.T.U 36 60 13,000 B.T.U 39 65 14,000 B.T.U 42 70 15,000 B.T.U 45 75 Standardization of Calorimeter. The accuracy of all calorN metric determinations depends upon the correct determination of the water equivalent value of the apparatus. Standardize the apparatus by means of naphthalene and benzoic acid whose heats of combustion have been determined by the U. S. Bureau of - Standards. The pure substances should show the following heat- ing values: Per Gram Naphthalene 9622 calories* Benzoic acid 6329 calories * Calories per gram XI. 8 = B.T.U. per pound. ANALYSIS OF FUELS 179 The determination is made under exactly the same conditions as in the determination on coal. Instead of using 1 gram of naphthalene or benzoic acid, however, as in the case of coal, the following weights are taken : Naphthalene 0.8450-0.8550 gram Benzoic acid 1.2850-1.2950 grams Make a pill of the naphthalene of the above weight and imme- diately weigh it accurately and place in the bomb. Also make the benzoic acid into pills, but place these pills in a desiccator over cone. H2SO4 for about eight hours. Then take the required weight and place immediately in the bomb. CALCULATION. Let a = Wt. of material multiplied by its heat- ing value in calories ; b = Heating value of iron wire in calories;* c= Heating value of titer in calories;* d = Corrected rise in temperature, C.; e = Wt. of water in grams ; and g = Water equivalent of apparatus in grams ; a+6+c then g = e. d The average of the naphthalene and benzoic acid results should agree within approximately 2 grams. Take for the water equiva- lent in grams of the apparatus the average of the results by naph- thalene and by benzoic acid. The final water equivalent factor is the water equivalent in grams plus the weight of water in grams, multiplied by 1.8. This factor is to be used for converting the corrected temperature rise of the coal to B.T.U. per pound. NOTE. " H "Value: If it is assumed that the calorific value of coal is due to combustion of organic matter and sulfur, it would seem probable that in coals of like character the calorific value would be proportional to the amount of these substances present. Therefore, if the sum of these percentages of moisture, ash and sulfur be subtracted from 100, the remainder would be approximately the organic matter in the coal ; and if the calorific value of the sulfur be subtracted from the calorific value of the coal as determined, the remainder should be the calorific value of the organic matter present. The calorific value of coal cal- culated on a moisture-ash-sulfur-free basis is commonly designated as " H." * B.T.U. -i- 1.8 = calories. 180 TECHNICAL METHODS OF ANALYSIS This " H " value differs for different grades of coal but for the same kind of coal from the same seam is fairly constant. The value of " H," being known for various grades of coals, serves as an approximate check on the analysis. Calculation of " H." Let A =B.T.U. as determined; B = per cent of sulfur expressed as decimal; C = per cent of moisture expressed as decimal; and D = per cent of ash expressed as decimal; A-4Q5QB then "H" = . 1-(B+C+D) REFERENCE. The procedures in this method are essentially those of the American Society for Testing Materials described in its Book of Standards. COAL Ultimate Analysis Apparatus and Chemicals. (A) The combustion is made in a 25-burner combustion furnace of the Glaser type. (B) The purifying trains through which the air and oxygen are passed before they enter the combustion tube are arranged in duplicate, one part for air, the other for oxygen, both being connected to the combustion tube by means of a Y-tube. The purifying reagents, arranged in the order of the flow of oxygen or air through them, are: (1) cone. H 2 S0 4 , (2) 30% KOH solution, (3) soda-lime, and (4) granular CaC^. The oxygen and air are allowed to bubble through about 0.25 inch of the reagents (1) and (2), which are in gas-washing bottles. Reagents (3) and (4) are in U-tubes. (C) The combustion tube is about 40 inches long and about f inch (16 mm.) internal diameter, made of hard glass or silica. The tube extends beyond each end of the furnace for a distance of about 4 inches, the ends of the tube being protected from the heat of the furnace by closely fitting circular shields of asbes- tos. The rear end of the tube (the end next to the purifying train) is closed with a rubber stopper. As this end of the tube is kept cool by the protection of the circular shield and by the pas- sage of cool air and oxygen, there is very little danger of volatile products being given off by the rubber. The other end of the tube is closed by a well-rolled cork of -specially selected quality, the danger from overheating at this end of the tube being too great to permit of the use of the more convenient rubber stopper. ANALYSIS OF FUELS 181 The tube is filled as follows: A space of 5-5.5 inches is left empty at the end nearest the absorbing train. Then follow: (1) a plug of asbestos; (2) 4-5 inches of fused PbCrC>4 in small lumps; (3) an asbestos plug; (4) 14-16 inches of pure, recently ignited "wire" CuO (or a close coil of fine copper gauze thor- oughly oxidized by heating it in a stream of pure oxygen) ; (5) an asbestos plug; (6) the boat for holding the coal. There must be 12-14 inches of empty tube following the last asbestos plug, so that the part of the tube in which the boat is placed will be well in the furnace, and yet the tube itself project at least 4 inches out- side of the furnace. (D) The absorption train is as follows : The water is absorbed in a 6-inch U-tube, filled with granular CaC^ ; the C02 is absorbed by KOH (30% solution *) in an ordinary Liebig bulb, to which is attached a 3-inch U-tube containing soda-lime and granular CaC^, the bulb and U-tube being weighed up together. This is followed by a final guard tube filled with CaCb and soda-lime to prevent any back-pressure of CC>2 or H^O. The gases formed during combustion are drawn through the train by suction, a Marriott bottle being used to secure a constant suction head. (E) The oxygen used is kept over water and is supplied under small pressure. The supply of oxygen and the aspiration during a combustion are so regulated as to keep the difference in pressure between the inside and outside of the tube very small, the pressure inward being slightly greater. This reduces the danger of leaks to a minimum, and, if by chance any slight leakage does occur, it is inward rather than outward and the effect upon the deter- mination is small. Testing the Apparatus. Before beginning the determination test the apparatus for leaks by starting the aspirator at the rate of 3 bubbles of air per second through the KOH bulb and then shut- ting off the supply of air. If not more than 1 bubble of air per minute passes through the KOH bulb, the connections are suf- ficiently tight to proceed with the determination. Air is then admitted to the purifying apparatus, the tube heated to redness throughout and 1000 cc. or more of air aspirated. The KOH bulb and drying tube are then detached and weighed. They are again * To this should be added a little KMnO 4 to oxidize any possible oxidizable impurities. 182 TECHNICAL METHODS OF ANALYSIS connected up and 500 cc. of oxygen, followed by 1000 cc. of air, are aspirated through the train. On commencing the second aspiration, the burners under the rear portion of the tube are gradually turned down and finally entirely out, so that the empty portion of the tube into which the sample for analysis is to be inserted becomes nearly or quite cool by the time the aspiration is complete. The burners under the two-thirds of the CuO next to the PbCr04 are kept lighted and this portion of the CuO kept at a red heat. After aspiration of the 1000 cc. of air, the KOH bulb and drying tube are detached and again re weighed. If the gain or loss in weight is less than 0.0005 gram, the apparatus is ready for an analysis. Carbon and Hydrogen. Ignite and cool the boat. Weigh it empty in a glass-stoppered weighing bottle. Place in it about 0.2 gram of the finely pulverized and well-mixed air-dry coal. (The sample must be ground very fine, otherwise, in weighing so small a quantity, average results will not be obtained.) Place the boat in the bottle, quickly stopper and weigh accurately. Insert the boat quickly into its proper place in the combustion tube and connect up the apparatus. The tube should be cool after -the first 12 inches, the CuO should be red hot and the PbCrO4 at a dull red heat.* The boat should be transferred from the weighing bottle to the combustion tube as rapidly as possible to avoid change in moisture content and should be placed near the asbestos plug at the beginning of the CuO section. After connecting up the apparatus, start the aspiration with pure oxygen at the rate of 3 bubbles per second. Turn on one burner about 4 inches back from the boat and continue aspiration carefully until practically all moisture is expelled from the sample. Then increase the heat very gradually until all the volatile matter has been driven off. In doing this, the heat must be applied grad- ually in order to prevent a too rapid evolution of gas and tar which might either escape complete combustion or be driven back into the purifying train. The heat should be slowly increased by turning on more burners under the open part of the tube until the sample is ignited; then increase the temperature rapidly, * When silica tubes are used, the PbCrO 4 section should be kept slightly below even a dull red on account of danger of fusing the tube. ANALYSIS OF FUELS 183 taking care, however, not to melt the combustion tube. Any moisture collecting in the end of the combustion tube or in the rubber connection joining it to the CaCb tube is driven over into the CaCl2 tube by carefully warming with a piece of hot tile. The aspiration with oxygen is continued for two minutes after the sample ceases to glow, the heat is then turned off and about 1200 cc. of air aspirated through the tube. Finally, the absorption bulbs are disconnected, wiped with a clean cloth and allowed to cool in the balance room until they come to room temperature and then weighed. Calculate the per cent of carbon and of hydrogen by the following formulas: ' (increase in weight of KOH bulb) Weight of sample (increase in weight of CaCl2 tube) 70 H=11.19X . . - . . Weight of sample NOTES. (1) After the boat has been removed, the apparatus is ready for another determination, since any CuO will all have been reoxidized by the air current. (2) After removing the boat, weigh the ash (unless the ash has been deter- mined on another sample) and carefully inspect it for any unburned carbon. If such is found, the determination is worthless -and must be repeated. (3) The wire CuO should be used, as the ordinary granular oxide some- times contains carbon and carbonates. It should be examined for CaCOs or other carbonates which are likely to give off CO 2 on heating, and also for lime which may absorb CO 2 . The CuO may be tested for CaO by extracting it with a little dil. HNO 3 , adding NH 4 OH in excess and then testing the liquid with (NH 4 ) 2 C 2 O 4 . It should give no precipitate. (4) The asbestos used should be boiled in HC1, washed, dried and ignited, in order to remove traces of CaCO 3 or other carbonates sometimes present. (5) The PbCrO 4 should be in moderately coarse lumps from which all fine material has been sifted out with a 20- or 30-mesh sieve. It must be neutral and free from alkaline chromaies. The same PbCrO 4 can be used for many determinations; as long as it does not turn green for more than 20% of its length in the tube, it is perfectly safe. (6) The oxygen should be tested as to its purity and must not be kept in rubber bags or passed through long rubber tubes. (7) Check determinations should agree within 0.07% for H, 0.30% for C and 0.05% for N. Nitrogen. Weigh 1 gram of the finely pulverized coal into a Kjeldahl distilling flask and determine the nitrogen by the Gunning method as described on page 65. 184 TECHNICAL METHODS OF ANALYSIS Sulfur. Determine the sulfur as described on page 174. Oxygen. As no satisfactory method is known for the direct determination of the oxygen in coal, it is always determined " by difference," the sum of the percentages of H, C, N, S, and ash being subtracted from 100% and the remainder called oxygen. The result so obtained is always inaccurate, the error increasing with the percentages of ash and sulfur. The weight of the ash does not represent that of the mineral matter in the coal, the pyrite in the coal being burned to Fe20s and the sulfur passing off as 862. Thus 4 atoms of S in 2 FeS2 are replaced by 3 atoms of O in the Fe20 3 , and the loss of weight is equal to f of the S. For this reason many chemists use f of the S, instead of the total S found by analysis, in calculating the 0. As coals contain sulfur in other forms than FeS2, however, and also frequently other com- pounds that lose weight on burning, such as FeCOa and CaCOs, it is doubtful whether the results obtained in this way are any closer to the truth. REFERENCES. Lord: "Notes on Metallurgical Analysis," pages 165-170. Bulletin No. 9, Geological Survey of Ohio, pages 317-319. U. S. Dept. Agr., Div. Chem., Bulletin No. 46, Revised, pages 14-16 (1899). American Society for Testing Materials, Triennial Standards, 1918, page 694. PHOSPHORUS IN COAL AND COKE Ignite to ash in a platinum crucible exactly 10 grams of coal or 5 grams of coke. Add 15 cc. of cone. HNOs and about 5 cc. of HF. Evaporate carefully to dry ness under a good hood. Fuse the residue with 3 grams of Na2COs (if any unburned carbon is present in the ash, 0.2 gram of NaNOs should be mixed with the Na2COa). Leach out the melt with water, filter and wash. Ignite the residue and again fuse with Na2C03. Leach out this melt, filter and wash. Acidify the combined filtrates in a flask with a very slight excess of HNOs and concentrate to a volume of 100 cc. Neu- tralize with cone. NILiOH and add approximately 5 cc. in excess. Then make slightly acid with HNOs, bring the solution to a tem- perature of 80 C., add 60 cc. of ammonium molybdate solution and shake for five minutes. Complete the determination as under Phosphorus in Steel (page 112), dissolving the yellow ANALYSIS OF FUELS 185 precipitate in standard NaOH solution and titrating the excess with standard HNOs solution. REFERENCE. This method is similar to that of the American Society for Testing Materials, Triennial Standards, 1918, page 693. COAL-ASH AND REFUSE General. The usual determinations on samples of coal-ash or refuse are the amounts of moisture, combustible matter and ash, and the calculation of the B.T.U. per pound. Moisture. Dry 1 gram of the finely powdered sample in a weighed platinum crucible for one hour at 110 C., in the special water-glycerin bath described on page 173. Ash. Ignite the residue from the above moisture determina- tion in the platinum crucible in a muffle at approximately 900 C., until complete combustion is obtained. Cool in a desiccator and weigh. Combustible Matter. Subtract from 100% the percentages of moisture and of ash as above determined. The difference is combustible matter. B.T.U. per Pound. The heating value is calculated on the arbitrary basis of 14,600 B.T.U. per pound of combustible matter. In other words, multiply the percentage of combustible matter, expressed as a decimal, by 14,600. NOTE. It is often customary to report results on the dry basis. In such case divide the results obtained for combustible, ash, and B.T.U., respectively, on the " as received " basis by the difference between 1.0000 and the moisture percentage expressed as a decimal. In the case of very wet samples, requiring preliminary air-drying, proceed as under Coal, page 172. GASOLINE General. The essential desirable properties in gasoline for motor use are: 1. It should not contain too large a percentage of highly volatile products which tend to cause large evaporation losses and excessive danger in handling and storage, but should have sufficient volatile constituents to permit starting an engine under reasonably unfavorable conditions without pre-heating. 2. It should not contain any considerable percentage of 186 TECHNICAL METHODS OF ANALYSIS heavy or non-volatile constituents which, after atomization into the engine cylinders, cannot be completely vaporized and burned. 3. It should not contain any material which, after combustion, leaves a residue which collects in the motor. 4. It should be free from substances which attack metal, either before or after combustion. This includes unremoved acid used in refining. 5. Neither the gasoline nor its products of combustion should have a strong or markedly disagreeable odor. 6. It should be free from non-combustible material such as water, sediment, etc. Types of Gasoline. 1. " Straight " Refinery Gasoline. In general these are made by distilling crude oil in a fire still and taking a cut when the gravity of the product reaches some pre- determined mark. So-called crude naphtha or benzine is acid refined and steam distilled. Several products of different ranges of volatility may be produced, or the steam distillation may simply separate the product from the less volatile bottoms which go into the burning oil stock. " Straight " refinery gasolines are generally characterized by a low content of unsaturated and aromatic hydrocarbons, and by a distillation range free from marked irregularities. 2. Blended Casing-head Gasoline. Casing-head gasoline is obtained from natural gas by compression or absorption. It is too volatile for general use " straight " and, before being mar- keted, is generally blended with sufficient heavy naphtha. The resulting mixture is characterized in general by a volatility range showing a considerable percentage of constituents of low and high boiling points, but a lack of intermediate products. Frequently, however, the blending is done in a manner difficult to detect, the natural gas gasoline being used in moderately small proportion with heavy straight-run naphtha in order to make a product hav- ing a desirable percentage of volatile constituents. As regards chemical properties, blended casing-head gasoline seems to be identical with " straight " refinery products of the same distillation range. The characteristic physical properties of blended gasoline are due wholly to the details of blending. 3. Cracked or Synthetic Gasoline. These are marketed largely, if not altogether, in the form of blends with " straight " ANALYSIS OF FUELS 187 refinery and casing-head gasoline. The cracked gasolines are similar to " straight " refinery products in most physical and chemical properties but contain varying percentages of unsat- urated and aromatic hydrocarbons. Color and Odor. Properly refined gasolines are water-white and the color of a sample as seen in a 4-ounce sample cylinder should be noted. The sample also should be free from rank and disagreeable odors. Water and Foreign Matter. The sample should be free from water, sediment, and other foreign matter. Water is seldom present and is easy to detect, as an appreciable amount will sep- arate into a lower layer. Acidity. Shake 10 cc. of the gasoline with 5 cc. of distilled water and test the water with blue litmus paper, which should not turn red. The amount of acid may be determined quantitatively by taking a larger known amount of gasoline and titrating the water with 0.1 N or 0.01 N NaOH and phenolphthalein. Specific Gravity. In view of the various sources of gasoline now on the market, the sp. gr. is no longer of any particular value as an index of the quality. It may be determined with a pyc- nometer, Westphal balance, or hydrometer. Distillation. Conduct the distillation of 100 cc. of the sample in an Engler distilling flask (in case it is desired to deter- mine the sp. gr. of the different fractions, 200-500 cc. should be employed for distillation) . (A) APPARATUS. The apparatus for the distillation is as follows: (Fig. 10.) (I) Flask. The flask used is the standard 100 cc. Engler flask, described in the various textbooks on petroleum. The dimensions are as follows: Dimensions Cm. Inches Diameter of bulb 6.5 2 . 56 Diameter of neck 1.6 . 63 Length of neck 15.0 5.91 Length of vapor tube 10.0 3.94 Diameter of vapor tube 0.6 . 24 Position of vapor tube, 9 cm. (3.55 inches) above the surface of the liquid when the flask contains its charge of 100 cc. The tube is approximately in the middle of the neck. 188 TECHNICAL METHODS OF ANALYSIS The flask is supported on a ring of asbestos having a circular opening 1.25 inches in diameter; this means that only this limited portion of the flask is to be heated. The use of a sand bath is not approved. 1 I Bureau of Mines, Technical Paper 166 FIG. 10. Apparatus for Distillation of Gasoline (II) Condenser. The condenser consists of a thin-walled tube of metal (brass or copper) 0.5 inch internal diameter and 22 inches long. It is set at an angle of 75 from the perpendicular and surrounded with a water jacket of the trough type. The lower end of the condenser is cut off at an acute angle and curved down for a length of 3 inches. The condenser jacket is 15 inches long. NOTE. For ordinary purposes in comparing different samples an ordinary Liebig condenser with an inner tube at least 22 inches long may be used. The water running through it should be as cold as possible and the graduated cylinder collecting the sample should preferably be surrounded by ice water. ANALYSIS OF FUELS 189 (III) Thermometer. The accuracy of the distillation primarily depends upon the accuracy of the thermometer. It should be an accurate nitrogen-filled instrument with a short bulb (10-15 mm. long). The diameter of the thermometer should be between 5.5 and 7 mm. and the diameter of the bulb less than that of the ther- mometer tube. The total length should be approximately 380 mm. and the range from 0-270 C. ; with the length of the grad- uated portion between the limits of 210-250 mm. It should be scaled for total immersion with an accuracy of 0.5 C. The above requirements insure that nearly always the lowest temperatures registered will come above the cork of the distillation flask and variations because of the stem correction will always be prac- tically the same. The stem correction should not be applied, but it should be understood that results of distillations are expressed in terms of thermometer readings, not of actual temperatures. METHOD OF DISTILLATION. The flask connected with the con- denser is filled with a 100 cc. charge of gasoline, measured from a 100 cc. graduated cylinder. The same cylinder may be used, without drying, as the receiving vessel for the distillate. Apply heat to the flask in regulated degree, taking care that the whole distillation from beginning to end proceeds at a rate of not less than 4 nor more than 5 cc. a minute. Take the readings of the thermometer when the first drop falls from the end of the con- denser, also when 1 cc. has distilled over, and then continue to take readings as every 5 cc. comes over, beginning at 5 cc. and running up to 95 cc. Record also the dry point or highest tem- perature reading obtainable at the end of the distillation. Determine the distillation loss by adding the per cent of residue in the distilling flask, after cooling, to the per cent of total distil- lates held in the receiver. If the distillation loss is over 3%, make a check distillation, as excessive loss may indicate that the rate of distillation at the beginning was too rapid. In case the magni- tude of the loss is confirmed this fact is of importance in indicating that the gasoline contains very volatile constituents, particularly those derived from added casing-head gasoline. Have the condenser trough filled with a mixture of cracked ice and water (not dry cracked ice), and during the distillation keep sufficient ice in the trough to prevent the temperature of the cooling water exceeding 8 C. (46 F.) 190 TECHNICAL METHODS OF ANALYSIS If distillations are made at high altitudes or when barometric pressures are low, allowances may be made for this factor. In general, recording the barometric pressure read at the time of the distillation will suffice, and it is recommended that whenever there is possibility of dispute over the results of a distillation this should be done. The thermometer bulb should be covered with a thin film of absorbent cotton; this keeps the glass always wet with the con- densate from the vapor and thus prevents possible fluctuations in the temperature. It also tends to prevent superheating of the bulb at the end of the distillation and thus makes possible an accurate determination of the dry point. NOTE. The use of apparatus at least approximately as described is essen- tial, although the method is such that no considerable discrepancies will result if the apparatus is not exactly standard. The chief source of difficulty is the rate of heating and the speed should come within the above limits. For the most accurate work it is advisable to have the 100 cc. sample measured out at the same temperature as the condenser bath, that is 33 F. REFERENCES. Dept. of the Interior, U. S. Bureau of Mines, Technical Paper 166, Petroleum Technology 39, May, 1917. HEATING VALUE AND SULFUR CONTENT OF LIQUID FUELS General. This method is to be employed for determining the B.T.U. and sulfur content of such liquids as oils, coal tar products, gasoline, alcohol, etc. Heating Value British Thermal Units. Calorimetric deter- minations are made in an Emerson bomb calorimeter with a gold lining. Weigh the liquid directly in a dried, weighed, No. 00 gelatin capsule (Parke, Davis & Co.). In order to avoid incomplete combustion from scattering of material with explosive violence, it is absolutely requisite that the capsule be very carefully filled so that no air bubbles are enclosed. To insure this the liquid is well shaken and a little poured into a small open dish. Each half of the previously weighed capsule is filled, and while still immersed the halves of the capsule are fitted tightly together. Then remove the capsule, dry it thoroughly, weigh immediately and place in the fuel tray, which is lined with recently ignited asbestos. Coil the iron wire around the capsule. Place about ANALYSIS OF FUELS 191 10 cc. of distilled water in the bottom of the bomb. Screw the lid tightly down on the lead gasket and force oxygen slowly into the bomb. When testing liquids of the nature of alcohol, employ a pressure of 18-20 atmospheres; for those similar to gasoline, use a pressure of 30 atmospheres; and for materials like fuel oil or lubricating oil use a pressure of 40 atmospheres. The bomb is placed in the bucket which contains exactly 1795 grams of water at a temperature of approximately 5-5.5 C. lower than the room temperature. The actual determination is completed exactly as for Coal (page 176). CALCULATION. The result is calculated as described on page 177. The result thus obtained is further corrected as follows: Deduct from the total B.T.U. generated the weight of the capsule multiplied by its heating value in B.T.U. This result, divided by the weight of the liquid taken, gives the B.T.U. per pound of the sample. NOTES. (1) To calculate B.T.U. per gallon determine the sp. gr. of the liquid at 15.5 C. and multiply this by 8 * and then by the B.T.U. per pound. (2) In case it is found that the calorimetric value of the liquid cannot be determined with the liquid alone, recently ignited asbestos can be used as an absorbent. This method is resorted to when the liquid explodes violently and the determination is incomplete, as is shown by carbonaceous residues in the bomb. Ignite the asbestos in a hot muffle for 4 hours to insure the removal of any combustible matter. Weigh the capsule alone, fill with the ignited asbestos, and weigh again. Then fill the capsule with the liquid by means of a small pipette and weigh again to obtain the weight of the liquid taken. (3) Heating Value of Capsules. The capsules after being dried should be kept in a rubber-stoppered bottle. The calorific value of the dried capsules per gram is obtained by weighing five capsules and burning them together in the bomb. The average result of check determinations thus run is taken as the heating value per gram of the capsules. The heating value of the capsule used expressed in B.T.U. is deducted from the total number of B.T.U. de- veloped in the bomb. Sulfur. The method of determining the sulfur content of liquids from the bomb washings after combustion in the bomb is accurate, practicable and rapid, and is recommended in preference to all other methods. The pressures given in the section pertaining to the B.T.U. determinations are higher than is required for the calorific deter- * One gallon pure H 2 O at 15.5 C. weighs 8.3335 Ibs. 192 TECHNICAL METHODS OF ANALYSIS mination alone but by using these amounts it has been found that the carbon and sulfur are completely burned, as no trace of CO or S(>2 were evident when tested for. After the regular calorimetric determination has been made, as outlined above, cool down the bomb to tap water temperature. Then open and wash the contents into a beaker and titrate. Add 5 cc. of a saturated solution of Na2CC>3, heat to boiling for ten min- utes and then filter and wash 6 times with hot distilled water. To the filtrate add 10 cc. of saturated bromine water, make slightly acid with HC1 and boil to expel liberated Br and CO2. Add slowly from a pipette 10 cc. of a 10% solution of BaCl2 -2H20. Continue the boiling for fifteen minutes and let stand overnight just below the boiling point. Complete the sulfur determination as described under Sulfur in Coal, page 175. Run a blank sulfur determination on one of the capsules. n T> j. a (weight of BaSC>4 blank) 1 _ CALCULATION. Per cent S = * - . , x , -, X 13.73. weight ot sample REFERENCES. Bulletin 43, Bureau of Mines: " Comparative Fuel Values of Gasoline and Denatured Alcohol in Internal Combustion Engines." Strong and Stone, 1912. Technical Paper 26, Bureau of Mines: "Methods of Determining the Sulfur Content of Fuels, Especially Petroleum Products." Allen and Robert- son, 1912. CHAPTER VI ANALYSIS OF PAINTS AND PAINT MATERIALS H Dia. -x Dia. TURPENTINE General. The nature of a turpentine and whether or not it is adulterated can best be determined by careful distillation and examination of the various fractions as to sp. gr., index of refraction and boiling point. Procedure. - - Weigh 500 grams of the turpentine into a round-bottom flask of about 1 liter capacity. Connect the flask by means of a tightly fitting stopper to a Hempel column of the exact dimen- sions shown in Fig. 11. In the top of the Hempel. column place a thermometer with the bulb reaching to within 1 inch of the glass beads, and connect the side arm to a condenser. Drop into the flask a small flat coil of copper or nickel wire to prevent bumping and place the flask and contents on a sand bath. If the room temperature is very low, it FlG . n._ H empel Column, may be necessary to place a shield of asbestos board around the column. Distill the turpen- tine at a maximum rate of 2 drops per second. It is important that this rate shall not be exceeded. Collect the distillate in 193 3E. All Diameter Measurements are Inside Dimeusions 194 TECHNICAL METHODS OF ANALYSIS weighed flasks or cylinders of about 100 cc. capacity. While the temperature of distillation is changing rapidly the fractions col- lected should be small, varying from 4-5%, and when the tem- perature is slow and regular they may be increased to 10-12%. Usually it is unnecessary to continue the distillation after a tem- perature of 180 C. (corrected) has been reached, unless there is a large amount of residue at this stage. As soon as a fraction has been collected, the flask or cylinder in which it is contained should be immediately stoppered. Take the weight of each fraction and record the temperature at which the first drop of the fraction distilled. Also determine the sp. gr. at 15 C. and index of refraction at 15 C. of each fraction. BOILING POINT. Correct the boiling temperature as actually read on the thermometer: (1) For the prevailing barometric pressure, by adding 0.056 for every mm. which the barometer reads below 760 mm. and subtracting a corresponding amount for every mm. above 760 mm. (2) For the emergent stem of the thermometer, according to the following formula: B. P. = T+0.000143 (T-0 N, where B. P. is the corrected boiling point sought, T is the observed temperature, t is the mean temperature of the thermometer stem above the cork (measured by fastening by means of rubber bands to the thermometer in the Hempel column another ther- mometer with its bulb at the middle of the exposed thread of mercury), and N is the length, expressed in degrees, of the mercury column above the cork. SPECIFIC GRAVITY. Determine the sp. gr. by means of a West- phal balance, making the determinations at room temperature and correcting to 15 C. by using the factor 0.00083 for every degree C. of difference from this standard temperature. This correction is to be . added for temperatures above 15 C. and to be subtracted for temperatures below 15 C. For small frac- tions the sp. gr. may be taken with the Westphal balance by placing the liquid in a test-tube supported in a flat cork. Since the cor- rection for temperature is large, take special care to make the temperature readings accurate when the gravity is determined. ANALYSIS OF PAINTS AND PAINT MATERIALS 195 INDEX OF REFRACTION. Determine the index of refraction by means of the Abbe refractometer, taking the readings at room temperature and correcting by means of the factor 0.00047 for every degree of difference from the standard temperature of 15 C., adding the correction for temperatures above 15 C. and sub- tracting it for temperatures below. Recording Results. Plot the three curves for corrected boiling point, specific gravity and refractive index, respectively, on the same sheet of paper, plotting the percentage (by weight) of each fraction vertically and the other factor horizontally. ).56 0.88 0.90 0.92 0.94 150 160 170 180 190 1.464 1.468 1.472 1476 1,480 Specific Gravity Boiling Point Index of Refraction 0.86 0.88 0.90 0.92 0.91 150 160 170 c 180 190 o 1164X4681.4721476 1.480 a FIG. 12. Typical Curves Showing Specific Gravities. Boiling Points and Indices of Refraction of: A. Boiling Gum Turpentine. B. Wood Tupentine. Interpretation of Results. The typical curves of pure gum and wood turpentines are shown in Fig. 12. An unadulterated sample will show curves which may be slightly misplaced to the right or left but will be parallel to these curves. In general, for a pure gum turpentine the first fraction of the distillation should give figures within the following limits : Refractive index at 15 C 1.4700-1.4725 Sp. gr. at 15 C 0.864-0.866 Corrected boiling point 156-157.5 C. 196 TECHNICAL METHODS OF ANALYSIS Pure gum turpentines of which the first fraction falls outside these limits are rare, but if the rest of the curve is normal such a turpentine probably is not adulterated. REFERENCE. U. S. Dept. of Agriculture, Forest Service Bulletin 105. TURPENTINE ELECTRIC RAILWAY SPECIFICATIONS General. The material desired is pure gum or refined steam- distilled turpentine, free from adulteration. Requirements. Turpentine must meet the following require- ments : 1. Appearance. Clear and practically water-white. 2. Specific Gravity. Sp. gr. at 15.5 C. not less than 0.862 nor more than 0.872. 3. Distillation. When 200 cc. are distilled as below described, 95% should pass over below 170 C. Conduct the distillation in a 300 cc. flask 8 cm. in diameter with the side tube 8 cm. from the main bulb and the neck extend- ing 8 cm. above the side tube. The diameter of the neck is 2 cm. ; of the side tube, 5 mm. Fit into the stopper in the neck a thermometer (reading from 145-200 C.) with the bulb opposite the side tube of the flask and the 175 C. mark below the stopper. Conduct the distillation so that about 2 drops' of distillate come over per second. 4. Residue on Evaporation. When 1.0 cc. of the sample are placed in a glass crystallizing dish 2.5 inches diameter and 1J inches high and evaporated on the open steam bath with a full head of steam for three hours, the amount of residue shall not weigh over 0.15 gram. One drop allowed to fall on a clean white paper must completely evaporate at room temperature (20 C.) without leaving any stain. 5. Polymerization Test. When 5 cc. of the sample are treated with cone. H^SO* according to the following method, not over 0.50 cc. shall remain undissolved at the end of thirty minutes. The unpolymerized residue shall be viscous in nature and show a refrac- tive index between 1.50 and 1.52 at 15.5 C. METHOD. Add slowly 5 cc. of turpentine to 25 cc. of cone. H2S04 in an ordinary graduated narrow neck Babcock milk flask ANALYSIS OF PAINTS AND PAINT MATERIALS 197 (the smallest divisions on the neck of these flasks are 0.04 cc.); shake the flask with a rotary motion to insure gradual mixing, keeping cool, if necessary, in ice water and not allowing the tem- perature to rise above 60-65 C. .Agitate thoroughly and main- tain at about 65 C. with frequent agitation for one hour. Cool and fill the flask with cone. H^SCU, bringing the unpolymerized residue into the graduated neck. Let stand one-half hour and read off the volume of unpolymerized residue. Note its consistency and color and determine its refractive index at 15.5 C. NOTE. If the residue is water- white and limpid and does not show the proper refractive index, repolymerize with 38 N H 2 SO 4 (100.92% H^CX by weight, prepared by mixing cone. H 2 SO 4 with sufficient fuming H 2 SO 4 to give this strength) as described by Veitch in U. S. Dept. of Agriculture, Bureau of Chemistry, Bulletin 135, page 30 (or Circular 85). LINSEED OIL General. Raw linseed oil is the refined oil obtained from flax- seed or linseed, generally by hydraulic pressing. For use in paint, etc., to obtain quicker drying qualities, raw oil may be heated with very small amounts of certain " driers " such as oxides of Pb, Mn, Co, etc. Oil thus prepared is called boiled linseed oil. Oils from different countries vary somewhat in composition. Pure linseed oils from North America should show about the fol- lowing " constants " : Raw Boiled Max. Min. Max. Min. Sp. gr. at 15.5 C Refractive index at 25 C .... 0.938 1.4805 0.932 1.4790 0.945 0.935 Saponification number 195 189 195 185 Iodine number . ... 180 160 Unsaponifiable matter 1 50% 1 50% Acid number 6.0 10.0 Drying test 75 hrs. 24 hrs. A low iodine number accompanied by a high sp. gr. may indi- cate polymerization due to old age or excessive heating. High- grade oils should also be clear and free from " foots " or sediment. 198 TECHNICAL METHODS OF ANALYSIS Preparation of Sample. See page 230. Specific Gravity at 15.5 C. See page 230. Refractive Index at 25 C. See page 231. Saponification Number. S.ee page 241. Iodine Number. See page 241. Unsaponifiable Matter. Use Boemer's Method (page 262). Acid Number. See page 242. Drying Test. Flow some of the sample over a piece of clean glass and let drain in a vertical position at about 20 C., testing at intervals for " tackiness " with the finger. Note how long it requires for the film to become dry. TUNG OIL General. This oil is also known under various other names such as " China Wood Oil," " Japanese Wood Oil," "Wood Oil," or " Nut Oil." It has a strong characteristic odor. Most of it comes from China; a small amount comes from Japan. Cold drawn oil is pale yellow and called " White Tung Oil"; hot pressed oil is dark brown and called " Black Tung Oil." Very little of the latter comes to this country. The constants of pure Chinese oil are as follows: Sp. gr. at 15.5 C 0.939-0.943 Saponification number 190-200 Iodine number 150-166 Refractive index at 25 C 1.515-1.520 Heating test (Browne method) max. 12 minutes Iodine jelly test max. 4 minutes Japanese oil has a lower sp. gr. (about 0.933-0.935 at 15.5 C.) For further properties, see Lewkowitsch: " Chemical Tech- nology and Analysis of Oils, Fats and Waxes," Vol. II. Specific Gravity at 15.5 C. See page 230. Saponification Number. See page 241. Iodine Number. See page 241. Free Fatty Acids. See page 242. Refractive Index at 25 C. See page 231. Unsaponifiable Matter. See page 262. ANALYSIS OF PAINTS AND PAINT MATERIALS 199 Heating Test (Browne Method). Place in a 160X15 mm. test tube 5 cc. of water and make a mark at the 5 cc. point. Then dry the test tube and insert a cork perforated so that a glass rod of 3 mm. diameter can move freely. Fill a copper dish 12 cm. high by 6 cm. inside diameter with cottonseed oil to a height of 7.5 cm. Insert a thermometer so that the bulb is 1.5 cm. from the bottom of the dish. The thermometer should be nitrogen-filled, total immersion, length 4-^.5 inches, graduated from 210-310 C. in 2 intervals; length between 210 and 310 C. not less than 2.5 inches. If pre- ferred, a longer thermometer (30 cm. with graduations from 100- 400 C.) may be used, in which case make corrections for the emergent stem. (See U. S. Bureau of Standards, Stem Correction Sheet No. 44.) Heat the bath slowly to 293 C.; then place in it the tube containing 5 cc. of the oil sample, so that the bottom is level with the lowest part of the thermometer bulb. Note the time and remove the heat for about forty-five seconds; then re-apply the heat. Before two minutes have elapsed, the temperature of the bath will have fallen to 282 C. Hold at this point as steadily as pos- sible. When the sample has been in the bath about nine minutes, raise the glass rod at intervals of one-half minute and when the rod is firmly set, note the time. Remove the flask at once; heat the bath again to 293 C. and repeat the experiments with another portion of the sample. No stirrer is necessary in the bath, but it should be protected with a screen. When the cottonseed oil bath becomes tarry and viscid, it should be renewed. Pure tung oil should " set " within twelve minutes. Iodine Jelly Test. Weigh accurately 2.500 grams of sample into a wide-necked 200 cc. Erlenmeyer flask; add 10 cc. of CHCla from a pipette and stopper the flask immediately. Carefully insert into the flask a small glass vial so that it stands upright, and into this vial pipette 10 cc. of a solution of iodine in CHCls con- taining 0.035-0.036 gram of iodine per cc. Place the flask in a bath containing water at 25-26 C. and let stand a few minutes, keeping the flask stoppered except when necessary to remove it. Tilt the flask and rotate so that the vial is upset and the con- tents are thoroughly mixed, starting a stop watch at the same 200 TECHNICAL METHODS OF ANALYSIS time. Keep the flask in the bath at 25-26 C., and every fifteen seconds tilt the flask toward a horizontal position. Note the time required for formation of a jelly that does not flow but sticks to the bottom of the flask or slides in a mass. Record the time to the nearest quarter minute. Pure tung oil should require 2.75-3.25 minutes to jell. If the temperature of the laboratory varies more than 2 or 3 C. from 25 C., place the flask containing the iodine solution in the bath and let it remain for several minutes before pipetting out the 10 cc. for test. The CHC1 3 used to dissolve the oil and to prepare the iodine solution should conform to U. S. P. requirements and have a sp. gr. of 1.480-1.481 at 25 C. If the sp. gr. is too low, wash the CHC1 3 with water; if too high, add a little 95% grain alcohol. The iodine solution is prepared as follows: Treat an excess of iodine with warm CHCls; shake for a few minutes; cool to about 20 C. and filter through glass wool. Pipette 10 cc. of the solution into an Erlenmeyer flask containing 10 cc. of 10% KI solution and titrate with 0.1 N thiosulfate. Calculate the iodine content and dilute with CHCls so as to obtain an iodine content of 0.035-0.036 gram per cc. After dilution, titrate again to make sure the solution is of proper strength. The above method is empirical and details must be followed exactly. REFERENCES. Lewkowitsch :, "Chemical Technology and Analysis of Oils, Fats and Waxes," Vol. II; American Society for Testing Materials, Standard, Serial D-12-16. MIXED PAINTS AND PIGMENTS IN OIL General. Paints from the chemical point of view may be con- sidered as consisting of two parts, (1) the vehicle, and (2) the pig- ment. The vehicle may consist entirely of linseed oil, which is usually the case with individual paste pigments. Frequently, however, especially in the case of mixed paints, it also contains a certain amount of driers and of thinner (either turpentine or mineral spirits) ; and in the case of enamels and high gloss paints, the vehicle may also contain varnish or varnish gums. ANALYSIS OF PAINTS AND PAINT MATERIALS 201 The pigment may be a simple chemical compound, such as ZnO, or a mixture of several substances. Zinc oxide, basic car- bonate white lead, and basic sulfate white lead, are among the most important ingredients of paint pigments. Red lead and graphite are largely used as protective coatings for iron and steel. Numerous other substances are also used as paint pigments, many of which are by no means simple chemical compounds, such as mineral silicates, yellow ochre, umber, lithopone, etc. The most important paint pigments are the following : WHITE PIGMENTS. 1. Basic Carbonate White Lead is approx- imately 2PbC0 3 -Pb(OH) 2 . 2. Basic Sulfate White Lead is largely basic lead sulfate, often containing a little ZnO. 3. Zinc Oxide (Zinc White), ZnO. 4. Lithopone is made by simultaneous precipitation of ZnS and BaS04 and generally contains about 70% of the latter. 5. Barytes or Blanc Fixe is BaS04. 6. Silica (Silex). This may be ground silica or diatomaceous (infusorial) earth. 7. Asbestine is essentially a silicate of Mg and is made by grinding waste asbestos. 8. Clay ( Kaolin) is a hydrated aluminum silicate. 9. Gypsum (Terra Alba) is CaSO 4 -2H 2 0. Anhydrous CaSO 4 , or burnt gypsum, is used as an extender in Venetian Red. 10. Chalk, Paris White, Whiting, Alba Whiting, etc. These are all CaCOa but differ somewhat in their physical conditions. BLACK PIGMENTS. 1. Lampblack is a grayish black, bulky pigment. It is the soot produced by burning oils, resins, etc. 2. Gas Black (Carbon Black) is made from natural gas. It is much blacker than lampblack. 3. Bone Black is made by carbonizing bones. It contains 10-20% carbon, the remainder being largely calcium phosphate. 4. Ivory Black (Dro^ Black) is a high-grade bone black made from ivory waste. NOTE. The color of bone black and ivory black may be modified by the addition of Prussian blue. 5. Charcoal Black is produced from vegetable charcoal. 6. Graphite (Plumbago) is a form of carbon occurring as a 202 TECHNICAL METHODS OF ANALYSIS natural mineral. That used in paints may run up to 85% carbon, the remainder being silicious matter. RED PIGMENTS 1. Indian Red is essentially Fe20s dark purplish red. 2. Tuscan Red is Indian red enriched by an alizarin lake, giving a crimson shade. It is often toned down with BaSCU, CaCOs, or gypsum. 3. Venetian Red is also largely Fe20s, but is brick red in color and contains more or less gypsum (or, in inferior grades, CaCOs or BaS0 4 ). 4. Red Lead is a brilliant scarlet but is used as a protective coating and not as a tinting pigment. When c. P. it is PbsOi but the commercial product contains 70-99% PbsC^ (usually over 85%), the remainder being PbO incidental to manufacture, unless the material is intentionally adulterated. 5. Orange Mineral is a form of red lead with a lower sp. gr. and lighter color than the usual form. 6. English Vermilion is HgS. It is not much used now because of its high price. 7. American Vermilion (Chrome Red, Scarlet Lead Chr ornate) is a basic chromate of lead. 8. Lakes are formed by combining the coloring matter of cer- tain dyes with inorganic carriers, such as BaSCX, CaCOs or clay. They are generally used in paints with a large amount of other pigment for brightening or modifying the color. Among the most important lakes are the vermilions and scarlets made from para- red, and from alizarin. The dye (color) may be as little as 5% in the lake itself. YELLOW PIGMENTS. 1. Chrome Yellows are chromates of lead of varying color and composition. (a) Chrome Yellow Light (" Chrome Yellow Lemon " or " Canary ") contains more or less PbSCX or other insoluble Pb compound intimately mixed with PbCrO4. (b) Chrome Yellow Medium is pure PbCrCU. (c) Chrome Yellow Orange (Chrome Orange) is a basic lead chromate which may vary in color from pale orange to nearly scarlet. 2. Ochers (Yellow Ochers) are natural .earths whose color is due to hydrated iron oxide (limonite, 2Fe(OH)3-Fe2Os) and ANALYSIS OF PAINTS AND PAINT MATERIALS 203 varies from citron yellow to almost olive. The hydrated iron oxide may vary from 10-60%, the remainder being silicious matter or clay. Golden Ocher is yellow ocher modified with PbCrO 4 . 3. Raw Siennas resemble ochers in general composition, but are brownish yellow and generally contain a little manganese. BROWN PIGMENTS. 1. Burnt Sienna is made by calcining raw sienna, which changes the color to an orange red or red brown. 2. Raw Umber is a natural earth pigment of yellow-brown color inclining towards the olive and is similar in composition to sienna but contains considerable manganese oxide. 3. Burnt Umber, made by calcining raw umber, has a rich brown color, darker than the raw but free from red. 4. Vandyke Brown (Cassel Earth, Cologne Earth) is a natural pigment of a carbonaceous nature and is distinguished by its sol- ubility in dil. alkali. BLUE PIGMENTS. 1. Prussian Blue is made by precipitating a soluble ferrocyanide with FeSCX and oxidizing the precipitate. If the precipitate were pure and completely oxidized it would, of course, have the composition Fe4[Fe(CN)6]3. The commercial product, however, varies considerably in composition. 2. Ultramarine was formerly obtained from the semi-precious mineral lapis lazuli. It is now made artificially, however, and is quite cheap. It is essentially a double silicate of Al and Na with some sulfides or sulfates. It yields EkS when treated with HC1 and loses its color even with weak acids. 3. Cobalt Blue consists of oxides of Al and Co. Genuine cobalt blue is quite expensive and it is frequently substituted by ultramarine mixed with a little ZnO. GREEN PIGMENTS. 1. Chrome Green is an intimate mixture of Prussian blue and chrome yellow and in this form is sold as " chemically pure." It is also put out as " commercial chrome green" which generally contains about 25% of the pure color and 75% of BaS(>4 or a silicate. 2. Chrome Oxide Green is C^Oa, often more or less hydrated. It is an expensive pigment. 204 TECHNICAL METHODS OF ANALYSIS ANALYSIS It is impossible to outline a general method which will be applicable to all paints. The following procedures for the separa- tion and determination of the amount of vehicle and pigment, however, are applicable to most paints and to pigments ground in oil. Total Vehicle. The vehicle may be separated from the pig- ment either by continuous extraction or by dilution with a suitable solvent and whizzing in a centrifuge. (A) BY CONTINUOUS EXTRACTION. Mix the sample very thor- oughly [see note (1) below], taking particular pains to stir up any pigment which has settled. Partly fill a small beaker with the well-mixed paint and place in it a stirring rod. Insert a plug of cotton in a Soxhlet extraction thimble (of double thickness if the paint contains ZnO or other extremely fine pigments), dry at 100 C. and weigh. Stand the thimble in another small beaker, and remove the cotton. Weigh the beaker with the paint, then fill the thimble about three-quarters full, pouring the paint carefully down the stirring rod. Again weigh the beaker with the stirring rod and determine the amount of paint which has been transferred to the thimble. Plug the thimble lightly with the cotton, place it in a Soxhlet extractor, fill the extractor nearly full of ether and let the paint and the thimble soak in it for at least one hour before starting the siphoning. Continue the extraction until the ether extract comes over colorless. Then remove the thimble, let the pigment dry, grind it to break up lumps, return to the thimble and again extract for at least four hours (or overnight). Finally evaporate the ether from the extract and dry to constant weight at 105-110 C. Also dry the thimble with the pigment at about 105 C. and weigh it. The difference between 100% and the sum of the percentages of the pigment and of the ether extract is a rough indication of the amount of volatile thinner. In case only the total pigment and total vehicle are desired, the weight of the dried pigment and not of the extract should be used in making the calculations. (B) BY CENTRIFUGING. Weigh 2 empty cylinders (4 oz. oil sample bottles are convenient), fill each about one-quarter to one- ANALYSIS OF PAINTS AND PAINT MATERIALS 205 third full of the mixed paint, and again weigh. Add an equal volume of ether, mix thoroughly, place the cylinders in a cen- trifuge opposite each other, so that they counterbalance, and whiz until the pigment settles clear. Pour off or siphon off as much as possible of the clear liquid, refill the cylinders with ether, stir up the pigment and again centrifuge. Repeat the process until the pigment is free from oil. Remove the ether with a gentle air blast and dry the pigment to constant weight at 105 C. NOTES. (1) If the sample is a large one, received in the original can, weigh the can as a whole, without shaking, remove as much of the clear vehicle as possible, and transfer it to a cork-stoppered flask or bottle. Weigh the can to obtain the amount of vehicle removed. Then mix thoroughly the contents of the can and transfer to another container. Clean out the empty can and weigh it. From these figures the analytical results can be calculated back to the original material. (2) The method of centrifuging is more rapid than the continuous extrac- tion method and in the case of very fine pigment which cannot be held by the thimbles it is preferable. (3) In many cases it is preferable to use several solvents in sequence as follows: Extract with gasoline by means of the centrifuge, again extract with gasoline, then with benzene, and finally with ether, removing each time as much as possible of the clear extract. In the case of certain enamel paints, it is advisable to follow the gasoline treatment by a treatment with turpentine and then remove the turpentine with gasoline before treatment with benzene and ether. (4) In the case of paints which settle with difficulty, better results are some- times obtained by a mixture of solvents, such as a mixture of 6 parts of ben- zene and 4 parts of wood alcohol by volume. (5) No extraction process will remove the last traces of the vehicle. The insoluble portion is probably oxidized linseed oil or metallic soaps. Pigment. Grind the dried pigment finely and make a quali- tative analysis on a portion of it before attempting a quan- titative analysis. The general procedure for a quantitative analysis of a mixed pigment is indicated on page 207, but the pro- cedure may have to be considerably varied in individual cases. (See also methods for White Lead, Chrome Yellow, and Red Lead.) Vehicle. If the amount and condition of the sample will per- mit, make the analysis on the clear vehicle poured off from the original sample. (A) THINNER. Weigh 50-100 grams of the separated vehicle (or a correspondingly greater amount of the original paint) into 206 TECHNICAL METHODS OF ANALYSIS a 500 cc. flask connected with a spray trap and a vertical con- denser and distill with steam.* Heat the flask in an oil bath nearly to 100 C. before passing in the steam and then raise the tem- perature to 130 C. Collect the distillate in 100 cc. graduated glass-stoppered cylinders which have previously been weighed. When the distillate comes over clear, or at least 300 cc. of water have been collected, stopper the cylinders and weigh them. Read the volume of water and of thinner, subtract the weight of water (1 cc. of water =1 gram) and calculate the per cent of thinner, both by volume and by weight in the vehicle, and by weight in the original paint. Instead of collecting the distillate in graduated cylinders, the entire distillate may be collected in a small weighed separatory funnel. Then let the layers separate sharply, draw off the water and weigh the funnel containing the thinner. Determine the sp. gr. of the thinner at 15.5 C. and make a polymerization test and such other tests as will establish whether it is pure turpentine or a mixture of turpentine with mineral spirits. (See pages 193 and 196.) (B) ANALYSIS OF OIL. (1) Specific Gravity. Determine the sp. gr at 15.5 C. of the original vehicle with a hydrometer or Westphal balance and correct for the amount of thinner and its sp. gr. as above determined. (2) Ash and Driers. Determine the amount of ash in a por- celain crucible and its nature. If much Pb is present, part of it is lost in igniting to ash, and a quantitative analysis of the ash is therefore not accurate. For the quantitative determination of driers, see page 221. (3) Iodine Number. Determine the iodine number of the oil freed from water after the steam distillation, bearing in mind that the constants of linseed oil which has been mixed with pigment, especially Pb compounds, may be much altered and that an iodine number even as low as 100 is not an indication of the presence of other fatty oils. (See page 241.) (4) Mineral Oil. Determine the saponification number as described on page 241, and if the oil is not completely saponifiable, determine the amount of unsaponifiable oil. (See page 261.) * If alcohol, acetone or other water-soluble solvent is present, it is neces- sary in order to determine these to run an additional distillation without steam. ANALYSIS OF PAINTS AND PAINT MATERIALS 207 Water. Some specifications place a maximum limit on the permissible amount of water. The latter may be determined on the mixed paint by the Xylol Method as described on page 271, using about 100 grams of the sample. REFERENCE. Holley and Ladd: " Mixed Paints, Color Pigments and Var- nishes"; U. S. Dept. of Agriculture, Bur. of Chem., Bulletin 109, revised; Bureau of Standards, Circular 69. GREEN GRAPHITE POLE PAINT General. This method covers the analysis of Green Graphite Pole Paint delivered under specifications requiring a paint con- taining between 40 and 45% of pigment; the pigment to consist of amorphous graphite, silica, iron oxide and alumina, and tinting color, the latter composed of Prussian blue and chrome yellow. NOTE. Chrome yellows and chrome greens are sometimes toned down with PbSO 4 and barytes (BaSO 4 ). Small amounts of these substances, therefore, would not be cause for rejection. Pigment and Vehicle. Determine the proportions of pigment and vehicle by extracting a weighed quantity of the paint with ether or with low-boiling gasoline, either in a Soxhlet ex- tractor * or with a centrifuge. Weigh the dry pigment. Total Insoluble Matter. Boil 1 gram of the dry pigment in a 250 cc. beaker with 30 cc. of cone. HC1 for about thirty minutes. Add from time to time a drop of alcohol. Then add 50 ce. of water and boil again for fifteen minutes. Filter on a filter paper which has been dried and weighed in a weighing bottle. Wash the residue thoroughly with hot water; dry in the weighing bottle, cool in a desiccator, and weigh. Carbon. Ignite the above residue in a platinum crucible until all C is burned off, cool in a desiccator and weigh. Report the loss as carbon. Barium Sulfate. Moisten the residue in the crucible with a little water and add a few drops of dil. H2SO4. Then fill the crucible two-thirds full of HF. Evaporate this off on a hot plate and repeat the operation to insure removal of all * If an extractor is used, extract until siphonings are colorless, remove the pigment, dry, grind and re-extract for at least four hours. 208 TECHNICAL METHODS OF ANALYSIS Ignite the residue gently until all 80s has been driven off. Fuse this residue (consisting of Fe 2 O3, Al 2 Os and BaSO 4 ) with a con- siderable excess of KHSCU, using a very low flame to avoid sput- tering. Digest the fusion in water containing a little HC1. If completely soluble, no BaSO 4 is present. If there is any residue, filter, wash with hot water, ignite in a platinum crucible, and weigh as BaS0 4 . NOTES. (1) The presence of BaSO 4 should be confirmed by fusing this residue with Na^COs, boiling with water and filtering. Test the filtrate for SO 4 ; dissolve the residue in HC1 and test for Ba. (2) Save the filtrate from the KHSO 4 fusion for determination of insoluble iron and alumina. Total Lead. To the filtrate from the total insoluble, add NH40H until a precipitate begins to form and then just enough HC1 to redissolve it. Dilute to about 500 cc.; saturate with H 2 S, heat to boiling, let settle, and filter off the black PbS. Wash with H 2 S water; then dissolve in cone. HNOs, containing a little Br water. Filter out any sulfur, boil off bromine, dilute to about 200 cc., and make faintly alkaline with NIIiOH. Then add a slight excess of acetic acid. To the boiling solution add an excess of K 2 Cr 2 O7 solution. Boil for two or three minutes until the pre- cipitate settles clear. Filter on a weighed Gooch crucible, wash with hot water, dry at 110 C., set the Gooch crucible in a larger platinum crucible, ignite gently, cool in a desiccator and weigh as PbCr0 4 . NOTE. Pb may be also determined as PbSO 4 as described under Total Lead in Chrome Yellow, page 214. Lead Chromate. Boil the filtrate from the PbS until it con- tains no more H 2 S, then add a few drops of HNOs and boil again. Precipitate the hydroxides of Fe, Cr, and Al with a slight excess of NH4OH. Filter and wash with hot water. Dissolve in HC1 and make up to 250 cc. Pipette 100 cc. of this solution (equiva- lent to 0.4 gram of original pigment) into a beaker, add a slight excess of NH4OH, cool and add about 1 gram of Na 2 2 , keeping the beaker covered with a watch glass. Digest on the steam bath until evolution of gas has ceased; filter and wash with hot water. Redissolve the precipitate in HC1 and reprecipitate with NH^OH; ANALYSIS OF PAINTS AND PAINT MATERIALS 209 and again add Na2C>2, keeping the volume of solution small. Filter and wash. Combine both filtrates and make slightly acid with acetic acid and boil until all peroxide has been decomposed. Continue the determination as described under Lead Chromate in Chrome Yellow, using either the gravimetric or the volumetric method (see page 215). Iron Oxide and Alumina. Insoluble Fe 2 O3 and A1 2 O3 are determined in the solution of the KHS04 fusion after filtering out any BaSO" 4 or undecomposed silica. To this filtrate add a few drops of HNOs, heat to boiling and then add a slight excess of NILtOH and boil until the odor of the latter is nearly gone. Filter, wash with hot water, ignite strongly and weigh as Fe203+Al203. Soluble Fe203 and AbOs are determined on an aliquot of the filtrate from PbS. Pipette 100 cc. (which is equivalent to 0.4 gram of original pigment) into a beaker, add a few drops of HNOs and heat to boiling; precipitate with a slight excess of NH 4 OH; filter, wash with hot water, ignite over a blast lamp and weigh. The residue consists of Fe 2 03+Al203+Cr 2 03. Subtract from this the Cr 2 O 3 equivalent to the PbCrO 4 found. Add the difference to the insoluble iron oxide and alumina and r eport as total iron oxide and alumina. CALCULATION. PbCr0 4 X 0.2352 = Cr 2 3 . Silica. Si02 is obtained by calculation as follows : The residue in the crucible after burning off carbon consists of SiO 2 +BaS04 (if present) + insoluble Fe 2 C>3 and A^Os. From its weight, therefore, subtract the amount of other substances as determined above and report the difference as silica. Prussian Blue. Weigh 2 grams of dry pigment into a Kjeldahl flask, and add 10 grams of K 2 SO 4 or 7.5 grams of anhydrous Na 2 S0 4 (free from N) and 30 cc. of cone. H2SO4. Heat over a Tirrill burner, gently at first, and then to copious fumes for six to eight hours. Cool, dilute to about 250 cc. and add a few grains of granulated Zn and about 80 cc. of cone. NaOH solution.* Distill into 50 cc. of 0.1 N HC1 until at least 100 cc. have come over. Titrate the excess of acid with 0.1 N alkali. Calculate the acid consumed to Prussian blue, Fe 4 [Fe(CN)e]3. * Sufficient NaOH solution must be added to give a strong alkaline reac- tion after distillation is completed. See under determination of Nitrogen, page 65. 210 TECHNICAL METHODS OF ANALYSIS CALCULATION. 1 cc. 0.1 N acid = 0.004773 gram Prussian blue. NOTES. (1) Any excess of Pb beyond that required for lead chromate should be calculated to lead sulfate. PbOO 4 X 0.9383 = PbSO 4 . (2) Commercial Prussian blues seldom if ever conform strictly to the formula Fe 4 [Fe(CN) 6 ]3, but it is necessary to assume some formula in making the chemical calculation and the theoretical formula is the one usually taken. RED LEAD AND ORANGE MINERAL General. Red lead and orange mineral in the pure state are oxides of lead (approximately PbsOi), being probably mixtures of compounds containing varying proportions of PbO and Pb02. Red Lead (Pb 3 O 4 ). Weigh 1 gram of the very finely ground pigment into a 150 cc. Erlenmeyer flask. Mix in a small beaker 30 grams of crystallized sodium acetate, 2.4 grams of KI, 10 cc. of water and 10 cc. of 50% acetic acid. Stir until all is liquid, pour into the Erlenmeyer flask containing the red lead, and rub with a glass rod until all of the lead is dissolved; add 30 cc. of water, and titrate with 0.1 N sodium thiosulfate, using starch as indicator. A small amount of lead may escape solution at first, but when the titration is nearly complete this may be dissolved by stirring. The reagents should be mixed in the order given, and the titration should be carried out as soon as the lead is in solution, as otherwise there is danger of loss of iodine. CALCULATION. 1 cc. 0.1 N thiosulfate = 0.03428 gram Pb 3 O 4 . For purpose of calculation it may also be assumed that 1 cc. 0.1 N thiosulfate = 0.03348 gram PbO. This is, however, really the equivalent of PbsOi in terms of PbO. Litharge. Treat 1 gram with 20 cc. of cone. HC1. Cover and heat on the steam bath for fifteen minutes. Add 100 cc. of hot water and boil. (If there are any insoluble impurities, filter them out, wash with boiling water, dry and weigh.) Add NELiOH until a white permanent precipitate forms. Redissolve with a slight excess of acetic acid, heat to boiling and precipitate with K 2 Cr 2 O 7 as under Total Lead in White Lead, page 212. Weigh as PbCr0 4 . Calculate to PbO and subtract from the latter the equivalent of the PbsO4 in terms of PbO. The difference will be the PbO actually present as such in the material. ANALYSIS OF PAINTS AND PAINT MATERIALS 211 CALCULATIONS. PbCrO 4 X 0.6906 = PbO. Pb 3 O 4 X 0.9767 = PbO. NOTE. In impure materials the impurities or adulterations likely to be found are organic dyes and water soluble materials. The sample should also be tested for nitrate, nitrite, carbonate, and sulfate. Organic coloring matter may generally be detected by adding 20 cc. of 95% alcohol to 2 grams of the pigment; then heat to a boil and let settle. Pour off the alcohol, boil with water and let settle. Then use very dilute NH 4 OH. If either the alcohol, NH 4 OH or water is colored, it indicates organic coloring matter. The quan- titative determination of such adulteration is difficult and must be generally estimated by difference. REFERENCE. U. S. Dept. of Agriculture, Bur. of Chem., Bulletin 109, revised (1912). WHITE LEAD (BASIC CARBONATE) General. The theoretical composition of white lead is Pb(OH) 2 -2PbC0 3 . The analysis of this would show: . Total lead, calculated as PbO 86.33% Carbon dioxide, C0 2 11 .35% Equivalent to: Lead hydroxide, Pb(OH) 2 31 . 10% Lead carbonate, PbC0 3 68.90% It is furnished to the trade either dry or ground in oil. In the latter case, the amount of oil is generally about 8%. The pigment of commercial white leads generally shows between 11 and 13% of C02. There should be no appreciable acid-insoluble material and no acetates present. The pigment should contain at least 98% of white lead, calculated to the above formula. Vehicle (Oil). Pour about 50^75 grams of the thoroughly mixed material into a small beaker and weigh it in the beaker together with a stirring rod. Dry and weigh one of the specially hardened extraction thimbles with a plug of cotton wool in the end of it. Remove the cotton and fill the thimble about two-thirds full of the pigment, pouring it from the beaker by means of the stirring rod. Weigh the beaker again, and from the difference in weight obtain the amount of material taken. Plug the thimble with the 212 TECHNICAL METHODS OF ANALYSIS cotton and place in a Soxhlet extractor. Fill the extractor about two-thirds full of ether, and let stand at least an hour before start- ing the siphoning. Then extract until no more oil is removed. Dry the thimble with the extracted pigment and weigh it. NOTE. When it is not necessary to determine the relative proportions of oil and pigment, the oil can be extracted more quickly by stirring some of the material up with 86 naphtha, centrifuging and decanting the liquid several times until no more oil is left in the pigment. ANALYSIS OF PIGMENT Insoluble Matter. Dissolve 0.5 gram of the dry pigment (well mixed and ground) in hot dil. HNOs. If an appreciable amount of insoluble material remains, filter the solution, wash the residue with hot water, ignite and weigh. Total Lead. Dilute the filtrate from the insoluble material to about 250 cc. and add NILiOH until the solution is slightly alkaline. Then add dil. acetic acid in slight excess. Heat to boiling, and add a hot solution of K^C^O? in excess. Sufficient bichromate solution should be added so that when the PbCrO4 settles, the supernatant liquor is distinctly yellow or orange. Boil until the precipitate settles quickly (about two minutes generally suffices). Filter through a weighed Gooch crucible. Wash with hot water, dry at 110 C., set the Gooch crucible in a larger platinum crucible, ignite gently, cool in a desiccator and weigh as PbCr0 4 . Carbon Dioxide. For this determination use Mohr's alka- limeter (Fig. 13). Weigh out FIG. 13.-Alkalirneter for Carbon 5 - 10 S ramS f the dried pigment Dioxide Determination. and b sh it carefully by means of a funnel scoop into the lower chamber of the alkalimeter (A). Add a few cc. of water and ANALYSIS OF PAINTS AND PAINT MATERIALS 213 connect the acid chamber (B). Close the stop-cock (E) and fill this chamber nearly full of dil. HC1 or HNO 3 (1:4). In the chamber (C) should be just enough cone. H2S04 to make a seal. Weigh the whole apparatus. Open the stop-cock (E) and let the dilute acid run slowly upon the pigment. When nearly all the acid has run in, close the stop-cock. Complete the reaction by placing the whole apparatus in a beaker and setting on the steam bath. Make sure that all the pigment has been dissolved, then allow the alkalimeter to come to room temperature. Attach a small rubber tube to the outlet of the side arm at (D) and open the stop-cock. Suck air slowly and carefully through the appa- ratus until all the liberated CCb has been replaced by air; then weigh on the balance. The loss in weight is CC>2. Lead Acetate. Dry some of the extracted pigment at 105 C. Place a little on a watch glass and moisten with a few drops of KI solution. If acetates are present a yellow color will form. To determine the amount, weigh 5-10 grams of the pigment into a distilling flask, add 200 cc. of water and 5 cc. of sirupy H3PO4 and distill* until the distillate no longer comes over acid. Then titrate the distillate with 0.1 N NaOH and phenolphthalein. 1 cc. 0.1 N NaOH = 0.01897 gram Pb(C2H3O 2 )2-3H 2 O. CALCULATIONS. Calculate the PbCrO4 to PbO. Calculate the CO 2 to PbCO 3 . From the total PbCrO 4 subtract the equiva- lent of PbCO 3 and of Pb(C 2 H 3 2 )2 and calculate the difference to Pb(OH) 2 . PbCr0 4 X0.6906 = PbO. CO 2 X 6.0724 =PbCO 3 . PbCO 3 X 1.2095 = PbCrO 4 . Pb(C 2 H 3 O 2 ) 2 3H 2 O X0.8521 = PbCr0 4 . PbCr0 4 X 0.7464 =- Pb(OH) 2 . CHROME YELLOW General. The most important yellow pigment is chrome yellow, which varies from a light yellow color to deep orange. The lighter shades generally contain PbS0 4 as well as PbCr0 4 , while the deep orange contains some basic lead chromate. Pure chrome yellow should contain only PbCrO 4 , PbSO 4 (or white lead) 214 TECHNICAL METHODS OF ANALYSIS and possibly some basic lead in the darker shades. A chrome yellow should be considered adulterated if it contains anything in appreciable amount besides insoluble lead compounds. The following analytical method applies to the dry pigment or to the pigment of a paste after the vehicle has been removed. Moisture. Dry 2 grams in a watch glass or dish to constant weight at 105 C. NOTE. As the pigment of pastes is generally dried in extracting the oil, the determination of moisture in such cases is superfluous. Insoluble Impurities. Treat 1 gram in a beaker with 20 cc. of cone. HC1, cover and heat on a steam bath for fifteen minutes. Add 100 cc. of hot water, boil (the solution should be complete), filter, wash very thoroughly with hot water and ignite in a plati- num crucible, cool in a desiccator and weigh the insoluble impurities. If the impurities are considerable in amount, evaporate the resi- due twice with HF and a few drops of dil. H^SCU, drive off the H2S04, ignite and weigh. If the insoluble impurities are mostly volatile with HF, report as silica. If the residue from the HF treatment is considerable, fuse it with KHSO4 and dissolve in water containing a little HC1. Filter out, ignite and weigh any insoluble residue. If the treatment with HF was complete, this residue is probably BaSO4, but it should be confirmed qualitatively. The filtrate from the BaS04 will contain insoluble Fe2Os, A^Og, CaO, and MgO. Iron and alumina alone indicate clay or similar mineral matter; a considerable airount of MgO generally indicates talc. It is not generally necessary to determine the insoluble Fe, Al and Mg; the difference between the BaSC>4 (if any) and the total insoluble may be reported as " silicious mat- ter." Total Lead. Nearly neutralize the filtrate from the insoluble with NHjOH, and dilute to 350 cc. Pass in a rapid current of H2$ for ten minutes. Cover with a watch glass, place on the steam bath and let settle. Test a portion of the clear liquid for complete precipitation by diluting with an equal volume of water and passing in more H2$. If precipitation is not complete, dilute the whole to about 500 cc. and pass in more H2$. Finally filter out the PbS and wash rapidly with water containing a little H 2 S. Dissolve the PbS with hot cone. HNOs containing a little bromine water, dilute ANALYSIS OF PAINTS AND PAINT MATERIALS 215 to about 100 cc., filter out sulfur if necessary, cool, add 5 cc. of cone. H2SO4 and evaporate to strong SOs fumes. Cool and add 100 cc. of water and 50 cc. of alcohol. Let stand one hour or until the PbSO 4 settles clear. Filter on a weighed Gooch crucible, wash with 5% H 2 SO 4 and finally with alcohol, dry at 105 C., set the crucible in a larger platinum crucible, ignite until pure white, cool and weigh as PbSO 4 . NOTE. In the presence of Cr the solution must be very dilute and the H 2 S passed in rapidly to get a good separation. Iron Oxide. Boil the filtrate from the PbS until free from H 2 S. Add a few drops of HNOs and boil again. Add a very slight excess of NILtOH, boil, let settle, filter and wash with hot water. Save the filtrate. Dissolve the precipitate (hydroxides of Fe, Al and Cr) in HC1. Rinse into a 500 cc. flask and dilute to the mark. To 200 cc. of the above solution, add a very slight excess of NH^OH and then sufficient Na 2 2 to oxidize all the Cr. Boil until all the H20 2 is expelled. If there is any insoluble residue filter and wash slightly. Dissolve the precipitate in dilute HC1, dilute with water, add a slight excess of NEUOH and treat with peroxide as before. Filter, wash with hot water and add this filtrate to the previous one. Ignite the residue strongly in a platinum crucible, cool and weigh as Fe 2 O3. Divide by 0.4 and multiply by 100 to obtain the percentage of Fe 2 Os. Lead Chromate. Acidify with acetic acid the total filtrate from the Fe determination, boil until all H 2 2 is decomposed and treat by one of the following methods. In either case divide the result by 0.4 and multiply by 100 to obtain the percentage of PbCr0 4 . (A) GRAVIMETRIC METHOD. Add a slight excess of clear lead acetate solution (a basic solution can generally be cleared up by adding a few drops of acetic acid), digest on the steam bath until the precipitate settles clear, and filter on a weighed Gooch crucible, washing with hot water; dry. set the crucible inside of a larger platinum crucible, ignite gently, cool in a desiccator and weigh as PbCrO 4 . The precipitate should not be allowed to stand in the beaker an undue length of time after settling clear on account of the danger of becoming basic, nor should the solution be boiling when the lead acetate is added or at any time after. 216 TECHNICAL METHODS OF ANALYSIS (B) VOLUMETRIC METHOD. Cool the solution which has been freed from H202, make distinctly acid with H2SO4, add a meas- ured excess of 0. 1 N ferrous ammonium sulf ate solution and titrate the excess of the latter with 0.1 N K^toO?, using potassium ferricyanide as outside indicator. The end point is the disap- pearance of the blue color first produced. From the amount of ferrous ammonium sulfate oxidized, calculate the amount of PbCrO 4 corresponding to the chromium. CALCULATION. 1 cc. of 0.1 N Fe(NH 4 ) 2 (SO 4 )2 = 0.01077 gram PbCrO 4 Alumina. Alumina is very seldom present but may be deter- mined in an aliquot of the solution prepared above under Iron Oxide by oxidizing with Na2C>2, filtering, rendering the filtrate slightly acid with HC1 and then very slightly alkaline with NHiOH. This precipitates A1(OH) 3 . Filter out, wash with hot water, ignite in the blast, cool and weigh as A^Os. (After adding the NHUOH, the solution should not be boiled, as some Cr(OH)s may also thus be thrown down.) Zinc Oxide. Into the filtrate from the original precipitate of hydroxides of Fe, Al and Cr (which should be ammoniacal) pass H 2 S, heat to boiling, and filter through a hardened filter, washing with H 2 S water. Dissolve the ZnS in a slight excess of HC1, dilute to 200 cc., boil off the H 2 S, add a slight excess of NH 4 OH and then an excess of acetic acid. Heat to boiling and add an excess of ammonium phosphate solution. Stir thoroughly and let stand until clear; filter and wash with water on a porcelain Gooch crucible, dry, ignite and weigh as Zn2P 2 O7. Calculate to ZnO. CALCULATION. Zn 2 P 2 O 7 X 0.5339 = ZnO. Lime. (Lime is seldom present unless the material is adul- terated with gypsum.) To the ammoniacal filtrate from the ZnS add an excess of hot sodium or ammonium oxalate solution and determine the CaO in the usual way. Magnesia. The MgO is determined in the filtrate from the CaO in the usual way by precipitating as MgNHjPO 4 . It is almost never present in a chrome yellow in an acid-soluble form. Total SO 3 . Dissolve 1 gram of the pigment by heating with 20-25 cc. of HC1 as described under Insoluble Impurities. Filter and wash with hot water and dilute to about 400 cc. Heat to boiling and add an excess of hot BaCb solution, drop by drop; ANALYSIS OF PAINTS AND PAINT MATERIALS 217 boil for 0.5 hour and let stand on the steam bath until the pre- cipitate settles clear. Filter and wash with hot water, ignite in a platinum crucible and weigh as BaSO4. Calculate to SOs. CALCULATION. BaSO 4 X 0.3430 = S0 3 . NOTE. The solution must be kept hot and dilute to prevent contamina- tion of the precipitate with lead. Calculation of Results. The PbCr(>4 is obtained directly. Calculate the SOs to PbSO4 and subtract this from the total Pb calculated as PbSO4. From the remainder subtract the equiva- lent of the PbCrC>4. Calculate the remaining lead to PbO and report as basic PbO; or, if carbonates are present, calculate to white lead, 2PbCO 3 -Pb(OH) 2 . The small amounts of Fe 2 O 3 , CaO, etc., are reported as such, unless it is evident that calcium carbonate or sulfate is present. NOTE. Some chrome yellows contain barium phosphate, barium sulfate or calcium sulfate in place of the PbSO 4 ; sometimes also lead citrate is used. FACTORS. BaSO 4 Xl.2991 = PbSO 4 . PbCrO 4 X 0.9383 = PbSO 4 . PbS0 4 X 0.7360 = PbO. PbSO 4 X 0.8526 = White Lead. SO 3 X2.1504 = CaSO 4 -2H 2 O. REFERENCE. Bulletin 109 (Revised) U. S. Bur. of Chem., page 29. OIL VARNISHES General. The methods of analysis for varnishes are far from satisfactory. Several methods for the determination of gums have been proposed, but none of them is reliable. It is, therefore, advisable to omit this determination and report the combined percentage of oil and gums. In case it is desired to get an approx- imate estimate of the gums, Boughton ; s method as described in Technologic Paper No. 65 of the Bureau of Standards, may be used. Care should be used in passing final judgment on a varnish merely from chemical tests, as the gloss, working qualities and durability depend largely upon the quality of the gum used, the quality and treatment of the oil, the quality of the driers and 218 TECHNICAL METHODS OF ANALYSIS especially as to how the varnish was prepared as regards heat, method of cooking, aging, filtering, etc. In the manufacture of oil varnish the resins or " gums " are melted and at the proper time the oil and driers are added. The mass is mixed and heated, then cooled and thinned, filtered and run into settling tanks where it is aged. The time of heat- ing, temperature, etc., depend upon the nature of the varnish and vary greatly for different kinds. Many different resins are used, such as Kauri, Zanzibar, Pon- tianak, Manila and Colophony (rosin). Generally the harder resins are more valuable. The principal oils used are Linseed and China Wood (Tung Oil). Common thinners are turpentine and light mineral oils. Compounds of Pb and Mn (and some- times Co) are used as driers, and if rosin is present, lime is gen- erally added to harden it. The following tests are of value in judging the quality of a varnish: Appearance and Odor. Transfer the sample, if it is in a metal container, to a glass-stoppered cylinder or bottle and note its appearance, color, transparency, body, and whether any sediment is present. The presence of light petroleum oil or wood turpen- tine may often be detected by the odor. After making these observations, the sample should be thoroughly mixed before making the remaining tests. Specific Gravity. Determine with a pycnometer, hydrometer or Westphal balance at 15.5 C. Flash Point. Determine the flash point as described in the method for Lubricating Oils (see page 255), using a low flame; begin testing at 65 F. and stir the varnish while heating. A low flash point may indicate light petroleum oil. Turpentine flashes at about 93 F. Viscosity. The determination of viscosity is not generally necessary but may throw some light on its working qualities, par- ticularly in comparison with other varnishes. Sufficiently accurate comparative results may be obtained by using a 100 cc. Dudley pipette and comparing with water as described in the method for Glue (page 323). The viscosity should be run at 20 C. Volatile Thinners. Weigh 100 grams of the varnish into a 500 cc. flask, connect with a spray trap and a water condenser, and ANALYSIS OF PAINTS AND PAINT MATERIALS. 219 pass through it a current of steam, first heating the flask in an oil bath at 100 C. With the steam still passing through, raise the temperature of the bath to 130 C. Catch the distillate in weighed glass-stoppered graduates, and continue distillation until 300 cc. of water have been condensed. Do not fill any graduate above the top mark. Let the distillates stand until they separate into two layers, then read the volume of water and weigh each graduate. Subtract the volume of water and weight of empty graduates and the difference is the weight of volatile thinners. Filter the light oils through dry papers and examine for turpentine and petroleum oils. A slight error is caused by the solubility of turpentine in water; this amounts to about 0.3-0.4 cc. for each 100 cc. of water. When sufficient varnish is available, it is well to take another portion and distill without steam or spray trap, placing the weighed flask in an oil bath. Note the temperature of the bath at which distillation begins, and continue distillation at a tem- perature of 185 C. in the oil bath, finally raising the temperature to 200 C. This method generally tends to give lower results on volatile oils than the steam distillation method ; but the distillate can be tested for water-soluble volatile liquids, which would be lost by the steam distillation. Fixed Oils and Gums. The percentage of fixed oils and gums is obtained by subtracting the percentage of volatile thinners from 100. A check upon this determination is obtained by weighing the residue from the dry distillation. In case an actual determination of the approximate amount of gums is required, use Boughton's method previously referred to. Acid Number. Determine the acid number as described on page 242 using 10 grams of the varnish. Rosin. After determining the acid number, decant the alcohol, evaporate and apply the Liebermann-Storch test; or test the orig- inal varnish as follows: Pour about 5 cc. into 9, separatory funnel, add an equal volume of C$2, shake and add 10 cc. of acetic anhydride. Let stand until completely separated. Draw off the lower layer of anhydride. Pour 1-2 cc. of this into an inverted crucible cover, add carefully by means of a stirring rod 1 drop of H 2 SO4 (1 : 1) to the edge of the cover so that it will mix slowly with the anhydride. If rosin is 220 . TECHNICAL METHODS OF ANALYSIS present a characteristic fugitive violet color will result. Do not confuse this with the brown or reddish brown color often given by other gums. Ash. Determine the ash in 10 grams, using a quartz or por- celain dish and carrying out the incineration at a low heat, best in a muffle. Determine the reaction of the ash to litmus paper and make a qualitative analysis. It is frequently well to make a quantitative determination of CaO, a large amount of which indicates rosin. It is sometimes advisable to determine the percentages of Pb and Mn. Some Pb will, however, be lost in the ashing, and if a correct determination is required, proceed as under the determination of Drying Salts in Japan Driers (see below). Examination of Films. Flow the varnish on a carefully cleaned plate of glass and let dry at room temperature in a vertical position. Note the time required for the varnish to " set to touch," i.e., to lose its tackiness, and also the time required for drying hard. After twenty-four hours examine the film, noting transparency, hardness, elasticity and tendency to dust by scratching. After thoroughly drying, immerse the plate in water overnight, dry without heat and note the appearance. Practical Tests. The working quality of a varnish must be determined by application on wood and it is best to make this test on well-seasoned and perfectly smooth white pine. Apply a thin coating of the varnish, let dry and sandpaper down smooth and then apply the coat to be tested. Observe how the varnish works under the brush, character of coat, etc. This panel, after drying, may be used for further testing as to whether the varnish will stand rubbing, etc. NOTES. (1) " Short oil " varnishes contain considerably more resin than oil, and " long oil " varnishes vice versa. To test, place 10 cc. of varnish in a small beaker and add 50 cc. of benzine which has been previously cooled to 5 C. If the varnish is " short oil," the gums will be partly precipitated and light in color; if a " long oil " varnish, there will be very little precipitation and the solution will be dark colored. Interior varnishes should be " long oil " for the best quality of work and exterior varnishes should always be " long oil." Rubbing varnishes should be " short oil." (2) The Navy specification for Interior Varnish for Naval Vessels, 1906, requires: ANALYSIS OF PAINTS AND PAINT MATERIALS 221 (a) Flash above 105 F. (6) Set to touch in 6-8 hours. (c) Dry hard within twenty-four hours at 70 F. (d) Must stand rubbing with pumice stone and water within twenty-four hours after application without sweating and must polish in seventy-two hours with rotten stone and water. JAPAN DRIERS General. The analysis of Japan driers is in general conducted in the same way as that of Oil Varnish, with the following modi- fications: Volatile Thinner. Weigh quickly 5 grams from a weighing bottle into a glass Petri dish and dry for three hours at 100 C. Cool and weigh. The loss represents the amount of volatile thin- ner, which is generally about 65%. Drying Salts. The most generally used driers are linoleates, resinates or borates of Pb and Mn, or mixtures of these substances. As a rule, in light Japans manganese borate is used ; in dark Japans, manganese oxide; and medium Japans, lead oxide. The most common composition is a mixture of Mn borate and PbO. Zinc is now seldom used. Qualitative tests for Mn, Pb and Zn are generally sufficient, but if a quantitative determination is desired proceed as follows: Place 25 grams in a 250 cc. Erlenmeyer flask and add 25 cc. of a mixture of equal parts of gasoline and turpentine. Add 50 cc. of HNOs (1:1) and let stand one hour, shaking every ten min- utes. Then immerse in hot wate and shake gently. Keep away from any flame. When thoroughly hot, mix with a circular motion to get rid of most of the gasoline. When cool, pour into a separatory funnel, draw off the lower aqueous layer into a casserole and wash the upper oily layer 4-5 times with warm water. Add the washings to the casserole and evaporate to dryness under the hood. Dissolve by warming with dil. HNOs. Filter into a 250 cc. flask and make up to the mark. LEAD. To 100 cc. of the above solution add 5 cc. of cone. H2S04 and evaporate on the hot plate to strong fumes. Cool, add 100 cc. of water and heat to boiling to dissolve anhydrous FeSCU if present. Let stand until cooled and the precipitate has settled clear. Filter on a Gooch crucible and wash with 5% 222 TECHNICAL METHODS OF ANALYSIS H2S04. Transfer the filtrate, which should be about 150 cc., to a beaker and wash the PbSO 4 on the Gooch crucible with 50% alcohol. Do not collect the alcoholic filtrate in the main body of the filtrate as it is used merely to wash out the acid from the PbSC>4. Dry the Gooch crucible and contents and place in a larger platinum crucible. Ignite gently, cool in a desiccator and weigh as PbS0 4 . Calculate to PbO. CALCULATION. PbS0 4 X 0.7360 = PbO. ZINC. If present, Zn may be determined in the filtrate from the PbS0 4 . To the filtrate in the beaker add 50 cc. of 30% NaOH solution and electrolyze as described under Zinc on page 146, finally weighing as metallic Zn. MANGANESE. The manganese will be present as hydroxide in the solution after electrolysis of the Zn. Dilute this solution to about 300-400 cc., filter, and wash with hot water. Wash into a 200 cc. Erlenmeyer flask with 50 cc. of warm HNOs (1 : 3), cool, add about 0.5 gram of sodium bismuthate and proceed as described under Manganese on page 110. NOTES. (1) The following specifications of the P. & R. Railroad, 1906, will give an idea of the requirements of a good Japan. The material desired consists of a pure turpentine hardener and oil drier, conforming to the following: (a) When equal parts by weight of the Japan and of pure turpentine are thoroughly mixed and poured over a slab of glass, which is then placed nearly vertical at a temperature of 100 F., with a free access of air but not exposed to draft, the coating shall be hard and dry, neither brittle nor sticky, in not exceeding twelve minutes. (6) When thoroughly mixed with pure raw linseed oil at the ordinary temperature in proportions of 5% by weight of Japan to 95% by weight of raw linseed oil, no curdling shall result, nor any marked separation or settling on standing. (c) When the above mixture is flowed over a slab of glass, which is then placed nearly vertical at a temperature of 100 F., with free access to air but not exposed to draft, the coating shall dry throughout, neither brittle nor sticky, in not exceeding two hours. (d) When 5 cc. of the Japan are poured into 95 cc. of pure turpentine at the ordinary temperature, and thoroughly shaken, a clear solution shall result, without residue, on standing one hour. (e) After evaporation of the turpentine, the solid residue must be h&rd and tough, and must not " dust " when scratched with a knife. (/) Benzine or mineral oil of any kind will not be permitted. (2) The U. S. Navy Specification is as follows: ANALYSIS OF PAINTS AND PAINT MATERIALS 223 Japan drier must not flash below 105 F. (open tester), must be of the best quality and made from pure kauri gum, pure linseed oil, pure spirits of tur- pentine, and lead manganese driers, and be free from adulterants and all other foreign materials, must set to touch in from one-quarter to one hour, dry elastic in from eighteen to twenty-one hours at a temperature of 70 F., and must not rub up or powder under friction by the finger. When mixed with pure raw linseed oil in the proportion of 8 parts of oil to 1 part of drier, must remain clear for two hours and set to touch in from six to seven hours at a temperature of 70 F. REFERENCES. Bureau of Standards: Technologic Paper No. 65; Bureau of Chemistry: Bulletin 109, revised; Holley and Ladd: " Mixed Paints, Color Pigments and Varnishes." SHELLAC AND SHELLAC VARNISHES General. Dry shellac occurs on the market in two forms, orange shellac and bleached shellac. There are many different grades of orange shellac. Bleached shellac, however, is generally of a high grade of purity. In making shellac varnishes, the dry lac is dissolved in alcohol, generally in the proportions of about 5 Ibs. of the shellac to a gallon of alcohol. Formerly methyl alcohol was almost universally used but now denatured alcohol is largely employed. SAMPLING OF DRY SHELLAC Bleached shellac is sold in three forms, (1) as hanks or bars containing approximately 25% of water, (2) as ground bleached in pulverized form with about the same water content, (3) as bone-dry or kiln-dried shellac. The latter is prepared by drying the ground bleached shellac in the air or in vacuum driers at mod- erate temperatures. It may contain up to 10% of water or more. In sampling bone-dry bleached shellac, a fairly large portion (about 1 Ib.) should be taken from different parts of the barrel and finally ground by running quickly through a coffee mill. No attempt should be made to sieve it. It should be rapidly mixed and transferred to a Mason jar with a screw cap and a rubber ring seal. The jar should not be more than two-thirds full, leaving room for a thorough mixing by shaking the contents. It must be kept in a cool place and tested as promptly as possible. If too warm, the shellac may become partly caked, in which case the lumps must be broken up by shaking the bottle. 224 TECHNICAL METHODS OF ANALYSIS In sampling bars or hanks, it is recommended that a whole hank be taken. It should be crushed and ground as rapidly as possible. Ground bleached may be treated as above, bearing in mind that the large amount of moisture present makes rapid handling imperative. ANALYSIS OF DRY SHELLAC Moisture. Both the orange and bleached shellac give off vola- tile matter at temperatures approaching 100 C. Bleached shellac alters chemically at this temperature, losing its solubility in alco- hol. For these reasons the usual methods of determining water by heating in an air bath at 100-110 C. are not applicable. METHOD No. 1. Weigh 5-10 grams of the sample in flat- bottomed dishes about 4 inches in diameter or in watch glasses ground to fit and provided with a clamp. Then place the shellac in a desiccator freshly filled with cone. H^SCX. The con- tents of the dish should be spread out in a thin layer to expose as large a surface as possible. Exhaust the desiccator by a vacuum pump as completely as possible. With a good vacuum (3 mm. pressure or better) constant weight will be obtained in between twenty-four and forty-eight hours. Absolutely dry bleached shellac is quite hygroscopic and the final weight should be taken as rapidly as possible. METHOD No. 2. The same results may be obtained by drying the shellac in a well-ventilated air bath from three to six hours at 100-110 F. (38-43 C.). One or two electric -light bulbs pro- vide a convenient source of heat. The temperature should not be allowed to rise above 43 C., otherwise sintering may occur and retard drying. With poorly ventilated ovens the drying may take much longer. Completeness of drying should be ascertained by continuing the treatment to constant weight. (Check the accuracy of results obtained in the oven by comparison with a test made in a vacuum desiccator before relying exclusively on the oven.) Rosin. Introduce about 0.200 gram of the shellac (0.400 gram of bleached shellac) finely ground, into a 250 cc. dry bottle of clear glass with a ground-glass stopper; add 20 cc. of glacial acetic acid (99%) and warm the mixture gently until solution is ANALYSIS OF PAINTS AND PAINT MATERIALS 225 complete (except for wax). A pure shellac is rather difficult of solution. Solution is quicker according to the proportion of rosin present. Add 10 cc. of CHCls and cool the solution to 21-24 C. The temperature should be held well within these limits during the test. Add 20 cc. of Wijs solution from a pipette or burette having a rather small delivery aperture. Close the bottle and stand in a dark place, and note the time. It is convenient to keep the bot- tles during the test partly immersed in water, which should be kept as nearly as possible between 22-23 C. Pure shellac will scarcely alter the color of the Wijs solution. If in small amount, rosin will produce a slowly appearing red- brown color; in large amount, rosin causes an immediate colora- tion, increasing in intensity as time passes. At the end of one hour add 10 cc. of 10% KI solution and titrate immediately with 0.1 N sodium thiosulfate solution; 25-30 cc. may be run in imme- diately, unless the shellac is very impure, and the remainder added gradually with vigorous shaking. Just before the end, add a little starch solution. The end point is sharp, as the reaction products of shellac remain dissolved in the chloroform. Disregard any color returning after a half minute or so. Run a blank determination with 20 cc. of Wijs solution, 20 cc. of acetic acid, 10 cc. of CHCls and 10 cc. of 10% KI solution. The blank is necessary on account of the well known effect of temperature changes on the volume and possible loss of strength of the Wijs solution. Subtract the titration of the sample from that of the blank and calculate the iodine number of the sample (percentage of I absorbed) . From this' calculate the percentage of rosin. CALCULATION. Let 7 = percentage of rosin; A = iodine number of mixture ; M 18,* iodine number of shellac; TV = 228, iodine number of rosin; (A-M) then F=100 N-M NOTES. (1) In the case of grossly adulterated samples or in the testing of pure rosin, it is necessary to use, instead of the 0.2 gram of material, a smaller amount, say 0.15 gram, or even 0.1 gram, in order that the excess of iodine monochloride may not be too greatly reduced, as the excess of halogen * For bleached shellac use 10 instead of 18. 226 TECHNICAL METHODS OF ANALYSIS is one of the factors in determining the amount of absorption. It is safe to say that in case less than 25 cc. of thiosulfate solution are required, another test should be made, using a smaller amount of the shellac to be tested. In the case of bleached shellac, 0.4 gram should be taken. (2) The time and temperature are very important factors. Variations from the conditions prescribed lead to unreliable results. (3) The strength of the glacial acetic acid used in dissolving the shellac, and also in the Wijs solution, is equal y important and must be maintained within the required limits. The strength of acid adopted is 98.8-99.1%, with a melting point of 14.7-15 C., as determined by the Titer Method. (See page 246.) (4) In weighing shellac, some difficulty is at times experienced on account of its electrical properties; in very dry weather it may be found that the necessary handling to prepare it for weighing has electrified it and that it may be necessary to leave it on the balance pan at rest for a few minutes before taking the final weight. (5) The following table shows how the iodine number may be interpreted in judging the quality of the shellac: Iodine Number Rosin Per Cent Quality Orange Shellac Bleached Shellac 18 or less 10 or less None Good 18-23 10-15 0-2.5 Fair 23-28 15-21 2.5-5 Poor 28-34 21-26 5-7.5 Bad Over 34 Over 26 Over 7.5 Grossly adulterated (6) Shellac having an iodine number of 23 begins to show rosin by the acetic anhydride test. (See page 356.) (7) The iodine number of bleached shellac is less than that of crude shellac. Bleached shellac of good color will generally run about 8. Anything over 10 would point to adulteration. The values 10 and 228 for shellac and rosin should be used on bleached samples. (8) The above method is the one recommended by the Committee on Uni- formity in Technical Analysis, American Chemical Society. ANALYSIS OF SHELLAC VARNISH Specific Gravity. Determine the sp. gr. at 15.5 C. with the Westphal balance or pycnometer, as may be most convenient. Total Solids. Weigh from a weighing bottle about 25 grams into a flat evaporating dish, evaporate spontaneously and finally ANALYSIS OF PAINTS AND PAINT MATERIALS 227 dry in an oven at 40 C. to constant weight. (See under Moisture in Dry Shellac.) Rosin. Determine the iodine number of the dried residue and calculate the percentage of rosin as described in the analysis of Dry Shellac above. The residue must be entirely free from solvent which might affect the iodine number. Examination of Solvent. Distill a portion of the original and note the temperature at which the solvent comes over. Deter- mine the sp. gr. at 15.5 C. of the latter, evaporate a portion on filter paper or blotter and note the odor of the last portions. In the case of denatured alcohol, the first odor is methyl alcohol and the final odor ethyl alcohol. The boiling point will also indicate whether the solvent is wood or denatured alcohol. Calculate the pounds of dry gum shellac per gallon of the sol- vent. CALCULATION. Multiply the sp. gr. of the original by 8.33. This gives the weight per gallon of the varnish (A). This multi- plied by the percentage of total solids, expressed as a decimal, will then give the pounds of dry shellac per gallon of the varnish (B). Subtract (B) from (A) and the difference is the weight of solvent per gallon of varnish (C). Divide (C) by the sp. gr. of the solvent times 8.33; call this (D). (B) divided by (D) will give the pounds of dry shellac per gallon of solvent. REFERENCES. " The Determination of Rosin in Shellac." A. C. Lang- muir, J. Soc. Chem. Ind. 24, 12 (1905). Report of the Sub-committee on Shellac- Analysis. J. Amer. Chem. Soc. 29, 1221 (1907). " Analysis of Shellac." Parry, J. Soc. Chem. Ind. 20, 1245 (1901). " Method of Analyzing Shellac." Mcllhiney, J. Amer. Chem. Soc. 30, 867 (1908). Allen's " Commercial Organic Analysis," 3d edition, Vol. II, Part III, page 191. " The Determination of Moisture in Shellac." J. Ind. Eng. Chem. 7, 633 (1915). BLACK AIR-DRYING INSULATING VARNISH AND BLACK BAKING INSULATING VARNISH ELECTRIC RAILWAY SPECIFICATIONS General. The materials which these specifications cover are insulating varnishes for armatures, field coils, etc., in suitable form 228 TECHNICAL METHODS OF ANALYSIS for use as received. They are a combination of asphalts and drying oils as a base, with a solvent, preferably a hydrocarbon solvent such as benzine or naphtha. It is aimed to obtain a product which, under the effects of the continued heat of normal operating condi- tions, shall show the least loss of plasticity or lowering of its dielectric strength. It must not be sufficiently fluid under these conditions to be forced out of the coils. Quality of Base. The base must contain no pigment of any kind. (It should show no significant residue insoluble in benzene or CCU.) It must be entirely non-volatile and after application as a coating, evaporation of solvent and drying, must become a solid that does not, at 100 C., become sufficiently softened to flow. Quality of Solvent. The solvent must not contain any alcohol or other electrically conductive material. It must be entirely volatile without residue at 100 C. Amount of Solvent. The air-drying varnish should not con- tain more than 60% by weight of solvent, and the baking varnish not more than 50% when tested as follows: Place 400 cc. of water in a liter distilling flask. Pour in 100 grams of the varnish and distill with steam. Continue the dis- tillation until no more volatile oil distills over, taking care that there is always some water in the distilling flask so that the tem- perature shall not rise above 212 F. Collect the distillate in a graduated weighed cylinder and weigh it, subtracting the volume of water distilled with it; or, measure its volume and deter- mine its sp. gr. and calculate the weight. Drying Test. (a) AIR-DRYING VARNISH. Dip a piece of clean window glass in the varnish and dry in a vertical position at room temperature (70 F.). It should be set firmly in one hour and become free from tackiness in not more than three hours. (6) BAKING VARNISH. See under Dielectric Strength below. Flexibility. Place some of the varnish in a shallow pan and dip in it, one at a time, several sheets of glassine paper about 1 foot square. Hang the sheets up and let them dry perpendicu- larly overnight. (See below.) (a) AIR-DRYING VARNISH. When thoroughly dry the sheets should be flexible and should withstand ordinary bending without any of the varnish cracking, flaking or breaking off. ANALYSIS OF PAINTS AND PAINT MATERIALS 229 (b) BAKING VARNISH. Bake the sheets for ten hours at 100 C. They should then withstand bending back flat on themselves without any cracking. Continue the baking for ten days. They should still be flexible and withstand ordinary bending without cracking. Dielectric Strength. Use the above sheets for testing, unbaked sheets in the case of air drying varnish and sheets baked ten days at 100 C. in the case of the baking varnish. Apply an alternating current at low voltage and raise gradually until punc- ture occurs. Make as many tests as possible on each sheet and report the final figure as the average of at least 10 breakdowns. Also make at least 10 micrometer measurements to obtain the average thickness of the varnish film. The varnish should have a dielectric strength of at least 1000 volts for each mil of thickness. Oil Resisting Properties. Take a piece of ribbon copper, 4 inches long and 2 inches wide, and dip for 3.5 inches of its length in the varnish. Let drain fifteen minutes. Place in a baking oven and bake until thoroughly dry (20-24 hours at 100 C.). With a pair of heavy scissors cut 0.5 inch off the bottom so as to remove the bead and all thickening of the coat. Also trim 0.25 inch off each side of the copper for its full length. Take a dish (metal or enamel ware) and fill it about 4 inches deep with water. On top of this pour a layer of about 0.5 inch of fairly heavy engine oil. Place the coated sheet of copper in this mixture of oil and water until the oil comes to about the middle of the coat of varnish. Then boil the mixture vigorously. At the end of every fifteen minutes or so remove the piece and see whether the varnish has suffered any injury. The coating should withstand two hours' boiling without any effect and should withstand four hours' boiling without any material deterioration. CHAPTER VII ANALYSIS OF OILS, FATS, WAXES AND SOAPS ANIMAL AND VEGETABLE OILS AND FATS General. The usual determinations on materials of this class are specific gravity, refractive index, iodine number, saponification number and fatty acid; and, in the case of fats, melting point. Certain other tests are often of value in the case of particular oils or in attempting to identify mixtures. In this method are collected general procedures for such tests as are standard. In the case of individual oils, the special method for that particular oil should be consulted. A table of the so- called " constants " of various oils is given on pages 232-239. Preparation of Sample. In the case of solid fats, melt and filter, using a hot water funnel. Make analyses on the melted homo- geneous mass. Filter oils which are not clear. Keep samples in a cool place protected from light and air to avoid becoming rancid. It is best to weigh out at once as many portions as are needed for the various determinations. Specific Gravity. (A) AT 15.5 C. Determine the sp. gr. at 15.5 C. (60 F.) with a pycnometer or Westphal balance. The pycnometer should be standardized with distilled water at the same temperature. The Westphal balance should read 1.0000 in distilled water. If the reading differs from this, divide the reading in oil by the reading in distilled water to get the correct sp. gr. of the oil. (B) AT TEMPERATURE OF BOILING WATER. Fill a weighed sp. gr. bottle (25-50 cc.) with freshly boiled hot water. Place in boiling water and boil rapidly for thirty minutes. Replace any evaporation from the bottle by addition of boiling water. Then insert the stopper, previously heated to 100 C., remove the bottle, cool and weigh. 230 ANALYSIS OF OILS, FATS, WAXES AND SOAPS 231 Dry the bottle at 100 C., fill with the dry, hot freshly filtered fat, entirely free from air bubbles. Immerse in boiling water for thirty minutes. Insert the stopper previously heated to 100 C., cool and weigh. Divide the weight of fat by the weight of water previously found, to obtain the sp. gr. NOTES. (1) Instead of standardizing the bottle with boiling water, the following formula may be used for calculating the weight of water at 100 C. from the weight at some other temperature: Wt. at 100 C. = W-[1 +0.000026 (100-01 a WD = (1.0026-0.0000260; a where t = temperature at which water is weighed; D = density of water at 100 C.; d = density of water at t C.; and W = weight of water at t C. (2) The bottle with contents should always be weighed at room tempera- ture. Refractive Index. Place the instrument so that diffused day- light or any form of artificial light can be used for illumination. Circulate through the prisms a stream of water at constant tem- perature. For fats, the standard temperature is 40 C. ; for oils it may be 15, 20 or 25 C., depending on circumstances. (A) WITH ABBE REFRACTOMETER. Open the double prism by means of the screw head and place a few drops of oil on the prism. Let stand a few minutes to come to uniform temperature. Move the alidade on the side scale backward or forward until the field of vision is divided into light and dark portions. Rotate the screw head of the compensator until a sharp colorless line is obtained between the fields, then adjust this line so that it falls on the point of intersection of the 2 cross hairs. Read the refract- ive index directly on the scale. NOTES. (1) The correctness of the instrument should be checked by means of the quartz plate which accompanies it, using monobromonaphtha- lene. It may also be checked with distilled water, the index of refraction of which at 20 C. is 1.3330. Apply to all readings any correction found in standardization. 232 TECHNICAL METHODS OF ANALYSIS ANALYSIS OF OILS, FATS, WAXES AND SOAPS 233 ,S3 : : '5! -cow ilia "'S : CO i i C2 . i-lQOOOCO CO iO'- OO (M 00 -OOC ICO^-I -( -.co< -CO^ O ~- ^ C CO i i l^ O " 1 :S 1C to - O- (N -I 0) (M ^*< CV1 Ui Ca 5OS OiOiOi -CCc to r~ oo $80 oo > rH^t* ir^od > cc^-iiccoco . t^ . . . < OO CO Oi ill OOOQQ Q Qw = iliil!!lll Hliliikil: OCWaWWWKMW- K^nl^^WMJhJ 33 A. B. Se TECHNICAL METHODS OF ANALYSIS .C!>... D . - O CO O COCOCO CDlOO -C3 ^^i-Ht^-r^-C^I Ot^-C^IOi ^c^SS^SSSS^^Sc^SSS^^^^^^^^^ -^^ooSSSSoJ^SSSS^ '^O'-OOOO -^ . . C C^ O5 -5. C-oiO -ooiioooooog - - >^ 6w -s^ 6 ^siS ^ ^ l 6w -s^ 6 ^ III ANALYSIS OF OILS, FATS,' WAXES AND SOAPS 237 P>aoooooooSroooao?So?Sco r oa>8oooo 1 ooo'oo2~ 2 co "*' c ; ( IS . < c^ , 10 " s " 9 o 1 i *tf CO -^ --CO CO CO *O OP O 1 CN TI . . . . OO '1C IO i 1 " 1C OO 00 OO >< '"5 t~ t- .*T ' i . . . ^'"'tfi^ O '^t< -00 -O> '""'oOO OO5 II *Ci <3 I ] ( ill |13 CO M . . . . . . . . . . . . . oo i < I 3-8 ' IO CO kO * ' t^-'CO^OOO-Oi - O>- OOOOO-OO OC^IO-* Oi-^O . cO 1 O ; w 3 a M C 10 b- r?l NSOLU 1 P . . oo . CO ... . ^ 10 S C ^H "* S oo ' S oo S ^ O ^ ( -O5 -O ' * !j J9 C9(M COlOCO 11 'S.2* 1 5 ;;;:;; : : * * * 8 * fi "3 J \* G S5 < to 1 i .5 Ojo 0(8) ..J. . 1. 00 lO oo -oo Q .... II (3 H <; fe 2"" -S 6 u J < Is i-g Oi i \i\\\n\\\\ i \ \\\\ \ \ ['! : ! 1 1 fy N ^E do' n N 1 1 1 MI M n n ; i i i- n u ^i ; = 8 SSSSSSSSS SS^g^looo" 6o(2" S5 &2<3 oj ll if S5SS3SgSSS58S8S2SSSSSg8SSSS2^ 2 J2^22 u2 238 TECHNICAL METHODS OF ANALYSIS * .1 . . D@ :.H : : !S :$ : : S. c5 o . . . .g - &! it 1*1 1 GOO O CO II . . . - IM -f r- "? co O S 2 O l-*lOOO300O30O--^ ^^H^^H 'ft r* ^H -H -^--HCOt^ -H -H U7 r-H OS OO III) lOS^HOIIIIIIII ||COO&Ctlll o l ^-.4oco o-^i o oo co on -H ^ . .00 -CO -H-HCO .C^I-pOSLOI^ ; ^ OS OS ' L i'ST I'oSSIS i i'c)ii^ ' O O5 OS C5 O ^^O5 OS OS CO -4 1 ^^O O OS SOCt -H 03 03 OS OS OS I i ! i ! i ; ;g ;| :g ; i i g y :-__ y;| ;g ; i :j | j j j ANALYSIS OF OILS, FATS, WAXES AND SOAPS 239 O CO t*~ OO O3 O i ' < >r~ooo3 OH < t^OOOOi-l - : :^ ':% : .^^ < rt CM **! 's. 1" OP4 2| 11 IS s*" . Animal Oil K. Vegetable Oil E. Liv H. Blu I I " ' 13 oQ' 5 g| 1 g 1 ? SI O 10 -oo -on : :S 3 ! iljj^g !^3^3^ 5 O. D.-S -^ 3 e 2cocoa2a2cca2cccccQHE-i 240 TECHNICAL METHODS OF ANALYSIS (2) Refractive index varies directly with temperature, increasing as tem- perature falls and vice versa. Temperature correction may be made as follows : Let Ri = reading at h, and Rz reading at h ; then #1 = # 2 +0.000365 (fc-fc). The decimal in the formula represents the change in refractive index for each degree C. (B) BY ZEISS BUTYRO-REFRACTOMETER. Place 2 or 3 drops of filtered fat on the surface of the lower prisms. Close the prisms and adjust the mirror until it gives the sharpest reading. If the reading is indistinct after running water at constant temperature through the instrument for some time, the fat is unevenly dis- tributed on the surfaces of the prisms. The instrument should be carefully adjusted by means of a standard fluid supplied with it and the temperature kept constant during use. Convert degrees butyro to refractive indices from the following table: BUTYRO-REFRACTOMETER READINGS AND INDICES OF REFRACTION Reading Index Reading Index Reading Index Reading Index 40.0 1.4524 50.0 1.4593 60.0 .4659 70.0 1.4723 40.5 1.4527 50.5 1.4596 60.5 .4662 70.5 1.4726 41.0 1.4531 51.0 1.4600 61.0 .4665 71.0 1.4729 41.5 1.4534 51.5 1.4603 61.5 .4668 71.5 1.4732 42.0 1.4538 52.0 1.4607 62.0 .4672 72.0 1.4735 42.5 1.4541 52.5 1.4610 62.5 .4675 72.5 1.4738 43.0 1.4545 53.0 1.4613 63.0 .4678 73.0 1.4741 43.5 1.4548 53.5 1.4616 63.5 .4681 73.5 1.4744 44.0 1.4552 54.0 .4619 64.0 .4685 74.0 1.4747 44.5 1.4555 54.5 .4623 64.5 .4688 74.5 1.4750 45.0 .4558 55.0 .4626 65.0 .4691 75.0 1.4753 45.5 .4562 55.5 .4629 65.5 .4694 75.5 1.4756 46.0 .4565 56.0 .4633 66.0 1.4697 76.0 1.4759 46.5 .4569 56.5 .4636 66.5 1.4700 76.5 1.4762 47.0 .4572 57.0 .4639 67.0 1.4704 77.0 1.4765 47.5 .4576 57.5 .4642 67.5 1.4707 77.5 1.4768 48.0 .4579 58.0 .4646 68.0 1.4710 78.0 1.4771 48.5 .4583 58.5 .4649 68.5 1.4713 78.5 1.4774 49.0 .4586 59.0 .4652 69.0 1.4717 79.0 1.4777 49.5 .4590 59.5 .4656 69.5 1.4720 79.5 1.4780 ANALYSIS OF OILS, FATS, WAXES AND SOAPS 241 Saponification Number (Koettstorfer Number). Saponify about 5 grams (accurately weighed) under a reflux condenser for one hour or longer with 50 cc. of 0.5 N alcoholic KOH in a 250- 300 cc. Erlenmeyer flask. Run a " blank " on 50 cc. of the 0.5 N alcoholic KOH under the same conditions. Cool and titrate with 0.5 N HC1 and phenolphthalein. Always make duplicate deter- minations. Subtract the number of cc. of acid used to neu- tralize the excess of alkali after saponification from the number of cc. required by the " blank." Multiply the result by 28.06 and divide by the weight of sample to obtain the saponification number, which represents milligrams of KOH consumed by 1 gram of oil. Alcoholic KOH Solution. Dissolve 29 grams of pure KOH in 1 liter of 95% alcohol by volume. This alcohol should be re- distilled from NaOH, over which it has been standing for some time, or boiled for a short time under a reflux condenser with stick NaOH and then distilled. Mix thoroughly and let stand until all carbonate has settled out. Pour off the clear solution for use. , Unsaponifiable Matter. See page 261. Iodine Number. Determine the iodine number by the Wijs method. Weigh 0.2-0.25 gram of oil into a wide-mouthed, glass-stoppered bottle. (For drying oils, such as linseed, use between 0.15 and 0.20 gram.) Add 10 cc. of CHCls. Then add from a pipette 25 cc. of Wijs solution; let stand one hour; add 40 cc. of 10% KI solution, and titrate immediately with 0.1 N thiosulfate, running in rapidly at first until the color begins to fade, then adding starch solution and titrating more slowly until the blue color disappears. The bottle should be stoppered and shaken during titration to make sure that all excess of iodine is removed from the CHCls. Two "blanks" should also be run with each determination, using 10 cc. of CHCls, 25 cc. of Wijs solution, and titrating at the end of one hour exactly as above. Care should be taken to have the temperature the same at the end of the determination as at the beginning. Subtract the number of cc. of thiosulfate required in the oil titration from the average number of cc. required by the 2 " blanks," multiply by 1.269 and divide by the weight of oil taken. The iodine number represents the number of centigrams of iodine absorbed by 1 gram of oil, i.e., the percentage of iodine absorbed. 242 TECHNICAL METHODS OF ANALYSIS Wijs Solution. (I) Dissolve separately 7.9 grams of iodine trichloride and 8.7 grams of iodine in glacial acetic acid on a water bath, taking particular care that the solutions do not absorb water; then pour both solutions into a liter volumetric flask, rinsing the containers with glacial acetic acid; make up to the mark with glacial acetic acid and mix thoroughly. Or, (2) Dissolve 13.0 grams of resublimed iodine in 1 liter of c. p. glacial acetic acid. Titrate a portion with 0.1 N thiosulfate. Set aside about 25 cc. Into the remainder pass washed and dried chlorine gas until the original thiosulfate titration is just doubled. Then add the small portion of original solution to neutralize any free chlorine. Preserve in amber-colored glass- stoppered bottles. NOTES. (1) It is very necessary that the chlorine gas be passed through cone. H 2 SO 4 to dry it, as moisture spoils the solution. (2) Wijs solution should not be used after it is more than one month old. Acid Number (Total Fatty Acids) . Weigh 20 grams of sample into a 300 cc. Erlenmeyer flask. Add 50 cc. of 95% alcohol pre- viously neutralized with 0.1 NaOH and phenolphthalein. Heat to the boiling point on the steam bath, agitate the flask thoroughly, and titrate with 0.1 N KOH (or NaOH), shaking thoroughly until the pink color persists. The add number is expressed as milli- grams of KOH required per 1 gram of oil. CALCULATION. 1 cc. 0.1 N KOH = 5.61 mg. KOH. In case the percentage of fatty acids is desired, use the follow- ing factor: 1 cc. 0.1 N KOH = 0.02824 gram oleic acid. It may be noted that the percentage of oleic acid X 1.99 = acid number. Soluble Fatty Acids. This determination is made on the solu- tion after titrating for the saponification number. Place the flask on the water bath and evaporate off the alcohol. Add suf- ficient 0.5 N HC1 so that its volume, plus the amount used in titrating for the saponification number, will be 1 cc. in excess of the amount required to neutralize the 50 cc. of alcoholic KOH added. Place on the steam bath until fatty acids separate into a clear layer. Fill to the neck with hot water and cool in ice water until the cake of fatty acids is thoroughly hardened. Pour the liquid contents of the flask through a filter into a liter flask. ANALYSIS OF OILS, FATS, WAXES AND SOAPS 243 Fill the flask again with hot water, and set on the steam bath until the fatty acids collect on the surface. Cool by immersion in ice water and again filter the liquid into the liter flask. Repeat this treatment with hot water 3 times, cooling and collecting the washings in the liter flask after each treatment. Titrate the com- bined washings with 0.1 N NaOH and phenolphthalein. Sub- tract 5 (which corresponds to the excess of 1 cc. of 0.5 N HC1) from the number of cc. of 0.1 N NaOH used. The difference multiplied by 0.00881 gives the weight of soluble acids, as butyric acid. Calculate to percentage. Insoluble Fatty Acids (Hehner Number.) Drain the flask containing the cake of insolub'e fatty acids from the previous determination and the paper through which the soluble fatty acids have been filtered, and dry for twelve hours. Transfer the cake, with as much of the fatty acids as can be removed from the filter paper, to a weighed wide-mouthed beaker flask. Then place the funnel containing the filter in the neck of the flask and wash the paper thoroughly with hot absolute alcohol. Remove the fun- nel, evaporate off the alcohol, dry for two hours at 100 C., cool in a desiccator and weigh. Again dry for two hours, cool and weigh. If there is any considerable decrease, re-heat for two hours and weigh again. Calculate the final weight to percentage of insoluble fatty ac ds. Soluble Volatile Fatty Acids (Reichert-Meissl Number). This method is of particular value in the case of substances containing butter fat, which has a very high Reichert-Meissl number. As the determination is entirely empirical, it is neces- sary to follow the directions exactly. REAGENTS. (1) NaOH solution (1:1). The NaOH should be as free as possible from carbonates. Protect the solution from contact with CO2. Let it settle and use only the clear liquid. (2) KOH Solution Dissolve 100 grams of purest KOH in 58 cc. of hot water. Cool in a stoppered vessel, decant the clear solution and protect from contact with CO2. (3) 95% Alcohol by Volume. Distilled over NaOH. (4) Dilute H 2 S0 4 . Dilute 200 cc. of cone. H 2 SO 4 to 1 liter with water. (5) Barium (or Sodium) Hydroxide Solution. Standardize an approximately 0.1 N solution. 244 TECHNICAL METHODS OF ANALYSIS (6) Indicator. Dissolve 1 gram of phenolphthalein in 100 cc. of 95% alcohol. (7) Pumice Stone. Heat small pieces to white heat, plunge in water, and keep under water until used. SAPONIFICATION. Weigh 5.75 cc. (about 5 grams) of filtered sample into a 300 cc. Erlenmeyer flask;* add 10 cc. of 95% alcohol and 2 cc. of NaOH solut on and heat on the steam bath under a reflux condenser (a glass tube not less than 3 feet long may be used) until saponification is complete as shown by clear solu- tion. After saponification, remove the alcohol by evaporation on the steam bath. Avoid possible loss near the end by removing the flask and waving it back and forth in the air. Remove the last traces of alcohol by a stream of air free from CO2. DISTILLATION AND TITRATION. Dissolve the soap obtained above by adding 135 cc. of recently boiled water and warm on the water bath, with occasional shaking, until the solution is clear. Cool to 60-70 C., add 5 cc. of the dil. H 2 SO 4 , stopper loosely and heat on the water bath until the fatty acids form a clear transparent layer, which may take several hours. Cool to room temperature, add a few pieces of pumice stone and connect with a glass condenser by means of a bulb tube. Heat slowly with a free flame until ebullition begins and distill, regulating the flame so as to collect 110 cc. of distillate in as nearly thirty minutes as possible. Mix this distillate, filter through a dry filter, and titrate 100 cc. with standard barium or sodium hydroxide solution, using phenolphthalein indicator. The red color should persist for two to three minutes. Multiply the number of cc. of 0.1 N alkali used by 1.1, divide by the weight of fat taken and multiply by 5 to obtain the Reich- ert-Meissl number. Correct the result by the figure obtained in a " blank " determination. Insoluble Volatile Fatty Acids (Polenske Number). For this determination use the method described in J. Assoc. Official Agr. Chemists. Methods of Analysis (1916), page 308. Acetyl Value. Boil the sample (10-50 grams, depending on its nature) with an equal volume of acetic anhydride in an acetyliza- * The fat should be warmed if necessary to melt it. Use a warm Mohr pipette for measuring out the liquid fat into the flask, taking care to wipe off the adhering fat and to prevent any fat getting on the sides of the flask. Let come to room temperature and weigh accurately. ANALYSIS OF OILS, FATS, WAXES AND SOAPS 245 tion flask for two hours; pour the mixture into a large beaker containing 500 cc. of water and boil for thirty minutes. To prevent bumping, pass a slow current of CCb into the liquid through a finely drawn out tube reaching nearly to the bottom. Let the mixture separate into two layers, siphon off the water, and boil the oily layer with fresh water until it is no longer acid to litmus. Separate the acetylated fat from water, filter and dry at 105 C. Weigh 2-4 grams of the acetylatep! fat into a 300 cc. Erlen- meyer flask and saponify with an excess of alcoholic KOH as described under Saponification Number, measuring the alcoholic KOH solution exactly. Evaporate the alcohol after saponifica- tion and dissolve the soap in water. Then add to the soap solu- tion a quantity of standard H2S04 exactly corresponding to the amount of alcoholic KOH added; warm gently, filter off the free fatty acids which col- lect on top, wash with boiling water until the washings are no longer acid and titrate the fil- trate with 0.1 N KOH and phenolphthalein. Multiply the number of cc. of alkali by 5.61 and divide by the weight of acetylated oil used, to obtain the acetyl value. Melting Point. Determine the melting point by the closed capillary tube method. Draw the melted sample into a thin- walled capillary tube. Use a column of fat 1-2 cm. long, according to the length of the thermometer bulb. Seal one end of the tube and cool on ice twelve to fifteen hours. Attach FIG. 14. Melting Point Apparatus, the capillary tube with a rub- ber elastic to the bulb of an accurate thermometer; immerse in a large test tube of water surrounded by a beaker of water, 246 TECHNICAL METHODS OF ANALYSIS and heat very slowly. An apparatus similar to that shown in Fig. 14 may be used. Record the temperature at which the fat becomes transparent as its melting point. NOTE. In legal cases and cases of dispute use the Wiley metnod as described in J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 301. liter Test. (A) ALCOHOLIC OR AQUEOUS NAOH METHOD. (1) Standard thermometer. The thermometer must have a zero mark, 0.1 graduations between 10 and 60 C., and auxiliary reservoirs at the upper end and between the and 10 marks. The cavity in the capillary tube between the and 10 marks must be at least 1 cm. below the 10 mark, which must be about 3-4 cm. above the bulb, the total length of the thermometer being about 38 cm. The bulb should be about 3 cm. long and 6 mm. in diameter. The stem of the thermometer should be 6 mm. in diam- eter and made of the best thermometer tubing, with the scale etched on the stem, graduations clear cut and distinct. The thermometer should have been annealed for seventy-five hours at 450 C., and the bulb should be Jena normal 16" 1 glass, mod- erately thin, so that the thermometer will be quick-acting. (2) Determination. Saponify 75 grams of the sample in a metal dish with 60 cc. of 30% NaOH solution (36 Be.) and 75 cc. of 95% alcohol by volume or 120 cc. of water. Evaporate to dry- ness over a very low flame or on an iron or asbestos plate, stirring constantly. Dissolve the dry soap in 1 liter of boiling water and, if alcohol has been used, boil for forty minutes to remove it, adding sufficient water to replace that lost in boiling. Liberate the fatty acids by adding 100 cc. of 30% H 2 SO 4 (25 Be.) and boil until they form a clear, transparent layer. Wash with boiling water until free from H 2 S04, collect in a small beaker and place on the steam bath until the water has settled and the fatty acids are clear; then decant into a dry beaker, filter while hot and dry twenty minutes at 100 C. When dried, cool the fatty acids to 15- 20 C. above the expected titer and transfer to a titer tube, 25 by 100 mm. (1 by 4 inches) and made of glass about 1 mm. in thick- ness. Place in a 16-ounce wide-mouthed bottle of clear glass, 70 by 150 mm. (2.8 by 6 inches) fitted with a perforated cork so as to hold the tube rigidly when in position. Suspend a standard ANALYSIS OF OILS, FATS, WAXES AND SOAPS 247 thermometer so that it can be used as a stirrer, and stir the mass slowly until the mercury remains stationary for thirty seconds. Then let the thermometer hang quietly, with the bulb in the center of the mass, and observe the rise of the mercury column. The highest point to which it rises is regarded as the titer of the fatty acids. The titer test must be made at about 20 C. for all fats having a titer above 30 C. and at 10 C. below the titer for all others. Test the fatty acids for complete saponification as follows: Place 3 cc. in a test tube and add 15 cc. of 95% alcohol by volume. Bring the mixture to boiling and add an equal volume of dil. NH40H. A clear solution should result. (B) GLYCEROL-KOH METHOD. Heat 75 cc. of glycerol-KOH solution (25 grams KOH in 100 cc. of high-test glycerol) to 150 C. in an 800 cc. beaker; then add 50 cc. of the oil or melted fat, pre- viously filtered if necessary to remove foreign substances. Sapon- ification often takes place almost immediately, but heating with frequent stirring, should be continued for fifteen minutes, avoiding a temperature much above 150 C. When saponification is com- plete, as indicated by a perfectly homogeneous solution, pour the soap solution into an 800 cc. casserole containing about 500 cc. of nearly boiling water, add carefully 50 cc. of 30% H^SCU and heat the solution, with frequent stirring, until the layer of fatty acids separates out perfectly clear. Transfer the fatty acids to a tall separatory funnel, wash 3-4 times with boiling water to remove all mineral acids, draw the fatty acids off into a small beaker, and let stand on the steam bath until the water has settled out and the acids are clear. Filter into a dry beaker and heat to 150 C. on a thin asbestos plate, stirring continually with the thermometer; transfer to the titer tube, fill it to within 2.5 cm. of the top and determine the titer as directed above. SPECIAL TESTS Cholesterol and Phytosterol (in Mixtures of Animal and Vege- table Fats). Introduce 200-300 grams (accurately weighed, if a determination of unsaponifiable matter is desired) of melted fat into a flat-bottomed liter flask. Close the neck of the flask with a 3-hole stopper and insert through these holes: (1) a reflux con- denser; (2) a right-angled glass tube, one arm of which reaches to 248 TECHNICAL METHODS OF ANALYSIS a point 6 mm. above the surface of the melted fat, the other being closed a short distance from the flask by means of a short piece of rubber tubing and a pinch-cock; (3) a glass tube bent so that one arm reaches down to the bottom of the flask and the other serves as a delivery tube for a 700 cc. round-bottomed flask con- taining 500 cc. of 95% alcohol by volume. Place the flasks, containing the melted fat and alcohol, respect- ively, on a steam bath and heat so that the alcohol vapor passes through the melted fat in the liter flask and is condensed in a reflux condenser, finally collecting in a layer over the melted fat. After all the alcohol has passed in this manner into the flask containing the fat, disconnect the flask from which the alcohol has been distilled and attach the tube to a short piece of rubber tubing attached to a right-angled glass tube (as in (2) above) and siphon the alcohol layer back into the alcohol distillation flask. Re-connect as at first and again distill the alcohol as in the first operation. When all alcohol has been distilled, siphon it again into the distillation flask and extract in the same manner for the third time. Discard the fat and retain the alcohol which now contains practically all of the cholesterol and phytosterol originally pres- ent in the fat. Concentrate the alcoholic solution to about 250 cc. and add 20 cc. of KOH solution (1 : 1) to the boiling liquid. Boil for ten minutes to insure complete saponification of the fat, cool to room temperature and pour into a large separatory funnel containing 500 cc. of warm ether. Shake to insure thorough mix- ing and add 500 cc. of water. Rotate the funnel gently to avoid the formation of persistent emulsions, but mix the water thor- oughly with the alcohol-ether-soap solution. A clear, sharp separation takes place at once. Draw off the soap solution and wash the ether layer with 300 cc. of water, avoiding shaking. Repeat the washing of the ether solution with small quantities of water until all soap is removed. Transfer the ether layer to a flask and distill the ether until the volume of the liquid remaining in the flask is about 25 cc. Transfer this residue to a tall 50 cc. beaker and continue evaporation until all ether is driven off and the residue is perfectly dry.* * If desired, a tared beaker may be used and the weight of unsaponifiable matter determined at this point. ANALYSIS OF OILS, FATS, WAXES AND SOAPS 249 Add 3-5 cc. of acetic anhydride to the residue in the beaker, cover with a watch-glass and heat to boiling over a free flame. After boiling for a few seconds, remove the beaker from flame, cool and add 35 cc. of 60% alcohol by volume. Mix the contents of the beaker thoroughly, filter off the alcoholic solution and wash the precipitate with 60% alcohol. Dissolve the precipitate on the filter with a stream of hot 80% alcohol by volume and wash the insoluble portion well with 80% alcohol. Acetates of choles- terol and phytosterol are dissolved, while the greater portion of the impurities present (including paraffin oil if present) remains behind on the filter. Cool the combined filtrate and washings to a temperature of 10-12 C. and let stand at that temperature for two to three honrs. During this time the acetates of cholesterol and phytosterol crystallize from solution. Collect the crystals on a filter, wash with cold 80% alcohol and then dissolve them in a minimum amount of hot absolute alcohol. Collect the alcoholic solution of acetates in a small glass evaporating dish, add 2 or 3 drops of water to the solution and heat if not perfectly clear. Let the alcohol evaporate spontaneously, the contents of the dish being stirred occasionally to mix with the main body of liquid the deposit of crystals which forms upon the edges. As soon as a good deposit of crystals has formed, collect them upon a hardened filter, wash twice with cold 90% alcohol and dry by suction, drying finally at 100 C. for thirty minutes, and determine the melting point in the apparatus shown in Fig. 14, using H^SCU in the outer beaker and glycerol in the inner tube. The melting point of the first crop of crystals usually gives definite information as to the presence or absence of phytosterol but the conclusion indicated should be confirmed by re-crystallizing the crystals from absolute alcohol and again determining their melting point. If the crystals are pure cholesteryl acetate, the melting point of the second crop should agree closely with that of the first. If phytosteryl acetate is present, however, a higher melting point will be noted, as phytosteryl acetate is less soluble in alcohol than cholesteryl acetate. The melting point of choles- teryl acetate is 114 C., that of phytosteryl acetate is 125-137 C. Qualitative Test for Rosin Oil. Polarize the pure oil, or a definite dilution with petroleum ether, in a 200 mm. tube. Rosin oil has a polarization, in a 200 mm. tube, of from +30 to +40 250 TECHNICAL METHODS OF ANALYSIS on the sugar scale (Schmidt and Haensch), while most other oils read between +1 and 1. Halphen Test for Cottonseed Oil. Mix 82 containing about 1% of sulfur in solution, with an equal volume of amyl alcohol. Mix equal volumes of this reagent and the oil sample and heat in a bath of boiling, saturated brine for one to two hours. This is conveniently accomplished in an acetylization flask. In the presence of as little as 1% of cottonseed oil, a characteristic red or orange-red color is produced. Lard and lard oil from animals fed on cottonseed meal will give a faint reaction; their fatty acids also give this reaction. The depth of color is proportional, to a certain extent, to the amount of cottonseed oil present, and by making comparative tests with cottonseed oil some idea as to the amount present can be obtained. Different oils react with different intensities, and oils which have been heated from 200-210 C. react with greatly diminished intensity. Heating for ten minutes at 250 C. renders cottonseed oil incapable of giving the reaction. NOTES. (1) Blown cottonseed oil and old rancid oil cannot be identified by this test. (2) Kapok and Baoban oils also give similar color reactions. (3) A blank test should always be conducted under the same conditions on a pure sample of the oil being tested and also on a pure oil to which has been added a little cottonseed oil. Tests for Peanut Oil. (A) RENARD TEST. Weigh 20 grams of oil into an Erlenmeyer flask. Saponify with alcoholic KOH solution, neutralize exactly with dilute acetic acid and phenol- phthalein and wash into an 800-1000 cc. flask containing a boiling mixture of 100 cc. of water and 120 cc. of 20% lead acetate solution. Boil for a minute and then cool the precipitated soap by immersing the flask in water, occasionally giving it a whirling motion to cause the soap to stick to the sides of the flask. After the flask has cooled, decant the water and excess of Pb acetate solution and wash the Pb soap with cold water and then 90% alcohol by vol- ume. Add 200 cc. of ether, cork and let stand for some time until the soap is disintegrated. Heat on the water bath with a reflux condenser, and boil for five minutes. In the case of oils, most of the soap will be dissolved; with lards, which contain much stearin, part of the soap will be left undissolved. Cool the ANALYSIS OF OILS, FATS, WAXES AND SOAPS 251 ether solution of soap to 15-17 C. and let stand until all insoluble soaps have separated out (about twelve hours). Filter upon a Biichner funnel and thoroughly wash the insoluble Pb soaps with ether. Wash the ether-insoluble Pb soaps into a separatory funnel by means of a jet of ether, alternating at the end of the operation, if the soaps stick to the paper, with HC1 (1 : 3). Add sufficient HC1 (1 : 3) so that the total volume of the latter amounts to about 200 cc. and enough ether to make its total volume 150-200 cc. and shake vigorously for several minutes. Let the layers separate, run off the acid layer, and wash the ether once with 100 cc. of dil. HC1 and then with several portions of water until the water washings are no longer acid to methyl orange. If a few undecomposed lumps of Pb soap remain (indi- cated by solid particles remaining after the third washing with water), break these up by running off almost all the water layer and then add a little cone. HC1; shake and then continue washing with water as before. Distill the ether from the solution of insol- uble fatty acids and dry the latter in the flask by adding a little absolute alcohol and evaporating on the steam bath. Dissolve the dry fatty acids by warming with 100 cc. of 90% alcohol by volume and cool slowly to 15 C., shaking to aid crystalliza- tion. Let stand at 15 C. for thirty minutes. In the presence of peanut oil, crystals of arachidic acid will separate from solution. Filter and wash the precipitate twice with 10 cc. of 90% alcohol by volume, and then with 70% alcohol by volume, taking care to maintain the arachidic acid and wash solution at a definite temperature in order to apply solubility corrections given below. Dissolve the arachidic acid upon the filter with boiling absolute alcohol, evaporate to dryness in a weighed dish, dry and weigh. Add to the weight 0.0025 gram for each 10 cc. of 90% alcohol used in the crystallization and washing, if conducted at 15 C.; if conducted at 20 C., add 0.0045 gram for each 10 cc. The melting point of arachidic acid thus obtained is 71-72 C. Twenty times the weight of arachidic acid will give the approximate amount of peanut oil present. Arachidic acid has a characteristic appear- ance and may be identified microscopically. As little as 5-10% of peanut oil can be detected by this method. It is best to run a " blank " on pure peanut oil along with the test. 252 TECHNICAL METHODS OF ANALYSIS (B) BELLIER TEST. * Weigh 1 gram of the sample into a long test-tube. Add 5 cc. of alcoholic KOH solution. Boil gently over a small flame holding the tube in the hand until saponifica- tion is complete, as shown by homogeneous solution (generally three to five minutes). Add the proper amount of acetic acid (see below) to exactly neutralize the 5 cc. of alcoholic KOH. Mix well, cool rapidly in water at about 17 C. and let stand in the water for at least thirty minutes, shaking occasionally. Then add 50 cc. of 70% alcohol containing 1% by volume of cone. HC1 and again place in the water for one hour. If no peanut oil is present, a clear or opalescent liquid is formed. If more than 10% of peanut oil is present, a flocculent, crystalline precipitate remains. Even with 5% of peanut oil a distinct precipitate remains and separates on standing. SOLUTIONS. (1) Alcoholic KOH: Dissolve 8.5 grams of pure KOH in 70% alcohol, and dilute to 100 cc. with the alcohol. (2) Acetic Acid: This should be of such strength that 1.5 cc. will exactly neutralize 5 cc. of the above solution. The dilute acetic acid reagent (4 : 10) is approximately the correct strength but should be tested against the alcoholic KOH. Tests for Sesame Oil. (A) BAUDOUIN TEST. Dissolve 0.1 gram of sugar in 10 cc. of cone. HC1 in a test tube. Add 20 cc. of the sample to be tested. Shake thoroughly for one minute. and let stand. The aqueous solution separates almost immediately and in the presence of even minute quantities of sesame oil it is colored crimson. (B) VILLAVECCHIA AND FABEis TESTS. The original test as proposed by Baudouin has been modified by Villavecchia and Fabris and is usually carried out according to one of the following modifications : (1) Place 0.1 cc. of a 2% alcoholic solution of furfural in a test-tube. Add 10. cc. of the oil and 10 cc. of cone. HC1. Shake one-half minute and let settle. In the presence of even 1% of sesame oil the aqueous layer is a distinct crimson color, and in the absence of sesame oil the lower layer is either colorless or at most, in the case of rancid pure olive oils, a dirty yellow color. (2) Mix 0.1 cc. of 2% alcoholic furfural solution with 10 cc. of the oil and add 1 cc. of cone. HC1. Agitate thoroughly and add * Allen: "Commercial Organic Analysis," 4th edition, Vol. 2, 99. ANALYSIS OF OILS, FATS, WAXES AND SOAPS 253 10 cc. of CHCls. The aqueous layer will float on top and in the presence of sesame oil will be colored crimson. NOTE. We have found that this method produces a slight coloration with certain olive oils claimed to be pure. Furthermore, the oils in question gave negative results by the Bellier reaction described below. (C) BELLIER TEST.* To 50 cc. of water, add 100 cc. of cone. H2S04, cool and add 10 cc. of 40% formaldehyde solution. Mix equal parts of the oil and the above reagent by stirring. In the presence of sesame oil the emulsion slowly becomes colored a very intense, stable blue-black. With olive oil, peanut oil, cottonseed 011 and walnut oil, the emulsion produced is more or less yellow. It is stated that the per cent of sesame oil can be detected with this reaction when working with a comparatively pure olive oil. With olive oil containing 2% of sesame oil the color of the emul- sion is a rather dark gray after five or ten minutes. The acid which separates is a blackish brown. With 5% of sesame oil, the mixture becomes a very dark black gray and the acid which sep- arates is black tinged with blue. Emery Test for Beef Fat in Lard. Weigh 5 grams of melted fat into a glass-stoppered 25 cc. cylinder about 150-175 mm. tall. Add warm ether up to the 25 cc. mark, stopper securely and shake until the fat is completely dissolved. Let the cylinder stand for about eighteen hours at a temperature of 16-20 C., during which time some of the solid glycerides will crystallize out. Decant the clear solution carefully from the crystals, wash with three 5 cc. portions of cold ether, avoiding breaking up the deposit during the first two washings. Agitate the crystals with a third portion of ether and transfer to a small filter. Wash on the paper with successive small amounts of cold ether until 15-20 cc. have been used, then remove the last traces of ether by means of slight suction on the stem of the funnel. Break up any large lumps and let the deposit dry. When thoroughly dry, pulverize the glycerides and take their melting point in a closed 1 mm. tube, using an apparatus similar to that in Fig. 14 f, page 245. Heat the water in the beaker rapidly to about 55 C. and maintain that temperature until the thermometer carrying the melting-point tube registers 50-55 C., then heat * Ann. chim. anal., 1899, page 217. t Page 245. 254 TECHNICAL METHODS OF ANALYSIS again and carry the temperature of the outer bath somewhat rapidly to 67 C., and remove the lamp. The melting point of the crystals is regarded as that point where the fused substance becomes perfectly clear and transparent. A dark background placed about 4 inches from the apparatus will prove of advantage. When the melting point of the glycerides obtained by this method is below 63.4 C., the presence of beef fat should be suspected, while a melting point of 63 C. or below, can be regarded as posi- tive evidence that the sample is not pure lard. It is advisable to carry out this method with a control sample of pure lard in con- nection with each batch of samples analyzed. Fish and Marine Animal Oils in Vegetable Oils. This test is only applicable in the absence of metallic salts. Dissolve in a test tube about 6 grams of sample in 12 cc. of a mixture of equal parts of CHCls and glacial acetic acid. Add bromine drop by drop, until a slight excess is indicated by the color, keeping the solution at about 20 C. Let stand fifteen minutes or more and then place the test tube in boiling water. If vegetable oils only are present, the solution will become per- fectly clear, while fish oils will remain cloudy or contain a precip- itate due to the presence of insoluble bromides. REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), pages 299-315. LUBRICATING OILS Gravity. The gravity of lubricating oils is generally deter- mined as degrees Baume. Take the gravity by means of a Baume hydrometer if possible. If the oil is too light for this, use a Westphal balance. In the case of thick oils, use a hydrometer and let it remain in the oil at least one-half hour, so that it will sink as far as possible. Use a cylinder sufficiently large so that the hydrometer does not touch the sides, and also use a sufficient volume of oil so that the hydrometer does not come nearer than within 0.5 inch of the bottom when at rest. Take the reading of the hydrometer at the point where the lower meniscus of the oil touches the scale. If the oil is not too thick, bring it to exactly 60 F. before reading the hydrometer or balance; otherwise determine the temperature of the oil at the time of reading the ANALYSIS OF OILS, FATS, WAXES AND SOAPS 255 hydrometer and correct the reading to 60 F. (15.5 C.). Taglia- bue's " Manual for Inspectors of Coal Oil " gives readings at 60 F. for any gravity from 20-100 Baume between 20 F. and 109 F. In using the Westphal balance, be sure that the plummet is completely submerged and does not touch the sides or bottom of the cylinder containing the oil. Have the temperature of the oil at 60 F. From the sp. gr. found by the balance the gravity Baume may be calculated from the formula: 141 5 Baume = - 131.5.* sp. gr. NOTE. The Westphal balance should be leveled so that the plummet in air just balances the arm at zero. If the weights are accurate the reading in distilled H 2 O at 60 F. should be 1.0000. In case the reading in H 2 O is differ- ent from this, correct the observed reading on the sample by dividing it by the reading of the balance in water. Flash Point. For lubricating oils use the Cleveland open-cup tester. The apparatus consists of a spun brass or cast-iron cup, 1.38 inches high by 2.5 inches in diameter, in ah air bath heated from below by a Tirrill burner. Fill the cup with oil to within about j inch of the top. Use an accurate Fahrenheit thermometer scaled for l^inch immersion. Immerse the thermometer in the center of the oil at sufficient depth so that the surface of the oil is at the 1-inch mark. The rate of heating should be 10 F. per minute. At intervals of about 2, when approaching the flash point, sweep a tiny flame not more than 0.5 inch long slowly and steadily across the cup at a distance of about f inch from the surface of the oil. Take as the flash point the first point at which a puff of blue flame appears and runs round the cup and then goes out. Fire Points Continue heating after the flash point has been determined, applying the test flame at intervals of 2. Take as the fire point the point at which the oil will just take fire and con- tinue to burn. This varies from 15-80 higher than the flash point, depending upon the nature of the oil. * This is the formula used in making Tagliabue hydrometers. The U. S. Bureau of Standards employs the formula : 140 Baume = 130. sp. gr. 256 TECHNICAL METHODS OF ANALYSIS NOTE. It is important that the rate of heating should be even and the test flame should not be played directly upon the surface of the oil, since that will cause local superheating. The apparatus must be carefully protected from drafts. Cloud Test. The cloud test indicates the temperature at which solid matter crystallizes out. Cut off the neck and shoul- der of a 4-ounce oil bottle and fill one-quarter full of the oil, or sufficient to reach 0.25 inch above the bulb of the thermometer. The thermometer used is the so-called " cloud test thermometer," especially made for testing oils, with a bulb J-f inches long. Insert the thermometer through a perforated cork so that it is held centrally in the bottle with the lower end of the bulb 0.5 inch from the bottom. Then place the whole in a metal or glass jacket 4-5 inches high, having an inside diameter 0.5 inch larger than the outside diameter of the oil bottle. Place a disc of felt, cork, or wax 0.25 inch thick in the bottom of the jacket. The inner bottle must not touch the sides of the jacket at any point. Then place the whole apparatus in a freezing mixture and at every drop in temperature of 2 F., when near the expected cloud test, remove the oil bottle from the jacket and inspect it, taking care not to disturb the oil by removing the thermometer or otherwise. When the lower half of the sample becomes opaque through chilling, read the thermometer. This reading shall be taken as the cloud test of the sample. NOTE. If the oil contains any water, misleading results will be obtained. It should, therefore, be heated momentarily to 150 C. and then cooled imme- diately before making the test. Pour Test. The " pour test " indicates the temperature at which the oil will just flow. In making this test the same bottle and quantity of oil are used as for the cloud test and it may be made immediately after the cloud test. In practically all cases the cloud test is higher than the pour test. Place the sample in the oil bottle in the jacket as previously described and put the whole in a freezing mixture. At each drop in temperature of 5 F. remove the bottle from the jacket, and incline until the oil begins to flow. Do not tilt the bottle any more than necessary to determine if the oil will flow. Finally the bottle should be held at horizontal. When the oil has become ANALYSIS OF OILS, FATS, WAXES AND SOAPS 257 solid about the thermometer and will not flow, the previous 5 point shall be taken as the pour test of the sample. NOTES. (1) The cold should preferably be applied so that the pour test shall be completed in about 0.5 hour. (2) Materials used in the freezing mixture may be ice with calcium chloride crystals or salt, or solid CO 2 with acetone. For oils solidifying above 35 F. use cracked ice; from +35 to +15 use finely cracked ice with about 5% its volume of salt; from +15 to 5 use one part salt, 3 parts ice. The salt should be very dry and fine enough to pass a 20-mesh sieve. It is possible to reach 25 F. with a mixture of ice and calcium chloride crystals, but for temperatures below 5 it is better to use solid CO 2 and acetone as follows: Take a sufficient amount of dry acetone and put it into a covered copper or Pyrex beaker. Place the beaker in an ice-salt mixture, and when cooled to +10 F. or lower, add solid CO 2 , little by little, until the desired temperature is reached. Solid CO 2 is obtained by inverting an ordinary liquefied CO 2 cylinder; open the valve carefully and let the gas run out into a chamois or canvas bag. Temperatures of 80 F. may be reached by this method. Cold Test. Use the oil cylinder as described in the Cloud Test. Add to it 1 ounce of oil (28 cc.). Place the oil in the bottle directly in the freezing mixture (see note (2) above) and freeze solid with a cold test thermometer immersed in it. When the oil has become solid throughout, let stand one-half hour.* Remove the bottle from the freezing mixture. Stir the oil thoroughly, as soon as it has become soft enough to permit, grasping the bottle in such a way that the heat of the hand does not warm the oil. Continue stirring, tipping the bottle up at frequent intervals to an angle of 45 below the horizontal, and note the temperature at which the oil will just run from one end of the bottle to the other. This point is taken as the cold test. NOTES. (1) To get comparative results the above directions should be followed exactly, and even in this case there is a considerable personal factor entering into the determination. (2) The 4-ounce oil bottles used should be cut off as near the neck as pos- sible, each one having a mark on the side indicating how much oil is to be taken for the cold test. Determine the position of this mark by placing 28 cc. of water in the bottle and making a mark at the upper surface of the water. Be sure the bottle is on a perfectly level place when the mark is made. * The oil should be cooled to at least 10 below its cold test. If, therefore, the cold test is first found to be near the temperature to which the oil has been cooled for 0.5 hour, repeat the test, first cooling the oil 10 lower for 0.5 hour. 258 TECHNICAL METHODS OF ANALYSIS Viscosity. Viscosity determinations are to be made with the Saybolt Universal Viscosimeter. The temperature is generally specified for different grades of oils. Unless otherwise requested, the following temperatures are to be employed: 212 F. Calender, Crane, Crank Case, Cylinder and Valve Oils. 130 F. Black (Dark Lubricating Oil), Air Compressor and Journal Oils. 100 F. Automobile, Crusher, Cutting, Dynamo, Engine, Froth, Governor, Loom, Machinery, Shafting, Sperm, Spindle, Transformer, Turbine, and Whale Oils. MANIPULATION. Have the viscosimeter level. Bring the water bath of the viscosimeter to the required temperature. Strain the oil through muslin into a tin cup and heat in the cup to the required temperature, stirring with the thermometer. Clean out the tube of the instrument with some of the strained oil to be tested, using the plunger. Place the cork in the lower outlet coupling tube just far enough to make it air-tight but not far enough to nearly touch the small outlet jet of -the tube proper. Between J and \ inch should be enough. Pour the heated oil from the tin cup through the strainer into the tube of the viscosimeter until it overflows into the oil cup up to and above the upper edge of the tube proper. Again note that the water bath is at the proper temperature. Stir the oil in the upper tube with the thermometer until it is exactly at the correct temperature, then remove the thermometer and draw off from the overflow tube with a pipette all surplus oil down to and below the upper edge of the inner tube. This always insures the same starting head of oil. Place the 60 cc. flask beneath and directly in line with the outlet jet and as close to the tube as is practical to permit of room for drawing the cork. Hold a stop watch in the left hand, and with a twisting motion remove the cork quickly with the right hand, starting the watch simultaneously. Note the exact instant at which the oil rises to the 60 cc. mark in the flask, and stop the watch. The time elapsed for 60 cc. of oil to flow from the viscosimeter is its viscosity at the given tempera- ture. Express results in seconds. NOTES. (1) Before each test clean out the tube of the instrument with some of the oil to be tested, and before removing the cork note that the oil ANALYSIS OF OILS, FATS, WAXES AND SOAPS 259 and bath are at the proper temperature and that the 60 cc. flask is free from oil. (2) The Universal instrument is not intended to be used on oils at tem- peratures below 100 F. Saponifiable Oil. Ordinarily this determination is unnecessary, except on cylinder oils or other compounded oils. It should always be run in duplicate. For cylinder oils and other heavy dark-colored oils weigh out two portions of about 20 grams each into 300 cc. Erlen- meyer flasks, add to each flask 25 cc. of approximately 0.5 N alcoholic KOH, accurately delivered from a pipette, and 50 cc. of pure benzene. Into another similar clean flask add from the same pipette 25 cc. of alcoholic KOH solution and 50 cc. of benzene for a " blank." Connect the flasks to reflux condensers and boil gently with a low flame for at least sixteen hours, agitating occasionally. The agitation is important to insure mixing and prevent caking. Then remove the flasks and add to each 100 cc. of distilled water. Cool, and titrate each with 0.5 N acid and phenolphthalein until the pink color disappears permanently. From the titration required by the " blank " subtract the titra- tions required by each of the other solutions, respectively. The difference represents the amount of KOH absorbed by the saponi- fiable oil. Report the saponification value (milligrams of KOH per gram of oil) and also the per cent of saponifiable oil. CALCULATION. 1 cc. 0.5 N acid = 28.06 mg. KOH. = 0.1439 gram tallow oil. NOTES. (1) The above details are for heavy cylinder oils containing up to 10% of tallow oil. Lighter oils do not need to saponify so long, and the ben- zene may be omitted. The blank in this case should have benzene omitted. For a light cutting oil one or two hours is sufficient. For oils containing more than 10% saponifiable use less than 20 grams for the determination. With a little experience the analyst can judge about how much oil to weigh out and how long to saponify it. (2) It is customary to calculate the saponifiable oil to tallow oil (saponifi- cation value 195) although sometimes other oils are used. With the exception of wool grease, however, their saponification numbers are approximately the same and the above calculations will show the amount present. In the case of wool grease (average saponification value 102) the following factor should be used: 1 cc. 0.5 N acid = 0.275 gram wool grease. Wool grease may be recognized by its characteristic odor on heating. If, in compounding the oil, a mixture of wool grease and some other fatty oil is 260 TECHNICAL METHODS OF ANALYSIS employed, it is impossible by analysis to determine the relative proportions, and the titration should be calculated both as wool grease and as tallow oil. The actual amount present will lie between these figures. (3) The saponification may be conducted in pressure flasks instead of by the above procedure. In this case weigh the sample into the pressure flask, add the alcoholic potash (but not benzene), clamp tightly and place the flasks in a steam or hot water bath. Eight hours is sufficiently long for a cylinder oil. A blank should be run in a separate pressure flask. (4) In the case of oils containing wool grease the saponification should always be conducted in a pressure flask to obtain accurate results, as by the ordinary method it is very difficult to completely saponify the wool grease. Residue Insoluble in Gasoline. Unless otherwise specified carry out this test on cylinder oils as follows: Shake 5 cc. of the oil with 100 cc. of ordinary gasoline and let stand for one hour. There should be no deposit or precipitation of tarry or other foreign matter. The quantitative estimation is carried out as follows: Weigh out approximately 5 grams of oil into a small beaker. Transfer by means of 86 naphtha to a 100 cc. graduated cylinder. Fill to the mark with 86 naphtha. Agitate several times until oil and naphtha are thoroughly mixed. Stopper and let stand for whatever time is specified (generally one hour). Filter through a filter paper which has been dried at 100 C. for two hours, cooled and weighed. Wash the cylinder out with 86 naphtha and pour the washings through the filter paper. Wash the paper thoroughly with the naphtha to remove all traces of oil. Dry in the air and then at 100 C., cool in a desiccator and weigh. (If desired it may then be ignited and weighed to determine the amount of mineral matter, dirt, etc.) NOTE. 86 naphtha is highly inflammable and must not be used in the same room with any flame. Mineral Acid. Weigh 25-50 grams of oil into a 500 cc. sepa- ratory funnel, add 300 cc. of hot distilled water, recently boiled, and shake thoroughly. Titrate the water while still hot with 0. 1 N caustic and phenolphthalein. Run a blank on the same amount of hot water and subtract the blank titration from the previous. Calculate the difference to H2SO4. CALCULATION. 1 cc. of 0.1 N caustic = 0.0049 gram H2SO4. REFERENCES. Gill: "Oil Analysis." Lewkowitsch: "Chemical Tech- nology and Analysis of Oils, Fats and Waxes," Vol. 1. American Society for Testing Materials, Tentative Standards, 1917. ANALYSIS OF OILS, FATS, WAXES AND SOAPS 261 UNSAPONIFIABLE MATTER IN OILS General. The determination of unsaponifiable matter in oils falls into 2 general classes, (1) mineral oil in mixtures of mineral and fatty oils, (2) the natural unsaponifiable matter occurring in animal and vegetable oils. The method which would give satisfactory results for the first class is hardly sufficiently accurate to be applied to the determina- tion of the small amount of unsaponifiable matter in the second class. Of the two methods given below the first is applicable in all cases except where a very accurate determination of the un- saponifiable matter naturally occurring in animal or vegetable oils is desired. Method 1. Weigh out 3-5 grams of the sample into a 300 cc. Erlenmeyer flask, add 50 cc. of 0.5 N alcoholic KOH and saponify under a reflux condenser for several hours. Add a few drops of phenolphthalein to make sure that there is an excess of alcoholic KOH. In case there is not an excess, add 25 cc. more and resaponify. Transfer the contents of the flask to a beaker and evaporate off the bulk of the alcohol on the steam bath. Dissolve the residual soap in about 50 cc. of hot water and transfer to a separatory funnel, using about 20-30 cc. of water for rinsing the beaker. The total volume should be kept below 100 cc. to minimize the tendency to form emulsions. Cool, and add 50 cc. of ether and shake thoroughly. If the solution does not separate well, add a few cc. of alcohol. Run off the lower soap solution into another separatory funnel and shake this out with a* fresh portion of 50 cc. of ether. Repestt the process once again. Com- bine the 3 ethereal extracts and wash twice with about 15 cc. of water to remove any dissolved soap. Transfer the washed extracts to a weighed Soxhlet flask, distill off the ether on the water bath and dry the residue at 100 C. to constant weight. NOTES. (1) If it is suspected that the unsaponifiable matter is contam- inated with soap, dissolve it in ether. Transfer to a porcelain or platinum dish, evaporate off the ether and ash the residue. Then titrate the ash with 0.1 N acid and methyl orange and calculate the amount of soap present. CALCULATION. 1 cc. 0.1 N acid = 0.0320 gram potash soap. (2) Troublesome emulsions sometimes form during extraction. In such cases it will be found convenient to add a little alcohol or glycerol after shaking, 262 TECHNICAL METHODS OF ANALYSIS and then impart a slight rotary movement to the separately funnel without, however, agitating it. In other cases the addition of a little NaOH will break up the emulsion. Sometimes a flocculent layer will appear between the ether solution and the solvent. This, however, does not interfere with the correct determination of the unsaponifiable matter and should be drawn off and con- sidered as soap. Another method of breaking emulsions is to apply suction. Insert a one- hole stopper with a glass tube in the neck of the funnel and apply suction gently, increasing slowly until the ether boils. Care should be taken not to boil the ether too violently and to avoid sucking back when the suction is released. (3) 86 petroleum ether may be used in place of ordinary ether, but, on account of the danger of high boiling non-volatile constituents, it is not rec- ommended. In certain cases, however, notably with tallow, petroleum ether fails to extract all the unsaponifiable matter. Method 2 (Boemer). To 100 grams of oil in a 1000-1500 cc. Erlenmeyer flask add 60 cc. of an aqueous solution of KOH (200 grams dissolved in water and made up to 300 cc.) and 140 cc. of 95% alcohol. Connect with a reflux condenser and heat on the water bath, shaking at first until the liquid becomes clear. Then heat for one hour with occasional shaking. Transfer while yet warm to a 2000 cc. separately funnel, to which some water has been added, and wash out the Erlenmeyer flask with water, using in all 600 cc. Cool) add 800 cc. of ether and shake vigorously for one minute. In a few minutes the ether solution separates per- fectly clear. Draw off the soap and filter the ether (to remove the last traces of soap) into a large Erlenmeyer flask and distill off the ether, adding, if necessary, one or two pieces of pumice stone. Shake the soap solution three times with 400 cc. of ether, which add to the first ether extract. To the residue left after distilling the ether add 3 cc. of the above KOH somtion, and 7 cc. of 95% alcohol, and heat under a reflux condenser for ten minutes on the water bath. Transfer to a small separatory funnel, using 20-30 cc. of water and, after cooling, shake out with 2 portions of 100 cc. of ether. Wash the ether 3 times with 10 cc. of water. After draw- ing off the last of the water, filter the ethereal solution so as to remove the last drops of water, distill off the ether, and dry the residue in a water oven to constant weight at 100 C. REFERENCES. Lewkowitsch: '' Chemical Technology and Analysis of Oils, Fats and Waxes," Vol. 1, pages 364-367; Boemer-Ubbelohde : " Hand- buch der Ole u. Fette," pages 261-2. ANALYSIS OF OILS, FATS, WAXES AND SOAPS 263 CASTOR OIL General. Pure castor oil is nearly colorless or pale greenish, very viscous and of high sp. gr. It has also a very high acetyl value, and is strongly dextrorotary. Its saponification value is comparatively low. It is miscible in all proportions with glacial acetic acid and with absolute alcohol. The " constants " of a pure oil should come within the following limits: Sp. gr. at 15.5 C 0.958-0.970 Saponification number 176-187 Iodine number . 81-91 Acetyl value about 150 Free fatty acid about 1% Analysis. The procedures for the determination of each of the above constants are given on pages 230-244. Solubility Tests. (a) In glacial acetic acid. (6) In absolute alcohol. Mix in 2 test tubes equal parts of the sample with each of the above reagents, respectively. In each case the oil should dissolve completely and show no turbidity. NOTE. Castor oil is generally not adulterated. Adulteration with blown oils (linseed, rape, cottonseed, etc.) would be detected by turbidity in the solubility tests. In the- case of suspected addition of rosin oil, determine the unsaponifiable matter, which in pure castor oil is less than 1 %. All adul- terations would lower the acetyl value. SULFONATED OILS (TURKEY RED OILS) General. Sulfonated oils are made by treating various oils with H2SO4. The excess of acid is then neutralized with alkali and a certain amount of water added. Formerly castor oil was supposed to give a better sulfonated oil than any other but this is now open to dispute, as many other oils have been sulfonated and the resulting product is satisfactory for many purposes. Among them may be mentioned olive oil, maize oil, cottonseed oil, lard oil, and rosin oil. Sulfonated rosin oil especially has found considerable use in cutting oils. It is to be noted that the amount 264 TECHNICAL METHODS OF ANALYSIS of oil in commercial sulfonated oils is generally considerably over- stated. A so-called " 50% oil " will not generally contain more than 40% of oil, and a 70% oil from 60-65% of oil, etc. Moisture. Weigh out 30-40 grams and determine the moisture by the Xylol method as described on page 271. Ash. Weigh any convenient quantity into a dish or cru- cible, ignite gently, allowing the oil to burn, and finally heat until all the carbon is consumed. Cool in a desiccator and weigh. Unsaponifiable Matter (Mineral Oil, etc.). Weigh approx- imately 10 grams into a 300 cc. Erlenmeyer flask. Add 5 cc. of KOH solution (50 grams of KOH dissolved in H 2 and diluted to 100 cc.), then 45 cc. of ethyl alcohol and a few glass beads. Boil for one hour with a reflux condenser. Add 100 cc. of water and cool, transfer to a separatory funnel and shake out at least three times with petroleum ether (naphtha), using 50 cc. each time. Wash the ether layer at least three times with 50 cc. of water con- taining 10 cc. of ethyl alcohol. If emulsions are formed, add a little alcohol to break them. Finally evaporate the petroleum ether extract in a tared vessel, dry, cool and weigh. NOTE. The petroleum ether should have a boiling point of 40-75 C. and should leave no residue when evaporated on the steam bath. During the saponification, if the contents of the flask bump violently, turn out the flame and let the solution cool, and then add 25 cc. of the petroleum ether and con- tinue the boiling. Free Sulfur Trioxide. Weigh out approximately 4 grams into a separatory funnel, add sufficient petroleum ether to dissolve it and shake out several times with 25 cc. portions of a cone. NaCl solution, free from sulfates. Combine the washings, dilute, filter and determine the SOs in the filtrate (salt solution) in the usual way with BaCl2, finally weighing as BaSCU. Calculate to S0 3 . CALCULATION. BaS0 4 X 0.3430 = SO 3 . Combined Sulfur Trioxide. Weigh approximately 4 grams into an Erlenmeyer flask and boil forty minutes with 30 cc. of HC1 (1 : 5). Shake frequently, cool, transfer to a separatory funnel and shake out with petroleum ether. Draw off the aqueous layer and wash the ethereal layer with water. Combine the washings with the main aqueous portions, let the petroleum ether evaporate spontaneously until the odor is practically gone ANALYSIS OF OILS, FATS, WAXES AND SOAPS 265 and then heat on a steam bath. Finally transfer to a flame, heat to boiling, add 10 cc. of hot 10% BaCb solution, boil one- half hour, filter out the BaSO4, ignite and weigh. Calculate to 80s. This gives the amount of total SOs. To obtain the com- bined SOa, subtract the free SOa as previously determined. Total Fatty Oil. Calculate the total fatty oil by subtracting from 100% the sum of the moisture, ash and unsaponifiable matter. If an actual determination of the total fatty oil is desired, save the soap solution left after extracting with petroleum ether in the determination of unsaponifiable matter. Add an excess of HC1 to it and shake out with ether three times. Wash the ether extracts with H^O, evaporate the ether in a tared flask, dry and weigh as fatty acids. NOTES. (1) Report the results only to one decimal place. (2) This is the Provisional Method of the American Leather Chemists' Association. REFERENCE. Journal of American Leather Chemists' Association 14, 668 (1919). LARD OIL General. Lard oils are generally graded as follows: 1. Prime Lard Oil. Is prepared from prime steam lard and is light straw colored. Also known as Extra Winter Strained. Acidity is low; railroad specifications usually allow 2% oleic acid. 2. Pure Lard Oil. Also a light-colored oil made from No. 1 lard and white grease. 3. Extra No. 1 Lard Oil. Somewhat darker oil than the above; made from light yellow grease. Specifications generally require not over 10% oleic acid. 4. No. 1 Lard Oil. Deep yellow; made from yellow grease. This grade varies considerably; the oleic acid may run 15% or less, or may go as high as 20%. 5. No. 2 Lard Oil. Brown colored, prepared from brown and gut greases. 6. Crackling Lard Oil. Cheapest grade and made from crack- ling grease. 266 TECHNICAL METHODS OF ANALYSIS For the " constants " of pure oil, see page 234. Specific Gravity. See page 230. Saponification Number. See page 241. Iodine Number. See page 241. Free Fatty Acid. See page 242. Cold Test. When it is desired to deterine the cold test use the procedure described on page 257. Some specifications call for a cold test below 45 F. between October 1st and May 1st. OLIVE OIL General. Olive oil is obtained from the fruit of the olive tree and its quality depends upon the care with which it is prepared. The best grade of edible oil is known as " Virgin Oil." Lower grade oils are frequently known as " Salad Oil "or " Ordinary Table Oil." A still lower grade is used for soap-making. The color varies considerably, commercial oils being of shades from colorless to yellow and the lower grades generally quite green due to dis- solved chlorophyll. The " constants " vary somewhat with the grade of oil but generally are within the following limits : Sp. gr. at 15.5 C 0.914-0.920* Saponification number 185-200 Iodine number 77- 93 Acetyl value 5-30 Unsaponifiable matter . 5-1 . 5% The iodine number of high-grade oils should be generally between 81 and 85 and the free fatty acid less than 1%. Several tests have been recommended for detection of mixtures of other oils with olive oil. Some of these tests depend upon the production of a characteristic color. Such tests, however, should be used with caution and always should be accompanied by a blank test on olive oil of known purity and also on pure olive oil con- taining varying amounts of the suspected foreign oil. Specific Gravity, Saponification Number, Iodine Number, and Free Fatty Acid. See pages 230-242. Elaidin Test. Place in a test tube 10 grams of the oil, 5 grams of cone. HNOs and 1 gram of mercury, and dissolve the latter ANALYSIS OF OILS, FATS, WAXES AND SOAPS 267 by shaking continuously three minutes. Let the mixture stand for twenty minutes and again shake for one minute. The behavior of different oils after that time is recorded in the following table: Kind of Oil. Consistence. Olive Solidified after 60 minutes Arachis (Peanut) Solidified after 80 minutes Sheep's foot Solidified after 120 minutes Sesame Solidified after 185 minutes Colza (Rape) Solidified after 185 minutes Linseed Forms red dough-like scum Cod liver Becomes doughy, red and forms scum Whale Remains unchanged Hemp seed Remains unchanged NOTES. (1) The test must be made at a temperature not lower than 25 C. and the temperature must be uniform throughout the experiment. (2) The length of time required for solidification is of far greater importance than the ultimate consistency of elaidin formed. (3) The test cannot be made quantitative. Also the age of the oil and the length of time it has been exposed to air and light have an important bear- ing on the results. It is necessary to carry out the test side by side with an oil of known purity under exactly similar conditions. Maumene Test. Weigh accurately 50 grams of the oil into a 250 cc. beaker. Have ready a bottle of cone. H2S04, the exact strength of which has been determined by titration. Place the bottle of acid and the beaker of oil in a large vessel of water until both have acquired the same temperature, which should be about 20 C. Remove the beaker of oil, wipe the outside and place in a " nest " of cardboard having hollow sides stuffed with cotton wool, or in a large beaker lined with cotton wadding. Immerse the bulb of a centigrade thermometer in the oil and note the tem- perature. Then pipette 10 cc. of cone. H2SO4 and let it run rapidly into the oil. The time allowed for emptying of the pipette should be only one minute. During this time, stir the oil with the thermometer and continue stirring until no further rise of temper- ature is observed. The highest point is easily noticed, as the tem- perature remains constant for some little time before it begins to fall. The influence of the concentration of the acid on the result is shown in the following table: 268 TECHNICAL METHODS OF ANALYSIS Kind of Oil Rise of Temperature Observed with Acid Containing Per Cent of H 2 SO 4 97.38% 96.71% 95.72% 94.72% 93.75% 92.73% 91.85% Olive, genuine. . . . 42.25 43.25 42 39 36.5 34.50 31 28.00 29.25 Rape, genuine. . . . 62 63 61 58 54 50.25 47 40.5 43.0 Olive, impure .... 48.5 47.0 47.5 43.75 44.25 40.25 40.75 38.5 39.0 35.5 32.5 Color Tests. Halphen Test for Cottonseed Oil, see page 250. Baudoin Test for Sesame Oil, see page 252. Villavecchia Test for Sesame Oil, see page 252. Bellier Test for Sesame Oil, see page 253. Probable Adulterants. Cottonseed Oil. Shown by high iodine value (100-117), high Maumene figure (variously given from 61-84), and positive Halphen Test. Peanut Oil. Shown by high iodine value (96-109) and isola- tion of arachidic acid by Renard Test (see page 250). Rape Oil. Shown by high iodine value (94-105) and low saponification value (167-179). Sesame Oil. Shown by Baudoin Test. Poppy-seed Oil. Shown by high iodine value (134-142) and high Maumene figure (74-78). NOTE. Olive oil is characterized by low Maumene value and iodine num- bers and by solid elaidin. REFERENCES . Gill : "Oil Analysis, ' ' 5th Edition . Lewkowitsch : ' ' Chem- ical Technology and Analysis of Oils, Fats and Waxes." Allen: " Commer- cial Organic Analysis." TALLOW General. Tallow is the fat of beef or sheep. There is, how- ever, very little pure beef or mutton tallow on the market; most tallow is a mixture of the two. We have prepared in this labora- ANALYSIS OF OILS, FATS, WAXES AND SOAPS 269 tory samples of pure mutton tallow and pure beef tallow which, on analysis, showed the following results: Mutton Beef Melting point (open tube) 50 C. 39 C Iodine number (Wijs) Saponification number .... 32 198 51 198 Free fatty acid (as oleic) 48% 1 16% Ash. None None Moisture None None Color. When it is desired to determine the color, melt some of the sample in a flat Petri dish and let cool quietly so that it will form a cake with a smooth flat surface. Measure the color with the Ives colorimeter or tintometer, specifying in the report which instrument was used. Melting Point. Determine the melting point by the open- tube method. Take a small glass tube about 20 mm. long and 1.5 mm. inside diameter, open at both ends. Partly fill the tube by pushing one end of it into the tallow. Wipe any tallow off the outside and have the contents flush with the bottom of the tube. Fasten the tube to the mercury bulb of a thermometer by means of rubber elastics and suspend the thermometer in a beaker of water. The bottom of the thermometer should be 0.5 inch above the bottom of the beaker and there should be sufficient water to com- pletely cover the tube. Heat the water slowly and note when the tallow just begins to melt and draw up from the bottom of the tube. Report this temperature as the melting point. Iodine Number. Weigh out 0.4-0.5 gram into a wide-mouth, glass-stoppered bottle, and determine the iodine number by the Wijs method, as described on page 241. NOTE. It is convenient to weigh the tallow on a small watch crystal which can then be placed directly in the bottle. Saponification Number. Pour about 2 grams of the melted fat into a weighed Erlenmeyer flask of about 300 cc. capacity. Then take the exact weight of the flask and tallow. Determine the Saponification number as on page 241. 270 TECHNICAL METHODS OF ANALYSIS Free Fatty Acid. Weigh 10 grams into an Erlenmeyer flask as above. Add 60 cc. of ethyl alcohol, previously neutralized with 0.1 N caustic and phenolphthalein. Warm for one-half hour on the steam bath with occasional shaking. Titrate with 0.1 N caustic and phenolphthalein with vigorous shaking until the pink color persists for one minute. Calculate the percentage of oleic acid. CALCULATION. 1 cc. of 0.1 N caustic = 0.0282 gram oleic acid. Moisture. Melt some of the tallow in a clean dry test tube and heat until it begins to smoke. If any appreciable amount of moisture is present the melted fat will be turbid and will crackle. In such case, determine the moisture by the Xylol Method (see page 271). Ash. Ignite 2-5 grams gently in a weighed platinum dish, cool in a desiccator and weigh the ash For accurate results do not let the fat take fire and burn. Soap. Dissolve the ash obtained above in distilled water and titrate with 0.1 N HC1 and methyl orange. Calculate the titration (if any) to sodium stearate. CALCULATION. 1 cc. 0.1 N acid =^0.0306 gram Na stearate (soap). GREASES Types. Commercial greases may be divided into the follow- ing classes: A. TALLOW TYPE. These greases are made up of tallow and more or less of an alkali soap, commonly the Na or K soaps of palm oil, mixed with a smaller amount of mineral oil. These were the principal types of lubricating grease some years ago, but to-day are less used than the greases of type B. B. SOAP-THICKENED MINERAL OIL TYPE. These are the most common journal greases to-day, and are composed of mineral oil of various grades made solid by the addition of Ca or Na soaps. The former soap is more commonly used. C. PIGMENTED GREASES. Types A or B with the addition of a mineral lubricant, usually graphite, mica or talc. D. ROSIN-OIL TYPE. These consist of rosin oil thickened by CaO, or less commonly, PbO, to which is added more or less min- ANALYSIS OF OILS, FATS, WAXES AND SOAPS 271 eral oil, either paraffin or asphalt oils being used. These are sticky, usually contain 20-30% of water, and find their chief application as gear greases, where true lubrication is not so essential as pre- vention of wearing and rattling of the gears. Some very heavy bearings are occasionally lubricated with this type of grease. Tar, pitch, graphite and such fillers as wood pulp and ground cork are often put into these gear greases. E. NON-FLUID OILS. These are thin greases stiffened to some extent with aluminium oleate or a mixture of soaps, as for instance Na and Ca soaps. F. SPECIAL GREASES, such as a mixture of wood pulp and graphite ; thin greases of any of the above types mixed with wool or cotton fibers; freak greases containing rubber, etc. Of the above types A, B, and C are the most important as lubricants. General. Note first the odor and the color. The grease should melt to a clear homogeneous fluid and the oil should not melt away from the soap. Melting Point. For a grease melting below 100 C. use an open tube with an interior diameter of 4 mm. and about 80 mm. long. Plunge this into the sample so that a plug of grease about 1 cm. long is left in the glass tube. Then attach the latter by a rubber band to an accurate thermometer, so that the grease is alongside the mercury bulb. Immerse the thermometer with the tube attached in a beaker of water so that the bottom of the grease is about 5 cm. below the surface. Heat the water slowly at the rate of about 3-4 C. per minute. When the melting point is reached, the plug, which is under a pressure of 5 cm. of water, will slide up in the tube. Moisture. This is determined by the so-called Xylol Method. Weigh out 10 grams of the grease on a balanced filter paper and place the grease and paper in a 300 cc. Erlenmeyer flask. Add to this 75 cc. of xylol which has previously been saturated with water, as follows: A convenient quantity of commercial xylol, say 500 cc., is shaken up in a separatory funnel with water and the xylol drawn off and distilled slowly from a distillating flask. From this dis- tillate a small amount of water will separate. The xylol standing above the water is poured off into a glass-stoppered bottle, with 272 TECHNICAL METHODS OF ANALYSIS a tightly fitting stopper, and preserved for use in moisture deter- minations. Connect the flask containing the grease and xylol with a condenser, which must be perfectly dry. Heat the flask gradually in a bath of cylinder oil and distill the xylol and water slowly until the xylol comes over clear, collecting the distillate in a funnel tube with a stem graduated to 0.1 cc. The bulk of the water comes over with the first 10 cc. of distillate. After the distillation is completed, wash down the condenser with water- saturated xylol and tap the funnel tube gently until any small drops of water clinging to the sides are brought down to the bottom. Read the volume of water. Each 0.1 cc. is equivalent to 1% of H2O in the grease, using a 10-gram sample. NOTE. If the mixture gives trouble from frothing, on account of the soap present, add sufficient dry fused and powdered KHSO 4 to decompose the soap before distilling. Free Fatty Acid. Dissolve or disintegrate 5-10 grams of the grease in 50 cc. of ethyl alcohol which has previously been neu- tralized with phenolphthalein and 0.1 N KOH. Digest on the water bath until the alcohol beings to boil. Titrate with 0.1 N KOH until a pink color persists after vigorous shaking. CALCULATION. 1 cc. 0.1 N KOH = 0.0282 gram oleic acid. Ash. Ignite 2 grams in a porcelain crucible, gently at first and finally at a higher heat, until the ash is as nearly white as possible. Cool in a desiccator and weigh. Soap. To the ash add a few cc. of water and a drop of methyl orange and titrate with 0.5 N HC1. Test this solution quali- tatively for Pb, Ca, K and Na (see also page 280), and then cal- culate the titration to the proper soap by using one of the following factors : 1 ce. 0.5 N HC1 = 0.161 gram K stearate. = 0.153 gram Na stearate. = 0.152 gram Ca stearate. If the grease is made up of a Pb soap, the Pb may be deter- mined by decomposing the grease by boiling with a mixture of H2SO4 and HNOs, evaporating to strong fumes of SOs, cooling, diluting with water and alcohol and weighing as PbSO^. (See page 215). In the case of a pigmented grease determine the Pb in the ash. (PbS0 4 X2.55 = Pb Stearate.) ANALYSIS OF OILS, FATS, WAXES AND SOAPS 273 If a rosin soap is present, the figures obtained by using the above factors will be somewhat low, depending upon the amount of rosin soap present, because of the higher combining weight of abietic acid; but for ordinary purposes they will be sufficiently accurate. Mineral Matter. The amount of ash other than the alkalies from the soap may be determined by the difference between the total ash and the alkali found by titration. In this case the alkali should be calculated to K 2 CO 3 , Na 2 CO 3 or CaCOs,* as the case may be. In the case of greases of type C, shake about 5 grams of the grease with 2 separate portions of ether in a glass cylinder and then centrifuge. This removes free oils and fats. Decant the ethereal solution through a filter paper which has previously been weighed. Wash twice with ether by decantation. Shake the residue with alcohol and filter through the same filter paper. Wash with alcohol. This removes alcohol-soluble soaps. Dry the paper and residue at 100 C., cool in a desiccator and weigh. This gives the total amount of pigment, f If graphite is present, ignite the residue in a weighed porcelain crucible and subtract this final residue from the total pigment for the amount of graphitic carbon. In general, however, this last step is unnecessary, as all that is desired is the total amount of graphite. The above determinations are, as a rule, all that is necessary in an examination of the ordinary types of grease. It may be desirable, however, at times to make further determinations and these are given below: Unsaponified Oil or Fat. The unsaponified oil may consist of mineral oil together with unsaponified saponifiable oil or fat. This may be determined by extracting 5-10 grams of the grease with neutral ethyl (sulfuric) ether. If it is desired to know its nature, the ether extract may be subjected to an examination for saponification number, iodine number, etc. In most cases where grease contains unsaponified saponifiable matter, it has been * The lime.will be present in the ash as CaCO 3 rather than CaO on account of the large amount of organic matter, unless ignited over a blast. t This figure includes any alcohol-insoluble soaps. If it is desired to correct for these, acidify with dil. HC1 and shake out with ether. Filter the ether solution of the combined fatty matter, wash free from acid, evaporate off the ether and weigh the fat. Calculate to the appropriate soap. 274 TECHNICAL METHODS OF ANALYSIS made by a partial saponification of the fat in question; hence the original grease may be saponified directly and the total fatty acids examined. Unsaponifiable Mineral Oil, etc. These will be found in the ether extract of the grease, together with the unsaponified saponi- fiable matter and the proportions of the two can be ascertained by a determination of the saponification number, or by shaking out with ether after saponification (see page 261). Reporting Results. Under ordinary circumstances report results as follows: 1. Type. 2. Melting Point ( F.). 3. Moisture (%). 4. Saponifiable Oil (%). 5. Mineral Oil (%). 6. Total Mineral Matter (%). 7. Soap (kind and %). 8. Pigment (% and nature). Ordinarily the mineral oil is taken " by difference " after adding together items 3, 4, 7 and 8. DEGRAS (WOOL GREASE) The term " Degras " in its present commercial sense refers to crude wool grease. It is in no way related to the older types, of a similar name, which are now called " French Degras " some- times " Sod Oil." Crude wool grease, besides the grease arising from sheep's wool, contains an indefinite amount of fatty acids recovered from the soap used in washing the wool and obtained in the same operation to which the wool grease is subjected. It is also the custom in many foreign mills to treat the wash waters, arising from the scouring of cloth or yarn with soapy water, at the same time that the wool scourings are treated. It is obvious that the grease obtained as a result of these methods will in all cases contain some free fatty acid from soap ; and in case cloth and yarn waters are mixed before treatment, there will also be a certain further amount of fatty acid. In addition there may be a neutral oil or even a mineral oil which comes from the cloth or yarn scour- ing. It is apparent, therefore, that no standard composition can be ANALYSIS OF OILS, FATS, WAXES AND SOAPS 275 ascribed to degras. The best we can do is to take an average or approximate composition. In certain types of standard degras where the oil resulting from cloth washing does not come into the mixture, repeated extraction with hot 95% alcohol and pouring off the hot alcohol solution will finally leave a residue in the bottom of the beaker insoluble in hot alcohol. This we have found to be about 25% of the weight of the degras taken. In other cases this method is of no use. The best general method is to saponify the degras with excess of alcoholic KOH in a pressure bottle for at least eight hours. After evaporating off all of the alcohol and taking up with water, shake out the unsaponifiable matter* with ether, evaporate off the ether and weigh the residue. This should constitute in the neighborhood of 35% of the degras taken and should be com- pletely soluble in hot 95% alcohol, the latter showing absence of mineral oil. BEESWAX General. The common adulterants of beeswax are rosin, paraffin, ceresin, stearic acid, spermaceti, and Japan and carnaiiba waxes; while to a less extent are used tallow, starch, sulfur, and mineral fillers. Any appreciable amount of water must also be looked upon as an adulterant. Preliminary Tests. As a preliminary test for the detection of adulterants, dissolve a portion of the sample in cold CHCls. Ceresin, paraffin wax, carnaiiba wax, and wool wax are not com- pletely soluble and any considerable quantity may thus be detected qualitatively, as well as starch and mineral matter. It should, however, be borne in mind that bleached white beeswax is not readily soluble in CHCla. Pure beeswax should also be completely souble in turpentine. Specific Gravity. Melt the wax and drop it by means of a stirring rod upon a moist, cold, porcelain surface in such a manner as to obtain globules of about | inch diameter. Let cool for at least two hours, and then place in a glass cylinder of about 200 cc. capacity, and add a mixture of alcohol and water of sp. gr. about 0.960 at 15 C. If the globules sink in this mixture, add water; if they rise, add alcohol. Add water or alcohol with rapid agitation * For the determination of unsaponifiable matter, see page 261. 276 'TECHNICAL METHODS OF ANALYSIS until the globules neither rise nor sink but remain suspended in the liquid at exactly 15.5 C. Then take the sp. gr. of the liquid at this temperature with the Westphal balance. This gives the sp. gr. of the beeswax at 15.5 C. Melting Point. Thinly coat the bulb of an accurate 'thermom- eter with the wax and let stand twenty-four hours. Place the bulb of the thermometer in a large test tube, holding it in place by a cork stopper grooved on the sides so as to allow free access of air. Immerse the test tube containing the thermometer in a beaker of water and raise the temperature very gradually (1 in 2-3 minutes). The temperature at which a transparent drop forms on the end of the thermometer is taken as the melting point. Iodine Number. Determine the iodine number by the Wijs method as described on page 241. Acid Number. Weigh 8 grams into a 500 cc. Erlenmeyer flask and add 70 cc. of neutral alcohol. Heat on the water bath until the mixture is entirely melted. Then add 1 cc. of phenolphthalein solution and titrate the mixture quickly with 0.5 N alcoholic KOH solution, keeping it hot. Calculate the number of milligrams of KOH required to neutralize the free acid in 1 gram of wax. CALCULATION. 1 cc. of 0.5 N alkali = 28.06 mg. KOH. NOTE. Instead of titrating with 0.5 N alcoholic KOH, the determination may be made with 0.5 N or 0.1 N aqueous KOH. In this case, however, it is necessary to dilute with about 200 cc. of hot neutral alcohol before titrating and to keep this solution hot during titration. In any case the wax must be in a melted state during titration. Saponification Number. Saponify about 2 grams of wax (accurately weighed) for at least three hours with 25 cc. of 0.5 N alcoholic KOH, as described on page 241. Calculate the saponi- fication number. Ester Value and Ratio Number. Subtract the acid number from the saponification number. The difference is the ester value. Divide this by the acid number. The quotient is the ratio number. For pure beeswax the ratio number lies in the neighbor- hood of 3.8. NOTE. If the saponification number of the sample is below 92 while the ratio number is normal, then paraffin or ceresin must be present. If the ratio number exceeds 3.8, then adulteration may be suspected with Japan wax, tallow, insect wax, carnaiiba wax, or spermaceti. If, however, the acid value ANALYSIS OF OILS, FATS, WAXES AND SOAPS 277 is much below 20, Japan wax is absent. If the ratio number is less than 3.8, stearic acid or rosin may be present. Moisture. Determine moisture by the Xylol Method as described on page 271. Rosin. Test qualitatively for rosin as follows: Heat 1 gram of wax for a few minutes with 1 cc. of 50% alcohol (alcohol of this strength will not extract stearic acid if present), cool and filter. Evaporate the filtrate to dryness on the water bath and add 5 cc. of acetic anhydride. Heat to boiling, cool thoroughly, and then carefully let 1 drop of cone. H 2 S04 flow into the solution. Rosin will develop a fugitive violet color. Fatty Adulterants. Japan wax and other fatty substances (such as tallow, ceresin, stearic acid, etc.) may be detected by boiling 1 gram with 1.5 grams of borax and 20 cc. of water, where- upon the aqueous solution will become milky or gelatinous when cooled. With pure beeswax it remains clear or becomes but slightly turbid. Carnatiba wax and rosin behave similarly to pure bees- wax. NOTE. The above determinations are generally sufficient to determine whether a sample is adulterated or not. It must be borne in mind, however, that beeswaxes from different localities vary somewhat in composition, also that the sp. gr. of white bleached beeswax is somewhat higher, whereas the other " constants " may be either raised or lowered according to the method employed in bleaching. Indian beeswax (Ghedda wax) is softer and more plastic than normal beeswax from European and domestic sources. It has a very low acid value, and high ester value, and it is said to be rarely adul- terated. Care must be taken not to confuse specimens of this wax (and also Chinese beeswax) with adulterated beeswax. The following table gives the approximate figures which an unadulterated sample should show: Ordinary Beeswax Indian Beeswax Sp gr at 15.5 C 0.958-0.970 0.958-0.970 Melting point 62-67 C. 60-68 C. Iodine number 8-11 5-11 Saponification value 90-104 76-130 Acid number 17-21 4.5-10 Ester value 73-87 69-123 Ra/tio number 3 6-3 8 7 4-17 9 278 TECHNICAL METHODS OF ANALYSIS In an adulterated sample the following tests will be found useful : Stearic Acid. Boil 3 grams of wax for several minutes with 10 cc. of 80% alcohol and let cool to 18-20 C. to form a thick paste. After standing one hour, filter into a 200 cc. cylinder and dilute the filtrate with water to about 200 cc. If stearic acid is present, it separates into flakes and collects on the surface. The test is sensitive to 1%. If from 7-8% are present, a thick creamy mixture results. Pure beeswax should show no appreciable deposit after standing one to two hours. Paraffin. Melt 2-10 grams of wax in a porcelain dish, then add an equal weight of finely powdered KOH. Continue heating for a few minutes with continued stirring. Cool and powder the hard mass and mix the resulting powder with 3 times as much potash lime (1 part KOH: 2 parts CaO) as wax used. Then introduce this mixture into a thick walled tube, immerse in an oil bath and heat to about 250 C. for three to four hours. After cooling, finely powder the tube with its contents, place the mass in a Soxhlet apparatus and extract with 86 naphtha for several hours. Evap- orate off the naphtha, dry the residue at 100 C. and weigh. (Make sure the naphtha itself leaves no residue at 100 C.) By this treatment esters are converted into alcohols and these alcohols on heating with potash lime are in turn converted into the respect- ive acids, while hydrocarbons present are not affected and will extract with the naphtha. Pure beeswax contains naturally from 12.5-14% of hydrocarbons and any adulteration with paraffin or allied bodies will increase this percentage. NOTE. For further information in regard to analysis of doubtful samples, see Allen and Lewkowitsch. REFERENCE. Lewkowitsch: " Chemical Technology and Analysis of Oils, Fats and Waxes," Volume II, page 751. Allen: "Commercial Organic Analysis," Edition IV, Vol. II, pages 242-270. U. S. Dept. of Agriculture, Bureau of Chemistry, Bulletin 150, page 49. PARAFFIN WAX General. For ordinary white paraffin wax the only necessary determination is the melting point. Yellow waxes and paraffin scale, however, often contain a considerable amount of oil. Melting Point. (A) CAPILLARY TUBE METHOD. Take a piece of capillary tubing about 1 inch long, soften the wax in the ANALYSIS OF OILS, FATS, WAXES AND SOAPS 279 fingers if necessary and push one end of the tube into it, filling the tube about 0.25 inch from the end. Fasten the tube to a ther- mometer with a rubber band so that the wax is opposite the bulb. Immerse the thermometer bulb and tube in a beaker of water (it will save time if the water is previously warmed to about 90 F.). Heat slowly with a low flame without stirring and note the tem- perature at which the wax melts and runs out of the tube. (B) " ENGLISH TEST." The English Test gives in reality the solidifying point. Stir the melted wax in a small cup, about 2.5 inches in diameter by about 2 inches deep, until the latent heat given up by the crystallization of the wax arrests the fall of the mercury column momentarily. Take this reading as the melting point. (C) " AMERICAN TEST." This gives results about 3 F. higher than the English Test and is determined as follows: A hemispherical cup, 3.75 inches in diameter, is three-fourthfi filled with melted wax, which is allowed to cool without stirring until a thin film forms on the top and extends from the sides to a thermometer with a round bulb, 0.5 inch in diameter, suspended so that it is three-fourths immersed in the center of the cup. NOTE. As the American Test is slow, it is customary to take the Eng lish Test and add 3 F. for the American. Results should be reported in terms of Fahrenheit degrees. Oil. Chill the wax and powder as finely as possible, weigh out about 1 gram and digest for 1-2 hours in 10 cc. of acetone at ordinary temperature. Mix well, then place in a freezing bath and cool to 15 C. (5 F.) or lower. Filter through cotton wool in a funnel surrounded by a freezing mixture and wash with acetone which has been chilled. Evaporate the filtrate, dry at not over 100 C. and weigh the oil. REFERENCES. J. Soc. Chem. Ind. 25, page 139 (1906); Rogers- Aubert : "Industrial Chemistry," 1912 edition, page 527. SOAP Sampling. The preparation of the sample requires consider- able care, since the moisture content of the outer layer may be very different from that of the interior of the cake. 280 TECHNICAL METHODS OF ANALYSIS If the sample is comparatively dry, it is a good plan to run it through a meat chopper and reduce to fine particles, repeating the operation several times in order to obtain a homogeneous sam- ple. If the soap is too soft to permit this procedure, the cake should be cut in two diagonally, and thin shavings taken from fresh surfaces, care being taken to cut entirely across in order to obtain a fair proportion of the outer and inner parts. The sample in either case should be thoroughly mixed and kept in a tightly stoppered bottle. Moisture. Weigh out 20 grams of sample and dissolve in about 150 cc. of hot water. Transfer the solution while still hot to a 250 cc. graduated flask and dilute to the mark. Mix thor- oughly, pipette 25 cc. while hot into a weighed platinum dish and evaporate to dryness on the steam bath. Then dry to constant weight at 110 C. The percentage of solid matter subtracted from 100 gives the per cent of moisture and volatile matter. NOTE. The moisture as determined above may also include alcohol and essential oils if they are present in the soap, also naphtha or other volatile substances. In such cases the true moisture may be determined by the Xylol method (page 271), using about 20 grams of soap and acidifying with an excess of powdered anhydrous KHSO 4 before distilling. Total Alkali. Ignite the residue from the moisture determina- tion at a low red heat until all carbonaceous matter is burned off. Weigh the mineral residue, which consists of Na2COs or K^COs (also SiO2, NaCl, etc., if present). Pour boiling water into the dish and warm until the residue is dissolved. Cool, and titrate with 0.1 N acid and methyl orange. If the residue is not com- pletely soluble in water, the insoluble matter should be filtered off and the filtrate evaporated to dryness, ignited and weighed before titration. The alkali by titration, calculated to either Na2COs or K^COs, should check with the weight of soluble ash as above determined. If it does not, the presence of both Na and K is indicated (pro- vided NaCl, etc., are absent). To confirm the presence of both elements, take an amount of ash the size of a pinhead, dissolve in a few drops of HC1, put one drop on a microscope slide, evaporate to dryness over a micro burner and draw across it with glass rod a drop of saturated uranium acetate solution made slightly acid with acetic acid. If Na is present, clear cut light yellow tetra- ANALYSIS OF OILS, FATS, WAXES AND SOAPS 281 hedra of sodium uranium acetate will appear. Put another small drop on a slide and run in from the side a smaller drop of H 2 PtCle. Potassium will give yellow octahedra. CALCULATIONS. 1 cc. 0.1 N acid = 0.005300 gram Na 2 CO 3 . = 0.006910 gram K 2 CO 3 . = 0.003100 gram Na 2 0. = 0.004710 gram K 2 0. Na 2 CO 3 X 0.5848 = Na 2 O. K 2 CO 3 X 0.6816 = K 2 0. Total Fatty Matter. Pipette 200 cc. of the original hot soap solution (see Moisture) into a beaker,* add dil. HN0 3 until slightly acid, heat on the water bath until the fatty acids have collected in a clear layer on top and the solution below is perfectly clear. Cool in ice water or let stand overnight. Remove the layer of fatty acids to a beaker. Shake out the clear liquid in a separatory funnel with 2 portions of 50 cc. each of CHCls to remove the rest of the fatty matter. Transfer the CHCls extract to the beaker containing the fatty cake and dissolve the latter. Transfer the solution to a separatory funnel, rinsing the beaker with . Wash the CHCls extract -with 2 portions of 20 cc. of water. Evaporate off the CHCk, dry at 100 C. to constant weight and weigh as total fatty matter. NOTES. (1) Fatty acids are determined above as acids, whereas they exist in the soap as anhydrides, which for ordinary soap materials = acids X 0.9673 (Lewkowitsch: "Chemical Technology and Analysis of Oils, Fats and Waxes," 1895 edition, page 624). Usually analyses are reported giving the total fatty matter as fatty acids, since small losses are almost inevitable in this determination and for commercial analyses the direct figure obtained is usually the most satisfactory. (2) Total fatty matter as determined above may include fatty acids, uncom- bined fat, rosin and hydrocarbons. If it is desired to determine only the com- bined acids, the determination should be made on a portion of the dry soap previously extracted with ether or 86 Baume* gasoline, as described later under " Unsaponified Matter." (3) If fatty acids are liquid at ordinary temperature, a known weight of beeswax or stearic acid may be added to the hot liquid before chilling to form a solid cake. The weight of wax added should be about equal to that of soap employed and should be deducted of course from the weight of the cake. * Time may be saved, if necessary, by pipetting directly into the separatory funnel; then acidify and shake thoroughly before adding CHC1 3 . 282 TECHNICAL METHODS OF ANALYSIS Excess of H 2 O may be removed by pressing between filters and the cake dried in a desiccator (or in vacua) and weighed. (4) If further examination of the fatty acids is to be made, it is best to effect decomposition of the soap solution in a separatory funnel, running off the aqueous liquid through a wet filter and subsequently allowing the fatty acids to run on the filter, where they are washed with boiling water. (5) Cocoanut and palmnut oil soaps yield a fatty acid soluble to a slight extent in hot water. In such cases the separation of fatty acids should be made in as concentrated a solution as possible, saturated with common salt. The washing of the fatty acids should be limited and drying carried on with as little exposure to heat as possible. (6) If it is necessary to determine soluble fatty acids, consult Allen: " Commercial Organic Analysis," fourth edition, Vol. II, page 432. Free Caustic Alkali or Free Fatty Acids. Dissolve 5 grams of soap in warm neutral 95% alcohol. Filter, using a hot water funnel; wash with hot alcohol; titrate the nitrate with 0.1 N acid and phenolphthalein and calculate to NaOH or KOH ; as the case may be. If the filtrate is acid, titrate with 0.1 N caustic and calculate to oleic acid. CALCULATION. 1 cc. 0.1 N acid =0.00400 gram NaOH. Ice. 0.1 N acid =0.00561 gram KOH. . 1 cc. 0.1 N alkali = 0.0282 gram oleic acid. NOTES. (1) The presence of free caustic alkali may be detected by treat- ing a freshly cut surface of the soap with a few drops of phenolphthalein solution. If no red coloration appears, it may be assumed that free caustic alkali is absent. Either free caustic or sodium silicate will give a bright red coloration. In the presence of excess moisture, however, NasCO 3 will also give this color. (2) If the soap contains both free alkali and free fat, the above method is open to objection, since during heating with alcohol the free alkali will saponify some of the free fat. In cases where great accuracy is required, Devine's method should be used. For details see J. Am. Chem. Soc. 22, 693 (1900). Free Sodium Carbonate. Dissolve the residue from the alco- holic solution in the above determination by pouring boiling water through the filter. Wash, cool the solution and titrate with 0.1 N acid and methyl orange. Calculate to Na2COs or K^COs, as the case may be. (It is well to make a flame test with a platinum wire before dissolving the residue.) CALCULATION. 1 cc. 0.1 N acid = 0.005300 gram Na 2 CO 3 . ANALYSIS OF OILS, FATS, WAXES AND SOAPS 283 NOTE. If borax or sodium silicate is present, it is sufficiently accurate for commercial purposes to assume that they will remain insoluble in alcohol and be titrated with the free carbonate. In this case corrections should be made on the methyl orange titration of carbonate as follows : 1 cc. 0.1 N acid = 0.013 gram Na 2 Si 4 O 9 .* = 0.0191 gram Na 2 B 4 O 7 -10H 2 O. Combined Alkali (Na 2 O or K 2 O). Calculate the free caustic and free carbonate to Na2O or K2O. Deduct from the total alkali, calculated to Na 2 or K20, and the difference will be combined alkali (provided sodium silicate and borax are absent). CALCULATION. NaOH X 0.7748 = Na 2 0. KOH X 0.8394 = K 2 O. Na 2 C0 3 X 0.5848 = Na 2 O. K 2 CO 3 X 0.6816 = K 2 O. Sodium Silicate. The presence of sodium silicate is generally indicated by the fact that the weight of ash is greater than the weight calculated from titration, provided of course that insoluble abrasives are not present. The amount of sodium silicate may be ascertained by determining Si0 2 in the soluble ash after titration for total alkali. Add a slight excess of HC1, evaporale to dryness, and finally bake for one hour at 130 C. Take up with cone. HC1, dilute with hot water, filter, wash, ignite very strongly and weigh as SiO 2 . From this weight calculate to Na 2 Su09. CALCULATION. SiO 2 X 1.257 NOTE. If the soap contains insoluble matter, make a water solution, filter, evaporate the filtrate, burn off the organic matter and determine SiO 2 as above directed. Borax. Weigh 10 grams of soap (or 5 grams if more than 5% of borax is suspected) into a platinum dish and add 2.15 grams of fusion mixture (consisting of 200 grams Na 2 COs and 15 grams of Si0 2 finely powdered). To this mixture add 15 cc. of alcohol, mix with the aid of a glass rod and, after washing the rod with a little alcohol, evaporate the mass to dryness on the water bath; ignite until combustible material is destroyed, cover the dish with a * Theoretically 1 cc. 0.1 N acid = 0.01516 gram Na 2 Si 4 O 9 but commercial silicate varies somewhat in composition and 0.013 is taken as being nearer the actual average titration. 284 TECHNICAL METHODS OF ANALYSIS piece of platinum foil and fuse. Completely disintegrate the fusion by boiling with water and transfer the solution to a 250 cc. round-bottom flask. Acidify with 20 cc. of dil. HC1 (1:1). Heat nearly to boiling and add a moderate excess of dry precipitated CaCOs. Connect with a reflux condenser and boil vigorously. Filter out the precipitate through a folded filter, washing several times with hot water, keeping the total volume of liquid below 100 cc. Return the filtrate to the flask, add a pinch of CaCOs and again boil under a reflux condenser. Remove the flame and connect the top of the condenser with a suction pump. Apply gentle suction until the boiling has nearly ceased, cool to ordinary temperature, add 1 gram of mannite* or 50 cc. of neutral glycerol and titrate the solution with 0.1 N NaOH (free from carbonate) and phenolphthalein. After the end point is reached, add 1 gram more of mannite or 10 cc. more of neutral glycerol and again titrate. Repeat this process until the addition of mannite or gly- cerol causes no further action on the end point. The number of cc. of 0.1 N NaOH required, multiplied by 0.00955, gives the equiv- alent of borax (Na 2 B 4 O7 10H2O) present in the solution. NOTES. (1} This method is described in J. Ind. Eng. Chem., 5, 645 (1913). (2) It is always advisable to test the soap qualitatively before under- taking a quantitative determination. Qualitative Test for Borax. Place about 1 gram of the soap in a test-tube with 10 cc. of dil. HC1. Heat the mixture to boiling, which causes the fatty acids to rise to the surface. Cool under the tap and filter through a wet filter paper. Immerse a strip of turmeric paper in the mixture and dry it. If borax is present, the paper will acquire a deep red color when dry and this color will change to blue or green when treated with NHiOH or Na2COs solution. The test is sensitive to about 0.05% of borax in the soap. Insoluble Matter. Dissolve 5 grams of soap in 75-100 cc. of hot water. Filter on a Gooch crucible, wash with hot water, dry at 105-110 C., weigh and calculate the percentage of total insol- uble matter. Ignite the residue and calculate the percentage of insoluble mineral matter. * Mannite gives a sharper end-point than glycerol. ANALYSIS OF OILS, FATS, WAXES AND SOAPS 285 Chlorides. Evaporate the filtrate from the fatty acids to about 100 cc. Neutralize carefully with CaCOs, and titrate with 0.1 N AgN0 3 , using K 2 Cr0 4 as indicator. (See page 492.) CALCULATION. 1 cc. 0.1 N AgNOt = 0.00585 gram NaCl. = 0.00746 gram KC1. Glycerol. Dissolve 20-25 grams of soap in hot water, add a slight excess of H2SO4 and heat on the water bath until fatty acids separate in a clear layer. Remove the fatty acids and filter the acid solution into a graduated flask. Remove chlorides and sol- uble fatty acids by adding crystals of Ag2SO4. Cool, make up to the mark, mix, let settle and filter through a dry filter paper. Place an aliquot portion corresponding to about 5 grams of soap in a 100 cc. graduated flask, dilute slightly, add a little silver oxide, let stand ten minutes, and add a slight excess of basic lead acetate. Make up to the mark, filter through a dry paper, and place 25 cc. of the filtrate in a perfectly clean beaker. Add first 12 drops of H2SO4 (1:4) to precipitate the Pb, and then an accurately measured amount (40-50 cc.) of a solution of bichromate (made by dissolving 74.56 grams of pure K2Cr2O? in water and diluting to 1000 cc.), and then 15 cc. of cone. H2S04. Cover the beaker and heat for two hours in boiling water; then cool. Add an excess of standard ferrous ammonium sulfate solution and tirate back with standard bichro- mate solution containing 7.456 grams per liter. (The ferrous ammonium sulfate solution should contain about 240 grams per 1000 cc.) 1 cc. of the stronger bichromate solution corresponds to 0.01 gram of glycerol. In the presence of sugar the above method is not reliable, since sugar will also reduce bichromate. Consequently, when sugar is present, remove the fatty acids as before, neutralize an aliquot with milk of lime, evaporate to about 10 cc., add 2 grams of sand and milk of lime containing about 2 grams of Ca(OH)2, and evaporate almost to dryness. Treat the moist residue with 5 cc. of 96% alcohol, rub the whole mass into a paste, heat the mixture on the water bath, stirring constantly, and decant the liquid into a 250 cc. flask. Wash the residue 5 or 6 times with small portions of alcohol, cool the contents of the flask to 15 C., fill to the mark with 96% alcohol, mix, and filter through a dry paper. 286 TECHNICAL METHODS OF ANALYSIS Evaporate 200 cc. of the filtrate to a syrupy consistency on the water bath, transfer to a stoppered cylinder with 20 cc. of absolute alcohol, add 3 portions of 10 cc. each of absolute ether, mixing after each addition ; let stand until clear, pour off through a filter, and wash the contents of the cylinder on the filter with a mixture of 2 parts of absolute alcohol and 3 parts of absolute ether. Evap- orate to a syrup; dry for one hour at the temperature of boiling water, weigh, ignite, and weigh again. The loss multiplied by 1.25 is the weight of glycerol in the aliquot taken. (Instead of weighing the glycerol it may be titrated with bichromate after driving off the alcohol and ether.) Sugar. Dissolve 5 grams of soap in water, add an excess of HC1, and heat on the steam bath for thirty minutes. Cool and filter out the fatty acids, collecting the filtrate in a volumetric flask. Nearly neutralize the excess of acid with NaOH, make up to volume and mix. For analysis take an aliquot depending upon the amount of sugar supposed to be present. The aliquot must not contain more than 0.25 gram of sugar calculated as dextrose. Determine the total reducing sugars by Allihn's modification of Fehling's method (see page 407) and calculate as dextrose. Cane sugar = Dextrose X 0.95. NOTE. Sugar is often found in transparent toilet soaps, especially those sold as " glycerin soaps," to the extent of 20-30%. Such soaps also often contain alcohol and sodium acetate. Unsaponified Matter. Dry 10 grams of soap and extract in a Soxhlet extractor with ether or naphtha (boiling below 60 C.). Transfer the extract to a separatory funnel and wash twice with water. Evaporate off the solvent, dry at 100-105 C. and weigh. Naphtha. For the determination of naphtha or other hydro- carbons volatile with steam, place in a liter round-bottom wide- neck flask about 400 cc. of water and add about 50 cc. of H2SO4 (1:1). After cooling the solution, if necessary, introduce about 100 grams of soap, weighed to 0. 1 gram, into the flask. The soap should be weighed as one piece and cut quickly into as large pieces as will go through the neck.. Then introduce into the neck of the flask a 2-hole rubber stopper connecting through one hole by a bent glass tube to a condenser. Through the other hole run a glass tube nearly to the bottom of the flask. This tube is for admission of steam. Connect the outlet of the condenser by ANALYSIS OF OILS, FATS, WAXES AND SOAPS 287 means of an adapter to a long tube graduated to 0.1 cc. A gas burette with a good-sized bulb at the bottom is satisfactory. By means of a rubber tube connected to bottom of this burette water can be drawn off as it accumulates without disturbing the upper layer of naphtha. Surround the graduated tube with ice water. Heat the solution in the flask to boiling and then pass in steam. Continue the steam distillation until the condensate comes over perfectly clear. This generally requires one to two hours. Have sufficient water in the measuring tube so that the upper layer of naphtha remains in the graduated portion, carefully drawing off the underlying water as fast as necessary. After completing the distillation, let the contents of the tube come to 60 F. and read the volume of the upper layer. Then determine the sp. gr. at 60 F. of the volatile distillate. From this calculate its weight. Transfer the entire volatile distillate to an Erlenmeyer flask. Add 25 cc. of alcohol which has been neu- tralized to phenolphthalein, warm on the steam bath about fifteen minutes and titrate with 0.1 N NaOH and phenolphthalein. This gives the fatty acids volatile with steam. Calculate the total weight of fatty acids as oleic acid and subtract from the weight of the volatile distillate. Figure the difference to percentage of naphtha by weight. CALCULATION. 1 cc. 0.1 N NaOH = 0.028 gram oleic acid. Heavy Petroleum or Hydrocarbons Non-volatile with Steam. Dissolve 10-15 grams of soap in hot water, using as little as pos- sible, add 50 cc. of 0.5 N alcoholic KOH and evaporate to dryness. Dissolve in water, transfer to a separatory funnel and extract with ether. Wash twice with water, evaporate off the solvent, dry at 100-105 C. and weigh. Rosin. Dissolve 3 grams of the dry fatty acids, separated as under " Total Fatty Matter," in 30 cc. of absolute alcohol in a flask and pass dry HC1 gas through the solution in a moderate stream. Keep the flask cool by placing in a vessel of cold water. Continue passing the gas until the esters separate and no more gas is absorbed, which usually requires about forty-five minutes. Let the flask stand, stoppered, about one hour to complete the reaction. Dilute with 5 times the volume of water and boil until the acid solution is clear and the esters containing the rosin float on top. Transfer to a separatory funnel and wash the flask out with 288 TECHNICAL METHODS OF ANALYSIS 50 cc. of naphtha (74 Be*.). Run off the acid solution, wash the naphtha solution once with water, treat with a solution of 0.5 gram of KOH and 5 cc. of alcohol in 50 cc. of water and shake. The rosin is immediately saponified and the 2 layers will separate completely. Draw off the lower solution of rosin soap into another separatory funnel, acidify with dil. HC1 and shake out three times with ether. Wash the combined ether extract twice with water, draw into a weighed flask, evaporate off the ether, dry the rosin at 100 C. and weigh. Calculate the percentage in the total fatty matter and in the original soap. NOTE. The above method is that described by Twitchell, J. Soc. Chem. Ind. 13, 804 (1891), and depends upon the fact that aliphatic acids are converted into ethyl esters by HC1 in alcoholic solution, whereas rosin remains unchanged. Miscellaneous Substances. Metallic substances such as Pb, Hg, Cu, and Zn will be found in the filtrate from the Si02 deter- mination and may be determined by the usual methods. Dextrin, starch, gelatin, etc., are sometimes present in soap and they may best be determined in aliquot portions of the filtrate from the total fatty matter. In this case, however, H 2 S04 or HC1 should be used in place of HNOs to separate the fat. Carbolic acid and coal tar products may be determined by the method described in Allen's " Commercial Organic Analysis," 4th edition, Vol. II, page 426. CHAPTER VIII ANALYSIS OF WOOD, PAPER AND PAPER-MAKING MATERIALS CELLULOSE IN WOOD Sampling. Samples are best obtained from green wood. If the wood is dry it should be soaked in water. Place the wood in a vise and rasp across the grain with a woodworker's rasp, the idea being to obtain the sample in as fibrous a condition as possible, like mechanical pulp. Dry the rasped wood at 98-100 C. Weigh out samples of 5 grams each into 500 cc. casseroles. Analytical Procedure. Add 200 cc. of 1% NaOH solution to the fiber, cover with a watch glass and boil gently for one-half hour, washing the fiber down from the sides of the vessel several times. Filter with suction on a 1-inch perforated plate placed in a 5-inch funnel. A little of the fiber will run through the plate at first and must be poured back after a good mat, like an asbestos filter, is formed. In order to hold the plate firmly in the funnel during filtration and the subsequent manipulations, pass a piece of fairly stiff silver wire up through one of the holes in the center of the plate and securely fasten it by bending the end down. Let the silver wire extend down through the stem of the funnel, pro- jecting J inch beyond the end. By putting one or two slight bends in this wire it can be made to bear against the inside walls of the stem, holding the filter plate firmly in place. Wash the boiled wood with a good volume of hot water, suck dry, loosen up the fiber with a sharp-pointed glass rod, and attach the stem of the funnel to the rubber tube leading from a chlorine generator (under the hood) . Cover the funnel with a watch glass and pass a stream (between 1 and 2 bubbles per second) of washed chlorine gas up through the fiber. Continue the chlorination for one hour. Every fifteen minutes the fiber should be loosened and the lumps broken up with the pointed rod. 289 290 TECHNICAL METHODS OF ANALYSIS After chlorination return the funnel to the suction flask and wash with hot water to remove HC1. Place 150 cc. of 2% Na 2 SO 3 in a wash bottle. Invert the funnel over a 500 cc. casserole, push on the silver wire until the main mass of the fiber with the filter plate drops into the casserole, turn the funnel right side up and wash out all the fiber with a stream of Na2SOs solution from the wash bottle. Wash the wire and plate and add the remaining Na2SOs. Bring the mixture to a boil, add 3 cc. of 10% NaOH and boil for five minutes. Again collect the fiber on the filter plate, wash with hot water until the washings are colorless, loosen up and expose to chlorine as before. With the wood of some broad-leaved species all of the lignin is removed by the first chlorination and, instead of coloring yellow the second time in chlorine gas, the fiber bleaches to a pure white. With most coniferous woods the fiber turns yellow when exposed to chlorine the second time. In that event chlorination is con- tinued for one-half hour, the fiber washed and boiled in alkaline Na2SOs solution as before, and again exposed to chlorine. Except in very unusual cases the fiber will bleach when exposed to chlorine the third time. At whatever stage the fiber bleaches white, remove it immediately from the generator, wash on the filter plate with a large amount of water, transfer with the aid of dis- tilled water to a casserole, let stand under water for a short time, collect in a large sized tared Gooch crucible, and wash well with alcohol and finally with ether. Dry the product, at first at a gentle heat, and weigh it. NOTES. (1) It is almost impossible to wash out the last traces of acid, and unless the fiber is washed with alcohol and ether, the slow drying with the concentration of acid at the drying surfaces will cause a browning of the edge of the cellulose. (2) If for any reason there should be any mineral matter present which would not be removed by the treatment, ignite the dried product and sub- tract the weight of the ash, to obtain the pure cellulose. (3) This method was prepared by Arthur D. Dean in 1906, and has given good results in this laboratory. WOOD PULP SAMPLING AND TESTING General. The following procedures for sampling and testing wood pulp have been used in this laboratory for many years. The method for sampling machine-dried pulp in bales was originally WOOD, PAPER AND PAPER-MAKING MATERIALS 291 developed in this laboratory and has been adopted as the official method of the American Pulp and Paper Association and the Asso- ciation of American Wood Pulp Importers. A resume of the requirements of the above associations is given at the end of this method. At the present writing no method for sampling wood pulp in laps has been officially adopted for general use in this country, but the strip method as here described is extensively used. SAMPLING Pulp in Bales (Machine Dried). The sample shall be taken by boring into a bale to a depth of 3 inches with a special augur bit * or with a machine cutter using a single knife. The discs, which are approximately 4 inches in diameter, shall be removed and ten of them taken as a sample, selected as follows: One disc shall be taken from the second sheet of the wrapper, 2 discs from a depth of 1 inch, 3 discs from a depth of 2 inches, and 4 discs from a depth of 3 inches. At least 5% of all the bales in a lot shall be sampled, although in any case 10 shall be the minimum number of bales sampled. The holes to be bored shall be so located that in 5 successive bales they will represent a portion extending diagonally across the bale; the first hole to be bored at the corner, the edges of the cut being at a distance of 1 inch from the edge of the bale; the second cut shall then be made half way between the location of the first and the center, the third at the center, and so on until the fifth bale is sampled in the opposite corner in a position corresponding to the first. Pulp in Rolls. The sample shall be taken in the same manner as for pulp in bales. The position of the samples taken shall be determined according to the following plan: The first sample shall be taken so that the edge of the disc shall be within 1 inch of the end of the roll, the second half-way between the first and the center, the third at the center, and so on. Pulp in Laps (Wet Pulp). (A) CUTTING OF SAMPLE. The sample shall be taken by cutting a strip about 1 inch f wide from * These may be obtained from the Millers Falls Co., Millers Falls, Mass. t A narrower or wider strip may be adopted, provided the same width is cut from every bale sampled. 292 TECHNICAL METHODS OF ANALYSIS the center of the folded section to the middle of the outside edge; the cut shall be made in each case half-way through the lap. The cuts to be made shall be so located that in four successive laps they will represent a cross having its center at the center of the top face of the lap and each arm terminating in the center of one of the four edges. The diagram (Fig. 15) will show the posi- tion of the cuts. The fifth lap shall be sampled the same as the first, and so on. (B) SELECTION OF 'THE LAPS. When the pulp is loose in laps, the number of laps to be sampled shall be not less than one in every 2000 pounds of wet pulp. In sampling from a car, care shall be taken that the sample laps fully represent every portion of 1st Lap 2nd Lap 3rd Lap 4th Lap FIG. 15. Method of Sampling Pulp in Laps. the car. With loose pulp it is necessary to weigh the whole car- load. When the laps are bundled together in bales, the total number of bales shall be ascertained and not less than 2.5%* of this number shall be weighed and sampled. Each bale sampled shall be opened and one lap withdrawn ; the samples from these laps shall be cut as above described and the laps shall be taken from the bales as follows: From 20% of the bales sampled, withdraw the middle lap. From 40% of the bales sampled, withdraw the lap half-way between the outside and the center. From 35% of the bales sampled, withdraw the lap next to the outside. From 5% of the bales sampled, withdraw the outside lap. Those laps which are withdrawn from the outside of the bales shall be sampled on that side which formed the surface of the bale. * It is convenient to work with a multiple of 20. WOOD, PAPER AND PAPER-MAKING MATERIALS 293 TESTING OF SAMPLES All samples immediately upon being taken shall be placed in a suitable air-tight container, the cover of which shall not be removed until after weighing. The samples shall then be dried in a suitable oven provided with good ventilation, at a temperature of 204-220 F. (95-105 C.) until successive weighings made after an interval of not less than three hours show no further loss in weight. From the loss thus obtained, the total percentage of moisture shall be calculated, and the difference between this and 100% will represent the amount of bone-dry pulp. Air-dry pulp is understood to consist of 90% of absolutely dry pulp and 10% of water. The percentage of air-dry pulp should, therefore, be calculated by dividing the bone-dry percentage by 0.90. NOTE. The scales used for weighing the samples must be accurate to 1 gram or less. (If avoirdupois scales are used, they should be capable of weighing to 0.01 oz.) OFFICIAL METHOD OF AMERICAN PULP AND PAPER ASSOCIATION AND ASSOCIATION OF AMERICAN WOOD PULP IMPORTERS General. All tests must be made by a chemist duly authorized and approved by the Joint Committee representing the Associa- tion of American Wood Pulp Importers and the American Pulp and Paper Association, and must be made strictly in accordance with the following instructions otherwise the Committee reserves the right to withdraw the approval of any chemist at any time. Before proceeding to the weighing and sampling, the chemist must ascertain that not less than one-half of the parcel in question is available. Chemists must have proper and adequate equipment for weigh- ing and sampling the bales and for the weighing and drying of samples. All sampling of pulp must be done by or supervised by the approved chemist personally, or by his competent bona fide assistants. Each chemist must file with the Committee a complete list of his bona fide assistants, who will do the sampling, sucji list to have the approval of the Committee. The chemist will be held respon- sible for the correct sampling by his approved assistants. 294 TECHNICAL METHODS OF ANALYSIS The Committee shall at any time have the privilege of investi- gating the sampling done by chemists or their assistants. Every test certificate shall clearly state the name of the person who did the sampling. The test certificates hereafter shall be uniform and in accord- ance with forms to be approved by the Committee, a sample draft of which will be furnished by the Committee to each chemist. Number of Bales to be Sampled. Not less than 5%, nor more than 10%, of the entire shipment, but not less than 10 bales, shall be sampled; samples to be drawn only from sound and intact bales, from different sections of the entire shipment, and the analyst shall be careful to observe that no unusual conditions prevail in the selection of the bales. The accurate weight of all bales sam- pled is to be ascertained by a sworn weigher before sampling, or, whenever a sworn weigher is not available, by a competent person, who must make sworn affidavit that the weights are correct; and no other bales than those weighed are to be sampled, and whenever bales are numbered, the number is to be given in addition to the weight. Method of Sampling. The method of sampling is the same as previously described under Pulp in Bales. Weighing of Samples. All samples must be either weighed by accurate scales immediately after being drawn from the bales or, where this is impracticable, must be put into air-tight vessels, made of metal or glass with ground glass or metal stoppers, and due care must be used in the transportation of such samples until they can be properly weighed at the laboratory of the chemists The entire bulk of samples selected from the bales must be dried out for the test. The temperature in the drying oven shall be as near to 212 F. as possible, but shall not exceed 220 F., nor be less than 204 F. SULFATE COOK LIQUOR General. In the usual sulfate process for making wood pulp the principal ingredients of the liquor used for cooking the wood are NaOH, Na2S04, JS^COs, and Na2S. The active agents in the digestive process are NaOH and Na2$, and it is stated that these combine with roughly 50% of the weight of the dried wood WOOD, PAPER AND PAPER-MAKING MATERIALS 295 to form soluble organic sodium salts. The liquor used for cooking wood is generally referred to as White Liquor, the residual liquor after the cooking as Black Liquor. The waste sulfate liquor (Black Liquor), obtained from cooking pulp materials by the sulfate process, contains in addition to small amounts of silicate, various organic salts of sodium and also vary- ing proportions of Na 2 S, Na 2 SO 4 , Na 2 S03, NaOH and Na 2 CO 3 . BLACK LIQUOR Specific Gravity. Determine the sp. gr. at 15.5 C. with the Westphal balance and calculate the gravity Baume. Total Solids. Weigh out 100 grams of the material and dilute to' 1 liter with distilled water in a graduated flask. Mix thor- oughly and pipette into a weighed platinum dish 50 cc. (equivalent to 5 grams). Evaporate on the steam bath and then dry to con- stant weight at 105 C. There will be some loss of H 2 S but the result will be approximately correct. Ash. Ignite the above residue until the carbon is burned off. In case it is not possible to burn off all the carbon, dissolve in water and filter through an ashless filter into a beaker. Ignite the filter paper in the same platinum dish. Cool and add the filtrate to the dish. Evaporate to dryness, ignite gently, cool in a desiccator and weigh. Silica. Moisten the ash above obtained with cone. HC1, warm, add a few cc. of water, evaporate to dryness on the steam bath, bake at least one hour at 120 C., take up with dil. HC1, heat to boiling, filter through a quantitative filter, wash with hot water, ignite strongly in a platinum dish and weigh as SiO 2 . Save the filtrate. Sodium Sulfate. Pipette 25 cc. of the original sample into a beaker, add about 100 cc. of distilled water, make slightly acid with HC1 and heat to boiling. Add 10 cc. of hot BaCl 2 solution slowly, drop by drop, boil for 0.5 hour and filter, washing, with hot water. Ignite in a weighed platinum crucible. On account of the organic matter present, some of the BaS0 4 is likely to be reduced to BaS. Moisten the precipitate in the crucible with a few drops of dil. H 2 SO 4 . Again ignite, cool and weigh as BaSO 4 . Cal- culate to Na 2 SO 4 and also to Na 2 O. 296 TECHNICAL METHODS OF ANALYSIS CALCULATION. BaSO 4 X 0.6086 = Na 2 S0 4 . BaS0 4 X0.2656 = Na 2 0. Total Sodium. Take an aliquot of the original solution corre- sponding to 5 grams of the sample, add an excess of HC1, evaporate to dryness and ignite to a dull red heat. All the sodium salts except the sulfate are decomposed to NaCl. The temperature must be kept low to avoid volatilization of the latter. Leach out the residue with hot water and filter. Cool and titrate the NaCl with 0.1 N AgNOs solution in the usual way, using Na 2 Cr0 4 as an indicator. It is necessary to leach the residue thoroughly with small por- tions of hot water and to remove all the particles from the dish. For accurate work it is desirable to filter and ignite the residue on the filter gently in a platinum dish and again leach with water. The leaching should be continued until the filtrate gives no reac- tion with AgNOs. Calculate the titration to Na 2 0. CALCULATION. 1 cc. 0.1 N AgNO 3 = 0.003 100 gram Na 2 O. The sum of the Na 2 thus found and the Na 2 found present as sulfate gives the total sodium as Na 2 O. Sodium Sulfide (Volumetric Zinc Method). SOLUTIONS. (A) Standard Zinc Solution. Dissolve 16.75 grams of c. P. 30-mesh zinc powder in a small excess of HNOs. Add sufficient NH^OH to redissolve all the precipitate formed and then 50 cc. excess. Dilute to 2000 cc. (If any precipitate forms after dilution, insuf- ficient NELiOH was added.) (B) Ammoniacal Nickel Sulfate Indicator. Make an approx- imately 10% solution of nickel ammonium sulfate and add a slight excess of NILiOH. TITRATION. From the solution previously prepared for Total Solids pipette out 100 cc. (equivalent to 10 g^ams of the original). Dilute to about 250 cc. in a beaker with distilled water and run in from a burette the standard zinc solution. The end point is reached when a drop of the solution in the beaker added to three drops of the nickel sulfate indicator tested on a white spot plate no longer forms a black precipitate. From the number of cc. of zinc solution required calculate the amount of Na 2 S. CALCULATION. 1 cc. zinc solution = 0.0100 gram Na 2 S. Total Available Alkalinity. Evaporate 25 cc. of the original sample to dryness in a platinum dish, ash over a Tirrill burner WOOD, PAPER AND PAPER-MAKING MATERIALS 297 and leach out the soluble salts with hot distilled water as in the determination of Total Sodium above. Cool the filtrate and titrate with 0.5 N acid and methyl orange. Calculate the titration to Na2O and also to NaOH. This gives the alkalinity available for recovery. CALCULATION. 1 cc. 0.5 N acid = 0.01550 gram Na20. = 0.02001 gram NaOH. Free Caustic Soda. Pipette 100 cc. of the Black Liquor (cal- culate the weight from its sp. gr.) into a 500 cc. volumetric flask and add 50 cc. of 10% BaCb solution. Shake and dilute to the mark with distilled water, freshly boiled and free from CO2. Let settle clear. Pipette 50 cc. of the clear supernatant liquor (equal to 10 cc. of the original) into a beaker and titrate with 0.1 N HC1 and phenolphthalein. Calculate to NaOH. CALCULATION. 1 cc. 0.1 N acid = 0.004001 gram NaOH. NOTE. The above determination must be corrected for Na2S, if present, as the latter reacts to phenolphthalein when it is half neutralized, according to the reaction: Na 2 S+HCl = NaSH-f NaCl. Hence 1 cc. 0.1 N HC1 = 0.007806 gram Na 2 S. To apply the correction, therefore, calculate the weight of Na 2 S which would be present in the amount of liquor taken for titration and divide by 0.007806. This will give the number of cc. of 0.1 N HC1 required by the Na 2 S. Subtract this from the total titration and calculate the difference to NaOH. Sodium Carbonate. Pipette 25 cc. of the original liquor into a white casserole and add 200 cc. of distilled water. Titrate with 0.5 N HC1 and methyl orange. This titration gives the alkalinity due to Na2COs, NaOH, and Na2S.* Subtract the equivalents of NaOH and Na2S previously determined and calculate the differ- ence to Na2COs. CALCULATION. 1 cc. 0.5 N HC1 = 0.02650 gram Na 2 CO 3 . Sodium Silicate. The sodium silicate may be approximately calculated by multiplying the amount of Si02 by 1.257. This is for the formula Na2Si40 9 . In the presence of considerable amounts of sodium silicate the method of analysis is complicated and results by the above procedures are not strictly correct as regards NaOH and Na 2 CO 3 . Total Alkalinity. The total alkalinity of the liquor itself is obtained by direct titration with standard acid and methyl orange * Also one-half of Na 2 SO 3 , if the latter is present. 298 TECHNICAL METHODS OF ANALYSIS as described under the determination of Sodium Carbonate. Calculate the titration as Na 2 O and also as NaOH. CALCULATION. 1 cc. of 0.5 N acid = 0.01550 gram Na 2 O. = 0.02001 gram NaOH. WHITE LIQUOR General. The determination of total solids, ash, total sodium and available alkalinity are not generally required on white liquor, but, if desired, may be made by the same procedures as described under Black Liquor. Specific Gravity. Determine as under Black Liquor. Sodium Silicate and Silica. Pipette 50 cc. into a platinum dish, add cautiously an excess of HC1, evaporate to dryness, and then bake for one hour or more at 120 C. Take up with HC1, dilute with water, filter, wash with hot water, ignite the residue in plat- inum with a blast lamp and weigh as SiO 2 . An approximation of the sodium silicate (Na^uOo) present may be obtained by mul- tiplying the SiO 2 thus found by 1.257. Sodium Sulfate. Determine as under Black Liquor. The Na 2 SC>4 should be present only as an impurity. Sodium Sulfide. Pipette 25 cc. of the liquor into a dry beaker of about 300 cc. capacity. Do not dilute with water. Run in from a burette an amount of ammoniacal AgNOa solution as close to the saturation point as can be estimated. (This can be done from a preliminary titration.) Then shake the con- tents of the beaker rather vigorously for two or three seconds. This causes the black Ag 2 S to separate in thick lumps from the clear solution. Add another drop of the AgNOs solution from the burette. If this forms a heavy precipitate where the drop comes in contact with the solution, add a few more drops, give the beaker a shake and repeat the additions until only a faint cloud appears in the clear solution. If the beaker is held over a white sheet of paper, the end point may be easily determined within a single drop. The last drop necessary to complete the precipitation forms only a faint cloud in the clear -solution. If then another drop is run in after shaking, instead of forming a faint dark cloud, it will remain as a clear colorless spot surrounded by the faintly distributed pre- cipitate in the pale brownish solution. WOOD, PAPER AND PAPER-MAKING MATERIALS 299 NOTE. If the silver solution is added in excess it will all be distributed evenly through the solution and no amount of shaking will cause it to separate in lumps. A ryimoniacal Silver Nitrate Solution. Dissolve 55.29 grams of pure metallic silver in pure HNOa, or dissolve 87.07 grams of pure AgNOs in water. Then add 250 cc. of cone. NH^OH and dilute to 1 liter. (Keep protected from strong light and away from heat.) 1 cc. of this solution is equivalent to 0.02 gram of Na2S. Alkalinity. (A) TOTAL ALKALI EXPRESSED IN TERMS OF NA20. Pipette out 25 cc. of the liquor and titrate with 0.5 N acid and methyl orange. The number of cc. of acid used (call this A) represents the alkali existing in the solution as Na 2 COs, NaOH, Na 2 S, and one-half Na 2 S0 3 . Calculate it in terms of Na 2 O. CALCULATION. 1 cc. 0.5 N acid = 0.01550 gram Na 2 O. (B) SODA AS NAOH+NA 2 S. Pipette 25 cc. of the sample into a 100 cc. graduated flask. Add 25 cc. of a 10% solution of BaCl 2 and make up to the mark with freshly boiled distilled water. Shake for a few minutes and let settle. Cool and draw off 50 cc. of the clear liquid and titrate with 0.5 N acid and methyl orange. The number of cc. multiplied by 2 indicates the amount of acid necessary to neutralize the NaOH and Na 2 S in the sample. (Call this B.) The difference between A and B represents the number of cc. required to neutralize the Na 2 COs and one-half the Na 2 SOs, barium sulfite being practically insoluble. (C) SODA AS SULFIDE, SULFITE AND THIOSULFATE. Make a rough titration of 10 cc. of the liquor with 0.1 N iodine, after acidifying with acetic acid. Then run out from a burette 0.5 cc. less than the required amount of iodine into a beaker containing about 200 cc. of distilled oxygen-free water. Pipette into the mix- ture 10 cc. of the liquor, make acid with acetic acid and complete the titration with 0.1 N iodine, using starch as indicator. This titration indicates the amount of Na 2 S, Na 2 S 2 Os and Na 2 SOa in the sample. (Call this titration C.) CALCULATION. 1 cc. 0.1 N iodine = 0.003903 gram Na 2 S. = 0.003100 gram Na 2 O. (D) SODIUM THIOSULFATE AND SULFITE. Pipette 50 cc. of the liquor into a graduated 250 cc. flask. Add an excess of an alkaline solution of zinc chloride (made by adding NaOH to a 300 TECHNICAL METHODS OF ANALYSIS solution in sufficient excess to redissolve the precipitate formed). Make up to the mark, shake for a few minutes and let settle. Draw off 100 cc. of the clear solution with a pipette and neutralize with 0.1 N H2SO4, using methyl orange. This converts the sulfites present into acid sulfites. When acid sulfites are titrated with iodine the following reaction takes place: Hence 1 molecule of acid sulfite on titration with iodine liberates acid equivalent to 3 molecules of NaOH. Titrate the neutralized solution with 0.1 N iodine solution and starch. (Call this D.) Then decolorize with 1 drop of 0.1 N thiosulfate solution and titrate till neutral with 0.1 N NaOH. The number of cc. of NaOH, multiplied by 0.0042, gives the amount of Na2SOs in the aliquot; and this figure, divided by 0.0063, gives the number of cc. of 0.1 N iodine to which it is equivalent. Subtract this from the iodine titration previously obtained. Calculate the difference to sodium thiosulfate. CALCULATION. 1 cc. 0.1 N iodine = 0.01581 gram Na2S2Os. (E) FINAL CALCULATIONS. C gives the cc. of iodine equiv- 2 alent to the Na2S. (This is a check on the Na2$ by ammoniacal AgN0 3 .) A-B gives the number of cc. required by the Na2COs and one-half the Na2SOa. 1 cc. 0.1 N iodine = 0.003903 gram Na 2 S. 1 cc. 0.5 N acid =0.02650 gram Na 2 C0 3 . = 0.01550 gram Na 2 O. The titration B, expressed as Na20, minus the sodium sulfide titration (with ammoniacal AgNOs), expressed as Na 2 0, gives the Na2O present as NaOH. Calculate this to NaOH by multiply- ing by 1.291. NOTE. The above methods under Alkalinity are those of Otto Kress pub- lished in Paper, 17, No. 24, page 30 (1916). REFERENCES. The method for Sulfide in White Liquor is described by Carl Moe in Paper, 14, No. 22, page 19 (1914). See also article by Otto Kress in Paper, 17, No. 24, page 30 (1916); and articles by Carl Moe begin- ning in Paper, 18, No. 12, page 11 (1916), and by the same author in Paper, 13, No. 24, page 156 (1913). WOOD, PAPER AND PAPER-MAKING MATERIALS 301 SULFITE ACID General. " Sulfite Acid " or " Bisulfite Liquor " is the liquid used for cooking wood to make pulp by the well-known sulfite process. It consists of an aqueous solution of calcium and mag- nesium bisulfites with an excess of sulfurous acid. As the SO 2 gas from which it is made always contains a little 80s, the fin- ished acid liquor will contain more or less CaS04 as an impurity. Specific Gravity. Determine with the Westphal balance or by the pycnometer, preferably the former. Total Sulfurous Acid. Measure out 10 cc. of the acid and make up to 100 cc. in a graduated flask. Pipette 10 cc. of the diluted acid into an excess of 0.1 N iodine and titrate with 0.1 N thio, using starch solution as an indicator. Calculate to SO 2 . CALCULATION. 1 cc. 0.1 N iodine = 0.0032 gram SCb. Free Sulfurous Acid. Titrate 25 cc. of the diluted solution with 0.5 N KOH, using phenolphthalein as an indicator. Cal- culate to SC>2. CALCULATION. 1 cc. 0.5 N KOH = 0.016 gram SO 2 . Sulfuric Acid. Measure out 25 cc. of the original solution, add a slight excess of HC1 and boil until the 862 is completely expelled. To the boiling solution add 10% BaCk solution, drop by drop, in excess. Boil for thirty minutes, let stand till clear, filter, ignite and weigh as BaSO 4 . Calculate to 80s. CALCULATION. BaSO 4 X 0.3430 = SO 3 . Lime and Magnesia. Pipette out 25 cc. of the original solu- tion, add about 1 cc. of cone. H2SO4, evaporate to dryness in a platinum dish, ignite carefully, cool and weigh the mixed sulfates as CaSO 4 and MgS0 4 . Dissolve the residue in 25 cc. of dil. HC1, wash the solution into a beaker, make alkaline with NKiOH, heat to boiling and add enough (NILi)2C20 4 solution to completely precipitate the lime. Continue the boiling for two minutes, and let the precipitated CaC2O 4 settle for one-half hour. Filter, wash with hot water, ignite in a platinum crucible over a blast lamp to constant weight and weigh as CaO. Calculate the CaO to CaSO 4 , deduct from the mixed sulfates and calculate the MgSO 4 to MgO. CALCULATION. CaO X 2.4279 = CaSO 4 . MgS0 4 X 0.3349 = MgO. 302 TECHNICAL METHODS OF ANALYSIS NOTE. The Ca and Mg sulfates may also be separated as follows: Add 10 cc. of water and 1 or 2 drops of HC1, which will dissolve all the MgSO 4 . Add 1 or 2 drops of cone. H 2 SO 4 and 25 cc. of 95% alcohol. Let stand for one hour, filter and wash with 60% alcohol to remove acid. Finally wash with 40% alcohol as long as anything is dissolved. Ignite and weigh as CaSO 4 . Results. The results should be expressed as follows: Sulfurous Acid (SO 2 ) Sulfuric Acid (SO 3 ) Lime (CaO) Magnesia (MgO) Equivalent to: Calcium Sulfate (CaS0 4 ) Calcium Bisulfite (CaS 2 O 5 ) Magnesium Bisulfite (MgS 2 O5) Free Sulfurous Acid (802). The 80s is calculated to CaSO4. The remaining CaO is cal- culated to CaS 2 5 and the MgO is calculated to MgS 2 O 5 . The excess SO 2 is expressed as free S0 2 . The so-called "Mill Test " is expressed as follows: " Free " Sulfurous Acid. " Combined " Sulfurous Acid. Total Sulfurous Acid. The " free " sulfurous acid is the actual free S0 2 plus one-half of the SO 2 in the CaS 2 Os plus one-half of the SO 2 in the MgS 2 O5, and should check approximately with the figure obtained by the titration with KOH. It is, more strictly speaking, the " available SO2. " The " combined " sulfurous acid is the sum of one-half of the SO2 in the CaS 2 Os and one-half of the SO 2 in the MgS 2 5 . NOTES. (1) The " free " and " combined " SO 2 together should be the same as the total SO2 determined by titration with iodine. (2) Griffin and Little ("The Chemistry of Paper Making," page 228) give the following as a typical analysis of a well-made liquor prepared from dolomite : Sp. gr. at 15 C 1.0582 Total S0 2 4.41% SO 3 0.13% CaO 0.95% MgO 0.72% WOOD, PAPER AND PAPER-MAKING MATERIALS 303 Combined as: CaSO 4 0.22% CaS 2 5 2.84% MgS 2 O 5 3.04% Free SO 2 0.11% This would give a " Mill Test " as follows: Free SO 2 2.26% Combined SO 2 2. 15% Total S0 2 4.41% The tendency to-day is toward a high "free S0 2 ," 2.50-3.20%, and a lower combined SO 2 , 1.20-1.35%. For cooking dry wood a suitable ratio is (approximately): Total SO 2 : "free" S0 2 : "combined" SO 2 =4 : 3 : 1. For wet wood the " combined " SO 2 should be higher. Factors. The following factors will be found useful in cal- culating the final combinations: SO 3 X 1.700 =CaSO 4 . CaOX 3.285 =CaS 2 O 5 . MgOX4.178 = MgS 2 5 . CaS 2 O 5 X0.6956 = SO 2 . MgS 2 5 X 0.7606 = S0 2 . ALUM General. This method is for the " alum " used by paper- makers, which is not a true alum but a sulfate of alumina, A1 2 (S0 4 )3-18H 2 O. Insoluble Matter. Dissolve 25 grams of the sample in about 200 cc. of boiling water and filter rapidly through a weighed Gooch crucible with suction. Wash the residue on the filter with hot water, dry at 105 C. and weigh. SOLUTION No. 1. Transfer the filtered solution to a 500 cc. volumetric flask, cool, dilute to 500 cc. and thoroughly mix. SOLUTION No. 2. Transfer 100 cc. of Solution No. 1, equiva- lent to 5 grams of alum, to another 500 cc. flask, make up to the mark with distilled water, and thoroughly mix. Alumina. Pipette out 50 cc. of Solution No. 2, equivalent to 0.5 gram of alum, and acidify with 10 cc. of cone. HG1 and 4-5 drops of cone. HNO 3 . Heat to boiling. Add 10 cc. of 10% 304 TECHNICAL METHODS OF ANALYSIS NBUC1 solution and a pinch of tannic acid on the end of a spatula, and then NELiOH until barely alkaline. Boil until there is only a faint smell of NHs. Let the precipitate settle, filter and wash free from chlorides. Dry the precipitate in the oven in a weighed platinum crucible, then ignite over a blast lamp to constant weight, and weigh as Fe203+Al203. To obtain the amount of alumina, determine the Fe20s separately as below and deduct from this weight. , Sulfuric Acid. Acidify 100 cc. of Solution No. 2 with HC1 and heat to boiling. Then add 10% BaCl2 solution slowly, drop by -drop, as long as it produces a precipitate. Boil one-half hour. (Or, boil five minutes and let stand overnight.) Let the precipitate settle completely, filter, ignite and weigh as BaSC>4. (Test the filtrate with more BaCb to insure complete precipitation.) Cal- culate to 80s. CALCULATION. BaSO 4 X 0.3430 = S0 3 . Iron. Transfer 100 cc. of Solution No. 1 to a 200 cc. beaker and add 5 cc. of cone. H2SO4 and 0.1 N KMn(>4, drop by drop, till a permanent pink forms. Pass this solution through the Jones reductor (see page 148) and titrate to a faint pink with 0.1 N KMnC>4 solution.* Calculate the iron as Fe2Os. CALCULATION. 1 cc. 0.1 N KMnO 4 = 0.008 gram Fe 2 O 3 . NOTE. In case it is impossible to get the reductor so that a single drop of KMnO 4 will color the washing solution, run a " blank " on the reductor using the same amounts of reagents and washings and subtract the " blank " titration from the titration required by the sample. Zinc. Add (NH^S in excess to the filtrate from the AkOa precipitate and let stand some time. Filter, wash with H2S water and dissolve the ZnS precipitate in dil. HC1. Make alkaline with NHiOH, add 1 drop of litmus solution and HNOs until the litmus just turns red. Heat nearly to boiling and add slowly ammonium phosphate solution containing a weight of phosphate equal to 12 times that of the Zn to be precipitated. Keep the solution just below boiling point for about fifteen minutes, or until the precipitate has become crystalline. Let the solution cool for * For alums containing a very small amount of Fe use 0.01 N KMnO 4 for titrating both the sample and the " blank." If extreme accuracy is desired a larger sample may be used. WOOD, PAPER AND PAPER-MAKING MATERIALS 305 at least four hours and then filter on a Gooch crucible. Wash with a 1% solution of ammonium phosphate until free from sul- fate, then wash 5 times with alcohol. Dry at 105 C. to constant weight and weigh as ZnNHiPCX. (It may also be ignited and weighed as Zn 2 P207.) Complete precipitation of the Zn depends on the neutrality of the solution and it is absolutely necessary to test the filtrate for Zn by adding NKUOH and a few drops of (NILi) 2 S and allowing it to stand for some time. CALCULATION. ZnNH 4 PO 4 X 0.3663 = Zn. Zn 2 P 2 O 7 X 0.4289 = Zn. Lime. Boil the filtrate from the Zn precipitate to expel H 2 S, make alkaline with NH^OH and add (NH4) 2 C 2 O4 solution in excess. Filter, wash, ignite in the blast and weigh as CaO. Basicity or Acidity. The following method * is based on the fact that an excess of neutral KF decomposes aluminum salts, forming two stable compounds reacting neutral to phenolphtha- lein, while any free acid remains unaltered, thus: SOLUTIONS REQUIRED. (1) 0.1% alcoholic solution of phenol- phthalein. (2) KF solution, prepared by dissolving 1000 grams of pure KF in 1200 cc. of hot CO 2 -free distilled water, adding 5 cc. of phenolphthalein solution and neutralizing if necessary with KOH or H 2 SO 4 (or HF) until about 1 cc. in 10 cc. of dis- tilled water shows a faint pink color. Filter out any insoluble matter without washing and dilute the clear nitrate to 2000 cc. with C0 2 -free water. Preserve in a wax bottle or a glass bottle coated inside with wax. (3) Standard 0.5 N # 2 S0 4 and 0.5 N KOH, free from A1 2 O 3 and similar bases. Standardize the alkali against the acid in about 40 cc. of distilled water to which 10 cc. of the above KF solution have been added, using phenolphthalein indicator. PROCEDURE. Place exactly 68 cc. of Solution No. 1, equivalent to 3.40 grams of sample, in a 4-inch casserole. Add about 35 cc. of distilled water and heat to boiling. To the solution add exactly * J.Soc. Chem. Ind. 30, 184 (1911). 306 TECHNICAL METHODS OF ANALYSIS 10 cc. of 0.5 N H 2 S0 4 . Cool to room temperature; add 18-20 cc. of KF solution and 0.5 cc. of phenolphthalein solution. Titrate with 0.5 N KOH (do not use NaOH), adding it drop by drop until a slight pink persisting for one minute is obtained. The titration shows whether the material is basic or acid as follows : Basic AhOz. When the KOH back titration is less than the amount of H2SO4 added, then Free A1 2 3 = (cc. of H 2 S0 4 cc. of KOH) X0.25. Free # 2 $0 4 . When the KOH back titration is greater than the amount of H 2 SO 4 added, then Free H 2 S0 4 = (cc. of KOH cc. of H 2 S0 4 )X0.72. Neutrality. If the back titration is equal to the H 2 SO 4 added, then the alum is neutral. NOTE. Darkening of the solution during the back titration with KOH indicates an insufficient amount of KF added. In such cases, repeat the test with a fresh solution and a larger amount of KF. Ammonium salts if present must be expelled by boiling the sample with an excess of standard KOH and this excess determined. Also if much iron salts are present, an increased quantity of KF solution may be required. CALCULATION OF RESULTS. It is seldom that papermakers' alum contains Zn or more than traces of lime, titanium, etc. Under ordinary circumstances the results should be reported as follows: Insoluble Matter Alumina (A^Os) (soluble in water) Sulfuric Acid (S0 3 ) Iron, calculated as Ferric Oxide (Fe 2 0s) Equivalent to : Insoluble Matter Ferrous Sulfate (FeS0 4 ) Aluminum Sulfate [Al 2 (S0 4 )a] Basic Alumina (Al 2 Oa)* Water (by difference). The Fe 2 0s is all calculated to FeSO 4 , although as a matter of fact more or less of it is generally present in the oxidized state. The * Or, Free Sulfuric Acid (H 2 SO 4 ). WOOD, PAPER AND PAPER-MAKING MATERIALS 307 remaining SO 3 is calculated to Ab (804)3. Any A1 2 3 left over is reported as basic A1 2 O 3 . In case the sample contains free acid, the Fe 2 O 3 is calculated to FeSC>4 as above, the total A1 2 3 to Ah (864)3, and the remaining 80s to free H2S04. Factors. The following factors will be found useful: Fe 2 3 X 1.9026 =FeSO 4 . A1 2 O 3 X 3.3503 = A1 2 (SO 4 ) 3 . A1 2 (S0 4 )3 X0.2985 = A1 2 O 3 . A1 2 (SO 4 ) 3 X0.7015 = SO 3 . Fe 2 O 3 X 1.0028 =SO 3 in equiv. am't of FeS0 4 . 80s X 1.2251 =H 2 S0 4 . ANILINE DYES General. This method is for the testing of aniline dyes for use in coloring paper. Solutions. (A) The stock should be made up from bleached sulfite pulp to contain 1 gram of dry fiber to 200 cc. of water. Add 2.5% of rosin and 5.0% of alum * to the stock in the beater and beat into the pulp. (B) The dye solutions should be made up in the ratio of 1 gram of dye in 2000 cc. of water. (500 cc. make a convenient amount to handle.) In making up the dye solution, add the weighed powder to about 100 cc. of water in a 250 cc. beaker and boil very gently for a few moments. Rinse into a 500 cc. graduated flask and dilute to the mark. (C) Dilute solutions of sodium chloride, lead acetate and soda ash are also required as mordants for certain colors. Procedure. Take 200 cc. of the stock solution in a finger bowl or other container of similar shape, and add the dye solution from a pipette. The strength of dye to use depends both on the class and the individual characteristics of the color. In general the following is a guide : For basic colors, from 0.5-0.20% of the weight of dry fiber. For acid and direct colors, about 0.25% of the weight of dry fiber. * Paper-makers' alum, Al 2 (SOo) 3 -18H 2 O. Both of these percentages are based on the weight of dry fiber. 308 TECHNICAL METHODS OF ANALYSIS The stronger the tinctural power, the smaller is the percentage required. Make up two lots of equal proportions of the laboratory stand- ard and of the sample under examination. Also make up several standards with 10% differences in strength of dye, as 70%, 80%, 90%, 110%, etc. Form the sheets by the use of a hand mold. After forming the sheet couch it on a strip of absorbent paper, fold the absorbent paper over the sheet and mark it with the strength of the dye. Squeeze out the surplus moisture by passing through a wringer. Remove the dyed sheet, punch a hole near the edge and suspend it on a copper wire or glass rod in the drying oven at a temperature of about 85 C. After the samples are entirely dry, they should be compared, getting the light from different angles and noting both sides of sheets. It will be found an easy matter to decide which of the standard samples is the nearest match and the whole procedure should then be repeated, making up standards varying by 5% each side of the one most nearly matching the dye under examina- tion. Results are to be reported as : X parts sample = Y parts standard (X preferably being equal to 100). NOTES. (1) A convenient method of permanently marking sample sheets is to use an indelible pencil after passing the sheets through the wringer. (2) In testing the yellow dyes, it is a help to add methylene blue or safra- nine, using a constant quantity of about 25% of the amount of yellow used. Samples so made are compared for shade rather than strength. (3) This method was submitted by H. P. Carruth in 1911. BLANC FIXE General. Blanc Fixe is generally marketed as a paste of BaSO4 and water, less commonly in the dry form. The com- mercial blanc fixe often also contains more or less phosphate and impurities such as carbonates, sulfites, organic matter, silica, heavy metals and alkali metals. It is quite often the case that a blanc fixe is practically pure as it leaves the factory but takes up impurities on standing in the storage barrels. It is claimed that WOOD, PAPER AND PAPER-MAKING MATERIALS 309 it is bad to store it in oak barrels, as a yellow stain nearly always develops. It should preferably be stored in fir barrels. Blanc fixe in the pulp form (paste) should not contain more than 30% of moisture, and both pulp and dry forms should con- tain not less than 97.5% of BaS04 on the dry basis. As a general rule the determination of water, organic and volatile matter, and total BaSCX is sufficient, but it is sometimes desirable to determine the amount and nature of the impurities and accordingly the following complete scheme is given. Qualitative Tests. As the source of raw materials and methods of manufacture determine the presence of certain impurities, a qualitative analysis should be made before proceeding with the quantitative determination. Where the blanc fixe is intended solely for photographic pur- poses, test the sample first with AgNOs as follows : Spread a small sample on a glass plate with a spatula and apply a drop of 10% AgNOs solution. If a deep brown or blackish stain develops within five minutes in the dark, the sample is unfit for photo- graphic purposes. Failure to meet this test, however, does not disqualify it as a filler or for coating ordinary paper. Organic Matter. A rough qualitative test for organic matter is to heat about 1 gram in a test-tube with 5-10 cc. of cone. H2SO4 and let settle. If organic matter is present the acid will turn brown. Silica. If a half gram sample dissolves completely in 5-10 cc. of hot cone. H2SO4 the absence of SiO2 is indicated. Lead. Boil 5 grams with saturated ammonium acetate solution and filter. To the filtrate add a solution of K^C^O?. A yellow precipitate indicates Pb. Iron and Alumina. Heat 5 grams with 25 cc. of HNOs (1 : 1) and filter. Add a slight excess of NKLjOH to the filtrate and heat. A flocculent precipitate indicates alumina, if white, and iron and alumina, if brownish. Alkalies. Tests for the alkalies and ammonia are made on the water solution in the usual way. Moisture. Weigh out rapidly 5 grams of the sample and dry to constant weight at 100 C. Report the loss as moisture. Loss on Ignition. Ignite strongly the residue from the moisture determination, cool in a desiccator and weigh. The loss indicates 310 TECHNICAL METHODS OF ANALYSIS organic and volatile matter. Previous qualitative tests will indicate whether a part of this loss is due to carbonate. Water-soluble Material. Digest 5 grams of the sample in 150 cc. of hot water. Filter and wash with hot water. Evap- orate the filtrate to dryness in a weighed platinum dish and dry to constant weight at 100 C. Acid-soluble Material. Transfer the residue on the filter paper from the water-soluble determination to the beaker con- taining the remaining residue, and treat the whole with 125 cc. of hot HC1 (1 : 3). Filter and wash. Evaporate the filtrate and washings to dryness in a weighed platinum dish and dry at 100 C. to constant weight. Iron Oxide and Alumina. If the acid-soluble material is high, the presence of Fe 2 O 3 , A1 2 3 , BaCO 3 or Ba 3 (PO 4 ) 2 is indicated. If phosphate is absent, take up the acid soluble residue with a little HC1 and 100 cc. of water. Make the solution slightly alkaline with NE^OH; filter, wash with hot water, and ignite the precipitate, finally with a blast lamp; cool in a desiccator, and weigh as Al 2 O 3 +Fe 2 O 3 . If phosphate is present, add to the acid solution a known amount of Fe 2 3 in the form of chloride or sul- fate (see page 25), then make slightly -alkaline with NHiOH and proceed as above. Subtract the known amount of Fe 2 Os added. The resulting figure wiU be Fe 2 O 3 +Al 2 O 3 +P 2 5 in the sample. Determine the P 2 Os on a separate portion by the molybdate method as described below, and subtract to get Fe 2 O 3 -t-Al 2 3 . Barium Carbonate (and Phosphate). To the filtrate from the above add a slight excess of HC1, heat to boiling, and then add slowly an excess of dil. H 2 SO 4 . Boil until the precipitate settles clear; filter, ignite, cool in a desiccator and weigh as BaSO 4 . Calculate to BaC0 3 or Ba 3 (PO 4 ) 2 according to the qualitative analysis. CALCULATIONS. BaSO 4 X 0.8456 = BaCO 3 . BaSO 4 X 0.8599 = Ba 3 (PO 4 ) 2 . Lime and Magnesia. These are not often present in com- mercial blanc fixe but may be determined in the filtrate from the barium carbonate above in the usual way. Lead Sulfate. If the qualitative tests showed lead, boil with saturated ammonium acetate solution the sulfate residue from WOOD, PAPER AND PAPER-MAKING MATERIALS 311 the acid soluble treatment. Filter and wash. Add a few drops of acetic acid and then a slight excess of bichromate solution. Heat to boiling, filter through a weighed Gooch crucible, wash with hot water, dry at about 100 C., set the Gooch crucible in a larger platinum crucible, ignite gently and weigh as PbCr0 4 . Cal- culate to PbSO 4 . CALCULATION. PbCr0 4 X 0.9383 = PbSO 4 . Barium Sulfate. Dry the residue from the ammonium acetate treatment (or, if Pb is absent, the residue from the acid soluble treatment) to constant weight and calculate its percentage of the original sample. Then weigh out 1 gram of this dry residue, mix with 5-7 grams of pure Na2COs and fuse in a platinum crucible to a thin liquid. Digest the fusion with hot water until completely disintegrated; filter and wash thoroughly with hot water. Make the filtrate slightly acid with dil. HC1, dilute to 350 cc. and heat to boiling. Then add 25 cc. of 10% BaCl 2 solution, drop by drop from a pipette. Boil for five minutes, let stand overnight, filter, wash with hot water, ignite and weigh as BaSO 4 . Calculate to the original basis. NOTE. In case the complete analysis is not desired, determine BaSO 4 as above using 1 gram of the dry material from the moisture determination; then wash the residue from the fusion with hot water into a beaker and treat the filter paper with warm dil. HC1, collecting this in the same beaker. The whole precipitate should dissolve completely in HC1. Dilute this solution to about 350 cc., boil gently to expel the CO 2 and add a slight excess of dil. H 2 SO 4 , drop by drop; let stand overnight. Filter out the BaSO 4 , ignite and weigh as usual. These two determinations give the total SO 3 as BaSO 4 and the total Ba as BaSCh; and in a pure blanc fixe they should check closely. An excess of total Ba indicates soluble barium salts [in the absence of Ba ; (PO 4 ) 2 ] and an excess of total SO 3 indicates other sulfates. Barium Phosphate. Boil 2 grams of the original material with 150 cc. of water, filter and wash by decantation; return the residue to the original beaker and boil with 100 cc. of water and 5 cc. of cone. HNOs. Filter and wash with hot water. To the warm filtrate add an excess of ammonium molybdate solution. Let stand until the precipitate settles clear, then filter and wash with a 2% HNOa solution. Dissolve the yellow precipitate from the filter paper with dilute NELiOH. The total volume should be between 50 and 100 cc. Add 25 cc. of magnesia mixture, stir 312 TECHNICAL METHODS OF ANALYSIS well and let stand 1 hour. Filter on a weighed Gooch crucible and wash with 5% NHjOH. Ignite strongly and weigh as Mg 2 P 2 O 7 . Calculate to P 2 O 5 and to Ba 3 (P0 4 ) 2 . CALCULATIONS. Mg 2 P 2 O 7 X 0.6379 = P 2 5 . P 2 O 5 X4.2384 = Ba 3 (P0 4 ) 2 . Mg 2 P 2 O 7 X 2.7038 = Ba 3 (P0 4 ) 2 . General Calculations. If the material contains carbonates and phosphates, the acid-soluble Ba may be present in both forms and the amounts can be calculated from the total acid-soluble Ba and from the P 2 C>5, figuring the latter to Ba 3 (PO 4 ) 2 as above described and then the excess of Ba to BaC0 3 . If neither phosphate nor carbonate is present, the soluble Ba is probably in the form of chloride or sulfide. The Pb should be calculated to PbSO 4 . The Fe and Al may be present as sulfates but are generally reported as oxides. CALCULATION. Ba 3 (PO 4 ) 2 X 1.1630 = BaS0 4 . BaSO 4 X 0.8456 =BaCO 3 . REFERENCE. This method is based on that described by A. B. Hutchins of the Ansco Co., Research Laboratory, published in Paper, 20, No. 13, page 11 (1917), somewhat modified by experience in this laboratory. BLEACH (BLEACHING POWDER) Available Chlorine. Mix the whole sample quickly and thor- oughly, discarding the outside (top layer) which has lost more or less chlorine. Weigh out from a weighing bottle about 10 grams into a porcelain mortar, add a little water and rub the mixture to a smooth cream. Then stir in more water with the pestle, let settle for a few moments and pour off through a funnel into a liter flask. Again rub up the sediment with water and pour off as before. Repeat the operation until the whole of the material has been conveyed into the flask without loss and the mortar washed clean. Then fill the flask to the mark with water, mix well and pipette out immediately, without allowing the material to settle, 50 cc. into a beaker. Add 50 cc. of 0.1 N arsenious acid. Stir the mixture well and then titrate back the excess of arsenious acid with 0. 1 N iodine, adding a little starch indicator. Duplicate determinations should always be made. From the amount of As 2 3 consumed calculate the available Cl in the bleach. WOOD, PAPER AND PAPER-MAKING MATERIALS 313 CALCULATION. 1 cc. 0.1 N As 2 O 3 = 0.003546 gram Cl. NOTES. (1) It is generally specified that a good bleaching powder should contain not less than 35% of available chlorine. (2) A good bleach when mixed with water in the proportion of 6 parts of bleach to 94 of water and stirred thoroughly for about twenty minutes, should settle clear in not more than an hour. (3) Bunsen's method for bleaching powder is as follows: Pipette 25 cc. of the bleach solution prepared as above described into a beaker and add an excess of a 10% solution of KI. Dilute to about 100 cc., acidify with dil. HC1 and titrate the liberated iodine with 0.1 N thiosulfate, adding a little starch toward the end. The titration gives the amount of available chlorine directly. This method, however, is not accurate where calcium chlorate is present, since it liberates the chlorine in the latter which is of no value in bleaching. If the amount of chlorate is desired, the difference in chlorine obtained by the two methods represents the chlorine due to chlorate. REFERENCE. Sutton: " Handbook of Volumetric Analysis," 10th edition, page 177. BLEACH CONSUMPTION OF PULP Weigh accurately 5 grams of the pulp, which should be in the air-dry condition. Cut into pieces about an inch square and place in a suitable container, such as a granite-ware cup, together with sufficient cold water. Thoroughly disintegrate the pulp by means of an egg beater, taking care not to lose any by spattering. The disintegration should not take more than five minutes, and should be thorough, or there will be lumps in the test sheets subsequently made. Transfer the pulp to a funnel containing a perforated porcelain plate, re-running the first portions passing through, which usually contain a little fiber. Remove excess water with suction. Trans- fer the pulp to a we'ghed 6-ounce bottle and add about 25 cc. of distilled water. Then add the requisite number of cc. of stand- ardized bleach solution from a burette. (See Note (1) below.) Place the bottle on the scales and add sufficient distilled water to make the ratio of air-dry pulp to water 1 : 20. In other words, the weight of the contents of the bottle should be 105 grams, which includes pulp, bleach and water. Stir the contents thoroughly with a glass rod, stopper the bottle and place in a water bath maintained at 40 C. Let it remain for four hours, stirring fre- quently. Remove the bottle, and filter the contents through the 314 TECHNICAL METHODS OF ANALYSIS funnel and porcelain plate originally used. Refilter the first runnings if necessary. Wash several times, using suction, and test the contents of the filter flask for bleach by the addition of a crystal of KI. If no yellow color develops, the bleach was com- pletely consumed. If a color shows, add acetic acid, and titrate the liberated iodine with 0.1 N thiosulfate, using starch. The solution should, of course, be cool. The back titration, if any, expressed in percent of bleaching powder should be subtracted from the bleach originally added. Transfer the bleached pulp to a crock or small pan, add a small quantity of water, stir well, and make up hand sheets on a brass wire mould. Compare these sheets and ascertain the least amount of bleach required to pro- duce a good white. The strength of the original bleach liquor is determined as follows: Pipette off 1 cc. direct (or an aliquot representing 1 cc.) with a standardized pipette into a flask containing about 50 cc. of water and a few crystals of KI. Acidify strongly with acetic acid and titrate the liberated iodine with 0.1 N thiosulfate. The addition of starch is unnecessary. The end point can be deter- mined by the disappearance of the yellow color. 1 cc. 0.1 N thiosulfate = 0.00355 gram chlorine. Assuming bleaching powder to contain 33.3% of active chlorine, 1 cc. 0.1 N thiosulfate = 0.0 1065 gram bleaching powder. Express results as per cent of bleaching powder consumed by the air-dry pulp. NOTES. (1) It is highly desirable that 2, or better 3, parallel tests should be carried out simultaneously on the same pulp. Graduated amounts of bleach should be used, say 10%, 15%, and 20%. If preferred, a single test of say 20% of bleach may first be run; and, depending upon the result obtained, 2 or more may be subsequently run, to determine more definitely the bleach required to produce a good white. (2) It is essential that the pulp density or ratio of pulp to water be always constant. It should be emphasized that results are without value unless all conditions be maintained strictly uniform. (3) Care should be taken in the making of the test sheets that they be made as nearly the same thickness as possible. It is extremely difficult to compare test sheets for color unless the pressure used in drying (i.e., the surface), density, thickness, and in fact all conditions be carefully regulated. (4) The assumption of 33^% active chlorine in bleaching powder is pos- sibly slightly low. It is customary, however, to take this first figure in all our experimental bleaching operations because the amount of chlorine extracted WOOD, PAPER AND PAPER-MAKING MATERIALS 315 from bleaching powder in ordinary practice is in all cases less than the total available chlorine in the dry powder. With 35% available chlorine in the dry powder, about 33 % would be actually present in the pulp mixture. (5) The above method is the result of experimental work done in this laboratory by R. B. Roe. CASEIN General. Always note the appearance and odor of the sample. A good casein should have a clean, pleasant odor and should be neither moldy, musty nor rancid. The color should be light yellow or cream color and there should be no dafk brown portions or black specks. Bright orange particles indicate that the casein has been burned in drying. Moisture. Weigh 5 grams into a shallow porcelain dish and dry to constant weight at 100-105 C. The temperature should not be allowed to run over 105 C. NOTE. The moisture is not usually determined in casein. On ground caseins, however, the maximum allowable moisture is generally considered 12%. Ash. Weigh 5 grams into a porcelain dish and ignite until the ash is white or grayish white (casein must not be ashed in platinum since it attacks this metal). Cool in a desiccator and weigh. NOTE. The ash of naturally soured caseins generally runs between 5 and 8.5%, the ash of acid casein not over 6%. Natural caseins burn to a white ash with greater ease than acid caseins. Alkali. Add about 5 cc. of distilled water to the ash and warm. Then add 1 or 2 drops of phenolphthalein indicator. If the ash is alkaline, titrate it with 0.1 N HC1 until colorless, then add two drops of methyl orange and complete the titration to a pink color. Calculate to Na2O. CALCULATION. 1 cc. 0.1 N HC1 = 0.0031 gram Na 2 O. NOTE. The ash of a pure casein should not be alkaline to phenolphthalein or to methyl orange. If it shows alkalinity, test it qualitatively for borax, sodium carbonate and sodium phosphate. Starch (Qualitative Test). Warm a portion of the sample with distilled water. Cool, and add a few drops of very dilute iodine solution. A blue color indicates starch. 316 TECHNICAL METHODS OF ANALYSIS NOTE. If added alkali has been found in the casein, make the solution slightly acid before adding the iodine solution. Solubility (" Cutting Test ") Weigh out 50 grams of the ground casein into a 350 cc. beaker and add 7.5 grams of powdered borax (equivalent to 15 per cent of the weight of the casein). Add 250 cc. of water at a temperature of 70 C., and heat for fifteen minutes at this temperature on the water bath, stirring constantly. High-grade caseins will dissolve completely. Inferior samples will show more or less lumps and mineral impurities. If the lumps are brown or orange color, the casein has probably been burned in preparation. Gsit and dirt will settle to the bottom and the approximate amount should be noted. In carrying out this test always run at the same time a standard high-grade casein for comparison. The " Muriatic Flakeless " casein of Innis Speiden & Co. is a satisfactory standard. Italian casein is also generally a high-grade casein but gives a thinner solution than domestic. Solutions of acid casein are usually more viscous than those of natural caseins, muriatic caseins being specially viscous. Most of the foreign caseins give thinner solu- tions than domestic caseins. Clay-carrying Capacity. Dilute the casein solution obtained above with 250 cc. of hot water. 10 cc. of this solution will then contain 1 gram of casein. Keep the solution hot during use. Weigh out 100 grams of bone-dry clay (standard D. Y. coating clay, which has been dried at 105 C., is a suitable clay for this use) and mix thoroughly with 65 cc. of water in a small sauce pan or casserole. When the mixture is perfectly smooth and free from lumps, add 50 cc. of the casein solution. This is equivalent to 5% of casein on the dry weight of the clay. Stir until smooth. Using a paint brush about j inch wide, remove a brush-full of the mixture and spread it evenly on one end of a strip of paper which should be about 2 feet long and from 7-9 inches wide. A smooth-surfaced, fairly heavy wrapping paper is suitable. Add 10 cc. more of the casein solution to the clay and mix thoroughly. This will make a mixture containing 6% of casein on the weight of the clay. Stir this mixture until smooth and brush on the paper as before. Continue in this manner until mixtures have been made and applied to the paper which contain 5, 6, 7, 8, 9, 10, 11, 12, 13, and 14% of casein, respectively, on the WOOD, PAPER AND PAPER-MAKING MATERIALS 317 basis of the weight of the clay. Let the coated paper dry over night at room temperature, or for about three hours at 130 F. (55 C.) When dry, determine the " critical clay-carrying point " in the following manner: Soften a piece of sealing wax by heating until it can be easily worked with the fingers for a distance of 0.5 inch up the stick. A convenient method of accomplishing this is to hold the stick of sealing wax about 0.5 inch above an electric hot plate until soft and then let it cool for about fifteen seconds. When this condi- tion is attained, hold the sealing wax 0.5 inch above the hot plate for fifteen seconds, remove it and press it down for an instant on the strip of clay-casein coating which contains 5% of casein. Then pull up sharply and observe whether the coating alone comes off or whether some fiber from the sheet adheres to it. Heat the wax as before for fifteen seconds, press down on the strip of coating which contains 6% of casein and pull up as before. Repeat until that coating is reached which adheres strongly enough to the paper so that some fibers pull away from the sheet when the wax is pulled up. This is the " critical clay-carrying point." Repeat in the same manner, starting from the 14% casein coating and run- ning down in the opposite direction, in order to check the critical point. If they do not check, the operation should be repeated. Report the clay-carrying capacity as the number of parts of clay which one part of casein will carry. This is obtained by dividing 100 by the percentage of casein (on the basis of the clay) in the coating where the critical point is found. For example: If the critical point is found in the coating containing 9% casein on the basis of the clay, the clay-carrying capacity is 100-5-9 = 11.1. NOTES. (1) Always carry out parallel tests on a standard sample of casein for comparison. (2) The clay-carrying capacity of a high-grade casein is about 11. Acidity. Weigh 20 grams of casein into a casserole, and add 80 cc. of water. Warm, and titrate with 0.5 N KOH, running in a little at a time with alternate warming on the steam bath until the end point is reached. Use strips of litmus paper as an outside indicator and apply a drop at a time of the casein solution to the blue paper until the red color is faint. Then apply to both blue and red paper until the end point is reached. Report the results 318 TECHNICAL METHODS OF ANALYSIS in terms of cc. of normal caustic solution required to neutralize 1 gram of casein. NOTE. (1) This last determination is not usually required. (2) The tests for solubility and clay-carrying capacity are of especial value in testing casein for use in coating paper. CLAY FOR PAPER FILLER General. Clay or kaolin for use as a filler in high grade papers should be pure white and free from artificial bluing. It should not have an appreciably gritty feel when rubbed between the teeth. The best clays will not show more than 1-2% residue by the flotation test nor more than a few tenths of 1% grit (generally less than 0.25%) by the 200-mesh sieve test. They should not contain an excessive amount of free moisture when shipped. Moisture. Dry 10 grams to constant weight at 100-105 C. The loss in weight is considered moisture. Grit. (A) FLOTATION TEST. Measure a depth of 2 inches from the bottom of a 500 cc. beaker and make a mark on the beaker to indicate this height. Weigh 20 grams of the clay into this beaker, mix thoroughly with water and fill up to the mark. Let settle for exactly one minute and pour off the milky water. Repeat the process until the supernatant water can be poured off prac- tically clear at the end of a minute. Place the beaker on the steam bath until perfectly dry, brush out the settled grit into a bal- anced watch-glass with a camel's hair brush and weigh it. (B) 200-MESH SIEVE TEST. Place 10 grams of the clay in a 200-mesh sieve and wash by means of a slow stream of running water and finally with a wash bottle until all the clay has been washed through the sieve. The residue which does not pass through is then washed into a small beaker and, after evaporating off the water on the steam bath, is brushed into a balanced watch- glass and weighed. NOTE. In both grit tests it is permissible and necessary to break up any lumps of clay with a rubber " policeman," but no hard object should be used for this purpose which might crush particles of the grit. Color. Spread out a small amount of the clay (preferably dried at 100 C.) on white paper by the side of, and in contact with, a similar amount of a standard sample. Each should be pressed WOOD, PAPER AND PAPER-MAKING MATERIALS 319 down with a spatula to give a smooth surface. The color is then compared. When the clays are of the same color it is impossible to see any line of demarcation between them. If a permanent record is desired, the color should be measured by a colorimeter or tintometer. Artificial Coloring. In one of two similar white porcelain dishes place a measured amount of freshly prepared, saturated lime water and in the other dish an equal amount of clear, dis- tilled or spring water. Then into each of these liquids dust, from the end of a knife or spatula, a little at a time, equal amounts of clay. If the clay has been artificially blued, the lime water will remove the bluing. After letting stand for a few moments the excess liquid should be siphoned off and the moist clay examined. If the clay has been blued, the residue from the lime water treat- ment will show the original color, which will be quite different from the color shown by the moist clay treated with water alone in the other dish. NOTE. The so-called " turpentine test " for the presence of artificial bluing in clay is unreliable, since certain natural clays when tested with turpentine behave similarly to those which have been artificially blued. CROWN FILLER General. Crown filler, used as a loading material or paper filler, is a special form of hydrated calcium sulfate. Total Moisture. Ignite about 10 grams over a Tirrill burner for thirty minutes. The loss in weight is considered the total moisture in the sample. Combined Moisture. Calculate the amount of water which will combine with the residue of anhydrous CaSCU, as found above, to form the crystallized form, CaSCX 2H2O, by multiplying the weight of residue by the factor 0.2646. Mechanical Moisture. Subtract the combined moisture from the total moisture to obtain the free or mechanical moisture. NOTE. This same procedure may also be applied to Gypsum or Terra Alba. 320 TECHNICAL METHODS OF ANALYSIS GLUE Moisture. Weigh 2-3 grams of the glue into a platinum dish, add -10-15 cc. of hot water, mix thoroughly with a small stirring rod and rinse the rod off into the dish with a wash bottle. Evap- orate on the steam bath to dryness and then dry to constant weight at 105 C. The loss in weight of the original glue repre- sents the moisture. A good glue contains not less than 8% nor more than 16% of moisture. Low moisture indicates over- heating in the manufacture. Ash. Ignite the above residue in the platinum dish. Use a low heat at first, as too rapid burning will make it very difficult to get rid of the last of the carbon. Then raise the heat and ignite to constant weight at the full heat of a Tirrill burner. Report the residue as ash. In making the ash determination, note whether the ash has fused or not. The ash of bone glue fuses and the aqueous solution is neutral, containing phosphates and chlorides. The ash of hide glues does not fuse owing to traces of lime which render the aqueous solution slightly alkaline. This solution is also free from phosphates or chlorides. The ash will vary from 2-8%, according to the quality of the glue. The ash may also show the presence of alum, which is sometimes added to impart a fictitious " body " to glues. Acidity or Alkalinity. Dissolve 1 gram of the glue in 500 cc. of distilled water. Add a few drops of phenolphthalein indicator and, if alkaline, titrate with 0.5 N acid. Calculate to Na2O. CALCULATION. 1 cc. 0.5 N acid = 0.01550 gram Na2O. In case the glue is acid instead of alkaline, titrate the total acidity with 0.1 N NaOH and phenolphthalein. It is also often necessary to make a separate determination of the volatile and organic acids in addition to titrating the total acidity. For this determination suspend in a stoppered flask 50 grams of the sample with 80 cc. of cold distilled water for about ten hours (or over- night) . Distill off the volatile acids by means of steam and collect the distillate in a 500 cc. graduated cylinder. After about 100 cc. have distilled add a few drops of phenolphthalein. If the distillate is found to be alkaline it is not necessary to proceed further with WOOD, PAPER AND PAPER-MAKING MATERIALS 321 the distillation. Otherwise, when 300 cc. have come over, stop the distillation and titrate the distillate with 0.1 N NaOH and phenolphthalein. This volatile acidity represents HC1 and H 2 SOs and should not exceed 0.2% for a good glue. Report as " volatile acidity, calculated as sulfurous acid, H^SOs." Subtract from the original titration of the glue the titration required for the volatile acids and calculate the difference as H2SO4. Report as " non-volatile acids calculated as sulfuric acid, H 2 SO 4 ." CALCULATIONS. 1 cc. of 0.1 N NaOH = 0.0041 gram H 2 SO 3 . 1 cc. of 0.1 N NaOH = 0.0049 gram H 2 SO 4 . Active Sulfur. When glue is to be used for sizing anti-tarnish papers, it is desirable to know the amount of active sulfur present, i.e., sulfur in the form of sulfides or sulfites. The determination of this is conducted in exactly the same way as the determination of the volatile acidity above described, with the exception that 5 cc. of syrupy HsPO4 should be added to the glue solution just before distilling with steam and the distillate should be collected in Br water. After distillation, boil off the Br from the distillate, make acid with a slight amount of HC1 and precipitate at boiling temperature with BaCl 2 solution. Filter, ignite and weigh the BaSO4 and calculate to SO 2 . Report the results as " active sulfur, calculated as sulfur dioxide." CALCULATION. BaS0 4 X 0.274 = S0 2 . Jelly Test. A very good idea of the value of a single glue sample, or the comparative values of different samples, may be gained by testing the stiffness of the jellies formed. The jellies from the unknown glues are compared with jellies from the stand- ard varieties of glues as made by the Peter Cooper Glue Factory.* The following grades should be used as standards: No. 1, No. 1-X, No. 11, No. If, No. 1J, No. If, No. If, and No. 2. The No. 1 is a very high grade glue and No. 2 a cheap glue. (The Cooper factory also puts out two higher grades than No. 1, namely, No. 1 extra and A extra. Certain other manufacturers also claim to make a glue even superior to A extra.) The following will give an idea of the relative values of the glues as purchased in large quantities: * Main works at Gowanda, New York. 322 TECHNICAL METHODS OF ANALYSIS 1912 Jan. 1920 Cents per Pound Cents per Pound No. H About 14 30-35 No. If About 12 24-28 No. 1| 8-9 22-25 No. 2 6-7 18-21 The method to be used in the jelly strength test is as follows: First, decide as to whether the unknown sample is probably a low-grade or a high-grade glue and then select several Cooper standards that will include the grade of the unknown sample. Weigh out 50 grams of each sample of air-dried glue, place in a 300 cc. beaker and add 200 cc. of cold water. Let the glues soak in the cold water until soft throughout. In the case of a ground glue this will require not over one hour. In the case of sheet glues, several hours will be necessary. When thoroughly soaked, transfer all the samples at the same time to a steam bath which has been previously regulated. Let the mixtures heat to 160 F. and hold at this temperature until thoroughly dissolved, stirring them from time to time to prevent a skin forming on the top. All the samples should be cooked the same length of time. Fit a 6-inch funnel with cheese cloth and filter the glue solutions into jelly glasses to within 0.5 inch of the top. Set aside to cool, taking care that the samples are not disturbed during jellying. In hot weather it is generally necessary to set the glasses in a pan of water to cool them; otherwise the jellies will not be stiff enough for satisfactory tests. This procedure may also be used when results are desired quickly. It is most convenient to cook the samples late in the afternoon and allow them to set overnight. The glasses containing the unknown and standard samples are placed in a row so that the labels cannot be seen. Using the third ringer of the left hand, compare the jellies for firmness or strength and rearrange the glasses until the jellies finally stand in order of their firmness. Then examine the labels. The standard samples should, of course, have been arranged in correct order in reference to each other. From the way the unknown samples are distributed among the standard samples they can be rated as to WOOD, PAPER AND PAPER-MAKING MATERIALS 323 quality. (The grades of glue ordinarily met with in paper mill work are those from No. 1 J to No. If.) Report as " Peter Cooper standard, No " Viscosity. Dissolve 50 grams of the glue in 200 cc. oi water by means of heat. Determine the number of seconds which is required for a 100 cc. Dudley pipette to deliver its volume of water at approximately 180 F. Heat the water to about 190 and fill the pipette somewhat above the mark. Have a rubber tube with a pinch cock on top of the pipette and by means of the pinch cock set the water at the mark. Then with a stop watch in the left hand open the pinch cock suddenly and at the same time, start the stop watch. Note the length of time required for the water to run from the upper to the lower mark. Repeat the operation 2 or 3 times and take the average. The opening of the pipette should be so adjusted that it will deliver 100 cc. of water in approximately 35 seconds. Determine the viscosity of the glue solution in the same way, i.e., heat the solution slightly above 180 F.; fill the pipette; set it carefully on the mark and determine the length of time it requires for the glue solution to run from the upper to the lower mark. Run at least 2 determinations and take the average, if they agree reasonably well. Divide this time by the time required for water and report the result as the specific viscosity of the glue solution at 180 F. Grease. Make a strong water solution of aniline blue. Use 25% solutions of the different samples of glue to be tested. To 25 cc. of the glue solution add 2 cc. of the dye, and with a 1.5 inch flat cameFs-hair brush draw a smooth streak of the dyed glue across wrapping paper. White spots indicate grease. A com- parison of the white spots will indicate the relative amount of grease in the different samples. The grease may be determined quantitatively by extracting 5-10 grams of the ground and dried material with anhydrous ether in a Soxhlet extractor, evaporating the ether and weighing the residue. 324 TECHNICAL METHODS OF ANALYSIS LIME General. There are two general classes of lime: (1) quicklime and (2) hydrated lime. (1) QUICKLIME. Of this there are two grades: (A) selected lime, which should be well burnt lime picked free from ashes, core clinker, or other foreign material; (B) run-of-kiln, which should be well burnt lime but is taken without selection. Quicklime is generally shipped in two forms: (1) lump lime, the size coming from the kiln, and (2) pulverized lime, which is lump lime reduced to pass a 0.25 inch screen. Chemically, quicklimes are divided into four types: 1. High Calcium Lime (over 90% CaO). 2. Calcium Lime (85-90% CaO). 3. Magnesia Lime (10-25% MgO). 4. Dolomitic Lime (not under 25% MgO). The Am. Soc. for Testing Materials specifications require that selected quicklime shall contain not under 90% CaO and not over 3% C02) run-of-kiln quicklime shall contain not under 85% CaO+MgO and not over 5% CO 2 . (2) HYDRATED LIME. Hydrated limes are divided chemically into the same four types as the quicklimes. It is important in draw- ing the sample for testing that it should be taken from the surface to the center of the package and at least 3% of the packages should be sampled. The sample should be stored in an air-tight con- tainer. The Am. Soc. for Testing Materials specifications demand that the non-volatile portion of hydrated lime shall contain not under 92% CaO + MgO, and hydrated lime shall contain not over 5% CO2 and sufficient water to fully hydrate the CaO present. A pat test is further specified as follows: Pat Test. A 3-inch pat 0.5 inch thick in the center, tapering to a thin edge, shall be made on a clean glass plate from a paste composed of equal parts by weight of hydrated lime and volume- constant Portland cement with only sufficient water to make the mixture workable. This pat, after hardening twenty-four hours in moist air, shall, when exposed in a convenient manner to steam above boiling water in a loosely closed vessel for five hours, WOOD, PAPER AND PAPER-MAKING MATERIALS 325 show no signs of popping, checking, cracking, warping or disinte- grating. Loss on Ignition. Ignite 1 gram of the powdered lime in a platinum crucible, gently at first, and then over the blast lamp, until the weight is constant. The loss in weight is moisture and CO2 and is expressed as " Loss on Ignition." Insoluble Matter. Transfer the ignited lime to a beaker and pour over it 50 cc. of water and enough dil. HC1 to dissolve the sample. Boil for five minutes, filter, wash with hot water, ignite and weigh. Iron and Aluminum Oxides. To the filtrate add 5 cc. of cone. HC1 and then NELtOH in very slight excess. Keep the solution near the boiling point until the odor of NHs is barely perceptible. Filter off the iron and aluminum hydroxides while hot, collecting the filtrate in a 250 cc. graduated flask. Wash with hot water, dry, ignite in a platinum crucible, blast, cool in a desiccator and weigh as Al2O3+Fe2O3. Lime. Make the filtrate from the preceding determination up to 250 cc. Mix thoroughly and pipette out 100 cc. into a 350 cc. beaker, heat to boiling and add slowly a boiling solution of 3 grams of (NH4)2C2O4 in water. Continue the boiling two or three minutes and let the precipitated CaC2C>4 settle for one-half hour. In the case of magnesia limes, decant through a filter, redissolve the CaC204 in the beaker and from the filter with HC1, washing the filter 5 times and finally washing NH^OH solution through it. Dilute the solution to 250 cc., bring to boiling, add 1 cc. of (NEU) 20264 solution and NKtOH in slight excess. Boil for two or three minutes and set aside for one-half hour. Filter off the CaC204 on the filter first used, wash with hot water and ignite in a platinum crucible over a Tirrill burner, and finally over a blast lamp to constant weight. Cool in a desiccator and weigh as CaO. Divide this weight by 0.4 and multiply by 100 to obtain the percentage of lime (CaO) in the original . sample. Since CaO absorbs moisture from the air, it should be weighed as rapidly as possible. If the lime contains only a small amount of MgO, one precip- itation is sufficient. NOTES. (1) If it is desired to complete the analysis in as short a time as possible, a portion of 50 cc. of the filtrate from the Iron and Aluminum 326 TECHNICAL METHODS OF ANALYSIS Oxides determination should be precipitated in the usual way with excess of (NH 4 ) 2 C 2 O4. Boil for about five minutes and let the CaC 2 O 4 settle clear. Decant through a qualitative filter and cool (with ice water if possible) . Add ammonium phosphate in large excess and 5-10 cc. of cone. NH.OH. Stir rapidly with a rubber "p n ' cern an." From the amount of precipitate thus formed one can judge whether the lime contains sufficient magnesia to require a double precipitation or not. For accurate work, if there is more than a slight amount of magnesia, a double precipitation should be carried out, using a fresh 100 cc. aliquot. (2) It is allowable and sometimes preferable to titrate the CaC 2 O 4 instead of igniting it. Take a portion of the filtrate from the iron and alumina determination corresponding to 0.2-0.25 gram of the material and precipitate the CaC 2 O4 as above described. In a 400 cc. beaker place about 125 cc. of distilled water and add 5-7 cc. of cone. H 2 SOi. Drop into this the moist filter paper containing the CaC 2 O 4 and heat to about 70 C. Stir to effect decom- position but avoid excessive disintegration of the paper. Titrate the solu- tion with constant stirring with 0.1 N KMnO 4 until a permanent pink color forms. Calculate to CaO. CALCULATION. 1 cc. 0.1 N KMnO 4 = 0.002804 gram CaO. Magnesia. Acidify the filtrate from the CaC2C>4 precipitate (or the combined filtrates, in case of a magnesia lime) with HC1 and evaporate until the salts begin to crystallize. Dilute until the salts are again in solution. Add a volume of dil. NIL^OH equal to one-third the volume of the solution. Chill the solution and add slowly and with constant stirring 2 grams of Na2HPO4 or (NH4)2HPO4 dissolved in 10 cc. of water. Let stand until com- pletely precipitated. Four hours are usually sufficient, but if possible, it is best to let the solution stand overnight. If the analysis is urgent, stir for one-half hour and the precipitation will be complete. Filter through a weighed Gooch crucible (pre- viously ignited) and wash with a mixture of 1 part NEUOH (sp. gr. 0.96), 1 part alcohol and 3 parts water. Dry at 105 C. in the oven. Ignite slowly and finally at the highest heat of the Tirrill burner until the weight is constant, cool in desiccator and weigh as Mg2P2O?. CALCULATION. Mg 2 P20 7 X 0.3621 = MgO. Divide the weight of MgO by 0.4 and multiply by 100 to obtain the percentage of MgO in the original sample. NOTES. (1) The sample weighed out for analysis should be finely ground and representative of the whole sample. (2) The amounts of (NH 4 ) 2 C 2 O 4 (3 grams) and sodium or ammonium phos- WOOD, PAPER AND PAPER-MAKING MATERIALS 327 phate (2 grams) as given in the procedure are ample to insure complete pre- cipitation of the lime and magnesia. (3) The residue insoluble in acid may contain some calcium and magnesium combined with the silica, but for all commercial purposes it is sufficient to report it as " Insoluble Matter." LIMESTONE General. Since limes are made from limestones by burning off the C02, the analyses of the two are in general quite similar. For the determination of Loss on Ignition, Insoluble Matter, Iron and Aluminum Oxides, Lime, and Magnesia, follow the same procedures as given above in the method for Lime. Final Calculations. Calculate the MgO and the CaO also to carbonates and include these figures in the final report. MgOx2.0915 = MgCO 3 . CaO X 1.7849 = CaC0 3 . NOTES. (1) If it is desired to determine CO 2 directly, this may be done in an alkalimeter as described under White Lead on page 212. For very accurate results, however, the CO2 should be set free with acid and absorbed and weighed in potash bulbs or soda lime tubes. (2) In case phosphorus is desired, dissolve 10 grams in dil. HNOs and determine the phosphorus by the molybdate method as described on page 529. ROSIN General. Rosin is graded according to color. The grades are as follows: WW, WG, N, M, K, I, H, G, F, E, D and B. The WW is the best and palest grade. B is the cheapest and darkest grade. Grades G, F, and E are most frequently used for paper making. Yaryan Extract Rosin grades between E and F and is ruby red in color. Grade. To determine the grade of rosin a set of standard cubes of rosin of the various grades must be available. The rosin under test is cast into a cube in a mold of sheet aluminum and compared as to color with the standards by looking through the cubes toward the light. Care must be taken to heat the rosin only just enough to pour, since overheating darkens the color. Dirt and Foreign Matter. Unless the rosin is quite dirty, no quantitative estimation is necessary. In case a quantitative 328 TECHNICAL METHODS OF ANALYSIS estimation is desired, dissolve 25 grams of rosin in warm alcohol; filter through a tared filter paper; wash with alcohol, dry and weigh the residue. Saponification Number. Weigh 2 grams of powdered rosin into an Erlenmeyer flask of 300 cc. capacity. Add 25 cc. of 0.5 N alcoholic KOH and boil for 2 hours under a reflux condenser. Shake the flask frequently with a swirling motion to prevent the rosin from sticking to the sides of the flask above the liquor line. Cool and titrate the excess KOH with 0.5 N acid and phenol- phthalein. Calculate the milligrams of KOH consumed per gram of rosin. This is the saponification number. In each case run a blank on the KOH solution by boiling 25 cc. of the solution for two hours and titrating in exactly the same manner that the saponification proper is carried out. CALCULATION. 1 cc. 0.5 N KOH = 28.06 mg. Acid Number. Dissolve 1 gram of powdered rosin in warm alcohol (neutral to phenolphthalein) ; cool and titrate the solution with 0.5 N KOH, using phenolphthalein. Express the result as milligrams of KOH consumed per gram of rosin. This is the acid number. It is sometimes customary to report the per cent of acid. This should be calculated as abietic acid. CALCULATION. 1 cc. 0.5 N KOH = 0.1512 gram abietic acid. Ester Value. The ester value is the difference between the saponification number and the acid number. Unsaponifiable Matter. Weigh about 5 grams of rosin and saponify by boiling two hours with excess of 0.5 N alcoholic KOH. Evaporate most of the alcohol, add about 100 cc. of water and extract in a separatory funnel with acid-free ether, exactly as in the determination of free rosin in rosin size (page 329). Ash. The determination of ash is seldom necessary. It is accomplished by igniting 5 grams in a platinum crucible to a white or light gray residue. Cool in a desiccator and weigh. Practical Sizing Tests of Rosin. It is sometimes desirable to make a practical sizing test of rosin in comparison with a rosin which is regarded as standard. For this purpose a small beating engine is desirable, although the work can be done by using a cream whipper. 25 or 50 grams (dry weight) of unbleached sulfite pulp are thoroughly disintegrated in the beater or cream whipper and 2% of rosin sizing added in the form of a thin milk. WOOD, PAPER AND PAPER-MAKING MATERIALS 329 After thorough mixing, 3% of a standard alum is added in solution and thoroughly mixed. The pulp is then thinned, made into hand sheets and the dried sheets are tested for ink penetration (page 354). Two sets of sheets are made, one with size made from a standard rosin and the other with size made from the rosin under test. The size is made by cooking a given weight of powdered rosin, in a container surrounded by boiling water, for four hours with that Weight of soda ash (Na2COs) which will yield a size containing 25% free rosin on the dry basis, i.e., with suf- ficient Na2COs to neutralize 75% of the free acid in the rosin. The water used in making up the size should be sufficient to give a finished thick size containing 30% dry matter. The thick size is diluted to a milk by stirring with water at 70 F. before adding to the pulp. ROSIN SIZE AND ROSIN SIZE MILK General. Rosin size consists essentially of a mixture of rosin-sodium soap with free rosin and more or less water. Thick, viscous sizes generally contain from 25-50% of water and hard, dry sizes from 1-10% of water. For paper mill use they are very much diluted with water and used in the form of rosin size milk, which may contain anywhere from 80-99% of water. ROSIN SIZE Free Rosin. Weigh out approximately 10 grams of size and mix with 30 cc. of water. (With solid or dry size, 6-8 grams will be sufficient.) Wash with as little water as possible into a 500 cc. separatory funnel, free from any trace of acid or alkali. Extract with 25 cc. of add-free ether. Draw off the watery layer into a second separatory funnel, extract this with a second portion of add-free ether and add the ether extract to that in the first funnel. Wash the combined ether extracts with two 25 cc. portions of water, adding the wash waters to the solution in the other funnel. Pour the washed ether extract into a weighed Soxhlet flask. Finally extract the water solution a third time with 25 cc. of ether, first using the ether to rinse out the funnel which contained the ether extracts. Draw off the watery layer into another separatory funnel, wash the third ether extract twice 330 TECHNICAL METHODS OF ANALYSIS with 25 cc. portions of water. Draw off the water each time into the funnel containing the soap solution. Pour the washed ether extract into the Soxhlet flask containing the main ether extract. Distill off the ether and dry the flask at not over 105 C. to con- stant weight. Cool in a desiccator and weigh the free rosin. NOTES. (1) It is especially important that all ether used in this deter- mination shall have been specially prepared by washing once with Na 2 CO 3 solution, once with water and then redistilled, to free it from all acid. It should be tested with a moist piece of sensitive blue litmus paper, which should not change color when completely submerged in it for fifteen minutes. (2) A convenient way to distill off ether is to connect the flask with a Soxhlet extractor and distill the ether up into the extractor. Moisture. Run the residue from the free rosin determination into a 250 cc. graduated flask (or a 500 cc. flask if necessary). Dilute to the mark and mix thoroughly. Pipette an aliquot equivalent to one-tenth of this solution into a weighed platinum dish, evaporate to dryness on the water bath, place in a water oven and dry at 105 C. to constant weight. Two hours' drying ought to be sufficient. Divide this weight by 0. 1 of the weight of the sample taken and multiply by 100. This gives the per cent of dry matter in the size, exclusive of free rosin. Add the per cent of free rosin as above determined, and subtract the sum from 100; the difference will be the per cent of water. Total Alkali. Ignite the residue from the moisture determina- tion until all carbonaceous matter is burned off. Dry in a desic- cator and weigh. Dissolve the residue in a few cc. of water and titrate with 0. 1 N acid and methyl orange. Calculate the titration directly to Na2CO3. This weight should check the weight of ash reasonably closely, unless the size contains insoluble or other foreign matter. Calculate the titration also to Na2O; divide the weight thus obtained by 0.1 of the original sample taken, and multiply by 100. This gives the per cent of Na2O in the size. CALCULATION. 1 cc. 0.1 N acid = 0.0031 gram Na 2 0. = 0.0053 gramNa 2 CO 3 . Combined Rosin. Pipette an aliquot representing four-fifths of the soap solution from the determination of free rosin into a separatory funnel and acidify with 10 cc. of dil. H2SO4 (1 : 5). Add 25 cc. of ether, shake well and let stand until the 2 layers are completely separated. Draw off the water solution into the WOOD, PAPER AND PAPER-MAKING MATERIALS 331 second separately funnel and wash the ether with two 25 cc. por- tions of water, drawing off the water into the second funnel and pouring the ether extract into a weighed Soxhlet flask. Rinse the first funnel with 25 cc. of ether into the second funnel. Shake well and draw off the water layer into the first funnel. Wash as above with two 25 cc. portions of water. Repeat once more. Evaporate the ether from the combined extracts as in the free rosin determination. Dry to constant weight at not over 105 C. Divide the weight obtained by 0.8 of the original weight of sample taken and multiply by 100 to obtain the per cent of combined rosin. NOTES. (1) The ether in this case does not need to be specially purified, though it should be free from any non-volatile residue. (2) The combined rosin thus obtained is weighed as rosin acids, whereas in the combined state it is actually present as anhydride. There are usually slight losses in manipulation and it is customary to report the combined rosin as actually weighed. Since rosin anhydride = rosin acid X 0.97, the usual rosin analysis should add up somewhat over 100%, depending upon the amount of combined rosin present. Free Carbonate. Weigh out 10 grams of size and dissolve in 200 cc. of add-free absolute alcohol. Let the solution stand eight to ten hours, or overnight if possible, protected from acid fumes and moisture. Filter on a weighed dry filter and wash thoroughly with absolute alcohol. Pour boiling water through the filter, and after cooling, titrate the aqueous solution with 0.1 N acid and methyl orange. Calculate to Na2CO3. CALCULATION. 1 cc. 0.1 N acid = 0.0053 gram Na 2 CO 3 . NOTES. (1) The filter need not be weighed if the insoluble matter is not to be determined. (See below.) (2) The determination of free alkali by this method is subject to a slight error on account of the solubility of Na2COa in alcohol. Tests made in this laboratory show that when 0.5 gram of anhydrous Na 2 CO 3 was allowed to stand sixteen hours in 200 cc. of 95% alcohol, 0.0075 gram went into solution. In the case of absolute alcohol 0.0050 gram dissolved. With 10 grams of ordinary rosin size and 200 cc. of absolute alcohol, the moisture in the size dilutes the alcohol to about 95% and the solubility of Na 2 CO 3 would cause results about 0.07% too high. Consequently, this figure may be used as a negative correction where greater accuracy is desired. Insoluble Matter. Any insoluble matter will be left on the weighed filter in the above determination of free Na2COs and may 332 TECHNICAL METHODS OF ANALYSIS be dried and weighed. In order also to determine whether this is mineral matter, it may be ignited and the mineral matter weighed. ROSIN SIZE MILK Specific Gravity. Take the sp. gr. with a pycnometer at a definite temperature; and for the various determinations pipette out definite volumes of the sample at the same temperature. Total Solids. Pipette 100 cc. into a 'weighed platinum dish, evaporate to dryness, then dry to constant weight at 105 C., cool in a desiccator and weigh. Total Alkali. Ignite the residue from the moisture deter- mination above and titrate with 0.1 N or 0.01 N acid and methyl orange (see Total Alkali above, under Rosin Size). Total Free Rosin. Pipette 100 cc. into a separatory funnel and extract with 50-75 cc. of acid-free ether without excessive shaking (see Free Rosin under Rosin Size above) . NOTE. If difficulty is experienced from emulsions, add 5-10 cc. of neutral grain alcohol or very cautiously apply suction to the top of the separatory funnel by means of a cork stopper with a glass tube running through it. Inert Free Rosin. If the milk appears to contain suspended rosin which settles on standing, boil 300 cc. for thirty minutes, filter on a filter paper which has been dried at 100 C. and weighed, wash with hot water and weigh the dried residue. Total Rosin. Pipette 50 cc. of the milk into a separatory fun- nel, add 10 cc. of very dilute H 2 SO4 (1%), then 50 cc. of ether and proceed as under Combined Rosin described above under Rosin Size. Combined Rosin. Subtract the total free rosin from the total rosin and report the difference as combined rosin. SATIN WHITE General. Satin White is a paste composed of water, alum- inum hydroxide, calcium hydroxide and hydrated calcium sulfate and is made by adding milk of lime to alum in solution. Aluminum Hydroxide. Weigh out rapidly 10 grams of the paste; dissolve in dil. HC1 and hot water. If any insoluble matter remains after boiling, it should be filtered out, ignited and weighed. WOOD, PAPER AND PAPER-MAKING MATERIALS 333 Make the filtrate up to 500 cc. and pipette 50 cc. into a beaker. Make slightly ammoniacal, heat to boiling, filter out the A1(OH) 3 , wash thoroughly with hot water, ignite over a Tirrill burner and then in the blast lamp and weigh as A1 2 O 3 . Calculate this to A1(OH) 3 . CALCULATION. A1 2 O 3 X 1.5284 = A1(OH) 3 . Total Lime. Heat the filtrate from the A1(OH) 3 determination to boiling and add a slight excess of (NH4)2C2O4 solution. Let settle, filter, wash with hot water, ignite in the blast lamp and weigh as CaO. NOTE. If preferred, the CaC 2 O 4 precipitate may be titrated with 0.1 N KMnO 4 (page 326). Total SO 3 . Pipette 50 cc. of the original solution into a beaker, dilute with 100 cc. of water, and heat to boiling. Then add, drop by drop, a slight excess of boiling BaCl2 solution and boil five minutes. Let stand overnight. Filter out the BaSO 4 , wash with hot water, ignite and weigh as usual. Calculate the BaS0 4 to CaSO 4 -2H 2 0, subtract the CaO equivalent of this from the total CaO, and calculate the remainder to Ca(OH) 2 . CALCULATIONS. BaS0 4 X 0.7375 = CaS0 4 2H 2 O. BaSO 4 X 0.3430 = SO 3 . CaSO 4 2H 2 O X 0.3257 = CaO. CaO X 1.3214 =Ca(OH) 2 . SO 3 X 2. 1504 = CaSO 4 - 2H 2 O. Moisture. It is generally customary to take the moisture " by difference." Add together the insoluble matter, A1(OH) 3 , CaSO 4 -2H 2 and Ca(OH) 2 , and subtract the sum from 100, reporting the difference as moisture. NOTES. (1) The moisture cannot be accurately determined by direct drying at 105 C. because the CaSO 4 -2H 2 O loses some of its water of crystal- lization on heating. If desired, the total loss on ignition may be determined and the uncombined water calculated from this. For accurate results it is necessary to use the blast lamp to make sure all the combined water is driven off. CALCULATIONS. Combined H 2 O from CaSO 4 2H 2 O = CaSO 4 2H 2 O X 0.2093. Combined H 2 O from Al (OH) 3 = Al (OH) 3 X 0.3460. Combined H 2 O fom Ca(OH) 2 = Ca(OH) 2 X0.2432. 334 TECHNICAL METHODS OF ANALYSIS (2) If any CaCOs is present, as is sometimes the case when Satin White is made from " lime mud," it is, of course, necessary to determine the amount of CC>2 present and take this into account in making the calculations. C ALCUL ATION . CO 2 X 2 .2743 = CaCO 3 . TALC FOR PAPER FILLER General. Pure talc is a hydrated magnesium silicate. The commercial article, however, is generally a double silicate of Mg and Al in which the Mg predominates. It contains, as natural impurities, Fe20 3 , CaCO 3 , and sand. The CaC0 3 is due to the difficulty in separating the talc from the limestone in which it occurs. For use as a paper filler, talc should show an analysis within the following limits : Sp.gr 2.7-2.9 CaC0 3 Not over 4% Fe 2 3 Not over 2%. Talcs containing up to 10% of CaCO 3 are not necessarily adul- terated, but if they contain more than 4% can only be used for cheap packing paper and pasteboard. Loss on Ignition. Weigh 1 gram in a platinum crucible, ignite at bright red heat to constant weight and calculate the loss in weight. For normal talc this is about 4%. Calcium Carbonate. Place 1 gram in 400 cc. of distilled water and add about 2.5-3 cc. of cone. HC1. Boil gently for fifteen to twenty minutes; filter and wash. To the filtrate, add 10 cc. of cone. HC1. Then make slightly ammoniacal, and filter out any alumina which may precipitate. To the filtrate add (NH4)2C 2 O 4 solution in excess; and after the precipitate has settled, filter it out, ignite and weigh as CaO. Calculate to CaCOs. CALCULATION. CaO X 1.7848 = CaCO 3 . Iron Oxide. Weigh 1-2 grams in a large platinum crucible, add a few drops of dil. H2SO4 and evaporate 2 or 3 times with 10-15 cc. of HF. Then fuse the residue with anhydrous KHS0 4 . Dissolve in water and filter if necessary. Add sufficient H2SO4 to make a 5% solution, and run through the Jones' reductor. Cool and titrate immediately with 0.1 N KMnO4. CALCULATION. 1 cc. 0.1 N KMnO 4 = 0.008 gram Fe 2 O 3 . WOOD, PAPER AND PAPER-MAKING MATERIALS 335 Specific Gravity. Fill a 200 cc. beaker about two-thirds full of xylene (commercial xylol). Bend a fine platinum or steel wire around a platinum crucible in such a way that it may be sus- pended from the hook above the balance pan. Weigh the crucible completely submerged in the xylene. From a weighing bottle partly filled with the talc, and previously accurately weighed, pour into the crucible (which should be about half full of xylene) approximately 1 gram of the talc, very slowly and carefully, to avoid loss through any of the powder being blown away. Tap the crucible gently to dislodge any air bubbles. Again submerge and weigh in xylene. The increase in weight represents the weight of the talc in xylene. The weight of talc is obtained by again weighing the weighing bottle. The difference between the weight of talc in air and in xylene represents the weight of xylene dis- placed. Determine the sp. gr. of the xylene in the beaker imme- diately with the Westphal balance. Calculate the result from the formula : Wt. talc taken Sp. gr. of talc = - -X sp. gr. of xylene. Wt. xylene displaced NOTE. Talc is often adulterated with barytes and with feldspar and if the sp. gr. of the talc is above 2.9, it indicates that the talc is adulterated with some heavy material such as these. Gypsum. Boil 2 grams in a beaker with 25 cc. of HC1 (1:4) and filter into a 250 cc. volumetric flask, washing the residue thoroughly with hot water. Cool the filtrate and dilute to the mark. Test about 100 cc. of this solution with BaCb. If it gives an appreciable precipitate, the amount should be deter- mined. For this purpose heat to boiling an aliquot of 100 cc. and add a boiling hot solution of 10 cc. of 10% BaCb solution a drop at a time. Boil for thirty minutes, let stand until clear and filter out the BaSO4. Ignite and weigh. Calculate to gyspum. CALCULATION. BaSO 4 X 0.7375 = CaSO 4 2H 2 O. NOTE. The amount of CaO present as gypsum should be subtracted from the total CaO as previously determined before calculating to CaCO 3 . Grit. A good idea of the relative amount of grit in two samples can be obtained by placing alternately portions of one and then the other on the tongue and rubbing the talc between the teeth. Quantitative estimation is carried out by means of the 336 TECHNICAL METHODS OF ANALYSIS flotation test and the 200-mesh sieve test as described in the method for Clay (page 318). REFERENCE. Paper, 6, No. 4, page 13 (1912.) ULTRAMARINE Strength of Color. Weigh out 1 gram of ultramarine and 10 grams of barytes (BaSO4) in a 1-ounce bottle together with some small round lead shot and agitate until a uniform mixture is ob- tained, using great care not to get any of the ultramarine on the stopper until it is fairly well mixed with barytes. This can be accomplished by giving the bottle a rotating motion at first. Then take out a little of the mixture with a spatula and spread it upon a piece of white paper beside a similar portion of the standard with which the comparison is made. A series of standards should be made up in a similar manner, mixing with 10 grams of BaSO4 the following amounts, respectively, of standard ultramarine: 0.70, 0.75, 0.80, 0.85, 0.90, 0.95, 1.00, 1.05 and 1.10 gram. In reporting the results state whether the sample is stronger or weaker than the standard and give the parts of sample equivalent to /me part of the standard. NOTE. It is important that the same barytes should be used for the standard as for the sample under examination and it should be dried before weighing. Resistance to Alum. Weigh 0.2 gram of the sample into a test-tube and add 10 cc. of a 10% solution of alum.* Shake for about one minute, then allow it to stand with an occasional shaking. Treat the standard ultramarine in a similar manner and compare with the sample under examination. Some ultramarines are decomposed by alum, resulting in a decided weakening of the shade. NOTE. In some instances, if the ultramarine is very resistant, it is neces- sary to warm the solution very slightly. Shade. Ultramarine should be tested for shade by comparing the sample and the standard undiluted, since with dilution the shade appears dull. * Paper-makers' alum, A1 2 (SO 4 )3-18H 2 O. . WOOD, PAPER AND PAPER-MAKING MATERIALS 337 FIBERS IN PAPER Quantitative Estimation. Tear small bits from different por- tions of the sample, using in all 1 or 2 square inches. Tear these pieces into smaller pieces about -ft to J inch square and place in a test-tube about one-third filled with 1% NaOH solution. Boil briskly for a few moments. Fill the test-tube with cold water several times, decanting off the liquid each time. Finally fill the tube completely with water and invert on the palm of the hand. By raising the edge of the tube cautiously, allow all the water to drain out, leaving the pieces of paper in the hand. Roll these into a hard ball between the thumb and finger and return to the test-tube. Make sure that all the NaOH has been washed out of the paper fibers. Fill the tube about one-third full of water and shake vigorously. This should completely disintegrate the fibers, making a uniform suspension. Remove a portion of the suspended fibers on the end of a micro- scope needle and blot off the excess moisture on a clean filter paper, holding the needle horizontally and making sure that the point does not remove any fibers from the filter paper. Place the fibers on a clean glass microscope slide and add two drops of fiber stain.* Pull the fibers apart, using two clean needles until a uniform mixture is obtained, free from any lumps or clots of fibers. Place a clean cover glass over the fibers and press down gently with the needles; blot off the excess liquid with a clean filter paper. The slide is then ready for microscopical examination. For microscopical examination a magnification of about 100 diameters is desirable. An 18 millimeter (f inch) objective and a 25 millimeter (1 inch) eye-piece is a satisfactory combina- * The stain is made up as follows : Solution A : 20 grams anhydrous zinc chloride, 10 grams water. Solution B: 2.1 grams potassium iodide crystals, 0.1 gram iodine, 5.0 grams water. Cool each solution and mix slowly, keeping cool. Allow the precipitate to settle overnight and pour off the supernatant liquid, or filter through glass wool. Add a small flake of iodine and preserve the stain away from the light or in a brown glass-stoppered bottle. The stain should be allowed to stand a day or two before using. i 338 TECHNICAL METHODS OF ANALYSIS tion. In general, the different fibers used in paper are colored by the stain as follows : Rag (cotton, linen, hemp) : pale red to brownish or pur- plish red. Bleached chemical wood (sulfite> soda, etc.): deep blue. Ground wood: bright yellow, sometimes pale yellow. Jute and manila: vary all the way from blue through violet and red violet to brownish-yellow and greenish-yellow. Straw and esparto: generally dark blue, but sometimes reddish or yellowish. For the characteristics of the different fibers the standard books should be consulted; particularly good illustrations and descriptions are given in Herzberg's " Papier-priifung," 3d edition (pages 88 and 89, and tables in the back of the book). In estimating the amounts of the various constituents, standard slides * should be prepared from papers of known composition (page 339), and frequently compared with the samples under examination. Qualitative Test for Ground Wood. There are several reagents which, when applied to the surface of a paper, show characteristic colors if ground wood is present. Such tests, however, are of questionable value in arriving at any estimation of the amount of ground wood and microscopical examinations should always be made. The principal test solutions for the presence of ground wood are as follows : (1) Aniline Sulfate. Dissolve 5 grams of aniline sulfate in 50 cc. of distilled water and add a drop of cone. H2SO4. This solution is not permanent, but decomposes rather easily, taking on a violet coloration. It should be used only when it is colorless. When applied to paper containing ground wood, it gives a bright yellow color. (2) Phloroglucinol. Dissolve 1 gram of phloroglucinol in 50 cc. of alcohol and add about 25 cc. of cone. HC1. This gives a pale yellow solution which keeps fairly well when protected from air and light. A fresh solution works more sharply and quickly than an old one. The color which it gives on paper containing ground wood is a bright magenta. (CAUTION. Jute, manila and unbleached sulfite sometimes give a pale pink coloration. An indication of traces of ground WOOD, PAPER AND PAPER-MAKING MATERIALS 339 wood by this test should therefore be confirmed by a micro- scopical examination.) (3) Dimethyl-p-phenylenediamine (Wiirster's Reagent). A water solution of this compound, when dropped upon a paper containing ground wood, gives an orange red spot after a short interval. (4) Nitric Acid. Cone. HNOs when applied to a paper con- taining ground wood gives a deep yellow color. STANDARD PAPERS FOR FIBER ANALYSIS General. Before making up any standard papers, the raw materials should be tested to see that they are pure sulfite, soda, rag, etc. Sulfite-Rag Standards. Pure sulfite pulp and unfilled pure rag paper should be used for the sulfite-rag standards. After determining the moisture in each, weigh them out into 100-gram bundles and place in a tightly covered can. When it is desired to make up a certain sulfite-rag standard, the amount of bone-dry stock required should be calculated from the moisture figures, the 100-gram bundles reweighed, and correction made for any change in moisture content which may take place during storage. The proper weight of the stock having been determined, add the sulfite to the beater and partly cut down; then add the rag paper and beat the two for. some time, with the knives down. Raise the knives and stir the contents of the beater for one and one-half to two hours. Make up sheets from this stock on a hand mold, press between blotters, and mark for identification. Soda-Rag Standards. Follow the same procedure as for the sulfite-rag standards, using, however, pure soda pulp in place of sulfite pulp. Other Standards. For other standard papers follow the same general procedure, taking special precaution to make certain of the purity of the original raw materials. CHEMICAL ANALYSIS OF PAPER General. The analyses described in this method do not include fiber analysis, which is described of page 337. For the deter- mination of tarnishing properties and sizing tests see pages 362 and 355, respectively. 340 TECHNICAL METHODS OF ANALYSIS Mineral Matter (Ash). Ignite 2 grams in a porcelain crucible until all the carbon has been burned off, brush the ash carefully into a balanced watch glass and weigh it. Since the ash of paper is generally very light, great care must be taken that none of it is blown out of the crucible by drafts of air. For this reason the crucible should always be kept covered after the main por- tion of the organic matter has been burned off. NOTES. (1) If the paper is filled with crown filler, multiply the ash as actually determined fr" 1.265 to obtain the equivalent of CaSO 4 -2H 2 O as it existed in the sheet. (2) Unfilled papers sometimes show a blue ash, due to the presence of ultramarine. The amount of the latter may be determined in many cases by carefully igniting a considerably quantity of the paper until all the carbon has been burned off, then taking a quantity of clay or calcium sulfate equal to the weight of the ash and determining how much of a standard sample of ultramarine must be mixed with this to produce the depth of color shown by the ash. Filler. As a rule, the presence of more than 1% of ash in a white paper indicates that it contains added filler. It should be borne in mind, however, that the ash will not indicate the actual amount of filler in the paper since the filler always contains more or less moisture and volatile matter which is driven off when the paper is ashed. The nature of the filler will be indicated by qualitative tests as follows : Boil the ash with dil. HC1, filter on a quantitative filter and wash. Test a portion of the filtrate for sulfates by boiling with a small amount of BaCl2 solution. Test the remainder of the filtrate for Al by adding an excess of NHiOH. Then filter and test the filtrate for Ca by adding (NHi) 20264. The presence of any considerable amount of Ca and of sulfate at this point indicates that the paper is filled with crown filler, CaSO4-2H2O; whereas only a small amount of Ca would probably be due to lime in clay or other insoluble filler. Transfer the residue insoluble in HC1 to a platinum crucible, burn off the filter paper, and then fuse with Na2COs to a clear fusion. Treat the fusion with hot water until it can be removed from the crucible. Dissolve a small portion of the residue insoluble in hot water in dilute HC1 ; if it gives a clear solution, this indicates the absence of BaSO4. In this case dissolve the entire fusion in WOOD, PAPER AND PAPER-MAKING MATERIALS 341 HC1 and evaporate to dryness. Take uj) with warm, dilute HC1 and filter. The insoluble residue will be SiCV In case the original fusion does not dissolve in HC1, boil it with water, filter and wash thoroughly Dissolve the residue insoluble in hot water in dilute HC1 and add dilute H2S04. A white precipitate shows the presence of Ba. Acidify a portion of the water solution with HC1 and add BaCb solution. A white precipitate shows the presence of sulfates. To the bulk of the HC1 solution of the fusion, or to the bulk of the water solution (after adding an excess of HC1 and boiling) add a slight excess of NHiOH and boil. A white gelatinous precipitate shows Al. Filter, and to the filtrate add an excess of (NHi) 20264 solution. A white precipitate shows the presence of Ca. Filter, and acidify the filtrate with HC1 and evaporate to the point of crystallization. Cool, and add an excess of ammo- nium phosphate solution. Stir thoroughly. A white crystalline precipitate shows the presence of Mg. INTERPRETATION OF RESULTS. A considerable amount of CaO and SO 3 soluble in HC1 indicates crown filler, CaSO 4 -2H 2 O. The presence of Ba and of SOs in the insoluble residue indicates blanc fixe, BaS04. The presence of Ba soluble in HC1 with effervescence indicates BaCOs. An alkaline ash showing lime soluble in HC1, but not showing SOs, indicates CaCOs (whiting or chalk). Con- siderable amounts of silica and alumina, with or without small amounts of Ca and Mg, indicate china clay, but if the amount of Mg is considerable the filler is probably agalite, talc or asbestine. A microscopical examination of the ash will often give a clue as to the nature of the filler. Agalite crystallizes in needles, asbestine is fibrous. NOTE. It is not usually possible to distinguish between talc and clay con- taining a little magnesia without a quantitative determination of the amount of MgO in the ash of the paper. It is sufficiently accurate to treat the ash in a platinum crucible with H 2 SO 4 and evaporate once or twice with HF to remove SiO 2 . Then take up the residue in warm dil. HC1 (it should practically all dissolve unless Ba salts are present), add NH 4 OH, boil until the odor of NH 3 is nearly gone and then add, without filtering, a slight excess of (NH 4 ) 2 C 2 O 4 . Filter, wash with hot water, add a slight excess of HC1 and evaporate till crystallization begins. Dilute until the crystals just dissolve and precipitate the Mg as MgNH 4 PO 4 in the usual way. Ignite and weigh as Mg 2 P 2 O 7 and calculate the per cent of MgO in the weight of paper ash 342 TECHNICAL METHODS OF ANALYSIS taken. Clay rarely contains over a few per cent of MgO, whereas talc contains approximately 31 per cent. Retention. The retention is generally figured from the ash, although for accurate results, the ash should be corrected for the natural moisture in the filler. The calculation of retention is as follows : Subtract the percentage of filler from 100% to obtain the per- centage of fiber in the sheet; divide the pounds of fiber fur- nished by the percentage of fiber in the paper to obtain the pounds of paper made from the engine. From this figure subtract the pounds of fiber furnished and the balance is the pounds of filler retained. This figure, divided by the pounds of filler furnished, gives the percentage of retention. Acidity. Papers containing free acid are rare. For practical purposes, however, the acidity due to excess of alum is to be con- sidered free acid. TOTAL ACIDITY. For purposes of comparison, total acidity may be determined as follows : Warm 5 grams of paper with 250 cc. of distilled water, pour off the water and titrate while still hot with 0.1 N NaOH and phenolphthalein until a permanent pink color is obtained. Run a " blank " on an equal amount of the water. Subtract the " blank " titration from the other titration and calculate the difference to 80s. CALCULATION. 1 cc. 0.1 N NaOH = 0.004 gram S0 3 . SULFUBOUS ACID. Tear up 50 grams of the paper and place in a 500 cc. distilling flask with about 350 cc. of distilled water. Distill until 150 cc. have come over and collect the distillate in a flask con- taining about 50 cc. of water with a few drops of bromine water added to it. Acidify with dil. HC1 and boil off the Br; then add a slight excess of BaCb solution. Boil till the precipitate settles clear. Filter out the BaSO4, ignite, and weigh in the usual manner. From the weight of BaSC>4, calculate the amount of 862 . CALCULATION. BaSO 4 X 0.2744 = S0 2 . NOTE. Qualitative tests for the acidity of the paper are generally suf- ficient. A satisfactory paper should not give more than a slight acid reaction when pressed in contact with moistened blue litmus paper between sheets of filter paper for fifteen minutes. Sulfurous acid in papers that have been exposed to the air for any length of time is oxidized to sulfuric acid. WOOD, PAPER AND PAPER-MAKING MATERIALS 343 Free Chlorine. Free chlorine is seldom found in papers. If present, it can be detected by placing the paper in a dilute solution of KI made weakly acid with H2S04, to which has also been added a few drops of starch solution. Free chlorine in the paper will liberate iodine, which will give a blue color with the starch. If desired, the amount can be determined by titrating the iodine with 0. 1 N thiosulf ate solution. Chlorides. Free chlorine or hypochlorites from excess of bleach in the paper do not remain as such for any length of time but are reduced to chlorides. The presence of the latter, however, in appreciable amount is also objectionable. To determine the amount of chlorides present, weigh 5-10 grams of the paper, tear it up in small bits and heat to boiling with distilled water in a beaker. Pour off the water and repeat the boiling with two fresh portions of water. Filter the combined extracts if necessary. Add a few drops of dil. HNOa and a slight excess of AgNOs solution. Filter the AgCl on a Gooch crucible. Wash with hot water, dry and weigh. Calculate to Cl. CALCULATION. AgCl X 0.2474 = Cl. NOTE. If preferred, the solution may be titrated with 0.1 N or 0.02 N AgNOs, using K2CrO4 solution as an indicator. (See page 509.) COATED PAPERS The coating consists of a mineral substance or substances with an adhesive binder to make a coherent film upon the surface of the paper. When papers are properly coated, the coating should not " pick " or lift off when the moistened thumb is pressed momentarily against the coating and then withdrawn with a quick jerk. In the case of poorly coated papers, more or less of the coating will adhere to the thumb when thus tested. Papers may be single coated (coated on one side) or double coated (coated on both sides). Amount of Coating. Weigh a piece of the paper cut exactly 2X5 inches and place in a flat glass dish. The dishes used for developing in photography are convenient for this purpose. Cover with water containing 1% of NKiOH and set aside in. a warm place (two or three hours is generally sufficient to loosen the coating). Remove the paper to a large watch glass, rub the sur- 344 TECHNICAL METHODS OF ANALYSIS face with a small camel's hair brush cut off square, and wash the coating into a beaker. If the paper is double coated, turn it over and repeat on the other side. Continue the operation until all the coating is washed into the beaker. Dry the paper and weigh it under the same conditions as those under which the original paper was weighed. The loss in weight is the weight of coating. Calculate this to per cent of the original sample and also figure the weight of coating on the basis of a ream of 500 sheets 24X36 inches (see table on page 351). Analysis of Coating. Heat the mixture in the beaker to boiling and filter through a qualitative filter. Test the filtrate for glue, casein and starch. For the qualitative tests for glue and starch, see page 355. Casein can generally be detected by the odor and also by the fact that it will precipitate from the solution when carefully neutralized with HC1. A confirmatory test for casein may be made on the original paper as follows : Boil 1 or 2 square inches of the paper with 5 cc. of water to which is added 5 drops of dil. H 2 SO 4 . Decant the liquid from the paper into a test tube, cool and add 1 drop (not more) of 3% formaldehyde. Mix and pour this solution gently down the side of an inclined test tube containing 2 cc. of cone. H 2 S04 to which has been added 1 drop only of 10% FeCls solution, taking care that the water solution flows on the top of the H 2 SO4 without mixing with it. If casein is present, a violet ring forms at the junction of the liquids. Glue does not give more than a brownish color and rosin does not react. A blank test should be conducted, using a small amount of casein, or preferably, a paper which is known to contain casein. The usual substances to be looked for in the mineral part of the coating are (1) blanc fixe, BaS04j (2) satin white, a mixture of CaSO 4 -2H 2 O, Ca(OH) 2 and A1(OH) 3 ; (3) china clay, aluminum silicate (approximately Al 2 Si 2 O7 H 2 O) ; and (4) chalk, CaCOs. Wash the residue on the filter paper into a beaker with a stream of water, first punching a hole in the paper. Pour 10-15 cc. of hot dil. HC1 through the filter into the same beaker. Heat the solution to boiling (note whether the acid causes effervescence of CO 2 ) . Let settle, decant through a quantitative filter and heat the residue in the beaker with a second portion of 100 cc. of dil. HC1. Filter through the same filter, washing with hot water. WOOD, PAPER AND PAPER-MAKING MATERIALS 345 Divide the filtrate into two portions. Test one for sulfates with BaCU solution and the other for Al and Ca. The presence of Ca, Al and 864 indicates that satin white was used in the coating. If Ca is found but no S04 and if the coating gives an effervescence with HC1, it contains chalk or whiting (CaCOs). If the coating contains carbonates but is free from Ca (and Mg), test the HC1 solution for Ba, as in some special papers BaCOs is used. Ignite the filter containing the residue from the HC1 treatment above, and fuse with excess of Na2COs in a platinum crucible. Boil the fusion with water and filter, washing well with hot water. To the nitrate add HC1 in excess and boil off all CO 2 . Test a portion of the solution for sulfate with BaCb solution. Evaporate the remainder to dryness, take up with HC1 and note whether any SiCb is present. If so, filter out and test the filtrate for A^Os by boiling with an excess of NH^OH. A slight precipitate here may be due to alum from the paper; a heavy precipitate indicates china clay, if silica was also found present. Add (NH4)2C204 solution to the filtrate from the alumina, boil and filter out any CaC2(>4 that may precipitate. Test this filtrate for Mg, which in considerable amount might indicate the presence of talc. Dissolve the water-insoluble portion of the fusion with HC1. Test a portion of it for Ba with H2SO4. If a precipitate is obtained, confirm this by dipping a platinum wire into the remainder of the HC1 solution and holding it in the flame. Barium will give a green coloration to the flame. If both Ba and sulfate are found in the fusion of the residue insoluble in HC1, the coating contains blanc fixe. NOTE. After removing the coating, the uncoated paper may be tested for ash and filler as previously described. The test for rosin, if called for, should be made on the original paper. REFERENCES. Herzberg: " Papier-priifung "; Griffin and Little : "The Chemistry of Paper Making 'J; U. S. Dept. of Agriculture, Report No. 89, by F. P. Veitch. PHYSICAL TESTING OF PAPER General. Before cutting the samples for the physical tests, it is necessary to determine the machine and cross directions of the sheet. Depending on the individual papers, one of the following methods can be used : 346 TECHNICAL METHODS OF ANALYSIS (a) Cut a circle 1 or 2 inches in diameter from the sheet and moisten on one side. The paper will curl about an axis corre- sponding to the machine direction of the paper. (6) Cut a strip 1 cm. wide and 15-20 cm. long from each direc- tion of the sheet, hold the ends together and let the strips bend of their own weight, first to the right and then to the left. The paper is less flexible in the machine direction, hence when the machine direction is uppermost there will be an opening between it and the lower strip. When bent the other way the two strips will lie together and the cross direction strip will then be on top. (c) The direction can sometimes be told, especially on cylinder- made papers, by examining the surface of the sheet. The fibers will have a very marked tendency to be drawn out along one direction, namely the machine direction. (d) Certain papers tear straight in the machine direction and very unevenly in the opposite direction. Tensile Strength. Make the determination on the Schopper machine (Fig. 16). Cut strips from both directions of the paper exactly 15 mm. wide and about 240 mm. long, if it is possible to obtain strips of this length without folds. A strip of this length can be used where the jaws are 180 mm. apart, in which case the percentage stretch can be read directly from the scale. Cut at least 5 strips in each direction and report the average of the 5 tests. If the individual tests vary considerably, run at least 10 in each direction. In placing the strip in the machine, first clamp it in the lower jaw, then pull it through the upper jaw and clamp. Adjust the stretch device catch and release the locks on the upper jaw and on the pendulum previous to starting the machine. Results are obtained in kilograms per 15 mm. Calculate to pounds per inch. CALCULATION. Kg. per 15 mm.X3.72 = lbs. per inch. For the calculation of breaking length, weigh each strip previous to testing and also determine its total length in millimeters. Cal- culate as follows : Breaking length (yards) = Tensile Strength (Kg. per 15 mm.)Xlength of strip (mm.) X 1.094 Wt. of strip (grams) WOOD, PAPER AND PAPER-MAKING MATERIALS 347 The breaking factor or tensile factor is the tensile strength (pounds per inch) in each direction divided by the weight in pounds of a ream of 500 sheets, 24X36 inches. FIG. 16. Schopper Tensile Machine. Stretch Test. If a 180 mm. strip is used for testing, the stretch will be given directly by the machine, in percentage. If a shorter strip is used, the percentage stretch will have to be cal- culated from the actual stretch in millimeters. Folding Tests. Determine the folding strength in double folds on the Schopper folding machine (Fig. 17). Cut the strips 348 TECHNICAL METHODS OF ANALYSIS for testing 15 mm. 'wide and about J inch longer than the strip gauge accompanying the folding machine. Fasten the strip in the jaws after releasing the tension of the jaw heads. After the strip is in place, pull back the jaw heads, set the gauge at zero and release the lock on the wheel. The machine should run at 120 R.P.M. Test at least 5 strips in each direction. (Test 10 if the individual results vary widely.) The folding result is affected by the relative humidity of the atmosphere. For purposes of record, therefore, the relative FIG. 17. Schopper Folding Machine. humidity and the temperature should always be recorded at the time the tests are made and it is preferable to conduct the tests in a room where temperature and humidity are maintained con- stant. The folding factor is the folding strength in each direction divided by the weight in pounds of a ream of 500 sheets, 24X36 inches. NOTE. It is very necessary that the small steel wheels which support the clamping jaws be perfectly round, well oiled and revolve easily as the jaws move back and forth, as otherwise results may be seriously affected. WOOD, PAPER AND PAPER-MAKING MATERIALS 349 Mullen Test (Bursting Strength) and Thickness. Use the small Mullen machine (Fig. 18) for light weight papers and the large (Jumbo) machine for heavier weight papers and cardboard. Make at least 5 tests and report the average. The reading of the gauge is in pounds per square inch. Place the sheet over the diaphragm and clamp securely. Make sure that the sheet is smooth and that no wrinkles or creases occur within the space to be tested. Set the pressure gauge at zero and apply pressure by turning the wheel with a FIG. 18. Perkins Mullen Tester. smooth, rapid motion free from jerks. When the paper breaks, immediately reverse the wheel to remove pressure on .the dia- phragm and then take reading of the gauge. Determine the thickness by a number of tests, using a thickness gauge (Fig. 19), and taking the test near each Mullen break. Average the Mullen tests and thickness results. The average Mullen test, divided by the thickness in ten thousandths of an inch, is the strength ratio, or bursting ratio. Example. Mullen test = 20 Ibs. and thickness = 0.0018 inch. Then the strength ratio = 20-5- 18 = 1.11. The average Mullen strength divided by the weight in pounds of a ream of 500 sheets, 24X36 inches, is the strength factor or the bursting factor. 350 TECHNICAL METHODS OF ANALYSIS Ream Weight. If a sufficiently large sample of the paper is available, weigh a sheet of the desired size on the ream weight scales (Fig. 20), which give directly the weight of a ream of 500 sheets or 480 sheets, as the case may be. For instance, if the ream weight 24X36 500 is desired, one sheet 24X36 inches weighed on the ream weight scales would give the ream weight in pounds direct. In case the size of the sample is limited, cut a small piece of the paper to accurate measurement and weigh on a precision FIG. 19. Thickness Gauge. balance/ and calculate the ream weight from this weight. A piece 2X5 inches (i.e., 10 square inches in area) is a convenient size. The standard ream for writing and printing papers is 500 sheets each 24X36 inches, and unless otherwise specified the cal- culations should be on this basis. If a piece of 10 sq. in. is employed, then: Wt. in grams X 95. 2 = ream weight. The designation of the size of the ream, etc., is as follows: Ream weight 24X36 500, 50 Ibs. For wrapping papers the ream is generally 480 sheets instead of 500 sheets, WOOD, PAPER AND PAPER-MAKING MATERIALS 351 The following table gives the " factors " with their logarithms for calculating the ream weights, on several different bases, from the weight in grams of a piece 2X5 inches. Size of Sheet Sheets in Ream " Factor " Logarithm Inches 24 X36 ' 500 95.2 .9786 24 X36 480 91.4 .9608 8|X11 500 10.31 .0131 12|X16 500 22.05 .3433 121X16 480 21.16 .3255 14 X17 500 26.24 .4190 14 X17 480 25.19 .4012 16 X21 500 37.0 .5686 16 X21 480 34.6 .5385 17 X22 500 41.2 .6153 17 X22 480 39.6 .5975 17 X26 500 48.7 .6877 17 X26 480 46.8 .6699 17 X28 500 52.5 .7199 17 X28 480 50.4 .7021 18 X24 500 47.6 .6778 18 X24 480 45.7 1.6600 19 X24 500 50.3 1.7013 19 X24 480 48.3 1.6835 20 X25 500 55.1 1.7413 20 X25 480 52.9 1.7235 24 X38 500 100.6 2.0026 24 X38 480 96.6 1.9848 25 X38 500 104.7 2.0199 25 X38 480 100.5 2.0023 27 X39 500 116.1 2.0650 27 X39 480 111.5 2.0472 34 X46 500 172.5 2.2368 34 X46 480 165.6 2.2190 40 X48 500 211.7 2.3257 40 X48 480 203.2 2.3079 352 TECHNICAL METHODS OF ANALYSIS Substance Number. In October, 1916, the Writing Paper Manufacturers Association adopted the " Substance Number " system for designating the weights of paper. The folio size (17X22 inches) is taken as the basis and the weight of a ream of 500 sheets of folio size is the substance number. By specifying FIG. 20. Ream Weight Scales (Fairbanks Co.). the substance number, therefore, the weight of paper desired is definitely established and the weight per ream of any other size, or the substance number of any paper, the ream weigh ; of which is known on some other basis, may be readily ascertained by reference to the following table : WOOD, PAPER AND PAPER-MAKING MATERIALS 353 TABLE SHOWING ACTUAL WEIGHTS (FIGURED TO 0.5 LB.) OF STANDARD SUBSTANCE NUMBERS* Size Substance No. 13 No. 16 No. 20 No. 24 No. 28 No. 32 No. 36 No. 40 No. 44 Inches 14X17 14X34 8.5 16.5 10.0 20.5 12/5 25.5 15.5 30.5 18.0 35.5 20.5 40.5 23.0 46.0 25.5 51.0 28.0 56.0 15X19 10.0 12.0 15.0 18.5 21.5 24.5 27.5 30.5 33.5 16X21 16X26 16X42 11.5 14.5 23.5 14.5 18.0 29.0 16.0 18.0 22.0 36.0 21.5 26.5 43.0 25.0 31.0 50.5 28.5 35 5 57.5 32.5 40.0 64.5 36.0 44.5 72.0 39.5 49.0 79.0 17X22f 13.0 20.0 24.0 28.0 32.0 36.0 40.0 44.0 17X26 17X28 17X44 17X56 15.5 16.5 26.0 33.0 19.0 20.5 32.0 40.5 23.5 25.5 40.0 51.0 28.5 30.5 48.0 61.0 33.0 35.5 56.0 71.5 38.0 40.5 64.0 81.5 42.5 46.0 72.0 91.5 47.5 51.0 80 102.0 52.0 56.0 88.0 112.0 18X23 18X46 14.5 29.0 17.5 35.5 22.0 44.5 26.5 53.0 31.0 62.0 35.5 71.0 40.0 79.5 44.5 88.5 '48.5 97.5 19X24 19X26 19X28 19X30 19X48 16.0 17.0 18.5 20.0 31.5 19.5 21.0 23.0 24.5 39.0 24.5 26.5 28.5 30.5 49.0 29.5 31.5 34.0 36.5 58.5 34.0 37.0 40.0 42.5 68.5 39.0 42.5 45.5 49.0 78.0 44.0 47.5 51.0 55.0 88.0 49.0 53.0 57.0 61.0 97.5 53.5 58.0 62.5 67.0 107.5 20X28 20X56 19.5 39.0 24.0 48.0 30.0 60.0 36.0 72.0 42.0 84.0 48.0 96.0 54.0 108.0 60.0 120.0 66.0 132.0 21X32 21X33 23.5 24.0 29.0 29.5 36.0 37.0 43.0 44.5 50.5 52.0 57.5 59.5 64.5 66.5 72.0 74.0 79.0 81.5 22X34 2 19.5 26.0 2i.O 32.0 30.0 40.0 36.0 48.0 42.0 56.0 48.0 64.0 54.0 72.0 60.0 80.0 66.0 88.0 23X28 23X31 23X34 23X36 22.5 25.0 27.0 29.0 27.5 30.5 33.5 35.5 34.5 38.0 42.0 44.5 41.5 45.5 50.0 53.0 48.0 53.5 58.5 62.0 55.0 61.0 67.0 71.0 62.0 68.5 75.5 79.5 69.0 76.0 83.5 88.5 76.0 84.0 92.0 97.5 24X38 24X48 31.5 40.0 39.0 49.5 49.0 61.5 58.5 74.0 68.5 86.0 78.0 98.5 88.0 111.0 97.5 123.0 107.5 135.5 26X32 26X33 26X34 26X38 29.0 30.0 30.5 34.5 35.5 36.5 38.0 42.5 44.5 46.0 47.5 53.0 53.5 55.0 56.5 63.5 62.5 64.0 66.0 74.0 71.0 73.5 75.5 84.5 80.0 82.5 85.0 95.0 89.0 92.0 94.5 105.5 98.0 101.0 104.0 116.5 27X40 37.5 46.0 .58.0 69.5 81.0 92,5 104.0 115.5 127.0 28X34 28X38 28X40 33.0 37.0 39.0 41.5 40.5 45.5 48.0 51.0 51.0 57.0 60.0 63.5 61.0 68.5 72.0 76.5 71.5 79.5 84.0 89.0 81.5 91.0 96.0 102.0 91.5 102.5 108.0 114.5 102.0 114.0 120.0 127.5 112.0 125.0 132.0 140.0 30X38 39.5 49.0 61.0 73.0 85.5 97.5. 109.5 122.0 134.0 31X53 57.0 70.5 88.0 105.5 123.0 140.5 158.0 175.5 193.5 34X44 52.0 64.0 80.0 96.0 112.0 128.0 144.0 160.0 176.0 * Paper, Dec. 6, 1916, page 17. f Folio size, basis of standard. 354 TECHNICAL METHODS OF ANALYSIS Penetration or Sizing Tests. (a) FOR WRITING PAPERS. Float a piece of paper 1 or 2 inches square on the surface of ink, arid note the time in minutes until the ink begins to be visible on the upper surface. The time in minutes is taken as a basis for comparing the relative sizing of different papers. (See also page 358.) NOTE. In order to avoid discrepancies due to the variation in the ink, the latter should be made according to the following formula given by Schluttig and Neuman in " Die Eisengallustiten ": Gallotannic acid .' 23.4 grams Gallic acid 7.7 grams Gum arabic 10.0 grams Hydrochloric acid 2.5 grams* Ferrous sulfate crystals (FeSO 4 -7H,O) 30.0 grams Methylene blue 2.0 grams Water to make up to 1 liter. Let settle several days and then decant from any sediment. (6) FOR PAPERS OTHER THAN WRITING PAPER. For all papers other than writing papers use the so-called Ferrocyanide Test as follows : Float a piece of paper about 2 inches square on the surface of a 5% solution of KiFe(CN)6 and note the time. Then test the upper surface of the piece of paper from time to time by stroking (across the machine direction) with a small camel's hair brush moistened with a solution of FeCls (5-10%). When the ferro- cyanide has soaked up through the paper sufficiently to come in contact with the FeCls, it will react the moment the latter is applied and give a blue color. The penetration is then consid- ered complete, the time is again noted, and the length of time since the paper was laid upon the surface of the solution is taken as the measure of its resistance to penetration. In stroking the paper with the camel's hair brush, take care to select a place on the paper which has not previously been wet with the FeCla solution. Report the results to the nearest minute, or, if the time is very short, in seconds. Blotting Paper. The absorption test on blotting paper is carried out as follows : Cut strips of the paper, in both directions, about 0.5 inch * 2.5 grams of actual HC1 are equivalent to 5.8 cc. of cone. HC1. WOOD, PAPER AND PAPER-MAKING MATERIALS 355 wide and at least 3.5 inch long, and make a pencil mark about J inch from one end. Then, starting from this mark, make a series of pencil marks at intervals of f inch on the paper for a distance of about 2.5 inches. Suspend the strips in water in such a manner that the surface of the water coincides with the first mark. When the strip is first placed in the water, note the time and at the end of three, five, and ten minutes, respectively, ascertain the rise of the water in the paper by means of the marks upon it. Report the result in sixteenths of an inch, i.e., a paper which shows an absorption of 1 inch will be reported as " Absorption, 16." If necessary, a small quantity of ink may be added to the water to aid in determining the exact rise of the liquid. Test several strips of each sample cut from each direction and report the average absorption. SIZING IN PAPER General. Blotting paper and so-called " water-leaf paper " are unsized and rapidly absorb ink or any other liquid by capillary attraction. Papers which are to be used for writing or printing purposes must be sized with some material which will prevent this capillary absorption. Rosin and glue (" animal size ") are most commonly used. Starch is also used to some extent, generally together with rosin and glue, particularly in the case of blueprint papers. Paper containing a considerable proportion of rags sometimes contains a small amount of starch due to its incom- plete removal from the rags. It is advisable to make qualitative tests before proceeding with quantitative determinations. Qualitative Tests. (A) GLUE. Boil several grams of the paper with water until the volume of the latter is only a few cc. (as soon as the water has come to boiling, a portion of it may be poured off and saved for the starch test). Filter and cool thor- oughly. To this add about an equal volume of a cold 10% NaCl solution nearly saturated with tannic acid and freshly filtered. A light grayish yellow, flocculent precipitate indicates the presence of glue. (It should be remembered that casein will also give this test.) NOTES. (1) If the paper has been treated with formaldehyde in addition to glue, the latter will be rendered more or less insoluble in hot water. In this 356 TECHNICAL METHODS OF ANALYSIS case boil the sample with dil. NaOH solution, pour off the solution and make slightly acid with HC1. Cool, and test this solution with tannic acid. Run a " blank " with the NaOH to make sure that it contains nothing which will precipitate tannin after acidification. (2) In case the paper contains starch and gives a precipitate by the above test, the precipitate may be due to the starch. In this case, repeat the test and, before adding the tannic acid solution, add sufficient HC1 to the concentrated water extract of the paper to give about a 2% solution. Digest on the steam bath until the starch is all converted to dextrose and a drop of the solution gives no blue color when added to 5 cc. of very dilute (about 0.001 N) iodine solution. Then cool the solution and add the tannic acid-NaCl solution. (B) ROSIN (LIEBERMANN-STORCH TEST). Place several small pieces (representing about 1-2 grams) of the paper in a clean dry test-tube. Cover with pure acetic anhydride and boil down to about 1 cc. The fumes of anhydride are extremely irritating and after they begin to come off from the test-tube it is well to hold the mouth of the latter near a flame so that the fumes will burn as fast as they are driven off. Pour the liquid residue into a clean dry test-tube and cool thoroughly. If any waxy particles separate, they should be filtered off. Let one drop of cone. EbSCX run carefully down the side of the test-tube. A fugitive rose violet coloration, formed when the acid meets the anhydride, indicates rosin. (C) STARCH. Boil a portion of the paper with water. Cool and filter if necessary. Add one drop of very dilute iodine solution (about 0.01 N). A blue coloration indicates starch. There are some papers not sized with starch which will give a faint violet coloration, but this should be disregarded. NOTE. Applying iodine solution directly to the paper may give mislead- ing results. Quantitative Tests. (A) GLUE. Weigh 3-5 grams of the paper, tear into small pieces, and place in a 500 cc. Kjeldahl digestion flask. Determine the nitrogen by the Gunning method as described on page 65. Calculate to glue by multiplying the nitrogen found by 5.6. (B) ROSIN. Cut 5 grams of the paper into strips approximately 0.5 inch wide and fold in numerous small crosswise folds. The folding is essential to secure complete and quick extraction. Do not tear the paper into small pieces, since it will then stick together WOOD, PAPER AND PAPER-MAKING MATERIALS 357 and not be completely extracted. Place the folded strips in a Soxhlet extractor and fill the latter with acidulated alcohol. This is made by adding to 100 cc. of 95% alcohol 10 cc. of acidulated water, consisting of 3 cc. of glacial acetic acid to 100 cc. of water, and thoroughly mixing. Connect the extractor to an unweighed Soxhlet flask; add sufficiently more of the acidulated alcohol to insure having the flask about one-quarter full. When the extractor is filled, start the extraction and syphon at least 10 times. Wash the alcoholic extract into a beaker and evaporate to 10 cc. or less on the steam bath. Cool, take up in about 25 cc. of ether and transfer to a 300 cc. separatory funnel containing about 150 cc. of water, to which has been added a little salt to prevent emulsion. Shake thoroughly and allow to separate. Draw off the water into a second separatory funnel and repeat the treatment with a fresh 25 cc. portion of ether. Combine the ether extracts which contain the rosin and wash twice with 100 cc. portions of water or until the ether layer is perfectly clear and the line between the ether and the water sharp and distinct. If glue, which is extracted from the paper by the alcohol, should interfere by emulsifying with the ether, wash the com- bined ether extracts with a 10-15% NaCl solution (twice if nec- essary), and then wash with three 100 cc. portions of distilled water to remove the NaCl. Finally transfer the ether extract to a weighed flask and evap- orate off the ether. Dry at 100 C. in a water oven for exactly one hour. Cool in a desiccator and weigh. (C) STARCH. Prepare from the paper a solution containing approximately 1% of reducing sugar (dextrose).* This solution must not contain over 1% and if, after the analysis is completed, it is found that this limit is exceeded, the analysis must be repeated, using a sriialler amount of paper. The procedure is as follows: Weigh carefully 3-5 grams of the paper, cut in very small pieces (about J inch square), into a 250 cc. Erlenmeyer flask; add 150 cc. of water, and boil for about five minutes. Cool to about 50 C. and shake violently until the fibers are well separated. Then add 10 cc. of filtered saliva solution. (See note (1) below.) Place the flask in a water bath at 40-50 C. and stir frequently for one hour; then add 1 drop of 0.1 N iodine solution. * If there is no indication *Starch-sized papers generally contain 2% or less of starch. 358 TECHNICAL METHODS OF ANALYSIS of starch being present (blue color) , filter on a Gooch crucible and wash with hot water. Transfer the filtrate, which should have a volume of about 200 cc., to a beaker; add 15 cc. of dil. HC1 (5 : 4) and boil gently for forty-five minutes, covering the beaker with a watch glass. Cool immediately. Nearly neutralize the acid with NaOH and complete the neutralization with Na 2 CO 3 and litmus paper. Filter the neutral solution into a 250 cc. flask; make up to the mark, and determine the dextrose according t6 the Allihn method on page 407. The weight of dextrose multiplied by 0.9 gives the weight of starch. Multiply this weight by 5, divide by the weight of paper taken and multiply by 100 to obtain the percentage of starch in the paper. At least two 50 cc. portions of the neutral dextrose solution should be thus analyzed and the average result taken. NOTES. (1) The saliva is obtained by chewing paraffin wax and collect- ing the saliva in a beaker. Then add one-half to two-thirds its volume of distilled water and filter. The saliva should be tested with starch to be sure that it is active. (2) No method is known which will accurately ascertain the amount of starch in papers. The above method, however, is probably the most nearly accurate, as cellulose is not appreciably affected by saliva. (3) This method depends upon the fact that the diastase in saliva will hydrolyze starch but will not hydrolyze the cellulose of paper. It some- times happens, however, that saliva is not active towards starch and it should, therefore, always be tested by running a "blank" on a little starch and noting by means of the iodine test whether the starch is completely hydrolyzed. Detection of Faulty Sizing.* Draw a strip of the paper over the surface of an iron tannate ink and allow it to drain and dry naturally. Examine the inked surface under the microscope. A well-sized paper will show no indication of the fiber having absorbed ink, and the entire surface will be uniformly and lightly colored, as indicated in Fig. 21. In more poorly si&ed paper blotches of fibers absorb the ink, as shown in Fig. 22 and Fig. 23. Fig. 236 shows water-leaf filter paper, indicating by the darker colors that the fibers have absorbed the ink. NOTES. (1) All the papers shown in these plates (except Fig. 23&) are, according to the ordinary methods of testing, well-sized and practically identical. (2) A paper well-sized tlyoughout should also show a uniform coloring when the surface has been rubbed with an ink eraser, the loose particles brushed off and the paper treated with ink as above. * United States Dept. of Agriculture, Bureau of Chem. Cir. No. 107. WOOD, PAPER AND PAPER-MAKING MATERIALS 359 FIG. 21. Appearance of Well-sized Paper after Test with Iron Tannate Ink. 360 TECHNICAL METHODS OF ANALYSIS . FIG. 22. Appearance of Poorly-sized Papers in Which Blotches of Fibers Absorb the Ink. WOOD, PAPER AND PAPER-MAKING MATERIALS 361 FIG. 23a. Appearance of Paper with Inferior Sizing. FIG. 236. Water-leaf Filter Taper, Showing Ink Absorbed. 362 TECHNICAL METHODS OF ANALYSIS TARNISHING TEST FOR PAPER General. A paper which is to be used for wrapping silver- ware should be essentially free from active sulfur compounds. The method of testing so-called " anti-tarnish" paper consists, in general, of comparing the sample to be tested with special papers impregnated with 0.001% and 0.0001% Na 2 S solutions, the sulfide test in each case being made under prescribed conditions by a hydrogen evolution method and lead acetate paper. Preparation of Special Impregnated Papers. Make the special papers from 10 cm. best white filter paper, each of which weighs approximately 0.6 gram. Prepare the following solutions: (a) Dissolve 3 grams of fresh sodium sulfide crystals in 100 cc. of distilled water. (3 grams of Na2S-9H2O are equivalent to 1 gram of Na2S.) (b) Dilute 1 cc. of solution (a) to 1 liter to make a 0.001% solution. (c) Dilute 10 cc. of solution (b) to 100 cc. to make a 0.0001% solution. Saturate the filter paper in solutions (b) and (c) and dry in air. Considerable quantities of these papers may be made at one time and stored in separate, tightly stoppered bottles labeled: " 0.001% Na 2 S Paper for Tarnishing Test." " 0.0001% Na 2 S Paper for Tarnishing Test." The papers may also be torn into four equal segments, each seg- ment (0.15 gram) being sufficient for one test. Materials Required. (1) Four 500 cc. flat bottom flasks, approximately 7 inches high ; (2) Granulated zinc (arsenic free) ; (3) 15% HC1 solution; (4) Lead acetate test paper, moistened; (5) Absorbent cotton. Method. Into each flask put 2 grams of granulated zinc and 0.15 gram of paper torn into small pieces. The 4 flasks are for the following papers: (1) Sample; (2) Pure filter paper (for a "blank"); (3) 0.001% Na 2 S paper; and (4) 0.0001% Na 2 S paper. Add to each flask 25 cc. of 15% HC1 (free from As). Into the neck of the flask insert a loose plug of cotton to a depth of about 1.5 inches. Above the cotton place a piece of moistened lead acetate test paper about 1 inch square, and cover WOOD, PAPER AND PAPER-MAKING MATERIALS 363 this loosely with a final plug of cotton. Set the 4 flasks in a pan or tub containing water at room temperature to a depth of 0.25- 0.50 inch, in order to prevent any considerable rise of temperature of the contents of the flask. The liberated hydrogen will carry any H^S evolved up to the lead acetate paper, which will darken. Examine the 4 lead acetate papers at the end of thirty, sixty and ninety minutes and record their comparative appearances. Interpretation of Results. It has been found that the 0.001% Na2$ paper causes some tarnishing when held in contact with a polished 10-cent piece for five weeks. Commercial papers known to have caused tarnishing of polished metal goods have been found to be more reactive under this test than the 0.001% Na2S paper. Therefore, a paper to be acceptable should show up as well as the 0.0001% Na2$ paper (which should show slight discolora- tion in about sixty minutes). A paper between 0.0001% and 0.001% Na2$ papers is dangerous; while those that are inferior to 0.001% Na2$ paper should be unquestionably rejected. In reporting, a paper superior to 0.0001% Na2S paper should be classed as "safe"; those between 0.0001% and 0.001% as "questionable"; and those inferior to 0.001% COTTON CELLULOSE (COTTON LINTERS) FOR NITRATION General. U. S. Navy Specification 65C5, July 10, 1913, requires that cotton for nitrating for smokeless powder shall con- tain not over 7% of water, not over 0.4% of ether extract, not over 0.8% of ash and only traces of lime salts; and that it shall be free from hypochlorites, dirt and foreign matter. This is essentially the same as the Joint Powder specifications which require the use of a bleached cellulose containing not more than 0.4% of extractive matter, not more than 0.8% of ash, and state that it should not contain more than " traces " of lime, chlorides and sulfates. For some commercial grades of nitrocellulose unbleached cotton is used, but the methods of analysis are the same as for bleached cotton. The routine analysis of cotton includes the determination of moisture, ether extract, ash, solubility in 95% H2SO4 and solu- bility in 10% KOH solution. The furfural value is also fre- 364 TECHNICAL METHODS OF ANALYSIS quently determined, and on crude fiber the amount of " dust " is determined by a sieving test. It is also sometimes desirable to determine the copper number. Sampling. In sampling fiber in bales, take a section extending from one side to approximately the center from each bale sampled, and take samples from not less than one-tenth of the bales in the lot. In sampling crude fiber, make special note of, and take sam- ples from, any bales showing large proportions of moldy fiber or of very oily fiber, as indicated by a strong yellow color. Blend the samples for moisture quickly and thoroughly by hand and place a sample of about 20 grams in a previously weighed glass or tin vessel with a tightly fitting cover. Open up the remainder of the sample by hand, or in a mill or picker if available, and after being thoroughly blended, reduce to proper size by quar- tering, and dry at 105 C. Make all determinations, except moisture and dust, on the dry sample. Moisture. Take about 20 grams of the sample prepared for moisture determination and weigh under conditions to avoid changes in moisture content. Dry at 105 C. for three hours, or if the moisture is high, as may happen with samples taken from bales that have been exposed to rain, until constant weight is reached. Calculate the loss in weight to per cent of the sample as received. NOTE. Instead of determining the moisture on a large scale and using portions of this dried material for subsequent tests, it is permissible to run the tests on the sample as received and calculate results to the dry basis; pro- vided, however, that the moisture is not excessive (5-10%) and that the sample is kept where it does not lose or gain moisture. Ash. Place at least 5 grams of dry material, prepared as above in a platinum or fused silica dish of 80-100 cc. capacity and ignite in a muffle or over a burner to complete incineration at a low red heat, taking care to avoid loss of ash by air currents during handling. Finally cool the dish and contents in a desiccator and weigh. Calculate the result to per cent of the dry weight of the cotton. NOTE. The Powder Specifications call for the digestion of a 1.5 gram sample with a little pure HNO 3 and incineration at a red heat. The use of the smaller sample of cotton and a higher temperature of incineration, how- ever, is likely to give lower results. WOOD, PAPER AND PAPER-MAKING MATERIALS 365 Ether Extract. Extract thoroughly about 5 grams of the dry material with pure ethyl ether in a suitable extraction apparatus (preferably Knorr's; see Eimer & Amend Catalog, No. 3185) for about eight hours or until further extraction removes no additional substances soluble in ether. Weigh the extractive matter after drying at 100 C. to constant weight, and calculate the result (after deducting the weight of any residue in the ether) to per cent of the dry weight of the material. Take care to have the extractive matter free from fine particles of fiber which may be carried through mechanically. After extraction with ether, the sample may be dried and used for the determination of non-cellulose, as below. Non-Cellulose. Tho H 2 S0 4 used must be within 0.5% of 95% strength. Make the determination by treating 5 grams of the dry sample at about 20 C. with 50 cc. of the acid at the same temperature. In the case of crude fibers, it is important to remove the oils by extraction with ether before making this determination. Stir the fiber vigorously in the acid for five minutes, then slowly pour into 1 liter of cold distilled water. Heat the aqueous solution on a hot plate for at least four hours, with frequent stirring. It is important to keep the temperature at 99-100 C. and to main- tain the level of the liquid constant. Then filter out the insoluble matter on a Gooch crucible with a carefully prepared asbestos mat. Thoroughly wash the contents of the Gooch crucible with boiling distilled water to remove the last traces of H 2 SC>4 and then dry for three hours at 102-105 C. Cool and weigh the non-cellulose and calculate the result to per cent on the dry weight taken. It is important to have the fiber well opened up and free from lumps, as, if lumps are present, a higher result may be expected. Approximate Cellulose. Calculate the " approximate cel- lulose " in the fiber by adding together the percentages of ash, ether extract, and non-cellulose and subtracting from 100%. Solubility in Caustic. Cellulose is insoluble in alkalies, so that in a crude fiber the solubility in KOH or NaOH is a measure of the non-cellulose present. In a purified fiber it is a measure of the severity of bleaching, and indicates the amount of hydrocel- lulose and oxycellulose present. Prepare a solution of pure KOH within 0.1% of 10% concen- tration by dissolving the proper weight of the purest obtainable 366 TECHNICAL METHODS OF ANALYSIS KOH in distilled water.* Carefully check the strength of the solution by titration with standard acid and phenolphthalein. It must be carefully protected from CO2, as carbonates do not act. Dry approximately 2 grams of the sample in a wide-mouth weighing bottle to constant weight at 102-105 C., transfer the contents of the bottle to a 250 cc. Pyrex glass or porcelain beaker, add 100 cc. of the solution, cover with a watch glass and heat at 100 C. for three hours. Heating on a steam bath is not satis- factory for this purpose, as it does not give a sufficiently high temperature. Care must be taken to avoid concentration of the solution or undue oxidation of the fiber due to exposure of the alkali-soaked fiber to the air. It is important that the tempera- ture be kept within 99-101 C., since variations in temperature affect the result materially. After the heating is completed, pour the contents of the beaker into a 2-liter beaker containing 1 liter of distilled water and wash any residue in the small beaker into the larger. Then neutralize the alkali with a decided excess of acetic acid, the excess of acid being necessary in order to break up the combina- tion of alkali and cellulose. Filter the undissolved cotton into a weighed Gooch crucible having an asbestos mat and thoroughly wash successively with hot water, alcohol and ether. Then dry rapidly to constant weight at 102-105 C. Calculate the loss of weight to per cent of the dry material. NOTE. In making this determination on crude fiber, the amount soluble in hot water alone is deducted from the total and expressed separately, and a further correction must be made for the per cent of oils extracted by ether and the per cent of ash which goes into solution in the acetic acid, though these corrections are not necessary on bleached fibers. In order to determine the amount of ash which goes into solution, an ash determination must be made on the fiber after treatment. Furfural Value (Pentosans). Preparation of Reagents. Test the purity of the phloroglucinol by dissolving a small quantity in a few drops of acetic anhydride, then heat almost to boiling and add a few drops of cone. H 2 SO4. A violet color indicates the pres- ence of diresorcin. If the phloroglucinol gives more than a faint coloration it should be purified by the following method : * A solution of pure NaOH within 0.1% of 7.14% may be used instead of a 10% KOH solution. WOOD, PAPER AND PAPER-MAKING MATERIALS 367 Heat in a beaker about 300 cc. of 12% HC1 (1 : 2) and 11 grams of the commercial phloroglucinol. Add the latter in small quantities at a time, stirring constantly until it has almost entirely dissolved (some impurities may resist solution). Pour into a sufficient quantity of the same HC1 (cold) to make the volume 1500 cc. Let stand at least overnight, better several days, to allow the diresorcin to crystallize out, and filter immediately before using. The solution may turn yellow, but this does not interfere with its usefulness. In using, add the volume containing the required amount to the distillate. Determination. Place a weighed quantity of the material, chosen so that the weight of furfural phloroglucid obtained shall not exceed 0.300 gram, in a flask together with 100 cc. of 12% HC1 and several pieces of recently heated pumice stone. Place the flask on a wire gauze, connect with a condenser, heat rather gently at first, and so regulate as to distill over 30 cc. in about ten min- utes, the distillate passing through a small filter' paper. Replace the 30 cc. driven over by a like quantity of the dil. HC1, added by means of a separatory funnel in such a manner as to wash down the particles adhering to the sides of the flask, and continue the process until the distillate amounts to 360 cc. To the total distillate grad- ually add a quantity of phloroglucinol (purified if necessary) dis- solved in 12% HC1 and thoroughly stir the resulting mixture. The amount of phloroglucinol used should be about double that of the furfural expected. . The solution first turns yellow, then green, and very soon an amorphous greenish precipitate appears, which grows rapidly darker, till it finally becomes almost black. Make' the solution up to 400 cc. with 12% HC1 and let stand overnight. Filter the amorphous black precipitate through an asbestos felt in a Gooch crucible which has been previously dried and weighed in a weighing bottle. Wash carefully with 150 cc. of water in such a way that the water is not entirely removed from the crucible until the very last, then dry for four hours at the temperature of boiling water, cool and weigh in the weighing bottle, the increase in weight being reckoned as furfural phloroglucid, Calculate to furfural, using the following formulae given by Krober: (a) For weight of phloroglucid, w, less than 0.030 gram: Furfural = (w+0.0052) X0.5170. 368 TECHNICAL METHODS OF ANALYSIS (6) For weight of phloroglucid, w, more than 0.300 gram: Furfural = (w+0.0052) X0.5180. (c) For weight of phloroglucid, w, between 0.030-0.300 gram: Furfural = (w+0.0052) X0.5185. Copper Value. Cut 3 grams of the sample into small pieces and mix with 150 cc. of boiling water. Heat to boiling 50 cc. of Fehling's copper solution and 50 cc. of Fehling's alkaline tartrate solution * and add to the sample. Then boil the whole with con- tinuous stirring for fifteen minutes, filter and wash with hot water. Warm the residue with dil. HNOs, to dissolve out the absorbed Cu, and filter. Determine the Cu electrolytically on a rotating cathode. Calculate the " copper value " or " copper number," which is the weight in grams of metallic Cu thus obtained per 100 grams of dry material. NOTES. (1) When this determination is used as a basis of acceptance or rejection, 6 samples shall be selected from various parts of the bale. The copper value of each of these samples shall be determined as above and the highest copper value of each of the 6 shall be considered the copper value of the lot. (2) Most specifications require that the copper value shall not exceed 2. (3) The copper value is a measure of the oxy-cellulose and indicates whether the material has been over-bleached. Mitchell and Prideaux: "Fibers in the Textile Industry," page 93, give representative copper numbers as follows: Surgical Cotton Wool 1.6 Parchment Paper 4.2 Bleached Sulfite Pulp 3.9 Overbleached Sulfite Pulp . 193 REFERENCES. Part of the above method was originally furnished us by W. F. Allen of the Meridian Cellulose Co., and is the method used by E. I. Du Pont de Nemours & Company. The procedure for Furfural Value is the tentative method of the Assoc. of Official Agricultural Chemists pub- lished in its Journal, Methods of Analysis (1916), page 110. WOOD DISTILLATE PRODUCTS General. The chief hard woods used in this country for dis- tillation are beech, birch and maple. The products of distilla- tion are gas, crude pyroligneous liquor and charcoal. This method is concerned only with the analysis of the crude liquor, which * Use the Soxhlet modifications (see page 3) . WOOD, PAPER AND PAPER-MAKING MATERIALS 369 consists essentially of acetic acid, crude wood alcohol, tar, and water. ANALYSIS OF CRUDE LIQUOR Tar. In commercial practice the bulk of the tar is mechan- ically separated, but the liquor always contains more or less dis- solved or suspended tar. Weigh the whole sample and determine its volume. Separate mechanically as much of the tar as possible, weighing the amount thus separated. If desired, its sp. gr. may be determined by the Westphal balance or a pycnometer. Determine the dissolved tar as later described and add the amount to the amount obtained by mechanical separation. Acetic Acid. Place 100 cc. of the liquor in a weighed retort or distilling flask, set up in an oil bath and connect with a Liebig condenser. Place a thermometer in the oil and gradually heat the oil bath to 140 C., collecting the distillate in a 250 cc. graduated flask. Keep the temperature of the bath at 140 C. until nothing more comes over. In the flask there will remain about 10 cc. of tar which still contains some acetic acid. The last traces of acid must be blown over by a current of steam, keep- ing the oil bath at 150 C. and collecting the distillate in the flask with the main distillate. Make up the distillate to the mark at 15.5 C. Pipette out 50 cc. at the same temperature and titrate with N NaOH and phenolphthalein. Calculate to acetic acid and to calcium acetate. CALCULATIONS. 1 cc. N NaOH = 0.06004 gram HC 2 H 3 2 . = 0.07907 gram Ca(C 2 H 3 2 )2. NOTE. For ordinary purposes, instead of blowing over the last traces of acid with steam, satisfactory results can be obtained by cooling down the flask, adding 50 cc. of water and again distilling until nothing more comes over at 140 C. (See U. S. Dept. of Agriculture, Forest Service, Bulletin 129, page 6.) Dissolved Tar. The residue in the flask from the distillation of the acetic acid represents the dissolved tar. Cool and weigh, and add the amount to the tar mechanically separated above. Crude Wood Alcohol. Place 1 liter of the crude liquor in a round bottom 1500 cc. flask. Set the flask in an oil bath and 370 TECHNICAL METHODS OF ANALYSIS connect to a vertical condenser. Distill until 500 cc. of distillate have been collected. Neutralize the distillate with NaOH and distill again from a smaller flask until 50% has distilled off (250 cc.). This distillate is still too dilute to estimate the alcohol accurately, especially as it still contains methyl acetate which, owing to its high gravity, would make the results come low. Hence, the dis- tillate should again be neutralized with NaOH and redistilled, collecting in a graduated 100 cc. flask. Stop the distillation just before 100 cc. have come over, cool to 15.5 C. and make up to volume accurately with distilled water. Determine the sp. gr. of the distillate at 15.5 C. with a Westphal balance, or preferably with a pycnometer. From the sp. gr. calculate the percentage of methyl alcohol in this distillate, both by weight and by volume, and figure back to the original sample. (See page 458.) NOTES. (1) Great care must be exercised in making the distillation not to lose alcohol. A vertical condenser is preferable to a horizontal one and the end of the condenser should run well down into the neck of the flask. (2) By this method all the substances which accompany methyl alcohol, such as acetone, acetaldehyde, allylalcohol, etc., are calculated as methyl alcohol. The results, therefore, should be reported as Crude Wood Alcohol and if the determination of the true methyl alcohol content is desired, use the method of Zeisal and Stritar. (Zeit. fur Anal. Chem. 29, 359; 42, 579; 43, 387.) ANALYSIS OF PRODUCTS OF CRUDE LIQUOR Acetic Acid. Two kinds of acetic acid from wood are recog- nized : (1) Crude wood acid, containing about 6% of acetic acid. (2) Rectified wood acid, so-called " light vinegar/' which has a similar strength. The first product is obtained from the crude wood liquor by distilling off the alcohol and diluting the residue in the still with water until it has an acid content of 6%. The second product is made from a single distillation of the crude wood acid, which yields a pale yellow liquid still containing about 6% acid. The determination of the strength of the acetic acid can be made in the case of tar-free wood acid (product 2) by titrating with N NaOH and phenolphthalein. In the case of the raw acid, however, the end point is often difficult to obtain with certainty and it is best to dilute it 1 : 10 before attempting titration. WOOD, PAPER AND PAPER-MAKING MATERIALS 371 The tar contained in crude liquor gives a strong coloration with NaOH and it is often impossible to get an accurate titration. This may be overcome by using litmus or phenolphthalein as an outside indicator on a white porcelain tile. For accurate results, however, it is preferable to make a distillation and titrate the dis- tillate as described in the first part of the method. Acetone. The acetone content of crude wood alcohol is determined by the well-known Messinger method, which is described on page 72. Total Ketones. For general purposes a determination of the acetone content, as obtained by the Messinger method, gives sufficient information. In case the total ketone content is desired, the method of Deniges (Comp. rend. 127, 963) may be employed. CHAPTER IX ANALYSIS OF TEXTILES AND TEXTILE FIBERS STRUCTURAL ANALYSIS OF TEXTILE FABRICS General. The following is a brief outline of the determina- tions usually desired in making structural analyses of textile fab- rics and a description of the procedures employed. Fiber Composition. See page 377. Ash (Mineral Weighting). Determine the ash on 2-5 grams of sample in a porcelain crucible and report the percentage as mineral weighting. The weighting usually consists principally of tin salts, Fe 2 O3, Prussian blue, Si0 2 and P2O5. Tin phosphate is not uncommon. The estimation of the mineral matter is of special importance in the examination of such fabrics as water- proof rain-coats, window shades, bookbinders' cloth, fireproof cloth, heavily weighted silks, etc. Black silks sometimes contain 250% of weighting, based on the weight of the silk. NOTE. In waterproof goods ammonium salts often occur and these, of course, would be volatilized in ashing and must be tested for on a separate portion. Sizing Materials. The sizing usually consists of starch, gums, glue, oil, fat or wax. Nearly all textiles will show the presence of small amounts of various sizing materials which it is necessary to use in dyeing and finishing in order to produce the desired feel and finish. It is seldom necessary to determine this quantitatively. Boiling hot water or a solution of " diastafor " will remove starch. Gums are also as a general rule soluble in water and so is glue. Oils and fats can be removed by ether extraction and waxes by alcohol extraction. When the sizing consists of starch and British gum (dextrin), the amount may be determined as follows: Dry approximately 5 grams to constant weight at 100 C. Determine the loss. 372 ANALYSIS OF TEXTILES AND TEXTILE FIBERS 373 Boil the dried sample in water for ten minutes. Rinse well and digest two hours at 60 C. in a solution of 15 cc. of commercial " diastafor " in 500 cc. of water. Wash well in hot EkO, boil one hour in 500 cc. of distilled water, wash again, dry at 100 C. and weigh. total loss moisture loss CALCULATION. % size =- . . -X100. wt. original sample Bursting Strength. Bursting strength is determined on the Perkins Mullen tester (Fig. 18, page 349) and reported in pounds per square inch. The average of at least 5 tests, and preferably 10, should be taken. Tensile Strength and Stretch. There are two methods of determining tensile strength as follows : (A) STRIP METHOD. Cut pieces of fabric, both in the warp and in the filling direction, each piece 8 inches long and 1.5 inches wide. Ravel down to the desired width, or the specified number of threads per inch, by removing approximately an equal number of threads from each side. Insert these strips in the test machine. (In case stretch is to be determined, 2 parallel marks should be made on the specimen 3 inches apart, and the lower edge of the jaws should be clamped even with these marks.) After clamp- ing the jaws, apply power to the machine to make the sepa- ration of the jaws proceed at a uniform rate of 20 inches per minute. From the breaking strength of the strip and its width calculate the tensile strength of the fabric in each direction in pounds per inch. Also measure the elongation or stretch at time of breaking between the lines drawn 3 inches apart. Express the stretch results in percentage on 3 inches. (B) GRAB METHOD. It is necessary to use this method in case of hosiery and knitted goods which cannot be raveled down. Tests should be made on the Scott tester. Test specimens should be 5 inches long by 2 inches wide and cut in each direction. On each piece as it lies flat without tension on a smooth surface, draw pencil lines along the thread vertically and 0.5 inch from each edge so that 1 inch of fabric will be between them. Then draw pencil lines along the thread horizontally and 1 inch apart in the center of the specimen. The clamps of the machine consist of 2 374 TECHNICAL METHODS OF ANALYSIS jaws, one at least 2 inches wide and the other 1 inch wide. Clamp the specimen in the 2 jaws of the machine securely, taking care that the threads being tested are parallel to the direction of pull, and the cross threads are at right angles to it. Have the length of the test specimen between jaws exactly 1 inch. Report the tensile strength in pounds per inch and the increase in length of the 1 inch portion as the 'elongation, either in units of length or as percentage of 1 inch. NOTE. On tensile and elongation tests the average of at least 5 and prefer- ably 10 tests in each direction should be made. Folding Endurance. Materials such as silk, book-cloth, win- dow shades, etc., are subject to folding to a considerable extent in actual use. In such cases a folding endurance test will show to what extent they may be expected to resist deterioration from this cause. Make the test on the Schopper folding machine (Fig. 17, page 348). Cut strips from each direction about 30 mm. wide and ravel down until they are exactly 15 mm. wide. Insert in the stand, under a machine and determine the number of double folds they will tension of 1 kilo, before breaking. At least 5 tests, and preferably 10, should be made in each direction and the average figures reported. Thread Count. This term is used in reporting the number of threads per inch in the fabric in each direction. The greater the number of threads per inch, the finer or closer is the texture of the fabric. Count the threads in each direction with a special counting glass and report the number of threads in the filling as " picks per inch " and in the warp as " threads per inch." Twist. This term is used in determining the number of twists per inch in a thread. It is also customary to designate them as right or left twists. Size of Yarn. The terms " yarn number " and " size " are commonly used to indicate the length of yarn per unit of weight. There are many systems of numbering in use which are for the most part based on arbitrary quantities, differing according to the kind of material or the locality, or even the preference of the individual. For use in this laboratory, we have adopted the " Fixed Weight System " in which the count represents the length (hanks or yards) of a fixed weight. In this system the count number is inversely proportional to the size of the yarn. The ANALYSIS OF TEXTILES AND TEXTILE FIBERS 375 basis of a No. 1 yarn on the " Fixed Weight System " is shown in the following table : FIXED WEIGHT SYSTEM Material Unit Usage Woolen 1,600-yd. lengths per pound Anglo-American Cotton 840-yd. lengths par pound World Worsted 560-yd. lengths pee pound Anglo-American Linen 300-yd. lengths per pound World Raw Silk 1-yd. lengths per ounce England A No. 2 cotton would contain 2X840 yards per pound and a No. 3 cotton, 3X840. Ply yarn is numbered to indicate the number of strands and the size of a single yarn, thus: 2-40 means that 2 strands of size 40 are placed together to form a 2-ply yarn. Spun silk is an excep- tion to this rule, the first number indicating the count or size of the ply yarn and the second the number of strands, thus: 20-2 spun silk indicates that 2 strands of 40 size have been doubled or twisted making a 2-ply yarn equal to a single 20. Silk is usually numbered either by the dram or the denier system. The dram system represents the number of drams per 1000 yards; the denier system represents the number of grains per 638 yards. By weighing a given length of yarn it is possible to calculate the number of unit lengths per unit weight, and the size of the yarn. Weight. The weight of fabrics is reported as ounces per square yard. In case of hosiery it is reported as weight per dozen pairs of hose. Weigh a piece of definite size and calculate the weight per square yard. CALCULATION. Grams per sq. in.X45.71 = oz. per sq. yard. Special Tests. (1) FASTNESS TO LIGHT. The rational test for determining fastness to light of dyed goods is exposure to bright sunlight. This is not practical, however, as it requires too much time. A fairly satisfactory indication of the relative fast- ness to light can be obtained by exposing a small piece of the goods 376 TECHNICAL METHODS OF ANALYSIS for seven hours to ultra-violet light and comparing with a piece of the original goods. In the case of very fast colors it is sometimes desirable to make a much longer exposure. (2) FASTNESS TO WASHING. (A) Cotton. Dissolve 2 grams of Ivory soap in 1000 cc. of water and immerse a piece of the sample in a portion of this solution for one-half hour at" 60 C., together with a piece of white cotton and a piece of white woolen goods. If the color strips off the sample and stains the white material, it is not fast to washing. (B) Wool. Dissolve 2 grams of Ivory soap and 0.5 gram of Na2COa in 1000 cc. of water and wash as in the case of cotton for one hour at 50 C. (3) FASTNESS TO PERSPIRATION. Dissolve 50 grams of 50% acetic acid and 100 grams of NaCl in a liter of water. Immerse a portion of the sample in this solution for fifteen minutes, let dry and repeat the operation twice. Compare the final dry piece with the original and note any change. (4) FASTNESS TO MUD SPOTS. Splash a sample of material with street mud, let dry, brush off the mud and note the effect. (5) HOT IRONING TEST. Iron with a hot iron and note if heat changes the color. This shows the sensitiveness of colors to heat and in many cases, especially with light shades, colors will be found to change to a considerable extent. Most of them assume their original shade on cooling, but in some cases the change is perma- nent. Tlie test is also of value on certain classes of fabrics, such for instance as Bolivia cloths, where hot ironing, especially in the presence of moisture, tends to destroy the surface finish of the fabric. In the case of heavily weighted silks, continued hot ironing will have a tendency not only to change the shade of some of the colors, but to cause the material to deteriorate and, in some instances, to crack badly. Hosiery. In hosiery analysis the terms " wales " and " courses " are used and correspond to warp and filling, respect- ively, in woven fabrics. The term " reinforcing thread," used in toe and heel, applies to an extra thread twisted in to reinforce and give additional strength. The weight is reported in ounces per dozen pairs. ANALYSIS OF TEXTILES AND TEXTILE FIBERS 377 FIBERS IN CLOTH AND YARNS General. The following method gives the procedure to be used in analyzing a complex mixture of textile fibers. It is seldom that more than two fibers are found in the same fabric, and where the fibers are known, as for instance in a wool-cotton mixture, the procedure can be very much shortened. Moisture. Weigh 2-5 grams of sample in a glass-stoppered weighing bottle and dry at 100-110 C. to constant weight. Report the loss as moisture. Sizing Materials, etc. Use the dry material from the moisture determination and boil thoroughly in very dil. HC1. The solu- tion should not be over 1% in strength and care must be taken not to disintegrate the fibers. If the fibers are much weakened or disintegrated, it is well to repeat, using weaker acid. Repeat the boiling until no residue is found upon evaporating a few drops of the liquid on a watch glass. This removes mineral loading matter soluble in HC1, finishing materials, and more or less dye- stuff, as well as other materials soluble in water. Next extract with alcohol, and finally with ether, until all soluble material is removed. Dry and weigh. This gives the weight of anhydrous fibers, together with any mineral matter not soluble in HC1. Divide immediately into 2 parts, weighing each. Ash one portion. This will give the natural insoluble ash of the fibers and any insoluble loading materials, such as certain mordants and tin salts. Cal- culate the weight back to percentage of the original material. Silk. Heat to boiling about 60 cc. of basic ZnCk solution,* and immerse in it the second portion of the above extracted sam- ple. Remove the flame, stir the sample thoroughly for about one minute and again bring to boiling. Transfer immediately to a previously weighed Gooch crucible, using the insoluble fibers (cotton and wool) for the mat of the crucible. Remove the insol- uble fibers to a beaker containing cold, dil. HC1 (5%). Thor- oughly wash in the beaker by agitation and again filter. Repeat the process several times to remove all ZnC^, using weaker acid solution at each subsequent washing. Finally remove to the * The zinc chloride solution is made as follows : Add an excess of Zn metal to cone. HC1 and let stand twenty-four hours. Filter through glass wool and slightly acidify with HC1. 378 TECHNICAL METHODS OF ANALYSIS crucible and finish the washing with several portions of hot water until free from Cl. Dry at 100-105 C., and weigh. The loss in weight represents silk (and in certain cases a portion of the sizing material). If the material is pure natural silk, no residue is left at this point of the analysis. NOTE. This treatment also removes artificial silk. Wool. Treat the above residue with a solution of approxi- mately 5% KOH and boil gently for ten to fifteen minutes. Pour into 500 cc. of cold water, let stand until the fibers have settled, and decant carefully the supernatant liquor, taking care to avoid loss of any fibers. Filter through a weighed Gooch crucible and wash with hot water, and finally with about 10 cc. of alcohol. Dry and weigh. The loss in weight by the KOH treatment repre- sents wool. Cotton, etc. The residue in the crucible at this stage is cotton, or vegetable fibers (wood fiber, jute, linen and similar material). Ignite and weigh. The loss represents the various fibers, the nature and approximate proportions of which must be determined microscopically on another portion. The residue in the crucible may consist of asbestos, mineral wool, etc., and should be exam- ined qualitatively and microscopically. Cotton- Wool Mixtures. In the case of fabrics consisting only of cotton and wool, determine the moisture as above described and remove sizing materials, if necessary, by boiling with 1% HC1 solution. Wash out the acid and dry at 105 C. to get the total weight of fibers. Boil the residue for fifteen to twenty minutes with 5% KOH solution and proceed as directed above under Wool, the final residue being cotton. Calculation of Results. About 5% of cotton is soluble in KOH solution. To obtain the true weight of bone-dry cotton, there- fore, divide the actual weight of cotton by 0.95 and also subtract this increment from the wool as figured by loss to KOH. Cal- culate the percentages of the various fibers on the bone-dry basis. As the natural moisture of cotton fiber is different from that of wool, where the fabric has a cotton thread in one direction and wool in the other, the threads should be separated and the moisture determined on each. In cases where wool and cotton are twisted together, however, this is not possible. The U. S. Government ANALYSIS OF TEXTILES AND TEXTILE FIBERS 379 allows a re-gain of 11% in wool for army blankets, etc. For silk the re-gain allowed should also be 11%, and for cotton and vege- table fibers 7%. Where it is not possible to determine the moisture of the different fibers, the air-dry percentage should be figured by dividing the percentages of wool, cotton (vegetable fiber) and silk by 0.89, 0.93 and 0.89, respectively. If the sum then exceeds 100% (showing that the fabric was below normal moisture), correct these figures by multiplying each percentage thus obtained 100 by -T-, where A is the sum total. A. EXAMPLE : Preliminary analysis: Per Cent Moisture (loss at 100 C.) 8 . 74 Sizing, etc. (loss to HC1) 3.28 Fibers (by difference) 87 . 98 Analysis of bone-dry fibers : Silk (loss to basic ZnCl 2 ) 22. 10 Wool (loss to KOH) 30.64 Cotton (residue) 47 . 26 Cotton (corrected, 47.26n-0.95) 49.75 Wool (corrected, 30.64-2.49) 28. 15 Air-dry analysis : Silk (22.10-^-0.89) 24.83 Wool (28.15^-0.89) 31.63 Cotton (49 .75-^-0.93) 53.49 Sizing, etc 3.28 Total 113.23 100 Final corrected analysis (air-dry analysis figures X ) 11 o.Jo Silk 21.93 Wool 27.93 Cotton 47.24 Sizing, etc 2 . 90 100.00 Distinction between Natural and Artificial Silks. The princi- pal varieties of artificial silk are: (1) Pyroxylin Silk (Chardonnet Silk *) : made from a solution of nitrated cellulose in a mixture of alcohol and ether, the cel- *Lehner Silk is also a pyroxylin silk made by a process somewhat different from that of Chardonnet. 380 TECHNICAL METHODS OF ANALYSIS lulose being usually afterwards denitrated with dil. HNOs, Feds and (NH 4 ) 2 HP0 4 . (2) Cupr ammonium Silk (Pauly Silk) : made from a solution of cellulose in ammoniacal copper solution (or sometimes ammonia- cal chloride of zinc). (3) Viscose Silk: made from a solution of alkaline cellulose xanthate prepared by the action of NaOH and 82 on mercerized cellulose. (4) Gelatin Silk: made from gelatin filaments rendered insolu- ble by treating with formaldehyde. (5) Acetate Silk (Celestron Silk, Lustron Silk): made from cellulose acetate. A cold solution of chromic acid dissolves all artificial silks, whereas real silk dissolves but slowly and cotton and other vege- table fibers are unaffected. KOH solution does not dissolve col- lodion or cellulose silks, but in a boiling solution gelatin silk and real silk are soluble. Schweitzer's reagent dissolves collodion and cellulose silks, as well as natural silk, whereas gelatin silk is insoluble and stains the liquid bright violet. Loew's reagent dissolves real silk immediately at 80 C. It will dissolve Tussah and gelatin silk when boiled for one minute; other artificial silks are not affected. The best solution for separating real silk from wool, cotton, and artificial silk is Loew's reagent. The differentiation between the three most common varieties of artificial silks, namely, Pyroxylin, Cuprammonium and Vis- cose silks, can be quickly made with 2 reagents, Fehling's solu- tion and zinc chloride-iodine solution, as follows: Heat 0.2 gram of the silk with 2 cc. of Fehling's solution on the water bath for 10 minutes in a test-tube and then fill the test tube with water. Pyroxylin silk produces a green color, whereas the other two give a clear blue. Furthermore, on the threads of the pyroxylin silk there will be noticeable a yellowish precipitate of cuprous oxide or hydroxide. The reaction depends upon the different reducing powers of the artificial silks. Only in the case of the pyroxylin silk (nitrocellulose) is the reducing power appreci- ably increased. To further distinguish between cuprammonium and viscose silks, cover equal weights of the silks in a test-tube with zinc chloride-iodine solution and after a few seconds pour off the excess ANALYSIS OF TEXTILES AND TEXTILE FIBERS of the reagent. Then fill the test-tube with water, pour the water off and repeat this washing process until the water is only light yellow or colorless. Cuprammonium silk is only weakly colored under these conditions and loses the brown shade very quickly when washed, whereas viscose silk is colored a bluish-green and retains the color a longer time. Cellulose acetate silk is soluble in a mixture of 5 parts of chloro- form and 2 parts of denatured alcohol by volume. As a confirm- atory test it may be saponified with KOH, forming potassium acetate, which in turn will yield acetic acid when treated with H 2 SO 4 . Pyroxylin silk which has not been denitrated can also be saponi- fied with KOH, forming KNOs which may be identified by the common qualitative test for nitrates. It is advisable in making the above identity tests to run com- parison tests with artificial silks of known origin. REAGENTS. The different reagents are made up as follows : (a) Schweitzer's Reagent. Dissolve 5 grams of copper sulfate crystals in 100 cc. of boiling water, add NaOH solution to com- plete precipitation, wash the precipitate thoroughly, and then dissolve in the least quantity of cone. NKUOH. This should give a deep blue solution. (6) Loew's Reagent. Dissolve 16 grams of copper sulfate in 150 cc. of water and add 10 grams of glycerol. Then add care- fully a solution of NaOH until the precipitate which at first forms is just re-dissolved. (c) Fehling's Solution. The Fehling's solution is made by mixing just before use equal volumes of the Soxhlet modifications of Fehling's copper solution, and Fehling's alkaline tartrate solu- tion, the formulas of which are given on page 3. (d) Zinc Chloride- Iodine Solution. Dissolve 2 grams of KI and 0.1 gram of iodine in 5 cc. of water. Add to this a solution of 20 grams of ZnCk in 10 cc. of water. Let settle and use the clear solution. REFERENCES: J. M. Mathews: "The Textile Fibers"; Wochenblatt fur Papierfabrikation, November 30th, 1907; Worden: " Nitrocellulose Industry," Volume 1, page 560. 382 TECHNICAL METHODS OF ANALYSIS CHEMICAL TESTS OF ROPES AND TWINES General. The chemical tests usually desired on ropes and twines are: (1) Moisture. (2) Ether extract, to show the amount of oilor tar. (3) Water extract, to show whether the material has been treated with any chemicals, such as CaCk solution. (4) Ash. Sampling. It is very important to get a representative sample. On twine and small rope, samples should be obtained by cutting pieces 2 or 3 inches long from sufficient portions to represent the whole sample. If the twine is wound on a spindle or card, the abnormally dry outside layer should be discarded and samples taken from the interior. On large ropes and cables, the material should be unstranded and portions taken, not only from each strand but also representative of the outside and inside of the twists in each strand. In making the analysis it will save cal- culation if an even number of grams is weighed out. Ether Extract. Weigh out 10 grams and extract (preferably in a straight extractor) with ether for ten to sixteen hours. Dry the extract to constant weight at not over 100 C. Moisture. Dry the residue from the ether extract to constant weight at 100 C. and calculate the total percentage loss from the original weight. This loss will be moisture plus ether extract. Subtract from this the ether extract; the difference will be moisture. Water Extract. Place the dried residue from the ether extract in a beaker and heat to boiling with distilled water. Decant the water through a filter, catching the filtrate in a liter volumetric flask. Drain off as much water as possible and again heat to boiling with a fresh portion of distilled water. Repeat this 5-8 times, or until the filtrate is colorless, using about 100 cc. of water each time. Make up to the mark with distilled water and pipette out an aliquot of 200 cc. (representing 2 grams of the original). Evaporate to dryness on the steam bath in a weighed platinum dish and then dry to constant weight in the oven at 100 C. Cool in a desiccator and weigh. Report the result as Total Water-Soluble Matter. Ignite the above residue at not over dull red heat until all ANALYSIS OF TEXTILES AND TEXTILE FIBERS 383 carbon is burnt off. Cool in a desiccator and weigh. Report the result as Water-Soluble Mineral Matter. NOTES. (1) If merely the total water-soluble is desired, the residual fiber after extracting may be dried to constant weight at 100 C. and the loss to water reported as Water-Soluble Matter. (2) In case the ash fuses, let it cool, dissolve in hot water, filter through a quantitative filter, ignite the filter paper in the weighed platinum dish, then add the filtrate, evaporate to dryness, ignite gently, cool in a desiccator and weigh. Ash of Fiber. If the residual fiber from the water extract has been dried to constant weight, weigh out 2 grams quickly and ignite in a weighed platinum crucible to a white ash; cool in a desiccator and weigh. Calculate the ash to the original basis. If the residual fiber from the water extract has not been dried, drive off the bulk of water on the steam bath and ignite in a weighed platinum crucible, in small portions at a time, taking care to avoid drafts which would blow any of the light ash from the crucible. (For the same reason the fiber should not be allowed to take fire and burn but should be smoked down to a char before raising the heat.) In this case, in calculating the percentage, use the original weight of 10 grams. NOTE. The ash thus obtained is the natural ash of the fiber. The total ash would be the sum of this ash and the water-soluble mineral matter. Fiber. The fiber is obtained " by difference." Add together the percentages of ether extract, moisture, water soluble and ash of the fiber, and subtract the sum from 100%. DIFFERENTIATION OF ROPE AND CORDAGE FIBERS General. In this country ropes are generally made from the following fibers: (1) Manila, fine and coarse. (2) Sisal. (a) Mexican. This is the most common and cheapest. (b) East African, which is now out of the market. (c) Java. (d) Bahama. (3) New Zealand flax. This is not a true flax. 384 TECHNICAL METHODS OF ANALYSIS (4) Mauritius. (5) Tampico or Istle. (6) Flax, hemp and jute; used for twines but not much for rope. Abroad hemp is used mixed with other fibers. An experienced eye can distinguish between manila and sisal but not, generally speaking, between sisal from different sources. New Zealand flax may generally be distinguished under the microscope by the presence of fibrillse which give it a ragged ap- pearance. Istle or Tampico is distinct because the fiber is of a horny character and is nearly round like a horse hair. The length is usually under 2 feet, much shorter than other rope fibers. Distinction between Sisal and Manila. For distinguishing between sisal and manila, use the Swett color reaction * with bleach followed by NELiOH. This test was developed in this laboratory and is as follows : Submerge the suspended fibers in a solution of chloride of lime (bleaching powder) made distinctly acid with acetic acid. This solution should contain 3-6% of available chlorine. After the fibers have soaked for five rr.inutes, remove, rinse with water and immerse in NH^OH (1 : 1). The resulting color should be examined at once, as after five rrinutes it is likely to change. Manila gives an umber brown, hemp a faint pink, and most of the other fibers a cherry red. The umber brown color appears to be characteristic of manila. For applying this test to rope, it is advisable to remove the oil with ether previous to staining. Microscopic Examination. To prepare a slide from a rope or yarn for microscopic examination, take a strand and remove the oil with ether. Then boil with NaOH solution (about 5%) for two or three minutes, rinse, stain if desirable, and finally rinse again. While the fibers are still damp and relatively soft, trim the end with a razor or a sharp knife. Then, resting the bundle on a piece of wood, cut the end across. The bundle usually sticks together as a unit. The length of the sections need not be over 1 mm. Separate these cut ends on a slide, add a drop of 50% glycerol and press with the flat side of a knife blade to break up the fiber bundles. Finally cover with a cover glass and examine under the microscope, using 100-300 diameters magnification. *J. Ind. Eng. Chem. 10, 227 (1918). ANALYSIS OF TEXTILES AND TEXTILE FIBERS 385 If spiral vessels are seen, it means that the sample contains some fiber other than manila. ASBESTOS COTTON TWINE General. In a twine consisting of asbestos and cotton, a determination of the loss on ignition will give an approximation of the amount of cotton present, provided the loss on ignition of the asbestos itself can be ascertained. As this is not generally possible, however, and since the loss on ignition of asbestos from different sources varies widely, it is generally necessary to deter- mine the cotton by dissolving it out with Schweitzer's reagent. (See page 381.) The reagent should always be tested out with absorbent cotton to see that it works properly. The directions for making up should be followed exactly and the final solution should dissolve cotton completely in the cold. The solution readily decomposes and should be made up rapidly, and only in small quantities as needed. Cotton. Weigh out 1 gram of the twine * and treat in the cold with Schweitzer's reagent, using at least 100 cc. of the reagent per gram of the twine. Let stand for several hours, preferably over- night; dilute to several times its volume with distilled water and decant through linen cloth on a Buchner funnel. (The best linen for this purpose is that used in the determination of crude fiber.) Wash with warm water, and then with cold, to remove the copper clinging to the asbestos fibers. Remove the residue from the linen and again digest with Schweitzer's reagent, washing as before. Dry the fibers in a tared weighing bottle, cool in a des- iccator and weigh (with the bottle stoppered) . Repeat the process of digestion, washing and weighing until the twine ceases to lose weight upon further treatment with the reagent. The loss to Schweitzer's reagent is the cotton plus the free moisture, which is determined on a separate portion and sub- tracted., * Many asbestos cotton twines are reinforced with a brass or copper wire This wire should be removed before making the analysis and the results reported on the twine itself, unless otherwise instructed. 386 TECHNICAL METHODS OF ANALYSIS NOTE. If the fibers appear blue after washing with cold water, the last traces of copper may be removed by washing with an extremely dilute solu- tion of acetic acid followed by. a further washing with water before drying. Moisture. Determine the moisture on a separate sample by drying 1 gram to constant weight at 105 C. in a weighing bottle. As the material is hygroscopic, it must be weighed with the weighing bottle stoppered. Loss on Ignition. The dried sample from the moisture deter- mination may be used for the loss on ignition, which is conducted in the usual way in a platinum crucible. NOTES. (1) All results should be reported on the moisture-free basis. (2) This method is based on analyses carried out in this laboratory. CHAPTER X ANALYSIS OF FOODSTUFFS AMMONIA IN EGGS SET up the apparatus as shown in Fig. 24. If the eggs are odorless, weigh out about 22 grams. If the odor is rather strong, weigh out 10-12 grams. In weighing out the sample pour A B C FIG. 24. Apparatus for Determining Ammonia in Eggs. some of the thoroughly mixed eggs into a beaker and weigh the two. Then pour the desired amount from the beaker into the tall cylinder " B " and again weigh the beaker and contents. In the 250 cc. graduated flask " C " place about 60 cc. of dis- tilled water and add 1 cc. of 0.2 N HC1 or H 2 S0 4 . To the eggs in the cylinder add 5 cc. of a solution of equal parts of a 10% solution of K 2 CO3 and a 15% solution of K^C^. The solutions of carbonate and of oxalate should be made up separately and mixed just before using. 387 388 TECHNICAL METHODS OF ANALYSIS After adding the above solution to the eggs, pour in sufficient mineral oil (a heavy engine or machine oil is most satisfactory) to form a layer about 0.5-0.75 inch deep on the surface of the eggs. This is to prevent frothing. Connect up the apparatus as shown in the figure. The tube in the bottle " A " should dip below the surface of the H^SCU to remove moisture and any NHs. In the cylinder " B " the inlet tube should reach nearly to the bottom. Connect the tube at " E " to an air blast and blow a rapid current of air through the apparatus for at least two hours. There should be a " trap " bottle between " E " and the blast to prevent the blowing over of any dust or other foreign matter into the cone. H 2 SO 4 in " A." After air has been blown through the eggs for a sufficient length of time to carry over all the NHs, remove the flask " C " and make the solution up to 250 cc. Nesslerize an aliquot of this solution and compare it with the regular Nessler standards as used in the determination of NHs in sanitary water analysis (page 500). The aliquot taken for comparison will depend upon the amount of NHa present in the eggs. It should not develop a color more intense than that given by 5 cc. of the standard NH 4 C1 solution. The nesslerization is conducted as follows : The standard NILtCl solution used in sanitary water analysis contains 0.000012 gram of NHs per cc. It is made by dissolving 3.822 grams of c. P. NHiCl in 1000 cc. of ammonia-free water, and diluting 10 cc. of this solution to 1000 cc. with ammonia-free water. The Nessler solution is made by dissolving 61.75 grams of KI in 250 cc. of distilled H2O and adding a saturated solution of HgCl2 cautiously, till a slight permanent red precipitate appears. Dissolve this slight precipitate by adding 0.75 gram of powdered KI. Then add 150 grams of KOH dissolved in 250 cc. of H 2 O. Make up to 1 liter and settle overnight. Carry out the nesslerization in 50 cc. tall Nessler tubes. For the standards take 0.1, 0.3, 0.5, 0.7, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, and 5.0 cc., respectively, of the above standard NELiCl solu- tion, each in a separate tube. In another tube place the aliquot of the solution to be tested. Add 1 cc. of the Nessler solution to each tube and dilute to 50 cc. with distilled water. Mix thor- oughly, let stand ten minutes, and match up the sample with the ANALYSIS OF FOODSTUFFS 389 standard giving the same color. Calculate the amount of NHs in the original eggs in parts per 100,000. NOTES. (1) This method was furnished by H. C. Lythgoe, Chemist of the Mass. Board of Health, and has been used in our laboratory with suc- cess for several years. (2) The Nessler solution described on page 499 may be used in place of the one above described. (3) The amount of NH 3 in fresh eggs varies with the fat content. The ratio of NH 3 (expressed as parts per 100,000) to fat (expressed as percentage) should be about 1 I 2.5. In other words, for fresh eggs the NH 3 should not be over 4 parts when the fat is 10%, and not over 8 parts when the fat is 20%. (The fat may be determined by drying the egg material in a Hof- meister Schaelchen and extracting with anhydrous ether.) COLORING MATTER IN FOODS General. The following procedures are the tentative methods of the Association of Official Agricultural Chemists published in its Journal, Vol. II, Methods of Analysis (1916), page 155. Coloring matters may be divided into two classes: (1) Insoluble Pigments and (2) Soluble Dyes and their Lakes. (1) PIGMENTS The insoluble pigments, ultramarine, lampblack, etc., are most commonly used as facings and may be separated by washing the sample with water and letting the washings settle. The par- ticles of coloring matter can be identified by microscdpic exam- ination and by chemical tests of the residue or purified coloring matter.* Most of the common pigments other than lakes, such as the yellow, brown and red ochers and umbers, are derivatives of the heavy metals and contain Fe, Mn, etc. Others, such as vari- ous green and blue compounds, including the green chlorophyll derivatives, contain Cu. (2) SOLUBLE COLORING MATTERS AND THEIR LAKES (I) Coal-tar Dyes. (A) WOOL DYEING TEST. (a) Wines, fruit juices, distilled liquors, flavoring extracts, vinegars, beers, syrups, non-alcoholic beverages and similar products. Dilute * See also Schultz: " Farbenstofftabellen," 5th German Ed., 1911-14. 390 TECHNICAL METHODS OF ANALYSIS 20-200 cc. of the sample with 1-3 volumes of water and boil, or heat on the steam bath, with a small piece of white woolen cloth (nun's veiling). When the mixture contains much alcohol, heat until most of the alcohol has been removed; in other cases, take out the wool after five to fifteen minutes and rinse with water. Then treat the liquid with 3-4 drops of cone. HC1 for each 100 cc. and warm again for ten to twenty minutes with a clean piece of wool. The basic dyes go on the fiber best from neutral or faintly ammoni- acal solutions and, if present, will appear on the first piece of wool. Acid colors dye from neutral solutions, but more readily from those containing free acid. If the wool takes up any considerable amount of coloring matter in either case, the presence of coal-tar dyes is indicated. The lichen colors (Archil, Cudbear, Litmus) go readily on wool, however, and many other natural colors, such as Turmeric, will dye the fiber, if present in considerable amount. On the other hand, a few coal-tar dyes, especially Auramine O and Naphthol Green B, are quite unstable and, if present in small amounts, may give no distinct dyeing. Acid dyes are much more frequently used than basic dyes and in most cases may be removed from wool without much decom- position by " stripping " the latter with dil. NIUOH. By the action of the alkali many natural colors are destroyed, while others remain for the most part on the fiber. If the behavior with wool in neutral and acid solutions indicates the presence of acid dyes, rinse the colored cloth thoroughly with water, cover with 2% NH40H in a casserole, boil for a few minutes, remove the cloth and squeeze out the adhering liquid. Boil the ammoniacal solution to remove excess of NHs, drop in a piece of clean, wet wool, make distinctly but not strongly acid with HC1 and boil again. If acid coal-tar dyes are present, they will usually give a fairly clean, bright dyeing on the second piece of wool. A further puri- fication may be carried out by repeating the stripping and redye- ing, though generally accompanied by corresponding loss of dye. (b) Candies and similar colored sugar products. Dissolve about 20 grams of sample in 100 cc. of water and treat the solution as directed under (a). When the coloring matter is on the surface of the candy, pour off the solution before the colorless inner portion has dissolved. (c) Jams and jellies. Boil a mixture of 10-20 grams of the ANALYSIS OF FOODSTUFFS 391 sample and 100 cc. of water with wool in neutral and also in acid solution as directed under (a). For thick jams it is usually better, though less easy, to first extract the coloring substances by treating the product as directed below under (d). (d) Canned and preserved fruits and vegetables, sausage casings, smoked fish, coffee, spices, etc. Macerate 20-200 grams of sample with 4^5 times its weight of 80% alcohol. After standing a few hours pour off the solvent as completely as possible and repeat the extraction, using 70% alcohol containing about 1% of NH.3. (1) Examine separately the filtered alcoholic extracts as directed under (a); or, (2) boil the ammoniacal solution until practically neutral, complete the neutralization with acetic acid, add the neutral 80% alcohol extract, continue the evaporation until most of the alcohol is removed, and boil with wool as directed under (a). (e) Cocoa and chocolate products. Treat cocoa as directed under (d). The alcoholic extract will contain a large amount of natural coloring matter and several dyeings and strippings may be necessary to get rid of this in order to show the presence of coal-tar dyes. Chocolate may be treated similarly but the following procedure is preferable: Wash 20-200 grams of the well-divided sample with gasoline on a filter until most of the fat has been removed; if the gasoline is colored, reserve for the examination of oil-soluble dyes as directed below under Oil-Soluble Dyes. Remove most of the adherent solvent from the residue by evaporation or pressure between layers of absorbent paper and digest with alcohol as directed under (d). Coal-tar dyes may also be detected in chocolate and cocoa products by mixing directly with 3^ times their weight of hot water and immediately boiling the magma with wool, as directed under (a). Because of the presence of large amounts of fatty and protein materials, this method is not very satisfactory. (/) Cereal products. Proceed as directed under (d), in most cases working with a large amount of the sample, 200-300 grams, and a relatively smaller amount of alcohol. Where tests are to be made only for the acid dyes, the extraction with neutral 80% alcohol may be omitted advantageously. (II) Oil-Soluble Dyes. Prepare an alcoholic solution of the oil- soluble dye by one of the following methods which are to be 392 TECHNICAL METHODS OF ANALYSIS applied to the oil or fat obtained by extraction with ether or gaso- line if the nature of the substance requires it : (a) Shake the oil or melted fat with an equal volume of 90% alcohol. The alcohol after separation will contain Aniline Yellow, Butter Yellow, Aminoazotoluene and Auramine, if present. (6) Saponify 20-200 grams of the oil or fat with alcoholic KOH and, after removal of most of the alcohol on the steam bath, extract the soap with ether or gasoline. Most of the common dyes are removed by this treatment, though the digestion with strong alkali may cause some decomposition and make the extraction rather troublesome. (c) Dilute 20-200 grams of the oil or melted fat with 1-2 vol- umes of gasoline and shake out successively with 2-4% KOH or NaOH solution, 12-15% HC1, and phosphoric-sulfuric acid mix- ture, prepared by mixing 85% HsPC^ with about 10-20% by volume of cone. H^SC^. The dil. alkali extracts Sudan G and Annatto. The dil. HC1 extracts Aniline Yellow (7)*, Aminoazotoluene, and Butter Yellow (16) the first two forming orange-red, the latter a cherry- red solution in this solvent. Benzeneazobetanaphthylamine and homologues also come in this group, though they are not extracted very readily and decompose rapidly on standing in strongly acid solution. The HsPC^ mixture is necessary for the extraction of Sudan I (11), Sudan II (49), and Sudan III (143), and the homo- logue of the last, Sudan IV. The procedure is not very suitable in the presence of Auramine, but this dye is seldom found in oils. Neutralize the alkaline and dil. HC1 solutions; dilute the H 3 PO4 mixture and partially neutralize, cooling the liquid during this operation; and extract the dyes by shaking with ether or gasoline. For the direct dyeing test use the alcoholic solution obtained as directed in (a). Evaporate to dryness the ether or gasoline solutions, obtained as directed in (6) and (c) and dissolve the residue in 10-20 cc. of strong alcohol. Add some strands of white silk and a little water and evaporate on the steam bath until the alcohol has been removed or until the dye is taken up by the * The numbers following the names of the dyes in parenthesis are the numbers by which that dye is designated in " A Systematic Survey of the Organic Coloring Matters," 1904, by A. Green, translated from Schultz and Julius. ANALYSIS OF FOODSTUFFS 393 silk. The dyeing test is sometimes unsatisfactory and in all cases a small portion of the alcoholic solution should be tested by treat- ment with an equal volume of cone. HC1 and SnCl2 solution, respectively. The common oil-soluble coal-tar dyes are rendered more red or blue by the acid and are decolorized by the reducing agent. Most of the natural coloring matters become slightly paler with the acid and are little changed by the SnCl2 solution. NOTE. For the separation and identification of the permitted coal tar colors, see J. Assoc. Official Agr. Chemists, Methods of Analysis (1916); also U. S. Dept. of Agr., Bureau of Animal Industry, Circular 180 (July, 1911), and J. Ind. Eng. Chem. 8, 1123 (1916), CRUDE FIBER General. The official method for determination of crude fiber as directed by the U. S. Dept of Agriculture, Bur. of Chem., Bulletin 107, page 56, is a long and tedious process. It has been somewhat modified by M. O. Sweeney (Bulletin 137, page 157), so that the time consumed is considerably shortened. It has been claimed, however, that for certain kinds of feed-stuffs, especially those rich in protein, the Sweeney method is not entirely satis- factory. This has, therefore, been further modified by Cornelia Kennedy [J. Ind. Eng. Chem., 4, 600 (1912)]. The three methods are given below, but, except for special cases, the Kennedy modification is the one to be employed in routine work. Official Method.* Prepare solutions of H2SO4 and of NaOH. each of exactly 1.25% strength, as determined by titration. Extract a quantity of the sample representing about 2 grams of the dry material with ordinary ether (or, in case the oil in the sample has been determined, use the residue from that deter- mination). To the extracted residue in a 500 cc. flask add 200 cc. of boiling 1.25% H^SC^. Connect the flask with an inverted condenser, the tube of which passes only a short distance beyond the rubber stopper into the flask, or, if a tall conical flask is used, simply cover with a watch glass or short-stem funnel, boil at once and continue boiling gently for thirty minutes. A blast of air conducted into the flask will reduce the frothing of the liquor. Filter through linen, and wash with boiling water until the wash- * J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 118. 394 TECHNICAL METHODS OF ANALYSIS ings are no longer acid. Rinse the residue back into the flask with 200 cc. of a boiling 1.25% solution of NaOH, which should be free, or nearly so, from carbonate. Boil at once and continue boiling gently for thirty minutes, in the same manner as directed above for the treatment with acid. Filter at once rapidly. (A Gooch crucible will be found convenient.) Wash with boiling water until the washings are neutral. Dry at 110 C. until it ceases to lose weight. Weigh, incinerate completely and weigh again. The loss of weight is considered to be crude fiber. NOTE. If filtration proceeds very slowly through a Gooch crucible, the last filtration may be made through linen or a tared filter paper. If a linen filter is used, rinse the crude fiber, after washing is completed, into a flat-bottomed platinum dish by means of a jet of water; evaporate to dryness on a steam bath, dry to constant weight at 110 C., weigh, incinerate completely and weigh again. The loss in weight is considered to be crude fiber. If a tared filter paper is used, weigh in a weighing bottle. In any case the crude fiber after drying to constant weight at 110 C. must be inciner- ated and the amount of ash deducted from the original weight. Sweeney Modification. Place 2 grams of the oil-free material in a wide-mouth Erlenmeyer flask of liter size, inserting a small air condenser in the mouth of the flask to prevent concentration due to loss of steam. To the sample in the flask add 200 cc. of a boiling 1.25% solution of H2SO4, as in the Official Method. Heat to boiling and after gently boiling for thirty minutes treat as follows : Neutralize with a 10% solution of NaOH, using a few drops of phenolphthalein as indicator. Approximately 25 cc. of NaOH solution are required. Add at once 200 cc. of a boiling 2.656% solution of NaOH. This solution should be prepared as accurately as possible by titration. Continue the digestion at the boiling point for thirty minutes longer in the same manner as in the treatment with acid. Then filter the alkaline solution con- taining the fiber residue rapidly through a linen cloth and wash repeatedly with boiling water. Transfer the fiber residue to a weighed platinum Gooch crucible and wash with alcohol and finally with ether. Dry at 100 C. to constant weight. Ignite the dried residue and again weigh. The loss in weight gives the weight of crude fiber. Kennedy Modification. Digest 2 grams of the fat-free sample with 200 cc. of a 1.25% H2SO4 solution by boiling thirty minutes in ANALYSIS OF FOODSTUFFS 395 a liter Erlenmeyer flask connected to an air condenser exactly as in the Sweeney Modification. Then add directly 200 cc. of a 3.52% solution of NaOH, the solution being made of this exact strength by titration. Boil the whole for thirty minutes, filter through linen and wash free from alkali with hot water. Then wash thoroughly with boiling 1.25% H2SO4 which will remove any material precipitated by the addition of the alkali. Wash free from acid. Transfer from the linen filter to a weighed Gooch crucible, wash with alcohol, then with ether, and dry to constant weight at 100 C. Weigh the residue, ignite and reweigh. The loss indicates the amount of crude fiber. SULFUR DIOXIDE IN FOODSTUFFS General. Free sulfurous acid in the form of sulfur fumes is extensively employed to bleach molasses, disinfect wine casks and to bleach and preserve dried fruits. The process is known as " sulfuring." The sulfurous acid salts most commonly employed as preservatives are the bisulfites of Na and of Ca. The normal sulfites (Na, K or NHi) are most commonly used as preservatives in fruit juices, ketchup, fruit and vegetable pulp, wines, malt liquors and meat products. They are frequently mixed with other antiseptics, such as salicylates and benzoates. Determination. The same methods are used for the qualitative detection of 862 as for its quantitative determination, except that in the former case weighed quantities need not be employed. (A) DISTILLATION METHOD. This method is adapted to all food products whether solid or liquid. Place 50-200 grams of the material in a 500 cc. flask, add water, if necessary, and 5 cc. of a 20% solution of HaPC^, and distill in a current of CO2 into about 100 cc. of water containing a few drops of bromine, until 150 cc. have passed over. If sulfides are present, as is true of decomposed meat products and possibly other foods, the steam from the distilling flask before entering the condenser should be passed through a flask containing 40 cc. of a 2% neutral solution of CdCl2 or a 1% solution of CuSCX. These solutions effectually remove the H^S without retaining any appreciable amount of 862. To avoid escape of 862, the condenser tube should dip below the surface of the bromine solution. 396 TECHNICAL METHODS OF ANALYSIS The method and apparatus may be simplified without material loss in accuracy by omitting the current of CO 2 , adding 10 cc. of HsPC^ instead of 5 cc., and dropping into the distilling flask a piece of NaHCOs, weighing not more than a gram, immediately before attaching the condenser. When the distillation is finished, boil off the excess of Br, dilute to about 250 cc., add 1 cc. of cone. HC1, heat to boiling and add, drop by drop, while boiling, an excess of BaCl2 solution. Let stand overnight in a warm place, filter, wash with hot water, ignite at a dull red heat and weigh as BaS04. Calculate to 862 and report as parts per million. CALCULATION. BaSO 4 X 0.2744 = SO 2 . (B) DIRECT TITRATION METHOD. This method is applicable to Sauternes and other white wines and to beer, but should not be used for other materials unless found by experiment to yield accurate results. To 25 grams of the sample (finely divided in water, if solid or semi-solid) in a 200 cc. flask, add 25 cc. of a N solution of KOH; shake thoroughly and set aside for at least fifteen minutes, with occasional shaking; and then add 10 cc. of H2S04 (!: 3) with a little starch solution and titrate the mixture with 0.02 N iodine solution, introducing the latter quite rapidly, until a fixed blue color is produced. CALCULATION. 1 cc. 0.02 N iodine = 0.00064 gram of S0 2 . REFERENCE. Leach: " Food Inspection and Analysis" (1913 edition), page 840. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 150. REDUCING SUGARS AND SUCROSE General. This method describes general procedures to be used in determining sucrose and different kinds of reducing sugars in various materials. In employing these methods, however, the analyst should make sure that there are not special methods or precautions to be used with the particular substances under analysis. The methods described below are official methods of the Association of Official Agricultural Chemists, unless otherwise designated. ANALYSIS OF FOODSTUFFS 397 (I) SUCROSE There are two general classes of methods for the determination of sucrose: (1) Optical methods, and (2) Chemical methods. (1) Optical Methods. The rules of the International Commis- sion for Unifying Methods of Sugar Analysis have been adopted as a tentative method of the Assoc. Official Agr. Chemists for raw sugars as follows: (A) GENERAL DIRECTIONS FOR RAW SUGARS. " In general, all polarizations are to be made at 20 C. The verification of the saccharimeter must also be made at 20 C. For instruments using the Ventzke scale 26 grams of pure dry sucrose, weighed in air with brass weights, dissolved in 100 metric cc. at 20 C. and polarized in a room, the temperature of which is also 20 C., must give a saccharimeter reading of exactly 100.00. The tem- perature of the sugar solution during polarization must be kept constant at 20 C. " For countries where the mean temperature is higher than 20 C., saccharimeters may be adjusted at 30 C. or any other suitable temperature, under conditions specified above, provided that the sugar solution be made up to volume and polarized at this same temperature. " In effecting the polarization of substances containing sugar employ only half -shade instruments." The saccharimeter used may be either single or double wedge and should be a half -shadow instrument with either double or triple field. " During the observation keep the apparatus in a fixed position and so far removed from the source of light that the polarizing Nicol is not warmed. As sources of light employ lamps which give a strong illumination, such as triple gas burner with' metallic cylinder, lens and reflector; gas lamp with Auer (Welsbach) burner; electric lamp; petroleum duplex lamp; sodium light." Whenever there is any irregularity in the sources of light, such as that due to the convolutions of the filament in the case of an electric light or to the meshes of the gauze in the case of the Wels- bach light, place a thin ground-glass plate between the source of light and the polariscope, so as to render the illumination uniform. " Before and after each set of observations the chemist must satisfy himself of the correct adjustment of his saccharimeter by 398 TECHNICAL METHODS OF ANALYSIS means of standardized quartz plates. He must also previously satisfy himself of the accuracy of his weights, polarization flasks, observation tubes and cover-glasses. (Scratched cover-glasses must not be used.) Make several readings and take the mean thereof, but no one reading may be neglected." Such plates are standardized to read to the second decimal point and by their use a quick and at the same time accurate test can be made. In using such plates for testing saccharimeters, it is necessary that the instrument, as well as the plate, be at 20 C., before making a reading. Different points of the scale, preferably 20, 50, 80, and 100 (sugar scale) should be tested against the plates. " In making a polarization use the whole normal weight for 100 cc. or a multiple thereof for any corresponding volume. " As clarifying and decolorizing agents use either basic acetate of lead, alumina cream, or concentrated solution of alum. Bone black and decolorizing powders are to be excluded." Whenever reducing sugars are determined in the solution for polarizing, use only neutral lead acetate for clarification, as basic lead acetate causes precipitation of some of the reducing sugars. In addition to these clarifying agents neutral lead acetate and basic lead nitrate (Herles' solution) have been made official by the Association. " After bringing the solution exactly to the mark at the proper temperature, and after wiping out the neck of the flask with filter paper, pour all of the well-shaken clarified (sugar) solution on a rapidly acting filter. Reject the first portions of the filtrate, and use the rest, which must be perfectly clear, for polarization." It is advisable to reject the first 20 cc. that run through, then cover the funnel with a watch glass and use the remainder for polarization. In no case should the whole solution or any part be returned to the filter. If cloudy after the 20 cc. have been rejected, begin a new determination. " Whenever white light is used in polarimetric determinations, the same must be filtered through a solution of potassium bichro- mate of such a concentration that the percentage content of the solution multiplied by the length of the column of the solution in centimeters is equal to nine." This concentration must be doubled in reading carbohydrate materials of high rotation dispersion, such as commercial glucose, etc. (B) PREPARATION AND USE OF CLARIFYING REAGENTS (TEN- ANALYSIS OF FOODSTUFFS 399 TATIVE). (a) Basic lead acetate solution. Boil 430 grams of neu- tral lead acetate, 130 grams of litharge, and 1 liter of water for thirty minutes. Let the mixture cool and settle, and dilute the supernatant liquid to sp. gr. 1.25 with recently boiled water. Solid basic lead acetate may be substituted for the normal salt and litharge in the preparation of the solution. (6) Alumina cream. Prepare a cold saturated solution of alum (potassium aluminum sulfate) in water. Add NH40H with constant stirring until the solution is alkaline to litmus; let the precipitate settle and wash by decantation with water until the wash water gives only a slight test for sulfates with BaCk solution. Pour off the excess of water and store the residual cream in a stoppered bottle. (c) Dry basic lead acetate (Horne Method). This clarify- ing agent is obtained as a dry powdered salt and should con- tain 72.8% of Pb, which corresponds to a composition of 3Pb (C2H302)2 2PbO. Dissolve the normal or half-normal weight of the sugar solution in a flask with water and complete the vol- ume. Add a small quantity of the dry salt and shake, then add more and shake again, repeating until completely precipitated, but avoiding any excess. Of this salt 0.1346 gram is equivalent to 1 cc. of the basic lead acetate solution, described under (a). When molasses or any other substances producing a heavy pre- cipitate is being clarified, some dry, coarse sand should be added to break up the balls of basic lead acetate and the precipitate. (This method is to have equal weight with the use of a solution of basic lead acetate in clarifying cane, sorghum, and beet products.) (d) Neutral lead acetate. Prepare a saturated solution of neu- tral lead acetate and add it to the sugar solution before completing to volume. Its use is imperative when determining the reducing sugars in the solution used for polarization. (e) Basic lead nitrate (Herks' Solution).- (1) Dissolve 250 grams of Pb(NOs)2 in water and make up to 500 cc. (2) Dissolve 25 grams of NaOH in water and make up to 500 cc. Add equal amounts of (1) and (2) to the sugar solution, shake, and add more if complete precipitation has not occurred, but avoid any excess. Then complete the volume with water. When this solution is used for clarification, the factor in the Clerget determination becomes 143.5 instead of 142.66. 400 TECHNICAL METHODS OF ANALYSIS (C) DETERMINATION OF SUCROSE IN THE ABSENCE OF RAF- FINOSE BY POLARIZATION BEFORE AND AFTER INVERSION WITH HCL. * Dissolve the normal weight (26 grams) of the substance in water, add basic lead acetate carefully, avoiding any excess, then 1-2 cc. of alumina cream, shake and dilute to 100 cc., filter, rejecting the first 20 cc. of the filtrate, cover the filtrate with a watch glass and, when sufficient filtrate is collected, polarize in a 200 mm. tube. The reading so obtained is the direct reading (P of the formula given below) or polarization before inversion. For the invert reading, remove the Pb from the solution either (1) by adding anhydrous K 2 C204, a little at a time, to the remaining solution, avoiding an excess, and removing the precipitate by filtration; or (2) by adding anhydrous Na2CC>3 under the same conditions. Introduce 50 cc. of the lead-free filtrate into a 100 cc. flask (if Na2COa was used for removing the Pb, neutralize carefully the excess with a few drops of dil. HC1) and add 25 cc. of water. Then add, little by little, while rotating the flask, 5 cc. of cone. HCL Heat the flask, after mixing, in a water bath kept at 70 C. The temperature of the solution in the flask should reach 67- 69 C. in two and one-half to three minutes. Maintain a tem- perature of as nearly 69 C. as possible for seven to seven and one- half minutes, making the total time of heating ten minutes. Remove the flask, cool contents rapidly to 20 C., and dilute to 100 cc. Polarize this solution in a tube provided with a lateral branch and water jacket, keeping at 20 C. This reading must be .doubled to obtain the invert reading. If necessary to work at a temperature other than 20 C., which is allowable within narrow limits, the volumes must be completed and both direct and invert polarizations must be made at exactly the same temperature. The inversion may also be accomplished as follows: (1) To 50 cc. of the clarified solution, freed from Pb, add 5 cc. of cone. HC1 and set aside for twenty-four hours at a temperature not below 20 C.; or, (2) if the temperature be above 25 C. set aside for ten hours. Make up to 100 cc. at 20 C. and polarize as directed above. Calculate the sucrose by one of the following formulas (Clerget): * In the presence of much levulose, as in honeys and fruit products, the optical method for sucrose gives too high a result. ANALYSIS OF FOODSTUFFS 401 (1) For substances in which the invert solution contains more than 12 grams of invert sugar per 100 cc. : The following formula is to be used when substances like raw 100 (P-7) sugars are polarized : S = , 142.66-- in which $ = per cent of sucrose; P = direct reading of normal solution ; 7 = invert reading of normal solution; and T = temperature at which readings are made. (2) For substances in which the concentration of the invert solution is less than 12 grams per 100 cc. : . The following formula, which takes into account the concen- tration of the sugar in solution, should be used in all other cases: 100 (P-7) T f T V 142.66--- 0.0065 142.66--- (P-7) in which S = per cent of sucrose, P = direct reading of normal solution, 7 = invert reading of normal solution, and T = temperature. (D) DETERMINATION OF SUCROSE AND RAFFINOSE (of value chiefly in analysis of beet products). If the direct reading is more than 1 higher than the per cent of sucrose as calculated by the formula given above, raffinose is probably present. Calculate sucrose and raffinose by the following formula of Herzf eld : 0.5124 P-7 P-S -0839- * = in which P = direct reading of normal solution; 7 = in vert reading of normal solution; /S = per cent of sucrose; and R = per cent of anhydrous raffinose. 402 TECHNICAL METHODS OF ANALYSIS The above formula assumes that polarizations are made at exactly 20 C. If the temperature (T) is other than 20 C., the following formula should be used : P (0.4724 +0.002 T)-/ o = Having calculated S, then R = 0.899-0.003 T P-S 1.852' (2) Chemical Methods. (A) DETERMINATION OF SUCROSE FROM REDUCING SUGARS BEFORE AND AFTER INVERSION (TENTA- TIVE). Determine the reducing sugars (clarification having been effected with neutral lead acetate, never with basic lead acetate), as directed below under Munson and Walker Method and calculate to invert sugar from the M. and W. tables. Then invert the solution as directed above (under Polarization before and after Inversion with HC1), exactly neutralize the acid and again determine the reducing sugars; but calculate them to invert sugar from the same table as just referred to, using the invert sugar column alone. Deduct the per cent of invert sugar determined before inversion from that obtained after inversion, and multiply the difference by 0.95; the result is the per cent of sucrose. The solution should be diluted in both determinations so that not more than 245 mg. of invert sugar are present in the amount taken for reduction. It is also important that all Pb be removed from the solution with K 2 C 2 O4 before reduction. (II) TOTAL REDUCING SUGARS General. The determination of reducing sugars by copper reduction depends upon the reduction of an alkaline copper solution by the action of the reducing sugars, precipitating red Cu2O. Since the extent of the reduction varies under different conditions, it is necessary that the directions be strictly adhered to. Of the common sugars, sucrose is the only one which has no direct reducing action on alkaline copper tartrate; but on under- going inversion, it is converted into reducing sugars which can be readily determined. ANALYSIS OF FOODSTUFFS 403 There are various methods, all based on the reduction of Cu salts, but varying in detail. For all ordinary use the Munson and Walker method is the most convenient, since the Cu2O can be calculated directly to dextrose, invert sugar, maltose or lactose, as the case may be. It is not well adapted, however, to the deter- mination of lactose (milk sugar) in the presence of sucrose, for instance in sweetened condensed milk or in sweetened milk choco- late, on account of partial inversion of the sucrose by the boiling. In such case it is necessary to use the Defren-O'Sullivan method. (1) Munson and Walker Method (Tentative). (A) REAGENTS. (a) Fehling's copper sulfate solution.* Dissolve 34.639 grams of carefully selected crystals of pure CuSO4-5H2O in distilled water. Dilute to exactly 500 cc. and filter through prepared asbestos. (b) Fehling's alkaline tartrate solution.* Dissolve 173 grams of Rochelle salts (KNaC^Oo 4H 2 O) and 50 grams of NaOH in water, dilute to exactly 500 cc., let stand two days, and filter through prepared asbestos. (c) Asbestos. Prepare the asbestos (which should be of the amphibole variety) by first digesting with HC1 (1:3) for two or three days. Wash free from acid and digest for a similar period with 10% NaOH solution. Then treat for a few hours with hot alkaline tartrate solution of the strength employed in the sugar determinations. (Old alkaline tartrate solutions that have stood for some time may be used.) Then wash free from alkali. Finally digest with dil. HNOs for several hours, and, after washing free from acid, shake with water for use. In preparing the Gooch crucible, load it with a film of asbestos 0.25 inch thick. Wash this thoroughly with water to remove fine particles of asbestos. Finally wash with alcohol and then with ether. Dry for thirty minutes at 100 C., cool in a desiccator and weigh. It is best to dissolve the Cu20 with HNOs each time after weighing and use the same felts repeatedly, as they improve with use. (B) PROCEDURE. Transfer 25 cc. each of the copper and the alkaline tartrate solutions to a 400 cc. beaker of alkali-resisting glass, and add 50 cc. of the reducing sugar solution, which must be neutral or slightly alkaline; or, if a smaller volume of sugar solu- tion be used, add water to make the final volume 100 cc. Heat * Soxhlet modification. 404 TECHNICAL METHODS OF ANALYSIS the beaker on an asbestos gauze over a Bunsen burner; so regulate the flame that boiling begins in four minutes, and continue boiling for exactly two minutes.* Keep the beaker covered with a watch glass throughout the heating. Without diluting, filter the Cu2O at once on the asbestos felt in a porcelain Gooch crucible, using suction. Wash the Cu20 thoroughly with water at about 60 C., then with 10 cc. of alcohol, and finally with 10 cc. of ether. Dry for thirty minutes in the water oven at 100 C., cool in desiccator and weigh as Cu2O. Calculate from the weight of the Cu 2 O (see note (3) below) the amount of reducing sugars according to Munson and Walker's tables. These tables may be found in Leach: " Food Inspection and Analysis," 4th Ed., page 623; also in J. Assoc. Official Agr. Chemists, 2, Methods of Analysis (1916), pages 88-96. NOTES. (1) The number of milligrams of Cu reduced by a given amount of reducing sugar differs when sucrose is present and when it is absent. In the tables, the absence of sucrose is assumed, except in the columns under invert sugar, where the total amount of sugar in 50 cc. of solution is given. (2) Blank determination: Always conduct a blank determination, using 50 cc. of reagent and 50 cc. of water; and if the weight of the Cu2O obtained exceeds 0.0005 gram, correct the result of the reducing sugar determination accordingly. The alkaline tartrate solution deteriorates on standing and the amount of Cu 2 O obtained in the blank increases. (3) The method of direct weighing of the Cu2O should be used only for determinations in pure sugar solutions; in all other cases the Cu of the Cu 2 O should be determined by one of the following methods, since the Cu2O is very apt to be contaminated with foreign matter. (C) DETERMINATION OF COPPER REDUCED. (a) Electrolysis from #2$04 solution (tentative). Filter the Cu2O in a Gooch crucible and wash the beaker and precipitate thoroughly with hot water without transferring the precipitate to the filter. Wash the asbestos film and adhering Cu20 into the beaker by means of hot dil. HNOs. After all the Cu is in solu- tion, refilter through a thin mat of asbestos in a Gooch crucible and wash thoroughly with hot water. Add 10 cc. of H2SO4 (1:4) and evaporate the filtrate on the steam bath until the Cu salt has * It is important that these directions be strictly observed. In order to regulate the burner for this purpose, it is advisable to make preliminary tests, using 50 cc. of the reagent and 50 cc. of water, before proceeding with the actual determination. ANALYSIS OF FOODSTUFFS 405 largely crystallized. Heat carefully on a hot plate or over asbestos until the evolution of white fumes shows that the excess of HNOs is removed. Add 8-10 drops of cone. HNOs and rinse into a 100-125 cc. platinum dish. Deposit the Cu upon the dish by electrolysis. Wash thoroughly with water, then break the cur- rent, wash with alcohol and ether successively, dry at about 50 C., and weigh. If preferred, the electrolysis may be conducted in a beaker, the Cu being deposited upon a weighed platinum electrode. (b) Electrolysis from H 2 S04 and HNOs solution (tentative). Filter and wash as directed above. Transfer the asbestos mat from the crucible to the beaker by means of a glass rod and rinse the crucible with about 30 cc. of a boiling mixture of dil. sulfuric and nitric acids, containing 65 cc. of cone. H2SO4 and 50 cc. of cone. HNOs per liter. Heat and agitate until solution is com- plete; filter and electrolyze as above. (c) Electrolysis from HNOs solution (tentative). Filter and wash as directed above. Transfer the asbestos mat and adhering Cu2O to the beaker. Dissolve the oxide still remaining in the crucible by means of 2 cc. of cone. HNOs, adding it with a pipette and collecting the solution in the beaker containing the asbestos. Rinse the crucible with a jet of water, letting the rinsings flow into the beaker. Heat the contents of the beaker until all Cu is in solution; filter, wash, dilute the filtrate to 100 cc. or more, and electrolyze. When a nitrate solution is electrolyzed, the first washings of the deposit should be made with water acidulated with H2SO4 in order to remove all HNOs before the current is interrupted. (2) Defren-O'Sullivan Method.* (A) REAGENTS. The solu- tions used in this procedure are the same as those for the Munson and Walker method above described. The asbestos used should be of the long fiber variety and should be specially prepared as follows: Boil first with HNO 3 (about 1:6). Wash out the acid with hot water and then boil with a 25% solution of NaOH, and finally wash out the alkali with hot water. Keep the asbestos in a wide-mouth flask or bottle and transfer it to the Gooch crucible by shaking up in the water and pouring it quickly into the crucible while under suction. * This method is described in Leach: " Food Inspection and Analysis," 4th Ed., page 618. 406 TECHNICAL METHODS OF ANALYSIS (B) PROCEDURE. Mix 15 cc. of Fehling's Cu solution with 15 cc. of the tartrate solution in a 250 cc. Erlenmeyer flask, and add 50 cc. of distilled water. Place the flask and contents in a boiling water bath and let remain five minutes. Then run rapidly from a burette into the hot liquor in the flask 25 cc. of the sugar solution to be tested (which should contain not over 0.5% of reducing sugars). Let the flask remain in the boiling water bath exactly fifteen minutes after the addition of the sugar solution; remove, and with the aid of suction filter the contents rapidly in a Gooch crucible containing a layer of prepared asbestos fiber about 1 cm. thick, the Gooch crucible with the asbestos having been previously ignited, cooled and weighed. Wash the Cu2O precipitate thor- oughly with boiling distilled water till the filtrate is no longer alkaline. Dry the Gooch crucible with contents in the oven, and finally heat to dull redness for fifteen minutes, during which the red Cu2O is converted into black CuO. If a platinum Gooch crucible is used (which is preferable) it may be heated directly over the low flame of a burner. If the Gooch crucible is of porcelain, con- siderable care must be taken to avoid cracking, the heat being increased cautiously and the operation preferably conducted in a radiator or muffle. After oxidation as above, transfer the crucible to a desiccator, cool, and weigh quickly. From the milligrams of CuO, calculate the milligrams of dextrose, etc., as the case may be, and calculate to per cent of the original sample. (See table in Leach, page 619.) (Ill) INDIVIDUAL REDUCING SUGARS (A) Invert Sugar. The methods for total reducing sugars pre- viously given, also apply to the determination of invert sugar.* The following method is the tentative approximate volumetric method for rapid work of the Association of Official Agricultural Chemists. (1) REAGENT: Soxhlet's modification of Fehling's solution. Prepare by mixing, immediately before use, equal volumes of reagents (a) and (6) described previously under the Munson and Walker Method. * See also under (D) below (page 408) . ANALYSIS OF FOODSTUFFS 407 (2) STANDARDIZATION OF COPPER SOLUTION. Since the factor for calculation varies with the minute details of manipulation, every operator must determine a factor for himself, using a known solution of the pure sugar that he desires to determine, and keeping the conditions the same as those used for the determination. Standardize the solution for invert sugar in the following man- ner: Dissolve 4.75 grams of pure sucrose in 75 cc. of water, add 5 cc. of cone. HC1 and invert with HC1 as described above under Polarization (Sucrose in the Absence of Raffinose). Neutralize the acid with NaOH solution and dilute to 1 liter. Ten cc. of this solution contain 0.050 gram of invert sugar, which should reduce 10 cc. of the reagent. The strength of the copper solution should never be taken as constant, but should be checked against the sugar. (3) DETERMINATION. Place 10 cc. of the reagent in a large test-tube and add 10 cc. of water. Heat to boiling, and add grad- ually small portions of the solution of the material to be tested until the copper has been completely reduced, boiling after each addition to complete the reaction. Two minutes' boiling is required for complete reduction when the full amount of sugar solution has been added in one portion. When the end is nearly reached and the amount of sugar solution to be added can no longer be judged by the color of the solution, remove a small portion of the liquid and filter rapidly into a small porcelain crucible or on a test plate; acidify with dilute acetic acid, and test for Cu with dilute potassium ferrocyanide solution. The sugar solution should be of such strength as will give a burette reading of 15-20 cc. and the number of successive additions should be as small as possible. (B) Dextrose: Allihn's Gravimetric Method (Tentative). (1) REAGENT: Allihn's modification of Fehling's solution. Prepare by mixing, 'immediately before use, equal volumes of (a) and (6). (a) Copper sulfate solution: Dissolve 34.639 grams of CuS0 4 -5H 2 O in water and dilute to 500 cc. (6) Alkaline tartrate solution: Dissolve 173 grams qf Rochelle salts and 125 grams of KOH in water and dilute to 500 cc. (2) DETERMINATION. Place 30 cc. of the CuSCU solution, 30 cc. of the alkaline tartrate solution and 60 cc. of water in a beaker and 408 TECHNICAL METHODS OF ANALYSIS heat to boiling. Add 25 cc.* of the solution of the material to be examined, prepared so as not to contain more than 0.25 gram of dextrose, and boil for exactly two minutes, keeping the beaker covered. Filter immediately through asbestos, and obtain the weight of Cu by one of the procedures given under the Munson and Walker method above. From the Allihn tables determine the cor- responding weight of dextrose. The Allihn tables are given in J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), pages 107-108; also in Leach: " Food Inspection and Analysis," 4th Ed., page 633. NOTE. This method may also be used for various other reducing sugars as described below. (C) Maltose, Lactose, Dextrose. Use the Munson and Walker general method as described above and calculate from the table the weight of maltose, lactose or dextrose equivalent to the weight of the Cu reduced. (D) Levulose, Invert Sugar, Arabinose, Xylose, Galactose. For the determination of these reducing sugars use the Allihn method as described above and multiply the weight of dextrose found in the Allihn tables by the following factors: Levulose 1.093 Invert sugar 1 . 046 Arabinose 0.969 Xylose 1.017 Galactose 1.114 REFERENCES. U. S. Dept. of Agriculture, Bureau of Chemistry, Bulletin 107, page 241; Leach: "Food Inspection and Analysis"; J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), pages 81-109. SACCHARINE PRODUCTS General. This method applies to the usual determinations common to all Saccharine Products, i.e., those containing one or * If necessary, the volume of the sugar solution added may be greater or less than 25 cc., provided the volume of water added is correspondingly varied from 60 cc., so that the total volume of sugar solution and water added is always 85 cc. ANALYSIS OF FOODSTUFFS 409 more sugars. For complete analysis of special products see the method for that particular material, e.g., Sugar, Honey, Maple Products, Sucrose in Beets, etc. The methods here given are, unless otherwise specified, the official methods of the Association of Official Agricultural Chemists. Preparation of Sample (Tentative). (a) Liquids (molasses, syrups, etc.) Mix materials of this class thoroughly. If crystals of sugar are present, dissolve them either by heating gently or by weighing the whole mass, then adding water, heating until com- pletely dissolved and, after cooling, re-weighing. Calculate all results to the weight of the original substance. (b) Semi-solids (jellies, jams, etc.). Weigh 50 grams of the sample into a 250 cc. graduated flask. Treat with water, fill to the mark and mix thoroughly. If insoluble material remains, mix uniformly by shaking before taking aliquots for the various determinations. (c) Solids (sugar, confectionery, etc.). Grind and mix thor- oughly materials of this class to secure uniform samples. Moisture. (A) SUGAKS. Dry 2-5 grams in a flat dish (nickel, platinum, or aluminum) at not over 110 C. for ten hours; cool in a desiccator and weigh; then dry again for an hour, or until there is only a slight change in weight. (B) MASSECUITES, MOLASSES AND OTHER LIQUID AND SEMI- LIQUID PRODUCTS (1) Drying upon pumice stone (tentative). Prepare pumice stone of 2 grades of fineness, one of which will pass through a 1 mm. sieve, the other through a 6 mm. sieve. Make the determination in flat metallic dishes or in shallow, flat- bottomed, weighing bottles. Place a layer of the fine pumice stone, 3 mm. in thickness, on the bottom of the dish, then a layer of the coarse pumice stone 6-10 mm. in thickness, dry and weigh. Dilute the sample with a weighed portion of water so that the diluted material shall contain 20-30% of solid matter. Weigh into the dish, prepared as described above, an amount of the diluted sample to yield approximately 1 gram of dry matter. If this weighing cannot be made rapidly, use a weighing bottle pro- vided with a cork through which a pipette passes. Dry in vacuo a.t 70 C. to constant weight, making trial weighings at intervals of two hours. For substances containing little or no levulose or 410 TECHNICAL METHODS OF ANALYSIS other readily decomposable substance, the drying may be made in a water oven at the temperature of boiling water. (2) Drying upon quartz sand (tentative). Digest pure quartz sand with cone. HC1, wash, dry and ignite. . Preserve in a stoppered bottle. Place 6-7 grams of the prepared sand and a short stirring rod in a flat-bottomed dish. Dry thoroughly, cool in a desiccator, and weigh. Then add 3^ grams of the molasses, mix with the sand (if necessary to thoroughly incorporate the two, add a little water), dry in a water oven at the temperature of boiling water for eight to ten hours, stirring at intervals of an hour; cool in a desiccator, and weigh. Stir, heat again for an hour, cool, and weigh. Repeat the heating and weighing until the loss of weight in an hour is not greater than 0.003 gram. Specific Gravity, Water, and Total Solids. (A) BY MEANS OF A SPINDLE. The density of juices, syrups, etc., is most conveniently determined by means of the Brix hydrometer. For rough work, or where less accuracy is desired, the Baum^ hydrometer may be used. The Brix spindle should be graduated to tenths. The range of degrees recorded by each individual spindle should be as limited as possible. The solution should be as nearly as prac- ticable of the same temperature as the air at the time of reading, and, if the variation from the temperature of the graduation of the spindle amounts to more than 1, a correction must be applied according to Table II. Before taking the density of a juice, let it stand in the cylinder until all air bubbles have escaped, and until all fatty or waxy matter has come to the surface and been skimmed off. The cylinder should be large enough in diameter to allow the hydrometer to come to rest without touching the sides. A 20 table of sp. gr. at - and per cent by weight of sucrose is given on page 125 of the Journal of the Assoc. Official Agri. Chemists, Vol. 2, Methods of Analysis (1916); and a table for the compar- 17 5 ison of sp. gr. at with degrees Brix (per cent by weight 17.5 of sucrose) and degrees Baume is given below in Table I. If the sample is too dense to determine the gravity directly, dilute a weighed portion with a weighed quantity of water, or dis- solve a weighed portion and dilute to a known volume with water. ANALYSIS OF FOODSTUFFS 411 TABLE I 17 5 Comparison of specific gravities at ' C., degrees Brix and Baum6. l / .o Degrees Baume = 146 78-^ . sp. gr. Degree Degree Degree Brix or Brix or Brix or Per Cent byWt. Specific Gravity Degree Baume Per Cent byWt. Specific Gravity Degree Baume Per Cent byWt Specific Gravity Degree Baum6 of Su- of Su- of Su- crose crose crose 1.0 1.00388 0.6 33.0 1 . 14423 18.5 65.0 1.31989 35. b 2.0 1.00779 1.1 34.0 1.14915 19.05 66.0 1.32601 36.1 3.0 1.01173 1.7 35.0 1.15411 19.6 67.0 1.33217 36.6 4.0 1.01570 2.3 36.0 1.15911 20.1 68.0 1.33836 37.1 5.0 1.01970 2.8 37.0 1.16413 20.7 69.0 1.34460 37.6 6.0 1.02373 3.4 38.0 1.16920 21.2 70.0 1.35088 38.1 7.0 1.02779 4.0 39.0 1.17430 21.8 71.0 1.35720 38.6 8.0 1.03187 4.5 40.0 1.17943 22.3 72.0 1.36355 39.1 9.0 1.03599 5.1 41.0 1.18460 22.9 73.0 1.36995 39.6 10.0 1.04014 5.7 42.0 1.18981 23.4 74.0 1.37639 40.1 11.0 1.04431 6.2 43.0 1.19505 23.95 75.0 1.38287 40.6 12.0 1.04852 6.8 44.0 1.20033 24.5 76.0 1.38939 41.1 13.0 1.05276 7.4 45.0 1.20565 25.0 77.0 1.39595 41.6 14.0 1.05703 7.9 46.0 1.21100 25.6 78.0 1.40254 42.1 15.0 1.06133 8.5 47.0 1.21639 26.1 79.0 1.40918 42.6 16.0 1.06566 9.0 48.0 1.22182 26.6 80.0 1.41586 43.1 17.0 1.07002 9.6 49.0 1.22728 27.2 81.0 1.42258 43.6 18.0 1.07441 10.1 50.0 1.23278 27.7 82.0 1.42934 44.1 19.0 1.07884 10.7 51.0 1.23832 28.2 83.0 1.43614 44.6 20.0 1.08329 11.3 52.0 1.24390 28.8 84.0 1.44298 45.1 21.0 1.08778 11.8 53.0 1.24951 29.3 85.0 1.44986 45.5 22.0 1.09231 12.4 54.0 1.25517 29.8 86.0 1.45678 46.0 23.0 1.09686 13.0 55.0 1.26086 30.4 87.0 1.46374 46.5 24.0 1.10145 13.5 56.0 1.26658 30.9 88.0 1.47074 47.0 25.0 1.10607 14.1 57.0 1.27235 31.4 89.0 1.47778 47.45 26.0 1.11072 14.6 58.0 1.27816 31.9 90.0 1.48486 47.9 27.0 1.11541 15.2 59.0 1.28400 32.5 91.0 1.49199 48.5 28.0 1.12013 15.7 60.0 .28989 33.0 92.0 1.49915 48.9 29.0 1 . 12488 16.3 61,0 .29581 33.5 93.0 1.50635 49.4 30.0 1 . 12967 16.8 62.0 .30177 34.0 94.0 1.51359 49.8 31.0 1.13449 17.4 63.0 .30777 34.5 95.0 1.52087 50.3 32.0 1.13934 17.95 64.0 .31381 35.1 412 TECHNICAL METHODS OF ANALYSIS In the first instance the per cent of total solids is calculated by the following formula : WS Per cent of solids in undiluted material = , w in which S = per cent of solids in diluted material ; W = weight of diluted material ; and w = weight of sample taken for dilution. When the dilution is made to a definite volume, the following formula is to be used : Per cent of solids in the undiluted material = , W in which V = volume of diluted solution at a given temperature ; D = sp. gr. of diluted solution at same temperature ; S = per cent of solids in diluted solution at same tempera- ture; and W = weight of sample taken for dilution at same tempera- ture. If the spindls reading be made at any other temperature than 17.5 C., the result should be corrected according to Table II. (B) BY MEANS OF A PYCNOMETER. 20 (1) Specific gravity at C. 4 20 Determine the sp. gr. of the solution at C. by means of a pycnometer and ascertain the corresponding per cent by weight of sucrose from official tables. When the density of the substance is too high for a direct determination, dilute and calculate the sucrose content of the original material as directed above. 17 5 3 (2) Specific gravity at ~ C. 17 .o 17 5 Determine the sp. gr. at - ' C. with a pycnometer and 17.5 ascertain the corresponding per cent by weight of sucrose from Table I. NOTE. Pycnometer determinations must not be made at any other temperature than those given. ANALYSIS OF FOODSTUFFS 413 TABLE II Corrections of Brix spindle readings for temperatures other than standard (17.5 C.) Tempera- ture C. 5 10 15 20 25 30 35 40 50 60 70 75 0.17 0.30 0.41 0.52 0.62 0.72 82 0.92 0.98 1.11 1.22 1.25 1.29 5 0.23 0.30 0.37 0.44 0.52 0.59 0.65 0.72 0.75 0.80 0.88 0.91 0.94 10 0.20 0.26 0.29 0.33 0.36 0.39 0.42 0.45 0.48 0.50 0.54 0.58 0.61 11 0.18 23 0.26 0.28 0.31 0.34 0.36 0.39 0.41 0.43 0.47 0.50 0.53 12 0.16 0.20 0.22 0.24 0.26 0.29 0.31 0.33 0.34 0.36 0.40 0.42 0.46 13 0.14 0.18 0.19 0.21 0.22 0.24 0.26 0.27 0.28 0.29 0.33 0.35 0.39 14 0.12 0.15 0.16 0.17 0.18 0.19 0.21 0.22 0.22 0.23 0.26 0.28 0.32 15 0.09 0.11 0.12 0.14 0.14 0.15 0.16 0.17 0.16 0.17 0.19 0.21 0.25 16 0.06 0.07 0.08 0.09 0.10 0.10 0.11 0.12 0.12 0.12 0.14 0.16 0.18 17 0.02 0:02 0.03 0.03 0.03 0.04 0.04 0.04 0.04 0.05 0.05 0.05 0.06 18 0.02 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 03 0.03 0.02 19 0.06 0.08 0.08 0.09 0.09 0.10 0.10 0.10 0.10 0.10 0.10 0.08 0.06 20 0.11 0.14 0.15 0.17 0.17 0.18 0.18 0.18 0.19 0.19 0.18 0.15 0.11 21 0.16 0.20 0.22 0.24 0.24 0.25 0.25 0.25 0.26 0.26 0.25 0.22 0.18 22 0.21 0.26 0.29 0.31 0.31 0.32 0.32 0.32 0.33 0.34 0.32 0.29 0.25 23 0.27 0.32 0.35 0.37 0.38 0.39 0.39 0.39 0.40 0.42 0.39 0.36 0.33 24 0.32 0.38 0.41 0.43 0.44 0.46 46 0.47 0.47 0.50 0.46 0.43 0.40 25 0.37 0.44 0.47 0.49 0.51 0.53 0.54 0.55 0.55 0.58 54 0.51 0.48 26 0.43 0.50 0.54 0.56 0.58 0.60 0.61 0.62 0.62 0.66 0.62 0.58 55 27 0.49 0.57 0.61 0.63 0.65 0.68 0.68 0.69 0.70 0.74 0.70 65 0.62 28 0.56 0.64 0.68 0.70 0.72 0.76 0.76 0.78 0.78 0.82 0.78 0.72 0.70 29 0.63 0.71 0.75 0.78 0.79 0.84 0.84 0.86 0.86 0.90 0.86 0.80 0.78 30 0.70 0.78 0.82 0.87 0.87 0.92 0.92 0.94 0.94 0.98 0.94 0.88 0.86 35 1.10 1.17 1.22 1.24 1.30 1.32 1.33 1.35 1.36 1.39 1.34 1.27 1.25 40 1.50 1.61 1.67 1.71 1.73 1.79 1.79 1.80 1.82 1.83 1.78 1.69 1.65 50 2.65 2.71 2.74 2.78 2.80 2.80 2.80 2.80 2.79 2.70 2.56 2.51 60 3.87 3.88 3.88 3.88 3.88 3.88 3.88 3.90 3.82 3.70 3.43 3.41 70 5.17 5.18 5.20 5.14 5.13 5.10 5.08 5.06 4.90 4.72 4.47 4.35 80 6.62 6.59 6'. 54 6.46 6.38 6.30 6.26 6.06 5.82 5.50 5.33 90 8.26 8.16 8.06 7.97 7.83 7.71 7.58 7.30 6.96 6.58 6.37 100 10.01 9.87 9.72 9.56 9.39 9.21 9.03 8.64 8.22 7.76 7.42 NOTE. For temperatures below 17.5 C. subtract the correction and for temperatures above, add it. Ash. METHOD 1. Heat 5-10 grams of the sample in a 50-100 cc. platinum dish at 100 C. until water is expelled, add a few drops of pure olive oil, and heat slowly over a flame until 414 TECHNICAL METHODS OF ANALYSIS swelling ceases. Then place the dish in a muffle and heat at low redness until a white ash is obtained. METHOD 2. Carbonize the mass at a low heat, dissolve the soluble salts in hot water, burn the residual mass as directed above, add the solution of soluble salts, evaporate to dry ness at 100 C., ignite gently, cool in a desiccator, and weigh. METHOD 3. Saturate the sample with H2SO4, dry, ignite gently, then burn in a muffle at low redness. Deduct one-tenth of the weight of the ash, and calculate the percentage. Soluble and Insoluble Ash (Tentative). Ash the material as directed above under Method 1 or Method 2, add water to the ash in the dish, heat nearly to boiling, filter, and wash with hot water until the combined filtrate and washings are about 60 cc. Return the filter and contents to the dish, ignite carefully, and weigh. Calculate the percentages of water-soluble and water- insoluble ash. Alkalinity of Soluble Ash (Tentative). Cool the filtrate from the above and titrate with 0. 1 N HC1 and methyl orange. Express the alkalinity as the number of cc. of 0.1 N acid per gram of sample. Alkalinity of Insoluble Ash (Tentative). Add an excess of 0.1 N HC1 (usually 10-15 cc.) to the ignited insoluble ash in the platinum dish, heat to boiling over an asbestos plate, cool and titrate the excess of HC1 with 0.1 N NaOH and methyl orange. Express alkalinity as above. Mineral Adulterants in the Ash (Tentative). Mix 100 grams of molasses, syrup, honey, or the confectionery solution (prepared as directed above under Preparation of Sample (6)), and evaporate to a syrupy consistency with about 35 grams of cone. H2SO4 in a large porcelain evaporating dish. Pass an electric current through it while stirring by placing one platinum elec- trode in the bottom of the dish near one side and attaching the other to the lower end of the glass rod with which the contents are stirred. Begin with a current of about 1 ampere and gradually increase to 4. In ten to fifteen minutes, the mass is reduced to a fine dry char, which may be readily burnt to a white ash in the original dish over a free flame or in a muffle. This method is preferred to the ordinary method of heating with H2S04, especially in the case of molasses, because, if properly ANALYSIS OF FOODSTUFFS 415 manipulated, the material comes quietly into the form of a very finely divided char or powder, particularly adapted for subsequent quick ignition. If an electric current is not available, treat in a large porcelain dish 100 grams of the saccharine solution, evaporate to a syrupy consistency with sufficient cone. H2&O4 to thoroughly carbonize the mass, and ignite in the usual manner. The following adulterants may be present: Salts of tin, used in molasses to bleach it; mineral pigments, such as chromate of lead in yellow confectionery; oxide of iron, sometimes used to simulate the color of chocolate; and copper. These elements may be detected by the usual qualitative tests. Nitrogen (Tentative). Determine the nitrogen in 5 grams of the material by the Kjeldahl, Gunning, or Kjeldahl-Gunning- Arnold Method as directed on pages 64-66, using a larger quantity of H2S04, if necessary, for complete digestion. Sucrose. METHOD 1 (TENTATIVE). (Substances in which the volume of the combined insoluble matter and precipitate from clarifying agents is less than 1 cc. from 26 grams.) Determine sucrose by polarization before and after inversion with HC1 as directed on page 400, under Determination of Sucrose in the Absence of Raffinose. NOTE. All products which contain dextrose or other reducing sugars in the crystalline form, or in supersaturated solution, exhibit the phenomenon of birotation. The constant rotation only should be employed in the Clerget formula, and to obtain this the solutions prepared for direct polarization should be allowed to stand overnight before making the reading. If it is desired to make the direct reading immediately, the birotation may be destroyed by heating the neutral solution to boiling for a few minutes or by adding a few drops of cone. NH 4 OH before completing the volume. METHOD 2. DOUBLE DILUTION METHOD (TENTATIVE.) (Substances in which the volume of the combined insoluble matter and precipitate from clarifying agents is more than 1 cc. from 26 grams.) Weigh out a half normal weight (13 grams) of the sample and make up the solution to 100 cc., employing the appropriate clarifier (basic lead acetate for dark-colored confectionery or molasses, and alumina cream for light-colored confectionery). Also weigh out the normal weight (26 grams) of the sample and 416 TECHNICAL METHODS OF ANALYSIS make up a second solution with the clarifier to 100 cc. Filter and obtain direct polariscopic readings of both solutions. Invert each solution with HC1 and obtain its invert reading. The true direct polarization of the sample is the product of the two direct readings, divided by their difference. The true invert polarization is the product of the two invert readings divided by their difference. Calculate the sucrose from the true polariza- tions thus obtained, by the formula given on page 401 under Determination of Sucrose in the Absence of Raffinose. Commercial Glucose (Approximate). METHOD 1 (TEN- TATIVE). (Substances containing little or no invert sugar.) Commercial glucose cannot be determined accurately, owing to the varying amounts of dextrin, maltose, and dextrose present. In syrups, however, in which the amount 'of invert sugar is so small as not to appreciably affect the result, commercial glucose may be estimated approximately from the polarization by the following formula: (o-S) 100 in which G= per cent of commercial glucose; a = direct polarization ; and S = per cent of cane sugar. Express results in terms of commercial glucose polarizing -hi 75 V. METHOD 2 (TENTATIVE). (Substances containing invert sugar.) Prepare an inverted half -normal weight solution of the sub- stance as directed on page 400 under Determination of Sucrose in the Absence ' of Raffinose by Polarization before and after Inversion with HC1, except that after inversion, cool the solution, make neutral to phenolphthalein with NaOH solution, slightly acidify with HC1, and treat with 5-10 cc. of alumina cream before making up to the mark. Filter and polarize at 87 C. in a 200 mm. jacketed tube. Multiply the reading by 200 and divide by the factor 163 to express the amount of glucose present in terms of glucose polarizing +175 V. Reducing Sugars. Determine reducing sugars, either as dex- trose or invert sugar by the Soxhlet method or. Munson and Walker method as described on pages 406 and 403. ANALYSIS OF FOODSTUFFS 417 Starch (Tentative). Measure 25 cc. of a solution or uniform mixture, prepared as directed previously under Preparation of Sample (6), which represents 5 grams of sample, into a 300 cc. beaker; or, introduce 5 grams of finely ground sample (previously extracted with ether if it contains much fat) into the beaker; add sufficient water to make the volume 100 cc., heat to about 60 C. (avoiding, if possible, gelatinizing the starch) and let stand for about an hour, stirring frequently to secure complete solution of the sugars. Transfer to a stout wide-mouthed bottle, rinse the beaker with a little warm water; cool, add an- equal volume of 95% alcohol, mix, and let stand at least an hour. Centrifuge until the precipitate is closely packed on the bottom of the bottle and decant the supernatant liquid through a hardened filter. Wash the precipitate with successive 50 cc. portions of 50% alcohol by centrifuging and decanting through the filter until 3 or 4 drops of the washings give no test for sugar with cc-naphthol when treated as described below. Transfer the residue' from the bottle and the hardened filter to a large flask and determine starch by the modified Sachsse method, described on page 442. a-Naphthol Test for Sucrose. Introduce into a test-tube a few drops of the liquid ; add 4 or 5 drops of 20% alcoholic a-naphthol solution and 2 cc. of water; shake well, tip the tube and let 2-5 cc. of colorless cone. H 2 S04 flow down the side. Then hold the tube upright and, if sucrose is present, a color varying from a faint to a deep violet will occur at the junction of the two liquids. On shaking the whole solution becomes a blue-violet color. Ether Extract in Confectionery. (A) CONTINUOUS EXTRACTION (TENTATIVE). (1) Measure 25 cc. of a 20% mixture or solu- tion, prepared as directed under Preparation of Sample (b), above, in-to a very thin, readily frangible, glass evaporating shell (Hofmeister Schalchen), containing 5-7 grams of freshly ignited asbestos fiber; or (2) if impossible to obtain a uniform sample, weigh 5 grams of the mixed finely divided sample into a dish, and wash with water upon the asbestos in the evaporating shell, using, if necessary, a small portion of the asbestos fiber on a stirring rod to transfer the last traces of the sample from the dish to the shell. Dry to constant weight at 100 C., cool, wrap 418 TECHNICAL METHODS OF ANALYSIS loosely in smooth paper, crush into rather small fragments between the fingers, transfer carefully the crushed mass, exclusive of the paper, to an extraction tube or a fat-extraction thimble. A thin lead disk (bottle cap) may be substituted for the Schalchen. The disk may then be cut into small pieces and placed in the extraction tube. Extract with anhydrous ether or petroleum ether (b. p. 45-60 C. and without weighable residue) in a con- tinuous extraction apparatus for at least twenty-five hours. In most cases it is advisable to remove the substance, from the extractor after the first twelve hours, grind with sand to a fine powder and re-extract for the remaining thirteen hours. Trans- fer the extract to a tared flask, evaporate the solvent and dry to constant weight in an oven at 100 C. (B) ROESE-GOTTLIEB METHOD (TENTATIVE). Substances such as butter-scotch, invariably yield extremely inaccurate results by the above method. In such cases introduce 4 grams of the material, or an amount of a uniform solution equivalent to this amount of the dry substance, into a Rohrig tube or similar appa- ratus, make up to a volume of 10 cc. with water, add 1.25 cc. of cone. NH40H and mix thoroughly. Add 10 cc. of 95% alcohol and mix. Then add 25 cc. of washed ether and shake vigorously for half a minute; then add 25 cc. of petroleum ether (b. p. below 60 C.), and shake again for half a minute. Let stand twenty minutes, or until separation between the liquids is complete. Draw off as much, as possible of the ether-fat solution (usually 0.5-0.8 cc. will be left) into a weighed flask through a small, rapid filter. (The flask should be weighed with a similar one as a counterpoise.) Again extract the liquid remaining in the tube, this time with 15 cc. each of ether and petroleum ether, shake vigorously half a minute with each, and let settle. Proceed as above washing the tip of the spigot and the filter with a few cc. of a mixture of equal parts of the 2 ethers (previously mixed and free from deposited water). For absolutely exact results the extraction must be repeated. This third extraction usually yields not more than about 1 mg. of fat, if the previous ether-fat solutions have been drawn off closely, or an amount averaging about 0.02% on a 4-gram charge. Evaporate the ether slowly on a steam bath, then dry the fat in a boiling water oven to con t ant weight. Test the purity of the fat by dissolving in a little petroleum ANALYSIS OF FOODSTUFFS 419 ether. Should a residue remain, wash the fat out completely with petroleum ether, dry the residue, weigh and deduct the weight. Paraffin in Confectionery (Tentative). Add to the ether extract in the flask, as above obtained, 10 cc. of 95% alcohol and 2 cc. of NaOH solution (1 : 1); connect the flask with a reflux condenser, and heat for one hour on the water bath, or until saponification is complete. Remove the condenser and let the flask remain on the bath until alcohol is evaporated and the residue is dry. Dissolve the residue as completely as possible in about 40 cc. of water and heat on the bath, shaking frequently. Wash into a separatory funnel, cool, and extract with four suc- cessive portions of petroleum ether, which are collected in a tared flask or capsule. Evaporate the petroleum ether and dry in the oven to constant weight. Any phytosterol or cholesterol present in the fat would be extracted with the paraffin. The amount is so insignificant that it may be disregarded generally. The character of the final residue should, however, be confirmed by determining its melting point, sp. gr., and refractive index. Alcohol in Syrups used in Confectionery (" Brandy Drops ") (Tentative). Collect in a beaker the syrup from a sufficient number of pieces to yield 30-50 grams of syrup. Strain into a tared beaker and weigh. Introduce the syrup into a 250-300 cc. distilling flask, dilute with half its volume of water, attach the flask to a vertical condenser and distill almost 50 cc. or as much of the liquid as possible without causing charring. Foaming may be prevented by adding a little tannin or a piece of paraffin about the size of a pea to the contents of distillation flask. Cool the distillate, make up to volume with water, mix well, determine the sp. gr. of the liquid very carefully with a pycnometer at standard temperature and obtain the corresponding weight of alcohol in the 50 cc. of distillate from standard alcohol tables.* Calculate the per cent by weight of alcohol in the candy filling. Coloring Matter. Proceed as under Coloring Matter in Foods, page 389. * These tables may be found in J. Assoc. Official Agr. Chemists, Methods of Analysis, (1916), pages 194-207; also Leach: " Food Inspection and Analysis," 3d Edition, pages 661-674; and Van Nostrand's " Chemical Annual." * 420 TECHNICAL METHODS OF ANALYSIS REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), pages 121-132. HONEY General. The following methods, unless otherwise indicated, are the tentative methods of the Association of Official Agricul- tural Chemists.* Preparation of Sample. (a) LIQUID OR STRAINED HONEY. If the sample is free from granulation, mix thoroughly by stirring or shaking before drawing the weighed portions for analysis. If the honey is granulated, place the container, with the stopper loose, in a water bath, and heat at not over 50 C. until the sugar crystals dissolve; mix thoroughly, cool, and weigh out portions for analysis. If sediment is present, su'ch as particles of comb, wax, sticks, bees, etc., heat the sample to 40 C. in a water bath and filter through cheesecloth before weighing. (6) COMB HONEY. Cut across the top of the comb, if sealed, and separate completely from the comb by straining through a 40- mesh sieve. If portions of the comb or wax pass through the sieve, heat the sample as in (a) and strain through a cloth. If the honey is granulated in the comb, heat until the wax is liquefied, stir, cool, remove the wax and take the clear liquid for analysis. Moisture. Weigh 2 grams into a tared, flat-bottomed, plat- inum, or aluminum dish of about 60 mm. diameter and containing 10-15 grams of fine quartz sand (which has been previously washed, dried and ignited) and a small glass stirring rod; add 5-10 cc. of water and thoroughly incorporate it with the sand and honey mixture by means of the rod; dry the dish and contents to constant weight in a vacuum oven at not over 70 C. Ash (Official). Weigh 5-10 grams into a weighed platinum dish, add a few drops of pure olive oil to prevent spattering and heat carefully till swelling ceases and then ignite to a white ash at not above dull redness. Soluble and Insoluble Ash. Add water to the ash in the plat- inum dish, heat nearly to boiling, filter through an ashless filter paper, and wash with hot water until the combined filtrate and *See J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 133. ANALYSIS OF FOODSTUFFS 421 washings measure about 60 cc. Return filter paper and contents to the dish, ignite carefully, and weigh. Calculate the percentages of water-soluble and water-insoluble ash. Alkalinity of Soluble Ash. Cool the filtrate from the above determination and titrate with 0.1 N HC1 and methyl orange. Express alkalinity in terms of cc. of 0.1 N acid per gram of the sample. Polarization. I. DIRECT POLARIZATION. (a) Immediate direct polarization. Transfer 26 grams of the honey to a 100 cc. volu- metric flask with water, add 5 cc. of alumina cream, dilute to the mark with water at 20 C., filter and polarize immediately in a 200 mm. tube. (6) Constant direct polarization. Pour the solution from the tube used in reading (a) back into the flask, stopper, and let stand twenty-four hours. Then again polarize the solution at 20 C. in a 200 mm. tube. (c) Birotation. The difference between (a) and (6) gives the birotation. (d) Direct polarization at 87 C. Polarize the solution obtained in (b) at 87 C. in a jacketed 200 mm. tube. II. INVERT POLARIZATION. (a) At 20 C. Invert with HC1 50 cc. of the solution (obtained under Direct Polarization above) as described on page 400 under the determination of Sucrose in the Absence of Raffinose, and polarize at 20 C. in a 200 mm. tube. (6) At 87 C. Polarize the above solution also at 87 C. in a 200 mm. jacketed tube. Reducing Sugars. Dilute 10 cc. of the solution used for direct polarization to 250 cc. Pipette out 25 cc. and determine the reducing sugars by the Munson and Walker or the Allihn method as described on pages 403 and 407 and calculate the result to per cent of invert sugar. Sucrose. Take 10 cc. of the solution obtained for invert polarization, dilute with a small amount of water, neutralize with Na 2 COs, and make up to 250 cc. with water. Pipette 50 cc. of this solution and determine the total sugars as reducing sugars by the Munson and Walker method. Calculate to invert sugar. Deduct the per cent, of invert sugar obtained previously and mul- tiply the difference by 0.95. The result is the per cent of sucrose. 422 TECHNICAL METHODS OF ANALYSIS Levulose. Multiply the direct polarization reading at '87 C. by 1.0315 and subtract the product from the constant direct polarization at 20 C.; divide the difference by 2.3919 to obtain the grams of levulose in the normal weight (26 grams) of honey. From this figure calculate the per cent in the original sample. Dextrose. Subtract the per cent of levulose above determined from the per cent of invert sugar found under Reducing Sugars, to obtain the approximate per cent of dextrose. The dextrose can be determined more accurately by multi- plying the per cent of levulose by the factor 0.915, which gives its dextrose equivalent in copper reducing power. Subtract this figure from that of the reducing sugars calculated as dextrose, to obtain the per cent of dextrose in the sample. NOTE. Owing to the difference in the reducing powers of different sugars, the sum of the dextrose thus found and the levulose as calculated above will be greater than the total amount of invert sugar obtained under Reducing Dextrin (Approximate). Transfer 8 grams of sample (4 grams in the case of dark-colored honey-dew honey) to a 100 cc. volumetric flask (using not more than 4 cc. of water) by letting the sample drain from the weighing dish into the flask and then dis- solving the residue in 2 cc. of water. After adding this solution to the contents of the flask, rinse the weighing dish with two 1 cc. portions of water to which a little alcohol is added subsequently. Fill the flask to the mark with absolute alcohol, shaking con- stantly. Set the flask aside until the dextrin has collected on the sides and bottom and the liquid is clear. Decant the clear liquid through a filter paper and wash the residue in the flask with 10 cc. of 95% alcohol, pouring the washings through the same filter. Dissolve the dextrin in the flask with boiling water and filter through the filter paper already used, receiving the filtrate in a tared dish, prepared as follows : Digest pure quartz sand with strong HC1, wash, dry and ignite. Place 6-7 grams of the pre- pared sand and a short stirring rod in the dish. Dry thoroughly, cool in a desiccator and weigh. Rinse the flask and wash the filter several times with small portions of hot water, add the washings to the tared dish, and evaporate the whole on a water bath. Dry to constant weight in vacuo at 70 C. ANALYSIS OF FOODSTUFFS 423 After determining the weight of the alcoholic precipitate, dis- solve the latter in water and make up to volume, using 50 cc. of water for each 0.5 gram of precipitate or part thereof. Determine the reducing sugars in the solution, both before and after inversion, by the Munson and Walker method, expressing results as invert sugar. Calculate the sucrose from the results thus obtained and subtract the sum of the reducing sugars before inversion and of the sucrose from the weight of the total alcoholic precipitate to obtain the weight of the dextrin. Free Acid. Dissolve 10 grams of honey in water and titrate with 0.1 N NaOH and phenolphthalein. Express results in terms of cc. of 0.1 N NaOH required to neutralize 100 grams of the sample. Glucose. QUALITATIVE TEST. Dilute the honey with an equal volume of water, then add a few cc. of iodine solution (1 gram of iodine, 3 grams of KI, 50 cc. of water). In the presence of glucose the solution turns red or violet, the depth and character of the color depending upon the quality and nature of the glucose em- ployed. A blank test with a pure honey of about the same color should be made in order to secure an accurate color comparison. Should the honey be dark and the percentage of glucose very small, precipitate the dextrin which may be present by adding several volumes of 95% alcohol. Let stand until the precipitate settles (do not filter), decant the liquid, dissolve the residue of dextrin in hot water, cool and apply the above test to this solution. A negative result is not proof of the absence of glucose, as some glucose, especially of high conversion, does not give any reaction with iodine. QUANTITATIVE TEST. An approximate determination can be made by Browne's formula as follows: Multiply the difference in the polarizations of the invert solution at 20 C. and at 87 C. by 77 and divide this product by the per cent of invert sugar after inversion found in the sample. Multiply the quotient by 100 and divide the product by 26.7, to obtain the per cent of honey in the sample; 100 per cent minus per cent of honey gives per cent of glucose. Commercial Invert Sugar. (A) BRYAN'S MODIFICATION OF FIBRE'S QUALITATIVE TEST. Dissolve 1 gram of resorcinol in 100 cc. of cone. HC1. Introduce 10 cc. of 50% honey solution into 424 TECHNICAL METHODS OF ANALYSIS a test tube and add 5 cc. of ether. Shake gently and let stand for some time until the ether layer is clear. Transfer 2 cc. of this clear ether solution to a small test tube and add a large drop of the resorcinol solution. Shake and note the color immediately. In the presence of artificial invert sugar, the resorcinol assumes immediately an orange-red color, turning to dark red. (B) FEDER ANILINE CHLORIDE TEST. To 100 cc. of c. P. aniline add 30 cc. of 25% HC1. Introduce 5 grams of the honey into a porcelain dish and add 2.5 cc. of the aniline reagent. A bright red color indicates the presence of commercial invert sugar. Diastase. Mix 1 part of honey with 2 parts of sterile water. Treat 10 cc. of this solution with 1 cc. of 1% soluble starch solu- tion and digest at 45 C. for an hour. At the end of this time test the mixture with 1 cc. of iodine solution (1 gram of iodine, 2 grams of KI, 300 cc. of water). Treat another 10 cc. portion of the honey solution, mixed with 1 cc. of the soluble starch solution, without heating to 45 C., with the reagent and compare the colors produced. If the original honey had not been heated sufficiently to kill the diastase, an olive-green or brown coloration will be produced in the mixture that has been heated at 45 C. Heated or artificial honey becomes blue. MAPLE PRODUCTS General. The procedures given herewith (unless otherwise indicated) are the tentative methods of the Association of Official Agricultural Chemists.* Preparation of Sample. (A) MAPLE SYRUP. Determine the moisture as given below. If it is less than 35% and there is some mineral sediment, pour the clear syrup into a beaker, washing the sediment also into the beaker with water. Then concentrate the syrup by boiling to a moisture content of about 35% (b. p. 104 C.). Set aside until cool, or preferably let the covered material stand overnight, and pour off the clear liquid for the analytical work. Where no sediment is present, the sample * See J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 136. ANALYSIS OF FOODSTUFFS 425 is ready for analysis after careful mixing. Where sugar has crystallized out, warm to dissolve the sugar before starting the analysis. It is desirable, in order to compare results upon dif- ferent samples, to reduce all results other than moisture to a dry substance basis as determined in the clear syrup. (B) MAPLE SUGAR, MAPLE CREAM, ETC. Determine the moisture as described below in the sample in its original condition by thoroughly mixing, if semi-plastic, or by rubbing up in a mortar representative portions of the product, if solid. For all other analytical determinations use a solution prepared as follows: Weigh roughly 100 grams of the product into a beaker and dissolve by boiling with 200 cc. of water. Decant the resulting syrup while hot through a muslin filter, concentrate by boiling to a moisture content of 35% (b. p. 104 C.); cool, or preferably let the covered material stand overnight; set aside until clear, and use this clear syrup for analysis. It is desirable, in order to com- pare results upon different samples, that all results except moisture be expressed upon the dry basis. Moisture. Weigh 2 grams of the sample into a tared flat- bottom platinum dish having a diameter of about 60 mm. and con- taining 10-15 grams of fine quartz sand which has been pre- viously washed, dried and ignited, and a small glass stirring rod. Add 5-10 cc. of water and incorporate thoroughly with the sand and syrup or sugar mixture by means of the rod. Dry the dish and contents to constant weight in a vacuum oven at a tem- perature not exceeding 70 C. Polarization. (A) DIRECT AT 20 C.; (B) INVERT AT 20 C. Make these determinations as described under the determination of Sucrose in the Absence of Raffinose by Polarization before and after Inversion with HC1, page 400. (C) INVERT AT 87 C. Proceed as described under Com- mercial Glucose, Method 2, on page 416. Reducing Sugars as Invert Sugar. (A) BEFORE INVERSION. Proceed as directed on page 403, using the Munson and Walker method. Employ an aliquot of the solution used for the direct polarization (as above) and use only neutral lead acetate for clar- ification. (B) AFTER INVERSION. Proceed as above, using an aliquot of the solution used for the invert polarization. 426 TECHNICAL METHODS OF ANALYSIS Sucrose. (A) BY POLARIZATION. Proceed as described on page 400 under the determination of Sucrose in the Absence of Raffinose by Polarization before and after Inversion with HC1. (B) REDUCING SUGARS BEFORE AND AFTER INVERSION. Follow the Munson and Walker procedure on page 403, clarifying with neutral lead acetate. Total Ash. Follow the procedure on page 413. Soluble and Insoluble Ash. See page 414. Alkalinity of Soluble and Insoluble Ash. See page 414. Lead Number (Winton). (A) REAGENT: Standard Basic Lead Acetate Solution. Boil 430 grams of normal lead acetate and 130 grams of PbO (or 560 grams of Home's dry basic lead acetate) for thirty minutes, with 1 liter of water, cool, let settle and dilute the supernatant liquid to sp. gr. 1.25. To a measured amount of this solution add four volumes of water and filter if not perfectly clear. The solution should be standardized each time a set of determinations is made. If the directions for preparing the basic lead acetate are not carried out carefully, the use of Home's dry basic lead acetate is preferable. (B) DETERMINATION OF LEAD IN THE BLANK. Transfer 25 cc. of the standard basic lead acetate to a 100 cc. flask, add a few drops of acetic acid and make up to the mark with water. Shake and determine Pb as PbS04 in 10 cc. of the solution, as directed below. The use of the acid is imperative in this case to keep the lead in solution when diluted with water. (C) DETERMINATION. Transfer 25 grams of the sample to a 100 cc. flask by means of water. Add 25 cc. of the standard basic lead acetate and shake; fill to the mark, shake and let stand for at least three hours before filtering. Pipette 10 cc. of the clear filtrate into a 250 cc. beaker, add 40 cc. of water and 1 cc. of cone. H2SO4, shake and add 100 cc. of 95% alcohol. Let stand over- night, filter on a tared Gooch crucible, wash with 95% alcohol, dry in a water oven, and ignite in a muffle or over a Tirrill burner. Apply the heat gradually at first, and if a flame is used, set the Gooch crucible inside of a platinum crucible. Cool and weigh. Subtract the weight of PbSC>4 so found from the weight of PbSO 4 found in the blank above and multiply by the factor 27.325. The use of this factor gives the lead number directly without the vari- ous calculations otherwise required. ANALYSIS OF FOODSTUFFS 427 Malic Acid Value (Cowles Method). Weigh 6.7 grams of the sample into a 200 cc. beaker, add 5 cc. of water, then 2 cc. of a 10% calcium acetate solution and stir. Add gradually, and with constant stirring, 100 cc. of 95% alcohol, and agitate the solution until the precipitate settles; or let stand until the supernatant liquid is clear. Filter off the precipitate and wash with 75 cc. of 85% alcohol. Dry the filter paper and ignite in a platinum dish. Add 10 cc. of 0.1 N HC1 and warm gently until all the lime dissolves. Cool and titrate back with 0.1 N NaOH and methyl orange. The difference in cc. divided by 10 represents the malic acid value of the sample. Previous to use the reagents should be tested by a blank determination and any necessary corrections applied. BUTTER AND BUTTER SUBSTITUTES (BUTTERINE, OLEOMARGARINE, ETC.) General. Butter fat is the fat of milk or butter. It has a peculiar and complex composition, the characteristic constituent being the radicle of butyric acid. Butter consists of a mixture of about 80-90% of butter fat with variable proportions of water, curd and salt. Coloring matter is often added and Na2COs is sometimes employed to prevent rancidity. In the United States the legal title for all butter substitutes is Oleomargarine (in England and several other countries it is Margarine). In preparing these substitutes various fats are used, including lard and tallow. The proper consistency is often obtained by adding sesame, peanut or cottonseed oil. The fat is then usually incor- porated with milk and salt, and colored. Sometimes more or less real butter is added; and palm oil and purified cocoanut oil have also been used. Preparation of Sample. If large quantities of butter are to be sampled, use butter trier or sampler. Melt completely the por- tions thus drawn, 100-150 grams, in a closed vessel at as low a temperature as possible and, when melted, shake the whole violently for some minutes, cooling at the same time, until the mass is homogeneous and sufficiently solidified to prevent the separation of the water and fat. Then pour a portion into the vessel from which it is to be weighed for analysis. It should 428 TECHNICAL METHODS OF ANALYSIS nearly or quite fill the vessel and should be kept in a cool place until analyzed. Moisture. Place 1.5-2.5 grams in a dish with a flat bottom having a surface of at least 20 sq. cm., dry at the temperature of boiling water and weigh hourly until it ceases to lose weight. The use of clean dry sand or asbestos is admissible. . Fat (Ether Extract). (a) INDIRECT METHOD. Dissolve in the dish with absolute ether or petroleum ether the dry butter obtained in the moisture determination (in which no absorbent was used), transfer to a weighed Gooch crucible with the aid of a wash bottle filled with the solvent and wash until free from fat. Dry the crucible and contents at the temperature of boiling water until the weight is constant and calculate the fat by difference. (6) DIRECT METHOD. From the dry butter obtained in deter- mining the moisture, either wither without the use of an absorbent, extract the fat with anhydrous alcohol-free ether, receiving the solution in a weighed flask. v Evaporate the ether, dry the extract at the temperature of boiling water, and weigh hourly until it ceases to lose weight. Casein, Ash, and Chlorine. Cover the crucible containing the residue from the fat determination by the Indirect Method (a) above and heat, gently at first, gradually raising the temper- ature to just below redness. The cover may then be removed and the heat continued until the contents of the crucible are white. The loss in weight represents casein, and the residue in the crucible mineral matter. In this mineral matter, dissolved in water slightly acidulated with HNOs, determine chlorine either gravi- metrically or volumetrically. Salt. Weigh in a counterpoised beaker 5-10 grams of the material, using portions of about 1 gram from different parts of the sample. Add about 20 cc. of hot water and, after it is melted, transfer the whole to a separatory funnel. Insert the stopper and shake for a few minutes. Let stand until the fat has all collected on top of the water, then draw off the latter into a flask, being careful to let none of the fat globules escape. Again add hot water to the beaker and repeat the extraction 10-15 times, using each time 10-20 cc. of water. The washings will contain all but a mere trace of the NaCl originally present in the butter. Determine its amount in the whole or an aliquot of the ANALYSIS OF FOODSTUFFS 429 liquid by the volumetric AgNOs method with K^CrCU as indi- cator. CALCULATION. 1 cc. 0.1 N AgNO 3 = 0.00585 gram Nad. Examination of Fat. (a) PREPARATION OF SAMPLE. Melt the butter and keep it in a dry place at about 60 C. for two or three hours, until the water and curd have entirely separated. Filter the clear supernatant fat through a dry filter paper in a hot- water funnel or in an oven at about 60 C. Should the filtered liquid fat not be perfectly clear, it must be refiltered. (6) REICHERT-MEISSL NUMBER. Determine the Reichert- Meissl number as described under Animal and Vegetable Oils, page 243. (c) SPECIFIC GRAVITY. Determine with a pycnometer the sp. gr. at 40 C. compared with water at the same temperature. NOTES. (1) The average Reichert-Meissl number of pure butter fat is 26. To calculate the approximate amount of butter fat in a butter sub- stitute, divide the Reichert-Meissl number of the separated fat by 26 and multiply by the percentage of total fat in the butter substitute. (2) U. S. Standard for Butter: It shall contain not less than 82.5% of butter fat. U. S. Standard for Butter Fat: It shall have a Reichert-Meissl number 40 C of not less than 24 and the sp. gr. at 4QO c shall be not less than 0.905. Renovated Butter and Oleomargarine. (a) FOAM TEST (SPOON TEST) Heat 2-3 grams of the sample, either in a spoon or a dish, over a free flame. True butter foams abundantly, whereas process butter will bump and sputter without foaming. Oleo- margarine behaves like process butter, but the chemical tests will determine whether the sample is oleo or butter. (6) MELTED FAT TEST. Melt 50-100 grams of the sample at 50 C. The curd from butter will settle, leaving a clear super- natant fat; with renovated butter, the supernatant fat remains more or less turbid. NOTES. (1) For the examination of butter for artificial coloring, see Leach: " Food Inspection and Analysis." (2) The above methods are those of the Association of Official Agricultural Chemists. 430 TECHNICAL METHODS OF ANALYSIS COCOA, CHOCOLATE AND CACAO PRODUCTS General. Plain or bitter chocolate (chocolate liquor) accord- ing to the Federal Standard (Bureau of Chemistry, Circular 19) is the solid or plastic mass obtained by grinding cacao nibs without the removal of fat or other constituents, except the germs; and contains not over 3% of ash insoluble in water, 3.5% of crude fiber, 9% of starch, and not less than 45% of cocoa fat.* By extraction of part of the fat cocoa is obtained. The fat in cocoa generally runs from 18 to 24%. Sweetened chocolate is made by adding sucrose (and some- times cocoa butter, spices, etc.) to bitter chocolate, and according to Bureau of Chemistry, Circular 19, it should contain, on a sugar and fat-free basis, no higher percentage of either ash, fiber, or starch than is found in the sugar and fat-free residue of bitter chocolate. Chocolate is further modified by the addition of milk or milk solids and is sold as " Milk Chocolate/' either sweetened or unsweetened. According to Food Inspection Decision 136, both sweetened and unsweetened milk chocolates should contain at least 12% of milk solids. (For determination see page 427.) For direct comparison of different kinds of chocolate and cocoa it is best to convert the results of analysis to a " moisture-, fat-, and sugar-free basis." This is accomplished by dividing the analytical figures by 1 (M-f-F+S), where M, F, and S are the percentages of moisture, fat and sugar, respectively, each expressed as a decimal. The following procedures where marked Official or Tentative are those of the Association of Official Agricultural Chemists. Preparation of Sample (Tentative). Mix powdered products thoroughly and preserve in tightly stoppered bottles. Chill sweet or bitter chocolate until hard and reduce to a finely granular condition by grating or shaving. Mix thoroughly and preserve in a tightly stoppered bottle in a cool place. Moisture (Official). Dry about 2 grams in a flat dish (prefer- ably platinum, although nickel or aluminum is allowable) at the * To convert the analysis of a cocoa or chocolate to the "45% fat basis," multiply the analytical results by , where F= % fat found by analysis. 100 r . ANALYSIS OF FOODSTUFFS 431 temperature of boiling water for ten hours. Cool in desiccator and weigh. Then dry again for one hour, or until there is only a slight change in weight. NOTE. In the case of sweetened chocolate and cocoa it is preferable to use 3-4 grams for moisture and ash determination. f Ash (Official). Char the residue from the moisture determina- tion in the platinum dish and burn until free from carbon at not exceeding a dull red heat. If impossible thus to burn the carbon off, exhaust the charred mass with hot water, collecting the insol- uble residue on a filter. Burn the residue in the dish until the ash is white or nearly white, and then add the filtrate to the ash and evaporate to dryness. Heat at low redness until the ash is white or grayish white, cool and weigh. Ash Insoluble in Acid (Tentative). Boil the ash above obtained with 25 cc. of 10% HC1 for five minutes. Filter on a Gooch crucible or ashless filter. Wash with hot water, ignite and weigh. Soluble and Insoluble Ash (Tentative). Ash the material as directed above, employing sufficient sample to contain approx- imately 1 gram of water-, sugar-, and fat-free material. Add water to the ash, heat nearly to boiling, filter through a quanti- tative paper and wash with hot water until the combined filtrate and washings measure about 60 cc. Return the filter paper and contents to the platinum dish, ignite carefully and weigh. Cal- culate percentages of water-soluble and water-insoluble ash. Alkalinity of Soluble Ash (Tentative). Cool the filtrate from the above and titrate with 0.1 N HC1 and methyl orange. Express the alkalinity in terms of the number of cc. of 0.1 N acid per gram of sample. Alkalinity of Insoluble Ash (Tentative). Add excess of 0.1 N HC1 (usually 10-15 cc.) to the ignited insoluble ash obtained above in the platinum dish. Heat to boiling over an asbestos plate, cool and titrate the excess of HC1 with 0.1 N NaOH and methyl orange. Express the alkalinity in terms of cc. of 0.1 N acid required per gram of sample. Total Nitrogen (Official). Determine nitrogen by the Kjeldahl or Gunning or Kjeldahl-Gunning- Arnold method as described on page 64. 432 TECHNICAL METHODS OF ANALYSIS Crude Fiber (Tentative). Determine crude fiber according to the method on page 393; in cases of dispute use the Official Method of Bulletin 107, revised (Bureau of Chemistry) . For the crude fiber determination employ sufficient sample to contain approximately 1 gram of water-, sugar-, and fat-free material. Both filtrations should be made upon paper, the. washed fiber either being weighed on a tared 'filter in the usual way or rinsed from the paper into a tared Gooch crucible, and then dried and weighed. NOTE. The residue after the fat extraction may be used directly for crude fiber determination in the analysis of commercial cocoa and other finely ground or pulverized cocoa products. If, however, the material is at all granular, it should be reduced to an impalpable powder, otherwise results will be much too high. The pulverization may be satisfactorily performed by grinding with ether as described later under the determination of starch, treating the extracted residue with hot H 2 SO 4 (1.25% solution) and proceeding from that point in the usual way. Crude Starch, Direct Acid Hydrolysis (Tentative).* Weigh 4 grams of sample, if unsweetened, or 10 grams if sweetened, into a small porcelain mortar, add 25 cc. of ether and grind. (See note.) After the coarser material has settled, decant the ether, together with fine suspended matter, upon an 11 cm. paper of sufficiently fine texture to retain crude starch. Repeat this treatment until no more coarse material remains. After the ether has evaporated from the filter, transfer the fat-free residue to the mortar by means of a jet of cold water and rub to an even paste, filtering on the paper previously employed. Repeat the process until all sugar is removed. In case of sweetened products the filtrate should measure at least 500 cc. With unsweetened mate- rial less washing is necessary. Determine the crude starch in the extracted residue as follows : ' Wash the residue from the filter into a 500 cc. Erlenmeyer flask with 200 cc. of water. To the solution so prepared, either with sweetened or unsweetened goods, add 20 cc. of dil. HC1 (5 : 4) and heat for 2J hours in the flask with a reflux condenser. Cool and nearly neutralize with NaOH. Add 5 cc. of basic lead acetate solution and dilute to exactly 250 cc. Filter, and to 100 cc. of the filtrate add 1 cc. of H 2 SO 4 (60%). Filter off the PbSO 4 * The crude starch by this method will include pentosans and other carbo- hydrate bodies present which are converted into reducing sugars by HC1. ANALYSIS OF FOODSTUFFS 433 and determine reducing sugars in 25 cc. of the filtrate by the Munson and Walker procedure as directed on page 403. The weight of dextrose multiplied by 0.90 gives the weight of starch. NOTE. If the fat is to be determined, use 3 grams of the residue from the fat extraction (see below) for the starch determination and calculate back to the original sample, correcting also for moisture. Pure Starch, Diastase Method (Tentative). Remove fat and sugar from 4 grams of material, if unsweetened, and 10 grams if sweetened, as directed under Crude Starch. Wash the wet residue into a 350 cc. beaker with 100 cc. of water, heat over asbestos to boiling with constant stirring, and continue boiling and stirring for thirty minutes. Replace the water lost by evap- oration and immerse the beaker in a water bath kept between 55-60 C.; cool to the temperature of the bath, add 20 cc. of freshly prepared malt extract * and digest the mixture for two hours with occasional stirring. Boil again for thirty minutes, dilute, cool and digest as before with another 20 cc. portion of malt extract. Heat again to boiling, cool, transfer to a 250 cc. flask, add 3 cc. of alumina cream, make up to the mark and filter through dry paper. The residue on the paper should show no signs of starch when treated with weak iodine solution and examined microscopically. Conduct the hydrolysis of 200 cc. of the filtrate with 20 cc. of HC1 for 2 . 5 hours and determine the reducing power of an aliquot of the solution as directed under Crude Starch. (Omit the addition of the lead acetate and the H 2 SO 4 .) Correct for the dextrose due to added malt extract as determined by an accompanying blank analysis upon 20 cc. of malt extract carried through the same procedure. Fat (Tentative). Dry 2 grams or more of sample over H 2 S04 in a vacuum desiccator until practically all moisture is removed. (Products rich in fat show a tendency to cake at the temperature of boiling water. Hence, drying by means of heat must be avoided.) Extract with anhydrous ether in a continuous extractor until no more fat is removed (generally eight to sixteen hours). Grind and repeat the extraction (four to eight hours is generally enough). Evaporate the ether and dry the residue to constant weight at 100 C. *Malt Extract. Digest 10 grams of freshly and finely ground malt for two or three hours at room temperature with 200 cc. of water, and filter. 434 TECHNICAL METHODS OF ANALYSIS NOTE. The rapid centrifugal method, though useful and accurate under ordinary conditions, is unreliable during the summer months or in warm latitudes and has not been approved. Sucrose and Lactose (Tentative). Prepare the sample by chilling well and shaving as finely as possible with a knife. Trans- fer 26 grams of this material to an 8-ounce nursing bottle, add about 100 cc. of petroleum ether and shake for five minutes. Cen- trifugalize until the solvent is clear. Draw off by suction and repeat the treatment with petroleum ether. Place the bottle containing the de-fatted residue in a warm place until residual traces of petroleum ether are practically expelled. Add 100 cc. of water, shake until all chocolate is loosened from the sides and bottom of the bottle and then shake three minutes longer. Add basic lead acetate solution from a burette to complete precipitation, then sufficient water to make the total volume of the liquid 110 cc. Mix thoroughly and filter through a folded filter. Make a direct polariscopic reading in a 200 mm. tube. Call this A. Precip- itate the excess of Pb from solution with anhydrous K2C204, a little at a time, avoiding excess. Filter out the precipitate. Introduce 50 cc. of lead-free filtrate into a 100 cc. flask and add 25 cc. of water. Then add, little by little, while rotating the flask, 5 cc. of cone. HC1. After mixing, heat the flask in a water bath at 70 C. The temperature of the solution in the flask should read 67-69 C. in two and one-half to three minutes. Maintain the temperature as near 69 C. as possible for seven to seven and one-half minutes, making the total time of heating ten minutes. Remove the flask and cool the contents rapidly to 20 C. and dilute to 100 cc. Polarize this solution in a tube provided with a lateral branch and a water jacket maintaining a temperature of 20 C. Multiply the invert reading by 2 to correct for dilution. Call this B. From the figures obtained calculate the percentages of sucrose (S) and lactose (L) by the following formulas: (A-B) (110+x) S=- , 142M-- ANALYSIS OF FOODSTUFFS 435 i In the first formula t is the temperature at which readings were made and x is obtained from the equation : 0.2244 (A-2ld) 1- 0.00204 (A -21d)' and d in this equation is obtained from . 142.66-^ NOTE. Incase the determination of sucrose is not desired, the amount of lactose may be determined by the Defren-O'Sullivan Method (see page 405). In the case of sweetened products the percentage of lactose thus found should be corrected by subtracting 0.7% for invert sugar formed from the sucrose. Casein in Milk Chocolate (Tentative). It is unnecessary to de-fat the chocolate. Weigh 10 grams of sample into a 500 cc. Erlenmeyer flask and add exactly 250 cc. of 1% Na2C2C>4 solution. Heat to boiling and boil gently for a few minutes, then cool, add 5 grams of MgCOs and filter. Determine nitrogen in 50 cc. of this filtrate. Pipette 100 cc. of the filtrate into a 200 cc. volu- metric flask and dilute almost to the mark with water. Then precipitate the casein by addition of 2 cc. of glacial acetic acid or 1 cc. of cone. H 2 SO4. Make to volume, shake, filter, and determine nitrogen in 100 cc. of the filtrate. The difference between the two nitrogen determinations gives the nitrogen derived from the casein, which, multiplied by 6.38, gives the amount of casein present in 2 grams of sample . NOTE. Casein is approximately 80% of the total proteins in milk; hence to find the total milk protein, divide by 0.8 the percentage of casein obtained above. Theobromine and Caffeine. Boil 10 grams of the powdered sample and 5 grams of calcined MgO for thirty minutes with 300 cc. of water; filter by suction on a Buchner funnel, using a round disk of filter paper. Transfer the material and paper to the original flask, add 150 cc. of water, boil for fifteen minutes, filter as before and repeat the operation of boiling with 150 cc. of water, 436 TECHNICAL METHODS OF ANALYSIS and filter. Wash twice with hot water, evaporate the united filtrates (with ignited quartz sand if sugar be present) to complete dryness in a Hoffmeister Schalchen, or other suitable thin glass dish of about 300 cc. capacity; grind the dish with contents to a coarse powder in a mortar; transfer to the inner tube of a Weber extractor,* dry thoroughly in a water oven and extract with CHC1 3 for eight hours in a weighed flask. Distill off the CHCls and dry the residue to constant weight at 100 C. Treat the residue in the flask for some hours at room temperature with 50 cc. of ben- zene. Filter through a small paper into a tared dish, evaporate to dryness and dry to constant weight at 100 C., thus obtaining the amount of caffeine. Determine theobromine as follows: Add to the residue and filter paper 150 cc. of water, enough NHUOH to make the solution slightly alkaline and an excess of 0.1 N AgNOs, accurately meas- ured. Boil to half volume, add 75 cc. of H^O and repeat boiling. If any NHs remains, repeat the adding of water and boiling until the solution is perfectly neutral. Filter off the insoluble silver theobromine compound and wash with hot water. To the fil- trate and washings add 5 cc. of a saturated solution of iron alum and a few cc. of HNOs (free from the lower oxides of N). Titrate the excess AgNOs with 0.1 N KSCN until a permanent light brown color appears. Subtract the amount of AgNOs thus determined from the original amount added and calculate the difference to theobromine. CALCULATION. 1 cc. 0.1 N AgNOs = 0.01801 gram theo- bromine. Other Nitrogenous Substances. Add the percentages of nitrogen present as theobromine and caffeine, subtract the sum from the per cent of total N and multiply the remainder by 6.25. CALCULATIONS. Theobromine X 0.31 1 1 = Nitrogen. Caffeine X 0.2886 = Nitrogen. Other Nitrogen Free-Substances. To show a complete anal- ysis, subtract from 100% the sum of the percentages of moisture, ash, theobromine, caffeine, other nitrogenous substances, pure starch, and fat, and report the difference as " other nitrogen-free substances." * Any other form of extractor which permits of hot extraction may be used. ANALYSIS OF FOODSTUFFS 437 Fat Constants. Separate the fat in a manner similar to that described above under Sucrose and Lactose and determine the melting point, refractive index (at 40 C.), saponification number, iodine number and Reichert-Meissl number. NOTES. (1) Melting point determinations on this material do not become normal until the fat has been kept for at least twenty-four hours in a cool place, preferably in a desiccator. (2) The constants for fat extracted from pure cocoa are as follows: Melting point 23-28 C. Refractive index at 40 C 1 . 4566-1 . 4579 Iodine number 32-41 . 7 Reichert-Meissl number . 2-0 . 8 Saponification number 192-200 Milk Fat in Milk Chocolate. In the case of milk chocolate the .extracted fat will consist of cocoa fat and butter. As the Reichert- Meissl number of cocoa fat is approximately 0.5 and that of butter 28.25, the approximate percentage of butter fat in the total fat can be calculated after determining its Reichert-Meissl number. If A = grams of butter fat in 5 grams of mixed fat, B = 5 A = grams of cocoa fat in 5 grams of mixed fat, arid C = Reichert-Meissl number of extracted fat ; 28.25 A +0.55 27.75A +2.5 then C=- - = - , C-0.5 and A "5T' C 5 Whence, per cent butter fat = per cent total fat X . 27.75 Milk Solids in Milk Cocoa or Chocolate. The milk solids are calculated as the sum of the butter fat, total milk proteins, lac- tose, and milk ash. The amount of milk ash is calculated by taking 5%* of the sum of the butter fat, milk proteins, and lactose. * According to Leach: "Food Inspection and Analysis" this would be 5.9%, but the Food Laboratories of the U. S. Dept of Agriculture use 5%. 438 TECHNICAL METHODS OF ANALYSIS As a check on the above calculations, it may be noted that the average composition of milk solids is approximately as follows: Per Cent Total protein 29.9 Lactose 35 . 5 Butter fat 28.4 Ash 5.5 Citric acid 0.7 100.0 Casein 23 . 7 REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 327; Leach: " Food Inspection and Analysis." FEED STUFFS AND MIXED GRAINS Preparation of Sample. Grind the sample so that it will pass through a sieve having circular holes 1 mm. in diameter. This is very important in the case of mixed grains. A suitable mill is the grinding and pulverizing mill 059 made by the Enterprise Mfg. Co., Philadelphia. The mill should be screwed down tightly and so-called spice grinders used in it. Pass the material through the mill, sieve it on the millimeter sieve, and re-grind the residue until it all passes the sieve. If it is impossible to grind it fine enough to go through the sieve, pound up the residue in an iron mortar. Finally mix the ground sample very thoroughly before removing portions for analysis. In the case of soft or sticky feeds that cannot be ground, reduce the sample to as fine a state as possible. NOTE. For occasional samples the No. mill is satisfactory, although this is not made to run by power. The Bureau of Chemistry at Washington uses the Enterprise Mill, serial 2962, which is run by a three-quarter horsepower motor. 1 Moisture. Dry a portion of the ground material, representing about 2 grams of dry substance, at the temperature of boiling water to constant weight (approximately five hours) in a current of dry hydrogen or in vacuo. If the substance is in a glass vessel the latter should not be in contact with the boiling water. ANALYSIS OF FOODSTUFFS 439 Ash. Char about 2 grams of the dry material in a weighed platinum dish and burn until free from carbonate at the lowest possible heat (not above dull redness). If a carbon-free ash can- not be obtained in this manner, exhaust the charred mass with hot water. Collect the insoluble residue on a filter, ignite it in the dish until the ash is white or nearly so, then add the filtrate to the ash in the dish and evaporate to dryness. Heat the whole to low redness till the ash is white or grayish white, cool in a desic- cator and weigh. Crude Protein. Determine the nitrogen by the Kjeldahl method or the Gunning method as described on page 64. Mul- tiply the percentage of N by 6.25 to obtain crude protein. Albuminoid Nitrogen. (A) STUTZER'S REAGENT. Dissolve 100 grams of pure copper sulfate in 5 liters of water, add 2.5 cc. of glycerol and then dil. NaOH solution until the liquid is just alkaline. Filter, rub the precipitate up with water containing 5 cc. of glycerol per liter, and wash by decantation or filtration until the washings are no longer alkaline. Rub the precipitate up again in a mortar with water containing 10% of glycerol, thus preparing a uniform gelatinous mass that can be measured with a pipette. Determine the quantity of Cu(OH)2 per cc. of this mixture. (B) DETERMINATION. Place 0.7 gram of the sample in a beaker, add 100 cc. of water and heat to boiling; or, in case of sub- stances rich in starch, heat on the water bath for ten minutes. Add a quantity of the Stutzer's reagent containing about 0.5 gram of the Cu(OH)2, stir thoroughly, filter when cold, wash with cold water and, without removing the precipitate from the filter, determine the nitrogen according to the Kjeldahl, Gunning, or Kjeldahl- Gunning-Arnold method, as described on page 64, adding suf- ficient K 2 S solution to completely precipitate all of the Cu and Hg. The filter paper used must be essentially free from N. If the material (such as seeds, seed residue, or oil cake) is rich in alkaline phosphates, add 1-2 cc. of cone, potash or soda alum solution, free from NHs, then the Cu(OH) 2 and mix well by stirring. If this is not done, copper phosphate and free alkali may be formed and the protein-copper precipitate partially dis- solved in the alkaline liquid. 440 TECHNICAL METHODS OF ANALYSIS Amido Nitrogen. Subtract the amount of albuminoid N from the amount of total N to obtain the amido N. Crude Fat (Ether Extract). (A) REAGENT. Prepare anhy- drous ether as follows: Wash the commercial ether with 2-3 successive portions of water. Add solid NaOH or KOH and let stand until most of the water has been extracted. Decant into a dry bottle. Add carefully cleaned metallic sodium cut into small pieces and let stand until there is no further evolution of hydrogen. The ether thus dehydrated must be kept over metallic sodium in a lightly stoppered bottle to allow any accumulated hydrogen to escape. It may be drawn off with a pipette as required. (B) DETERMINATION BY DIRECT METHOD. Large quantities of soluble carbohydrates may interfere with the complete extraction of the fat. In such cases extract with water before proceeding with the determination. Extract about 2 grams of material, dried as previously described, with the anhydrous ether for six- teen hours. Dry the extract at the temperature of boiling water for thirty minutes, cool in a desiccator and weigh. Continue the alternate drying and weighing at thirty-minute intervals to con- stant weight. For most feeds a period of one to one and one-half hours is required. (C) INDIRECT METHOD. Determine the moisture as previously directed, then extract the dried substance for sixteen hours with anhydrous ether, dry again and regard the loss in weight as ether extract. Crude Fiber. Determine the crude fiber as directed on page 393. Carbohydrates. In feed stuffs, free from sugar, and in mixed grains the carbohydrates are generally taken by difference. Add together the moisture, ash, protein, fat and crude fiber and sub- tract the sum from 100% for the carbohydrates. Total Sugars.* (A) PREPARATION OF SOLUTION. Place 10 grams of the material in a 250 cc. graduated flask. If the sub- stance has an acid reaction, add 1-3 grams of CaCOs and boil on a steam bath for one hour with 150 cc. of 50% alcohol by volume, using a small funnel in the neck of the flask to condense the vapor. Cool and let stand several hours, preferably overnight. Make up to volume with neutral 95% alcohol, mix thoroughly, let * -Particularly applicable to cattle foods. ANALYSIS OF FOODSTUFFS 441 settle, transfer 200 cc. to a beaker with a pipette and evaporate on the steam bath to a volume of 20-30 cc. Do not evaporate to dryneas; a little alcohol in the residue does no harm. Transfer to a 100 cc. graduated flask and rinse the beaker thoroughly with water, adding the rinsings to the contents of the flask. Add enough saturated neutral lead acetate solution to produce a floc- culent precipitate. Shake thoroughly and let stand fifteen min- utes. Make up to the mark with water, mix thoroughly and filter through a dry filter. Add sufficient anhydrous Na2COs to the filtrate to precipitate all the Pb. Again filter through a dry filter and test the filtrate with a little anhydrous Na2COs to make sure that all the Pb has been removed. (B) REDUCING SUGARS. Determine the reducing sugars by the Munson and Walker Method (see page 403), employing the Soxhlet modification of Fehling's solution and using 25 cc. of the solution prepared as above directed (representing 2 grams of the sample). Express the results as dextrose or invert sugar. (See note.) (C) SUCROSE. Introduce 50 cc. of the solution prepared as above directed into a 100 cc. graduated flask; add a piece of litmus paper, neutralize with acetic acid, add 5 cc. of cone. HC1 and let the inversion proceed at room temperature (for twenty-four hours at a temperature of 20-25 C. or for ten hours, if the temperature be above 25 C.). When inversion is complete, transfer the solu- tion to a beaker. Neutralize with Na2COs, return the solution to the 100 cc. flask, dilute to the mark with water, filter if necessary and determine reducing sugars in 50 cc. of the solution (represent- ing 2 grams of the sample), as directed under Reducing Sugars above, and calculate the result as invert sugar. Subtract the per cent of reducing sugars before inversion from the per cent of total sugar after inversion, both calculated as invert sugar, and multiply the difference by 0.95 to obtain the per cent of sucrose present. NOTE Since the insoluble material of grain or cattle food occupies some space in the flask as originally made up, it is necessary to correct for this volume. Results of a large number of determinations on various materials have shown the average volume of 10 grams of material to be 7.5 cc. and therefore to obtain the true amount of sugars present all results must be multiplied by the factor 0.97 (since 7.5 cc. is 3% of 250 cc.). 442 TECHNICAL METHODS OF ANALYSIS Starch (Direct Acid Hydrolysis, Modified Sachsse Method.) By this method there will be included as starch the pentosans and other carbohydrate bodies present which undergo hydrolysis and inversion to reducing sugars on boiling with HC1. PROCEDURE. Stir a quantity of the sample representing 2.5-3 grams of the dry material in a beaker with 50 cc. of cold water for one hour. Transfer to a filter and wash with 250 cc. of cold water. Heat the insoluble residue for two and one-half hours with 200 cc. of water and 20 cc. of dil. HC1 (5 : 4) in a flask pro- vided with a reflux condenser. Cool and nearly neutralize with NaOH. Add the proper clarifying agent and complete the volume to 250 cc. Filter and determine the dextrose in an aliquot of the filtrate as directed by the Munson and Walker Method (page 403), or the Allihn Method (page 407). The weight of dextrose obtained, multiplied by 0.90, gives the weight of starch. NOTE. The factor 0.90 is the theoretical ratio between starch and glucose, but, according to Noyes and other investigators, the factor 0.93 more nearly approaches the actual yield. Pentosans. The determination of pentosans is seldom neces- sary. No official method has yet been adopted. The tentative method of the Association of Official Agricultural Chemists is described under Furfural Value of Cotton Cellulose on page 366. Galactan. The same remarks apply to galactan as to pen- tosans. The tentative method is described in J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 118. Water-soluble Acidity (Tentative). Weigh 10 grams of the sample into a shaking bottle. Add 200 cc. of water and shake for fifteen minutes. Filter the extract through a folded filter, pipette out 200 cc. (equivalent to 1 gram), dilute with 50 cc. of water and titrate with 0.1 N NaOH and phenolphthalein. State the result in terms of cc. of 0.1 N NaOH required for neutralization. REFERENCE. The above are essentially the official methods of the Assoc. Official Agr. Chemists (except those marked Tentative) described in its Jour- nal, Methods of Analysis (1916), pages 79-120. There have been, slight changes* made to conform to the usage of this laboratory. ANALYSIS OF FOODSTUFFS 443 MILK AND CREAM MILK General. The sp. gr. of pure milk at 60 F. generally ranges from 1.027-1.035. Its average composition is approximately as follows : Per Cent Butter Fat 3.6 Solids not Fat: Casein 3.0 Other nitrogenous sub- stances 0.8 \ 9.1 Lactose 4.5 Citric Acid 0.1 Ash 0.7 Total Solids 12.7 According to the above figures, the composition of the solids should be as follows: Per Cent Casein 23.7 Lactose 35 . 5 Fat 28.4 Ash 5.5 Total Protein 29.9 It will be seen that casein is about 80 per cent of the total protein. The U. S. federal standard for pure milk (Bureau of Chem., Circular 19) is as follows: Per Cent Solids not Fat minimum 8 . 5 Milk Fat minimum 3 . 25 Skimmed milk should contain not less than 9.25% of milk solids. The usual determinations in ascertaining the nutritive values of milk are Specific Gravity, Total Solids, Fat, Protein, Lactose, and Ash. The following methods, unless otherwise indicated, are the official methods of the Association of Official Agricultural Chemists. 444 TECHNICAL METHODS OF ANALYSIS Specific Gravity.* Determine the sp. gr. at 60 F. with a pycnometer, first shaking the sample well. Total Solids. Heat 3-5 grams of the milk at the temperature of boiling water until it ceases to lose weight, using a tared, flat- bottomed dish of not less than 5 cm. diameter. If desired, previ- ously place 15-20 grams of pure, dry sand in the dish. Cool in a desiccator and weigh rapidly to avoid absorption of hygro- scopic moisture. Ash. Weigh about 20 grams of the milk in a tared porcelain dish, add 6 cc. of HNOs, evaporate to dryness and ignite at a temperature just below redness until the ash is free from carbon. Protein. Place about 5 grams of the milk in a Kjeldahl diges- tion flask and determine (without evaporation) the total N by the Kjeldahl or Gunning or the Kjeldahl-Gunning- Arnold method (page 64). Multiply the percentage of N by 6.38 to obtain the percentage of N compounds or total protein. Casein. This determination should be made while the milk is fresh or nearly so. If it cannot be made within 24 hours, add 1 part of formaldehyde to 2500 parts of the milk and keep in a cool place. METHOD No. 1. Place 10 grams of the milk in a beaker 'with 90 cc. of water at 40-42 C. and add at once 1.5 cc. of 10% acetic acid. Stir and let stand 3-5 minutes. Then decant and filter, wash by decantation 2-3 times with cold water and trans- fer the precipitate to the filter. Wash once or twice on the filter. The filtrate should be clear, or very nearly so. If the first por- tions of the filtrate are not clear, repeat the filtration, after which complete the washing of the precipitate. Determine N in the washed precipitate and filter paper as directed above, multiply by 6.38 and calculate the percentage of casein. In samples of milk which have been preserved, the acetic acid should be added in small portions, a few drops at a time, with stirring, and the addition continued until the liquid above the precipitate becomes clear, or very nearly so. METHOD No. 2. To 10 grams of the milk add 50 cc. of water at 40 C., then 2 cc. of alum solution, saturated at 40 C. or higher. Let the precipitate settle, transfer to a filter and wash with cold water. Determine N in the precipitate and filter paper * Not included in the A. 0. A. C, method. ANALYSIS OF FOODSTUFFS 445 as directed above, multiply by 6.38 and calculate the percentage of casein. Lactose. Dilute 25 grams of the milk with 400 cc. of water in a 500 cc. volumetric flask. Add 10 cc. of Fehling's copper sulfate solution (Soxhlet modification) and about 7.5 cc. of KOH solution of such strength that 1 volume is just sufficient to pre- cipitate completely the Cu from 1 volume of the Fehling's solu- tion. (Instead of KOH of this strength, 8.8 cc. of 0.5 N NaOH solution may be used.) After the addition of the alkali solution, the mixture must still have an acid reaction and contain copper in solution. Dilute the solution to the mark, mix, filter through a dry filter and determine lactose in an aliquot of the filtrate by the Munson and Walker method (page 403). Fat (Roese-Gottlieb Method). Weigh 10-11 grams of the milk into a Rohrig tube or some similar apparatus, add 1.25 cc. of cone. NELtOH (2 cc. if the sample is sour) and mix thoroughly. Add 10 cc. of 95% alcohol by volume and mix well. Then add 25 cc. of washed ether and shake vigorously for 30 seconds, then 25 cc. of petroleum ether (re-distilled slowly at a temperature below 60 C.) and shake again for 30 seconds. Let stand 20 minutes, or until the upper liquid is practically clear. Draw off as much as possible of the ether-fat solution (usually 0.5-0.8 cc. will be left) into a weighed flask through a small, quick-acting filter. (The flask should always be weighed with a similar one as a counter- poise.) Re-extract the liquid remaining in the tube, this time with only 15 cc. of each ether, shake vigorously 30 seconds with each and let settle. Draw off the clear solution through the small filter into the same flask as before and wash the tip of the spigot, the funnel and the filter with a few cc. of a mixture of the two ethers in equal parts. For absolutely exact results the re-extraction must be repeated. This third extraction yields usually not more than about 1 mg. of fat (about 0.02% on a 4-gram charge) if the previous ether-fat solutions have been drawn off closely. Evaporate the ethers slowly on the steam bath, then dry the fat in a boiling water oven to constant weight. Confirm the purity of the fat by dissolving in a little petroleum ether. Should a resin remain, remove* the fat completely with petroleum ether, dry the residue, weigh and deduct the weight. 446 TECHNICAL METHODS OF ANALYSIS Finally correct this weight by a blank determination on the reagents used. Fat (Babcock Method). (I) APPARATUS. (a) Standard Babcock test bottles. The standard Babcock test bottles for milk and cream shall be as follows : (1) 8%, 18 GRAM, 6-lNCH MILK TEST BOTTLE. The total per cent graduation shall be 8. The total height of the bottle shall be 150-165 mm. (5|- 6i inches) . The capacity of the bulb up to the junction with the neck shall be not less than 45 cc. The graduated portion of the neck shall have a length of not less than 63.5 mm. (2.5 inches) and the neck shall be cylindrical for at least 9 mm. below the lowest and above the highest graduation marks. The gradu- ations shall represent whole per cents, halves and tenths of a per cent. (2) 50%, 9 GRAM, 6-lNCH CREAM TEST BOTTLE. The total per cent graduation shall be 50. The total height of the bottle shall be 150-165 mm. (5|-6| inches). The capacity of the bulb up to the junction with the neck shall be not less than 45 cc. The graduated portion of the neck shall have a length of not less than 63.5 mm. (2.5 inches) and the neck shall be cylindrical for at least 9 mm. below the lowest and above the highest graduation marks. The graduations shall represent five per cents, whole and halves of a per cent. (3) 50%, 9 GRAM, 9-lNCH CREAM TEST BOTTLE. Same as (2) except that the total height of the bottle shall be 210-225 mm. (8J-8f inches.) (6) Centrifuge. (c) Pipettes. Graduated to deliver 17.6 cc. of water at 20 C. in 5-8 seconds. (d) Graduates. Capacity 17.5 cc. or a Swedish acid bottle delivering that amount. (II) CALIBRATION OF APPARATUS. (a) Graduation. The unit of graduation for all Babcock ANALYSIS OF FOODSTUFFS 447 glassware shall be the true cc. (0.998877 gram of water at 4 C.). With bottles, the capacity of each per cent on the scale shall be 0.20 cc. With pipettes and graduates, the delivery shall be the intent of the graduation; and the graduation shall be read with the bottom of the meniscus in line with the mark. (6) Testing. The method for testing Babcock bottles shall be calibration with mercury (13.5471 grams of clean, dry mercury at 20 C., to be equal to 5% on the scale), the bottle being previously filled to zero with mercury. (The mercury and cork, alcohol and burette, and alcohol and brass plunger methods may be employed for the rapid testing of Babcock bottles, but the accuracy of all questionable bottles shall be determined by calibration with mercury.) The method for testing pipettes and graduates shall be calibration by measuring in a burette the quantity of water (at 20 C.) delivered. (c) Limit of error. For standard Babcock milk bottles the error at any point of the scale shall not exceed 0.1%. For standard Babcock cream bottles the error at any point of the scale shall not exceed 0.5%. For standard milk pipettes the error shall not exceed 0.05 cc. For acid measures the error shall not exceed 0.2 cc. (Ill) DETERMINATION. Pipette 17.6 cc. of the carefully mixed sample into a test bottle and add 17.5 cc. of commercial H2SO4 (sp. gr. 1.82-1.83). Mix and, when the curd is dis- solved, centrifugalize for 4 minutes at the required speed for the machine used. Add boiling water, filling to the neck of the bottle, and whirl for 1 minute; again add boiling water so as to bring the fat within the scale on the neck of the bottle, and, after whirling for 1 minute more, read the length of the fat column, making the reading 57-60 C., at which temperature the fat is wholly liquid. The reading gives directly the per cent of fat in the milk. 448 TECHNICAL METHODS OF ANALYSIS Gelatin (Qualitative Test Tentative). To 10 cc. of the milk add an equal volume of acid Hg(NOa)2 solution (Hg dissolved in twice its weight of cone. HNOs and this solution diluted to 25 times its volume with water), shake the mixture, add 20 cc. of water, shake again, let stand 5 minutes and filter. If much gel- atin is present, the nitrate will be opalescent and cannot be obtained quite clear. To a portion of the filtrate contained in a test-tube, add an equal volume of saturated aqueous picric acid solution. A yellow precipitate will be produced in the presence of any con- siderable amount of gelatin, while smaller amounts will be indi- cated by a cloudiness. In the absence of gelatin the filtrate will remain perfectly clear. Preservatives. For these tests see J. Assoc. Off. Agr. Chem. Methods of Analysis, 1916, page 141. To test for salicylic or benzoic acid, first acidify 100 cc. of the milk with 5 cc. of HC1 (1 : 3), shake until curdled, filter and proceed with the clear filtrate. Coloring Matter (Leach Method Tentative). Warm about 150 cc. of milk in a casserole over a flame and add about 5 cc. of acetic acid, then slowly continue the heating nearly to the boiling point while stirring. Gather the curd, when possible, into one mass with a stirring rod and pour off the whey. If the curd breaks up into small flecks, separate from the whey by straining through a sieve or colander. Press the curd free from adhering liquid, transfer to a small flask and macerate for several hours, prefer- ably overnight, in about 50 cc. of ether, the flask being tightly corked and shaken at intervals. Decant the ether extract into an evaporating dish, remove the ether by evaporation and test the fatty residue for annatto as directed on page 392. The curd of an uncolored milk is perfectly white after complete extraction with ether, as is also that of a milk colored with annatto. If the extracted fat-free curd is distinctly colored an orange or yellowish color, a coal tar dye is indicated. In many cases, upon treating a lump of a fat-free curd in a test-tube with a little strong HC1 the color changes to pink, indicating the presence of a dye similar to aniline yellow or butter yellow or perhaps one of the acid azo yellows or oranges. In such cases, separate and identify the coloring matter present in the curd as directed on page 389. If aniline yellow, butter yellow, or other oil-soluble dye is present, the greater part will be found in the ether extract containing the ANALYSIS OF FOODSTUFFS 449 fat. In such cases proceed as directed on page 391 under Oil- soluble Dyes. In some cases the presence of coal tar dyes can be detected by treating about 10 cc. of the milk directly with an equal volume of cone. HC1 in a procelain casserole, giving the dish a slight rotary motion. In the presence of some dyes the separated curd acquires a pink coloration. CREAM Total Solids. Follow the same procedure as for Milk, using 2-3 grams of the sample. Ash, Protein, and Lactose. Proceed as above under Milk. Fat. Follow the method given above under Milk. For the extraction method, weigh 4-5 grams of the homogeneous sample into a Rohrig tube or similar apparatus, dilute with water to about 10.5 cc. and proceed as directed. For the Babcock method, weigh 9 or 18 grams, depending upon the fat content, into a standard Babcock cream bottle and proceed as directed. Gelatin. Follow the same procedure as for Milk. Coloring Matter. Follow the procedure on page 389, par- ticularly looking for oil-soluble dyes and annatto. Preservatives.^-Proceed as directed on page 448. REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916- 17), pages 287-293. CONDENSED OR EVAPORATED MILK General. The common meaning of " condensed milk " in this country is milk which has been condensed and preserved with cane sugar. Milk which has been condensed without added sweetening is often termed " evaporated cream," although this term is a misnomer. The U. S. Standard for condensed milk requires not less than 25.5% of milk solids and not less than 7.8% of milk fat (Food Inspection Decision 158). In the case of unsweetened condensed milk, dilute 40 grams of the homogeneous sample with 60 grams of water and make the 450 TECHNICAL METHODS OF ANALYSIS determinations as directed under Milk and Cream, (page 443) correcting the results for the dilution. The following procedures are to be used on sweetened products. Preparation of Sample. Warm the sample to 30-35 C., transfer to a dish sufficiently large to stir thoroughly and make the whole mass homogeneous. Weigh 100 grams into a 500 cc. volumetric flask and dilute to the mark with water. If the milk will not dissolve, weigh out each portion for analysis separately. Total Solids and Moisture. Pipette 10 cc. of the above solu- tion into a small weighed platinum dish containing 15-20 grams of pure, dry sand or asbestos fiber, which has previously been ignited and weighed with the dish. Dry to constant weight at the temperature of boiling water, cool in a desiccator and weigh rapidly. Consider the difference between the percentage of total solids and 100% to be moisture. Ash. Ignite the residue from the total solids carefully, cool and weigh. Protein. Pipette 10 cc. of the prepared solution into a Kjeldahl flask and determine nitrogen (without evaporation) by the Kjel- dahl or Gunning or Kjeldahl-Gunning- Arnold method as described on page 64. Multiply the N by 6.38 to obtain the protein. Lactose (Milk Sugar). Dilute 100 cc. of the prepared solu- tion in a 250 cc. volumetric flask to about 200 cc. Add 6 cc. of Fehling's copper sulfate solution (Soxhlet modification) and dilute to the mark. Filter through a dry filter and determine lactose by the Munson and Walker Method (page 403). Fat. (1) RESE-GOTTLIEB METHOD. Weigh 4-5 grams of the original homogeneous sample into a Rohrig tube or similar apparatus, dilute with water to about 10.5 cc. and proceed as directed under Milk and Cream (page 445). (2) ADAMS METHOD. Prepare strips of soft white filter paper about 4X12 inches, of the quality of S. and S. No. 597, by soaking 2 or 3 hours in alcohol and then, after thoroughly drying in the oven, extract several hours with ether or until no residue is left from the ether as it comes through. Distribute 5 cc. of a 40% solution of the condensed milk carefully over the whole surface of the thoroughly dried paper. (This is best done by attaching one end of the paper to some object and holding the other end out straight so that the pipette can be emptied by pass- ANALYSIS OF FOODSTUFFS 451 ing the point back and forth over the whole surface.) To dry the paper, suspend it over a copper wire in the drying oven, where it will thoroughly dry out in 2 hours, or much more rapidly than if coiled up or put in a tube. After drying, roll up in a coil, wind with thread or small copper wire, place in the Soxhlet extractor, and extract with ether for not less than 8 hours. .Remove the coil from the extractor, loosen the wire or thread and let the ether evaporate. Suspend in 500 cc. of water for 2 hours, then return the coil to the oven, dry as before, and extract again for not less than 5 hours. Sucrose. Determine sucrose " by difference," deducting from 100% the sum of the moisture, ash, protein, lactose and fat. Milk Solids. These are the total solids, minus the sucrose. REFERENCES. The above are the official methods of the Association of Official Agr. Chem. as published in its Journal, II, Methods of Analysis (1916-17), page 293, except the Adams Method for Fat, which is described in Leach: " Food Inspection and Analysis " (1914 ed.), page 154. CHEESE General. The U. S. Government standard for cheese accord- ing to Circular No. 19 is as follows: Cheese is made from milk or cream by coagulating the casein with rennet or lactic acid, with or without the addition of ripening ferments and seasoning, and contains in the water-free substance not less than 50% of milk fat. It may also contain added coloring matter. Cheese is sometimes adulterated by preserving with boric acid or borax. (See page 284 for detection.) The following procedures, unless otherwise indicated, are the official methods of the Association of Official Agricultural Chemists. Sampling. When the cheese can be cut, take a narrow wedge- shaped segment reaching from the outer edge to the center. Cut into strips and pass three times through a sausage machine. When the cheese cannot be cut, take the sample with a cheese trier. If only one plug can be obtained, take it perpendicular to the surface of the cheese at a point one-third the distance from the edge to the center, extending either entirely or half-way through. When possible, draw 3 plugs: 1 from the center, 1 from a point 452 TECHNICAL METHODS OF ANALYSIS near the outer edge, and 1 from a point half-way between the other two. For inspection purposes reject the rind, but for inves- tigations requiring the absolute amount of fat in cheese include the rind in the sample. Either grind the plugs in a sausage machine or cut very finely and mix carefully, preferably the former. Moisture (Tentative). Place 2-3 grams of very short fiber asbestos (long fiber may be made suitable by rubbing through a fine sieve) in a flat-bottomed platinum dish, 6-7 cm. in diameter, and pres's the asbestos down firmly. Place in the dish a glass rod about 5 mm. in diameter and slightly longer than the diameter of the dish. Ignite, cool and weigh the dish and contents. Then weigh into the dish 4-5 grams of sample, prepared as above, and mix the cheese and asbestos intimately with the glass rod until the mass is homogeneous. Leave the mass in as loose a condi- tion as possible to facilitate drying. Dry the mixture in an oven at 100 C. and weigh at intervals of one to one and one-half hours until the weight becomes constant. (Three weighings are usually sufficient.) Ash. Ignite the residue from the moisture determination cautiously to avoid spattering and remove the heat while the fat is burning off. When the flame has died out, complete the burning in a muffle at low redness. Cool in a desiccator and weigh the ash. Salt. Dissolve the ash from the above determination in water slightly acidified with HNOs, and determine the chlorine, either gravimetrically as AgClj or volumetrically by the chromate method as on page 492. Calculate to NaCl. CALCULATION. AgCl X 0.4078 = NaCl. 1 cc. 0.1 N AgNO 3 = 0.005846 gram NaCl. Nitrogen. Determine nitrogen by the Kjeldahl or Gunning or Kjeldahl-Gunning- Arnold method as on page 64, using about 2 grams of cheese, and multiply the per cent of N by 6.38 to obtain the per cent of nitrogen compounds. Acidity (Tentative). To 10 grams of finely divided cheese add water at 40 C. until the volume equals 105 cc. Shake vigor- ously and filter. Titrate 25 cc. portions of the filtrate, representing 2.5 grams of sample, with 0.1 N NaOH and phenolphthalein. Express results in terms of lactic acid. CALCULATION. 1 cc. 0.1 N NaOH = 0.009 gram lactic acid. ANALYSIS OF FOODSTUFFS 453 Coloring Matters (Tentative). Proceed according to page 389. Fat. (A) PREPARATION OF SAMPLE (TENTATIVE). (1) Alka- line extraction. Treat about 300 grams of cheese, cut into frag- ments the size of a pea, with 700 cc. of 5% KOH solution at 20 C. in alarge, wide-necked flask. Shake vigorously to dissolve the casein. In five to ten minutes the casein will be dissolved and the fat will rise to the surface in lumps. Collect into as large a mass as possible by shaking gently. Pour cold water into the flask until the fat is driven up into the neck and remove with a pipette. Wash the fat thus obtained with just sufficient water to remove any alkali. The fat is not perceptibly attacked by the alkali in this treatment and is practically all separated in a short time. It is then prepared for chemical analysis by filtering through a dry filter paper in a hot water funnel at about 60 C. Refilter, if necessary, and dry. (2) Acid extraction Pass the cheese through a grinding machine; transfer to a large flask and cover with warm water, using 1 cc. for every gram of cheese. Shake thoroughly and add H 2 SO4 (sp. gr. 1.82-1.825) slowly and in small quantities, shaking after each addition. The total amount of acid used should be the same as the amount of water. Remove the fat, which separates after standing a few minutes, by means of a separatory funnel, wash free from acid, filter and dry as above. (B) EXAMINATION OF FAT. Make such tests as are necessary on the fat by procedures described on pages 230, 241, and 243. As a general rule the sp. gr., refractive index, iodine number, and Reichert-Meissl number will be sufficient. (C) QUANTITATIVE ESTIMATION. Cover the perforations on the bottom of an extraction tube with dry asbestos and place on this a mixture containing equal parts of anhydrous CuSCU and pure dry sand to a depth of about 5 cm., packing loosely. Cover the upper surface of this mixture with a layer of asbestos. Place on this 2-5 grams of the sample and extract with anhydrous ether for five hours in a continuous extraction apparatus. Remove the cheese and grind it with pure sand in a mortar to a fine powder. Return the mixed cheese and sand to the extraction tube, wash out the mortar with ether, add the washings to the tube and continue the extraction for at least ten hours. Dry the fat at 105 C., cool in desiccator and weigh. 454 TECHNICAL METHODS OF ANALYSIS REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 296. VINEGAR Preparation of Sample. Before starting the analysis note the appearance, color, odor and taste. For microscopical examina- tion employ the original sample, but for chemical analysis filter if turbid. Calculation of Results. Express all results in grams per 100 cc., unless otherwise noted. Specific Gravity. Determine the sp. gr. at 15.5 C. by means of a pycnometer, a small accurately graduated hydrometer, or a Westphal plummet on the analytical balance. If a pycnometer is used, it should be warmed quickly to room temperature after filling and before weighing, to prevent the error due to collection of moisture on the outside. A small hole filed in the cap will permit the necessary expansion in the volume of the liquid. Alcohol. Measure 100 ce. of the sample into a round-bottom distilling flask and make faintly alkaline with saturated NaOH solution. Add a small scrap of paraffin, distill nearly 50 cc. and make up the distillate to 50 cc. at the temperature of the sample. Filter if necessary and determine the sp. gr. with a pycnometer. Calculate the grams of alcohol per 100 cc. from standard alcohol tables * bearing in mind that the alcohol strength of the distillate is twice that of the original vinegar. GlyceroL When it is necessary to determine the amount of glycerol, use the method described in J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 253. Total Solids (Tentative). Measure 10 cc. of filtered vinegar into a weighed flat-bottomed platinum dish of 50 mm. bottom diameter. Evaporate on a boiling water bath to a thick syrup t and dry for exactly 2.5 hours in the drying oven at the temperature of boiling water. Cool in a desiccator and weigh. NOTE. It is essential to use a flat-bottomed dish. We have found, how- ever, that silica dishes may be used in place of platinum. * Leach: "Food Inspection and Analysis," 3d Ed., page 661, 4th Ed. page 690; Van Nostrand's " Chemical Annual"; J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 194; Bureau of Standards, Circular 19. fThis should require 30 minutes. ANALYSIS OF FOODSTUFFS 455 Total Reducing Substances before Inversion. Proceed accord- ing to Munson and Walker's Method (page 403), using 10 or 20 cc. of the sample; express results as grams of invert sugar per 100 cc. Malt vinegar should be clarified with sodium phospho- tungstate. In the case of malt vinegar, express the results as dextrose; in all other cases, as invert sugar. Reducing Sugars before Inversion, after Evaporation. Evaporate 50 cc. on the water bath to 5 cc. Add 25 cc. of water and evaporate to 5 cc. Again add 25 cc. of water and evaporate to 5 cc. Transfer to a 100 cc. volumetric flask, make up to the mark and proceed as in the preceding paragraph, using a quantity equivalent to 10 or 20 cc. of the sample. Reducing Sugars after Inversion. Proceed as in the preceding paragraph and, after the last evaporation to 5 cc., transfer to a 100 cc. flask with 70 cc. of water. Clarify with basic lead acetate solution and alumina cream, and invert the solution with HC1 as described on page 400 under the Determination of Sucrose in the Absence of Raffinose by Polarization before and after Inver- sion. Nearly neutralize with NaOH solution, make up to the mark and determine the reducing sugars by the Munson and Walker method (page 403), taking an aliquot equivalent to 10 or 20 cc. of the sample. Lead Precipitate. To 10 cc. of the sample in a test tube add 2 cc. of 20% lead acetate solution, shake and let stand 30 minutes. Describe the precipitate as turbid, light, normal, heavy or very heavy. Polarization. If the lead precipitate is normal, add to 50 cc. of the sample 5 cc. of basic lead acetate solution. Shake and let stand 30 minutes; filter and polarize, preferably in a 200 mm. tube, correcting for dilution. If basic lead acetate gives only a turbidity, add to the sample, already treated with basic lead acetate, 10 cc. of alumina cream, shake, let stand 30 minutes, filter and polarize, correcting for dilution. (See pages 397-399.) In the case of malt vinegar, treat 100 cc. of the sample with 5 cc. of 10% phosphotungstic acid solution and filter. To 50 ec. of the filtrate add 5 cc. of the basic lead acetate solution, filter and polarize, correcting the reading obtained for dilution. (See also Leach: "Food Inspection and Analysis" (1911 Edition), pages 578 and 769). 456 TECHNICAL METHODS OF ANALYSIS Ash. Measure 25 cc. into a weighed platinum dish. Evapo- rate to dryness on the steam bath and heat to a char at low heat. Treat the charred portion with a few cc. of water and filter through a quantitative filter. Ignite the filter paper and carbon in the platinum dish at low heat. Add the filtrate and evaporate to dryness. Ignite gently and weigh. At no time allow the temper- ature to exceed a dull red. NOTE. Useful information may often be obtained by noting the odor given off by the solids during charring. Soluble and Insoluble Ash and Alkalinity. Add to the above ash 10-15 cc. of water, bring nearly to a boil and filter through a 9 cm. quantitative filter. Wash with successive portions of hot water until the filtrate measures about 60-75 cc. Dry and ignite the filter with the undissolved residue at low red heat in the platinum dish. Cool, weigh and calculate as insoluble ash. Subtract from the total ash to obtain the soluble ash. Cool the filtrate and titrate with 0.1 N HC1 and methyl orange. Express the alkalinity of the soluble ash as the number of cc. of 0.1 N HC1 per 100 cc. of sample. Soluble and Insoluble Phosphoric Acid. Determine the P20s in the water-soluble and water-insoluble portions of the ash by the official method for Fertilizers, (page 525), dissolving the water- insoluble portion in about 50 cc. of boiling HNOa (1 : 9). Express results as milligrams of P205 per 100 cc. of vinegar. Total Acids. Dilute 10 cc. of the sample with recently boiled and cooled water until it appears very slightly colored and titrate with 0.5 N NaOH and phenolphthalein. Calculate results to acetic acid. For most purposes it is customary to report as Total Acidity, Calculated as Acetic Acid. CALCULATION. 1 cc. of 0.5 N NaOH = 0.0300 gram acetic acid. NOTE. Instead of using 0.5 N NaOH to titrate 10 cc., the 10 cc. may be diluted in a graduated flask to 250 cc and 50 cc. of this solution titrated with 0.1 N NaOH- (1 cc. 0.1 N NaOH = 0.00600 gram acetic acid). Fixed Acids. Measure 10 cc. into a 200 cc. porcelain casserole. Evaporate just to dryness, add 5-10 cc. of water, and again evapo- rate. Repeat until at least five evaporations have taken place and no odor of acetic acid can be detected. Add about 200 cc. ANALYSIS OF FOODSTUFFS 457 of recently boiled, distilled water and titrate with 0.1 N NaOH and phenolphthalein. Express the result as malic acid. CALCULATION. 1 cc. 0.1 N NaOH = 0.0067 gram malic acid. Volatile Acids. Calculate the fixed acid as acetic and sub- tract from the total acid, also calculated as acetic. Report the difference as Volatile Acids, Calculated as Acetic Acid. Esters.* Dilute 200 cc. of the sample with 25 cc. of water and distill slowly into a graduated 200 cc. flask until nearly filled to the mark. Complete the volume and mix well. Exactly neutralize 50 cc. of the distillate to phenolphthalein with 0.1 N NaOH, and add a measured excess of 25-50 cc. of the 0.1 N NaOH over the amount required for neutralization. Either boil for one hour with a reflux condenser, or let stand overnight in a stoppered flask, and then heat with a tube condenser for one-half hour below the boiling point. Cool, and titrate the excess of NaOH with 0.1 N acid and phenolphthalein. Multiply the number of cc. of 0.1 N NaOH consumed in the saponification by 0.0088, thus obtaining the grams of esters, calculated as ethyl acetate. Mul- tiply this by 2 to bring to the basis of 100 cc. Also report the ester number, which is the milligrams of KOH required to saponify the esters in 1 gram of the sample. For this purpose 1 gram of vinegar may be considered equivalent to 1 cc. Formic Acid (Fincke Method). To 100 cc. of the sample add 0.4-0.5 gram of tartaric acid and distill with steam. Pass the outflowing stream through a boiling mixture of 15 grams of CaCOs and 100 cc. of water and keep this volume constant throughout the process. Keep the volume of the solution in the sample flask down to 30-40 cc. by heating with a small Tirrill flame. Collect 1000 cc. of distillate. Discard the distillate, disconnect the apparatus, filter the CaCOs mixture and wash with a little hot water. Make the filtrate slightly acid with HC1, add 10-15 cc. of mercuric chloride solution (10 grams of HgCb and 3 grams of NaCl to 100 cc.). Heat in a boiling water bath for two hours, filter on a weighed Gooch crucible, wash the precipitate thoroughly with cold water, then with a little alcohol and finally with ether. Dry in a boiling water oven for thirty 'minutes, cool in a desic- cator and weigh as mercurous chloride. Calculate to formic acid. * This is not a method of the Association of Official Agricultural Chemists. 458 TECHNICAL METHODS OF ANALYSIS CALCULATION. Hg 2 Cl 2 X 0.097 = formic acid. NOTES. (1) The determinations usually required on a sample of vinegar, are sp. gr., total solids, and total acidity calculated as acetic acid. (2) For other determinations than above described (alcohol precipitate, pentosans, tartaric acid, free mineral acid, dextrin, spices, coloring matters, preservatives, etc.) see J. Assoc. Official Agr. Chemists, Methods of Analysis (1916; page 253. (3) This method conforms to the requirements of the Massachusetts Board of Health's Regulations for Testing Vinegar. (4) The Massachusetts standards for Cider Vinegar (Chapter 145, Gen- eral Acts of the Legislature, 1918) are as follows: " It shall contain no added or artificial coloring and not less than 4 grams of acetic acid in 100 cc. " These are the same requirements as the Federal standard (Circular 19) which in addition states that it shall be laevo-rotatory; that not more than 50% of the solids shall be reducing sugars; that the ash shall not be less than 0.25 gram in 100 cc.; and that the water-soluble ash from 100 cc. shall contain not less than 10 mg. of ?2O5 and require not less than 30 cc. of 0.1 N acid to neutralize its alkalinity. (5) The above methods, unless otherwise indicated, are the tentative methods of the Association of Official Agricultural Chemists (J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 253). LEMON AND ORANGE EXTRACTS Specific Gravity. Determine the sp. gr. at 15.5 C. with a pycnometer, compared with H^O at the same temperature. Alcohol. Dilute 50 cc. of the extract, measured at 15.5 C., in a 200 cc. volumetric flask nearly to^the mark. Let stand until the oil separates to a clear layer on top (or if necessary, centrifugal- ize). Then make up to the mark at 15.5 C., using the lower meniscus of the oil. Pour the mixture into a dry Erlenmeyer flask containing 5 grams of light MgCOs, stopper, shake well and filter quickly through a large dry folded filter. Determine the alcohol in the filtrate as follows: (a) BY WEIGHT. Place 150 cc. of the filtrate, measured at 15.5 C., in a 300-500 cc. distilling flask. Attach the flask to a vertical condenser and distill, catching the distillate in a 100 cc. graduated flask. Stop the distillate when nearly 100 cc. have been distilled. Make up to the mark with water at 15.5 C. and mix thoroughly. Determine the sp. gr. accurately with a pycnometer. Ascertain from standard alcohol tables the percentage of alcohol by weight corresponding to the sp. gr. of the distillate. Multiply ANALYSIS OF FOODSTUFFS 459 this figure by the weight of the distillate (its volume Xsp. gr.) and divide by the weight of the sample in the aliquot taken, i.e., 37. 5 X its sp. gr. (6) BY VOLUME. From the sp. gr. of the distillate, obtained as above, ascertain the per cent of alcohol by volume from the alcohol tables. Multiply this figure by the volume of distillate and divide by the volume of the original sample, i.e., 37.5. NOTES. (1) References to standard alcohol tables are given on page 454. (2) In case alcohol tables are used which are based on a gravity at some other temperature than 15.5 C., the sp. gr. of the original sample and of the alcohol distillate must both be determined at the same temperature as that on which the tables are based. Glycerol. Evaporate 100 cc. of the extract * in a porcelain dish on a water bath to a volume of about 10 cc., treat the residue with about 5 grams of fine sand and 3-4 cc. of milk of lime (containing about 15% of CaO) for each gram of extract present, and evaporate almost to dryness. Treat the moist residue with 50 cc. of 90% alcohol by volume, remove the sub- stance adhering to the sides of the dish with a spatula, and rub the whole mass to a paste. Heat the mixture on the water bath, with constant stirring, to incipient boiling and decant the liquid through a filter into a small flask. Wash the residue repeatedly by decantation with 10 cc. portions of hot 90% alcohol until the filtrate amounts to about 150 cc. Evaporate the filtrate to a syrupy consistency in a porcelain dish, on a hot, but not boiling, water bath; transfer the residue to a small glass stoppered grad- uated cylinder with 20 cc. of absolute alcohol and add three por- tions of 10 cc. each of absolute ether, thoroughly shaking after each addition. Let stand until clear, then pour off through a filter and wash the cylinder and filter with a mixture of 1 part of absolute alcohol and 1.5 parts of absolute ether, pouring the wash liquor also through the filter. Evaporate the filtrate to a syrupy consistency, dry for one hour at the temperature of boiling water, weigh, ignite and weigh again. The loss on ignition gives the weight of glycerol. * If 100 cc. contain more than 0.4 gram of glycerol, use a smaller amount. 460 TECHNICAL METHODS OF ANALYSIS Lemon and Orange Oils. (a) BY POLARIZATION (MITCHELL). Polarize the extract at 20 C. without dilution in a 200 mm. tube and divide the reading in degrees Ventzke by 3.2, in the case of lemon extract, and by 5.2, in the case of orange extract. In the absence of other optically active substances, the result will be the percentage of oil by volume. A small amount of cane sugar is occasionally present ; if so, determine it as directed under Sucrose and correct the reading accordingly. (6) BY PRECIPITATION (MITCHELL). Pipette 20 cc. of the extract into a Babcock milk bottle; add 1 cc. of dil. HC1 (1 : 1); then add 25-28 cc. of water previously warmed to 60 C,; mix and let stand in water at 60 for five minutes; whirl in a centrifuge for five minutes; fill with warm water to bring the oil into the graduated neck of the flask; repeat the whirling for two minutes; stand the flask in water at 60 C. for a few minutes and read the per cent of oil by volume. In case oil of lemon is present in amounts over 2%, add to the percentage of oil found 0.4%, to cor- rect for the oil retained in solution. If less than 2% and more than 1% is present, add 0.3% for correction. To obtain the per cent by weight from the per cent by volume, as found by either of the above methods, multiply the volume percentage by 0.86* and divide the result by the sp. gr. of the original extract. Total Aldehydes (Chace Method). (a) REAGENTS. (1) Aldehyde-free Alcohol. Allow alcohol (95% by volume), containing 5 grams of m-phenylenediamine hydrochloride per liter, to stand for twenty-four hours with frequent shaking. Boil under a reflux condenser for at least eight hours (longer if neces- sary), let stand overnight and distill, rejecting the first 10% and the last 5% which come over. Keep in a dark, cool place in well-filled bottles. 25 cc. of this alcohol on standing for twenty minutes in a cooling bath at 14-16 C. with 20 cc. of fuchsin solution should develop only a faint pink coloration. If a stronger color is developed, treat it again with m-phenylenediamine hydro- chloride. (2) Sulfite-fuchsin Solution. Dissolve 0.5 gram of fuchsin in 250 cc. of water, add an aqueous solution of 862 containing 16 grams of the gas and let stand until colorless or nearly so. * This is for lemon oil; use 0.85 for orange. ANALYSIS OF FOODSTUFFS 461 Then make up to 1 liter with distilled water. This solution should stand twelve hours before using. It should be kept in a refrigerator and discarded after three days. (3) Standard Citral Solution. Use 0.5 or 1.0 mg. of c. P. citral per cc. in 50% by volume of aldehyde-free alcohol. (6) MANIPULATION. Weigh in a stoppered weighing bottle approximately 25 grams of extract, transfer to a 50 cc. volumetric flask, and make up to the mark at room temperature with alde- hyde-free alcohol. Measure at room temperature 2 cc. of this solution into a comparison tube. Add 25 cc. of the aldehyde- free alcohol (previously cooled to 14-16 C.), then 20 cc. of the sulfite-fuchsine solution (also cooled) and finally make up to the 50 cc. mark with more aldehyde-free alcohol. Mix thoroughly, stopper, and keep at 14-16 C. for fifteen minutes. Prepare a standard for comparison at the same time and in the same manner, using 2 cc. of the standard citral solution, and compare the colors developed. Calculate the amount of citral present. Repeat the determination, using a quantity sufficient to give the sample approximately the strength of the standard. From this result calculate the amount of citral in the sample. If the comparisons are made with Nessler tubes, standards containing 1, 1.5, 2, 2.5, 3, 3.5, and 4 mg., respectively, of citral should be prepared and the trial comparisons made against these, the final comparison being made with standards between 1.5 and 2.5 mg. with 0.25 mg. increments. NOTE. It is absolutely essential to keep the reagents and the comparison tubes at the required temperature. Comparisons should be made within one minute after removing the tubes from the bath. Give the samples and standards identical treatment. Citral (Hiltner Method). (a) REAGENTS. (1) m-Phenylenediamine Hydrochloride Solution. Prepare a 1% solution of m-phenylenediamine hydrochloride in 95% ethyl alcohol. Decolorize if necessary by shaking with fuller's earth and filter through a double filter. The solution should be bright and clear, free from suspended matter, and practically colorless. It is well to prepare only enough solution for the day's work, as it darkens on standing. (2) Alcohol. For the analysis of lemon extracts, 90-95% alcohol should be used, but for terpeneless extracts alcohol of 462 TECHNICAL METHODS OF ANALYSIS 40-50% strength is sufficient. Filter to remove any suspended matter. The alcohol need not be purified from aldehyde. If not practically colorless, render slightly alkaline with NaOH and distill. (6) MANIPULATION. All of the operations may be carried on at room temperature. Weigh 25 grams of the extract into a 50 cc. graduated flask and make up to the mark with alcohol. Stopper the flask and mix the contents thoroughly. Pipette 2 cc. of this solution into a Nessler tube, add 10 cc. of ra-phenylene- diamine hydrochloride reagent and complete the volume to 50 cc. with alcohol. Compare at once the color with that of the standard prepared at the same time, using 2 cc. of standard citral solution and 10 cc. of the m-phenylenediamine reagent, and diluting to volume with alcohol. From the result of this first determination calculate the amount of standard citral solution that should be used in order to give approximately the same citral strength of the sample under examination, then repeat the comparison against the new standard. Total Solids. Evaporate nearly to dryness 10 cc. of the sam- ple in a flat platinum dish on a water bath at a low temperature. Heat the residue for two and one-half hours in a drying oven at the temperature of boiling water, cool and weigh. Ash. Burn the residue from 10 cc. of the extract to a white ash at the lowest possible heat, cool and weigh. Sucrose. Neutralize the normal weight of the extract, evap- orate to dryness, wash several times with ether, dissolve in water and determine sucrose in the usual manner by means of a polari- scope (page 400), or by copper reduction (page 402). Methyl Alcohol (Riche and Bardy). Test the distillate ob- tained in the determination of Alcohol above for methyl alcohol as follows: Place 10 cc. in a small flask with 15 grams of iodine and 2 grams of red phosphorus. Keep in ice water for ten to fifteen minutes until action has ceased. Distill off on the water bath the methyl and ethyl iodides formed into about 30 cc. of water. Wash with dilute NaOH to eliminate free iodine, separate the heavy oily liquid which settles and transfer to a flask containing 5 cc. of aniline. If the action is too violent, place the flask in cold water; if too slow, warm the flask. After one hour, boil the product with water and add about 20 cc. of 15% NaOH solution. ANALYSIS OF FOODSTUFFS 463 When the bases rise to the top as an oily layer, fill the flask up to the neck with water and draw them off with a pipette. Oxidize 1 cc. of the oily liquid by adding 10 grams of a mixture of 100 parts of clean sand, 2 of common salt and 3 of cupric nitrate. Mix thoroughly, transfer to a glass tube and heat to 90 C. for eight to ten hours. Exhaust the product with warm alcohol, filter and make up to 100 cc. with alcohol. In the presence of ethyl alcohol free from methyl alcohol the liquid has a red tint, but in the presence of 1% of methyl alcohol it has a distinct violet shade; with 2.5% the shade is very distinct, and still more so with 5%. To detect more minute quantities of methyl alcohol, dilute 5 cc. of the colored liquid to 100 cc. with water, and dilute 5 cc. of this again to 400 cc. Heat the liquid thus obtained in a porcelain dish and immerse in it a fragment of white merino yarn (free from sulfur) for thirty minutes. If the ethyl alcohol is pure, the wool will remain white, but if methyl alcohol is present, it will become violet, and the depth of the tint will give a fairly approximate indication of the proportion of methyl alcohol present. NOTE. The presence of methyl alcohol and its approximate amount may also be ascertained by determining the refraction at 20 C. with the im- mersion refractometer. (See J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 247.) Detection of Coloring Matter. (a) LEMON AND ORANGE PEEL COLOR (ALBRECH METHOD). Place a few cc. of the extract in a test-tube and add slowly 34 volumes of cone. HC1. Place a few cc. of the extract in a second tube and add several drops of cone. NH4OH. If the color is due to the presence of lemon or orange peel only, it is materially deepened in both cases. (6) TURMERIC. Evaporate 25-50 cc. of the extract upon a small piece of filter paper, dry at low temperature, and moisten with a weak solution of boric acid containing a small amount of HC1. Upon drying a second time a cherry red color, changing to green when spotted with NELiOH, develops in the presence of turmeric. REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 261. 464 TECHNICAL METHODS OF ANALYSIS ORANGE OIL AND LEMON OIL Specific Gravity. Determine the sp. gr. at 15.5 C. with a Westphal balance. Refractive Index. Determine the refractive index at 20 C. with the Abbe refractometer. Optical Rotation. Determine the rotation at 20 C. with any standard instrument, using a 50 mm. water-jacketed tube and sodium light. The results should be stated in angular degrees on a 100 mm. basis. If instruments having the sugar scale are used, the reading on orange oil is above the range of the scale, but read- ings may be obtained by the use of standard laevo-rotatory quartz plates or by the use of a 25 mm. tube. The true rotation cannot be obtained by diluting the oil with alcohol and correcting pro- portionately. Citral (Hiltner Method). Proceed as directed under Lemon and Orange Extracts (page 461), weighing 2 grams of lemon oil or 8 grams of orange oil into a 100 cc. volumetric flask, diluting to 100 cc. with 95% alcohol by volume and using 2 cc. of this solution for the comparison. Total Aldehydes. Weigh a small quantity of the sample into a small stoppered flask and dilute with aldehyde-free alcohol in the proportion of 2 grams of lemon oil or 4 grams of orange oil to 100 cc. of solution. Determine the total aldehydes as described on page 460 for Orange and Lemon Extracts (Chace Method), expressing the result as citral. REFERENCE. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916), page 264. OIL OF PEPPERMINT Specific Gravity. Determine the sp. gr. at 20 C. with a Westphal balance. Refractive Index. Determine the refractive index with the Abbe refractometer at 25 C. Specific Rotation, [a]. Determine the angle of refraction by means of the polariscope. Have the temperature exactly 20 C. ANALYSIS OF FOODSTUFFS 465 and use a sodium light.* Calculate the specific rotation, [a]??, from the following formula : 20 _^ 100 -Z X sp.gr.at20 d where a = angle of refraction, and L = length of tube in mm. Dimethyl Sulfide. Distill 1 cc. from 25 cc. of oil and pour the distillate on an aqueous solution of HgCk. A white film forming after a short time at the zone of contact indicates dimethyl sulfide, which is found in non-rectified oils. Menthyl Esters. Saponify 8-10 grams of oil by boiling for one hour with 25 cc. of 0.5 N alcoholic KOH in a flask provided with a reflux condenser. After cooling, titrate with 0.5 N HC1 and phenolphthalein. Multiply the number of cc. of KOH consumed in the saponi- fication by 9.9 and divide the product by tfee weight of oil taken. This gives percentage of esters as menthyl acetate. Total Menthol. Wash the residual oil from the ester deter- mination with water a number of times, transfer to an acetyliza- tion flask (preferably a flask with a ground glass tube condenser), add .10 cc. of acetic anhydride and 1 gram of anhydrous sodium acetate, and boil gently for one hour. Cool, wash the acetylized oil with water, and then with very dilute NaOH solution, until the mixture is slightly alkaline to phenolphthalein. Separate the oil from any water in a separatory funnel, pour into a small flask, add a few pieces of fused CaCk, let stand for several hours, and filter. Weigh out 3-6 grams of this dry oil in an Erlenmeyer flask and saponify by boiling for one hour with 50 cc. of 0.5 N alcoholic KOH. After cooling, titrate the excess alkali with 0.5 N HC1. Each cc. of 0.5 N alcoholic KOH corresponds to 0.0781 gram of menthol, or 0.0991 gram of menthyl acetate. In order, there- fore, to obtain the percentage of menthol in the original oil (not acetylized but freed from ester), it is necessary to deduct 0.0210 * In case the reading is taken with the Ventzke Saccharimeter, using gas or other white light, calculate the saccharimeter reading to angular reading by multiplying the former by 0.3468. 466 TECHNICAL METHODS OF ANALYSIS gram (the difference between 0.0781 and 0.0991) for every cc. of 0.5 N alkali consumed. The following formula, therefore, gives the total menthol content (free and ester) : 7.81a P = S-0.0210a' where P = total menthol content; S = grams of acetylized oil ; and a = cc. of 0.5 N alkali consumed. By subtracting the amount of ester as first determined, the amount of free menthol is obtained. NOTE. The chemistry of the foregoing method is as follows: Menthol is a saturated secondary alcohol of the following structural formula. H 3 C CH 3 CH I CH HiC\. xCHa CH I CH 3 The ester has the formula: H 3 C CH 8 Y CH CH H 2 Cf // \CHO-C 2 H 3 CH I CHi ANALYSIS OF FOODSTUFFS 467 In the ester determination the acetate radical is removed and menthol formed by saponification. The menthol is then all transformed to the ester by the action of acetic anhydride and sodium acetate. This compound is then saponified back to menthol, and the amount of alkali consumed gives the amount of menthol by the above calculation. REFERENCES. United States Pharmacopoeia; United States Dispensa- tory; Gildemeister and Hoffmann: " The Volatile Oils." PEPPERMINT, SPEARMINT AND WINTERGREEN EXTRACTS Alcohol. Since these extracts usually contain only about 1% of oil, the alcohol can in most cases be calculated from the sp.gr. of the extract. If, however, the extract is high in solids, deter- mine the alcohol as follows : Add 25 cc. of the extract to 75 cc. of a saturated solution of NaCl in a separatory funnel and extract twice with 50 cc. por- tions of petroleum ether (b. p. 40-60 C.). Collect the petroleum ether extract in a second separatory funnel and wash twice with two separate 25 cc. portions of saturated brine. Combine the original salt solution with the washings, add a little powdered pumice, and distill into a 100 cc. volumetric flask. When almost 100 cc. have been distilled, make up to the mark at a definite temperature and determine the alcohol from the sp. gr. as de- scribed on page 458. NOTE. A 50% solution of CaCl 2 may be used in place of saturated brine. Oil (Modified Howard Method). Pipette 10 cc. of the extract into a Babcock milk bottle, add 1 cc. of CS2, mix thoroughly, and then add 25 cc. of cold water and 1 cc. of cone. HC1. Close the mouth of the bottle and shake vigorously. Centrifugalize for six minutes and remove all but 3-^4 cc. of the supernatant liquid, which should be practically clear, by aspirating through a glass tube of small bore. Connect the stem of the bottle with a filter pump, immerse for three minutes in water kept at about 70 C., remove from the bath every fifteen seconds and shake vigorously. Continue in the same manner for forty-five seconds, using a boiling water bath. Remove from the bath and shake while cooling. Disconnect from the suction and fill the bottle to the neck with saturated NaCl solution at room temperature. Cen- trifugalize for two minutes and read the volume of the separated 468 TECHNICAL METHODS OF ANALYSIS oil from the top of the meniscus. Multiply the reading by 2 to obtain the per cent of oil by volume. In the case of wintergreen use as a floating medium a mixture of 1 volume of cone. H2SO4 and 3 volumes of saturated Na2S04 solution. Methyl Salicylate in Wintergreen Extracts (Modified Hort- vet and West Method). Mix 10 cc. of the extract with 10 cc. of 10% KOH solution. Heat on the steam bath until the volume is reduced about one-half. Add a distinct excess of HC1 (1 : 1), cool and extract with 3 portions of ether, 40, 30 and 20 cc., respect- ively. Filter the extract through a dry filter into a weighed dish. Wash the paper with 10 cc. of ether and allow the nitrate and washings to evaporate spontaneously. Dry in a desiccator con- taining H2&04 and weigh. Multiply the weight of salicylic acid thus found by 9.33 to obtain the per cent by volume of winter- green oil in the sample. (Sp. gr. of wintergreen oil = 1.178.) REFERENCE. The above are tentative methods of the Association of Official Agricultural Chemists as published in its Journal, Methods of Analysis (1916), page 268. VANILLA EXTRACT Specific Gravity. Determine at 15.5 C. by means of a West- phal balance. Alcohol. Weigh into a distilling flask 20-25 grams of the sample and dilute with 100 cc. of water. Distill into a 100 cc. volumetric flask through a vertical condenser, stopping the dis- tillation just short of 100 cc. Make up to volume at 15.5 C. and determine the sp. gr. of this distillate accurately with a pyc- nometer at the same temperature. Determine the per cent of alcohol by weight and by volume as described on page 458. Determination and Identification of Vanillin and Coumarin. The determination of vanillin, coumarin, normal lead number and residual color in the filtrate are all to be made on one weighed portion, using 50 grams of the sample. For the determination of coumarin and vanillin employ the modified Hess and Prescott method as follows: Weigh 50 grams of the extract directly into a tared 250 cc. beaker with marks showing volumes of 80 and 50 cc. Dilute to ANALYSIS OF FOODSTUFFS 469 80 cc. and evaporate to 50 cc. in a water bath kept at 70 C. Dilute again to 80 cc. with water and evaporate to 50 cc. Trans- fer to a 100 cc. volumetric flask, rinsing the beaker with hot water, add 25 cc. of 8% lead acetate solution, make up to the mark with water, shake, and let stand eighteen hours (overnight) at 37- 40 C. This can best be obtained by means of an incubating oven. Decant on a small dry filter, pipette off 50 cc. of filtrate, and extract, shaking four times in a separatory funnel, using 50 cc. of ether first and then 15 cc. three times. Wash the combined ether solutions four or five times with 2% NHs solution, using 10 cc. the first time and 5 cc. thereafter. Slightly acidulate the combined ammoniacal solutions with dil. HC1, cool and extract in a sepa- ratory funnel with 4 portions of ether, using about 40 cc. in all. Evaporate the ethereal solutions at room temperature, dry over cone. H2S04 (more quickly accomplished in a vacuum desiccator) and weigh. If the. residue is considerably discolored or gummy, re-extract in the dry state with boiling petroleum ether (b. p. 40 C. or below) not less than fifteen times; evaporate the solvent, dry and weigh. The residue should now be white, crystalline vanillin with a m. p. of approximately 80 C. A small amount of this residue dissolved in 2 drops of cone. HC1, upon the addition of a crystal of resorcinol, should develop a pink color. Evaporate the original ether extract of the sample, after removal of the vanillin with NHiOH, at room temperature, and dry over cone. H2SO4. The residue, if pure coumarin, should melt at approximately 67 C., and should respond to Leach's test for coumarin as follows: A small portion of the residue dissolved in not more than 0.5 cc. of hot water should develop a brown pre- cipitate upon the addition of a few drops of 0.1 N iodine solution. This precipitate finally gathers in green flocks, leaving a clear brown solution. The reaction is especially marked if the reagent is applied with a glass rod to a few drops of the solution on a white plate or tile. NOTES. (1) For extracting the coumarin and the vanillin, use ether washed with water at least twice to remove alcohol. (2) The method is not applicable to cone, preparations in which the amount of vanillin and coumarin present exceeds the quantity dissolved by 100 cc. of H 2 O at 20 C. In such cases use a smaller amount of the sample and dilute to 50 cc. 470 TECHNICAL METHODS OF ANALYSIS Normal Lead Number. To an aliquot of 10 cc. of the filtrate from the lead acetate precipitate obtained above, add 25 cc. of water, 0.5-1.0 cc. of cone. H2SO4 and 100 cc. of 95% alcohol. Let stand overnight, filter on a Gooch crucible, wash with 95% alcohol, dry at a moderate heat and ignite at low redness for three minutes, taking care to avoid the reducing flame. (It is better to place the Gooch crucible inside a platinum crucible during the ignition.) Cool in a desiccator and weigh. Calculate the normal lead number from the following formula : In the above formula P = normal lead number (grams of metallic Pb in the precipitate from 100 cc. of the sample); S = grams of PbS(>4 corresponding to 25 cc. of the standard lead acetate solution as determined in a blank analysis made on water containing 4-5 drops of glacial acetic acid; and W = grams of PbSC>4 in 10 cc. of the filtrate from the lead acetate precipitate as just described. Residual Color in Filtrate after Precipitation with Lead Acetate. Determine the color value in terms of red and yellow of a por- tion of the filtrate from the lead acetate precipitate obtained in the determination of vanillin and coumarin, using a 1-inch Lovi- bond cell. Multiply the reading by 2, thus reducing the results to the basis of the original extract. In case the actual read- ing of the solution is greater than 5 red and 15 yellow, as may happen if the extract is highly colored with caramel, use the 0.5 or 0.25 inch cell and multiply the readings by 4 or by 8, respect- ively. Divide the figures for red and yellow, respectively, by the cor- responding figures obtained by measurement of the original extract and multiply the quotients by 100, thus obtaining the per- centages of the two colors remaining in the lead acetate filtrate. Calculate also the ratio of red to yellow in both extract and lead filtrate. NOTE. Determine the color value of the original extract as follows: Pipette 2 cc. into a 50 cc. graduated flask and make up to the mark with a mixture of equal parts 95% alcohol by volume and water. Determine the color value of this diluted extract in terms of red and yellow by means of the ANALYSIS OF FOODSTUFFS 471 Lovibond tintometer, using the 1-inch cell. To obtain the color value of the original extract multiply the figures for each color by 25. Per Cent of Color Insoluble in Amyl Alcohol (Marsh Test). Pipette 25 cc. of the extract into a porcelain dish and Qvaporate just to dryness on the steam bath. Transfer to a 50 cc. flask by means of 95% alcohol by volume and water, using a total of 26.3 cc. of 95% alcohol. Dilute to the mark with water. Trans- fer 25 cc. of this solution to a separatory funnel. Add 25 cc. of the Marsh reagent and shake lightly to avoid emulsification. Let the layers separate and repeat the shaking and standing twice again. After the layers have separated clearly, run off the lower aqueous layer into a 25 cc. cylinder and make up the volume with 50% (by volume) alcohol. Compare in a colorimeter with the remaining 25 cc. portion which has not been extracted and express the results as percentage of color insoluble in amyl alcohol. Marsh Reagent. Mix 100 cc. of amyl alcohol, 3 cc. of sirupy phosphoric acid and 3 cc. of water. Shake immediately before using. If the reagent becomes colored on standing, the amyl alcohol should be redistilled over 5% phosphoric acid. Total Solids. Determine the total solids on 10 grams of the sample by evaporating in a dish containing a weighed amount of quartz sand as described on page 410. Ash. Evaporate 10 grams of the extract, char and ignite until free of carbon at the lowest possible heat, not exceeding dull redness. Sucrose. Place 200 cc. of the extract in a porcelain dish, exactly neutralize to litmus paper with approximately N NaOH and evaporate to about a quarter of the original volume. Trans- fer to a 200 cc. volumetric flask, add sufficient normal lead acetate to clarify and dilute to the mark with water. If necessary, add also 1-2 cc. of alumina cream before dilution. Shake and filter through a folded filter. Polarize the filtrate at 20 C. in a 200 mm. tube. Free about half of this filtrate from Pb by treating with dry K2C2O4 added a little at a time. Shake after each addition and avoid an excess. Filter through a dry paper, pipette 50 cc. of the lead-free filtrate into a 100 cc. volumetric flask and add 25 cc. of water. Then add little by little, while rotating the flask, 5 cc. of cone. HC1. Heat the flask, after mixing, in a water bath at 70 C. The temperature of the solution in the flask should 472 TECHNICAL METHODS OF ANALYSIS reach 67-69 C. in two and one-half to three minutes. Maintain a temperature of approximately 69 C. during seven to seven and one-half minutes, making the total time of heating ten minutes. Remove the flask, cool the contents rapidly to 20 C., and dilute to 100 cc. Polarize this solution at 20 C., and multiply the invert reading by 2 to correct for dilution. (If it is necessary to polarize at a temperature other than 20 C., both direct and invert polarizations and dilution to volume should be made at exactly the same temperature.) Calculate the sucrose by the following formula : _ 100 (A -B] 26 /\ T 142.66-4; 2 where S = per cent of sucrose; A direct reading; B = invert reading; and T= temperature at which readings are made. NOTE. In calculatL-g the percentage of sucrose, the relation of the amount of sample to the normal weight (26.048 grams per 100 cc.) must be taken into consideration. This is corrected for in the above formula by the second fraction. Vanilla Resins (Qualitative Tests). Evaporate 50 cc. of the extract in a glass dish on a water bath until the alcohol is removed and make up to about the original volume with hot water. If alkali has not been used in the manufacture, the resins will appear as a flocculent red to brown residue. Acidify with acetic acid, let stand a short time, collect the resins on a filter, wash with water and reserve the filtrate for further tests. Place a portion of the filter with the attached resins in a few cc. of dil. KOH solution. The resins are dissolved, giving a deep red solution. Acidify and the resins are reprecipitated. Dissolve a portion of the resins in alcohol. To one fraction add a few drops of FeCls solution; no striking coloration should be produced. To another portion add HC1 ; little change in color should result. To a portion of the filtrate obtained above, add a few drops of lead subacetate solution. A very bulky precipitate should ANALYSIS OF FOODSTUFFS 473 result and the filtrate from this precipitate should be practically colorless. Test another portion of the filtrate from the resin for tannin with a solution of gelatin. Tannin should be present in varying but small quantities and should not be present in great excess. Glycerol. Heat to boiling in a flask 100 cc.* of the extract and treat with successive small portions of milk of lime (containing about 15% of CaO) until it becomes first darker and then lighter in color. When cool add 200 cc. of 95% alcohol, let the precipitate subside, filter, and wash with 95% alcohol. Evaporate the fil- trate in a porcelain dish on the water bath to about 10 cc. and proceed as in the determination of Glycerol in Lemon Extract, page 459. NOTES. (1) This method embodies the tentative and official methods and recommendations of the Association of Official Agricultural Chemists. (2) The limits of composition of pure standard vanilla extracts as set forth in Bureau of Chemistry Bulletin 163, page 89 are as follows: (a) Percentage of vanillin should be not less than 0.10 nor more than 0.35. (6) Normal lead number should be not less than 0.40 nor more than 0.80. (c) The percentage of color insoluble in acidified amyl alcohol (Marsh's reagent) should be not more than 35 and will seldom exceed 25%. * The amount of glycerol in the sample taken for analysis should be be- tween 0.10 and 0.40 gram. CHAPTER XI MISCELLANEOUS ANALYSES LEATHER Preparation of Sample. Reduce the sample for analysis to as fine a state of division as practicable, either by cutting or grinding. Moisture. Dry 10 grams for sixteen hours between 95 and 100 C. Cool in a desiccator and keep the dish or weighing bottle tightly covered while weighing. Report the loss in weight as moisture. Fat and Oil. Extract completely 5-10 grams of the air-dry sample in a Soxhlet apparatus with petroleum ether* boiling below 80 C. Evaporate off the ether and dry the extract to approx- imately constant weight. Or, if preferred, extract 30 grams of leather as described above. In the latter case, the extracted leather, when freed of solvent, may be used for the determination of water-soluble material. Ash. Incinerate 10-15 grams in a tared dish at a dull red heat until carbon is consumed. If it is d fficult to burn off all the car- bon, treat the ash with hot water, filter through an ashless filter and ignite the filter and residue in the dish. Then add the filtrate, evaporate to dryness and ignite. Cool in a desiccator and weigh the mineral matter. Water-soluble Matter. Digest 30 grams of fat-free leather with water in a percolator overnight, then extract with water at 50 C. for three hours, using successive small portions. The total volume of solution should be 2 liters. Pipette out 100 cc. into a weighed flat-bottom dish of a diameter of 2.75-3 inches. Evapo- rate to dryness and dry to constant weight at 100-105 C. From this weight calculate the per cent of water-soluble material. * The extraction will require 8-12 hours, depending upon the amount of fat. If convenient, let the extraction run overnight. Place a layer of fat- free cotton above and below the ground leather in the extractor. 474 MISCELLANEOUS ANALYSES 475 Glucose. Place 200 cc. of the water extract prepared above in a 500 cc. flask, add 25 cc. of a saturated solution of normal lead acetate ; shake frequently five to ten minutes, and filter through a dry filter. Keep funnels and beakers covered to prevent evapo- ration. Add to the filtrate an excess of solid K2C2O4. Mix frequently for fifteen minutes and filter, returning the filtrate until it runs through clear. Pipette 150 cc. of this clear filtrate into a 500 cc. Erlenmeyer flask. Add 5 cc. of cone. HC1 and boil under a reflux condenser for two hours. Cool, add a small piece of litmus paper and neutralize with anhydrous Na 2 CO3. Transfer to a 200 cc. graduated flask and dilute to volume. Filter through a dry double filter, rejecting the first filtrate. The final filtrate must be clear. Determine the dextrose immediately in 50 cc. of this solution, using the Munson and Walker method as described under Total Reducing Sugars on page 403. Calculate the results to the original leather and report as " glucose, calculated as dextrose." In making the determination place 25 cc. of the CiiSO4 solution and 25 cc. of the alkaline tartrate solution in a 400 cc. beaker. Add 50 cc. of the prepared leather solution, heat to boiling in exactly four minutes, and boil for two minutes. The burner should be adjusted by a preliminary trial so that these conditions are fulfilled. Filter immediately without diluting through a weighed Gooch crucible containing an asbestos mat. Wash thor- oughly with hot water, then with alcohol, and finally with ether. Dry for one-half hour at the temperature of boiling water and weigh as Cu2O. NOTE. If the above conditions of dilution, etc., are all fulfilled, 50 cc. of the clarified and neutralized solution will correspond to 0.5 gram of the orig- inal leather (assuming exactly 30 grams taken for the water-soluble deter- mination) . Nitrogen. Determine N on 0.7 gram of leather by the Gun- ning modification of the Kjeldahl method as described on page 65. Multiply the nitrogen by 5.62 to obtain the amount of " hide substance." Chromium in Chrome Leather. For this determination see page 478. Free Mineral Acid. Most leathers when moistened are acid to litmus but this, unless extremely marked, is no evidence of free. 476 TECHNICAL METHODS OF ANALYSIS mineral acid. A marked acid reaction, however, especially in the presence of much sulfate, is suspicious. On the other hand, even if the leather contains free H2S04, the ash will usually be alkaline, since acid is driven off on ignition and a portion of the sulfates is reduced to sulfides, and even to carbonates or free bases, especially in the case of lime salts. There is no method which will give with certainty a true figure for free H2S04. The two methods which are probably the best are described below: (1) JEAN METHOD. Dry 10 grams or more of finely divided leather at not over 95 C. (drying in a vacuum desiccator is prefer- able) and extract with absolute alcohol in a Soxhlet extractor, placing about 0.5 gram of Na^COs powder in the flask to combine with any acid dissolved by the alcohol. Extract until the alcohol comes over colorless. Distill off the alcohol, transfer to a platinum dish and ignite to a char. Leach out with hot water, filter, and determine the total SOs in the solution by acidifying first with HC1 and then adding to the boiling solution an excess of BaC^ solution. Filter and weigh the BaSO 4 in the usual way. Calcu- late to H 2 SO 4 . CALCULATION. BaSO 4 X 0.4202 = H 2 SO 4 . NOTE. The alcohol used must be absolute and freshly prepared by distilling from quick-lime. The method is based on the fact that H^SCX is soluble in absolute alcohol, but sulfates are not. For this reason also the leather must be dried, but extreme drying, and especially the use of a high temperature, must be avoided, as this will cause the acid to attack and com- bine with organic constituents of the leather. If the above conditions are complied with, a positive result may be taken as proof of the presence of free acid; but a negative result is not conclusive evidence that acid is not present, even in injurious quantities, since it has been proved that even a long extraction may fail to remove all the acid from the leather. (2) PROCTOR AND SEARLE METHOD. Moisten 2-3 grams of leather in a platinum dish with exactly 25 cc. of accurately stand- ardized 0.1 N Na2COs solution; evaporate to dryness and char at a gentle heat until thoroughly carbonized. This drives off practically all organic sulfur, while the reduction of sulfates is very slight. Pul- verize the carbonaceous mass with a glass rod and leach with boiling water, filtering through a small quantitative filter paper. Dry the filter paper and return it to the mass in the dish, and ignite the whole until all or nearly all carbon has disappeared. Cool MISCELLANEOUS ANALYSES 477 and treat the ash with 25 cc. of 0.1 N HC1, accurately standard- ized against the 0.1 N Na2COs. Wash the whole into the beaker containing the filtrate of the charred mass and add methyl orange. If the liquid now shows an acid reaction, titrate with 0.1 N alkali and calculate the titration to KkSCU (unless it is known that the acid is HC1). If the reaction is alkaline, H^SCU is absent, and no notice need be taken of the alkalinity. NOTE. The two above methods for free acid are chiafly of value in com- paring different leathers, although if carefully carried out they give results which are reasonably accurate. Probably the most accurate method at pres- ent known is that of Wuensch, but it is a troublesome and time-consuming procedure. It is described in H. R. Proctor: " Leather Industries Laboratory Book of Analytical and Experimental Methods/' 1908 edition, page 371. REFERENCE. The above procedures, with the exception of those for free mineral acid, are essentially the methods of the American Leather Chemists' Association. CHROMIUM IN CHROME SALTS AND LEATHERS General. The most accurate method for the determination of chromium in chrome salts is to oxidize to the chromate condition and then determine the latter either gravimetrically, by precip- itation as PbCrC>4, or volumetrically. Chrome Liquors (Chrome Tanning Liquors, Chromium Ace- tate, etc.). (a) ONE BATH CHROME LIQUORS (USED IN TAN- NING).* Dilute a weighed quantity of the liquor with water to a definite volume, so that the dilution contains from 0.15 to 0.25% of Cr 2 O 3 . Pipette 10 cc. of this solution into a 300 cc. Erlenmeyer flask and add about 50 cc. of water and 2 grams of Na2(>2. Boil gently one-half hour, adding water if necessary to keep the volume from falling below about 15 cc. Cool, neutralize with cone. HC1 and add 5 cc. excess. Again cool and add 10 cc. of 10% KI solution. After one minute run in from a burette 0.1 N thiosulfate until the iodine color nearly disappears, then add a few cc. of starch solution and titrate to the disappearance of the blue. Calculate to C^Os. CALCULATION. 1 cc. of 0.1 N thiosulfate = 0.002533 gram Cr 2 O 3 . * J. Am. Leather Chem. Assoc. 14, page 667 (1919). 478 TECHNICAL METHODS OF ANALYSIS (6) CHROMIUM ACETATE SOLUTIONS (USED IN MORDANTING). Weigh out about 10 cc. accurately and dilute to 100 cc. in a volumetric flask. Mix thoroughly and pipette 10 cc. into a 300 cc. Erlenmeyer flask. Then proceed as above. (c) ALTERNATIVE METHOD. If the material is free from sul- f ates, the C^Oa may be determined gravimetrically as follows : Oxidize with peroxide as above. Filter out any insoluble matter. Heat nearly but not quite to boiling. Make acid with acetic acid and add an excess of lead acetate solution containing a few drops of acetic acid. Let stand on the steam bath until clear, but avoid long standing on account of the danger of forming basic compounds. Filter through a weighed Gooch crucible, wash with hot water, and dry at 105 C. Set the Gooch crucible inside of a larger platinum crucible, ignite to very dull redness, cool and weigh as PbCrCU. Calculate to C^Os. CALCULATION. PbCrO 4 X 0.2352 = Cr 2 O 3 . NOTE. In the case of chrome liquors containing impurities such as organic matter, dyestuffs, etc., oxidation cannot be effected with peroxide alone. The solution must be rendered alkaline with NaOH and boiled with cone. KMnO 4 solution. The violet color of the permanganate disappears on boiling and more permanganate must be added and the solution again boiled. Repeat this process until a violet red color finally persists even after boiling and then remove the small excess of KMnO 4 by warming the solution with 2 drops of alcohol. Filter and wash and determine the CraOs in the filtrate. Chromium in Chrome Leather.* Ash 3 grams of leather. Mix the ash thoroughly with 4 grams of a mixture of equal parts of Na2COs, K2COs and powdered borax glass and fuse in a platinum crucible for thirty minutes. Dissolve the cooled fusion in hot water and slightly acidify with HC1. Filter. If there is any residue on the filter, ash it and treat the ash with 1 gram of the above fusion mixture in the same manner as the original ash, adding the solution to the first. Make up to 500 cc., and pipette out 100 cc. of this solution into an Erlenmeyer flask. Neutralize with cone. HC1, add 5 cc. in excess and then add 10 cc. of 10% KI solution and titrate as previously described. * J, Am. Leather Chem. Assoc. 14, page 668 (1919). MISCELLANEOUS ANALYSES 479 SUMAC EXTRACT Solutions Required. (1) Potassium Permanganate: 0.50 gram pure KMnO^ per liter. (2) Indigo Carmine: 5 grams pure indigo carmine and 50 grams cone. H2S04 per liter. (3) Gelatin: 20 grams pure gelatin per liter. (4) Acid Salt Solution: a saturated solution of NaCl containing 50 cc. of cone. H2S04 per liter. STANDARDIZATION. The indigo carmine solution must be fil- tered and should give a pure yellow, free from any trace of brown, when oxidized with KMnO4. Dilute 25 cc. of this solution in a large white porcelain dish with about 750 cc. of tap water and add the KMnC>4, drop by drop, from a burette until a pure yellow is obtained, stirring the liquid constantly. The dropping should be as nearly as possible at a similar rate for each experiment and should be slower toward the end of the titration. The final end-point must be approached cautiously by adding the KMnCX very slowly until the pure yellow liquid shows a faint pinkish rim, which can most clearly be seen on the shaded side. At least two titrations should be made. Total Astringency. Weigh out 2.5-3.5 grams of the sumac extract and dilute to 500 cc. Place 5 cc. of this solution in a large porcelain dish with 25 cc. of the indigo carmine solution and 750 cc. of tap water. Titrate with the KMnC>4 solution as above described, running at least two determinations. Subtract from this titration the amount of KMnO4 required by the indigo solu- tion in standardizing, and from the known strength of the KMnC>4 solution calculate the number of cc. of 0.1 N KMnC>4 reduced by the total astringency. From this calculate the total astringency as tannins. CALCULATION. 1 cc. of 0.1 N KMnO4 = 0.004 157 gram tannin. Astringent Non-tannins. To 50 cc. of the dilute sumac solution above referred to add 25 cc. of the gelatin solution, 25 cc. of the acid salt solution and 10 grams of china clay or kaolin. Shake thoroughly for about five minutes and filter through a dry filter. This removes the tannins. Titrate 10 cc. of the filtrate (corresponding to 5 cc. of the original sumac extract solution) for non-tannins in exactly the same manner as the total 480 TECHNICAL METHODS OF ANALYSTS astringency was determined, and calculate the per cent of non- tannins by means of the same factor. Tannin. Subtract the percentage of non-tannins from the percentage of total astringency and report the difference as the percentage of tannin in the extract. REFERENCES. H. R. Proctor: " Leather Industries Laboratory Book of Analytical and Experimental Methods," 2d Ed., page 227. Leach: " Food Inspection and Analysis," page 282. 20 PER CENT PARA RUBBER COMPOUND General. This method is the procedure recommended by the Underwriters' Laboratories for Code Rubber and should be used for all rubber insulation on wires and cables other than 30% Para Compound. All determinations should be performed in duplicate and must check within 0.2%,. calculated on the total sample. The average value should be taken as the true value. Preparation of Sample. Take a sufficient length of wire to give at least 15 grams of rubber compound. Remove the outer coverings and sandpaper the surface of the rubber sufficiently to remove all irregularities, and in any case to a depth of not less than 0.003 inch. (This is to get rid of extraneous matter absorbed from the impregnating compound.) . Wipe with a dry cloth. In some cases after the mechanical cleaning it may be neces< sary to clean further with a cloth dampened with ether, taking care to avoid allowing the ether to penetrate the compound to any marked extent. Strip all the rubber from the wire and cut into small pieces. Grind this entire sample in a mill until it is finely divided, avoiding heating. Spread the sample out on a piece of glazed paper, pass over it a strong magnet to remove any metal which may have come from the grinder, and then mix thoroughly. Acetone Extract. Extract a 2-gram sample with c. P. acetone until the sample is practically free from matter soluble in acetone (not less than five hours), the speed of condensation being such that at least 150 drops per minute fall directly upon the sample. Distill off the acetone, dry the flask with the extract to constant weight at 95-100 C., and weigh. Extraction Apparatus. A very satisfactory extractor is that recommended by the Joint Rubber Insulation Committee and MISCELLANEOUS ANALYSES 481 shown in Fig. 25, but any extractor may be used which conforms to the following specifications: 1. The extraction cup shall be surrounded by the vapor of the solvent at its boiling point. 2. The condensed solvent shall fall directly on the sample. Block tin tubing, 8 mm. outside diameter All dimensions in millimeters FIG. 25. Extraction Apparatus for Rubber. 3. The outlet from the extraction cup shall be at the bottom only. 4. No rubber or cork stoppers shall come in contact with the solvent. 5. The sample shall be put directly into the extraction cup without the use of a paper thimble, a disk of filter paper or its 482 TECHNICAL METHODS OF ANALYSIS equivalent at the bottom of the cup being depended upon to hold back the solid particles. Chloroform Extract. Without drying the residue from the acetone extraction, extract with CHCls for three hours. If at this time the solution is not coming through colorless, continue until it is colorless, provided the total extraction does not exceed five hours. Distill off the CHCla, dry the flask with the extract to constant weight at 95-100 C., and weigh. Alcoholic Potash Extract. Prepare a normal alcoholic KOH solution * by dissolving c. P. KOH in 95% alcohol which has been freshly distilled over KOH. Let the solution stand overnight and filter. Dry the residue from the CHCk extraction at 50-60 C., place in a 200 cc. Erlenmeyer flask and boil with a reflux con- denser for four hours with 50 cc. of normal alcoholic KOH. Filter the solution into a beaker; wash first with 100 cc. of absolute alcohol, then with 50 cc. of hot water, and evaporate to approximate dryness. Add a little water, transfer into a separatory funnel, dilute to 100 cc. and acidify with dil. HC1. Shake out with 20 cc. portions of ether until the last portion is colorless, after which shake out twice more. Shake out the ether solution twice with 50 cc. of water. Filter the ether solution into a small weighed beaker and wash the filter with a little ether; evaporate off the ether without boiling, dry to constant weight at 95-100 C., and weigh. Total Sulfur. Mix a 0.5 gram of sample with 4 grams of Na202 (sulfur free) and 6 grams of ^COs in a dry 15 cc. iron crucible; cover the crucible, insert into a close-fitting hole in an asbestos board and place about 10 cm. above a flame turned very low. Gradually increase the flame until the mixture fuses, pro- ceeding cautiously, as rapid heating will cause an explosion, and then apply the full heat for fifteen to twenty minutes. Rotate the crucible while the melt solidifies. When cool put the cru- cible and cover into a 200 cc. casserole filled with water, add 5-10 cc. of bromine water, and boil until the melt is dissolved and the bromine expelled. Let the precipitate settle, with the addi- tion of MgO if necessary; decant the liquid through a thick filter and wash the residue by decantation with hot water until prac- * 56.1 grams of KOH per liter. MISCELLANEOUS ANALYSES 483 tically neutral. Acidify the filtrate with HC1, evaporate to dry- ness and dehydrate the silica; take up in 400 cc. of water, add 5 cc. of dil. HC1 and filter. Bring to a boil and add slowly a slight excess of hot 10% BaCl2 solution. Let stand overnight, filter, wash, ignite and weigh the BaSC>4. Calculate to S. . CALCULATION. BaS0 4 X 0. 1373 = S. NOTE. In case of disagreement on total sulfur, the Carius' Method shall be used. (See Gattermann: "Practical Methods of Organic Chemistry." Sulfur (Alternative Method). Fuse a 0.5 gram sample with a mixture of 2 parts by weight of KNOs, 3 parts of Na2COs and 1 part of K2COs in a porcelain crucible. The heating should at the start be conducted at a very low temperature. After fusion, disin- tegrate the melt with water. Filter, boil with bromine water, and acidify with excess of HC1. Boil down in order to break up nitrates, dilute to at least 300 cc. with water and finally precipitate while hot with an excess of BaCb solution. Let stand overnight, filter and wash, ignite and weigh the BaSC>4. Ash. Place in a small porcelain dish a weight of acetone extracted compound which, before such extraction, had a weight of 1 gram. Burn off the rubber completely at as low a tempera- ture as possible, without letting it catch fire. (High temperatures must be carefully avoided.) This may be done in either of two ways: First, by inserting the dish into a close-fitting hole in an asbestos board and heating carefully over a small flame, the dish afterwards being placed on a triangle and just sufficient flame applied to the sides to remove any condensation; or, second, by placing the dish containing the sample near the mouth of a suitably heated muffle furnace. Cool in a desiccator and weigh the ash. Free Sulfur. The free S will be in the acetone extract. Add to the flask containing the acetone extract 10-15 cc. of fuming HNOs. Place the flask on the water bath until all is in solution. Transfer to a hot plate and add small portions of KClOa, until the solution is decolorized. Evaporate to dryness, carefully avoiding any SOs fumes which would foe absorbed by the solution. Take up the residue with 50 cc. of water, add 3 cc. of cone. HC1 and filter into a 600 cc. beaker. Dilute to about 400 cc., heat to boiling and precipitate with BaCl2 in the usual manner. Let stand over- night, protected from SO 3 fumes. Filter off the BaSO 4 . Ignite, cool in a desiccator and weigh. Calculate to S. Subtract from 484 TECHNICAL METHODS OF ANALYSIS this result any S found in a " blank " which should be carried through simultaneously. NOTE. The procedure for free sulfur is not included in the Under- writers' methods. CASE-HARDENING COMPOUNDS General. For case-hardening steel a great variety of sub- stances is used. Among the more common ones are materials containing coke, charcoal, flour, grain, fibers, BaCOa, cyanides, Na 2 C03, crushed bone, powdered glass, rosin, gums, etc. The sample should first be examined under the microscope to note its general appearance and to detect if possible, any of the fore- going substances. The following is an attempt at a general scheme of analysis but may have to be modified to suit conditions. Ether Extract. Mix the sample thoroughly and quarter it down to about 50 grams. Extract 10 grams or more with ether in a Soxhlet extraction thimble, collecting the extract in a weighed Soxhlet flask. Evaporate off the ether by placing the flask in warm water, dry the residue at 100 C., and weigh. If the extract is appreciable in amount it should be tested for rosin and its iden- tity ascertained, if possible, by determining the saponification number, iodine number, refractive index, etc. Moisture. Dry 2 grams (using the extracted material if the original contains oil) at 100 C. until the weight is constant. The loss indicates moisture. Volatile Matter. The volatile matter is determined in exactly the same manner as the volatile matter in the Proximate Analysis of Coal (page 173). Grind 2 grams of the material, on which the moisture has been determined, rapidly to a fine powder and place 1 gram in a covered platinum crucible over a Chaddock's burner, gas pressure 1.1 inches, and burn for exactly seven minutes. Cool and weigh. The loss is the volatile matter. Ash. Ignite 2 grams in a porcelain crucible until the ash is free from carbonaceous matter. Dry in a desiccator and weigh. Loss on Heating at 850 C. for Six Hours. This should be done in a porcelain dish, 3 inches in diameter and 2 inches deep. MISCELLANEOUS ANALYSES 485 Weigh the dish empty and then fill it with the compound, heaping it up and packing down. Weigh again to determine the amount of substance taken. Place in a muffle, the temperature of which has been ascertained by means of a pyrometer, and heat at 850 C. for six hours. Cool in a desiccator and determine the loss in weight. Analysis of Ash. The ash should be analyzed in the ordinary manner for Ba (probably present as carbonate if at all), CaO, P20s, SO4, Na and K. It should be noted whether the ash is alkaline or acid and, if so, the extent should be determined by titra- tion. The presence or absence of carbonates should also be noted. If Caa(PO4)2 is found, the material probably contains ground bone. Water Extract. Extract 10-20 grams or more with hot water (after extracting with ether if any oil is present) in a Soxhlet extractor, collecting the extract in a weighed flask. Evaporate off the water; dry at 100 C. and weigh. If the water extract is appreciable it should be analyzed. The best .procedure is to make it up to a given volume and ascertain its nature qualitatively and then make quantitative determinations on separate aliquots. It should be remembered that it is unnecessary to test for any- thing that is insoluble in water. Nitrogen. (a) TOTAL NITROGEN. This is determined by the Kjeldahl method, modified to include nitrogen present as nitrates, using 1-4 grams of the material (see page 67). (b) NITROGEN AS AMMONIA. Place 1-5 grams in a 500 cc. round-bottom flask. Add about 200 cc. of water and 5-10 grams of MgO, free from carbonates. Connect with a con- denser and distill into 50 cc. of 0.1 N HC1. Titrate the excess of HC1 with 0.1 N NaOH and calculate the difference to NH 3 . (c) PROTEIN NITROGEN AND NITRATES. A qualitative test on the water extract will show whether or not nitrates are present. If they are not present, the difference between the total N and theN present as NHs, will give the N present as protein, and this figure multiplied by 6.25 will give the amount of protein. (See note.) If nitrates are present it is generally sufficient to report the difference between the total nitrogen and the ammoniacal nitrogen as " protein and nitrate nitrogen." The microscopical examina- tion and the odor on burning will generally give indications as to whether protein is present or not. 486 TECHNICAL METHODS OF ANALYSIS (d) NITROGEN AS CYANIDE. Cyanide may be detected quali- tatively by acidifying the original material and very cautiously noting the odor. If cyanide is present, digest 10 grams of the original material with warm water and wash by decantation once or twice, pouring the liquid through a filter. Finally transfer the material to a filter and wash thoroughly with warm water. Cool to room temperature and determine cyanogen by titrating with 0.1 N AgNOs as described on page 31. NOTE. If cyanide is present, the total N as previously determined will include the N of the cyanide. Carbide. The presence of calcium carbide may be quali- tatively detected by the odor of acetylene when the material comes in contact with water. An approximately quantitative estimation may be obtained by collecting the acetylene from a weighed amount of the sample under water in a nitrometer tube. 1 cc. of C2H2 at C. and 760 mm. pressure is equivalent to 0.002866 gram of CaC 2 . Sulfide (Qualitative). This may be detected by the odor of H2S when the material is acidified with HC1, or in small amounts by the blackening of filter paper moistened with lead acetate solution. The previous examination of the sample will generally give indications as to the nature of the sulfide. The amount of sulfur present as sulfide may be determined by treating 2 grams of the material with dil. HC1 and bromine water, filtering, and deter- mining the total sulfate in the filtrate by means of BaCb. On a separate portion determine the sulfur present as sulfate. The difference between these two determinations shows the amount of sulfur present as sulfide. CUTTING COMPOUNDS General. These materials usually consist of a mixture of some fatty oil with a mineral oil, emulsified by means of soap and, as a rule, contain 40-60% of water. The soap is often, though not always, a potash soap. Whale oil and lard oil are the fatty oils generally used. Moisture. Dry approximately 5 grams in a flat platinum dish at 105 C. to constant weight. The loss represents moisture and volatile matter. MISCELLANEOUS ANALYSES 487 The moisture may also be determined by the Xylol Method. (See page 271.) Ash. Ignite the residue from the moisture determination until the ash is white or light gray in color. Cool in a desiccator and weigh. Soap. Titrate the ash with 0.1NHC1 and methyl orange. Test the HC1 solution with a platinum wire in a flame to see whether the base is Na or K. Calculate the titration to the proper soap. CALCULATION. 1 cc. 0.1 N HC1 = 0.0304 gram Na soap. 1 cc. 0.1 N HC1 = 0.0321 gram K soap. NOTE. As a check the titration should also be calculated to sodium or potassium carbonate and the weight thus calculated should agree approx- imately with the weight of the ash. CALCULATION. 1 cc. 0.1 N HC1= 0.00530 gram Na 2 CO 3 . = 0.00691 gram K 2 CO 3 . Total Oily Matter. Weigh out about- 10 grams into a 200 cc. beaker, add 100 cc. of water, warm on the steam bath and add an excess of dil. H2S04. Cool the mixture and transfer to a separatory funnel. Wash the beaker finally with CHCls and add the wash- ings to the separatory funnel. Shake out 3 times with portions of about 30 cc. of CHCl^. After each treatment allow the two liquids to separate completely and clearly and then draw off the CHCla extract into a weighed Soxhlet flask. Evaporate the chloroform from the combined extracts in the Soxhlet flask, dry to constant weight at 105 C. and weigh the total oily matter. Unsaponifiable Oil. Weigh approximately 10 grams into a 300 cc. Erlenmeyer flask, add 50 cc. of 0.5 N alcoholic KOH. Saponify for two hours, or longer, over a low flame, using a reflux condenser. Add a few drops of phenolphthalein solution. If the mixture in the flask is not still alkaline, add 25 cc. more of alcoholic KOH and saponify for another two hours. The mixture should be alkaline after the saponification is complete. Add about an equal volume of water, connect the flask with an ordinary Liebig condenser and distill off at least half of the liquid. Cool the residue and transfer to a large separatory funnel and dilute with several times its volume of cold water. Extract the solution 4 times with portions of about 50 cc. of ether. If the mixture is violently shaken it is likely to form an emulsion which can be broken up 488 TECHNICAL METHODS OF ANALYSIS only with difficulty. Therefore mix the two by giving a rotating motion to the funnel or by laying it on its side and rolling it. Draw off the ether each time and wash once with water. Pour the combined extracts into a weighed Soxhlet flask, evaporate off the ether and weigh. NOTE. Instead of ether, 86 Baume naphtha may be used. This is less likely to form emulsions, but care should be taken to use naphtha which leaves no residue on evaporation at 100 C. Uncombined Fatty Oil. This is obtained by calculation. Calculate the amount of fat in the soap present by using the fol- lowing factors: Combined Fat = Na Soap X 0.93. = K Soap X 0.88. Add together the unsaponifiable oil and the fatty oil combined as soap and subtract the sum from the total oily matter. The difference represents the uncombined fatty oil. Free Fatty Acid. Weigh 20 grams into a 300 cc. Erlenmeyer flask, add 100 cc. of alcohol which has previously been warmed and made neutral to phenolphthalein with 0.1 N NaOH. Warm the mixture on the steam bath for about one-half hour, titrate with 0.1 N KOH and phenolphthalein and calculate the titration to oleic acid. CALCULATION. 1 cc. 0.1 N KOH = 0.0282 gram oleic acid. METAL POLISHES General. Liquid metal polishes may be divided into 2 general classes : (1) Naphtha Polishes; (2) Fireproof Polishes. The former consist essentially of naphtha containing an abrasive powder held in suspension by means of soap and the odor of the naphtha covered up to some extent by an essential oil such as citronella or nitrobenzol. The second class is for the most part water emulsions of various solvents holding an abrasive powder in suspension. Carbon tetrachloride has been used to some extent either alone or in combination with naphtha. An examination of a metal polish should in general include a determination of the amount and character of the solvents, the soap and the abrasive. Solvents. Weigh out 100 grams of polish into a 250 cc. dis- tilling flask provided with a thermometer and connected with a MISCELLANEOUS ANALYSES 489 long condenser. Start the distillation in a water bath, taking care that the flame is amply protected from any uncondensed vapors which may issue from the end of the condenser, and collect the dis- tillate in a 25 cc. tared cylinder. When active distillation ceases, or if none takes place, substitute an oil bath for the water bath and continue the distillation up to 250 C., collecting the distillate in tared cylinders, as before. The following fractions should be taken: (1) below 100 C.; (2) 100-150 C.; (3) 150-200 C.; (4) 200-250 C. The boiling points, odors, refractive indices and specific grav- ities of the various fractions will give indications of the character of the solvents. Those which are likely to be present are: turpentine, pine oil, rosin spirit, rosin oil, naphtha, kerosene, carbon tetra- chloride, alcohol and water. The approximate constants of these solvents are given below : '. Specific Gravity at 15.5 C. Distilling Temperature C. Refractive Index American turpentine. Wood turpentine .... Pine oil Rosin spirit Rosin oil 0.860-0.870 0.855-0.910 0.935-0.947 0.856-0.880 0.96 -1.10 155-170 159-165 190-220 Gradual rise (about 50% below 120) 300-360 1.468-1.473 @20C. 1.468-1.475 @20C. 1.484-1. 486 @15.5C. Naphtha . 700-0.750 50-150 Kerosene CC1 4 0.775-0.800 1.60 150-300 77 Alcohol, wood Alcohol, denatured.. . About 0.800 About 0.816 About 65 About 78 NOTES. (1) Water will separate from all other constituents except alcohol. If both water and alcohol are present the amounts can be roughly approxi- mated by taking the volume and sp. gr. of the aqueous distillate and cal- culating from the sp. gr. tables for alcohol. (2) If water is present, the distilling temperatures will be no indication of the boiling points of the other substances present. Soap. Dry 15 grams of the polish on a Hofmeister Schalchen, crush and extract the residue with ether in a Soxhlet extractor. 490 TECHNICAL METHODS OF ANALYSIS After the ether-soluble materials are removed, extract the soap with alcohol (95%). Evaporate off the alcohol, dry at 105 C., and weigh. Examine the soap to determine whether it is a sodium, potassium or ammonium soap. In case no alcohol-soluble soap is obtained, it is probable that a lime soap is present, in which case the contents of the capsule should be acidified with 1 cc. of cone. HC1 and extracted with ether. Evaporate off the ether, weigh the liberated fatty acids and cal- culate them as calcium oleate. CALCULATION. Fatty acids XI. 07 = calcium oleate. Boil the acidified contents of the capsule with hot water, filter, and make a qualitative test for calcium. Free Ammonia. If an ammonium soap is present or the polish smells of NHa, weigh out 2-5 grams of the original polish into a flask, add 200 cc. of water and distill into 0.1 N acid. A spray trap should be used and a few drops of methyl red added to the acid. Titrate the excess acid with 0.1 N NaOH. CALCULATION 1 cc. 0.1 N acid = 0.0017 gram NH 3 . Combined Ammonia. Dilute the solution remaining in the flask with 200 cc. of water, add 10 cc. of cone. NaOH, and distill as before. Calculate to NB^ soap. CALCULATION. 1 cc. 0.1 N acid = 0.03 gram NH 4 soap. Abrasive. Weigh out 10 grams of the metal polish, evaporate to dryness, taking proper precautions if the solvent is an inflam- mable one, ignite at a low red heat to burn off any organic matter, and weigh. Correct the ash thus formed for the ash of the soap, unless an ammonia soap is present. Examine the ash microscop- ically to ascertain if it consists of tripoli or ground rock. If the latter, a qualitative analysis may be made ; although the physical condition of the abrasive is of more importance than its chemical composition. NOTE. It should be remembered that all of the determinations are more or less approximate, due to various causes, and consequently calculations should be made to the nearest even percentage. SOLDERING PASTE General. Soldering pastes generally consist of a mixture of and NH4C1 with petrolatum. While they vary somewhat MISCELLANEOUS ANALYSES 491 in composition their proportions are generally within the following limits : ZnCl 2 15-20% NH 4 C1 1- 5% Water Less than 5% Since the salts are likely to settle out on standing, especially if the paste has been in a warm place, it is necessary before start- ing the analysis to make sure that the material is very thoroughly and completely mixed. Water. Determine the water by the Xylol Method as in the analysis of Greases (page 271). Total Ammonia. Weigh out 30-40 grams of the paste in a beaker and transfer to a separatory funnel by means of a hot 5% solution of HNOs. About 200 cc. of the acid should be employed. Shake out the paste thoroughly with the acid and let the layers separate. Draw off the solution into a liter graduated flask. Repeat the shaking out of the melted paste in the funnel 3 times with fresh portions of about 200 cc. of the hot acid. Cool the solution in the flask to room temperature and dilute with distilled water to the mark. Use aliquots of this solution for the deter- mination of total NHs, total Cl and Zn. For the NHs determination, pipette 200 cc. into a Kjeldahl dis- tilling -flask, add a few drops of methyl orange and make distinctly alkaline with NaOH, adding at the same time a few grains of metallic zinc to prevent bumping. Distill in the usual manner, collecting the distillate in 50 cc. of 0.1 N acid. Titrate back the excess of acid with 0.1 N alkali and calculate the acid neutralized during the distillation to NHs and also to NHiCl. Divide the weights thus obtained by 0.2 of the weight of the original sample and multiply by 100 to obtain the percentages. CALCULATION. 1 cc. of 0.1 N acid = 0.001703 gram NH 3 . = 0.005350 gram NH 4 C1. Zinc Chloride. Pipette 100 cc. of the acid solution into a 350 cc. beaker. Add a slight excess of NEUOH and heat to boiling. Filter if there is a precipitate of iron and aluminum hydroxides. Make the filtrate faintly acid with acetic acid; then add a large excess of ammonium phosphate and boil the solution until the precipitate of ZnNHiPC^ is crystalline. Let settle until clear. 492 TECHNICAL METHODS OF ANALYSIS Filter on a Gooch crucible; wash with hot water and dry in the oven. Finally ignite strongly to constant weight and weigh as Zn 2 P207. Calculate to ZnCb; divide by 0.1 of the weight of the original sample taken and multiply by 100 to obtain the per- centage of ZnCb. CALCULATION. Zn 2 P 2 O 7 X 0.8943 = ZnCl 2 . Total Chlorine. It is not generally necessary to determine the total chlorine unless it is desired to check up the previous deter- minations or unless NaCl is suspected to be present. For the determination, pipette 50 cc. of the HNOs solution into a clean por- celain dish or casserole. Add sufficient pure CaCOs to neutralize all the acid present and still have an excess of the CaCOs undis- solved. Then add about 5 cc. of K 2 CrC>4 indicator solution and titrate the Cl with 0.1 N AgNOs until a permanent reddish color appears in the solution. The end point can best be determined in the presence of artificial light. Calculate the amount of chlorine to which the titration corresponds. Divide this weight by 0.05 of the weight of the original sample and multiply by 100 to obtain the percentage of Cl. CALCULATION. 1 cc. 0.1 N AgNO 3 = 0.003546 gram Cl. NOTE. The total Cl of course should check closely the amount of Cl equivalent to the ZnCl 2 and NH 4 C1 determined. If it is in excess, the mate- rial should be tested for Na, and, if present, the excess Cl calculated to NaCl. CALCULATIONS--NH4C1 X 0.6628 = Cl. ZnCl 2 X0.5.204 =C1. ClX 1.6486 =NaCl. SANITARY ANALYSIS OF WATER AND SEWAGE General. Before making a sanitary analysis of a sample of water, the general characteristics of the water should be noted. All substances which are likely to undergo change should be determined as quickly as possible. This includes free and albu- minoid ammonia, nitrites, nitrates and oxygen consumed (also free C02 and alkalinity or acidity when desired) . The determinations included in the usual sanitary water analysis are given below and the form of reporting and the num- ber of significant figures to be reported are also indicated : MISCELLANEOUS ANALYSES 493 Sediment X.X Turbidity X.X Odor '... " .... .- Color XX Parts per 100,000 Residue on Evaporation X.XX Nitrogen as Free Ammonia X.XXXX Albuminoid Ammonia X.XXXX Nitrites X.XXXX Nitrates ' X.XXXX Oxygen Consumed X.XX Chlorine X.XX Hardness X.X Iron X.XXXX Odor. The observation of the odor of cold and hot samples of surface waters is very important, as the odors are usually connected with some organic growths or with sewage contamination, or both. The odor of ground waters is often caused by the earthy constituents of the water-bearing strata. The odor of a con- taminated well water is often a contributary evidence of its pol' lution. COLD. ODOR. Shake the sample violently in one of the gallon collecting bottles, about half or two-thirds full of the water at room temperature (about 20 C.). Remove the stopper and smell the odor at the mouth of the bottle. HOT ODOR. Into a tall 400 cc. beaker without lip, pour about 150 cc. of the sample. Cover the beaker with a well-fitting watch glass, place on a hot plate and bring the water to just below boiling. Remove the beaker from the plate and let it cool for not more than five minutes. Then shake with a rotary movement, slip the watch glass to one side and note the odor. EXPRESSION OF RESULTS. Express the quality of the odor by some such descriptive term as the following; vegetable, aromatic, grassy, fishy, earthy, moldy, musty, disagreeable, peaty, sweetish, sulfide, etc. Sediment and Suspended Matter. Ordinarily the sediment may be expressed qualitatively as slight, flocculent, considerable, or heavy. If quantitative results are desired, filter a measured 494 TECHNICAL METHODS OF ANALYSIS quantity of the water after shaking (500 cc. are suitable) through a filter paper which has previously been dried in a weighing bottle in the oven to constant weight. Then dry the filter containing the sediment to constant weight, at the same temperature. This gives the total suspended matter. If the mineral suspended matter is desired, ignite in a platinum crucible and weigh. Turbidity. The turbidity is due to suspended matter such as clay, silt, finely divided matter, microscopic organisms, etc. The standard of turbidity adopted by the U. S. Geological Survey is a water which contains 100 parts of silica per million in such a state of fineness that a bright platinum wire, 1 mm. in diameter, can just be seen when the center of the wire is 100 mm. below the surface of the water and the eye of the observer is 1.2 meters (about 4 feet) above the wire; the observation being made in the middle of the day, in the open air but not in sunlight, and in a vessel so large that the sides do not shut out the light. The turbidity of such a water is arbitrarily fixed at 100. The number obtained by dividing the weight of suspended matter in the sample, in parts per million, by the turbidity, is called the Coefficient of Fineness. If greater than 1, it indicates that the matter in suspension is coarser than the standard; if less than 1, that it is finer. SILICA STANDARDS. To prepare the standard, use precipitated fuller's earth, dry and sift through a 200-mesh sieve. One gram of this preparation in 1 liter of distilled water makes a stock sus- pension containing 1000 parts of silica per million, which should have turbidity of 1000. Test this suspension, after diluting a portion of it with 9 times its volume of distilled water, with a platinum wire as described below to ascertain if the silica has the necessary degree of fineness and if the suspension has the necessary degree of turbidity. If not, correct by adding more silica or more water as required. Prepare standards for comparison from the stock suspension by diluting with distilled water. For turbidity readings below 20, keep standards of 0, 5, 10, 15, and 20 in the same size bottles as are used in collecting the samples. For readings above 20, keep standards of 20, 30, 40, 50, 60, 70, 80, 90, and 100 in 100 cc. Nessler tubes approximately 20 mm. in diameter. In comparing the water under examination with the standards, take an amount MISCELLANEOUS ANALYSES 495 equivalent to the standard, view the liquids sidewise towards the light, looking at some object, and note the distinctness with which the margins of the object may be seen. Keep the standard stop- pered and shake both sample and standard thoroughly before making the comparison. NOTE. In order to prevent growth of bacteria or algae in the standards, a small amount of mercuric chloride may be added to them. PLATINUM WIRE METHOD. The platinum wire method of determining turbidity may be used directly on the sample instead of preparing standards as above. It is the method to be used in ascertaining the correctness of the stock suspension above de- scribed. This method requires a rod with a platinum wire of 1 mm. diam- eter (0.04 inch, No. 18 B. & S. gauge) inserted in it about 1 inch above the end of the rod and projecting from it at least 25 mm. at a right angle. Near the end of the rod, at a distance of 1.2 meters (about 4 feet) from the platinum wire, place a wire ring directly above the wire. When making the examination look through this ring with the eye directly above it. Graduate the ring as follows : At a distance of 100 mm. from the center of the wire, mark 100. If desired, other graduations may be made according to the table below or the distance may be measured in mm. and the turbidity calculated. Procedure. Lower the rod vertically into the water as far as the wire may be seen, and then read the level of the surface of the water on the graduated scale (or mark the point and measure the distance). The following precautions must be taken to insure correct results: Observations must be made in the open air, preferably in the middle of the day and not in direct sunlight. The wire must be kept bright and clean. If for any reason observations cannot be made under natural conditions directly, a pail or tank may be filled with water and the observation taken in that, but in this case care must be taken that the water is thoroughly stirred before the observation is made, and no vessel is to be used for this purpose unless its diameter is at least twice as great as the depth to which the wire is immersed. Waters which have a turbidity above 500 must be diluted with clear water before the observations 496 TECHNICAL METHODS OF ANALYSIS are made; but in case this is done, the degree of dilution used should be stated and form a part of the report. Turbidity, Parts per Million Vanishing Depth of Wire, mm. Turbidity, Parts per Million Vanishing Depth of Wire, mm. 7 1095 70 138 8 971 75 130 9 983 80 122 10 794 85 116 11 729 90 110 12 674 95 105 13 627 100 100 14 587 110 93 15 551 120 86 16 520 130 81 17 493 140 76 18 468 150 72 19 446 160 68.7 20 426 180 62.4 22 391 200 57.4 24 361 250 49.1 26 336 300 43.2 28 314 350 38.8 30 296 400 35.4 35 257 500 30.9 40 228 600 27.7 45 205 800 23.4 50 187 1000 20.9 55 171 1500 17.1 60 158 2000 14.8 65 147 3000 12.1 The wire method is to be used for testing the degree of fineness of the standard silica, and this should be such that when added to distilled water in an amount equal to 100 parts per million, the wire observed under standard conditions can be just seen at a depth of 100 mm. below the surface of the water. MISCELLANEOUS ANALYSES 497 Color. The " true color " of water is considered that part of the apparent color which is due only to substances in solution, i.e., it is the color of water after filtration. The " apparent color " is the color as viewed by. inspection of the original sample. In stating the results, the word " color " means the color of the fil- tered sample, unless otherwise expressed. PLATINUM-COBALT STANDARD. The platinum-cobalt method of measuring color is the standard and the unit of color is that pro- duced by one part of platinum per million. Prepare the standard solution (which has a color of 500) as follows: Dissolve in water 1.246 grams of K^PtCle (containing 0.5 gram of Pt), 1 gram of crystallized CoCb-GH^O (containing 0.25 gram of Co) and 100 cc. of cone. HC1, and dilute with dis- tilled water to 1 liter. Place varying aliquots of this solution in Nessler tubes and dilute with distilled water to the 100 cc. mark. These diluted solutions should have colors of 0, 5, 10, 15, 20, 25, 30, 35, 40, 50, 60, and 70, respectively. The tubes must be of such size that the 100 cc. mark is between 20 and 25 cm. above the bot- tom and is uniform for all tubes. Protect these standard tubes from dust when not in use. Procedure. Fill a clean Nessler tube with the sample to the height equal to that in the standard tubes, and compare the color with the standards. The observation must be made by looking vertically downward through the tubes upon a white surface placed at such an angle that light is reflected upward through the column of liquid. Estimate the color of the water from the standards which most nearly match it. NOTES. (1) Waters that have a color darker than 70 must be diluted before making the comparison. (2) Water containing suspended matter must be filtered until no visible turbidity remains before the observation is made. If the suspended matter is coarse, use filter paper; if fine, a Berkefeld filter is recommended. The Pasteur filter must not be used, as it has a decolorizing effect. Chemical Analysis. Express results of chemical analysis in parts per 100,000. RESIDUE ON EVAPORATION (TOTAL SOLIDS). Evaporate 100 cc. of the sample in a weighed platinum dish on the water bath. If the water has a high Mg content, add 25 cc. of 0.02 N Na2COs solution,* correcting for this in the final calculation. Dry the * This prevents the hydrolysis of MgClz and loss of HQ. 498 TECHNICAL METHODS OF ANALYSIS residue for one-half hour at a temperature of about 103 C. Cool in a desiccator over cone. H2SO4 and weigh. Loss ON IGNITION. Heat the platinum dish with the residue in a " radiator/' which consists of another platinum dish large enough to allow an air space of about 0.5 inch between the inner and outer dishes, the inner dish being supported by a triangle of platinum wire laid on the bottom of the outer dish. Suspend over the inner dish a disk of platinum foil large enough to cover the outer. Heat the large dish to bright redness until the residue is white or nearly so. Let cool, moisten the residue with a few drops of distilled water, dry in the oven for one-half hour, cool and weigh. The difference between this and the total solids is the " loss on igni- tion." NOTES. (1) An electric muffle furnace may be used in place of the radiator. (2) It is not customary to report the loss on ignition but it is of value in interpreting the analysis. The manner in which the residue behaves as to odor and color upon ignition should be noted, as it often gives a helpful clue to the character of the organic matter. NITROGEN AS FREE AMMONIA General. There are two methods for estimating nitrogen as free ammonia: (A) by distillation, and (B) by direct nessleriza- tion. The former is recommended for most waters, while the latter is preferable for sewages, sewage effluents, and highly pol- luted surface waters. I. Free Ammonia by Distillation. Connect a liter round- bottom flask to a block tin or aluminum condenser in such a way that the distillate may be conveniently delivered into Nessler tubes. Free the apparatus from NHs by boiling distilled water in it until the distillate shows no further traces of free NHs. Then empty the distilling flask and measure into it 500 cc. of the sample, or a smaller portion diluted to 500 cc. Apply the heat so that the distillation will be at the rate of 6-10 cc. per minute. Collect three * Nessler tubes of the distillate, 50 cc. to each portion. These contain the free NHs, to be measured as described below. * If the free NH 3 is unusually high, it may be necessary to collect more than 3 tubes of distillate. The last tube should be free or practically free from NH 3 . MISCELLANEOUS ANALYSES 499 NOTE. If the sample is acid or if the presence of urea is suspected, add about 0.5 gram of Na2COs previous to distillation. Omit this when possible as it tends to increase bumping. Use only Nessler tubes which do not show a variation of more than 6 mm. (0.25 inch) in the distance which the 50 cc. graduation mark is above the bottom. The tubes must be of clear white glass with polished bottoms. The comparison of the distillates may be made either (1) against nesslerized solutions containing known quantities of N as NHiCl or (2) against permanent standard solutions in which the colors of nesslerized standard ammonia colors are duplicated by solutions of Pt and Co chlorides.. (1) Comparison with Ammonia Standards. (A) REAGENTS. For comparison with ammonia standards, prepare the following reagents : (a) Ammonia-free water. (b) Standard NH 4 Cl solution. Dissolve 3.82 grams of NILiCl in 1 liter of distilled water and dilute 10 cc. of this to 1 liter with ammonia-free water. 1 cc. of the final solution equals 0.00001 gram of N. (c) Nessler' s reagent. Dissolve 50 grams of KI in a minimum quantity of cold water. Add a saturated solution of HgCk until a slight permanent precipitate forms. Add 400 cc. of 50% KOH solution, made by dissolving the KOH and allowing it to settle clear before using. Then dilute to 1 liter, let settle and decant. This solution should give the required color with ammo- nia within five minutes after addition and should not precipitate with small amounts of NHs within two hours. (B) PROCEDURE. Prepare a series of 16 Nessler tubes which contain the following number of cc. of the standard NH4C1 solu- tion, diluted to 50 cc. with ammonia-free water, namely, 0.0, 0.1, 0.3, 0.5, 0.7, 1.0, 1.4, 1.7, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0 and 6.0. These will contain 0.00001 gram of N for each cc. of the standard solution used. Nesslerize the standards and also the distillates from the sample by adding approximately 2 cc. of Nessler's reagent to each tube. Do not stir the contents of the tubes. Have the temperature of the tubes practically the same as that of the standard, otherwise the colors will not be directly comparable. Let the tubes stand for at least ten minutes after nesslerizing. Compare the color produced in these tubes with that of the stand- 500 TECHNICAL METHODS OF ANALYSIS ards by looking vertically downward through them at a white sur- face placed at an angle in front of the window so as to reflect the light upwards. In case the color obtained by nesslerizing the distillates is greater than that of the darkest standard, mix the contents of the tube thoroughly and pour out one-half. Make up the remain- der to the original volume with ammonia-free water. Then make the color comparison and multiply the result by 2. If the color is still too dark, repeat the process of division until a read- ing can be made. In case the color of the distillates is too high, this process may be shortened by mixing together all of the distillates in one sample before making the comparison, then taking an aliquot portion for comparing with the standards. If a precipitate is formed, dilute the reagent before adding it to the tubes. After the readings have been made and recorded, add together the results obtained on separate tubes of the sample to get the total number of cc. of standard required. If 500 cc. of the sample were distilled, multiply this sum by 0.002 to get the parts of N as free NHa per 100,000 parts of the sample. (2) Permanent Standards. Permanent standards for com- parison may be prepared as below and the colors of the nesslerized distillates compared with these after the former have stood about ten minutes. (A) REAGENTS. (1) Platinum Solution. Dissolve 2 grams of K2PtCl6 in a small amount of distilled water. Add 100 cc. of cone. HC1 and dilute to 1 liter.' (2) Cobalt Solution. Dissolve 12 grams of CoCl 2 -6H 2 O in distilled water, add 100 cc. of cone. HC1 and dilute to 1 liter. (B) PROCEDURE. For the standards place varying amounts of these two solutions in Nessler tubes and fill up to the 50 cc. mark with distilled water as shown in the table on page 501. The amounts stated here are approximate, and the actual amount necessary will vary with the character of the Nessler solu- tion used, with the color sensitiveness of the analyst's eye, and with the other incidental conditions. The final test of the stand- ard is best obtained by comparing it with nesslerized standards and modifying the tint accordingly. Such a comparison should be made for each new batch of Nessler solution and should be checked by each analyst. MISCELLANEOUS ANALYSES 501 Equivalent Volume of Standard NH 4 C1 Solution cc. Platinum Solution cc. Cobalt Solution cc. 0.0 1.2 0.0 0.1 1.8 0.0 0.2 2.8 0.0 0.4 4.7 0.1 0.7 5.9 0.2 1.0 7.7 0.5 1.4 9.9 1.1 1.7 11.4 1.7 2.0 12.7 2.2 2.5 15.0 3.3 3.0 17.3 4.5 3.5 19.0 5.7 4.0 19.7 7.1 4.5 19.9 8.7 5.0 20.0 10.4 6.0 20.0 15.0 7.0 20.0 22.0 It is necessary to use tubes which have the 50 cc. mark not less than 20 nor more than 22 cm. above the bottom. These standards may be kept for several months if protected from dust. The method of calculating results is precisely the same as with the ammonia standards. II. Free Ammonia by Direct Nesslerization. (1) REAGENTS. (a) CuS0 4 (10% solution) (6) Lead acetate (10% solution) (c) NaOH or KOH (50% solution) (d) MgCl 2 (10% solution). (2) PROCEDURE (FOR SEWAGE). Mix 50 cc. of the sample with an equal volume of water, place in a short Nessler tube and add a few drops of CuSC^ solution. After thoroughly mixing, add 1 cc. of the caustic solution and again thoroughly mix. Let stand for a few moments. A heavy precipitate should fall to the bottom, 502 TECHNICAL METHODS OF ANALYSIS leaving a colorless supernatant liquid. Nesslerize an aliquot por- tion of this clear liquid. Many samples containing H^S require the use of lead acetate in addition and others require a few trials before the right com- bination of the three solutions to bring about the best results can be made. In place of adding CuSO4 to sewages of high Mg con- tent, satisfactory clarification may often be obtained by mixing with the caustic alone. The amount of N as free NHs is computed after comparison with standards in the same manner as in the distillation procedure. NITROGEN AS ALBUMINOID AMMONIA General. The addition of an alkaline permanganate solution to liquids containing nitrogenous organic matter causes the forma- tion of NHs, the amount of which can be measured upon distillation of the treated sample and nesslerization of the distillate. In sewages and other liquids and substances containing considerable nitrogenous organic matter, the percentage of nitrogenous material which is ammonia-forming is decidedly variable. For this reason albuminoid NHs results in such cases are less valuable than the total organic N, sometimes called the Kjeldahl nitrogen. Hence, for sewage work, including the analyses of both the influents and the effluents of purification plants, as well as the study of highly polluted streams, it is recommended that albuminoid NHs deter- minations be omitted and in their place the total organic N be determined. For ground waters and surface waters containing but little pollution, the N as albuminoid NHs quite uniformly approxi- mates about one-half of the total organic N. Accordingly the continuance of albuminoid NHs determinations for this class of work is approved. Nevertheless the inferiority of such results to those of total organic -N is recognized. Reagent. Alkaline KMnO. Pour 1200 cc. of distilled water in a porcelain dish holding 2500 cc., boil ten minutes and turn off the gas. Add 16 grams of c. P. KMnC>4 and stir until dissolved. Then add 800 cc. of a 50% clarified solution of NaOH or KOH and enough distilled water to fill the dish. Boil down to 2000 cc. Test each batch of this solution for albuminoid NHs by making a blank determination and correct for this blank in the analysis. MISCELLANEOUS ANALYSES 503 Procedure. Interrupt the distillation (made as already described) after the collection of the distillate for free NHs, add 40 cc. or more of alkaline KMnCU and continue the distillation until at least 4 portions of 50 cc. each, and preferably 5 portions, of the distillate have been collected in separate tubes. Have enough permanganate solution present to insure the maximum oxidation of the organic matter. Determine the N in the dis- tillates as previously described and express the results as in the case of free NHs. NOTES. (1) Dissolved N as albuminoid NH 3 may be determined from a sample from which suspended matter has been removed by filtration either through filter paper, or, if finely divided matter is present, through a Berkefeld filter. (2) Suspended N as albuminoid NHs may be obtained by taking the dif- ference between the total and dissolved N. TOTAL ORGANIC NITROGEN Procedure for Water. Boil 500 cc. of the sample in a round- bottom flask until free from NHs. This usually requires the dis- tillation of about 200 cc. of the sample, which is to be collected for the determination of free NHs. Add 5 cc. of c. P. cone. H2SO4, free from N, together with a small piece of ignited pumice. Mix by shaking and place over a flame under a hood. Digest until copious fumes of SOs are given off and until the liquid chars and finally becomes colorless. Remove from the flame and add KMnO4 crystals in small portions until a heavy green precipitate persists in the liquid. Cool, dilute with about 100 cc. of ammonia-free water and neutralize with 10% ammonia-free Na2COs solution. Distill off the NHs, collect in Nessler tubes, nesslerize and compare with standards as already described. Procedures for Sewage. METHOD I. Distill the free NHs by passing live steam through 100 cc. or less of the sample. Collect the distillate, and determine the free NHs in it. Add 5 cc. of H2S04 and 1 cc. of 10% CuSCU solution, and digest for one-half hour after the whole has become colorless. Add 0.5 gram of KMnO 4 crystals to the hot acid residue in the flask and dilute to 500 cc. in a volumetric flask. Place 10 cc. or more of this liquid in a 200 cc. Kjeldahl distilling flask. Dilute with 100 cc. 504 TECHNICAL METHODS OF ANALYSIS of water, neutralize with 10 cc. of 10% Na2COs solution, distill with steam, and nesslerize. In this determination care must be taken to digest thoroughly, to add KMn04 to the point of precipitation; to sample carefully after dilution, and to add enough Na 2 C03 to insure the separation of the NHa from the precipitated manganese hydroxide. KMnC>4 must not be added during digestion because it causes loss of N. METHOD II. Omit the separation of free NHs and determine both the N as free NHs and organic N exactly as described above. Upon a separate sample determine the free NHs by direct nessler- ization as already described. Subtract the latter to obtain the organic N. NITROGEN AS NITRITES General. The formation of nitrites is the second intermediate step by which nitrogenous matter passes from crude organic matter to mineral matter (nitrates) . Nitrites may be also formed by the reduction of nitrates. The following is the standard method of procedure for water and sewages: Reagents. (1) Sulfanilic Acid Solution. Dissolve 8 grams of the purest sulfanilic acid in 1000 cc. of 5 N acetic acid (sp. gr. 1.041). This is practically a saturated solution. (2) a-Amidonaphthalene Acetate Solution. Dissolve 5.0 grams of solid a-amidonaphthalene * in 1000 cc. of 5 N acetic acid and filter the solution through washed absorbent cotton. (3) NaN0 2 Stock Solution. Dissolve 1.1 gram of AgNO 2 in nitrite-free water; precipitate the Ag with NaCl solution and dilute the whole to 1 liter. (4) Standard NaN0 2 Solution. Dilute 100 cc. of solution (3) to 1 liter, then dilute 10 cc. of this solution to 1 liter with sterilized nitrite-free water, add 1 cc. of CHCls and preserve in a sterilized bottle. 1 cc. = 0.0000001 gram nitrogen. (5) Fuchsin Solution. 0. 1 gram per liter. Procedure. Measure out into a Nessler tube 100 cc. of the decolorized sample (decolorized by adding aluminum hydroxide free from nitrite see under Chlorine), or a smaller portion diluted to 100 cc. The Nessler tubes must be of clear white glass, with the * a-Naphthylamine. MISCELLANEOUS ANALYSES 505 100 cc. graduation marks not varying more than 6 mm. in distance above the bottom. At the same time make a set of standards by diluting various volumes of the standard nitrite solution in Nessler tubes to 100 cc. with nitrite-free water, for example, 0, 1, 2, 4, 7, 10, 14, 17, 20 and 25 cc. Add 2 cc. of reagents (1) and (2) above to each 100 cc. of the sample and to each standard. Mix and let stand ten minutes. Compare the samples with the standards. Do not allow the samples to stand over one-half hour before being compared. Make a blank determination in all cases to correct for the presence of nitrite in the air, the water, and the reagents. Dilute all samples which develop more color than the 25 cc. stand- ard before comparing. Mixing is important. When 100 cc. of the sample are used, then 0.0001 times the number of cc. of the standard gives the parts of N as nitrite per 100,000 parts of water. NOTES. (1) The solution must be acid. HC1, which formerly was in quite general use in this country as a solvent for the naphthylamine, permits satisfactory results to be obtained, but the speed of the reaction is much slower than in the case of acetic acid. For this reason the latter acid is preferred. (2) The nitrite standards made up as described above may, as an expedient in routine work, be matched by eye by diluting the fuchsin reagent (5) to the required depth of color. For waters high in nitrite and for all sewage work, these fuchsin standards have been found to be approximate. They should be checked once a month and if kept out of the bright sunlight are more permanent than the dilute nitrite standard. (3) For cases not involving court testimony, the work can be considerably shortened by comparing the sample with colored squares of paper printed on a white background instead of with the standard nitrite tubes, NITROGEN AS NITRATES General. No single method appears to be applicable to all classes of water and sewage and there is no method which is not subject to considerable error. Where the amount of Cl is less than 3 parts per 100,000, the phenolsulfonic acid method is recom- mended; in other cases the reduction method, particularly in sewage work. Phenolsulfonic Acid Method (for Waters Low in Cl). REAGENTS. (1) Phenolsulfonic Acid. Mix 30 grams of pure white synthetic phenol with 370 grams of c. p. cone. H2S04 in a round-bottom flask. Put this flask in a water bath, supported in 506 TECHNICAL METHODS OF ANALYSIS such a way that it is completely immersed in the water, and heat for six hours. (2) Ammonium Hydroxide (1 : 1). Dilute cone. NtLiOH with an equal volume of water. (KOH may be used.) (3) Standard Nitrate Solution. Dissolve 0.72 gram of pure recrystallized KNOs in 1 liter of nitrate-free distilled water. Evaporate cautiously 10 cc. of this strong solution to dryness on the water bath in a porcelain dish. Moisten quickly and thor- oughly with 2 cc. of phenolsulfonic acid, stirring with a small glass rod, and dilute to 1 liter for the standard solution. 1 cc. = 0.000001 gram nitrogen. PROCEDURE. Evaporate 20 cc. of the sample in a small porcelain evaporating dish on the water bath, removing it from the bath just before it has come to dryness. Let the last few drops evaporate at room temperature in a place protected from dust. When the sample is suspected to contain a large amount of nitrate, evaporate less than 20 cc. If it is suspected to contain but little, evaporate more. If the sample has a high color, decolorize before evaporating by the use of washed A1(OH)3, as directed under the determination of Chlorine. Add 1 cc. of phenolsulfonic acid and rub this quickly and thor- oughly over the residue with a glass rod. Add about 10 cc. of distilled water and stir with the glass rod until mixed. Add enough NH^OH solution (or KOH if the operation must be carried on in a room where NHs distillations are made) to render the liquid alkaline. Transfer the liquid to a 100 cc. Nessler tube and fill the tube to the 100 cc. mark with distilled water. If nitrates are present there will be formed a yellow color. This may be compared with permanent standards made for the purpose by putting the following quantities of the standard solution into 100 cc. tubes and making up to the 100 cc. mark with dis- tilled water, adding 5 cc. of cone. NHiOH or KOH to each tube; namely, 0, 1, 2, 4, 7, 10, 15, 20, 25, 30, 35 and 40 cc. These standards may be kept for several weeks. Compare the sample treated as above described with these standards by looking down vertically through the tubes at a white surface so placed in front of a window that it will reflect the light upward through them. Divide the figure (cc. of standard) obtained by this comparison by 10 times the number of cc. of the water evaporated. This will MISCELLANEOUS ANALYSES 507 give the parts of N in the form of nitrates in 100,000 parts of water. Reduction Method (for Waters and Sewage High in Cl). REAGENTS. (1) NaOH or KOH Solution. Dissolve 250 grams of the caustic in 1250 cc. of distilled water. Add several strips of Al foil and let the action proceed overnight. Boil down to 1 liter. (2) Aluminum Foil. Use strips of pure Al about 10 cm. long, 6 mm. wide and 0.3 mm. thick, and weighing about 0.5 gram. PROCEDURE. Measure 100 cc. of the sample into a 300 cc. casserole. Add 2 cc. of the caustic solution and boil down to about 20 cc. Pour the contents of the casserole into a test tube about 6 cm. long and 3 cm. in diameter and of about 100 cc. capacity. Rinse the casserole several times with N-free water and add the rinsings to the solution already in the tube, making the volume approximately 75 cc. Add a strip of Al foil, and close the tube with a rubber stopper through which passes a glass tube about 5 mm. in diameter, bent into a " V." Make the short end of the tube flush with the lower side of the rubber stopper while the other end extends below the surface of distilled water contained in another test-tube. This apparatus serves as a trap through which the evolved hydrogen escapes freely. The amount of NHs escaping into the trap is slight and may be neglected. Let the action proceed for a minimum period of four hours, or overnight. Pour the contents of the tube into a distilling flask, dilute with 250 cc. of ammonia-free water, distill, collect in Nessler tubes and nesslerize. When the nitrate content is high, collect the distillate in a 200 cc. flask and nesslerize an aliquot portion. If the super- natant liquid in the reduction tube is clear and colorless, the solu- tion may be diluted to a definite volume and an aliquot part nesslerized without distillation. OXYGEN CONSUMED General. " Oxygen consumed " means the oxygen which the organic compounds of sewage and water consume when treated in an acid solution with KMnOi. It is also called " oxygen re- quired " and " oxygen absorbed." It is the C and not the N in organic matter which is thus oxidized by KMnO4. This deter- mination is hence frequently referred to as an indication of the carbonaceous organic matter present. It indicates only a certain 508 TECHNICAL METHODS OF ANALYSIS portion of the carbon, however, varying in different samples of water and of sewage. Furthermore, it does not directly differ- entiate the C present in unstable organic matter from that which might be called fairly stable organic matter, such as is sometimes referred to as residual humus matter. If nitrates, ferrous salts, sulfides or other unoxidized mineral compounds are present, they will increase the O consumed and a correction should be made for them when studying carbonaceous organic matter. The determination of oxygen consumed is an empirical pro- cedure and details must be strictly followed to obtain concordant results. Reagents. (1) Dilute Sulfuric Add. One part of cone. H2SO4 to 3 parts of distilled water. This must be freed from oxidizable matters by adding KMnC>4 until a faint pink color persists after standing several hours. (2) Standard KMnO* Solution. Dissolve 0.4 gram of the crystals in 1 liter of distilled water. Standardize against the oxalate solution. 1 cc. is equivalent to approximately 0.0001 gram of available oxygen. (3) Ammonium Oxalate Solution. Dissolve 0.888 gram of (NH 4 )2C204-H 2 O in 1 liter of distilled water. 1 cc. = 0.0001 gram of oxygen. Procedure. Measure into a flask 100 cc. of the water, or if high in organic content, a similar portion diluted to 100 cc. Add 10 cc. of the H2S04 solution and bring to the boiling point. Then add 10 cc. of KMnC^ solution and boil for exactly five minutes, agitating the liquid constantly with a small current of air to guard against bumping. Then immediately add 10 cc. of the ammo- nium oxalate solution and titrate hot with the KMnC^ solution until a faint but distinct pink is obtained. (Accurate 10 cc. pipettes should be used for measuring the liquids.) Run a blank under exactly the same conditions, using 100 cc. of distilled water in place of the sample. Subtract the KMnCU required by the blank from that required by the sample. One-tenth the difference expressed in cc. gives Oxygen Consumed in parts per 100,000. NOTES. (1) If 10 cc. of the KMnO 4 is insufficient for complete oxidation, repeat the determination, using 15 cc. or more. There should be an excess of at least 5 cc. of the KMnO 4 when the oxalate is added. (2) In connection with sewage works analysis, the KMnO 4 solution should MISCELLANEOUS ANALYSES 509 be added to the sample before heating in order to include the O consumed by volatile compounds. (3) If the sample contains sulfides, nitrites, or ferrous salts in appreciable quantity, correct the oxygen consumed figure for the KMnO 4 reduced by these as follows: Digest another 100 cc. portion of the water with 10 cc.* of the KMnO 4 at room temperature for three minutes. Add about 1 cc. of 10% KI solution (free from iodates), mix well, and titrate the liberated iodine with a weak thiosulfate solution (1.0 gram per liter). Run a blank on 10 cc. of the KMnO 4 and 100 cc. of distilled water in the same way to establish the relation between the KMnO 4 and the thiosulfate solution. The amount of KMnC>4 consumed by the 100 cc. of sample in the cold should be subtracted from the amount consumed hot, before calculating the Oxygen Consumed CHLORINE General. Chlorine in waters and sewages has its origin for the most part in the common salt, which comes, generally speaking, from mineral deposits in the earth, from the ocean vapors carried inland by the wind, or from polluting materials like sewage and trade wastes, which contain the salt used in the household and in manufacturing. Comparison of the chlorine content of a water with that of other waters in the general vicinity known to be unpolluted, frequently affords useful information as to its sanitary quality. Reagents. (1) Standard Salt Solution. Dissolve 16.50 grams of fused NaCl in distilled water and dilute to 1 liter. Dilute 100 cc. of this stock solution to 1 liter in order to obtain a standard solu- tion, each cc. of which contains 0.001 gram of Cl. (2) Standard AgNOs. Dissolve 2.396 grams of AgN0 3 crystals in 1 liter of distilled water. One cc. of this will be equiva- lent to approximately 0.0005 gram of chlorine. Standardize against the standard salt solution. (3) Potassium Chromate Indicator. Dissolve 50 grams of neutral K^CrCU in a little distilled water. Add enough AgNOs to produce a slight red precipitate. Filter and make up the filtrate to 1 liter with distilled water. (4) Aluminum Hydroxide. Dissolve 125 grams of potash alum or ammonia alum in 1 liter of distilled water. Precipitate the A1(OH) 3 by cautiously adding NH 4 OH. Wash the precipitate in a large jar by the successive addition of distilled water and by decantation until free from Cl, nitrites and NH 3 . * Or more if necessary. See Note (1). 510 TECHNICAL METHODS OF ANALYSIS Procedure. Use 50 or 100 cc. of the sample in a white 6-inch porcelain evaporating dish, where the Cl is not extremely low or very high. If the Cl is very high, use 25 cc. (or even a smaller quantity) diluting the volume taken to 50 cc. with distilled water. A satisfactory end-point cannot be obtained when more than 8-10 cc. of the AgNOs solution are required. When there are over 100 parts of Cl per 100,000, a gravimetric determination should be made. When the sample is very low in Cl more accurate results may be obtained by using 50 or 100 cc. of the sample and adding, prior to titration, 1 cc. of standard NaCl solution, correcting for this in the calculation. If the sample has a color greater than about 30, it must be decolorized by heating it to the boiling point with washed Al(OH)a (3 cc. to 500 cc. of the sample). Make the determination on a portion of the clarified sample, filtered, if necessary. Before titrating the Cl add 2 or 3 drops of phenolphthalein indicator. If a red color appears, neutralize with approximately 0.1 or 0.05 N H2SO4. If the water is acid to methyl orange add a slight excess of pure CaCOs (or just neutralize with 0.1 or 0.05 N Na 2 C0 3 ). Rotate the neutral liquid in the dish to make sure that no por- tion of any residue on the side walls remains undissolved, rubbing if necessary with a rubber-tipped glass rod. Add 1 cc. of the K2Cr04 indicator and titrate with the AgNOs solution, under similar conditions as to volume, light and temperature as were used in standardizing the AgNOs. The detection of the end- point is facilitated by frequent comparison of the contents of the porcelain dish in which the determination is being made with those of another dish placed alongside and containing the same quantity of chromate solution in the same volume of distilled water as the volume of the sample taken for titration. It is also preferable to make the titrations in a darkened room provided with a yellow light. HARDNESS General. The hardness of water is caused chiefly by the salts of Ca and of Mg. It is commonly measured by the soap-destroying power of the water. The addition of K or Na soap causes its MISCELLANEOUS ANALYSES 511 decomposition and produces insoluble Ca and Mg soaps. The solubility of carbonates of Mg and Ca in a water beyond certain limits depends upon the presence of CO2 and results in bicarbonates. On boiling the CC>2 is removed and the normal carbonates pre- cipitated. Precipitation, however, is not complete, as the normal carbonates themselves have a slight solubility. The hardness of the water removed by boiling is called " temporary hardness. " The hardness which still remains is termed " permanent hardness," and is due largely to sulfates and chlorides of Ca and Mg and to the traces of carbonates still held in solution. It is generally sufficient to determine the total hardness of the water. In case the permanent hardness is also desired, boil gently a known volume of the water for one-half hour, let cool, and then restore to its original volume with recently boiled and cooled distilled water. Then filter the water and determine the per- manent hardness in the filtrate by the soap method as below. This, subtracted from the total hardness, will give the temporary hardness. Total Hardness by the Soap Method. REAGENTS. (1) Stand- ard CaCk Solution. Dissolve 0.2 gram of pure calcite (CaCOs) in a little dil. HC1, being careful to avoid loss of solution by spatter- ing. Evaporate to dryness several times to expel excess of acid, dissolve in distilled water and make up to 1 liter. 1 cc. = 0.0002 gram CaCOs. (2) Standard Soap Solution. Dissolve 100 grams of dry white castile soap in 1 liter of 80% alcohol and let stand several days before standardizing. From the above stock solution dilute with 70% alcohol such a quantity that the resulting diluted soap solution will give a permanent lather when 6.40 cc. of it are properly added to 20 cc. of standard CaCb solution. Usually from 75-100 cc. of the stock soap solution are required for making 1 liter of the standard soap solution. Pure potassium oleate, made from lead plaster and K^COs, may be used to advantage in place of castile STANDARDIZATION. Pipette 20 cc. of the CaC ] 2 solution into a 250 cc. glass-stoppered bottle and dilute to 50 cc. with distilled water which has been recently boiled and cooled. Then add from a burette 0.2-0.3 cc. of soap solution at a time, shaking the bottle vigorously after each addition until a lather over the entire surface 512 TECHNICAL METHODS OF ANALYSIS of the water is formed, which remains continuous for five minutes after the bottle is laid upon its side. When the soap solution is of the strength above stated, then the quantity of CaC0 3 equivalent to each cc. of the soap solution is indicated in the following table: TABLE OF HARDNESS SHOWING THE PARTS OF CaCO 3 PER 100,000 FOR EACH 0.1 cc. OF SOAP SOLUTION WHEN 50 cc. OF THE SAMPLE ARE USED. Soap Solution 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 cc. cc. cc. cc. cc. cc. cc. cc. cc. cc. cc. 0.0 0.0 0.16 0.32 1.0 0.48 0.63 0.79 0.95 1.11 1.27 1.43 1.56 1.69 1.82 2.0 1.95 2.08 2.21 2.34 2.47 2.60 2.73 2.86 2.99 3.12 3.0 3.25 3.38 3.51 3.64 3.77 3.80 4.03 4.16 4.29 4.43 4.0 4.57 4.71 4.86 5.00 5.14 5.29 5.43 5.57 5.71 5.86 5.0 6.00 6.14 6.29 6.43 6.57 6.71 6.86 7.00 7.14 7.29 6.0 7.43 7.57 7.71 7.86 8.00 8.14 8.29 8.43 8.57 8.71 7.0 8.86 9.00 9.14 9.29 9.43 9.57 9.71 9.86 10.00 10.15 PROCEDURE. Measure 50 cc. of the water into a 250 cc. bottle and add soap solution in small quantities and in precisely the same manner as described under the standardization of the soap solution. From the result obtained, calculate from the table above the total hardness of the water, expressed as parts of CaCOs per 100,000. When adding the soap solution to waters containing Mg salts, it is necessary to avoid mistaking the false end-point * for the true one. Consequently, after the titration is apparently finished, read the burette and add about 0.5 cc. more of soap solution. If the end-point was due to Mg, the lather now disappears. Soap solution must then be added until the true end-point is reached. Usually the false lather persists for less than five minutes. When more than 7 cc. of soap solution are required for 50 cc. of the water, it is necessary to take less of the sample and dilute to 50 cc. with distilled water which has been recently boiled and cooled. This step reduces somewhat the disturbing influence of Mg salts, which consume more soap than do equivalent weights of * Also called the Magnesium end-point. MISCELLANEOUS ANALYSES 513 Ca salts. At best the soap method is not a precise test on account of the varying amounts of Ca and Mg present in different waters. For hard waters, especially in connection with processes for puri- fication and softening, it is advisable not to use this method but to calculate the total hardness (in terms of CaCOs) from the amount of CaO and MgO found by chemical analysis. NOTES. (1) When free CO 2 is present in the sample in considerable amount, it should be removed by aeration. (2) The strength of the soap solution should be determined from time to time, to make sure that it has not materially changed while standing. Unless otherwise stated, record all results in terms of CaCO 3 . (3) English degrees of hardness, Clark's scale, are equivalent to grains of CaCO 3 per imperial gallon, and are multiplied by 1.43 to give parts per 100,000. (4) French degrees of hardness represent parts of CaCO 3 per 100,000, and are the same as obtained by the above table. (5) German degrees of hardness represent parts of CaO per 100,000 and are multiplied by 1.78 to give parts of CaCO 3 per 100,000. IRON General. Iron in ground waters is usually in the soluble fer- rous state, sometimes as carbonate or sulfate and also combined as organic matter. Most waters, especially those which have been exposed to the air, contain the iron in the form of a colloidal hydroxide. Total Iron. REAGENTS. (1) Standard Fe Solution. Dissolve 0.7 gram of Fe(NH4)2(S04)2-6H 2 in 50 cc. of distilled water and add 20 cc. of dil. H 2 S0 4 . Warm slightly and add KMnO 4 , little by little, until the Fe is completely oxidized. Dilute to 1 liter. 1 cc. = 0.0001 gramFe. (2) KSCN Solution. Dissolve 20 grams in 1 liter of distilled water. (3) Dilute HCl.Md to cone. HC1 (free from HN0 3 ) an equal volume of distilled water. (4) 0.2N KMnO. Dissolve 6.30 grams in distilled water and dilute to a liter. (5) Cone. HCl. Free from iron. PROCEDURE. Evaporate 100 cc. of the sample to dryness. (Use the residue from the total solids determination, if convenient.) If it contains much organic matter destroy this by gentle ignition. 514 TECHNICAL METHODS OF ANALYSIS Cool and add 5 cc. of cone. HC1,* moistening the whole of the inner surface of the dish. Warm for two to three minutes and again moisten the whole inner surface with acid. Rinse down the sides of the dish with 5-10 cc. of distilled water and let stand on the water bath for about three minutes. Wash the hot acid solution into a 100 cc. Nessler tube. Filter the sample if necessary, first washing the filter paper with hot water. Add a drop or two of KMn04 solution to be sure that the Fe is oxidized. The pink color should persist for at least five minutes. If not, add more, a drop at a time. To the cooled solution add 10 cc. of KSCN solution, dilute to 100 cc. and mix thoroughly. Compare the color immediately with a series of standards, prepared side by side with the sample in 100 cc. Nessler tubes. These standards should be prepared from the standard Fe solution, containing amounts ranging from 0.5-4.0 cc. of the latter. Dilute these amounts with distilled water to about 50 cc. Add 5 cc. of cone. HC1 and a drop or two of KMnO4 and then 10 cc. of KSCN solution. Finally dilute all the standards to 100 cc. NOTE. For a single sample, it is more convenient to run standard Fe solu- tion from a burette into a Nessler tube containing the acid, distilled water and KSCN, until, after mixing, the color matches the sample. From the reading of the burette, calculate the amount of Fe. When using standards, the color comparisons must be made immediately. REFERENCES. American Public Health Association, " Standard Methods of Water Analysis," 1913. J. Assoc. Official Agr. Chemists, Methods of Analysis (1916-17), pages 35-39, INDUSTRIAL WATER General. This method applies to the mineral analysis of waters to be used for boilers and for general manufacturing pur- poses. It is fully realized that it is open to criticism from a strictly chemical standpoint (e.g., it takes no account of the possible presence of potassium salts or bicarbonates) but the results give the desired information for practically all technical purposes. The quantity of water which must be taken for analysis depends upon the amount of dissolved solids which it contains. As a general rule, about 1 liter is a suitable amount of water for * Dilute HC1 will not dissolve all the Fe 2 O 3 . MISCELLANEOUS ANALYSES 515 evaporation. One U. S. gallon of pure water at 60 F. contains 58,334.9 grains. If, therefore, 583.3 cc. of water are taken for analysis, the weight in centigrams of the various substances found will give directly the number of grains per U. S. gallon. For ordinary waters, therefore, evaporate 1167 cc. and divide by 2 the number of centigrams obtained. The results will be grains per gallon. It is convenient to have several flasks specially calibrated to contain 1167 cc., 583.3 cc., 291.7 cc. and 116.7 cc., respectively. These may be prepared by selecting ordinary volumetric flasks of 1000 cc., 500 cc., 250 cc. and 100 cc., respectively, choosing flasks where the graduation mark is low on the neck. Weigh each flask, place in it the proper number of grams of water at 20 C. and make a new mark at the meniscus of the water; or, fill each to its original mark, add the proper number of cc. of water from a burette to bring the volume to the proper figure, and make the new mark, preferably by etching. A suitable flask for the largest size is the kind with a bulb in the neck and 2 graduations at 1000 and 1100 cc. If 1000 cc. are taken for analysis, multiply the number of cen- tigrams found by 0.5833 to obtain grains per gallon. Free Carbon Dioxide. To 291.7 cc. of the sample (measured in a specially graduated flask) contained in a 400 cc. Erlenmeyer flask add 1 cc. of phenolphthalein indicator solution, and titrate rapidly with 0.01 N Na 2 COs till a faint pink color persists after gentle shaking. Take care not to breathe into the flask and do not use a flask unnecessarily large. Calculate the titration to CO2, according to the reaction Na 2 CO 3 + C0 2 + H 2 O = 2NaHC0 3 . CALCULATION. 1 cc. 0.01 N Na 2 CO 3 = 0.00022 gram C0 2 . Multiply by 2 the number of centigrams of CO 2 obtained and the result will be grains per U. S. gallon. NOTE. If the sample contains over 46 grains of NaHCO 3 per gallon or if the total carbonate hardness is over 10 (corresponding to the equivalent of 5.8 grains of CaCOs per gallon) it must be diluted with CCVfree water before titration, to prevent the precipitation of CaCO 3 and MgCO 3 with consequent liberation of COa. 516 TECHNICAL METHODS OF ANALYSIS Suspended Matter. If the sample is decidedly turbid or contains an appreciable amount of suspended matter, shake thoroughly and measure out 1167 cc. in a specially graduated flask. Filter through a filter paper* which has previously been dried at 100 C. and weighed in a weighing bottle, collecting the clear filtrate in a clean beaker or flask. Wash the residue on the filter paper once with distilled water and dry at 100 C. in the same weighing bottle. Cool in a desiccator and weigh. Divide by 2 the number of centigrams to obtain the grains of total suspended matter per gallon. Ignite the filter paper in a weighed platinum crucible, dry in a desiccator and weigh. This weight will give the amount of non- volatile suspended matter. Total Solids. Evaporate the filtrate obtained above in a weighed platinum dish on the steam bath. Finally dry in the steam oven at 100 C. for about one-half hour. Cool in a desic- cator and weigh. NOTE. The evaporation may be hurried by gentle boiling on a hot plate or over a free flame on an asbestos mat, taking care not to lose any by spattering and not to let it go to dryness before transferring to the steam bath. In urgent cases clamp a horizontal glass tube about 0.5 inch above the top of the dish, with the open end at about the center, and connect the other end to suction or a gentle air blast to remove the vapors. Organic and Volatile Matter. Manipulate the dish in a Tirrill flame with a pair of tongs until the organic matter is burned off at as low a temperature as possible. Cool in a desiccator and weigh. Report the loss as " organic and volatile matter." NOTE. This result will generally be too high, due to loss of CO 2 from car- bonates, and is used only as a rough check on the analysis, the actual organic and volatile matter in the final calculation being taken "by difference." Total Mineral Matter as Sulfates. Add a few drops of dil. H2S04 to the above residue in the platinum dish (note here whether there is an effervescence of carbonates). Manipulate the dish so that the acid will come in contact with all of the residue. *A Gooch crucible may be used if preferred, although, if the water con- tains much organic matter or fine silt, these may have a tendency to clog the asbestos mat. MISCELLANEOUS ANALYSES 517 If there is a strong effervescence, it will be advisable to add a few more drops of the acid. Evaporate to dryness on the steam bath and then to fumes of S0 3 on the hot plate. (If no white fumes are obtained, insufficient H2S04 has been added.) Finish the heating carefully over a free flame until the evolution of S0 3 fumes ceases. Then heat for some time at a red heat to decompose any FeSO4 into Fe20 3 . Cool in a desiccator and weigh. The combined weight represents the mineral matter as Fe2O 3 , SiO 2 , A1 2 3 , CaSO 4 , MgSC>4 and alkaline sulfates. For all ordinary purposes the latter may be considered Na2S04. Silica. Add a few cc. of cone. HC1 to the above residue and warm until all the Fe2O 3 dissolves. Dilute with about twice the volume of water and continue the heating for a short time. During the heating loosen the insoluble matter adhering to the dish with a rubber policeman and stir to hasten the solution. Filter through a small ashless filter and wash the residue with hot water. Place the filter paper containing the insoluble matter in the original platinum dish and burn off the filter paper, taking care to protect the dish from drafts of air. Cool in a desiccator and weigh as SiC>2. NOTE. If the water is to be used for boiler purposes, it is not necessary to determine the silica separately unless a large amount is present. Iron Oxide and Alumina. Heat the filtrate from the Si0 2 (or, in case the SiO2 is not to be determined separately, the unfil- tered solution of the .sulfates) nearly to boiling. Add a slight excess of NEUOH and digest on the steam bath until the odor of NH 3 is nearly gone. Filter through a small filter. Ignite strongly in a platinum crucible, cool in a desiccator and weigh as Fe 2 3 +Al 2 3 (+Si0 2 ). NOTE. If Mn is suspected to be present, add 5 cc. of strong bromine water and boil before adding the NH 4 OH to precipitate iron and alumina. In this case the Mn will come down with the iron and alumina after treatment with bromine water and NH^OH and if it is desired to know the amount, it must be determined in a separate portion of the water and the amount (calculated as Mn 3 O 4 ) subtracted from the total weight of the NH 4 OH precipitate. (See pages 110 and 148, under Manganese.) Iron Oxide. If it is desired to determine the Fe 2 O 3 separately, fuse the residue in the crucible with a little KHS04. Dissolve the 518 TECHNICAL METHODS OF ANALYSIS fusion in water and determine the Fe20s colorimetrically with KSCN, matching the color against a standard Fe solution. (See under Total Iron on page 513.) Alumina. Ordinarily A1 2 O3 is not determined separately. If, however, the colorimetric determination of Fe leaves considerable to be accounted for in the NH^OH precipitate, the difference may be considered as A^Os. Lime. Heat to boiling the filtrate from the NH^OH precipitate. Add 5 cc. of (NEU) 20264 reagent solution and digest on the steam bath until the precipitate has settled. Filter through a small ashless filter, wash well with hot water, ignite thoroughly in a platinum crucible and weigh as CaO. NOTE. Instead of igniting the CaC 2 O 4 it may be titrated with 0.1 N KMnO 4 as follows: In a 400 cc. beaker place about 125 cc. of distilled water and add 5-7 cc of cone. H 2 SO 4 . Drop the moist filter paper containing the CaC 2 O 4 into this. and heat to about 70 C. Stir to effect the decomposition of the CaC 2 O 4 but avoid excessive disintegration of the paper. Titrate the hot solution, with constant stirring, with 0.1 N or 0.01 N KMnO 4 until a permanent pink color forms. CALCULATION. 1 cc. 0.1 N KMnO 4 = 0.002804 gram CaO. Magnesia. Make the filtrate from the CaO determination slightly acid with HC1 and evaporate until crystallization starts. Dissolve any crystals formed with a small amount of distilled water. Cool, and add 25 cc. of a 10% solution of (NH^HPO^ (It is permissible to use Na 2 HPO 4 or NaNH4HP0 4 .) Then add slowly with constant stirring cone. NELjOH until neutral, finally adding about 2 cc. in excess. This is best done from a burette or a Mohr pipette. Let stand overnight, if possible, or cool in ice water and stir for one-half hour. Filter through a weighed Gooch crucible and wash with magnesia wash solution.* Dry the precipitate and then finally ignite strongly, cool and weigh as Mg 2 P20 7 . Calculate to MgO. CALCULATION. Mg 2 P2O7X 0.3621 = MgO. Alkalies. For all ordinary purposes the alkalies may be regarded as consisting wholly of Na 2 O and are obtained by cal- culation as described below. *Magnesia wash solution: Add 200 grams of NH 4 N0 3 to 400 cc. of cone. NH 4 OH; mix and dilute with water to 1 liter. MISCELLANEOUS ANALYSES 519 Sulfur Trioxide. To 291.7 cc. of the sample (measured in a specially graduated flask), contained in a 400 cc. beaker, add 5 cc. of dil. HC1. Heat to boiling and then add, drop by drop, 5 cc. of 10% BaCb solution. Let stand overnight, protected from SO 3 fumes. Filter, wash with hot water, ignite the precipitate and weigh as BaS0 4 . Calculate to SO 3 and multiply the number of centigrams of SO 3 by 2 to obtain grains per gallon. CALCULATION. BaSO 4 X 0.3430 = SO 3 . Chlorine. Determine chlorine by the procedure given on page 509. If 116.7 cc. of the water are used, then the number of cc. of the standard AgN0 3 solution, divided by 4, will give directly grains of Cl per U. S. gallon. If the water contains over 60 grains of Cl, make the determi- nation gravimetrically, weighing the AgCl. Calculation of Results. Calculate the CaO and MgO to CaSO 4 and MgSO4, respectively. Add to the sum of these the amount of Fe2O 3 , SiC>2 and A^Os and subtract this sum from the total weight of mineral matter calculated as sulfates, as previously determined. Consider the difference as Na2S0 4 and calculate the equivalent amount of Na20. Calculate the total Cl to NaCl, if there is enough Na2O to combine with it. If, however, there is an excess of Cl over the Na 2 O, calculate the remainder to MgCl 2 . If the Na 2 O is in excess of the Cl, calculate the excess to Na2SO 4 and then the remainder, if any, to Na2C0 3 . If there is insufficient Na2O left from NaCl to combine with all the SOs, calculate the excess of the latter to CaSO 4 . If the SO 3 is still in excess, combine it with any MgO not satisfied with Cl, and if there is still an excess, calculate it to H2SO 4 . (In this case, of course, the water would show an acid reaction to methyl orange and litmus.) Calculate any MgO or CaO not accounted for by Cl and SOs to CaCO 3 and MgCO 3 , respectively. If the Na 2 O is in excess of both Cl and SO 3 , calculate the excess to Na 2 C03. The Fe2O 3 is probably present as FeSO 4 but for practical purposes this calculation is unnecessary. The SiO2 is usually reported as such, although in water containing a considerable amount, and also containing Na2CO 3 ,* it may be present as sodium silicate. * Water containing Na 2 CO 3 will be alkaline to phenolphthalein, although, even if the water is neutral to this indicator, it may still contain NaHCO 3 , in which case it will be alkaline to methyl orange. 520 TECHNICAL METHODS OF ANALYSIS Report the final results as follows : Grains per U. S. gallon Free Carbon Dioxide (CO 2 ) Suspended Matter: Organic and Volatile Non-volatile Total On Filtered Sample Silica (SiO 2 ) Alumina (A1 2 O 3 ) Iron Oxide (Fe 2 O 3 ) Lime (CaO) Magnesia (MgO) Sodium Oxide (Na 2 O) Sulfur Trioxide (SO 3 ) Chlorine (Cl) . Probably combined as follows: Silica (SiO 2 ) Iron Oxide (Fe 2 O 3 ) Alumina ( A1 2 O 3 ) Sodium Chloride (NaCl) Sodium Sulfate (Na 2 SO 4 ) Sodium Carbonate (Na 2 CO 3 ) .... Magnesium Chloride (MgCl 2 ) . . . Magnesium Sulfate (MgSO 4 ) .... Magnesium Carbonate (MgCO 3 ) . Calcium Chloride (CaCl 2 ) ' Calcium Sulfate (CaSO 4 ) Calcium Carbonate (CaCO 3 ) Residue on Evaporation NOTES. (1) In waters which are to be used for boiler purposes it is suf- ficient ordinarily to report Fe 2 O 3 , A1 2 O 3 and SiO 2 together, and not separately. (2) The following are incrusting or scale-forming solids : Oxides of iron and aluminum. Calcium sulfate and carbonate. Calcium chloride (corrosive). Magnesium sulfate (forms scale only in the presence of CaCO 3 ). Magnesium chloride (strongly corrosive). Magnesium carbonate. The following are non-incrusting solids: Chlorides, sulfates and carbonates of the alkalies. Organic matter. MISCELLANEOUS ANALYSES 521 (3) The general grading of waters for boiler purposes, according to the contents of scale-forming matter is as follows: 10-20 grains Fair 20-30 grains Poor 30-40 grains Bad Over 40 grains Very bad (4) Grains per gallon X-V-= lbs - P er 10,000 gallons. Factors. The following factors will be found useful in cal- culating results: Given i Wanted Factor Log CaCl 2 CaO 0.5052 9.70346 CaO CaCO 3 1.7849 0.25161 CaS0 4 2.4279 0.38523 CaS0 4 CaO 0.4119 9.61479 Cl CaCl 2 1.5650 0.19451 KC1 2.1025 0.32274 MgCl 2 1.3429 0.12805 NaCl 1.6486 0.21712 CO 2 NaaCO, 2.4090 0.38184 Fe 2 3 FeCO 3 1.4510 C. 16167 KC1 K 2 O 0.6317 9.80050 K 2 SO 4 1 . 1686 0.06766 K 2 O K 2 CO 3 1.4671 0.16647 K 2 SO 4 1.8499 0.26715 MgCl 2 MgO 0.4234 9.62675 MgO MgC0 3 2.0915 0.32046 MgS0 4 2.9857 0.47504 Mn 3 O 4 MnCO 3 1.5071 0.17814 NaCl Cl 0.6066 9.78290 Na 2 0.5303 9.72452 522 TECHNICAL METHODS OF ANALYSIS Given Wanted Factor Log Na 2 CO 3 Na 2 SO 4 1 . 3401 0.12713 Na 2 O NaCl Na 2 CO 3 Na 2 Si0 3 Na 2 SO 4 1.8858 1.7098 1.9726 2.2913 0.27550 0.23295 0.29504 0.36008 Na 2 S0 4 Na 2 CO 3 Na 2 O S0 3 0.7462 0.4364 5636 9.87286 9 . 63989 9.75097 Si0 2 C0 2 * Na 2 SiO 3 0.7298 2.0282 9.86320 0.30711 S0 3 CaSO 4 MgS0 4 Na 2 SO 4 1.7004 1.5036 1.7744 0.23055 0.17712 0.24905 REFERENCES. Low: "Technical Methods of Ore Analysis." Tillmann and Henblein, Z. Nahr. Genussm., 24, 429; C. A. 7, 38. BOILER SCALE General. Break up pieces as representative as possible of the scale and grind in an iron or porcelain mortar. Then quarter down and pulverize a small sample finely in an agate mortar for analysis. Moisture. Dry 5 grams of the powder at 105 C. to constant weight. Report the loss in weight as moisture. Oil. Transfer the dry powder left from the moisture deter- mination to an extraction thimble and extract with ether in a Soxhlet extractor in the usual way, collecting the extract in a weighed Soxhlet flask. Distill off the ether and dry the flask at 100 C. to constant weight. Organic and Volatile Matter. Weigh out 1 gram of the pow- dered material in a platinum crucible and gently ignite until the organic matter is burned off. Cool the crucible in a desiccator and weigh. From this weight subtract the moisture previously MISCELLANEOUS ANALYSES 523 determined and report the difference as Organic and Volatile Matter. [See note (2).] Silicious Matter. Transfer the residue from the above deter- mination with a little water to a 250 cc. beaker. Cover the beaker with a watch glass and add HC1 in excess. Dilute to abo'ut 150 cc., heat to boiling and filter. Wash thoroughly with hot water. Ignite the residue in a platinum crucible, first over a burner and then in a blast lamp. Cool in a desiccator and weigh. Iron and Aluminum Oxides. To the filtrate from the silicious matter add a few drops of cone. HNOs and heat to boiling. Add 10 cc. of NILiCl solution and then NH4OH in excess. Boil until there is only a faint odor of NHs. Filter and wash thoroughly. Ignite, blast, and weigh as Fe2O3+Al2Os. Lime. To the filtrate from the above add a few cc. of NEUOH, heat to boiling and add (NH4)2C2O4 solution in excess. Boil until the precipitate becomes granular. Let stand in a warm place until the solution becomes clear. Filter and wash with hot water. Ignite the precipitate first in a Tirrill burner and then in a blast lamp to constant weight. Cool in desiccator and weigh rapidly as CaO. (If preferred, the CaC2O4 may be titrated with 0.1 N KMn04 as on page 518.) Magnesia. Make the filtrate from the CaO determination slightly acid with HC1 and evaporate until crystallization begins. Add sufficient water to just redissolve any crystals which form. Cool thoroughly. Add at least 10 cc. of 10% sodium phosphate or ammonium phosphate solution. Stir briskly with a stirring rod and rubber policeman until precipitation starts, and add about one-quarter the volume of strong NKiOH. Let stand overnight, if possible. If the analysis is urgent, place the beaker in ice water and stir for one-half hour. Filter the precipitate on a weighed Gooch crucible and wash with magnesia wash water (see page 518). Dry the precipitate in the oven and ignite strongly over a Tirrill burner to constant weight and weigh as Mg2?207. Calculate to MgO. CALCULATION. Mg 2 P20 7 X0.3621 = MgO. Sulfur Trioxide. Boil 1 gram of the finely powdered material with cone. HC1, dilute with two volumes of water and boil again. Filter and wash thoroughly with hot water. Heat the filtrate to boiling and add slowly 10 cc. of 10% BaCl2 solution. Boil gently 524 TECHNICAL METHODS OF ANALYSIS until the precipitate settles clear. Filter, wash free from chlorides with hot water, ignite the precipitate, and weigh as BaSO 4 . Calculate to SO 3 . CALCULATION. BaS0 4 X 0.3430 = SO 3 . Chlorine. Weigh out 5 grams of the dry powdered scale. Add 100 cc. of water and heat to boiling. Filter and wash thoroughly, catching the filtrate in a porcelain dish. Cool, add a few drops of K 2 Cr0 4 indicator and titrate with standard AgN0 3 solution to the appearance of a reddish color. Since the amount of chlorides present is generally small, a weak AgN0 3 solution (0.01 N) should be used in titrating. If the scale forms an alkaline solution, it must be exactly neutralized with dil. H 2 S04 and phenolphthalein before titrating. CALCULATION. 1 cc. of 0.01 N AgN0 3 = 0.000355 gram Cl. Final Calculations. Calculate the Cl (if any is present) to NaCl. Calculate the SO 3 to CaS0 4 ; and if an excess of S0 3 remains, calculate it to MgS0 4 . In case the SO 3 is insufficient to combine with the CaO, calculate the remainder of the CaO to CaCO 3 . Report any MgO which is in excess of the S0 3 as MgO and not as MgC0 3 , since the latter is decomposed at the tempera- ture of the boiler. NOTES. (1) Other substances such as copper, zinc, etc., from local sources are occasionally found in boiler scales. If such are present, they may be determined by the usual quantitative methods. (2) The determination of organic and volatile matter is usually too high, owing to the partial loss of CO2 from carbonates, so it is generally advisable to take the organic and volatile " by difference " merely using the actual deter- mination as a rough check upon it. Any oil present is, of course, included in the organic and volatile matter. FERTILIZERS Mechanical Analysis of Bone and Tankage. Transfer 100 grams of the original material to a sieve having circular openings 0.02 inch (0.5 mm.) in diameter and sift, breaking the lumps by means of a soft rubber pestle if the material has a tendency to cake. Weigh the coarse material remaining on the sieve and determine the fine portion by difference. Preparation of Sample. Reduce the gross sample by quarter- ing to an amount sufficient for analysis. Transfer to a sieve having circular openings 0.04 inch (1 mm.) in diameter and sift, breaking MISCELLANEOUS ANALYSES 525 the lumps with a soft rubber pestle. Grind in a mortar the part remaining in the sieve until all the particles will pass through. Mix thoroughly and preserve in tightly stoppered bottles. Grind and sieve as rapidly as possible to avoid loss or gain of moisture. Moisture. Heat 2 grams, prepared as above, for five hours in a water oven at the temperature of boiling water. With potash salts, NaN0 3 , and (NEU) 2 SO 4 heat 1-5 grams at about 130 C. to constant weight. The loss in weight is considered as moisture. Nitrogen. Test for nitrates as follows: Mix 5 grams of the fertilizer with 25 cc. of hot water and filter. Cool and add to a portion of the solution 2 volumes of cone. H2S04, free from HNOs and nitrous oxides. Let the mixture cool and add cautiously a few drops of a cone, solution of FeSO4 down the side of the tube so that the fluids do not mix. If nitrates are present, the junction shows at first a purple, then a brown color; or if only a very minute quantity is present, a reddish color. To another portion of the solution add 1 cc. of a 1% solution of NaNOs and test as before to determine whether sufficient H2S04 was added in the first test. If nitrates are present, determine the various forms of nitrogen as described on page 64. If nitrates are absent, determine merely organic and ammoniacal nitrogen Phosphoric Acid (Gravimetric Method). PREPARATION OF REAGENTS. (a) Ammonium citrate solution. Dissolve 370 grams of commercial critic acid in 1500 cc. of water; nearly neutralize with commercial NHiOH; cool, add NILiOH until exactly neutral (testing with litmus or azolitmin paper), and dilute sufficiently to make the sp. gr. 1.09 at 20 C. The volume "will be about 2 liters. (b) Molybdate solution. Dissolve 100 grams of molybdic acid in 144 cc. of cone. NHiOH, and 271 cc. of water; slowly and with constant stirring, pour the solution thus obtained into 489 cc. of cone. HNOs and 1148 cc. of water. Keep the mixture in a warm place for several days, or until a portion heated to 40 C. deposits no yellow precipitate of ammonium phosphomolybdate. Decant the solution from any sediment and preserve in glass-stoppered bottles. (c) Ammonium nitrate solution. Dissolve 200 grams of commercial NHiNOs, free from phosphate, in water and dilute to 2 liters. 526 TECHNICAL METHODS OF ANALYSIS (d) Magnesia mixture. Dissolve 22 grams of recently ignited calcined MgO iirdil. HC1, avoiding an excess of acid. Add a little calcined MgO in excess, and boil a few minutes to precipitate Fe, Al and P 2 O 5 ; filter; add 280 grams of NILtCl, 261 cc. of cone. NELiOH, and dilute to 2 liters. Instead of the solution of 22 grams of calcined MgO in HC1, 110 grams of crystallized MgCl2-6H2O may be used. (e) Dilute NHOH for washing. Dilute 100 cc. of cone. NILtOH to 1 liter. (/) Magnesium nitrate solution. Dissolve 320 grams of cal- calcined MgO in HNOs, avoiding an excess of acid; then add a little calcined MgO in excess; boil, filter from the excess of MgO, Fe(OH)s, etc., and dilute with water to 2 liters. TOTAL PHOSPHORIC ACID. (a) Methods of making solution. Treat 2.5 grams of the sample by one of the methods given below. After solution, cool, dilute to 250 cc., mix, and pour on a dry filter. (I) Ignite and dissolve in HCL (II) Evaporate with 5 cc. of Mg(NOs)2 solution, ignite and dissolve in HC1. (III) Boil with 20-30 cc. of cone. H 2 SO 4 in a Kjeldahl flask, adding 2-4 grams of NaNOa or KNOs at the beginning of the digestion and a small quantity after the solution has become nearly colorless, or adding the nitrate in small portions from time to time. After the solution is colorless add 150 cc. of water and boil for a few minutes. (IV) Digest in a Kjeldahl flask with strong H2S04 and such other reagents as are used in either the plain or modified Kjeldahl or Gunning method for estimating nitrogen. (See page 64.) Do not add any KMnO4, but after the solution has become color- less add about 100 cc. of water and boil for a few minutes. (V) Dissolve in 30 cc. of cone. HNOs and a small quantity of HC1 and boil until organic matter is destroyed. (VI) Add 30 cc. of cone. HC1, heat, and add cautiously, in small quantities at a time, about 0.5 gram of finely pulverized KClOa to destroy organic matter. (VII) Dissolve in 15-30 cc. of cone. HC1 and 3-10 cc. of HNO 3 . This method is recommended for fertilizers containing much iron or aluminum phosphate. MISCELLANEOUS ANALYSES 527 (b) Determination. Take an aliquot of the solution prepared above corresponding to 0.25 gram, 0.50 gram, or 1 gram; neutralize with NH4OH and clear with a few drops of HNOs. In case HC1 or H 2 SO4 has been used as a solvent, add about 15 grams of dry NH4NOs or a solution containing that amount. To the hot solution add 60-80 cc. of molybdate solution for every 0.1 gram of P2Os that is present. Digest at about 65 C. for one hour and test the clear supernatant liquor for complete precipitation by the addition of more molybdate solution. Filter and wash with cold water or, preferably, with NEUNOs solution. Dissolve the pre- cipitate on the filter with NH^OH and hot water and wash into a beaker to a bulk of not more than 100 cc. Nearly neutralize with HC1, cool, and add magnesia mixture from a burette, slowly (about 1 drop per second), stirring vigorously. After fifteen minutes add 12 cc. of cone. NKUOH. Let stand till clear (two hours is usually enough), filter on a weighed Gooch crucible, wash with the weak NHs solution until practically free from chlorides; ignite to whiteness or to a grayish white and weigh as Mg2P20r. Calculate to P2Os. CALCULATION. Mg 2 P2O 7 X 0.6379 = P 2 O 5 . WATER-SOLUBLE PHOSPHORIC ACID. Place 2 grams of the sample on a 9 cm. filter, wash with successive small portions of water, allowing each portion to pass through before adding more, until the nitrate measures about 250 cc. If the nitrate be turbid, add a little HNOs. Make up to any convenient definite volume, mix well, use an aliquot * and proceed as under Total Phosphoric Acid, above. CITRATE-INSOLUBLE PHOSPHORIC ACID. (a) Determination in Acidulated Samples. Heat 100 cc. of strictly neutral ammonium citrate solution (sp. gr. 1.09) to 65 C. in a flask placed in a warm water bath, keeping the flask loosely stoppered to prevent evapo- ration. The level of the water in the bath should be above that of the liquid in the flask. When the citrate solution in the flask has reached 65 C., drop into it the filter containing the washed residue from the water-soluble P20s determination, close tightly with a smooth rubber stopper and shake violently until the filter paper is reduced to a pulp relieving the pressure by momentarily * An aliquot corresponding to 0.5 gram of the original sample is generally a suitable amount. 528 TECHNICAL METHODS OF ANALYSIS removing the stopper. Place the flask in the bath and maintain it at such a temperature that the contents of the flask will stand at exactly 65 C. Shake the flask every five minutes. At the expiration of exactly thirty minutes from the time the filter and residue are introduced, remove the flask from the bath and imme- diately filter the contents as quickly as possible through a rapid filter paper. Wash with water at 65 C. until the volume of filtrate is about 350 cc., allowing time for thorough draining before adding new portions of water. Return the filter with contents to the digestion flask, add 30-35 cc. of cone. HNOs, 5-10 cc. of cone. HC1 and boil until all phosphate is dissolved. Dilute the solution to 200 cc., mix well, filter through a dry filter; take a definite portion of the filtrate and proceed as under Total Phos- phoric Acid, above. (b) Determination in Non-acidulated Samples In case a determination of citrate-insoluble P2Os is required in non-acid- ulated samples, treat 2 grams of the phosphatic material without previous washing with water, precisely in the way above described, except that ; in case the substance contains much animal matter (bone, fish, etc.), the residue insoluble in ammonium citrate is to be dissolved by any one of the processes (II), (III), or (IV) described under Total Phosphoric Acid. CITRATE-SOLUBLE PHOSPHORIC ACID. The sum of the water- soluble and citrate-insoluble, subtracted from the total, gives the citrate-soluble phosphoric acid (P2Os). Phosphoric Acid (Optional Volumetric Method). PREPARA- TION OF REAGENTS. (a) Molybdate solution. To 100 cc. of molybdate solution, prepared as directed above under Gravimetric Method, add 5 cc. of cone. HNOs- This solution should be filtered each time before using. (b) Standard NaOH or KOH solution. Dilute 323.8 cc. of normal alkali, free from carbonates, to 1 liter. 100 cc. of the solution should neutralize 32.38 cc. of normal acid. 1 cc. is equal to 0.001 gram of P 2 O5 (1% of P2Os on a basis of 0.1 gram of substance). (c) Standard acid solution. The strength of this solution is the same as, or one-half of, the standard alkali solution, and is determined by titrating against that solution, using phenolphthal- ein indicator. Either HC1 or HNOs may be used. MISCELLANEOUS ANALYSES 529 (d) Phenolphthalein solution. Dissolve 1 gram of phenolphtha- lein in 100 cc. of alcohol. TOTAL PHOSPHORIC ACID. (a) Methods of Making Solution. Dissolve according to methods (II), (V), (VI), or (VII), as described above under Gravimetric Method [preferably by (V), when these acids are a suitable solvent] and dilute to 200 cc. with water. (6) Determination. (I) For percentages of 5 or below use .an aliquot corresponding to 0.4 gram of substance; for percentages between 5 and 20 use an aliquot corresponding to 0.2 gram; and for percentages above 20 use an aliquot corresponding to 0.1 gram. Add 5-10 cc. of HNOs, depending on the method of solu- tion (or the equivalent in NEUNOs); nearly neutralize with NH 4 OH; "dilute to 75-100 cc., heat in a water bath to 60-65 C., and for percentages below 5 add 20-25 cc. of freshly filtered molyb- date solution; whereas for percentages between 5 and 20 add 30-35 cc. of molybdate solution. For higher percentages add sufficient to insure complete precipitation. Stir, let stand about fifteen minutes, filter at once and wash once or twice with water by decantation, using 25-30 cc. each time, agitating the precipitate thoroughly and allowing to settle. Transfer to the filter and wash with cold water until two fillings of the filter yield a pink color upon the addition of phenolphthalein and 1 drop of the standard alkali. Transfer the precipitate and filter to a beaker or flask, dissolve in a small excess of the standard alkali, add a few drops of phenolphthalein solution, and titrate with standard acid. (II) Proceed as directed in (I), with this exception: Heat in a water bath at 45-50 C., add the molybdate solution, and let remain in the bath with occasional stirring for thirty minutes. (III) Proceed as in (I) to the point where the solution is ready to place in the water bath. Then cool the solution to room temperature, add molybdate solution at the rate of 75 cc. for each 0. 1 gram of P2Os present, place the stoppered flask containing the solution in a shaking apparatus and shake for thirty minutes at room temperature. Filter at once, wash, and titrate as in the preceding method. WATER-SOLUBLE PHOSPHORIC ACID. Dissolve according to directions given under the Gravimetric Method for Water-soluble To an aliquot portion of the solution corresponding to 0,2 530 TECHNICAL METHODS OF ANALYSIS or 0.4 gram, add 10 cc. of cone. HNOs and then NH 4 OH until a slight permanent precipitate is formed. Dilute to 60 cc. and proceed as under Determination (I) for Total P20s, on page 529. CITRATE-INSOLUBLE PHOSPHORIC ACID. Make the solution according to the directions given under the Gravimetric Method (IV) and determine the ?2O5 in an aliquot corresponding to 0.4 gram as directed under Total P205, Determination (I). CITRATE-SOLUBLE PHOSPHORIC ACID. The sum of the water- soluble and citrate-insoluble, subtracted from the total, gives the citrate-soluble P2O5. Potash. The K^O is determined as on page 41. NOTE. This is the procedure of the Association of Official Agricultural Chemists as given in its Journal, Methods of Analysis (1916), page 1. The methods are official, excepting the mechanical analysis of bone and tankage, which is tentative. CARBOLINEUM AND SIMILAR WOOD-PRESERVING OILS General. About one quart of oil is required for a complete analysis. It is possible to make a single analysis on about a pint, but the above amount provides for a check determination in case anything happens to the first. Before analyzing, thor- oughly liquefy the sample and mix it well by shaking and stirring. Specific Gravity at 38 C. Fill the hydrometer cylinder with the liquefied oil and place the cylinder in water, then heat until the temperature of the oil is several degrees higher than 38 C. Let the water cool until the oil has reached a temperature of 38 C. Thoroughly stir the oil, place the hydrometer in the cylinder, and carefully observe the reading. Take care that the hydrometer does not touch the bottom or sides of the cylinder when the reading is made. Condition at 38 C. Heat about 100 cc. of the oil in a beaker to a temperature of not less than 45 C. and let the oil cool grad- ually. When a temperature of 38 C. is reached, examine the oil carefully by means of a glass stirring rod. No solid crystalline particles should appear on the rod when withdrawn from the oil. A cloudiness of the oil may be disregarded. Flash Point. Place an evaporating dish, about 4 or 4.5 inches in diameter, on an asbestos diaphragm of sufficient size to MISCELLANEOUS ANALYSES 531 extend several inches beyond the dish. Cut a hole in the center of the diaphragm about half the maximum diameter of the dish and set the dish in it. Cover the bottom of the dish with dry sand to a depth of about 0.25 inch and place an evaporating dish, about 3 inches in diameter, on the sand. Fill the remaining space between the two dishes with sand until it reaches nearly to the rim of the inner dish. Arrange a thermometer so that the bulb is inside and about 0.25 inch above the bottom of the inner dish. Pour some of the liquefied oil into the dish until it is about three- quarters full. Place a low flame beneath the sand bath, with a suitable guard to protect it from draughts, which should be care- fully excluded from the vicinity of the testing apparatus. Heat so that the temperature of the oil will increase about 3 C. per minute. Apply a small flame just above the surface of the oil for every two degrees' rise of temperature until the flame flashes across th 3 surface of the oil. The temperature of the oil when this occurs is the flash point. Report results in Centigrade degrees. Burning Point. Continue the heating and the application of the testing flame until the oil ignites and burns for five seconds or more. The temperature at which this occurs is the burning point. Water. Weigh out 10 grams of the original oil into a 300 cc. Erlenmeyer flask. Add to this 75 cc. of Xylol which has pre- viously been saturated with water and proceed as directed on page 271. Fractionation. Arrange the apparatus for distillation as shown in Fig. 26.. For the distillation use a 200 cc. Jena * glass round-bottomed fractionating flask which shall satisfy the following requirements: It shall hold not less than 190 cc. nor more than 210 cc. when filled to the base line of the neck. The bulb shall be 2| inches in diameter. The neck shall be yf inch in diameter and 4 inches in length. The side arm shall be taken off at a point equidistant from the top and base of the neck. Connect an air condenser 14 inches long and 0.5 inch in diam- eter to the side arm of the flask, as shown in the figure, in order to insure complete condensation of the distillate. * Or Pyrex. 532 TECHNICAL METHODS OF ANALYSIS Before setting up the apparatus, weigh the flask accurately and then pour into it, by means of a stirring rod, exactly 100 grams of the thoroughly liquefied and mixed oil. Before beginning the distillation, fit the thermometer through the stopper in the neck of the flask so that the top of the bulb comes just below the lower edge of the side arm tube. Do not move the thermometer during the distillation. Rest the flask in a hole 1.75 inches in diameter in an asbestos board supported on the ring stand. To protect the flask from draughts, provide an asbestos shield surrounding the flask, resting on the asbestos Iron Support Ajr Con Ease of Stand FIG. 26. Distillation Apparatus for Carbolineum. board, and rising to the level of the top of the bulb. Play an open flame on the bottom of the flask and apply the heat so reg- ulated that the distillate passes over at the rate of about 1 drop per second. It is to be understood that the distillation has commenced as soon as the first drop of the distillate appears in the delivery tube of the flask and ended when a temperature of 360 C. is reached. NOTE. The distillation of some samples of oil may be accompanied, espe- cially during the first few moments, by a violent spattering of the liquid and small quantities may go over into the test-tubes. This can usually be avoided by inserting a small piece of unglazed porcelain in the flask before commencing the distillation. Distillate. Weigh up accurately 7 test tubes properly labeled for identification. Collect the fractions of the distillate in these MISCELLANEOUS ANALYSES 533 test-tubes according to the temperatures given in the following table: Test-tube No. Temperature C. Fraction 1 Up to 205 Water, tar acids, naphtha- lene (if present) 2 3 205-235 \ 235-270 J Tar acids, naphthalene 4 270-300 5 6 300-315 315-330 Anthracene, anthracene oil 7 330-360 Residue above 360 Residue in flask After completion of the distillation, drive over all the material which is in the air condenser by playing a flame over the condenser. Weigh up each of the test-tubes and also weigh the flask with the residue after it is cool. Calculate the percentages of the different fractions and add them together. The sum should be within 0.5% of 100%, except when there is a considerable amount of water present. If the loss, after carefully checking the figures, is found to be large when a sample containing little water or other low-boiling constituents is distilled, repeat the distillation. Sulfonation Residue. Carry out this test on the fractions of the carbolineum distilling between 300 and 330 C. (For Dead Oil of Coal Tar the test should be carried out on the fractions dis- tilling between 300 and 360 C.) Transfer the fractions to a large-mouthed round flask of about 250 cc. capacity. Then add 4-5 volumes of cone. H2SO4 for each volume of distillate under test. Heat for about five minutes with frequent shaking; then let stand about one-half hour, shaking at intervals. Slowly pour the mixture of acid and oil into a liter beaker containing enough water to dissolve the sulfonic acids formed. When cool, transfer from the beaker to a separatory funnel and let settle for one hour, or longer if necessary, until the undissolved oil separates clearly. Draw off and reject the water containing the sulfonic acids. 532 TECHNICAL METHODS OF ANALYSIS Before setting up the apparatus, weigh the flask accurately and then pour into it, by means of a stirring rod, exactly 100 grams of the thoroughly liquefied and mixed oil. Before beginning the distillation, fit the thermometer through the stopper in the neck of the flask so that the top of the bulb comes just below the lower edge of the side arm tube. Do not move the thermometer during the distillation. Rest the flask in a hole 1.75 inches in diameter in an asbestos board supported on the ring stand. To protect the flask from draughts, provide an asbestos shield surrounding the flask, resting on the asbestos Iron Support o Thermometer-^tt Cork Stopper Air Condenser Flask Ease of Stand FIG. 26. Distillation Apparatus for Carbolineum. board, and rising to the level of the top of the bulb. Play an open flame on the bottom of the flask and apply the heat so reg- ulated that the distillate passes over at the rate of about 1 drop per second. It is to be understood that the distillation has commenced as soon as the first drop of the distillate appears in the delivery tube of the flask and ended when a temperature of 360 C. is reached. NOTE. The distillation of some samples of oil may be accompanied, espe- cially during the first few moments, by a violent spattering of the liquid and small quantities may go over into the test-tubes. This can usually be avoided by inserting a small piece of unglazed porcelain in the flask before commencing the distillation. Distillate. Weigh up accurately 7 test tubes properly labeled for identification. Collect the fractions of the distillate in these MISCELLANEOUS ANALYSES 533 test-tubes according to the temperatures given in the following table: Test-tube No. Temperature C. Fraction 1 Up to 205 Water, tar acids, naphtha- lene (if present) 2 3 205-235 \ 235-270 / Tar acids, naphthalene 4 270-300 5 6 300-315 315-330 Anthracene, anthracene oil 7 330-360 Residue above 360 Residue in flask After completion of the distillation, drive over all the material which is in the air condenser by playing a flame over the condenser. Weigh up each of the test-tubes and also weigh the flask with the residue after it is cool. Calculate the percentages of the different fractions and add them together. The sum should be within 0.5% of 100%, except when there is a considerable amount of water present. If the loss, after carefully checking the figures, is found to be large when a sample containing little water or other low-boiling constituents is distilled, repeat the distillation. Sulfonation Residue. Carry out this test on the fractions of the carbolineum distilling between 300 and 330 C. (For Dead Oil of Coal Tar the test should be carried out on the fractions dis- tilling between 300 and 360 C.) Transfer the fractions to a large-mouthed round flask of about 250 cc. capacity. Then add 4-5 volumes of cone. H^SCU for each volume of distillate under test. Heat for about five minutes with frequent shaking; then let stand about one-half hour, shaking at intervals. Slowly pour the mixture of acid and oil into a liter beaker containing enough water to dissolve the sulfonic acids formed. When cool, transfer from the beaker to a separatory funnel and let settle for one hour, or longer if necessary, until the undissolved oil separates clearly. Draw off and reject the water containing the sulfonic acids. 534 TECHNICAL METHODS OF ANALYSIS Transfer all the oily residue to a tube graduated to 0.1 cc. and measure its volume. (See note.) If there should be a heavy insoluble residue, it is probable that the sulfonation has been incomplete. In this case, add com- mon salt to the water until the residue rises to the top of the water solution. Separate the unsulfonated oil and submit it again to the sulfonating treatment. In case the residue exceeds the maximum allowed by the specifications, treat it with a 10% NaOH solution. If the residue is soluble in this reagent, the sul- fonation test may be regarded as giving negative results. Separate and measure the residue of oil remaining after the treatment with NaOH solution. The amount of residue thus found is reported as " sulfonation residue." It is a measure of any mineral oil present. NOTE. As the sulfonation residue is generally very small, we have found it convenient to draw off all but about 10 cc. of the lower acid layer from the separatory funnel and then drain the liquid remaining in the funnel into a " color carbon tube " graduated 0-25 cc. in 0.1 cc. Rinse the funnel out with ether and drain this into the tube. Then immerse the tube in warm water to a level slightly above the level in the tube. Blow a gentle current of air into the tube, by means of a smaller glass tube reaching nearly to the surface of the inner liquid. Slowly raise the temperature of the outside water by pouring in hot water (keep flames away) until the ether boils. When all ether has been evaporated off, cool to room temperature and read the volume of the upper layer. Tar Acids. Carry out this determination on the fractions distilling below 300 C. Transfer them to a beaker of about 300 cc. capacity and thoroughly mix by stirring. Add 30 cc. of NaOH solution (17-18%) to the oil and heat the whole gently for two minutes, with frequent stirring. Transfer the mixture to a large separatory funnel and shake vigorously for at least one minute. Then let the mixture settle and draw off and save the NaOH solution which forms the lower liquid layer. Repeat this operation a second and third time, using 20 cc. of NaOH solution each time. Mix the 3 NaOH solutions together, boil vigorously for five minutes and let cool. Transfer to a special separatory funnel with a tube graduated to 0.1 cc. Add dilute H2SO4 (1:3) until the solution is acid to litmus paper. About 35 cc. of acid will be required and care must be taken that the MISCELLANEOUS ANALYSES 535 solution be kept cool while adding the acid. Let stand until the tar acids are well separated and read their volume (upper layer). Compute the per cent by dividing this volume by the volume of the oil taken for distillation; viz., the original weight of oil divided by its sp. gr. Insoluble in Benzene. Weigh out a 10-gram portion of the original oil into a small beaker (about 100-150 cc. capacity), add 20 cc. of benzene (commercial benzol) and filter through a weighed alundum (or Gooch) crucible with suction. Wash the residue in the crucible with benzene until the washings run color- less, thus indicating that no more substances will dissolve; dry the crucible at 110 C., weigh and compute the per cent of insoluble material. Ash. Weigh carefully into a weighed porcelain crucible about 2 grams of the oil, and heat cautiously over a low flame until the oil ignites. When the flame from the burning oil ceases, increase the heat, and after all organic matter has apparently burned off, cool and weigh. Reheat and reweigh, until a constant weight is obtained. Acetic Acid and Acetates (Qualitative Test). Place about 5 cc. of the oil in a test-tube and add about one-half of its volume of alcohol. Then add cone. H2SO4 slowly, drop by drop, with fre- quent shaking until the liquid is hot to the touch. If acetic acid or acetate is present, the characteristic odor of ethyl acetate will develop. To make certain of this odor always run a blank at the same time, adding to another portion of 5 cc. of the original oil in the test-tube a drop of acetic acid, and then carry through the test as above described. NOTES. (1) The apparatus should be thoroughly cleaned after each distillation. To remove the residue from the retort, add some of the oil and heat until the residue is all dissolved. Then pour out the oil and wash the flask with hot water and Sapolio. (2) The above method, with the exception of the determination of water, is ob tamed from Specification 1^-3432 of the Western Electric Company: "Analy- sis of Carbolineum and Similar Wood Preserving Oils," November 10, 1911. 536 TECHNICAL METHODS OF ANALYSIS GYSPY MOTH CREOSOTE Specific Gravity. Determine sp. gr. at 15.5 C. with a West- phal balance. Fractional Distillation. Distill 200 cc. of the creosote in a 250 cc. side-neck distilling flask having a thermometer inserted through the cork stopper in the neck with the bulb opposite the side tube of the flask. Conduct the distillation at the rate of approximately 2 drops per second, collecting each fraction in a small weighed flask. Take the following fractions and report the per cent by weight: 150-200 C. 200-245 C. 245-270 C. 270-320 C. Residue above 320 C. Examine the residue above 320 C. and note whether heavy oils have been added to the creosote. Tar Acids. Employ the distillate obtained above, by dis- tilling up to a temperature of 320 C., and determine the tar acids as described on page 534. Report the per cent of tar acids by volume. COAL-TAR ROOFING PITCH General. The methods for testing materials like coal-tar pitch must necessarily be more or less empirical. The following procedures are from specifications prepared for the Supervising Architect of the U. S. Treasury Department and are valuable in comparing different samples where straight run coal-tar pitch is desired. Melting Point. Determine the melting point by the Cube Method as described on page 542. Insoluble in Benzene. Digest 10 grams of the pitch in c. P. toluene on the steam bath, and decant through a filter cup con- sisting of two No. 575 C. S. & S. hardened filter papers (previously dried and weighed). Transfer the residue to this filter, and extract with c. P. benzene in any form of extraction apparatus MISCELLANEOUS ANALYSES 537 which insures hot extraction, until the washings run through practically colorless. Dry and weigh the papers plus the residue. Specific Gravity at 60 F. Make a small ball of the material weighing several grams. Make sure that no air bubbles are oc- cluded within it. Weigh this carefully on a tared watch glass (A). Attach a silk thread to it and again weigh (B). Then weigh the ball of material immersed in water, which is at 60 F., by sus- pending the thread from the stirrup of the balance arm (C). Subtract (C) from (B). This gives the loss in weight due to the buoyancy of water, i.e., the weight of an equal volume of water. Loss on Evaporation. Determine the loss in weight of 100 grams of pitch placed in a flat nickel dish 2 inches in diameter, and subjected to a temperature of 325 F., for seven hours. Specific Gravity of Distillate to 670 F. Distill from a side- neck flask an unweighed quantity of the sample (50-100 grams). Have the bulb of the thermometer opposite the side-neck of the flask and continue the distillation until the temperature reaches 670 F. Determine the sp. gr. of the distillate with a pycnometer at 140 F., compared with water at the same temperature. NOTES. (1) This method is obtained from Bulletin No. 154 of the Amer- ican Railway Engineering Association, Vol. 14 (Feb., 1913), page 850, which in turn is obtained from U. S. Government Specifications. (2) The specifications referred to above are as follows : The pitch shall be straight run residue obtained from the distillation of coal tar and shall meet the following requirements : (A) Melting point: 135-155 F. (B) Matter insoluble in benzol: 15-35%. (C) Sp. gr. at 60 F.: 1.25-1.35. (D) Evaporation loss, seven hours, at 325 F.: Maximum 9% for pitch of 145-155 F. m.p., and 11% for pitch of 135-145 m.p. (E) Sp. gr. of distillate to 670 F.: Minimum 1.07, determined at 140 F. as compared with water at 140 F. BITUMINOUS AND ASPHALTIC ROAD BINDERS General. Asphalt road binders are generally bought by specification which prescribes the method of testing. Unless otherwise specified, however, the following procedures should be employed : 538 TECHNICAL METHODS OF ANALYSIS Water. Weigh out approximately 10 grams of material on a filter paper, place the whole in a 300 cc. Erlenmeyer flask and determine the water by the Xylol Method as described on page 271. Specific Gravity. Determine the sp. gr. at 25/25 C. by one of the following methods, depending upon the consistency of the sample. (A) HYDROMETER METHOD (FOR THIN, FLUID MATERIALS). Bring the material to 25 C. in a hydrometer cylinder, place the hydrometer in it, and when it comes to rest, take the reading. In case the hydrometer sinks slowly, give it sufficient time to come to a definite resting point. Check this point by raising the hydrometer and letting it sink a second time. Never push the hydrometer below the point at which it naturally comes to rest until the last reading has been taken. Then push below the reading for a distance of 3 or 4 small scale divisions, whereupon it should immediately begin to rise. If it fails to do so, the material is too viscous for the hydrometer method. The direct sp. gr. reading thus obtained is based on water at 15.5 C. as unity. To correct to water at 25 C. multiply by 1.002. (B) PYCNOMETER METHOD (FOR Viscous AND SEMI-SOLID MA- TERIALS). Use a special pycnometer of the Hubbard type. First weigh the clean, dry pycnometer empty. Call this A. Fill with freshly boiled distilled water at 25 C. and again weigh. Call this B. Bring the material to a fluid condition with the least possible heating and pour into the dry pycnometer, which may also be warmed, and fill J to f full without allowing the material to touch the sides of the tube above the desired level. Cool to room tem- perature and weigh with the stopper. Call this C. Then pour in distilled water at 25 C. until the pycnometer is full. Insert the stopper and cool the whole to 25 C. by immersing completely for one-half hour in a beaker of distilled water at this temperature. Remove all surplus moisture with a soft cloth and weigh. Call this D. Calculate the sp. gr. of the material by the following formula: 25 C. C-A SP. gr. at 2^5;= (B _ A) _ (D _ c) - Results by this method should be accurate to 0.002. NOTE. The sp. gr. of fluid material may be determined in the ordinary MISCELLANEOUS ANALYSES 539 manner by completely filling the pycnometer with the material and dividing the weight of material taken by that of an equal volume of water. (C) DISPLACEMENT METHOD (FOR HARD, SOLID MATERIALS). For materials which are hard enough to be broken and handled in fragments at room temperature, weigh a small piece suspended by means of a silk thread from the hook on the balance arm about 1.5 inches above the pan. Call this weight A. Then weigh immersed in water at 25 C. by placing a beaker about two-thirds full of water on a support over the balance pan, but not touching it. Call this weight B. Calculate from the formula: ' Penetration Test. Determine the penetration with a standard No. 2 Roberts needle, acting for five seconds under a total load of 100 grams, the temperature of the material being at 77 F., and report results in terms of hundredths of a centimeter, avoiding decimals. (A) APPARATUS. The standard needle is made from round, polished, annealed steel drilling rod, diameter 0.0405-0.0410 inch. The rod is tapered to a sharp point at one end with the taper extending back 0.25 inch. The container for holding the material is a flat-bottom cylin- drical dish 2j^ inches in diameter and If inches deep (this require- ment is fulfilled by American Can Company's Gill style 3-ounce ointment box, deep pattern). The penetration apparatus consists of a standard needle inserted in a short brass rod, which in turn is held in the aluminum rod of the apparatus by a binding screw. The frame, aluminum rod and needle weigh 50 grams without any weight on the bottom of the frame. For test with 100-gram load put on the 50-gram weight. (B) PROCEDURE. Warm the sample sufficiently to flow and pour it into the tin box to a depth of not less than f inch. Transfer to a glass crystallizing dish or other suitable dish- and cover with as much water at 77 F. as convenient without spilling. Let cool one-half hour at room temperature protected from dust, then immerse in water at exactly 77 F., and keep at that temperature^ 540 TECHNICAL METHODS OF ANALYSIS one and one-half hours. Place the dish containing the tin holder with the material on the shelf of the machine; make sure that the binding screw of the needle holder is tight and that the tin dish is firm so that no rocking motion can occur; lower the rod until the point of the needle almost touches the surface of the sample; finally very cautiously adjust until the needle point just comes in contact with the surface of the sample. This can best be seen by having a light so situated that upon looking through the sides of the glass cup, the needle will be reflected from the surface of the sample. After thus setting the needle, move the counterweight slowly until the foot of the rack rests on the head of the rod and take the reading of the dial. With one hand open the clamp by pressing the button and with the other hand start the chronometer. At the end of exactly five seconds release the clamp, lower the rack until it rests on top of the rod and again read the dial. The difference between the first and second readings in hundredths of a centimeter is the penetration under the above conditions. Make at least three tests on points on the surface of the sample not less than f inch from the side of the container and not less than | inch apart. After each test return the sample and dish to the water bath at 77 F., and carefully wipe the needle toward its point with a clean, dry, cloth to remove all adhering asphalt. The penetration reported shall be average of at least three tests whose values do not differ by more than four points. NOTES. (1) The point of the needle should be examined from time to time with a magnifying glass to see that it is not injured in any way. If it is found defective it may be removed by heating the brass rod and withdrawing with pliers. A new needle may then be inserted in the heated brass, and held firmly in place by a drop of soft solder. (2) A cup suitable for holding the box containing the test material during penetration is conveniently made from a glass crystallizing dish 4 inches in diameter with straight sides about 2.5 inches high. Three right triangles, with right angle sides 0.4 and 2 inches, respectively, are cut from ^ inch sheet metal, some solid bitumen is melted in the bottom of the dish forming a layer about | inch thick, into which the triangles are placed, resting on the side 2 inches long. Their apexes should meet the center with their short sides divid- ing the circumference of the dish into three equal parts. When the bitumen is hardened, the triangles give a firm support for the circular boxes and the possibility of any rocking motion, and consequent faulty results, is avoided Volatility. Weigh out about 50 grams of sample in a tin box inches in diameter by about If inches deep (3 ounce Gill style MISCELLANEOUS ANALYSES 541 ointment box, deep pattern), first carefully weighing the box; then adjust the weight of sample so that it does not vary more than 0.2 gram from 50. It may be necessary to warm some of the material in order to handle it conveniently, after which it must be allowed to cool before determining the accurate weight. Before making the test, the interior of the oven should show a temperature of 163 C. (325 F.). Heat the material in the oven for five hours, remove,, cool in a desiccator and weigh. Cal- culate the per cent loss. NOTES. (1) For strictly accurate work the New York Testing Laboratory oven should be used. (See U. S. Dept. of Agriculture, Bulletin 314, page 19.) (2) In case it is not desired to determine the penetration of the residue, tests should be run on 20 grams of material in a tin container, 6 cm. in diam- eter by 2 cm. deep. In any case, the amount of material taken should be stated. Float Test. This test is always made on viscous and semi- solid refined tars and often on viscous and semi-solid petroleum and asphalt products, although, when penetration tests can be employed on the latter, the float test is not always considered necessary. For more fluid products make the tests at 32 C.; for semi-solid materials, at 50 C.; and in certain cases, on unusu- ally hard materials, at 100 C. The float apparatus consists of 2 parts, an aluminum float or saucer and a conical brass collar. Place the brass collar with the small end down on a brass plate previously amalgamated with Hg by rubbing it first with a dilute solution of mercuric chloride or nitrate and then with Hg. Heat a small quantity of the mate- rial in a metal spoon until fluid, taking care that it suffers no appre- ciable loss by volatilization and that it is kept free from air bubbles. Pour into the collar in a thin stream until slightly more than level with the top. Cool to room temperature and remove the surplus with a spatula which has been slightly heated. Place the collar and plate in ice water at about 5 C. for at least fifteen minutes. Meanwhile place a 500 cc. cup or beaker, nearly filled with water, over a flame and heat to the test temperature. At the end of fifteen minutes or more remove the collar with contents from the brass plate and screw into the aluminum float, taking care to screw it in as far as it will go.. Float the apparatus on the sur- face of the water, at the same instant starting a stop watch. 542 TECHNICAL METHODS OF ANALYSIS When water first breaks through the plug of bituminous material, stop the watch. The time in seconds between placing the appa- ratus on the water and when the water breaks through is the " float test." Flash Point. Determine the flash point in the open cup tester as described on page 255. Softening Point. Bituminous materials have no true melting point. Any method of determining the " melting point " of these materials must be arbitrary. The two in most common use are the following : (1) CUBE METHOD (Not applicable to pitches having a melting point above 77 C.). First melt the material in a spoon by gentle application of heat until sufficiently fluid to pour readily, taking care that it suffers no appreciable loss by volatilization. Stir thoroughly, avoiding incorporating air bubbles in the mass. Then pour into an 0.5 inch brass cubical mold, which has been amal- gamated with Hg, and which is placed on an amalgamated brass plate. The brass may be amalgamated by washing it first with a dilute solution of mercuric chloride or nitrate, after which the Hg is rubbed into the surface. By this means the bitumen is to a con- siderable extent prevented from sticking to the sides of the mold. The hot material should slightly more than fill the mold, and when cooled the excess should be cut off with a slightly heated spatula. (A) Pitches having softening points between J$ and 77 C. Fill a 600 cc. low form Griffin beaker to a depth of about 3.75 inches with freshly boiled distilled water at 15.5 C. Place the cube of pitch on an L-shaped right-angled hook made of No. 12 B. & S. gauge copper wire. The foot of the L should be 1 inch long and should run through the center of the cube so that one edge of the cube, not its surface, is parallel to the bottom of the beaker and exactly 1 inch above it. The upper edge of the cube should be 2 inches below the surface of the water. Let it remain in the water for fifteen minutes before applying heat to the beaker set on a wire gauze. Suspend the thermometer so that the bottom of the bulb is level with the bottom edge of the cube and within 0.25 inch of but not touching the cube. Apply heat so that the temperature of the water is raised 5 C. (9 F.) per minute. The rate of rise must be uniform and should not be averaged over the period of test, The maximum MISCELLANEOUS ANALYSES 543 permissible variation for any minute period after the first three minutes is 0.5 C. (1 F.). If the rate of rise exceeds this limit, the test must be rejected. The temperature recorded by the thermometer at the instant the pitch touches the bottom of the beaker is the softening point of the sample. NOTES. (1) The burner should be protected by a shield to avoid draft. (2) The use of freshly boiled distilled water is essential to prevent air bub- bles forming on the cube and retarding sinking. (3) Rigid adherence to the prescribed rate of heating is absolutely essential for accuracy. (4) A sheet of paper placed on the bottom of the beaker and weighted down will prevent the pitch from sticking to the glass and save considerable time and trouble in cleaning. (5) The limit of accuracy of the test is 0.5 C. (1 F.). (6) The thermometer should be graduated from 0-80 C., preferably in J divisions and the top of the mercury column at the time of reading should be above the surface of the water. (B) Pitches having softening points below 43 C. Use the same method as given above, except that the water when placed in the beaker should be at a temperature of 4 C. instead of 15.5 C. Let the cube remain fifteen minutes in this water before applying heat. (2) RING AND BALL METHOD. (A) Apparatus. This consists of a brass ring exactly f inch in diameter, J inch deep, and ^ inch wall, suspended exactly 1 inch above the bottom of a beaker; a steel ball | inch in diameter, weighing between 3.45 and 3.55 grams; a standardized thermometer; and a low form Griffin glass beaker of about 600 cc. capacity. (B) Procedure. Carefully melt the sample, as in the Cube Method above, and fill the ring with excess of the material to be tested. During filling, rest the ring on amalgamated brass to prevent sticking. After cooling, remove the excess with a slightly heated spatula. Fill the beaker to a depth of about 3.25 inches with freshly boiled distilled water at 5 C.* Place the ball in the center of the upper surface of the material and suspend in water so that the lower surface of the filled ring is exactly 1 inch above the bottom of the beaker and the upper surface is 2 inches below the * For materials having a softening point above 90 C. use glycerine instead of water. 544 TECHNICAL METHODS OF ANALYSIS surface of the water. Let remain in the water for fifteen minutes before applying heat. Suspend the thermometer so that the bottom of the bulb is level with the bottom of the ring and within 0.25 inch of, but not touching, the ring. Apply heat uniformly, so that the temperature of the water rises 5 C. (9 F.) per minute. The rate of rise must be uniform and is not to be averaged over the period of test. The maximum permissible variation for any minute period after the first three shall be 0.5 C. (1 F.). Reject any tests where the rate of rise exceeds these limits. The temperature recorded by the thermometer at the instant the sample touches the bottom of the beaker is its softening point. (See notes under Cube Method above.) Total Bitumen (Soluble in Carbon Bisulfide). Prepare a Gooch crucible (the best size is 1.75 inch at the top, 1 inch deep, and 1.5 inch at the bottom) with an asbestos mat which will just show light through it. Suck dry, heat a few minutes in the oven, ignite over a Tirrill burner, cool in a desiccator and weigh. Place 1-10 grams of the sample, depending upon the amount of insoluble matter, in a 150 cc. Erlenmeyer flask, which has been previously weighed, and weigh accurately; then pour 100 cc. of CS2 into the flask in small portions with continual agitation until all lumps disappear and nothing adheres to the bottom. Cork and set aside for fifteen minutes or longer. Decant the CS2 solution very carefully through the asbestos in the Gooch crucible without suction, with care not to stir up any precipitate. At the first sign of any sediment coming over, stop the decantation, and let the filter drain. Wash a small amount of CS2 down the sides of the flask, bring the precipitate upon the mat and remove all adhering matter from the flask to the crucible with a policeman which is not attacked by CS2. Wash the contents of the crucible with CS2 until washings are colorless. Apply suction until no odor of CS2 remains. Clean the outside of the crucible with a soft cloth moistened with a little CS2. Dry at 100 C. for about twenty minutes, cool in a desiccator and weigh. (If any appre- ciable amount of insoluble matter adheres to the flask, it should also be dried and weighed and any increase over the original weight added to that of the insoluble matter in the crucible.) The total weight of insoluble material may include both MISCELLANEOUS ANALYSES 545 organic and mineral matter. Ignite at a red heat until no incan- descent particles remain. Cool and weigh the mineral matter. Report the difference between the total weight of material insoluble in CS2 and the weight of the substance taken (both expressed in percentages) as " total bitumen." Also report the per cent of mineral matter as " ash." NOTES. (1) In certain natural asphalts it is practically impossible to retain all finely divided mineral matter on an asbestos mat. It is, therefore, generally more accurate to obtain the result for total mineral matter by direct ignition of 1 gram in a platinum crucible or to use the result for ash obtained in the fixed carbon test. Then determine the total bitumen by deducting from 100% the sum of the per cent of total mineral matter and of organic insoluble matter. If the presence of carbonate mineral is suspected, the per cent of mineral matter may be most accurately determined by treating the ash from the fixed carbon determination with a few drops of ammonium car- bonate solution, drying at 100 C., then heating for a few minutes at dull red heat, cooling and weighing again. (2) When unusual difficulty in filtering is experienced, it is necessary to let stand much longer than fifteen minutes. In such cases it is preferable to proceed as follows: Weigh 2-15 grams (depending on richness in bitumen) into a 150 cc. Erlenmeyer flask, which .has been previously weighed, and treat with 100 cc. of 082. Cork the flask loosely and shake from time to time, until practically all large particles have been broken up. Set aside undisturbed for forty-eight hours. Decant the solution into a similar flask that has been previously weighed, as much of the solvent being poured off as possible with- out disturbing the residue. Treat the first flask again with fresh 082, and shake as before. Put away with the second flask undisturbed for forty-eight hours. At the end of this time carefully decant off the contents of the two flasks upon a weighed Gooch crucible fitted with an asbestos filter, the contents of the second flask being passed through the filter first. The asbestos filter should be made of ignited long-fiber amphibole, packed in the bottom of the Gooch crucible to a depth of not over ^ inch. After passing the contents of both flasks through the filter, shake the two residues with more fresh CS 2 and set aside for twenty-four hours without disturbing, or until good subsida- tion has taken place. Again decant the solvent off upon the filter. Con- tinue this washing until filtrate or washings are practically colorless. Dry the crucible and both flasks at 125 C. and weigh. Evaporate the filtrate containing the bitumen, burn the bituminous residue, and add the weight of ash thus obtained to that of the residue in the two flasks and cruci- ble. The sum of these weights deducted from the weight of substance taken gives the weight of bitumen extracted. (3) In the analysis of tars the insoluble organic matter is commonly known and reported as "free carbon." 546 TECHNICAL METHODS OF ANALYSIS Bitumen Insoluble in 86 Naphtha. This determination is made in the same general way as the Total Bitumen, using instead of C&2 100 cc. of naphtha, at least 85% of which distills between 35 and 65 C. It is advisable to heat the sample after it has been weighed into the flask and let it cool in a thin layer around the lower part. Not more than half the total amount of naphtha required should be used until the sample is entirely broken up; then add the remainder, swirl the flask, mix thoroughly, cork and set aside thirty minutes or more. In making the filtration use the utmost care to avoid stirring up any of the precipitate, and make the first decantation as complete as possible. Suction may be applied when filtration by gravity almost ceases, but should be used sparingly as it tends to clog the filter. The material on the felt should never be allowed to run dry until washing is com- pleted as shown by a colorless filtrate. When considerable insoluble matter adheres to the flask, make no attempt to remove it completely, merely wash until free from soluble matter and dry the flask (after removing naphtha) for about one hour at 100 C., after which cool and weigh. The per cent of bitumen insoluble in naphtha is reported upon the basis of the total bitumen taken as 100. NOTE. The difference between the amounts insoluble in CS 2 and in naph- tha is the bitumen insoluble in naphtha. If, for instance, the insoluble in CS 2 is 1% and the total insoluble in naphtha is 10.9%, the calculation of % bitumen insoluble in naphtha is as follows: Bit, insol. in naphtha 10.9-1 _ 9.9 Total bitumen ~ 100-1~ 99~ Fixed Carbon and Ash. Determine fixed carbon and ash on 1 gram as in Coal, page 174. (See also under Total Bitumen above.) Distillation. From the sp. gr. of the material calculate the weight of 100 cc. and pour this amount into a tared 250 cc. Engler distillation flask, after warming in a tin cup if necessary to make it sufficiently fluid. For the procedure in distilling and apparatus used see page 551. Report results both as per cent by weight and by volume, or as required. Ductility. Form a briquette of the sample by pouring the molten material into a briquette mould. The dimensions of the MISCELLANEOUS ANALYSES 547 briquette shall be: 1 cm. (0.394 inch) in thickness throughout its entire length; distance between clips or end pieces, 3 cm.; width of asphalt cement section at mouth of clips, 2 cm. ; width at min- imum cross-section, half-way between clips, 1 cm. The center pieces are removable, the briquette mold being held together during molding with a clamp or wire. The molding of the briquette is to be done as follows: The two center sections must be well amalgamated to prevent the sample from adhering to them. Then place the briquette mold on a freshly amalgamated brass plate. Pour the sample to be tested, while in a molten state, into the mold, adding a slight excess to allow for shrinkage on cooling. When the material in the mold is nearly cool, cut off the briquette level with a warm knife or spatula. When thoroughly cooled to the proper tem- perature, remove the clamp and the two side pieces, leaving the briquette held at each end by the ends of the mold, which now play the part of clips. Keep the briquette in water for thirty minutes at 4 C. (39 F.) or 25 C. (77 F.) before testing, depend- ent on the temperature at which the ductility is desired. Place the briquette with clips attached in the ductility test machine, filled with water at one of the above temperatures to a sufficient height to cover the briquette not less than 5 cm. (1.97 in.). The machine consists of a rectangular water-tight box, having a mov- able block working on a worm gear from left to right. The left clip is held rigid by placing its ring over a short metal peg pro- vided for this purpose; the right clip is placed over a similar rigid peg on the movable block. The latter is provided with a pointer which moves along a centimeter scale. Before starting the test, adjust the centimeter scale with the pointer at zero. Then apply power by the worm gear, pulling from left to right at a uniform rate of 5 cm. per minute. The distance in centimeters reg- istered by the pointer on the scale at the time of rupture of the thread of asphalt material is taken as the ductility of the material. Paraffin Scale. The determination of paraffin scale is seldom required. The procedure is described in U. S. Dept of Agriculture, Bulletin 314, page 32. Petroleum or Asphalt Products in Tar. Take fractions of the distillation from 270-300 C., from 300-350 C., and from 548 TECHNICAL METHODS OF ANALYSIS 350-375, C. respectively. Stir each fraction separately and, if necessary, warm to dissolve any solids. To 4 cc. of each fraction in tubes graduated to 0.1 cc. add 6 cc. of dimethyl sulfate. Shake well and let stand thirty minutes. Read off the volume of any oil separating on top of the liquid. This is due to asphalt or petroleum products. Calculate the vol- ume per cent in each fraction and report as follows : Fractions C. Per Cent Distillate Per Cent Distillate Insoluble in Dimethyl Sulfate 270-300 300-350 350-375 The test is mainly qualitative but will indicate as little as 3% of petroleum or asphalt products in tar. REFERENCES. Sp. gr., flash point, float test, penetration, melting point (Cube method) and bitumen: U. S. Dept. of Agriculture, Bulletin 314. Melting point (Ring and Ball method): American Soc. for Testing Materials, Proceedings 1916, page 549. Volatility: U. S. Dept. of Agriculture, Bulletin 314, page 19 and Bulletin 555, page 36. Ductility and Paraffin: Am. Soc. Civil Eng., 1914, Proceedings, pages 3047 and 3049. See also articles by J. M. Weiss in J. Ind. Eng. Chem., 1918, " Methods of Analysis used in the Coal Tar Industry." CRUDE COAL-TAR AND WATER-GAS TAR General. The analysis of crude coal and water-gas tars gen- erally involves the following determinations: Specific gravity, free carbon, water, fractional distillation, and tar acids. Sampling. Tar is best sampled when being unloaded from the tank car or barge. A pet cock, with a nipple projecting about one- third of its diameter, should be placed in the pipe line and a con- tinuous stream of tar drawn off into a barrel during the time of unloading. The pet cock should be so regulated that the sample will represent approximately 0.1% of the shipment. The tar may MISCELLANEOUS ANALYSES 549 then be stirred up and a sample taken from the barrel. Samples of tar should be placed in heavy clear bottles or screw top tin cans. When necessary to sample from storage tanks, or wells, it should be done by means of a " thief." This is particularly necessary when different shipments of tar of widely different gravities have been run into the same tank. A simple and efficient apparatus may be made from a piece of 2-in. pipe provided with a lever handle cock. This may be closed by means of a small iron rod. By cutting away part of the cock and half of the plug, an opening nearly as large as the interior of the pipe is produced. In taking the sample, the cock is opened and the " thief " lowered slowly to the bottom of the tank, well, or car, the " thief " having previously been rinsed with the liquid to be sampled. The cock is then closed, the " thief " is withdrawn, and the sample run into a bottle. This operation is repeated until a sample of about 1 gallon is obtained, after which the contents should be thoroughly mixed and a portion taken to serve as a smaller sample for analysis. It should be noted that this method cannot be used with horizontal cylindrical tanks. In the case of tar where there is always a certain amount of water or ammoniacal liquor floating on the surface, it seems best to attempt to locate the level of the water or liquor, taking a sample at this point, and then sample a lower portion of the tar which is reasonably free from water, and by calculation, estimate the total quantity of water present. Specific Gravity. The sp. gr. of thin tar, such as water-gas tar, free from water, may be determined by the Westphal balance. The measurements should be made at 25 C. where possible. If this is not possible, correct from the observed temperature to 25 C. as follows: Take the reading with the Westphal balance at the tempera- ture t C. Balance the plummet in distilled water at the same temperature and take the reading. Divide the first reading by the second. This is the sp. gr. of the material at t C. If t is greater than 25, then Sp. gr. at 25 C. = sp. gr. at t C.+ 0.00068 (Z-25). In case t is less than 25, then Sp. gr. at 25 C. = sp. gr. at t C.- 0.00068 (25- 1). A hydrometer may be used in place of the Westphal balance. 550 TECHNICAL METHODS OF ANALYSIS Since hydrometers are standardized for water at 15.5 C., the hydrometer reading in the tar at 25 C. should be multiplied by 1.002 to bring it to the basis of H 2 O = 1 at 25 C. For accurate work and for thick tars the modified Hubbard's sp. gr. bottle should be used. In using this bottle the following weights are necessary in the order given : A = Weight of empty bottle; B = Weight of bottle filled with water to the mark at 25 C.; and C = Weight of bottle filled with tar at 25 C. ; 25 C C A then sp. gr. of tar at r^f = ]^[- NOTE. The Hubbard method is not accurate for tar containing water. In such case the tar should be dehydrated, as described below under Distilla- tion, and the sp. gr. determined on the dehydrated tar. The sp. gr. of the original tar may be calculated as follows : Let A = Sp. gr. of dehydrated tar; J3 = Per cent of H^O expressed as decimal; and X = Sp. gr. of original tar; then X = B+A (1-5). Moisture. For most purposes the Xylol Method is suitable, as described on page 271. In cases of dispute, however, use the method described in the Gas Chemists' Hand Book (1916), page 191, as follows: Measure 50 cc. of coal-tar naphtha or light oil (which must be tested to determine that it is free from water whenever a new supply is required) in a 250 cc. graduated cylinder. Add 200 cc. of the tar, transfer the contents of the cylinder to a copper still (see below under Distillation) and wash the cylinder with 50-75 cc. more of naphtha, adding the washing to the contents of the still. Attach the lid and clamp, using a paper gasket. Distill through a still head, connected to a condenser, and collect the distillate in a separatory funnel having a graduated stem to which 15-20 cc. of benzene have been previously added. Apply heat to the still by means of the ring burner and distill until the temperature as indicated by the thermometer has reached 205 C. The bulb of the thermometer must be opposite the side neck of the still head. The reading of the volume of water is made after twirling the funnel and letting the water settle for a few minutes. Figure the per cent by volume. MISCELLANEOUS ANALYSES 551 Free Carbon. Fur accurate work the tar should be dried before testing and after drying passed hot through a 30-mesh sieve to remove foreign substances. This ordinarily, however, is not necessary as such foreign substances could easily be detected during analysis. From materials of 5% or more carbon take about 5 grams, with lesser percentages take 10 grams approximately. Weigh out in a 100 cc. beaker and digest with about 5 cc. of c. p. toluene on the steam bath for not over thirty minutes. If the solution is kept hot and constantly stirred, digestion can be completed very rapidly. Weigh in a weighing bottle a filter cup prepared as described below and place in a carbon filter tube over a beaker or flask. Decant the toluene tar mixture through the thimble and wash with hot toluene until clean. After transferring all the ma- terial from the beaker, wash once with hot c. P. benzene, drain, cover with a cap of filter paper and extract with benzene in an extractor of the Cottle or Rubber Insulation Committee type as used in rubber analysis (see Fig. 25, page 481). Continue extrac- tion until the descending benzene is colorless; remove the thimble, discard the cap, dry at 105 C. and weigh in the same weighing bottle as originally used. The residue is carbon. Report the result as " free carbon (toluene-benzene method)." NOTES. (1) If the carbon is contaminated by dirt, ignite in a crucible and weigh the inorganic residue. Subtract this from the original weight of residue. (2) Filter cups: The filter cups or thimbles are made of 15 cm. hardened filter paper. To make a cup, two circles should be taken and one cut down to a diameter of about 14 cm. A round stick about 1 inch hi diameter is used as a form. The stick is placed in the center of the circles of filter paper, the smaller inside; the papers are then folded symmetrically round the stick to form a cup of about 2.5 inches in length. After being made, they are soaked in benzene to remove any grease due to handling, drained, dried in a steam oven and kept in a desiccator until used. Distillation. (1) SAMPLING. The sample as received must be thoroughly stirred and agitated, warming, if necessary, to insure complete mixture before the portion for analysis is removed. (2) DEHYDRATION. If the presence of water is suspected or known, dehydrate the material before distillation. Place about 00-400 cc. in a copper still (A. H. Thomas Co. ; Catalog No. 552 TECHNICAL METHODS OF ANALYSIS 20416) provided with a distilling head connected with a water- cooled condenser. Use a ring burner, starting with a small flame at the top of the still, and gradually lowering, if necessary, until all water has been driven off and the thermometer reads 170 C. (The bulb of the thermometer should be opposite the side tube of the distilling head.) Collect the distillate in a 200 cc. separatory funnel with the tube cut off close to the stopcock. When all water has been driven over and the distillate has separated clear, draw off the water and return the oils to the residue in the still. (If crystals separate, warm the mixture until they go into solution.) Let the contents of the still cool to below 100 C. before the oils are returned; stir well and mix with the residue. (3) APPARATUS. The apparatus shall consist of the following standard parts: (a) Flask. The distillation flask shall be a 250 cc. Engler distilling flask, of the following dimensions : Diameter of bulb 8.0 cm. Length of neck 15 . cm. Diameter of neck 1.7 cm. Surface of material to lower side of tubulature 11 .0 cm. Length of tubulature 15 . cm. Diameter of tubulature 0.9 cm. Angle of tubulature 75 A variation of 3% from the above measurements is allowed. (b) Thermometer. The thermometer shall be of hardened glass, filled with inert gas under pressure and provided with an expansion chamber at the top; it shall read from to 400 C. or 450 C., shall be graduated in single degrees Centigrade, and shall have the following dimensions : Diameter of stem 6.5-7.5 mm. (approximately). Length of thermometer. 385 mm. Length from to 400 marks 285-305 mm. Length of bulb 10-15 mm. Diameter of bulb 5 mm. and not exceeding diameter of stem. Distance from zero to bottom of bulb . 25-35 mm . When the thermometer is taken at a temperature of 26 C. and plunged into a free flow of live steam, the meniscus must pass the 90 mark in not more than six seconds. MISCELLANEOUS ANALYSES 553 (c) Condenser. The condenser tube shall have the following dimensions : Length of tube 500 mm. Width of tube 12-15 mm. Width of adaptor end of tube 20-25 mm. (d) Stands. Two iron stands are required, one with a universal clamp for holding the condenser, and one with a light grip arm with a cork-lined clamp for holding the flask. (e) Burner and shield. The Bunsen* burner shall be provided with a tin shield 20 cm. long and 9 cm. diameter, having a small hole for observing the flame. (/) Cylinders. The cylinders used in collecting the distillate shall have a capacity of 25 cc., and shall be graduated to 0.1 cc. (4) SETTING UP APPARATUS. Connect the distilling flask containing the tar to the air condenser with a one-hole stopper. Insert the thermometer through a one-hole stopper in the top of the flask in such a way that the top of the bulb is opposite the middle of the side arm opening of the flask. Have the burner so adjusted that the shield will completely protect the flame and also the bulb of the distilling flask. (5) PROCEDURE. Weigh exactly 100 cc. of the dehydrated material into the distilling flask, which has been previously weighed. Adjust the thermometer, shield, condenser, etc. Commence distillation, so regulating the rate that 1 cc. passes over every minute. Change the receiver as the mercury column just passes the fractionation point. Report the temperature at which the first drop comes over and then report the per cent of each of the following fractions, both by weight and by volume : ,Up to 110C. 110-170 C. 170-235 C. 235-270 C. 270-315 C. 315-355 C. Residue above 355 C. *0r Tirrill. 554 TECHNICAL METHODS OF ANALYSIS The residue is determined by cooling the flask after distillation and weighing it. During the distillation the condenser tube should be warmed when necessary to prevent deposition of any sublimate. NOTE. This method is based on standard method D20-18 of the Amer- can Society for Testing Materials, 1918. Tar Acids. Distill a known volume of the tar up to 315 C. For each 100 cc. of distillate add 40 cc. of an approximately 20% solution of NaOH. Warm slightly while stirring and place in a separatory funnel. Shake vigorously and let stand until the oil and soda solutions separate, and draw off the latter containing most of the tar acids. Make a second and third extraction, using 75% and 50% of the original volume of NaOH solution, respectively. Unite the three alkaline extracts in a 200 cc. graduated cylinder (see below) and acidify with dil. H 2 S(>4. Let cool and read the volume of tar acids. From the amount of tar taken calculate the per cent by volume of tar acids in the original tar. NOTE. The cylinder or separatory funnel should be of special form with a stopcock and graduated stem so that the tar acids may be drawn down in the graduated portion and the volume accurately determined. Sulfur. See page 191. Road Tars. The examination of road tars is usually made to conform to special specifications which describe the method. If methods are not specified, follow the procedures on page 537. REFERENCES. Gas Chemists' Handbook, 1916. Am. Soc. for Testing Materials: Standards, 1918, page 669. SPENT OXIDE General. The material used for the removal of H 2 S from illuminating gas is known under various names, such as oxide, iron oxide, iron mass, iron sponge, etc. After it has been in use for some time, it is generally referred to as " spent oxide." It consists of wood shavings mixed with hydrated ferric oxide. The action of H2S on this material is probably represented by the following equations: Fe 2 3 H 2 O+3H 2 S = Fe 2 S 3 +4H 2 O, and Fe 2 O 3 H 2 O+3H 2 S = 2FeS+S+4H 2 O. MISCELLANEOUS ANALYSES 555 After the material has become " foul/' it is revivified or reox- idized by exposure to air. The reactions here are probably as follows : 2Fe 2 S 3 +3O 2 = 2Fe 2 O 3 +6S, and 4FeS+3O2 = 2Fe 2 3 +4S. There are also present Prussian blue and other cyanogen and carbonyl products. In selling the material for its cyanogen con- tent, it is customary to calculate the latter in terms of crystallized potassium ferrocyanide, K4Fe(CN)6-3H 2 0. Moisture. When the material arrives in the laboratory, it is usually in such condition, due to moisture and tarry matter, that it must be dried before grinding. Mix the sample thoroughly and weigh out 100 grams (or more if possible) into a rectangular tin dish and dry to constant weight, keeping the temperature as nearly as possible between 90 and 95 C. When ferrocyanide is to be determined, dry for nine hours at 50-60 C. It is unsafe to hasten drying by raising the heat, as this will volatilize some of the free sulfur, and may decompose Prussian blue. Preparation of Sample. Grind the dried sample in an ordinary coffee mill, repeating the grinding until the material is as fine as possible. By bringing the grinding surfaces closer together each time, it is possible to reduce the mass almost to a powder. For accurate work, determine the residual moisture in the ground sample and make such corrections as are necessary in figures obtained in the subsequent analysis to bring them to the basis of the original material. NOTE. In grinding it is very essential, in case ferrocyanide is to be deter- mined, that the temperature should not get above 50-60 C., as excessive heat- ing will cause loss of cyanogen. Free Sulfur. Since, for commercial purposes, S combined as iron sulfide may be considered in the same category as free S, the analysis is shortened by determining these two forms of S together. The method is based on the fact that free S is dis- solved by CS 2 . Since the latter, however, will dissolve only free S, some of the sample should be spread out and exposed to air for " revivification " before extracting. The sample must also be dry, since moisture interferes with the extraction. 556 TECHNICAL METHODS OF ANALYSIS Weigh 5 grams of the ground dry material on a watch glass and heat in a water oven at not over 95 C. to constant weight in order to remove the last traces of moisture. Transfer to a Soxhlet thimble and extract with recently distilled c. P. 082 * until no further material is extracted, collecting the extract in a weighed flask. Distill off the 082 on the water bath through a Liebig condenser. Dry the residue to constant weight at not over 100 C. The weight gives the amount of tar and free S. Add to the flask 50 cc. of fuming HNOa, evaporate on the hot plate or sand bath to half its volume, then add, little by little, 3 grams of KClOs, and evaporate to dryness. Bake on the hot plate for one-half hour, cool and add 30 cc. of HC1 (1 : 1). JBoil, filter and wash with hot water. Heat the filtrate to boiling and add slowly a boiling 10% solution of BaCk in excess. Boil for one-half hour, or let stand overnight; filter hot, and ignite and weigh the BaSC^ in the usual way. Calculate the weight to sulfur. CALCULATION. BaSO 4 X 0.1373 - S. NOTES. (1) The CS2 extraction is a slow one; the length of time required depends upon the percentage of S and the amount of tar. Usually thirty hours will be sufficient. (2) The free S in spent oxides may run from a few per cent to as high as 55 or 60%. Tar. The weight of S subtracted from the total weight of C$2 extract gives the weight of tar. Combined Sulfur. If it is desired to determine the combined S, this may be done on the residue from the C$2 extraction, after making sure that all C$2 has been expelled. After oxidizing with fuming HNOs, as described under Free Sulfur, and filtering, dry the undecomposed material on the filter paper and fuse it with equal parts of Na2COs and NaNOs. Disintegrate with hot water, filter and wash. Dilute the filtrate to 250 cc. and acidulate with HC1. Boil off CO2 and precipitate hot with excess of BaCl2 solution. Add the S thus found to the S found by the HNOs treatment. The sum gives combined S. * CS 2 is very inflammable and in the gaseous state, when mixed with a certain percentage of air, is highly explosive. The extraction apparatus should never be disconnected until it has cooled down to room temperature. MISCELLANEOUS ANALYSES 557 Potassium Ferrocyanide (Modified Knublauch Method). MOISTURE. Dry 30 grams of the material for nine hours at 50-60 C. EXTRACTION OF PRUSSIAN BLUE. Grind the dried oxide until it all passes an 80-mesh sieve, taking care to avoid heating. Introduce 10 grams of this fine material into a 250 cc. volumetric flask. Add 50 cc. of 10% KOH solution and .let stand 15-16 hours at room temperature, shaking frequently. Then make up to 250 cc. and add 5 cc. more to compensate for the volume of the oxide. Shake vigorously and filter through a dry filter. Some free S may come through the filter paper, but most of this can be removed by refiltering through the same paper. What remains will do no harm. Pipette out 100 cc. of this filtrate and let it run slowly into 50 cc. of a boiling solution of FeCla.* Boil the mixture a few minutes to complete the precipitation of the Prussian blue. After this settles a little, filter and wash with boiling water until the wash- ings are free from acid. Transfer the blue, together with the filter paper, into a 400 cc. beaker and add 25 cc. of 10% KOH solution. After complete decomposition, transfer to a 250 cc. volumetric flask and make up to the mark. Shake and filter through a dry filter. Use 100 cc. of this filtrate for titration. ' PREPARATION OF SOLUTIONS. (1) Zinc Sulfate Solution. Weigh out 10 grams of c. P. ZnSC^-TEbO, dissolve in water, add 10 cc. of cone. H2S04 and dilute to 1 liter. (2) Potassium Ferrocyanide Solution. Weigh out exactly 5 grams of K4Fe(CN)6-3H20. This should be chemically pure and contain the full amount of water of crystallization. If the latter is more or less, a correction must be made. Dissolve in water and make up to 250 cc. STANDARDIZATION OF ZNSO* SOLUTION. Measure out 25 cc. of the ferrocyanide solution into a beaker. Add about 50 cc. of water and 10 cc. of 10% H 2 SO 4 . Titrate with the ZnSO 4 solution from a burette. As an outside indicator use a 3% solution of ferric alum on * Dissolve 60 grams of FeCl 3 crystals in water, add 100 cc. of cone. HC1 and dilute to 1 liter. 558 TECHNICAL METHODS OF ANALYSIS Schleicher and Schull's drop reaction paper No. 601.* Place 1 drop of ferric alum solution on the paper and let it spread as far as it will. Place a drop of the solution being titrated so that its extreme edge, after spreading, just meets the edge of the ferric alum drop. If the two over-run, faulty results will follow. The end-point of the titration is reached when a blue coloration at the point where the 2 drops meet does not appear for a space of one minute. Questionable end-points may be detected by holding the test paper so that strong sunlight passes through it. The faintest trace of blue is readily detected in this way. Titrations should not be attempted by artificial light or on dull cloudy days. At least three titrations should be made, the first to find the approxi- mate amount of ZnSO 4 solution required. The second and third titrations should check each other closely. From the titration calculate the value of 1 cc. of ZnSO 4 solu- tion in terms of K 4 Fe(CN) 6 -3H 2 O. TITRATION OF PREPARED SOLUTION. Place in a beaker 100 cc. of the solution prepared as described under Extraction of Prussian Blue. Add a drop or two of methyl orange indicator, neutralize with 10% H 2 SO 4 and then add 10 cc. excess. Titrate with stand- ard ZnSO 4 solution, using ferric alum as outside indicator, exactly as in the standardization. From the strength of the ZnSO 4 solution calculate the per cent of K 4 Fe(CN)6-3H2O both in the sample as received and on the dry basis. NOTE. The entire success of the method depends upon obtaining the correct end-point. Titrations must be carried out in strong sunlight, never by artificial light. REFERENCES. Proceedings of American Gas Institute, 7, 761; R. H. Royle: " Chemistry of Gas Manufacture," page 174; " Gas Chemists' Handbook. " * This is the only paper which we have found satisfactory. Other papers, such as Whatman's, give a blue color directly with the reagent. They can be made satisfactory, however, by treatment with very dil. HC1 as follows: Place the entire sheet in a large beaker and cover it with HC1 of between 2 and 5% strength. Boil gently or digest on the steam bath for about one- half hour. Wash free from acid; then continue washing until free from chlo- rine. Dry in the air protected from chemical fumes. MISCELLANEOUS ANALYSES 559 MORTAR AND CONCRETE General. Structural concrete consists of sand, gravel and cement. In mortar the gravel (large aggregate) is omitted. In some mortars also lime is used. These lime mortars generally contain an amount of lime about equal to the amount of cement. The setting of cement is a process of hydration; therefore, in order to obtain the composition of the original material, the analysis must be made on the sample after ignition.. Gravel. Any stone or coarse aggregate which will not pass through a sieve with J-inch openings * is considered gravel; material finer than J-inch is considered sand. If the composition contains gravel, the whole sample should be weighed and disin- tegrated with a hammer or mortar and pestle, taking care not to crush or break the stone or sand particles. Knock off the cement from the large particles. Then determine the percentage of the whole sample which will not go through a J-inch sieve and report as " gravel." Thoroughly mix the finer portion, and weigh out a known quantity, generally 10-50 grams, depending upon the size of the particles. Ignite and determine loss on ignition. Then proceed with the determinations of Sand, etc., as below, beginning " Treat with a considerable volume of dil. HC1." Sand. If the original material is fine and appears to be fairly uniform (as in the case of mortars and surfacings), crush gently and mix thoroughly. Ignite a portion and weigh out 5-10 grams of the ignited sample. Treat with a considerable volume of dil. HC1 (not stronger than 1 part cone, acid to 10 parts water) and warm on the steam bath. Decant the liquid through a filter into a graduated flask. (If CaO and MgO are not to be determined, discard the filtrate; see Note 2.) Add a second portion of dil. HC1, again warm and decant. Continue this until all soluble matter has been removed. Wash with hot water, finally trans- ferring all the residue to the filter. Ignite in a weighed crucible, cool in a desiccator and weigh. This gives the sand. NOTES. (1) Some of the SiO 2 from the cement may be thrown out of solution as silicic acid. The amount, however, is small if dilute acid is used * This is not a 4-mesh sieve, i.e., a sieve with four openings to the inch. 560 TECHNICAL METHODS OF ANALYSIS and is largely compensated by the small amount of iron and alumina which is dissolved from the sand. (2) In case of concrete and mortars free from added lime or limestone, it is sufficiently accurate to consider the rest of the material as cement. From the loss on ignition of the sand plus cement calculate what the weight of the entire sample would have been after ignition and on this weight figure the percentages of gravel and sand, and take cement " by difference." Total Lime and Magnesia. If the material contains added lime or limestone, make the filtrate up to volume and take an aliquot representing about 2 grams of the original ignited portion. Make slightly alkaline with NH^OH, boil, and filter out any Fe and Al hydroxides. To the hot filtrate add an excess of (NH4) 20264 solution. Heat to boiling and let stand until the precipitate settles clear. Filter, wash with hot water and ignite over a blast lamp. Cool in a desiccator and weigh as CaO. (The CaC204 may also be titrated with 0.1 N KMnO4 in the usual way instead of igniting.) Test a little of the filtrate from the CaO determination for MgO. If any appreciable amount is present it must be determined as follows: Return the port 'on of the filtrate on which the qualitative test was made to the main filtrate. Make slightly acid with HC1 and evaporate until crystallization begins. Cool and dilute suf- ficiently to redissolve any crystals. Add a considerable excess of a solution of Na or NH4 phosphate and then make strongly ammo- niacal. Let stand overnight (or cool in ice water and stir for about one-half hour). Filter through a weighed Gooch crucible, wash with a mixture of 1 part of NH40H (1 : 1), 1 part of alcohol, and 3 parts of water. Ignite very gently at first and finally blast thoroughly. Cool in a desiccator and weigh as Mg2?2O7. Cal- culate to MgO. CALCULATION. Mg 2 P 2 O 7 X 0.3621 = MgO. NOTE. Since the entire procedure gives of necessity only approximate results, it is not necessary to make a double precipitation of the CaC 2 O 4 before determining Mg. CALCULATIONS. If the brand of cement is known, look up the amount of CaO it contains. If the brand is not known, assume that it contains 62% CaO. Then, assuming that the ignited material consists entirely of sand, cement, and free lime (and MgO), MISCELLANEOUS ANALYSES 561 and that the cement itself contains 62% CaO (+MgO), calculate the proportions in the sample as follows: Subtract the per cent of sand and gravel from 100%. The difference is cement + CaO + MgO. (This is B below.) Let A = Total CaO+MgO (found by analysis) ; B = Cement + free CaO+free MgO; and X = Cement ; then B-X = Free (CaO+MgO), and A = 5-^+0.62 X = B-0.38X. Or 0.38X = B-A; NOTE. If high-grade lime was used, there will be very little MgO present; but if dolomitic lime was used, there will be considerable MgO. * SAMPLING AND PHYSICAL TESTING OF PORTLAND CEMENT General. The following procedure is based on specification C 9-17 of the American Society for Testing Materials. Portland cement is there defined as " the product obtained by finely pulver- izing clinker produced by calcining to incipient fusion an intimate and properly proportioned mixture of argillaceous and calcareous materials, with no additions subsequent to calcination excepting water and calcined or uncalcined gypsum." Portland cement is usually purchased on the basis of physical tests only, and unless expressly requested, no chemical tests are necessary. Specification for Physical Requirements. The standard speci- fications of the A. S. T. M. were revised in 1916 to become effective January 1, 1917. Unless otherwise instructed, cement is to be tested according to the revised specifications. As some engineers, however, request tests according to the old specifications, both requirements are given below: 562 TECHNICAL METHODS OF ANALYSIS Old Specifications New Specifications Maximum Minimum Maximum Minimum Specific Gravity: Ordinary Portland cement. White Portland cement. . . Fineness, per cent: Residue on 100-mesh sieve. . Residue on 200-mesh sieve . Initial set: Gillmore needle . 3.10 3.10 3.07 8%" 25% 22% 60 minutes 45 minutes Vicat needle Final set: Gillmore needle . 30 minutes 10 hours 10 hours Vicat needle 10 hours 1 hour 175 Ibs. 500 Ibs. 600 Ibs. 200 Ibs. 275 Ibs. Tensile strength (neat) 24 hours 7 days 28 days Tensile strength (1:3 mortar) 7 days 200 Ibs. 300 Ibs. 28 days A pat of neat cement shall remain firm and hard and show no signs of distortion, cracking, checking, or disintegration in the steam test for Soundness described below. A bag of cement shall contain 94 Ibs. net; a barrel, 376 Ibs. net. NOTES. (1) If the sample under test falls below the sp. gr. requirement, a second test may be made on an ignited sample. The sp.gr. test is not required by the revised specifications and will not be made unless requested. (2) The average tensile strength of standard mortar at twenty-eight days shall always be higher than the strength at seven days. (3) At least ten days from the time of sampling shall be allowed for the completion of the seven-day test and thirty-one days for the twenty-eight-day test. (4) The Gillmore needle is used on setting tests in this laboratory unless otherwise requested. MISCELLANEOUS ANALYSES . 563 Rejection. Cement may be rejected if it fails to meet any of the requirements of the specifications. It shall not be rejected on account of failure to meet the fineness requirement if, upon retest after drying at 100 C. for one hour, it meets this require- ment. Cement failing to meet the soundness test in steam may be accepted if it passes a retest, using a new sample, at any time within twenty-eight days thereafter. Packages varying more than 5% from the specified weight may be rejected; and if the average weight of 50 packages taken at random from any shipment is less than that specified, the entire shipment may be rejected. Sampling. Tests may be made on individual or composite samples. Each test sample should weigh at least 8 Ibs. Samples should preferably be shipped and stored in air-tight con- tainers. (a) INDIVIDUAL SAMPLE If sampled in cars, take one test sample from each 50 bbls. or fraction thereof. If sampled in bins, take 1 sample from each 100 bbls. (6) COMPOSITE SAMPLE. If sampled in cars, take one sampler- ful from 1 of each 40 sacks (or from 1 of each 10 bbls.) and com- bine them to form 1 test sample. If sampled in bins or ware- houses, take 1 test sample for each 200 bbls. or fraction thereof. (c) SAMPLING AT MILL. Use any of the following methods that may be practicable, as ordered : (1) From the conveyor delivering to the bin. Take at least 8 Ibs. of cement from approximately each 100 bbls. passing over the conveyor. (2) From filled bins by means of proper sampling tubes. Tubes inserted vertically may be used for sampling cement to a maximum depth of 10 feet. Tubes inserted horizontally may be used where the construction of the bin permits. Samples shall be taken from points well distributed over the face of the bin. (3) From filled bins at points of discharge. Sufficient cement shall be taken from the discharge openings to obtain samples representative of the cement contained in the bin, as determined by the appearance at the discharge openings of indicators placed on the surface of the cement directly above these openings before drawing of the cement is started. 564 TECHNICAL METHODS OF ANALYSIS Preparation of Sample. Pass the sample through a sieve hav- ing 20 meshes per linear inch in order to thoroughly mix the sample, break up lumps and remove foreign material. Specific Gravity. Determine the sp. gr. with the Le Chatelier apparatus, standardized by the Bureau of Standards. Use kero- sene, free from water, or benzine not lighter than 62 Baume. Fill the flask with either of these liquids to a point on the stem between and 1 cc. and introduce slowly 64 grams of cement of the same temperature as the liquid, taking care that the cement does not adhere to the inside of the flask above the liquid and to free the cement from air by rolling the flask in an inclined position. After all the cement is introduced, the level of the liquid will rise to some division of the graduated neck. The difference between readings is the volume displaced by 64 grams of the cement. Cal- culate the sp. gr. from the formula: g r == _ 64_ displaced volume (cc.)' Keep the flask immersed in water during the operation to avoid variations in the temperature of the liquid in the flask (which should not exceed 0.5 C.) and report results to 2 decimal places. Check determinations should agree within 0.01. NOTE. The sp. gr. is to be determined on the cement as received; if it falls below 3.10 (or 3.07 for white cement), make a second determination after igniting the sample in a muffle at a temperature between 900 and 1000 C. Ignite at this temperature for fifteen minutes and then for periods of five minutes to constant weight. Fineness. Place 50 grams of cement on a clean, dry 200-mesh sieve with pan and cover attached, if desired, and hold in one hand in a slightly inclined position so that the sample will be well distributed over the sieve, at the same time gently striking the side about 150 times per minute against the palm of the other hand on the upstroke. Turn the sieve every 25 strokes about one-sixth of a revolution in the same direction. Continue the operation until not more than 0.05 gram passes through in one minute of continuous sieving. Weigh the residue on the sieve in grams and multiply by 2 to obtain the per cent of residue. NOTES. (1) A permissible variation of 1% will be allowed and all results between 22.0 and 23.0% shall be reported as 22.0%. MISCELLANEOUS ANALYSES 565 (2) Routine sieving is done in this laboratory on a Ro-tap mechanical sifter. The sifter is run for fifteen minutes, the fine material in the pan discarded and then the sieving continued for five-minute periods until no more material comes through the sieve. In case, however, any sample fails to pass on the Ro-tap tester, it shall not be rejected until checked by the hand method above described. (3) The 200-mesh sieve used must be standardized by the U. S. Bureau of Standards. It should have 200 wires per inch and the number of wires in any whole inch should not be less than 192 nor more than 208. No opening be- tween adjacent parallel wires should be more than 0.0050 inch. The diameter of the wire should be 0.0021 inch and the average diameter should not be less than 0.0019 nor more than 0.0023 inch. The value of the sieve as determined by sieving tests made in conformity with the standard specification for these tests on a standardized cement which gives a residue of 25-20% on the 200-mesh sieve, or on other similarly graded material, should not show a variation of more than 1.5% above or below the standards maintained at the Bureau of Standards. (4) The old specifications required a 100-mesh sieve test. This test is made in the same way as the 200-mesh test, using a 100-mesh sieve of the following specifications : Diameter of wire, inch 0. 0042-0 . 0048 Meshes per linear inch Warp 95-101 Woof 93-103 Preparation of Neat Paste or Mortars. The quantity of dry material to be mixed at one time must be between 500 and 1000 grams. Weigh out the dry materials to the nearest gram, place on a non-absorbent surface (glass is satisfactory), thoroughly mix dry, if sand is used, and form a crater in the center, into which pour the proper percentage of clean water. Turn the material on the outer edge into the crater by means of a trowel. After an interval of one-half minute for the absorption of the water, complete the operation by continuous, vigorous mixing, squeezing and kneading with the hands for at least one minute. Protect the hands by rubber gloves during the mixing. NOTES. (1) The temperature of the room and the mixing water should be maintained as nearly as practicable at 21 C. (70 F.). (2) In order to secure Uniformity in the results of tests for the time of setting and tensile strength, the manner of mixing as above described should be carefully followed. At least one minute is necessary to obtain the desired plasticity, which is not appreciably affected by continuing the mixing for several minutes. The exact time necessary depends upon the personal equa- tion of the operator. Any error in mixing should be on the side of over-mixing. 566 TECHNICAL METHODS OF ANALYSIS Normal Consistency. Use the standard Vicat apparatus (Fig. 27), which consists of a frame bearing a movable rod weighing 300 grams, one end being 1 cm. in diameter for a distance of 6 cm. and the other having a removable needle 1 mm. in diameter, 6 cm-, long. The rod is reversible and may be held in any desired posi- tion by a screw, and has midway between the ends a milli- meter scale attached to the frame. The paste is held in a conical, hard-rubber ring, 7 cm. in diameter at the base, 4 cm. high, resting on a glass plate about 10 cm. square. FIG. 27. Vicat Apparatus. In making the determination, weigh out 500 grams of cement and knead into a paste with a measured quantity of water as described previously. Quickly form into a ball with the hands, completing the operation by tossing it six times from one hand to the other, maintained about 6 inches apart. Press the ball, resting in the palm of one hand, into the larger end of the rubber ring held in the other hand, completely filling the ring with paste; then remove the excess at the larger end by a single movement of the palm of the hand. Place the ring on its larger end on the glass plate and slice off the excess paste at the MISCELLANEOUS ANALYSES 567 smaller end (top) of the ring by a single oblique stroke of a trowel held at a slight an*gle with the top of the ring. During these operations take care not to compress the paste. Place the paste confined in the ring, resting on the plate, under the rod and bring the larger end of the rod in contact with the sur- face of the paste; read the scale and quickly release the rod. The paste is considered to be of normal consistency when the rod settles to a point 10 mm. below the original surface in one-half minute after being released. The apparatus must be free from all vibrations during the test. Make trial pastes with varying percentages of water until the normal consistency is obtained. Express the amount of water required in percentage by weight of the dry cement. Standard Mortar (1 : 3). The consistency of standard mortar depends upon the amount of water required to produce a paste of normal consistency from the same sample of cement. Having determined the normal consistency of the cement sample, the con- sistency of standard mortar made from the same sample shall be as indicated in the table below, the values being given in per- centage of the dry weights of the cement and standard sand. PERCENTAGE OF WATER FOR STANDARD MORTARS Normal Consistency Per Cent of Water for Neat Cement Paste Per Cent of Water for 1 : 3 Mortar Normal Consist- ency Per Cent of Water for Neat Cement Paste Per Cent of Water for 1 : 3 Mortar 15 16 9.0 9.2 23 24 10.3 10.5 17 18 9.3 9.5 25 26 10.7 10.8 19 20 9.7 9.8 27 28 11.0 11.2 21 22 10.0 10.2 29 30 11.3 11.5 Soundness. Make a pat of cement paste of normal consistency about 3 inches in diameter, 0.5 inch thick at the center and taper- 568 TECHNICAL METHODS OF ANALYSIS ing to a thin edge, on a clean glass plate about 4 inches square and store it in moist air for twenty-four hours. 'In molding the pat, first flatten the cement paste on the glass and then form the pat by drawing the trowel from the outer edge toward the center. Then place the pat in an atmosphere of steam at a temperature between 98 and 100 C. upon a suitable support .1 inch above boiling water for five hours. At the end of this time, examine the pat for shrinkage, absorption, cracking, checking or dis- integration. The pat should be firm and sound. NOTE. Should the pat leave the plate, distortion may be detected best with a straight edge applied to the surface which was in contact with the plate. Time of Setting. The time of setting may be determined either with the Vicat needle, previously described, or with the Gillmore needle. Unless otherwise directed, tests in this labora- tory are made with the Gillmore needles. (a) With Gillmore needles. Make a pat of neat cement about 3 inches in diameter and 0.5 inch in thickness with a flat top, mixed to normal consistency, and keep in moist air at a temperature main- tained as nearly as practicable at 21 C. Test at intervals with the standard Gillmore needles. (Fig. 28.) The cement shall be con- sidered as having acquired its initial set when the pat will bear, without appreciable indentation, the Gillmore needle T V inch in diameter, weighing 0.25 Ib. The cement has acquired its final set when the pat will bear without an appreciable indentation, the Gillmore needle -$ inch in diameter, weighing 1 Ib. In making the test hold the needles in a vertical position and apply lightly to the surface of the pat. (6) With Vicat needle. Make a paste of neat cement of normal consistency and mold it into the hard rubber-ring as previously described under Normal Consistency above. Place it under the rod and then carefully bring the smaller end of the rod in contact with the surface of the paste and quickly release the rod. The paste has reached the condition of initial set when the needle ceases to pass a point 5 mm. above the glass plate in one-half minute after being released ; and final set, when the needle does not sink visibly into the paste. The test pieces must be kept in moist air during the test. The needle must be kept clean, as the collection of cement on the sides MISCELLANEOUS ANALYSES 569 retards the penetration, and cement on the point may increase the penetration. NOTE. Time of setting is affected not only by the percentage and tem- perature of the water, but by the temperature and humidity of the air, and its determination is, therefore, only approximate. Tensile Strength. The tensile strength tests are made on briquettes having a cross section of 1 square inch. For the neat tests use a neat paste made up to normal consistency. For mortar Rat w'tb Top Surface Flattened for Determining Time of Setting by Gillroore Method. Q (Z>) Gil (more Needles. FIG. 28. tests make a standard mortar according to the table previously given under Normal Consistency, using 1 part of cement to 3 parts of standard Ottawa sand, by weight. Use standard briquette molds as specified by the A. S. T. M. Make 4 briquettes for each 28-day test and 3 for each of the other tests called for. Immediately after mixing, place the paste or mortar in the molds (wipe the molds with an oily cloth before using) , pressing in firmly with the thumbs and smoothing off with a trowel without ramming. Then heap up additional mortar (or paste) above 570 TECHNICAL METHODS OF ANALYSIS the mold and smooth off with a trowel. Draw the trowel over the mold in such a manner as to exert a moderate pressure on the ma- FIG. 29. Fairbanks Cement Testing Machine. DIRECTIONS FOR USE Hang the cup F on the end of the beam D, as shown in the illustration. See that the poise R is at the zero mark, and balance the beam by turning the ball L. Fill the hopper B with fine shot (of which a bag is provided with each machine) . Place the briquette in the clamps N N. Tighten the hand wheel g sufficiently to cause the graduated beam D to rise to the stop K. Only enough pressure should be exerted to hold the beam firmly against the stop; not enough to transmit any strain to the specimen. Open the auto- matic valve J to allow the shot to run into the cup F. At the point where the spout joins the reservoir there is a small valve, by which the flow of shot may be regulated. Better results will be obtained by allowing the shot to run very slowly into the cup. When the briquette breaks, the beam D will drop and automatically close the valve J. terial. Then turn the mold over and repeat the operation of heap- ing, thumbing and smoothing off. Make the tests on the standard MISCELLANEOUS ANALYSES 571 testing machine (Fig. 29), which should be frequently calibrated in order . to determine its accuracy. The briquettes should be tested as soon as they are removed from the water. See that the bearing surfaces of the clips and briquettes are free from grains of sand or dirt. Carefully center the briquettes and apply the load continuously at the rate of 600 Ibs. per minute. In reporting results give the breaking strength of each briquette and the average of the 3 briquettes. NOTES. (1) The fourth briquette on the 28-day test is made up only for use in case anything goes wrong with one of the other briquettes. If any of the first three 28-day briquettes appears to be faulty or gives a breaking strength widely at variance with the other two, then use the fourth briquette in place of the faulty one. Otherwise do not break the fourth briquette. (2) Briquettes which are manifestly faulty or which give strengths differing more than 15% from the average value of all test pieces made from the same samples and broken at the same period, will not be considered in determining the tensile strength. (a) Storage of test briquettes. The moist closet may consist of a soapstone, slate or concrete box, or a wooden box lined with metal. If a wooden box is used, the interior should be covered with felt or broad wicking kept wet. The bottom of the moist closet should be covered with water. The interior of the closet should be provided with non-absorbent shelves on which to place the test pieces. Unless otherwise specified, all test pieces, immediately after molding, shall be placed in the moist closet for twenty to twenty- four hours. The briquettes should be kept in the molds on glass plates in the moist closet for at least twenty hours. After twenty- four hours in moist air the briquettes should be immersed in clean water in storage tanks of non-corroding material. The air and water should be maintained as nearly as possible at a temperature of 21 C. The briquettes for the various tests shall be stored as follows : 24-four-hour test: Twenty-four hours in moist air. 7 r day test : One day in moist air, 6 days in water. 28-day test : One day in moist air, 27 days in water. (6) Standard Ottawa Sand. Standard sand- shall be natural sand from Ottawa, Illinois (obtained from the Ottawa Silica Com- pany), and screened to pass a No. 20 sieve, but not a No. 30 sieve. The sand, having passed the No. 20 sieve, shall be considered 572 TECHNICAL METHODS OF ANALYSIS standard when not more than 5 grams pass the No. 30 sieve after one minute of continuous sieving of a 500-gram sample. ' . REFERENCE. American Society for Testing Materials, Triennial Stand- ards (1918), page 503, ff. CHEMICAL ANALYSIS OF PORTLAND CEMENT General. Portland cement, according to Le Chatelier, con- sists of a mixture of dry calcium silicate and dry calcium aluminate. It may also contain, and generally does contain, small amounts of magnesia and of calcium sulfate. The average analysis of 13 samples of different American Portland Cements * is as follows: Per cent Silica, Si0 2 21.85 Iron oxide, Fe 2 3 2.62 Alumina, A^Oa 7 . 03 Lime, CaO 62.50 Magnesia, MgO 2 . 06 Sulfur trioxide, S0 3 1 .38 Loss on ignition 1 . 80 It is seldom necessary, however, to make a complete chemical analysis. A partial analysis will show whether the cement has been adulterated or is unsatisfactory. Specifications. The American Society for Testing Materials, under specifications C9-17, gives the following chemical require- ments for Portland cement: Maximum Per cent Loss on ignition 4 . 00 Insoluble residue . 85 Sulfur trioxide, S0 3 2.00 Magnesia, MgO 5 .00 The methods given herewith, so far as they apply to* the above determinations are according to the American Society for Testing Materials requirements. Loss on Ignition, Heat 1 gram in a weighed, covered platinum * Meade: " Portland Cement " (1906), page 16. MISCELLANEOUS ANALYSES 573 crucible of 20-25 cc. capacity, using either of the following methods as ordered: (A) Place the crucible in a hole in an asbestos board, clamped horizontally so that about three-fifths of the crucible projects below, and blast at a full red heat for fifteen minutes with an inclined flame. Cool in a desiccator and weigh. Check the loss in weight by a second blasting for five minutes. Take care to wipe off any particles of asbestos that may adhere to the crucible when withdrawn from the asbestos board. (B) Place the crucible in a muffle heated to 900-1000 C. for fifteen minutes and cool in a desiccator and weigh. Check the weight by a second heating for five minutes. NOTE. A permissible variation of 0.25% is allowed and all results between 4.00 and 4.25% shall be reported as 4.00%, when the cement is bought to A. S. T. M. specifications. Insoluble Residue. Treat 1 gram of the sample in a beaker with 10 cc. of water and 5 cc. of cone. HC1 and warm until effer- vescence ceases. Dilute to 50 cc. and digest on the steam bath or hot plate until decomposition is complete. Filter the residue and wash with cold water. Digest the filter paper and contents in about 30 cc. of a 5% solution of Na2CO3, keeping the liquid at just below boiling for fifteen minutes. Filter this residue, wash with cold water, then with a few drops of hot HC1 (1 * 9) and finally with hot water. Ignite at red heat, cool in a desiccator and weigh. NOTE. A permissible variation of 0.15% will be allowed and all results between 0.85 and 1.00% shall be reported as 0.85%, when the cement is pur- chased to A. S. T. M. specifications. Sulfur Trioxide. If the insoluble residue has been determined, use the acid filtrate for the determination of SOs. Otherwise, dissolve 1 gram of the cement in 10 cc. of HC1 (1:1) with gentle warming. When the solution is complete, add 40 cc. of water. Filter and wash the residue thoroughly with water. Dilute to 250 cc., heat to boiling and add 10 cc. of a hot 10% solution of BaCl2, slowly, drop by drop, from a pipette, and continue boiling for fifteen minutes. Digest on the steam bath until the pre- cipitate has settled. Filter and wash with hot water. Place the paper and contents in a weighed platinum crucible and slowly char 574 TECHNICAL METHODS OF ANALYSIS the paper until consumed without burning. Then ignite, cool in a desiccator and weigh as BaSC>4. Calculate to 80s. CALCULATION. BaSO 4 X 0.3430 = SO 3 . NOTE. A permissible variation of 0.10% will be allowed and all results between 2.00 and 2.10% shall be reported as 2.00%, when the cement is pur- chased to A. S. T. M. specifications. Silica. Place 0.5 gram of the cement in an evaporating dish, add 10 cc. of water to prevent lumping and then 10 cc. of cone. HC1. Heat gently and agitate until decomposition is complete. Then evaporate to complete dry ness on the steam bath. Heat the residue to about 150 C. for one-half to one hour. Take up with 20 cc. of HC1 (1 : 1), cover the dish and digest for ten min- utes on the steam bath. Dilute and filter through a quantitative filter, washing thoroughly with hot water. Evaporate the filtrate again to dryness on the steam bath. Take up with HC1 (1:1), digest for ten minutes on the steam bath and filter through a fresh quantitative filter, washing with hot water. Place both papers in a weighed platinum crucible, dry, blast to constant weight and weigh as SiO2- NOTE. If the silica determination is not required, it is not necessary to make the second evaporation, nor, of course, to weigh the SiCV Iron Oxide and Alumina. To the filtrate from the SiO 2 deter- mination (about 250 cc.), add 5 cc. of cone. HC1 and sufficient bromine water to precipitate any Mn which may be present. Make alkaline with NH40H, and boil until the odor of NH 3 is nearly but not quite gone. Let the precipitate settle and wash once by decantation and then slightly on the filter paper. Set aside the filtrate and transfer the precipitate by a jet of hot water to the original beaker. Dissolve in 10 cc. of hot HC1 and extract the filter paper with acid, adding the solution and wash- ings to the main solution. Then reprecipitate the iron and alumina at boiling heat by NH4OH and bromine water in a volume of about 100 cc. Collect the precipitate, washing on the filter previously used, if this is still intact. Transfer to a weighed platinum crucible, ignite in a blast lamp, cool and weigh. The above precipitate consists of Fe2OsH-Al2O3+Mn3O4. For general purposes it is sufficient to report this as " iron oxide and alumina." If desired, the amounts of Fe20a and Mn 3 04 may be determined and subtracted from the total precipitate to MISCELLANEOUS ANALYSES 575 determine the amount of A^Os- The amount of Mn is generally insignificant, and the iron may be determined by fusing the ignited precipitate with KHSCU and passing the solution of the fusion (made acid with fl^SCU) through a Jones reductor, titrating the iron with standard KMnO4 solution. (See page 148.) Lime. To the combined filtrates from the iron and alumina, somewhat evaporated if necessary, add 1 cc. of cone. NH 4 OH, heat to boiling and add 25 cc. of a saturated, boiling solution of (NH4)2C2C>4. Boil until the precipitate settles well, let stand for one hour, filter and wash with hot water. Place the filter while still wet in a platinum crucible (unweighed) and burn off the paper over a low flame, finally igniting until the paper is consumed. Dissolve this residue in HC1 and dilute to 100 cc. Add NH^OH in slight excess, heat to boiling and reprecipitate the lime with (NH4)2C204,* let stand until settled clear; then filter and wash with hot water. The precipitate of CaC2O4 may be dissolved in H2SO4 and titrated hot with 0.1 N KMnO4 (see page 326) or ignited strongly in a weighed platinum crucible, cooled in a desiccator and weighed rapidly as CaO. Magnesia. To the combined filtrates from the CaC2O4, add a slight excess of HC1, concentrate on the steam bath to about 150 cc., make slightly alkaline with NH 4 OH, boil, and filter, if necessary. (There may be a slight precipitate of iron and alumina and perhaps calcium salts.) When cool, add 10 cc. of a saturated solution of NaNH4HPO4, with constant stirring. When the crystalline NH^MgPCU has formed, add a moderate excess of cone. NH40H. Set aside for several hours, preferably overnight, filter and wash with, water containing 2.5% of NHs. Dissolve the precipitate in a small quantity of HC1, dilute to about 100 cc. and add 1 cc. of a saturated solution of *NaNH4HPO4 "and then cone. NELiOH, with constant stirring, until the crystalline precip- itate is again formed and the NH^OH is in moderate excess. Let stand for about two hours. Filter through an ignited and weighed Gooch crucible, washing as before. Ignite the precipitate to constant weight over a Meker burner or blast lamp not strong enough to soften or melt the pyrophosphate. Cool in a desiccator and weigh as Mg2?2O7. Calculate to MgO. CALCULATION. Mg 2 P 2 O 7 X 0.3621 = MgO. 576 TECHNICAL METHODS OF ANALYSIS NOTES. (1) A permissible variation of 0.4% will be allowed and al) results between 5.00 and 5.40% shall be reported as 5.00%, when the cement is bought to A. S. T. M. specifications. (2) In case only the MgO .is desired all the steps in the above procedure must be followed beginning with " Place 0.5 gram of the cement in an evap- orating dish," etc., except that a double evaporation of the silica solution is not necessary, nor, of course, is it necessary to do anything with the iron and alumina precipitates or the second CaC 2 O 4 precipitate. REFERENCE. American Society for Testing Materials, Triennial Stand- ards (1918), pages 506-509. MECHANICAL TESTING OF SAND AND GRAVEL FOR USE IN REIN- FORCED CONCRETE General. Sand and gravel for use in reinforced concrete con- struction have been the subject of considerable study, but thus far no official specifications for testing have appeared. The following methods, however, have been used for some time in this laboratory with good results. It is to be understood that they apply only to material for use with cement in concrete construction. Definitions. (1) SAND OR FINE AGGREGATE. This con- sists of natural sand, crushed stone or gravel screenings, graded from fine to coarse, and completely passing a quarter-inch screen in the dry condition. It should be free from dust, loam, soft particles, clay lumps, and organic matter. Not more than 6% should pass a 100-mesh sieve and it should be free from an excessive amount of mica. (2) GRAVEL OR COARSE AGGREGATE. This consists of natural gravel, crushed stone, cinders, or slag, graded from small to large particles, none of which, however, will pass through a quarter-inch screen. It should be clean and free from sticks, leaves, roots, or other organic matter. Bank gravel should be screened on a quarter-inch screen before mixing in construction work. (3) SLAG. When slag is used it should be clean, air-cooled blast furnace slag weighing not less than 75 Ibs. per cubic foot and containing not over 1.3% of sulfur as sulfides. (4) CINDERS. Where cinders are used as coarse aggregate they should be composed of hard, clean, vitreous clinker, free from unburned coal or ashes and from sulfides. General Appearance. Examine the sample as received and MISCELLANEOUS ANALYSES 577 note whether or not it is clean and contains any of the impurities mentioned above. Mechanical Analysis (Fineness). The sample must be dry before starting screening tests. COARSE OR MIXED AGGREGATE. If the sample looks as though it would practically all pass a J-inch sieve, weigh out 500 grams. If it contains many particles coarser than J-inch, weigh out a- con- siderably larger amount, depending upon the proportion and size of the large particles. For samples containing particles as large as 2 inches, as much as 5-10 kilograms should be weighed. Pass the sample successively through the various screens, beginning with the largest and continuing down to and including the J-inch screen. Weigh the residue retained on each screen. Calculate the amounts passing each screen as follows : Subtract the amount remaining on the largest screen from the total amount of sample weighed out. This gives the weight passing this screen. From this weight subtract the amount remaining on the next screen. This gives the weight passing that screen. From the latter weight subtract the amount retained on the third screen; and so on. Calculate these weights to percent- ages of the original sample. Mix that portion of the sample which passed the J-inch screen and weigh out 200 grams. Make sieving tests on this as described below under Sand. NOTES. (1) The screens generally used for coarse aggregate are 3, 3, 2, 2, l\, 1, f, , and Hnch, respectively. They should have square openings of uniform size. The above figures refer to the size of the openings; the 1-inch screen, for example, has openings 1 inch square. (2) The screening tests on the screens of Hnch and larger are calculated on the sample as received, whereas the figures on sieves smaller than J-inch should be calculated on the basis of the sand (material finer than |-inch) . (3) The Hnch sieve is taken as the dividing line between sand and gravel. SAND. If the sample is all finer than J-inch, weigh out 200 grams; otherwise weigh out 200 grams of the material which has passed a J-inch sieve. Place it on a 6-mesh sieve which is the top- most of a series of 7 brass sieves fitting into each other to form a nest. The other sieves in the nest are 8, 10, 20, 30, 50, and 100- mesh, respectively. Put the cover on the top sieve and a receiver under the bottom one, and place the nest in a Ro-tap mechanical 578 TECHNICAL METHODS OF ANALYSIS sieving machine. After five minutes' sieving, remove and weigh the residue in the bottom pan (passing 100-mesh). Return the pan and continue sieving for five minutes longer. Continue this until the increase in the amount of the fine residue in the pan is not more than 1 gram between subsequent shakings. Finally weigh the residue on each sieve and calculate the percentage passing each mesh as previously described. The following example shows the method of tabulating the figures and the form of the final report. The figures to be reported are in the last column. SCREENING TEST OF SAMPLE AS RECEIVED (5000 GRAMS TAKEN) Screen Grams Retained Grams Passing Per Cent Passing 3 inch None 5000 100.0 3 inch 500 4500 90.0 2 inch 750 3750 75.0 2 inch 1000 2750 55.0 U inch 250 2500 50.0 1 inch 100 2400 48.0 f inch 275 2125 42.5 \ inch 500 1625 32.5 \ inch 500 1125 22.5 SIEVING TEST OF SAND PASSING I-INCH (200 GRAMS TAKEN) Sieve Grams Retained Grams Passing Per Cent Passing ^ inch None 200 100.0 6-mesh 24 176 88.0 8-mesh 33 143 71.5 10-mesh 27 116 58.0 20-mesh 35 81 40.5 30-mesh 30 51 25.5 50-mesh 21 30 15.0 100-mesh 21 9 4.5 NOTES. (1) Some contractors require also a 200-mesh sieve test. In this case use the cement sieve described on page 565. (2) Sieving may be done by hand, using the sieves consecutively, always MISCELLANEOUS ANALYSES 579 starting with the largest; but much time is saved by the Ro-tap and it elim- inates the personal factor of the operator. (3) The sieves used should be accurately made and should be carefully tested out unless certified by the U. S. Bureau of Standards. The A. S. T. M. specifications for standard sieves * are as follows: Permissible Variations above and below Mesh Desig- nation Acutal Mesh (per inch) Opening (inch) WireDiam. (inch) Standard Mesh Diameter (per inch) (inch) 10 9.9 0.079 0.022 0.1 0.002 20 20.3 0.0335 0.0157 0.5 0.0006 30 30.5 0.0197 0.0130 1.0 0.0005 40 40.6 0.0142 0.0102 1.5 0.0004 50 50.8 0.0114 0.0083 2.0 0.0004 80 78.7 0.0067 0.0059 3.0 0.0003 100 99.1 0.0055 0.0046 3.0 0.0003 200 200.7 0.0029 0.0021 8.0 0.0002 Tensile Strength. Make up seven 1 : 3 mortar briquettes with standard Ottawa sand and any of the standard brands of Portland cement, f as described on page 569 under Tensile Strength. Make up seven more briquettes using the same cement, but, instead of standard Ottawa sand, use the sand in question (after removing any material coarser than J inch). Both sets of bri- quettes should be made up to standard mortar, based on the normal consistency of the cement used, and the per cent of water used should be stated. In the case of bank sand or crushed stone containing considerable fine material, it will be found that more water is required to make a mortar of the proper consistency than in the case of the standard Ottawa sand. Break 3 briquettes of each set at the end of seven days and * American Society for Testing Materials: Triennial Standards (1918), page 663. f The cement should be tested according to A. S. T. M. specifications (see page 561), and not used if it does not meet these specifications. 580 TECHNICAL METHODS OF ANALYSIS 3 at the end of twenty-eight days, respectively. The fourth briquette of the 28-day sets is only to be broken in case any- thing goes wrong with one of the other three [see page 571, under Note (1)]. Divide the average strength of the sample briquettes by the average strength of the standard sand briquettes broken at the same time, and multiply by 100 to obtain the per- centage strength. The following example shows the method of reporting: Tensile Strength, Ibs. per square inch: 1 cement: 3 sand Seven days 248-270-264 Average 261 Twenty-eight days 380-340-355 Average 358 Per cent of water used 11 . 0% 1 cement: 3 standard sand Seven days 238-258-250 Average 259 Twenty-eight days 370-328-341 Average 346 Per cent of water used 10 . 2% Percentage Strength of Standard Sand: Seven days 105% Twenty-eight days 103% NOTES. (1) A well-graded, clean sand should show a tensile strength more than 100% that of the standard sand at both periods. It is usually recommended that a sand which shows less than 70% of the standard sand's strength at the end of twenty-eight days should be rejected for use in reinforced concrete work. A satisfactory sand should also show a greater actual strength at the end of twenty-eight days than at the end of seven days. (2) Any obviously defective briquette should not be counted in the average nor any result which varies more than 50 Ibs. from the average. Color Test for Organic Matter. The method is based on a comparison of the color produced by the reaction of NaOH upon the organic matter in the sand with standard colors produced from known amounts of alkaline sodium tannate. To 200 grams of the dry sample (passing J-inch), add 100 cc. of 3% NaOH solution and digest at room temperature with occasional stirring for twenty-four hours. Filter and refilter, if MISCELLANEOUS ANALYSES 581 necessary, until the filtrate is absolutely clear. Place 10 cc. of the final clear filtrate in a 50 cc. Nessler tube and dilute to 50 cc. with distilled water. Mix thoroughly and let stand until all foam and bubbles disappear. Determine the color value of the liquid by comparing it with tubes containing standard solutions of alkaline sodium tannate, looking through the full depth of the solution with the cylinders held toward a good natural light. Standard Tannate Solution. The preparation of the standard solution for comparing the colors should be begun at the same time as the treatment of the sand. Add 10 cc. of a 2% solution of tannic acid in 10% alcohol to 90 cc. of a 3% solution of NaOH and let stand twenty-four hours at room temperature. Place 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 cc., respectively, of this solution in 50 cc. Nessler tubes, dilute to the mark with distilled water and mix. The following table shows the amount of tannic acid in each cylinder and the color value of the solution expressed in parts of tannic acid per million parts by weight : Tannate Solution cc. Tannic Acid mgs. Color Value 1 2 100 2 4 200 3 6 300 4 8 400 5 10 500 6 12 600 7 14 700 8 16 800 9 18 900 10 20 1000 It is desirable to have good sunlight for comparing the colors. If sunlight is not available the amount of tannic acid in each of the standard tubes may be decreased by one-half and the other values of the table modified accordingly. In case the solution obtained by digesting the sand with the NaOH is very dark, use less than 10 cc. for the comparison, and make the necessary modifications in the calculation. With very 582 TECHNICAL METHODS OF ANALYSIS light-colored solutions, use more than 10 cc. of the filt ate. The depth of the color of the solution decreases on standing, and, there- fore, the solution should be made up fresh for each day's work. CALCULATION. Using 10 cc. of the filtrate from the 200-gram sample, the color value of the sand is the same as that of the standard tube which it most nearly matches. If 5 cc. of the fil- trate are used, the color values should be doubled. Similar cor- rections should be made if any other volume than 10 cc. is em- ployed. NOTES. (1) This color test was developed by Abrams and Harder, published in Circular No. 1, Structural Materials Research Laboratory, Lewis Institute, Chicago, (1917). (2) Sand showing a color value greater than 250 should be looked upon with suspicion for use in reinforced concrete. Most good sands give color values between and 100. TABLES IN the following tables the weights and calculations are based on the table of International Atomic Weights for 1920. In com- puting the molecular weights in Table I the number of decimal places has been governed by the least number of decimal places in any one of the atomic weights of the elements entering into the chemical formula. For example, in the case of H^PtCle the cal- culated molecular weight would be 2X1.008+195.2+6X35.46 = 409.976. Since, however, the atomic weight of Pt is given only to one decimal place, the molecular weight of the compound should also contain only one decimal place and therefore will be found in table I as 410.0. In the tables of equivalents of volumetric solu- tions the values have, in most cases, been carried out to a sufficient number of places to give an accuracy of at least 0.1%; and for purposes of uniformity the factors in Table III have, as a rule, been calculated to four decimal places. In a few cases, where the values are more or less approximate or empirical, they are carried to a lesser number of decimal places. Five-place logarithms have been used throughout the tables for uniformity and greater accuracy. In most cases, however, sufficient accuracy would be obtained from four-place logarithms and the analyst is advised to "round off" the last place of the logarithms given, unless unusual accuracy is required. 583 584 TECHNICAL METHODS OF ANALYSIS TABLE I INTERNATIONAL ATOMIC WEIGHTS, 1920 Element Symbol Atomic Weight Logarithm Aluminum Al Antimony Sb Argon A Arsenic As Barium Ba Bismuth Bi Boron B Bromine Br Cadmium Cd Caesium Cs Calcium Ca Carbon C Cerium Ce Chlorine Cl Chromium Cr Cobalt Co Columbium ^ Cb Copper Cu Dysprosium Dy Erbium Er Europium Eu Fluorine F Gadolinium Gd Gallium Ga Germanium Ge Glucinum Gl Gold Au Helium He Holmium Ho Hydrogen H Indium In Iodine , I Iridium Ir Iron Fe Krypton Kr Lanthanum La Lead Pb Lithium -. . Li Lutecium Lu Magnesium Mg Manganese Mn 27.1 120.2 39.9 74.96 137.37 208.0 10.9 79.92 112.40 132.81 40.07 12.005 140.25 35.46 52.0 58.97 93.1 63.57 162.5 167.7 152.0 19.0 157.3 70.1 72.5 9.1 197.2 4.00 163.5 1.008 114.8 126.92 193.1 55.84 82.92 139.0 207.20 6.94 175.0 24.32 54.93 1.43297 2 . 07990 1.60097 1 . 87483 2.13789 2.31806 1.03743 1.90266 2.05077 2.12323 1.60282 1.07936 2.14691 1.54974 1 . 71600 1.77063 1 . 96895 1.80325 2.21085 2 . 22453 1 . 18184 1.27875 2.19673 1.84572 1.86034 0.95904 2.29491 . 60206 2.21352 0.00346 2.05994 2.10353 2.28578 1.74695 1.91866 2.14301 2.31639 0.84136 2.24304 1 . 38596 1 . 73981 TABLES 585 TABLE I INTERNATIONAL ATOMIC WEIGHTS, 1920 Element Symbol Atomic Weight Logarithm Mercury Hg Molybdenum Mo Neodymium Nd Neon Ne Nickel Ni Niobium Nb Niton (radium emanation) Nt Nitrogen N Osmium Os Oxygen O Palladium Pd Phosphorus P Platinum Pt Potassium K Praseodymium Pr Radium Ra Rhodium Rh Rubidium Rb Ruthenium Ru Samarium Sa Scandium Sc Selenium Se Silicon Si Silver Ag Sodium Na Strontium Sr Sulfur S Tantalum Ta Tellurium Te Terbium . Tb Thallium Tl Thorium Th Thulium Tm Tin Sn Titanium Ti Tungsten W Uranium U Vanadium V Xenon Xe Ytterbium (Neoytterbium) Yb Yttrium Yt Zinc Zn Zirconium. . Zr 200.6 96.0 144.3 20.2 58.68 2.30233 1.98227 2.15927 1.30535 1.76849 (See Columbium) 222.4 14.008 190.9 16.000 106.7 31.04 195.2 39.10 140.9 226.0 102.9 85.45 101.7 150.4 44.1 79.2 28.3 107.88 23.00 87.63 32.06 181.5 127.5 159.2 204.0 232.15 168.5 118.7 48.1 184.0 238.2 51.0 130.2 I'K.S 89.33 65.37 90.6 2.34713 1 . 14638 2.28081 1.20412 2.02816 1.49192 2.29048 1.59218 2 . 14891 2.35411 2.01242 1.93171 2.00732 2.17725 1 . 64444 1.89873 1.45179 2.03294 1 . 36173 1.94265 1.50596 2.25888 2 . 10551 2 . 20194 2.30963 2.36577 2.22660 2.07445 1.68215 2.26482 2.37694 1.70757 2.11461 2.23930 1.95100 1.81538 1 . 95713 586 TECHNICAL METHODS OF ANALYSIS TABLE II MOLECULAR AND ATOMIC-GROUP WEIGHTS Name Formula Molecular Weight Logarithm Acetaldehyde CH 3 CHO 44 042 64387 (Acetate Radical) 59 034 77110 Acetone (CH 3 ) 2 CO 58 063 76390 Acetic Anhydride (Acetyl Radical) ... (CH 3 CO) 2 O C 2 H 3 O 102.068 43 034 .00888 63381 Acetylene 26 026 41541 Acid, Abietic 302 34 2 48050 Acetic HC 2 H 3 O 2 60 042 1 77845 Arachidic 312 42 2 49474 Arsenic H 3 AsO 4 -H 2 O 150 99 2 17895 Arsenic, Anhyd H 3 AsO 4 141 98 2 15223 Benzoic HC 7 H 5 O 2 122 083 2 08666 Bichromic H 2 Cr 2 O 7 218 2 33846 Boric H 3 BO 3 61 9 1 79169 Butyric HC 4 H 7 O 2 88 084 1 94490 Carbonic H 2 CO 3 . . . 62 021 1 79254 Chlorauric, Anhyd HAuCL, 340 2 53148 Chlorauric, Cryst Chloric, Anhyd HAuCl 4 -4H 2 O HC1O 3 412.1 84 47 2.61500 1 92670 Chloric Cryst . HC1O 3 7H 2 O 210 58 2 32342 Chlorplatinic, Anhyd. . . . H 2 PtCl 6 410 2 61278 Chlorplatinic, Cryst H 2 PtCl 6 6H 2 O 518 1 2 71441 Chlorplatinous H 2 PtCl 4 339 1 2 53033 Chromic H 2 CrO 4 118 2 07188 Citric, Anhyd H 3 C 6 H 5 O 7 192 09 2 28351 Citric, Cryst Fluosilicic . . H 3 C 6 H 5 7 -H 2 H 2 SiF 6 210.11 144 3 2.32245 2 15927 Formic HCHO 2 46 021 1 66296 Hydriodic : HI 127 93 2 10697 Hydrobromic HBr 80 93 1 90811 Hydrochloric HC1 36 47 1 56194 Hydrocyanic HCN 27 021 1 43171 Hydrofluoric HF 20 1 30103 lodic HIO 3 175 93 2 24534 Lactic HC 3 H 5 O 3 90 063 1 95455 Malic 134 068 2 12732 Molybdic, Anhyd H 2 MoO 4 . 162 2 20952 Molybdic, Hydrated Nitric H 2 MoO 4 -H 2 O HNO 3 180.0 63 016 2.25527 1 79945 TABLES 587 TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm Acid Nitrous HNO 2 47 016 1 67224 Oleic 282 36 2 45080 Oxalic, Anhyd. . . . H 2 C 2 O 4 . 90.026 1.95437 Oxalic Cryst H 2 C 2 O 4 -2H 2 O. 126 058 2 10057 Palmitic HCi6HsiO 2 256.34 2.40882 Perchloric, Anhyd HC1O 4 100.47 2.00204 Perchloric, Cryst Persulfuric HC1O 4 -2H 2 O H 2 S 2 O 8 136.50 194.14 2.13513 2.28812 Phosphoric Hypo- H 2 PO 3 81 06 1 90881 Phosphoric, Meta- HPO 3 80.05 1 90336 Phosphoric, Ortho- . H 3 PO 4 98 06 1 99149 Phosphoric Pyro- H 4 P 2 O 7 178 11 2 25069 Phosphorous, Hypo- H 3 PO 2 66.06 1 . 81994 Phosphorous, Ortho-. H 3 PO 3 82.06 1 91413 Phosphotungstic P 2 O 5 -12WO 3 -42H 2 O 3682 8 3 56618 Prussic (see Hydrocyanic) Pyrosulfuric Salicylic H 2 S 2 O 7 HC 7 H 5 O 3 . 178.14 138 083 2.25076 2 14014 Selenic Anhyd H 2 SeO 4 145 2 2 16197 Selenic, Cryst Selenious H 2 SeO 4 -H 2 H 2 SeO 3 .... 163.2 129 2 2.21272 2 11126 Silicic Meta- H 2 SiO 3 78 3 1 89376 Silicic, Ortho- H 2 Si0 4 96.3 1.98363 Silicotungstic 4H 2 SiO 3 '12WO 3 - Stearic 22H 2 3494.6 284 38 3.54340 2 45390 Sulfocyanic HCNS 59 08 1 77144 Sulfuric . . H 2 SO 4 98 08 1 99158 Sulfurous H 2 SO 3 82 08 1 91424 Tannic 322.15 2 50806 Tartaric, Anhyd Tartaric Cryst xi^^-^-H^Oe H 2 C 4 H 4 O 6 -H 2 O 150.068 168 084 2.17629 2 22553 Tungstic H 2 WO 4 250 2 39794 Alum K 2 A1,(SO 4 ) 4 -24H 2 O 949.0 2.97727 Alumina (see Aluminum Oxide) Aluminum Acetate . . A1(C 2 H 3 O 2 ) 3 ... 204 2 2 31006 Aluminum Chloride Anhyd A1C1 3 . 133 5 2 12548 Aluminum Chloride, Cryst A1C1 3 -6H 2 O 241 6 2 38310 Aluminum Fluoride, Anhyd . . . A1F 3 84 1 1 92480 Aluminum Fluoride, Cryst A1F 3 -3|H 2 O 147.2 2.16791 Aluminum Hydroxide A1(OH) 3 78.1 1.89265 588 TECHNICAL METHODS OF ANALYSIS TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm Aluminum Oleate Al(Ci 8 H 33 O 2 ) 3 871 2 2 94012 Aluminum Oxide A1 2 O 3 102 2 2 00945 Aluminum Phosphate. . A1P0 4 .. 122.1 2.08672 Aluminum Potassium Fluoride (See Potassium Aluminum Fluoride) Aluminum Potassium Silicate (See Potassium Aluminum Silicate) Aluminum Potassium Sulfate (Se Aluminum Silicate e Alum) Al 2 Si 2 O 7 -H 2 O 240 8 2 38166 Aluminum Sulfate Anhyd A1 2 (SO 4 ) 3 342 4 2 53453 Aluminum Sulfate, Cryst (Amino Radical) A1 2 (SO 4 ) 3 -18H 2 O.... NH 2 666.7 16 024 2 . 82393 1 20477 Ammonia NH 3 17 032 1 23126 (Ammonium Kadical) NH 4 18 040 1 25624 Ammonium Acetate NH 4 C 2 H 3 O 2 77 074 1 88691 Ammonium Alum (NH 4 )oAl 2 (SO 4 ) 4 - 24H 2 O 906 9 2 95756 Ammonium Bichromate (NH 4 ) 2 Cr 2 O 7 252.1 2 40157 Ammonium Bromide NH 4 Br 97 96 1 99105 Ammonium Carbonate (NH 4 ) 2 CO 3 - Ammonium Carbonate, Cryst . . '. Ammonium Chloride Ammonium Chlorplatinate .... NH 4 CO 2 -NH 2 .... (NH 4 ) 2 C0 3 -H 2 0.... NH 4 C1 (NH 4 ),PtCl 6 174.154 114.101 53.50 444 2.24094 2.05729 1.72835 2 64738 Ammonium Chromate (NH 4 ) 2 CrO 4 152 1 2 18213 Ammonium Chrome Alum Ammonium Citrate (NH 4 ) 2 Cr 2 (S0 4 ) 4 - 24H/3 (NH 4 ) 3 C 6 H 6 7 956.7 243.19 2.98078 2.38594 Ammonium Citrate Ammonium Copper Chloride (S Ammonium Ferric Alum (See Fe Ammonium Ferrous Sulfate (See Ammonium Fluoride 24H 2 O (NH 4 ) 3 C 6 H 6 7 ee Cupric Ammoniui rric Ammonium Alum Ferrous Ammonium NH 4 F 956.7 243.19 n Chloride ) Sulfate) 37.0 35.048. 144.96 2.98078 2.38594 ) 1.56820 1.54466 2.16125 Ammonium Hydroxide Ammonium Iodide NH 4 OH NTU Ammonium Iron Alum (See Ferric Ammonium Alun Ammonium Magnesium Chloride, etc. (See Magnesium Ammonium Chloride, etc.) Ammonium Molybdate Ammonium Nickel Chloride (See Ammonium Nitrate (NH 4 ) 2 MoO 4 196.1 Chloride) 80.048 299.39 124 . 090 142.106 2.29248 1.90335 2.47624 2 . 09374 2.15261 Nickel Ammonium C NH 4 NO 3 Ammonium Oleate NH 4 Ci 8 H 33 O 2 Ammonium Oxalate, Anhyd .... Ammonium Oxalate, Cryst (NH 4 ) 2 C 2 4 (NH 4 ) 2 C 2 4 -H 2 0... TABLES TABLE II MOLECULAR WEIGHTS 589 Name Formula Molecular Weight Logarithm (Ammonium Oxide Radical) .... Ammonium Persulfate (NH) 2 O 52.080 228.20 132.13 115.10 1860.1 1877.2 137.09 209.14 132.14 51.11 68.14 76.11 Dsphate) 130.147 88.121 93.094 242.6 171.7 297.5 400.7 320.4 304.4 226.6 288.4 336.6 imonyl Tar 150.105 138.96 261.92 214.04 310.22 229.92 122.96 181.34 77.98 197.92 1.71667 2.35832 2.12100 2.06108 3.26953 3.27351 2.13701 2.32043 2.12103 1.70851 1.83340 1.88144 2.11444 1.94508 1.96892 2.38489 2.23477 2.47349 2.60282 2.50569 2.48344 2.35526 2.46000 2.52711 trate) 2.17639 2.14289 2.41817 2.33049 2.49167 2.36158 2.08977 2.25850 1.89198 2.29649 (NH 4 ) 2 S 2 O8 Ammonium Phosphate, Di- Ammonium Phosphate, Mono-. . Ammonium Phosphomolybdate, Di- (NH 4 ) 2 HP0 4 NH 4 H 2 P0 4 (NH 4 ) 2 HPO 4 - 12MoO 3 . . . Ammonium Phosphomolybdate, Tri- (NH 4 ) 3 PO 4 .12MoO 3 NH 4 NaHPO 4 Ammonium Sodium Hydrogen Phosphate, Anhyd Ammonium Sodium Hydrogen Phosphate, Cryst NH 4 NaHPO 4 -4H 2 O (NH 4 ) 2 SO 4 . Ammonium Sulfate Ammonium Sulf hydrate Ammonium Sulfide NH 4 SH (NH 4 ) 2 S. . . Ammonium Sulfocyanate Ammonium Zinc Phosphate (See Amyl Acetate. . . . NH 4 CNS Zinc Ammonium Ph< Amyl Alcohol C 6 HnOH Aniline Antimonic Oxychloride C 6 H 6 .NH 2 SbOCl 3 .. Antimonous Oxychloride Antimony Pentachloride . SbOCl . SbClg Antimony Pentasulfide Sb 2 S 6 Antimony Pentoxide Sb 2 O 6 . . Antimony Tetroxide Antimony Trichloride Antimony Trioxide Sb 2 4 SbCl 3 Sb 2 O 3 . . Antimony Trisulfide Sb 2 S 3 Antimonyl Potassium Tartrate Arabinose (See Potassium Ant (Arsenate Radical) AsO 4 (Arsenate Radical, Pyro-) Arsenic Disulfide As 2 O 7 As 2 S 2 Arsenic Pentasulfide As 2 S 5 Arsenic Pentoxide As 2 O 5 .... (Arsenite Radical) Arsenous Chloride AsO 3 AsCl 3 Arsenous Hydride . Arsenous Oxide AsH 3 As 2 O 3 .... 590 TECHNICAL METHODS OF ANALYSIS TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm Arsenous Sulfide Arsine (See Arsenous Hydride) Auric Chloride, Anhyd As 2 S 3 AuCl 3 246.10 303.6 339.6 197.38 208.29 244.32 253.4 279.7 171.39 315.51 261.39 153 . 37 169.37 313.50 602 . 19 233.43 169.43 78.078 77.070 155.6 61.013 216.0 394.0 484.1 259.5 303.0 304.0 512.2 464.0 69.8 154.194 196 . 220 158.152 183.32 219.35 146.42 366.24 2.39111 2.48230 2.53097 2 . 29531 2.31867 2.38796 2.40381 2.44669 2.23399 2.49901 2.41729 2 . 18574 2.22884 2.49624 2.77937 2.36816 2.22899 1.89253 1.88689 2.19201 1.78542 2.33445 2.59550 2 . 68494 2.41414 2.48144 2.48287 2.70944 2.66652 1.84386 2.18806 2 . 29274 2 . 19907 2.26321 2.34114 2.16560 2.56377 Auric Chloride, Cryst Barium Carbonate AuCl 3 -2H 2 O. BaCO 3 Barium Chloride, Anhyd Barium Chloride, Cryst BaCl 2 BaCl 2 -2H 2 O BaCrO 4 BaSiF 6 Ba(OH) 2 Ba(OH) 2 -8H 2 O Ba(NO 3 ) 2 BaO BaO 2 . Barium Chromate Barium Fluosilicate . ... Barium Hydroxide Anhyd Barium Hydroxide, Cryst Barium Nitrate Barium Oxide Barium Peroxide, Anhyd Barium Peroxide, Cryst Barium Phosphate, Tri- Barium Sulfate BaO 2 -8H 2 O Ba 3 (PO 4 ) 2 BaSO 4 Barium Sulfide Benzene BaS C 6 H 6 (Benzyl Radical) (Biborate Radical) . . . B 4 O 7 (Bicarbonate Radical) HC0 3 Cr 2 O 7 . . (Bichromate Radical) ... . Bismuth Nitrate, Anhyd Bismuth Nitrate, Cryst Bismuth Oxy chloride Bismuth Phosphate Bi(NO 3 )s Bi(NO 3 ) 3 -5H 2 O BiOCl BiPO 4 Bismuth Sub-Nitrate Bismuth Sulfide . . BiONO 3 -H 2 O Bi 2 S 3 Bismuth Trioxide Bleaching Powder (See Calcium ( Bone Phosphate (See Calcium PI Borax (See Sodium Tetraborate, Boric Oxide ... . . . . Bi 2 O 3 3xychloride) losphate, Tri-) Cryst.) B-O 3 ... . Borneol C 10 H 17 OH CioHnC 2 H 3 O 2 (C 3 H 7 CO) 2 O Bornyl Acetate Butyric Anhydride Cadmium Chloride, Anhyd ..... Cadmium Chloride, Cryst Cadmium Hydroxide CdCl 2 CdCl 2 -2H 2 O Cd(OH) 2 CdI 2 Cadmium Iodide TABLES 591 TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm Cadmium Nitrate, Anhyd Cd(NO 3 ) 2 236 42 2 37369 Cadmium Nitrate, Cryst Cd(NO 3 ) 2 -4H 2 O. . 308.48 2.48923 Cadmium Oxide CdO 128.40 2 10857 Cadmium Sulfate, Anhyd CdSO 4 208 46 2 31903 Cadmium Sulfate, Cryst Cadmium Sulfate, Cryst Cadmium Sulfide. ... CdSO 4 -fH 2 O....... CdSO 4 -4H 2 O CdS 256.50 280.52 144 46 2.40909 2.44796 2 15975 Caesium Carbonate Cs 2 CO 3 325 63 2 51272 Caesium Chloride Caesium Hydroxide . . CsCl CsOH 168.27 149 . 82 2.22601 2 17557 Caesium Nitrate CsNO 3 194 82 2 28963 Caesium Sulfate Cs 2 SO 4 361 . 68 2 . 55833 Caffeine C 8 Hi N 4 O 2 194 152 2 28814 Calcium Acetate, Anhyd Ca(C 2 H 3 O 2 ) 2 158 14 2 19904 Calcium Acetate, Cryst Ca(C 2 H 3 O 2 ) 2 -H 2 O.. . 176.16 2.24591 Calcium Bicarbonate Calcium Bisulfite (Meta-) CaH 2 (CO 3 ) 2 CaS 2 O 5 162.10 184 19 2.20978 2 26527 Calcium Carbide CaC 2 64 08 1 80672 Calcium Carbonate CaCO 3 100.08 2.00034 Calcium Chloride Anhyd CaCl 2 . 110 99 2 04528 Calcium Chloride, Cryst. (Hexahydrate) CaCl 2 -6H 2 O 219.09 2.34062 Calcium Chloride, Cryst. ( Mono hydrate) CaCl 2 -H 2 O 129 01 2 11062 Calcium. Fluoride CaF 2 78 1 1 89265 Calcium Hydroxide Ca(OH) 2 74 09 1 86976 Calcium Hypochlorite, Anhyd . . . Calcium Hypochlorite, Hyd Calcium Nitrate, Anhyd Calcium Nitrate, Cryst Ca(ClO) 2 Ca(ClO) 2 -4H 2 O. . .. Ca(N0 3 ) 2 Ca(NO 3 ) 2 -4H 2 O 142.99 215.05 164.09 236 15 2.15531 2.33254 2.21508 2 37319 Calcium Oleate Ca(Ci 8 H33O 2 )2 602 . 78 2 . 78016 Calcium Oxalate Anhyd CaC 2 O 4 128 08 2 10748 Calcium Oxalate, Cryst . CaC 2 O 4 -H 2 O 146.10 2.16465 Calcium Oxide CaO 56.07 1 . 74873 Calcium Oxychloride CaOCl 2 126 99 2 10377 Calcium Peroxide CaO 2 72.07 1 . 85775 Calcium Phosphate, Di-, Anhyd CaHPO 4 136.12 2.13392 Calcium Phosphate Di- Cryst CaHPO 4 -2H 2 O 172.15 2 . 23591 Calcium Phosphate, Mono-, Anhyd CaH 4 (PO 4 ) 2 234.18 2.36955 592 TECHNICAL METHODS OF ANALYSIS TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm Calcium Phosphate, Mono-, Cryst . . CaH 4 (PO 4 ) 2 -H 2 O. .. Ca 3 (P0 4 ) 2 CaSiO 3 252.20 310.29 116.4 606.81 136.13 172.16 72.13 120.13 156.16 76.13 44.005 28.005 153.85 60.005 388.28 172.25 326.27 434.37 328.50 568.68 712.81 83.46 119.39 408.0 116.0 158.4 266.5 103.0 238.0 373.1 400.2 152.0 392.2 482.3 2.40175 2.49177 2.06595 2.78306 2.13396 2.23593 1.85812 2.07965 2.19357 1 . 88156 1.64350 1.44724 2.18710 1.77819 2.58915 2.23616 2.51358 2.63786 2.51654 2.75487 . 2.85298 1.92148 2.07696 2.61066 2.06446 2.19976 2.42570 2.01284 2.37658 2.57183 2.60228 2.18184 2.59351 2.68332 Calcium Phosphate, Tri- Calcium Silicate Calcium Stearate . . . Ca(C 18 H 36 2 ) 2 CaSO 4 Calcium Sulfate, Anhyd Calcium Sulfate, Cryst Calcium Sulfide CaS0 4 -2H 2 0.. CaS Calcium Sulfite, Anhyd CaSO 3 Calcium Sulfite, Cryst Carbon Bisulfide CaSO 3 -2H 2 O CS 2 Carbon Dioxide CO 2 Carbon Monoxide CO Carbon Tetrachloride (Carbonate Radical) CC1 4 CO, Carborundum (See Silicon Carbide) Caustic Potash (See Potassium Hydroxide) Caustic Soda (See Sodium Hydroxide) Ceric Nitrate .... CWNO,^ Ceric Oxide Cerous Nitrate, Anhyd. . CeO 2 Ce(NO 3 ) 3 Cerous Nitrate, Cryst Ce(NO 3 ) 3 -6H 2 O. ... Ce 2 O 3 Cerous Oxide Cerous Sulfate, Anhyd Ce 2 (S0 4 ) 3 Ce2(SO 4 ) 3 -8H 2 O. ... C1O 3 ... Cerous Sulfate, Cryst (Chlorate Radical) Chloroform CHC1 3 (Chlorplatinate Radical) (Chromate Radical) . . . PtCl 6 CrO... Chrome Alum (See Potassium Chrome Alum) Chrome Orange (See Lead Chromate, Basic) Chrome Yellow (See Lead Chromate) Chromic Chloride, Anhyd CrCL Chromic Chloride, Crvst . . . CrCl 3 -6H 2 O Chromic Hydroxide Cr(OH) 3 Cr(NO 3 ) 3 Chromic Nitrate, Anhyd Chromic Nitrate, Cryst Cr(N0 3 ) 3 -7H 2 0... Cr(NO 3 ) 3 -9H 2 O. ... Cr 2 O 3 Chromic Nitrate, Cryst Chromic Oxide Chromic Sulfate, Anhyd Chromic Sulfate, Cryst Cr 2 (SO 4 ) 3 Cr 2 (SO 4 ) 3 -5H 2 O.... TABLES TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm Chromium Trioxide CrO 3 Cr(C 2 H 3 2 ) 2 Cr(C 2 H 3 O 2 ) 2 -H 2 O... CrCl 2 100.0 170.1 188.1 122.9 68.0 148.1 274.2 152.178 165.35 109.99 165.94 240.91 129.89 237.99 182.99 291.08 74.97 155.03 281.14 468.63 540.69 oride) ate) 146.093 1014.11 277.52 187.54 123.58 221.17 344.75 134.49 170.52 97.59 187.59 241.63 295.68 79.57 2.00000 2.23070 2.27439 2.08955 1.83251 2.17056 2.43807 2.18235 2.21841 2.04135 2.21995 2.38186 2.11358 2.37656 2.26243 2.46401 1.87489 2.19041 2.44892 2.67083 2.73295 2.16463 3.00609 2.44329 2.27309 2.09195 2.34473 2.53751 2.12869 2.23177 1.98941 2.27321 2.38315 2.47082 1.90075 Chromous Acetate, Anhyd Chromous Acetate, Cryst Chromous Chloride Chromous Oxide . CrO Chromous Sulfate, Anhyd Chromous Sulfate, Cryst Citral CrSO 4 CrSO 4 -7H 2 O C 9 H 15 -CHO CoCl 3 Cobaltic Chloride Cobaltic Hydroxide Co (OH) 3 Cobaltic Oxide Co 2 O 3 Cobalto-cobaltic Oxide Cobaltous Chloride, Anhyd Cobaltous Chloride, Cryst Cobaltous Nitrate, Anhyd Co 3 O 4 . . . CoCl 2 CoCl 2 -6H 2 O Co(NO 3 ) 2 Cobaltous Nitrate, Cryst Co(NO 3 ) 2 -6H 2 O. ... CoO Cobaltous Oxide Cobaltous Sulfate, Anhyd CoSO 4 Cobaltous Sulfate Cryst CoSO 4 -7HoO Copper (See also Cupric and CL Copper Arsenate, Anhyd iprous) Cu 3 (AsO 4 ) 2 Copper Arsenate, Cryst Cu 3 (AsO 4 ) 2 -4H 2 O... Potassium Cupric Ch otassium Cupric Sull ric Chloride) C 9 H 6 O 2 . . Copper Potassium Chloride (See '. Copper Potassium Sulfate (See P Corrosive Sublimate (See Mercuj Coumarin Cupric Aceto-arsenite Cu 3 (AsO 3 ) 2 -2As2O 3 - Cu(C 2 H 3 O 2 ) 2 Cupric Ammonium Chloride .... Cupric Arsenite CUC12-2NH4C1- 2H 2 O CuHAsO 3 CuCO 3 Cupric Carbonate Cupric Carbonate, Basic Cupric Carbonate, Basic Cupric Chloride Anhyd CuCO 3 -Cu(OH) 2 ... 2CuCO 3 -Cu(OH) 2 . . CuCl 2 Cupric Chloride, Cryst CuCl 2 -2H 2 O Cupric Hydroxide Cu(OH) 2 Cupric Nitrate Anhyd . . . Cu(NO 3 ) 2 Cupric Nitrate, Cryst Cupric Nitrate Cryst Cu(N0 3 ) 2 .3H 2 0.... Cu(NO 3 ) 2 -6H 2 O.... CuO Cupric Oxide 594 TECHNICAL METHODS OF ANALYSIS TABLE II MOLECULAK WEIGHTS Name Formula Molecular Weight Logarithm Cupric Sulfate, Anhyd . . CuSO 4 159.63 249.71 95.63 80.58 143.14 159.20 121.64 26.013 52.026 198.142 116.100 383.4 29.050 88.084 46.058 964.38 162.22 270.32 859.11 106.86 241.86 404.00 159.68 177.70 150.88 n) 399.86 562.00 207 . 86 231.52 295.76 173.91 245.97 392.14 115.85 126.76 198.82 591.36 2.20311 2.39744 1.98059 1.90623 2.15576 2.20194 2.08507 1.41519 1.71622 2.29697 2.06483 2.58365 1.46315 1.94490 1.66330 2.98425 2.21010 2.43188 2.93405 2.02882 2.38357 2.60638 2.20325 2.24969 2.17863 2.60191 2.74974 2.31777 2.36459 2.47094 2.24033 2.39089 2.59344 2.06390 2.10298 2.29846 2.77185 Cupric Sulfate, Cryst CuSO 4 -5H>O Cupric Sulfide CuS Cuprous Hydroxide CuOH . . . Cuprous Oxide Cu 2 O Cuprous Sulfide Cu 2 S Cuprous Sulfocyanate CuCNS (Cyanide Radical) CN Cyano gen C 2 N 2 . . . Dextrose, Cryst Dimethylglyoxime . L/6Hi 2 Oe H 2 O (CH 3 ) 2 C 2 (NOH) 2 ... Er 2 O 3 Erbium Oxide (Ethyl Radical) Ethyl Acetate Ethyl Alcohol QjHsOH Ferric (Ammonium) Alum Ferric Chloride, Anhyd. . . Fe 2 (NH 4 ) 2 (S0 4 ) 4 - 24H 2 O FeCl 3 Ferric Chloride Cryst FeCl 3 -6H 2 O Ferric Ferrocyanide Fe 4 [Fe(CN) 6 l 3 Ferric Hydroxide Fe(OH) 3 Fe(NO 3 ) 3 Ferric Nitrate, Anhyd Ferric Nitrate, Cryst Fe(N0 3 ) 3 -9H 2 0.... FezOs Ferric Oxide Ferric Oxide, Hydrated Ferric Phosphate, Anhyd FeaOs-HzO FePO 4 Ferric Potassium Alum (See Poi Ferric Sulfate, Anhyd bassium Ferric Aim Fe 2 (SO 4 ) 3 Ferric Sulfate Cryst Fe2(SO 4 ) 3 -9H 2 O. ... FejzSa Ferric Sulfide * Ferroso-f erric Oxide . . . Fe 3 O 4 Ferroso-f erric Sulfide Fe 3 S 4 Ferrous Acetate, Anhyd Fe(C 2 H 3 2 ) 2 Fe(C 2 H 3 2 ) 2 -4H 2 0.. Fe(NH 4 ) 2 (S0 4 ) 2 - 6H 2 O Ferrous Acetate Cryst Ferrous Ammonium Sulfate Ferrous Carbonate FeCO 3 Ferrous Chloride Anhyd FeCl 2 Ferrous Chloride, Cryst Ferrous Ferricyanide FeCl 2 -4H 2 O Fe<[Fe(CN) 6 ] 2 * See also Iron Sul-fide TABLES TABLE II MOLECULAR WEIGHTS 595 Name Formula Molecular Weight Logarithm Ferrous Hydroxide. . . . Fe(OH) 2 89.86 71.84 151.90 278.01 87.90 142.3 30.021 96.057 180.126 92.079 34.016 34.08 17.008 63.04 174.92 158.92 162.38 333.84 233.30 119.96 647.36 360.252 326.0 685.60 325.27 608.51 1422.20 379.32 347.17 899.52 267.21 775.64 278.12 1.95357 1.85637 2.18156 2.44406 1.94399 2 . 15320 1.47743 1.98253 2.25557 1.96417 1.53168 1.53250 1.23065 1.79962 2.24284 2.20118 2.21054 2.52354 2.36791 2.07904 2.81115 2.55660 2 . 51322 2.83607 2.51224 2.78427 3.15296 2.57900 2.54054 2.95401 2.42686 2.88966 2.44423 Ferrous Oxide FeO: Ferrous Sulf ate, Anhyd Ferrous Sulfate, Cryst FeSO 4 FeSO 4 -7H 2 O FeS Ferrous Sulfide (Fluosilicate Radical) SiF 6 Formaldehyde HCHO Furfural C 4 H 3 O-CHO.. Fusel Oil (See Amyl Alcohol) Galactose CH,oO. Glauber Salts (See Sodium Sulfate, Cryst.) Glycerol |C 3 H 5 (OH) 3 . . Gold Chloride (See Auric Chloride) Gypsum (See Calcium Sulfate, Cryst.) Hydrogen Peroxide HoOo Hydrogen Sulfide H 2 S... (Hydroxyl Radical) HO (Hypophosphate Radical) (lodate Radical) PO 2 IO 3 Iodine Dioxide IO 2 Iodine Monochloride IC1 Iodine Pentoxide I 2 O 5 Iodine Trichloride IC1 3 Iron Alum (See Ferric Ammoniu Iron Di-sulfide tn Alum) * FeS2 Iron Sulfide Fe 7 S 8 roso-ferric Oxide) Ci 2 H22Oii-H 2 O Iron Oxide, Magnetic (See Fen Lactose, Cryst Lanthanum Oxide LaizOs Lead, Red Pb 3 O 4 Lead, White (See Lead Carbon. Lead Acetate, Anhyd ite, Hydrated) Pb(C 2 H 3 O 2 ) 2 Lead Acetate, Basic Lead Acetate, Basic t Pb 2 (C 2 H 3 2 ) 3 -OH... 3Pb(C 2 H 3 O 2 ) 2 -2PbO Pb(C 2 H 3 2 ) 2 -3H 2 0.. PbHAsO 4 .. Lead Acetate, Cryst Lead Acid Arsenate. . . . Lead Arsenate . . Pb 3 (AsO 4 ) 3 Lead Carbonate PbCO 3 ... . Lead Carbonate, Hydrated 2PbCO 3 -Pb(OH) 2 ... PbCL Lead Chloride *For other compounds of iron see under Ferric and Ferrous. t Home's Reagent. 596 TECHNICAL METHODS OF ANALYSIS TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm Lead Chromate PbCrO 4 323 2 2 50947 Lead Chromate, Basic PbCrO 4 -PbO . . 546 4 2 73751 Lead Cyanide Pb(CN) 2 259 23 2 41368 Lead Hydroxide Pb(OH) 2 241 22 2 32842 Lead Iodide PbI 2 461 04 2 66374 Lead Molybdate PbMoO 4 367 2 2 56490 Lead Monoxide PbO. . 223 20 2 34869 Lead Nitrate Pb(NO 3 ) 2 331 22 2 52012 Lead Oleate Pb(Ci 8 H 33 O 2 )2 769 21 2 88644 Lead Peroxide PbO 2 239 20 2 37876 Lead Stearate Pb(Ci 8 H, fi O 2 ) 2 773 94 2 88870 LeadSulfate PbS0 4 303 26 2 48181 Lead Sulfate Basic PbSO 4 PbO 526 46 2 72137 LeadSulfide PbS 239 26 2 37887 Lead Tungstate pbWO 4 455 2 2 65820 Levulose C 6 Hi 2 O 6 180 126 2 25557 Lime (See Calcium Oxide) Litharge (See Lead Monoxide) Lithium Acetate, Anhyd .... LiC 2 H 3 O 2 65 97 1 81935 Lithium Acetate, Cryst LiC 2 H 3 O 2 -2H 2 O . . 102 01 2 00864 Lithium Bromide LiBr 86 86 1 93882 Lithium Carbonate LioCO 3 73 89 1 86859 Lithium Chloride LiCl 42 40 1 62737 Lithium Hydroxide LiOH 23 95 1 37931 Lithium Nitrate, Anhyd LiNO 3 68 95 1 83853 Lithium Nitrate, Cryst Lithium Oxide LiNO 3 -3H 2 O Li 2 O . . 123.00 29 88 2.08991 1 46538 Lithium Phosphate, Anhyd Lithium Phosphate, Cryst Lithium Sulfate, Anhyd Li 3 P0 4 Li 3 P0 4 -H 2 Li 2 SO 4 115.86 133.88 109 94 2.06393 2.12672 2 04116 Lithium Sulfate, Cryst Li 2 SO 4 -H 2 O 127 96 2 10707 Magnesia (See Magnesium Oxide) Magnesium Acetate, Anhyd .... Magnesium Acetate, Cryst Magnesium Acid Sulfite Mg(C 2 H 3 2 ) 2 Mg(C 2 H 3 2 ) 2 -4H 2 0. MgH 2 (SO 3 ) 2 142.39 214.45 186 46 2 . 15348 2.33133 2 26059 Magnesium Ammonium Chloride MgCl 2 -NH 4 Cl- 6H 2 O 256 84 2 40966 Magnesium Ammonium Phos- phate, Anhyd MgNH 4 PO 4 137 40 2 13799 . Magnesium Ammonium Phos- phate, Cryst MgNH 4 PO 4 -6H 2 O. . 245 . 50 2.39005 TABLES TABLE II MOLECULAR WEIGHTS 597 Name Formula Molecular Weight Logarithm Magnesium Ammonium Sulfate. Mg(NH 4 ) 2 (S0 4 ) 2 - 6H 2 O 360 62 2 55705 Magnesium Bicarbonate MgH 2 (CO 3 ) 2 146 35 2 16539 Magnesium Bisulfite (Meta-). . . . MgS 2 O 6 168 44 2 22644 Magnesium Carbonate MgCO 3 84 33 1 92598 Magnesium Carbonate, Basic . . . Mg(OH) 2 .4MgC0 3 - 5H 2 O 485 74 2 68641 Magnesium Chloride, Anhyd. . . . MgCl 2 95 24 1 97882 Magnesium Chloride, Cryst Magnesium Hydroxide Magnesium Nitrate, Anhyd MgCl 2 -6H 2 O Mg(OH) 2 Mg(NO 3 ) 2 203.34 58.34 148 34 2.30822 1.76597 2 17126 Magnesium Nitrate, Cryst Magnesium Oxide Mg(NO 3 ) 2 -6H 2 O.... MgO 256.43 40 32 2.40897 1 60552 Magnesium Phosphate, Anhyd. . Mg 3 (PO 4 ) 2 263 04 2 42002 Magnesium Phosphate, Cryst. . . Magnesium Pyroarsenate Magnesium Pyrophosphate Mg 3 (PO 4 ) 2 -4H 2 O.. . Mg 2 As 2 O 7 Mg 2 P 2 O 7 335.10 310.56 222 72 2.52517 2.49214 2 34776 Magnesium Silicate (Meta-) .... MgSiO 3 100 6 2 00260 Magnesium Silicate (Ortho-) MgSiO 4 116 6 2 06670 Magnesium Sulfate, Anhyd Magnesium Sulfate, Cryst MgS0 4 MgSO 4 -7H 2 O 120.38 246 49 2.08056 2 39180 Magnesium Sulfite, Anhyd MgSO 3 104 38 2 01862 Magnesium Sulfite Cryst MgSO 3 -6H 2 O 212 48 2 32732 Maltose, Cryst C 12 H 22 On-H 2 O 360 252 2 55660 Manganese Carbonate MnCO 3 114 94 2 06047 Manganese Chloride, Anhyd. . . , MnCl 2 125 85 2 09986 Manganese Chloride, Cryst MnCl 2 -4H 2 O 197 91 2 29647 Manganese Dioxide MnO 2 ... 86 93 1 93917 Manganese Heptoxide Mn 2 O 7 221 86 2 34608 IVlanganese Hydroxide Mn(OH) 2 88 95 1 94915 Manganese Nitrate, Anhyd Mn(NO 3 ) 2 . 178 95 2 25273 Manganese Nitrate, Cryst . Mn(NO 3 ) 2 -6H 2 O 287 04 2 45794 Manganese Pyrophosphate Mn 2 P 2 O 7 283 94 2 45323 Manganese Silicate MnSiO 3 131 2 2 11793 Manganese Sulfate, Anhyd MnSO 4 150 99 2 17895 Manganese Sulfate, Cryst MnSO 4 -4H 2 O 223.05 2 34840 Manganese Sulfide MnS 86 99 1 93947 Manganese Trioxide MnO 3 . . . 102 93 2 01255 Magnanic Oxide .... Mn 2 O 3 157 86 2 19827 Mangano-manganic Oxide Mn 3 O 4 228 79 2 35944 Manganous Oxide MnO. . 70 93 1 85083 598 TECHNICAL METHODS OF ANALYSIS TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm Menthol Ci Hi 9 OH 156 210 2 19371 Menthyl Acetate .... 198.236 2.29718 Mercuric Bromide HgBr 2 360 4 2.55678 IVlercuric Chloride HgCl 2 271 5 2 43377 Mercuric Cyanide Hg(CN) 2 252.6 2.40243 Mercuric Hydroxide Mercuric Iodide Hg(OH), HgI 2 234.6 454.4 2.37033 2.65744 Mercuric Nitrate, Anhyd Mercuric Nitrate, Cryst Mercuric Oxide Hg(N0 3 ) 2 Hg(N0 3 ) 2 -H 2 HgO 324,6 342.6 216.6 2.51135 2.53479 2.33566 Mercuric Sulfate HgSO 4 296.7 2.46232 Mercuric Sulfide HgS 232.7 2.36680 Mercurous Chloride Hg 2 Cl 2 472.1 2.67403 Mercury Fulminate HgC 2 N 2 O 2 284.6 2.45423 Methane CH 4 16.037 1.20512 (Methyl Radical) CH 3 15.029 1 . 17693 Methyl Acetate CH 3 C 2 H 3 O 2 74.063 1.86960 Methyl Alcohol CH 3 OH 32.037 1 . 50565 Methvl Salicvlate.. . CH.C7H.O.,.. 152.104 2.18214 Microcosmic Salt (See Ammoni Cryst.) Minium (See Lead, Red) (Molybdate Radical) . . um Sodium Hydrog MoO 4 3n Phosph 160.0 128.0 240.0 160.1 144.0 336.6 291.20 118.69 129.60 237.70 110.71 182.77 288.86 182.69 290.79 74.68 154.74 90.74 109.70 ate, 2.20412 2 . 10721 2.38021 2.20439 2.15836 2.52711 2 . 46419 2.07441 2.11261 2.37603 2.04419 2.26191 2.46069 2.26172 2.46358 1.87320 2.18960 1.95780 2.04021 ]Vlolybdic Dioxide MoO 2 Molybdic Oxide ... Mo 2 O 3 Molybdic Sulfide MoS 2 Molybdic Trioxide MoO 3 Neodymium Oxide Nd 2 O 3 Nickel Ammonium Chloride .... Nickel Carbonate NiCl 2 -NH 4 Cl-6H 2 O NiCO 3 Nickel Chloride Anhyd NiCl 2 Nickel Chloride, Cryst . ... NiCl 2 -6H 2 O Ni(CN) 2 Ni(CN) 2 -4H 2 O NiC 8 H 14 N 4 O 4 ... Nickel Cyanide, Anhyd Nickel Cyanide, Cryst Nickel Glyoxime Nickel Nitrate Anhyd Ni(NO 3 ) 2 Nickel Nitrate, Cryst Nickel Oxide Ni(NO 3 ) 2 -6H 2 O. . .. NiO Nickel Sulfate NiSO 4 Nickel Sulfide NiS Nickelic Hydroxide Ni(OH) 3 TABLES TABLE II MOLECULAR WEIGHTS 599 Name Formula Molecular Weight Logarithm Nickelo-nickelic Oxide Nickelous Hydroxide Ni 3 O 4 240.04 92.70 162.178 62.008 30.008 108.016 46.008 76.016 44.016 88.010 48.000 122.7 99.46 118.93 162.110 95.04 174.08 79.04 208.34 142.08 137.42 110.08 337.0 427.1 266.1 mide) 98.13 136.17 120.17 258.4 279.1 332.4 100 11 294.2 78.1 2.38028 1.96708 2.20999 1.79245 1 . 47724 2.03348. 1.66283 1.88091 1.64361 1.94453 1.68124 2.08884 1.99765 2.07529 2.20981 1.97791 2.24075 1.89785 2.31877 2.15253 2.13805 2.04171 2.52763 2.63053 2.42504 1.99180 2.13408 2.07979 2.41229 2.44576 2.52166 2.00047 2.46864 1.89265 Ni(OH) 2 Nicotine Ci Hi 4 N 2 (Nitrate Radical) NO 3 Nitric Oxide NO Nitrogen Pentoxide N 2 O 6 Nitrogen Tetroxide NO 2 Nitrogen Trioxide N 2 O 3 Nitrous Oxide N 2 O Orpiment (See Arsenous Sulfide) (Oxalate Radical) C 2 O 4 ... . Ozone O 3 Palladium Monoxide Paris Green (See Cupric Aceto- (Perchlorate Radical) PdO irsenite) C1O 4 (Permanganate Radical) Phloroglucinol, Cryst. . . MnO 4 C 6 H 3 (OH) 3 .2H 2 O... P0 4 P 2 O 7 (Phosphate Radical) (Phosphate Radical, Pyro-) (Phosphite Radical) PO 3 ... Phosphoric Anhydride (See Pho, Phosphorus Pentachloride sphorus Pentoxide) PC1 6 Phosphorus Pentoxide Phosphorus Trichloride Phosphorus Trioxide P 2 5 PC1 3 P 2 O 3 Platinic Chloride, Anhyd Platinic Chloride, Cryst PtCU PtCl 4 -5HoO Platinous Chloride PtCl 2 Potash (See Potassium Oxide) Potash Alum (See Alum) Potassium Silver Cyanide (See Potassium Acetate Silver Potassium Cyj KC 2 H 3 O 2 Potassium Acid Sulf ate KHSO 4 Potassium Acid Sulfite Potassium Aluminum Fluoride... Potassium Aluminum Silicate . . . Potassium Aluminum Sulfate Potassium Antimonyl Tartrate . . . Potassium Bicarbonate KHSO 3 K 3 A1F 6 KAlSisOg (See Alum) KSbOC 4 H 4 O 6 ^H 2 O KHCO 3 Potassium Bichromate K 2 Cr 2 O 7 Potassium Bifluoride KHF 2 600 TECHNICAL METHODS OF ANALYSIS TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm Potassium Binoxalate KHC 2 O 4 128.12 2.10762 Potassium Bitartrate KHC 4 H 4 O 6 188 16 2 27453 Potassium Bromide . . KBr 119 02 2 07562 Potassium Carbonate K 2 CO 3 138 21 2 . 14054 Potassium Chlorate KC1O 3 122 56 2 08835 Potassium Chloride KC1 74 56 1 87251 Potassium Chlorplatinate .... K 2 PtCl 6 .. . 486 2 2 68681 Potassium Chromate K 2 CrO 4 194 2 2 28825 Potassium Chrome Alum Potassium Cobaltinitrite, Anhyd. Potassium Cobaltinitrite, Cryst.. Potassium Cupric Chloride Potassium Cupric Sulf ate K 2 Cr 2 (SO 4 ) 4 -24H 2 O. K 3 Co(NO 2 ) 6 K 3 Co(NO 2 ) 6 -HH 2 O. K 2 CuCl 4 -2H 2 O K 2 Cu(SO 4 ) 2 -6H 2 O... 998.8 452.32 479.34 319.64 441 . 99 2.99948 2.65545 2.68065 2.50466 2.64541 Potassium Cyanide KCN 65 11 1 81365 Potassium Ferric Alum K 2 Fe 2 (SO 4 ) 4 -24H 2 O 1006.50 3.00282 Potassium Ferricyanide Potassium Ferrocyanide, Anhyd . Potassium Ferrocyanide, Cryst . . Potassium Fluoride K 3 Fe(CN) 6 K 4 Fe(CN) 6 K 4 Fe(CN) 6 -3H 2 0... KF 329.22 368.32 422.37 58.1 2.51749 2.56622 2.62569 1.76418 Potassium Fluosilicate ..... K 2 SiF 6 220.5 2 34341 Potassium Hydrosulfide KHS 72 17 1 85836 Potassium Hydroxide KOH 56.11 1.74904 Potassium lodate KIO 3 214 02 2 33045 Potassium Iodide. . . . KI 166.02 2 . 22016 Potassium Nitrate KNO 3 101.11 2 00479 Potassium Nitrite. KNO 2 85.11 1 . 92998 Potassium Oleate .... 320 45 2 . 50576 Potassium Oxalate, Anhyd K 2 C 2 O 4 166.21 2 . 22066 Potassium Oxalate, Cryst Potassium Oxide K 2 C 2 O 4 -H 2 O K 2 O 184.23 94 20 2.26536 1 97405 Potassium Perchlorate KC1O 4 138.56 2 . 14164 Potassium Permanganate KMnO 4 158.03 2 . 19874 Potassium Persulf ate K 2 S 2 O 8 270 . 32 2.43188 Potassium Phosphate, Di- Potassium Phosphate Mono- K 2 HPO 4 KH 2 PO 4 174.25 136 16 2.24118 2 . 13405 Potassium Phosphate (Ortho-) . . . K 3 PO 4 212.34 2 . 32703 Potassium Silicate K 2 SiO 3 154.5 2 . 18893 Potassium Silver Cyanide KAg(CN) 2 199 01 2.29887 Potassium Sodium Carbonate, Anhyd KNaCO 3 122.11 2.08676 TABLES TABLE II MOLECULAR WEIGHTS 601 Name Formula Molecular Weight Logarithm Potassium Sodium Carbonate, Cryst KNaCO 3 -6H 2 O KNaC 4 H 4 O 6 -4H 2 O.. KCisHssOz K 2 SO 4 230.20 282.22 322.47 174.26 110.26 200.34 158.26 194.29 97.17 226.25 235.26 335.4 160.27 329.8 261.92 174.08 385.8 296.9 594.426 209.3 186.90 266.96 251.4 183.14 348.8 136.2 143.2 127.2 111.2 2.36211 2.45059 2.50849 2.24120 2.04242 2.30177 2.19937 2.28845 1.98753 2.35459 2.37155 2.52556 2.20485 2.51825 2.41817 2.24075 2.58636 2.47261 2.77410 2.32077 2.27161 2.42645 2.40037 2.26279 2.54258 2.13418 2.15594 2.10449 2.04610 Potassium Sodium Tartrate Potassium Stearate Potassium Sulfate Potassium Sulfide, Anhyd K 2 S Potassium Sulfide, Cryst Potassium Sulfite, Anhyd Potassium Sulfite Cryst K 2 S-5H 2 O K 2 SO 3 K 2 SO>-2H 2 O Potassium Sulfocyanate KCNS Potassium Tartrate, Anhyd Potassium Tartrate, Cryst Potassium Tetrasilicate .... K 2 C 4 H 4 6 K 2 C 4 H 4 O 6 ^H 2 O.... K 2 Si 4 O 9 . . . Potassium Xanthogenate KS 2 COC 2 H 5 Praseodymium Oxide Pr,O, . Prussian Blue (See Ferric Ferrocyanide) Prussic Acid (See Acid, Hydrocyanic) Pyrites (See Iron Bisulfide) Pyrites, Magnetic (See Iron Sulfide) (Pyroarsenate Radical) AsoO^ (Pyrophosphate Radical) Radium Bromide P 2 7 RaBr 2 Radium Chloride RaCl 2 Raffinose, Cryst C 18 H 32 Oi C -5H 2 O RhCl 3 Realgar (See Arsenic Disulfide) Red Lead (See Lead, Red) Rhodium Chloride Rochelle Salts (See Potassium Rubidium Oxide Sodium Tartrate) Rb 2 O Rubidium Sulfate Rb 2 SO 4 Ruthenium Oxide Ru 2 O 3 Saccharin C 7 H 5 SO 3 N Sal Soda (See Sodium Carbonate, Salt (See Sodium Chloride) Samarium Oxide Decahydrate) SaaOa Scandium Oxide ScoO 3 Schlippe's Salt (See Sodium Thi< (Selenate Radical) (Selenite Radical) Dantimonate) SeQ 4 SeO 3 Selenium Dioxide Silica (See Silicon Dioxide) Se0 2 602 TECHNICAL METHODS OF ANALYSIS TABLE II MOLECULAR WEIGHTS Name Formula Molecular Weight Logarithm (Silicate Radical Meta-) SiO 3 76.3 92.3 257.2 40.3 60.3 104.3 187.80 275.77 143.34 331.8 133.89 234.80 169.89 231.76 199.01 311.82 247.82 165.95 540.0 494.78 82.03 136.08 103.05 157.10 916.8 263.0 1.88252 1.96520 2.41027 1.60531 1.78032 2.01828 2.26370 2.44055 2.15637 2.52088 2.12675 2.37070 2 . 23017 2.36504 2.29887 2.49391 2.39414 2.21998 2.73239 2.69441 1.91397 2.13380 2.01305 2.19618 2.96227 2.41996 (Silicate Radical, Ortho-) SiO 4 (Silicate Radical, Tetra-) Silicon Carbide S^Og SiC Silicon Dioxide SiO 2 SiF 4 Silver Bromide AgBr Silver Carbonate Ag2CO 3 Silver Chloride AgCl . Silver Chromate Ag 2 CrO 4 Silver Cyanide AgCN Silver Iodide Ael Silver Nitrate AgNO 3 Silver Oxide Ag 2 O Silver Potassium Cyanide Silver Sulfate AgK(CN) 2 Ag 2 SO 4 Silver Sulfide Ag 2 S Silver Sulfocyanate AgSCN Silver Thioantimonite Ag 3 SbS 3 Silver Thioarsenite . . AgSi 4 O 9 0758 8 87967 * Phenolphthalein Indicator, t Methyl Orange Indicator. J By Saponification with Caustic Solution. 618 TECHNICAL METHODS OF ANALYSIS TABLE IV VOLUMETRIC SOLUTIONS (0.5 N HCl Continued) 1 cc. of 0.5 N HCl is equivalent to: Substance Gram Logarithm Nicotine, CioHi 4 N2. 08109 8 9089710 Protein (NX6.25) 0.04378 8.64128 Protein (N X5 . 70) 03992 8 60119 Prussian Blue, Fe 4 [Fe(CN) 6 ] 3 0.02386 8.37767 Tallow Oil 1439 9 15806 Wool Grease H . 275 9 43949 1 cc. of 1 .0 N NaOH is equivalent to: Abietic Acid,* HC 20 H 29 O 2 Acetic Anhydride,* (CH 3 CO) 2 O 0.3023 05103 9.48044-10 8 70783 A1 2 (SO 4 ) 3 . . 05707 8 75641 Amyl Acetate, J C 3 H u C 2 H 3 Oo 13015 9 11445 B 2 (X * 03490 8 54283 B 4 O 7 * 03890 8 58995 Benzoic Acid,* HC 7 H 5 O 2 12208 9 08665 Bornyl Acetate, J Ci H 17 C 2 H 3 O 2 19622 9 29274 Butyric Acid, * HC 4 H 7 O 2 . . . 08808 8 94488 C0 2 * 04400 8 64345 Ca(C 2 H 3 O 2 ) 2 07907 8 89801 Ca(Ci 8 H 33 O 2 ) 2 . 3014 9 47914 Ca(Ci 8 H 35 O 2 ) 2 3034 9 48202 Citric Acid, Cryst , HoC 6 H 5 O 7 H.O 07004 8 84535 Cl 03546 8 54974 Ethyl Acetate,| C 2 H 6 C 2 H,O 2 08808 8 94488 Formaldehyde, HCOH 0.03002 8.47741 Formic Acid,* HCO 2 H 04602 8 66295 Glycerol, || C 3 H 5 (OH) 3 03069 8 48700 H 3 BO- * 06190 8 79169 HBr 08093 8 90811 HC 2 H 3 O 2 * 06004 8 77844 H 2 C 2 O 4 * 04501 8 65331 H 2 C 2 O 4 -2H 2 O* 06303 8 79955 HCl 0.03647 8.56194 HI 0.12793 9.10697 * Phenolphthalein Indicator; t Methyl Orange Indicator. t By Saponification. Saponification Value 195. f Saponification Value 102. II By Saponification after Acetylization. TABLES 619 TABLE IV VOLUMETRIC SOLUTIONS (1.0 N NaOH Continued) 1 cc. of 1.0 N NaOH is equivalent to: Substance Gram Logarithm HNO 3 06302 8 79948 10 H 3 PO 4 * 04903 8 69046 H 3 PO 4 f 09806 8 99149 H 2 SO 3 * 04104 8 61321 H 2 SO 4 04904 8 69055 3205 9 50583 3225 9 50853 KHCO 3 10011 9 00047 KHCoO 4 * 12812 9 10762 Lactic Acid,* HC 3 H 6 O' . 09006 8 95453 Malic Acid * H 2 C 4 H 4 O 5 06703 8 82627 Menthol,!] C 10 H 19 OH 15621 9 19371 Menthyl Acetate, | CioHi 9 C 2 H 3 O 2 19824 9 29719 Methyl Acetate,! CH 3 C 2 H 3 O 2 0.07406 8 86958 N 014008 8 14638 NO 3 06201 8 79246 N 2 O 5 05401 8 73247 Na2B 4 O 7 * 05040 8 70243 Na 2 B 4 O 7 10H 2 O * 09545 8 97978 NaC 7 H 5 O 2 14408 9 15860 NaC 7 H 5 O 3 16008 9 20433 NaC] 8 H 3 *O 2 3044 9 48344 NaCi 8 H 35 O 2 3064 9 48629 NaHCO 3 08401 8 92433 NaOH 04001 8 60217 Oleic Acid,* HC 18 H 33 O 2 2824 9 45086 Pb(C 2 H 3 O 2 ) 9 -3H 2 O 18966 9 27798 Salicylic Acid,* HC 7 H 6 O 3 . 13808 9 14013 SO 2 * 03203 8 50556 SO 3 04003 8 60239 SO 4 04803 8 68151 Stearic Acid,* HCi 8 H 36 O 2 2844 9 45393 Tallow Oil 0.2877 9 45894 Tartaric Acid, Anhyd.,* H 2 C 4 H 4 O 6 07503 8 87523 Tartaric Acid, Cryst.,* H 2 C 4 H 4 O 6 H 2 O Wool Grease If 0.08404 5501 8.92449 9 74044 * Phenolphthalein Indicator, t Methyl Orange Indicator. J By saponification. Saponification Value 195 7 Saponification Value 102. II By Saponification after Acetylization. 620 TECHNICAL METHODS OF ANALYSIS TABLE IV VOLUMETRIC SOLUTIONS (Continued) 1 cc. of 0.5 N NaOH is equivalent to: Substance Gram Logarithm Abietic Acid,* HG>oH 29 O 2 1512 9 17955 10 Acetic Anhydride,* (CH 3 CO) 2 O 025517 8 40683 A1 2 (SO 4 ) 3 02853 8 45530 Amyl Acetate J C 5 HnC 2 H 3 O2 06507 8 81338 Benzoic Acid,* HC 7 H 5 O 2 06104 8 78561 B->O 3 * 01745 8 24180 B 4 O 7 * 01945 8 28892 Bornyl Acetate,} Ci Hi 7 C 2 H 3 O 2 Butyric Acid,* HC 4 H 7 O 2 0.09811 04404 8.99171 8 64385 CO 2 * 02200 8 34242 Ca(C 2 H 3 O 2 ) 2 3954 8 59704 ' Ca(Ci 8 H 33 O 2 ) 2 1507 9 17811 Ca(Ci8H 3 5O 2 ) 2 1517 9 18099 Citric Acid, Cryst , H 3 C 6 H 5 O 7 H 2 O 03502 8 54432 Cl 0.01773 8 24871 Ethyl Acetate,} C 2 H 5 C 2 H 3 O 2 04404 8 64385 Formaldehyde HCOH 01501 8 17638 Formic Acid,* HCO 2 H 02301 8 36192 Glycerol, C 3 H 5 (OH) 3 015347 8 18603 H 3 BO 3 * '03095 8 49066 HBr 04047 8 60713 HC 2 H 3 O 2 * 0.03002 8 47741 H 2 C 2 O 4 * . ... . 022507 8 35232 H 2 C 2 O 4 -2H 2 O * 03151 8 49845 HC1 018235 8 26091 HI 06397 8 80598 HNO 3 03151 8 49845 H 3 PO 4 * 024515 8 38943 H 3 PO 4 f 0.04903 8 69046 H 2 SO 3 * 02052 8 31218 H 2 SO 4 02452 8 38952 1602 9 20466 1612 9 20737 KHCO 3 05006 8 69949 KHC 2 O 4 * . . 06406 8 80659 Lactic Acid,* HC 3 H 5 O 3 Malic Acid,* H 2 C 4 H 4 O 5 0.04503 03352 8.65350 8 52530 * Phenolphthalein Indicator. t Methyl Orange Indicator. t By Saponification. By Saponification after Acetylization. TABLES 621 TABLE IV VOLUMETRIC SOLUTIONS (0.5 N NaOH Continued) 1 cc. of 0.5 N NaOH is equivalent to: Substance Gram Logarithm Menthol d Hi 9 OH 0.07811 8 89271-10 Menthyl Acetate { CioHi9C 2 H 3 O2 09912 8 99616 Methyl Acetate, } CH 3 C 2 H 3 O 2 0.03703 8 . 56855 N 0.007004 7 84535 NO 3 03100 8 49136 NoO 5 0.02700 8.43136 Na 2 B 4 O 7 * 0.02520 8 40140 Na 2 B 4 O 7 10H 2 O * 04773 8 67879 NaC 7 H 5 Oo 0.07204 8.85757 NaC 7 H 5 O 3 0.08004 8 90331 NaCi 8 H 33 O 2 1522 9 18241 NaCi 8 H 35 Oo 0.1532 9 . 18526 NaHCOs 0.04201 8 62335 NaOH 0.020005 8.30114 Oleic Acid,* HCi 8 H 33 O 2 0.1412 9 14983 Pb(G>H 3 O 2 )o-3H 2 O . . 0.09483 8 97695 Salicylic Acid * HC 7 H 5 O 3 06904 8 83910 SO 2 * 0.016015 8 . 20453 SO 3 020015 8 30136 SO 4 0.024015 8.38048 Stearic Acid,* HCi8H 35 O 2 Tallow Oil H 0.1422 1439 9.15290 9 15806 Tartaric Acid, Anhyd ,* H 2 C 4 H 4 O6 0.03752 8.57426 Tartaric Acid, Cryst.,* H 2 C 4 H 4 O 6 -H 2 O Wool Grease !| .0.04202 2751 8.62346 9 43949 1 c.c. of 0.1 N AgNO 3 is equivalent to: Ag 0.010788 8.03294 10 AgNO 3 0.016989 8 23017 BaCl 2 010415 8 01766 BaClo-2H 2 O 012216 8 08693 Br 0.007992 7 . 90266 CaCl 2 005550 7 74429 CdCl 2 009166 7 96218 CdI 2 '. 0.018312 8.26274 * Phenolphthalein Indicator. J By Saponification. By Saponification after Acetylization. I Saponification Value 195. II Saponification Value 102. 622 TECHNICAL METHODS OF ANALYSIS TABLE IV VOLUMETRIC SOLUTIONS (0.1 N AgNO s Continued) 1 cc. of 0.1 N AgNO 3 is equivalent to: Substance Gram Logarithm Cl 0.003546 CN* 0.005203 FeCl 2 0.006338 FeCl 3 0.005407 HBr 0.008093 HC1 0.003647 HCN * ! 0.005404 HI 0.012793 I 0.012692 KBr 0.011902 KC1 , 0.007456 KCN * 0.013022 KI 0.016602 K 2 0.004710 KSCN 0.009717 LiCl 0.004240 MgCl 2 0.004762 MgCl 2 -6H 2 O 0.010167 NaBr 0.010292 NaBr-2H 2 0.013895 NaCl 0.005846 NaCN * 0.009802 Nal 0.014992 NaI-2H 2 O 0.018595 Na 2 0.003100 NH 4 Br 0.009796 NH 4 C1. 0.005350 NHJ 0.014496 NH 4 SCN 0.007611 PbCl 2 0.013906 SrCl 2 0.007928 SrCl 2 -6H 2 0.013333 Theobromine, C 7 H 8 N 4 O 2 0.018013 ZnCl 2 0.006815 7.54974- 7.71625 7.80195 7.73296 7.90811 7.56194 7.73272 8.10697 8.10353 8.07562 7.87251 8.11468 8.22016 7.67302 7.98753 7.62737 7.67779 8.00719 8.01250 8.14286 7.76686 7.99131 8.17586 8.26940 7.49136 7.99105 7.72835 8.16125 7.88144 8.14320 7.89916 8.12493 8.25558 7.83347 -10 * Liebig Method (See page 31). TABLES 623 TABLE IV VOLUMETRIC SOLUTIONS (Continued) 1 cc. of 0.1 N Iodine is equivalent to: Substance Gram Logarithm Acetone (CH 3 )oCO 0009677 6 98574-10 As 0.003748 7.57380 AsO 3 006148 7.78873 As 2 O 3 004948 7 69443 As 2 O 5 005748 7 75952 Br 007992 7 90266 CaOCl 2 (Bleach) 006350 7 80277 Cl 003546 7 54974 CrO 3 003333 7 52284 Cr 2 O 3 002533 7 40364 Cu 006357 7 80325 CuO 0.007957 7.90075 CuSO 4 015963 8 20311 CuSO 4 -5H 2 O . . 024971 8 39744 Fe'" 005584 7 74695 Fe 2 O 3 007984 7.90222 HNO 2 ... 002351 7 37125 H 2 S 001704 7 23147 H 2 SO 3 004104 7 61321 Iodine . 012692 8 10353 KC1O 3 002043 7 31027 K 2 CrO 4 006473 7 81111 K 2 Cr 2 O 7 ... 004903 7 69046 KMnO 4 003161 7 49982 KNO 2 004256 7 62900 Na 2 CrO 4 005400 7.73239 Na^Cr-sOr 004367 7 . 64018 Na-iCraOr^HijO 004967 7 69609 NaNO 2 003451 7 53794 Na 2 S 0.003903 7.59140 Na 2 S-9H 2 O Na e SO 3 0.012010 006303 8.07954 7 79955 Na 2 S(X 7HO 012609 8 10069 Na 2 S 2 O 3 015812 8 19899 Na2S 2 O 3 -5H..O 024820 8 39480 Nag&Os 004753 7 67697 (NH 4 ) 2 CrO 4 005070 7 70501 Oxveen. . 0.0008000 6.90309 624 TECHNICAL METHODS OF ANALYSIS TABLE IV VOLUMETRIC SOLUTIONS (0.1 N Iodine Continued] I cc. of 0.1 N Iodine is equivalent to: Substance Gram Logarithm PbCrO 4 ' . 010773 8 03234 10 PbO 2 0.011960 8 . 07773 Pb 3 O 4 0.03428 8 53504 S 001603 7 20493 SO 2 0.003203 7 50556 Sb . 006010 7 77887 SboOg 0.007210 7 85794 Sn .... 005935 7 77342 1 cc. of 0. 1 N Na 2 S 2 O 3 is equivalent to: Acetone (CH 3 ) 2 CO 0009677 6 98574 10 Br 007992 7 90266 CaOCl 2 (Bleach) 0.006350 7 80277 Cl 003546 7 54974 CrO 3 0.003333 7 52284 Cr 2 O 3 0.002533 7 . 40364 Cu 0.006357 7 80325 CuO . 007957 7 90075 CuSO 4 015963 8 20311 CuSO 4 -5H 2 O . 024971 8 39744 HNO 2 002351 7 37125 Iodine 0.012692 8 10353 K 2 CrO 4 006473 7 81111 K 2 Cr 2 O 7 0.004903 7 69046 Na 2 CrO 4 005400 7 73239 Na 2 Cr 2 O 7 004367 7 64018 Na 2 Cr 2 O 7 -2H 2 O . 004967 7 69609 NaNO 2 003451 7 53794 Na^Os 015812 8 19899 Na 2 S 2 O 3 5H 2 O . 024820 8 39480 (NH 4 )oCrO 4 005070 7 70501 PbCrO 4 010773 8 03234 PbOo 011960 8 07773 Pb 3 O 4 03428 8 53504 s 001603 7 20493 SO 2 0.003203 7 50556 Sb 0.006010 7 77887 SbzOs 007210 7 85794 Sn 0.005935 7 77342 TABLES 625 TABLE IV VOLUMTERIC SOLUTIONS (Continued) 1 cc. of 0.1 N KMnO 4 is equivalent to: Substance Gram Logarithm BaO 2 008469 7 92783 10 BaO 2 -8H 2 O 015675 8 19521 CaCO 3 005004 7 69932 CaO 0028035 7 44770 CaO 2 0036035 7 55672 CaSO 4 006807 7 83296 CaSO 4 .2H 2 O 008608 7 93490 Fe 005584 7 74695 Fe(NH 4 ) 2 (SO 4 ) 2 -6HoO 039214 8 59344 FeO 007184 7 85637 FeoO 3 007984 7 90222 Fe 3 O 4 007717 7 88745 FeSO 4 015190 8 18156 FeSO 4 -7H 2 O 027801 8 44406 HCOoH (Formic Acid) 002301 7 36192 H 2 C 2 O 4 004501 7 65331 H 2 C 2 O 4 -2H 2 O 006303 7 79955 H,O 2 003402 7 53173 Iodine 012692 8 10353 KMnO 4 003161 7 4QQS2 KNO 2 004256 7 62900 K 2 Cr 2 O 7 004903 7 69046 K 2 S 2 O 8 013516 8 13085 Mn 0010986 7 04084 MnO 0014186 7 15186 MnO 2 004347 7 63819 MoO 3 * 0049 7 6Q020 Na 2 CoO 4 006701 7 82614 NaNO 2 003451 7 53794 Na 2 S2O 8 011906 8 07577 (NH 4 ) 2 C 2 O 4 006205 7 70074 (NH 4 ) 2 C 2 O 4 -H 2 O 007105 7 85156 (NH 4 ) 2 S 2 O 8 011410 8 05729 P * .. 000089 f) Q4Q3Q P 2 O 5 * 000203 6 ^07^0 Sb 006010 7 77SS7 Sn 005935 7 77^49 Tannin, C 14 Hi O 9 004157 7 R1S7S * From titration of yellow phosphomolybdate after reduction (See page 25). 626 TECHNICAL METHODS OF ANALYSIS TABLE IV VOLUMETRIC SOLUTIONS (Continued) 1 cc. of 0. 1 N K 2 Cr 2 O 7 is equivalent to: Substance Gram Logarithm CrO 3 . 003333 7 52284 10 Cr 2 O 3 002533 7 40364 Fe" 005584 7 74695 FeO 007184 7 85637 Fe 3 O 4 '. . 007717 7 88745 FeSO 4 015190 8 18156 FeSO 4 -7H 2 O 027801 8 44406 Glycerol C 3 H 5 (OH) 3 0006577 6 81803 K 2 Cr 2 O 7 004903 7 69046 PbCrO 4 010773 8 03234 Zn . 003269 7.51441 BIBLIOGRAPHY The following is a list of those books which we have found to be of great value in Analytical Work. It does not claim to be a complete list, but a representative selection of books which will answer the questions that arise in the course of routine or special analytical work. Prices are not given, because they are changing from time to time and it will doubtless be some time before they reach a stable condition. Handbooks HODGMAN, C. D. Handbook of Chemistry and Physics. 7th ed. 1918. Chemical Rubber Co., Cleveland, Ohio. LINDELL, D. M. Metallurgists' and Chemists' Handbook. 2d ed. McGraw- Hill Book Co., Inc. MEADE, R. K Chemists' Pocket Manual. 3d ed. 1918. Chemical Pub- lishing Co., Easton, Pa. OLSEN, J. C. Van Nostrand's Chemical Annual. 1918 ed. D. Van Nos- trand Co. SEIDELL, A. Solubilities of Inorganic and Organic Substances. 2d ed. 1919. D. Van Nostrand Co. Qualitative Analysis BASKERVILLE, C., and CURTMAN, L. J. Course in Qualitative Analysis. 2d ed. 1916. The Macmillan Co. . COHN, A. I. Indicators and Test Papers. 2d ed. Wiley & Sons. COHN, A. I. Tests and Reagents. 1903. Wiley & Sons. MERCK, VON E. Chemical Reagents, their Purity and Tests. Translated by Henry Schenck. 2d ed. 1914. D. Van Nostrand Co. MULLIKEN, S. P. Methods for the Identification of Pure Organic Compounds by a Systematic Analytical Procedure Based on Physical Properties and Chemical Reactions. 1904-1916. Wiley & Sons. Volume I. Compounds of Carbon with Hydrogen and Oxygen. II. Nitrogenous Compounds. III. Commercial Dyestuffs. MURRAY, B. L. Standards and Tests for Reagent Chemicals. 1920. D. Van Nostrand Co. NOTES, A. A. Qualitative Analysis of Inorganic Substances. 6th ed. 1915. The Macmillan Co. 627 628 BIBLIOGRAPHY PRESCOTT, A. B., and JOHNSON, O. C. Qualitative Chemical Analysis. 7th ed., revised by John C. Olsen. 1916. D. Van Nostrand Co. STEIGLITZ, J. O. Elements of Qualitative Analysis. 2 vols. 1911. The Century Co. TREADWELL, F. P. Analytical Chemistry. I. Qualitative Analysis. 4th English ed. Translated by W. T. Hall. 1915. Wiley & Sons. Quantitative Analysis CLASSEN, A. Ausgewahlte Methoden der analytischen Chemie. 1901. Fried- rich Vieweg und Sohn, Braunschweig, Germany. MAHIN, E. G. Quantitative Analysis. 2d ed. 1919. McGraw-Hill Book Co., Inc. OLSEN, JOHN C. Textbook of Quantitative Analysis. 5th ed. 1916. D. Van Nostrand Co. SUTTON, F. Volumetric Analysis. 10th ed. 1911. Blakiston's Son & Co. TREADWELL, J. C. Analytical Chemistry. II. Quantitative Analysis. 4th English ed. Translated by W. T. Hall. 1916. Wiley & Sons. Technical Analysis Inorganic AMERICAN SOCIETY FOR TESTING MATERIALS. Triennial Standards. 1918. See also supplementary volume for 1919. LUNGE, GEORGE. Technical Methods of Chemical Analysis. Translated by C. A. Keane. 3 vols. 1908-1914. D. Van Nostrand Co. SCOTT, W. W. Standard Methods of Chemical Analysis. 2d ed. 1917. D. Van Nostrand Co. VILLAVECCHIA, V. Treatise on Applied Analytical Chemistry. Translated by T. H. Pope. 2 vols. 1918. Blakiston's Son & Co. Technical Analysis Organic ALLEN, ALFRED H. rCommercial Organic Analysis. 9 vols. 1909-1917. A Treatise on the Properties, Proximate Analysis, Analytical Examina- tion, etc. Blakiston's Son Co. GATTERMANN, LUDWIG. Practical Methods of Organic Chemistry. Latest ed. The Macmillan Co. MULLIKEN. See under Qualitative Analysis. SHERMAN, H. C. Methods of Organic Analysis. 3d ed. 1912. The Mac- millan Co. SPECIAL SUBJECTS Agriculture ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS. Official and Tentative Methods of Analysis. 1920. This covers also such subjects as: Foods and Feeding Stuffs, Soils, Fertilizers, Insecticides, Fruits, Beverages, Meals, Dairy Products, Food Preservatives, etc. WILEY, HARVEY W. Principles and Practice of Agricultural Analysis. III. Agricultural Products. 2d ed. 1914. Chemical Publishing Co., Easton, Pa. BIBLIOGRAPHY 629 Alloys BLOUGH, E. Analysis of Aluminium and Its Commercial Alloys. 1910. Aluminum Co. of America, Pittsburgh, Pa. JOHNSON, C. M. Rapid Methods for the Chemical Analysis of Special Steels, Steel Making Alloys and Graphite. 3d ed. 1920. Wiley & Sons. LORD, N. W., and DEMOREST, D. J. Metallurgical Analysis. 4th ed. 1916. McGraw-Hill Book Co., Inc. PRICE, W. B., and MEADE, R. K. Technical Analysis of Brass and the Non- ferrous Alloys. 2d ed. 1911. Wiley & Sons. Asphalt ABRAHAM, H. Asphalt and Allied Substances. 1918. D. Van Nostrand Co. Dyes GREEN, A. G. Analysis of Dyestuffs and their Identification in Dyed and Colored Materials, etc. 1915. Lippincott Co. MULLIKEN. See Qualitative Analysis. SCHULTZ, G., and JULIUS, P. Systematic Survey of the Organic Coloring Matters. 5th ed. 1914. 42d Street Commercial Studio, New York City. WEISS, J. M. Methods of Analysis of the Coal-tar Industry. 1918. The Barrett Co. Cement MEADE, RICHARD K. Portland Cement. 2d ed. 1911. Chemical Pub- lishing Co., Easton, Pa. Drugs and Chemicals PHARMACOPOEIA OF THE UNITED STATES. IX Decennial revision. 1916. Blakiston's Son & Co. Foods ALLEN. See Technical Analysis Organic. ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS. See Agriculture. LEACH, A. C., and WINTON, A. L. Food Inspection and Analysis. 4th ed. 1920. Wiley & Sons. WINTON, A. L. Microscopy of Vegetable Foods. 2d ed. 1916. Wiley & Sons. WOODMAN, A. G. Food Analysis. 1915. McGraw-Hill Book Co., Inc. Fuels BACON, R. F., and HAMOR, W. A. American Petroleum Industry. 2 vols. 1916. McGraw Hill Book Co., Inc. CROSS, R. Handbook of Petroleum, Asphalt, and Natural Gas. 1919. Kansas City Testing Laboratory. GILL, A. H. Gas and Fuel Analysis for Engineers. 8th ed. 1917. Wiley &Sons. 630 BIBLIOGRAPHY HOLDE, D. Examination of Hydrocarbon Oils. Translated by E. Mueller. 1915. Wiley & Sons. Gas Analysis GAS CHEMISTS' HANDBOOK. 2d ed. 1920. American Gas Association, New York City. HEMPEL, WALTHER. Methods of Gas Analysis. Translated from 3d Ger- man ed. 1906. The Macmillan Co. WHITE, ALFRED H. Technical Gas and Fuel Analysis. 2d ed. 1920. McGraw-Hill Book Co., Inc. Iron and Steel BLAIR, A. A. The Chemical Analysis of Iron. 8th ed. 1918. Lippin- cott Co. See also under Alloys. Leather PROCTOR, H. R. Leather Industries' Laboratory Book of Analytical and Experimental Methods. 2d ed. 1908. Spon & Chamberlain. Minerals and Ores Low, A. H. Technical Methods of Ore Analysis. 8th ed. 1919. Wiley & Sons. Oils, Fats and Soaps FRYER, F. E., and WESTON, P. J. Technical Handbook of Oils, Fats and Waxes. 2d ed. 1918. University Press, Cambridge, Eng. Putnam's Sons. GILL, A. H. Short Handbook of Oil Analysis. 8th ed, 1918. Lippincott. LAMBORN, LLOYD L. Modern Soaps, Candles and Glycerin. 1906. D. Van Nostrand Co. LEWKOWITSCH, J. Chemical Technology and Analysis of Oils, Fats and Waxes. 5th ed. 3 vols. 1913-1915. The Macmillan Co. PICKERING, G. F. Aids in the Chemical Analysis of Oils, Fats and Their Commercial Products. 1917. Lippincott Co. Paints and Varnishes GARDNER, H. A., and SCHAEFFER, JOHN A. Analysis of Paints and Painting Materials. 1911. McGraw-Hill Book Co., Inc. HOLLEY, C. D. Analysis of Paint Vehicles, Japans and Varnishes. 1920. Wiley & Sons. HOLLEY, C. D., and LADD, E. F. Mixed Paints, Color Pigments and Var- nishes. 1908. Wiley & Sons. TOCH, M Chemistry and Technology of Paints. 2d ed. 1916. D. Van Nostrand Co. Paper and Cellulose BROMLEY, H. A. Outlines of Stationery Testing. 1913. Lippincott Co. GRIFFIN, R. B., and LITTLE, A. D. Chemistry of Papermaking. 1894. G. E. Stechert, New York City. BIBLIOGRAPHY 631 HERZBERG, W. Papierpriifung. 4th ed. 1915. G. E. Stechert, New York City. SCHWALBE, C. G., and SIBBER, R. Chemische Betriebskontrolle in der Zellstoff- und Papierindustrie. 1919. G. E. Stechert, New York City. STEVENS, H. P. Paper Mill Chemist. 2d ed. 1919. D. Van Nostrand Co. WITHAM, G. S. Modern Pulp and Papermaking. 1920. Chemical Catalog Co. WORDEN, E. C. Technology of Cellulose Esters. Vol. VIII. 1916. D. Van Nostrand Co. Rubber HEIL, A., and ESCH, W. Manufacture of Rubber Goods. 2d ed. 1919. Griffin, London. WEBER, C. O. Chemistry of India Rubber. 1903. Lippincott Co. Sugar BROWN, C. A. Handbook of Sugar Analysis. 1912. Wiley & Sons. PRINSEN GEERLIGS, H. C. Chemical Control in Cane Sugar Factories. 2d ed. 1917. N. Rodger, London. Textiles BARKER, A. F., and MIDGELY, E. Analysis of Woven Fabrics. 1914. D. Van Nostrand Co. KNECHT, E., LOWENTHAL, R., and RAWSON, C. Manual of Dyeing. 3d ed. 2 vols. 1916. Lippincott Co. MATTHEWS, J. M. The Textile Fibers. 3d ed. 1913. Wiley & Sons. MITCHELL, C. A., and PRIDEAUX, R. M. Fibers Used in Textile and Allied Industries. 1910. Scott, Greenwood & Son, London. Water AMERICAN PUBLIC HEALTH ASSOCIATION. Standard Methods of Water Analysis. 3d ed. 1917. American Public Health Association, Boston. MASON, W. P. Examination of Water. 5th ed. 1917. Wiley & Sons. PRESCOTT, S. C., and WINSLOW, C.-E. A. Elements of Water Bacteriology, with Special Reference to Sanitary Water Analysis. 3d ed. 1913. Wiley & Sons. RICHARDS, ELLEN, Conservation by Sanitation. 1911. Wiley & Sons. GOVERNMENT PUBLICATIONS The following bulletins from the various Government Bureaus mentioned below deal in whole or in part with analytical methods which, as mentioned in the preface, have been freely used in the preparation of the methods of analysis given in this book. Bureau of Chemistry BULLETIN 66. Fruits and Fruit Products: Chemical and Microscopical .Examination. 1902. 632 BIBLIOGRAPHY BULLETIN 79. Testing of Road Materials, Including Methods Used and Results Obtained in Road Material Laboratory in Collaboration with Office of Public-road Inquiries. 1903. BULLETIN 91. Mineral Waters of United States. 1905. BULLETIN 108. Commercial Feeding Stuffs of United States, their Chemical and Microscopical Examination. 1908. BULLETIN 109. Some Technical Methods of Testing Miscellaneous Supplies, Including Paints and Paint Materials, Inks, Lubricating Oils, Soaps, etc. 1908. BULLETIN 114. Meat Extracts and Similar Preparations, Including Studies of Methods of Analysis Employed. 1908. BULLETIN 135. Commercial Turpentines, their Quality and Methods for their Examination. 1911. BULLETIN 147. Coal-Tar Colors Used in Food Products. 1912. CIRCULAR 25. Coloring Matters for Foodstuffs and Methods for their Detection. 1905. CIRCULAR 63. Identification of Food Colors, Tentative Report on Solubility and Extraction of Certain Colors and Color Reactions of Dyed Fiber and of Aqueous and Sulfuric Acid Solutions. 1911. CIRCULAR 89. Quantitative Separation of Mixtures of Certain Acid Coal-Tar Dyes. 1912. CIRCULAR 107. Detection of Faulty Sizing in High-grade Papers. 1913. CIRCULAR 113. Quantitative Separation and Determination of Subsidiary Dyes in Permitted Food Colors. 1913. Bureau of Mines BULLETIN 12. Apparatus and Methods for the. Samplings and Analysis of Furnace Gases. 1911. BULLETIN 42. The Sampling and Examination of Mine Gases and Natural Gas. 1913. BULLETIN 97. Sampling and Analysis of Flue Gases. 1915. BULLETIN 116. Methods of Sampling Delivered Coal, and Specifications for the Purchase of Coal for the Government. 1916. BULLETIN 125. The Analytical Distillation of Petroleum. 1916. TECHNICAL PAPER 8. Methods of Analyzing Coal and Coke. 1913. TECHNICAL PAPER 25. Methods for the Determination of Water in Petro- leum and its Products. 1912. TECHNICAL PAPER 26. Methods for the Determination of the Sulfur Content of Fuels, Especially Petroleum Products. 1912. TECHNICAL PAPER 31. Apparatus for the Exact Analysis of Flue Gas. 1913. TECHNICAL PAPER 49. The Flash Point of Oils; Methods and Apparatus for its Determination. 1913. TECHNICAL PAPER 148. The Determination of Moisture in Coke. 1917. TECHNICAL PAPER 166. Motor Gasoline; Properties, Laboratory Methods of Testing, and Practical Specifications. 1917. TECHNICAL PAPER 186. Methods for Routine Work in the Explosive and Physical Laboratory of the Bureau of Mines. 1918. BIBLIOGRAPHY 633 TECHNICAL PAPER 212. The Determination of Combustible Matter in Sil- icate and Carbonate Rocks. 1919. TECHNICAL PAPER 214. Motor Gasoline; Properties, Laboratory Methods of Testing, and Practical Specifications. 1919. Bureau of Standards CIRCULAR 19. Standard Density and Volumetric Tables. 1916. CIRCULAR 25. Standard Samples General Information. 1917. CIRCULAR 33. United States Government Specification for Portland Cement. 1917. CIRCULAR 38. The Testing of Rubber Goods. 1915. CIRCULAR 41. Testing and Properties of Textile Mater als. 1918. CIRCULAR 45. The Testing of Materials. CIRCULAR 48. Standard Methods of Gas Testing. 1916. CIRCULAR 62. Specifications for and Methods of Testing Soaps. 1919. CIRCULAR 95. Inks Their Composition, Manufacture, and Methods of Testing. 1920. Geological Survey BULLETIN 700. Analysis of silicate and carbonate rocks. 1919. INDEX Abb refractometer, use of Abietic acid in rosin Abrams and Harder color test for sand. . Abrasive in metal polish Absorption test of blotting paper Acetaldehyde, standard solution of .... Acetate of lime (See Calcium acetate) . Acetate silk Acetates in carbolineum , qualitative test for Acetic acid in acetate of lime carbolineum ethyl alcohol lead arsenate "light vinegar" pyroligneous acid. . , vinegar wood acid wood distillate wood preserving oils , crude, analysis of , qualitative test for , reagent , anhydride, analysis of , aniline method for , boiling point of , direct titration of Acetin method for glycerol. . Acetone in wood alcohol 72, , Messinger method for extract of rubber Acetyl number (See Acetyl value). value of oils Acetylene equivalent of calcium carbide Acetylization of glycerol oils Acid (See also Acidity). in leather lubricating oils vinegar , abietic, in rosin , acetic (See Acetic acid). , arachidic, melting point of , arsenious, 0.1 N solution of , benzoic,' heating value of , butyric, in oils , carbolic, in soap (reference) 231 328 580 490 354 76 380 535 535 32 535 74 49 370 369 457 370 369 535 370 535 1 82 83 84 82 85 371 72 480 244 486 85 244 475 260 456 328 251 11 178 243 288 Acid, carbonic (See Carbon dioxide). , chlorplatinic, preparation of 16 , formic, analysis of 81 , , in vinegar 457 , hydrochloric, in formic acid 81 , , reagent 8 , , volumetric solutions of 7, 8 , lactic, in cheese 452 , lead, preparation of 138 , malic, in vinegar 456 , mineral, in leather 475 , , lubricating oils 260 , nitric, reagent 4 , , red fuming, reagent 4 , , test for groundwood 339 number of beeswax 276 fats and oils 242 rosin 328 , oxalic, 0. IN solution of 8 , picric, test for gelatin 448 , phosphoric (See Phosphoric acid and Phosphoric anhydride). , prussic (See Hydrocyanic acid). , stearic, in beeswax 278 , sulfite, analysis of 301 , , composition of 301, 302 , sulf uric (See also Sulfur trioxide) . , , in aluminum sulfate 304 , , fdrmic acid 81 , , leather 475 , , sulfite acid 301 , , free, in aluminum sulfate 305 , .fuming (See Oleum). , , reagent 5 , , 38 Normal 197 , sulfurous (See also Sulfur dioxide). , , in paper 342 , , sulfite acid 301 , tannio, analysis of 95 , value (See Acid number). , , malic, of maple products 427 , wood, analysis of 370 Acidity of aluminum sulfate 305 casein 317 cheese 452 ethyl alcohol 74 f eedstuffs and grains 442 635 636 INDEX Acidity of gasoline 187 glue. . 320 leather 475 paper 342 Acids, fatty (See Fatty acids). , tar (See Tar acids). Acorn oil, constants of 232 Adams method for fat in condensed milk 450 Agalite in paper 341 Aggregate for concrete 576 Air-compressor oil, testing of 254 Alba whiting, composition of 201 Albrech test for lemon peel color 463 orange peel color 463 Albumin, analysis of 93 , blood-serum 93 , egg 93 , soluble coagulable, determination of 94 Albuminoid nitrogen in feedstuff s 439 water 502 Alcohol, aldehyde-free, preparation of 75, 460 , amyl (See Fusel oil) , ethyl, in brandy drops 419 , , confectionery syrups 419 , , lemon extract 458 , , orange extract 458 , , peppermint extract 467 , , spearmint extract 467 , , syrups 419 , , vanilla extract 468 , , vinegar 454 , , wintergreen extract 467 , , analysis of 74 , by volume 459 , weight 458 tables (reference) 74, 419, 454 , grain (See Alcohol, ethyl). . , methyl (See Methyl alcohol). , wood, crude, in wood distillate. . . . 369 Alcoholic potash extract of rubber .... 482 Alcoholic potash, 0. 5 N solution of . . . 1, 29 Aldehyde in alcohol, silver nitrate test for 75 ethyl alcohol 75 lemon extract 460 orange extract 460 , Chace method for 460 free alcohol, preparation of 75, 460 , sulfite f uchsin method for 75 Alizarin lakes, composition of 202 Alkali in casein 315 rosin size 330 rosin size milk 332 soap 280, 282, 283 Alkalies, analysis of 19 Alkalies, in water 518 , qualitative tests for 20 Alkalimeter, Mohr, for COz determina- tion (Fig. 13) 212 Alkalinity of black sulf ate liquor . . 296, 297 cocoa 431 glue 320 white sulf ate liquor 299 Allihn method for dextrose 407 Fehling's solutions 3, 407 Alloy steel, analysis of 116 Alloys, aluminum, analysis of 160 , Babbitt, analysis of 157 , brass and bronze, analysis of 142 , German silver, analysis of 152 , nickel silver, analysis of 1 52 , solder, analysis of 154 , type metal, analysis of 158 , white, analysis of 154, 160 , zinc amalgam, analysis of 163 Almond oil, constants of 232 Alternate shovel method of reducing samples 170 Alum, paper makers', analysis of 303 , resistance of ultramarine to 336 Alumina in aluminum sulf ate 303 blanc fixe 310 boiler scale 523 chrome yellow 216 Portland cement 574 salt 25 water 518 , basic, in aluminum sulfate 305 cream, preparation of 2, 399 Aluminum in aluminum alloys 160 brass and bronze 151 alloys, analysis of . . . : 160 foil method for nitrates 507 hydroxide for clarifying water 510 in satin white . . 332 oxide (See Alumina) sulfate, analysis of 303 Amalgam, zinc, analysis of 163 Amalgamated zinc, preparation of. ... 149 Amalgamation of brass 542 American Elect. Ry. Eng. Assoc. tin- ning test 166 Pulp and Paper Assoc. method for wood pulp testing 293 Tel. and Tel. Co. tinning test 166 test on paraffin wax 279 vermilion, composition of 202 Amido nitrogen in feedstuffs and grains. 440 Amidonaphthalene acetate reagent for nitrites 504 Aminoazotoluene, separation of. 392 Ammonia (See a.lso Ammonium hy- droxide), INDEX 637 Ammonia, analysis of 23 , aqua, U. S. P 23 in ammonia liquors 23 ammonium hydroxide 23 case-hardening compounds 485 eggs 387 fertilizers 525 lead arsenate 49 metal polishes 490 organic materials 68 soldering paste 491 , nesslerization method for 501 , titration with methyl orange 66 liquor, analysis of 23 nitrogen in sewage 501 , , albuminoid, in water 502 , , free, in water 498 , , magnesia method for 68 water, analysis of 23 and nitric nitrogen 68 Ammoniacal silver nitrate solution. . . . 299 Ammonium acetate reagent 2 carbonate reagent 2 chloride reagent 2 in soldering paste 491 citrate solution, preparation of 525 hydroxide, analysis of 23 reagent 2 molybdate solution for phosphorus in steel 113 P 2 O 5 in fertilizers 525, 528 reagent 2 nitrate reagent 2 oxalate reagent 2 phosphate reagent 3 phosphomolybdate, preparation of . . 113 polysulfide reagent 3 sulf ate reagent 3 , moisture in 525 sulfide reagent 3 sulfocyanate, 0. 1 N solution 11 Amyl alcohol (See Fuseloil). Analytical factors, tables of 608, 615 Anchovy oil, constants of 232 Anhydrides, analysis of 82 Aniline chloride test for invert sugar.. . 424 dyes, testing of 307 method for acetic anhydride 83 sulf ate test for ground wood 338 yellow, separation of 392 Animal and vegetable oils and fats, analysis of 230 size (See Glue). Annatto, separation of 392 Antimonious acid (See Antimony oxide). Antimony, golden sulfide of 34 in Babbitt metals 157 brass and bronze. . . . 150 Antimony in solder 155 type metals 159 white metals 155 , permanganate method for 150, 155 oxide in antimony sulfide 35 sulfide, analysis of 34 Anti-tarnish paper, testing of 362 Apricot kernel oil, constants of 232 Aqua ammonia, U. S. P 23 regia 157 Arabinose, Allihn method for 408 Arachidic acid, melting point of 251 method for peanut oil 250 Arachis oil (See Peanut oil). Archil, dyeing test for 390 Armature insulating varnish, specifica- tions for 227 , testing of 227 Arnold-Kjeldahl-Gunning method for nitrogen 66 Arsenate of lead (See Lead arsenate) . Arsenic, determination of small amounts of 35 , distillation method for 55, 149 , Gutzeit (modified) method for 37 in alloys 149 beer 35, 39 Bordeaux mixture with lead arsenate . 54 with Paris green 53 brass and bronze 149 chemicals 35, 39 drugs 35, 39 foods 35, 40 lead arsenate 47 lead arsenate with Bordeaux mix- ture 54 paper 40 Paris green 55 Paris green with Bordeaux mix- ture 53 textiles 40 wall papers 40 , Marsh test for 36 , Sanger-Black-Gutzeit method for. . 37 , titration with iodine 47 , water-soluble, in lead arsenate 49 bronze, analysis of 149 mirrors, preparation of standard .... 37 oxide (See also Arsenic). in lead arsenate 47 Arsenious acid, 0. 1 N solution of 11 oxide in Paris green 56 , C. C. Hedges method for 56 , C. M. Smith method for 57 , sodium acetate-soluble, in Paris green 57 , water-soluble, in Paris green... . 57 638 INDEX Asbestine, composition of 201 in paper. 341 Asbestos in asbestos-magnesia pipe covering 63 , purification of. 403, 405 cotton twine, analysis of . . 385 magnesia pipe covering, analysis of 62 , composition of 62 Ash of albumin 94 asphalt 546 bituminous materials 546 black sulfate liquor 295 butter 428 carbolineum 535 casein 315 cheese 452 chocolate and cocoa 431 coal 174 cotton linters 364 glue 320 grease 272 honey 420 indigo 99 leather 474 milk 444 paper 340 rope and twine 383 rubber compounds. " 483 saccharine products 413 , electrolytic method for... . 414 sulfonated oil 264 sulfur 17 tallow 270 textile fabrics 372 tobacco 102 varnish 220 vinegar 456 wood-preserving oils. 1 535 , sulfate method for 94, 414 , sulfated, of albumin 94 Ashes, coal, analysis of 185 Asphalt products in tar, test for 547 Asphaltic road binders, analysis of .... 537 Assaying of tin ores 136 Association of American Wood Pulp Importers' method for wood pulp testing 293 Astringency of sumac extract 479 Atomic weights, table of 584 group weights, table of 586 Auramine, separation of 392 Auramine O, behavior on dyeing 390 Automobile oil, testing of 254 Available chlorine in bleach 312 organic nitrogen 69 Azotometer method for nitrogen (refer- ence) 67 B.t.u. (See British thermal units). Babbitt metals, analysis of 157, 158 Babcock method for fat in milk and cream 446 test bottles 446 Ball and ring method for softening point 543 Bardy and Riche test for methyl alcohol 462 Barium carbonate in blanc fixe 310 paper 341 chloride in salt 26 reagent 3 hydroxide reagent 8 phosphate in blanc fixe 310, 311 sulfate in blanc fixe 311 sulfate in paint pigments 207 paper 341 paper coating 344 Barytes (See also Barium sulfate). , composition of 201 Basicity of aluminum sulfate 305 Battery zinc amalgams, analysis of . . . . 163 Baudouin test for sesame oil 252 Baume and Brix comparison table 411 gravity, temperature corrections (reference) 255 and specific gravity, comparison table 411 of oils, relation of 255 Bear oil, constants of 232 Beechnut oil, constants of 232 Beef fat in lard, Emery test for 253 marrow, constants of 232 tallow, analysis of 268 , constants of 232, 269 Beer, arsenic in 39 , coal-tar dyes in 389 , sulfur dioxide in 396 Beeswax, adulterants of 275, 277 , analysis of 275 , constants of 232, 277 .bleached 275 , Indian 277 Beet products, sugar in 401 Bellier test for peanut oil. 252 . sesame oil 253 Ben oil, constants of 232 Benzene-insoluble matter in carbolin- eum 535 coal-tar roofing pitch 536 wood-preserving oils 535 Benzeneazobetanaphthylamine, separ- ation of 392 Benzine 186 Benzoic Acid, heating value of 178 Benzol (See Benzene). Bertrand and Javillier method for nico- tine. .. . 101 INDEX 639 Beverages, coal-tar dyes in 389 Bibliography of reference books 627 Bicarbonate in alkalies 20 Bichloride of mercury (See Mercuric chloride) . Bichromate method for- glycerol 89 metallic zinc 141 of potash (See Potassium bichro- mate) . soda (See Sodium bichromate). Bichromates, volumetric method for. . . 30 Binder, asphalt road, analysis of 537 Birotation of honey 421 , overcoming of 415 Bismuthate method for manganese . 1 10, 147 Bisulfite liquor, analysis of 301 , composition of 301, 302 Bitumen in bituminous materials 544 insoluble in 86 naphtha 546 soluble in carbon bisulfide 544 Bituminous road binders, analysis of . . 537 Black-Gutzeit-Sanger method for ar- senic 37 Black insulating varnish, analysis of. . . . 227 liquor (sulfate), analysis of 295 oil, testing of 254 pigments, classification of 201 Blanc fixe, analysis of 308 , composition of 201, 308 for photographic purposes 309 in paper 34 1 paper coating 344 . silver nitrate test of 309 Bleach, analysis of 312 , Bunsen's method for 313 consumption of pulp 313 liquor, analysis of 314 Bleaching, determination of excessive. . 368 powder, analysis of 312 Blood serum albumin, analysis of 93 Blotting paper, absorption test of 354 Blue pigments, classification of 203 Bluing in clay 319 , turpentine test for 319 Boemer method for unsaponifiable mat- ter in oils 262 Boiler scale, analysis of 522 water, analysis of 514 , grading of 521 Boiling point of turpentine 194 Bomb calorimeter (See Calorimeter). Bone black, composition of 201 fat, constants of . . . . : 232 fertilizers, analysis of . 524 oil, constants of 232 Books, reference 627 Borates, qualitative test for 284 Borax in soap 283 Bordeaux mixture, analysis of 50 , standard 50 with lead arsenate, analysis of ... 54 Paris green, analysis of 52 Borneo tallow, constants of 132 Boron (see also Boric acid and Borates). Bottlenose oil, constants of 232 Boughton method for gums in varnish (reference) 217 Brandy drops, alcohol in 419 Brass, amalgamation of 542 y, analysis of 142 Brazilnut oil, constants of 232 Breaking factor of paper 347 length of paper 346 Brimstone (See Sulfur). Briquettes, cement, preparation of .... 569 , , storage of 571 British gum (See Dextrin). British thermal units and calories, rela- tion of 178 in coal 176 coal ash 185 liquid fuels 190 per gallon 191 Brix and Baume gravity comparison table 411 Brix hydrometer 410 , temperature corrections for 413 Bromine absorption of wood alcohol. . . .73 test for fish oils 254 Bronze, analysis of 142 Brown pigments, classification of.,. ... 203 Browne heating test for tung oil 199 Bryan-Fiehe test for invert sugar 423 Bunsen's method for chlorine in bleach. 313 Burning point (See Fire point). test (See Fire point) . Burnt sienna (See Sienna, burnt). Bursting factor of paper .' 349 ratio of paper 349 strength of paper 349 textile fabrics 373 Butter, analysis of 427 , composition of 427 , standard 429 fat, constants of 232 in butter substitutes 428 milk .chocolate 437 , standard 429 , kokum, constants of 234 , nutmeg, constants of 236 , renovated, test for 429 scotch, fat in 418 , shea, constants of 238 substitutes, analysis of 427 , preparation of 427 yellow, separation of 392 640 INDEX Butterine, analysis of 427 Butyric acid in oils 243 anhydride, analysis of 84 Butyro-refractometer degrees to refract- ive index, conversion of 240 Zeiss, use of 240 Cacao butter (See Cocoa butter) products, analysis of 430 , artificial coloring in 391 Cadmium, electrolytic method for 141 in zinc 140 , sulfate method for 140 Caffein in cocoa and chocolate 435 Calcium acetate, analysis of 32 carbide in case-hardening com- pounds 486 , acetylene equivalent of 486 carbonate in paper 341 talc 334 chloride reagent 3 hydroxide reagent 3 in satin white 333 hypochlorite (See Bleach). oxide (See Lime). sulfate, analysis of 319 in antimony sulfide 35 talc 335 , separation from magnesium sul- fate 301 Calender oil, testing of 254 Calories and B.t.u., relation of . . 178 Calorimeter, standardization of 178 Calorimetric analysis of coal 176 liquids 190 Calophony (See Rosin). Calophyllum oil, constants of 232 Cameline oil, constants of 232 Canari oil, constants of . 232 Canary chrome yellow, composition of. 202 Candies, analysis of 409 , artificial coloring in 390 Candlenut oil, constants of . . , 232 Canned goods, artificial coloring in. ... 391 Capsules for calorimetric work 191 Carapa fat, constants of 232 Carbide, calcium, acetylene equivalent of. 486 Carbohydrates in feedstuffs and grains 440 Carbolic acid in soap (reference) 288 Carbolincum, analysis of 530 Carbon, colorimetric method for 131 , combustion method for 108 in coal 182 iron, colorimetric method for. ... 131 , solution-combustion method for... 130 Carbon in steel 108 tar 551 , combined, in iron 131 , , steel 110 , fixed, in bituminous materials 546 , , coal 174 , free, in tar 545, 551 , graphitic, in iron 131 , , paint 207 , , steel 110 bisulfide extract of bituminous ma- terials 544 spent oxide 555 test of sulfur 17 black, composition of 201 dioxide, alkalimeter method for. . . . 212 in alkalies 20, 22 Bordeaux mixture 51 water 515 white lead 212 , solid, preparation of 257 furnace (Fig. 6) 108 steel (See Steel). tetrachloride, purification of 77 Carbonate in alkalies 20 Carbonates, titration of, with methyl orange 7, 22 Carbonic acid (See Carbon dioxide). Carnaiiba wax, constants of 232 Case-hardening compounds, analysis of 484 Casein, analysis of 315 , clay-carrying capacity of 316 , "cutting test" of 316 , foreign and domestic 316 , formaldehyde test for 344 in albumin, qualitative test for 94 butter 428 milk 444 milk chocolate 435 paper coating 344 , muriatic flakeless 316 , solubility of 316 Cassell earth (See Vandyke brown) . Cassiterite 134 Cast iron, analysis of 129 , sampling of 106, 129 Castor oil, analysis of 263 , constants of 232, 363 Cattle foods, analysis of 438 Caustic alkali in soap 282 , test for 20 alkalies, analysis of 19 potash (See Potassium hydroxide) . soda (See Sodium hydroxide) . solutions, volumetric 8 Cedar nut oil, constants 232 Celestron silk . 380 INDEX 641 Cellulose for nitration, analysis of 363 in cotton linters 365 wood 289 , caustic method for 365 , sulfuric acid method for 365 acetate silk 380 Cement, Portland, chemical analysis of . 572 , , composition of 572 , , in mortar and concrete 560 , , physical testing of 561 , , sampling of 563 . specifications for 5C1, 572 briquettes, preparation of 569 , storage of 571 testing machine (Fig. 29) 570 Cereal products, artificial coloring in. . 391 Ceresin, constants of 232 Chace method for aldehyde in extracts 460 citral 460 Chalk, composition of 201 in paper 341 paper coating 344 Charcoal black, composition of 201 iron, analysis of 129 , sampling of 106, 129 Chardonnet silk 379 Chaulmoogra fat, constants of 232 Cheese, analysis of 451 , standard for 451 Cherry kernel oil, constants of 232 laurel oil, constants of 232 Chicken fat, constants of 232 China clay (See Kaolin). wood oil (See also Tung oil). Chinese wax, constant%of 232 wood oil (See also Tung oil). , constants of 198, 234 Chlorate in bleach 312 method for manganese 148 Chlorides (See also Chlorine) . in cyanides 32 paper 343 soap 285 sodium silicate 30 water and sewage 509 , gravimetric method for 27, 81 , volumetric method for 509 Chlorination method for cellulose 289 Chlorine (See also Chlorides). in bleach 312 - boiler scale 524 butter 428 paper 343 sewage 509 soldering paste 492 water 509 , available, in bleach 312 , volumetric method for 509 Chloroform extract of rubber 482 Chlorplatinic acid, preparation of 16 Chocolate, analysis of 430 , bitter, standard 430 , coloring matter in 391 , milk, standard 430 , plain, standard 430 , sweetened, standard 430 liquor, standard 430 Cholesterol in fats -. . . 247 Cholesteryl acetate, melting point of. . . 249 Chromate method for lead. . . .47, 208, 212 Chrome green, composition of 203 leather, chromium in 478 liquid, analysis of 477 orange, composition of 202 oxide green, composition of 203 red, composition of 202 salts, chromium in 477 steel, analysis of 116 tanning liquors, chromium in 477 vanadium steel, analysis of 116 yellow, analysis of 213 yellows, composition of 202 Chromium in chrome leather 478 tanning liquors 477 yellow 215 chromium acetate 477 salts 477 steel 122, 125 , lead chromate method for. . ., . . 215, 478 , permanganate method for 122 , peroxide method for 208, 215, 477 , volumetric method for 216, 477 ajid vanadium, double titration of. . 125 steel, analysis of 116 Chrysalis oil, constants of 232 , Japanese, constants of 232 Cigarette papers, composition of 103 tobacco, nicotine in 103 Cigars, nicotine in 103 Cinders for concrete 576 Citral in lemon extract 461 orange extract 461 , Chace method for 460 , f uchsin method for 460 , Hiltner method for 461 , metaphenylenediamine method for 461 Citrate-soluble phosphate 528, 530 solution, neutral, preparation of. ... 525 Clarifying reagents for sugar solutions . 398 Clark's scale of water hardness 513 Clay for paper filler, testing of 318 in paper 341 coating 344 , composition of 318 carrying capacity of casein 316 , China (See also Kaolin). 642 INDEX Clerget formula for sucrose 400 Cleveland open cup flash tester 255 Cloth (See also Fabrics, textile). , arsenic in 40 , fiber analysis of 377 , structural analysis of 372 Cloud point (See Cloud test). test of lubricating oils 256 Coal, B.t.u. in 176 , "H" value of . . .-. 179 , heating value of 176 , nitrogen in 183 , phosphorus in 184 , preparation of laboratory sample of 170, 172 , proximate analysis of 172 , sampling of 167 , sulfur in 174 , ultimate analysis of 180 ash, analysis of 185 refuse, analysis of 185 tar, crude, analysis of 548 colors in foods 389 products in soap (reference) 288 roofing pitch, analysis of 536 , specifications for 537 Coated paper (See Paper, coated). Coating on paper, analysis of 343 Cobalt blue, composition of 203 platinum method for color in water 497 Nessler standards 500 Cochineal solution, preparation of .... 100 Cocoa, analysis of 430 , artificial coloring matter in 391 butter, constants of 232, 437 in cocoa and chocolate 433 Cocoanut fat, constants of 232 Cod liver oil, elaidin test of 267 , constants of 232 Code rubber, analysis of 480 Coffee beans, artificial coloring in 391 berry oil, constants of 232 Coke, phosphorus in 184 Cold test of lubricating oils 257 Cologne earth (See Vandyke brown). spirits (See Alcohol ethyl) . Color of clay 318 tallow 269 ultramarine 336 vanilla extract 470 water 497 , platinum-cobalt method for 497 , artificial, in clay 319. lakes, separation of 389 test of nicotine 105 sand and gravel 580 Colorimetric method for carbon in iron. 131 iron.. . 513 Colorimetric method for titanium 128 Coloring, artificial, in clay 319 matter in foods 389 lemon and orange extracts. . . . 463 milk 448 , organic, in red lead 211 Colors, coal-tar, separation of 389 Colza oil (See Rape oil). Combustible matter in coal ash 185 Combustion apparatus for ultimate analysis 180 furnace, carbon, (Fig. 6) 108 , operation of 108, 180 Compressor oil, testing of 254 Concentration of tin ores 135 Concrete, analysis of 559 Condensed milk (See Milk, condensed). Confectionery, analysis of 409 , artificial coloring in 390 "Constants" of oils, fats and waxes. . . 232 Cooper, Peter, standard glues 321 Copper as copper oxide, determination of 119 , electrolytic, determination of 57, 145, 152, 404 in aluminum alloys 161 Babbitt metals 157 Bordeaux mixture 52 with Paris green 52 brass and bronze 145 nickel silver 152 Paris green 57 solder 157 steel 119 , thiosulf ate method for 58 method for oxycellulose 368 number (See Copper value). potassium chloride solution 130 reduction methods for sugars 402 steel, analysis of 116 value of cellulose 368 Cordage, chemical tests of 382 , fibers in 383 , sampling of 382 Corn oil (See Maize oil). Corrosive sublimate (See Mercuric chlo- ride). Cotton, copper value of 368 for nitrating, U. S. Navy Specifica- tion for 363 .joint powder specification for 363 in asbestos-cotton twine 385 cloth and yarns 378 , regain of 379 , sampling of 364 , surgical, copper value of 368 asbestos twine, analysis of 385 INDEX 643 Cotton cellulose for nitration, analysis of 363 fibers, stain for 337 hull fiber, potassium in 43 linters for nitration, analysis of .... 363 wool mixtures, analysis of 378 yarn, sizes of 375 Cottonseed oil, constants of 232 , Halphen test for 250 stearin, constants of 234 Coumarin in vanilla extract 468 , Hess and Prescott method for 468 Courses in hosiery 376 Cowles method for malic acid value. . . 427 Crane oil, testing of 254 Crank case oil, testing of 254 Cream, analysis of 449 Creosote, gypsy moth, analysis of 536 Croton oil, constants of 234 Crown filler in paper 340, 341 , analysis of 319 Crude fiber 393 in cocoa and chocolate 432 Crusher oil, testing of 254 Cube method for softening point of bitumens 542 Cudbear, dyeing test for 390 Cuprammonium silk 380 Cuprous oxide method for nitrogen (reference) 67 Curcas oil, constants of 234 Cutting compounds, analysis of 486 oil, testing of 254 "Cutting test" of casein 316 Cyanide, determination of 31 in case-hardening compounds 486 method for formaldehyde 80 nickel in steel 119 tin ore assaying 136 of potash (See Potassium cyanide). soda (See Sodium cyanide). Cyanides, analysis of 31 Cyanogen, determination of 31 Cylinder oil, testing of 254 Dairy salt, analysis of 24 , U. S. standard for 24 Dark lubricating oil, testing of 254 Dead oil of coal tar, analysis of 530 Defren-O'Sullivan method for reducing sugars 405 Degras, analysis of 274 , composition of 274 , constants of 234 , French 274 Dehydration of tar 551 Denaturant, methyl alcohol, analysis of 71 , nicotine, analysis of 103 Denaturant, special No. 4, analysis of .. ... 103 Denier system of silk numbering 375 Density (See Specific gravity). Dextrin, analysis of 91 in albumin, detection of 94 honey 422 textile fabrics 372 , removal of, by diastafor 372 Dextrose, Allihn method for 407 in dextrin 92 honey 422 leather 475 , Munson and Walker method for. . . 403 Diastafor method for dextrin, starch, etc 372 Diastase in honey 424 method for starch 433 Dielectric test of insulating varnish . . . 229 Dika fat, constants of 234 Dimethylglyoxime method for nickel . . 121, 153, 162 Dimethylparaphenylenediamine test for ground wood 339 Dimethylsulfate test for petroleum or asphalt products in tar 547 Dimethylsulfide in oil of peppermint . . . 465 Disc method for pulp sampling 291 Displacement method for specific gravity 539, 549 Distillation of carbolineum 531 coal-tar, crude 551 creosote, gypsy moth 536 gasoline 187 nicotine solution 103 road binders 546 turpentine 193, 196 water-gas tar, crude 551 wood-preserving oils 531 Dogfish oil, constants of 234 Dolomitic lime 324 Dolphin oil, constants of 234 Dram system of silk numbering 375 Dress goods, arsenic in 40 Dried milk (See Milk, dried). Drier, Japan (See Japan drier). Driers (See also Drying salts). in paint vehicles 206 varnish 218, 220 Drop .black, composition of 201 Drugs, arsenic in 39 Drying salts in Japan drier 221 test for insulating varnish 228 Japan drier 223 linseed oil 198 oil varnish 220 Ductility test of bituminous materials . 546 Dudley viscosity pipette 92, 323 644 INDEX Dyeing tests for artificial colors 390 on paper, comparative 307 Dyes for paper making, testing of 307 in foods 389 . red lead 211 , acid, behavior of 390 , aniline, testing of 307 , basic, behavior of 390 , coal-tar, separation of 389 , oil-soluble, test for 391 Dynamo oil, testing of 254 Egg albumin, analysis of 93 oil, constants of 234 Eggs, analysis of 387 Elaidin test on oils 266 Elazy oil, constants of 234 Elderberry oil, constants of 234 Electrolytic determination of ash in saccharine products 414 cadmium 141 copper 57, 145, 152, 404 lead 137, 145 z i nc 146, 153, 162 Elliott evolution method for sulfur. . . . 132 Elongation (See also Stretch). Emery test for beef fat in lard 253 Emulsions, overcoming of 261 Enamel paints, analysis of 200 Engine oil, testing of 254 Engler flask 187, 552 English scale of water hardness 513 test on paraffin wax 279 vermilion, composition of 202 Eschka method for sulfur 174 mixture 175 Esparto fibers, stain for 337 Ester value of beeswax 276 rosin 328 Esters in ethyl alcohol 74 vinegar 457 wood alcohol 73 Ether, acid-free, preparation of 330 , anhydrous, preparation of 440 extract (See also Fat and Oil). of butter 428 cheese 453 chocolate and cocoa 433 confectionery 417 cotton linters '. 365 feeds and grains 440 Ethyl acetate in ethyl alcohol 74 vinegar 457 alcohol (See Alcohol, ethyl). Evaporated cream 449 milk (See Milk, evaporated) . Evaporation test (See also Volatility). of coal-tar roofing pitch 537 Evolution method for sulfur 115, 132 Expansion coefficient of normal solu- tions 87 Extract, lemon, analysis of 458 , orange, analysis of 458 , peppermint, analysis of 467 , spearmint, analysis of 467 , sumac, analysis of 479 , tobacco, analysis of 100 , wintergreen, analysis of 467 , vanilla, analysis of 468 Extractor, rubber (Fig. 25) 481 Extracts, flavoring, coal-tar dyes in ... 389 Fabrics, textile, arsenic in 40 , fastness to ironing 376 , , light 375 , , mud spots 376 , , perspiration 376 , , washing 376 , , fiber analysis of 377 , , folding endurance of 374 , , moisture in 377 , , picks per inch 374 , , sizing materials in 372, 377 , , strength of 373 , , stretch of 373 , , structural analysis of 372 , , thread count of 374 , , weight of 375 , , weighting in 372 Fabris and Villavecchia tests for sesame oil 252 "Factor" of volumetric solutions 6 Factors, gravimetric, table of 608 , volumetric, table of 615 Fastness tests on fabrics 375 , ultraviolet light 375 Fat in butter 428 butter scotch 418 cheese 453 cocoa and chocolate 433 condensed or evaporated milk . . . 450 confectionery 417 cream 449 cylinder oils 259 eggs 389 feedstuffs 440 grains 440 greases 273 leather 474 milk 445, 446 soap 281 , Adams method for 450 , Babcock method for 446 , Roese-Gottlieb method for 418, 445 , bone, constants of 232 , butter, constants of 232 INDEX 645 Fat, carapa, constants of 232 , chaulmoogra, constants of 232 , chicken, constants of 232 , cocoanut, constants of 232 , dika, constants of 234 , goose, constants of 234 , hare, constants of 234 , horse, constants of 234 , human, constants of 234 , lard, constants of 234 , laurel, constants of 234 , macassar, constants of 236 , nux vomica, constants of 236 , rabbit, constants of 236 , sawarri, constants of 238 , ucuhuba, constants of 238 Fats, animal and vegetable, analysis of 230 , constants of (table) 232 Fatty acids, constants of (table) 233 , free, in cutting compounds 488 , , grease 272 , , soap 281 , Hehner number 243 , insoluble, in oils 243 , , volatile, in oils 244 , Polenske number 244 , Reichert-Meissl number 243 , soluble, in oils 242 , , soap (reference) 282 , , volatile, in oils 243 , titer test of 246 , total, in oils 242 , , soap 281 Fatty matter in soap 281 Fatty oil in sulfonated oil 265 Feder aniline chloride test for invert sugar 424 Feedstuffs, analysis of 438 Fehling's solutions (Allihn modifica- tion) 3, 407 (Soxhlet modification) 3, 403 Ferric chloride reagent 3 nitrate indicator 12 oxide (See Iron oxide). Ferrocyanide, Knublauch method for. . 557 , titration of, with ZnSO 4 557 sizing test for paper 354 Ferrous sulfate-zinc-soda method for nitrate nitrogen 69 Fertilizers, analysis of 524 , available organic nitrogen in 70 Fiber, crude 393 (total) in rope and twines 383 analysis of paper 337 , standard papers for 339 stain for paper analysis . . . . 337 Fibers in cloth and yarns 377 paper 337 Fibers in rope and cordage 383 Fiehe-Bryan test for invert sugar 423 Filler in paper 340 , retention of 342 Fincke method for formic acid 457 Fineness test of clay 318 Portland cement. 564 sand and gravel 577 Fir-seed oil, constants of 234 Fire point of carbolineum 531 lubricating oils 255 wood preserving oils 531 Fireproof metal polishes 488 Fish, smoked, artificial coloring in 391 oils in vegetable oils, test for 254 , bromide test for 254 Fixed weight system of yarn sizes 374 Flash point of carbolineum 530 lubricating oils 255 turpentine 218 varnish 218 wood preserving oils 530 test (See Flash point). tester, Cleveland 255 Flashing point (See Flash point). Flask, distillation, for nicotine analysis (Fig. 5) 103 , Engler 187, 552 Flavoring extracts, coal-tar dyes in. ... 389 Flax fiber, New Zealand 384 wax, constants of 234 Flexibility test for insulating varnish. . 228 Float test of bituminous materials. ... 541 Flotation test for grit in clay 318 Foam test for butter 429 Folding factor of paper 348 machine, Schopper (Fig. 17) 348 test of paper 347 textile fabrics 374 Folio size of paper 352 Foods, arsenic in 35, 39 , coloring matters in 389 , pigments in 389 , cattle, analysis of 438 Foodstuffs, analysis of 387 Formaldehyde, cyanide method for. ... 80 , Hehner test for 78 , hydrogen peroxide method for 79 , Leach test for 78 , morphine sulf ate test for 78 solutions, analysis of 79 test for casein 344 sesame oil 253 Formic acid, analysis of 81 in vinegar 457 , Fincke method for 457 , permanganate method for 81 Fractionation (See Distillation). 646 INDEX Freezing mixtures 257 French degrees of water hardness 513 Froth oil, testing of 254 Fruit juices, coal-tar dyes in 389 Fruits, canned, coal-tar dyes in 391 , preserved, coal-tar dyes in 391 Fuchsin method for aldehydes 75 . citral 460 nitrite standards. . . 505 Fuels, liquid, analysis of 190 Fuller's earth as a water turbidity standard 494 Fuming nitric acid (See Nitric acid, red fuming). sulf uric acid (See Oleum) . Furfural in ethyl alcohol 76 , phloroglucid method for 366 test for sesame oil 252 value of cellulose 366 cotton linters 366 Furnace, combustion carbon 108, 180 Fusel oil in ethyl alcohol 77 , permanganate method for 77 Galactan in feedstuffs and grain (refer- ence) 442 Galactose, Allihn method for 408 Gallon, grains per U. S 515 Galvanizing on iron and steel 163 test, Preece method 164 Garden cress oil, constants of 234 rocket oil, constants of 234 Gas black, composition of 201 engine oil, testing of 254 Gasoline in soap 286 , analysis of 185 , desirable properties of 185 , types of 186 test on cylinder oils 260 Gauze, corrosive sublimate in 62 Gear grease 271 Gelatin in albumin, qualitative test for . 94 milk, qualitative test for 448 silk 380 German degrees of water hardness. . . . 513 silver (See Nickel silver) . Ghedda wax (See Indian beeswax). Gillmore needles for cement testing (Fig. 28b) 568 Glucose by polarization at 87 C 416 in honey 423 leather 475 saccharine products 416 syrups 416 , qualitative test for 423 Glue, analysis of 320 in albumen, qualitative test for .... 94 344, 355, 356 Glue, in the presence of starch, test for 356 , Peter Cooper standard 321 , qualitative test for 94, 355 , tannin test for 94, 355 Glycerine (See also Glycerol). , analysis of 85 jacketed drying oven 173 soap 286 Glycerol, acetin method for 85 , bichromate method for 89, 285 , determination of 85 in lemon extract 459 orange extract 459 the presence of sugar 285 soap 285 lyes 85 vanilla extract '. 473 vinegar (reference) 454 potash saponification 247 Glyoxime method for nickel. .121, 153, 162 Golden ochre, composition of 203 sulfide of antimony 34 Goose fat, constants of 234 Gottlieb-Roese method for fat. . . .418, 445 Governor oil, testing of 254 "Grab method" for tensile strength... . 37 Grain alcohol (See Alcohol ethyl). Grains, mixed, analysis of 438 per U. S. gallon 515 U. S. gallon to pounds per 10,000 gallons 515 Grams per square inches to ounces per square yard, factor for 375 Grapeseed oil, constants of 234 Graphite, composition of 201 in greases 273 grease 271 Graphitic carbon in iron 131 steel 110 Gravel, color test of 580 , mechanical testing of 576 , organic matter in 580 for concrete 576 in mortar and concrete 559 Gravimetric factors, table of 608 Gravity (See also Specific gravity). of oils 230, 254 Baum6 and specific gravity, relation of 255 Gray acetate (See Calcium acetate). Grease in glue 323 Greases, analysis' of 270 , types of 270 Green pigments, classification of 203 pole painj, analysis of 207 Grinding mill for feedstuffs 438 Grit in clay or kaolin 318 INDEX 647. Grit in talc 335 , flotation test for 318 , sieve test for 318 Ground wood, qualitative tests for. . . . 338 pulp fibers, stain for 337 Gums in albumin, detection of 93 varnish 219 , Boughton method for (reference) . . 217 Gunning method for nitrogen 65 , modified for nitrates 67 Gutzeit-Sanger-Black method for ar- senic 37 Gypsum in talc 335 , analysis of 319 , composition of 201 , burnt, composition of 201 Gypsy moth creosote, analysis of 536 " H " value of coal 179 Haddock oil, constants of 234 Halphen test for cottonseed oil 250 Harder and Abrams color test for sand . 580 Hardness of water 510 , Clark's scale 513 , English degrees 513 , French degrees 513 , German degrees 513 , permanent 511 , soap method for 511 , temporary 511 Hare fat, constants of 234 Hazelnut oil, constants of 234 Heating test of tung oil 199 value of benzole acid 178 coal 176 liquid fuels 190 naphthalene 178 Hedges' method for arsenious oxide . . . .56 Hehner number of oils 243 test for formaldehyde 78 value (See Hehner number). Hemp fibers, stain for 337 Hempel column (Fig. 11) 193 Hempseed oil, constants of 234 , elaidin test of 267 Herle's solution for clarifying 399 Herring oil, constants of 234 Herzfeld formula for sucrose and rafn- nose 401 Hess and Prescott method for coumar- in and vanillin 468 Hide substance in leather 475 Hiltner method for citral 461 Honey, analysis of 420 , artificial 424 - , mineral adulterants in 414 Home method for clarifying sugar solu- tions 399 Horse fat, constants of 234 marrow, constants of 234 oil, constants of 234 Horsefoot oil, constants of 234 Hortvet and West method for methyl salicylate in wintergreen extract. . . . 468 Hosiery, testing of 376 Howard method for essential oil in mint and wintergreen extracts 467 Hubbard pycnometer method for specific gravity 538, 550 Human fat, constants of 234 Hydrocarbons, non-volatile, in soap . . . 287 , volatile, in soap 286 Hydrocellulose, caustic method for .... 365 in cellulose 365 cotton linters 365 Hydrochloric acid (See also Chlorides) . in formic acid 81 reagent 3 , volumetric solutions of 7, 8 Hydrocyanic acid (See also Cyanides). Hydrogen in coal 182 peroxide (See also Peroxide). method for formaldehyde 79 Hydrometer method for specific gravity 410, 538 Hydrosulfite (See also Sodium hydro- sulfite) method for indigotin 97 Hydroxide, qualitative test for 20 Hypochlorite of lime (See Bleach). Hyposulfite (See Thiosulf ate) . Ignition loss of cement 572 lime 325 talc 334 Incrusting solids in water 520 Index of refraction (See Refractive index) . Indian beeswax 277 red, composition of 202 Indicators, preparation of 12 Indigo, analysis of 97 , natural, impurities in 97 , sulfonation of 98 , synthetic 97 carmine solution, preparation of .... 96 Indigotin in indigo 97 , hydrosulfite method for 97 Ink, standard iron tannate 354 sizing test of papers 354, 358 Insoluble residue in cement 573 Insulating compound, rubber (See Rub- ber). varnish (See Varnish, insulating) . Insulation, asbestos-magnesia, analysis of 62 Inversion of sugar solutions 400 648 INDEX Invert sugar (See Sugar, invert). Iodides (See also Iodine) . Iodine, 0.1 N solution of 10 jelly test of tung oil 199 number of oils 241 rosin 225 shellac 225 , Wijs method for 225, 241 value (See Iodine number). Iron (See also Iron oxide). , analysis of 129 in alkalies , 22 aluminum alloys 162 sulf ate 304 brass and bronze 148 wa ter 513,517 zinc 139 , Jones reductor method for 148 , sampling of 106, 129 , sulfocyanate colorimetric method for 513 , cast, analysis of 129 , charcoal, analysis of 129 , galvanized, testing of 163 , pig, analysis of 129 , sheradized, testing of 163 , tinned, testing of 166 oxide in aluminum sulf ate 304 blanc fixe.. 310 boiler scale 523 chrome yellow 215 green pole paint 209 Portland cement 574 presence of phosphates 25 salt 25 talc 334 water 517 and alumina in lime 325 Ironing test on textile fabrics 376 Istle fiber (See Tampico). Ivory black, composition of . , 201 Jamba oil, constants of 234 Jams, analysis of 409 , artificial coloring in 390 Japan drier, analysis of 221 , P. & R. R. R. specifications for. 222 , U. S. Navy specifications for. . . 222 wax, constants of 234 Japanese wood oil (See also Tung oil) . , constants of 234 Javillier and Bertrand method for nico- tine 101 Jean method for free acid in leather .... 476 Jelly, analysis of 409 , artificial coloring in 390 test'of glue 321 tung oil 199 Jones reductor, use of 148 Journal grease, analysis of 270 oil, testing of 254 Juices, fruit, coal-tar dyes in 389 Jute fibers, stain for 337 Kainite, potash in 41 Kaolin, composition of 201 for paper filler, testing of 318 Kapok oil, constants of 234 Kennedy method for crude fiber 394 Ketones in crude wood alcohol (refer- ence) 371 Kissling method for nicotine 100 Kjeldahl method for nitrogen 64 , modified for nitrates 67 Gunning- Arnold method for nitro- gen 66 Knublauch method for ferrocyanide. . . . 557 Koeme oil, constants of 234 Koettstorfer number 241 value (See Koettstorfer number). Kokum butter, constants of 234 Kraft pulp (See Sulf ate pulp). Lactic acid in cheese 452 Lactose in cocoa and chocolate. . . .403, 434 condensed or evaporated milk . . . 450 milk 445 , Defren-O'Sullivan method for 405 , Munsen and Walker method for ... 408 Lakes, composition of 202 in foods 389 Lallamantia oil, constants of 234 Lampblack, composition of 201 Lard, Emery test for beef fat in 253 fat, constants of 234 oil, analysis of 265 , constants of 234 , grades of 265 Laurel fat, constants of 234 Leach method for coloring matter in milk 448 test for formaldehyde 78 Lead in aluminum alloys 161 Babbitt metals 157 blanc fixe, qualitative test for . . . 309 Bordeaux mixture and lead arsen- ate 54 brass and bronze 145 chrome yellow 215 greases 272 green paint 208 Japan drier 221 nickel silver 153 red lead 210 solder 156 type metals 159 INDEX 649 Lead in white lead. metals . . . 212 156 zinc 137 , chromate method for 47, 208, 212 , electrolytic method for 137, 145 , lead acid method for 138 , sulfate method for. . . 138, 145, 156, 214 , red, analysis of 210 , composition of 202 , white, analysis of 211 , composition of 201, 211 acetate in white lead 213 , dry basic, as clarifier 399 reagent 4, 399 solution, basic 4, 90, 399 acid, preparation of 138 method for lead 138 arsenate, analysis of 46 , composition of 46 with Bordeaux mixture, analysis of 54 carbonate, basic (See White lead) . chromate in chrome yellow 215 green paint 208 , scarlet, composition of 202 nitrate solution, basic 399 number of maple products 426 vanilla extract 470 , Winton method for. 426 oxide (See also Lead). in Bordeaux mixture with lead arsenate 54 lead arsenate 47 red lead 210 , water soluble, in lead arsenate . . 49 precipitate test for vinegar 455 soap in greases 272 sulfate in blanc fixe 310 Leather, analysis of 474 , chromium in 477 Lehner silk 379 Lemon chrome yellow, composition of. . 202 extract, analysis of 458 oil, analysis of 464 in lemon extract 460 peel color, Albrech test for 463 Levol method for tin ore assaying 136 Levulose in honey 422 , Allihn method for 408 Lichen colors, behavior on wool 390 Liebermann-Storch test for rosin. .219, 356 Light, fastness of fabrics to 375 Lime (See also Calcium oxide) , analysis of 324 , classification of 324 in aluminum sulfate 305 blanc fixe 310 boiler scale. .. . 523 Lime in concrete 560 lime 325 lime-sulfur solution 61 mortar 560 Portland cement 575 salt 26 satin white 333 sulfite acid 301 water 518 , specifications for 324 , volumetric permanganate method for 326, 518 , acetate of (See Calcium acetate). , dolomitic 324 , hydrated, grades of 324 , , specifications for 324 sulfur solution, analysis of 59 water reagent 3 and magnesia, sulfate method for . . 301 Limestone, analysis of 327 Limonite, composition of 202 Lindo-Gladding method for potassium 41 Linen fibers, stain for 337 yarn, sizes of 375 Linotype metals, analysis of 158 Linseed oil, analysis of 197 , constants of 197, 234 , drying test of 198 , elaidin test of 266 , boiled 197 , raw 197 Linters, cotton, analysis of 363 Liquors, distilled, coal-tar dyes in 389 Litharge (See Lead oxide). Lithopone, composition of 201 Litmus, dyeing test for 390 Loading in paper (See Filler). Loew's reagent for silk 381 Loewenthal-Proctor method for tannin 95, 479 " Long oil" varnish 220 Loom oil, testing of 254 Lubricating oils, testing of 254 Lustron silk : 380 Macassar fat, constants of 236 Machine oil, testing of 254 Machinery oil, testing of 254 Madia oil, constants of 236 Mafura tallow, constants of 236 Magnesia in asbestos-magnesia cover- ing 62 blanc fixe 310 boiler scale 523 lime 324, 326 mortar and concrete 560 paper ash 341 Portland cement . . . 575 650 INDEX Magnesia in salt 26 sulfite acid 301 water 518 and lime, sulfate method for 301 asbestos pipe covering, analysis of . . 62 , composition of 62 lime 324 method for ammoniacal nitrogen ... 68 mixture for phosphates in fertilizers . 526 reagent 4 wash solution 4, 518 Magnesium in aluminum alloys 162 ammonium chloride solution (See Magnesia mixture). carbonate in asbestos, magnesia cov- ering 63 oxide (See Magnesia) . sulfate, separation from calcium sul- fate 301 Maize oil, constants of 236 Malabar tallow, constants of 236 Malic acid in vinegar. 456 value of maple products, Cowles method 427 Malt extract, preparation of 433 Maltose, Munson and Walker method for 408 Manganese, bismuthate method for 110, 147, 161 , chlorate method for 148 in alloy steel 116 aluminum alloys 161 brass and bronze 147 high chrome steels 116 iron 131 Japan drier 222 steel 110 Manihot oil, constants of 236 Manila fibers, stain for 337, 383 , Swett test for 383 and sisal fibers, distinction between . 383 Maple cream, analysis of 425 products, analysis of 424 sugar, analysis of 425 syrup, analysis of 424 Margarine 427 Marine animal oils, bromine test for. . . 254 Marrow, beef, constants of 232 , horse, constants of 234 Marsh reagent for testing vanilla 471 test for arsenic 36 Massecuites, analysis of 409 Maumene test on oils 267 Mechanical moisture in crown filler .... 319 testing of sand and gravel for con- crete 576 Melting point (See also Softening point) . , closed capillary tube method for. 245 Melting point, drop method for 276 of beeswax 276 fats and oils 245 greases 271 paraffin wax 278 tallow 269 , open tube method for 271, 278 , titer test 246 , Wiley method for (reference) ... 246 Menhaden oil, constants of 236 Menthol in oil of peppermint 465 , structure of 466 Menthyl acetate in oil of peppermint. . 465 Mercuric bromide paper for arsenic test. 37 chloride in gauze 62 reagent 4 Alercury in zinc amalgam 163 , sulfide method for 62 Messinger method for acetone 72 Metal polishes, analysis of 488 Metals, type, analysis of 158 , white, analysis of 154 Methyl acetate in wood alcohol 73 alcohol (denaturant) , analysis of ... 71 in grain alcohol 78 lemon extract 462 orange extract 462 wood distillate (reference) .... 370 , methyl violet test for 462 , qualitative tests for 78, 462 , Riche and Bardy test for 462 orange indicator 12 red indicator 12 salicylate in wintergreen extract. . . . 468 violet test for methyl alcohol 462 Milk, analysis of .* 443 , composition of 443 , standard 443 chocolate, analysis of 430 , condensed, analysis of 449 , , standard 449 , evaporated, analysis of 449 fat in milk chocolate 437 , skimmed, standard 443 solids, composition of 438, 443 in condensed milk 451 milk chocolate 437 sugar (See Lactose) . " Mill test" of sulfite acid 302 'Mineral acid in oils 260 , free, in leather 475 matter (See also Ash). in greases 273, 274 oil in creosote, etc. (See Sulfonation residue). fatty oil mixtures 261 grease 274 paint vehicle . 206 INDEX 651 Mineral oil in sulfonated oils 264 weighting of textile fabrics . 372 Mitchell method for lemon and orange oils 460 Mohr's alkalimeter for carbon dioxide . . 212 Moisture by drying in vacuo without heat 224 upon asbestos 452 pumice stone 409 quartz sand 410, 425 in acetate of lime 32 albumin 94 alkalies 20 ammonium sulfate 525 asbestos cotton twine 386 asphalt road binders 538 bicarbonates 20 bituminous road binders 538 blanc fixe 309 Bordeaux mixture 50 butter 428 calcium sulfate, hydrated 319 case-hardening-compounds 484 casein 315 cheese 452 chocolate and cocoa 430 clay 318 coal 172, 173 coal ashes 185 condensed or evaporated milk . . . 450 cotton linters 364 crown filler 319 cutting compounds 486 dextrin 91 feedstuff s and grains 438 fertilizers 525 glue 320 greases 271 honey 420 indigo 97 lead arsenate 47 leather 474 maple sugar and syrup 425 massecuites 409 molasses 409 Paris green 55 potash salts 525 potassium carbonate 20 pulp 293 road binders 538 rope and twine 382 rosin size : 330 saccharine products 409 satin white 333 shellac 224 soap 280 soda ash 20 sodium bicarbonate. .. 20 Moisture in sodium nitrate 525 silicate 29 spent oxide 555 sugar 409 sulfonated oils 264 sulfur 17 syrups .- 409 tallow, qualitative test for 270 tar 550 textile fabrics 377 water glass 29 wood pulp 293 , xylol method for 27f Molasses, analysis of 409 , mineral adulterants of 414 Molecular weights, table of 586 Molybdate method for phosphates.525, 528 ; phosphorus 112, 114, 143 solution (See Ammonium molyb- date solution). Molybdenum in steel 127 , lead molybdate method for 127 , qualitative test for 127 steel, analysis of 116 Monotype metal, analysis of 158 Morphine sulfate test for formaldehyde 78 Mortar, analysis of 559 , cement, preparation of 565 , standard 1:3 567 Mowrah seed oil, constants of 236 Mud spot test of textile fabrics 376 Mullen test of paper (See Bursting strength) . tester for paper, Perkins (Fig. 18) . . 349 Munson and Walker method for reduc- ing sugars 403 Mustard oil, constants of 236 Mutton tallow, analysis of 268 -, constants of 236, 269 Myrtle wax, constants of 236 Naphtha, crude 186 in soap 286 metal polishes 488 Naphthalene, heating value of 178 Naphthol green B, behavior on dyeing. 390 a-Naphthol test for sucrose 417 a-Naphthylamine acetate reagent for nitrites 504 Neatsfoot oil, constants of 236 Needle, standard Roberts, for penetra- tion test 539 Nessler standards, platinum-cobalt. . . . 500 Nessler's reagent '. . . . 388, 499 Nesslerization method for ammonia . . . 501 New York- Liverpool test of alkalies. . . 21 New Zealand flax fiber 383, 384 Newcastle test of alkalies . . 21 652 INDEX Nickel, cyanide method for 119 , dimethylglyoxime method for 121, 153, 162 , hydroxide method for 151 in aluminum alloys 162 brass and bronze 151 nickel silver 153 steel 119 silver, analysis of 152 steel, analysis of 116 Nicotine, Bertrand and Javillier method for 101 ' in tobacco and extracts 100 , Kissling method for 100 , silicotungstate method for 101 flask (Fig. 5) 104 solution (denaturant), analysis of 103 Niger seed oil, constants of 236 Nitrates, absolute (CuO) method for (reference) 67 , ferrous sulfate-zinc-soda method for 69 , Gunning method for 67 in ethyl alcohol 79 fertilizers 525 sewage 507 water, phenolsulfonic acid meth- od for 505 , reduction method for 507 , Kjeldahl method for 67 , nitrogen in 69 , phenolsulfonic acid test for 79 , qualitative test for 525 , Ulsch-Street method for 68 , zinc-iron method for 68 Nitration, analysis of cotton for 363 Nitric acid reagent 4 , red fuming 4 test for ground wood 339 Nitrite of soda (See Sodium nitrite). Nitrites, fuchsin standards for 505 in water and sewage 504 , titration of 28 Nitrogen, absolute method for (refer- ence) 67 , azotometer method for (reference) . . 67 , cuprous oxide method for (refer- ence) 67 , ferrous sulfate-zinc-soda method for ? ... 69 , Gunning method for 65 in case-hardening compounds 485 coal 183 f eedstuffs and grains 439 fertilizers 64, 70, 525 leather 475 nitrate salts 68, 69 organic matter 64 sewage 498 Nitrogen in water 498 , Kjeldahl method for 64 Gunning- Arnold method for 66 , magnesia method for 68 , Ulsch-Street method for 68 , zinc-iron method for 68 , albuminoid 439, 502 , , in feeds 439 , , water 502 , ammoniacal in water 502 , amido 440 , ammoniacal, in sewage 501 , , water 498 , , magnesia method for 68 , and nitric 68 , nitrate, in sewage and water 505 , nitric and ammoniacal 68 , and organic 67 , nitrite, in sewage and water 504 , organic 64, 69 , , available 69 , , total, in sewage and water .... 503 , and nitric 67 , , total 67 Normal consistency of cement 566 solutions, temperature corrections for 13 , expansion coefficient of 87 Nut oil (See Tung oil). Nutmeg butter, constants of 236 Nux vomica fat, constants of 236 Ocher, golden, composition of 203 Ochers, composition of 202 Odor of water 493 Oil in boiler scale 522 cotton linters 365 cutting compounds 487 leather 474 paints 204, 206 paraffin wax 279 rope and twine 382 sulf onated oil 264 white lead 211 , unsaponifiable matter in 261 , acorn, constants of 232 , air compressor, testing of 254 , almond, constants of 232 , anchovy, constants of 232 , apricot kernel, constants of 232 , arachis (See Peanut oil). , automobile, testing of 254 , bear, constants of 232 , beechnut, constants of 232 , ben, constants of 232 , black, testing of 254 , bone, constants of 232 , bottlenose, constants of 232 INDEX 653 Oil, Brazilnut, constants of 232 , calender, testing of 254 , calophyllum, constants of "232 , cameline, constants of 232 , canari, constants of 232 , candlenut, constants of 232 , carbolineum, analysis of 530 , castor, analysis of 263 , , constants of 232, 263 , cedar nut, constants of 232 , cherry kernel, constants of 232 , laurel, constants of 232 , China wood (See Tung oil). , chrysalis, constants of 232 , , Japanese, constants of 232 , codliver, constants of 232 , , elaidin test of 267 , coffee berry, constants of 232 , colza (See Rape oil). , compressor, testing of 254 , corn (See Maize oil). , cottonseed, constants of 232 , , Halphen test for 250 , crane, testing of 254 , crank case, testing of 254 , creosote, analysis of 536 , croton, constants of 234 , crusher, testing of 254 , curcas, constants of 234 , cutting, testing of 254 , cylinder, testing of 254 , dark lubricating, testing of 254 , dogfish, constants of 234 , dolphin, constants of 234 , dynamo, testing of . 254 , egg, constants of 234 , elderberry, constants of 234 , elazy, constants of 234 , engine, testing of 254 , firseed, constants of 234 , fish, test for 254 , froth, testing of 254 , fusel, in ethyl alcohol 77 , garden cress, constants of 234 , rocket, constants of 234 , gas engine, testing of 254 , governor, testing of 254 , grapeseed, constants of 234 , haddock, constants of 234 , hazelnut, constants of 234 , hempseed, constants of 234 , , elaidin test of 267 , herring, constants of 234 , horse, constants of 234 , horsefoot, constants of 234 , jamba, constants of 234 , Japanese wood (See also Tung oil) . , , constants of 234 Oil, journal, testing of 254 , kapok, constants of 234 , koeme, constants of 234 , lallamantia, constants of 234 , lard, analysis 265 , , constants of 234 , , grades of 265 , lemon, analysis of 464 , , in lemon extract 460 , linseed, analysis of 197 , , constants of 197, 234 , , drying test of 198 , , elaidin test of 267 , loom, testing of 254 , lubricating, testing of 254 , machine, testing of 254 , madia, constants of 236 , maize, constants of 236 , manihot, constants of 236 , menhaden, constants of 236 , mineral, in creosote, etc. (See Sul- f onation residue) . , , fatty oil mixtures 261 , , grease 274 , , sulf onated oil 264 , mowrah seed, constants of 236 , mustard, constants of 236 , neatsf oot, constants of 236 , niger seed, constants of 236 , nut (See Tung oil). , olive, analysis of 266 , , constants of 236, 266 . , elaidin test of, 267 , kernel, constants of 236 -, orange, analysis of 464 , , in orange extract 464 , ostrich, constants of 236 , owala, constants of 236 , palm, constants of . 236 , nut, constants of 236 , peach kernel, constants of 236 , peanut, constants of 232 , , elaidin test of 267 , , tests for 250 , peppermint, analysis of 464 , , in peppermint extract 467 , perilla, constants of 236 , persimmon seed, constants of 236 , pistachio nut, constants of 236 , plum kernel, constants of 236 , poppy seed, constants of 236 , porpoise, constants of 236 , porpoise jaw, constants of 236 , pumpkin seed, constants of 236 , quince, constants of 236 , radish seed, constants of 236 , rape, constants of 238 , , elaidin test of 267 654 INDEX Oil, rape, Maumene test of 268 , ravison, constants of 238 , red, constants of 238 , rice, constants of 238 , rosin, constants of 238 , , test for 249 , saffron, constants of 238 .salad 266 , salmon, constants of 238 , sanguinella, constants of 238 , sardine, constants of 238 , seal, constants of 238 , senega root, constants of 238 t sesame, constants of 238 , , elaidin test of 267 , , tests for 252 , shafting, testing of 254 , shark, constants of 238 , sheepsfoot, constants of 238 , , elaidin test 267 , skunk, constants of 238 , sod (See also Degras). , soja bean, constants of 238 , spearmint, in spearmint extract . . . 467 , sperm, constants of 238 , , testing of 254 , spermaceti, constants of 238 , spindle, testing of 254 , sterculia, constants of 238 , sturgeon, constants of 238 , sulfonated, analysis of 263 , sunflower, constants of 238 , table :.. 266 , tallow, constants of 238 , tallowseed, constants of 238 , tea seed, constants of 238 , tobacco seed, constants of 238 , transformer, testing of 254 , tung, analysis of 198 , , constants of 232 , turbine, testing of 254 , turkey red, analysis of 263 , , constants of 238 , turtle, constants of 238 , tsubaki, constants of 238 , ungnadia, constants of 238 , unsaponifiable, in cutting com- pounds 487 , , greases 274 , , sulfonated oil 264 , valve, testing of 254 , virgin 266 , walnut, constants of 238 , whale, constants of 238 , , elaidin test of 267 , , testing of 254 , wheat, constants of 238 , wintergreen, in wintergreen extract 467 Oil, wood (See Tung oil). , wool, constants of 238 resisting test for insulating var- nish 229 soluble dyes in foods 391 .separation of 391 varnish (See also Varnish). , analysis of 217 Oils, constants of (table) 232 , heating value of 190 , sulfur in 191 , unsaponifiable matter in 261 , animal and vegetable, analysis of. . 230 , fish, bromine test for 254 , lubricating, testing of 254 , marine animal, bromine test for. . . 254 , mineral, testing of 254 , wood preserving, analysis of 391 Oleic acid, constants of 236 in oils 242 Oleomargarine, analysis of . . 427 , constants of 236 in butter 420 Oleum, analysis of 17 Olive kernel oil, constants of 236 oil, adulterants of 268 , analysis of 266 , constants of 236, 266 , elaidin test of 267 , Manmene test of 267 Orange extract, analysis of 458 mineral, analysis of 210 , composition of 202 oil, analysis of 464 in orange extract 460 peel color, Albrech test for 463 Ore, tin, analysis of 134 Organic coloring matter in red lead. . . 211 matter in blanc fixe, qualitative test ^ for ' 309 sand and gravel . 580 ultimate analysis 180 Ostrich oil, constants of 236 Ottawa sand (See Sand, standard Ottawa). Ounces per square yard from grams per square inch, factor for 375 Oven for moisture in coal 173 Owala oil, constants of 236 Oxalic acid, 0. 1 N solution of 8 Oxide, spent, analysis of 554 Oxycellulose, caustic method for 365 , copper method for 368 in cellulose 365 cotton linters : . . 365 Oxygen in coal 184 absorbed (See Oxygen consumed). consumed in sewage 507, 509 INDEX 655 Oxygen consumed in waters required (See Oxygen consumed). 507 Paint, green graphite pole, analysis of 207 pigments, classification of 201 Paints, mixed, analysis of 200 Palm nut oil, constants of 236 oil, constants of 236 Panning of tin ores 135 Paper, arsenic in 40 , breaking factor of 347 , length of 346 , bursting factor of 349 , ratio of 349 , strength of 349 , chemical analysis of 339 , cross direction of 345 , fiber analysis of 337 .filler in 340 , folding factor of 348 , test of 347 , glue in 355, 356 , machine direction of 345 , microscopic analysis of 337 , Mullen test of 349 , physical testing of 345 , ream weight of 350 , retention of filler in 342 , rosin in 356 , sizing in 355 , , detection of faulty 358 , (penetration) tests of 354 , starch in 356, 357 , strength factor of 349 , ratio of 349 , tests of 346 , stretch test of 347 , substance number of 352 , sulfur in 362 , dioxide in 342 , tarnishing test of 362 , tensile factor of 347 , strength of 346 , thickness test of 349 , anti-tarnish, testing of 362 , blotting, absorption test of 354 , cigarette, composition of 103 , coated, analysis of 343 , parchment, copper value of 368 , wall, arsenic in 40 , writing, sizing tests of 354, 358 testing machines 347, 348, 349, 350, 352 Papers, standard fiber, preparation of. 330 Para red lakes, composition of 202 rubber, 20% compound, analysis of. 480 Paraffin in beeswax 278 confectionery 419 Paraffin scale in bituminous materials (reference) 547 wax, analysis of 278 , constants of 236 Parchment paper, copper value of. ... 368 Paris green, analysis of 54 , composition of 54 with Bordeaux mixture, analysis of 52 white, composition of 201 Paste pigments, analysis of 200 Pat test on hydrated lime 324 Pauly silk 380 Peach kernel oil, constants of 236 Peanut oil, arachidic acid test for 250 , Bellier test for 252 , constants of 232 , elaidin test of 267 , Renard test for 250 Penetration test of bituminous mater- ials 539 paper 354, 358 Pentosans in cellulose 366 cotton linters 366 feeds and grains 442 Peppermint extract, analysis of 467 oil, analysis of 464 Perilla oil, constants of 236 Perkins' Mullen (bursting strength) tester (Fig. 18) 349 Permanganate (See also Potassium permanganate) . method for antimony 150, 155 calcium 326, 518 chromium in steel 122 formic acid 81 nitrites 28 sumac extract 479 tannins 479 vanadium in steel 123 Peroxide method for chromium 208, 215, 477 formaldehyde 79 sulfur 59, 482 titanium 128 Persimmon seed oil, constants of 236 Perspiration test on textile fabrics 376 Peter Cooper standard glues 321 Petroleum hydrocarbons in soap 287 products in tar, test for 547 Phenacetolin indicator solution 100 Phenol (See also Carbolic acid). Phenolphthalein indicator 12 Phenolsulfonic acid, preparation of. ... 505 method for nitrates 505 m-Phenylenediamine method for citral 461 Phloroglucid method for furfural 366 pentosans 366 656 INDEX Phloroglucinol, purification of 366 test for ground wood 338 Phosphate method for zinc. . .147, 216, 304 Phosphates (See also Phosphoric acid). Phosphoric acid (See also Phosphoric anhydride). in fertilizers 525, 528 vinegar 456 .molybdate method for, gravi- metric 525 , , volumetric 528 , citrate-insoluble 527, 530 , -soluble 528, 530 water-soluble 527, 529 anhydride (See also Phosphoric acid). in blanc fixe 311 salt 25 Phosphorus (See also Phosphoric anhy- dride). in alloy steel . 118 brass and bronze 143 coal and coke 184 iron 132 limestone 327 steel 112 , gravimetric molybdate method for 114 , reduction method for 25 , volumetric molybdate method for 112, 143 Phytosterol in fats 247 Phytosteryl acetate, melting point of . . 249 Picric acid test for gelatin 448 "Picking" test of coated papers 343 Picks per inch in textile fabrics 374 Pig iron, analysis of 129 , sampling of 106, 129 Pigment in greases 273 mixed paints . 204 Pigmented greases 270 Pigments, composition of 201 in foods, classification of 389 oil, analysis of 200 , black, classification of 201 , blue, classification of 203 , brown, classification of 203 , green, classification of 203 , red, classification of 202 , white, classification of 201 , yellow, classification of 202 Pipe covering, asbestos magnesia, analysis of 62 , , composition of 62 Pistachio nut oil, constants of 236 Pitch, asphaltic, analysis of 537 , coal-tar roofing, analysis of 536 Plant ashes, potassium in 41 Platinum, care of 14 chloride solution, preparation of 16 cobalt method for color of water 497 Nessler standards 500 residues, recovery of 16 wire method for turbidity of water. 495 Plum kernel oil, constants of 236 Plumbago (See Graphite). Ply yarn, sizes of 374 Polarizarion of animal and vegetable oils 250 honey 421 maple products 425 rosin oil 249 sugars. 397 vinegar . . . 455 Pole paint, green graphite, analysis of . . 207 Polenske number of oils 244 value (See Polenske number). Polishes, metal and brass, analysis of . . 488 , , composition of 488 Polymerization test of turpentine 196 Polysulfide sulfur in lime sulfur solu- tion 60 Poppy seed oil, constants of 236 Porpoise oil, constants of 236 jaw oil, constants of 236 Portland cement (See Cement, Port- land). Potash (See also Potassium and Potas- sium hydroxide). , alcoholic, . 5 N solution of 1 , caustic, analysis of 21 , water-soluble, in ashes 43 salts, moisture in 525 , potassium in 41 Potassium in fertilizers 41 kainite 41 plant ashes 41 potash salts 41 rocks 44 soils , 44 , Lindo-Gladding method for 41 , J. Lawrence Smith method for. ... 44 , qualitative test for 281 bicarbonate, analysis of 19, 22 in alkalies 20, 22 bichromate, analysis of 30 ,0.1 N solution 10 reagent 4 carbonate, analysis of 19, 22 in alkalies 20, 22 chromate indicator 12 cyanide, analysis of 31 f erricyanide reagent 4 f errocyanide in spent oxide 557 , Knublauch method for 557 reagent 4 INDEX 657 Potassium fluoride method for testing aluminum sulfate .105 hydroxide, analysis of 19, 21 in potassium carbonate 22 ,0.1 N solution of 8 permanganate, 0. 1 N solution of . . . 8 sulfocyanate, 0. 1 N solution of. ... 11 reagent 4 thiocyanate (See Potassium sulfo- cyanate). Pounds from grains per gallon, calcula- tion of 521 Pour test of lubricating oils 25G Preece test for galvanizing 164 Prescott and Hess method for couma- rin and vanillin 468 Preservatives in milk (reference) 448 Preserves, artificial coloring in 391 Proctor-Loewenthal method for tannins 95, 479 and Searle method for free acid in leather 476 Proof of alcoholic liquids 74 Protein in f eedstuffs and grains 439 milk 444 Proximate analysis of coal 172 Prussian blue, composition of 209 , extraction of, from spent oxide. . . 557 in green paint 209 Pulp (See also Wood pulp). , bleach consumption of 313 , sulfite, copper value of 368 Pumice stone method for moisture. . . . 409 Pumpkin seed oil, constants of 236 Pycnometer method for specific gravity. 230 , Hubbard, for specific gravity. 538, 550 Pyroligneous liquor, analysis of 368 Pyrox, composition of 52 Pyroxylin silk 379 Quartering method of sampling 167 Quartz sand method for moisture . . . 410, 425 Quicklime, specifications for grades of. 324 Quince oil, constants of 236 Rabbit fat, constants of 236 "Radiator" in water analysis 498 Radish seed oil, constants of 236 Raffinose in beet products 401 Rag in paper, detection of 338 fibers, stain for 337 soda standard papers 339 -sulfite standard papers 339 Rape oil, constants of 238 , elaidin test of 267 , Maumen6 test of 268 Ratio number of beeswax 276 Ravison oil, constants of 238 Reagents, preparation of 1 Ream weight of paper. 350 scales, Fairbanks (Fig. 20) 352 substance number table 353 Red lead, analysis of 210 , composition of 202 oil, constants of 238 pigments, classification of 202 Reducing substances in vinegar 455 sugars, determination of 396 in vinegar 455 Reduction method for nitrates 507 phosphorus 25 Reductor, Jones (See Jones reductor). Reference books 627 Refractive index of oils and fats 231 with Abbe refractometer 231 butyro-refractometer 240 , temperature correction for 240 Refractometer, Abbe, use of 231 , Zeiss butyro-, use of 240 Refuse (coal ash), analysis of 185 Regain of silk, wool and cotton 379 Reichert-Meissl number of butter fat . . 429 cocoa butter 437 fats 243 value (See Reichert-Meissl num- ber). Renard test for peanut oil 250 Resazurin indicator. 103, 105 Residue on evaporation (See also Solids). of water 497 Resorcinol test for invert sugar 423 Retention of filler in paper 342 Rice oil, constants of 238 Riche and Bardy method for methyl alcohol 462 Ring and ball method for softening point 543 Road binder, bituminous and asphaltic, analysis of 537 tars, analysis of 554 Roasting of tin ores 135 Roberts needle for penetration 539 Rocks, potassium in 44 Roese-Gottlieb method for fat 418, 445 Roofing pitch, coal-tar, analysis of. ... 536 Rope fibers, classes of 383 , microscopic examination of 384 Ropes, chemical tests of 382 , fibers in 383 , sampling of 382 Rosin, analysis of 327 , constants of 232 , grades of 327 in beeswax, test for 277 < paper 356 658 INDEX Rosin in rosin size ............... 329, 330 -- shellac ....................... 224 --- varnish .................... 227 -- soap 287 varnish, qualitative test for ..... 219 , iodine number of ................ 225 , Liebermann-Storch test for ---- 219, 356 , sizing tests of ................... 328 , Twichell method for ............. 287 i yaryan ....................... . 327 anhydride ..................... . . 331 oil, constants of .................. 238 -- , polarization of ................ 249 -- , qualitative test for ............ 249 -- grease ........................ 270 size, analysis of .................. 329 -- , preparation of, from rosin ...... 329 -- milk, analysis of ............... 332 Rotation, optical, of lemon and orange oils ........................... 464 , specific, of peppermint oil ........ 464 Rubber, code, analysis of ............ 480 extraction apparatus (Fig. 25) ...... 481 insulation compound, Underwriters' Laboratory method for an- alysis of ................. 480 --- , 20% para, analysis of ........ 480 Saccharimeter ...................... 397 , Ventzke, angular degrees from ..... 465 Saccharine products, analysis of ...... 408 Sachsse method for starch ............ 442 Saffron oil, constants of ............. 238 Salad oil ........................... 266 Salicylate, methyl (See Methyl salicylate). Saliva method for starch in paper ...... 357 Salmon oil, constants of .............. 238 Salt (See also Sodium chloride). in butter and substitutes ......... 428 -- cheese ........................ 452 -- soap ............ ............. 285 , U. S. Standard for ............... 24 , dairy, analysis of ................ 24 , table, analysis of ................ 24 Sampling butter .................... 427 cheese ..................... . ____ 451 coal ............................ 167 cotton (linters) ............... .' . . 364 fertilizers ....................... 524 iron ......................... 106, 129 lime ............................ 324 Portland cement ................. 563 rope and twine ......... ......... 382 rubber ..... . ................... 480 shellac ........................... 223 soap ............................ 279 steel ........................ 106, 107 Sampling tar 548 white metals 154 wood pulp, A. P. and P. A. and A. A. W. P. I. method 293 , Little method 290 zinc 137 , alternate shovel method of 167 , quartering method of 167 Sand, color test for 580 for concrete 576 in mortar and concrete 559 , mechanical testing of 576 , organic matter in 580 , standard Ottawa 571 method for moisture 410, 425 Sanger-Black-Gutzeit method for ar- senic 37 Sanguinella oil, constants of 238 Sanitary analysis of water and sewage . . 492 Saponifiable oil in mixed oils 259 Saponification of wool grease 275 under pressure 260 with aid of benzene 259 number of oils 241, 259 rosin 328 wool grease 259 value (See Saponification number). Sardine oil, constants of 238 Satin white, analysis of 332 , composition of 332 in paper coating 344 Sausage, artificial coloring in 391 Sawarri fat, constants of 238 Saybolt universal viscosimeter 258 Scale, boiler, analysis of 522 forming solids in water 520 Scarlet lead chromate, composition of . 202 Schlippe's salt 34 Schopper paper testing machines (Figs. 16, 17) 347, 348 Schweitzer's reagent for silk 381 Screening test (See Sieve test). Screens for sand and gravel 577 Seal oil, constants of . 238 Searle and Proctor method for free acid in leather 476 Sediment in water 493 Senega root oil, constants of 238 Sesame oil, Baudouin test for 252 , Bellisr test for 253 , constants of 238 , elaidin test of 267 , formaldehyde test for 253 , furfural test for 252 , sugar test for 252 , Villavecchia and Fabris test for 252 Setting test of Portland cement 568 tung oil 199 INDEX 659 Sewage, analysis of 402 Shafting oil, testing of 254 Shark oil, constants of 238 Shea butter, constants of . 238 Sheepsfoot oil, constants of 238 , elaidin test of 267 Shellac, analysis of 223 , iodine number of 225 , sampling of 223 , bleached, analysis of 224 , , forms of 223 , orange, analysis of 224 varnish, analysis of 226 , preparation of 223 Sheradizing on iron and steel 164 , testing of 163 "Short oil" varnish 220 Sienna, burnt, composition of 203 , raw, composition of 203 Sieve test of bone and tankage 524 clay or kaolin 318 gravel and sand 577 Portland cement 564 Sieves, specifications for 565, 579 Silex, composition of 201 Silica in barium sulfate, qualitative test for 309 black sulfate liquor 295 boiler scale 523 green pole paint 209 Portland cement 574 sodium silicate 29 water 517 white sulfate liquor 298 dishes for evaporation of vinegar . . . 484 standard for turbidity of water .... 494 Silicate of soda (See Sodium silicate) . Silicious matter in boiler scale 523 Silicon in aluminum alloys 160 iron 132 steel 114 tungsten steel 126 mixture 115 Silicotungstate method for nictotine. . 101 Silk, basic zinc chloride method for. . . 377 in cloth and yarns 377 , numbering of 375 , regain of 379 , acetate 380 , artificial, varieties of 379 , and natural, distinction be- tween 379 , celestron . 380 , cellulose acetate 380 , Chardonnet 379 , cuprammonium 380 , gelatin 380 , Lehner 379 Silk, lustron 380 , Pauly 380 , pyroxylin 379 , raw, sizes of 375 , spun, sizes of 375 , viscose 380 Silver carbonate, preparation of 89 nitrate, 0. 1 N solution of 11 reagent 4 solution, ammoniacal 299 test for aldehyde in alcohol 75 blanc fixe 309 Sirup (See Syrup). Sisal and Manila fibers, distinction between 383 Size of yarn, determination of 374 , animal (See Glue). , rosin, analysis of . . . 329 , , preparation of 329 Sizing in paper, faulty, detection of ... 358 materials in cloth and yarns. . . . 372, 377 paper 355 test of paper .354, 358 rosin 328 Skunk oil, constants of 238 Slag for concrete 576 Smith (C. M.) method for arsenious oxide 57 (J. Lawrence) method for potassium 44 Soap, analysis of 279 , glycerine 286 in cutting compounds 487 greases 272 metal polishes 489 tallow 270 , transparent 286 lyes, glycerol in 85 method for hardness of water 511 solution, standard 511 thickened mineral oil grease 270 Sod oil (See Degras). Soda (See also Sodium oxide). ash (See also Sodium carbonate) . , analysis of 20 lime for carbon dioxide absorption : . 109 pulp fibers, stain for 337 rag standard papers 339 Sodium in water 518 , qualitative test for 280 , total, in black sulfate liquor 296 arsenite, 0.1 N solution of (See Arsenious acid.) biborate (See Borax). bicarbonate, analysis of 19, 22 , conversion to carbonate 7 in alkalies 20, 22 bichromate, analysis of 30 carbonate, analysis of 19, 22 662 INDEX Sucrose in lemon and orange ex- tracts 462 maple products 426 saccharine products 415 vanilla extract 471 , optical methods for 397 solutions, specific gravity of 410 Sudan G, separation of 392 I, separation of 392 II, separation of 392 III, separation of 392 IV, separation of 392 Sugar (See also Sucrose). , a-naphthol test for 417 , determination of 396 in albumin, detection of 94 soap 286 , polarization of 397 , invert, determination of 406, 408 , , Feder aniline chloride test for. . . 424 , , Fiehe-Bryan test for 423 , , in honey 423 , , maple products 425 , -, resorcinol test for 423 , , volumetric method for 406 , maple, analysis of 424 , raw, analysis of 396 , reducing, Allihn method for 407 , , by copper reduction 402 , , Defren-O' Sullivan method for.. 405 , , determination of 402 , , in cattle foods 441 , , vinegar 455 , , Munson and Walker method for. 403 , total, in cattle foods 440 test for sesame oil 252 Sugars in vinegar 455 Sulfate of alumina (See Aluminum sul- fate). ash of albumin 94 process for wood pulp . 294 pulp cook liquor, analysis of 294 Sulfates (See also Sulfur trioxide). in alkalies 22 lime sulfur solution 61 sodium silicate 30 Sulfide in sodium sulfide 28 of antimony (See Antimony sulfide). sulfur in lime sulfur solution 60 test of tinned iron and steel. 166 Sulfides (See also Hydrogen sulfide). , sulfites, and thiosulfates, determi- nation of 299 Sulfite acid, analysis of 301 , composition of 301, 302 fuchsin method for aldehydes. . .75, 460 pulp, copper value of 368 fibers, stain for 338 Sulfite-rag standard papers 339 Sulfites (See also Sulfur dioxide). Sulfocyanate, 0. 1 N solution of 11 Sulf onated oil, analysis of 263 , preparation of 263 Sulf onation of indigo 98 residue in carbolineum 533 dead oil of coal tar 533 , Western Elec. Co. method for.. . 533 Sulfur (See also Sulfur trioxide). as polysulfide 60 , analysis of 17 , calorimetric method for 191 , Elliott method for 132 , Escha method for 174 , evolution method for 115 in alcohol 79, 191 alloy steel 118 antimony sulfide 34 coal 174 foods 395 fuel oil 190 gasoline 190 glue 321 iron 132 lime sulfur solution 59 liquid fuels 191 paper 362 rubber compounds 482 spent oxide 555 steel 115 , sodium peroxide method for .... 59, 482 , active, in glue 321 , , paper 362 , free, in antimony sulfide 34 , , rubber compounds 483 , total, in rubber compounds 482 compounds in alcohol 79 , hydrogen peroxide test for 69 dioxide, distillation method for .... 395 in bisulfites 301 foods 395 glue 321 paper 342 . sulfite acid 301 trioxide (See also Sulf uric acid). in aluminum sulfate 304 boiler scale 523 chrome yellow 216 oleum 17 Portland cement 573 salt 26 satin white 333 sodium silicate 30 sulfonated oils 264 water 519 Sulfuretted antimony (See Antimony sulfide). INDEX 663 Sulfuric acid in aluminum sulfate 304 formic acid 81 glue 321 leather 475 sulfite acid 301 , free, in aluminum sulfate 305 , , leather 475 , fuming (See Oleum). reagent 5 , 38 normal 197 " Sulfuring " of foods 395 Sulfurous acid (See also Sulfur di- oxide). in glue 321 paper 342 sulfite acid 301 Sumac extract, analysis of 479 Sunflower oil, constants of 238 Surgical dressings, mercuric chloride in. 62 Suspended matter in water. , 493, 516 Sweeney method for crude fiber 394 Swett test for Manila fiber 383 Syrup, coal-tar dyes in 389 , mineral adulterants in 414 , maple, analysis of 424 , sugar, analysis of 409 Table oil, ordinary 266 salt (See Salt). Talc, analysis of (for paper filler) 334 , composition of 334 in paper 341 Tallow, analysis of 268 , beef, constants of 232 , , in lard, Emery test for 253 , Borneo, constants of 232 , Mafura, constants of 236 , Malabar, constants of 236 , mutton, constants of 236, 269 , vegetable, constants of 238 grease 270 oil in cylinder oil 259 , constants of 238 seed oil, constants of 238 Tampico fiber 384 Tankage, analysis of 524 Tannate solution for color test of sand . . 581 Tannic acid (See also Tannin). , analysis of 95 in tannins 96 Tannin in sumac extract 479 , Loewenthal-Proctor method for . 95, 479 , permanganate method for 95, 479 test for glue 94, 355 Tanning liquor, chrome, CroOs in 477 Tannins in tannic acid 95 Tar, asphalt products in 547 in cylinder oils 260 Tar in pyroligneous liquor 369 spent oxide 556 wood distillate 369 , petroleum products in 547 , sampling of 548 , coal, analysis of 548 , , in soap (reference) 288 , road, analysis of 537 , water-gas, analysis of. . 548 acids in carbolineum 534 crude tar 554 gypsy moth creosote 536 wood preserving oils 534 , Western Elec. Co. method for . . 534 Tarnishing test for paper 362 Tea seed oil, constants of 238 Temperature coefficient of normal solu- tions 87 corrections for Brix gravities 413 normal solutions 13 refractive index of oils and fats. 240 specific gravity of oils and fats . 231 Tensile factor of paper 347 machine, Schopper (Fig. 16) 347 , cement (Fig. 29) 570 strength of paper 346 Portland cement 569 sand and gravel 579 textile fabrics 373 Terra alba (See also Gypsum). , analysis of 319 , composition of 201 Textiles (See also Fabrics, textile). , arsenic in 40 , fibers in 377 , structural analysis of 372 Theobromine in cocoa and chocolate . . 435 Thermometer for gasoline distillation. 189 tar distillation 552 titer test 246 stem correction 194 Thickness of paper 349 gauge for paper (Fig. 19) 350 "Thief" for tar sampling 549 Thinner of paints, separation of 205 Thinners in Japan drier 221 varnishes 218 " Thio " (See Sodium thiosulf ate) . Thiosulfate (See also Sodium thiosulf ate) . in lime sulfur solution 61 method for copper 58 , sulfite and sulfide, determination of . 299 Thread, twists per inch in 374 count of textile fabrics 374 number 374 size 374 Tin in aluminum alloys 161 Babbitt metals l. r >7, 158 664 INDEX Tin in brass and bronze 144 solder 154 in tin ores 134 type metals 158 white metals 154 , gravimetric method for 144, 154 , iodine volumetric method for 155 ores, assay of 134 oxide, natural 134 , purification of 144, 154 , test for 134 -stone 134 Tinning test, Am. Elect. Ry. Eng. Assoc. method for 166 , Am. Tel. and Tel. Co. method for 166 Titanium, colorimetric method for 128 in steel 128 , peroxide method for 128 , qualitative test for 123, 128 steel, analysis of 116 sulfate, preparation of standard.. . . 129 Titer test of fatty acids 246 Tobacco, analysis of 100 , nicotine in 103 extract, analysis of 100 seed oil, constants of 238 Toluene-benzene method for free car- bon 551 Transformer oil, testing of 254 Tsubaki oil, constants of 238 Tung oil, analysis of 198 , constants of 198, 232 Tungsten in steel 126 steel, analysis of 116 Turbidity of water 494 , platinum wire method for 495 , standard of 494 Turbine oil, testing of 254 Turkey red oil (See also Sulf onated oil) . , constants of 238 Turmeric, dyeing test for 390 in lemon extract, test for 463 orange extract, test for 463 Turpentine, analysis of, electric rail- way specifications 196 , , Forest Products Laboratory method 193 in paints 206 varnish 218 , temperature expansion coefficient of 194 test for artificial color in clay 319 Turtle oil, constants of 238 Tuscan red, composition of 202 Twichell method for rosin 287 Twine, asbestos cotton, analysis of. ... 385 Twines, chemical tests of 382 , fibers in 377, 383 .sampling of 382 Twist in thread 374 Type metals, analysis of 158 Ucuhuba fat, constants of 238 Ulsch-Street method for nitric and ammoniacal nitrogen 68 Ultimate analysis of coal 180 Ultramarine, composition of 203 in paper 340 , resistance to alum 336 , testing of 336 Ultraviolet light test for fabrics 375 Umber, burnt, composition of 203 , raw, composition of 203 Underwriters' Laboratory method for analysis of rubber compound 480 Ungnadia oil, constants of 238 Unsaponifiable matter in oils 261 rosin : . 328 soap 286 sulf onated oils 264 oil in greases 274 Unsaponified oil in greases 273 Uranium acetate test for sodium 280 Vacuum moisture method 224, 438 Valve oil, testing of 254 Vanadium in steel 123, 125 , permanganate method for 123 , qualitative test for 123 and chromium, double titration of. . 125 steel, analysis of 116 Vandyke brown, composition of 203 Vanilla extract, analysis of 468 , U. S. standard for 473 resins, tests of 472 Vanillin in vanilla extract 468 , Hess and Prescott method for 468 Varnish, black insulating, analysis of. . 227 , , specifications for 227 , interior, U. S. Navy specifications for. 220 , oil, analysis of 217 , , composition of 218 , , physical tests of 220 , , short or long oil test of 220 , shellac, analysis of 226 gums 218 Vegetable oils and fats, analysis of .... 230 , constants of (table) 232 tallow, constants of . 238 Vegetables, canned, coal-tar dyes in. . . 391 Vehicle of paints, analysis of 204 , separation of 204 in white lead paste 211 Venetian red, composition of 202 Ventzke saccharimeter, angular degrees from 465 , polarization with , 397 INDEX 665 Vermilion, American, composition of . . 202 , English, composition of 202 Vicat apparatus for cement testing (Fig. 27) 566 Villavecchia and Fabris tests for ses- ame oil 252 Vinegar, analysis of 454 , coal-tar dyes in 389 , standards for 458 , light, analysis of 370 Virgin oil 266 Viscose silk 380 Viscosimeter, Saybolt universal, opera- tion of 258 Viscosity of dextrin solutions 92 glue solutions 323 lubricating oils 258 varnish 218 with Dudley pipette 92, 323 Volatile acids in glue 320 vinegar 457 matter in coal 173 thinners in Japan drier 221 oil varnish 218 Volatility (See also Evaporation test). of bituminous materials 540 Volumetric solutions, preparation of . . . 6 , standardization of 6 , table of equivalents of 615 , temperature corrections for .... 13 Wales in hosiery 376 Wall paper, arsenic in 40 Walnut oil, constants of 238 Washing test on textile fabrics 376 Water (See also Moisture). , clarification of, by aluminum hy- droxide 510 , mineral analysis of 514 , sanitary analysis of 492 , weight per gallon of 191, 515 , boiler, analysis of 514 , , grading of 521 , industrial, analysis of 514 :, manufacturing, analysis of 514 equivalent of calorimeter, deter- mination of 178 extract (See Water-soluble matter). gas tar, analysis of 548 glass (See Sodium silicate). soluble matter in case hardening compounds 485 lead arsenate 48 leather 474 ropes and twines 382 Wax, carnauba, constants of 232 , Ghedda (See Ghedda wax). , flax, constants of 234 Wax, Japan, constants of 234 , myrtle, constants of . . 236 , paraffin (See Paraffin wax). Waxes, constants of (table) 232 Weight of textile fabrics 375 , ream, of paper 350 Weighting in textile fabrics 372 Western Electric Co. method for car- bolineum 530 Westphal balance, use of 230, 255, 549 Whale oil, constants of 238 , elaidin test of 267 , testing of 254 Wheat oil, constants of 238 White lead, analysis of 211 , basic carbonate, composition of 201, 211 , sulfate, composition of 201 liquor (sulfate), analysis of 298 metals, analysis of 154 pigments, classification of 201 Whiting in paper 341 , alba (See Alba whiting). , composition of 201 Wijs method for iodine number 241 solution, preparation of 5, 242 , special for shellac analysis 225 Wiley method for melting point (refer- ence) 246 Wine, coal-tar dyes in 389 , sulfur dioxide in 396 Wintergreen extract, analysis of 467 oil in wintergreen extract 467 Winton method for lead number 426 Wire, galvanized, testing of 163 , tinned steel, testing of 166 Wood, cellulose in 289 acid, crude, analysis of 370 alcohol (See also Methyl alcohol). , analysis of 71 in crude wood liquor 369 pyroligneous liquor 369 distillate products, analysis of 368 liquor, crude, analysis of 368 oil (See Tung oil). preserving oils, analysis of 530 pulp, bleach consumption of 313 , copper value of 368 in bales, sampling of 291 rolls, sampling of 291 laps, sampling of 291 , sampling of, disc method 291 , .official method of A. P. and P. A. and A. A. W. P. 1 293 , , strip method 291 , and testing of, Little method. 291 1 ground (See Ground wood pulp). 666 INDEX Wood pulp, mechanical (See Ground wood pulp). fibers, stain for 337 spirit in ethyl alcohol 78 Wool in cloth and yarns 378 , regain of 379 cotton mixtures, analysis of 378 dyeing test for colors in foods 389 grease in cylinder oil , . . 259 , crude (See Degras) . , saponification of 275 oil, constants of 238 Woolen yarn, sizes of 375 Worsted yarn, sizes of 375 Writing papers, sizing tests of 354, 358 Wuensch method for free acid in leather (reference) 477 Wuerster's reagent for ground wood. . . . 339 Xylene method for specific gravity .... 335 Xylol (See also Xylene). method for water 271 Xylose, Allihn method for 408 Yarn, count number of . . . 374 374 fibers, microscopic examination of. . 384 number 374 Yarns, fibers in ,. 377 Yaryan rosin 327 Yellow ochers, composition of 202 pigments, classification of 202 Zeiss butyro-refractometer, use of 240 Zinc, analysis of 137 , electrolytic determination of 146, 153, 162 , grades of 137 in aluminum alloys 162 sulfate 304 Babbitt metals . 157 brass and bronze 146, 151 Japan drier 222 nickel silver 153 solder 157 soldering paste 491 zinc dust 141 , phosphate method for. 147, 216, 304, 491 , sampling of 137 , amalgamated, preparation of 149 , available, in zinc dust 141 amalgam, analysis of 163 chloride solution, basic 377 iodine fiber stain 337 dust, analysis of 141 , bichromate method for 141 iron method for nitrogen 68 oxide in chrome yellow 216 white, composition of 201 MINERAL TEGHNGLbbY LIBRARY UNIVERSITY OF CALIFORNIA LIB] BERKELEY Return to desk from which borrowed. This book is DUE on the last date stamped below. LD 21-100m-9 > '48(B399sl6)476 460150 UNIVERSITY OF CALIFORNIA LIBRARY