THE- OPTICAL ROTATING POWER OF Organic Substances, and Its Practical Applications BY '' ' J '^ DI DOCTOR HANS^AjroOL'f 5 # PROFESSOR OF CHEMISTRY IN THE UNIVERSITY OF BERLIN ASSISTED BY DR. 0. SCHONROCK, DR. P. LINDNER, DR. F. SCHUTT, DR. L. BERNDT, and DR. T. POSNER UNlVERSn SKCOXD EDITION OF .CALII- AUTHORIZED ENGLISH TRANSLATION WITH ADDITIONS BY DR. JOHN H. LONG, PROFESSOR OF CHEMISTRY IN NORTHWESTERN UNIV., CHICAGO ILLUSTRATED EASTON, PA.: THE CHEMICAL PUBLISHING CO. i got COPYRIGHT, 1902, BY EDWARD HART. PREFACE TO THE FIRST EDITION Although the subject of the optical rotating power of organic substances, in theoretical as well as in practical relations, has been growing in importance for a. long time, chemical litera- ture possesses thus far no work w^bkrir 'O'^erts a complete dis- cussion of the whole field, and studies in this lire may be car- ried out only by the aid of articles "Scattered t*i:MdgK t'i<5 jour- nals. I have attempted in the present book to supply the want in this direction, and the incentive to the work was fur- nished by an article which I published some time since in Lie- big's Annalen der Chemie, Vol. 189. This article dealt with the determination of the specific rotation of solid substances, and was introduced by a short general discussion of optical activity. From several quarters I was urged to enlarge the article, and, especially by the addition of a description of all the new polarization instruments and the practical uses which may be made of them, to work up a monograph as complete as possible of the subject of optical rotation. Such an under- taking appears all the more inviting since in the last few years different observers have carried out investigations in this field which have increased our knowledge considerably, and which make it possible to give a certain degree of completeness to the treatment of the subject. From a theoretical standpoint the optical activity of organic substances possesses this great interest, that it is a consequence of a peculiar arrangement of the atoms within the chemical molecule, and therefore stands in close relation to the question of constitution. However, we are just at the beginning of investigations of the relations between rotating power and chemical structure, and for future study there is material at hand abundant and full of promise. To lessen the labor in work of this kind, especially in such cases where results may be obtained only by aid of exact measurements, it became nee- 102026 iv PREFACE essary to go fully into the discussion of the methods for the determination of specific rotation. Therefore, the use of the different kinds of polarization instruments called for detailed treatment, and besides this the determination of other experi- mental data, the specific gravity, for example, should be ex- plained. In this work I have taken pains to give methods of the greatest possible exactness and to indicate always the limits of accuracy which may be reached in the numerical results. If for any special purpose less rigid care is permissible in work the observer will see for himself how the procedure may be simplified^ - ^ U J :| t In a pfactical directionvas is well known, optical rotation has loa^ .Til mikoportent application in the determi- nation of sugar ; 'arid recently the optical analysis of other sub- stances, especially that of the cinchona alkaloids, has been also developed. The methods applied in such cases are fully dis- cussed in the book ; for the sugar chemist the description of the various saccharimeters and the corrections which must be applied in using them may be of interest, and in part new. The optical introduction, which possibly may be of interest to many chemists, has been kept as concise and elementary as possible. I have also touched but briefly on the relations be- tween crystalline form and rotating power, as this topic ap- pears to belong in the field of physical crystallography. Finally, the table of contents gives full information concern- ing the subjects treated. HANS LANDOLT. A \t HKX, January, 1879. PREFACE TO THE SECOND EDITION The first edition of this work, which appeared in 1879, pre- sented a general view of our knowledge of optical rotation as it existed at that time. Now, after the lapse of eighteen years, when a new edition corresponding to our present position is called for, we have to deal with a task of entirely different di- mensions. The progress which has been made in the last two decades in the field of optical activity rests, so far as the theo- retical side is concerned, mainly upon the great interest aroused among chemists by the hypothesis of van't Hoff and LeBel on the relation between rotating power and the atomic structure of carbon compounds. Since 1879, when this doctrine was still in its infancy, numerous investigations suggested by it have been carried out, the results of which have abundantly confirmed the theoretical requirements in all cases, so that to- day the theory may be presented in complete and fully devel- oped form. A marked widening of our knowledge may be observed in other directions also ; for example, with reference to multirotation, the causes of variation in specific rotation, rotation dispersion, etc. In a practical direction progress has been made in the improvement of polarization apparatus and in the development of methods of optical analysis. In addi- tion, the number of optically active substances known has in- creased since 1879 from 300 to over 700. From the whole range of material now available I have worked up certain parts only myself, which are in brief, the subjects discussed in the following theoretical portions of the book : Part I. General Conditions of Optical Activity. Part II. Physical Laws of Circular Polarization. Part III. Numerical Values for the Rotating Power. Spe- cific Rotation. In a section of Part I dealing with the decomposition of ra- VI PREFACE cemic bodies by fungi I have received valuable assistance from Herrn Prof. Dr. P. Lindner, Department Director of the In- stitute for Fermentation Industries in Berlin ; he has had the kindness to prepare a chapter on the subject of the fungi and the forms suitable for resolution, the methods of pure culture and experiments in resolution. I am no less indebted to Dr. W. Marckwald and Prof. Dr. H. Traube, who have rendered me assistance in many questions of stereochemistry and crys- tallography. In order to secure a more rapid completion of the following parts of the book I was obliged to seek the cooperation of others, and to the extent as now to be explained : Part IV. ' ' Apparatus and Methods for the Determination of Specific Rotation," by Dr. O. Schonrock, Assistant in the Physikalisch-Technischen Reichsanstalt. This section pre- sents, first, a description of the different polariscopes and saccharimeters, which in the last few years, and especially through the work of Lippich, have reached such a degree of perfection that they may now be classed among the most ex- act of instruments for physical measurements. In order to understand these instruments and the methods of using them, it was necessary to discuss, not only their construction, but also, especially, the optical theory on which their use is based, and it was further necessary to go into an accurate definition of the kinds of light employed in determining angles of rota- tion, particularly the sodium light. The author has discussed this subject in detail which had never before been handled as a connected whole. This was all the more desirable since there can be no doubt that the marked discrepancies found in the determination of the specific rotation of the same substance by different observers do not always depend on impurities in the material used, but very largely on improper manipulation of the polariscope, or on variations in the character of the sodium light employed. It may be further remarked here that a con- sideration of these sources of error carries with it no increased difficulty in the methods of observation, as one might, at first thought, assume. The section embraces in addition a discussion of all known methods for determination of rotation dispersion, a subject PREFACE Vll which up to the present time, has not been covered in its re- lated bearings. Finally, the chapters on sodium lamps and polarization tubes follow, and something on the preparation of solutions and the determination of specific gravity, which, as compared with the corresponding parts of the earlier edition, are greatly enlarged. In the preparation of this whole section the author was in position to make use of the experience gained in many investi- gations of the Physikalisch-Technischen Reichsanstalt. Part V. "The Practical Applications of Optical Rotation" was written by Dr. F. Schiitt, Royal Councilor and Permanent member of the Patent Office. As regards saccharimetry, which makes up the larger part of the section, it may be remarked that it was considered the most satisfactory to follow the methods officially adopted in the German Sugar Tax Law of May 27, 1896, and to present these literally. Among other methods of polarimetric analysis those concerned with the de- termination of cinchona alkaloids are shortened as compared with what was given in the first edition, while, on the other hand, some new methods have been added. Part VI. "The Constants of Rotation of Active Bodies " has been compiled mainly by Dr. L. Berndt and Dr. Th. Pos- ner, formerly assistants in the II Chemical Laboratory of the Berlin University. The section on ethereal oils was prepared by Dr. Rimbach, and Prof. Dr. Thierfelder had the kindness to assist in securing the data on bile acids and proteids. A few chapters remained in my hands. In the preparation of this collection of experimental data it was not found possible to include all the statements given in the literature, as this would have unduly increased the size of this part of the work. It was sufficient in many cases to quote observations in part and refer in a note to the original articles. For the same reason any data on the methods of prep- aration of the substances observed had to be omitted, although such information might often have been found valuable. The search through the literature, in which, however, the possi- bility of overlooking certain data could not be wholly excluded, was complete to about the middle of 1896. From that date it was only partial. In arranging the substances in order the Vlll PREFACE most consistent system may not always have been followed, but by aid of the alphabetical index it will be possible to find any body described. It is hoped that the work in this new edition also will fill its place as a text- and handbook in presenting a complete resume of our knowledge on the subject of optical rotation. H. LANDOLT. BERLIN, December, 1897. TRANSLATOR'S PREFACE The two editions of this work which appeared in Germany in 1879 and 1898 enjoyed there a great and well-deserved pop- ularity. A translation of the first edition was brought out in England in 1882, under the title : " Handbook of the Polaris- cope and Its Practical Applications," and contributed not a little to the advance of methods of optical analysis in that country and the United States. Both the original edition and this translation have been, however, long out of print. The scope of the second edition, a translation of which I have the honor of presenting to American and English readers, is much wider than that of the first ; the main points of differ- ence are made plain in the author's preface, but attention may be called to the fact that the detailed discussions in Sections IV and V of Part I on the relations between the rotating power and the chemical constitution of carbon compounds, along with the full numerical data on constants of rotation, etc., render the work of the highest value to investigators in many fields of pure organic chemistry. Some of the most important ad- vances in this direction are those which have been made in the methods for the resolution of racemic compounds. This sub- ject is thoroughly treated by the author, and permission was also given to include still more recent work in the English edition. The sections which I have added in this connection relate to the resolution of asymmetric nitrogen and sulphur compounds and to several new general processes of resolution. I have made also many additions to the numerical values in Part VI, on Constants of Rotation. The data for a few of these were sent me by the author, while the others were taken from the journals of the three years following the publication of the German text. By an arrangement between the publishers, the cuts for the illustrations of the original work become the property of the X PREFACE American publisher and are used in the translation. This will account for the appearance of some German words in connec- tion with a few of the illustrations. In conclusion I wish to acknowledge my indebtedness to my colleagues, Professors Crew and Dains, of Northwestern Uni- versity, and to Dr. H. W. Wiley, of Washington, for several suggestions of value, and to my assistant, Mr. Frank Wright, for help in the reading of proof. J. H. LONG. CHICAGO, January, 1902. TABLE OF CONTENTS GENERAL CONDITIONS OF OPTICAL ACTIVITY I. Introduction, Definitions and Formulas of Calculation 1 . Active Bodies i ^ 2. Measure of Rotating Power, Specific Rotation 2 3. Molecular Rotation 6 4. Historical 6 II. Classification of Active Substances 5. Preliminary Remarks. Relation of Crystalline Form to Rota- tion 7 6. First Class. Bodies which Possess the Power of Rotating the Plane of Polarized Light in the Crystal Condition only 9 List of Bodies in this Class 9 and 13 Behavior of Crystalline Mixtures 1 1 Behavior of Active Crystals in Powdered Condition 1 1 7. Second Class. Bodies which Rotate in both Crystalline and Amorphous Form 14 8. Third Class. Bodies which are Active in Amorphous Condi- dition only (natural liquids or solutions) 18 List of Active Carbon Compounds 1 . Hydrocarbons 19 2. Monohydric Alcohols and Derivatives 19 3. Dihydric Alcohols and Derivatives 20 4. Trihydric and Tetrahydric Alcohols 20 5. Pentahydric Alcohols 20 6. Hexahydric Alcohols 20 7. Heptahydric Alcohols 21 8. Octahydric Alcohols 21 9. Nonahydric Alcohols 21 10. Acids with 2 Atoms of Oxygen and Derivatives 21 11. " 3 " " " 21 12. 4 " 22 13- " 5 " " " 23 14- 6 " 24 15- 7 " " 24 16. 8 " 25 17. Acids with More than 8 Atoms of Oxygen and Deriva- tives 26 Xll TABLE OF CONTENTS 1 8. Oxyaldehydes, Aldoses, Aldehyde Sugars 26 19. Oxyketones, Ketoses, Ketone Sugars 28 20. Disaccharides 28 21. Trisaccharides 28 22. Polysaccharides 28 23. Carbohydrates 28 24. Gums 28 25. Pectin Bodies 28 26. Alcohols and Acids of Unknown Structure 28 27. Terpenes 29 28. Camphors and Derivatives 30 29. Ethereal Oils, Essential Oils 33 30. Resin Acids 34 .; i . Aromatic Amines 34 32. Alkaloids 34 33. Glucosides 36 34. Bitter Principles, Coloring-Matters 37 35. Bile Acids 37 36. Protein Substances 38 37. Derivatives of Asymmetric Nitrogen 38 38. Derivatives of Asymmetric Sulphur 38 Enumeration of Active Bodies 38 III. Nature of the Rotating Power $ 9. Distinction between Crystal Rotation and Liquid Rotation. Rotation of Vapors. Molecular Rotation 39 10. Optical Theory of Circular Polarization in Quartz 41 ii. Optical Constitution of Active Liquid Substances 43 '',. 12. Investigations of Pasteur. Molecular Asymmetry 44 v IV. Relations between Rotating Power and Chemical Constitution of Carbon Compounds 13. van't Hoff-LeBel Theory 47 $ 14. Asymmetric Nitrogen and Sulphur 52 V. Optical Modifications 15. General Remarks 54 A. Calculation of Number of Optical Modifications of a Compound from the Number of Asymmetric Carbons Atoms in It I 16. Numerical Results 55 B. Physical and Chemical Behavior of the Optical Modifications a . Behavior of the Antipodes '/ 17- Physical Properties 67 \ 18. Different -r of the Antipodes on Combination with Active Substances 68 'i 19. Physiological Differences between the Antipodes 71 TABLE OF CONTENTS Kill b. Properties of Racemic Compounds and Distinctions between Them and Active Modifications i. Crystallized Racemic Compounds 20. Molecular Weight 77 21. Crystalline Form and Water of Crystallization 78 22. Density 79 23. Solubility 80 24. Melting-Point 83 2. Liquid Racemic Compounds 25. Are These to be Considered as Compounds or Mixtures 86 26. Results of Discussion 89 C. Formation of Racemic Bodies 27. Production of Racemic Bodies by Combination of the Anti- podes. Transition Temperature 90 28. Production of Racemic Bodies from One of the Active Forms by Heat 93 29. Racemization by Conversion of Asymmetric Bodies into Asym- metric Derivatives 96 30. Production of Racemic Compounds by Conversion of Sym- metric Bodies into Asymmetric 97 31. Racemic Compounds from Right- and Left-Rotating Isomers of Different Configurations 98 D. Resolution of Racemic Bodies 32. i. Resolution by Crystallization, Spontaneous Resolution 99 33. 2. Resolution by Active Compounds 102 a. Resolution of Racemic Acids by Alkaloids 103 b. Resolution of Racemic Bases by Tartaric Acid no c. Resolution by Stronger Acids 113 Resolution by Esterification or Saponification 115 34. 3. Resolution by Aid of Fungi, and Data on the Fungi Suita- ble for Resolution 117 List of Active Forms Obtained by Aid of Fungi 127 E. Formation of Active Isomers 35. i. From Inactive Materials. Artificial Preparation of Active Compounds 130 36. 2. From Active Materials 132 37- 3- Formation of Active Bodies in the Animal or Vegetable Cell 13? F. Transformation of the Active Isomers 38. Reciprocal Transformation of the Antipodes 138 39. Reciprocal Transformation of Active Isomers of Different Configurations 14 xiv TABLE OF CONTENTS G. Inseparable Modifications of Inactive Configuration \ 40. Different Classes of These Bodies 140 fc 41. Differences in the Properties of Racemically Inactive and Structurally Inactive Isomers 144 SECOND PHYSICAL LAWS OF CIRCULAR POLARIZATION 42. Relation of Rotation to Length of Column 146 43. Dependence of the Angle of Rotation on the Wave-Length of the Ray. Rotation Dispersion 146 \ 44. Rotation Dispersion of Crystals 148 \ 45. Rotation Dispersion of Liquids and Dissolved Substances 154 \ 46. Anomalous Rotation Dispersion 157 F/\RT THIRD NUMERICAL VALUES FOR THE ROTATING POWER. SPECIFIC ROTATION \ 47. Biot's Conception of Specific Rotation 165 I. Constant Specific Rotation of Dissolved Substances \ 48. Original Biot Law 166 II. Variable Specific Rotation of Dissolved Substances A. Dependence of the Specific Rotation on the Concentration 49. Recognition of Variation in Specific Rotation 169 >. Determination of True Specific Rotation 170 \ 51. Reduction Formulas 175 $ 52. Experimental Proof of Biot's Formulas 176 I 53. True Specific Rotation of Solid Active Substances 190 'i 54. Slight Changes in Specific Rotation by Variations in Concen- tration 194 't 55. Specific Rotation in very Dilute Solutions 196 I 56. Minimum Value of Specific Rotation 197 8 57. Reversal in the Direction of Rotation by Change in Concen- tration 2OI g 58. Increase or Decrease in Specific Rotation with Increasing Di- lution of Solutions 203 B. Dependence of the Specific Rotation on the Nature of the Solvent '',. 59. Specific Rotations in Different Solvents 206 C. Dependence of the Specific Rotation on the Temperature 60. Effect of Increase of Temperature on Liquid and Dissolved Active Substances 207 TABLE OF CONTENTS XV D. Causes of the Changes in Specific Rotation 61. a. Electrolytic Dissociation in Aqueous Solutions 215 62. b. Formation or Decomposition of Molecular Aggregations of Simple Structure 227 63. c. Presence of Complex Polymerized Molecules (Crystal Mole- x cules) in the Solution 232 64. d. Combinations of the Active Body with the Solvent. Hydrates 234 65. /, 131. Description of the Instrument 385 2. Half-Shadow Saccharimeters \ 132. Construction of the Instruments 387 % 133. Half-Shadow Saccharimeters with Single Wedge Compensa- tion 388 | 134. Half-Shadow Saccharimeters with Double Wedge Compensa- tion 388 >/. 135. Beet Juice Saccharimeter with Enlarged Scale 389 xvill TABLE OF CONTENTS $ 136. Half-Shadow Saccharimeter of Peters 391 \ 137. Half-Shadow Saccharimeter of Fric 392 c. Illuminating Lamps i. Lamps for White Light \ 138. Schmidt and Haensch Gas Lamps 393 139. The Hinks Petroleum Lamp 394 | 140. Lamps with VVelsbach Light 394 $ 141. Lamp for Electric Light 394 142. The Zirconium Light 394 2. Lamps for Homogeneous Light \ 143. Simple Sodium Light Lamps 395 \ 144. Pribram's Sodium Lamp 396 \ 145. Landolt's Sodium Lamp 397 146. Intense Sodium Light 398 3. Purification of the Sodium Light. Optical Center of Gravity 147. Lippich Sodium Light Filter 399 148. Optical Center of Gravity of Sodium Light 402 \ 149. Spectral Purification of Sodium Light ' 405 \ 150. Dependence of the Optical Center on the Brightness 407 \ 151. Absolute Determination of the Rotation of Sodium Light for Quartz 413 2 152: Relation of the Angles of Rotation, a D and a.j 415 8 ; 53- Optical Center of Gravity of White Light 416 d. Determination of Rotation Dispersion \ 154. Method of Broch 419 ^ . Method of v. Lang 423 $ 156. Method of Lippich 425 \ 157. Method of Lommel 427 \ 158. Method of Landolt with Ray Filters 429 $ 159. The Arons-Lummer Mercury Lamp 433 B. Construction of Polarization Tubes and the Meas- urement of Their Length \ 160. Construction of the Tubes and Method of Closing Them by End-Plates of Glass ... 436 fc 161. Water-Jacket Tubes and Water-Heating Apparatus 438 \ 162. Calculation of Specific Rotation with Consideration of Tem- perature 441 163. Schmidt and Haensch Control Tube 441 \ 164. Measurement of the Tube Length 443 C. Determination of Percentage Strength of Solutions | 165. Reduction of Weighings to Vacuo 444 f 166. Preparation of Solutions by Weighing 446 . 167. Change in Percentage Strength on Filtration of Solutions 448 TABLE OF CONTENTS xix D. Determination of Specific Gravity 168. Construction and Use of the Pycnometer 449 169. Calculation of the Specific Gravity - 454 170. Variations in Specific Gravity with Temperature 456 E. Determination of the Concentration of Solutions 171. Calculation of the Concentration from the Specific Gravity and Percentage Strength 458 172. Preparation of Solutions in Measuring Flasks 459 F. Effect of the Different Errors of Observation on the Specific Rotation 173. Calculation of Errors 461 F/\RT PRACTICAL APPLICATIONS OF OPTICAL ROTATION I. Determination of Cane Sugar. Saccharimetry A. Determination of Sugar with Instruments Having a Circular Graduation \ 174. Calculation of Concentration. Formulas 463 | 175. Concentration and Variable Rotation 463 B. Determination of Cane Sugar with Application of Wedge Compensation Instruments \ 1 76. Preliminary Remarks 466 | 177. Practical Methods of Saccharimetry According to the Provi- sions of the German Sugar Tax Law 467 II. Determination of Milk Sugar << 178. Constants for Milk Sugar 488 | 1 79. Determination of Sugar in Milk 489 III. Determination of Glucose 180. Calculation of Formulas 491 \ 181 . Sugar in Diabetic Urine 493 IV. Determination of Maltose 182. Calculation of Formulas 495 V. Determination of Galactose \ 183. Calculation of Formulas 496 VI. Determination of Camphor 184. Formulas and Method 497 XX TABLE OF CONTENTS VII. Determination of Cinchona Alkaloids 185 Calculation of Formulas 49 8 VIII. Determination of Cocaine 186. Calculation of Formulas 5! IX. Determination of Nicotine 187. Formulas and Method 503 SI^CTH CONSTANTS OF ROTATION OF ACTIVE BODIES Group 1. Hydrocarbons Ethylamyl, Propylamyl, etc 505 2. Alcohols with One Atom of Oxygen Amyl Alcohol and Derivatives 506 Hexyl Alcohol, Methylhexyl Carbinol 510 3. Alcohols with Two to Four Atoms of Oxygen 510 4. Alcohols with Five Atoms of Oxygen Pentitols : Arabitol, etc 511 5. Alcohols with Six Atoms of Oxygen Hexitols : Mannitol, Sorbitol, etc 511 6. Alcohols with Seven Atoms of Oxygen Mannoheptitol, etc 512 7. Acids with Two Atoms of Oxygen Valeric Acid, Caproic Acid 513 8. Acids with Three Atoms of Oxygen and Derivatives Lactic Acid and Salts 515 Oxybutyric Acid, Leucin, etc 518 Mandelic Acid and Derivatives 520 9. Acids with Four Atoms of Oxygen and Derivatives Glyceric Acid and Salts, etc 523 10. Acids with Five Atoms of Oxygen Malic Acid, Salts and Esters 526 Oxysuccinic Acids and Derivatives 536 Shikimic Acid and Derivatives 543 n. Acids with Six Atoms of Oxygen Arabonic Acid, Ribonic Acid, etc 544 Tartaric Acids 546 12. Acids with Seven Atoms of Oxygen Gluconic Acid, Gulonic Acid, etc 565 13. Acids with Eight Atoms of Oxygen Glucoheptonic Acid, Saccharic Acids 569 TABLE OF CONTENTS XXI 14. Acids with Nine Atoms of Oxygen Glucooctonic Acid, etc 572 15 Acids with Ten Atoms of Oxygen Gluconononic Acid, etc. 573 16. Oxyaldehydes, Aldoses, Aldehyde Sugars Arabinose, Xylose, Glucose, etc 574 17. Oxyketones, Ketoses Fructose 589 18. Invert-Sugar 591 19. Disaccharides, Saccharoses Cane-Sugar 596 Milk-Sugar, Malt- Sugar, etc 599 20. Trisaccharides and Polysaccharides Raffinose, Melitose, etc 605 21. Carbohydrates Soluble Starch, Dextrines, etc 607 22. Gums Arabin, Wood-Gum 61 1 23. Camphors and Terpenes A. Aliphatic Terpenes 612 B. Terpan Group 614 C. Camphan Group 627 D. Polyterpenes 658 24. Ethereal Oils 660 25. Resin Acids 666 26. Alkaloids Of Aconite Species, etc 667 Cinchona Alkaloids 671 Of Coca Leaves 698 Of Opium 701 Strychnos Alkaloids 706 Other Alkaloids and Bases 707 27. Glucosides Salicin Helicin, Amygdalin, etc 713 28. Bitter Principles and Indifferent Bodies Santonin Group 715 Other Vegetable Substances 717 29. Biliary Substances 718 30. Gelatinous Substances 723 31. Protein Bodies Albumins, Albumoses, etc 724 General Index 729 Index of Active Substances 737 PART FIRST General Conditions of Optical Activity 1. INTRODUCTION, DEFINITIONS, AND FORMULAS OF CAL- CULATION i. Active Bodies. Those substances which possess the property of rotating through a certain angle the olane of polarization of a ray of polarized light which passes through them are desig- nated as optically active, or circular ly^pnlari^ing, . V-4Ie the property itself is described as optical rotating power. The property of optical activity is shown by : i . A number of inorganic and organic substances in crystalline condition. 2. By a large number of carbon compounds when exposed to the polarized ray in liquid or dissolved condition. In bodies of the first class the cause of the optical activity is due to peculiarity of crystalline structure, while in the second it is due to an unsymmetrical arrangement of the atoms within the molecule. According to the direction in which the rotation of the plane of polarization takes place active bodies are either : Dextrorotatory, with the sign -f- or d, Laevorotatory " or /. If an organic ^-compound be subjected to chemical trans- formation, the derivatives may be in part also right rotating, or they may be in part even left rotating. In order to indicate the derivation from the original parent substance, the letter, d, is retained as a prefix for all bodies of the group, without, however, expressing by it the direction of rotation in the derivative. If this, also, is to be shown, it can be done by the addition of the + or sign. The expressions, d (-f-) and d ( ) , indicate right and left rotating derivatives of a dextro- parent substance, while / ( ) and / (-f ) indicate the direction of rotation of the derivatives of a laevorotating substance. 2 GENERAL CONDITIONS OF OPTICAL ACTIVITY Many bodies occur in isomeric forms with optically different behaviors. There are recognized : /. Active modifications, found always in two forms con- , stituting the so-called optical antipodes, inasmuch as under like conditions, one form rotates as strongly to the right as the other to the left. These are designated as the ^-form and the /-form. 2. Inactive modifications which are mixtures or compounds of the active antipodes in equal proportions, and which may be split up into these by the action of certain agents. For these so-called mow//'; bodies, the symbols;- or e = *" 5) ' M "= + 44 '*>- (glacial acetic acid, p 39.72, a 4 - 1.0113), []-: + 47-I& As experience lias shown there are certain substances for which the specific rotation, as calculated from solutions of GENERAL FORMULAS FOR ROTATION 5 different concentrations, remains constant or at most suffers but a very slight change. For such bodies, for example, cane- sugar or milk-sugar dissolved in water, the angle of rotation is exactly proportional to the concentration, and therefore by the following formula, I OCX* the strength of a solution of the substance may be found if the value of [<*] is known. Polarimetric analysis, especially of sugars, is based on this fact. With the great majority of active substances, however, it has been observed that the specific rotation increases or decreases, with increasing dilution of the solution, and at very different rates for different substances ; sometimes the direction of rotation, even, may change. In such cases, there must be an alteration of the nature of the substance by the action of the solvent. If the value of [<*], for a number of solutions of different strengths has been found, it w r ill be possible to express the dependence of the specific rotation on the factors p and gby the following formulas, in which the constants, A, B, and C, or a, b, and c are determined by experiment : (V) [a] =A + Bq (VI) [a] == a + bp, or (V) [a] '=A + JB?+ Cf (VI') [>] == a + bp + #* In the formulas, (V) and (V), the constant A expresses the specific rotation of the active substance in undiluted con- dition, w r hile B and C show the change produced by i per cent. of the inactive solvent. If q is taken equal to 100, the specific rotation in infinite dilution is given. In the formulas, (VI) and (VF), the constant a corresponds to the specific rotation in infinitely great dilution, while the value for the pure substance is given when/> is taken equal to 100. If it be desired to replace the constants, A and B, by a and in the formulas, (V) and (VI), and vice versa, we take 6 GENERAL CONDITIONS OF OPTICAL ACTIVITY A a + ioo b a = A -f 100 # B= b b= B In the 3-term formulas, we have : ^ = a -f ioo b -f 10,000 ^ a A + ioo ^ -h 10,000 C /? - b 200 c. b = B 200 C. C=c. c = C. If, further, the constants of the equation, [a] =a + ty+ & 2 , have been determined for an active substance, with molecular weight J/, and if it be desired to alter them, so as to make them apply for a derivative (hydroxide or salt, for example), with the molecular weight M 1 ', which leads to the formula, then, we place : M 3. Molecular Rotation. This term is applied to the product of the molecular weight and specific rotation of a body, and is represented by the symbol [M]. But, to avoid the use of inconveniently large numbers, it is customary to take the one-hundredth part of this product ; thus : ioo In this case, [J/] expresses the rotation which would follow, if each cubic centimeter of the solution contained i gram- molecule of the active substance, and the length of the liquid column were i millimeter. The molecular rotation is applied in making stoichiometric comparisons. 4. Historical. The rotation of the plane of polarization was noticed first in quart/ plates by Arago, in 1811. In 1815 Biot and Seebeck discovered the optical activity of certain organic substance^ ' oil of turpentine, and aqueous solutions :^ar and tartaric acid). Through a long series of investi- gations, extending over a period of 47 years ( from 1 8 1 3 to 1 860) , CLASSIFICATION OF ACTIVE SUBSTANCES 7 Biot established the important physical laws and the general nature of the phenomena observed. In 1823, Fresnel published a theory of the effects as noticed in quartz in which he intro- duced the term circular polarization. In 1831, Herschel dis- covered the important relation existing between the rotating power in quartz crystals, and the development of their faces. A further fundamental discovery in this field w r as made by Pasteur, who found in 1848, in the examination of tartaric and racemic acids, that one and the same active substance may occur in oppositely rotating and in inactive modifications. The last great advance in the subject was brought about in 1874 by van 't Hoff and LeBel, who independently discovered the relation existing between the rotating power of organic substances and their atomic constitution, in the discussion of which the notion of asymmetric carbon atoms was introduced in the science. II. CLASSIFICATION OF ACTIVE SUBSTANCES 5. Preliminary Remarks, Relation of Crystalline Form to Rotation. All bodies which in crystalline condition or in solution have the power of rotating the plane of polarization of light crystallize, as was shown by Pasteur, 1 in so called hemihedral forms, and the crystals of the right and left modifications of an active substance are enantiomorphous. A hemihedral crystalline polyhedron is not superposable on its mirror image. The original form and that corresponding to its image are related as is the right hand to the left ; they exhibit the peculiarity described in crystallography as enantio- morphism. Hemihedral forms, from a geometric standpoint, can possess axes of symmetry only, but no center of symmetry and no planes of simple or compound symmetry. 2 In the thirty-two possible crystalline groups, eleven are found with hemihedral forms, and these are given below with 1 Pasteur : Compt. rend., 26, 535 ; 27, 401 ; 35, 180. Compare also Becke' : Min. und petrogr. Mitth. v. Tschermak, 10, 414 (1889); 12, 256 (1891). - For details on the symmetry relations in enantiomorphous forms consult the works of Th. Liebisch : Grundriss der physikalischen Krystallographie, Leipzig, 1896. P. Groth : Physikalische Krystallographie, 3 Aufl., Leipzig, 1895. 8 CLASSIFICATION OF ACTIVE SUBSTANCES the nomenclature of Liebisch, and also that of Groth in parenthesis. /. Regular System. 1. Plagihedral-hemihedral (pentagon-icositetrahedral) group. 2. Tetartohedral (tetrahedral-pentagondodecahedral) group. II. Hexagonal System. 3 . Trapezohedral-hemihedral ( hexagonal- trapezohedral ) group. 4. First hemimorph-tetartohedral( hexagonal-pyramidal) group. 5. Trapezohedral- tetartohedral (trigonal- trapezohedral) group. 6. Octahedral (trigonal-pyramidal) group. ///. Tetragonal System. 7. Trapezohedral-hemihedral (trapezohedral) group. 8. Hemimorph-tetartohedral (pyramidal) group. IV. Rhombic System. 9. Hemihedral (bisphenoidal) group. V. Monoclinic System. 10. Hemimorphic (sphenoidal) group. VI. Triclinic System. 11. Hemihedral (asymmetric) group. In all cases in which a complete determination of the crystalline symmetry of the bodies under consideration could be carried out the statement made above, that optical rotation in a crystal is always associated with enantiomorphism, has been confirmed. 1 As experience has shown, this rule is not reversible ; that is, if the crystals of a body are found to be hemihedral the conclusion can not be drawn that in either the solid or dissolved form it will show circular polarization. For example, the following compounds crystallize in hemihedral forms, NH 4 C1, Ba(NO s ) 2 , Li 2 SO 4 + H f O, NiSO 4 + 6H 2 O, Sr(CHO,), -f- 2H 2 O, but they are inactive in solid form as well as in solution. 2 1 I.iebisch : I/x:. cit., p. 41 and 426. Objections to Pasteur's law have been raised by Wyrouboff : Ann. chim. phys. [6], 8, 416; [7], i, 10. Also by Walden : Ber. d. chem. Ges., 39, 1692. Traube replied to this : /bid., 39, 2446. OPTICALLY ACTIVE CRYSTALS 9 Optically active substances may be divided into three essen- tially distinct classes : 6. First Class. Bodies which possess the property of rotating the plane of polarized light only in the crystallized condition, and which lose this property when brought into the amorphous condition by fusion or solution. Circular polarization has been noticed only in crystals belonging to the regular, hexagonal, and tetragonal systems ; that is, in groups i to 8 of the above scheme. At present the following organic and inorganic bodies are known to belong here : Regular System. f Sodium chlorate NaClO 3 . Group 2 J Sodium bromate XaBr0 3 . Sodium sulphantimonate Na 3 SbS 4 -f 9H 2 O. [ Sodium uranyl acetate NaUO 2 (C 2 H 3 O 2 ) 3 ? Hexagonal System. f Potassium lithium sulphate KLiSO 4 . Ammonium lithium sulphate (NH 4 )LiSO 4 . Rubidium lithium sulphate RbLiSO 4 . (^ Potassium sulphate lithium chromate K 2 SO 4 -f- Li 2 CrO 4 . f Quartz SiO,. Cinnabar HgS. Potassium dithionate K 2 S 2 O 6 . Rubidium dithionate Rb 2 S 2 O 6 . Calcium dithionate CaS 2 O 6 4H 2 O. Strontium dithionate SrS 2 O 6 -f- 4H 2 O . Lead dithionate PbS 2 O 6 + 4H 2 O. Benzil C 6 H 5 .CO.CO.C 6 H 5 . (Rubidium and cesium tartrates, laurel camphor, and matico-camphor belong to Class II. ) Group 6 \ Sodium periodate NaIO 4 -[- 3H 2 O. Tetragonal System. f Ethylene diamine sulphate (N 2 H 4 .C2H 4 )H 2 SO 4 . I Guanidine carbonate (CH 5 N 3 )H 2 CO 3 . Group 7 \ Diacetyl phenolphthalein C 20 H 12 ( C 2 H 3 O 2 ) 2 O 4 I Sulphobenzene trisulphide ( C 6 H 5 .SO 2 .S) 2 S. [ Sulphotoluene trisulphide (Strychnine sulphate belongs in Class II.) Group 5 10 CLASSIFICATION OF ACTIVE SUBSTANCES The optically active single-refracting crystals belonging to the regular system show rotation of the plane of polarization equally strong in all directions, as was shown especially by Sohncke, 1 in the case of sodium chlorate. In the cases of hexagonal and tetragonal crystals, which are uniaxially double- refracting, the phenomenon of circular polarization may be observed only in the direction of the optical axis, and plates for this purpose must therefore be cut perpendicularly to this axis. In normal biaxial crystals belonging to the rhombic, monoclinic and triclinic systems the rotating power has not yet been observed, and as Wiener 2 has shown, it cannot be found in all those cases where there is at the same time strong double refraction. All the active crystals which have been mentioned above Fig. i. Fig. 2. occur in right- and left-rotating varieties, which exhibit equal activities for equal thicknesses of layers passed by the light. The direction of rotation stands in relation to the condition of enantiomorphism in the crystal, which is often shown in the geometric development of the latter by the appearance of so-called hemihedral or tetartohedral surfaces, oppositely located in different individual crystals. The best known illus- tration of this is found in hexagonal, trapezohedral-tetarto- hedral quart/., in which the tetartohedral surfaces, s and x, very often appear, and in such a manner that in right-rotating Mis CFig. 2), the surface, s, lies to the right of x, while in left-rotating crystals ( i'\ K . i ) , it lies to the left of x. The one 1 Sohncke: Wit-d. Ann., 3, 530. Wiener: Wied. Ann., 35, i. OPTICALLY ACTIVE CRYSTALS II form is the mirror image of the other. Analogous relations are found in other crystals of the class. 1 The extent of rotating power, for equal thicknesses of layer passed by the light, is very different in different active crystals. The following table gives the data thus far found for plates of i millimeter thickness, and for light of different wave lengths. The latter, expressed in millionths of a millimeter (jUA*), correspond either to the Fraunhofer lines, C, D, E, F, G, to mean yellow light,/, or to the lines from lithium, sodium, and thallium. The bodies are arranged in the order of their rotating power for the line D. 2 Crystalline mixtures of isomorphous active crystals exhibit a rotation which is nearly proportional to the percentage com- position and amount of rotation of the components. The proof of this rule has been given mainly by Bodlander 3 from investigations of various crystallizations of lead dithionate and strontium dithionate. Behavior of active crystals in pou'dered condition. The ques- tion as to whether the rotating power of fine particles is the same as that of the larger crystal, or whether it decreases when the particles have reached a certain degree of fineness has been tested by Landolt* w T ith sodium chlorate. In crystalline plates i millimeter in thickness the salt shows a rotation for white light, ctj = 3.54, and as its specific gravity is d - 2.488 the specific rotation must be \_a~]j a/d = =b 1.42. If the crystals are rubbed as fine as possible and the powder so obtained be suspended in a mixture of absolute alcohol and carbon disulphide, the composition of which may be varied until a clear liquid is obtained, that is, until the mixture has the same refractive index as the salt particles, then it is found on examination in a polarization tube 1 All these cases are discussed in the works of Groth and I,iebisch, already referred to. - Besides these the following observations have been made which refer to green light not specially denned : Full data are given for quartz and sodium chlorate in the chapter on "Rotation Dispersion." 3 Bodlander : Inaug. Diss. Breslau, iSSa ; Wied. Beib., 7, 396. 4 Landolt : Sitzungsber. der Berl. Akad., 1896, 785 ; Ber. d. chem. Ges., 29, 2404. 12 CLASSIFICATION OF ACTIVE SUBSTANCES (i to 1.5 decimeter in length) that the suspended substance shows the right or left rotation of the original crystals. In order to keep the powder evenly distributed through the liquid it is necessary to rotate the polarization tube on its axis. In such experiments in which the diameters of the particles were mostly between 0.004 mm. and 0.012 mm. or even between 0.003 an d 0.007 mm - ^e specific rotation 1 was found as [a]y = =fc 1.36 to 1.49, or, in the mean, 1.41, referred to a layer i mm. in thickness. As this value agrees exactly with that found for the large crystals it follows that in powdering to the degree mentioned the active crystalline structure has not suffered the slightest change. The question, what size the smallest particle which still shows circular polar- ization must have, and of how many single molecules it must be composed, remains unanswered. The optical activity of sodium chlorate disappears completely in aqueous solution, and even when this is in supersaturated condition. If the salt be precipitated by addition of alcohol, the crystalline precipitates formed, and from solutions of either right or left rotating crystals, are found to be, when examined in the suspended condition, either inactive or to show one of the two directions of rotation. The specific rotation is, however, always smaller than the normal ( zb 1.41), from which it follows that the precipitates are mixtures of right and left salts. Which of these predominates depends on the direc- tion of rotation of the particles which first separate. Bearing on this it may be said that when one of two portions of a saturated solution is treated with a trace of the solid right- rotating salt, and the other with a trace of the left-rotating salt and then precipitated by alcohol, the precipitates formed are found to rotate accordingly. On the cause of rotation in crystals see 9 and 10. 1 The specific rotation of solid active particles is expressed, as in the case of solu- tions by [a] = or///, when a. is the angle of rotation, /the length of tube, v the volume of the same and / the weight of the powder suspended. OPTICALLY ACTIVE CRYSTALS _ - s ^S- = fo OJ *"1 ^. jj * i 2.8 ~ 3 i & 3? ~ s " S>^s? Q^-ls^g -ft flfif*2? S ?^* M 3 SS o ?lglj-tl ?!lB!! * 2 Q r* ^ O OC M ^ \p "po rt ^ " ifcBi^S 1 *^ miiti* ? i i a s " ? & * - S -2- i g ? NJ -; - - ^ ?effin iljufls L os r = & ? 8 31 rt b^ - - p ^ . 2. H. n of the persion. 139, 224 157, 122 5 ( eic 2. of t rsio 9, 2 7, 1 (188 h*;i 1 ^ cj B . ^. ! c S B B ^ d gB^B : : ! ! 5*.*s. ^ ' ' & '. i : tf ; . Os^J . W '-" b : in ocvo M 10 K> W 1-1 M M \D (M 4" OOCn W O* KJ . MM in i. M ft ft ~ C O 9? g^^^ .... K> 00 . -P>. . . . . <*> bo ; : : : : OJ 00 -F. k) M C vO CC^J 14 CLASSIFICATION OF ACTIVE SUBSTANCES 7. Second Class. Bodies which rotate in both crystalline and amorphous form, in solution or in fused condition. But six compounds of this class are known : Matico camphor, Rubidium tartrate, Patchouli camphor, Cesium tartrate, Laurel camphor, Strychnine sulphate. 1 It is interesting to compare in these substances the rotation in the crystalline form with that shown by an equally thick plate of the amorphous compounds (H. Traube). In the last case the action of the single molecules alone comes into play, while in the other the effect of a particular crystalline structure is to be added to this. In this second case the molecular rotation and crystalline rotation must be added together, or, if they have opposite directions they may partially neutralize each other. The following observations have been made on these sub- stances : Matico camphor. C^H.^O.* The rotation of the hexagonal trapezohedral-tetartohedral crystals was discovered by Hintze 3 and measured by him, and also, later by H. Traube. 4 The following angles were measured for a plate i mm. in thickness: Hintze : a u = - 1.68, Na - - 2.07, <* T , = 2.47. Traube: a Na = - 1.81, 1.96, 1.86. Mean, a n = 1.92. Right rotating crystals have not yet been found. H. Traube 5 has measured the rotating power of the sub- stance, which melts at 94, in the liquid condition, in a tube i decimeter in length and at different temperatures, as follows : Temp. Rotation for Sp. gr. at t. Specific rotation. T. i dm. a D d\. WP. 108 26.29 0-924 28.45 H5 25.59 0.901 28.40 126 24.74 0.874 28.32 135 23.86 0.845 28.24 i Recently W. J. Pope (J. Chem. Soc., 69, 971) has added two bodies to this list. These are d-?r-camphanic acid, O.C 8 Hi, i. 08 The angle of rotation of the amorphous and crystalline sub- stances are related thus as 29 : 178 ; and if we assume that in the last number the molecular rotation is included, the part due to the crystalline structure must be 178 29 = 149. It follows therefore that the rotatory polarizing behavior of the crystals is about ^ due to the molecular rotation and f to the crystalline rotation. Patchouli camphor, C 15 H 26 O. The optical rotation of the hexagonal crystals, first noticed by v. Seherr-Thoss 1 was measured by H. Traube, 2 who found a rotation of OL D = 1.325 for i mm. Montgolfier 3 found \of} D = - 118 as the specific rotation of the fused substance at 59 (corresponding to about 1 1 8 . 3 for the ordinary temperature ) . From alcoholic solution in which the rotation decreases with the dilution he derived the value, [ct]^ 124.5 for the pure substance. From these two values, by taking the specific gravity as 1.051, according to Gal 4 (for the crystals), the rotation of a layer i mm. thick is calculated as tx D = 1.24 and 1.31, which values agree very well with that observed in the crystals ; viz., 1.325. We have here a case in w r hich the crystalline rotation coincides nearly with the molecular rotation, and one therefore in which the effect of the first is scarcely, if at all, perceptible. Laurel camphor, C 10 H 16 O. The crystals, according to H. Traube, hexagonal trapezohedral-tetartohedral, show, by 1 Private communication. - H. Traube : Sitzungsber. der Berliner Akad., i, 195 (1895). 3 Montgolfier: Bull. soc. chim. [2], a8, 414 (1877). 4 Gal: Ztschr Chem., 220 (1869). 1 6 CLASSIFICATION OF ACTIVE SUBSTANCES the measurements of v. Seherr-Thoss 1 a right rotation as follows, for a plate of i mm. thickness : White light. Ray B. Ray D. Transition tint. Ray G. 0.875 0.455 0-65 0.73 1.818 From solutions of the camphor in different liquids the maxi- mum specific rotation of the pure amorphous substance is found to be \_oi] D - -f- 55.5 ; 2 therefore, if the specific gravity be taken as 0.998, the same as for the crystallized camphor, the angle of rotation for a layer of i mm. thickness is found to be []/> = : -f 0.55. This value is but little lower than that found for the crystals, a D = -f 0.65, from which it follows that the activity of the latter appears to depend almost wholly on the molecular rotation. Right and left rubidium tartrate, Rb 2 C 4 H 4 O 6 . According to H. Traube both salts are hexagonal trapezohedral-tetarto- hedral. As Wyrouboff 3 found and Traube 4 later the crystals formed from ordinary d- tartaric acid are left rotating, while those from /- tartaric acid are right-rotating. They observed for plates of i mm. thickness : d-Acid salt. /-Acid salt. Wyrouboff (* D -10.7 flf^ -f 10.5 Traube -10.24 +10.12 If the left-rotating crystals be dissolved in water the solu- tion shows right rotation and vice versa. The specific rotating power is diminished with increasing dilution, and at a rate which is shown, according to Rimbach, 5 by this formula, derived for the right rotating salt O] =a + 25.63 -0.06123 ?, in which q represents the percentage amount of water present. This formula was derived from observations on solutions con- taining up to 64.5 per cent, of the salt, and the constant, 25.63, may therefore be taken as satisfactorily representing the specific rotation of the amorphous anhydrous substance. Using with this the known specific gravity of the crystallized v. Seheir Thoss : Ztachr. fur Kryst., 33, 583 (1894). Mean of the determinations of I^andolt : Ann. Chem. (Liebig), 189, 332, and Rim- bach: Ztvrhr. ihys. Chem., 9, 698. 3 Wyrouboff : Jour, de physik. [3], 3, 451 (1894). H Tr.ui>.. Sitzungsber. der Berliner Akad.. i, 195 (1895). Rimbach : Ztschr. phys. Chem., 16, 671 (1895). OPTICALLY ACTIVE CRYSTALS 17 salt, d = 2.694, we obtain the rotation of a plate i mm. in thickness from a D = 0.2563 X 2.694 = + 0.69. Regarding the phenomenon that the crystals show a rotation opposite in direction from that of the solutions it may be said that this depends on the activity of the molecules. It is well known that with many substances the direction of rotation is dependent on the concentration of their solutions. For example, malic acid, sodium malate and barium malate show a negative rotation in dilute solutions which decreases with increasing concentration, becomes zero (for malic acid in 34 per cent, solution) and then goes over into a right-hand rota- tion. This would remain if the substances were finally brought to the solid condition. The same phenomenon is exhibited by solutions of = -f 0.70 and for the iron salt, with c = 66.8 [tx~\ D = -[-0.53 were found. 8. Third Class. Bodies which are active only in amorphous con- dition (natural liquids or solutions). The substances of this class are carbon compounds exclu- sively ; no inorganic substances are known which belong here. The following tables give a general summary of the active DCS Cloizeaux : Pogg. Ann., 102, 477 (1857). H. Traube : Lamlolt-Boernstein's phy. chem. Tab., and Ed. p. 460. Wyrouboff: Bull. Soc. Min., 7, 10 (1884). Ix: Bel: Ber. d. chem. Ges., 5, 391 (1872). Wyrouboff: Ann. chim. phys. [61, 8, 340 (1886). H. Traube: N. Jahrb.fur Min., Beil., y, 625. ACTIVE ORGANIC COMPOUNDS 1 9 organic compounds with their optical modifications. Of the last the right-rotating are designated by -j-, the left-rotating by , the racemic bodies, which may be split into their anti- podes, by r, and the inactive forms, which can not be split up by i. The letter i is used also with some inactive bodies of unknown constitution, which on account of the connection are grouped with other substances. In the symbols, d(-\-}, d( ), /(-}-), /( ), d and / indicate the derivation from a right- or left-rotating parent substance, while -f- and indicate the direction of rotation. For those bodies which have been studied in solution the solvent is mentioned, because this sometimes exerts an influ- ence on the direction of rotation. W = water. A alcohol. E = ether. Ac acetone. B = benzene. C = chloroform. Aa = acetic acid. Bo = borax. / indicates that the body is in itself a liquid. As far as space permits the constitutional formulas of the substances are given and in a manner which will indicate the asymmetric carbon atom, and that is by placing the four groups connected with it in parentheses. The data on specific rotations will be given later in the section on constants of rotation. i. Hydrocarbons. Methylethylpropylmethane; (C 3 H 7 )(C 2 H 5 )C(H)(CH 3 ) / Diamyl,(C 2 H 5 )(CH 3 )(H;C(CH 2 -CH 2 )C(H)(CH 3 )(C 2 H 5 ) / Phenylamyl, (C 6 H 5 CH 2 )(C 2 H 5 )C(CH 3 )(H) ........... / Isobutylamyl, (C 2 H 5 )(CH 3 )C(H)(C 4 H 9 ) ............... / Ethylamyl, (C 2 H 5 )(CH 3 )C(H)(C 2 H 5 ) ................. / 2. Monohydric Alcohols and Derivatives. Methylethylcarbinol, (C 2 H 5 )(CH 3 )C(H)(OH) .......... Derivatives : chloride, iodide .................... Amyl alcohols : Methylethylcarbincarbinol , common active amyl alcohol, (C 2 H 5 )(CH 3 )C(H)(CH 2 OH) .......... .............. f 20 CLASSIFICATION OF ACTIVE SUBSTANCES Derivatives: (a) From /-amyl alcohol : chloride, bro- mide, iodide, cyanide, sulphydrate, thiocyanate, ether, ester, amylsul- ^ -f- phuric acid, diamylamine, triamyl- i aniine and salts of amylamine J Amyl amine (d) From f, Ch Propionyl-, butyrylmalic acid, and an- hydride Ch Methylmalic acid, (CH 3 )(COOH)(H)C C(OH)(H) (COOH) W Chlorsuccinic acid, (C1)(H)C(COOH)(CH 2 COOH) . . . . W Methoxysuccinic acid, (CH 3 O)(H)C(COOH)(CH 2 COOH) W Ethoxysuccinic acid W Urimidosuccinic acid, (H)(NH.CO.NH)C(CO)(CH 2 I J COOH) W Uramidosuccinamide, (H) (CH 2 .CO.NH 2 ) C (NH.CO. NH 2 ) (COOH) W Asparticacid, (H)(NH 2 )C(COOH)(CH 2 COOH) W Common aspartic acid, in alkaline solution , in acid solutions -}-. /3-Asparagin, (H)(NH 2 )C(COOH)(CH 2 .CONH 2 ) W /-Asparagin, in alkaline solution , in acid solu- tion. ... Oxyglutaric acid, (H)(OH)C(COOH)(C 2 H 4 .COOH). . . W Glutaminic acid, (H)(NH 2 )C(COOH)(C 2 H 4 .COOH) W, acids {/-Acid W+ alkalies {/-Amide HC: Pyroglutaminicacid, (H)(NH)C(C 2 H 4 CO)(COOH), and amide W a-Isotrioxystearic acid, C 17 H 32 fOH) 3 COOH -r -r + + + - + - CLASSIFICATION OF ACTIVE SUBSTANCES 14. Acids with Six Atoms of Oxygen and Derivatives. [The aldehydes of these monobasic acids are described as oxyaldehydes (sugars) below.] Tartaric acid, (COOH)(OH)(H)C C(H)(OH)(COOH) W + r Derivatives: Of 7 iy \ ACTIVE ORGANIC COMPOUNDS Derivatives: Thymol- and dichlorthymolglucoronic acid ? rf-Gulonic acid, C<5H 12 O 7 ; lactone li' /-Gulonic acid, C 6 H 12 O 7 ; lactone W Phenylhydrazide W -+- z-Gulonic acid, C 6 H 12 O 7 W * ' rf-Galactose, C 6 H 12 O 6 W + Derivatives: Oxime ( W), a-methylgalactoside ( W}, 0-methylgalactoside ( W -f borax) ethylgalactoside ( W), pentacetate (C) Anilide, toluidide, phenylhydrazone A Mercaptal W Osazone /-Galactose, C 6 H 12 O 6 W Derivatives : Phenylhydrazone W -j- Osazone ? i-Galactose, C 6 H 12 O 6 W r Talose, C 6 H 12 O 6 W + r Sorbose, C 6 H 12 O 6 W Derivative: Methylsorboside W Idose, C 6 H 12 6 W +? +? r Rhamnohexose, C 7 H U O 6 W c) Heptoses: a-Glucoheptose, C 7 H U O 7 W Mannoheptose, C 7 H 14 O 7 W T r Rhamnoheptose, C 8 H 16 O 7 W + d) Octoses : a-Glucooctose, C 8 H 16 O 8 W rf-Mannooctose, C 8 H 16 O 8 sirup W + 28 CLASSIFICATION OF ACTIVE SUBSTANCES e) Nonoses: Glucononose, QH^O,, ........................... sirup W Mannononose, QH^O., ............................... W 19. Oxykctones, Ketoses, Ketone Sugars. rf- Fructose, levulose, C 6 H 12 O 6 ............. ........... W Derivatives: Oxime, anilide (A), osazone (Aa} pentacetate ....................... C /-Fructose, C 6 H 12 O 6 ..................... .............. W Derivative: Osazone ............................. Aa /-Fructose, a-acrose, C 6 H 12 O 6 .......................... Invert sugar {/-glucose (dextrose) ^/-fructose (levu- lose) ............................................. W 20. Disaccharides, C 12 H 22 O n . Cane sugar W Milk sugar ( -f H 2 O) W Maltose (+ H 2 O) W Isomaltose W Trehalose (my cose) ( -j- 2 H 2 0) W Melebiose U r Turanose II' Lupeose W Cyclamose W Agavose // ' 21. Trisaccharides, C 18 H S2 O 16 . Meletriose, raffinose ( -f 5 H 2 0) W Melezitose ( + 2H 2 O) W 22. Polysaccharides. Gentianose, C 36 H fl2 O 3l ? W Lactosin, C 36 H 62 O 31 ? W Stachyose, ? W 23. Carbohydrates, (C 6 H 10 O 5 ) fl . Amorphous soluble starch . W Crystallized soluble starch, amyloilextrine W Achrodextrine [V Maltodextrine W Wood dextrine W Fermentation gum, dextrane W y-Galactan // a-Galactan W Glycogen II' Fermentation mucin W Cellulosin W i i Cellulose (CuO-ammonia, HC1) a and ft Amylan W Gelose W + acids Graminin W Inulin W Irisin W L/evosin W Levosan W Sinistrin W Triticin W Levulin (synanthrose) W 24. Gums. Arabin, arabic acid W Wood gum, xylan W Vegetable mucilage W Wine gum W 25. Pectin Bodies W, -{-,, *'. 26. Alcohols and Acids of Un- known Structure. Quebrachol, C 20 H 34 O C Cupreol, " " " A Cinchol, c Cholestol, C 22 H S8 A Cholesterin , C 26 H 44 O A Phytocholesterin, C 26 H 44 O . A Isocholesterin, " " " . E Paracholesterin, , c Caulosterin, *. Quinovic acid, C 3! ,H 48 O 6 , K-salt W Quinaethonic acid, C 14 H, 8 O 9 W Atractylic acid, Cjo^SjO^, K-salt W ACTIVE ORGANIC COMPOUNDS 27. Terpenes, C JO H 16 . 1. Pinene (Australene 4-, Terebenthene ) / Hydrochloride, C 10 H ]6 HC1 A Hydrobromide, C 10 H 16 HBr ' A rf-Dibromide, Ci H 16 Br 2 / rf-Nitrcscchloride, C K H 16 NCC1 C /-Aldehyde, C 10 H 14 O / 2. Caniphene A /-Hydrochloride, C 10 H 16 HC1 A a- and -Camphene phosphonic acids, C 10 H 15 H 2 Po 3 ----- A 3. Fenchene f 4. Limonene f -f- Hydrochloride, C 10 H 16 HC1 / Tetrabromide, C 10 H 16 Br 4 C a-Xitrosochloride, C 10 H 16 NOC1 C d ( + ) I ( ) . . , / (-) a-Benzoyl nitrosochloride, C ]0 H 16 NOC1. C 7 H 5 O acetic ether d ( ) and esters A Cyancampholic acid, C 10 H 17 O 2 .CN, and oxy- camphocarbaminic acid A Camphoronic acid, C 9 H U O 6 W Fenchone, C 10 H 16 O A Fenchyl alcohol (fenchol), C 10 H 17 OH A Fenchone oxime,. C 10 H 16 .NOH. A Derivatives of fenchone oxime: Fenchone nitrile, C 10 H 15 N A Fenchylamine, C 10 H 17 NH 2 A Benzylidene fenchylamine A Formyl, acetyl, propionyl, butyryl fen- chylamine C o- and />-Oxy- and methoxybenzylidene fenchylamine C Pulegone, C 10 Hi 6 O / Pulegone hydrobromide, Ci H 16 O.HBr A Pulegone oxime, v C 10 Hi 6 NOH, and hydro- chloride A Thujone, Tanacetone, C 10 H, 6 O / Carvol, C, H U O / Hydrogen sulphide carvol, C, H U O.H 2 S C Dihydrocarveol, C, Hj 8 O '/ Dihydrocarvone, Ci H 18 O, /, and dihydro- carvoxime, C 10 H, 6 NOH (?) Eucarvol, Ci H 14 O / 1 Carvacrol. / (-) rf(-h) *(+) ' (+) <*(+) i (-) / (-) r r ^(f) / (+) {(-) ,, + , d(-) + *(+) i (-) l.CH rf(-) r I? ACTIVE ORGANIC COMPOUNDS 33 _ 2 Aliphatic Camphors and Terpenes. Licareol (Aurantiol, lavendol, nerolol, lina- lool), C 10 H 18 0, [(CH 2 = C.CH 3 -CH 2 - CH = CH)(H)C(CH 3 )(CH 2 .CH 2 .OH)]. .. j Coriandrol ( d licareol ) C 10 H 18 O / -f Rhodinol (citronellol ) C 10 H 18 O / -j- 1 OYm/,C 10 H 16 / lonone, C 13 H 20 O / Irone, C 13 H 20 O / + Citronellal, C 10 H 18 O / + 29. Ethereal Oils. Essential Oils. + -r and Angelica Oil Asafoetida Oil Andropogon Oil Basil Carrot Cajeput " Bergamot Copaiba ' ' Cedarwood " Betel Cubeb Eucalyptus Calamus Curled mint Fir needle Caraway Elemi Sandal wood " Cardamom " Frankincense Turpentine " Cascarilla Geranium " Celery Ginger " Chamomile Gurjun balsam " . Chekenleaf " Hemlock Coriander Hemp Costus Juniper Dill Kneepine Fennel Lavender ' ' Lemon ' ' Onion Lime Parsley Mace Rose Marjoram " Rue " Mandarin Silver fir Mastic Storax Muscat Tansy Myrtle Thuja " Orange " Thyme " Orange blossom " Wormseed Para-coto bark " Ylang-ylang Pine needle " Poley Savin Sassafras Spike Star anise " 1 From licareol acetate, mellisa oil or citronella oil. From rose oil. 3 Geraniol. 34 CLASSIFICATION OF ACTIVE SUBSTANCES 30. Resin Acids. + + + + + _L -f + T + + + + + _l_ rf:(-f) rf( + ) rf(+) d(-) <*(-) rf(-f) + / (-) I (-) ' (~) I (-) / ( \ r r r r r r r r i i i i Svlvir arid C H O . A T?1*mi flpiH OHO > A 31. Aromatic Amines. Diphenylethylenediainine (NH 2 )(C 6 H 5 )(H) CPi'WUPTIVNHI E a-Tetrahydronaphthylamine hydrochloride- . W 1-5 Tetrahydronaphthylenediamine hydro- chloride w 32. Alkaloids. a. Alkaloids, of which Several Optical Modifica- tions are Known, and Related Bodies. a-Methylpiperidine (a-pipecoline), C S H IO NCH S / /3-Methylpiperidine (-pipecoline), C 5 H 10 NCH, / a Fthv1rir*riHin* P H VP TT f o-Propylpiperidine, a-conine, C 5 H 10 NC 3 H 7 Conhydrine (conydrine) and pseudoconhy- drine C H NO f Coniceine, C 8 H, 5 N(a- P- and y forms in- Paraconine, C 8 H, 7 N (isopropylpiperidine) - / a-Isobutyl-piperidine C H XC H f Eccronine C H XO livdrorliloridp US Cinnamyl-ecgonine methyl ester (C.^H;. C H ( ) C' H \ T () \ hvflrnrhlorirlp US Benzoyl-ecgonine methyl ester (cocaine), C.w.NO, . r Anhydro-ecgonine, C 9 H, 3 NO 2 from f and Ecgoninic acid. C : H,,N() : ,, from and Tropinic acid, C H H, S NO 4 , from and \* \ ) I (4-) Atropine, C, 7 H 2S NO : , A Hvoscvamine. C..H...NO. . , A ACTIVE ORGANIC COMPOUNDS 35 Pseudohyoscyamine, C, 7 H 23 NO 3 .............. A Hyoscine, C 17 H, 1 NO 4 ........................ A Tropine, C 8 H 15 XO 3 .......................... A b. Alkaloids, of which Only One Optical Modification is Known. Nicotine, C 10 H U N, .......................... / Salts ................................... /r + Solid Alkaloids. The direction of rotation is given for the free bases, dissolved in alcohol or chloroform, and for the salts dissolved in water. Alkaloids which have been found to be inactive are also included. Alkaloids of Cinchona Bark. * * Paricine, C 16 H 1S X 2 O ........ Cinchotenine, C^H^X^ Cinchotenicine, Cinchotenidine, Cinchonine, C 19 H,,X,O /3-and S-Cinchonine, '' Benzoylcinchonine, ll a-Isocinchonine, 1 ,, Cinchoniline, * -Isocinchoirine, ) Cinchonigine, / Apocinchonine , Diapociuchonine, Isoapocinchonine, Apoisocinchonine, Homocinchonine, Pseudocinchonine, Cinchonifine, Cinchonicine, Dicinchonine, Cinchonidine " -and7-Cinchonidine, " Apocinchonidine, " Homocinchonidine, " Apoquinamine, Cincholeuponic acid, a- and /3-Oxycinchonine, .... ....... ..... C 19 H 2a X 3 2 Apoquinidine, Apoquinine, 4- - - Cupreine, C 19 H,,X,O, Chitenine, Ci 9 H 2 ,X, ) O 4 Cinchonamine, Ci 9 H 24 N 2 O Cinchotine, " Hydrocinchonidine, " Quinamine, C 19 H 24 X 2 O 2 Conquinamine, " Quinamicine, " Quinamidine, " Geissospermine, " Quinidine, C 20 H., 4 X 2 O 2 Quinicine, Quinine, Acetyl- and propionylquinine Xitrocamphorquinine Hy droconquinine, C 20 H 26 X 2 O 2 Hydroquinicine, Hydroquinine, Chairamine, C.,,H.^ 2 O t Conchairamine, Chairamidine, Conchairamidine, " Cusconine, C 23 H, 6 X,O 4 Concusconine, Aricine, - CLASSIFICATION OF ACTIVE SUBSTANCES .-llkaloids of Opium. Morphine, C 17 H 19 NO 3 Codeine, C, 8 H 21 NO 3 Methocodeine, Ci 9 H 23 NO s . Pseudomorphine, C^HggN Thebaine, C, 9 H,,XO 3 Narcotine, Laudanosine, C 21 H 27 NO 4 Laudanine, C 20 H 25 XO 4 Cryptopine, C 21 H 23 XO 5 Papaverine, C 20 H 21 XO 4 Narceine, C 23 H 27 NO 8 Pseudonarceine, C 23 H 27 XO 8 Alkaloids of Strychnos Species. Strychnine, C 21 H N < > Chlorstrychnine, C 21 H 21 C1N 2 2 Brucine, C 23 H 26 X 2 O 4 Other Alkaloids. Aconine, Aconitine, C^H^XO^ Isoaconitine, napelline, Lycaconitine, C 27 H 34 N 2 O 6 . Arginine, C 6 H 14 X 4 O 2 ........ Bulbocapnine, C 84 H M N 2 O 7 (?) Corydaline, C 22 H 27 XO 4 ...... Oxyacanthine, C W H, 9 XO :J Pelletierine, C h H, 3 XO ....... Pilocarpine, C 23 H :U X 4 O 4 ..... Quebrachine, C 21 H 2fi N 2 O : , .... Hyoscyamine, C, 7 H 23 NO S ---- Pseudohyoscyaniine, C I7 H 23 N0 3 .............. Hyoscine, C 17 H 2 ,NO 4 ....... Aspidospermine, C T2 H^. 2 O 2 Aspidosperm atine, C r2 H 28 N 2 2 ............. Colchicine, C M H 25 NO H ...... - Echitamine (Ditaine), C 22 H 28 N 2 4 Hydrastine, C 21 H 21 NO 6 -.- Imperialine, C85H80NO4C?). Hydronicotine, C 10 H 16 N 2 . Paytine, C 21 H 24 N 2 O Sparteine, C 15 H 26 N 2 Aribine, C 23 H 20 N 4 Berberine, C 20 H 17 NO 4 Chelerythrine, C, 7 H 13 NO 4 Cevadine, C S2 H 49 NO 9 Delphinine, C J2 H.} 5 NO 6 . Delphinoidine, Ci-jH^N-jO- Hydrastinine, C U H U NO 2 . . Methylhydrastine, Piperine, C 17 H 19 NO 3 Staphisagrine, C, 2 H 33 NO 5 Cytisine (Sophorine), C n H 14 N 2 33. Gluco sides. Apiin, C 27 H 32 1B ........... A o- and 0-Chinovin, CH 4h 8 (?) ............ ./ Coriamyrtin, CsoH.^0,0 ..... A AmyKflalin, C, H, T X() n .... /r Coniferin, C 16 H 22 O B Glycovanillin,C 14 H,.< ). Convallamarin, C, : II,,( ),., Glucoside from iv leaves, W It' . / -:-. Helicin, C, : ,H,,.O 7 A Hesperidin, C,,II, 8 O 12 W Xaringin (Aurantin), C 21 H 2fi O, Phloridzin, C,,H,,( ), A Populin, C, H,,o. W Salicin, C 1;( H, H O 7 W Tetrabutyrylsaponin, c^ii^o.^c.iLo 4 w .... Aa Thevetin, C 54 H H4 () 21 .... The synthetic glucosides and analogous compounds with other sugars are given under the latter head. ACTIVE ORGANIC COMPOUNDS 37 34. Bitter Principles (Santonin Group], Coloring-Matters and Unclassified Compounds. Santonin Group. Santonin, C 15 H 18 O 3 ...... C, A o- and -Metasantonin, C 13 H 18 O 3 .............. A Santonide, C^H^O^ ...... C, A Parasantonide, C 15 H 18 O 3 ---- Metasantonide, C 15 H 18 O 3 . . Santonousacid (and esters), Isosantonous acid (and es- ters), C^H^O,, ....... A Disantonous acid (C 15 H 19 3 ) 2 ........... A Santoninic acid (and salts) C 15 H 20 4 .............. A Santonic acid (and esters), C 15 H 20 4 .............. C Santonyl chloride, C 13 H 19 S C1 ............ C Santonyl bromide, C 13 H 19 8 Br C + Santonyl iodide, C 13 H 19 O 3 I C Parasantonic acid (and es- ters). C 15 H 20 4 C Metasan tonic acid, C 13 H 20 4 C -j- Dehydrophotosantonic acid, C 13 H 20 O 4 A i Photosan tonic acid, C 15 H 22 5 A, C r a-Ethylphotosantonate ... A ft. "< " " A Isophotosan tonic acid, C 15 H 2 ._A A Hydrosantonic acid, C 13 H 22 4 A Echicerin, C 30 H^O, C, E Echitin, C 32 H 52 O 2 C, E Echitein, C 42 H 70 O 2 C, E Echiretin, C 35 H^O 2 E Euphorbon, C 15 H 24 O C Lactucerin, C, S H 44 O 2 (?)... E Lactucol, C 13 H 20 O E Other Bodies. I Quassiin, C 32 H 42 O 10 (?)- - ! Asebotoxin, C 31 H 31 10 (?)- C W, A Picro toxin, C 12 H U O 5 A Erythrocentaurin, C 27 H 24 O 8 A Ostruthin ( C 14 H 17 O 2 ) W A Hematoxylin, C 16 H 14 O 6 . A 35. Bile Acids. Cholanic acid, C 20 H 28 O 6 A Cholalic acid (and salts and esters) , C 24 H 40 O 5 A Desoxycholic acid Isocholanic acid, C 25 H 38 O 7 ... A Dehydrocholic acid, C 25 H 36 O 3 A Bilianic acid, C^H^Og A Glycocholic acid, C 26 H 43 NO 6 and Na salt A a- and /3-Hyoglycocholic acid, i C 26 H 43 NO 5 , Na salt A a- and j9-Hyoglycocholic acid, C 26 H 43 N0 5 , Na salt W Taurocholic acid, C, 6 H 45 NSO 7 and Na salt. W -J- Lithobilic acid, CgoH^Oe - A 38 CLASSIFICATION OF ACTIVE SUBSTANCES 36. Proteid Substances. Egg albumin U' - Syntonin W -f HC1 Serum albumin W Propeptone W Casein, W( -- HClorNaOH) Protalbumose W Serum globulin . NaCl sol Deuteroalbumose W Lactalbumin W Glutin W Chondrin //' \ NaOH Hemielastin W Paralbumin //'-- NaOH Mycoprotein W+ NaOH Heteroalbumose W+ NaCl Fibrin peptone W Elastin peptone H' Vegetable peptone W Fibrinogen W+ NaCl 37- Derivatives of Asymmetric Nitrogen. Methylethylpropylisobutylammonium chloride, N(CH 3 )(C 2 H 5 ) (C S H 7 )(C 4 H 9 )C1....' '. W Compounds of the chloride with PtCl 4 and AuCl :! W Acetic acid salt of the base W Sulphuric acid salt of the base // a-BenzylphenylallylmethyW-camphor sulphonate, C 6 H 5 .CH 2 .N (C 6 H 5 )(C 8 H 5 )CH 8 .S0 3 C 10 H 15 !...."... }\' Iqdide, C 6 H 5 .CH,-N(C 6 H 5 )(C 3 H 5 XCH 3 )I Bromide, C 6 H 5 .CH 2 N(C 6 H 5 )(C 3 H 5 )(CH 3 )Br A 38. Derivatives of Asymmetric Sulphur. rf-Methylethylthetine ^/-camphor sulphonate W rf-Methylethylthetine d-brom camphor sulphonate H (/-Methylethylthetine platinichloride W As the table shows, a great many active bodies are known, and mainly from the animal or vegetable organism, which have been found only in the one form, either right or left rotating. On the other hand, of the synthetically prepared active bodies, the constitution of which has been established, the greater number are already known in the two active modi- fications, and also in the racemic form. An enumeration of the substances given in the tables shows this result : Right rotating only 286 \ Left rotating only 237 I 625 Right and left rotating 102 > Kacemic forms 73 If the salts and c^tt-rs of the active acids and bases, which NATURE OF THE ROTATING POWER 39 were not included, be counted in, the number of active bodies known at the present time [1898] will be over 700. At the time of the appearance of the first edition of this book, 1879, the number was about 300. l On the relation of crystalline form to direction of rotation of bodies of these groups consult 12. III. NATURE OF THE ROTATING POWER 9. Distinction between Crystal Rotation and Liquid Rotation Rotation of Vapors Molecular Rotation. The fact that bodies of the first class exhibit rotating power in the crystalline con- dition only, and lose this power completely when brought into solution, shows that the cause of the rotation must depend on the crystalline structure; that is, on a definite arrangement of molecular groups (crystal molecules) . In fusing or dissolving the body, this is destroyed, and the optical activity disappears. The phenomenon is here, therefore, a purely physical one. Substances of the second and third classes, on the contrary, rotate in the liquid condition. It is probably true of bodies in this condition that the smallest amount of substance acting as a unit, does not consist in single chemical molecules, but in molecular aggregations. There are grounds for believing that, at least in concentrated solutions of a solid body in a liquid, the solid is not completely separated into single molecules or further into its ions, but that molecular groups or aggregations exist. If then, a liquid is found to possess rotating power it is conceivable that the origin of this, as in the case of crystals, should be found in a definite structure of these molecular groups. In this event, the phenomenon would fall in the field of physics. If the cause just suggested is sufficient, it follows that the rotating power of an active substance must disappear as soon as the same is decomposed into single molecules ; that is, as soon as it is brought into the condition of a true vapor. This import- ant experiment was undertaken first by Biot" in the year 1817. He allowed turpentine vapor to pass through a metallic tube 1 Since the above was written nearly 100 new active compounds, mainly esters and complex substitution products, have been added to the list. Tr. (1900). - Biot : Mem. de 1'Acad., 2, 114. 40 NATURE OF THE ROTATING POWER 30 meters in length, and closed at both ends with glass plates, and observed that it possessed the power to produce rotation in tne polarized ray. Exact measurements could not be made as the vapor suddenly became inflamed and destroyed the apparatus. It was not until 1864 that the experiment was repeated, and by Gernez, 1 who, by the aid of excellent instru- ments, determined the rotation of a number of active liquids with increasing temperature and finally in the state of vapor. The substances tested were sweet orange oil ( + ), bitter orange oil (-(-), turpentine oil ( ), and camphor (-J-). Iu all cases the specific rotation, [<*] (that is the rotation cal- culated for unit density and unit length of active layer), decreased with increase of temperature, and finally when the vapor was examined it was found that the specific rotation had decreased to an extent, corresponding to the increase in temperature. In illustration of this, the following numbers obtained from oil of turpentine and camphor are given : Density Observed Length of the State of aggregation. Tempera | referred to angle of observation ture. water. rotation, tubes in deci- d. a. meters. /. Turpentine oil (left rotating). LiQuid r 11 0.8712 J rvS f\ Tnnfi 15-97 T A A n^ 0.5018 36.53 V&oor ] 9 0.7990 1 I,S4 0.7505 168 n nninS? 14.47 13-50 c -?fi 0.50215 0.50237 36.04 35-Si Observed vapor density at Theoretical vapor density . 5-7 168 ^4 . ..11=4 40.61 981. 700. 35-49 Camphor (right rotating). Fused ............ I 204 0.812 31.46 0.5509 70.33 Vapor ............ 220 0.003843 10.98 40.63 70.31 Observed vapor density at 220 = 5.369. Theoretical vapor density ---- = 5.252. The density of the vapor at the temperature of the experi- ments, is, as is readily seen, very nearly the same as the theo- retical density, and it follows from this that single molecules, mainly, and not molecular aggregations, must have acted on the polarized ray. 2 As, moreover, the specific rotation 1 Gernez: Ann. scient. de 1'Ecole norm, sup., i, i. * Ph. A. Guye and P. do Amaral have recently observed (Arch, sc phys. de Geneve [3], 33, 409, 513; Wied. Beibl. 1895, 792, 894) agreement in Die specific rotation OPTICAL THEORY OF CIRCULAR POLARIZATION 41 remains undiminished, the optical activity must be a property inherent in the molecule and must have its origin in the arrange- ment of the atoms in the same. The phenomenon, therefore, belongs in the domain of chemistry. The optical activity of crystals on the one hand, and of liquids on the other, are, accordingly, two quite distinct phenomena, and to indicate that the latter resides in the individual molecule, Biot gave to it the name molecular rotation. But this term has already been applied, as mentioned in 3, to the product of the specific rotation by the molecular weight ; M that is, the quantity [M] = - [] . To avoid confusion it may be better in the latter case to use with the term, molecular rotation, the symbol [M] . 10. The Optical Theory of Circular Polarization in Quartz was first enunciated by Fresnel. 1 This theory assumes that parallel to the principal axis in quartz a peculiar kind of double refraction takes place, and of such a character that a linearly polarized entering ray is decomposed into two rays which move forward in helical paths, one being inclined toward the left and the other toward the right. On leaving the crystal these circularly polarized rays unite to form again a linearly polarized ray, but if they had moved through the crystal medium with unequal velocities, it would follow that the new plane of oscillation would be different from that of the enter- ing ray. It would be turned in the clock-hand direction, that is, to the right, when the polarized ray deviated in the same direction moved with a greater velocity than the other, and vice versa. The existence of these two rays in quartz was first shown experimentally by Fresnel, and later by Stefan, 2 and also by Dove, 3 who found that they are absorbed by colored quartz (amethyst) in unequal proportions. The theory of of a number of amyl derivatives in the liquid and vapor condition. An exception noted that valeraldehyde as vapor, rotates only about half as much as it does as a liquid may be accounted for by the chemical change or racemization which takes place by change of temperature in this substance. 1 Fresnel : Ann. chim.phys. [i], 28, 147. - Stefan : Pogg. Ann., 124, 623. 3 Dove: Pogg. Ann., no, 284. 42 NATURE OF THE ROTATING POWER circular polarization has received a very full mathematical treatment at the hands of many physicists, and for this reference must be made to other works. 1 In regard to the structure, which a crystalline medium must have in order that it may effect a rotation of the plane of polarization, the theory assumes an uneven condensation of the ether around the molecules of the body, and to such an extent that this can not be considered as infinitesimally small as compared with the wave length of the transmitted light. This naturally depends on a definite molecular structure of the substance. The connection of direction of rotation in active crystals with the existence of right or left hemihedral planes has led to the hypothesis, that in these crystals the particles are arranged with reference to each other in the form of a right- or left-handed screw (spiral stair form). This view expressed by Pasteur, 2 Rammelsberg, 3 and others, has received a great degree of probability through an experiment first tried by Reusch, 4 and later followed up by Sohncke. 5 If a number of thin plates of optically biaxial mica (12 to 36) are so placed, one on top of the other, that the principal axis of each one makes always the same angle (45, 60, 90 or 120) with the preceding one, a column is produced which, like an active crystal, has the power of rotating the plane of transmitted polarized light, and either to the right or left as opposed to the direction of the twist in the column of plates. The optical behavior of such mica combinations 6 has been thoroughly studied by Sohncke who found that by sufficiently diminishing the thickness of the mica plates, a combination is secured, the rotating power of which follows exactly the laws that hold for active crystals ; that is, the amount of rotation is pro- portional to the length of the column, and nearly proportional inversely to the square of the wave length. 7 Based on a theory of crystal structure developed by himself, Sohncke has further 1 See Winkelmann: " Handbuch d. I'hysik.," Breslau, 1894, Vol. II, part i, page 784; Ketteler : "Theoretische Optik.," Braunschweig, 1885; Verdet: "kecons d'Optique physique." Pasteur : Consult 12. kninmelsberg: Her. d. chem. Ges., a, 31. Reusch : Pogff. Ann., 138,628. Sohncke : Ibid.. Supplement, 8, 16. These columns may be obtained from Steeg and Renter in Homburg. L. Sohncke : " Theory of Crystal Structure." Leipzig, 1879. OPTICAL CONSTITUTION OF ACTIVE LIQUID SUBSTANCES 43 shown 1 that not only in the trapezohedral-tetartohedral group of the hexagonal system, but also in others of the hexagonal and of the tetragonal and regular systems, a certain spiral- stair arrangement of the crystal particles can exist which is accompanied by rotation of the plane of polarization ( See 5 ) . By means of mica plates of different thicknesses, arranged one upon the other with the axes inclined at different angles, it is possible to imitate the right- and left-handed forms of such active structures. A conception very similar to that of Sohncke, as to the cause of crystal rotation has been developed by Mallard ;~ Wyrouboff also assumes the building up of layers of biaxial plates. 3 ii. Optical Constitution of Active Liquid Substances. For a given thickness of layer, active liquids possess the same rotating power in all directions ; they exhibit, therefore, the same behavior observed in active regular crystals. The property of circular double refraction is inherent in the latter, and if the analogy with active liquids is complete, the same property should be expected in these also. This question was definitely decided, after Dove 4 in 1860 had made some unsuc- cessful experiments, by E. v. Fleischl 5 in 1884, and according to the method of Fresnel, w r ho determined the double refraction of quartz in the direction of its principal axis by the aid of a combination of right- and left-rotating quartz prisms. 6 The apparatus used by v. Fleischl consisted of a glass trough in the shape of a parallelopipedon, 543 mm. long, and 20 mm. wide, open above, and divided by means of glass plates set diagonally into 20 hollow prisms, with refractive angles of 120, and two end prisms with angles of 60. Fig. 3 gives a shortened illus- tration of the arrangement. The 22 compartments were filled alternately with right- and left-rotating 1 Ztschr. fur Krystallog., 19, 529 ; 13, 214 ; 14, 426. 2 Trait6 de Cristallog, a, 313 (1884). : Wyrouboff: Ann. chim. phys. [6], 8, 340; Jour. de. Physik. [2], 5, 258 (1886); Bull. Soc. Min., 13, 215 (1890). 4 Dove : Pogg. Ann., no, 290. " E. v. Fleischl : Wiener Sitzungsber., 90, II, 478 (1884); also Wied. Ann., 24, 127. (> See the text-books of physics. Fig- 3- 44 NATURE OF THE ROTATING POWER substances, which, by proper dilution had been brought to possess exactly the same refractive indices. In a first series of experi- ments, solutions of saccharose and levulose were used, and in a second series, right orange-peel oil and left turpentine oil. When now a ray of light from a very fine opening (pin-hole) was passed through this system of prisms and examined by a reading telescope, two bright spots instead of one, were seen at the other end. If now the decomposition of the original light (ordinary or plane polarized) had followed as in the Fresnel experiment, the two emerging rays must be found circularly polarized, and in opposite directions. It was, in fact, shown by a well known method, employing a quarter wave length mica plate and rotating nicol, that the two rays had been transformed into two linear polarized rays with planes at right angles to each other. In two positions of the nicol, 90 apart, first one and then the other of the bright spots dis- appeared. It is therefore apparent, through these experiments, that the optical cause of activity, that is to say, the manner of the wave motion of the ether, must be the same for liquids as for isotropic crystals. In all directions in both media, two waves are propagated, which are circularly polarized in opposite directions, and which move forward with unequal velocities. It has been mentioned that in such crystals this peculiarity is found, that they possess neither a plane of symmetry nor a center of symmetry, and further that they are found in enan- tiomorphic forms of which the one turns the plane of polari- zation to the right, and the other to the left. The same is to be assumed concerning active liquids, and as here the seat of the activity is found in the single molecules, it follows finally that an asymmetric structure must be assigned to the latter themselves. Tliis conception had been already reached in another way, and through the investigations of Pasteur carried out in 1848, which led to the following conclusions : 12. Investigations of Pasteur. Molecular Asymmetry. As Biot and Seebeck 1 recognized in 1815, common tartaric acid rotates 1 Biot and Seebeck : Bull. Soc. Philom., 1815, 190. INVESTIGATIONS OF PASTEUR 45 to the right, and in 1842 it was further observed by Mitscher- lich 1 that racemic acid, isomeric with tartaric, is inactive. Pas- teur found next, that from a solution of racemic acid, rhom- bic-hemihedral crystals of a double salt having the composition, NH,NaC^H 4 O 6 .4H 2 O, similar to the tartrate, could be obtained by slow concentration at a low temperature. But these crystals are not all identical in crystalline form, for two different structures may be easily recognized. Often the crystals appear developed as illustrated in Figs. 4 and 5, and in this case the differences may be easily recognized even by q 2 ft Fig. 4. Fig. 5- one not specially trained in crystallography. 3 If the crystals are so placed that the two narrow surfaces, q and q 2 , are turned toward the observer, it will be seen on some individuals that the small surface, o 1 , is to the right of q and q 2 (Fig. 4), and on others, it will appear that this surface is to the left of q and q 2 (Fig. 5). There is exhibited here as in the case of quartz, the phenomenon of enantiomorphism, or "non- superposable hemihedry," as it was called by Pasteur ; one crystal figure is the mirror image of the other, and cannot be covered by it. When Pasteur 4 had separated these two kinds of crystals 1 Mitscherlich : Monatsber. der. Berl. Akad., 1842. 2 Pasteur: Ann. chim. phys. [3], 24, 442; Compt. rend. 26, 535; 27, 367, 401, (1848) ; 29, 297 (1849) ; Ann. chim. phys. [3] 28, 56 (1850) ; Compt. rend. 31, 480 (1850) ; 33, 217, 549 (1851) ; Ann. chim. phys. [3], 31, 67 (1851). a Crystals with the same surfaces may appear also in forms, other than those shown ; in such a case, measurements of angles are necessary to distinguish one kind from the other. * An explanation of the manner in which he was led to his discovery is given by Pasteur in his " Recherches sur la dissymelrie moleculaire des produits organiques naturels," Soc. chim. de Paris. Lecons de chimie professes in 1860. Paris 1861. See Alembic Club Reprint, No. 14- 46 NATURE OF THE ROTATING POWER mechanically, he found that those with the surface, o l , to the right of q and q 2 , when dissolved in water and examined in the polariscope, exhibited a right-hand rotation, while those with o 1 to the left showed a left-hand rotation. From the two crystallographically different sodium-ammonium salts he obtained, on the one hand, dextro-, and on the other, levo- tartaric acid, and by mixing equal parts of these in aqueous solution he obtained an inactive liquid which, on evaporation, furnished crystals of racemic acid. Analogous relations were later found among many other active carbon compounds. In this way, it was for the first time shown that an active substance may exist in two forms, right rotating and left rotating, the rotating power being under like conditions the same. From the observations, it was further apparent that the opposite asymmetric characteristics which the two kinds of crystals of sodium-ammonium tartrate possess belong to their molecules also, inasmuch as after solution in water they show right and left rotation. This led Pasteur to the view that the invidual molecules, as all other material objects, in respect to their forms and repetition of identical parts, fall naturally into two classes : i . Those which are superposable on their mirror images (as a straight-stair, a cube) ; 2. Those, whose mirror images can not be covered by the originals and which may appear in two oppositely constructed (enantiomorphic) forms (spiral-stair, irregular tetrahedron, right and left screw, right and left hand ) . Molecules of the first class possess a symmetrical structure ; in the second the atoms are asymmetrically ordered, and these should show optical activity. In relation to racemic acid, and the two tartaric acids Pasteur 1 remarked : ' ' Are the atoms of the right acid grouped in the form of a dextrogyrate helix, or do they stand at the corners of an irregular tetrahedron, or are they found ar- ranged in some other asymmetric form ? We are not able to an- swer these questions. But of this there can be no doubt: That an i metric arrangement of the atoms must exist in such a manner as would furnish a non-superposable image. It is just ertain that the atoms of the left acid are arranged in a manner exactly the reverse of those in the right, and finally 1 " Recherches ur la dissymetrie moleculaire," Alembic Club Reprint, p. 24. THEORY OF VAX 'T HOFF AND LE BEL 47 we know that racemic acid results from the combination of these two inversely asymmetric atomic groups." Through these considerations, Pasteur introduced a new conception, that of molecular asymmetry, into the science. Before, however, this could bear fruit, a much wider develop- ment of organic chemistry was necessary, and only after the constitutional formulas of a large number of carbon compounds had been determined, was it found possible to trace a con- nection between the atomic structure of molecules and their optical activity. 1 IV. RELATIONS BETWEEN ROTATING POWER AND CHEMICAL CONSTITUTION OF CARBON COMPOUNDS 13. Van't Hoff-LeBel Theory. One of the most important advances in our knowledge of optical rotation was made in 1874, when J. H. van 't Hoff,"' then in Utrecht, and a few weeks later J. A. LeBel, 3 in Paris, furnished the proof that optical activity has a definite connection with the structure of carbon compounds. The fundamental conception founded on this notion, to which van't Hoff was led through the assumption of a tetrahedral arrangement of the atoms, Le Bel, on the 1 An essentially different hypothesis to account for the activity of liquids as well as crystals, which is based on the assumption of the rotation of the molecules, has been proposed by Fock (Ber. d. chem. Ges.. 24, 101). Wyrouboff (Ann. chim. phys. [7], i, 5 ; Chem. Centrbl., i, 260 (1894)) has sought to show that the rotation of crystalline organic substances in solution bears relation to the crystalline structure, and is not merely dependent on the nature of the chemical molecule. First published in the paper : Voorstel tot uitbreiding der tegenwoordig in de scheikunde gebruikte structur-formules in de ruirnte ; benevens en daarmee samen- hangende optnerking omtrent het verband tusschen optisch actief vermogen en chemische constitutie vanorganischeverbindingen; Utrecht, iS;4. At the end the pa per is signed September 5, 1874, J. H. van 't Hoff. An abstract from this article appeared in 1875 in Bull. Soc. Chim. [2], 23, 295. Then followed : i. La chimie dans 1'espace, par J. H. van 't Hoff, Rotterdam, 1875 : 2 - Die Lagerung der Atome im Raurn, a German translation of the last by Dr. F. Hermann, Braunschweig 1877; 3. Dix annees dans 1'histoire d'une theorie, par J. H. van 't Hoff, Rotterdam 1887.; 4. Stereochimie, by W. Meyerhofer, a German edition,, essentially of the Dix annees, etc., Leipzig and Vienna, 1892 ; 5. Die Lageruug der Atorne im Raum, von J. H. van 't Hoff, 2nd. ed., Braunschweig, 1894. 3 Le Bel : First paper : Sur les relations qui existent entre les formules atomiques des corps organiques et le pouvoir rotatoire de leurs dissolutions. Bull. Soc. Chim., [2] 22,337, November number, 1874. Then following papers: Bull. Soc. Chim. [2], 23. 338 (2875) ; 25, 546 (1876) ; 27, 444 (1877) ; 33, 106 (1880) ; 37, 300 (1882) ; [3!, 7, 164 ; 8,613(1892); Compt. rend., 89, 312 (1879); 92, 843, (1881) ; no, 144 (1890); 112, 724 (1891) ; 114, 504,417 (1892). 48 ROTATING POWER AND CHEMICAL CONSTITUTION contrary, through Pasteur's idea of molecular dissymmetry, found gradually decided confirmation through experiment, and, as is well known, the later developments of the theory, especially the views advanced by van 't Hoff on the arrange- ment of the atoms in space, have created a new epoch in the science, that of stereochemistry. In this book we are concerned only with those portions of the van't Hoff-LeBel theory which are directly connected with optical activity, and these will be but briefly discussed as they are found explained in all text-books of stereochemistry and organic chemistry. The fundamental points in the theory are as follows : i. Consider in a compound of the type, CR 4 , the carbon atom situated in the center, and the four elements or groups joined to it situated at the corners of a tetrahedron, then in case the four groups are all different, the resulting solid formula CCR^RgRJ will possess no plane of symmetry, and must exist in two non-superposable forms of which one is the mirror image of the other. According to this view, everybody whose structural formula possesses a so-called asymmetric carbon atom, that is, one which is combined with four differ- ent atoms or groups, must be optically active and appear in a right- and left- rotating form of equal rotating pow r er. Experience has shown further that equal weights of the two modifications can unite to form an inactive compound or mixture (racemic body) which by various means can be split up into the active components. Asymmetric carbon atoms (*C) can appear in all direct methane derivatives, and chain structure molecules ; in benzene derivatives they can exist only in the side chains, but in hydrated cyclic compounds also in the nucleus. Examples are found in the list of active substances given in 8 ; a few other cases may be referred to here, from which it will be seen that of the four radicals, two may be combined between them- selves (propylene oxide), or one of the same with two different asymmetric carbon atoms (phenoxacrylic acid) ; further, that the asymmetry of a carbon atom may depend on remotely situated groups, and not necessarily on those imme- diately connected (limonene, menthene) : THEORY OF VAN 'T HOFF AND LE BEL 49 Limonene. Propylene oxide. CH 3 x ,CH 2 Menthene. CH O f H CH, / \ I *PW *O TT/ Xr^TT \-jn~ v, H CH 2 /x /x H 2 C CH 2 H 2 C CH 2 Phenoxacrylic acid. C 6 H 5V /C0 2 H HC CH 2 H 2 C CH Vc *c< \\/ " \^ H/ X/ X N C C O I I CH 3 C 3 H 7 2. In substances which contain two asymmetric carbon atoms and whose molecules, like that of tartaric acid, CO 2 H *CHOH *CHOH CO 2 H, are built up of two similar halves, there must be, according as these halves show the same or opposite rotations, besides the right- and left-rotating forms, a third inactive form depending on this intramolecular com- pensation, and which cannot be resolved into active com- ponents. Such an inactive form is not possible when the half molecules are dissimilarly constructed, but in this case, four active isomers may be expected, each two possessing equally strong, but oppositely directed rotations. If a compound con- tains several asymmetric carbon atoms, by addition or sub- traction of the effects of the single groups, a large number of unequally strong active modifications may result, two of which again in each case belong together as antipodes, and finally the existence of some definite number of inactive compensation forms may also be expected. In all such cases, consideration will show how many of these optical isomers^must exist when the structural formula of the substance is known. The method of making such a computation will be shown in the next chapter on Optical Modifications. With the ethylene derivatives having four different radicals, R 1 R 2 C=CR 3 R 4 , the four groups must lie in one plane if we consider the carbon atoms united by an edge of the tetra- hedrons containing them, and no asymmetry is possible. In fact, all ethylene derivatives have been found to be inactive, 1 1 Le Bel (Bull. Soc. Chim. [3], 8, 613) had considered optical activity possible in unsaturated compounds, and Perkin (Jour. Chem. Soc., 53, 695) believed he found this in chlorfumaric and chlormaleic acids, CO 2 H CC1=CH CO 2 H. Walden, how- ever, showed the error in these observations (Ber. d. chem. Ges., 26, 210). 4 50 ROTATING POWER AND CHEMICAL CONSTITUTION even when made from active compounds, as for example, fumaric and maleic acids from malic acid, bromcinnamic acid, C 6 H 5 CBr=CHCO. 2 H, from the dibromide, C 6 H 5 CHBr CHBr.CO,H, and others. Likewise, no asym- metry is possible when in bodies of the type, R,R 2 C=CR 3 R 4 an even number of doubly linked carbon atoms is introduced. But, on the other hand, as van 't Hoff remarked, 1 asymmetry and optical activity appear if the number of added carbon atoms is uneven, inasmuch as the four radicals then stand crossed, as is the case in the tetrahedron. The simplest bodies of this kind would be the propadiene (allene) deriva- tives ; observations on such substances are wanting as yet. As van 't Hoff pointed out, 2 cyclic compounds present cer- tain definite conditions of asymmetry, and to begin with, we have the derivatives of tri- and tetramethylene, but active bodies belonging here are not yet known. But many such appear in the six member rings, that is in the di-, tri- and hexa-hydrated benzene derivatives. Among the last inositol, :i /CH.OH CH.OH, CH.OH< >CH.OH, X CH.OH CH.OH/ offers an example in which the existence of asymmetric carbon is not apparent from the formula, and in which the asymmetry and the mirror image form appear only when the position of the H and OH above and below the plane of the carbon ring, that is to say, the cis and trans isomerism, is taken into consideration; anything further concerning this belongs in the field of stereochemistry. Benzene derivatives which are not hydrides can hold asymmetric carbon atoms in the side chains. Confirmation of the van 't Hoff -Le Bel theory has come gradually, and in many different ways. It has been found that without exception, activity is connected with the pres- ence of asymmetric carbon, and that in bodies in which this is lacking, rotating power is not found. For a number of bodies of the last class, such as N-propyl alcohol, styrol, /2-picoline and others, in which activity was claimed, it was found that 1 " Lagerung der Atome im Raum," 2nd. ed. (1894), pp. 68 to 76. * Loc. cit.. p. 83 1094. * See Bouveault : Bull. Soc. Chim. [3], n, 144 (1894). THEORY OF VAN 'T HOFF AND LE BEL 51 this assertion was an error. Through direct experiments, the appearance or disappearance of optical activity by formation or destruction of asymmetric carbon atoms was further shown. Le Bel first proved this by the conversion of active amyl iodide, CH 3 .HC*.C 2 H 5 .CH.J, into inactive methyldiethyl methane, CH 3 .C.H.C 2 H 5 .C 2 IV Then Just 2 obtained from the same amyl iodide, by action of zinc and hydrochloric acid inactive dimethylethyl methane, (CH 3 ) 2 .C 2 H 5 .C.H, but by action of ethyl iodide and sodium he obtained active methyl- ethylpropyl methane, CH 3 .C 2 H 5 .C 3 H..*C.H ; alsb, by heating with sodium, active diamyl, C 2 H 5 .CH 3 .H.*C. CH 2 CH 2 . *C.H.CH 3 .C 2 H 5 . Further, it was shown that for the existence of optical activity the nature of the four radicals combined with the asymmetric carbon atom is a matter of no consequence. It was formerly observed that the introduction of a halogen led often to a dis- appearance of activity ; thus from left-rotating malic acid, inactive bromsuccinic acid, (Kekule), 3 from left- rotating mandelic acid, inactive phenylbromacetic acid, C 6 H 5 .*CHBr.CO,H ( Easterfield ) / and iromd- and /-isopropyl- phenylglycolic acid, inactive isopropylphenylchloracetic acid, (C 6 H 4 .C 3 H 7 )(H)^C(Cl)(CO 2 H),(Fileti), 5 were obtained. As was later found, the cause of the inactivity of these products lay in the fact that racemic forms w r ere produced by reason of the high reaction temperature. By keeping this as low as possible, these halogen bodies were obtained in rotating con- dition. This was shown particularly by Walden, who prepared an active chlorsuccinic acid from malic acid by action of phos- phorus pentachloride with addition of chloroform, and later from sarcolactic acid, ethyl tartrate, and mandelic acid, a large number of chlorine and bromine derivatives, such as methyl chlorpropionate, ethyl brommalate, phenylchloracetic acid, and others which all possessed optical activity. 6 Finally, the 1 Le Bel : Bull. Soc. Chitn. [2], 25, 546 (1876). 2 Just : Ann. Chem. (Liebig), 220, 146 (1883). * Kekule : Ann. Chem. (Liebig), 130, 25 (1864). 4 Easterfield : Jour. Chem. Soc., 59, 75 (1891). 5 Fileti : Gazz. Chim., 22, II, 405; J. prakt. Chem. [2], 46, 562. 6 Walden : Ber. d. chem. Ges., 26, 214 (1893). See further Le Bel : Bull. Soc. Chim., 9, 674 (1893) and Ber. d. chem. Ges., 28, 1923 (1895) ; also Walden : Ber. d. chem. Ges., 28, 2766. 52 ROTATING POWER AND CHEMICAL CONSTITUTION fact was fully explained why many bodies exist which contain asymmetric carbon atoms, but are nevertheless inactive. In some cases it was shown that they are racemic forms, inasmuch as they are resolvable into active compounds ; in other cases, as mesotartaric acid, dulcitol ; and mucic acid, they contain two similarly constituted halves, and the inactivity follows from the opposite rotating power of these. In a third group of asymmetric substances it was found that they possess a very weak rotating power, and in order to recognize this, either a very long column must betaken, or, as in the case of mannitol, some indifferent substance (boric acid) must be added to increase it. It appears therefore, that in all cases the views of van 't Hoff and Le Bel are found to agree with experience, and that when apparently a contradiction was found (limonene), 1 later investigations removed this. The doctrine of asym- metric carbon atoms may be looked upon as one of the best established of chemical theories. 14. Asymmetric Nitrogen and Sulphur. Compounds of triad nitrogen with radicals different from each other appear always to be inactive ; attempts to split up the tartaric acid salts of ethylbenzylamine (Kraft),-' benzyl hydroxylamine (Behrend and Konig), 3 methyl aniline, tetrahydroquinoline, and tetra- hydropyridine (Ladenburg), 4 have led to no result. On the other hand, an active compound of pentavalent nitrogen, isobutylpropylmethylethylammonium chloride has been obtained. The inactive salt directly obtained was split up by Le Bel 5 by aid of the fungus culture method, and a left-rotating chloride ([]= -7 to 8) was obtained, which was further converted into active chlorplatinate, chlor- mercurate, and acetate. The chloraurate, rotating very feebly to the left, became dextrorotatory after addition of hydro- chloric acid. The sulphate was found to be inactive. More recently another active compound of pentavalent nitrogen has been produced. Wedekind 6 attempted to resolve von Baeyer : Ber. d. chem. Gcs., 37, 436; Tiemann and Semmler : /bid., 28, 3495. Kraft : Ber. d. chem. Ges., 23, 2780 (1890). Behrend and Konif? : Ann. Chem. (Liehi^), 263, 184 (1891). Ladenburg : Ber. d. chem. Ges., 26, 864 (1893). LeBel : Compt. rend., 112, 724 (1891). Wedekind : Ber. d. chem. Ges., 32, 517. ASYMMETRIC NITROGEN AND SULPHUR 53 a'-benzylphenylallylmethylammonium hydroxide by com- bination with tartaric and camphoric acids, but without success. Pope and Peachey, however, by using the much stronger dextrocamphor sulphonic acid, which they have applied in several other cases, succeeded in effecting a perfect resolution. 1 They mixed the iodide of <*-benzylphenylallylmethyl- ammonium with the silver salt of dextrocamphor sulphonic acid in molecular proportion, and boiled in a mixture of acetone and ethyl acetate. After separating silver iodide by filtration, the liquid left deposited, on cooling, a crystalline mass of the dextro- and levo-benzylphenylallylmethyl- ammonium dextrocamphor sulphonates. This was crystallized from acetone, the less soluble dextro constituent being readily obtained in colorless plates melting at 170, and giving \_a~\ D -- -f- 44.4. For the /-salt separated from the mother- liquors in less pure form, the rotation [tf]/>= - 18.6 was found. By double decomposition of the camphor sulphonates, C 6 H 5 .CH 2 .N(C 6 H 5 )(C 3 H 5 )(CH 3 ).S0 3 .C 10 H 15 0, the authors obtained the d- and /-iodides and bromides, C 6 H 5 .CH 2 .N(C 6 H 5 )(C 3 H 5 )(CH 3 )I, or Br. In recent years many attempts have been made to resolve asymmetric racemic sulphur compounds, but for a time with- out success. See for example the work of Aschan. 2 Very lately, however, and just as this translation is going to the press, Pope and Peachey" have applied their camphor sulphonic acid process to the resolution of methylethyl- thetine and have succeeded in separating an active body with the sulphur as the asymmetric element. The authors conclude from their work that a large number of other elements may be found to behave as asymmetric centers of optical activity. Some details of their process will be given in a following chapter. Ammonium derivatives containing two similar radicals as the chlorides of dimethylethylpropyl-, methylethyldipropyl-, ethyldipropylisobutyl-, and ethylpropyldiisobutylammonium 1 Pope and Peachey : J. Chem. Soc., 75, 1127. - Aschan : Ber. d. chem. Ges., 32, 988. 3 Pope and Peachey : J. Chem. Soc., 77, 1072. 54 OPTICAL MODIFICATIONS cannot be converted, as Le Bel 1 found, into active forms by the action of fungi. Further consideration of the subject of asymmetric nitrogen belongs in the field of stereochemistry. V. OPTICAL MODIFICATIONS 15. The fact that a body can exist in a right rotating, a left rotating, and an inactive form was first recognized, as mentioned, by Pasteur in 1848, in the case of tartaric acid. The number of bodies acting similarly was increased very slowly, and in 1879, at the time of the publication of the first edition of this book, only three other examples could be given; mz., malic acid, camphor, and camphoric acid. Already in 1875, van 't Hoff, in his " Chimie dansPespace," had developed the general formulas by which the number of possible stereoisomers and hence, also, optical isomers of a body could be calculated from the number of asymmetric carbon atoms in its molecule. For a long time the observations available, from which these formulas could be tested, were entirely too scanty, and only in the last few years, the great investigations of E. Fischer, on the members of the sugar group, have furnished material which demonstrated com- pletely the correctness of the theoretical predictions. Obser- vations were multiplied also, in other classes of compounds and as the table of active substances given in 8 shows, there are now over 100 such bodies known in different optical modifications. However, there is still a very large number of active bodies, over 300 in fact, which are known only in one form, some right, some left rotating. This is true of whole groups of bodies as the polysaccharides, natural glucosides, starch varieties, alkaloids, bitter principles, bile acids, and proteids. Without doubt, most of these contain several asymmetric carbon atoms, and must exist in different forms with different rotating powers as well as in inactive modifications. 1 Le Bel : Compt. rend., na, 724 CALCULATION OF NUMBER 55 A. Calculation of the Number of Optical Modifications of a Compound from the Number of Asymmetric Carbon Atoms Contained in It. 16. If we divide the compounds consisting of a chain of singly linked carbon atoms into the three classes given below, the number of possible stereoisomers or optically active and inactive forms is shown in the following expressions in which : n = the number of asymmetric carbon atoms in the compound. N the whole number of possible isomers, which are divided into i 3= inactive, non-separable modifications, and a = active forms, which occur in pairs as optical antipodes with equally strong opposite rotations. These lead to r inactive separable racemic modifications. First class : n even or odd. Structural formula not in two equal halves. If RR V represent the terminal radicals, and a, d, the radicals combined to the middle carbon atoms, the general type is For example : Malic acids CO 2 H *CH.OH CH 2 CO 2 H Phenyl-tf-chlorlactic acids C 6 H 5 *CH.OH *CHC1 CO 2 H Pentoses CH 2 OH *CH.OH *CH.OH *CH.OH CHO Hexonic acids CH 2 OH (*CH.OH) 4 CO 2 H Bodies belong here in th chain of which there is at some point, a carbon atom which is not asymmetric (CH 2 , CO 2 , CO). For example : Butylchloralaldol CH 3 *CHC1 CC1 2 *CH.OH *CH(CHO) *CH.OH CH 3 In all such cases we have (I) 1 N = 2 n a = 2 i == o. 1 The expressions (I) and (II) were first proposed by van 't Hoff (" I^a chimie dans e," 1875, p. 9 and 12) and the second one is also found in this form : n N= - 2 "- 22 or2^- 2 Attention was first called to formula (III) by E. Fischer (Ann. Chem. (Uebig), 270, 67 (1891)). These formulas are derived from the theorems on permutation and combinations, attention being paid to the conditions obtaining for various reversed and reflected image forms. OPTICAL MODIFICATIONS Therefore : n = i 2 3 4 5 6 N = a = 2 4 8 16 3 2 64 r = I 2 4 8 16 32 Second class : n even. Structural formula in two equal halves. The type is R(*C*l>) t , vv .-R For example : Tartaric acid CO,H *CH.OH *CH.OH CO 2 H Syra. dimethylsucciuic acid . . . . CO 2 H *CH(CH 3 ) *CH(CH 3 ) CO a H Hydrobenzoin C 6 H 5 *CH.OH *CH.OH C 6 H 5 Hexitols CH,OH ( *CH.OH) 4 CH..OH Tetraoxydicarboxylic acid CO 2 H (*CH.OH) 4 CO,H. Bodies are found here with symmetrically halved structural formulas which contain in the middle, an even number of non-asymmetric carbon atoms. For example : DimethyladipicacMs, CO 2 H-*CH(CH 8 )-CH 2 -CH 2 -*CH(CH 8 )-CO 2 H Diallylbromides, CH,Br *CHBr CH, CH, *CHBr CH,Br. We have in the second class : (II) a = 2' Therefore, when "I " 4 6 8 <=] i= I r ~ \ 3 2 I I 10 8 2 4 36 32 4 16 136 128 8 64 Third Class : n uneven. Structural formula equally halved, after excluding the middle carbon group. The type is * (*CaJ) 3 , 5l7 ,... R Examples : Trioxyglutaric acids, CO 2 H *CH.OH CH.OH *CH.OH CO a H a-Glucoheptitol, CH,.OH (*CH.OH), CH.OH-(*CH.OH) 2 -CH. 2 .OH CH :< H CH 3 Dimethyltricarballylic acid H *C C *C 1 1 CO a H C0 2 H CO,H CALCULATION OF NUMBER 57 In these cases C, the middle atom of the chain, is : Asymmetric (active) when the other parts of the chain, the equal halves, are asymmetric similarly, that is, have the same direction of rotation ; Symmetric (inactive) when the two other parts of the chain are oppositely asymmetric, and therefore neutralize each other in their rotating power. Under both circumstances when the middle atom, C. is included in the number n, we have the following formula for calculating the isomers : (III) N=2 l 1= 2 n i 2 . From this it follows that for : n = 3 5 7 9 N = 4 ]6 64 256 a = 2 12 56 240 i = 2 4 8 16 r I 6 28 120 In deriving the different stereoisomers of a body, it is con- venient to employ the method of representation proposed by E. Fischer, 1 which consists in this, that the solid model of the molecule (built up by the aid of the well-known rubber carbon atom models) is placed in such a manner over the plane of the paper that all the carbon atoms are found in a straight line, and the radicals combined with them (H and OH ) , stand to the right and left above the plane. Then the projection of such a structure, for example, that of X X x x X X X X /?! R-L The Inactive Non- Separable Modifications are distinguished in the following configuration formulas by this, that the latter may always be cut by a horizontal line into two equal halves of which the lower one is the mirror image of the upper. The compensation existing within the molecule is illustrated by this, fpr if in Diagram I, below, a spiral be drawn through the four radicals in the direction, * ^ R ( x &) in the u PP er half of the figure it will be turned to the right, and in the lower to the left. In compounds which have in the chain an uneven number of carbon atoms, the cut passes through the middle (not asymmetric) one. Such inactive molecules may be represented by the following diagrams : I R X 2 R X X 3 R X X 4 R x x X 5 R X X X. x R X x R x R x x R X x R In what follows the derivation of the possible optical modi- fications for chain structure molecules with n i, 2, 3, 4, 5 will be carried through and illustrated. I. n=i. First Class : N= a = 2 i= o r= i For example : CH, CH 3 CH 2 .COOH CH 2 .COOH H C OH HO C H H C NH 2 NH 2 C H III I CH 2 OH UI,OH COOH COOH Right and left propyleneglycol. Right and left aspartic acid. CALCULATION OF NUMBER 59 II. n = 2. A. First Class: N = a = 4 i = o r = 2 The four possible active combinations, of which each pair form antipodes, are: 12 34 R R R R x x x x x x x x An example of this is furnished by cinnamic acid dibromide of which the four active as well as the two racemic forms are known: C 6 H 5 C 6 H 5 C 6 H 5 BrH HBr HBr BrH BrH HBr BrH HBr COOH COOH COOH COOH Which of these configurations belongs to each isomer has not been established. The same conditions must appear with: Phenyl-a-chlorlactic acid C 6 H 5 CH.OH CHC1 CO 2 H, Phenyl-jS-chlorlactic acid C 6 H 5 CHC1 CH.OH CO 2 H, Trioxybutyric acids CH 2 OH CH.OH CH.OH CO 2 H, etc. B. Second Class: N 3 a = 2 t = i r = i If R = R l of the four combinations given under A, i and 2 will be identical, as can be shown by rotating the diagram (turning it upside down), and there remain: R R R x x x x x -x R R R Inactive. Oppositely active. 6o OPTICAL MODIFICATIONS Example : COOH COOH COOH 1 HOH HOH HOH HOH HOH HOH COOH COOH COOH Meso- Right Left tartaric acid, tartaric acid, tartartic acid. III. n == 3. A. First Class: N= a = 8 We have: = o -f- tartaric acid tartaric acid Racemic acid. r== 4 I 2 34 56 78 R R R / ? R R R R x X , X X X -X X X x x X X X X X X X* X x x x x x x R, R, A> /> T) r> r> r> 1 'l **\ **! -*M **| ^"" ' ' V J Example: RI R Pentoses .. . .CH OH (Cl H.OH) 3 CHO Pentonic acids . . . -CH OH (C] S.OH) 3 COOH I 2 3 4 5 6 7 8 R R R R R R R R HOH H OH HOH HO H HOH H HOH HOH HOH H HOH H OH OH HOH II HOH H OH HOH HO OH HOH H HOH HOH HOH HOH A, j t| R l J\ 'i ^i ^i A '\ ^i /-Ribose. Unknown. /-Arabinose. rf-Arabi-J Unknown. /-Xylose. Lyxose. Unknown. /-Ribonic /-Arabonic nose. /-Xylonic lyyxonic acid. acid. acid. acid. B. Third Class: TV" == 4 a 2 1 = 2 r = l. Of the configurations given under A the following are identical when R = R r - i with 2 4 with 7 3 " 8 5 " 6 CALCULATION OF NUMBER 6l and there remain: 5 R x x X- R Inactive. Oppositely active. Inactive. I 3 4 R R R x X X x x X x . X X R R R Example : Pentitols Trioxyglutaric acids . R R R HOH HOH HOH HOH HOH HOH HOH HOH HOH R Adonitol. Ribotrioxy- glutaric acid. Inactive. # R /-Arabitol. /-Trioxy- glutaric acid. Unknown. CH,OH (CH.OH) 3 CH 2 OH COOH (CH.OH) 3 COOH R HOH HOH HOH R Xylitol. Xylotrioxy- glutaric acid. Inactive. IV. n 4. A. First Class: N = a == 16, i = o, r = i 2 3 4 5 6 7 8 R R R R R j? J? R x X X x x X x X x X x X X x x X X- X X- X x X X X x X x X X* X x X R\ *1 ^1 Kl *, ^1 ^1 *, 9 10 II 12 13 14 15 16 y? R ^ A R R R /? x X X x X X X x x X x x X X X x. X X X X X x X X ^1 X *, X #, X *, x X X *1 x x* 62 OPTICAL MODIFICATIONS Of these sixteen active configurations, in the R Hexoses .................. CH 2 OH (CH.OH), CHO and Hexonic acids ............. CH 2 OH (CH.OH), COOH the following are known: 5 6 7 8 9 R R R R R HOH H OH HOH HOH HOH HOH HO H HOH HOH HOH HOH H OH HOH HOH HOH HOH HOH HOH HOH HOH Glucose, gluconic acid. d I d Talose, Gulose, gulonic acid, talonic acid. II 12 13 M 15 16 R R R R R R HOH HOH HOH HOH j HJOH HOH HOII HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH #* R\ ^i #1 ^i * Mannose, mannonic acid. Idose, idonic acid. B. Second Class: N= 10 a = 8 I = 2 Galactose, galatonic acid. = 4 This number of isomers follows from the configurations given under A when R is taken equal to /? p because then there become identical: i with 2 3 " 10 4 " 9 5 with 8 6 " 7 15 " 16 CALCULATION OF NUMBER There remain then: i (3, 4) (5, 6) (u, 12) (13, 14) 15. We have here Hexitols CH 2 OH (CH.OH) 4 CH 2 OH and Saccharic acids COOH (CH.OH) 4 COOH, of the last of which the ten forms are known : I R 3 4 R 1 5 R 6 R HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH R Inactive Allo- mucic acid. R I R d R R I d Sorbitol. Saccharic acid. Talitol. Talomucic acid. ii R 12 R 13 R 1 14 R 15 R HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH OHH HOH HOH HOH HOH HOH R I R d R d R I R Inactive. Dulcitol. Mucic acid. Mannitol. Mannosaccharic acid. Iditol. Idosaccharic acid. V. = 5 . A. First Class : N == a = 32 i = o r = 16. The thirty-two possible comb inations are as follows, leaving out the symbols, R and R l : I 2 3 4 5 6 7 8 x X X x x X x X x X x X X x x X x X x X x X X x x X x X X X x^ X x X x X x X x X 6 4 OPTICAL MODIFICATIONS 9 io X- X x x x x X X x x II 12 x x x x x -x X X 13 H X X X X x x x x >o x 21 22 x x X X X X x x 29 30 X X X oX X X x x X X 15 16 X X x x X X x x X* X 23 24 x ^x X X x x X X x x 31 32 x x x x x x X X X X 17 18 X X x x x x X X x x 19 20 X X x x x x x x X X 25 26 x x X X x x x x X X 27 28 x x x x X X X X x x Of bodies belonging here, but ing to : 8 8 CHO COOH . HOH HOH few are known, correspond- 16 16 CHO COOH HOH HOH HOH HOH HOH HOH HOH HOH HOH HOH 1 1 OH HOH HOH HOH 1 1 OH HOH HOH HOH CO.;OH o-Gluco- heptose. CH./)H a-Gluco- heptonic acid. CH 2 OH /3-Gluco- heptose. CH 2 OH jS-Gluco- heptonic acid. B. Third Class : N - - 1 6 12 1 = 4. r -- 6 When R -- R are identical : I with 2 3 " 12 4 " ii 5 " TO 6 " 9 7 " 8 of the forms given under A, the following 13 with 32 1 8 with 25 M " 3i 19 " 20 15 " 30 21 " 28 16 " 29 22 " 27 17 " 26 23 " 24 CALCULATION OF NUMBER 65 There remain then, as can be calculated from the formulas, sixteen configurations ; viz. , six active pairs and four inactive forms. These are : I 3 4 5 6 7 13 14 x X x x X x^ X x x X X X x x X x X X X x X X x X x x X X X x x X X x X X X x x X Inactive. Active. Active. Inactive. Active. 15 16 17 18 19 21 22 23 X x X x X x X x x X x X x X x X X x x X x X x x x X X x x x X X x v. X x ^ X X x X x Active. Active. Inactive. Active. Inactive. Of such configurations, the following are known : 7 16 COOH COOH HO H HO H HOH HOH HOH HOH HOH HOH HOH HOH COOH COOH a-Glucopentoxy- p-Glucopentoxy- pimelic acid. pimelic acid. Inactive. Active. A discussion of the methods and considerations leading to the determination of the positions O and OH as they have been developed by E. Fischer for the bodies of the sugar group, does not fall within the plan of this book. Asymmetric Molecules with Chain Structure which contain a double carbon linkage must, in consequence of the resulting cis-trans isomerism, show d- and /-modifications for each of the two forms. According to Walden, 1 who called attention to 1 Walden : Ber. d. chem. Ge Transform. J H.C.(CH a ) 7 .COOH Right-rotating ) CH V (CH 2 ) 5 .*CH(OH).CH 2 .CH ricinoleic acid. V Cisform. j COOH.(CH,) 7 .C.H The left-rotating antipodes are as yet unknown. Asymmetric bodies having a triple carbon linkage can yield only one pair of optical antipodes ; thus, from the above acids only a single well known, the structure of camphoric acid has not been finally settled ; the above relations point to the existence of two asymmetric carbon atoms. (Bredt's formula contains three, those of Tiemann and Semmler four, *C.) The same conditions appear to obtain with other cyclic compounds, as, for example, borneol, the composition of which i^ not definitely known. Further details concerning the relations of optical modi- fications will be found in the chapter on the "Development of Active Isomers." 1 Goldsobcl : Her. d. chem. Ges., 37, 3121. * Aschan : Ibid., 37, 2001. PHYSICAL AND CHEMICAL BEHAVIOR 67 B. Physical and Chemical Behavior of the Optical Modifications a. BEHAVIOR OF THE ANTIPODES 17. Physical Properties. Of these, only such can be different for the two antipodes in which the contrast of -f and is essentially inherent. Besides optical right and left rotation, and the enantiomorphism of crystals, \.\\e pyroelectricity of the latter belongs here. The phenomenon of opposite electrical poles in a certain class of crystals is disclosed only when they are heated, or cooled, or subjected to one-sided strain or pressure. Among such, which in solid or dissolved condition show optical activity, the following have been found to exhibit pyroelectricity : Hexagonal System : quartz, potassium-lithium sulphate, sodium-lithium sulphate, potassium bromate, potassium peri- odate, ^-antimonyl-strontium tartrate, aMead tartrate. Tetragonal System : ^-antimonyl-barium tartrate. Monoclinic System : d- and /-tartaric acid, ^-potassium, ammonium and strontium tartrates, cane-sugar, milk-sugar, d- and /-carvoxime, d- and /-fenchone oxime, d- and l-a- carvone pentabromide. For further information on the subject of pyroelectricity of crystals, the reader is referred to the excellent discussion of this condition in Liebisch's C 4 H 4 O 6 2H 2 O iH 2 O Sodium tartrate and racemate, Xa,C 4 H 4 O 6 2H.,O 2H 2 O Thallium tartrate and racemate, T1,C 4 H 4 O 6 -JH 2 O Sod. -ammon. tartrate and racemate, Xa XH 4 )C 4 H 4 O 6 -f H.,O 4H 4 O Potass, lithium tartrate and racemate, KLiC 4 H 4 O H H.,O H,,O Potass, antimon. tartrate and racemate, KSbOC 4 H 4 O 6 - |H 2 O 2 H 2 O Calcium mannonate, Ca(C 6 H H O 7 ) 2 - 2H. 2 O 1 Strontium glycerate, Sr(C 3 H-< iH,O 3H..O'- Barium glycerate, Ba(C 3 H 5 O 4 ), }H,O 3H,O- 22. Density. The specific gravities of the active isomers and of the racemic form of a number of crystalline bodies have been determined by Liebisch 3 and by Walden. 4 The observa- tions have shown that the densities for the d- and /-modifica- tions, which were always found to be the same, were in most instances smaller than that for the corresponding racemic form. In some cases the reverse was found to be true, while perfect agreement may also occur. In the following table, which pre- sents these three cases, the amount of increase or decrease in passing from the active to the racemic form is given. The observations by Walden give values of d\ 1 Fischer: Ber. d. chem. Ges., 23, - Frankland and Appleyard: J. Chem. Soc., 63, 310. The i,i, Mg, Ca, Zu and Cd salts of active and racemic glyceric acid have the same water of crystallization. s I^iebisch (see Wallach): Ann. Chem. (tiebig). 286, 139. 4 Walden: Ber. d. chem. Ges., 29, 1692. 8o OPTICAL MODIFICATIONS Active forms. Ra- cemic form. Change in per cent. Ob- server. 759 755 595 .188 .243 2.134 1.108 1.117 i-i34 2.2428 1.128 1.788 1.778 1.601 1.228 1.249 2.225 1.126 1.142 1.180 2.2495 1.131 + i.7 + 1-3 + 0.4 + 3-4 + 0.5 + 4-3 + 1.6 + 2.2 + 4-1 + 0.3 + 0.3 L. W. P. 1 W. W. W. L. L. L. L. L. L. 1.687 2.093 1.538 i.34i 1.679 2.073 1.511 1.300 -0.5 i.o -1.8 -3-i W. W. W. W. Carvone tribromidc from hydrocarvonc* 1.958 1.958 L. In those cases in which the specific gravity of the racemic bodies is greater or less than those of the components, it may be safely assumed that the first represent true chemical com- pounds. If they were merely mechanical aggregations, the same behavior should be expected as with isomorphous mixtures, the densities of which, as is well known, are additive properties, that is, as the densities of the isomers are the same, that of the raceme body should also be the same. A contraction or dilatation, on the contrary, points to chemical combination. 23. Solubility. As far as observations have shown the racemic compounds, as a rule, are less readily soluble than the active forms. Numerical results have been given for the following substances: Perkin : J. Chem. Soc., 51 , 366. SOLUBILITY 100 Parts of solvent dissolve at t c . 8l / Active forms. Racemic form. Observer. Tartaric acid and anhydrous racenric acid in water. 20 40 60 80 100 115.0 pts. 139.4 " 176.0 " 217.6 " 273-3 " 343-5 ' 8. 2 pts. 17.0 " 37-0 ' 64-5 ' 98.1 " 137.8 " Leidie: C. r., 95, 87. Glycenc fMg(C,H,O 4 ), acid salts Ca (anhy- J Ba " drous) in y n water. |.J 20 20 20 20 20 43-05 " 9.32 " 50.15 " 39-03 " 85.00 " 22.78 " 3.85 " 6.60 (( 3.87" 4-43 " Franklandaud Apple- yard: J. Chem.Soc., 63, 310. Potassium methoxy- succinate in wafer. Calcium methoxy- succinate in water. Calcium ethoxysucci- nate in water. 16 14 15 di4-4, '13-9 5-41 " 4-15 " 3-0 " 0.46 " 0.63 " Purdie and Walker: J. Chem. Soc., 63, 222, 233. Calcium gulonate in water. 15 5-8 " 1.6 " E. Fischer: Ber., 25, 1028. Calcium galactonate, in water. 100 50 " 2.2-2.5" E. Fischer and Hertz: Ber., 25, 1253. Leucine in water. 2-44 " 0.98 " Walden: Ber., 29, 1702. Inosite in water. ca. 50 " 4-56 " Walden: Ber., 29, 1702. Camphoric acid in water " alcohol 20 15 6. 9 6 J " 112 " o.2 39 2 - 33" 1 Jungfleisch. 2 Aschan. From Beilstein, Org. Ch. Isocamphoric acid in water. 2O 0-347 " 0.203 " Aschan: Stud. Campher- gruppe. Camphoronic acid in water. 20 16.8 " 3-72 " Aschan: Ber., 28, 16. Isopropylphenylglycol- ic acid in alcohol. 13 47-4 " 21.6 " Fileti: J. prakt. Chem., [2], 46, 561. The lower solubility of the racemic compounds is shown ^ many cases by the formation of a crystalline precipitate when concentrated solutions of the active forms are mixed. This is 82 OPTICAL MODIFICATIONS true, for example, of concentrated aqueous solutions of tartar ic acids (Pasteur), 1 of alcoholic solutions of camphoric acids (Chautard) 2 and of many derivatives of the limonenes as the nitrosochlorides, nitrosates, hydrochlornitrolbenzylamines, hydrochlornitrolanilides, dissolved in alcohol (Wallach), 1 ' and the flr-nitrolpiperidines dissolved in petroleum ether (Wallach). 4 Likewise, from a mixture of the methyl alcohol solutions of d- and /-dimethyl tartrate, the less soluble racemate ester crystallizes at once (Anschiitz). 5 This rule of the lower solubility of the racemic compounds is, however, not without exceptions. Thus, the two active or-limonenebenzoylnitrosochlorides are less soluble in acetic ether than the crystalline dipentene compound (Wallach). 6 8.64 parts of the active mandelic acid, and 15.97 parts of the racemic acid dissolve in 100 parts of water at 20 (Lewkowitsch). 7 100 parts of water dissolve only 0.084 P art of /-silver valerate at 20, but 1.181 parts of the racemic salt at the same temperature (Schiitz and Marckwald). 8 d- and/- dimethyldiacetyltartrat.es are less soluble in benzene, than the diacetylracemate ester (Anschiitz). 9 d- and /-barium cam- phoronates form difficultly soluble precipitates, while the r-salt is easily soluble in water (Aschan). 10 Nearly complete agree- ment in solubility was found for d- and r-conine, 100 parts of water dissolving 1.80 parts of the first and 1.93 parts of the second at 19.5 (Ladenburg). 11 The solubility of the racemic compounds may be modified also by the fact, that even in a concentrated solution they are not in unchanged condition, but are partly dissociated into their antipodes. This has been shown for racemic acid, as mentioned, and also for its sodium-ammonium salt which, even in strong aqueous solution, may be completely split up into the two tartrates. See 27 : Temperature of Transformation. Pasteur : Dissymmetric moleculaire, 1860. Chautard: Compt. rend., 56, 698. Wallach : Ann. Chem. (I_iebig), 270, 195. Wallach: Ibid., 253,125. Anschiilz: Her. d. chem. Ges., 18, 1398. Wallach: Ann. Chem. (Liebig), 370, 177. Lewkowitsch : Her. d. chem. Ges., 16, 1566. Schiitz and Marckwald : Ibid., 39, 58. Anschiitz: Ann. Chem. (L,iebig), 347, 116. 10 Aschan : Her. d. chem Ges., 38, 16. 11 Ladenburg : Ibid., 38, 165. MELTIXG-POINT 24. Melting-Point. The determinations of the melting-points of active and racemic forms of many crystallizable substances have failed to show any regularities. In some cases, the melting-point of the racemic compound is higher than that of the active components, while in other cases the reverse is true. The melting-points may also be the same. As may be seen from the table below, quite different groups of bodies are found in each of the three classes : Active **~. Diff - Observer. A. Melting-point higher for the racemic than for the active form. Tartaric acid, racemic acid H O 170 20d 1A Walden : Ber., 29, 1701. IOO I "JO ^ o<+ 10 ^ Walden : Ber., 29, 1698. Dimethyl tart, and racem . . 45 80 *a"0 8 5 108 O^'O 40 IQ Anschiitz : Ber., 18, 1307. Purdie and Marshall : Ch. Isopropylphenylglycolic l = -i e Ts;6 C A 7 Soc. 63, 217. Filetti : Gazz. ch. , 22, II, Galactonic acid, lactone- < go 92 1 1 6"^ ^ 122 125'-' l68 33 395- ^chnelle and T. : L. Ann., 271,83. 2 Fischer: Ber., 25, 1247. iv O'G 187 20 1 5 16 383- Smith ' L Ann 272 180 Mannonic acid, phenylhy- drazide 2jc **->3 2 ?() r c E Fischer Ber 23 378 Mannoheptonic acid, phen- 22O *&* 2 C X D Smith ' L/ A.nn 272 i8s Mannoheptosephenylosa- 2O7 -*:) 186. Galactosephenylosazone Gulonic acid, phenylhydra- zide ^*O 195 148 206 I c/1 II 5 188. E. Fischer: Ber., 25, 1256. E Fischer Ber 25 1020 187 1 O4 T C \Valden * Ber 29 1700 Isocamphoric acid. 10/ 171 ^ IOO ^ X 3 IQ Walden * Ber 29 1701 Limonenetetrabromide. Limonene-a-nitrolpiperi- dide.. 105 QA ya 124 I^A. A V 19 60 Wallach :Ber., 24, 1559. Wallach : Ber.. 24, I.S.SQ. OPTICAL MODIFICATIONS Active forms. Ra- cemic form. Diff. Observer. L,imonene-0-nitrolpiperi- 72 "7.5 92.5 03.5 137 152.5 86.5 168 190 116.5 152 126 140 109.5 150 160.5 169.5 97 178 205 42 13 21 22.5 17 46.5 17 10.5 10 15 II Wallach : Ber., 24, 1559. Wallach : Ber., 24, 1559. Wallach : Ber., 24, 1559. Wallach : L. Ann., 270, 194. Wallach : L. Ann., 270, 192 \Vallach : L. Ann., 270, 192. \Vallach : L. Ann., 272, 108. Wallach: L. Ann., 272, 108. Wallach : L. Ann., 286, 123. Baeyer : Ber., 28, 640. Marckwald : Ber., 29, 46. Marckwald: Ber., 29, 46. Limonene-a-nitrolanilide Limonenehydrochlornitrol- Limonene-a-nitrolbenzyl- Limonenehydrochlornitrol- Fenchylphenylsulfourea- . . 0-Carvonepentabromide . . . Caronesemicarbazide a-Pipecolinehydrochloride - B. Melting-point of the racemic lower than for the active forms. 176 172 202 /I54 162.5 196- 200 207 132.8 153 ' 135 109.5 153-5 160.5 198 85 132.5 175-5 191 118 149 125 22. 5 ii. 5 4 23 16 14.8 4 10 19-5 Walden: Ber., 29 1699. Walden: Ber., 29, 1699. Walden: Ber., 29, 1700. Fischer and Smith: L,. Ann., 272, 182. Fischer: Ber., 23, 2620. Fischer: Ber., 23, 2226; L. Ann., 272, 182. Fischer: Ber., 27, 1524. Lewkowitsch: Ber., 16, 1566. Wallach: Ber., 24, 1559. Wallach: Ber., 24, 1559. " L. Ann., 270, 176. Mannoheptonic acid, Glucosediphenylhydra- Man noheptose-phen y 1- Mannitoltribenzacetal .... Limonene-/3-nitrolanilide . ' ' -hydrochlorcarvoxirm " a-benzoylnitrosochlo- MELTING-POINT Active forms. Ra- cemic form. Diff. Observer. 40 160.5 II4-5 95 121 142.5 162.5 I3I.5 194 147 34 159 98.5 64-5 108 125 155 118.5 186 131 6 1-5 16 30-5 13 17.5 7-5 13 8 16 Wallach: L. Ann., 272, 108. " L. Ann. ,272, 108. " L. Ann., 272, 108. " L. Ann., 272, 108. " L. Ann. ,286, 121. " L. Ann., 286, 122. v. Baeyer: Ber., 28, 640. Marckwald: Ber., 29, 46. Marckwald: Ber., 29, 46. Marckwald: Ber., 29, 46. Oxybenzylidene-fen- /3-Carvonetetrabromide . . . a-Carvonepentabromide. Carvonesernicarbazide .... a-Pipecoline salts, C H N H \uCl (C 6 H 13 N) 2 .H 2 PtCl 6 ... (C 6 H 13 N.HI)CdCl 2 ... C. Melting-points of the active and racemic forms the same. Galactonicacidphenylhy- i drazide "...'.. 2oo~-2os Gulonic acid, phenylhy- drazide Phenylgulosazone Limonene-o-nitrosochlo- Fischer: Ber., 27, 3225. Fischer: Ber., 27, 3225. Fischer: Ber., 25, 1030. ^-Monobromcamphor . . . 92.4-92.7 uaiiacn: L,. Ann., 252, in, 125. Kipping and Pope: J. Chem. Soc., 67, 372. In the cases of those racemic bodies whose melting-points are found to be lower than those of the components there is the possibility that they are not compounds but mixtures, as this relation is characteristic of mixtures. In fact this has been proved in the case of gulonic acid lactone. A mixture of equal parts of the antipodes gave on evaporation of the aque- ous solution at first an inactive crystal mass, the melting-point of which (160) was much lower than that of the active forms ( 1 8 1 ) . But it was recognized that this could not be a racemic compound because by repeated fractional crystallization it could be split up into antipodes. 1 Possibly the same condi- tions will be disclosed by fuller investigations of other sub- stances in group B. See 32. 1 Fischer and Stahel: Ber d. chem. Ges., 34, 534; Fischer and Curtiss: Ber. d. chem. Ges., 35, 1025. 86 OPTICAL MODIFICATIONS In those cases in which racemic bodies have about the same melting-point as have the components, it is also probable that they are merely physical mixtures. The conditions then resemble those which Kuster 1 found in isomorphous mixtures where the melting-points lie between those of the constituents. Then again, a change in the melting-point has been found in certain substances. Wallach observed 2 that the melting- point of racemic y#-carvonepentabromide, which was about 96-98 at the start, and quite different from that of the components (86-87), becomes lower by repeated crystal- lization of the preparation. It appears here that a partial decomposition of the racemic compound must have taken place. Walden has called attention to the parallelism between specific gravity and melting-point of the optical modifications. 3 A consideration of the above tables shows in fact, that with those bodies in which the specific gravity of the racemic form is greater than that of the active components, the same relation holds for the melting-points, and vice versa. 2. Liquid Racemic Compounds 25. On mixing equal amounts of the antipodes of liquid substances it may happen that immediate solidification to a crystalline mass takes place, which is inactive in solution ; in such case a true racemic compound certainly is formed. This occurs, for example, by pouring together of d- and /-carone- oxime (v. Baeyer). 4 If, on the other hand, the inactive mixture remains liquid, it is uncertain whether it should be considered as a racemic compound. There is evidently no ground for this, if it be found that the physical properties of the mixture are' the same as those of the components. This has been observed, for example, with limonene, also with carvone, in respect to specific gravity and boiling-point (Wallach), 5 conine, with respect to specific gravity, boiling-point and solubility (Laden- burg), 6 further with the esters of /- and inactive glyceric acid Kiister : Ztschr. phys. Chem., 5, 601 ; 8, 577. Wallach: Ann. Chem. (Uebitf), 386, 138. Walden : Ber. d. chem. Ges., 29, 1704. v. Baeyer : /bid., 28, 640. Wallach : Ann. Chem. (I^ebig), 286, 138. I^adenburg : Ibid., 247, 81 ; Ber. d. chem. Ges., 28, 163. LIQUID RACEMIC COMPOUNDS 87 and diacetylglyceric acid Frankland and (MacGregor). 1 The agreement in the specific gravities of liquid inactive and active isomers, leads to the conclusion that the inactive forms must be considered as mechanical mixtures, as I. Traube has shown by the aid of his atomic constants and molecular covolume data. 2 Such inactive mixtures may differ in chemical behavior, from their components, d- and also /-limonene are trans- formed by bromination into rhombic-hemihedral crystals of the tetrabromide, melting at 104, while the mixture of the two isomers (dipentene) furnishes rhombic crystals melting at 124, which are less soluble in ether than the first, and w r hich form a racemic compound (dipentene tetrabromide) (Wallach). 3 The racemization probably begins here, however, with the for- mation of the crystalline compound, and the behavior noted gives no proof therefore that the dipentene is already a real combination of the antipodes. Elevation of the boiling-point could betaken as an indication of raceme formation, but this has not yet been observed in a single case with certainty, experiment showing always practical agreement in this factor. Besides, even with true racemic compounds, the possibility is present of finding not their own boiling-point, but that of the components, because at the high temperature necessary, dissociation into the last may take place. The temperature changes observed on mixing active isomers have been employed to decide the question under discussion. In this regard, the following facts are to be considered : If the two antipodes unite to form a real racemic compound, which immediately separates in crystalline form, the increase of temperature observed is due partly to the heat of formation and partly to change in the state of aggregation. This is, for example, the case w r hen concentrated solutions of d- and /- tartaric acid (Pasteur), or d- and /-limonenetetrabromide in ether, are mixed. 4 With the tartaric acid there is also the heat of hydra tion, in consequence of forming racemic acid, c,H 6 o 6 + H,O. 1 Frankland and MacGregor : J. Chem. Soc., 63, 511! - I. Traube : Ber. d. chem. Ges., 29, 1394. Wallach : Ann. Chem. (Liebig), 286, 138. 4 l,adenburg: Ber. d. chem. Ges., 28, 1994. 88 . OPTICAL MODIFICATIONS If the two antipodes are themselves liquid substances, and if no solid racemic body separates on mixing them, then a change of temperature may be due : first, to the heat of solu- tion or dilution, or, secondly, to the possible formation of a liquid racemic compound. Regarding the heat of solution observed on pouring together two miscible liquids, this may be shown according to the nature of the substances, and the proportions even, in a decrease of temperature as well as by an increase. As investigations of Bussy and Buignet, 1 and also of Favre, 2 have disclosed, the heat effect bears no definite relation to change of density which takes place. Thus, alcohol and ether mixed with contraction, carbon disulphide and chloroform with expan- sion, but in both cases there is a decrease in temperature. If alcohol and chloroform be mixed in different proportions, there is always contraction, but in spite of this, reversed temperature changes may be found, as shown by the following example : Mixing proportions. Temp, change. 6 niols. chloroform - i mol. alcohol 2.5 i^ mols. chloroform i mol. alcohol o.o i mol. chloroform -j- 6 mols. alcohol 4 4.2 In general, as late investigations of Ladenburg' have also shown, large changes in volume seem to correspond to large temperature changes. If the contraction or dilatation is very small, that is, if the specific gravity of the mixture is very nearly that of the mean calculated from the mixing pro- portions, a very small temperature change is in general noted, as in the case of methyl and ethyl alcohol, isobutyl and iso- amyl formate, xylene and toluene, and soon (Ladenburg). But, on the other hand, large temperature changes have been observed ; thus, on mixing two volumes of ether with three volumes of carbon disulphide there is no change of volume, but notwithstanding this, there is a fall of temperature of 3.6 (Bussy and Buignet), and this is the case with some very similar liquids, such asd- andAconine, with which Ladenburg 1 1 Bussy and Buignet : Jahr^sbericht, 1864, 62 1069. Ann. chim. phys., [4], 4, 5. a Favre : Jahresbericht, 1864, 66. Compt. rend., 59, 783. * L,adenburg : Ber. d. chem. Ges., 28, 1991. I^adenburg : /bid., 28, 164. LIQUID RACEMIC COMPOUNDS 89 found a fall of temperature of 1.4. But d- and /-limonene mix without appreciable change of temperature. 1 As the above relations show, the temperature changes accompanying mixing or solution are of such different kinds, that the heat effects in consequence of a real combination of the antipodes may be quite uncertain. Therefore, neither an increase nor decrease of temperature which may be observed on mixing two active isomers, may be taken as definitely indicating the existence of racemization. Finally, it may be remarked, that it has been found impossible through cryoscopic measurements also to prove racemization in liquid inactive bodies. Thus, Frankland and Pickard 2 found the simple molecular weight for the inactive methyldibenzyl ester of glyceric acid dissolved in acetic acid, benzene, nitrobenzene or ethyl bromide. 26. The following may be given as the principal results of all of the above comparisons of the properties of racemic com- pounds with those of the active modifications : r . True racemic compounds are found only among crystal- lizable substances. They have always a different crystalline form, and often an amount of water of crystallization different from that of the active isomers. Further, they show, as a rule, deviations in respect to specific gravity, melting-point and solubility, but often in different directions. 2. Inactive crystalline masses may be sometimes simple aggregations or growths of mixed enantiomorphic active crystals. 3. The existence of liquid racemic'compounds is improbable, and up to the present time has not been shown. The question as to whether an inactive body is a racemic compound or racemic mixture is besides, in many cases, only of secondary importance. Both forms may be split up into optical antipodes in the same w r ay, and in this we have all that is really important in the respective substances, that is, their distinction from inactive configuration isomers. 1 I^adenburg : Ber. d. chem. Ges., 28, 1994. - Frankland and Pickard : J. Chem. Soc., 69, 128. QO FORMATION OF RACEMIC BODIES C. Formation of Racemic Bodies Racemic bodies may be produced in the following ways : 1 . By direct combination of the active antipodes. 2. From one of the active forms by action of higher tem- perature. 3. By the chemical transformation of asymmetric bodies into asymmetric derivatives. 4. By the conversion of inactive symmetric bodies into asymmetric compounds. 27. Production of Racemic Bodies by Combination of Equal Amounts of the Antipodes. Temperature of Transition. Certain phenomena may appear here which belong in the class of chemical equilibrium reactions, the important characteristic of which is, that for them a definite temperature exists which, if passed in the one direction or in the other, leads to the transition of one system of bodies into another or the reversion of the latter into the former. Thus, the breaking up of a racemic body into its antipodes is possible, or, on the other hand, the reproduction of the racemic from the antipodes. The transition temperature has been determined in but two cases : First : with sodium ammonium racemate and the two cor- responding tartrates. When Pasteur permitted a solution of racemic acid which was saturated, half with soda and half with ammonia, to evaporate spontaneously, he obtained separate crystals of the d and / double tartrate salt. But on repeating the experiment, it did not always succeed ; thus, Staedel 1 observed only the formation of the racemate. Scacchi' 2 first noticed that the kind of crystal which separates depends on the tem- perature at which the evaporation takes place, and WyroubofF then determined 28 as about the limiting temperature above which racemate, and below which, on the contrary, the tartrates, crystallize out. A complete explanation of the phenomenon was first given by van' t Hoff and Deventer, 4 when 1 Staedel: Ber. d. chem. Ges., n, 1752 (1878). 1 Scacchi: Rendiconti dell Accad. di Napoli. 1865, 250. 1 Wyrouboff : Bull. Soc. Chim., 45, 52 (1886) ; Compt. rend., 102, 627. 4 Van't Hoff and Deventer: Ber. d. chem. Ges., 19, 2148(1886); Ztschr. phys. Chem., i. 165. TRANSITION TEMPERATURE 91 they showed that we have to deal with a process here which can also take place outside of solution. If a finely powdered mixture of equal parts of right and left sodium ammonium tartrate, NaNH 4 C 4 H 4 O 6 -f 4H 2 O, be sealed in a tube and exposed to a temperature which is kept below 7 27, it remains quite unchanged ; but above 27 the formation of crystals of the racemate, NaNH 4 C 4 H 4 O 6 -f H 2 O, sets in, water being liberated at the same time which partially liquefies the mass. On the other hand, if powdered sodium ammonium racemate be mixed with water below 27 in the proportion of 2(XaNH 4 C 4 H 4 O 6 + H 2 O) : 6H 2 O, the semi-fluid mass which forms at first, solidifies after a time to a dry mixture of the two tartaric acid salts. Above 27 this does not happen. The two reactions may be expressed by the following equations in which T= C 4 H 4 O 6 : Stable below 27. Stable above 27". d NaXH 4 T 4H 2 O ) \r [NaNH 4 T + H 2 O] 2 / NaNH 4 T 4H 2 O / \ 6H 2 O The change of one salt into the other may be recognized by aid of a dilatometer, which is filled partly with a mixture of the two tartrates and partly with oil; the slow formation of the racemate between 26.7 and 27.7 is accompanied by a marked increase in volume. Further investigations of van't Hoff, Goldschmidt and Jorissen 1 showed that the sodium ammonium racemate (XaNH 4 C 4 H 4 O 6 -f- H 2 O), when heated to about 35, under- goes a further change as it breaks up into sodium racemate and ammonium racemate, according to the equation: 2[NaNH 4 r.H 2 0] 2 = [N'a, T], - [(NH 4 ) 2 ^] 2 + 4 H 2 O. Sodium ammonium racemate can exist in solution, therefore, only within the narrow temperature limits of 27 to 35. Again, it can happen that the separation of the double racemate may fail when a mixture of d- and /-sodium ammo- nium tartrates is exposed to a temperature higher than 27, for example, to 30, and instead a separation of crystals follows, consisting of a mixture of sodium racemate and ammonium racemate as in the last case. This is especially true when 1 van't Hoff, Goldschmidt and Jorissen: Ztschr. phys. Chem., 17, 49. 92 FORMATION OF RACEMIC BODIES crystals of the last named salts are added to the solution. The change may be represented in this way: d 2[NaNH 4 T + 4 H 2 0] V f /2[N.NH 4 7'-f4H f O]J l The methods by which the transition temperatures of 30 and 35 were established consisted in tensimetric observations and determinations of solubility. Analogous relations, according to experiments of van't Hoff, Goldschmidt and Jorissen, 1 have been found to exist in the case of the potassium sodium racemate [KNaC 4 H 4 O 6 + 3H 2 O] 2 , described by Wyrouboff, 2 and d- and /-Rochelle salt, KNaC 4 H 4 O 6 -f- 4H 2 O. A solution formed from the racemate or from the two tartrates is found in the following conditions according to the temperature. 1. Below about 6 d- and /-potassium sodium tartrate exist in solution together: 2. From about 6 on the combination to potassium sodium racemate begins, and especially on an addition of small crystals of this salt: d KNa T - 4H 2 ) f [KNa T + 3H,O] 2 / -r 4H 2 O I 2H 2 O 3. The double racemate exists up to the temperature of 41; from that point it decomposes into the sodium racemate, [Na., T J 2 , and potassium racemate,, [K 2 7^ -f- 2H 2 O] 2 according to the equation: 2[KNa7- + 3H,0], = [Na 2 T] 2 + [K,7> 2H 2 O] 2 -f 8H 2 O. 4. If the formation of the potassium sodium racemate from the two tartrates fails, which may happen by excluding every trace of the first named salt, a conversion of the tartrates into potassium racemate and sodium racemate is possible. This takes place at a temperature of 33: d 2(KNar+ 4 H,0) \ Therefore under varying conditions very different salts may crystallize from the solution. ! van 't Hoflf, Goldschmidt and Jorissen: /tschr. phys. Chem., 17, 505. 3 Wyrouboff: Ann. chini. phys. [6], 9, 224. PRODUCTION OF RACEMIC BODIES BY HEAT 93 Exact knowledge of transition temperatures extends only to the two salts just described. With a number of substances it has been observed that when mixed solutions of their antipodes are allowed to evaporate under ordinary conditions of temperature active crystals only, never racemic, are formed. Thus, Piutti 1 did not succeed in uniting d- and /-asparagine to form a racemic compound, while it was possible with d- and /-aspartic acid. As already remarked, according to Fisher and Curtiss, 2 the lactone of d- and /-gulonic acid separates in independent enantiomorphous crystals, w T hile, on the other hand, a racemic calcium salt and phenylhydrazide may be obtained. It has likewise been found impossible to obtain by crystallization racemic forms of several other bodies, as the zinc ammonium salt of active lactic acid, glutaminic acid, glutaminic acid hydrochloride, homo-aspartic acid and camphoric acid. In all of these cases it cannot be assumed that the antipodes do not possess the property of uniting to form a racemic body; knowledge of the transition temperatures simply is lacking. 28. Production of Racemic Bodies from One of the Active Forms by Heat. By long application of heat, many active substances suffer a gradual decrease in their rotating power, without undergoing a change in composition. It was at first supposed that a destruction of the active property took place, but the phenomenon was later recognized as one of racemization. In consequence of the enlarged atomic motion by increase of temperature, a conversion into molecules of the opposite modi- fications follows until finally a condition of equilibrium is established, in which just as many molecules of the ^-form are changed into the /-form as vice versa, that is, in which the mixture consists of equal amounts of the two antipodes. Van 't HofF has treated the question from the standpoint of thermo- dynamics. The phenomenon was first observed by Pasteur 4 in the case of tartaric acid, when he heated ^-cinchonine" tartrate five or 1 Piutti: Ber. d. chem. Ges., 19, 1694. - Fischer and Curtiss : Ber. d. chem. Ges., 25, 1025. :! van 't Hoff : ^agerung der Atome im Raum, 2nd ed., p. 33. 4 Pasteur : Compt. rend., 37, 162 (1853). Q4 FORMATION OF RACEMIC BODIES six hours to a temperature of 165 to 175 in an oil-bath. The alkaloid was changed first, being converted mainly into cin- chonicine, and then the production of racemic acid gradually followed, which was recovered by conversion into the calcium salt. /-Cinchonine tartrate behaves in a similar manner. The alkaloid takes no part in the reaction, it only protects the tartaric acid from destruction by the heat. The formation of racemic acid follows also when dry tartaric acid is heated to 170 to 180, and likewise, by boiling its aqueous or weak hydrochloric acid solution several days, which, however, brings about a conversion of only a few per cent. 1 Racemiza- tion follows almost completely on heating 30 grams of tartaric acid with 3 or 4 cubic centimeters of water for thirty hours to 175 (Jungfleisch). 2 Further, tartaric acid, in the form of its ethyl ester, is very easily changed by boiling into racemic acid (Pasteur). 3 The addition of several bodies aids the change ; when tartaric acid is mixed with some aluminum tartrate and heated in an autoclave to 140 a conversion into racemic acid follows quickly, but some mesotartaric acid is produced at the same time (Jungfleisch). 4 Racemization by high and long heating in closed tubes has been observed in other substances, as : l-Aspartic Acid. The transition takes place easily when an aqueous solution of the hydrochloric acid compound is heated some hours to 170 or 180 (Michael and Wing). 5 d- and l-Mandelic Acid. Small amounts (3 or 4 grams) of the dry substance, thirty hours to 160 (Lewkowitsch).* d- and l-Isopropylphenylglycolic Acid. Heating for forty hours with water to 180 or 200 (Fileti). 7 d-Camphoric Acid. Heating with a little water to 170 or 180 (Jungfleisch, 8 Friedel). 9 The active terpenes suffer a decrease in rotation when Dessaignes: Jahresbericht, (1856), 463; (1863), 301. Jungfleisch: Compt. rend., 75, 439, 1739. Pasteur : Loc. i it. Jungfleisch : Compt. rend., 85, 805. Michael and Wing: Her. d. chem. Ges., 17, 2984. Lewkowitsch : Ibid., 16, 2721. Fileti : Gazz. chim. ital., aa, II, 395. Jungfleisch : Jahresbericht, (1873), 631. Friedel : Compt. rend., 108, 978. PRODUCTION OF RACEMIC BODIES BY HEAT 95 heated to about 250 to 300, but show at the same time an elevation in the boiling-point and specific gravity, from which it follows that polymerization as well as racemization has taken place, d- as well as /-limonene and pinene are converted into dipentene, partly also into terpinene (Wallach). The temperature of racemization is not in all cases as high as for the bodies given above. As Wallach 1 found, d- and /-limonene hydrochloride have the property of changing gradually even at the ordinary temperature. A preparation which was examined when produced, and after having been kept several weeks showed a decrease in rotation from \oi\ D = + 39-5 to 6.2, which, as could be demonstrated, was due to the conversion into the dipentene compound. Besides, the boiling-point of the substance, which was originally 97.5, increased to about 170 (under j.i mm.), which indicated polymerization at the same time. In some cases it has been observed that racemization is hastened by addition of certain substances and is completed at a much lower temperature. As already mentioned tartaric acid in presence of aluminum tartrate is easily converted into racemic acid. Active amyl alcohol when heated alone must be maintained a long time at 250 to 300 to lose its rotating power, but on treatment with sodium or with caustic potash the change is much more rapid; from the inactive product obtained dextroamyl alcohol could be separated by aid of fungi (Le Bel). 2 On repeating this experiment Borucki 3 found that when amyl alcohol was heated with potash in an auto- clave during ten hours to 160 the angle of rotation, a D for i dm., decreased from 1.01 to o. 10, and that by heating under ordinary pressure with renewed alkali, complete inactivity resulted only after 465 hours. According to Walden 4 almost perfect inactivity may be reached much more rapidly by dis- solving one-tenth its weight of sodium in the amyl alcohol and heating then 3^ hours to 200 or 220 in an autoclave. Erlen- meyer and Hell 5 have observed that by treating Erlenmeyer and Hell: Ann. Chem. (Iiebig), 160, 302. 96 FORMATION OF RACEMIC BODIES from fermentation amyl alcohol with a few drops of strong sulphuric acid and heating then ^ hour in a sealed tube to 250 complete inactivity resulted. Whether this was due to racemic formation or not was not established. Active leucine which is not changed by heating with w r ater to 170-! 80 is converted by addition of baryta water at 150-! 60 into the r- compound (Schulze and Bosshard). 1 The racemization of terpenes is hastened by addition of sulphuric acid, either con- centrated or diluted with alcohol (Wallach). 2 29. Racemization by Conversion of Asymmetric Bodies into Asymmetric Derivatives. If an active molecule retains its con- stitution in a chemical change, that is, for example, if a change of an acid into ester, salt or amide, or of an alkaloid into a combination with an acid is concerned, the activity is always retained. This is the case also with simple transitions, as of active amyl alcohol into valeric acid, camphor into camphoric acid, asparagine into aspartic acid and maleic acid, amygdalin into amygdalic or mandelic acid, etc. But under certain conditions the product may be inactive. This is particularly the case in the production of derivatives in which the atoms directly united to the asymmetric carbon take a part in the reaction. If, for example, in active malic acid, CO,H *CH.OH CH 2 CO 2 H, the hydroxyl group in combination with the *C be replaced by Br by aid of hydro- bromic acid, racemic bromsuccinic acid, CO 2 H *CH.Br CH, CO 2 H, results (Kekule). 3 Further, /-valeric acid, (C 2 H 5 ) *CH (CH 3 )(CO. 2 H), is converted completely into racemic bromvaleric acid, (C 2 H 5 ) *CBr fCH s )(CO 2 H), by treatment with bromine and phosphorus at a low temperature followed by heating to 100 (Schiitz and Marckwald). 4 The same active valeric acid when oxidized in aqueous solution by potassium permanganate yields racemic oxy valeric acid, (C,H 6 ).*COH.(CH 3 ) (CO 2 H). In changes like the above, by keeping the temperatures as low as possible, the racemization may be often prevented and 1 Schulze and Bosshard: Ztschr. physiol. Chem., 10, 135. 2 Wallach: Ann. Chem. (I Skraup : Ann. Chem. (I == -+ 64) and /-allocimiamic acid dibromide ([]/> = - 70), a residue which on treatment with carbon disulphide could be separated into the components. E. Fischer* has attempted to combine compounds with each other whose configurations present only partially the object- reflection relation, as is the case with : ^-Gluconic acid. /-Mannonic acid. COOH COOH I I H C OH H C OH HO C H H C OH H C OH HO C H H C OH HO C H CH.OH CH.OH From a mixture of equal parts of the two acids which was evaporated to a sirup, pure /-mannonic acid lactone only 1 lyiebermann : Her. d. chem. Ges., 37, 2045. 8 Fischer : Ibid., 27, 3226. RESOLUTION OF RACEMIC BODIES 99 separated, and from a mixture of the calcium salts, ^-calcium gluconate crystallized first. The oppositely rotating forms of camphoric acid and iso- camphoric acid do not unite to yield a racemte compound (Aschan). 1 A tendency, therefore, toward the formation of such half- racemic bodies does not appear to exist. D. Resolution of Racemic Bodies For the decomposition of racemic bodies into the anti- podes, we have as yet three methods, w r hich were all dis- covered by Pasteur, and first applied to racemic acid. These are : 1. Resolution by crystallization. 2. Resolution by aid of active compounds. 3. Resolution by aid of fungi. The principles underlying these methods have been explained already and it remains to discuss in this chapter the practical methods of carrying out the processes. i. Resolution by Crystallization. Spontaneous Resolution 32. This process depends on the facts that racemic com- pounds in solution, under proper temperature conditions, break up into their antipodes and that these last may be separated by evaporation of the solution in the state of enantiomorphic crystals which may be sorted out by aid of the characteristic planes into active right and left forms. Equal amounts of the two antipodes are .obtained. The transition temperatures explained in 27 must be considered as these are different for different substances. The method was first applied by Pasteur in 1848 for the splitting of racemic acid in the form of sodium-ammonium salt and may be best carried out in the following manner : :{ An aqueous solution of racemic acid is divided into two parts, one being saturated with sodium carbonate or hydroxide and the other with ammonia in excess. The mixture is then evapo- 1 Aschan : Ber. d. chetn. Ges., 27, 2001. - Pasteur: Ann. chim. phys. [3], 24, 442 (1848). 3 Pasteur: Ibid., 28, 56 (1850). 100 RESOLUTION OF RACEMIC BODIES rated on the water-bath until crystallization begins on cooling. The crystals are redissolved by the aid of water and a little ammonia after which the solution is allowed to stand in a wide crystallizing dish for spontaneous evaporation. As explained in 27 the temperature must be kept below 27. It is advi- sable to remove a part of the crystals each morning, because, in consequence of the rise of temperature during the day, partial solution might follow with loss of the hemihedral faces. As the solution gradually loses ammonia a little must be added from time to time, sufficient to maintain a weak alkaline reaction. The separated crystals which are illustrated in Figs. 4 and 5, and which often reach a length of several centimeters, are removed from time to time by the aid of pincers and sorted out after examination with a magnifying glass. If the indi- vidual crystals have grown together or are united into groups it is best to bring them into solution by adding a little water and warming and then repeat the whole crystallization. The separation of the d- and /-sodium ammonium tartrate is facili- tated by placing the two kinds of crystals in the evaporated liquid at the start and as far apart as possible. Very often the crystallographic examination is difficult because the hemihedral faces are but imperfectly developed, or cannot be found at all. In such a case the question as to whether a crystal consists of the right or left tartrate or of the racemate may be decided by a procedure suggested by Anschiitz 1 which depends on this that the calcium racemate is much less soluble than the two calcium tartrates. The mother liquor is separated from the crystals by washing with a little water, a fragment is dissolved in a small volume of water and the solution is divided into two parts. One part is mixed with about 3 cc. of a saturated solution of dextro-calcium tartrate and allowed to stand some time; if a precipitate appears it shows the presence of the left tartrate. If a precipitate does not form and if the other half of the liquid gives a precipitate with a solution of levo-calcium tartrate the presence of the right tartrate is indicated. Finally, if precipitation takes place in both cases, racemic acid is shown to be present in thecrystal. As Pasteur 2 found, the d- and /-tartrates separate always in 1 Anschiitz: Ann. Chem. (I y iebig), 226, 193. Pasteur: Ann. chim. phys. [3], 34, 458. BY CRYSTALLIZATION IOI exactly equal amounts at any point m he -crystallization, for if at any time the whole mass of crystals -be -removed and dis- solved in water the solution will be foutid to -have no rotating power, and the mother-liquor likewise not. The two salts must therefore possess exactly the same solubility. Jung- fleisch, on the contrary, believes 1 that the right salt is less soluble than the left, inasmuch as he found more of the first in the early part of the crystallization, and more of the second in the last part. The separation of the two tartrates may be accomplished as Gernez 2 found, in this way: The original solution prepared by heat is carefully kept free from crystals of the salt or from dust and allowed to cool down until it reaches the super- saturated condition, w r hen a few crystals of the d- or /-sodium ammonium tartrate are throw r n in. Then the separation of the corresponding salt follows, while the other remains in solution. Besides racemic acid the following racemic bodies have been resolved by crystallization: r- Fermentation lactic acid, in the form of its zinc-ammonium salt, ZnNH 4 (C 3 H 5 O 3 ) 3 -f- 3H 2 O, which breaks down into the d- and /- lactates, ZnNH 4 (C 3 H 5 O 3 ) 3 + 2H 2 O (Purdie.) 3 This takes place when to a concentrated solution of the racemic compound, brought to supersaturation by cooling, crystal frag- ments of one of the lactates are added, which brings about a separation of the corresponding salt. The crystal fragments needed for this are prepared by splitting the r-lactic acid by means of the more easily performed strychnine method. If the solution of the r-zinc-ammonium salt alone be allowed to crystallize spontaneously at the ordinary temperature most of the salt separates as such. Solutions of the following four compounds furnish simultaneously crystals of the d- and /-modifications, which can be distinguished crystallographically, when their aqueous solutions are allowed to evaporate at the ordinary temperature. dl-Gulonic acid lactone, formed by mixing the antipodes (Fischer and Curtiss). 4 Racemic crystals are not formed. i Jungfleisch: Bull. Soc. Chim. [2], 41, 226. - Gernez: Cotnpt. rend., 63, 843. 3 Purdie: J. Chem. Soc., 63, 1144 to 1151. 4 Fischer and Curtis : Ber. d. chem. Ges., 25, 1046. 102 RESOLUTION OF RACEMIC BODIES by, auction of ammonia on esters of maleic and fumaric Qcid .(Korner and Menozzi), 1 or from the antipodes (Piutti).-' Racemic crystals are not formed under the same conditions. r-Homoaspartic acid, formed by the action of alcoholic ammonia on the ethyl esters of citra-, mesa- and itaconic acids (Korner and Menozzi ) t ? r-GIutaminic rtf/V/fromr-glutamide (Menozzi and Appiani). 4 The holohedric crystals of the r-acid on repeated crystallization from water yield crystals with right- and left-hemihedral sur- faces. 2 . Resolution by Active Compounds 33- This method is based, as explained in 18, on the un- equal solubilities shown by the salts of a d- and /-acid with the same active base (alkaloid), or of a d- or /-base with the same acid {e.g. tartaric acid). It was first applied by Pasteur 5 in the decomposition of racemic acid, by which he found that when a hot aqueous solution of the acid was mixed with molecular proportions of different cinchona bases to saturation there separated on cooling, not the racemic acid, but the tartaric acid salt ; with quinine or quinicine the first crystallization con- tained = 4- 68.3 for 12 to 15 per cent, solution in alcohol. From the mother-liquor, after a second treatment w r ith strychnine (i of base : 2 of acid), more of the d-sa\t and then the salt of the /-acid may be obtained. The latter acid has been secured, however, with rotation up to [<*]/> = 45.8 only. Allocinnamic acid dibromide could be separated by aid of cinchonidine in benzene solution from which the salt of the /-acid crystallized. The highest observed polarization of the acid was [<*]/?= -83.2. The af-acid could not be obtained pure (Liebermann 1 ). Phenyl-a fi-dichlorpropionic acid, (cinnamic acid dichloride) with strychnine, i molecule of base, (40 grams) and i molecule of acid (45 grams), dissolved in 500 cc. of 99.5 per cent, alcohol gave crystals of the NH By the aid of X^H 2 CH 2 af-tartaric acid, the af-base may be obtained pure (Ladenburg).' 2 /CH,-CH.C 3 H 7 a-N-Propylpiperidine, CH./ >NH ^CH 2 CH 2 Synthetic conine. This was obtained by Ladenburg 3 from a-propylpiperidine as follows : To the concentrated, but not sirupy solution of the ^/-bitartrate, small crystals of the df-conine-df-bitartrate were added. (These, as obtained by Schorm 4 from natural conine, were rhombic crystals having the composition, C 8 H 15 N.C 4 H 6 O 6 -j- 2H 2 O.) A crystal magma formed from which, by pressing and recrystallizing, the debase was obtained pure, and showing the same rotating power as the natural conine, []/>+ 18.3. It is also possible to evaporate the moderately dilute solution of the tartrates of the racemic base at the ordinary temperature, and then purify by repeated crystallization, the crystals of d?-conine-NH. By evaporating the \*CH.C 2 H 5 -CH/ aqueous solution of the tartrates, the salt of the = -25.9. The d- base could not be obtained pure (Levy and Wolffenstein). Propylenediamine, CH 3 *CH.NH 2 CH 2 NH r One mole- 1 I,adenburg : Ber. d. chem. Ges., 37, 75. 2 leaden burg : Ann. Chem. (L,iebig), 247, 7'- 3 I^adenberg : Ber. d. chem. Ges., 19, 2582 ; Ann. Chem. (i,iebig), 347, ^5. 4 Schorm : Ber. d. chem. Ges., 14, 1768. 6 L,adenburg : Ibid., 37, 3065. Levy and WolfTenstein : Ibid., 38, 2270. BY STRONG SULPHONIC ACIDS 113 cule of the base dissolved in water with two molecules of tar- taric acid deposits crystals on evaporating, which contain the /-base, [#]/?= -20.96. The == + 56. The /-base is not known (Ladenburg). 2 [See below. Tr] . i, 5-Tetrahydronaphthylenediamine. The aqueous solution H NH. f t ^ ie ^-bitartrate, evaporated to a sirup, and H \/ treated with a small crystal of ^-conine- /\ */\ ^-bitartrate, furnished a deposit of crystals HC C CH which held the /-base. Rotation of the salt I I l a ']z> = 7-5 fc> r / = 3-96. After standing >v xv x - several months, the separated mother-liquor C c furnished crystals of the bitartrate of the XH, H 2 ^-base. Rotation of this salt : [<*] D -f 8. 15 f or p = 2 . 44 ( Bamberger ) . 3 a-Phenylethylamine, C 6 H 5 *CH.CH,.NH 2 . The separation as bitartrate was tried first by Kraft, 4 but without success, and later by Loven 5 who succeeded. The concentrated hot solu- tion gave, on cooling, needle-shaped crystals with i| H.,O, from which the impure, slightly dextrorotatory base was separated ; from the mother-liquor prismatic anhydrous crystals slowly separated which furnished a strongly left-rotating base. c. Resolution by Stronger Acids Pope and Peachey 6 have recently suggested the use of strong optically active acids as agents of resolution in place of tartaric acid, ^-^-chlorcamphorsulphonic acid, ^-or-bromcamphorsul- phonic acid and ^-camphorsulphonic acid are compounds which suffer relatively great dissociation in aqueous solution 1 Baumann : Ber. d. chem. Ges., 28, 1179. 2 Ladenburg : Ibid., 27, 76. ! Bamberger : Ibid., 23, 291. * Kraft : Ibid., 23, 2783. '' IvOvn : Ibid., 29, 2313. Pope and Peachey: J. Chem. Soc., 73, 893 and 75, 1066. 8 114 RESOLUTION OF RACEMIC BODIES and in their action with weak bases may be compared to the mineral acids. Among other applications the following may be quoted: Tetrahydropapaverine, C. 20 H., 5 NO 4 . By combining the racemic base in aqueous solution with the calculated amount of */-ar-bromcamphor sulphonic acid and warming to effect solu- tion the combination, C^H^NO^.C^I^BrO.HSOg, is formed. On cooling long needles of the /-salt separate, as this is the less soluble. On concentrating and cooling again more of the salt may be secured. For the purifiedcry stals [<*]/> = - 30 was found. In the mother-liquor the = - 143.4. From the resinous df-salt the corre- sponding base with \<*\ D = - -j- 153.7 was secured (chloroform). In the same resolution df-tf-chlorcamphorsulphonic acid was employed also with good results. The authors tried Reychler's camphorsulphonic acid, but the salts formed remained in a very soluble sirupy form and could not be well separated. Tetrahydroquinaldine, C 10 H 13 N. This was resolved by the aid of ^-a-bromcamphorsulphonic acid as described above. An alcoholic solution of the /-salt gave \_<*\ D =' + 41 -5 - For the base, separated by distillation in a current of steam with a slight excess of soda, [V]^ = -- 58.12 was found. Camphoroxime , C, H 16 .NOH. The racemic oxime was resolved by action of = - 3.3, showing that it now contained 0.72 gram of /-mandelic acid. In the paper quoted, the authors describe the separation and identification of this acid in pure form. Experiments were then made with the mixture of esters and unchanged menthol left in ether solution after separation of the uncombined mandelic acid. The ether was evaporated, the residue mixed with 3.5 grams of potassium hydroxide in solution and boiled some hours. This liquid was evaporated and treated with water to dissolve the potassium salt. The solution obtained was heated to drive out traces of menthol, treated with sulphuric acid and extracted with ether. In this way a mandelic acid was obtained as a saponification product, and amounted to 8. 7 grains Its specific rotation was \_a] D = 1 v. Meyer : Her. d. chem. Ges., 27, i5 Fig. 6. Mucor Mucedo. Form and ramifications of a full grown M rct'/non, developed from the spore, a. fix the fruit stem and s the sporangium. After Brefeld. gives a clear representation of these relations. This shows a kind of mucor Mucor Mucedo}, which may be easily obtained from bread, malt, horse-dung and decaying fruits kept in moist BY AID OF FUNGI 121 places. At a is shown the developed spore ; / is an air-hypha which at its end is in the act of forming a spore-case or so-called sporangium. With darkening in this spherical organ maturity is reached. If we touch it with a previously sterilized needle, the sporangium skin breaks and a large number of spores cling to the needle. If this is now dipped into sterile nutritive solution, a new development of spores begins similar to that just described. In a few days the whole liquid is permeated by the fungus filaments, and on the surface, new air-hyphge appear and grow toward fructifi- cation. Figure 7 shows the structure of the sporangium. The external wall or skin is covered with numerous small crystalline needles of calcium oxalate. The club-like bunch shown in the center is the so-called columella (little column) ; it is shown free in Fig. 8 after the sporangium has been broken and emptied. The columella is originally a simple transverse wall which separates the sporangium from the sporangium stem. Fig. 7. Mucor Mucedo. Young fruit stem. Fig. 8. Old fruit stem. Brefeld. If in the study of the air a spore of Penicillium glaucum had become imbedded we would not observe as fine a develop- ment of air-hyphae. Fructification begins early and a very low turf only is formed which becomes covered with masses of white spores turning later to bluish green. Under the micro- scope we can see on the ends of the air-hyphae the growth of lateral bunches or branches with finger-like spore-supporting organs, so-called sterigmata. When a spore is fully developed a new one is immediately formed, which remains loosely united to the first. This process repeats itself and by undisturbed growth gives rise to beautiful spore chains resembling strings 122 RESOLUTION OF RACEMIC BODIES of pearls and containing fifty or more members. See Figs. 9 Fig. 9. Penicillium glaucum. Hvphse. Fig. 10. (From Biefeld. Hot. Unters. liber Penicillium. Schimmelpilze, i. Heft, 1872.) Piece of asexually fructifying urn. Mvcelia. From Brefeld. and 10. Fig. 10 further shows that this mold, as distinguished from Mucor Mucedo, contains transverse walls in the stem in large numbers; in other words the stems are divided. In order to take these spores from the spore case by the aid of a needle this is best dipped first in the nutrient medium or in sterile water. * The spores are dry and would not cling to a dry needle very well; but, as distinguished from Mucor spores, they are not moistened immediately as the contents, and apparently the membrane also contains fat. They distribute themselves uniformly over the droplet on the needle-point in the form of a thin dry layer. In consequence of these pecu- liarities this mold is spread very rapidly through the air ; the physiologist attempting to produce pure cultures has much to fear from its ubiquity. In appearance and behavior Aspergillus glaucus, Fig. n, stands close to Penicillium glaucum. Here, also, long spore chains grow on sterigmata, but the latter are all situated on club-like expansions of the undivided fruit stalk. Several 1 The distilled water of the laboratory is not sterile; it often contains 100,000 or more germs to the cubic centimeter. BY AID OF FUNGI 123 Fig. ii. varieties have also branched sterigmata, the Sterigmatocysti. Frequently mycelia develop on the gela- tine unaccompanied by fructification, even with Penicillium and Aspergillus varieties. As broken-off pieces of the filaments are ca- pable of growth, it is sufficient to pick up a little of the mycelium with a needle and de- posit it in a nutrient medium. While the common molds as Penicillium glaucum, Aspergillus glaucus, and others, may nearly always be obtained from the air, yeasts are found less often. As these find the most favorable conditions of growth in nature on sweet fruits, it follows that they are most commonly met with in the autumn. As compared with the mold my- 1 .1 Aspergillus glaucus. celia, they grow almost invisibly in the (Burotiumherberionun) , , . ,, , , c c Spore-chain, m Mvcel. nutrient gelatine, and especially in the form After Kn y , wan charts. of small, white pinhead-like colonies. It is only occasionally that one finds colonies that are spread out superficially ; these consist then generally of aerobiotic mold yeasts. It is very difficult, even by the aid of the microscope, to determine directly what form of yeast one has found. In order to reach a certain comparison with forms already known, there is required usually a long series of culture experiments which may consume weeks or months. If one desires to further culti- vate one of the colonies found in the gelatine in a new nutri- ent solution, the inoculation must take place with a sterilized needle. It is sufficient if only a part of a colony remains clinging to the needle, as this much will contain thousands of cells capable of development. That which appears to us very strange in the behavior of the yeasts is the fact that physiologically very differently acting forms exhibit almost no differences in the appearance of their cells. After having mixed four or five different varieties in a little drop, we are often no longer able from the micro- scopic image to pick out the separate cells. The cell forms, 124 RESOLUTION OF RACEMIC BODIES Figs. 12 to 14, which represent ,S. ellipsoideus, may be found in a large number of beer and wine yeasts. Simple pictures or drawings of the cells are therefore quite insufficient as a means of characterizing a definite yeast variety. Fig 12. Sacchar. ellibs. Conidia Fig. 13. Sacchar. ellips. Developed spores. After in which internal formation Brefeld from F. v. Tavel : "Morphologic of spores has begun. derPilze." Jena 1892. The distinction between yeast cells in budding condition offers the same difficulties as the distinction between mold mycelia in which no special seed forms have been developed. With the yeasts, besides by bud- ding, there is a second method of fructification ; viz. , through the for- Fie 14 Sacchar. ellips. Sporangia . - , f^. . the spores in which are swollen matlOn of endogenous Spores. TlllS form of seed which is shown in the two illustrations never comes to development in fermenting liquids, and then also on account of its morphological simplicity it affords few characteristic points of differentiation to settle the question of species. In germination the spore passes imme- diately to the budding condition again. In making use of the yeast colonies obtained in air investi- gations it is necessary to recognize that while in all probability we have secured pure material, a certain guarantee for it is lacking. This can be secured only by the method introduced by Hansen of cultivation from a cell. With yeasts this method of cultivation may be applied without difficulty. But with the bacteria it often fails because of their extreme minuteness. However, the colonies grown in gelatine, as distinguished from the yeasts, present often characteristic color differences, so that by the eye alone a conclusion may be reached as to different forms present by the simple variations in their gross appearance. A greater variety in general is observed also with respect to shape and size of the single cells. By aid of the methods of pure culture it has been found, as with the yeasts and molds, that what were formerly regarded as simple forms BY AID OF FUNGI 125 may be resolved into several species. So, for example, the Bacterium tentio of the older authors is no longer regarded as a single species, but the name is used as a collective descrip- tion. The different varieties of Proteus, as P. vulgaris, Zenkeri, hominis, exhibit about the same behavior that was formerly given as characteristic of B. termo. For bacteria also the rule is true that it is easier to obtain a pure culture than to deter- mine its species. From the remarks just made, it appears clear that the investigations of the older authors / on the action of organisms on race- & mic bodies must be repeated under rj /I such conditions that only pure cul- || tures and germ-free solutions may be employed, and further that care must be taken to prevent the ingress 1 ^ < ^* m J ^5ffi i of any infection during the progress Pringsh : jahrb. 27, i (1895). of the experiment. To make a nutrient solution germ-free is not difficult ; it is simply necessary to boil it a long time or frequently for short intervals, the neck of the flask being closed by a wad of cotton. This last may be dispensed w r ith in the so-called Pasteur flask, the neck of which is continued by a long bent tube. Formerly, experiments on resolution or splitting were usually carried out by mixing the racemic body, 3 to 5 grams to the liter, with nutrient salts ; e. g. , with one gram of potas- sium phosphate and 0.2 gram of magnesium phosphate. In working with molds, a little phosphoric or sulphuric acid was added to prevent the rapid growth of bacteria. A small amount of the special organism used was sowed in. This must then grow gradually and develop its splitting power. It appears to me to be better to add no nutrient salts to the solution of the racemic body, but to seed the organism employed in larger amount. This should be previously grown in a specially good nutrient solution. In the case of yeasts, for example, the cells could be cultivated in sterilized wort or wine-must, the fermented liquid poured off, and the residue washed with sterilized and cooled distilled water. To the 126 RESOLUTION OF RACEMIC BODIES vessel containing the racemic body, the so-purified yeast material is then added. The action is much more rapid, and the development of accidentally admitted germs avoided. Molds may be similarly separated from the nutrient solution. But the preparation of an active bacterial sediment offers greater difficulties. I employ for this purpose long glass tubes, 5 or 6 centimeters wide, narrowed at each end, which are completely filled with the turbid nutrient solution and placed in a horizontal position. The sediment finds now a large surface for settling and collects here as a relatively firm layer which almost wholly remains when the liquid is poured away. The sterilization of the tube is effected by steaming and sub- sequent addition of a sterilized cotton plug. The nutrient solution poured off from the sediment (accomplished by forcing air in through the cotton filter), may be replaced directly by the solution of the racemic body. After sufficiently long action the liquid may be drawn off and fresh added. This arrangement of the experiment which would be suitable also for yeasts and molds, as the air necessary for the growth of the organisms may easily be forced in through the cotton filter, permits continuous operation to a certain degree. With the molds, the tube should not be quite filled with liquid, but an air space should be left in which an active mold surface may be formed. With these conditions it is likely that the solution under experiment could be allowed to flow through slowly and continuously. The action here should be, for this reason, a very rapid and complete one, as organism and liquid offer a large contact surface. Information concerning the preparation of nutrient solu- tions and nutrient gelatine, the sterilization of vessels and liquids, the production of pure cultures and the further culti- vation of the same in larger apparatus on the technical scale may be found in the work of Lindner : " Mikroskopische Betriebs- controle in den Gahrungsgewerben mit einer Einfiihrung in die Hefereincultur, Infectionslehre, und Hefenkunde. Mit vier Lichtdrucktafeln und 105 Textabbildungen. Verlag, Paul Parey, Berlin." On molds and their cultivation much information will be found in the work of Wehmer : ' ' Beitrage zur Kc-nnlniss einheimischer Pilze." In Part I, the citric acid- BY AID OF FUNGI 1 27 producing molds are discussed and the green molds thus far described of the genera Aspergillus {Eurotium) Sterigmato- cystis, Penicillium and Citromyces are compared in tabular form. Part II treats specially of the rotting of fruit and the varieties of fungi which grow preferably on or in solutions of organic acids. Part III (which has not yet appeared, 1898) will con- tain a monograph of the genus Aspergillus. Part I was pub- lished in 1893 by Hahn, Hannover and Leipzig, the following parts by Gustav Fischer, Jena. Finally, attention must be called to the book by Zopf, " Die Pilze," Breslau, 1890, pub- lished by Edward Trewendt, as the chapter on physiology has been well worked out. To chemists who wish to concern themselves with the study of racemic bodies it may be recommended to visit fermentation laboratories such as are found in Berlin, Munich, Vienna, Copenhagen, Xew York, Chicago, and elsewhere. It may be possible to obtain from these even large quantities of pure cultures of yeasts or fungi materials. According to a compilation by Winther 1 the following active forms have been obtained by the aid of fungi: /- Tartaric acid by Pasteur 2 from ammonium racemate by addition of a spontaneously fermented ammonium tartrate solution, and then by aid of Penicillium glaucum? d- Tartaric acid was obtained by Lewkowitsch 4 by the action of an unidentified schizomycete (vibrio), occurring in impure Penicillium cultures, on ammonium racemate. l-Gly eerie acid from the racemic ammonium salt by aid of Pen icilliu m gla ucu m ( Le wko witsch ) . 5 d-Gly eerie acid was obtained by Frankland and Frew 6 from r-calcium glycerate, to whose solution peptone, salts and calcium carbonate were added, by the action of Bacillus ethaceticus. Left-rotating calcium glycerate is produced which yields right-rotating free glyceric acid by treatment with oxalic acid. After long heating on the water-bath, solutions 1 Winther: Ber. d. chem. Ges., 28, 3022. - Pasteur: Compt. rend., 46, 615 (1858). 3 Pasteur : Ibid., 51, 298 (1860). 4 I^ewkowitsch : Ber. d. chem. Ges., 16, 1572. 5 1,6 wko witsch: Ibid., 16, 2720. 6 Frankland and Frew: J. Chem. Soc., 59, 96. 128 RESOLUTION OF RACEMIC BODIES of the acid yield a slightly soluble left-rotating anhydride. d- Lactic acid was obtained by Lewkowitsch, 1 also by Linossier from fermentation ammonium lactate by the action of Penicillium glaucum, P. Frankland and MacGregor 3 observed the formation of left-rotating calcium lactate, which furnished the right-rotating free acid, in the spontaneous fermentation of solutions of the racemic calcium salt to which calcium car- bonate, peptone, and nutritive salts had been added. Sarcolactic acid rotates to the left. d-Ethoxysuccinic add was obtained by Purdie and Walker 4 from the r-ammonium salt by Penicillium glaucum. The salts also are right-rotating. l-Aspartic acid. A moldy solution of the r-acid in the air becomes left-rotating (Engel). 5 l-Glutaminic acid split off by Penicillium glaucum (Schulze and Bosshard, 6 Menozzi and Appiani). 7 Active leucine, rotating to the right in water, and to the left in hydrochloric acid solution, was obtained from the racemic product by Schulze and Likiernik 8 by the aid of Penicillium. glaucum, the racemic compound having been formed by the action of hydrocyanic acid on isovaleraldehyde ammonia. It was obtained also from the r-compound produced from fer- mentation of caproic acid (Schulze). 9 By the use of beer yeast the following have been obtained from racemic sugars: l-Glucose (E. Fischer), 10 l-Mannose (E. Fischer), 11 l-Galactose (E. Fischer and Hertz), 12 I- Fructose (E. Fischer). 13 Lewkowitsch : Ber. d. chem. Ges., 16, 2720. I.inossier : Bull. Soc. Chim., [3], 6, 10. Frankland and MacGregor: J. Chem. Soc., 63, 1028. Purdie and Walker : Ibid., 63, 229. Engel : Compt. rend., 106, 1734. Schulze and Bosvh.ml /.tschr. physiol. Chem., 10, 143. Menozzi and Appiani : Chem. Centrbl., 1894, i, 674. Schulze and I.ikiernik : Ber. d. chem. Ges., 34, 671. Schulze: Ibid... 36, 56 ; Ztschr. physiol. Chem., 10, 138. -cher: Ber. d. chem. Ges., 33, 2620. Fischer: Ibid., 33, 382. '-' Fischer and Hertz: Ibid., 35, 1259. 18 Fischer: Ibid., 33, 389 ; 37, 2031. RESOLUTION BY AID OF FUNGI I2Q l-a-Propyleneglycol was formed in a 3 per cent, aqueous solution of the racemic compound made from glyceric acid after sowing an impure fungus culture from cheese, in w T hich Bacterium termo was abundantly present. The preparation must be perfectly freed from empyreumatic matters. Propionic and lactic acids were found as products of the fermented part ( LeBel ) . * The following active alcohols have been made from dilute aqueous solutions of the synthetic preparations and all by aid of Penicillium g la ucum : l-Methylethyl carbinol (Combes and Le Bel), 2 ' 3 l-Methyl-N-propyl carbinol (Le Bel), 3 l-Methylbutyl carbinol (Combes and Le Bel), 2 d-Ethylpropyl carbinol (Combes and Le Bel), 2 ' 3 d-Methyl-N-amyl carbinol (Le Bel). 3 d-Methylethyl carbincarbinol was obtained through fungi from a mixture of r- and /-amyl alcohol with destruction of the latter (Le Bel). 4 d-Mandelic add was obtained by Lewkowitsch 5 by addition of a few spores of a pure Penicillium culture to a solution of 3 grams of the r- ammonium salt in a liter of water containing i. 25 grams of nutrient salts with a little sulphuric or phos- phoric acid and carefully sterilized. l-Mandelic acid was formed in nine out of twelve experi- ments by application of an impure Penicillium culture, in which ,5*. ellipsoideus and an undetermined fungus (vibrio?) w 7 ere finally recognized in the liquid (Lewkowitsch). 6 The acid from amygdalin is left-rotating. d-Cinnamic acid dichloride has been separated by aid of Aspergillus fumigatus and also by yeast (Stavenhagen and Finkenbeiner). 7 l-hobutylpropylethylmethyl ammonium chloride was obtained by Le Bel 8 from the synthetic compound by use of a Penicillium culture which was not quite pure. 1 Le Bell: Bull. Soc. Chim., [3], 9, 678; Compt. rend., 93, 53?. 2 Combes and Le Bel : Bull. Soc. Chim., [3], 7, 551. 3 LeBel: Ibid., [3], 9,676. 4 LeBel : Compt. rend., 87, 213. r> Lewkowitsch : Ber. d. chem. Ges., 15, 1505 ; 16, 1569. r - Lewkowitsch : Ber. d. chem. Ges., 16, 1571. 7 Stavenhagen and Finkenbeiner: Ber. d. chem. Ges., 27, 456. 8 Le Bel: Compt. rend., 112, 725. 9 130 FORMATION OF ACTIVE ISOMERS E. Formation of Active Isomers 35. i. From Inactive Materials. Artificial Preparation of Active Compounds. The phenomenon, that active bodies as occurring in nature were always found to be inactive when prepared syn- thetically, led formerly to the opinion that substances endowed with the power of optical rotation could be produced only within the animal or vegetable cell. This view which was maintained particularly by Pasteur 1 had to be abandoned when in 1873 Jungfleisch" succeeded in producing tartaric acid com- pletely by synthesis, starting with ethylene which was con- verted into the dibromide, ethylene cyanide, succinic acid, dibromsuccinic acid and finally racemic acid, \vhich w r as split up into active components. Since it has been recognized that the inactivity of asymmetric synthetic compounds depends on their racemic structure, many of them have been produced in active forms. Among such which are found in nature, conine may be especially referred to, the complete synthesis of which was accomplished by Ladenburg in the following manner: 8 Starting with acetic acid this is converted into acetone, isopro- pyl alcohol, glycerol, allyl bromide, trimethylene bromide, tri- methylene cyanide, pentamethylene diamine, piperidine, pyri- dine, o'-picoline, or-ally! pyridine and finally into a-propyl piperidine = r conine, from which by splitting with tartaric acid the af-form, identical with the natural conine, is secured. A special method for the synthesis of active compounds, has recently been tried by Boyd 4 as he attempted to determine whether, when asymmetric bodies were formed in a magnetic field, one of the antipodes w r ould not be predominant, leaving the product endowed with rotating power and not racemic. Benzoyl formic acid was converted into mandelic acid by treatment with sodium amalgam in vessels kept in a magnetic field of 7,000 to 8,000 C. G. S. units, but it was found that the product was quite inactive, and also when the reaction was carried out in presence of active substances, such as df-tartaric acid or /-mandelic acid. Likewise, in the bromi- nation of stilbene, racemic stilbene bromide was formed. The 1 Pasteur: Compt. rend., 81, 128. * Jungfleisch: Bull. Soc. Chim. [2], 19, 194; Compt. rend., 76, 286. Ladenburg: Ber. d. chem. Ges., 32, 1403; Ann. Chem. (Uebig), 347, 80. Boyd : Inaug. Dissertation, Heidelberg, 1896. FROM INACTIVE MATERIALS 131 magnetic rotation to which the molecules are subjected during formation, leaves no permanent result. Relative to the synthesis of active substances, these con- ditions follow from the doctrine of asymmetric carbon atoms as first pointed out by van't Hoff (Chimie dans 1'espace, 1875, p. 20-28) : a. If a compound formed from ;a symmetric substance has but one asymmetric carbon atom, then a single racemic body always results. The same is true when the molecule of a resulting active compound contains two asymmetric carbon atoms, but consists of two similar halves. For example, from succinic acid only one racemic dibromsuccinic acid may be made. b. If a molecule containing two asymmetric carbon atoms, but not consisting of two similar halves, is derived from an inactive substance, the production of two racemic pairs is possible, the splitting of which must lead to four active isomers. It is not necessary that the two racemic bodies should be formed in equal amounts, but, because of different degrees of stability, the formation of one may be favored, and that of the other entirely suppressed. For example, cinnamic acid, C 6 H 5 CH = CH.CO. 2 H, may yield the following cinnamic acid dibromides : 5 65 65 65 II II H C Br Br C H H C Br Br C H H C Br Br C H Br--C H H C Br II I I CO 2 H CO 2 H CO 2 H CO,H First racemic body. Second racemic body. According to Liebermann and Hartmann, 1 the inactive bromination product of cinnamic acid does not probably con- tain the two racemic pairs, but only one, the splitting of which up to the present time has led to antipodes with rotations, \_oi\ D =-. : -j- 55 and 41. The other racemic compound has been obtained by Liebermann 2 by bromination of the labile allocinnamic acid, and decomposed approximately into com- ponents with \_a\ - -- -f 64 and 71. 1 Liebermann and Hartmann : Ber. d. chem. Ges., 26, 1664. a Liebermann : Ibid., 27, 2037. 132 FORMATION OF ACTIVE ISOMERS 36. 2. From Active Materials. The following cases may appear : a. A body with one asymmetric carbon atom may be converted into a compound with two such atoms. According to the van 't Hoff-Le Bel theory, an active molecule, CR 1 R 2 R 3 R 4 , whose right-rotating configuration is shown by I below, must furnish two isomeric products, la and Id, which are not corresponding i antipodes, by the introduction of the group, R 5 C R 6 . They possess, therefore, different properties, and may be pro- duced in unequal amounts. Likewise from the left-rotating configuration, II, the not-antipode isomers, Ila and lid are formed : Id. J R 4 -C-R 2 R 6 -C-R 5 R 3 II. Ila. lid. I 1 I' I' R 2 C R 4 R 2 C R 4 R, C R 4 RS RS C R 6 R 6 C RS R 3 RS In this case the original right- or left-rotating substance does not yield a racemic body by the chemical change, but a mixture of two active isomers which must possess unequal rotating powers. The relations are therefore different from what they are in the syntheses from inactive bodies. Among the above four isomers, each two form true antipodes ; viz., la with 11^, and Id with Ila, which may be combined to form two racemic compounds. A mixture of these two last must result when the racemic form, I -f- II, of the original substance is subjected to the chemical reaction, the final product being then inactive. As a matter of course, the case remains the same if the new asymmetric carbon atom is produced, not through addition , I. la. RI RI R 4 -C-R 2 R 4 -C-R 2 R 3 R 5 -C-R 6 1 R 3 FROM ACTIVE MATERIALS 133 but from one of the symmetric carbon atoms already present in the molecule. The above relations pointed out by van't Hoff, may be expressed in the following manner if the two asymmetric groups in the four isomers be represented by A and B : f i_ A I n The original -f compound furnishes the bodies: _ ^_ g -compound " and the two pairs of antipodes are : + A + B\ , + A -A-B] and -A + If the specific rotations of the four isomers are known, the rotating powers of the groups A and B may be calculated. These theoretical predictions have been frequently confirmed by experiment. If, for example, limonene be converted into the nitrosochloride compound, C 10 H 16 .NOC1, by treatment with amyl nitrite and hydrochloric acid with addition of acetic acid, which, with great probability, gives, CH 3 CH 2 CH 3 CH 2 Y Y *CH *CH /\ /\ H 2 C CH 2 H 2 C CH 2 I | 1 1 H 2 C CH H 2 C CH.NO Y *CC1 1 1 CH 3 CH 3 Limonene. Lirnonenenitrosochloride. there are always formed from the latter substance, according to Wallach, 1 two isomers (a and fi} which may be easily sepa- rated by their different solubilities in ether. There results from + Limonene {-nitrosochloride [a] f = + 3,3.4 fa- " " V4.8 - Limonene | ^ = _ 242.2 The two -compounds on the one hand, and the two /?-com- pounds on the other, represent optical antipodes. 1 Wallach: Ann. Chem. (Liebig), 252, 108 ; 270, 171. 134 FORMATION OF ACTIVE ISOMERS In the reaction the a-nitrosochlorides are formed in larger amounts than the ^-products; the first crystallize in well- formed monosymmetric prisms which dissolve easily in ether, but decompose quickly on keeping; the latter are finely crys- talline, difficultly soluble in ether, and much more stable. Chemically the a- and y#-isomers behave in a similar manner; the latter, however, in benzene solution show a double molec- ular weight (Wallach). 1 Dipentene (^-/-limonene) is an inactive product which is a mixture of two racemic bodies (Wallach). 2 Of the two asymmetric carbon atoms in the limonene nitroso- chlorides, one may be easily destroyed. This happens when each compound is treated with alcoholic potash solution by which, with loss of H -f- Cl, they become transformed into, carvoximes, Cj H l4 .NOH : CH 3 CH 2 CH 3 CH 2 Y C 1 1 *CH *C /\ /\ H 2 C CH 2 H 2 C CH 2 1 1 ' 1 1 H 2 C CH.NO HC C=NOH *CC1 Y 1 | CH 3 CH 3 Limonenenitrosochloride. Carvoxime. The carvoxime appears only in tw r o forms (d and /), and therefore, from the a- and yS-nitrosochloride of the same kind, the same product must result. This was found by Wallach 5 to be the case, a change in the direction of rotation following: From + j-_nitrosochloride j ^^ _ carvoxime [a]z) = _ 39 ^ { < " -h carvoxime " == + 39.7. b. If in a body which contains several asymmetric carbon atoms ; the number of the latter be increased by one, the same conditions which have just been discussed must obtain, the new substance 1 Wallach: Ber. d. chem. Ges., 28, 1308. 2 Wallach : Ann. Chem. (Liebig), 270, 175. Wallach : Ibid., 346, 227. FROM ACTIVE MATERIALS 135 must exist in two isomeric forms. E. Fischer 1 has furnished the experimental proof of this as he found that from <*-glucoheptose, C 7 H U O 7 , ([]/> = - 19.7, c= 10, water) two different glucooctonic acids, C fe H 16 O 9 , may be obtained by the cyanide reaction, of which the one, as lactone, has the rotation, [**]/> =- -f 45-9, and the other -j- 23.6 (in water, r 10). In these cases, it has been well established that the configuration formulas are : COOH COOH CHO H C OH HO C H H C OH H C OH H C OH I I I H C OH H C OH H C OH I I I HO C H HO C H HO C H I I I H C OH H C OH H C OH H C OH H C OH H C OH I I I CH 2 OH CH.OH CH 2 OH a-Glucoheptose. Glucooctonic acid. In the same way two isomeric rhamnohexonic acids, C 7 H U O : , whose lactones have the rotations [d] D =-. -f 83.8 and +43.3, are formed from rhamnose, C 6 H 12 O 5 (Fischer and Piloty). 2 On the contrary from mannose, C 6 H 12 O 6 , by aid of the cyanide reaction, only one of the mannoheptonic acids, C T H 14 O 6 , could be obtained ; ^-mannose furnished a left- rotating acid lactone ([0?]^ = - 74.2) ; /-mannose a right- rotating ([of\ D = -\- 75. 2). The two corresponding antipodes were therefore formed which united to produce a crystalline compound. This case shows that the formation of one of the two possible isomers may be particularly favored ( Fischer, 8 Smith, 4 Hartmann). 5 c. Increase of the number of asymmetric carbon atoms takes place also by the combination of active bases with active acids, in which case two pairs of antipodes with different properties must 1 Fischer : Ann. Chem. (lyiebig), 270, 64. '- Fischer and Piloty : Ber. d. chem. Ges., 23, 3104. 3 Fischer and Hirschberger : Ibid., 22, 370; Fischer and Passmore : Ibid. ,23, 2226. 4 Smith : Ann. Chem. (Liebig), 272, 182. 5 Hartmann : Ibid., 272, 190. 136 FORMATION OF ACTIVE BODIES result. Of salts of this kind, the following have been made : Marckwald 1 has combined d- and /-or-pipecoline with d- and /-tartaric acid, forming acid tartrates, the water of crystalli- zation, melting-points, and crystalline forms of which he determined. He did the same with the salts from racemic acids and bases. The results were : Melting-point. -f- Tartaric acid -f pipecoline The acid tartrates of d- and l-limonene-a-nitrolbenzylamine, C 10 H 16 .NO.NH.CH 2 C 6 H 5 , were examined by Wallach and Conrady 2 with respect to their rotation. The solutions in aqueous alcohol contained 0.97 to 1.38 per cent of salt : Water of Anhy- crystalli- Hydrated drous zation. Crystalline form. salt. salt. {monoclinic } hemimorphous > enantiomorphous j 65- 66 III- II2 C " ? 45- 126 " I 46 I " monociinic 85 -f tartaric acid -f- base, [a] D = 49.9 ( { " = + 69-6 " = - 69.9 From this the rotations follow : -f- base = 60 -|- acid = + 10 " = = + 60 " - 10 In the two bases a change of rotation follows on combination with tartaric acid. The following observations have been made by Fileti 3 on salts of d- and /-isopropylphenylglycolic acid ( \oi\ D = zb 135) with quinine and cinchonine : Melting-point- [<*]z> -f Isopropylphenylglycolic acid quinine, 192-193 79-4 204-205 - 118.4 -f- " " " -f cinchonine 201 -f- 136.8 167 + 83.4 From this the differences in the two not-antipodic isomers are apparent. 1 Marckwald : Ber. d. chem. Ges., 39, 43. 2 Wallach and Conrady : Ann. Chem. (Uebig), 353, 148. * Fileti : Gazz. chim. ital., aa, II, 395 ; Ber. d. chem. Ges., a6, IV, 89. IN THE ANIMAL OR VEGETABLE CELL 137 37. 3. Formation of Active Bodies in the Animal or Vegetable Cell. In the production of asymmetric bodies in the vegetable cell from inactive materials, it might be expected, as in artificial syntheses, that the two antipodes would be formed and racemic bodies result. Secondly, it might be possible that of the different configurations of a molecule, several, or indeed all might be formed at the same time. As far as experience has shown, up to the present time, however, of such possible plant isomers, only one is produced, and this generally one of the active forms. Of the hexoses only the right-rotating ^- glucose, of the ketoses, only the left-rotating ^-fructose appears (E. Fischer), 1 of the four tartaric acids, only the active right-rotating form. The same thing is seen in whole groups of vegetable substances, such as the bitter principles and alkaloids, which have all been found in one of the two active forms only. In the turpentine oils from the different species of pine, as well as in other ethereal oils there is found either ^-pinene and aMimonene, or, on the other hand, /-pinene and /-limonene ; but at the same time dipentene may also be contained in them, and we have here a case of the formation of a racemic body in the plant. A suggestion of the manner in which new active bodies may be made from others already present in the plant-cell has been given by E. Fischer. 2 This is based on the fact that in the artificial building up of sugars from others of a smaller number of carbon atoms by aid of the cyanhydrin reaction, the once existing asymmetry of the molecule is further continued. Imagine, for example, the conversion of mannose by the addition of cyanhydric acid three times into mannononose, and this then so split up that the original hexose would be reproduced, then the second compound w r ith three carbon atoms would be also an active system ; the first active molecule has produced a second one. In the same manner from the active substances in the chlorophyl grains, which are held by vegetable physiologists to be the seat of the formation of sugar, this latter body could be formed by the taking up of carbonic acid or formaldehyde, condensation, and final splitting 1 E. Fischer: Ber. d. chem. Ges., 27, 3230. 2 E. Fischer : Ibid., 27, 3231. 138 TRANSFORMATION OF THE ACTIVE ISOMERS off. As sugar, in turn, is used by the plant in the formation of other organic substances, these furnish the material for the production of new chlorophyl grains which again build up sugar, and thus, a direct and continuous creation of asym- metric molecules takes place. Similar views have been expressed by Stohmann. 1 In what manner the first active substance is formed in the plant-cell can not, of course, be explained, nor also, the reason why in one body the formation of a right-rotating and in another, of a left-rotating modification is preferred. The assumption that both forms are simultaneously produced and one immediately destroyed, that is, used to build up other substances, appears untenable, as the last process could not take place momentarily, and racemic bodies should then be found in part in plants, which, as remarked above, is very seldom the case. In the animal organism which is formed mainly of asym- metric substances, and receives such as food, the production of new active compounds by addition and decomposition can follow. With this, there is the possibility that inactive bodies present may take part in the changes, and so be con- verted into active substances. This is shown, for example, by the observation that brombenzene taken into the organism ap- pears in the urine as active bromphenylmercapturic acid. 2 It is remarkable that the proteid bodies of the animal kingdom, and also of the vegetable, show, without exception, left rotation. On the other hand, the bile acids nearly all ro- tate to the right. F. Transformation of the Active Isomers 38. Reciprocal Transformation of the Antipodes. A general method by which active bodies may be half converted into oppositely rotating antipodes consists in converting them into racemic forms and splitting these. In this way Lewkowitsch 3 obtained from /-mandelic acid the ^-acid. Piutti 4 by treating 1 Stohmann : Ztschr. f. Biologic, Jahrg., 1894. 2 Jaff6 : Her. d. chem. Ges., ia, 1092 : Baumann and Preusse : Ztschr. physiol. Chem., 5, 309; Baumann : Ber. d. chem. Ges., 15, 1731. 3 Lewkowitsch: Ber. d. chem. Ges., 16, 2722. I'iutti: Gaz. chim. Hal., 17, 126; Jahresbericht, (1887), 1664. TRANSFORMATION OF THE ACTIVE ISOMERS 139 = - 5 to 5.3 in acetone, c = 13 to 16) with phosphorus pentachloride with addition of chloroform, thus avoiding high temperature, active mono- chlorsuccinic acid results and in the right-rotating form. By now replacing the chlorine in this by hydroxyl (by boiling the aqueous solution, neutralized with potassium carbonate, with silver nitrate) malic acid results, and this rotates just as much to the right as the original acid did to the left. On the other hand, the */-acid so obtained yields /-monochlorsuccinic acid on treatment with phosphorus pentachloride and from this /-malic acid may be reproduced. We have thus a perfect cycle of changes. The conversion of either of the two malic acids into the other ma} r be brought about by converting the corre- sponding dimethyl ester, w T hich has the same rotation into the dimethyl chlorsuccinic ester by aid of phosphorus pentachloride, and in this conversion the direction of rotation changes. An explanation of the peculiar change in rotation and mol- ecular configuration which follows in the substitution of hy- droxyl by chlorine under the action of PC1 5 has been attempted by Armstrong. 2 Of a different kind are a number of observed changes of active bodies into oppositely rotating isomers, inasmuch as the latter are not the true antipodes of the original substances, which follows from the fact that racemic compounds do not result as end products. The inversion of /-menthone into d'-menthone, or the reverse, which follows by the action of weak or strong sulphuric acid, hydrochloric acid or alcoholic 1 Walden: Ber. d. chem. Ges., 29, 153. 2 Armstrong: Proceedings Chem. Soc., (1896), page 45. On the change of d-lactic acid into /-lactic acid, see Purdie and Williamson : J. Chem. Soc., 69, 837. 140 MODIFICATIONS OF INACTIVE CONFIGURATION potash, and which was discovered by Beckmann 1 is a case in point. In the same way /-ecgonine and /-cocaine may be con- verted by heating into right-rotating isomers (Einhorn and Marquardt). 2 See further the chapter on the direction of rotation of the derivatives of active bodies. 39. Reciprocal Transformation of Active Isomers of Different Configurations. Such changes were discovered, as is known, by E. Fischer, 3 and especially in the acids of the sugar group. They occur when these acids are heated with quinoline or pyridine to I3o-i50, in which the addition of the bases is employed mainly to prevent the production of lactones which interfere with the reaction. According to experience up to the present, a change of position of the H and OH on the asymmetric carbon attached to thecarboxyl group follows, and this reaction appears to be always reversible, so that the product is a mix- ture of the original acid with the one newly formed. Thus, the following stereoisomeric acids have been reciprocally con- verted, the one into the other. 4 C 5 H 10 6 /-Arabonic acid *1 9 v> ^TL d-Gluconic acid A X x 1 X X X " /-Gluconic acid X df-Galactonic acid X X X . . x x X /-Ribonic acid A xxx X d-Mannonic acid A XX v ^ XX A /-Mannonic acid A X X x XX' ^/-Talonic acid A ^ xxx A- /-Gulonic acid |^ /-Idonic acid A * x * ' ** X X X X X X X Besides these cases others are known. G. Inseparable Modifications of Inactive Configuration 40. As already explained in the chapter on the number of isomers possible in bodies with asymmetric carbon atoms ( 16) 1 Beckmann: Ann. Chem. (Liebig), 250, 342; 289, 362. 2 Einhorn and Marquardt: Ber. d. chem. Ges , 23, 468, 979. 8 E. Fischer : Ibid., 23, 799 ; 24, 2137, 3622 ; 27, 3193. * In the formulas A = CH 2 OH, X = CO 2 H, = H, X = OH. MODIFICATIONS OF INACTIVE CONFIGURATION 14! this type may appear in such molecules whose formulas may be divided into two equal halves. The inactivity may be explained by the equally strong but oppositely directed rotating power of the two halves which compensate each other. The correctness of this view follows from the fact that when the symmetry of such a molecule is destroyed, an active pro- duct results. This was shown first by E. Fischer. 1 As he found, the inactive mucic acid, CO 2 H (CHOH) 4 CO 2 H, yields by reduction racemic galactonic acid, CHO (CHOH) 4 CO 2 H, which may be split into active components, and conversely these, as well as the galactoses, may be con- verted by oxidation into inactive non-separable mucic acids. Inactive isomers of this kind are found in the following classes of symmetric bodies : i. In chain structure compounds with an even number of car- bon atoms. In this case, according to 16, the number, i, of n inactive modifications is given by the formula, i = 2 2 , where n is the whole number of asymmetric carbon atoms. The bodies of this group which are known are mainly these : Among those with n = 2, where i=i, there is, first of all, the meso- or antitartaric acid discovered by Pasteur 2 in 1853, in which the impossibility of resolution and consequent dif- ference from racemic acid (paratartaricacid) was shown. The configuration of this, as also that of erythritol, whose inactivity has also been hown, must be given, according to 16, by COOH CH 2 OH I I H C OH H C OH I I H C OH H C OH COOH CH 2 OH Mesotartaric acid. Erythritol. A further number of such symmetric bodies as the di- and tetrasubstituted succinic acids, also erythritol derivatives, hydrobenzoin, etc. , are undoubtedly likewise inactive. Among compounds with n =4, and consequently i = 2 we have : 1 E. Fischer: Ber. d. chem. Ges.. 25, 1247, 1260. 2 Pasteur : Compt. rend., 37, 162. See Pryzbytek : Ber. d. chem. Ges., 17, 1415. 142 MODIFICATIONS OF INACTIVE CONFIGURATION CH 2 OH COOH COOH I I I H C OH H C OH H C OH I 1 1 HO C H HO C H H C OH I 1 1 HO C H HO C H H C OH | | 1 H C OH H C OH H C OH 1 1 1 CH 2 OH COOH COOH Dulcitol. Mucic acid. Allomucic acid. With the octitols and their corresponding acids, four isomers with inactive configuration are possible, but we are not yet acquainted with them. Naturally no change follows in the above relations, if an even number of carbon atoms with two similar radicals attached (CH a groups) are introduced into the middle of the molecule, as is the case with : Dimethyladipic acids, ( CO 2 H *CH. CH 3 ) CH 2 CH 2 ( *CH. CH, CO 2 H ) . Diallyl bromides, (CH 2 Br *CHBr) CH 2 CH 2 (*CHBr CH. 2 Br). and others. 2 . In chain-structure compounds with an uneven number of carbon atoms. Here the number of possible isomers is dependent on the form of combination of the radicals with the middle carbon atom. a. If the middle carbon atom besides being united with the two symmetric groups is joined to two other radicals, similar to each other, as in the tf-dimethylglutaric acids, (CO,H *CH.CH 3 ) CH 2 (*CH.CH 3 -C0 2 H), then the number of inactive isomers may be calculated by the formula used for the bodies of the last group. b. If, on the other hand, the middle carbon atom, besides being attached to the symmetric groups, is joined to two other dissimilar radicals, then the number of inactive isomers may be found by the formula, i = 2 , where n is the number of asymmetric carbon atoms, the middle carbon atom being included as one. In reality, however, this atom should not be con- sidered as asymmetric, because, as a glance at the following configuration formulas will show, a plane of symmetry may be MODIFICATIONS OF INACTIVE CONFIGURATION 143 passed through the same, and the dissimilar radicals joined to it. If this middle carbon atom is excluded, the number of n inactive isomers is given by the formula, t = 2 2 The following are bodies of this class : CH,OH CH 2 OH C0 2 H C0 2 H 1 1 i H C OH H C OH H C CH, H C CH 3 H C OH HO C : H H C CH 3 H 3 C-C H H C OH H C OH H C CH 3 H C CH 3 CH 2 OH C :H,OH C0 2 H CO 2 H Adonitol (ribitol"). Xvlitol. , Ribotrioxylglutaric Xylotrioxyglutaric Trimethylglutaric acids, acid. acid'. Bodies which contain five, or according to the above con- siderations four, asymmetric carbon atoms, furnish four inac- tive isomers of which at present tf-glucoheptonpentoldiacid, OH OH H OH OH I I I I I C0. 2 H C C C C C C0 2 H I I I I I H H OH H H and ar-glucoheptitol are known. 3. In cis-form cyclic compounds. The simplest case is given by the i,2-trimethylenedicarboxylic acids, where, as is known, three isomers are possible: Trans forms: Cis form: III. H CO 2 H CO.,H H CO 2 H CO 2 H J I C- -C f\ H /\ H \|/ H 1 I -I Active. Active. Inactive. I. : C0 2 H 1 c II. CO,H H \ H , ' H A\?/ / C0, Racemic form. The asymmetric symbols, I and II, which stand in the relation of object and image to each other, correspond to the active forms, w r hile the third symbol possesses a plane of symmetry, passing through the group CH 2 , and therefore represents an inactive type. In an analogous way are related the hexahydro- 144 MODIFICATIONS OF INACTIVE CONFIGURATION phthalic acids, the hexahydroisophthalic acids, the ^-tetrahydro- terephthalic acids, and others, in which the racemic modification (trans form) and the structural!}' inactive (cis) form are known. 41. Differences in the Properties of Racemically Inactive and Structurally Inactive Isomers. These have been observed with respect to : a. Water of Crystallization. For example : Calcium mesotartrate CaC 4 H 4 O 6 -f 3CH 2 O (Anschiitz) 1 Calcium racemate CaC 4 H 4 O 6 -f 4H,,O Calcium d- and /-tartrate CaC 4 H 4 O 6 -f 4H 2 O Free mesotartaric acid, like racemic acid, crystallizes with one molecule of water ; the inactive tartaric acids, on the contrary, are anhydrous. b. Melting -Point. As may be seen from the following obser- vations, the melting-point of the racemic modification is gen- erally higher than that of the inactive modification, although the reverse relation also appears. In the last cases, however, including erythritol and some of its derivatives, it is possible that mixtures, rather than racemic compounds, were examined : Racemic- ally inactive. Configu- rationally inactive. Diff. RC Observer. Racemic or tartaric acid, anhyd. 205-206 IQd 140-143 1 2O + 6 4 1 7.1 f Bisch. \ andW. 2 Walden 3 1^4 128 T /4 1 f\A L Tf* y Q.J/ oon^ 1 4 1 Afifn} < < 229 122 127 QQ IOI i 4\yj 1 2/1 C t< 3 T^rvthritol 72 yy iui rT Q I '^O A(\ /* 8l I 7^ 40 << 5 O Mo 0^ t < i , 2-Trimethylenedicarboxylic acid y i75 *33 '34 139 37-5 + 36 + 27 Buchner s ^ 1 v) iy^ 2 3 J 2 -Tetrahydroterephthalic acid . . 220 150-155 43 + 67 v. Baeyer 9 1 Anschiitz : Ann. Chem. (I^iebig), 326, 197. 2 Bischoff and Walden : Ber. d. chem. Ges., aa, 1815. Walden : Ztschr. phys. Chem., 8, 467. /bid., p. 487- Griner : Compt. rend., 116, 723; 117, 553. Buchner : Ber. d. chem. Ges., 33, 703. v. Baeyer : Ann. Chem. (Liebig), 358, 218. Perkin : J. Chem. Soc., 59, 813. v. Baeyer : Ann. Chem. (Uebig), 351, 308. MODIFICATIONS OF INACTIVE CONFIGURATION 145 c. Solubility. In the cases which have been studied, the compounds inactive by configuration have been always more readily soluble than the racemic forms. One hundred parts of water dissolve : Racemic- ally inactive. Confi^u- rationally inactive. Active. Observer. Tartaric acids, Acid potassium anhydrous at 15 tartrate at 19 ... 17.1 pts. 0-555 " 125.0 pts. 12.5 " 132.2 pts. 0.535" Bischoff& Walden 1 z7 2 Tetrahydrote '" > cold water . . 0.170 " 2.70 " 1 v.Baeyer 2 d. Constant of Dissociation, K, determined from electrical conductivity. The following observations exhibit no regularity in this relation: Racemic- ally inactive. Configu- rationally inactive. Diff. RC. Observer. Racemic acid, mesotartaric acid, 0.097 0.060 -f 0-037 _i_ o 0068 Walden 5 4 Diethylsuccinic acid O O2d^ O O'i/lt o 0008 ,, o 026 o 006 4< OOQCC O OO^^ o an d take A B - = a and - = b, a oc we have next to determine the value, a from which the desired wave-length, expressed in millionths of a millimeter (A " " " Joo , o. t = a (i -f 0.031463 1 + o.o 7 329 /) 20 " 100. For low temperatures, Soret and Guye 4 found : ' + o C 1 -f 0.031326 /), holding between 55 and +23, <*'-ho (i + 0.031265 /), " 72" + 18. According to Le Chatelier 5 the rotation increases irregularly at high temperatures. Based on observations made by him, that quartz undergoes a sudden change in dimensions at 570, he calculates the following formulas : Between o and 570 : / = v (i + o.o 4 96 / + o.o 6 2i7 f 1 }. At 570 there is an increase, d a = 0.043 a o- Above 570 this expression holds : 6 a = u (1.165 + o.o 4 is (t 570)). The angles of rotation found by Le Chatelier are : / Diff. ry ^P A = 518 A = 500 A = 448 MM ' Mg-line 20 17-25 21.72 28.62 30.78 39-24 280 260 18.06 22.68 29.82 32.16 40.80 415 135 18.60 23.40 30.60 32.90 42.OO 560 145 19.38 24.30 32.04 34.56] 44.10 600 40 20.10 25.26 .33.18 35.76 45.60 900 300 25.32 [33.24 36.00 45-84 In these figures the sudden strong increase between 560' and 600 may be recognized. v. Lang : IJDC. cit. Sohncke : Wied. Ann., 3, 516. Joubert : Compt. rend., 87, 497. Soret and Guye : Ibid., 115, 1295 ; 116, 75. Le Chatelier : Ibid., 109, 244 (1889). Confirmed by Gumlich : I^oc. cit., p. 230. ROTATION DISPERSION OF CRYSTALS 153 According to v. Lang, 1 Sohncke 2 and Le Chatelier 3 the tem- perature coefficient remains constant for rays of all wave- lengths, but more exact measurement would probably disclose differences here. The rotation of quartz at low temperatures has been care- fully investigated by Soret and Guye. 4 For a plate having the thickness of i mm. at 12 the following rotations of the sodium ray w r ere found: t= 71.5 -55-3 -42.3' o 21.655 + 17-7 21.719 + 22. 7 C , = 51.470 21.505 21.537 21.655 21.719 21.730. These observations may be expressed with satisfactory accuracy by the interpolation formula of Joubert given above. 2. Sodium Chlorate, NaClO 3 . The angles of rotation of the regular crystals were determined by Guye 5 and also by Sohncke 6 for different rays and temperatures. Based on a thickness of i mm. these values were found: Guye. Line. Guye. a at 2O Sohncke. a at 21. Relation of quartz to sodium chlorate. a 2.070 6.908 B 2.273 2.38 6.927 C 2.503 2.52 6.915 D, 3.128 D 3.16 6.9361 A 3-I32 6.9362 B 3-944 3.96 6.982 F 4.670 4.61 7.013 G 6.005 5-89 7.089 h 6.675 7.H5 H 7-174 6.86 7.134 For increase of rotation with temperature : Guye. . ... a, = a (i -f 0.000586 /), for t = -f 5 to Sohncke... a, = o (i + 0.00061 /), " *?=+ l6 " 28 1 Loc. cit. 2 Loc. cit. 3 Loc. cit. * Soret and Guye: Arch. sc. phys. et. nat. Geneve [3], 29, 243 (1893). 5 Guye: Arch, de Geneve, [3"], 22, 130(1889). The values for fourteen ultraviolet lines are omitted from the table. Sohncke: Wied. Ann., 3, 529 (1878). 154 PHYSICAL LAWS OF CIRCULAR POLARIZATION Among other active crystals the following have been inves- tigated with respect to their rotating power: Sodium periodate, NaIO 4 -f- 3H 2 O. Potassium dithionate, K 2 S 2 O 6 . Lead dithionate, PbS 2 O 6 -f- 4H 2 O. The data for these have already been given in 6. 45. Rotation Dispersion of Liquid and Dissolved Substances. As yet there are but relatively few observations on record in this direction, of which the older ones (on cane sugar, santonin, turpentine, bile acids, etc. ) were made by the Broch method 1 and are based on the Fraunhofer lines, while the later obser- vations have been made by the ray filter method. 1 The numerical data will be found in part VI, "Constants of Rotation." With substances which are in themselves liquid the rotation, as far as is known, shows a normal behavior ; that is, the amount of rotation increases with the refrangibility of the light. This is the case with the terpenes and also in a series of derivatives of amyl alcohol, lately studied by Guye and Jordan," which all have the simple molecular \veight. 3 These bodies possess very different dispersive powers, which for each substance is a characteristic constant and may be expressed by the difference [#] violet [^]red, and designated as the specific rotation dispersion. The same normal behavior is observed with many substances in solution (cane sugar, dextrose, santonin and derivatives, bile acids and others). The following table will serve for the comparison of the dispersive powers of different substances with each other, and also with that of quartz, the figures referring to the rotations for the lines, B, C, D, E, /^and G : 4 1 See Part IV: Determination of Rotation Dispersion. 8 Guye and Jordan: Cotnpt. rend., 122, 883 (1896). 3 Ramsay and Shields: Ztschr. phys. Chem., 12, 433. 4 Cane sugar in water,/ = 101030. Stefan : Wien. Sitzber, 52, II. 486. Cholalic acid in alcohol, c = 3. Hoppe-Seyler : J. prakt. Chem., [i], 89, 257. Cholesterol in petroleum, c = 10. Lindenmeyer : J. prakt. Chem., [i], 90, 323. Turpentine oil and lemon oil, G. Wiedemann : Pogg. Ann., 82, 222. Santonin in chloroform, c =31090. Nasini : Accad. d. Iincei, [3], 13, (1882). ROTATION DISPERSION OF LIQUIDS, ETC. 155 BCD E F G Quartz a i mm 15.75 17.31 21.71 27.54 32.76 42.59 Cane-sugar [a] i dm -j- 47.56 52.70 66.41 84.56 101.18 131.96 Cholalic acid-., [a] " + 28.2 30.1 33.9 44.7 52.7 67.7 Cholesterol [a] " - 20.63 25.54 31.59 39.91 48.65 62.37 Turpentine oil .. a " - 21.5 23.4 29.3 36.8 43.6 55.9 Lemon oil a " + 34.0 37.9 48.5 63.3 77.5 106.0 Santonin [a] " +484.0 549.0 754.0 1088.0 1444.0 .... If we calculate from these figures how much more strongly the rays C, D, E, F and G are rotated than is the ray B, the following so-called coefficients of dispersion result : B C D E F G Quartz i i .09 Cane-sugar Cholalic acid Cholesterol . . . Terpentine oil Lemon oil Santonin i. n 1.07 1.24 1.09 i. ii ,1.1 .38 1.75 2.08 2.70 .40 1.78 2.13 2.77 .20 1.59 1.87 2.40 .53 1.93 2.36 3.02 .36 1.71 2.03 2.60 .43 1.86 2.28 3.12 .56 2.25 2.98 From this, it appears that the relation between quartz and cane-sugar is nearly constant, these substances having nearly the same rotation dispersion, while other bodies disperse either more or less strongly than quartz. This fact is applied in the construction of polarimeters with quartz-wedge compensation (Soleil's color saccharimeter and the Schmidt and Haensch half shadow saccharimeter), when, as is usual, white light is employed. The construction of these instruments presupposes equality in the dispersive power of the active substance investigated with that of the quartz- w r edges, and, therefore, substances which depart much from this relation cannot be used or studied with them. In studying the dispersion ratio between cane sugar and quartz some new observations of Seyrfart 1 may be used which are based on the rotation of sugar solutions for seven artificial spectral lines. If, as a basis, a solution be taken which rotates the red hydrogen line (Ha) through the same angle as a quartz plate i mm. in thickness, that is 17.31, then for the other colored rays the following angles are found which are given along with the angles for quartz for comparison. (For some of the last, marked with a *, the values were found from the 1 Seyffart: Wied. Ann., 41, 113, 128 (1890). 156 PHYSICAL LAWS OF CIRCULAR POLARIZATION above quoted measurements of Soret and Sarasin by aid of the Boltzmann interpolation formula.) Angle of rotation. Color. Line. Red ............ //a(C) Yellow ......... Na(D) Green .......... 77 Greenish-blue .. ffp(F) Blue ........... Sr Wave-length. Quartz. Cane-sugar. Difference. 656. 2MM 17.31 I7.3I 589.2 21.72 21.78 0.06 535-0 26.64* 26.81 0.17 486.1 32.76 32.98 O.22 460.7 36.77* 37.18 0.41 434-1 41.88* 42.44 0.56 420.2 45-oo* 45.78 0.78 Violet .......... Rbn From these figures the following dispersion coefficients are calculated : Ho. Na 71 H$ Sr H y Rbn Quartz ...... I 1.255 1.539 I - 8 93 2 - I2 4 2.419 2.600 Cane-sugar., i 1.258 1.549 1-905 2.148 2.452 2.645 It is seen that the dispersive power of sugar exceeds that of quartz, but the difference is appreciable only in case of the blue and violet rays, which, on account of their low luminosity, are but little used in the quartz wedge saccharimeters. The dependence of the specific rotation of cane-sugar on the wave-length, A, of the light employed, may be expressed by a formula derived by Seyffart 1 from observations on a 20 per cent, solution. A is expressed in millimeters. iyu _ 2. 16036 5.47276 l^/ 5 io ls .A* As examples of substances with larger dispersion, coupled with high specific rotation, we have the following santonin bodies, investigated by Nasini :* Santonide Parasantonide Fraun- Wave- length. Santonin Ci5HigO3 Solution in alcohol. ClsHjgOg CiftHjgOg Solution in chloro- Solution in chloro- fomi . f nrtn hofer Angstrom. c 1.782 c = 3 to 30 c _ - t o 'c O lines. MM t = 2Q /=20 / = 20 to ] w B 686.7 110.4 1. 00 + 484 I.OO + 580.5 I.OO C D 656.2 589.2 118.8 161.0 1.08 1.46 549 754 i? 655.6 891.7 1.13 i-54 E 526.9 222.6 2.02 1088 2.25 1264 2.18 b. 518.3 237- I 2-15 1148 2.37 1334 2.30 F 486.1 261.7 2-37 1444 2.98 1666 2.87 e 438.3 380.0 3-44 2201 4.55 2510 4-32 g 422.6 2610 i 5.39 2963 5.10 Seyffart : Loc. cit. Nasini : Accad. d. Uncei, Cl. sc. fis. mat. e. nat, [3], 13, 18*2. ROTATION DISPERSION OF LIQUIDS, ETC 157 If an active body is dissolved in different liquids, the rotation dispersion remains the same in all the solutions. This was shown by Gennari 1 with mixtures of nicotine with water, methyl alcohol, ethyl alcohol and benzene. Finally, as regards the effect of temperature on rotation dis- persion, Gernez 2 has observed that the dispersion suffers no change in turpentine oil, orange oil, bitter orange oil and cam- phor, and not even when, by aid of heat, the substances are brought into the condition of vapor. 46. Anomalous Rotation Dispersion. With certain substances in solution the phenomenon is observed that the rotation of the plane of polarization does not change regularly with increasing refrangibility (or decreasing wave-length) of the rays, but that for some color lying between the red and violet ends of the spectrum it has a maximum, or, also, a minimum. Further than this, the rotation may be the same for a number of different rays. Such anomalies, which, as is known, may appear also in respect to refractive dispersion have been observed in the following substances. d- Tartaric Add, The irregularities in the rotation dispersion of this substance were discovered by Biot 3 and later investi- gated by Arndtsen. 4 The latter determined the increase in the specific rotation of the acid in solutions of decreasing per- centage strength, for different Fraunhofer lines, employing the method of Broch, and calculated the following interpolation formulas, holding for a temperature of 24, q representing the percentage amount of water present: Spectral Wave-length, lines. fifji C 656.2 [a] = -f 2.748 -f 0.0945 q D 589-2 [a] = -r 1.950 + 0.1303? E 526.9 [a] = + 0.153 -f O.I75I q d l 518.3 [a] = 0.832 + 0.1915 q F 486.1 [a] = - 3.598 + 0.2398 q e 438.3 [3 = - 9- 6 57 + 0.3144 q If from these formulas, which hold to q = 40 (the strongest solution of tartaric acid which can be made at 24 contains 60 1 Gennari : Ztschr. phys. Chern., 19, 130 (1896). Gernez : Ann. de 1'ecole norm., i, i. 3 Biot. Mem. del'Acad., 15, 93 (1838). 4 Arndtsen : Ann. chim. phys., (3), 54, 403. Pogg. Ann., 105, 31*. 158 PHYSICAL LAWS OF CIRCULAR POLARIZATION per cent, of the acid), but which certainly apply for q = 30, we calculate the specific rotations of solutions containing from 30 to 90 per cent, of water, we obtain the following numbers, given in order of increasing concentration: In 100 parts of solution. Red. Yellow. Green. Green. Greenish blue. Blue. Water 9 Tartaric acid. We []* M* a]* 0> [3. 9 10 11.25 13-68 15.92 16.40 I7.98 18.64 80 20 10.30 12.37 14.16 14.49 15-59* 15-49 70 3 9-36 11.07 i?.4r 12.57 I3-I9* 12.35 60 40 8.42 9-77 10.66 10.66 10.79* 9- 21 50 50 7-47 8.47 8.91* 8.74 8.39 6.06 40 60 6-53 7.16* 7.16* 6.83 5.99 2.92 30 70 5.58 5-86* 5-41 4.91 3.60 0.23 From this table it is evident that for some particular color each solution shows a maximum in its rotating power which is indicated by a *. For the weakest solution, with ten per cent, of tartaric acid, the maximum occurs normally at the color of greatest refrangibility, e\ but with increase of con- centration it passes gradually toward the red end of the spectrum, the 50 per cent, solution exhibiting the maximum rotation for the green ray, and the 70 per cent, .solution for the yellow. In these cases the rotation, after passing the maxi- mum point, decreases with increase in refrangibility and becomes finally negative for blue light with the most concen- trated solution. This solution must therefore be perfectly inactive for some color between F and e. It is also evident that for certain concentrations different rays are rotated through the same number of degrees; thus the specific rota- tion of the solution with 40 per cent, of acid is 10.66 for E as well as for b, and for the 60 per cent, solution it is the same for D and E, vis., 7.16. The left-hand rotation which with a 70 per cent, solution appears for blue light, would increase and show even with the less refrangible rays, if it were possible to pass to more con- centrated solutions. The anhydrous acid, whose specific rotation is expressed by the first constants of the Arndtsen ROTATION DISPERSION OF LIQUIDS, ETC. 159 formulas, must, as shown by the sign, exhibit right-hand rotation for the rays C, D and E, and left-hand rotation for b, P and e. In fact, Biot 1 was able to observe such opposite rotation with plates made by pouring a mixture of tartaric acid melted with a little water ; and Arndtsen 2 found that left rotation for the strongly refrangible rays is shown when con- centrated alcoholic solutions of the acid are used. The anomalies in the rotation dispersion of tartaric acid disappear when the solutions are examined at a higher tem- perature (Krecke), 3 or when a little boric acid is added (Biot). They are not shown with the salts of tartaric acid (Biot), which fact has recently been confirmed by Rimbach 4 in the case of rubidium tartrate. Malic Acid. The rotating power for different kinds of light has been investigated by B. A. Woringer 3 by aid of the ray- filter method, and using solutions, the amount of water in which, q, varied from 49 per cent, to 93 per cent. The follow- ing interpolation formulas were derived, based on a tem- perature of 20 : Red \ = 665.9 W> [ a ] = 4-605 0.0709 q Yellow "-=5919" " = 6.544 0.0957" Green "=533.0" " = 8.349 0.1128" Light blue "= 488.5 " " =10.1210.1298" Dark blue "=448.2 " " =14.971 0.1730" The specific rotations calculated from these are : TABLE I. Malic acid. Water. Dark- P Q .K.CU i e 1OW ij re en bh ie blue 50 5 4- 1.06 + I .76 + 2.71 + 3 .63 + 6.33 40 60 -fo.35 + 81 4- I. 58 + 2 33 + 4-59 30 70 0.56 -r 0.15 - 45 + I 03 4- 2.86 20 80 - 1.07 i ii* 67 O .27 4- 1.13 IO 90 -1.78 2.07* I. 80 I .56 -0.59 8 92 -1.92 2.26* 2. 03 I 84 -0.95 5 95 2.13 - 2.55* 2. 37 2 23 - 1-47 1 Biot : Ann.chim. phys , [3], 28, 351. 2 Arndtsen: Ann. chim. phys., [3], 54, 415. 3 Krecke: Arch. Neerland., 7, (1872). 4 Rimbach : Ztschr. phys. Chem , 16, 671. 5 Woringer : Investigations in the author's laboratory, not yet published (1898). i6o PHYSICAL LAWS OF CIRCULAR POLARIZATION In another series of observations carried out by Nasini and Gennari, 1 in which the ray-filter method was employed, the yellow light, however, being that of the sodium flame, the following specific rotations, for a temperature of 20 were obtained : TABLE II. No. 4* P q [1* [ ]* Mir [] EJ- i 1.3454 72.79 27.21 + 1.80 + 2.86 + 3-90 + 5-20 + 6. 39 2 1.2723 59.02 40.98 + 1-35 + 2.08 + 3-5 + 4.21 -f 5.63 3 1.1861 42.80 57-20 + 0.19 + 0.55 -f 1.18 + 2.08 + 3-29 4 1.1423 34-27 65-73 - 0.18 + 0.07 + o.5i + 1.64 + 2.20 5 LI395 33-24 66.76 - 0.41 - 0.31 + 0.07 + 0.46 -f 0.86 6 1.1239 30.02 69.98 - 0.51 - 0.42 - 0.05 + 0.29 + 0.72 7 1.1193 28.72 71.28 - 0.79 0.67 0.46 - 0.22 + 0.29 8 1.1034 25.67 74.33 - o.94 - 0.81 - 0.69 - 0-39 + 0.14 9 1.0663; 16.84 83.16 - 1.07 - 1.28* - 1-05 - 0.62 0.00 10 1.0635! 16.24 83.76 - 1.28 - 1.46* - 1.30 0.91 -0.36 ii 1.0304 8.23 91.77 - 1.09 - 1.09 - i .08 - 1.09 - 1.08 12 1.0156 4.61 95.39 - 1.87 - 1.17 - 2.56 - 2.45 - 2.51 From these two tables the following appears : 1. Concentrated solutions of malic acid, with amounts of water varying from 27 to 60 per cent., exhibit right hand rota- tion for all colors, which increases normally with the refrangi- bility of the rays. 2. In solutions with q = 66 to 75 per cent. (4 to 8 in Table II) the less refrangible rays rotate to the left and the stronger to the right. The point of inactivity which is passed here moves, with increasing dilution, toward the violet end of the spectrum. 3. Anomalous dispersion is shown in solutions containing more than 80 per cent. (Table I) or 83 to 84 per cent, of water (Table II, No. 9 and 10); a maximum rotation (-)-) is shown for yellow rays. 4. In Table II the phenomenon appears that in a solution with about 92 per cent, of water there is equally strong rota- tion for all colors, and that one with q = 95 shows a minimum rotation for yellow light, and then increasing left-hand rotation toward the violet end. i Nasini and Gennari : Ztschr. phys. Chem., 19, 113 (1891). ROTATION DISPERSION OF LIQUIDS, ETC. 161 There is less certainty about these observations however, because in such weak solutions the observed angles of rotation are very small. Such anomalies do not appear in the results calculated for q = 92 and 95 from the interpolation formulas of Woringer, but a maximum is shown for the yellow rays. Boric acid does not correct the irregularities in the dispersion of malic acid (Nasini and Gennari). Equimolecular Mixture of Nicotine and Glacial Acetic Acid, Treated with Water, In such combinations, Gennari 1 has observed the appearance of anomalous dispersion, but only within very narrow limits of concentration. The specific rotation of nicotine was found in the following mixtures by means of the ray-filter method (for yellow, the sodium flame however) at a temperature of 20 : [Jred [a]^ [L. [] [1- Pure nicotine | d\ = 1.0107 j -123.37-162.84 209.78 250.71 -317.79 Equimolecular mixture] 143.54 acetic acid. J Mixtures of the same with water. Orijrinal mixture. Water g. 81.26 18.74 9.20 12.57 -1683 21.99 26.68 78.87 21.13 -4-13 5.82 8.20 10.92 77-84 22.16 1.70 2.73 4.39 -6.63 9-93 77-45 22.55 0.12 0.54 1.48 2.80 5.00 76.39 23.61 0.78 -ro.52 -0.75 76.30 23.7 "-I.36 -0.98 -fo-74 1.40 76.23 23.77 -| 5.90 +7-44 -8.85 -9.87 76.10 23.90 -12.92 4-16.74 -20.72 +24.56 28.00 Specific rotation of nicotine acetate Salt. Water. 53-72 46.28 -} 16.44 -21.36 25.81 -29.05 . .. 44-3 55.70 -14-30 18.85 -22.83 -26.57 31.37 26. d8 73-52 V2I 17-35 -21.23 -23.98 24.28 75.72 i ;.co 1696 20.41 -23.50 25.84 1 Gennari : Ztschr. phys. Chem., 19, 130. II 162 PHYSICAL LAWS OF CIRCULAR POLARIZATION As immediately apparent, pure nicotine and the mixture of this with one molecule of acetic acid possesses a normal left- hand rotation, increasing with the refrangibility of the light. As nicotine acetate rotates to the right it appears that a part o the nicotine in the mixture must exist in the uncombined condition. The peculiarities in dispersion which appear when water is added to the equimolecular mixture of acetic acid and nico- tine, can be best seen from the following diagram. This extends from the red only to light blue, as [a] for dark blue had to be omitted on account of incompleteness in the observations: Left-rotation. (a) blue +-** red. 2220 18 16 14 12 10 8 6 4 2 2 4 6 Right-rotation. red *>-* blue. (a] 8 10 12 14 16 18 20 22 24 26 q = 18,74 21 13 2216 22.55 2361 2370 2377 23,90 It will be seen that: 1. In mixtures with 18.74 to 22 -55 P er cent, of water there is normal left-hand rotation for the different colors and that this decreases, the dispersion also, with increasing amount of water. 2. The solutions containing from 23.77 to 23.90 per cent, of water possess right rotating power, which, with the disper- sion, increases in marked degree by the slight increase in the amount of water. 3. With an amount of water varying from 23.61 to 23.70 per cent., the rotation and also the dispersion are very small, the first solution showing complete inactivity for green rays and the second for light blue rays. We have here then the case of appearance of a minimum of rotation with increasing refrangibility of the light. ROTATION DISPERSION OF LIQUIDS, ETC. 163 Nicotine acetate shows normal dispersion in solutions con- taining 46 to 76 per cent, of water, as appears from the above table. Of interest finally are the conditions of dispersion in solu- tions of nicotine in acetone, ethyl alcohol and propyl alcohol, for which Nasini and Gennari 1 found the following specific rotations: P C [a] red []/> [Lr M* []< Acetone 22 OO 2O 77 A Q1 6 01 7 IO 7 C1 8 90 ^VJ. // ^Vo /*** /OO Ethyl alcohol 21.40 19.09 -5-73 -7.09 9.01 -9.71 - 10.32 Propyl alcohol 21.15 19.06 -3-30 -3-62 - 3-92* -3.88 - 3-07 The rotation and dispersion of nicotine in these solutions are seen to be very much decreased. With the propyl alcohol there is even anomalous dispersion as the green ray appears to have suffered the greatest deviation. In regard to the explanation of anomalous rotation disper- sion, Biot 2 has shown at length that the phenomenon must appear when the ray of light passes two liquid layers which rotate the plane of polarization in opposite directions, and which at the same time possess different dispersive powers. He employed in experiments left-rotating turpentine and a solution of right-rotating camphor in acetic acid, which were contained in tubes placed one behind the other, and showed by calculation how r the length of the column of the camphor solution and its concentration must be changed to reach a maximum or minimum of rotation for a given ray. That a perfect achromatism, that is equally strong rotation for all rays is possible, has been shown by investigations of Nasini and Gennari. 3 The same conditions naturally obtain when two oppositely rotating liquids are mixed in different proportions. Thus, v. Wyss 4 was able to show a maximum of rotation for yellow 7 light of w T ave-length 565 up in mixtures of right and left turpentine oil. Further, Genuari 5 found that if to a solution 1 Nasini and Gennari: Ztschr. phys. chem., 19, 117. 2 Biot: Ann. chim. phys.. [3], 36, 405 (1852). ' Xasini and Gennari: loc. cit., p. 121. 4 v. Wyss : Wied. Ann., 33, 554 (1888). 5 Ztschr. phys. Chem., 19, 134 (1896). 164 PHYSICAL LAWS OF CIRCULAR POLARIZATION of right-rotating nicotine sulphate, left-rotating nicotine be added gradually, in small portions, the rotation for the different colors grows less and less, and that finally the left rotation of the nicotine appears in increasing degree for all rays. There is here the same condition as with the above-mentioned mixtures of nicotine, acetic acid and water, and the phenomena which these exhibit may be explained by assuming that the different solutions contain at the same time right-rotating nicotine acetate and left-rotating free nicotine in variable proportions. But on what the anomalous dispersion of the aqueous solu- tions of malic acid and tartaric acid depends, has not yet been definitely shown. An attempt at an explanation will be referred to later in 63. Finally, a special class of anomalous dispersion phenomena must be mentioned, which appear when the active solutions are colored. If white light is sent through these, absorption of some of the rays follows, and the rotation of those remaining does not then change regularly with the .wave-lengths. Cotton 1 has investigated such conditions in solutions of copper tartrate and chromium tartrate in potash. i Cotton : Compt. rend., 120, 989, 1044 ; Ann. chim. phys., [7], 8, 347 (1896). PART THIRD Numerical Values for the Rotating Power. Specific Rotation 47. In the measurement of the optical activity of liquids and dissolved bodies, Biot 1 introduced in 1835 the conception of specific rotation , indicating by this term, the angle of rotation \_oi\ , which a liquid would show if it contained in a volume of I cubic centimeter, i gram of active substance, and should act on the polarized ray through a column i decimeter in length. As already explained in the introduction, 2, the specific rotation of bodies, in themselves liquids, is expressed by the formula W M=777 in which a = the angle of rotation measured for a ray of definite wave-length, / = the length of the observation tube in decimeters, d = the specific gravity of the liquid, referred to water at 4 as standard. In the measurement of these three quantities, the tem- perature must always be considered, and should be kept the same for all. As normal temperature, 20 C. is usually taken. The specific rotation of an active liquid for a given light and temperature, for example, [], expresses a characteristic con- stant for this substance. For solid active substances which are brought into solution by aid of an inactive solvent, the specific rotation may be derived in two ways, on the assumption that a is proportional to the concentration : i. With determination of the concentration, c, by which is 1 Biot : Mm. de 1'Acad., 13, 116 (1835). Ann. chim. phys., [3], 10, 5. 1 66 SPECIFIC ROTATION understood the number of grams of active substance dissolved to make 100 cc. of solution, we have the formula : 2. With the determination of the percentage strength, p, of the solution, by which is understood the number of grams of active substance in 100 grams of the solution, and further of the specific gravity, d, of the solution. We have then the formula : .. a . ioo in which p d = c, In many cases, especially where we are concerned with single values for the specific rotation, the experimentally simpler determination by formula (II) is sufficient. A solution is made in a flask holding ioo true cc. at 20, and this is polar- ized at the same temperature. But if it is desired to follow changes in the specific rotation corresponding to changes in the composition of the solution, it is necessary to prepare the latter by weighing the active substance and the solvent, as the percentage amounts of both must be known. It then remains to find the specific gravity of the solution at 20 referred to water at 4, determine the angle tf, and calculate according to formula (III). The details of practical methods are given in Part IV. As experience has shown the specific rotation calculated from solutions is seldom a constant number ; with most sub- stances the value changes through several influences in a more or less marked degree, being dependent on: 1. The concentration of the solution. 2. The nature of the solvent. 3. The temperature. A detailed discussion of these relations will be given in the following chapters. I. CONSTANT SPECIFIC ROTATION OF DISSOLVED SUBSTANCES 48. Cane-Sugar was the first substance whose specific rotation was determined by Biot in 1835 and he found that for a con- CONSTANT SPECIFIC ROTATION 167 stant length of the observation tube the angle of rotation was proportional to the amount of sugar in solution; in other words, the same value resulted for [] whatever concentration of the solution was employed. The same was found for mix- tures of turpentine and ether. Biot, assuming from such results, that the amount of rotation is simply proportional to the number of active molecules passed by the light in going through the solution, reached the following law: ' ' When an active substance is dissolved in an inactive sol- vent, which produces no chemical change, the angle of rota- tion observed is proportional to the amount by weight of the active substance in the unit volume of solution, and the specific rotation is therefore a constant quantity." According to our present knowledge, however, perfect con- stancy in the values for [a] are shown by but few substances. Even with cane-sugar later more exact investigations have shown that with decreasing concentration of the aqueous solu- tion the specific rotation increases slightly, it being in fact found that by changing the percentage strength from 65 to 2 per cent, of sugar, the value [a] D increases uniformly from 65.62 to 66.80, that is about 1.8 per cent. (See 52.) On the other hand in certain cases even w ? ith great variations in the concentration no such regular change in the rotation could be found, for example: Milk-Sugar which was investigated in 32 aqueous solutions varying in strength from/ = 2.35 to p = 36 per cent., gave for the specific rotation numbers which varied irregularly between the limits [<*] = 51.94 and 53.18. The mean of all determinations was [<*]"= : + 52.53 for C 12 H 22 O U .H 2 O (Schmoger). 1 Rhamnose shows between the limits c = 3 to 30 the constant specific rotation [>] = = + 8.50 for C 6 H 12 O 5 .H 2 O (Schnelle and Tollens). 2 Parasantonide in Chloroform. Of this very strongly rotating substance 13 solutions, varying in strength from/ = 0.14 to 48 per cent., gave rotations between [or]3 = : + 887.9 and 1 Schmoger: Ber. d. chem. Ges., 13, 1922. - Schnelle and Tollens: Ann. Chem. (I^iebig), 271, 64. 1 68 SPECIFIC ROTATION 896.5. The mean of all observations was [] 2 = -f- 890.9 which differs from the extreme values by only 0.5 per cent. The rotation is constant for other light rays than D. The temperature likewise did not seem to exert any appreciable influence. In solutions of the body in alcohol, however, for the slight change in concentration from^> = 0.26 top = 8.5 a change from [or]3 = = -f 880 to [>]'- -f- 833.9 followed (Nasini). 1 Santonide, dissolved in chloroform, for all concentrations between c = 3 and 30 and for all kinds of light shows a con- stant specific rotation. For example, for yellow, [#]/> + 754 (Nasini). 2 Nicotine, which possesses the specific rotation [a] = -- 164 dissolved in benzene gave the following numbers : Nicotine. [a]fc Nicotine. [a]K p = 84.36 - 164.29 p = 19.00 - 163.95 48.02 164.14 16.36 163.88 25.47 - 164.10 8.52 - 163.67 Mean [a] = - 164.00. There is therefore here but a very small change in the specific rotation, amounting to only 0.2 percent. (Hein). 3 In like manner the specific rotation of nicotine remains almost unchanged when it is dissolved in ether or acetone, while with aniline and toluidine there is a slight decrease, to about [<*]/> = - 156.5 (Hein). The decrease in ethyl or propyl alcohol is somewhat more, while with water it is very marked. (See 52.) Cocaine, dissolved in chloroform, shows for solutions vary- ing from p=io to 25, specific rotations which are always between [>] = - 16.28 and 16.36 (Antrick). 4 With several other substances, for example, d-camphor dis- solved in. almond oil or olive oil (Aignan), 5 and l-a-camphol dissolved in alcohols, acetone, acetic ether or hydrocarbons (Haller) 6 constancy in the specific rotation was found, but the Nasini: Her. d. chem. Ges., 14, 1512. Nasini: Accad. d. Lincei, [3], 13 (1882). Hein: Inaug. Diss.. Berlin (1896). Antrick : Ber. d. chem. Ges., ao, 321. Aignan : See next paragraph. Haller : Compt. rend., 112, 143. VARIABLE SPECIFIC ROTATION 169 experiments covered but small variations in concentration. As is apparent, therefore, the Biot law given above possesses but a limited applicability. II. VARIABLE SPECIFIC ROTATION OF DISSOLVED SUBSTANCES A. Dependence of the Specific Rotation on the Concentration of the Solutions 49. In the investigation of aqueous solutions of tartaric acid, Biot 1 found in 1838, that the specific rotation of this sub- stance was the larger, the more dilute the solution employed. This case was considered a long time as an exception, until in 1852, with the aid of better polarization apparatus, Biot 2 recognized that the phenomenon appears with other substances also. Thus, with increasing dilution of its solutions in alcohol or acetic acid camphor showed a decrease in specific rotation, turpentine on the contrary, by increasing additions of alcohol or olive oil an increase, and finally, even with sugar, a slight increase was observed with an increase in the amount of water. Besides this, the influence of the solvent was brought to light, as in the case of camphor different values for [] were obtained by dissolving it in alcohol or in the same amount of acetic acid. Biot then noted clearly the fact, that in general the values of specific rotations calculated from solutions are more or less variable numbers, and that consequently the molecules of the active substances seem to suffer some altera- tion by the presence of the inactive solvent particles. 3 For a long time, however, this widening of our knowledge of specific rotation remained largely unnoticed. The fact that in solutions of cane-sugar, the angle of rotation is almost exactly proportional to the concentration, and that accordingly from the observed angle of rotation, the amount of sugar can be calculated, led to the construction of the optical saccharim- eters which soon became extensively used, and were also 1 Biot : M6m. de 1'Acad., 15, 93 ; Ann. chim. phys., [3], 10, 385. 2 Biot : Ann. chim. phys., [3], 36, 257. 3 Several French investigators as Aignan (Pouvoir rotat. spec. d. corps act. dissous. These, Paris, 1893, and Freundler, (Ann. chim. phys., [7], 4, 244, (1895)) refer to the constancy of [a] as Biot's law and state the case as if I (I,andolt: Ann. Chem. (L,iebig), 189, (1877)) had first shown its inaccuracy, whereas, this had long since been shown by Biot himself. 170 SPECIFIC ROTATION employed in the investigation of other substances. In spite of the fact that Biot 1 in 1860 had published an extended paper, which contained a resume of all his work in this field, and again called attention to the correct relations, the opinion was still in the main held that active substances in general behave as does cane-sugar ; it was therefore considered sufficient in the optical examination of a substance to determine the rotation for a single solution and then, from this, by aid of formula II or III, to calculate the specific rotation and con- sider this as constant. In this way, a great many specific rotations have been determined, which, generally with no mention of the concentration or solvent employed, have slipped into the physical and chemical handbooks and have been retained for a long time. In 1873 Oudemans" independently found that the specific rotation can assume very different values when different inactive liquids are used as solvents. Hesse 3 published in 1875 a large number of determinations of rotations, of over fifty active substances, and in solutions of different concentrations. Even with only small variations in the latter (between i and 10 grams in 100 cc. ) nearly all bodies exhibit marked differences in specific rotation, and in most cases a decrease in this with increase in concentration. Still greater differences, amounting often to over 50 per cent., were found by varying the, liquids employed as solvents. The numerous investigations which have been carried out in the last twenty years on the rotation of new organic substances have led to the same results. 50. The Determination of the True Specific Rotation of Dissolved Substances, According to Biot. Only such value can be given to the specific rotations calculated from a single solution as attaches to a constant obtained under special conditions. But Biot* showed long ago in his investigations on the rotation of tartaric acid how a definite meaning may be given to these variable numbers and this is explained in the following con- siderations: 1 Biot: Ann. chim. phys., [3], 59, 206. 2 Oudemans: Pogg. Ann., 148, 337 : Ann. Chem. (Liebig), 166, 65. 8 Hesse: Ann. Chem. (Liebig), 176, 89, 189. 4 Biot: Mm. de PAcad., 15, 205 (1838); 16, 254 (1838); Ann. chim. phys. [3], 10, 385 (1844); 38, 215 (1850); 36, 257 (1852); 59, 219 (1860). TRUE SPECIFIC ROTATION 171 For liquid active substances the specific rotation may be directly determined, and for a given temperature this is a con- stant. If now such a body, for example turpentine, be mixed in different proportions with an indifferent liquid, such as alcohol, and then from the percentage amount, the specific gravity and observed rotation, the specific rotations be calcu- lated, values result which differ more or less widely from that found for -the pure substance. The original rotation of the body undergoes then a change by reason of the presence of the inactive molecules, most active substances showing an increase in rotation; a few, however, a decrease in specific ro- tation by increase in the amount of the solvent. 1 If the active body is solid it can be investigated only in solu- tion, and, according to the constitution of the latter, different numbers for the specific rotation are obtained which do not represent the real specific rotation of the pure substance but values modified by the presence of the inactive solvent and differing from the first to an extent which is quite unknown. If a pure homogeneous liquid is employed as solvent so that indifferent molecules of one and the same kind only affect the molecules of the active body, changes in the specific rotation may be followed most readily by graphic representation by taking the percentage amounts of inactive solvent (^) as ab- scissas in a coordinate system, while the corresponding values for [a~\ are taken as ordinates. An increase or decrease in the specific rotation is often shown then as a straight line which inclines in proportion to changes in q and which may be repre- sented by the general formula (I) [>]=A + B?, the constants in which, A and B, may be calculated from the experiments. In other cases, on the contrary, the line obtained is not straight, but is a curve, ordinarily a part of a parabola or hyperbola, in which case the dependence of the 1 This could be seen by aid of a polarization apparatus placed vertically, the obser- vation tube being left open above. I,et turpentine oil, for example, be poured into the tube and say i cm. in height, and the rotation then observed. Now by adding increased amounts of alcohol, and observing, greater and greater rotations will be found. The number of active molecules is the same throughout, but they are distrib- uted through a lengthened column. By employing nicotine and diluting gradually with water a constantly decreasing rotation would be found. 172 SPECIFIC ROTATION specific rotation on q is shown by an expression of the form (II) [a]=.4 or by some other equation with several constants. In these formulas, A represents the specific rotation of the pure substance, and the values for B, formula (I), and for B and C, formula (II) , represent the increase or decrease which A suffers by the presence of i per cent, of inactive solvent. If q = o we have the specific rotation of the pure substance ; if, on the other hand, in equations I and II we put q = 100, there results for [] a value which must be looked upon as the specific rotation of the body in solution of infinite dilution. If we assume that in the case^ 100, the active body has entirely disappeared, and the liquid consists of the inactive solvent only, the rotation then must become o. This, accord- ing to Biot, 2 may also be derived from the above expressions, by putting them equal to the equation [or] ' ' , which represents the specific rotation as calculated directly from the observed rotation a. If in the last equation, in place of p, 1 The three constants, A, Band Cof the formula [a] = A + may be found ac- cording to Biot (Ann. chim phys., [3], n, 96, 69) in the following way, if for three solutions with ft, q<> and q% per. cent, of inactive solvent, the corresponding specific rotations, [a]i [a] 2 and [o] 3 have been determined. If we take then the values for a and c follow from these equations : (Mztfa Mi ft) + (Ma Mi) c (ft ft) = > (Ms ft Mi ft) + (Ms [*[\)c (ft- ft) a = o, and therefore * from each of the following equations : (Mi ) (?i +c) = *, (Mi ) (ft + c) = *, . (Ms ) (ft + c) = b, Finally we have Biot brings the equation also into the form : [a]=A + i ^ q c q in which ff = ~- and C = --. Instead of q in the above formulas we can naturally introduce/ and write M =** +S(ioop), [a] =A + ff (zoo p) + C (100 /)2. Biot : Ann. chim. phys., [3], 10, 399, 59 ; 59, 224, 15. TRUE SPECIFIC ROTATION 173 inasmuch as p + q = 100, we put the value 100 there follows = i.d \A + (B - \g 63. 904 + 0.064686?- 0.00039864*. 1 The calculation of A from equation (II) follows by taking in equation (I) q = o and in (II) p = 100 and then equating the two. On the other hand a follows from equation (I) when in (I) we take q = 100 and in (II) p = o. As P + 9 = loo we have further the equations: A + B (too p) -j- C (100 p)*- = A + 100 B + 10,000 C + bp + cp*, a. + b (100 9) - c (loo q)- = a + 100 + 10,000 c -f Bq + Q 2 , from which the relations between B and b, and Cand c follow. Toilens : Her. d. chem. Ges., 10, 1410 ; 17, 1757. I 76 SPECIFIC ROTATION Landolt 1 found for camphor dissolved in benzene within the limits q = 37 to q = 76 : [ ( *]D = 55-21 0.1630?, from which follows for p = 24 to 63 per cent., [>]/> = 38-91 + 0.1630 p. It is sometimes required to change the constants of the equation [a] = a + bp + #', for a substance with molecular weight J/ so that they will apply for a derivative (hydrate, salt, etc.) with the molecular weight MI . Then the above formula becomes : [a] = a, + b,p + c,p\ and the constants of this are found : 2 i. In the case, M < M l from 2. In the case, M^> M l from For example, the following equation was established by Tollens 3 for anhydrous glucose, C 6 H 12 O 6 , M= 180: M/? 5 2 -5 + o. 018796 p 4- 0.00051683 / 2 . From this there may be derived for the hydrate, C 6 Hj.,O 6 -f H 2 O, M l == 198, the following formula, taking into considera- tion that M = 47-73 + o-oi5534 P + 0.00038830 p\ 52. Experimental Proof of Blot's Formulas. With what degree of certainty the true specific rotation of a substance may be calculated from observations on its solutions can be determined by experiments on active liquid bodies. First, the specific rotation is found directly, and then a number of mixtures with inactive liquids are prepared and from the observed rotations in these the constants in the formula [a] = A -f Bq or [or] = 1 I^andolt: Ann. Chem. (Uebig), 189, 334. - In the introduction, 2, the calculation is carried out only for the case M < M\. 3 Tollens: Ber. d. chem. Ges., 17, 2238. EXPERIMENTAL PROOF OF BIOT S FORMULAS 177 A -(- Bq -f Cq' are derived. The values obtained for A by using different solvents must all agree very closely with the observed value \_at~] , and it remains to see how far this agree- ment is diminished when only solutions of low concentration are employed in the calculations, as is the case with bodies of slight solubility. Experiments of this kind have been made with right and left turpentine, nicotine, and ethyl tartrate (Landolt). 1 In the following observations, which for purpose of illustra- tion are given in full, the angles of rotation were found mostly by aid of the Wild polaristrobometer and as the means of ten single observations. The normal temperature of 20 employed was secured by jacketed tubes; the densities d are reduced to water at 4 as standard. 2 /. Left Turpentine Oil. The French oil with boiling-point 160 to 162 was used and the rotation found in two tubes of different length : d; I in dm a 21 M~ 0.8629 0.9992 -31.91 37.00 2.1979 70.20 37.02 Mean. 37.01. a. Mixtures with Alcohol. The specific gravity of the alcohol used was, = 0.7957: Mixture Turpentine oil. No. p Alcohol. J 20 q U 4 a for / = 2.1979 dm. MS I 90-05 9-95 0.8556 -62. 7 2 - 37.04 II 69.94 30.06 0.8392 - 48.05 37.25 III 49-97 50.03 0.8254 - 34-04 - 37-55 IV 29.97 70.03 0.8127 20.29 - 37-90 V 10.01 89.99 O.SoiI - 6. 7 8 - 38.49 1 Landolt: Ann. Chem. (Liebig), 189, 311 (1877). 2 In the following tables all the numbers, which in the original paper were carried out to four places of decimals for/> and q, to five places for d and to three places for a and [a], have been shortened. In consequence a recalculation of [a] might lead in some instances to values differing by one or two units in the last decimal from those now given. But this is of no consequence for the present purpose. 12 178 SPECIFIC ROTATION b. Mixtures with Benzene The benzene used had a boiling-point of 80.4 and a specific gravity, d = 0.8803 : Mixture No. Turpentine oil. Benzene. P 1 rf? a for / = 2. 1979 dm Ms I 89.92 10.08 0.8634 -63.47 -37.19 II 77-93 22.07 0.8644 55.50 37.49 III 65.06 34.94 0.8656 -46.79 -37.80 IV 51-05 48.95 0.8677 -37-18 -38.18 V 36.90 63.10 0.8705 27.20 -38.52 VI 22.06 77-94 0.8738 -17.21 39-03 VII 9.98 90.02 0.8771 7-59 -39-45 c. Mixtures with Acetic Acid The acetic acid used had a density, ^f sponding to 99.8 to 99.9 per cent, of real acid. 1.0502, corre- Mixture No. Turpentine oil. P Acetic acid. 9 d? a for r _,-i 20 /= 2.i979dm. i L"J^ 1 I 90.16 9.84 0.8757 -64.46 -37.15 II 78.07 21.93 0.8917 -57.23 37-41 III 64.86 35-14 0.9116 49.24 -37.89 IV 50.97 49.03 0-9353 -40.27 -38.43 V 22.96 77-04 0.9918 -19.86 -39.67 VI 9.84 90.16 1.0233 - 8.90 40.22 As the above observations show, the specific rotation of the turpentine increases in all cases with increase in the amount of inactive solvent q, and in the curves shown in the graphic illustration (Fig. 16) that for acetic acid ascends the most rapidly, that for benzene less, and that for alcohol the least. The curvature of these is not great, but they differ too much from a straight line to permit the application of the formula [a] A -f- Bq. If the constant A is determined from two mixtures, values are found which are always smaller than the specific rotation of the pure turpentine oil (37.01), and which depart the more widely from this, the more dilute the solu- tions are that are used in the observations. This is shown, for example, by the following figures : EXPERIMENTAL PROOF OF BIOT'S FORMULAS 179 Fig. 16. From the mixtures There results Deviation from Extrapolation, with alcohol. A= 37 01. Percent. I and II 36.93 0.08 10 II " III 36.79 0.22 30 III " IV 36.66 -0.35 50 IV " V 35.87 - 1.14 70 If, on the other hand, the formula \_a~\ = A -\-Bq-\- Cq 1 ', be used and the constants be calculated from solutions with the smallest, a mean, and the largest value for q, there results for A a number which is very near the specific rotation of the pure turpentine oil, and further, the formula agrees in a very satisfactory manner with the whole determined curve from q = 10 to 90. There follow from the mixtures with a. Alcohol, calculated from solutions I, III, V, M/> = 36-97 4- 0.004816 q -f 0.0001331 f dm W 8 -*9 I 0.4982 " a. Mixtures with Alcohol. Specific gravity of the alcohol d 0.7962 : Mixture No. Ethyl tartrate. P Alcohol. 9 d~ a. for /= 2.199 dm. MS I 77.98 22.02 1.0837 I6.3I5 C 8.78 II 35-74 64.26 0.9089 6.87 9.62 III 22.33 77.67 0.8634 4.17 9-85 The specific rotation increases gradually, therefore, with the amount of alcohol, and the change may be represented by a curve of very slight curvature (Fig. 19) which almost coin- cides with a straight line. For the formula [a] A + Bq there follows : From mixtures I and II A = 8.34 II " III 8.52 I " III 8.36 = -(- O.OI98 O.OIJO O.OI92 and in the mean, /> = 8.41 -f 0.0187 q. The formula with three constants gives A = 8.27. 1 The imperfect purity of the substance employed was without consequence for this investigation. One could just as well take a mixture of the active body with any inactive substance as a basis, and determine how exactly the original specific rotation might be found from observations on solutions. 1 86 SPECIFIC ROTATION Fig. 19. b. Mixtures with Methyl Alcohol, Mixture No. Ethyl tartrate. P Methylalcohol. 9 df a for /= 2.198 dm. E]s I 77.46 22-54 1.0882 17.88 9-65 II 56.65 43-35 1.0007 12.97 10.41 III 39.92 60.08 0.9381 8.98 10.92 IV 26.97 73-03 0.8946 5.87 11.07 V 15.31 84.69 0.8568 3.23 II. 21 The slight change which the specific rotation of the ethyl tartrate suffers by reason of the presence of the methyl alcohol is not quite proportional to the dilution but may be represented by a curve, at first rather strongly and later less strongly inclined. (This does not show in the figure because of the small scale on which it is drawn.) From mixtures I, II, and III we have [>]/> = 8.42 + 0.0625 q 0.0003479 q\ the constant A agreeing fairly well with the rotation of the original ethyl tartrate. If we use for the calculation the EXPERIMENTAL PROOF OF BIOT'S FORMULAS I8 7 dilute solutions III, IV, and V we obtain A = 10.25, which differs rather widely from the true value of 8.31. c. Mixtures with Water These were polarized immediately after making. Mixture No. Ethyl tartrate. Water. Q rfr / in dm. Of Ms I II III 69.69 39-82 13.89 30-3I 60.18 86.11 1.1508 1.0884 1.0292 2.198 2.199 2.198 24.68 19.27 7.92 14.00 20.22 25.20 The very strong increase in the specific rotation of the tar- trate on addition of water is almost proportional to the amount of the latter added. For the constants of the formula, [a] = A -f- Bq we obtain : From mixtures I and II A = 7.69 B = -j- 0.2082 II " III 8.66 0.1920 " " I " III 7.92 0.2007 and in the mean, [] D = 8. 09 -|- 0.2003?. The marked deviation in the constant A from the specific rotation of the original tartrate (8.31) may be a result of the beginning saponification of the latter. After forty-eight hours, the above solutions gave rotations smaller by o. i to 0.2. From the above investigations with turpentine, nicotine, and ethyl tartrate as well as from many other experiments, the following relations have been established : i . The specific rotation of an active body on increasing dilution with an indifferent inactive liquid suffers no sudden change, but a gradual progressive alteration. Whether the latter is in the nature of an increase or decrease depends on the nature of the active body ; thus, oil of turpentine and ethyl tartrate on being mixed with different solvents show always an increase, while nicotine and camphor (for which experiments follow in 53) show a decrease in the specific rotation. But on one and the same active body different solvents act in very different degrees, so that if the results were represented graphically curves i88 SPECIFIC ROTATION would be obtained, which, starting from the origin , of coordinates, representing the rotation of the pure substance, would radiate from each other. The weaker, therefore, a solution of an active substance is, the greater is the deviation of its specific rotation from that which it shows in pure condition. The whole of the changes which have been found here may be shown by calculating from the interpolation formulas, the limits for^ = o (pure substance) and q = 100 (maximum of dilution). With the bodies investigated we obtain the following numbers : Active substance. Solvent. MB Of the pure substance g = o. M For maximum dilution g = 100 Difference. Turpentine oil (left-rotating) alcohol benzene acetic acid 36.97 36.97 3689 38.79 39-79 40.72 + 1.82 4- 2.82 + 3-83 Turpentine oil (right- rotating) alcohol 14.17 15-35 + 1.18 Nicotine (left-rotating) alcohol water 160.83 161.29 138.59 74-13 - 22.24 -87.16 Ethyl tartrate (right-rotating) alcohol wood alcohol water 8.2 7 8.42 8.09 10.19 11.19 28.12 + i.9 2 + 2.77 + 20.03 It appears, therefore, that the specific rotation of an active substance is changed by different solvents to very different degrees. 2. From the specific rotation of a number of solutions it is possible to calculate that of the pure substance. The degree of certainty with which this is true is different for different substances, and is dependent on the following conditions : a. Upon the extent of the changes made in the specific rotation by the inactive solvent. The greater these are, the more unfavorable in general, are the conditions for calculation (as in the case of nicotine for example), b. Upon the manner in which these changes take place by increase in the amount of inactive EXPERIMENTAL PROOF OF BIOT'S FORMULAS 189 solvent, that is, upon whether they may be represented by a straight line, or by one more or less strongly curved, c. Upon the concentrations of the solutions used. The stronger these are the greater is the certainty in the calculations. The above investigations show that in cases where the formula, [a] = A -f Bq, is applicable, the constant ^4 agrees accurately with the true rotation of the pure substance (or within a few tenths of a degree), when the strongest solution contains about 50 per cent, of active substance. If, on the other hand, the use of the formula, [<*] = A + Bq -f- Cq*, is necessary, vari- ations of more than a degree can appear if solutions containing less than 80 per cent, of active substance are taken as the basis of calculation. j. In the calculation of the original specific rotation of a sub- stance, the same value is always obtained independently of which indifferent liquid is employed as a solvent. The numbers found for the active bodies investigated (the constants A} are given below : I. TURPENTINE OIL (LEFT). Directly observed \_ a ]D 37-oi Calculated from the mixtures with alcohol " 3697 0.04 " benzene " 36.97 0.04 11 " " " " acetic acid.. ' 36.89 +0.12 II. TURPENTINE OIL (RIGHT). Directly observed " 14.15 Calculated from the mixtures with alcohol *' 14.17 0.02 III. NICOTINE (LEFT). Directly observed " 161.55 Calculated from the mixtures with alcohol " 160.83 0.72 " " " " " water " 161.29 0.26 IV. ETHYL TARTRATE (RIGHT). Directly observed " 8.31 Calculated from the mixtures with alcohol " 8.27 0.04 " methyl alcohol " 8.42 + o.n " " " " " water " 8.09 0.22 The differences found are so small that they evidently must be results of errors of observation. #. In the comparison of the rotations of different dissolved bodies, only those values may be used which hold for the pure 190 SPECIFIC ROTATION substances, that is, the constants A. If specific rotations which embrace the changes due to the solvents be taken as the basis for comparison, possible relations will appear, the less dis- tinctly, the weaker the solutions from which the numbers were obtained. In certain cases the constant A can have, besides, a different meaning from that of simply expressing the original rotation of the single molecules of the active body. This would be true when the active substance forms definite compounds with the solvent, or when the active molecules unite to form aggre- gations which as such possess rotation. See 63 and 64. 53. Determination of the True Specific Rotation of Solid Active Substances. The method to be employed here is suggested by what has just been given. First of all it is essential to pre- pare solutions of the greatest possible concentration, and as the nature of the inactive solvent is immaterial one must be chosen which best permits the fulfilling of this requirement. With the aid of such a solvent it is necessary to prepare at least three solutions of different concentrations and to determine their rotating power. If the relation between the specific rotation, [a] , and the percentage amount of the solvent, q, be expressed graphically and it is seen that the three points lie in a straight line, that is, that [] changes directly with q, then the constant, A, calculated from the formula [<*] == A -f- Bq will express the specific rotation of the pure substance. But if the middle point lies higher or lower than the others then a larger number of solutions must be investigated in order to establish the curve as completely as possible, for which then a corresponding interpolation formula ([#] == A -f Bq -f- Cf t or analogous one) is to be calculated. It is also possible to obtain by the graphic method, that is by prolonging the curve obtained to the abscissa q = o, a value which approximates more or less closely to the specific rotation of the pure sub- stance. It must be understood that numbers obtained by such extra- polations must be received with caution. To secure greater certainty it must not be neglected to carry out the investiga- tion with several different solvents; if the values so found for OF SOLID ACTIVE SUBSTANCES igi the constant A agree among themselves closely the mean of these will be taken as the sought- for specific rotation of the substance, but if they do not agree the whole calculation must be rejected. In consequence of their conditions of solubility the calcu- lation of the specific rotation of the original substances is made very difficult in many cases. According to experience as referred to above it will be possible to obtain reliable num- bers only in such instances where solutions may be made with at least 50 per cent, of active substance, and where further the curve of rotation obtained does not vary too much from a straight line. With all difficultly soluble bodies there is no prospect of finding the true specific rotation in pure condition. As an illustration of obtaining the true specific rotation of an active solid substance, a series of experiments with ordinary camphor may be given here. A number of solutions in differ- ent liquids were made and from these the following figures secured by observation: Solvent. No .of solu- tion. Camphor. P Solvent. q < a^for /= 2.1979 dm. Ms Acetic acid. I II III 65.2519 39-7I83 I5.88I9 34.7481 60.2817 84.1181 0.98983 I.OII28 1.03389 72.II7 41.652 15.887 50.801 47.181 44.021 Acetic ether. I II III 53.7260 34.5489 14.9221 46.2740 6545II 85.0779 0.93269 0.91987 0.90686 58.492 36.520 15.290 53-109 52.283 51.408 Mono- chloracetic ether. I II III 54.2184 3I.399 14.2332 45.7816 68.6010 85.7668 1.04206 1.08670 I.I2243 65.356 38.340 17.543 52.631 51-123 49.961 Benzene. I II III 63.1250 49- 6 359 24.3169 36.8750 50.3641 75.6831 0.93067 0.91920 0.89910 63.575 47.097 20.638 49.236 46.966 , 42.948 Dimethyl - aniline. II III 57.1519 36.0428 15.1028 42.8481 63.9572 84.8972 0.95997 0.95914 0.95813 59-533 35.151 13.708 49-370 46.263 1 43.101 Methyl alcohol. I II III 49-3866 3 .3I54 11.2590 50.6134 69.6846 88.7410 0.88093 0.85318 0.82700 46.840 ' 26.820 9.382 48.996 47-179 45.844 Alcohol. I II III IV V 54.7281 49.8142 30.1620 15.0920 9.6883 45.27I9 50.1858 69.8380 84.9080 90.3117 0.8S02I 0.87194 0.84031 0.81752 o.Sf>943 50.634 44.806 25.013 11.840 7.378 47-823 46.934 44.901 43-66i 42.806 1 9 2 SPECIFIC ROTATION Fig. 20. The specific rotations with all these solutions decreases as the amount of solvent increases, but in very different degrees with different inactive solvents. The graphic illustration (Fig. 20) shows that these changes may be represented almost exactly by straight lines when as solvents acetic acid, acetic OF SOLID SUBSTANCES 193 ether, monochloracetic ether, benzene and dimethylaniline are used; the formula, [ar] = A -f- Bq, is then applicable in these cases. But with ethyl alcohol and methyl alcohol the devia- tion from the straight line is too large, and the formula [or] = A -f- Bq -f Cf, was used for the calculations. The following table contains : (i), the values obtained for the constants A and B from the different solutions ; (2), the mean interpolation formulas calculated from these ; ( 3 ) , the calculated specific rotations of the solutions employed, obtained by these formulas, and the differences between these and the observation values given in the preceding table. i 3 Solvent. [*].= A-Bg Means. Solu- tion []/> calcu- lated Difference from obser- vations. Calculation! from A solutions B Acetic acid land II 55.73 II and III 55.17 I and III 55.58 0.1418 0.1326 0.1373 WD =55-49 0.13723? I II III 50.72 47.22 43-95 -0.39 + 0.04 0.07 Acetic ether land II 55.11 II and III 55.21 I and III 55.14 0.04307 0.04458 0.04384 !>]/?= 55-15 0.04383? I II III 53-12 52.28 5I-4I -f o.oi o.oo o.oo Mono- chlor- acetic ether land II II and III I and III 55-65 55-77 55-69 0.06608 0.06769 0.06677 [>]/> = 55.70 0.06685 q I II III 52.64 51.12 49-97 -f O.OI 0.00 -f O.OI Benzene land II II and III I and III 55' 45 54.96 55-21 0.1683 0.1587 0.1620 []/> = 55-21 0.1630?- I II III 49.19 47-00 42.87 0.05 -r 0.03 0.08 Di- methyl- aniline land II 55.68 II and III 55.92 I and III 55 76 0.1472 0.1510 0.1491 []/> = 55.78 0.1491? I II III 49-40 46.25 43-13 + 0.03 O.OI + 0.03 Methyl alcohol ill [a]/? =56.15 0.1749? 0.0006617 q* I Alcohol III V O]/> = 54.38 0.1614 q -f 0.0003690 q 1 l II IV 47-21 43-33 -f 0.28 -0.33 The formula [a] 13 A + Bq 42.8799 -^ gives W D = 54.83 - 23 V 82 ^. 194 SPECIFIC ROTATION If we compare with each other the values obtained for the constant A, from the different solutions an agreement is found which, in consideration of the great extrapolations necessary (from q = o, amounting to 25 to 50 per cent.), must be con- sidered as a very close one ; we can look upon the mean of these numbers, therefore, as representing the true specific rotation of the pure camphor. The values for B, depending on the solvent, vary on the contrary, very greatly. If we calculate from the formulas the specific rotations for the two limits, g = o and q = 100, the following results are obtained, from which may be seen to what extent the different solvents influence the rotating power of camphor : Solvent. [a] D for q = o = A D Pure substance. [tf] D for q = loo Infinite dilution. Total change. ;; ; 41 8 11 7 oDO re 2 co 8 1 0'/ 4 A Monochloracetic ether OO"* 55-7 ec 2 49.0 78 Q 6-7 16 T. OO"* re 8 o-V i = 55-4, w r ith a mean error of 0.4. 54. Slight Changes in Specific Rotation by Variations in Concentra- tion. Such are observed, as will be shown later in the chapter in which changes in the rotation are more fully discussed, in cases in which an action of the solvent on the active body is as far as possible excluded, and where, therefore, molecular aggregation, dissociation or hydrolysis, etc., can not take place. A body of this kind is cane-sugar for which Tollens 1 and also Schmitz 2 have shown by accurate investigations that the specific rotation undergoes a slight but regular increase with decrease in the concentration. The observations of 1 Tollens: Ber. d. chem. Ges., 10, 1403 (1877). * Schmitz: Ibid., 10, 1414. SLIGHT VARIATIONS 195 Schmitz which were carried out by methods described in Part IV, using a Wild and also a half shadow polarimeter, are given here as an illustration of careful experiments : Grams i Grams a 30 sugar in sugar in Specific U D W" Difference No. 100 grams solution. 100 cc. gravity, solution. for P ,=* Biot- Mem. de 1'Acad., 13, 131 (1835). * Tollens: Ber. d. chem. Ges., 10, 1413 (1877). * Arndtsen : Ann. chim. phys., [3], 54, 403 ; Pogg. Ann., 105, 312. 4 Pribram : Ber. d. chem. Ges., 20, 1846. * Bremer : Rec. Trav. chim. Pays-Bas., 6, 258. MINIMUM VALUE OF SPECIFIC ROTATION 197 centration is diminished from 69 per cent., and reaches a maximum at p = 1 8. 86 per cent. (66.528) ; there is then a de- crease, so that for/ = i.i, [<*] = 66.402. Direct experiments with dilute solutions 1 with p = 10 to i per cent, of sugar gave ir- regular values between [<*] = 66.499 and 66.276, the varia- tions in which from those calculated by the above formula lie within the errors of observations, but show in general a slight decrease. Pribram 2 in dropping from p = 3.659 to 0.222 in five steps found a continuous decrease in the specific rotation from [**] = 66.531 to 66.213. On the other hand, Nasini and Villavecchia 3 observed for five concentrations lying between p= 1.253 an d 0.824, a regular increase from [<*] = 67. 37 to 68.24. The results are therefore contradictory ; but for practical use, the Tollens formula may be looked upon as satisfactory for all concentrations. Dextrose. According to Tollens* the formulas, Anhydrous dextrose M /> = 5 2 -5 + o.oi8796/> -f o.ooo5i68/ 2 Dextrose hydrate .' [<*] = 47-73 + 0.015534^ + 0.0003883^, derived from concentrated solutions, satisfy the results of observation down to a decreased concentration of p = i. It has not been certainly shown, therefore, with any of the above substances that the character of the curve expressing the dependence of the specific rotation on the concentration undergoes a change for very great dilutions. A case, on the other hand, where the behavior is essentially different will be discussed in the following paragraph. (See Nicotine.) 56. A Minimum Value of Specific Rotation. This peculiar phe- nomenon has been recognized in the following cases : Nicotine, dissolved in water. As already shown in 52, the specific rotation decreases in a very marked degree as the dilution increases, and this was followed to ?= 91 per cent., at which point the original rotation of the nicotine, [] # = - 161.55 na d sunk to 75-53- Pribram 5 first noticed that when q is increased from 96 to 99 per cent., an increase in 1 Tollens : Ber. d. chem. Ges., 17, 1751. * Pribram : Ibid., ao, 1848. 3 Nasini and Villavecchia: Wied. Beib., 16, 366. 4 Tollens : Ber. d. chem. Ges., 17, 2238. 6 Pribram : Ibid., ao, 1848. 1 9 8 SPECIFIC ROTATION the rotation from [ar] = 77.03 to 79.32* follows. This behavior has been more fully investigated by Hein, 1 who determined the point of minimal rotation ; he employed a sample of nicotine with [] = 164.00 and examined eight dilute solutions at temperatures of 5, 15 and 20. These tests were combined with a series of cryoscopic molecular weight determinations of nicotine (C 10 H U N 2 =162) in the same or nearly same concentrations. The values obtained are the following : In JOG parts by weight. M5> Ms MS Molecular weight determination. Nicotine. P Water. q Nicotine in roo parts of solution. Molec- ular weight found. Varia- tion front 162. 15.592 84.408 - 73.39 -76.18 - 77.59 I3-736 275 113 II. 206 88.794 73-05 75-96 77.01 11.512 26l 99 10.258 89.742 72.78* 75-59* 76.89 8.307 235 73 8.307 91.693 73-07 75.76 76.84* 5.700 209 47 5.700 94.300 73.81 76.00 76.96 3.016 180 18 3.016 96.984 74.46 76.27 77.25 2.042 168 6 2.042 97.958 74-74 76.35 77.32 1.225 165 3 1.061 98.939 74-79 76.83 77.66 0.346 163 i It appears, therefore, that at the temperatures of 5 and 15 a minimum specific rotation (marked *) occurs when the amount of nicotine is about 10 per cent. , and at a temperature of 20 this minimum is found in the solution of about 8 per cent, strength. From the molecular weight determinations it is seen that nicotine in very dilute solutions is present as a normal molecule while in the more concentrated, hydrates and, likely, molecular aggregations exist. To these a much lower rotating power must be ascribed than to the pure nicotine, but if these are broken up by increasing dilution, as suggested by the decrease in the molecular weight, more and more fresh nicotine appears by which the decrease in the specific rotation is gradually arrested and finally an increase in the rotation 1 Hein : "Ueber das specifische Drehungsvermogen und das Moleculargewicht des Nicotins in l,6sungen." Inaug. Diss. Berlin, 1896. (Investigations carried out in the author's laboratory.) MINIMUM VALUE OF SPECIFIC ROTATION 199 must follow. In this manner it may be possible to explain the occurrence of the minimum in the case in hand. 77.8 77.6 Per cent, of water. 85 51 90 100 Per cent, of acid. Minimum at 63.5 per cent. acid. Fig. 21. In Fig. 21, above, the rotation for 20 is graphically shown. Camphor, dissolved in isovaleric or caproic acid. The follow- ing solutions have been investigated by H. Vogel 1 and a mini- mum rotation (*) found for large concentrations : 1 H. Vogel: "Ueber das optische DrehungsvermogendesCamphers." Inaug. Diss. Berlin. (Investigations carried out in the author's laboratory.) 200 SPECIFIC ROTATION Isovaleric acid. Caproic acid. Camphor. P Val. acid. Q Ms Camphor. P Capr. acid. 1 Ms 52.37 47- 6 3 + 53.43 49.84 50.16 + 53-67 46.71 53.29 53.29 47-88 52.12 53-63 43-03 56.97 53.16 43.30 56.70 53-46 38.55 61-45 53-10* 36-48 63-52 53.22* 36.71 63.29 53.2 26.19 73.81 53.42 32.22 66.78 53.28 18.49 81.51 53-70 27.06 72.94 53-40 8-53 91.47 56.51 18.28 8t.72 54-51 2.31 97.69 67.19 12.63 87.37 56.02 9.46 90.54 58.17 4.36 95.64 68.41 3.05 96.95 76.44 As seen from the above numbers, and also from the graphic illustration in Fig. 21, the specific rotation of the camphor decreases at first in very slight amount, from the point of greatest concentration on, with an increase in the percentage of acid, but reaches a minimum when the amount of iso valeric acid is 61.45 P er cent, or the amount of caproic acid is 63.52 per cent. Then a slow increase in the rotation begins, and only after considerable dilution does a sudden and marked change appear. Corresponding to the minimum point in the rotation there is no simple molecular proportion between the amounts of acid and camphor. The dependence of the specific rotation on the amount of acid can be represented, according to Vogel, by the following formulas : Isovaleric acid [or] ^ = 57.15 0.12572 q -\- o.ooiooo? 2 Caproic acid [**];? 58.90 o. 16846^ -\- o.ooi 279 q 1 Solutions of camphor in other fatty acids do not exhibit the appearance of a minimum or a great increase in the specific rotation with strong dilution, but [or] decreases regularly with increased addition of acid, as was true with the solvents described in 53. Vogel gives the following observations: REVERSAL IN THE DIRECTION OF ROTATION 201 Formic acid. Acetic acid. Propionic acid. N. Butyric acid. q MS E* Ms q Ms 35-89 + 39-93 c 47.25 + 49-37 60.92 + 50.53 49.10 + 52.49 49.11 35.o8 61.16 47.20 69.40 49.70 63.07 51.35 57.68 32.38 65-65 46.60 76.93 48.98 68.39 50.94 65.62 29.85 70.40 45-93 86.30 48.34 76.76 50.07 75.95 26.91 80.78 44-24 88.34 49.67 79.56 26.03 83-58 43-87 .... 93-44 49-56 9i-5o 43.36 .... .... .... .... The effect of these fatty acids on the original rotation of the camphor (+ 55.4) is decreased, as easily seen, with increased molecular weight of the acids. It is not clear upon what the peculiar behavior of the valeric and caproic acids depends. 57. Reversal in the Direction of Rotation by Change in Concentra- tion. In the case of bodies whose specific rotations decrease with increase in the amount of solvent, it may happen that the rotation will sink to zero and then, with increased dilution, rise in the opposite direction. This behavior was first noticed by Schneider 1 in aqueous solutions of ordinary malic acid and some of its salts, for which the following results were obtained: Malic acid. 'Sodium acid malate. Sodium malate. Barium malate. 9 MS q Ms q MS q Ms 29.88 + 3-34 39-45 -f 0.15 34-47 -f 4.72 90.62 + 1.81 40.01 -7-231 50.46 1.71 44-74 + 2.15 91.50 + 1.61 50.13 + 1.38 60.28 -3.27 51.21 + 0.50 95-01 4- 0.54 53-53 -f i. oo 69-98 -4.26 53-16 0.16 98.04 O.II 62.47 + 0.17 79.81 -5-57 57.78 - 1.26 63.34 + 0.09 80.05 - 5-64 66.09 -3.43 64-74 0.04 .... 70.01 -4-34 70.31 0.34 .... .... 74-73 -5.28 .... 70.94 0.63 85.34 -6.98 .... 83.35 -1.58 .... 94.73 -8.39 Qi.68 - I'lo 1 G. H. Schneider: Ann. Chem. (I^iebig), 307, 257. 202 SPECIFIC ROTATION The relation of the specific rotation to the amount of water, g, can be expressed by the formulas, Malic acid [a] = 5.89 0.0896 q Sodium acid malate [**]/> 9-37 0.2791 q + 0.001152 q~ Sodium malate [or] " = 15. 20 0.3322 q -f 0.000814 q' 1 , from which by taking A -f Bq -f Cq* =o, the values of q are obtained at which the solution becomes inactive. Malic acid. q = 65.76 Sodium acid malate. Sodium malate. 40.25 52.57 A change in the direction of rotation does not occur with the other alkali malates ; they all exhibit increasing left rotation with changes in concentration from greatest strength to extreme dilution. The causes on which these changes possibly depend are explained in 63. The same phenomenon has been observed with /-sodium lactate dissolved in alcohol (Purdie and Walker), 1 and with aqueous solutions of the barium salts of ^-methoxysuccinic acid (Purdie and Marshall) 2 and ^-ethoxysuccinic acid (Purdie and Walker). 3 This may be seen from the following table in which c is the number of grams of active substance dissolved in 100 cc. of solution. At the foot of the table are given the concentrations corresponding to the inactive points : /-Sodium lactate (in alcohol). rf-Barium methoxysuccinate (in water). d-Barium ethoxysuccinate (in water). c MS c MS c M* 9 23.21 2.28 C 26.13 - 14.27 21.48 -4.37 19.79 - 2.22 12.42 - 7-36 10.77 + 2.46 11.20 - 0.80 5-75 2.21 4.56 + 6.37 9.29 - 0.48 MS + 3.16 . 7-47 -f 1-34 . . . 5-60 + 2.50 . . . . 2.24 + 8-93 . . . . 1. 12 f 10.36 . . . . 0.56 + 20.53 8.81 3.85 o 14.63 1 Purdie and Walker: J. Chem. Soc., 67, 631. In the original, the values of [a] are given. 2 Purdie and Marshall : J. Chem. Soc., 63, 227. Purdie and Walker : Ibid., 63, 235. INCREASE OR DECREASE IN SPECIFIC ROTATION. 203 A change of sign takes place with the following substances, investigated by Freundler, 1 when they are brought from the pure condition (Y = 100) into solution: d-Propyl dicaproyltartrate. rf-Ethyl diacetyltartrate in chloroform. In bromoform. In benzene. c M. c [}. c M, 100 -f 5-o 100 + 2.2 100 + 2.2 50 4 30 - 5-3 - 5-9 - 6.5 23.16 5.88 2.39 -5.2 8.0 19-59 10.83 5-45 2.O -3-6 -4-3 20 - 7.1 1.54 -5-4 IO 5 - 7-5 8.8 .... .... ... 2.5 10.0 ... 58. Increase or Decrease in Specific Rotation with Increasing Dilu- tion of Solutions. As shown by the examples thus far given with some substances there is an increase, and with others a decrease in the specific rotation as the amount of solvent is increased. In order to show at a glance the relations found, a number of observations are given in diagram form, the increase or decrease of rotation being indicated by the direction of arrows. The solvents used are given in parentheses: 2 A. Increase in Rotation. i. Active Acids and Their Salts. Methyl alcohol 1 O u '/ 1 II I 4- Q T. 1 ^ 4 1 A*' 1 + IO 4. V'O 4-72 4- ; ; jjthvl alcohol + Q 6 4- 6 a 4-^6 y.w -4-86 "o 4- o- u 4- 2 d -1- 8 ; I j-j 4- -z 8 _|_ i -i I -o 4-64 o -j 4-47 O' 1 + 1 7 I u -5 4- a 8 4- 06 I Q 1 O' _l_ 7 A 4- 06 "7 2 I Metliylene bromide \ O-4 _|_ T 7 4- 2'd * 1 + T 2 I A J .4 + 1 2 1.4 O I 4-3 A O 2 6 - a 8 71 0' 7- 1 It is apparent that the specific rotations of the original sub- stances are sometimes increased, and sometimes diminished by 1 Freundler: Compt. rend., 117, 556; Ann. chim. phys., [7], 4, 244. DEPENDENCE ON TEMPERATURE 2Oy the inactive liquids, and further, that the order in which the solvents stand with reference to their action is almost the same for the three substances. Even bodies which exhibit but slight changes in specific rotation, such as cane-sugar, show quite appreciable deviations when dissolved in different liquids. Tollens 1 investigated solu- tions of the following composition : 10 parts sugar 90 parts water [or] D = + 66.67 C ethyl alcohol " = -p 66.83 10 " " -f 23 " " -f 67 parts { acetone " =-- + 67.40 I methyl alcohol " = 68.63 It may also happen that a change in the direction of rotation can take place by application of different solvents. Illustra- tions of this have been shown by tables already given, and others are furnished by substances described below. d- Tartaric Acid, which rotates to the right when dissolved in water, exhibits left-rotation when dissolved in a mixture of acetone and ether (Landolt). 2 Pribram 3 found very marked variations by use of the following liquids in which the amount of substance dissolved was always 5 grams to make 100 cc. Solvent: [>]" Water -f 14.40 Alcohol * -f- 3.79 Equal vols. of alcohol and mononitrobenzene -+- 3.17 " " " " " nitroethane + 3-9 " " " " " mononitrotoluene - 0.69 " " " " ethylbromide - 3.62 " " " " benzene - 4." " " " " " toluene - 6.19 " " " " xylene - 6.52 " " " " " cymene - 7.91 " " " " " monochlorbenzene - 8.09 As far as explanations of this behavior are possible they will be given in the chapter on the causes of changes in specific rotation, 61 to 65. C. Dependence of the Specific Rotation on the Temperature. 60. Increase of temperature affects different active bodies in 1 Tollens : Ber. d. chem. Ges., 13, 2303. 2 I^andolt: Ibid., 13, 2332. 3 Pribram: Ibid., 22, 6. 208 SPECIFIC ROTATION different ways; with some there is an increase, with others a decrease in the rotation and in different degrees. Among active crystals, as remarked in 44, only quartz and sodium chlorate have been investigated in this direction. With both, an increase in the rotation follows on warming. The change here is, therefore, the reverse of that in the ordi- nary refraction of light which is diminished by increase of temperature. Liquid Active Bodies. If such a substance is contained in an observation tube, which, to accommodate expansion, is fur- nished with a lateral opening, then on application of heat the density must decrease and consequently the number of mole- cules in the active column, causing a diminution of the angle of rotation. But, on the other hand, the length of the tube has increased which exerts an action of the opposite kind. In calculating the specific rotation, [or] = - -> 'these influences are eliminated if density and length of tube are found for the same temperature at which the rotation is determined, and in case these alone come into question the values for [or] should remain constant. But according to experience this is not quite true in the case of any known body, and it follows that heat must exert some special effect on optical activity. On the effect of temperature on bodies which are in them- selves liquids we have but few observations. Increase in specific rotation with increase in temperature has been found, for example, in the following bodies: a. Nicotine, left-rotating. According to investigations of Landolt 1 which are here given in full to illustrate the changes in the different factors observed, there were found: Temp. I d 4 I Of/) [ a ]z> 10.2 1.01837 99-9I4* 163.776 - 160.96 20.0 I.OIIOI 99-923 163.204 l6l.55 30.0 1.00373 99-932 2 - 162.450 - 161.96 It is seen that while the angle of rotation has become 1 I^andolt: Ann. Chem. (I, according to Frankland and MacGregor, 1 and for ordinary temperatures the following amounts for i : . ( methyl . . . 0.016 Glycerate of { * isopropyl . . . 0.065 meth 1 oo- Diacetyl- 1 isobutyl ..... 0.054 Diacetyl- j * y ' ^ 3 glycerate of j N-heptyl ..... 0.067 glycerate of \ N ; Q L N-octyl ...... 0.043 c. For the right-rotating esters of tartaric add, Pictet 2 found : Temperature. 20 100 Methyl tartrate ..... \oi\ D = -f 2.14 [a] D + 6.00 Ethyl tartrate ....... " = -f 7.66 " =+13.29 N-propyl tartrate .... " - 12.44 " = + I 7- 11 I-propyl tartrate ..... " =4-14.89 " =4-18.82 If the increase in [a] was proportional to the change in temperature it amounted to about 0.05 to 0.07 for i. d. An increase in the angle of rotation has been found in the following bodies: ' Right-rotating isobutyl isoamyloxide (Le Bel 3 and Colson), 4 di-isoamyl oxide .................. (Colson),* " methyl isoamyloxide .......... ---- ( Colson), 4 amyl acetate ...................... ( Colson) , 5 Left-rotating methyl lactate ..................... (Le Bel). 3 With reference to the question whether the change in rota- tion could follow from polymerization with lower temperature, LeBel 6 observes that according to the investigations of Ramsay ( i ) ethyl tartrate possesses the simple molecular weight for all temperatures, (2) the same is true of isobutylamyl oxide between 23 and -f- 125, (3) that, on the other hand, propyl 1 Frankland and MacGregor : J. Chem. Soc., 65, 760 (1894). 2 Pictet: Arch, de Geneve, [3], 7, 82 (1882). 3 I^eBel: Compt. rend., 118, 916 (1894). 4 Colson: Ibid., 116, 319 (1893). 5 Colson: Ibid., 119, 65, 1894. From the papers of LeBel and Colson, it is not clear whether the data refer to the observed angle of rotation or to the specific rotation. 6 I^eBel: Compt. rend., 118, 916; 119, 226 (1894). 14 2IO SPECIFIC ROTATION glycol has the double molecular weight above 100 and the quadruple weight at the ordinary temperature, without show- ing any change in activity corresponding to this polymerization. This supposed cause appears therefore to be insufficient. A decrease in specific rotation with elevation of temperature was observed by Gernez 1 in several essential oils. It was found that the decrease could be expressed by the following formulas, holding from o to 150: Right turpentine oil [or] D = 36.61 o. 0x54437 / Orange oil [/*]/> = II 5-3 I 0-1237 / o.o 4 i6 P Bitter orange oil [] D 118.55 0.1175 / o.o. 2 2i6 & The decrease goes still further when the temperature of the boiling-point is passed and the body becomes a vapor (see 9). On the other hand, as Gernez found, the dispersion suffers no marked change by heat. Dissolved Active Bodies. As experiment has shown, with these also there may be a change, not only in the observed angle of rotation, but in the specific rotation, which may undergo either an increase or decrease by change in tempera- ture. In the following table in which bodies already men- tioned are included (and marked with an *) all the known relations are given: CHANGE IN THE SPECIFIC ROTATION WITH INCREASED TEMPERATURE. * Increase in left rotation +~m *Nicotine *Esters of glyceric acid *Esters of diacetylglyceric acid Malic acid in dilute aqueous solution Decrease in right rotation *Right turpentine oil *Orange oil *Bitter orange oil Malic acid in strong solution Cane-sugar in water Milk-sugar " " Maltose " " Galactose " " Atabinose " " Rhamnose " " Cinchonicine in alcohol Quinidine " " Quinidine sulphate in water Tartar emetic " " 1 Gernez: Ann. de 1'Ecole normal, i, i (1864). 1 The extent of the changes in the rotation of the different substances, as far as observations reach, will be given in the chapter on "Constants of Rotation." DEPENDENCE ON TEMPERATURE 211 - o - Decrease in left rotation - Turpentine oil Fructose (levulose) in water Invert sugar Saccharin Mandelic acid Sodium santoninate " " Quinine in alcohol Quinine sulphate and disulphate in alcohol Cinchonidine in alcohol Thebaine " " Glutin " water -+ Increase in right rotation *Isobutylamyl oxide *Methylisoamyl oxide *Diisoamyl oxide *Amyl acetate *Methyl lactate *Tartaric acid esters Tartaric acid in water Alkali tartrates in water Glucuronic acid " " Xylose " " N-propyl dibutyryl tartrate in monobromethylidene bromide A change in the direction of rotation by elevation of tem- perature, when a point of inactivity is passed is shown in the following table : Aspartic acid in water. Malic acid " " Tartaric acid " " , Invert sugar " " On the last four substances we have the following more complete data : Aspartic Acid. Aqueous solutions which are perfectly free from other acids as well as from alkalies, exhibit right rotation at the ordinary temperature. But, as found by Ellen Cook, 1 this decreases with elevation of temperature, and passes finally into increasing left rotation, the point of inactivity being passed at 75. The following specific rotations for white light, \_a~\j (converted into \_a~\ D by multiplication with 0.89), were found by use of supersaturated solutions. The values of the concentration, c, were found by determination of the density at the temperatures at which the rotations were observed. 1 Ellen P. Cook : Ber. d. chem. Ges., 30, 294. 212 SPECIFIC ROTATION Specific rotation. Solution. No. Amount. Temp. Specific gravity. Concentr. Observed. P t d( c */ M/ E3* I 0.528 20 I.OOI85 0.53 1 4- 0.103 + 4.90 4- 4.36 II 1.872 32 1.0043 1.880 + 0.320 4- 4-25 -f3-78 II 1.872 40 I.OOI5 1.875 4- 0-255 + 3-40 4- 3.4 II 1.872 50 I.OOO4 1.873 4- 0.130 4- i.74 -f 1-55 II 1.872 60 0.9917 1.857 4- 0.102 4- 1-37 4- 1.22 II 1.872 75 0.9821 1.838 o O II 1.872 77 0.9800 1.835 0.050 -0.68 0.61 II 1.872 80 0.9777 1.830 0.062 -0.85 0.76 II 1.872 90 0.9747 1.825 -0.155 - 2.12 - 1.86 Malic Acid (common}. Concentrated solutions show right rotation which decreases with elevation of temperature; dilute solutions are levorotatory and more strongly, the higher the temperature. In solutions of a certain strength right rotation appears at a low temperature, and left rotation at a high tem- perature and the point of inactivity changes with alterations in the percentage amount, p, of acid. This behavior, which is shown by neutral sodium malate also, can be seen in the following observations of Th. Thomsen: 1 Malic acid . . . p t = 10 t = 20 t = 30 [53-75 !>]/> = + 2.52 []/> = 4- 1-73 [ =4- 0.94 I 0.44 4- 1.31 + 0.54 -0.12 I 28.67 4- 0.33 0.35 0.83 [21.65 0.44 0.90 -1.43 Sodium malate . 42.75 4-0.38 -0.89 2.04 Tartaric Acid. In examining tartaric acid melted with a little water in a glass vessel with parallel walls, Biot 2 observed at first right rotation which, with falling temperature, decreased and on solidification passed into left rotation. The right rotation characteristic of aqueous solutions of tartaric acid increases with heat in a marked degree, as appears from the following observations of Krecke: 3 1 Th. Thomsen: Ber. d. chem. Ges., 15, 441. 2 Biot: Ann. chim. phys., [3], 59, 206, 11 (1860). * Krecke: Arch. Nerland, 7, 97 (187*). DEPENDENCE ON TEMPERATURE 213 Amount of tartaric acid in solution. Temp- 40 Per cent. 20 Per cent. 10 Per cent. |>]0 =+ 5-53 [>]* = + 8.66 MD = -f 9-95 10 7-49 9.96 10.94 20 8.32 H-57 I2.2 5 30 9.62 12.49 13-93 40 11.03 13-65 15-68 5 12.27 15.01 17.11 60 12.63 16.18 18.31 70 13-38 17.16 19.42 80 14.27 18.40 20.72 90 I5-9I 19.99 22.22 ICO 17.66 21.48 23-79 Of the salts of tartaric acid, according to Krecke, disodium and sodium potassium tartrate show a slight increase, but potassium antimonyl tartrate a decrease in rotation by elevation of temperature. Invert Sugar. As Tuchschmid 1 found, an aqueous solution with 17.21 grams in 100 cc. shows a diminution of its left rotation with elevation of temperature, according to the formula : []# = 27.9 + 0.32 /. Therefore, the rotation must become zero at 87.2, to pass into right rotation at a still higher temperature. In agreement with this, v. Lippmann 2 found the point of inactivity to be at 87.8, and Casamajor 3 at 88. If alcohol is added to the invert sugar, which causes a decrease in the left rotation, the change in direction follows on moderately warming. If, for example, a solution of 19 grams of cane-sugar in 1 5 cc. of water and 5 cc. of glacial acetic acid is inverted by heating in the water-bath, and diluted afterwards to 100 cc. with absolute alcohol, the liquid which now contains 20 grams of invert sugar, shows the following angles of rotation : / 20 3 4 5 60 a/j for 2 dm. 1.9 0.9 -j-o.2 +1-3 2.2 The point of inactivity is therefore found to be about 38 (Landolt). 4 The change in the direction of rotation may be explained 1 Tuchschmid : J. prakt. Chem., [2], a, 235 (1870). v. I,ippmann : Ber. d. chem. Ges., 13, 1822 (1880). 3 Casamajor : Wied. Beib., (1879), 804. 4 I^andolt : Ber. d. chem. Ges., 13, 2335 (1880). 214 SPECIFIC ROTATION when the active substance consists of two oppositely rotating components which are affected to different extents by heat. This is the case with invert sugar. As first shown by Dubrunfaut, 1 and later more particularly by Honigand Jesser, 2 the rotating power of levulose decreases rapidly on warming ( a D for i C. about 0.67 ), while that of dextrose is but slightly altered ; the direction of rotation of the latter becomes then gradually apparent. As another case, Aignan 3 has shown that a mixture of left turpentine and right camphor in benzene may change its direction of rotation with elevation of tem- perature, and with different kinds of light at different degrees. The numbers given in the following table are the observed angles of rotation in a 2 dm. tube : Temperature. Red light. Yellow light. Green light. -r-I3 33 to 38 5o " 51 61 " 62 65 " 72 81 " 90 2.62 - 1-53 -0.83 0-35 + 0.18 + 0.57 -0.72 0.40 + 1-50 + I-98 r 2.67 + 3- + 2 .40 + 4-08 -f 5-10 i- 5-55 f 5-90 + 6.72 Finally, it is of interest to determine, whether, with the same substance, elevation of temperature and increasing dilution exert like changes in the specific rotation. Thus far, but few bodies have been studied in both directions, and with several of them, the alterations observed have been very slight. What has been found is tabulated below. a. Corresponding Changes in the specific rotation through increasing dilution or elevation of temperature occur in : 3 + " "*"""" 1 Dubrunfaut : Compt. rend., 43, 901 (1856). Honig and Jesser : Zeit. Ver. f. Riibenzucker-Ind., (1888), p. 1039. * Aignan : Compt rend., 116,725, (1893). ELECTROLYTIC DISSOCIATION IN AQUEOUS SOLUTIONS 215 b. Opposite Effects are observed in the following cases: Increasing temperature. - + Increasing dilution. -0 + Quinine in alcohol .. m ~ Ouinidine * * Nicotine in water or alcohol Xvlose " ^/-Turpentine jjj alcohol /- '* benzene, acetic acid It is evident, therefore, that increased temperature and dilution have sometimes the same and sometimes the opposite actions and that no definite regularity is apparent with respect to the nature of the substances examined. Those cases in which an explanation of this behavior is in any degree possible will be discussed in the following chapter. D. Causes of the Changes in Specific Rotation. The many variations which are exhibited in the specific rotation of dissolved substances depend, according to the nature of the substances, on essentially different causes, and all changes which are in general possible with solutions, as electrolytic dissociation, formation or breaking up of molecular aggregations, hydrolysis and other less clearly defined effects may come into play. Besides this the explanation of the phenomenon is made difficult by the fact that of the nature of concentrated solutions almost nothing is known. Up to the present time the following data have been accumulated with reference to each one of these influences. 61. a. Electrolytic Dissociation in Aqueous Solutions. In 1873 in the examination of a number of neutral salts of tartaric acid, it was remarked that they agreed very closely with each other in molecular rotation, from which it was clear that differences in the metals combined had but little effect, 216 SPECIFIC ROTATION and this not in any relation with the atomic weight (Landolt). 1 Oudemans* appeared to find in solutions of cinchona alkaloids in different dilute acids which were added in increasing molec- ular weights, a difference in the effects of these additions, but in 1879, in pursuing further investigations with quinamine, he found that the rotation of this alkaloid remains almost unchanged in whatever acid it is dissolved. 3 The law then stated, " the specific rotation of the alkaloids is modified in the same manner by different acids, provided the salts formed repre- sent the same condition of saturation of the alkaloid by the acid" was later confirmed by Oudemans 4 in the study of quinidine- amine and. by Tykociner 5 with brucine, strychnine, morphine and codeine. Finally, Oudemaus 6 showed that the rule holds good for the active acids as he found that podocarpic acid and quinic acid, after saturation with different bases, or in the form of dilute salt solutions, retain nearly the same rotating powers. The following table contains some of the results referred to, the rotation for equal molecular weights being expressed by (M) for salts or by \_a] D for the active group: Tart rates (Landolt). Quinates (Oudemans). Quinamine (Oude- mans). Strychnine (Tyko- ciner). i mol. of acid to i mol. of base. In 100 cc. In loo cc. In 100 cc. 7.69 grams tartaric acid. IXJs of the salt. In loo cc. 2 .6 grams quinic acid C;H 12 6 . M/> of the acid. Acids. 1.56 grams base. MS of the base 0.84 gram base. of the base. U 2 .C 4 H 4 6 (NH 4 ) 2 63*0 K.C 7 H U 6 Na 48.8 48.9 HC1 HN0 3 + 114-4 116.5 -34.1 34-1 Na, 59-9 NH 4 47-9 HC10 3 116.1 K 2 64.4 Ba(C 7 H u 6 ) 2 46.6 H 2 S0 4 116.4 35-3 . Na.NH 4 61 *" Sr 48.7 H 3 P0 4 II7-3 34-4 K.NH 4 63-8 Ca 48.7 H 3 As0 4 33-9 K.Na 62.3 Mg 47.8 CH 2 2 114.7 34-0 K.AsO 58.8 Zn 51.0 C 2 H 4 2 116.2 34-0 K.C 2 H 5 64.6 C 2 H 2 4 118.1 33-1 Bar/2C 2 H 5 63.0 33 9 Mg 61.7 Landolt : Ber. d. chem. Ges., 6, 1077 (1873). Oudemans: Ann. Chem. (Uebig), 182, 33, 58 (1876). Arch. Neerland, 10, 193. Oudemans: Ann. Chem. (Uebig), 197, 48, 66 (1879). Arch. Neerland, 13, 155. See further, Rec. trav. Chim. Pays-Bas, i, 18 (1882). Oudemans: Ann. Chem. (Uebig), aop, 38 (1881). Tykociner: Rec. trav. Chim. Pays-Bas., i, 144 (1882). Oudemans: Ibid., 4, 166 (1885). ELECTROLYTIC DISSOCIATION OF AQUEOUS SOLUTIONS 2 17 In investigations on the effect of concentration on the rota- tion of salts of malic and camphoric acids it was shown further that their molecular relations, which are very different in strong solutions, become nearly the same for the salts of the same acid as the dilution increases. Thus, the following molecular rotations, [-$/], are calculated from the interpolation formulas found by Schneider 1 for a number of alkali malates, the water present amounting to about 40 to 90 per cent: Water in 100 parts by weight of solution. Mol. Wt 40 50 60 70 80 90 Malic acid C 4 H 6 O 5 134 -h3.io + 1.89 + 0.68 - 0.59 - 1.71 - 2.91 Acid salts UP "R O 140 151 156 I 7 2 i So 8f*\ 10.16 11.28 - 9-45 8 72 11.77 9Ro (NH ^ " 3.O2 -7.72 -0.08 44 -8.15 -2.65 c Q 7 -8.58 5/1/1 ^jsn 4 ; j^ a .09 10.02 n f\R K " .U4 -6.83 7-5 7-77 o. /^ - 8.74 4.92 o-/ Neutral salts U,.C 4 H 4 5 (NH 4 ), " Na, " K 2 " 146 168 178 210 + 5-90 -7.29 + 5-73 -5-15 0.26 -8.1 3 + 1.16 -7-43 -5-62 -9.17 -3-17 -9-47 10.09 10.35 - 7.16 11.32 13-77 16.54 11.74 13.27 10.93 14-31 12.87 14.26 As plainly evident, the rotations of these salts in concen- trated solutions are very different among themselves, and even differ in sign, but with increasing dilution they become more and more uniform. Further, it is seen that the free malic acid shows a molecular rotation very different from that of the malates. An explanation of these phenomena was first given by Hadrich 2 in 1893, an d based on the theory of electrolytic dis- sociation which meanwhile had been proposed by Arrhenius. Hadrich makes it clear, that the closely agreeing rotations which the different salts of an active acid or base exhibit in equimolecular solutions, have a meaning as soon as it is assumed that these bodies, as electrolytes, become largely 1 Schneider: Ann. Chem. (I^iebig), 207, 257. - Hadrich : Ztschr. phys. Chem., 12, 476. 218 SPECIFIC ROTATION dissociated into their ions on sufficient dilution, because then from each one the same amount of active acid or base ions would be formed. In accordance with this view he gave to the law expressed by Oudemans, this extension : The rotating power, not only of salts, but of electrolytes in general, is, in approximately completely dissociated solutions, independent of the inactive ion. A confirmation of this law was given by Hadrich by show- ing that when an alkaloid is neutralized by different acids, and the solutions then are treated with increasing amounts of water, ( i ) , the molecular rotations become constant for each salt from a certain concentration on, and, (2), that the constant values for the molecular rotations of the different salts agree also among themselves, as shown by the observations already given. In the following table, giving some of these experiments, the numbers show the molecular rotations of the salts 1 and under V is given the volume of water in liters, in which one gram equivalent of the salt is dissolved : Quinidine. Morphine. V Hydro- chloride. Nitrate. Sulphate. Hydro- chloride. Nitrate. Sulphate. 10 ... ... -359 -36l -357 20 + 703 + 703 + 702 364 364 3 6 4 30 712 710 710 365 365 365 40 717 717 717 37i 368 368 60 719 723 719 370 370 365 80 726 726 726 374 369 374 120 723 723 723 ... 160 726 726 726 ... ... r/ Brucine. Strychnine. Hydro- chloride. Nitrate. Hydro- chloride. Nitrate. Sulphate. Amygdalate. 10 -I 5 6 -I 5 6 20 141 141 -H3 -II 4 -"3 -113 30 138 138 114 114 114 114 40 136 136 "3 "3 H3 "3 i Instead of giving the molecular rotations, the observed angles for a given tube length could have been given, as the solutions contain in equal volumes, the same amount of the corresponding alkaloid. ELECTROLYTIC DISSOCIATION IN AQUEOUS SOLUTIONS 2 19 The quinidine and morphine salts show at first an increase in rotation with increasing dilution and constant values are reached when the dilution of the first, F, is brought to 80 liters and of the second to 40 liters. The alkaloid iron possesses, therefore, in both cases a greater rotating power than the un- dissociated salt molecule. With brucine the opposite is the case; and the strychnine salts appear to be already dissociated in N/20 solution. The agreement in molecular rotation in dilute aqueous solu- tions has been observed in many other salts containing active bases or acids, although sometimes with considerable varia- tions, the cause of which is found partly in insufficient dilu- tion, and partly in the unequal degrees of dissociation for the different salts. Further, it must be remembered that in dilute solutions, as the observed angles are very small, experimental errors exert a great influence on the final result. Observations have been made on the following substances: Alkali salts of malic acid (Schneider), 1 tartrates of different metals (von Sonnenthal, 2 Pribram), 3 tartratesof substituted amines (Kanno- nikoff), 4 alkali salts of methyl and ethyl tartaric acid (Fayollat), 5 salts of active gly eerie acid (Frankland and Appleyard), 67 salts of quinic acid (Cerkez), 8 alkali salts of active valeric acid and compounds of valeric acid with inactive organic bases in alcoholic solutions (Guye and Rossi), 9 salts of amyl sulphuric acid and salts of active di-isoamyl amine (Carrari and Gennari), 10 conine hydrochloride and hydro- bromide (Zecchini), 11 nicotine salts (Schwebel, 12 Carrari), 13 Schneider: Ann. Chem. (Liebig), 207, 257. v. Sonnenthal: Ztschr. phys. Chem., 9, 656. Pribratn: Wien. Monatsh., 14, 742. Kannonikoff : J. russ. phys.-chem. Ges., 22, 36. Fayollat: Compt. rend., 117, 632. Frankland and Appleyard: J. Chem. Soc., 63, 296. Frankland and Appleyard, J. Chem. Soc., 63, 311, observed, especially with the magnesium, zinc and cadmium salts of active glyceric acid, variations in the molecular rotation from that found with the alkali salts, and in consequence of this were inclined to question the universality of the Oudemans rule. . But they used 10 per cent, solutions, in which the dissociation was incomplete, and which likewise for the differ- ent salts, especially those of the dyad metals, might be very unequal in extent. 8 Cerkez: Compt. rend., 117, 174. 9 Guye and Rossi: Bull. Soc. Chim., [3], 13, 465. 10 Carrara and Gennari: Ztschr. phys. Chem., 17, 561. 11 Zecchini: Ztschr. phys, Chem., 16, 246. 12 Schwebel: Ber. d. chem. Ges., 15, 2850. la Carrara: Ztschr. phys. Chem., 14, 562, 16, 244. 220 SPECIFIC ROTATION cinchonidine salts (Schuster), 1 salts of d- and /-ruenthyl- amine (Binz). 2 Of observations of this character, those which Walden* made on tf-bromcamphorsulphonic acid and its salts, may be given here because there was found at the same time the extent of dissociation by means of determination of the electrical con- ductivity. The solutions contained equivalent amounts, and c indicates the number of grams in 100 cc. a is the angle of rotation found for a column 4 dm. in length at a temperature Of 20. 5 V Dissociation a D 1>_U L ID in per. cent. Free-a-bromcamphor- 14.952 + 55-20 + 92.3 + 287 68. 5 sulphontc acid. 1.0366 3.64 87-7 273 92.7 C 10 H u BrO.SO 3 H 0.5183 1.795 86.6 269 94-4 M= 311. 0.2592 0.901 86.9 270 95-5 Potassium salt. i 1633 3.644 78.3 273 83-6 C 10 H 14 BrO.S0 3 K 0.5817 1-793 77.1 269 8 7 .2 M = 349. 0.2908 0.898 77.2 269 90-3 Thallium salt. I.7I34 3.633 53-i 273 83.9 C 10 H 14 BrO.S0 3 Tl 0.8567 1.817 52.9 272 87-3 M= 5 1 4 . 0.4283 0.903 52.7 271 90-5 Zinc salt. 1.1417 3.620 79-3 272 71.5 K(C 10 H M BrO.SO.) 1 .Zn] 0.5709 1-795 78.6 269 77.2 M = 342.5. 0.2854 0.900 78.8 270 81.8 Barium salt. 1.2617 3-630 71.9 272 69.8 H(C 10 H u BrO.S0 3 ) 2 .Ba) 0.6309 1.807 71-6 271 74-8 ^=378.5 0.3154 0.895 70.9 269 79-4 As can be seen, the molecular rotations of all these bodies in dilute solution approach rapidly the constant value of 269 to 270, even the salts of dyad metals, although these are less dissociated than the alkali metal salts or the free acid. In 1 Schuster: Wien. Monatsh., 14, 573. * Binz: Ztschr. phys. Chem., la, 734. 3 Walden : Ztschr. phys. Chem., 15, 196. Kipping and Pope, also, (J. Chem. Soc., 63, 548) have investigated some salts of the acid. * The original paper contains observations for the sodium and glucinum salts. ELECTROLYTIC DISSOCIATION OF AQUEOUS SOLUTIONS 221 general, a relatively small change in the molecular rotations is observed, corresponding to the increasing degree of dis- sociation. Free acids which behave as good electrolytes, must show, in sufficiently dilute solution, the same molecular rotation as the neutral salts, since the concentration of the active ions is finally the same. This is illustrated with tf-bromcamphor- sulphonic acid described above. If, on the other hand, the acid is a poor electrolyte and at the same time is dibasic, as tartaric acid or malic acid, then in consequence of the lower degree of dissociation and also of the formation of different active ions (for example, C 4 H 5 O 5 and C 4 H 4 O 5 from C 4 H 6 O 5 ) the observed molecular rotation will depart widely from that of the neutral salts. 1 The acid salts also may not agree with these, because here different conditions of dissociation obtain. These differences are shown in the following table which em- braces observations of Schneider* on malic acid and malates in 5 per cent, solutions, and of Landolt 3 on tartaric acid and tartrates: Malic acid. IM\ D Tartaric acid. [*]* Free acids. C 4 H 6 5 - 3.2 C 4 H 6 6 + 21.1 Li.C 4 H 5 5 11.9 Li.C 4 H 5 6 +.42.8 Acid salts. Na K 10.5 10.2 Na K 4L2 42.5 NH 4 " 10.1 NH 4 " 42.8 U 2 C 4 H 4 5 -17.7 Li. 2 C 4 H 4 6 + 58.1 Neutral Naj 16.0 Na 2 " 59-9 salts. K 2 " 14.8 I K 2 64.4 (NH 4 ) 2 " 14.1 (NH 4 ) 2 '< 63.0 1 From the measurements of Ostwald (Ztschr. phys. Chem., 3, 371) on the elec- trical conductivity of tartaric acid, it follows that this acid in a concentration of 0.3 gram per liter is only about one-half dissociated, while the extent of dissociation for the neutral tartrates can be taken as above 95 per cent. 2 Schneider: Ann. Chem. (Liebig), 207, 257. The numbers given are calculated from the interpolation formula given for q = 95. 3 lyandolt: Ber. d. chem. Ges., 16, 1076. The concentrations employed were equiva. lent to 7.69 grams of tartaric acid in 100 cc. for the neutral tartrates. Weaker solutions were used for the acid salts. 222 SPECIFIC ROTATION The acid salts in respect to their rotation stand between the free acids and the neutral salts. From them, for example from the acid malates, at first, in the main, the ion C 4 H 5 O 5 separates, which with greater dilution passes into C 4 H 4 O 5 . Finally the same molecular rotation should be expected as with the neutral salts but sufficient observations are lacking to show this. 1 With the salts of very weak bases and acids besides the electrolytic, hydrolytic dissociation may also take place, by which the number of atomic aggregations in the liquid is still further increased. Such complicated changes appear to- take place with the di-hydrochlorides of the cinchona alkaloids, inasmuch as these do not, like the monohydrochlorides, give a constant end value for the molecular rotation by increasing dilution. The following numbers for \_M~\ D were found by Hadrich, 2 in which i gram-molecule of substance was con- tained in V liters: V 10 20 40 80 160 3 Cinchonidine- dihydrochloride - 525 - 521 - 516 - 504 - 465 monohydrochloride - 356 - 381 - 400 - 402 Quinidine- dihydrochloride- . . . -+- IOT1 + 1028 -1043 -h 1049 4- 1123 + 1225 monohydrochloride + 703 + 717 + 726 + 726 The great differences between the mono- and dihydrochlorides are without doubt caused not only by differences in the nature of the electrolytic dissociation but also, especially with quini- dine, by the existence of other kinds of dissociation. If the degree of dissociation of an active body be diminished by adding to the solution other substances which also behave as electrolytes, a change in the rotating power follows. Acids must produce such an action and in fact it has been found that the specific rotation of tartaric acid experiences a decrease when the aqueous solution is treated with hydrochloric, nitric, sulphuric, or acetic acid. Equivalent amounts of these acids 1 The interpolation formulas of Schneider gave for q = too very great differences between the acid and neutral malates. 2 Hadrich: Ztschr. phys. Chem., la, 491. HYDROLYTIC DISSOCIATION 223 exert influences in different degrees (Landolt). 1 Oudemans 2 observed the same phenomenon on treating di-acid alkaloids with i, 2, 3, ..., molecules of different acids. At first an increase in the specific rotation follows and continues to a maximum, which appears when somewhat more acid is present than is required for formation of the neutral salt, and then a continuous decrease takes place. Of numerous observations the following may be given: Cinchonine. In 100 cc. of solution 5 mg. molecules of alkaloid -f- n m g- molecules of acid. The maximum values are shown by * : Mol. acid to i mol.base. Hydrochloric acid. Nitric acid. Formic acid. I [>]2 = -f 201.0 + I9I.7 .... 2 2-54-1 253.4 242.2 4 259.0* 257.3 243-9 3 258.7 257-8* 245-6 4 257-7 254.6 250.7 6 253-3 252.1 256.6 10 252.1 251.8 257.8 20 246.0 258.9* 45 .... .... 257-9 92 254.0 The increase in the rotation at the beginning may possibly be explained by the assumption of hydrolytic dissociation on addition of small amounts of acid. The behavior of the acids depending on their different degrees of affinity is also evident, inasmuch as to reach the maximum rotation unequal amounts, of hydrochloric acid, 2^ mols.,of formic acid, on the other hand, 20 mols. , are required. Alkalies act in the same manner as acids. From observa- tions of Th. Thomsen 3 it appears that the specific rotation of neutral sodium tartrate undergoes a progressive decrease on addition of increasing amounts of sodium hydroxide, while by addition of water, on the other hand, it increases. In the same way the changes in specific rotation which 1 I^andolt: Ber. d. chem. Ges., 13, 2331. 2 Oudemans: Rec. trav. Chim. Pays-Bas, i, 28. 3 Th. Thomsen: J. prakt. Chem. [2], 35, 145; also Aignan: Compt. rend., na, 1009. 224 SPECIFIC ROTATION follow by addition of salts ( 70) depend largely on alterations in electrolytic dissociation. If, finally, active electrolytes are dissolved in liquids which possess a smaller dissociating power than water, the specific rotation in comparison with that in the latter will assume a new value, which may be larger or smaller according as the active ion possesses a greater or less rotating power than the undissociated molecule. The same phenomenon must be observed when such a liquid, for example alcohol or acetone, is added to an aqueous solution of the body. Among the many observations on this point, the following by Walden 1 may be quoted in which the extent of dissociation has been calculated from the electric conductivity: Substance. Solvent. c ws> Dissociation. Per cent. Water \ 271 -Brom- camphor 7 parts water 1 ~f~ 93 P ar ts acetone * Water 1.0366 o ^187 T *'j 343 260 92.7 5-9 sulphonic acid 3.5 parts water "I ^ + 96.5 parts acetone I 0.5183 ^uy 326 94-4 4.1 I 2617 272 60 8 a-Brom- camphor 7 parts water ) 4- 93 parts acetone * ' Water 1.2617 o 6100 */ 328 271 8.1 7/1 8 sulphonate 3.5 parts water 1 -4-96.5 parts acetone f U>U O' J 7 0.6309 */* 301 74.0 5-0 Oudemans 2 found that some of the salts of the cinchona alkaloids rotate in alcoholic solution more strongly than in aqueous, others less strongly. In general many of the variations shown in the specific rota- tion of bodies dissolved in different solvents depend on differ- ences in the extent of electrolytic dissociation, provided dilute solutions are considered. 1 Walden: Ztschr. phys. Chem., 15, 205. 2 Oudemans: Rec. trav. Chim. Pays-Bas, i, 18. DISSOCIATION OF SALTS 225 Dissociation of Salts with Active Anion and Kation. For such bodies the experiments of Walden 1 have shown, as was to be expected, that the rotations in dilute solutions are equal to the sum of the rotations for the ions. For flf-bromcamphorsul- phonate of morphine, C 10 H u BrO.HSO 3 .C 17 H 19 NO 3 , dissolved in water, he found: C = 1.9867 \M]n = 100 c = 0.9933 \M]D = 101 For the morphine ion we have, according to the experiments of Hadrich, cited above, the values 365 to 374, in the mean 371 for \M\ D . For the ion of bromcamphor sulphonic acid we have \_M~\ D = -f 269 to -f- 273, in the mean 4-271. Hence as molecular rotation of the dissociated salt we must have \M\ D = - 37 1 + 271 = - 100, which agrees with the above observation. This behavior is shown also with quinidine tf-bromcamphor sulphonate, where both ions are right rotating. Behavior of Boryl, Arsenyl and Antimonyl Tartrates. These compounds which are formed by. heating acid tartrates with boric acid, arsenious oxide and antimonious oxide show marked deviations in their rotating power from the ordinary neutral tartrates. With the latter the rotation increases with increas- ing dilution and reaches a constant value w r hich corresponds to the completely separated ion, C 4 H 4 O 6 . Thus from the formula of Th. Thomsen 2 for ^/-sodium tartrate, [M^D = 60.56 0.04647^ o. 002216 p 2 , the following numbers may be calculated for solutions which contain F liters of water for i gram-mol. of salt (or in 100 parts of solution p grams of salt). These values for [M~\ D change but little. V i 2 ~ 4 8 16 32 p 0.1625 0.0884 0.0462 0.0238 0.0120 0.0060 [M]/> 59.22 59.98 60.30 60.45 6o -5o 60.53 The same end value, as remarked before, is found with the other neutral tartrates. On the other hand Hadrich found for certain alkali boryl 1 Walden: Ztschr. phys. Chem., 15, 206. - Thomsen: Jour, prakt. Chem. [2], 34, 80. 15 226 SPECIFIC ROTATION tartrates the following molecular rotations, when the same dilutions were employed (i gram-mol. of salt in V liters). 1 V. K(BO)C 4 H 4 6 . Na(BO)C 4 H 4 O 6 . NH 4 (BO)C 4 H 4 . I + 143 + 152 + H8 2 134 133 136 4 121 122 121 8 106 107 107 16 88 8 9 8 9 32 74 74 74 We have here to begin w r ith a much larger molecular rota- tion than with the simple tartrates, and this is explained by the circumstance that the separated ion is not C 4 H 4 O 6 , but C 4 H 4 O 6 .BO. Secondly, the rotation decreases rapidly with increasing dilution and for V - 16 is not yet approaching constancy. As appears from the investigations of Magnanini 2 on the conductivity of solutions of boro-tartaric acid this may be referred to gradual hydrolysis taking place at the same time which brings about a decomposition of the complex ion, C 4 H 4 O 6 .BO. Finally the ion C 4 H 4 O 6 , with \_M^ D = 60.5, must be present. The arsenyl tartrates act in the same way. Hadrich found for [AT\ D : V Na(AsO)C 4 H 4 6 . NH 4 (AsO)C 4 H 4 O ti . 2 4- 224 -f 230 4 185 186 8 131 132 16 79 80 32 63 63 With these compounds in the most dilute solution the rota- tion of the tartar ic acid ion (58 to 63 as already given) has been reached. Other phenomena are shown by potassium antimonyl tartrate (tartar emetic). Here we have very strong rotation which scarcely decreases by dilution. For the formula KSbOC 4 H 4 O a , Hadrich 3 found these numbers: 1 Hadrich: Ztschr. phys. Chem., 13, 494. * Magnanini: Ibid., 6, 67. * Hadrich: loc. cit. DISSOCIATION OF MOLECULAR AGGREGATIONS 227 1= 2.35 dm. 4 13.70 548 i6 32 64 6.85 3.42 1.70 0.85 548 546 544 544 Mol. conductivity, M ... 70.58 79.45 87.63 94.20 Some hydrolysis takes place here as shown by Hadrich from the manner of change in the conductivity, but only to a slight extent. An explanation of the slight change in the molecular rotations is still lacking. 62. b. Formation or Decomposition of Molecular Aggregations of Simple Structure. As is well known molecular weight deter- minations by the freezing- or boiling-point method have shown that many substances, liquid as well as solid, when dissolved in certain liquids appear as single molecules, while in others they exist as double molecules (for example, acetic acid in ether = C 2 H 4 O,, in benzene = (C 2 H 4 O 2 ) 2 , etc.). Experiments have accordingly been made to determine whether the influence which several solvents, or their concentrations, exert on the specific rotations of many substances corresponds to a change in the molecular weight of the latter. On this subject we have, mainly, the following investigations: Freundler 1 dissolved a number of tetra-substituted tartaric acid esters, the rotations of which in pure condition were known, in different liquids (c = 5 to 6) and determined the specific rotation and the molecular weight. He believes the following laws obtain for these bodies: i . In solvents which change the rotation of the esters but little or not at all, the latter show the normal molecular weight. For example: Molecular weight. Spec. rot. [<*] D . Solvent. Active substance. From Obser- In solu- Without - formula. vation. tion. solvent. - Propyl dipropionyl tartrate 346 342 + 5-4 + 5-5 Ethy- dibutyryl 374 363 + 5-5 + 5-2 lene " divaleryl " 402 389 + 3-6 + 3-6 bromide " dicaproyl 430 424 -h 2.4 + 2.2 Methyl divaleryl 346 348 -15-6 15-9 Benzene Isobutylamyl oxide 144 I4T + 1.4 + 1-3 1 Freundler: Ann. chim. phys., [7], 4, 256 (1895). 228 SPECIFIC ROTATION. 2. In solvents which bring about a marked change in the original specific rotation of the esters, anomalous numbers are found in the cryoscopic molecular weight determinations. For example : Solvent. Active substance. Molecular weight. Spec. rot. \_Ot\ D . From the formula. Obser- vation. In solu- tion c = 5 to 6 Without solvent. Benzene Propyl diacetyl tartrate 318 277 4- 1.2 + 13.4* " dipropionyl " 346 295 - 3-4 + 5-6 " dibutyryl " 374 304 1.4 + 5-2 i 11 divaleryl " 402 324 2.2 + 3-3 11 " dicaproyl " 430 345 - 4-3 4- 2.2 Nitro- benzene Isobutyl diacetyl " 346 3i8 -f- 12.0 -f 17.0 Nitro- benzene Ethyl dicaproyl " 402 376 - 5-i - 3-i Acetic acid Isobutyl dipropionyl " 374 287 -f 20.2 -f- IO.2 Ethylene bromide Ethyl diphenylacetyl tartrate 442 394 + 19.2 + 15-2 Ethylene - bromide Propyl " " " 470 406 + 23.3 + 20.9 Benzene (i 470 413 4- 15-7 + 20.9 Nitro- benzene i< > i 470 378 -f 14.6 + 20.9 Acetic acid i < < < < < K 470 377 + 27.2 + 20.9 In all these cases, the specific rotation of the dissolved sub- stance is markedly different from that of the original solvent- free body, being sometimes higher, sometimes lower, and sometimes showing a change in direction. The molecular weights, as determined, are all below the normal, which probably depends on dissociation of the compounds. Freundler also found substances whose molecular weights in solution are much larger than the formula weights and which show marked changes in rotating power. The explanation here may be found in polymerization. The following simple esters of From the formula. Observa- tion. In solution. Without solvent. + 2.14 Benzene Methyl tartrate 178 411 - 8.8 Benzene Propyl " 234 306 -f 20.1 + 12.44 Ethyl bromide 234 326 - 0.6 + 12.44 The rotation and molecular weight of nicotine in different solvents has been investigated by Hem, 1 and with concentra- tions at which the boiling-point method yields reliable results. It was found that by diminishing the percentage amount of nicotine, /, the specific rotation was also diminished, although w r ith several liquids, as ether, acetone, and benzene, in very small degree, and somewhat more with ethyl and propyl alcohol. The molecular weights appear from the observations to undergo a slight decrease with decrease in /, but the values are all very near the normal number. The following are the results obtained: Pure nicotine: a*% = 164.0. Molecular weight 162. Solvent. Decrease in percentage amount of nicotine. Corresponding decrease in specific rotation. Ms Molecular weight found. Ethyl alcohol . Propyl alcohol T^tVipr From 11.4 to 1.7 " 13.4 " 2.0 From 141.1 to 139.0 - 147.2 " 144.6 < 169 i " ifii 8 167 to 164 156 " 152 i v-y 4- u * ' 12 O *' 2 'Z " 163 3 " 162 6 192 177 188 " 172 " IA.A " 7.S " Tfi7 8 " fftl A T7C " T72 Solutions of nicotine in water show, on the other hand, a different behavior. As pointed out in 56 the specific rotation changes within the limits [or]" = 76.84 to 77.59 when the percentage strength sinks from/ = 15.59 to 1.06, with a mini- mum at/ = 9. The molecular weight, found cryoscopically, shows however, according to the observations in 56, a very strong decrease; it has for/ = 13.74 the value 275, which gradually sinks to the normal, 162, when/ is less than 2 per 1 J. Hein: Ueber das specif. Drehungsvermogen und das Moleculargewicht des Nicotins inLosungen, Inaug. Diss., Berlin (1896). 230 SPECIFIC ROTATION cent. In this case the great change in molecular weight has no influence on the rotating power of the substance. Rotation and molecular weight in solutions of different con- centrations have been further investigated by Frankland and Pickard 1 with the following substances: Solvent. Decrease in per centage amount of active sub- stance. Corresponding change in specific rotation, [U Molecular weight found cryoscopically. rf-Dibenzoylgly cerate of methyl [a] X J = -f 26.9 ; M == 328. Between 299 and 322 Benzene 34. i to 3.0 Inc. from +40.7 to 45.7 Nitro- benzene 28.1 " 2.4 j Dec. " 4-22.0 " 19.8 Ethylene- bromide 22.3 "3.3 11 " 4- 21.7 " 19.2 Acetic acid 18.6 " 1.7 Inc. " + 32.4 " 34-3 305 " 341 322 ' 359 305 " 34i /-Diacetylglycerate of ethyl [a] 'J = 16.31 ; M= 218. Benzene Acetic acid 29.8 to 5.3 25-0 " 3-4 Inc. from i i4.8to 17.2 19.4 " 28.7 Dec. from 216 to 209 " 194 " 136 With the first ester, no definite change in the molecular weight, corresponding to increase or decrease in the specific rotation with diminished concentration, is noticed ; for the molecular weight, irregularly varying numbers were found, which are not very far from the normal formula weight. The diacetylglycerate of ethyl dissolved in benzene shows an increase in rotation, but a decrease in the corresponding nearly normal molecular weight. In acetic acid, the rotation increases likewise and the molecular weight decreases, but the latter shows values which are much smaller than the normal, so that dissociation appears to have taken place here. With the ethyl ester of /-mandelic acid, the following values were found by Walden* for the specific rotation and molecular weight, the latter being determined by elevation of the boiling- point : 1 P. Frankland and Pickard: J. Chem. Soc., 69, 123. 2 Walden : Ztschr. phys. Chem., 17, 705. DISSOCIATION OF MOLECULAR AGGREGATIONS 23! Pure ester (superfused) [a] D = - 123.1; M= 180. Solvent. c Molecular weight found. 175-4 (4-2i subst. in 100 pts. solution) I. ID 07.1 Carbon disulphide 5.00 - 180.0 ,, M " 2.50 iSo.o The original specific rotation of the ester experiences, there- fore, in acetone a marked decrease, but in disulphide of carbon, on the other hand, an increase, while the molecular weight is normal in both solutions. Finally, o'-mononitrocamphor dissolved in carbon disulphide shows a strongly decreasing rotation with increase in concen- tration, but in alcohol only a slight change. In both solvents the substance possesses the normal molecular weight (Pescetta). 1 According to the above observations the following phe- nomena, in general, have been noticed: 1. A change in the rotation, with the molecular weight remaining normal (ethyl mandelate, tf-mononitrocamphor, nicotine in ethyl and propyl alcohol). 2. A change in the molecular weight while the rotation remains constant (nicotine in water) . In these two cases there can, naturally, be no relation between the constants. 3. Simultaneous changes in rotation and molecular weight. If here the molecular weight in solution is much greater than the normal, the cause of the modified rotation is probably found in a polymerization of the molecule ( simple tartrate esters). If the molecular weight is found to be smaller than the normal, the change in rotation is probably due to beginning dissocia- tion (diacetyl glycerate of ethyl in acetic acid). Whether or not the variations from the original specific rotation which were found in the tetra-substituted esters of tartaric acid, investigated by Freundler, when they were dis- solved in different solvents, have any connection with the decrease in molecular weight observed at the same time, can not be shown with certainty. 1 Pescetta: Gazz. chim. ital., 25, II, 418. 232 SPECIFIC ROTATION 63. c. Presence of Complex Polymerized Molecules (Crystal Mole- cules) in the Solution. While it is not clear how by association of two or only a few molecules the original rotation should be altered, some action should follow, on the other hand, when a large number of active molecules unite to produce a crystal structure, which in turn possesses asymmetric form. It has already been shown, in 7, that those bodies which are active in dissolved and in crystalline condition possess, in the latter form, a rotating power which is due to the combined activity of the single molecules and the crystal molecules. If the assump- tion may be made that in concentrated solutions, at least, of solid active substances, such complex aggregations are present, which by continued dilution gradually break down into normal molecules, then the corresponding changes in the rotation may be explained. The possibility of the occurrence of such crystal molecules in solutions has been frequently affirmed by Groth, 1 Fock, 2 Bell, 3 Wyrouboff 4 and others, but experimental proof is thus far wholly lacking. It is, however, possible that the following phenomena observed in aqueous solutions of malic and tartaric acids may be ascribed to this cause. Ordinary malic acid exhibits left rotation in dilute aqueous solutions, and this grows less with increasing concentration, passes through a point of inactivity, and finally turns to increas- ing right rotation ( 57). The same phenomenon is noticed on lowering the temperature ( 60). With ^/-tartaric acid, on the other hand, the rotation changes from right to left gradu- ally as the concentration becomes very great ( 46). For each spectrum color the point of inactivity appears at a certain and distinct concentration ( 46). These marked variations in the rotation can not be explained, as Nasini and Gennari 5 especially have pointed out, by ( i ) electrolytic dissociation, because this with malic acid and tar- taric acid is noticeable only in very dilute solutions, where an accurate observation of the rotation could no longer be made ; Groth: " Physikal. Krystallog.," Ill ed. (1895), p 268. Fock: " Einleitung in die cheniische Krystallographie " (1888), p. 19. Louis Bell: Silliman's Jour. [3], 7, 120. Wyrouboff: Compt. rend., 115, 832; Bull. Soc. Chim., [3], 9, 214. Nasini and Gennari: Ztschr. phys. Chem., 19, 113 PRESENCE OF COMPLEX POLYMERIZED MOLECULES 233 (2) by simple polymerization, as cryoscopic observations with malic acid in concentrations 9 and 24.5 have shown the normal molecular weight ; (3) by formation of hydrates of variable composition; for the reasons given in 64 these are in general not possible. The phenomenon may be understood, however, if w r e assume that the left-rotating single molecules of malic acid with increasing concentration gradually combine to form right- rotating aggregations, and the right-rotating tartaric acid molecules to form left-rotating groups. Accordingly, finally, in anhydrous condition, /-malic acid should exhibit right rotation and d- tartaric acid left rotation. With the first acid this has not been shown experimentally, 1 but in the case of tartaric acid it has been, as already mentioned in 46. That this condition can actually obtain when solid crystalline par- ticles separate is shown in the case of rubidium tartrate, which, as explained in 7, possesses right rotation in solution, but left rotation as salt. The assumption that in solutions of malic and tartaric acids, single molecules and molecular aggregations occur at the same time, and possess opposite rotations, would explain : (i) The anomalous rotation dispersion of the two substances (46); (2) the parallel change in rotation with increasing dilution or elevation of temperatures (s 60), as both causes would lead to a breaking down of molecular aggregates ; (3) the phenomenon referred to in 59 in which solutions of ^-tartaric acid in mixtures of alcohol and benzene or other hydrocarbons exhibit left rotation, inasmuch as these liquids, as is well known, have the power of favoring the formation of molecular combinations. A proof of aggregations by cryoscopic methods is not pos- sible, as these do not exist in dilute solutions. In such solutions, as shown in 20, tartaric acid has the normal molecular weight. The further changes observed in the rotation of tartaric acid, with great dilution ( 55), find their explanation in the now possible electrolytic dissociation. As may be finally remarked, phenomena different from those referred to above had already led to the view that molecular combinations exist in concentrated solutions which i This has since been shown by Walden. See Part VI, Constants of Rotation. Tr. 234 SPECIFIC ROTATION break down with increasing dilution. Hittorf explains in this way the abnormal behavior of cadmium salts on electrolysis. 1 64. d. Combinations of the Active Body with the Solvent. Hydrates. Biot 2 attempted to explain the changes in the specific rotation of tartaric acid on increasing dilution on the assumption of the formation of hydrates containing more and more water. But thus far, it has not been found possible with this substance or with others to positively prove the existence of such compounds, as the methods based on observations of osmotic pressure furnish here no information. As Nernst 3 has shown, a pro- gressive formation or decomposition of hydrates with increasing dilution is in general not possible, and for the following reasons: If a molecule, A, with n molecules of another substances B (w r ater) enters into the reversible reaction, A -f n B = A B n , and the corresponding concentrations are, c, c, c, then must, by the Guldberg-Waage law, T - n Kc = c l c. t . If the molecule species, B, represents the solvent, present in excess, then its concentration, c. 2> in comparison with ^ and c, is very large, and it will be but little changed in the reaction, whatever direction this takes. Consequently c 2 may be com- bined with the constant K and we have : - = const. ; c \ that is, for all concentrations, the relation of the hydrated to the non-hydrated molecules must remain the same. This law would naturally no longer obtain if the substance on solution should form several kinds of groups, A, by polymerization or chem- ical decomposition. Hydrates of definite composition are without doubt formed by the solution of certain active bodies in water. This is indi- cated, for example, by the strong liberation of heat in the case 1 See H. Jahn : " Grundriss der Electrochemie," Vienna, 1895, pp. 49 and 57. 2 Biot: M6m. de I'lnstitut., T. 15 (1838). 8 Nernst : Ztschr. phys. Chem., n, 345; " Theoretische Chemie," p. 370. COMBINATIONS OF THE ACTIVE BODY AND SOLVENT 235 of nicotine (15 for 24 grams of nicotine and 6 grams of water); also by the phenomenon that strong solutions separate, on heating, into the oily base and water. Further, as follows from the observations cited in 52, the density of the solutions increases with increasing addition of water, reaches a maxi- mum with the proportions 65.9 nicotine to 34.1 water (corre- sponding to C 10 H 14 N 2 .5H 2 O) and then rapidly decreases. This peculiarity in the variations in the specific gravity is not shown how r ever, in the continuous decrease exhibited by the specific rotation, and it may therefore be questioned if the latter is influenced by the nicotine hydrate. The changes in the rotation of aqueous solutions of malic acid, referred to in 57, have been accounted for by Bremer 1 on the assumption that the acid itself possesses right-hand rotation while the hydrates, COOH CH.OH CH 2 -C(OH) 3 , C(OH) 3 CH.OH CH 2 C(OH) S , show left-hand rotation. Further, the phenomenon that rhamnose hydrate, C 6 Hi 2 O 3 .H 2 O, dissolved in water on the one hand, and in cer- tain alcohols on the other, exhibits opposite rotation directions, has been explained by Rayman 2 on the hypothesis that the solution contains a hydrate, C 5 H n O 4 .CH(OH) 2 , in the one case and in the other, alcoholates, C 5 H U O 4 . *CH(OH)(OR), in which a new asymmetric carbon atom appears. The observed specific rotations referred to C 6 H 12 O 5 are the following: Water p = 5 to 40 [a~\ D = + 9- 2 to 9-4 3 Methyl alcohol . . p = 19 [#] D == 10.59* Ethyl alcohol p = 6.4; 9.3 \oi\ D 10.65;* IO - 5 Isobutyl alcohol . . p = 7.3 [<*] D 7.3* Amyl alcohol .... left-rotating 4 Isopropyl alcohol . \_ a ~\n = ~t~ ^' 6 ? 6 1 Bremer: Rec. trav. Chim. Pays-Bas., 3, 162, 336. 2 Rayman: Ber. d. chem. Ges., 21, 2050. 3 Rayman and Kruis: Bull. Sex:. Chim. (2), 48, 632; Schnelle and Tollens: Ann. Chem. (I = - 37, by addition of alcohol, benzene or acetic acid, which liquids finally yield a maximum value of [<*] = -38.8, -39.8, and 40.7 (52). Although in thesecases, as in many others, the increase or decrease in the rotating power is but small, it may still be followed with certainty. Parizek and Sulc: loc. cit.\ Sulc: Ber. d. chem. Ges., 37, 594. Fisher: Ber. d. chem. Ges., a6, 2400. Aignan: Pouv. Rot., Thesis, 1893, p. 24. Freundler: Bull. Soc. Chim., [3], 9, 683. Wyrouboff: Jour, de Phys. [3], a, 180; Ann. chim. phys. [7], i, i. Shields: Ztschr. phys. Chem., la, 167. Bredig : Ibid., 13, 322. COMPLEX SYSTEMS 237 Phenomena of this order, as already remarked in the first edition of this work, may possibly be accounted for by the hypothesis, that when between the molecules of a certain substance (turpentine) other molecules (alcohol) enter, certain modifications in the structure of the first result and of such a nature that in each molecule, the relative positions of the atoms, their arrangement in space and the conditions of their motions are somewhat altered. This will follow in greater degree, the larger the number of added inactive particles. Observations of other kinds of phenomena have also led to the same notions of possible slight perturbations in atomic equili- brium, not sufficient, however, to endanger the existence of the molecule. 1 E. Specific Rotation of Complex Systems 67. Solutions of an Active Body in Two Inactive Liquids If the change in the specific rotation of the body by each one of the solvents alone is expressed by the constants of the equations: m [] = + &J + ctf []. = a + b,g + <#>, in which a is nearly the same, the action of the mixture will be given by (II) [or] == a + (^P, + b,P,)q + (c,P v + cfjf, where 100 parts by weight of the solution of the active sub- stance contain q parts of the inactive liquid mixture, or i part by weight of the latter is made up of P l and P. 2 parts of the components. This formula applies, however, only when the two liquids mix with but slight change in volume or other physical property, as, in the other event, some modification in the behavior of the same with the active body might be expected. Rimbach 2 investigated the relations obtaining with solutions of camphor in mixtures of acetic ether and benzene. He found, as expressing the influence of the liquids separately: Camphor in acetic ether []^ = 56.54 0.0907 q -f o.ooo 401 g. 2 " benzene [] = 55-99 0.1847 q + o.ooo 269 q 1 Then the specific rotations were determined for a number of 1 See, for example, van 't Hc^T : " Etudes de dynamique chimique," 1884, p. 41. - Rimbach: Zeit. phys. Chem., 9, 698. 238 SPECIFIC ROTATION solutions which are given below in parallel with the values found by formula (II). For this calculation a was taken = 56.265. Mixture. 9 Acetic ether. A Benzene. /I Ms Cal. Obs. Observation. Calculation. 49-94 0.7509 0.2491 : 4- 51-76 51-49 -0.27 64.98 0.7509 0.2491 50.86 50.41 0.45 90.00 0.7509 0.2491 49-63 48.97 -0.66 46.21 0.5050 0.4950 50.88 50.66 0.22 64.96 0.5050 0.4950 49-35 48.77 -0.58 79.70 0.5050 0.4950 48.06 47.46 O.6o 89.49 0.5050 0.4950 47-32 46.67 0.65 40.16 0.2569 0.7431 50.35 50.31 O.O4 50.13 0.2569 0.7431 49.14 48.98 0.16 65.18 0.2569 0.7431 47.41 47-09 -0.32 So.OO 0.2569 0.7431 45.89 45.36 -0.53 89.69 0.2569 0.7431 44.86 44-30 0.56 With these mixtures the calculated specific rotation was always found a little less than that found by observation, the difference increasing with the dilution of the mixture. A second series of investigations made by Rimbach on solutions of right turpentine oil in mixtures of alcohol and glacial acetic acid showed very small differences between observation and calculation, and sometimes positive, some- times negative. While in the above illustrations, the specific rotation with mixtures has been found to lie between those found with the components, it has been noticed that in some cases the first may be considerably the larger. In this event, a maximum rotation is found for some definite mixture of the two liquids. The following are observations in this line : According to Hesse 1 cinchonidine gives in concentration, c= 2 : Dissolved in alcohol of 97 per cent by volume. [a~\ D = 106.9 "chloroform " - 83.9 " alcohol-chloroform (1:2) " 108.9 1 Hesse : Ann. Chem. (Liebig), 176, 219. COMPLEX SYSTEMS 239 For anhydrous cinchonidine nitrate and hydrochloride, Oudemans 1 obtained the following numbers : Hvdro- Solvent ^Nitrate, chloride, Water [<*~\D = '- 99-9 - 99-9 Absolute alcohol " 103.2 104.6 80 per cent.- alcohol 20 per cent.- water. " 127.0 128.7 89 " " ii " " " ... " 119.0 119.6 According to Oudemans quinidine hydrochloride in concen- tration c== 1.89, for the anhydrous salt, shows : Dissolved in water [#] D = -4- 190.8 " " absolute alcohol " 199-4 Dissolved in alcohol of 90.5 per. cent, by weight " 213.0 Hesse 3 has followed the changes in the specific rotation of quinine hydrochloride (with 2H 2 O) with variations in the proportions of water and alcohol used as a solvent, employing always the constant concentration, c= 2. From the follow- ing data, in which g gives the per cent, by volume of alcohol in the solvent, it appears that for g = 60, a maximum of rotation occurs : g = o 20 40 50 60 70 So 85 90 97 [a\ D = -138.8 166.6 182.8 187.5 187.8 182.3 174-8 168.3 160.8 143.9 Oudemans 4 gives the following observations on the specific rotation of cinchonine in mixtures of chloroform and alcohol : I 2 3456 Chloroform 100.00 99.66 98.74 94.48 86.95 82.26 Alcohol o 0.34 1.26 5.52 13.05 17.74 [#]/> +212.0 216.3 226.4 236.6 237.0 234.7 7 8 9 10 ii Chloroform 65.00 44.29 27.54 17.02 o.oo Alcohol 35-00 55.71 72.46 82.96 loo.oo [tx"] D 229.5 226.6 227.6 227.8 228.0 A maximum is found here which is shown by graphic interpolation to occur with the mixture containing 10 per cent, of alcohol. It is also observed that in an alcoholic solu- tion of cinchonine, about one-half of the alcohol may be replaced by chloroform without producing any marked change in the specific rotation, while on the other hand, if in a solu- 1 Oudemans: Ann. Chem. (I,iebig), 182, 49. 50. - By weight. 3 Hesse : Ann. Chem. (IL+A [].. A+A assuming that each body has no influence on the specific rota- tion of the other. If, however, some such action takes place the observed specific rotation must depart more or less widely from that calculated. An investigation of this question was undertaken by Ham- merschmidt 1 with the following substances : Mixtures of Right- and Left- Rotating Turpentine. Mixture No. In 100 parts of mixture. Observed rotation of the mixture. Ms Calculated specific rotation. Difference. Calc. Obs. Right oil. I,eft oil. 100 .... + 17-39 .... .... I 79-25 20.75 + 6.40 f 6. 4 I -f- o.oi II 60.40 39.60 - 3-54 - 3-55 + O.OI III 40.82 59-18 - 13-90 - 13-90 4- o.oi IV 20.83 79-77 28.82 28.80 0.02 .... 100 - 35-50 .... .... From these numbers it is evident that the specific rotations of mixtures of such similar bodies as two turpentine oils corre- spond exactly to the above mixture formula. It is further found by calculation that a mixture of 67.13 parts by weight of the right-hand oil with 32.87 parts by weight of the left- rotating oil must be inactive optically. 69. Solutions of Two Active Bodies in an Inactive Liquid. Let the mixture contain in 100 parts by weight : 1 Hammerschmidt: " Ueber das specifische Drehungsvermogen von Gemengen optisch activer Sui>stanzen." Inaug. Dissert. Rostock 1889. COMPLEX SYSTEMS 241 A^ per cent, of the first active substance, A 2 " " " " second " F " " " " inactive liquid, and let the effect of the solvent on the first active body be expressed by (I) [>], = , + b,p + c,p\ and that of the solvent on the second active body by (II) [a] f = a, + b t p + c,p\ in which formulas p gives the percentage amount of active substance in each solution. Then we substitute : In equation (I) for/ the value . 1-, ; (ID < p - j 00 ^. With the specific rotations [a] T and [a], so obtained, there follows for the mixture, but from the observed angle of rotation a , n we have the value, In order to judge of the difference between observation and calculation, it is preferable to compare the observed angles of rotation directly instead of the specific rotations, from which it will be seen w r hether or not the errors of observation are exceeded. The calculated angle of rotation follows by equating the last two formulas, as : ! 2 , d "!5o~ These deductions may be tested by some experiments which Hammerschmidt 1 carried out with aqueous solutions contain- ing cane-sugar and grape-sugar. The following tables give first, the observed data, and then the calculations, for which the interpolation formulas of Tollens are used in finding the specific rotations of the two sugars : 1 Hammerschmidt : Loc cit. 16 2 4 2 SPECIFIC ROTATION Cane-sugar Grape-sugar p = 66.386 + 0.015035 p 0.0003986 p- 7 , = 52.500 -7- 0.018796 p 0.0005168 p' 1 In 100 parts by weight. Solu'n No. Specific gravity of solution. Observed angle of rotation for / = i. 9992dm. Specific rotation. []-. Cane- Grape- Water sugar. A\ sugar. A F d O-m Ray D I 5-49 19.490 75.461 .09996 30.14 55.83 2 9.814 14.851 75-335 .10104 3I.8I 58.56 3 14.655 9.863 75.482 .10073 33-06 61.28 4 I9-5I7 4.892 75.591 .10054 34-25 63.78 5 19.558 4.855 75.587 .10056 34.30 63-85 Solu'n Cane- sugar. Grape- sugar. Cane- sugar. Grape-sugar. Calculated angle of rotation No. loo A] 100 A t [ a ]i L U J2 Rav n Calc.-Obs. A\ + F A z + F Ray/? RayD I 6.271 20.526 66.464 53-104 30.14 0.00 2 11.526 16.467 66.506 52.950 31.68 0.13 3 16.259 H-557 66.525 52.786 32.91 -0.15 4 20.521 6.078 66.527 52.633 34-23 0.02 5 20.556 6.035 66.527 52.632 34.25 -0.05 From the slight deviation of the calculated angle of rotation from the observed it follows that cane-sugar and grape-sugar do not sensibly affect each other in their rotating power. An agreement equally close is found w r ith aqueous solutions of mixtures of cane-sugar and raffinose (meletriose). The specific rotation of each, and especially of the latter, is but slightly dependent on the amount of water, and we can take as constants for : Cane-sugar [] D = -f- 66.5 Raffinose [#] n = -f 104.5 In such cases we can employ in the above mixture formula the concentrations c { and c. 2 in place of the weight per cents. , A^ and A^ that is, we can consider the number of grams of each substance dissolved in 100 cc. of solution, from which, and am = I OO Experiments by Creydt 1 have given the following results : 1 Zeit. Ver. fur Riibenzucker-Ind., 1887. p. 153. PRESENCE OF INACTIVE BODIES 243 TOO cc. solution contains. Angle of rotation [arl m for tube length 7=2 dm. Specific rotation. M- Cane- sugar. Raf- finose. Observed for D. Calculated Calc. Obs. Observed for D. Calculated. Calc. Obs. i6g 4g -f 29.61 + 29.64 + 0.03 + 74-ot -f 74-10 -ho.oS 17" 3" 28.92 28.88 0.04 7229 72.20 0.09 18" 2 " 28.11 28.12 + O.OI 70.28 70.30 -+- 0.02 19" I" 27-37 27.37 o.oo 68.43 68.43 0.00 Somewhat greater differences between observation and calculation were obtained by Hammerschmidt 1 with mixtures of ^/-camphor and /-santonin dissolved in chloroform. 70. Addition of Inactive Bodies to Solutions of Active Substances. According to the nature of the two substances mixed, the increase or decrease in the rotating power noted depends on a change in the chemical equilibrium, the degree of dissociation, or on the formation of new compounds. Most of the investi- gations carried out in this field deal with tartaric acid and malic acid, or with different sugars. Among these the following may be considered. A . Tartaric Acid and Malic Acid. a. Influence of Alkali Salts on the Rotation of Tartrates. A series of investigations carried out by lyong 2 relates to potas- sium sodium tartrate, KNaC 4 H 4 O 6 .4H 2 O, the specific rotation of which changes but little within the limits, c = 5 to 45, and which may be given as [] = 22.10. 20 grams of Rochelle salt with 5, 10, 15, or 20 grams of different alkali salts were dissolved to make 100 cc. of solution and the variations, J, from the value 22.10 were determined. These were found to be partly positive and partly negative, and increased with increased amounts of the alkali salts. In the following table the results are given which were found with 5 and 20 grams of the salts (or with other amounts designated in parentheses) : 1 Hammerschmidt : Loc. cit., p. 22. 2 Long : Am. ]. Sci. Arts, [3], 36, 351 (1888). 244 SPECIFIC ROTATION Increase in 22. 10 by A. Decrease in 22.10 by A. Given by Amount of salt. Given by Amount of salt. 5 grams. 20 grams. 5 grams. 20 grams. KC1 . A 0.62 0.62 0.19 0.36 0.50 0.38 0.42 0.47 0.48 0-37 A 1-33 1. 01 0.85 1.37 0.63 (10) o.73 1.02 1. 00 0.63 (I 5 ) 0.49 (10) NaCl A 0.30 0.21 0.38 0.43 0.19 0.20 0.24 0.15 A 2.35 1. 00 1.03 1.60 0.52 1.78 1.19 0.98 0.28(10) 1.41(10) TTTJr NaBr KI NaNOo . KNO Na SO . Kqr Na^HPO.-f I2aq NaH 2 PO 2 + aq . . Na 2 S 2 3 + 5aq.. NaC 2 H 3 2 -f 3 aq Na 2 B 4 O 7 + 10 aq Na. WO J OU 4 KSCN ~KC H O K,C 2 4 +aq.... NH Cl jjjj g r LiCl Tl SO . 1.67(6.75) 3-43 NH 4 SCN (NH 4 ) 2 C 2 4 + aq 0.41 The specific rotation of potassium sodium tartrate is accord- ingly increased, in the concentrations employed, by addition of potassium or ammonium salts, while, on the other hand, sodium salts, lithium chloride, and thallium sulphate cause a decrease. The reason for these opposite actions is not clear, and no investigations have been made to show whether or not they hold good in dilute solutions. The strongest effect is found with thallium sulphate. Neutral potassium tartrate also, as shown by Schiitt, 1 exhibits a slight increase in specific rotation by addition of potassium chloride, and a decrease with sodium chloride. The following mixtures were made and the polarization, p, found in a 2-dm. tube with a half-shadow instrument having the Ventzke sugar scale. From these values the specific rotations \a\n were calculated: 2 1 Schiitt: Ber. d. chem. Ges., 31, 2586. 2 On the assumption that 1 Ventzke (ray j) = 0.346 angular degree (ray /?). PRESENCE OF INACTIVE BODIES 245 In 100 cc. solution. /*. Diff. []*. Diff. 40 gm. 40 " tartrate 8 gm. i < KC1... 66.8 65 8 1.0 28.89 28 A.6 0-43 40 * -f- 8 gm. > 62.8 3-o 27.16 1.30 30 gm. 10 " tartrate -f 14 gm. i < KC1.. 49.9 ,.0 - 1.2 28.78 oft 08 0.70 o u 30 " -|- 14 gm. NaCL. 40.7 44.7 4.0 25.78 2.30 20 gm. 20 " tartrate 22 gm. KC1-. 33.3 ii 8 1.5 28.80 27 ^1 1.29 20 " + 22gm. NaCL. 51.0 27.5 4.3 */-j l 23-79 3-72 10 gm. IO ' ' tartrate -f- 25 gm. KC1.. 16.7 1. 1 28.89 1.90 10 " " + 25gm. NaCL. 13.2 2.4 o.yy 22.84 4.15 The effect varies with the proportions in which the substances are mixed. On this difference in behavior of potassium chloride and sodium chloride Schiitt has based a method for the quantitative analysis of a mixture of the two salts. See PartV. A slight decrease in the rotation of sodium tartrate by addition of sodium nitrate was observed by Th. Thomsen. 1 The specific rotation of tartar emetic (K.SbO.C 4 H 4 O 6 iH 2 O) for c = 5, [a] 20 :=-|- 141.27, was found by Long 8 to be diminished by addition of potassium, sodium and ammonium salts, and in greater degree, the more of the salts are present. The decrease by KC1, KBr, KNO 3 , NaCl, NaNO 3 , NH 4 C1, and NH 4 NO 3 is slight, while for sodium acetate, sodium phosphate, and sodium carbonate it is considerable, when these salts are added in amount insufficient to produce a precipitate. Thus, 10 grams of sodium acetate reduce the specific rotation given above to 123.59, and small amounts of sodium carbonate to 55.8 even. According to Long, the action of these salts depends on this, that the antimonyl-potassium tartrate is partly decomposed into alkali tartrates and compounds contain- ing SbO and K with acetic, phosphoric, and carbonic acids. 1 Thomsen : J. prakt. Chem., [2], 34, 83. 2 Long : Am. J. Sci. Arts, [3], 38, 264 ; 40, 275. 246 SPECIFIC ROTATION b. Influence of Boric Acid on the Rotation of Tartaric Acid. The marked increase in activity which is found here was observed first by Biot 1 in 1837 and later made the subject of lengthy investigations. 2 In order to follow the changes which occur where water and boric acid are both added, he showed first that the rotation of the tartaric acid is increased by each one of these bodies taken alone ; that is, first, by melting the tartaric acid with increasing amounts of boric acid to form glass-like masses, and secondly, by dissolving the acid in increasing quantities of water. If now an aqueous solution of tartaric acid be treated with boric acid, the observed specific rotation depends on these two conditions : i. On the relation of the tartaric acid to the boric acid. The latter increases the rotation, as borotartaric acid is formed, and this has a greater rotating power/* If the relation of tartaric acid to water is maintained constant, the increase which the specific rotation of tartaric acid experiences by addition of varying amounts of boric acid, /?, may be expressed by the formula in which the constants, A, B, C, are to be found by a series of observations. 2. On the amount of water. On the one hand, this acts to increase the rotation of the tartaric acid, but on the other, it causes hydrolytic decomposition of the borotartaric acid, and in consequence, a decrease in the rotating power. Experi- ments showed that as long as the mixture contained for i part of tartaric acid less than 0.088 part of boric acid, the rotation is increased by gradual increase in the amount of water. If the relation between tartaric acid and boric acid is exactly i : 0.088, the specific rotation remains the same for all dilutions, because then, through the increasing hydrolysis of the borotartaric acid, the activity is decreased in the same degree in which it would be increased by the influence of the i Biot : Mem. de 1'Acad., 16, 229. * Biot : Ann. chim. phys., [3], II, 82 (1844) ; 29, 341, 430 (1850) ; 59, 229 (1860). 8 Mono- or diboryltartaric acid. Not stable in solid condition. (Duve : Jahre.s- bericht, 1869, p. 540.) Dubrunfaut, Compt. rend.. 43,112, assumes the formation of the compound, H S BO 3 + 2C 4 HO 8 . PRESENCE OF INACTIVE BODIES 247 added water on the tartaric acid. Finally, if the mixture contains, for i part of tartaric acid, more than 0.088 part of boric acid, then the first action is the stronger and the rotation falls by increasing addition of water. In general, in these cases, the changes may be expressed by the formula [or] A -f Bq, in which q represents the amount of water in TOO parts by weight of the mixture. If all three com- ponents are varied, a maximum rotation is found for definite weight relations. The papers of Biot contain a large amount of numerical data, which are based on red light with a wave- length of about 635>^cf. ty With reference to the D ray, for which but few investiga- tions have been made, the extent of the influence of boric acid on the specific rotation of tartaric acid may be seen from the following numbers (Koch): 1 In 100 cc. of solution. Mol. relation. ^ ^ *^e tartaric acid. Tartaric acid. Grams. Boric acid. Grams. Tartaric acid. Boric acid. With Without boric acid. boric acid. 32.13 3.32 4 i - 29.80 + 10.86 29.11 24ol 16.63 4.01 5.06 6.87 3 2 I I I I 34.09 39.58 43-44 11.25 11.85 12.88 By determining the electrical conductivity of a large num- ber of different mixtures of tartaric acid, boric acid, and water, Magnanini 2 also was able to prove the existence of a boro- tartaric acid compound, which conducts well, and the electro- lytic dissociation of the same by increasing addition of water. On the rotation of the alkali boryl tartrates see 61. Magnanini found that other oxy-acids also experience an increase in conductivity by addition of boric acid ; thus, lactic acid, glyceric acid, oxybutyric acid, and malic acid. 3 An increase in optical activity might be expected therefore with these, which, with reference to malic acid, was already pointed out by Pasteur. 4 1 P. Koch: "Einwirkung weinsaurer Verbindungen auf polarisirtes Licht." Inaug.-Diss., Tubingen 1869. - Magnanini: Zeit. phys. Chem., 6, 67; Gazz. chim. ital., 20, 4535 2 " ^ I 34- 3 Magnanini : Gazz. chim. ital., 21, II, 215. Ber. d. chem. Ges., 24, III, 894. 4 Pasteur: Ann. chim. phys., [3], 59, 243. 248 SPECIFIC ROTATION c. Action of Molybdates and Tungstates on Tartaric Acid. On this question extended investigations have been carried out by Gernez which cover the following salts : Sodium molybdate Na 2 MoO 4 + 2 Aq, Compt. rend., 104, 783. Lithium molybdate Li 2 MoO 4 , Compt. rend., 108,942. Magnesium molybdate.. MgMoO 4 , Compt. rend., 108, 942. Ammonium molybdate.. (NH 4 ) 6 Mo 7 O 24 + 4 Aq, Compt. rend., 105, 803. Potassium tungstate- . K 2 WO 4 , Compt. rend., 106, 1529. Sodium tungstate Na 2 WO 4 + 2 Aq, Compt. rend., 106, 1527. In these investigations, solutions were employed which con- tained always in 100 cc., 2.5 grams of tartaric acid and increas- ing amounts of the salts, added in molecular proportions to the tartaric acid. Gernez reports only the angles of rotation found in a tube 1.057 dm. in length. The following table contains the complete numerical data for ammonium molybdate, and in the last column, the specific rotations (calculated by Dr. Berndt), which correspond to the tartaric acid in the different solutions : AMMONIUM MOLYBDATE. Mol.of salt to i mol. of Grams of salt to 2.5 grams of ff^for i< i iaru: aiziu in Vise mol. tartaric acid. / = 1.057 dm. i_ j " o.ooo 11' + 13-2 I 0.161 i 2 39-o 2 0^22 I 41 63-6 3 0.482 2 21 88.9 4 0.644 2 57 112 6 0.965 4 5 154 8 1.288 5 3 191 12 I-93I 6 52 260 16 2-575 8 49 334 24 3.863 13 22 506 32 = '/4 5.150 17 38 667 40 6.438 19 50 750 | 42.66 = >/ 6.866 20 39* 781* 48 7.725 2O 36 780 56 9.013 20 35 779 64 = V. 10.300 19 47 749 96 15.450 17 28 66 1 128 = i 20.600 16 44 633 192 = i% 30.900 , 16 33 626 PRESENCE OF INACTIVE BODIES 249 For the other salts Dr. Berndt has calculated the following specific rotations of tartaric acid from the data of Gernez: Added salt in 1 10 mol. for each mol. of tartaric acid. Tungstate of Molybdate of Potassium. w? Sodium. Sodium. WZ Lithium. Magnesium. 14.0 14.0 13.2 14.0 14.0 V* = \ : u mol. 28.4 2 7 .6 31.4 25.3 23-4 i=V " 41.6 40.5 5I.I 38.6 33-3 2 = V 6 " 69.3 65.6 89.7 62.4 53-3 3=V " 95.4 91.6 128 87-7 72.2 4 = Vi " 119 117 167 112 91.5 5 143 141 206 137 no 6 = */, " 169 I6 4 243 162 129 7 196 185 288 186 149 8 223 207 334 209 169 9 252 228 383 235 189 10 281 247 435 255 209 n 308 264 479 277 - 229 12 = I " 327* 277* 517* 299 248 13 3 J 8 271 330 272 14 .. .. 513 358 294 IS' 270 2*1 512 383 3i8 16 .. .. .. 413 343 17 .. 438 368 18 241 222 505 462 394 21 .. 210 483 463 24 = 2 211 I 99 498 484* 523* 36-3 " -. 170 482 468 509 48 = 4 " .. 154 473 457 495 60=, 5 " " 140 455 450 478 It is seen, therefore, that the specific rotation of tartaric acid increases on addition of increasing amounts of the salts to a maximum, after which a decrease follows which is marked with the tungstates, but slight with the molybdates. These maximum points correspond to definite molecular proportions between the tartaric acid and the salts, and in fact to : 3 C 4 H 6 6 i (XH 4 ) 6 Mo 7 O 24 4 aq i K 2 \VO 4 2 aq 1 Na-MoO 4 -f 2 aq 2 Li,MoO 4 2 MgMoO 4 The remarkably great increase in the specific rotation of tartaric acid, which, for example, with ammonium molybdate 250 SPECIFIC ROTATION reaches a value sixty times the original, may be, without doubt, ascribed to the formation of complex acids. Rosenheim 1 has already shown that tungstic, molybdic, and vanadic acids form such compounds with oxalic acid. d. Action of Molyb dates and Tungstates on Ordinary Malic Acid. Investigations on this point are also due to Gernez. They deal first with the behavior of : Ammonium molybdate. (NH 4 ) 6 Mo 7 O 24 -f 4Aq, Compt. rend., 109, 151. Sodium molybdate NajMoO^ -f- 2 Aq, Compt. rend., 109, 769. In each case i . 1 1 66 grams of malic acid were dissolved with increasing amounts of the salts, P, to make 12 cc., and the angle of rotation, a Dt was found in a tube 1.057 dm. l n g at a temperature of 17. The following table does not give all the solutions tested by Gernez, but only those with the numbers added ; in the fourth column the specific rotations of the malic acid, calculated by Dr. Berndt, are given : Ammonium molybdate. Sodium molybdate. No. of solution. Grams salt. P. 4 M* z No. of solution. Grams salt. P. <* M3. z. I 0.000 - 0.20 C - 2.0 I 0.000 0.20 2.0 2 0.013 - 0.40 4.1 3 0.084 1. 12 - 11.4 4 0.054 - 0.97 9-9 5 0.336 - 3-72 - 37-8 5 0.107 - 1.70 - 17.3 6 0.504 ' 5.48 - 55.7 6 0.191 - 2.75 - 28.0 7 0.672 - 7-25 - 73-7 8 0.282 - 3.82 - 38.8 9 i.ooS 6 - 9-07 - 92.1 M 10 0.429 - 4-95 50-3 10 1.176 5.20 - 52-9 1 H 0.572" - 5-32 - 54-i M, ii 1-344 1.52 15.5 16 0.644 - 4-93 - 50-1 12 I.5I3 -r 3.02 - 30.7 J?i 18 0.736 - 4.17 - 42.4 14 1.848 i- 11-03 4~ 112. 2 20 0.792 - 340 34-6 15 2.0I7 7 f 14.02 f 142.5 M. L 22 0.936 I.OO 10.2 17 2.353 + 8.68 -f 88.2 23 0.966 - 0.42 4.3 D 19 2.62 26.6 24 i .030 3 -f 0.83 + 8.4 J\. 20 2*857 4 0.32 + 3-3 26 1.144 4- 3.23 32.8 21 3.025* - 0.83 8.4 R-i 28 29 1.288 1-395 - 7.20 + 73.2 f 10.35 f 105.3 24 26 3.529 9 3-865 - i-55 I.OO - 15-8 10.2 M, 32 1.717* 4- 20.92 4- 212.7 27 4-033 1 " 0.50 5-1 36 2.146 f 36.22 f 368.2 28 4.201 o.oo 0.0 jf 40 2-S75 5 3.863 r 52.47 4- 72.00 f 533-5 - 731-9 30 4.538 5.042 4- 0.87 4- 2.27 4- 8.8 19-3 46 5.!5o fl r 72-80 f 74o.i M, 36 5.546 3-95 - 40.2 4 8 6.008 r 72.33 r 735.3 39 - 7.17 72.9 49 J 6.438 4- 7 2 - 00 + 731.9 41 8.067" + 10.25 f 104.2 Rosenheim : Ztschr. anorg. Chem., 4, 352 ; Her. d. chem. Ges.. 26, II, 1191. Equals Vi8 niol. Equals Vs mol. 9 Equals 1.75 mols. Equals '/io mol. 7 Equals i mol. 10 Equals 2 mols. Equals V mol. 8 Equals 1.5 mols. " Equals 4 mols. Equals V mol. PRESENCE OF INACTIVE BODIES 251 From the above, it is evident that the relations are much more complicated than with tartaric acid, inasmuch as shown by the letters under Z, not only are there points of maximum rotation ( J/) but points of change in the direction of rotation or reversal R ; increasing amounts of ammonium molybdate cause at first an increase in the original levorotation which grows to a maximum, then decreases, and finally changes to dextrorotation which increases very rapidly, but at last falls a little. With sodium molybdate there are found three inactive concentrations and three points of maximum rotation of which two are on the side of levorotation, and one on that of dextrorotation. The curve expressing these changes would have a zigzag form. Finally, it may be remarked that the characteristic points, M and R, frequently correspond to con- centrations at which there is a definite molecular relation between the malic acid and added salts. Further investigations of Gernez are concerned with the action of the following salts on the rotation of malic acid : Potassium tungstate .......... K.,\VO 4 \ ^ _,. - .. * f Compt. rend., 1 10, 1365. Sodium tungstate ............. >. a.A\ O^ i Lithium molybdate ........... Li MoO ^ Magnesium molybdate ........ MgMoO \ } C mpt rend " II0 ' ^ Sodium potassium molybdate. - K,Na 4 Mo 3 O 1 ., - 14 Aq 1 Compt. rend., Acid sodium molybdate ....... Xa 6 Mo : O 24 - 22 Aq I m, 792. Potassium phosphomolybdate K,P 2 Mo 5 6 24 Sodium phosphomolybdate ---- Na 3 P,Mo 5 6., 3 14 Aq I Compt. rend., Ammonium phosphomolybdate (XH )-P,AIo-O, * 112,226. )-P,AIo-O, The relations appearing here, are, in general, similar to those found with sodium and ammonium molybdates. B. Sugars. a. Changes in the Rotation of Cane-Sugar by Alkalies and Salts. As a great many investigations, carried out largely with reference to saccharimetry have shown, the following bodies all cause a decrease in the rotation : Hydroxides of the alkali and alkali-earth metals. Chlorides, nitrates, sulphates, carbonates, phosphates, acetates and citrates of the alkali metals. Borax, Magnesium sulphate, Chlorides of the alkali-earth metals. 252 SPECIFIC ROTATION The action of these substances increases with increased addition of the same, and so as to reduce the specific rotation of cane-sugar from -+- 66.7 to about 60. l On the behavior of the chlorides of the alkali and alkali- earth metals, the extended experiments of Farnsteiner 2 have given the following specific rotations [a]^ : i Part of cai le-sugar and Parts of water. 3 . 5 10 Without salt . 66.60 66.67 66.75 KC1 1 08-; 5-1 55 6/1 cc. NaCl 1 004 62 47 U ^OO 6* 80 LiCl i 8 l.WO 6l e.7 6-; 18 TCaOl IO7O Ac QC 66 10 66 ac. SrCl I OQ6 "o-Vj 64 12 Ac on UU 'OO 6^ 8c, OaOl o 006 62 Co 6l 42 U 0' v) M^Cl . u.yyu 12 ^O *oD 62 17 67 70 Ac 7Q VJ ! O V -' The effect of the salts becomes weaker with increased dilution. It is further evident that the decrease in rotation brought about by nearly equal weights of salts is greater, the lower the molecular weight of the chloride. Borax, like the other alkali salts, causes a decrease in the specific rotation of cane-sugar. Muntz :{ found the following values when he examined solutions obtained by mixing 10 grams of cane-sugar in 100 cc. with increasing weights of borax : Borax = o 0.5 i 2 3 4 5 7. 5 grams [a] ^=66. 7 65.9 65.0 63.5 62.5 61.6 61.1 60.5 The effect of increased addition of salts has been followed in most cases only to a low limit. More extended investigations have been carried out by Farnsteiner 4 with reference to cal- 1 A complete compilation of observations is found in I,ippmann's "Chemie der Zuckerarten." J Farnsteiner: Her. d. chem. Ges., 33, II, 3570. 8 Miintz : Ztschr. Riibenzuker-Ind., 26, 735. * Farnsteiner : Ber. d. chem. Ges., 23, 3572. PRESENCE OF INACTIVE BODIES 253 cium chloride, and these have shown that following the first observed decrease in the rotation, an increase later appears, when the salt addition passes a certain limit. To a solution of i part of sugar in 8.643 parts of water, the following amounts of calcium chloride were added : No. of the sol. i 2 3 4 5 CaCl . O O55 2 751 2 Qo8 parts t//| 17-5 (L _ . U -VOO f r . T .yiy ( __ '/DO f\1 At <*\D OO.74 05.41 64.50 ^3-5 03.41 No. of the sol. 6 7 8 9 10 CaCl 5 676 5087 Darts \ct\ I7 5 . .uq.u f.~ /J-J-jf 1 95 f- , _ 35 Ar A A J.U/U /:/; , c V/ r ai /:- CQ L" J D - 3- 2 3 6 345 05.06 o-35 The minimum (*) of rotation is found with solution No. 6. In solution No. 10 the sugar shows a stronger specific rotation than it does without addition of salt. It has been observed with ammonia also that in its concen- trated solutions (16 to 24 per cent. NH 3 ) it occasions an increase in the rotation of sugar, while in dilute solution it produces a very small, or possibly even no, decrease (Ost). 1 b. Dextrose and Calcium Chloride. It has been observed here that the rotation is increased by addition of the salt. Rimbach 2 found the following values for \oi\ D : Strength of the CaClo solution Dextrose in 100 parts by weight of the whole solution. used. Per cent. 3 20 IO 5 0.00 55-0 55-4 55.8 56.1 10.02 55-6 55-5 54-8 54-9 19.94 60.7 60.5 | 60, 60.4 With the 5 and 10 per cent, dextrose solutions a slight decrease in the rotations is at first observed. c. Action of Borax on Bodies of the Mannitol Group. While, as remarked above, the effect of borax is to decrease the rota- 1 Ost : Neue Ztschr. fur Riibenzucker-Ind., 9, 41. J Rimbach: Ztschr. phys. Chem., 9, 707. 254 SPECIFIC ROTATION tion of cane-sugar, it produces an increase of rotation in the alcohols of the glucose group. The phenomenon was first observed by Vignon with mannitol, and later E. Fischer found the same behavior in the pentitols, hexitols, etc., discovered by him. Many of these bodies, which from their constitution should be active and which are not racemes, show in pure aqueous solution either no rotation or a very small one; but this may be developed by addition of borax to the solution. The extent of the rotation caused in this way may be seen from the table below, the specific rotations being calculated in most cases from the data in the original papers: Without borax . Mi With borax. In 100 cc. solution. Observer. Sugar. Borax. Ma 0[-0.2 5 ]> < 1 0[-2.0 4 P + 2.3 -I.22 3 + 2.0 O 10.38 10 3 3 9.06 9.06 9.06 10 8 10.53 10 7-3 4 12.8 7-4 7-4 7-3 7-3 7-3 7-3 7-3 10 10 400 Fischer 5 Vignon 6 Fischer 7 14 v. Lippmann 8 ( Fischer and \ " Stahel 9 Fischer 10 f Fischer and X Passmore 11 Fischer 12 t< 13 .0 + 22. 5 f 28.3 -28.3 4- 1.5 -f 1.4 M - 5.5 + 4-8 -f- 6.0 + 2.6 d- " /. //-Sorhitol d- " /. " B O H (van't Hoff). 6 C , 254 Schmdger , < , 9IS iann Her <1 cliein <,, . M Mei-1: J. prmkt. Chem.. fa]. J 8 , * Pircun and Tollt o a S7 ,, SUGARS SHOWING MULTIROTATION 259 of inactivity, an increase in the right-hand rotation, while the change of y into ft is accompanied by a decrease in rotation. The preparation of the three modifications of the above sugars is given in 72. The rapidity with which a labile modification of a sugar changes into the stable form, increases in marked degree with rise in temperature. At the ordinary temperature a period of from six to twenty-four hours is usually required, but by boil- ing, the transformation is completed in a few minutes. This behavior was observed by Dubrunfaut with grape-sugar and with milk-sugar. After solution of a labile sugar form in water the change begins immediately and it is not, therefore, possible to estab- lish the beginning rotation with accuracy. Several minutes must elapse before the observation of the liquid in the polari- scope can commence, and with the greater rotating or-forms a value somewhat too low and with the lower rotating y-forms a value somewhat too high is always found. An accurate result is obtained only for the end product of the transformation, the stable /?- modification. 72. Resume of the Sugars Showing Multirotation. The follow- ing observations were all carried out by allowing the transfor- mation to proceed at the ordinary temperature (usually 20). First, in the shortest possible time after preparation of the solutions, the beginning rotation was determined and later the constant end rotation, the time required to complete the change being noted. The specific rotations given correspond to [<*] />, and the concentration, c, expresses the number of grams of sugar in 100 cc. of solution. 1 i. l-Arabinose, C.H 10 O 5 (two modifications) a- Modification. Ordinary crystallized arabinose. ft- Modification. This is obtained by precipitating a hot solution of the tf-form in an equal weight of water, by addition of 20 parts of absolute alcohol. In the crystallized condition it is not stable and reverts gradually to the <*-form. J 1 Many of the data based on density and percentage composition have been reduced to this form to correspond. - Tanret: Bull. Soc. Chim., [3], 15, 201. 260 SPECIFIC ROTATION Observed for c 9.73, t = 20 a. First rotation after 6.5 minutes 4-156.7 1 Decrease ft. Final rotation after 1 .5 hours + 104.6 l / 5 2 - l0 ft. (c= 13.8) + 105. (c - 2.5) + 104 (Tanret). 2. l-Xylose, C 5 H 10 O 5 (two modifications) a- Modification, ordinary xylose. ft- Modification, not yet obtained pure, but mixed with ; reverts to a in solid condition. Observation i. c = 10.235, * 20 a. First rotation after 5 minutes - 85.9 ) Decrease ft. Final rotation after 5 hours 18.6 - ) 6 7-3 Observation 2. c - 1 1.07, / - 20 a. First rotation after 4.5 minutes -J 78.6 \ Decrease ft. Final rotation after 2.5 hours I9.2 :! ) 59-4 3. d-Glucose, C,.H, 2 O 6 (three modifications) a- Modification. Ordinary form crystallized from water at mean room temperature as hydrate, C^H^O,. + H 2 O, or as anhydrous crystallized grape-sugar from alcohol or from water at 30 to 35. Immediately after solution, and referred to the anhydride, [<*]/>> -f- 105, then by change into /?, decreases to 52.5. ft- Modification. This remains usually as an amorphous hygroscopic mass by evaporation of dissolved - glucose. It may be obtained in crystalline condition by evaporating on the water-bath, and stirring constantly, dissolving the residue dried at 98 in half its weight of cold water and adding a large amount of alcohol cooled to o ; or it may be made by melt- ing ar-glucose, cooling to 100 and adding some crystals of /? glucose. It dissolves at 19 in half its weight of water and shows the con-tant rotation of -f- 52.5 (referred to C 6 H I2 O 6 ). On evaporation of its aqueous solution it yields -glucose, which, however, is only formed after, crystallization, as the mother-liquor slmw^ alway^ the rotation of p.* The y-Modification is formed when a concentrated solution ircusand Tollenn: Ann. Chcm (i,irt>ig), 357, 174. * Wheeler and Tollcns: Ann. Chcm. (I.iebin), 254,311. Parcus and Tollen* Ann Chnn (l.icbiK), 257, 175. According to Lobry de Bruyti ;m.i \lh,- r .|.i van l-:k-n>.trin HIT. Decrease Final rotation after 7 hours ................. 51. 4 01 / 43- x 8. d-Galadose, C 6 H, 2 O 6 (three modifications) a- Modification. Ordinary crystallized galactose. * ft- Modification. This is obtained by dissolving the a- form in i^ to 2 parts of boiling water and pouring the solution into eight times the volume of alcohol, stirring mean- while, which produces a crystalline precipitate. The conver- sion of a into ft is complete in strong solution only. It is sol- uble in 1.57 parts of water at 22, and reverts into the a- form on crystallization from water. The y- Modification is said to be produced when a solution of 12 grams of or-galactose in 30 grams of water is treated with 0.03 gram of sodium phosphate and a drop of dilute sulphuric acid, warmed on the water-bath a few minutes, then cooled, and gradually mixed with 240 cc. of absolute alcohol. On repeat- ing the process several times the rotation of the product, prob- ably still impure, could be reduced to 52.25. In aqueous solution at the ordinary temperature it is converted into ft inside of two hours, while for conversion into ct three times as long is required.' Change of a into ft. (c 10) a First rotation after 7 minutes .............. -j 1 1 7.4 ) Decrease ft Final rotation after 6 hours ................. + 80.3' / 37- lQ o First rotation (Tanret) ..................... l 35- \ Decrease ft Final rotation " .............. 85.3+ 81.6 53-4 Change of 7 into ft. ', First rotation after 5 minutes (Tanret) ...... 5 2 -3 \ Increase ft Final rotation (Tanret) .............. ...... 81.6 J 2 9-3 9. d-Galactoseoxime , C 6 H 12 O 3 .NOH (two modifications). 5-217) First rotation after 10 minutes ............... 20.6 } Decrease Final rotation after 20 hours ................. 14. 5 4 1 E. Fischer : Ber. d. chem. C.es., 33, 2619. - Tanret : Bull. Soc. Chim., [3], 15, 195. * Parcus and Tollens: Ann. Chem. (Liebig), 257, 169. 4 Jacobi: Ber. d. chem. Ges., 24, 698. SrciARS SHOWING MULTIROTATIOX 263 10. d-Mannoseoxime, C 6 H lvi O..NOH (two modifications) (c = 4.875) First rotation after io minutes 7.5 ) Decrease Final rotation after 6 hours - 3.2' / 4-3 11. d-Fructose, Levulose, C 6 H,.,O 6 C 10, / 20. First rotation after 6 minutes 104.0 \ Decrease Final rotation after 20 to 25 minutes - 92. 3 2 J Il -7 Final rotation 90.2* Final rotation - - 90. 7* 12. Rhamnose, isodulcite, C 6 H 12 O 5 ; hydrate, C^H^O^ (three modifications) a- Modification. Ordinary rhamnose, crystallized from water. As hydrate dissolved in water it shows a beginning rotation of - 5 to 7 (referred to C 6 H,.,O 3 ), then, by conversion into the /3-form, a decrease and afterwards a change to increasing right rotation which reaches about -(-10. ft- Modification. For the preparation of this i part of the a -form is dissolved in ^ part of boiling water and then mixed with 5 parts of absolute alcohol and 5 parts of ether. The first crystalline precipitate which separates is ^-rhamnose; this is removed and a new portion of ether is added, which throws down the /3-form, with which, however, a small amount of the y-iorm with lower rotation may be mixed. The (3- form crystallizes with J molecule of water in fine needles. It attracts moisture from the air and reverts into the tf-form. In aqueous solution it shows the constant right rotation of -(-9.1 to 10.1 or even 12.7, referred to C B H 12 O 5 . :> y- Modification. This is produced from the /^-rhamnose when it is dried in a desiccator, and then heated for several hours to 90. On crystallizing, it reverts to the tf-form. It begins with a rotation of -f 22.8, which gradually sinks to about -j- 10 through conversion into the y#-form. 6 1 Jacobi: Ber. d chem. Ges., 24, 699. Parcus and Tollens: Ann. Chem. (Liebig), 257, 166. Jungfleisch and Grimbert: Compt. rend., 107, 390. Honig and Jesser: Wien. Akad., Ber., 97, II, 547. Tanret: Bull. Soc. Chim.. (3), 15, 202. Tan ret : I.oc. cit. 264 SPECIFIC ROTATION Change of a into /3. Hydrate dissolved in water. Calculated for C 6 Hi 2 O 5 . 1. c io, / 20. a First rotation after 5.5 minutes 5 -> Decrease and } Final rotation after 66 minutes -f 9-4 cl J increase 14.4 2. c ----- 9.08, ordinary temperature. a First rotation after i minute 5.6 "i Decrease and ft Final rotation after i hour -f 9-2"' / increase 14.8 3. c 5 and io, / 13.5. a First rotation after 10.5 minutes 7.1 | Decrease and /9 Final rotation after i hour -f 9.I 03 J increase 16.2 Rhamnose anhydride dissolved in water (c= 9.5) shows at once the constant rotation -f- 8.7 (/?-form). 4 Change of y into /3. 7 First rotation 22. 8 C ) Decrease /3 Final rotation after 1.5 hours 10.1 With alcohol as solvent different rotations are found. Jacobi (loc. tit.) gives the following observations: 1. Hydrate, dissolved in alcohol, C 7.67. First rotation, after 15 minutes (for C 6 H 12 O 5 ) 12.6 ^ Decrease Final rotation after 16 hours " " - 10.0 / 2.6 2. Anhydride dissolved in alcohol, c 7.5. First rotation after 5 minutes (for C 6 H 12 O 5 ) . . ^ 3.4 1 Decrease and Final rotation after 24 hours " " . 9.0 /increase 12.4 For the explanation of the phenomenon that the hydrate and anhydride of rhamnose show a rotation in alcohol opposed to that in water see 64. 13. Rhamnoseoxime , C B H,.,O 4 .NOH (two modifications) c 9.863, ordinary temperature. First rotation after io minutes f 6. i ) i ncrea se Final rotation after 20 hours -f 13.6 5 / 7.5 14- Fucose, C 6 H 12 O 5 (two modifications) 6.916, / 20. I ; irst rotation after ir minutes -111.8 ) Decrease Final rotation after 2 hours - 77.0* f 34.8 1 Schnelle and Tollens : Ann. Chem. (I^iebij?), 371, 63. - Jacobi : Ann. Chem. (IJebig), ajj, 176. Tan ret : I.c. at. 4 Jacobi : Ann. Cht-in. (Ut-bix), 7, 177. ' Jacobi: Her. d. chrm < .<- 24, 698. (.iinther and Tollens: Ann. Chem (I.iebijf), 371,90. SUGARS SHOWING MUI/TIROTATION 265 15. RhamnohexosC) C.H,^ (two modifications) t 10, / 2 First rotation after \ hour ................. 82.9 ^ Decrease Final rotation after 12 hours ................ 61.4 ! j' 21.5 1 6. d-Mannoheptose, C T H 14 O. (two modifications) c= 10. 14 ordinary temperature. First rotation after lominutes .............. -{- 85. i ) Decrease Final rotation after 24 hours ................ -\ 68.6 - i 16.5 17. ct-Glucoheptose, C.H 14 O. (two modifications) c First rotation after 15 minutes .............. 25.0- ) Decrease Final rotation after some hours ............. I 9-7 3 f 5-3 1 8. a-Glucoctose, C s H 16 O b -j- 2 H 2 O (two modifications) c = 6.637, dissolved at 50 and cooled. First rotation .............................. 6 1 .4- ) Decrease Final rotation after 6 hours ................. 43 9 4 j 17.5 19. Milk-sugar, C^H^O,, (three modifications) a- Modification. Ordinary crystallized milk-sugar, C 12 H 22 O n -f- H.,O. This exhibits a beginning rotation of -j- 87 (anhydrous) and in aqueous solution is changed into the /3-form. ' ft- Modification. This may be obtained from a and /?, and is secured in crystalline form, C 12 H 22 O n -f 1/2 H 2 O, by adding three or four volumes of absolute alcohol to the warm concen- trated solution of either of these. 6 The ^-form is obtained by rapidly evaporating a solution of two or three grams of tf-milk-sugar in 10 cc. of water to dry- ness in a platinum-dish on a water-bath, and drying the resi- due at 98 to complete loss of the water of crystallization. 7 It may also be obtained by rapidly boiling down a solution of ordinary milk-sugar in a metallic vessel, when suddenly the whole liquid solidifies to a mass of small porous anhydrous crystals/ Pure anhydrous crystals may be secured by repeated 1 Fischer and Piloty: Ber. d. chem. Ges., 33, 3102. - Fischer and Passmore: Ber. d. chem. Ges., 23, 2230. 3 Fischer: Ann. Chem. (Liebig), 379, 75. 4 Fischer: Ann. Chem. (Liebig), 270, 97. * Erdmann and Schmoger : Ber. d. chem. Ges., 13, 1922. fi Tanret: Bull. Soc.Chim., (3), 13, 625. Schmoger : Ber. d. chem. Ges., 13, 1915- - Erdmann : Ber. d. chem. Ges., 13, 2180. 266 SPECIFIC ROTATION crystallization from anhydrous alcohol (Tanret). The first rotation is +36.2, which increases to that of the /f-form. The three forms show the same freezing-point depression. If ordinary powdered milk-sugar is dehydrated at 130, the residue shows the rotation phenomena of the or-form. Change of a into ft I- c '---'- 7-5 (hydrate), / = 20. Calculated for a First rotation after 4 minutes .. +84.0 88.4 \ Decrease Final rotation after 6 hours ---- - 52.5 55-3' * 33- lD 2. ^4.841 (hydrate). / = 20 a First rotation after 8 minutes. 4 82.9 -|- 87.3^ Decrease /3 Final rotation after 24 hours.. -\ 52.5 55-3" ' 32.0 Change of y into /3. c = 7.07 ; 7.72 ; ordinary temperature. y First rotation ................. - 34.4 -f 3 6 - 2 1 Increase /3 Final rotation after 24 hours ... 4 52.45 -j- 55.2' ) 19.0 20. Maltose, C 12 H 2a O u (two modifications) ; hydrate, ' C lr H B 0,, + H 2 1. c -- 14 to 19 (anhydride) / = 15. forCkAfti- First rotation after 4 minutes... -f 122.0 to 124.8 \ Increase Final rotation after 12 hours ......... -f 138. 3* >' 13.5 2. c = 9.2 to 9.8 (anhydride), / ~ 20. First rotation after 6 to 12 minutes 118.8 to 121.0 ^ Increase Final rotation after 5 to 9 hours 136.8 to 137. o 05 J 17.0 The absence of multirotation has been shown for inosite, sorbin, levosin, ethyl glucoside, methyl and ethyl galactosides, benzyl arabinoside, and the phenyl hydrazonesof galactose and rhamnose. 73. Rate of Change in Rotation. Urech 5 was the first to show that the Wilhelmy velocity formula for reactions of the first order, Schmoger : Ber. d. chem. Ges., 13, 1931. Parcus and Tollens . Ann. Chem. (Liebig), 357, 170. Schmoger : Ber. d. chem. C.es., 13, 1918. Meissl : J. prakt. Chem., (2) 35, 122. Parcus and Tollens : Ann. Chem. (Uiebig), 357, 172. I'rech : Ber. d. chem. Ges., 16, II, 2270 (1883) ; 17, I, 1547; 18, II, 3059. RATE OF CHANGE OF ROTATION 267 is applicable to the case of the change of rotation in milk-sugar and grape-sugar. New and careful investigations with refer- ence to dextrose have been carried out by Levy 1 and by Trey.' In consideration of the fact that the actual beginning rotation is a quantity which can never be accurately deter- mined here, the first of these observers changed the formula used in calculating the constant, C, (I) into (II) in which fifi., are the rotations corresponding to the times /,/ 2 , and (f) represents the constant final rotation. 3 In illustration the following series of observations by Levy may be given : AQUEOUS SOLUTION OF ANHYDROUS DEXTROSE OF 3.502 PER CENT STRENGTH d~ 1.0114, TEMPERATURE, 20.5 TO 20.9. Time in minutes after solution. Observed angle of rotation / = S dm. t / t . Temperature. ! Constant C. < formula II. 1 . /, = 25 ft 27.865 20.9 t, = 30 ft 27.060 5 20.9 0.00649 ', = 35 ft = 26.159 10 20.9 0.00719 A, = 40 ft 25.637 15 20.8 0.00644 >, 45 ft = 24.927 20 20.7 0.00662 /,. 50 ft = 24.369 25 20.6 0.00652 *i 55 ft = 23-895 3 20.5 0.00636 /, 60 ft = 23.166 35 20.5 0.00677 65 ft = 22.797 40 20.5 0.00656 t, = 70 ft- 22.171 45 20.5 0.00687 t-i = 75 ft = 21.837 50 20.5 0.00674 t. So ft 21.470 55 20.5 0.00671 >, = 85 ft = 21.088 60 20.5 0.00675 24 hours 16.692 Mean 0.00662 1 l,evy : Ztschr. phys. Chem., 17, 301 (1895). 2 Trey : Ibid., 18, 193.; 23, 424. 1 The same formula was derived somewhat earlier by P. Th. Miiller: Compt. rend., 118, 425. 268 SPECIFIC ROTATION From fifteen series of experiments made at a nearly constant temperature of 20, Levy found that for strengths of i to 5 per cent., the constant, C, is independent of the concentration. With increasing temperature, the value increases. It was found in the mean, that for about 20. 10 C=o.oo6io 20.25 C= 0.00637. Effect of Added Substances on the Velocity of Transformation The decrease in the rotation of aqueous dextrose solutions with the time, is sometimes hastened and sometimes retarded by the presence of other substances, such as acids, alkalies or salts, and the behavior of these bodies appears to be a cata- lytic one. Numerous investigations by Levy an.d Trey have shown the following facts : a. Bodies Which Hasten the Change i . Acids. The effect of these w r as first noticed by Krdmann 1 in the case of milk-sugar. According to Levy, 2 the velocity constant, C, (formula II) assumes the following values when dilute acids containing ^ mol. to the liter were employed as solvents : C Temperature. Relative acceleration. Water Water o 006^7 2O 2^ Hydrochloric acid 0.02300 o 0228^ ^'0 20.25 100 Trichloracetic acid . . 0.02325 o 01886 20.25 90.99 96.67 Dichloracetic acid Monochloracetic acid. 0.01670 0.01004 o 00716 20.2 20.25 2O 2 7 I -95 62.41 17.25 Propionic acid ... o 006^6 TO 8 4-/ u i f>-i 19.0 1.03 It appears, therefore, that in their accelerating behavior, the acids stand in the order of their affinity coefficients, as 1 Krdmann : Jaliresber., 1X55, j>. 671. 1 Levy : IJQC. ri/., p. 301. RATE OF CHANGE OF ROTATION 269 measured by electrical conductivity, the catalysis of methyl acetate or by their power of inverting cane-sugar. Trey 1 came to the same conclusion in experiments on the action of hydrochloric, sulphuric, oxalic, and cacodylic acids. But, on the other hand, he found a retarding effect with acetic and propionic acids. The value of the constant, C, increases as a matter of course, with the strength of the acid. For example, with T \ normal hydrochloric acid, the value is C = 0.00971. 2. Bases. These bring about an enormous acceleration in the velocity of transformation of all sugars, so that in a very few minutes, or immediately after the addition of the base, the normal end rotation is reached. This was first noticed by Urecrr in the case of milk-sugar and ammonia, and he found also that in time, the rotation sinks still lower, which may be accounted for by a chemical change in the sugar." The action of ammonia on different sugars has been thoroughly investigated by Schulze and Tollens. 4 They found that by application of o.i per cent, ammonia, after 5 to 10 minutes, even, the normal lower rotation is reached with dextrose, xylose, arabinose, galactose, rhamnose, levulose, and ordinary milk-sugar. This was found to be the case with ^-milk-sugar and maltose also, from which it is seen that not only the higher, but the lower rotation as well, is destroyed immediately. Schulze and Tollens observed, for example, the following changes: Xylose: [] D Maltose hydrate: \_ a ~\ D In water constant -f- 18.7 In water constant -\- 130 In ammo- f After 10 min. 14.8 In ammo- f After loniin. - 126.1 niaof 20.4^ After i day -f u.o niaof2O.4J After 7^ hours -j- I2 3-9 per cent, j After 3 days -f 5.7 per cent. [After i day 118.1 d 0.924 [After 5 days 5.9 In the same manner Trey 5 found in a solution, which contained in 100 cc., 2.25 grams of dextrose anhydride and 0.085 NH 3 , 1 Trey : Ztschr. phys. Chem., 18, 205 ; 22, 448. - Urech: Her. d. chem. Ges., 15, 2132 (1882). Trech : Ibid., 17, 1545. 4 Schulze and Tollens: Ann. Chem. (L,iebig), 271, 49. 5 Trey: Ztschr. phys. Chem., 22, 439. 2JO SPECIFIC ROTATION After 15 inin. [^J /> ~ "^ 5 2 - 2 (normal) After 24 hours -{- 49.6 After 65 days -f- 44.9 The alkali hydroxides have an equally strong action ; by means of ^^ normal potassium hydroxide solution the final rotation in dextrose is reached almost instantly. With greater concentration, a further progressive decrease is observed which may even pass into left rotation. Thus, Trey 1 found in a solution containing in 100 cc. 2.25 grams of dextrose, and 0.2 gram of sodium hydroxide, the following specific rotations : After 15 minutes []/> = + 52.7 (normal) " 24 hours [a]/, == -f- 36.7 " 48 " [a]/, = -f 26.0 " 34 days [a]/, = + 15.1 " 6 5 " [ a ]/> = - 0.4 It is evident that some chemical change takes place here. As Lobry de Bruyn and Alberda van Ekensteiir have observed, the reciprocal conversion of dextrose, levulose, and mannose into each other by the catalytic action of small amounts of alkali is possible. It is only with very weak bases that the fall in the multi- rotation may be quantitatively followed. This was done in a dextrose solution containing urea with which Levy 1 ' found the following values for the velocity constant, C (formula II) : ( 10 per cent. : C 0.008^ at about 20.2 Urea in solution < I 5 per cent. : C - 0.00749 at about 20.0 These numbers are but little larger than given above for pure water. j. Salts. These all, with the exception of sodium chloride, have an accelerating action on the fall of rotation in dextrose. Salts with an alkaline reaction bring about the normal end rotation within a few minutes. This is the case with sodium carbonate, with which, after long standing, no further decrease is observed ; this salt is therefore suitable for quickly destroy- ing the multirotation. Potassium cyanide sometimes produces a marked decrease below the normal end value (Trey). 4 1 Trey : Loc. cit. t p. 438. '-' Lobry de Bruyn and All>erda van Eckenstein : Her. d. chein. (ies., 28, III, ,^078; Kec. trav. chim. Pays-Bas., 14, 156, 203. * J*evy : Ztschr. phys. Chem., 17, 324. 4 Trey : Ibid., aa, 440. RATE OF CHANGE OF ROTATION 2JI Salts with a neutral reaction cause a slower decrease in the multirotation, so that the conversion may be followed and meas- ured. Levy 1 found the velocity constant C (formula II), for dextrose in the presence of the following salts : C. Temp, Pure water 0.00610 20. i Sodium sulphate, anhydrous, 10 per cent. .. 0.00844 20.0 Sodium sulphate, anhydrous, 5 per cent- . . 0.00800 20.2 Sodium acetate, 10 per cent 0.03897 20. i According to Levy, the action of these salts depend on this, that through partial hydrolysis, sodium hydroxide is formed which brings about the acceleration. This hydrolysis takes place to a greater extent with sodium acetate, than with the sulphate, and this explains its much stronger effect. Trey has demonstrated the accelerating action of the follow- ing salts on the fall of rotation in dextrose :'-' Sodium bicarbonate, Magnesium chloride, Potassium nitrate Magnesium sulphate, Potassium iodide, Aluminum chloride, Ammonium chloride, Cadmium iodide, Ammonium thiocyanate, Lead acetate, Barium chloride, Mercuric chloride. Trey found also that sodium sulphate lessens the effect of sulphuric acid. b. Bodies Which Retard the Change /. Sodium Chloride. The behavior of this salt, different from that of all others, was first noticed by Levy 3 who found the following values for the constant C (by formula II), which are smaller than that for water : Dextrose solution . C. Temp. In pure water 0.00610 20.1 In 10 per cent, sodium chloride 0.00533 20.0 In 5 per cent, sodium chloride 0.00586 20.2 Trey 4 confirmed this retarding action of common salt by the following series of observations, in which the changes in the specific rotation of dextrose w T ith the time of standing were found : 1 L,evy : Loc. cit. - Trey : Ztschr. phys. Chem., 22, 429. : I^evy : Ibid., 17, 320. 4 Trey : Ibid., 22, 429, 436. 272 SPECIFIC ROTATION In (Dextrose 2.250 9.000 cc - (xad c 0.2925 0.2925 0-585 1.170 Time, 5111111. 102.7 100.0 Time, ismin. 95-6 96.0 94.2 96.8 97-3 97-1 Time, 25inin. 87.3 90.0 86.4 89.6 89.9 89.7 Time, 35inin. 82.2 84.4 80.2 83-2 83.7 83.3 Time,45min. 78.2 80.4 74-6 77-9 78.5 78.1 Time, 55inin. 74-4 76.7 69.9 73.1 73-2 73-6 Time, 65min. 72.2 749 66.2 69.4 69.6 70.7 Time, -s: 50.7 5i.i 50.8 51.1 51-7 S2 2 In presence of common salt the decrease in the rotation follows more slowly than in pure water. It is seen also that a variation in the amount of salt is of little importance. An explanation of this peculiar behavior of" the salt is wholly lacking. 2. Alcohoh and Other Organic Substances. Levy obtained, according to formula II, lower values for the constant, C, with solutions of dextrose in aqueous ethyl alcohol than in pure water. Pure water 0.00610 1 ,,, mol. alcohol 0.00555 In i liter -' ' : , mol. alcohol 0.00521 i mol. alcohol 0.00510 Temp. 20.1 2O.O 20.0 2O. I In aqueous methyl alcohol, the retarding action of which had been already observed by Dubrunfaut, 1 Trey 2 found that the decrease in the rotation takes place somewhat more rapidly than in ethyl alcohol. Solutions of dextrose in pure methyl alcohol show like- wise, multirotation, and the decrease- in the same may be retarded by addition of other organic substances. Trey :i found the velocity constant from formula I for the following solutions : 1 Dubrunfaut: Compt. rend., 42, 228 (i '-' Trey: 7,tschr. phys. Chein., 17, 200. * Trey : Ibid,, 17, CAUSE OF THE MULTIROTATIOX OF SUGARS 273 In 100 cc. of the methyl alcohol solution. C. i-775 gram dextrose 4.95 i. 02 10 " " 0.1525 gram phenol 2.54 0.9615 " -0.1543 " naphthalene 1.77 0.8504 " --0.2725 " diphenylamine i.ii -$495 " " 0-3072 " succinamide 0.85 Finally, a retardation in the decrease in rotation in aqueous dextrose solutions has been noticed on addition of acetone and also of cane-sugar (Trey). 1 74. Cause of Multirotation of the Sugars. On this point, the following views have been advanced : 1. The change in the rotation may depend on this, that in the original freshly prepared solution, molecular aggregations 'crystal molecules) of active form may exist, which gradually break down into molecules of lower rotation (Landolt," Pribram, :1 Hammerschmidt, 4 WyroubofT) /' By cryoscopic methods, as already explained in 63, these aggregations cannot be shown, because they no longer exist in dilute solutions. 2. With dextrose, which shows multirotation as an anhy- dride as well as hydrate, the fall in rotation may be due to the taking up, or splitting off, of water. This view which was brought forward years ago by Erdmann (i and Bechamp, 7 and later abandoned by its authors has recently come to light again. K. Fischer is of the opinion that anhydrous grape-sugar dis- solves first as Q.Hj.,0^ (with high rotation) and is gradually transformed into the hydrate or heptahydric alcohol, C,.H 14 O 7 with lower rotation). According to Jacobi," this view is confirmed by the fact, that in birotating glucose anhydride solutions, the formation of the phenyl hydrazone takes place more rapidly than in the lower rotating solutions. On the Trey : Ztschr. phys. Chem., 22, 450, 456. I^andolt : " Optisches Drehungsvermogen," ist ed. : Pribram : Wien. Monatsheft, 9, 401. Hammerschmidt : Ztschr. f. Riibenzucker-Ind., 40, 939. Wyrouboff : Compt. rend., 115, 832. Krdmann : Jahresbericht, (1855), p. 672. Bechamp: Ibid., (1856), p. 639, 640; Compt. rend., 42, 640, 896. E. Fischer : Ber. d. chem. Ges , 23, 2626. Jacobi : Ann. Chem. (L,iebig), 272, 179. 18 274 SPECIFIC ROTATION other hand, Tollens 1 assumed that the anhydride on dissolving is transformed immediately into the hydrate (with high rotation), as evolution of heat follows, and that then gradually, in solution, a reformation of the anhydride (with low rotation) takes place. Such a reaction is, however, thermodynamically impossible. With dextrose hydrate, which dissolves without liberation of heat, the high rotation of the unchanged com- pound which at first appears, and the decrease in the rotation, would both have to be ascribed to the gradual formation of anhydride. Arrhenius, 1 and also Brown and Morris, :< have found that birotating and normally rotating dextrose solutions are cryo- scopically identical. But this observation does not decide the question as to the presence of hydrate or anhydride in solution, because the difference between the freezing-point depressions of the two would be so small as to fall within the limits of errors of experiment. 3. The differences in rotation of multirotating sugars may be due to the existence of different isomeric modifications, which, in solution, are gradually transformed into each other. Such an assumption was made as early as 1856 by Erdmann, 1 Dubrunfaut, 5 and also by Bechamp," and with the under- standing that of the different modifications, one is crystalline and the others amorphous. But since then, as explained in 71 and 72, Tanret has succeeded in obtaining tthe wo or three different rotating forms of several sugars in crystalline condition, and in showing the equality in their molecular weights, and also their reciprocal convertibility. There remains, therefore, only the question as to the constitution of these isomers, and on this point van't Hoff 7 called attention ( 1894) to the clue which may be found in the analogy existing between the sugars and the rotating lactone- forming acids, as gluconic acid, arabonic acid, saccharic acid, and others. With Tollens : Her. d. chem. Ges., 26, 1799. Arrhenius : Ztschr. phys. Chem., a, 500 (1888). Brown and Morris: Chem. News, 57, 196 (1888). Erdmann : Jahresbericht, (1855), 672; (1856), 639; Her. d. chem. Ges., 13,2180 (1880). Dubrunfaut : Compt. rend., 43, 739 (1856). Bichamp : Compt. rend., 43, 896 (1856) ; Bull. Soc. Chitn., (3), 9, 511 (1893). van't Hoff: " I, age rung der Atonic im Raume," 2d ed., p. in. CAUSE OF THE MULTIROTATIOX OF SUGARS 275 these latter, as will be explained in 75, the increase in rota- tion, which follows on solution in water, depends on the formation of their lactones, which possess a cyclic constitution. On the other hand, a decrease in rotation results when the lactones are changed into the acids, with destruction of the ring formation. According to this, for example, of the two modifications of xylose the higher rotating might have the formula CH.X)H CH (CH.OH) 2 CH.OH, 1 O I and the change into the lower rotating form might depend on the conversion into a body of this structure, CH 2 OH (CH.OH), CHO. Further, it was observed simultaneously by E. von Lipp- mann 1 and by Lobry de Bruyn and Alberda van Ekenstehr that in regard to the three differently rotating ^/-glucoses, the assumption of the ethylene oxide constitution of the same pro- vides for two stereoisomeric forms, and that the third or aldehyde structure may also appear. While it is true that the proof of which constitution must be ascribed to each of the three forms of glucose, etc. , is as yet a very difficult question, we may now consider this much as established, that the cause of the multirotation in the sugars is to be found in the transformations of their different isomeric modifications. 3 II. Multirotation of Ovy- Acids and Their Lactones. 75- In 1873 Wislicenus* made the observation that the rotating power of ^-lactic acid, in aqueous solution, is gradu- ally changed at the ordinary temperature, and he found the cause of this to lie in the slow hydration of anhydrous mole- cules which were contained in the solution. The same behavior was later noticed, especially by Tollens, in other oxy-acids, as saccharic acid, gluconic acid, rhamnonic acid and others, and 1 E. v. I^ippmann: Ber. d. chem. Ges., 29, 203; " Chemie der Zuckerarten," (1895) p. 130, 990, 992. '-' lyobry de Bruyn and Alberda van Ekenstein: Ber. d. chem. Ges., 28, 3081 (1895) 3 See a recent paper by Brown and Pickering (J. Chem. Soc., 71, 769) on this subject. 4 Wislicenus: Ann. Chem. (I^iebig), 167, 302. 276 SPECIFIC ROTATION also in their lactones. The latter have always a markedly greater rotation than the corresponding acids, and if the lactones are dissolved in water there is found a progressive decrease in the rotation until a constant value is finally reached. If, on the other hand, fresh solutions of the acids are made by using salts of the same and hydrochloric acid in addition, then a gradual fall of the rotation may be observed, and finally the same value is reached as is obtained by starting with the lactone. The cause of the phenomenon is found in this fact, that in aqueous solution, both the acid and its lactone suffer a recipro- cal decomposition which goes on very slowly and continues until a condition of equilibrium is reached when the liquid con- tains both substances in definite proportions. This ratio varies with the temperature, and in such a manner that with elevation of temperature there is an increase of the lactone molecules, and therefore of the rotation. On cooling, the rotation decreases to its former value. The velocity relations in the conversion of the oxy-salts into their lactones, and the reverse, have been studied especially by P. Henry 1 who made use of y-oxybutyric acid and lactone. CH.,OH ( CH, ), COOH CH, (CH.,),-CO o y-oxyvaleric acid and lactone. CH 3 CH.OH ( CH, i COOH CH,-CH (CH.), CO O - ' in which case by titrations the following relations were found : a. The transformation of the oxy-acids into lactones is hastened by addition of acids, which act catalytically. The velocity is expressed by the equation corresponding to a reaction of the first order, in which A represents the original amount of oxy-acid, x the amount of lactone formed in the time, /, and C the constant of the reaction. It was found that different acids added act 'P.Henry:/! Cliem., 10, 96 (1892). St-i- ftlio: K. Hj.lt: i:< r d. chem. Ges., 34, 1236 (1891) ; Unno Collar by* < IK-HI . 10, 130 (1*92). THE MULTIROTATIOX OF OXY-ACIDS AND LACTONES 277 with an intensity proportional to their affinity coefficients. 7-Oxybutyric acid is transformed only partially, but with 7-oxyvaleric acid the reaction is complete. b. In the case of spontaneous transformation of the oxy- acids into lactones, the relations observed correspond to the assumption that the oxy-acid still remaining, at any moment, can accelerate its own conversion ; that is, autocatalysis comes into play. The above formula no longer holds good, since in this case the velocity of the reaction is proportional to the amount of acid not dissociated, that is to C(A x}, con- sidered as an indifferent body, and also to the dissociated, cata- lytically acting acid, which at any moment is equal to y{A x), where y is a certain function which expresses the number of H-ions. The complete expression for the velocity is given then by the equation obtaining for a reaction of the second order, d.\ = Cy (A - *)', which was found to give values for C in accord with the results of experiment. For the derivation of y the original paper must be consulted. 1 c. Addition of bases to the solution of the lactone hastens the conversion of the latter into acid. When the two bodies are mixed in equivalent proportions the velocity of the reaction corresponds to the equation, It was found that the activity of the bases is proportional to the intensity of their basic character. Observations. On the multirotation of oxy- acids and lactones the most important results are, in the main, the follow- ing, and among these, those concerning saccharic acid are given first, because here the condition of equilibrium referred to above is most clearly discerned (c = concentration in 100 cc.). 2 1 See alsoH. Jahn : " Grnndriss der Electrochemie," (1895), p. 118. - In the calculation of the specific rotation of the several solutions, which contain acid and lactone at the same time, it is necessary to express the concentration in terms of either one or the other. In each case it is stated to which txxlies the numbers given refer. 2 7 8 SPECIFIC ROTATION d- Saccharic acid ',* Acid ammonium salt with HC1. c = 4.634 acid = 4.237 lactone. Saccharic acid lactone, C.H 8 O r Weighed directly. c = 10.213 lactone. After j day T V- A no JJCglillllllg . 2 days " 6 4 ' 16 d 44 n " o 32.3 30.5 26.3 24.9 24.1 23.2 23.1 22.9 22.9 22.8 22.5 Calculated for C H H K O 7 . With norisosaccharic acid, C 6 H 10 O,,, and isosaccharic acid, C S H 8 O 7 , almost the same rotation is likewise found after warm- ing the solutions (-(- 49 to -f- 52 ). 2 2. After Rhamnonic acid, c.H.,0.. Strontium salt and HC1. c. 6.204 acid. 10 18 33 2 3 5 minutes -- 7.7 C 44 n. i " 15-8 hours 27.9 da y s 29.4 29.2 Rhamnonic acid lactone Weighed directly. c = 5.680 lactone = 5.960 acid. At once - 34.3 After i day 34.3 " 2 days 33.7 3 33-7 \ hour heated .... 34.8 3 days 34.0 Calculated for C,.H,.,O. Heated \ hour at > 100 and cooled / ' After i day 30.7 44 3 days 30.1 " 5 " 30.1 Calculated for C B H 12 O. It appears, therefore, that in the condition of equilibrium the liquid contains mainly the lactone, and that probably the latter is not transformed into the acid at all. :! 1 Sohst and Tollens : Ann. Chem. (I,iebig), 345, 10, 12. 1 Tiemann : Ber. d. chem. C.es., 37, 137. hnelle and Tollens : Ann Chem. (Ijebivj), 371, 72. THE MULTIROTATIOX OF OXV-ACIDS AND LACTOXES 279 3 . d- Glucon ic a cid, ' C.H..O,. Calcium salt with oxalic acid. Mi At the beginning 1-74 After 14 to 21 days. ... 10 to 12 Heated to 100 23.4 After cooling 10 to 12 Calculated for Gluconic acid lactone, C.H..O.. c = 7.5 lactone. [L> After 10 minutes 55.9 " i day 47.6 " 3 days - 44.0 " 8 days .36.0 " ii days 33.6 C,;H,.,O 7 . " 47 days -7-18.9 Calculated for C 6 H 12 O 7 . The conversion of the lactone into the acid takes place, there- fore, very slowly and after forty- seven days is not yet complete. It may be further remarked, as with rhamnonic acid, that heating the acid leads to increased formation of lactone. 4. d-Galactonic add,' C.H.A. Calcium salt with HC1. c = 7.5? acid. []/. After 10 to 15 min. 10.6 5 hours 13.8 6 days 39.2 " 15 ^ays 45.9 Heated to 100 59.7 After 14 days 53.0 Calculated for C 6 H,,O 7 . Galactonic acid lactone, C B H, 6 . Weighed as hvdrate. C-.H^Oe + H 2 O. ' c = 6.85. M, After 10 min. 64.2 " 24 hours 63.7 " warming 63.9 2. c = 7.0 After 10 min 6.5.5 " 3 days 64.3 Calculated for QH 12 O 7 . With many bodies, the following for example, the observa- tions are still incomplete. 5 . A ra bon ic acid, ' C-.H.A, Strontium salt and HC1. c = ^.46 acid. M* After 10 min 8.50 ' ' 4 hours 30.0 ' ' 6 hours 37.0 " 20 hours 45.0 ' ' 2 days 45.9 " 2 months 48.2 Calculated for Arabonic acid lactone * C.HA- Weighed directly. c = 9.749 lactone. Beginning. . . . After 14 hours -65.9 -65.9 Calculated for C 5 H 10 6 . ( 73-9 calculated for i Schnelle and Tollens: Ann.Chem. (i.iebig), 271, 74. '-' Schnelle and Tollens: Ibid., 271, M. - Allen and Tollens: Ibid., 260, 312. 4 Hi O. t . II < ). whose rotation must be smaller than that of the acid. Among the experiments which \Vislicenus 1 has made, tin- following may be cited : 1 Allen and Tolh MS : Ann. Cht-in. (I,if1>itf), 260, JII. .m-llr and T>llfii> ; //./., 371, 66. I ischer : Ibid., 770, 4 U'i '"/., 167, 324 (1873) ; also 164, iM THE MULTIROTATION OF OXY-ACIDS AND LACTONES. 28 1 C 8 H 6 0, in 100 cc. \_(x\V Beginning 42.65 0.41 After 7 months 42.97 + 2.85 " 9 " 42.97 2.91 " dilution 15-74 M3 " 44 days 15.74 +1.84 Among bodies of Class II it is found, without exception, that the lactones possess a much stronger rotating power than their corresponding acids. In the constitution of the two groups, there is a marked difference, as we have for example : Saccharic acid. .Saccharic acid lactone. CO .OH CO . OH I I CH .OH CH . OH CH . OH ,CH CH . OH / CH . OH CH . OH \ CH . OH I \ I CO . OH \CO The higher rotation of the lactone may be attributed to the ring structure contained in it. See 84. Furthermore, it has not been shown in all cases that the decrease in rotation, which is observed with the lactones, depends on hydration. In the case of the ytf-glucohep tonic acid lactone, referred to above, Fischer was unable to discover any gradual formation of acid. Besides, the change in rotation may take place in solutions which are not aqueous. This was first observed by Colson 1 with the anhydride of diacetyl tar- taric acid, which showed at first, in acetone, a rotation of 12, and after half an hour only 10.18. ///. Multirotation of a Few Other Substances. 76- Here we have the following bodies: Formylfenchylaminc, C ]0 H 1T .XH.HCO. A solution in chloro- form showed a decrease in rotation which ceased after 12 hours.' i Colson : Bull. Soc. Chim., (3) 7, - Binz: Ztschr. phys. Chem., 12, 726. 282 SPECIFIC ROTATION /OH (i) p- Oxy-benzylidcnefcnchvlaminc, C K H 4 ^ \CH =. N.C 10 H 1T (4) With a solution in chloroform (p = 1.28, d -- 1.4905) there was found: 1 For the fresh solution [a] *; 77 After 18 hours [a-] = -f 72. Nicotine, dissolved in water. Pribram observed that a solu- tion of 20.169 per cent, strength, with a density of 1.0149, when allowed to remain in the polarization tube at the ordin- ary temperature, exhibited the following increase in rotation: []5? [] for / = 3.999 dm Fresh solution 7.188 87.81 After 12 hours 7.623 93-13 After 1 8 hours 7.903 96.55 After 48 hours 7.904 96.56 According to Pribram, this phenomenon depends on the formation of a hydrate, since nicotine and water mix with marked evolution of heat. In freshly prepared 20 per cent. solutions, the conversion does not seem to be completed at once, but to require a certain time. From the increase in the rotation it must be concluded that the hydrate is more active than the pure base. 2 G. Relations Between the Amount of Rotation and Chemical Con- stitution 77. Preliminary Remarks. A large number of investigations have been carried out with these relations in view, but up to the present time without yielding very satisfactory results. In establishing the rotating power of groups of bodies of given chemical constitution, certain regularities could be recognized, but almost always these were clouded by exceptions more or less numerous. The cause of this may rest in the uncertainty which is connected with the determination of the specific rotation of many substances. With solid bodies which can be investigated only in solutions, the rotation, as is well known, 1 Binz: IJH-. fit. 2 Pribram : Ber. d.chem. (its., 20, 1847. ROTATION AND CHEMICAL CONSTITUTION 283 is in a large measure dependent on the concentration as well as on the nature of the inactive solvent, and such bodies fre- quently cannot be brought into comparison. Investigation is largely limited, therefore, to active liquid bodies, and even here there may be sometimes doubts as to their applicability. If, in their preparation, high heat is applied, or a violent reaction ensues, a partial racemization of the product is pos- sible, and in this case too low a result for the rotation will be obtained. (See 28 and 29.) The reliability of the numerical value given for the specific rotation, if of a liquid substance, may be tested by examining several preparations made by different processes. In this direction, Purdie and Williamson 1 have carried out some experiments with esters of malic and lactic acids, which were made : 1. By action of alkyl iodides on the silver salts of the acids. 2. By esterification of a mixture of acid and alcohol by aid of hydrochloric or sulphuric acid. By using also some observations of Walker, 2 and of Anschiitz and Reitter, 3 the following results were obtained : MALIC ACID ESTERS Silver salt method. Acid method. Purdie and Williamson. Ms Purdie and Williamson. MS* Walker M? Anschiitz and Reitter. M* Methyl malate ' Ethyl malate , N-Propyl malate.. . N-Butyl malate 7.34 - 12.42 - 13-70 12.20 - 10.34 - 6.85 - 10.18 11.62 6.88 - 10.65 ii. 60 10.72 Ethyl acetyl malate Ethylbutyryl " - 23.00 22.70 - 21.58 - 22.52 - 22.22 22.60 1 Purdie and Williamson : J. Chem. Soc., 69, 818 (1896). - Walker : Ibid., 67, 914 (1895). 3 Anschiitz and Reitter: Ztschr. phys. Chem., 16, 493 (1895). 284 SPECIFIC ROTATION LACTIC ACID ESTERS Silver salt method. Acid method. Walker. Pnrdie and Williamson. Purdie and Williamson. - I3-46 ~ 49.87 - 10-33 - 49-75 + i9-4i Ethylacetyl lactate Ethvlchlor propionate / I 21.78 / The silver salt method, with the simple esters, gives, accord- ingly, somewhat higher values than the acid method. The lower rotation of the esters made by the latter method was not due to any racemization, since it was found that the acids separated from them showed no lower rotation than those originally employed. Besides, in the preparation of the above esters no high temperatures were applied (not above 100); in other cases possibly greater differences will be noticed. As a further cause of the uncertainty in the specific rotation of liquid bodies, possible changes in the products, through polymerization, have been suggested. But as has been shown, however, in 62, the effect of this reaction on the rotation has not been demonstrated with complete certainty. This has been confirmed lately by an observation made by Walden 1 on the diamyl ester of itaconic acid, as it was found that this sub- stance, in spite of gradually increasing polymerization, still showed no indication of any, or at most of only an unimpor- tant change in the optical activity. In different conditions this compound gave the following rotations for a layer i dm. in length: Freshly prepared, mobile liquid a, t 4.80 After 2 months, thick liquid, stringy 4.75 Completely hardened, colorless #lass 4-75 To test the relation between rotating power and chemical constitution, the esters of active amyl alcohol have been fre- quently employed. These have all been made from commercial left rotating amyl alcohol of different degrees of activity 1 Walden: Ztschr. phys. Chem., ao, 383 (1*96). ROTATION AND CHEMICAL CONSTITUTION 285 (for example, []/, = - 4.5, Guye; -- 4.78, Walden). Such preparations always contain a certain, but not deterrmn- able, proportion of inactive isomers, and are comparable among themselves to a limited extent only when secured from the same crude product. The different isomers of the crude amyl alcohol appear to be attacked by chemical reagents to not essentially different degrees, since Guye and Chavanne 1 found that the alcohols recovered by saponification of several esters, possessed a rotating power scarcely changed from that of the original. In the following comparisons w r hen bodies of different com- positions are dealt with, the molecular rotation, [J/] = M r^l, is alwavs employed, while for isomers, the value \oi\ 100 L is sufficient. ' /. horn eric Bodies An idea of the relations obtaining here, may be obtained from the following observations. 78. a. Metamerism. Structural Isomerism i . Trans location of the Active Radicals Aniylacetic acid [a~\ ,, = -f- 8.53 ) Walden : Ztschr. phys. Amylacetate 2.50 ) Chem., 15, 638. Diamylacetic acid - 18.27 ) Walden : Ztschr. phys. Amylamyloacetate 7.01 j Chem., 15,638. , \ Guye and Chavanne : Methvl valerate 16.83 - a [ Arch. phvs. nat., [4], Amyl formate 2.01 | ' J j 54- Iii these cases very marked differences appear. 2. Alcohol Radical Active. Isomerism in the Inactive Acid Radical. Amyl normal butyrate [^] D = 2 -97 "> Walden : Ztschr. phys. Amyl isobutyrate 2.83 J Chem., 15, 638. Amyl normal brombntyrate- 2.27) Walden : Ztschr. phys. Amyl isobrombutyrate 2.53 J Chem., 15, 638. The differences in rotation are small. j. Alcohol Radical Inactive. Isomerism in the Active Acid Radical. Methyl normal butyryl malate [a] /; - 22.44 \ Walden : Ztschr. phys. Methylisobutyryl malate 22.36 J Chem., 17, 245. 1 Guye and Chavanne: Bull. Soc. Chim., [3], 15, 275 (1896). 286 SPECIFIC ROTATION Ethyl normal butyryl malate. . 22.22 ) Walden : Ztschr. phys. Ethylisobutyryl malate ....... 21.99 Chem., 17, 245. Kthyl-a-brom 'normal bntyryl , Wa]( , en . ^^ h malate .................... 24. 76 v Ethyl-a-bromisobutyryl malate - 22 -57 ' The differences are likewise very small. 4. ^4V Radical Active. I so men sm in the Inactive Alcohol Radical Normal butvl valerate ........ fa],, = + 10.60 ) Guye and Chavanne: ' + 10.60 ) --10.48 Isobutvl valerate 1454- ^ Frankland and Mac Normal propyl glvcerate ..... 12.94 T 4 \ Gregor: J. Chem. Isopropyl glycerate ......... - 1 1.82 j ^ ^ ^ Normal butyl glycerate ...... 11.02^ Frankland and Mac Isobutyl glycerate .......... -14.231 Gregor: J. Chem. Secondary butyl glycerate... 10.58) Soc., 63, 524. Normal propyl malate ....... - 1 1 .62 \ Walden : Ztschr. phys. Isopropyl malate ............ 10.41 ^ Chem., 17, 245. Normal propyl tartrate ...... -f- 29.1 1 ( Freundler: Ann. chiin. Isopropyl tartrate ........... -f 34.83 ' phys., [7], 3, 433. Normal butyl diacetyl tartrate 8.0 ^ Freundler: Ann. chim. Isobutyl diacetyl tartrate ---- -f- 17.0 ' phys., [7], 3, 433. Normal butyl dipropionvl tar- ^ ^ Freundler: Ann. chim. trate ...................... -f 6.9 V ,., Isobutyl dipropionyl tartrate -f 11.4 J P yS>> "- 7 ^' 3 ' 433 ' Normal butyldibutvryl tar- Freundler: Ann. cliim. trate .................... + 6.0 \ Isobutyl dibutyryl tartrate... .-f 8.5 J P yS " L7J ' 3 ' 433- The iso compounds show sometimes a higher, sometimes a lower, rotation than the normal. b. Position Isomcrism in Benzene Derivatives 79- A number of investigations have been carried out to determine the influence of the o, m or p positions of an active and inactive group, or of two active groups in disubstituted benzene derivatives. i. H. Goldschmidt and Freund 1 have determined the specific rotation of the following solid bodies from chloroform solu- tions, the per cent, strength, p, within each group being nearly the same. In some groups the compounds, C 6 H 3 .R, were investigated 1 /tschr. phys. Chem., 14, 394 (1894). ROTATION AND CHEMICAL CONSTITUTION 287 to determine the effect of the addition of CH 3 by comparison with C 6 H 4 < \CH, The differences show the increase or decrease in molecular rotation from one member to another. /-Amylphenyl carbaminate - 4. 19 - 1 - 8.7 /-Amyl-0-tolyl carbaminate 2.66 5.9 m- " r ^Jtlx >.C 5 H n J L P 5-3 . 4 ^NH.COO.C 5 H U J + 4-47 + 9-9 f /-Menthylphenyl carbaminate 77.2 212.3 II. I /-Menthyl-0-tolyl carbaminate 65.9 190.4 ~~ p= 5 .6 - " m- r CH 3 H - 71.4 -206.4 + ' ' p- L " XH.COO.C 10 Hj- 7 2. 3 -208.9 Carbanilido-H- r CH, -1-298-88.83^ k " p- L 6 * NH.CO.NO : C ]n H u J - 30.8 +91.6 Benzoyl-*/-carvoxime 26.6 - 71.7 _ ^-Toluyl-^-carvoxime 27. i 4- 76.6 ~ *~ 9 *m- " r CH, 1 . ..4-26.Q -76.0" to 10 .p- f0-Broinbenzoyl-f/-carvoxime 26.0 -f 90.3 V. T,_ -, T o 26.8 TCH - CH:i "I - 26 '9 "4 76.o _"- L 6 CO.XO:C 10 H U J ..--23.4 -66.3 Dyl-o'-carvoxime 26.0 -f- 9-3 * % - J"" " TCH Br -i ... + * +63-5^;* 5 !/>- " CO.NO : C, H,J 14.9-51.9 f 0-Nitrobenzoyl-^ 68.0 79.0 II. 61.7 70.6 8-9 -52.8 61.0 8.2 ^-Compound . . M*- 102.8 23.8 9^-5 20.9 - 76.9 15-9 Kthyl ester. 0-Compound- . . r^-] 54-7 - 50-4 ^/-Compound . . /-Compound .. M, []/> 63.7 90.0 9.0 26.3 -58-7 -81.5 8-3 22.8 -69.5 Here the rotation increases in the order o, m, p, and the difference between o and ;;/ is always smaller than the difference between m and p. These relations remain true for the differ- ent temperatures. 3. The specific rotations of the following bodies have been determined by Bin//' from chloroform solutions: 0-Oxybenzylidenefenchylamine ( /> 2.5) 66.0 [oil -j 6 - C " Hl CH:N.C 10 H I7 J 0-Methoxybeii/ylidenefenchylaniine (/> 5 ) 59.4 ;C LH. N :C 10 4. Walden 3 obtained from solutions in glacial acetic acid (c i), for Malic acid di-^-toluide ............... f"n ] 66.5 Malic acid di-/>-toluide ............... 70.0 ^'* Iii the majority of these isomers the .para compouiul appears to have a greater rotation than the ortho. But cases are known in which isomeric bodies rotate in opposite directions. This is found' in Propylclibenzoyl glycerate. Metliyldiphenylui-tt \ 1 v;l\ cerate. CH.,.O.CO.C, H x / >c< v. H COO/ X>. CO.C,H CH,Coo- O.CO.CH,.C H [r];5 4-21.0 !_],, 16.1. 1 J. Chem. Soc., 69, 1309 (1896). 2 Binz: Ztschr. phys. Chem., 12, 727 (1893). U'al'lt-ii: X,tM )ir. phys. Chem., 17. * Frankland and MacC'regor: j cin-m. S.K- , 69, 104 (1896). ROTATION AND CHEMICAL CONSTITUTION 289 Also with the isomers in the santonin group, for which there was obtained from chloroform solutions: Santonin 1 [#]/>= 171.4 Metasantonin -f 1 18.8 Santonide -f~ 744.6 Metasantonide 223.5 Parasantonide -)- 897.3 c. Stereoisomeric Bodies 80. A series of observations carried out by Walden, 2 are concerned with the active amyl esters of fumaric and maleic acids and their derivatives, /-amyl alcohol was employed in their preparation. Diamyl ester of [/*]/> \.M~\ D FM. Fumaric acid 8 + 5.93 -f 15.17 Maleic acid +4.62 +11.82 Chlorfumaric acid 3 -j- 5.78 +16.78 Chlormaleic acid +4.03 -(-11.70 Bromfumaric acid -(- 5.99 -j- 20.07 Brommaleic acid -(- 4.58 -f- 15.36 Methylfumaric (mesaconic acid) +5-93 + 16.01 Methylmaleic acid (citraconic acid.) -j- 4.14 +11.17 Mean 4.50 In all these bodies, it is seen that the fumaroid form has a molecular rotation higher by about 4.5 than the maleinoid. On the other hand, if we consider compounds which are related to these types, but which are saturated (acids of the suc- cinic series), according to Walden, the above no longer holds true, as seen in the following : Diamyl ester of [**} D^] D FM /-Dimethylsuccinic acid 3.66 10.47 Antisuccinic acid 3.42 -f 9.79 Racemic acid -3-37 Hr 9-77 _ 4 06 Mesotartaric acid -7- 4.77 -j- 13.83 i Carnelutti and Nasini : Ber. d. chem. Ges., 13, 2208 (1880). - Walden : Ztschr. phys. Chem., 20, 377 (1896). 3 Earlier observations of Walden (Ztschr. phys. Chem., 15,638(1894)) gave the following values for these bodies : Diamyl esters of [<*]/> \M~\ D Diff. Fumaric acid + 5.69 + 14.56 Maleic acid + 4.35 :i.i3 Chlorfumaric acid 5.74 16.67 C 19 Chlormaleic acid . /o + 13.36 2QO SPECIFIC ROTATION But it must be remembered here, however, according to 18, that a racemic acid ester of active amyl alcohol does not exist. This is a mixture of the amyl esters of d- and /-tartaric acid, which bodies, since they are not reflection images of each other, cannot form a racemic compound. 77. Homologous Series 81. The relations appearing here can be seen from the follow- ing observations : CHANGES IN THE MOLECULAR ROTATION WITH INCREASE OF CH 2 \a\ D [M] D Diff . I. 1 Amyl formate + 2 - 01 + 2 -33 , ^ Amylacetate 2.53 3.29 ^ Amyl propionate 2.77 3.99 Amyl A^-butyrate 2.69 4.25 Amyl jV-valerate 2.52 4.33 Amyl A^-caproate 2.40 4.46 Amyl AMieptylate 2.21 4.42 Amyl A^-caprylate 2.10 4.49 Amyl AMionylate 1.95 4.44 Amyl laurate 1.56 4.21 Amyl palmitate . 1.28 4.17 Amyl stearate 1.27 4.49 2. 'Valeric acid +13.64 +13.91 , ^ Methyl valerate 16.83 J 9-53 ^'06 Ethyl valerate 13.44 17-47 6 jV-Propyl valerate u.68 16.82 ' 5 A^-Butyl valerate 10.60 16. 75 3. s Methyl glycerate -- 4.80 - 5.76 + 6 Ethyl glycerate 9.18 12.30 6 'g A'-Propyl glycerate 12.94 19.15 _^_ 2 ' 22 ^V-Butyl glycerate 13-19 21.37 Heptyl glycerate 11.30 23.05 Octyl glycerate 10.22 22.28 ~ -77 4. 4 Methyl diacetylglycerate 12.04 24.56 Ethyl diacetylglycerate 16.31 35.56 XI '^ A^-Propyl diacetylglycerate ^9-47 45- 1 7 5. 5 Methyl dibenzoylglycerate +26.89 +88.20 Ethyl dibenzoylglycerate 26.58 90.90 jg , A^-Propy 1 dibenzoylglycerate 2 1 .00 74-76 Guye and Chavanne : Compt. rend., lao, 452; Bull. Soc. Chim., (3), 15, 275 (1896). Guye and Chavanne: Compt. rend., 116, 1454 (1893). Frank land and MacGregor: J. Chem. Soc., 63, 1415 (1893). Frankland and MacGregor: J. Chem. Soc., 63, 1430. Frankland and MacGregor: J. Chem. Soc., 69, 104 (1896): ROTATION AND CHEMICAL CONSTITUTION 291 I. II. III. M, M, MX, 6. Methyl malate .. -- 6.85 - 6.88 - 7.34 Ethyl malate .... 10.18 10.65 12.42 JV-Propyl malate 11.62 11.60 13.70 jV-Butyl malate 10.72 12.20 Amyl malate ... 9.92 Capryl malate .. 6.92 \M~\ r^/i fji/i L AD \- \ D L J /? Methyl malate .. 11.10 11.15 11.89 Ethyl malate. . , . ! 9 . 3 5 + 5 20.23 + 9 ^ 23.56 + l1 ^ jV-Propyl malate 25.32 5 ' 97 25.29 ] 29.87 jV-Butyl malate .. 26.38 ~ " ItO9 30.01 Amyl malate ... 27.19 Capryl malate -.24.77 I. Formed by esterification from acid and alcohols by aid of sulphuric acid. 1 II. Produced in same way. 2 III. By action of the alkyl iodides on silver malate.* MK W\* Diacetyl malate of I. 4 II. 3 I. 4 II. 5 Methyl 22.92 22.86 46.76 , 46.64 Ethyl 22.52 22.60 52.25 T*'t? 52.43 N-Propy}... 22.85 22.68 59.40 ~] 58.96 JV-Butyl 21.88 19.93 63.01 + 31 57.38 M* 8. 6 Methyl ^/-tartrate -f 2.14 +3.8 Ethyl rf-tartrate 7.66 15.7 A'-Propyl ^-tartrate 12.44 29.0 9. 7 Methyl chlorsuccinate -{-41.42 +74.8 Ethyl chlorsuccinate 27.50 57.3 A r -Propyl chlorsuccinate 25. 63 60.6 ^V-Butyl chlorsuccinate 21.57 57.1 Amyl chlorsuccinate 21.56 63.1 2 act. rad. Walden: Ztschr. phys. Chem., 17, 245 (1895). Anschiitz and Reitter: Ztschr. phys. Chem., 16, 493 (1895). Purdie and Williamson: J. Chem. Soc., 69, 818 (1896). Walden ; Ztschr. phys. Chem., 17, 245 (1895). Anschiitz and Reitter: Ztschr. phys. Chem.. 16, 493 (1895). M. A. Pictet : Arch. phys. nat., (3), 7, 82 (1882) Walden : Ztschr. phys. Chem., 17, 245. 292 SPECIFIC ROTATION [oi\ D lo. 1 Santonic acid ......................... -- 70.31 -185.6 Methyl santonate ..................... 52-33 T 45-5 Ethyl santonate ...................... 45-35 132.4 ^ o jV-Propyl santonate .................. 39-34 120.4 Parasantonic acid .................... 89.51 260.1 Methyl parasantonate ................ 108.91 302.8 Ethyl parasantonate .................. 99.98 291.9 A^-Propyl parasantonate .............. 91.27 279.3 From chloroform solutions. ii. Fen chylamine (liquid)* ............... -- 24.89 - 38.0 Formylfenchylamine (^ = 3.9) ....... 3 6 -56 66.0 Acetyl " (fi=--4.6) ....... 46.62 90.7 [ ^ Propionyl " (^ = 5.0) ....... 53.16 110.9 Butyryl " ( = 1.8) ....... 53.11 118.2 Determined from chloroform solutions. The following relations appear by comparing the molecular rotations of the above compounds : With the homologous esters, the values of \_M~\ sometimes increase and sometimes decrease with increasing molecular weight of the alcohol radical. An increase is found in the esters of glyceric acid (No 3), diacetyl glyceric acid (No. 4), malic acid (No. 6), diacetyl malic acid (No. 7), and tartaric acid (No. 8) ; also with the amyl esters of the fatty acids (No. i), and the fenchyl deriv- atives (No. n). A decrease is found with the esters of valeric acid (No. 2), santonic and parasantonic acids (No. 10). The differences between the molecular rotations of two neighboring members of a series, decrease in most cases regularly (Nos. i, 2, 4, 6, n) ; the effect of the gradually increasing molecular weight of the alcohol radical becomes therefore, constantly less. From a certain molecular weight on, the influence seems to be exerted in the opposite direction, so that the increase is changed to a decrease, from which it follows that in the homol- ogous series a member is found which possesses a maximum rotation. Carnelutti and Nasini : Ber. d.chem. Ges., 13, 2208 (1880). 2 Binz : Ztschr. phys. Chem., 12, 731 (1893). ROTATION AND CHEMICAL CONSTITUTION 293 The geatest change is found in going from the acid to the first ester (Nos. 2 and 10). In Nos. 5 and 9 the rate of change is irregular. ///. The Effect of Linkage of the Carbon Atoms. The experimental results on this are as follows : a . Change from Single to Double Bond by Loss of 2 Atoms of H. 82. The changes of rotation in cases of this kind have been investigated by Walden 1 in a series of liquid esters of active amyl, CH(CH 3 )(C 2 H 5 ) CH 2 , in the preparation of which left-rotating amyl alcohol was used. C 3 H U = A. \0\ D [M] D Diff. Amyl w-butyrate CH 3 CH 2 CH 2 CO 2 A + 2.81 + 4.43- ^ i Amyl crotonate CH 3 CH^CH CO 2 A +4.24 + 6.62 CH 3 CH CO, A Amyl isobutyrate +3.10 + 4.90 CH, CH 2 =C C0 2 A Amyl methacrylate + 3.51 + 5.47 fCH, COXA Diamyl succinate +3-76 + 9.71 CH 2 -C0 2 A CH C0 2 A | Diamyl fumarate T- 5.93 + 15-^7 L CH CO,A CH 2 CO,A Diamyl chlorsuccinate | -h 3-75 + 10.98 CHC1 CO 2 A CH-C0 2 A Diamyl chlorfumarate +5.78 +16.78 C Cl CO 2 A CH 2 CO,A Diamyl methylsuccinate | +3.76+9.99 CH.CH 3 COjA CH CO, A Diamyl mesaconate + 5.93 + 16.01 C.CH :i CO 2 A f CH 2 CO,A Triamyl tricarballylate CH CO 2 A + 4.01 + 15.48 CH 2 -CO,A CH-C0 2 A I! Triamyl aconitate C CO, A + 6.16 + 23.66 I CH 2 CO 2 A 1 Walden : Ztschr. phys. Chem., 20, 569 (1896). 294 SPECIFIC ROTATION C 6 H n = A [>]/> z> DiflF. | A *J53* ' C 6 H 5 -CH 2 -CH 2 -C0 2 A + 2.26 + 4.98 ^ 1 Amyl cinnamate C 6 H 5 CH=CH CO,A + 7.51 -f 16.36 ^'^ In all these cases it is seen that the change from single link- age of carbon atoms to double linkage is followed by an increase in the rotating power. b. Change from Double to Triple Bond between Carbon Atoms. 83. For this case, we have as yet only the following illus- tration, likewise from Walden : l M/> \M\D Diff. ( Amyl cinnamate C 6 H 5 CH=CH -CO 2 A -j- 7.51 + 16.36 \Amyl phenylpropiolate C 6 H 5 C=C CO 2 A -f 5.58 -f 12.05 4 ' 3 * Here a decrease in activity follows. c. Change from a Chain Carbon Compound to a Cyclic Com- pound. 84. As van't HofF first showed, the ring structure exerts a very considerable influence on the degree of rotation, and it may even change its sign. Accurate numerical data are here hard to obtain, as most of the compounds which could be considered are solid, and their specific rotations, therefore, variable with the solvent and con- centration. The relations which may obtain may be seen from the following comparison of two dicarboxylic acids with their cyclic anhydrides: f water c = Diacetyl tartaric acid J water e = j ethyl alcohol c = [ ethyl alcohol c = f acetone c = Diacetyl tartaric anhydride \ . *. benzene c = Dibenzoyl tartaric acid f ethyl alcohol c = ( anhydrous ) \ methyl alcohol c = Dibenzoyl tartaric anhy- acetone c = dnde 3 17.95 7.37 3-27 11.66 4.40 2.09 1.05 4.76 4.63 4-64 M* = - teU- M*- + 23.0 19.3 23.6 21.5 59-7 62.0 58.7 63.1 117.7 122. 1 1429 1 Walden : Loc. eit. * van't Hoff : "Mgerung der Atome im Raumc," 1894, p. 109. M. A. Pictet : Jahresbericht, 1882, p. 856. ROTATION AND CHEMICAL CONSTITUTION 295 In both of these cases the rotation of the anhydride is greater than that of the acid, and of opposite direction. But another condition appears with the following bodies : l acetic ether c = 10 [a~] n = + 52.7 Chlorsuccmic acid 2 L"J D acetic ether c = 6.66 -f 52.9 acetic ether c = 10 [oH = Chlorsuccmic anhydride acet ic ether c = 5 In this case the rotation of the anhydride is smaller than that of the acid, but in the same direction. If we compare the lactone- form ing acids of the sugar group with the lactones themselves, it is found, as pointed out in 75, that the latter have always much the stronger activity. As the true rotating power of the acid can not be determined with certainty because of the existence of multirotation, Alberda van Ekenstein, Jorissen and Reicher 3 have investi- gated the neutral alkali salts, which do not show multirotation, and from which the rotation of the acid ion may be obtained. The lactones were tested as soon as possible after solution so as to avoid the effects of multirotation. In the experiments the concentration of the acid ions was from 2 to 6.5 grams in 100 cc. and of the lactones from 4 to 10 grams. With addition of a few data from Fischer, Tollens, and others, the authors mentioned give the following table. 4 \_M~\D , . Change in Acid ion. Iactone. rotation. Ribonic acid -+- 2 - 30 32 I. /-Amyl /-oxybutyrate - 8.5 II. /-Amyl /-oxybutyrate -f- 1.5 III. /-Amyl /-oxybutyrate - 7*3 I + II = - 7.0 AMYL AMYLACETATES/ [ar] D I. /-Amyl I. / : Amyl /-mandelate 96.46 II. /-Amyl /-mandelate -j- 2.76 III. /-Amyl /-mandelate 94.02 I + II = - 93.70 DlAMYI, CHI,ORSUCCINATES. 3 \_ a ~\D I. /-Amyl rf-chlorsuccinate -f 21.56 II. /-Amyl /-chlorsuccinate -f- 3.75 III. /-Amyl 6 [XU 7 I. /-Amyl I. /-Amyl /-valeryl /-tartrate + 2.44 II. /-Amyl d- valeryl /-tartrate + 3.48 III. /-Amy 1 /-valeryl rf-tartrate + 6.42 IV. /-Amyl l I. Amyl acetate + 3.25 II. Amyl acetic acid + 1 1.08 III. Amyl amylacetate + 14.02 I + II = + 14.33 The great differences which appear in the specific rotations of the isomeric sugars, for example in the hexoses, or in the hexonic acids, depend, undoubtedly, as van't Hoff 2 suggested, on the summation of the effects of the four asymmetric groups contained in them, the rotations of which are unequally strong and in opposite directions. Observations are not yet sufficiently numerous to establish the values of the group rotations, not even for the ions of the lactone-forming acids given some pages back. IV. Dependence of the Rotatory Power of an Active Atomic Complex on the Masses of the Four Radicals Joined to the Asymmetric Carbon Atom. The Hypothesis of Guye. 86. An attempt to determine the amount and direction of rotation from the composition of an active molecule was made in 1890, simultaneously by Ph. A. Guye 3 and Crum Brown, 4 a consideration of the tetrahedral form of the asymmetric complex, and the relative masses of the four groups being the common starting point in the discussion. The problem, which was handled in detail and brought into mathematical form especially by Guye, 5 was well calculated to arouse great interest, and it has been the incentive in the undertaking of numerous investigations. The hypothesis, as stated by Guye 6 in 1893, * n general form is as follows : Walden : Ztschr. phys. Chem., 15, 638 (1894). van't Hoff : " I^agerung der Atome im Raume," ad ed. p. 120. Guye : First paper: Compt. rend., no, 714 (1890). Crum Brown : Proc. Roy. Soc. Edin., 17, 181 (1890). Guye : Compt. rend, in, 745 (1891) ; 114, 473 (1892) ; 116, 1133, 1378, 1451, 1454 (1893); 119, 906 (1894) ; lao, 157, 452,632, 1274 (1895). These: Paris, 1891. Conferences de la Soc Chim., Paris, 1891, p. 149. Arch. sc. phys. nat. [3] a6, 97, 201, 333 (1891). Ann. chim. phys., [6], 35, 145 (1892). Bull. Soc. Chim., [3], 9, 403 (1893). 6 Guye : Compt. rend., 116, 1378, 1451. 300 SPECIFIC ROTATION As a measure of the amount and sign of the rotating power of an active molecule, the so-called product of asymmetry, P, may be taken, which, in general, may be defined as equal to the product of the six perpendiculars from the center of gravity of a tetrahedron to the six planes of symmetry of the original regular tetrahedron. According to the orientation of the four groups combined with the asymmetric carbon atom, Pis found in different ways : i . If the tetrahedron is regular and the radicals are found exactly at the angles of the same, the product of asymmetry depends only on the masses of the four groups, a, b, c, and d\ that is, on their formula weights. In this case, the following expression 1 is found for P. in which the constant factor (/. sin a)* may be dropped. 2 2. The masses, a, b, c, and d may be situated at different distances, /, m, n, and /, from the center of gravity of the original tetrahedron, but always in the direction of the straight lines from the center to the four angles. 3. The masses a, b, c and d are found at different distances, /, m, n and/ from the center of the original tetrahedron, and further, on account of their mutual attractions, they have undergone lateral displacements, so that the straight lines, /, m, n and p form different angles with each of the original planes of symmetry (ae l ... a 6 for / ; /?, ... /?,. for m ; y l . . y 6 for w ;,... <5 6 for/). The complicated formulas 3 for P in case 2 and the perfectly general case 3 can not be used for calculations because they contain undeterminable quantities (/, m, n, p at. . . /?. . . y . . . 6). We are, therefore, limited to formula I, under the assumption that the displacing influences mentioned in cases 2 and 3 are too small to cause appreciable disturbances. The above equation satisfies the conditions, that : a. The product P must be zero when two or more of the 1 For the derivation of the formula, the original paper must be consulted. In this, / is the distance of the four masses from thecenterof the tetrahedron, and a is the angle 54 44'. 8 See the original paper. ROTATION AND CHEMICAL CONSTITUTION 301 masses, a, b, c, and d are equal ; that is, when the asymmetry of the molecule is destroyed. b. The product must be the same but of opposite sign when two of the values a, b, c, and d are transposed, the one for the other. Such a change corresponds to the conversion of the right-rotating form of a body into the isomeric left-rotating. Changes in the rotating power must follow parallel with changes in the product, P, corresponding to variations in the weights a, b, c, and d. If the order of the weights of the groups is as follows : a>b>c>d and a is replaced gradually by smaller and smaller values, then, if the original body be assumed, for illustration, as right- rotating, the following conditions are to be expected : 1 . As long as a > b there must be, according to the numeri- cal relation between them, either a continuous decrease in the right rotation, or at first an increase, and then after passing a maximum, a decrease in the rotation. 2. a = b. Condition of inactivity. 3. a < b. Change to increasing left rotation to a maximum, then a decrease. 4. a = c. Second condition of inactivity. 5. a < c. Appearance of right rotation, which increases to a maximum and then decreases. 6. a = d. Third condition of inactivity. ^7. a < d. Increasing left rotation. In the experimental examination of these provisions, they seemed at first to be confirmed. Thus, it was possible to show in some homologous series the complete or nearly complete coincidence of a maximum point in rotation with a maximum point in the product of asymmetry, as is illustrated by the following table of Guye and Chavanne 1 based on observations by Frankland and MacGregor 2 on the rotation of esters of /-glyceric acid. The value of the product of asymmetry, P, is shown in parallel column, and multiplied by io 6 to give con- venient numbers for comparison. 1 Guye and Chavanne : Compt. rend., 116, 1454. 2 Frankland and MacGregor: J. Chem. Soc. 63, 524 (1893). 302 SPECIFIC ROTATION Glycerate of * C(COOR). (CH 2 OH.) (OH.) d (H.) OL. LM] D . P. 10.6 JV-Butyl . . . ioi 31 17 I II.02 1 - 17.9 347 AT.propyl.. 8? 31 17 I - 12.94! - 19.2 358 Ethyl 73 31 17 I - 9.18 - 12.3 345 Methyl .... 59 3 1 17 I - 4.80 - 5.8 289 A very near coincidence in the maximum points, with a dis- placement of only one term, is shown in the following valeric esters investigated by Guye and Chavanne. 2 Valerate of C(COOR). (C,H 6 .) (CH 3 .) d (H.) M*. [ = 21.4 (J. Chem. Soc., 63, 1417), which makes this example unsuitable for confirmation of the hypothesis. 1 Guye and Chavanne : Compt. rend., 116, 1454. Guye and Chavanne: Compt. rend., 119, 906. If the values of the molecular rotation \M\ be taken, the maximum is then found at amyl caprylate, that is, far removed from the maximum of the product of asymmetry. ROTATION AND CHEMICAL CONSTITUTION 303 in which besides Guye, 1 many chemists, 2 and especially Walden 3 have taken part led gradually to the discovery of numerous facts which can not at all be reconciled with the re- quirements of the hypothesis. The following discrepancies, especially, were brought out : i. Compounds in which two of the groups a, b, c, and d, have the same weight, and which, accordingly, should be inactive, often possess a strong rotating power. Walden 4 gives many cases, for example : M, ["] a bed CfCH COOCH ^ (CO OCH WO C H O^ (H^ . 73 59 59 i Methyl acetylmandelate C(C H ^ (CO OCH "> fO C H O^ (FH.. 22.9 146 J. 4"- 3Od 5 77 59 59 i Ethyl propionylmandelate C(C TT ^ (CO OC H "1 CO C H O\ CR\ . 117 7 268 -; 77 73 73 i Dimethyl propionylmalate C(CH CO OCTT ^ (CO OCH WO C H O^ (H^ 22 Q CQ O 73 59 73 i Dipropyl isovalerylmalate C(CH CO OC H 1 (CO OC.H,") (O C-HO^ (H") .. .. - 101 87 101 i 21.7 05-5 2. A change in the order {transposition} of two group weights, which, according to the theory, should be accompanied by a change 1 Guye has introduced another constant of rotation in addition to the specific and molecular rotation, and expresses it by the formula rj , a 3 77 [6] =~r A -5-, in which a is the observed angle of rotation, / the length of column, M the molecular weight, and d the density of the substance. Aignan criticized the applicability ot the formula (Compt. rend., 120, 723). 2 l,e Bel : Compt. rend., 114, 304; 119, 226. Bull. Soc. Chim., [3], 7, 613, 801. Colson : Compt. rend., 114, 175, 417; 115,729, 948; 116, 319, 818 ; 119, 65; lao, 1416. Bull. Soc. Chim., [3], 7, 802; 9, i, 87, 195. Friedel : Compt. rend.. 115, 763, 994 ; 116, 351. Freundler : Compt. rend., 115, 509, 866 ; 117, 556. Bull. Soc. Chim., [3], 7, 804 ; 9, 409,680; n, 305, 366,468, 470, 477. Ann. chim. phys. [7], 3, 487. Simon: Bull. Soc. Chim., [3], li f 760. Purdie and Walker : J. Chem. Soc., 63, 240. Frankland and MacGregor : J. Chem. Soc., 63, 1416, 1430; 65, 750. Piutti : Gazz. chim., a4, II, 85. Binz : Ztschr. phys. Chem., la, 733. Goldschmidt : Ztschr. phys. Chem., 14, 394 Wallach : Ann. Chem. (t,iebig), 376, 316, 322. 8 Walden : Ztschr. phys. Chem., 15, 638 ; 17, 245, 705. * Walden : Ztschr. phys. Chem., 17, 245, 712 (1895). 304 SPECIFIC ROTATION in the direction of rotation, often is not followed by this effect. Thus, we have according to Walden: 1 t Substance. Order of the weights of the groups. Direction of rotation of the substance. Sign of the product of asymmetry. | Mandelic acid C(C 6 H 5 )(CO.OH)(OH)(H) (i b c d 4_ i Amyl mandelate C(C 6 H 5 )(CO.OC 5 H H )(OH)(H) i Acetylmandelic acid C(C 6 H 5 )(CO.OH)(O.C,H 3 0)(H) d c b d 1 _L Dipropyl acetylmalate C(CH 2 .CO.OC 3 H 7 )(CO.OC 3 H 7 )(O.C,H 3 0)(H) 101 87 59 i.. Dipropyl chloracetylmalate C(CH 2 .CO.OC,H 7 ;(CO.OC,H 7 )(O.C. 2 H 2 C10)(H) 101 87 93.5 (i) abed a c b d + + J_ On the other hand, by transposition of the group weights a change may follow in the direction of rotation, while the sign of the product of asymmetry remains the same. For example : Substance. 1 Direction of rotation of the substance. Sign of the product of asymmetry. Mandelic acid C(C 6 H 6 )(CO.OH)(OH)(H) 11 AZ 17 I . . i 77 45 7 * Phenylbromacetic acid C(C 8 H 5 )(CO.OH)(Br)(H) c d b d 1 Phenylbromacetyl bromide C(C B H 5 )(CO.Br)(Br)(H) b c d d Dimethyl malate C(CH 2 .CO.OCH 8 )(CO.OCH 3 )(OH)(H) d b c d 4_ Dimethyl bromsuccinate C(CH,.CO.OCH 3 )(CO.OCH,)(Br)(H) c a b d 4- i Walden : Ztschr. phys. Chem., 17, 705. The compounds were made from /-mandelic acid and had, therefore, a direction of rotation the opposite from that given in the table. The same is true of the ester of malic acid. ROTATION AND CHEMICAL CONSTITUTION 305 j. In homologous series the changes in rotatory poiver and product of asymmetry are not parallel in the majority of cases , but subject to manifold deviations. From all these considerations it has become evident that the principles on which the product of asymmetry is based, are not satisfactory. It is clear, as Guye 1 also admits, that it is not alone the masses of the four groups which exert the influence, but also their relative positions, the actions which they have on each other, their configurations, and finally the nature of the elements themselves which are important in determining the direction and extent of rotation. On account of this complexity in the phenomenon, it is unlikely that, even through other means, will it ever be found possible to discover the numerical relations between amount of rotation and atomic structure of the molecule. 1 Guye and Chavanne : Bull. Soc. Chim., [3], 15, 195 (1895). Arch. phys. nat., [4], 1,54(1896). 20 PART FOURTH Apparatus and Methods for Determina- tion of the Specific Rotation 87. General Conditions. In the calculation of the specific rotation, the experimental determination of the following data is necessary : 1 . The measurement of the angle of rotation a for a definite light ray. 2. The measurement of the length / of the tube for the liquid, in decimeters. 3. The determination of the amount p of active substance in 100 grams of solution. 4. The determination of the specific gravity d of the solution. 5. The determination of the amount c of active substance in loo cubic centimeters of solution. A. MEASUREMENT OF THE ANGLE OF ROTATION 88. Ordinary and Polarized Light. While in an ordinary light ray the vibrations of the ether particles take place in all directions in a plane perpendicular to the line of propagation of the light, in a ray of plane polarized light the vibrations of the ether particles occur in a single direction only. Such a plane polarized ray is no longer symmetrically disposed around its axis. The plane in which the ray is polarized is known as its plane of polarization. The conversion of ordinary light into polarized light may be effected, to begin with, by reflection, which is accomplished by aid of the apparatus shown in Fig. 22. If a pencil of light be allowed to strike the black glass mirror A under an angle of 57, the rays will be reflected upwards and polarized in the ORDINARY AND POLARIZED LIGHT 307 plane of incidence. The proof of this may be given by aid of the second mirror B. The reflected rays first pass through the empty vessel F, the bottom of which is formed of a plate of plane glass, and strike the mirror B under the same incident angle of 57. By means of the lever D and the rack- work at E, B and the paper screen C may be rotated around a vertical axis. If the mirror B has a position parallel to A, the plane of incidence of the polarized rays reaching B coin- cides with their plane of polariza- tion, and in conse- quence there is a considerable reflec- tion toward C where a bright spot is formed by the pen- cil of light. On ro- tating the mirror B, however, the inten- sity of the light reflected from it gradually decreases until a position is reached 90 from the original one, when it is found that no more light is reflected and the screen C remains perfectly dark. 1 The plane of incidence of the rays 1 All of the light is refracted in the glass and absorbed by the dark back surface. Fig. 22. 308 POLARIZATION APPARATUS is now perpendicular to the plane of polarization. On further rotation of the mirror B, it is found that at 180, that is, in the position where the planes of incidence and polarization again coincide at B, there is a maximum and at 270 a minimum again of reflection. 89. Rotation of the Plane of Polarization. Let the mirror B be brought into the position of greatest darkness, so that the plane of incidence of the rays polarized by the mirror A, and reaching B, is vertical to the new plane of polarization. If the vessel F be now filled with a cane-sugar solution, for example, the remarkable phenomenon is exhibited in which the screen C becomes suddenly bright and remains so until the mirror B is rotated through a certain angle. Now, again, as a matter of course, the plane of incidence of the transmitted rays is perpendicular to their plane of polarization. It follows, therefore, that the plane of polarization of the rays leflected through the sugar solution has been turned or twisted through an angle equal to that through which B was turned. This angle is known as the angle of rotation. 90. Iceland Spar Prisms. A pencil of light may be linearly polarized by double refraction in crystals, especially in Iceland spar, much more perfectly than by reflection. If a ray of light falls perpendicularly on one of the faces of a natural Ice- land spar rhombohedron it is broken up, on entering the crystal, into two separate rays, unequally refracted and linearly polarized in planes perpendicular to each other. If we define as the optical axis of the crystal that direction parallel to which no double refraction and also no polarization takes place, and as the optical principal plane of the incident ray, that plane which includes the perpendicular at the point of incidence and also the optical axis, then the principal plane is at the same time the plane of polarization of the ordinary refracted ray, while the plane of polarization of the extraordinary ray is per- pendicular to the principal plane. This holds still accurately true when the incident ray instead of falling vertically upon the surface of the crystal strikes it at any angle, as long as the incident plane is at the same time a principal plane of the crystal. In the practical applications of these rays in polar- ICELAND SPAR PRISMS 309 ization instruments it is better to permit only one to emerge, in the direction of the incident light, while the other is elimin- ated. This may be accomplished in various ways, most per- fectly on converting the Iceland spar into a Nicol prism. The polarization prisms, 1 described at length below, are used in the modern forms of polarization inslruments for scientific as well as for technical purposes. i. NicoVs Prism. This, which is the most widely known type, is made in the following manner : A rhombohedron, abed (Fig. 23), the length of which is fully three times the w r idth, is cut from a clear crystal of Iceland spar ; the end surfaces, which make originally angles of 7 1 with the side edges, are polished off so that these angles, at a and c, be- come 68, and then the prism is sawed through in the direction b' , d' . After the angles a V d' and c d' b' are ground down to 90 and the sawed surfaces polished they are cemented together again in the original position by means of Canada bal- sam. Finally the side surfaces are blackened and the finished nicol is fastened into a brass frame by aid of cork. The optical principal plane of the prism for all rays falling on the ends is the plane vertical to the end surfaces and passing through the optical axis. In illustration, if a ray of light whose plane of incidence contains the optical axis falls upon one of the end surfaces, it is divided on entering into two rays polarized perpendicularly to each other. In case the entering ray makes but a small angle with the axis of length of the prism, the ordinary com- ponent suffers total reflection on the cement surface, is thrown to the dark side surface and is here largely absorbed, while the extraordinary component passes through the cement and emerges alone from the second end surface in a direction parallel to that of entrance. The plane of polarization of this emerging ray is vertical to the principal plane. If a small flame at some distance is observed through the nicol, under such condition that the entering rays make but a 1 Feussner : " Ueber die Prismen zur Polarisation des lyichtes," Ztschr. f. In- strum., 4,41 ( 1884). Grosse: "Ueber Polarisationsprismen," Ztschr. f. Instrum., 10,445 (1890). Halle: "Ueber Herstellung Nicol'scher Prismen," V. d. D. Ges. f. Mech. u. Opt., 143 (1896). Fig. 23. 3io POLARIZATION APPARATUS small angle with the axis of length of the prism, the eye perceives uniform illumination within a certain limited field, and the planes of polarization of the individual linearly polarized rays coming through the prism deviate very little from each other, that is, within certain limits, they are all polarized perpendicularly to the principal section of the nicol. It may then be briefly said (although not with absolute accuracy) that when a pencil of light passes through a nicol, the emerging light is linearly polarized, and in a direction vertical to the principal section. 1 The plane of polarization of the light emerging from a nicol can be found most simply, empirically, by aid of a revolving glass mirror. The light from the prism is allowed to strike the mirror under an incident angle of 57, and the glass is then rotated around the rays as an axis until all reflected light disappears. According to 88 the plane of polarization of the nicol is now vertical to the incident plane of the rays on the mirror. 2. Hartnack-Prazmowski Prism. From the natural crystal, ( b abed (Fig. 24), the prism a' b' c f d' is cut out and sawed in the direction a' c'. After the surfaces a' b' and c' d' and the sawed surfaces a' c f are ground and polished the latter are cemented together. The entering angle b' a' c' must vary according as Canada balsam, linseed oil or other transparent cement is employed. Although the loss of material on cutting the crystal is greater than in the Nicol prism, the Hartnack prism possesses, notwithstanding its shorter length, a much greater field of view. Be- sides this, it has the advantage of presenting straight end surfaces. In the prism, as de- scribed by Hartnack, the optical axis stands vertical to the plane of the section a' c' ; the optical principal section is, therefore, a plane through the axis of length and vertical to the cut surface. 3. Prism of Clan [-Thompson].' 2 A much greater loss of 1 For further details see I^ippich : " Ueber polaristrobometrische Methoden, Wien. Sitzungsber. II, 85, 268 (1882). 2 Lippich has constructed a similar prism. Wien, Sit/.miK.sber. II, 9i, 1079. (1885). The r.lan prism with air layer is not to be confounded with the above. Fig. 24. POLARIZER AND ANALYZER Fig- 25. material than in the last prism is suffered in that of Glan. On a symmetrical rhombohedral crystal, two surfaces are ground down parallel to each other and perpendicular to the optical axisof the crystal ABC (Fig. 25); vertically to these surfaces the prism a b c d is then cut out and sawed through in the direction bd. After the surfaces a b and c d, and the cut faces b d are ground and polished, the two halves are cemented to- gether. The angle a b d depends, as before, on the kind of cement C which is employed. The Glan prism surpasses the others described in having an es- sentially larger opening with corresponding length ; the field of view is also normal to and symmetrical with the axis of length of the prism. This prism may therefore be described as scientifically the most perfect form, and for this reason it has come into use as the polarizing prism in all good instru- ments. As the optical axis of the prism is adjusted parallel to the refractive edges b and d of the piece of spar, it follows that the optical principal section is a plane through the axis of length and vertical to the edge a b. 91. Polarizer and Analyzer. If ordinary light from any source is passed through a nicol the emerging light, as explained in the last paragraph, is linearly polarized per- pendicularly to the principal section. Such a linearly polarized ray may be decomposed, like a force, into two linearly polarized compo- nents, vertical to each other as regards their planes of polarization. If then linearly polar- ized light, the plane of polarization and ampli- tude of which is A B (Fig. 26), falls on a new Nicol, the principal section C D of which makes an angle ex with A B, this light may be broken up into two components, A E = A B cos a and A F = A B sin a. Only the latter component, which is perpendicular to the principal section, will pass Fig. 26. 312 POLARIZATION APPARATUS through, while the component A E, which is polarized in the principal section, is completely extinguished by the nicol. Of a linearly polarized light ray only that component can pass a Nicol prism which is perpendicular to the principal section, and this component is the smaller the smaller the angle which the plane of polarization of the entering ray makes with the principal section of the nicol. If now the nicol is rotated all light passes when C D is brought perpendicular to A B ; on further turning, the light passing becomes gradually weaker, and after reaching 90 (CD parallel with A B) complete darkness follows. On still further turning, the light reappears and reaches a maximum of brightness at 180, and so on. Let the following conditions be considered : We place two Nicol prisms between the eye and a small luminous surface at some distance, and one before the other in such position that their principal sections are parallel with the line of vision. The nicol nearest the light may be in fixed position while the one next the eye may be rotated around its axis of length. The first one is called the polarizer and the other the analyzer. The light reaching the polarizer from the luminous surface is, after passage, linearly polarized vertically to the principal sec- tion. This next reaches the analyzer. If this is at first turned so that its principal section is parallel with that of the polarizer, the already polarized light suffers no further change in passing through the analyzer, and the eye perceives the field of view brightly illuminated. This is also the case when the analyzer is turned through 180, which brings the prin- cipal sections into parallel position again. If, next, the analyzer be so placed that its principal section crosses that of the polarizer at right angles the polarized light will be com- pletely shut off, because now its plane of polarization and the principal section of the analyzer coincide. The rays entering the analyzer behave as ordinary polarized rays until the cement layer is reached and here they are thrown off by reflection. No light can, therefore, pass the analyzer and the field of view remains dark. The same is true after rotating 180. In all other cases in which the principal sections of the two nicols are neither parallel nor vertical, a part of the light entering the analyzer will be allowed to pass, and always that component POLARIZATION APPARATUS 313 which is polarized perpendicularly to the principal section of the analyzer. 92. Polarization Apparatus. The apparatus shown in Fig. 27 may be used for the observation of these phenomena. The horizontal bar d, supported on, a stand, carries at one end the polarizing Nicol, a, in fixed b jj position, and at the other the analyzer b, which may be turned with its receptacle, by means of the lever c, around its axis. A single or double pointer is turned with it over the graduated circle fastened also to the bar, d. Between the Nicols the tube, f, may be placed, the ends of which are closed by glass plates. The polarizer is first turned toward a source of light, which for the sake of greater sim- plicity in the phenomenon should be monochromatic, such as given, for example, by a Bun- sen burner and sodium carbonate bead. At first the tube re- mains empty. On looking through the analyzer and rotating it, a position is easily found in which the field appears at its greatest darkness. Assuming that the pointer is now at o on the circle, from what was said above it will appear that the second position of darkness will be at 180, and the two brightest positions at 90 and 270. For the observations, the darker positions are more suitable than the light ones, because with the former a small motion of the nicol makes a very perceptible change. The position of the analyzer at which the field has the greatest darkness, is called the zero point of the apparatus. In this position the plane of polarization of the light coming from the polarizer coincides with the principal section of the analyzer. If now the tube f be filled with a cane-sugar solution and placed in the apparatus, the plane of polarization of the light coming from the polarizer will undergo rotation through a 314 POLARIZATION APPARATUS certain angle, ar, by action of the sugar solution as explained above. In consequence of this, the field of view becomes bright. Darkness will come again when the principal section of the analyzer is brought into parallel position /with the rotated plane of polarization ; that is, when the analyzer, also, is turned through the angle a. This angle a may be read off on the graduated circle and is equal to the angle of rotation of the sugar solution. If after putting an optically active substance in the tube, it is necessary to turn the analyzer from o in the direction of the clock-hand motion to reach the point again where the light disappears, the substance is said to be right rotating ; if the analyzer is turned in the opposite direction to reach the same end, the substance is left-rotating. 93. Determination of the Direction and Angle of Rotation. With exception of the Wild polaristrobometer, which has four zero points, all polarization instruments have two zero points 1 80 apart. We assume, first, that we have to do with one of the latter forms. By the aid of such apparatus we have to determine the direction of rotation, and the amount of rota- tion of an active substance. After adjusting the apparatus to the zero point and putting the active substance in position, the analyzer must be turned say, through -f a ( a certainly less than 180), that is, in the clock-hand direction, to darken the field again. The angle a read off is not yet necessarily the angle of rotation ; as regards whole multiples of 180 it is yet quite undetermined. It can only be said that the angle of rotation of the substance is equal to ot n 180, where n is either o or a whole number to be determined. In the case of a solid substance, if the thickness of the layer is not above a few millimeters, or in the case of a liquid if the tube length is not above two decimeters, then, unless the substance is one possess- ing unusually great activity, the angle of rotation will be less than 1 80, so that the choice will lie between -f- ot and -f a - 1 80. In order to decide between these two angles the same substance must be examined in a layer of just one-half the thickness, or, in the case of solutions, one of just half the concentration, with the same tube length, may be employed ; now the angle of rotation will be half as great as before. Sup- pose the angle is now -f ft, a simple consideration will show DIRECTION AND ANGLE OF ROTATION 315 ot this : If @ = the substance is right rotating with an angle of a for the full thickness of layer. If ft = 90 + - then the substance is left rotating with an angle of rotation equal to a 180 for the full thickness of layer. The direction of rotation of a liquid may be very conve- niently found by filling it into the control observation tube, with variable length, of Schmidt and Haensch, to be later described. After putting the tube in place the analyzer is moved to the position of darkness ; the tube is then length- ened a little, which produces a brightening of the field. If it is now necessary to turn the analyzer a few degrees in the clock motion direction to secure darkness again, the liquid is right rotating ; but, on the other hand, if the analyzer must be turned in the opposite direction the liquid is left-rotating. The determination of the direction and amount of rotation by aid of the Wild instrument is more complicated, -because this possesses four zero points 90 apart. Under the assump- tion that the angle of rotation is less than 90, the direction and number of degrees of rotation may be found by a plan similar to that just outlined, by working first with a layer of full length and then with one of half the length. But if angles up to 180 are possible then a third length of layer of substance must be taken which is one-fourth the first length. Here also the application of the Schmidt and Haensch control observation tube would be advantageous, the length being first contracted to one-half and then to one- fourth. The position of the Nicol is then always observed in the first quadrant, between o and 90, and the graduation on the circle follows in this manner, that in the case of a small right-rotation of the plane of polarization, with observations in the first quadrant, small numbers close to the o are read off. The correctness of the following can then be easily demon- strated ; if the reading with full thickness of layer is , and if further, with one-fourth this thickness, it is = i then substance is right rotating with the angle o, 4 = 22.5 + i then substance is right rotating with the angle 90 -{- a, 316 POLARIZATION APPARATUS 45 H , then substance is left rotating with the angle 180 a, 4 = 67.5 H , then substance is left rotating with the angle 90 a; 4 Here the angles a, 90 -f , 180 - of, 90 - a refer to the full thickness of layer. a. Polarization Instruments 94. Polarization Apparatus and Saccharimeters. For the exact measurement of the angle of rotation different instruments have been constructed, which, according to their uses, are divided into two classes. These are : 1. The so-called Polariscopes or Polaristrobometers . These are used for scientific purposes in the investigation of all active substances. They have a circular graduation and require homogeneous light. 2. The Saccharimeters. These are specially constructed for the determination of the strength of sugar solutions. In place of the circular graduation, they have a quartz wedge compen- sation with linear scale and employ white light. They are used chiefly in the sugar industry. 95. Construction of the Polariscopes. Before taking up the descrip- tion of the special forms of instruments, a short discussion of the requirements in a good polariscope will be given, and also an explanation of the path of the light rays through the appa- ratus. ! The following considerations obtain for all polariscopes and Saccharimeters, with the exception of the Wild instrument, which, in principle, is different from all other forms of appa- ratus. Those optical parts which all polarization instruments have, in common, are shown in Fig. 28. The light from the lumi- nous body A passes through the lens B into the instrument and is linearly polarized by the polarizer C. Immediately in front of this is found the round polarizer diaphragm D, which is focused on. Then follow the round analyzer diaphragm E, the analyzer F, and a reading telescope. In the figure an i See, also Uppich: Wien. Sitzungsber., II, 85, 268 (1882) ; 91, 1059 (1885). These considerations on the construction of apparatus and the path of the light rays should be carefully followed, in all more exact work, if otic wishes to be certain of excluding bad systematic errors in the results. CONSTRUCTION OF THE POLARISCOPES 317 ordinary astronomical telescope is shown with the objective G, the ocular H, and the diaphragm J, in front of which the eye of the observer is placed. At the outset, the parts from B to J must be ac- curately adjusted with reference to the axis of the instrument. Inasmuch, as w r e shall presently see, as all light going through the instrument is limited by the diaphragms D and E, at any rate in the forms as now commonly constructed, and as all bodies to be investigated with reference to their rotating power are placed between D and E we shall understand as the axis of the apparatus from now on, the line which unites the centers of the diaphragms D and E. All other optical parts of the instrument must be exactly centered this line. Although the adjustment of the on illuminating lens B need be only approximately correct, the polarizer C must be so centered that its optical principal section is exactly parallel to this axis of the instrument. As regards the size of the diaphragms D and E, these must be corre- spondingly smaller than the cross dimensions of the prisms C and F, so that a sufficiently broad border of about two millimeters in diameter around the edges of the prisms should be ob- scured. 33 While in the case of the saccharimeters all the optical parts are fixed with exception of the ocu- lar H J, which is movable in the direction of the axis, in the polariscopes the parts E to J may be rotated around a common axis. This axis of rota- tion, which at most should not be inclined more than a few minutes, must coincide exactly with the axis of the apparatus. This may be easily secured in the smaller forms of apparatus, which, like the saccharimeters, can be worked out in the lathe, but is realized with greater difficulty in the larger instruments, which are composed of the distinct parts B to D and E to J. It may hap- 318 POLARIZATION APPARATUS pen here, that, in instruments most excellent in all other respects, the inclination of the axis of rotation with refer- ence to the axis of the apparatus may amount to as much as ten minutes or even more. In order to avoid such an error the diaphragm D must be attached after the other parts of the apparatus are fastened to the support, and then, if necessary, eccentrically with reference to the thread. The optical prin- cipal section of the analyzer F must be exactly parallel to the axis of the apparatus and the axis of rotation, while the requirement that the axis of rotation must be at the same time the optical axis of the telescope G H J, is one which can always be met satisfactorily. But, above all, care must be taken to have the adjustment of the prisms C and F, with reference to each other, a fixed and unchangeable one, as otherwise constant variations in the zero point would result. With the saccharimeters, therefore, the prism must be fixed once for all, and in the polariscopes the optical principal section of the rotating analyzer must remain always parallel to the axis of the instrument. We have the following two criteria by which to determine whether or not the prisms in the polariscope are properly adjusted and free from errors. First, the two zero points of the apparatus must be exactly 180 apart ; second, if a large angle, say 90 , is measured, the two final observation readings 180 apart must give exactly the same value for the angle of rotation. While with the smaller polariscopes the graduated circle generally remains at rest with the rotation of the analyzer, and the verniers only move, in the larger instruments the graduated circle rotates with the analyzer. In order to elimi- nate the unavoidable errors of graduation in the circle, the analyzer is furnished with a setting which may be rotated independently, or the shell to which the verniers are attached may be turned through 360, which is easily done. In this way the zero point of the apparatus may be brought to correspond to any part of the circle, and the rotation, therefore, measured with different parts of the graduation . In order to eliminate the eccentricity of the circle, that is, the error, which is due to the fact that the axis of rotation does not pass exactly through the center of the disk, two observation verniers, 180 apart, are PATH OF THE RAYS 319 always attached. Both of these must be read each time and the mean of the angles, as given by each vernier, taken ; of course, with the double reading the error of observation, also, is reduced. Besides, the plane of the graduated circle must be vertical to the axis of rotation, otherwise, different values for the same angle of rotation would be found on different parts of the graduation. But such an error is not greatly to be feared, as it is not a difficult matter for the instrument-maker to fulfil this requirement in a satisfactory manner. 96. Path of the Rays in the Polariscope. If one wishes to secure, in reality, the remarkable accuracy which may be reached in the best forms of polariscopes, it is, above all, necessary to pro- vide for a perfectly correct course of the rays through the instru- ment to the eye of the observer. In all accurate polariscopes or saccharimeters, the observer focuses on a field which is made up of two or more separate fields, the illuminations of which are compared with each other. If full advantage is taken of the deli- cacy of this method of reading, the brightness of each separate field must be perfectly uniform, and second, with the apparatus at rest, the degree of illumination on the several fields must remain absolutely constant, or, expressed differently, the dis- tribution of the illumination in the whole field of view must re- main always uniform. Both conditions could easily be reached if the source of light were uniform in intensity throughout. This is, however, never absolutely the case, and it must then be determined how the rays may be passed through the apparatus in order that the two requirements mentioned may be satisfied, notwithstanding changes and irregularities in the distribution of the luminosity of the source of light itself. It must be assumed, however, that every point in the source of light illuminates equally in all directions. This condition will always obtain, if the small surface of the illuminating lens, and only such can be considered here, is kept at a relatively great distance from the source of light, and this will be assumed in what is to come. In order to simplify the following discussion let us imagine first the two polarization prisms C and F of Fig. 28 removed and the two diphragms D and E brought close to the two lenses B and G, so that we have essentially only the luminous 320 POLARIZATION APPARATUS surface A (Fig. 29), the illumina- ting lens B, which is now focused on, and the telescope G H J left. We shall consider first the path of the rays in the case in which \hzpolarization prisms are placed in parallel light rays. In order to realize this condition we must choose an illumination lens of long focus and place the luminous body in its focal plane. A is, therefore, in the focal plane of the lens B. Let L M represent the axis of the apparatus. The degree of brightness under which an ele- ment of surface at the point N of the illuminating lens is seen, de- pends on the cone of rays a N b, supposing the luminous surface at A large enough to begin with, and that all the rays in the cone a N b actually pass through the reading telescope and reach the eye of the observer. Since all the rays in the cone a N b were, be- fore passing the leas B, in the cone aj N b, , therefore the bright- ness at N is proportional to the amount of light which is sent out from the part a, b, of the lumi- nous surface. If we consider a point at O, near the edge of B, its brightness is determined by the cone a O b. Remembering now that A is situated in the focal plane of B it follows that corresponding to the cone a O b we have, before passing the lens, the cone a, O b,, in which, for PATH OF THE RAYS IN THE POLARISCOPE 321 example, a O is parallel with a 2 N. The brightness at O is proportional, therefore, to the amount of light emitted b t L a i from a. 2 b. 2 of the luminous body. In the same manner the brightness at P, a point symmetrical with O, is propor- tional to the light emitted from the part a 3 b 3 of the luminous body. We see, therefore, that for every point of B, the corresponding part a n b n embraces a different portion of the luminous body. It follows, therefore, that unless the area b. 2 a 3 of the luminous body is uni- >] form in brightness, the lens B cannot appear uniformly bright, and further that the distribution of the illumina- tion at B, on account of changes in the luminous body, may be different at dif- ferent times ; but if the luminous sur- face is narrowed down to the portion b 3 a 2 , the luminosity in every part of B is then proportional to the light emitted from b 3 a 2 . Now, whatever the distribution of the light may be in the part b.< a 2 , the lens B will appear uniformly bright, as in principle it is required to be. If the diameter of the diaphragm B is represented by d lt and that of the diaphragm G by d 2 , the distance be- tween the two diaphragms by , reaches a maximum. If, besides this, the light rays pass in the meridian from north to south, the rotation of the plane of polarization is in the negative direction. In order to give an idea of the maximum value of the magnetic rotation, is 0.049", the rotation of a column of water, 20 cm. in length, is cp = 3.0"; SUMMER'S HALF-SHADOW INSTRUMENT 365 that is an amount which must be taken into consideration in exact investigations, certainly at least in the calculation of errors. It is always best to place the axis of the apparatus perpendicular to the magnetic meridian so as to wholly elimi- nate the magnetic effect. It may be remarked in conclusion, that the magnetic rotation in gases and vapors may be wholly neglected. 7. Lummer* s Half -Shadow Apparatus' 121. Description and Theory of the Instrument. As it has not yet been fully described, the illustra- tion in Fig. 50 is only diagrammatic, A and the discussion will be brief. The I j hypothenuse surface A B, of a right ^ angled glass prism ABC, as free as possible from any strain, is silvered and a part of the silver layer removed. Care is taken to have the two fields join each other with perfectly sharp edges, and that they have a position perpendicular to the refractive edge of the prism. The illuminating lens D is placed in front of one of the side surfaces of the prism A C, and between them the polarizing Nicol E, so that the light rays suffer total reflection on the hypothenuse surface A B. The analyzer F and the telescope G H J face the other cathetus surface B C, and the two hypothenuse fields are brought into focus. Imagine the polarizer placed at first so that its principal section is vertical to the plane of reflection ; that is, so that the plane of polarization forms zero angle with the plane of reflection, then the light reflected from the glass and silver surfaces, and passing through B C is rectilinearly polarized and in the plane of reflec- tion. The two fields appear uniformly light or dark with any position of the analyzer F. But if the plane of polarization of the polarizer E be a turned through the angle away from the position parallel to Ztschr. fur Instrum. 15, 293 (1895). L <=> J Fig. 50. 366 SACCHARIMETERS the plane of reflection, then the planes of polarization of the light reflected from the glass and silver surfaces, are sym- metrical to the plane of reflection and the two fields be- have as half-shadow fields, whose half-shadow is equal to a. Of course, a division into three or any number of parts may be made in the simplest manner in the field of view. But it is difficult to secure a glass prism perfectly free from dis- tortion strains, and to prevent such from appearing during an observation. There are noticed, therefore, even with small half-shadows, brighter or darker parts in the field of view. Further, the light leaving the prism is not absolutely linearly polarized, but also elliptically, and this the more strongly, the larger the half-shadow. For the reasons mentioned in 113, the apparatus should not, therefore, be used in exact measure- ments. b. Saccharimeters 122. Simple Wedge-Compensation. The saccharimeters largely used in practice, serve especially for the determination of the strength of sugar solutions. Such a determination may be made with any of the instruments described above, but they require homogeneous light ; to be able to use ordinary white light in practical work was the leading factor which led to the construction of the saccharimeters. This problem was solved in 1848 by the wedge-compensation of Soleil, which is the characteristic part in all saccharimeters. This will, there- fore, be described first, but it may be remarked that a full discussion of the theory of wedge-compensation would lead too far here, and it must be left for a special treatment. As already explained, quartz plates cut perpendicularly to the axis rotate the plane of polarized light, and rather strongly, since a plate I mm. in thickness turns the plane of sodium light about 21.72. In order to simplify the following con- siderations we shall assume any of the polarization instruments as illuminated by homogeneous light. We place the analyzer in the zero position, or as we may briefly express it with Fric, in the position of optical equilibrium. Then bring between the polarizer and analyzer any positively or negatively rotating substance and the optical equilibrium will be destroyed, which may be restored again, without turning the analyzer by adding, SIMPLE WEDGE-COMPENSATION 367 also between polarizer and analyzer, a negative or positive quartz plate of such thickness that the algebraic sum of the rotations of the active body and the quartz is equal to zero. It is then said that the rotation of the body is compensated by the opposite rotation of the quartz. The Soleil wedge-compen- sation is nothing but such a quartz plate, whose thickness, within certain limits, may be changed so L as to compensate any desired rotation within limits. Of two equally thick, plane parallel quartz plates, cut perpen- l dicularly to the axis, the negative one A B C D and the positive one E F G H, Fig. 51, imagine one, say the latter, divi- ded by a cut, vertical to the plane of the paper, into the two wedges, E J K H and J F G K, and that the last is enlarged to form the wedge L M N. The large and the small wedge have now the same wedge-angle. The whole wedge-compen- sation is situated between the polarizer and analyzer, so that the light rays pass in and out from the surfaces B C and E H ; the negative plate and the small wedge are fixed, while the large wedge may be moved along J K. In moving the long wedge, the surfaces E H and M N remain always parallel, so that the two wedges form a positive quartz-plate of variable thickness. In the position of the plates illustrated in the figure, the whole com- pensation gives the rotation zero, as E F = A B. If the long wedge is moved so that L approaches J, the thickness of the positive plate will exceed that of the negative, and the wedge- compensation will produce a positive rotation with which the negative rotation of an active body may be compensated. If, on the other hand, the large wedge is shoved in the opposite direction, so that N approaches K, then the thickness of the negative plate is in excess, and positive rotations which are not too large may be compensated. The greatest possible change in thickness is secured, the longer the large wedge is Fig- 368 SACCHARIMETERS made, and the greater the wedge angle. Practically the long wedge is not shoved along J K, but the two wedges are sepa- rated as shown in Fig. 51 below, and then the long wedge is so moved that the surface M N remains in the same relative position parallel to E H. But as the layer of air remaining between the two wedges effects a lateral displacement of the whole system of rays going through the apparatus, the two wedges are separated no further than is absolutely necessary to permit the free motion of the large wedge. This displace- ment of the rays by the air explains also why the wedge-com- pensation does not give exactly zero rotation when the thick- ness of the two wedges together is exactly equal to that of the negative plate. If the mounting of the large wedge is fur- nished with a scale, the displacement of this, with reference to a fixed vernier, is directly proportional to the change of thick- ness of the positive plate. Now, let us suppose the apparatus illuminated by white light, it being assumed, of course, that the polarizer permits this, which is the case with the accurate Lippich half-shadow instrument, and let the analyzer be turned to optical equilib- rium. Then on inserting the wedge-compensation again between polarizer and analyzer, it will be found that optical equilibrium is restored when the large wedge is turned so as to indicate zero rotation. This appears simultaneously for rays of all w r ave-lengths, because the rotation dispersion is the same for positive and negative quartz-plates. But, at the same time, it is found that because of this rotation dispersion, the rota- tion of such active bodies only, as have the same dispersion as quartz, may be compensated by the wedge-combination. The rotation of quartz-plates in the first place may be compensated, and also that of sugar solutions, since, as shown in 45, the rotation of cane-sugar is very nearly the same as that of quartz. It is because of this fact that the construction of saccharim- eters, which may be employed with white light, is possible. 123. Double Wedge-Compensation of Schmidt and Haensch. In the double wedge-compensation system introduced in saccharimetry by Schmidt and Haensch, the negative plate of Fig. 51 is replaced by two negative quartz-wedges of the same THE SUGAR SCALE 369 Fig. 52. angle, so that the negative as well as the positive plate pos- sesses a variable thickness. This double wedge-compensation is shown in Fig. 52. The smaller wedges are fixed, and the large ones, as before, movable. The thick ends of the large wedges face in the same direction, so as to eliminate as far as possible, the displacement of an air layer between the wedges. Ordinarily, the wedge angles of the negative wedges are made, as nearly as possible, equal to those of the positive. As may be readily seen, it is possible w T ith this combination to compen- sate negative as well as positive rotations. Its great advantage over the single wedge arrangement is found in the fact that it does not give the zero-rotation for a single position only of the large wedges as shown, for example, in Fig. 52, but for any position of the one wedges, a corresponding one of the other may be found for which the rotation is again zero. This follows from the fact that the thicknesses of the positive and negative plates are changed to the same extent by movement in the same direction. A small rotation may, therefore, be com- pensated several times by choosing different parts of the large wedges, so that it is possible to largely eliminate wedge errors by taking the mean of these several determinations. Other facts concerning double wedge-compensation can be brought out in the discussions of the following paragraphs. 124. Preparation of a Sugar Scale for Polariscopes with Circular Graduation. The following paragraphs deal with the sugar scale introduced into German saccharimetry by Ventzke, which is naturally the most important part of the saccharimeter. We enter upon a field in which, unfortunately, much uncertainty still exists, and in part must exist, since experimental investi- gations in this direction have not yet handled the subject with sufficient completeness or accuracy. For this reason, in the following, many points will be touched upon only briefly, and others not at all. Above all, it must be explained at the start, the method will not be indicated here which alone can lead to a scientifically unobjectionable sugar scale. Complete clear- ness and accuracy in this field can be expected only after the 24 370 SACCHARIMETERS conclusion of the extended investigations which have been undertaken by the Reichsanstalt. It is important in practice to have a scale which shows directly in per cent the amount of sugar in the substance investigated. It will be assumed in all that follows, that a constant tempera- ture of say 20 is maintained. Let a grams of pure sugar be dissolved in enough water to make a solution of exactly 100 cc. and then polarize the solution in a 20 cm. tube of the Lip- pich polarimeter, using sodium light ; let the observed angle be ft. We shall consider a as the normal weight and designate the solution made as the normal sugar solution. Next assume a solid or liquid substance which contains along with sugar, only such bodies as are inactive and soluble in water. In 100 grams of this substance let there be present p grams of pure sugar ; then p is the percentage amount of sugar and the determi- nation of this is the problem of optical saccharimetry. Dissolve now a grams of this substance in water, dilute to 100 cc. and polarize as before in the 20 cm. tube ; the observed angle is now y. For aqueous sugar solutions the rotation is known to be proportional to the concentration ; that is, to the number of grams in loocc. of the solution (as this is not absolutely true, the variations and corrections will be discussed later). As the concentration of the normal sugar solution is , and that of the second solution o.oi a p, then it follows that ft : y : : a : o.oi a p, from which p~ , or also, ft : y : : loo : p ; that is, the two angles of rotation are related, as are the percentage strengths of the original substances, the strength of the pure sugar being, of course, 100 per cent. Now, if we take ft = loo, y will be equal top, and in polarizing, we read directly the desired percentage strength. Therefore, in order to find the percentage amount of sugar in a substance, it is only necessary to polarize in a 20 cm. tube, a solution of the substance which contains the normal weight dissolved to make loo cc. As seen, this normal weight may be chosen quite arbi- trarily ; the length only of the sugar scale depends on it, as this is proportional to the normal weight. We shall apply this sugar scale of the polarimeter now to the saccharimeters. SUGAR SCALE FOR SACCHARIMETERS 371 125. Preparation of the Sugar Scale for Saccharimeters. Let the analyzer of the polarization apparatus, using white light, be brought into the position of optical equilibrium and insert the single Soleil wedge-compensation between the analyzer and analyzer diaphragm, so as to be able to compensate the positive rotation of sugar solutions. We assume the constant tempera- ture of say 20, and, for the moment, perfect equality in the rotation dispersion of sugar and quartz. Then let the large wedge be so moved that optical equilibrium still obtains, and at some convenient point on the setting of the large wedge make a mark to be designated as o. This mark prolonged to the fixed framework of the mounting, gives the fixed index position. In this way the zero-point of the apparatus is deter- mined. Then the normal sugar solution in the 20 cm. tube is inserted, and the large wedge is moved until optical equilib- rium is secured again. A second mark is now made on the large wedge, opposite the fixed index mark, and this is desig- nated loo. In this manner, the extremely important TOO mark of the instrument, upon the accuracy of which every- thing depends, is established. If, next, the interval between the o mark and the 100 mark is divided into a convenient number of exactly equal divisions, the instrument is ready for use. As the displacement of the large wedge is exactly pro- portional to the amount of rotation of the inserted sugar solu- tion, it follows that the percentage strength of pure sugar may be again read off directly on the scale. If the normal weight of a substance is dissolved in water, diluted to 100 cc., and if the solution is polarized in a 20 cm. tube, the number read on the scale gives the percentage amount of pure sugar present. Imagine next a positive quartz plate, perfectly plane paral- lel, and cut perpendicularly to the optical axis, and of such a thickness that it gives the 100 point in an accurately gradu- ated saccharimeter. By the aid of such a plate, it will now be much easier to control the 100 mark of instruments than through the use of a normal sugar solution, the exact prepa- ration of which requires the expenditure of much time. A quartz-plate so made is called a normal quartz-plate, and, as a matter of fact, all saccharimeters are in practice graduated by it. Normal quartz-plates are to be defined, naturally, by their 372 SACCHARIMETERS rotation and not by their thickness, because the latter cannot by any means be as accurately measured as the rotation. It need hardly be remarked that saccharimeters may, of course, be graduated with quartz-plates which do not polarize exactly loo, as long as it is known to what value on the sugar scale they correspond. 126. The Ventzke Sugar Scale. The normal weight has not been numerically defined in the preceding paragraphs, because it may be arbitrarily chosen, and in the course of time has been frequently changed. At the present time, only two sugar scales are found in common use, the German or Ventzke scale, and the French scale, of which the first only will be discussed. The French scale 1 must be considered as quite unsatisfactory as its hundred point is defined by the rotation of a quartz-plate i mm. in thickness. This use of the thickness of a quartz- plate in the definition of a sugar scale is 1 not alone wholly un- necessary, but it is also very unpractical, because as often as the absolute rotation of i mm. of quartz is differently deter- mined, it is necessary to correspondingly change the normal weight ; this explains why it is that the French normal weight has been changed at least once in every ten years. Ventzke' 2 proposed, first, a method for preparing the normal sugar solution, which was intended to render the use of a balance unnecessary. He defined as the normal sugar solution, a solution of pure sugar in water which should have at 17.5 the specific gravity of i.ioo, referred to water at 17.5. To determine then the polarizing sugar of any substance, it would be simply necessary to prepare a solution of it of this density by aid of an aerometer. But naturally this method could not give exact results, because the salts in the cane-sugars to be investigated have an effect, and have usually a density differ- ent from that of sugar itself ; it was, therefore, soon abandoned. But as the 100 point of many saccharimeters had already been fixed by aid of the normal sugar solution of i.i sp. gr., and 1 The French scale is described in this way : That sugar solution lias the- normal weight whose rotation in a 20 cm. tube is equal to that of a quartz plate i mm. in thickness. The normal weight thus defined ha^ suffered many changes in the course of time, as it has vark-d between the limits, 16.02 and 16.471 gram. These two weights differ by about 2.8 per cent. - Vent/.ke : Krdm. Jour, fur prakt. Chem., 35, 84 (1842) ; a8, m (1843). SUGAR SCALE FOR SACCHARIMETERS 373 as it was not desirable to change the scale once introduced, the concentration of the Ventzke normal solution at 17.5 was then determined. Investigations showed that 100 cc. of such a solution contains 26.048 grams of sugar weighed in air with brass weights. The normal weight should be then 26.048 grams. If then this weight of pure sugar is dis- solved to make 100 cc. and polarized in the 20 cm. tube, the 100 point on the Ventzke scale is again found. As this method was in use long before the Mohr cubic centimeter was sug- gested, the number 26.048 must certainly have been applied to true cubic centimeters. After the introduction of the Mohr graduated flasks, the instrument makers, especially Schmidt and Haensch in Berlin, began to employ a solution of 26.048 grams of sugar in 100 Mohr cubic centimeters in fixing the 100 mark of instruments. For a number of years, then, the defi- nition of the normal sugar solution has been this : That sugar solution has the normal strength which contains at 17.5 in 100 Mohr cubic centimeters, 26.04.8 grams of cane-sugar weighed in the air with brass weights. As the Mohr cubic centimeters have continually given rise to error, it was finally determined to go back to true cubic centimeters. This last definition of the normal sugar solution will then be changed so as to correspond to true cubic centi- meters. The following is the definition of the Mohr cubic centi- meter. 1 If 100 grams of water are weighed in the air with brass weights at 17.5, the volume is 100 Mohr cubic centimeters. If we take into consideration, the weight of the air displaced in weighing and the specific gravity of the water at 17.5, it follows that loo Mohr cc. == 100.234 true cc. 2 Therefore, the definition of the normal sugar solution for true cubic centi- meters is this : A sugar solution has the normal concentration when it contains 25.9872 grams of cane-sugar, weighed in air with brass weights, dissolved in water to make 100 cubic centi- meters at 17.5. As again appears from the lack of agree- ment in the specific gravity of sugar, and the substances whose sugar content is to be determined, it would be neces- 1 Mohr: Chemisch-analytische Titrirmethode, 44-50 (1886). - The Mohr flasks made by the firm of Schmidt and Haensch hold, as experi- ments of the author show, in fact within 0.03 per cent, of 100.23 true cubic centi- meters. 374 SACCHARIMETERS sary to reduce all weights to vacuo to secure a perfectly accurate result. As the specific gravity of pure sugar is 1.6, the 25.987 grams in real mass is 26.003 grams. The normal sugar solution contains then 26.003 true grams of sugar dissolved at 17.5 to make 100 true cubic centimeters. But in what follows we shall retain the air weighings, as the error made by neglecting the reduction amounts seldom to 0.05 per cent., which has no importance in practice. Finally, it must be added that the variations in the normal weight from changes in the density of the air amount to about 0.0007 grams for 26 grams of sugar, and are likewise unimportant in practice. Retaining then the long established definition of the 100 point, the preparation of the Ventzke sugar scale, and the determination of sugar in a substance are as follows : A normal sugar solution is made by dissolving 26 .04.8 grams of pure cane- sugar, weighed with brass weights in air, in a 100 cc. Mohr flask at 77.5, and this is polarized at 17.5 in a 20 cm. tube in the saccharimeter, care being taken to keep the wedge-compensa- tion also at 17.5 ; this establishes the 100 mark (100 V) of the saccharimeter . Then by weighing and dissolving 26.04.8 grams of an impure sugar under the same conditions, and polarizing in the same tube, the compensation still at 77.5, the Ventzke scale gives directly the amount of pure sugar in per cent. Now, in regard to the accuracy with which the saccharim- eters in the hands of sugar chemists agree with this definition of the true 100 mark, nothing definite can be said. 1 Certainty with reference to the accuracy of the 100 point on the Ventzke scale will be established by the investigations of the Reichsan- stalt. 127. The 100 Point of the Saccharimeter. Suppose the 100 point of a saccharimeter accurately determined by means of a correct normal sugar solution, and then a normal quartz-plate made which polarizes exactly 100 V in the so-graduated instrument. Then by aid of this plate, new saccharimeters may be much more accurately graduated than with freshly prepared sugar 1 The work of Nasini and Villavecchia (Sul peso normale pei saccarimetri, Puhhl. d. Iab. chim. centr. d. Gab. 1891. Oesterr.-Ungar. Zeitschrift f. Zuckerindustrie, I Heft, 1892) adds nothing to the discussion. THE 100 POINT OF THE SACCHARIMETER 375 solutions. In the first place such a solution would always vary more or less from the correct normal solution; and, secondly, it would be only with the greatest difficulty that the normal temperature required by the definition could be main- tained, especially in the quartz compensation. But the normal quartz -plate remains always applicable and correct, and its rota- tion in the saccharimeter is independent of the temperature, as long as care is taken to have the plate and compensation wedge at the same temperature, because the temperature coefficient of positive and negative quartz-plates is the same. For a long time, therefore, all saccharimeters have been graduated with such normal quartz-plates, which goes to explain the remark- able agreement in the 100 points of the instruments purchased from dealers. The loo points on the saccharimeters made by Schmidt and Haensch in Berlin, which are now used in all parts of the world, seldom differ by more than 0.05 to o. i V, and the dif- ferences are usually smaller. It was, therefore, very com- mendable that the firm of Josef-Jan Fric in Prague, which also began the construction of saccharimeters, should bring their instruments into exact agreement with those of Schmidt and Haensch. How then may the 100 point on these saccharim- eters be defined ? This can be most simply done through the rotation of the normal quartz-plate for some definite light, as the sodium ray. Experiments by Schonrock 1 with four quartz- plates of about 100.3, 93-5. 9 I -7 and 75-4 V in a Schmidt and Haensch half-shadow saccharimeter, and experiments by Josef-Jan Fric 2 with a normal quartz-plate and a plate of about 99.7 V, in a large number of saccharimeters, have shown in complete agreement that the normal quartz-plate at 17.5 rotates the sodium ray 34.68 o.o2. 3 It should be remarked that the quartz-plates and compensations were kept at the same temperature in the saccharimeters during the experiments, and that the result is independent of the composition of the white light used, as the differences for different lamps remained below 0.03 V. Therefore, we can take for quartz-plates : 100 Ventzke = 34.68 circular degrees for D at 17.5 C. 4 1 Schonrock : Ztschr. fur Instrum., 16, 242 (1896). 2 Personal communication to the author. 3 The optical center of gravity of sodium light = 589.3 MM- * Or, 100 V = 34.69 arc for D at 20. 376 SACCHARIMETERS The number 34.68 is called the factor of reduction. If a quartz-plate is polarized in a saccharimeter with its tempera- ture kept equal to that of the wedge-compensation, and if it shows a V, then the quartz- plate will have a rotating power for sodium light equal to 0.3468 a dr 0.0002 a circular degrees at 17.5. More about the reduction factors for substances other than quartz, will be found in 129 and 153. If the rotation of a quartz-plate i mm. thick, at 17.5, for sodium light is 21.714, the thickness of the normal quartz-plate must be about 1.597 mm - The following may be said about the dimensions commonly chosen in the quartz wedge-compensations used in saccharim- eters. The wedge angle of the quartz is usually taken at 3, so that the distance between the o and 100 points on the scale is about 30.5 mm. The whole interval is divided into 100 equal parts, and by a vernier, one-tenth Ventzke may be read off. In the simple wedge-compensation, the two wedges are usually positive, and the compensation plate negative and about 4.3 mm. in thickness. If the large wedge is then moved toward the 100 point, the thickness of the two wedges de- creases. 128. Testing the Saccharimeter Scale. As the agreement of dif- ferent saccharimeters at the 100 point leaves little to be desired, it may be asked now how the scale of a single instrument may be tested between the zero-point and 100 point. The error in graduation in the scale and vernier should remain everywhere below 0.05 V. The ivory scales formerly commonly used have been abandoned, and properly, because their length is influenced by the moisture in the air ; the changes amount to 0.3 V and more. The nickelin scales now commonly used are free from this error, and are also independent of changes in temperature as the change in length between o and 100, with a temperature fluctuation of 10, is only about 0.01 V. Furthermore, the wedge should give correct results at all parts of the scale; that is, it should be so made that when it is moved through the nth part of the whole distance ( 100 V) , the change of rotation produced in the compensation should be the wth part of the rotation of the normal plate corresponding to the loo point on the scale. This would be the case exactly if THE 100 POINT OF THE SACCHARIMETER 377 the large wedge were perfectly plane on both sides, optically homogeneous, and its motion without error. Errors in the construction and character of the small wedges and the com- pensation plate are of less importance as they are fixed in posi- tion, and because the wedge-compensation is close to the analyzer, while the polarizer is focused with the telescope ; the rays from the whole field of view 7 pass through the two small wedges and the plate and uniformly over their whole ex- tent. As regards the large w r edge, it is possible with sufficient care to make its surfaces so plane that the error occasioned here in the saccharimeters is inappreciable. Aso. i V cor- responds about to 0.0016 mm. in thickness of the w r edge, the notion has become common that the surfaces bounding the wedge cannot vary 16 /i ooo of a millimeter from absolute planes, if it is to give readings correct to within o. i V ; but as it is impossible to make the whole facing surfaces of the wedge, 30 mm. long, as perfectly plane as this, the errors often amount- ing to tenths of Ventzke degrees are supposed to be explained. But all of this is based on error ; the consideration would be true only when a single ray were dealt with. In reality, how- ever, we work with a whole field whose length is about a sixth of the whole wedge-length. If it is considered further that the wedge is placed right at the analyzer, it may be under- stood that a slightly convex or concave bounding surface on the wedge hurts nothing. Errors are occasioned only when the radius of curvature has not a constant sign ; that is, when points of inflection are present, but such surfaces are seldom met with. The optician can, in fact, prepare the wedge so that no errors need result from faults of construction. Then how may it be accounted for that variations amounting to several tenths of a Ventzke degree are almost the rule ? The answer to this question is simple, but not very encouraging. Nearly all errors which are found are due to the optical impurity of the quartz wedges. 1 It must, unfortunately, be admitted that quartz is a very poor material, so that one rarely comes into possession of optically faultless plates of i to 2 square centi- meters of surface ; it must even be considered a piece of good fortune to possess a pure quartz-wedge 3 cm. long and of cor- 1 See " Die Thatigkeit der physikalisch-technischen Reichsanstalt " under Brod- hun and Schonrock ; Ztschr. fur Instr., 17. (1897). 378 SACCHARIMETERS responding thickness. If now we examine the wedge-com- pensations further in respect of their optical purity, it must seem in many cases impossible that with such material such delicate observations could be made, when the mean error of an adjustment is only about =fc 0.03 V ; yet this is the case with the half-shadow saccharimeters. This is explained by the fact that the wedge-compensation, fortunately, is located at the analyzer, so that its image, with a correct path of the rays, according to 96, coincides with the plane of the pupil of the eye and the impurity therefore cannot be recognized. Now, if the large wedge be moved new and different parts are then passed by the rays, and the rotations of these may differ considerably from each other. 1 But as the irregular parts of the wedges appear only gradually in the portion of the field occupied by the rays, the errors in a saccharimeter do not show sudden changes from point to point, but are subject to gradual variations, as the author has had opportunity of observing. Therefore, if the errors are determined for say 5, 10, 15 V and so on, and if these are plotted on coordinate paper, with the Ventzke degrees as abscissas and the errors as ordinates, it will be possible to find from the curve passed through the tops of the ordinates, the errors for any intermediate points. The simplest and at the same time the best method of finding errors is through the examination of different parts of the scale with good quartz plates. But as this is not always practicable for the individual owners of saccharimeters, other methods of error determination have been devised. Accurate results are alone furnished by the control observa- tion tubes made by Schmidt and Haensch and described in 163. These give more accurate results than the preparation of a large number of solutions of proper rotation for the reason that in the control tubes, relative measurements only need be made as soon as the 100 point is once for all accurately deter- mined. The control observation tubes are constructed on the 1 The author onre had the opportunity of examining two plane parallel quartz- plates about 5 mm. thick and well made, which, when polarized in a half-shadow apparatus with sodium light and large half-shadow, that is, with rather bright field, gave pretty constant results, but when the plates were turned in their planes the angles of rotation assumed all values between o and 90. Although quartz-wedges of corresponding bad quality could hardly be found, it cannot be denied that the error of a saccharimeter must be largely ascribed to the optical impurity of the wedges. THE 100 POINT OF THE SACCHARIMETER 379 principle of making the length of the rotating liquid column variable and the variation measurable. As the deflection of the plane of polarization is proportional to the length of the liquid layer, the rotation decreases in the same degree as the length of the column. The examination with the control tube is then made in the following manner : To avoid unnecessary corrections, the zero of the vernier is brought into exact agree- ment with the zero of the scale, using the vernier correction in aid of this. Then the 100 point must be at exactly the right distance from the tw 7 o zero points, which, according to 127, is nearly always the case. Then the control tube is filled with a sugar solution whose rotation is somewhat greater than 100 Ventzke, when the tube is pulled out to its full length. The tube is then shortened until the rotation gives exactly a = 100 V, when the length / is measured accurately . If the conditions of experiment, and most important, the tem- peratures of the wedge-compensation and the sugar solution, are kept constant and the length of the tube then shortened to /', the rotation corresponding to this length is b = a I' II. If instead of reading off b, the value read is exactly b* V, then the error in the immediate neighborhood of the point ^ is given by/ = b b^ = (a I'll) b^ ; this error, /, is to be added algebraically to the observed value b^ in order to obtain the true value of d l in Ventzke degrees. In this way, the errors of any desired number of points on a considerable portion of the scale may be found, and the length of this is limited simply by the extent to which the tube may be shortened. With a properly constructed control tube, using a single solution, it is possible to determine the errors from 100 V down to 55 V. Then the tube is filled with a new solution whose rotation, at full length, is something over 55 V, and the point 55 is taken as the new starting point, because its error has been already determined, and then observations are made as before until the error at the 5 mark is reached. The following table shows how, with five different properly chosen solutions, the errors on the whole scale may be found i 1 1 Absolutely pure sugar is not necessary for the purpose. SACCHARIMETERS The tube length variable between 42 cm. and 22 cm. Cane sugar solution. No. Concentration, or number of grams of sugar in 100 cc. of solution. Starting point. o V. Errors found at the points. I 12.53 100 95, 90, 85 ..-60, 55 2 6.89 55 50, 45, 40, 35, 30 3 3-76 30 25, 20, 16 4 2.00 16 15, 10, 9 5 .I3 9 5 Of course, the accuracy of the tube scale itself must have been previously determined. By plotting the errors of the ' saccharimeter scale on coordinate paper the error curve is found. As may be readily seen, this method of rinding errors, in case it is expected to give really accurate results, requires quick and very close work. It is, therefore, recommended to make on different days two complete series of determinations to secure a picture of the accuracy obtainable. If this is sat- isfactory the mean curve is taken as the final one. Although the testing of a saccharimeter scale requires some pains, one should not hesitate to do the work, because the errors remain constant and are absolutely independent of time so long as no injury has hap- pened to the optical mechanism of the instrument. An error curve once accurately established holds good always. We turn now to a consideration of the double wedge-com- pensation of Schmidt and Haensch (123) on the value of which different views have been held. The zero point is usually found at the thinner end of the long wedge, so that the working wedge is negative and the control wedge positive. To avoid unnecessary zero point corrections the control wedge is placed exactly at o and optical equilibrium is secured by moving the working wedge ; if the zero point of the working scale does not coincide exactly with that of the vernier the dif- ference is corrected as before by means of the micrometer screw below the vernier corresponding to the scale. Then the working wedge is placed at 100 and a test is made as accurately as possible of the correctness of the 100 mark. If now optical equilibrium is again established by aid of the control wedge, THE 100 POINT OF THE SACCHARIMETER 381 this should stand exactly at 100. If this is not the case the scale has been carelessly made, and the apparatus should be simply sent back to the manufacturer. Naturally the two scale lengths are usually not the same, which is not necessary. If the two wedges are accurately made it must follow that when one of them is put on any point of its scale optical equilibrium must obtain w r hen the other is placed at the same point on its scale. If in this way the two wedges are tested from 5 to 5 and if the differences in readings actually remain below 0.03 V., it may be said with great prob- ability that the wedge-compensation does not possess any ap- preciable error, because it would be by extraordinary chance that any errors due to optical impurities in certain places in one wedge would be exactly compensated by corresponding errors in the other. In addition to all this, in order to be cer- tain, it is well to test at least a few points by aid of the control tube. As a rule, however, it will be found in testing the two wedges as above that the differences in the readings in certain parts of the scale amount to some tenths of a Ventzke degree, so that an accurate determination of errors is here also re- quired. For this purpose the reading differences from 5 to 5 V. must be accurately found in this way. The control wedge is placed at 95 exactly, and five adjustments at optical equilibrium are made with the working wedge, from which the mean is taken and subtracted from 95, to give the reading dif- ference. In this way results are obtained down to the 5 mark. Then the whole series of observations is repeated by placing the working wedge exactly at 95 and so on, and se- curing optical equilibrium by aid of the control wedge. The two series of observations should give the same differences with opposite signs, within the limits of errors of experiment, which is to be remembered, in general, in all error corrections. From each two corresponding differences the mean of the absolute values is taken which is to be used in the further calculations, with proper sign attached. The control wedge is next placed exactly at o and the complete determination of errors in the working w r edge is made by use of the control observation tube as fully explained above. From the curve of errors for the working wedge and the table of adjustment differences for the 382 SACCHARIMETERS two wedges the error curve of the control wedge may be found in a way easily recognized. The latter may be found also directly by aid of the control tube and positive sugar solutions, by placing the working wedge at exactly 100 and finding the errors in the control wedge from 5 to 95. This furnishes, at the same time, the best control of the accuracy of all the ob- servations made. If now the rotation of a solution is found with one wedge and controlled, after taking out the solution, with the other wedge, then after adding the corrections from the error curves the two wedges should give exactly the same result. // follows, therefore, that correct observations may be made with the double wedge combination and an actual complete control secured only when the error curves of the two wedges are known. It appears, therefore, that as accurate work may be done with the single as with the double wedge-compensation. 129. Observation of Solutions in the Saccharimeter In the fol- lowing considerations we shall keep the half-shadow saccharim- eters in mind as they have completely displaced the color in- struments. As the rotation dispersion of sugar does not agree absolutely with that of quartz, a slight but unimportant dif- ference in the colors of the field of view is found in testing high polarizing sugar solutions. Although this does not inter- fere with the accuracy of the reading for a practiced observer, the adjustments of different observers with different degrees of color sensitiveness, may vary considerably from each other. To eliminate the colors either a plate of potassium dichromate or a cell with solution of potassium dichromate is placed be- tween the light and illuminating lens ; a dichromate plate in the ocular is not used. As regards the changes in the rotation of sugar solutions for white light of different kinds of lamps it is found that appreciable differences occur only when ob- servations are made at one time with, and at other times with- out, dichromate plates. 1 Still this point requires fuller syste- matic investigation. With reference to testing the saccha. rimeter scale by aid of the control tube it must be further re- marked that the illumination must not vary during determina- tion of the curve of errors. The curve finally found is, how- ever, independent of the illumination, as the error in the 100 1 /tschr. fiir lustrum., 16, 243 (1896). SACCHARIMETER READING 383 point is placed equal to zero. If a change is made from one light to another, and a change in the 100 point follows, that is, if the rotation of a normal sugar solution varies by a V., then in a corresponding manner the point b of the scale would change by o.oi a b V. As complaints about continuous changes in the zero points of instruments are still heard in practice, reference must be again made to 96. These changes may be very easily avoided. The focal length of the illuminating lens must be chosen so as to be equal to half the distance between this lens and the ana- lyzer diaphragm, and the source of light is placed as far away from the illumination lens as the latter is distant from the analyzer diaphragm, or better so that after putting the absorp- tion cell in position a sharp image of the source of light will be thrown by the illuminating lens on the analyzer diaphragm. If this is the case observations may be made day after day without disclosing the slightest change in the position of the zero point, since in the saccharimeters the polarizer and analy- zer are fixed in definite positions, or at least should be ; a lat- eral displacement of the source of light cannot be followed by a zero point change. If the Ventzke reading in the investigation of a sugar solu- tion is to be converted into circular degrees referred to sodium light, this -may be done as explained in 127 by aid of the re- duction factor 0.3468, although this properly applies to quartz plates only. An absolutely accurate conversion is not always possible, because of the effects of temperature and the varia- tions in the rotation with the illumination and color sense of the observer. Landolt 1 has actually found in the observation of a cane-sugar solution in a Schmidt and Haensch half-shadow saccharimeter, with a gas lamp, that a rotation of 100 V. cor- responds to the rotation of 34.65 0.05 for sodium light. But if it is required to accurately measure the rotation of a sugar solution for sodium light this must be done in a polar- imeter actually illuminated by sodium light. 130. Effect of Temperature on the Saccharimeter Reading. In or- der to give an idea of the error w r hich results through temper- ature changes in the determination of cane-sugar the errors from several sources will be discussed on the basis of a tem- 1 I^andolt: Her. d. chem. Ges. ai, 194 (1888). 384 SACCHARIMETERS perature variation of 10. In the saccharimeter itself the wedge-compensation only is variable with the temperature ; the zero point is constant, but all other points change in value. As we can take the temperature coefficient of quartz (and sugar also) as constant through the interval considered, for all wave-lengths, that is, the rotation dispersion as independ- ent of the temperature, a simple calculation shows that a tem- perature variation of 10 changes the 100 point (100 V.) about -f 0.15 V. From this it will be recognized that the temperature of the compensation must be pretty accurately known if one expects to polarize within a few hundredths Ventzke. 1 Now let us consider the polarization of a cane-sugar solution in the saccharimeter, and in order to have a definite case in mind assume that we have a normal sugar solution in a 20 cm. tube, which must therefore polarize about 100 V. If the tube is of glass its length will increase with the temperature, and, therefore, the rotation of the sugar solution also. A tempera- ture elevation of 10 corresponds to a change of -fo.oi V; therefore, the change from this cause is inappreciable in ob- servations. But at the same time the concentration of the so- lution decreases with the increase in temperature, and the ro- tation must therefore decrease. As the coefficient of expan- sion of a normal sugar solution at about 20 is 0.000291, the increase of temperature through 10 produces a change in ro- tation of o. 29 V. Besides this the specific rotation of cane- sugar decreases with increase in temperature, so that the angle of rotation still further decreases ; the older observations of Tuchschmid, Seyffart, and Andrews, which gave the specific rotation of cane-sugar as constant at different temperatures, are not correct, since the decrease in rotation with increase in temperature is even quite appreciable/ On account of this 1 As regards the temperature of the wedge-compenation we are very much in the dark with the present o.n-tru tion^, ;is tin- temperature of the rather thick wedge and compensation plates follows variations in the room temperature very slowly. It may, therefore, be recomnx ndcd t<> pl:u-<- the- coniiirnvation box in ;i routined space enclosed by non-conducting materials :md have a tin rmometer, graduated in whole degrees, extend into the box. It is also well to have the light passing into the apparatus and that reaching the reading mirror also pa^s through a cell filled with water to absorb the heat rays and yield a cold light. 1 See " Die Thiitigkeit der physikalisch-technisi-hen Reichsanstalt " under Scln'in- rock, /Aschr. fiir lustrum., 17, (1897). [Hut, contrary to the statement in the text, this was previously accurately pointed out by Andrews, Technology Quarterly, May, 1889, p. 367. See also later, under Constants of Rotation. Tr. J SOLEIL-VENTZKE INSTRUMENT 385 change in specific rotation the rotation of the normal sugar solution is decreased about 0.21 V. more by the 10 in- crease in temperature. The whole change then is about 0.50 V. From this it is seen how necessary it is to control the tem- perature in exact wprk. It must be further noticed that the errors due to the wedge-compensation and the sugar solution are to be added to each other, as indeed the rotation of quartz increases while that of sugar decreases with increasing tem- perature, so that the whole error of the polarization of the normal sugar solution will be 0.65 V. If now it is expected to determine the loopointofa saccharimeter to within dzo.o>j V. by aid of a normal sugar solution, then, aside from all other sources of error, and especially those in the preparation of the so- lution (temperature again /), the temperature of the compensa- tion and of the solution in the tube should not differ from that in the definition of the 100 point by more than 0.5 C. Now, at least, one should be able to recognize how difficult the gradua- tion of a saccharimeter by aid of a normal sugar solution really is. As further, the mean adjustment error of the present half- shadow instruments is only about =0.03 V., and as these saccharimeters are used only in technical work there would be no sense in increasing their delicacy still further, since, at the present time, polarizations are made with a degree of accuracy which is wholly illusory. i. The Soleil- Ventzke Saccharimeter. 131. Description of the Instrument. In the description of the different saccharimeters to follow, we shall be very brief, since it is only necessary to add a wedge-compensation to any one of the polarimeters described above to have the corresponding saccharimeter. As explained, the first saccharimeter was made by the Paris optician, Soleil, and was later improved by Soleil and Duboscq. This is the so-called color saccharimeter, and is shown in Fig. 53. It may be made by setting the Robi- quet polariscope, described in 100, at the position of optical equilibrium and then inserting a simple wedge-compensation between polarizer and analyzer. The light, therefore, passes the following parts in going through the apparatus from right to left ; first, the regulator of the sensitive tint not described 25 3 86 SACCHARIMETERS above under the head of the Robiqtiet apparatus, and which must be referred to here, consisting of a nicol, A, which may be rotated, and a right or left quartz-plate, B, which is ground perpendicular to the axis, the illuminating lens C, the polari- zer D, the Soleil double plate E, the wedge-compensation F, the analyzer G, and the telescope H. As in the Robiquet apparatus, adjustment is made on the transition tint by move- ment of the long wedge ; that is, uniform color of the two fields is secured. In 102, the drawback of the Robiquet instrument, that after inserting the rotating substance the transition tint no longer has the same color observed in finding the zero point, is referred to. This fault appears in much less Fig- 53- degree in the saccharimeter because the rotation dispersion of the active substance is largely compensated by the quartz wedges. But, in order to keep the shade of the transition tint as nearly the same as possible after putting in the active sub- stance, even if somewhat colored, and in order to give the observer the power of choosing the shade for the transition tint, for which the eye is the most sensitive, the so-called regu- lator has been added to the saccharimeter. The light polari- zed by the nicol A passes into the quartz-plate B and suffers rotation dispersion, and on account of this, and the relation of the position of A to B, certain rays will not pass through the latter, or will pass with diminished intensity. By rota- ting the nicol A, the tone of the transition tint may, there- fore, be changed at will. The rotation of the tube which contains the regulator A B HALF-SHADOW SACCHARIMETERS 387 is accomplished by aid of a spur-wheel and pinion, the latter being worked by a rod attached to the button J. The brass frame holding the large wedge may be moved horizontally, and for this purpose is furnished with a lateral rack in which works a pinion wheel controlled by the button K. An inclined mirror, L, serves in reading the Ventzke scale, the image of the scale being thrown into the tube M which contains a magni- fying lens. In the illustration, a screw is shown at N by aid of which the analyzer may be turned and brought into the proper position with reference to the polarizer ; this movement must be made, however, only by the instrument-maker in adjusting the apparatus, so that the observer has nothing to do with the screw N. As regards the use of the instrument, reference is made to 97 and 102. The mean error of a reading is about 0.2 V. As the adjustment to equality of tint in the field of view is not possible with color blindness, inaccurate with deficient color sense, and for most eyes much more tiresome than the adjustment to uniformity of illumination, the much more sen- sitive half-shadow instruments have properly displaced the color instruments completely. It is quite in vain to attempt to make the color instruments more sensitive by giving a con- centric form to the Soleil double plate ; the part of these in saccharimetry has been played and for good. 2. Half- Shadow Saccharimeters 132. Construction of the Instruments. The half-shadow polar- iscopes contain, always, an illuminating lens, a polarizing mechanism, the analyzer, and a telescope. In the construction of saccharimeters, the following polarizing arrangements are usually employed. The firm of Schmidt and Haensch uses, in general, the Jellett polarizer (109) for the saccharimeters with double field; the double field Lippich polarizer (114) could, of course, be applied as well. In the saccharimeters with triple field, the triple Lippich polarizer is employed (115). If, now, either kind of polarization apparatus is adjusted to uniform shadow on the fields of view, and a simple or double wedge-compensation is introduced between the analyzer and the analyzer diaphragm, the corresponding half- 388 SACCHARIMETERS shadow saccharimeter is produced. The light passes then through the following optical parts : the illuminating lens, polarizing mechanism, wedge-compensation, analyzer, and tel- escope. Polarizer and analyzer must have a fixed position and be properly protected against any accidental displacement, as is the case, for example, in all the instruments made by Schmidt and Haensch. It must be considered an error of construction to give a saccharimeter a variable half-shadow. The zero point of the instrument varies naturally with the half -shadow. This should be made 5 to 8. The mean error of an adjust- ment for a saccharimeter with double field is about rb 0.06 V, and for the saccharimeters with triple field, about 0.03 V. Below a very condensed description of the commonly used types of saccharimeters will be given, with especial considera- tion of the mechanical construction. 133. Half-Shadow Saccharimeter with Single Wedge-Compensation and Double Field (Schmidt and Haensch). This apparatus is shown in Fig. 54. The movable quartz wedge is mounted in a frame .B 54- attached to a lateral rack in which a pinion wheel controlled by the button A works. The reading of the scale and ver- nier is accomplished by aid of the inclined mirror in B, and lens in the tube C. 134. Half-Shadow Saccharimeter with Double Wedge-Compensation and Triple Field (Schmidt and Haensch) ,i n this instrument, shown in Fig. 55, the black set screw A moves the working wedge while the yellow screw B moves the control wedge, and in such a manner that by aid of the lens C, the working scale is seen above and the control scale below. As regards the illu- mination of the scales this is accomplished by means of the BEET-JUICE SACCHARIMETER 339 mirror D, which is attached to the ball and socket mechanism at E, and which may be so turned as to receive light from the D Fig. 55- illuminating lamp and throw it through the matted upper sur- face of the glass plate F on the scales. The small screen G serves to protect the scales from outside lights. 135. Beet-juice Saccharimeter with Limited Enlarged Scale. 1 This instrument, the front part of which is illustrated in Fig. 56, was constructed by Schmidt and Haensch from a suggestion of Stammer. It is distinguished from the previously described instruments only by its limited scale extending from o to 35 V. A reading of higher degrees is not necessary, since it is used only for the determination of sugar in beets. In order to carry out the polarizations in such cases, where in a short time the largest possible number of tests must be made, and 1 Stammer : Ztschr. fiir Riibenzucker-Ind., 37, 474 (1887). Schmidt and Haensch : Jbid., 43, 1040 (1893). 390 SACCHARIMETERS to avoid the delay of reading by the telescope, the instrument is furnished with an enlarged scale which permits an easy reading to o. i of i per cent, with the unaided eye, and even at a distance. The mechanism consists of a segment of a cir- cle attached in upright position to the wedge-compensation, and on which the beet-juice graduation of o to 35 V is Fig. 56- marked. In the'center of the segment there is a small drum, R, which is kept in position by a spring ; this drum is attached to the movable wedge by a thin steel chain in such a manner that when the wedge is moved to and fro by the button K, the drum R is brought into motion, and then the long pointer joined to it moves over the graduation of the segment. If the diameter of the drum bears a certain relation to the length of the movable scale, then the circular segment graduation gives exactly the whole and tenths of per cent, of the scale. To secure accurate adjustment the apparatus is brought to the point of half -shadow uniformity ; if the scale then does not begin at o a correction is made by a key to be attached to V. In this position the pointer of the circular graduation should stand at o also ; any necessary correction of this zero PETERS' HALF-SHADOW SACCHARIMETER 391 point may be made by a slight movement of the small screw fastening the chain to the wedge, which is accomplished by aid of the little set lever 5". Any movement of the wedge ex- pressed in per cent, must be exactly duplicated on the circular graduation. If this is not the case a correction is made by turning the nut attached to R a little on or off, which has the effect of slightly increasing or decreasing the diameter of R, and, therefore, of making the deflection of the pointer less or more, to correspond. The whole adjustment must be revised from time to time. 136. The Half-Shadow Saccharimeter of Peters. 1 This instru- ment, shown in Fig. 57, is distinguished from the foregoing in several points which will be explained. It is a half- shadow Fig. 57- saccharimeter with Lippich polarizer and double wedge-com- pensation, and is supported on two unusually stable feet, which prevent upsetting the instrument. In the brass shell A, there is a wide glass tube with end plates, like a polarizing tube, which may be filled with a solution of potassium dichro- mate. The pointer at B is used to turn one of the nicols of the Lippich polarizer ; the scale is placed below here in order 1 Peters (Berlin): Ztschr. fur Riibenzucker-Ind., 44, 221 (1894). 392 SACCHARIMETERS to be able to illuminate the Yentzke scale above by mirror re- flection from the polarization lamp. The pointer at B is usually left in a fixed position and can be moved only by aid of a special key. But this possibility of regulating the half-shadow is without value in a technical instrument and may give rise to serious errors ; for as often as the index B is moved the ana- lyzer must also be turned to the point of equal illumination of the two halves of the field ; the technical saccharimeters must have, beyond question, a fixed half-shadow. The apparatus has no lid at C, but closing is effected, after laying in the ob- servation tube, by rotating a movable, half-open shell. The final adjustment screws are extended by means of universal joints so as to reach nearly to the table on which the apparatus rests, which makes it possible to move the compensation with- out lifting the arms. The screw D, which works the right scale, is placed somewhat lower than E, which turns the left one, so as to avoid possibility of confusion. 137. Half-Shadow Saccharimeter of Josef-Jan Fric. 1 This instru- ment with double wedge-compensation, the front partof which is shown in Fig. 58, is distinguished from the ordinary constructions in this that each one of the two scales is read by a special lens and Kig. 58. I'M*. 59- 1 Joseph-Jan Fric (Prague): Oesterr-Ungar. Ztschr. fiir Zuckerindustrie, V. Heft (1895). E. H. Sargent & Co., Chicago, are the American agents for these new instruments. GAS LAMPS 393 illuminated separately by a new and excellent arrangement, the details of which are shown in Fig. 59 diagrammatically. The light coming from the regular illumination lamp is re- flected by the movable mirror A and passes through the milk-glass plate B upon the scale C, the surface of which is so inclined toward the mirror D, that the light reach- ing the latter is thrown in the direction of the optical axis of the telescope E. The so illuminated metallic scale C ap- pears on account of the diffuse light from the milk-glass plate, as distinct as the old ivory scales. The left telescope with black mountings reads the working scale, while the right one with red mountings reads the control scale. Besides this, the control scale is seen by red light, since the milk-glass plate in this case is replaced by a plate of red glass, to exclude possibility of mistake in reading. c. Illuminating Lamps. i. Lamps for White Light. 138. Schmidt and Haensch Gas Lamps. Great importance need not be attached to perfect uni- formity in illumination as long as a correct passage of the rays through the apparatus is provi- ded for according to 96 and 129. Changes in the intensity of the light bring about no changes, then, in the zero point, and it is immaterial whether we use a flat burner or a round one. A very convenient form of lamp is shown in Fig. 60, which em- ploys gas, and is furnished with a triple flat burner, metallic chim- ney and a reflector. As these gas and petroleum lamps are usu- ally still furnished w r ith a so- called condensing lens which, as a matter of fact, has no importance, 1 1 The lens does not increase the intensity feres with the correct passage of the rays. Fig. 60. of the illumination, and always inter- 394 ILLUMINATING LAMPS in ordering a lamp one should be careful to secure a form which has in place of this lens a clear glass plate. 139. The Hinks Petroleum Lamp. For use with petroleum, the Hinks duplex lamp with metallic chimney, shown in Fig. 61, has given excellent satisfaction. Before lighting the lamp, the precaution must be taken to see that the cap used to extinguish- the flame is shoved back into its proper place, by aid of the lever on one side of the burner, so as to leave the wick per- fectly free to burn evenly. 140. Lamps with the Welsbach Incan- descent Gas Light On account of their intense light these can be strongly recommended for the illu- mination of saccharimeters. For this purpose the ordinary lamp with glass chimney is furnished with a second outer chimney of porcelain or asbestos, which at the proper point is provided with a suitably large opening to permit the passage of the light. The polariscope is naturally directed toward the brightest part of the glowing substance. It is not at all necessary to use a ground-glass chimney, as, with proper distance between lamp and instrument (129), the meshes of the mantle do not disturb in the slightest degree, since, in fact, the image of the glowing body produced by the illumination lens is formed at the analyzer diaphragm. 141. Lamp for Electric Light. This consists of a stand with the necessary connecting wires on which a very strong incan- descent lamp (of about 50 Hefner units with 100 to no volts) can be moved in vertical direction. The metallic cylinder sur- rounding the globe has anjopening in the right place to allow the best part of the light to pass. 142. Zirconium Light. The zirconium light is, by far, the GAS LAMPS 395 most intense white light. To produce it a Linnemann burner fed by oxygen and illuminating gas is used, the hottest part of the flame being directed against a plate of zirconia in a platinum support. 1 Instead of a zirconia plate a cylinder of the same substance may be used. 2 It is most advantageous to place the burner in a large sheet-iron box provided with windows, a door, and an inclined chimney. The lime light burners, fed by illuminating gas and oxygen, may also be strongly recommended. 3 2. Lamps for Homogeneous Light 143. Simple Sodium Flame Gas-Lamps. 4 In order to secure a sodium flame of considerable dura- tion, the lamp illustrated in Fig. 62 may be used. This consists of a Bun- sen burner which may be adjusted in vertical position, over which a me- tallic chimney is placed. The gas entrance, s, is found at one side where it will not become clogged by bits of salt dropping from the flame. The chimney may be adjusted by aid of the screw h at the proper height. At the top of the column, />, which may be rotated, a rod is at- tached horizontally w T hich carries at its end a bundle of fine platinum wires. These are so bent that they form a little pointed spoon. If this is filled with well dried salt and turned so as to rest in the front side of the flame, the melted salt is drawn ^j If up into the point of the spoon andl^ produces an intense yellow light on volatilization. By aid of the sheet metal shutter k, which has an opening, the light from the 1 lyinnemann : " Ueber ein neues Iveuchtgas-Sauerstoffgeblaesse und das Zir- konlicht," Wiener Sitzungsberichte II, 92, 1248 (1885). - From M. Wolz, in Bonn. 3 From Meckel, Kaiserstrasse 32, Berlin. 4 From Schmidt and Haensch, Berlin. Fig. 62. 396 ILLUMINATING LAMPS brightest part of the flame only may be allowed to pass. A second sodium lamp, which is very much r like the first, is shown in Fig. 63 ; the application will be read- ily understood. Instead of salt, well calcined so- dium carbonate may be used in these lamps ; the volatilization is slower, but the intensity of the light, at the same time, less. 144. Pribram's Sodium Lamp. 1 If it is necessary to work a long time with constant illumination, the Pribram lamp, shown in Figs. 64 and 65, is found useful. The gas led in at a emerges through the fine open- ings at b and mixes in the burner top c with the air which enters at d, the flow of which may be reg- ulated by turning a per- ::. forated disk by aid of the lever k. The gas is ig- nited at the gauze top of the burner at e. The chimney b, which is lined with asbestos, has four openings, through one of which, m, the light reaches the polariscope, while a second one, i, is furnished with a cap and serves for igniting and observing the flame. At g and h two little platinum boats, which are filled with fused salt, may be introduced into the flame. i Pribram : " l>l>er einen neuen Brenner fur Natriumlicht," Ztschr. anal. Chem., 34, 166 (1895), made by Schmidt and Haensch. Fig- 63. LAXDOLT'S SODIUM LAMP 397 -g Fig. 64. Fig. 65. 145. Landolt's Sodium Lamp. 1 A much stronger sodium light than that furnished by the burners described may be secured by the lamp shown in Fig. 66. A Muencke burner (Bunsen lamp with conical wire gauze top and so strong an air supply that the inner dark cone of the flame disappears) is supported on an iron stand, the upright rod of which carries a square chim- ney, B, made of sheet iron. The front side of this chimney has a round opening, over w ? hich the plate C, with three holes of 20, 15, and 10 mm. diameter, may be shoved. Two nickel wires, D, are laid across the sheet metal cylinder at the top of the lamp A, resting in notches in the cylinder, and around the middle of these wires pieces of nickel gauze are rolled. The meshes of this gauze are filled with salt, and most easily, as shown in Fig. 67, by laying them in a little trough of nickel foil in which the salt has been previously melted by aid of two Terquem burners. By placing the cylinder of the Muencke burner low, so that the salt is just over the wire gauze cone of the lamp, a very intense color is produced at the front and back of the flame. 1 I^andolt: "Xatriumlampe fur Polarisationsapparate," Ztschr. fur Instrurr., 4,390 , made by Muencke, Berlin. 398 ILLUMINATING LAMPS Fig. 66. If instead of common salt Fig. 67. dried sodium bromide is used, following the suggestion of Fleischl v. Marxow, 1 a very much more intense light is secured ; but the sodium bromide volatil- izes much more rapidly than the chloride and bromine vapors escape from the flame. In working with sodium bromide one must, therefore, place the burner under a good draft, as otherwise the polariscope may be completely ruined by the bromine vapors. 146. Intense Sodium Light. A very high illuminating power is found in the pencils suggested by du Bois, 2 consisting of sodium bicarbonate and sodium bromide in gum tragacanth, heated in the Linnemann oxygen blast-lamp. As these pen- cils give out bromine vapors, a good draft must be provided. Besides this, the greatly increased illuminating power calls for a great consumption of material, so that a rod about 4 mm. thick and 14 cm. long is completely burned in about ten min- utes. If one wishes, therefore, to work with these pencils, it is necessary to have an assistant or a clock-work mechanism to continually regulate the flame. These drawbacks are avoided by using, according to Gumlich, 3 rods of fused sodium 1 Fleischl v. Marxow : Wied. Ann., 38, 675 (1889). 2 I)u Bois : Ztschr. fiir lustrum., 12, 165 (1892). 8 Gumlich: /bid., 16, in (1896). PURIFICATION OF THE SODIUM LIGHT 399 carbonate about 6 mm. thick and 15 cm. long in the Linne- mann oxygen blast-lamp. No objectionable vapors are pro- duced, and the rods burn so slowly that in a period of about seven minutes it is usually not necessary to change their position. Although the intensity of the light produced is not quite equal to that frDrn the du Bois pencils, it is quite suffi- cient for nearly all purposes in polarimetry. As, furthermore, the use of these soda pencils is quite cleanly, the volatilization of fused rods in the oxygen lamp may be strongly recom- mended. j. Purification of the Sodium Light. Optical Center of Gravity 147. Lippich's Sodium Light-Filter. 1 In the following considera- tions, in order to have a specific kind of instrument in mind, we shall assume that all observations are made with a Lippich half-shadow apparatus with double or triple field. As the following paragraphs will take up also the comparison of polarimetric measurements, it may be remarked at the outset that the discussion can not be a complete one, because, in the first place, the limits of the book would not justify it, and, secondly, because in many cases it would not be possible to verify theoretical considerations by experimental data avail- able. Assume a substance having the power of rotating the plane of polarization and kept under constant conditions. It will then rotate lights of all wave-lengths through definite angles, depending only on these wave-lengths Let the rotation /? correspond to the perfectly homogeneous light of wave-length A.. Now, suppose the apparatus illuminated by mixed light made up of light of wave-length \ and light of wave-length A.,. Let Aj be smaller than A, and A. 2 greater than A., but so nearly the same that the eye may not recognize the difference in shade between them. Notwithstanding the consequent rotation dispersion, it will then be possible to find the angle of rotation fi l of the substance for this mixed light, as well as if the instrument were illuminated with perfectly homogeneous light. The same angle of rotation, /?,, would be 'found by using for illumination, a perfect^ homogeneous light of wave- 1 Lippich : Ztschr. fur Instrum., 12, 340 (1892). 400 SODIUM LIGHT length A 3 . // may be snid, therefore, that for polarimetric meas- urements, \ is the optical center oj gravity of the mixed lights of wave-lengths \^ and A 2 / that is, if this mixed light be em- ployed, rotations are obtained which actually correspond to the wave-length A 3 . As all polarimetric measurements depend finally on comparisons of brightness, it may be recognized directly that the optical center of gravity A 3 , depends not only on the wave-lengths A. t and A 2 but on the intensities of the two homogeneous components. For a definite condition of bright- ness in the two components we have, for example,. A 3 A. and consequently $ ft ; if now, the brightness of the component Aj increases, then the optical center of gravity A g will approach Aj from A. // may be now shown that the optical center of gravity A 3 depends simply and alone on the wave- lengths and degrees of brightness of the two homogeneous com- ponents in the source of light, and not on the rotation dispersion of the investigated substances, or on the amount of the angle of rotation or the size of the half -shadow angle chosen, as long as the observations are made with the Lippich apparatus, and the absorbing power of the substance is relatively the same for the two components. In investigating substances with considerable color, care must be taken to see that this last condition is ful- filled ; in what follows, uniform absorption is assumed. If we pass now to the general case, that is, if we assume that the source of light furnishes light of all wave-lengths, then, for polarimetric purposes, a definite optical center of gravity will correspond to this light also, as long as the field of view possesses the same color when the point of optical equilib- rium is found with the observed substance in position, as it had at the time of the zero-point determination. If this is not the case, then the optical center of gravity depends on the color sense of the observer, and the adjustment for reading becomes more and more inaccurate with increasing difference in color. With constant color , it may be shown, as before, that the optical center of gravity of the source of light depends simply and alone on the relative distribution of intensity in the spectrum of the source of light, as long as the absorption of the active substance is relatively the same for all wave-lengths in the light in ques- tion. We have then this important result, that under the PURIFICATION OF THE SODIUM LIGHT 401 assumed conditions, a definite optical center of gravity corresponds to any given source of light, which is the same for all Lippich instruments. Of course, other sources of light which show the same relative brightness in their spectra, have the same optical center of gravity. If the light is purified by passing through light filters, the real source of light must now be taken as that corresponding to the new optical center of gravity. In what follows the optical center of gravity will be given for the sources of light mentioned, as far as this is possible with the present meagre and inaccurate determinations. We shall consider first the most commonly used homogene- ous light, the sodium light. Every source of sodium light gives a continuous spectrum, in which, however, the light of the sodium lines is enormously in excess. If now 7 , smaller rotations are measured with the unpurified sodium light, the rather dark field of view will disclose no color. But as soon as large angles of rotation are to be measured the field of view appears distinctly colored and the sodium light must be puri- fied from foreign rays. The color of the field depends on the rotation dispersion of the active substance. As the analyzer is always so placed that the yellow sodium light is nearly ex- tinguished, it follows that with large rotations the blue rays, for example, w r hich are much more strongly rotated than the yellow, are able to pass through the analyzer but little weak- ened and, therefore, impart a blue color to the field of view. The shade depends on the rotation dispersion of the active sub- stance, and the amount of the angle of rotation. In the course of time a large number of absorbing substances have been sug- gested to purify the sodium light, 1 of which, up to the present, the Lippich sodium light filter is the best, if for the moment we leave the complete spectral purification out of consideration ; therefore this Lippich light filter only will be specially de- scribed. The Lippich sodium light filter is an absorption cell consist- ing of two chambers which the light passes in succession, and which are closed by plane plates. 2 The larger of the two chambers has a length of 10 cm., the smaller a length of 1.5 cm. The large chamber is filled with a filtered 6 per cent, so- 1 The light filters must be placed between the lamp and the illumination lens. 2 It is made by Schmidt and Haensch, Berlin. 26 402 SODIUM UGHT lution of potassium dichromate, in water. In the smaller cell there is a solution of uranous sulphate, US 2 O 8 . This is deep green and must be made by reduction of the corresponding uranyl salt, USO 6 . As the uranous salt solution passes into the other by oxidation in the air, a perfectly air-tight cell must be provided, and the solution must be renewed from time to time. The uranous sulphate solution is made as follows : 5 grams of pure uranic sulphate is dissolved in 100 cc. of water and 2 grams of pure zinc in powdered form added. Then 3 cc. of concentrated sulphuric acid is added in three portions, waiting after each addition until the reaction is nearly com- plete ; the flask must remain closed. After the addition of the last portion of acid the closed flask is allowed to stand about six hours ; the liquid is then filtered and filled into the cham- ber in such a manner as to leave the smallest possible air bub- ble. After a day the solution comes to rest and remains one or two months constant. The weights and volumes given above must be adhered to within l l loo f their amounts. While the potassium dichromate absorbs a part of the green rays and the blue rays, the uranous sulphate solution ha!s a wide and deep absorption band in the red which reaches nearly to the D lines. A spectrum is, therefore, obtained in which only a small band with the D lines in the middle is present. The two solutions produce so complete a purification of the sodium light that even with a rotation of 50 and a strong illumina- tion, difference in color- is scarcely perceptible. It is, therefore, desirable that chemists in general should employ this L,ippich filter for the purification of sodium light, and especially for the reason that thereby the results of different observers would be comparable among themselves. 148. Optical Center of Gravity of Sodium Light. Completely pu- rified sodium light consists of the light of the two D lines and extremely little light of adjoining wave-lengths. Following Bell 1 we shall take the wave-length of the less refrangible so- dium line Dp as 589.62 /^, and of the more strongly refrangi- ble one D 2 , as 589.02 ////. In close agreement the observa- tions of Soret and Sarasin," and of Lippiclv' have shown that Bell : Phil. Mag. [5] 35, 245, 350 (1888). 8 Soret and Sarasin: Compt. rend., 95, 635 (1882). 8 I^ippich: Wiener Sitzungsber, II, 99, 722 (1890). CENTER OF GRAVITY OF SODIUM LIGHT 403 for a quartz plate i mm. in thickness the difference in rotation of the two D lines corresponding to the 0.60 n/J is about i6o". 1 Therefore a difference in rotation for i mm. of quartz, J ft, corresponding to a change, J A, in the wave-length in the neighborhood of the D lines may be calculated from the equa- tion, J yff/J \ = 266 \ ^ sec \ If, therefore, we determine L ;ujw J as Lippich did, the rotation /? of a quartz plate for homogene- ous light of wave-length A == D 2 , for example, and then the rotation /?,, for an}' other sodium light, we are able in the sim- plest manner, by aid of the above equation, to calculate the optical center of gravity corresponding to /3 im But this optical center of gravity is not peculiar to the quartz alone, but it holds good for all substances, since, as was explained in the last paragraph, the optical center of gravity of a source of light is independent of the rotation dispersion and of the size of the angle of rotation. It is, therefore, possible to find, with the aid of a quartz plate, the optical center of gravity of the different kinds of sodium light. We ma} r now calculate this optical center of gravity for per- fectly purified sodium light, and it may be assumed that it contains only the light of the two D lines. According to Die- trich 2 the relation of the intensities of the two lines is D 2 /D 1 = 1.6. By aid of this value and the wave-lengths given above for the D lines the optical center of gravity of the fully puri- fied sodium light is found to be 589.25 ///* ; the method of cal- culation need not be discussed here. Lippich has experi- mentally determined the optical center of gravity for several sodium lights ; 3 the wave-lengths of the center given in the table below differ from those stated by Lippich in the paper cited, by a constant difference, because he assumes a wave- length for D. 2 different from that we have taken above. Finally, it is possible to calculate from some observations of Landolt* the optical center of gravity of unpurified sodium light (Lan- dolt sodium lamp with NaCl) as 588.06 /*/*. In the following table the determined optical centers are grouped for comparison. 1 About the same value is given by the Boltzmann dispersion formula for quartz. - Dietrich : Wied. Ann., 12, 519 (1881). 3 Lippich: Ztschr. fur Instrum., 12, 333 (1892). * I^andolt: Ber. d. chem. Ges., 27, 2885 (1894). 404 SODIUM LIGHT OPTICAL CENTERS OF GRAVITY OF SODIUM LIGHT, ASSUMING 589.02 /x/x AS THE WAVE-LENGTH OF D 2 . No. Source of Light. Purification. Wave- lengths in MM- I Bunsen burner with NaBr Layer of a 9 per cent, aqueous solu- tion of K 2 Cr 2 O 7 10 cm. thick 592.04 2 Bunsen burner with NaCl Layer of a 9 per cent, aqueous solu- tion of K 2 Cr. 2 O 7 10 cm. thick 589.48 3 Bunsen burner with NaCl or NaBr Lippich's sodium light filter, K 2 Cr 2 O- and US 2 O, 589-32' 4 Sodium light Perfectly purified spectrum light. The two D lines only 589-25 5 Landolt's sodium lamp with NaCl Layer of a 6 per cent, aqueous solu- tion of K 2 Cr 2 O- 1.5 cm. thick 588.94 6 Bunsen burner with NaCl Layer of a 9 per cent, aqueous solu- tion of K 2 Cr 2 O 7 10 cm. thick, and layer of a 13. 6 per cent, aqueous solution of CuCl 2 i cm. thick 2 588.91 7 Landolt's sodium lamp with NaCl Not purified 588.06 It must be again remarked that these optical centers of gravity are exact only when the active substances under in- vestigation do not appreciably alter the distribution of the intensity of the light from the source employed. If, in illus- tration, the rotation of a quartz plate is found to be 20 with light No. i, the same plate will show a rotation of 20.27 with light No. 7 ; the difference amounts to 16.2 minutes of arc, certainly a considerable amount. It is plain, therefore, how important it is in statements of rotations for sodium light to give at the same time the optical center of gravity of the light 1 The optical center of gravity is therefore found midway between the two D lines and it is further seen that, after purification with the Lippich filter, even large varia- tions in the intensity of the sodium light do not appreciably affect the optical center of gravity. 2 One gram of cupric chloride to 6.35 cc. of water. SPECTRAL PURIFICATION OF SODIUM LIGHT 405 used, as otherwise the rotations are uncertain to the extent of i per cent, or more. If we consider, foi example, the results which the different observers found in the determination of the Verdet constant for the electromagnetic rotation of the plane of polarization, it is evident that the measurements of these different observers are not directly comparable with each other. 1 Nearly every one of the seven observers employed a dif- ferent source of sodium light and method of purification, and besides this, several values have even been found with the Laurent half-shadow instrument (see 113) ; furthermore, the electromagnetic rotation dispersion for carbon disulphide and water is somewhat greater than the natural rotation dis- persion of quartz. 149. Spectral Purification of Sodium Light." In very exact work physicists w r ill always prefer spectral purification of light to that by means of filters. The spectral purification has always this great advantage over the filter method, that it permits the yellow rays to pass undiminished into the instruments while in the light filters, a certain amount of the light is always B A Fig. absorbed. For the purpose of spectral purification, it is not recommended to use the Wernicke liquid prism, because, in consequence of the heating effect, the course of the rays is subject to constant changes ; it is much better to use glass prisms. The following method of spectral purification of sodium light has been found by the author through long ex- 1 A tabular compilation of all the absolute determinations is found in Ztschr. fur lustrum., 16,283 (1896). Notice, especially, the almost impossible accuracy with which several observers claim to have established the value. - See, also, Brodhun and Schonrock : Ztschr. fur Instrum , 16, 244 (1896). 406 SODIUM LIGHT perience to work well ; it may be used also in the purification of any other homogeneous light. By aid of the lens B (Fig. 68), a sharp image of the source of light A is thrown on the slit-screen C. The light which passes through the slit falls on the achromatic lens D and is then decomposed by the flint glass prism E, placed in position of minimum deviation. A sharp image of the luminous slit C, with the spectrum, is thrown on the second slit-screen G by the lens D. In order to secure the greatest possible dispersion, the distance between D and G is taken rather great, say 2 to 3 meters. The rays are then sufficiently parallel in passing the prism E. Just in front of the slit G there is a lens F, which produces a sharp image of the lens D on the illuminating lens H of the Lippich apparatus. The slit G, which passes the purified sodium light, must naturally have such a position with reference to the lens H that the latter will produce a clear image of G at the analyzer diaphragm. As D is pictured at H, it follows that the former must be uniformly illuminated by the slit C, which however may be accomplished without difficulty. Aside from reflexions, there is, therefore, no loss of light on the way from C to H. The active sections of the lens D and prism E must be chosen large enough, so that the polarizer diaphragm will be quite filled with light. The width of the slit G is, of course, made only as great as required by the analyzer diaphragm and focal length of the illumination lens H in order to secure maximum brightness in the field of view ; and, accordingly, the slit C is made so narrow that its enlarged image formed by the sodium rays is just sufficient to fill the slit G. As regards the centering of the course of the rays, it may be remarked that we begin with H and place G, then F, and the following pieces in position. Of course, the pieces from B to G need not be united in one apparatus ; it is better to have them attached to small stands which may be moved into the right positions and then made fast to the table with wax. Although the light rays, if the dispersion is sufficiently great, must pass through a distance of 5 or 6 meters from the source of light to the eye of the observer, it is still pos- sible to build up the whole apparatus in a rather small room if a plane mirror is placed in the prolongation of the axis of the BRIGHTNESS OF THE LIGHT 407 apparatus between E and F. The pieces from A to E can then be built up parallel to the apparatus and but slightly removed from it, and the sodium light thrown into the instrument by aid of the mirror. The spectral purification of the light se- cured by this method is so perfect that even with an angle of rotation of 500, the slightest color in the field of view is not apparent. 150. Dependence of the Optical Center of Gravity on the Brightness; that is, on the Amount of Luminous Vapor in the Unit of Volume of the Source of Light. We have now to consider the question: Is the optical center of gravity of a purified homogeneous light a function of the brightness of the light ? Although the case of sodium light is somewhat complicated, we shall take it up first, since this homogeneous light is the one most commonly used. We shall assume then, in what follows, completely purified sodium light ; that is, light consisting of the two D 613 A in p. /u 589,62 589,02 Fig. 69. lines and extremely little of neighboring wave-lengths, such as is obtained by the method of spectral purification described in the last paragraph. The relative distribution of brightness in the spectrum of this sodium light corresponds approxi- mately to Fig. 69 in which, as abscissas, the wave-lengths A. are expressed in j^f^, while the ordinates represent the inten- sities for the corresponding wave-lengths. 1 The light of the line D 2 is, according to Dietrich, about 1.6 times as bright as that of D r After satisfactory spectral purification light be- 1 The false light is, of course, much too bright in the representation. 565 408 SODIUM LIGHT tween the wave-lengths 613 w and 565 HV, approximately, will enter the polarization apparatus ; as this spectrum is very long for the dimensions chosen in the figure, only the middle and end portions are represented. The optical center of gravity of purified sodium light from all sources is the same as long as the mea?i wave-lengths of the two D lines, and also their relations as regards brightness, are constant. 1 We may consider first, a source of sodium light of constant luminosity, which, after purification, reaches the illumination lens of the instrument with a perfectly definite optical center of gravity, and ask now if this center is altered if the light at any part of its path is uniformly weakened. This could be the case, only if the wave-lengths were variable with the in- tensity. Now, Lippich 2 has shown experimentally the con- stancy of wave-length with different values of the intensity to within Vioooooooo of the value of the wave-length. His conclusions were later confirmed by Ebert, 3 who obtained results by the method of high interferences which, with proper homogeneous light, showed constancy in wave-lengths to within V], 000.000 f r values of the intensity varying between the limits of i and 250. Ebert was able to show for the light of the two D lines especially, that a diminution to 3 per cent, of the original brightness did not change the mean wave-lengths Vsooooo ^ their value ; that is, not by o.ooi m*. We have then this important result, that with unchanged emission from the source of light, the optical center of gravity does not vary with the intensity. Now let us suppose the emission from the source of sodium light to change by altering, for example, the brightness. As far back as 1871, Zollner 4 showed that when by moving the salt globule more or less completely into the Bunsen flame, differ- ent amounts of sodium vapor were produced, the line D., in- creased in width more rapidly than D, with increasing bright- ness, and that D } widened more rapidly toward the side of greater wave-lengths than toward the other, while no such dis- 1 The continuous spectrum of thejfalse light in the sodium li^l't dots not i-Ji.-uin^ the optical center of gravity of the two D lines, because the brightness of the whole- range of the narrow portion of the spectrum front which light enters the apparatus may be taken as constant. -' I.ippich : Wiener Sit/.uiiKsber., II 72, 355 (1875). Kbert : Wied. Ann., 33, 337 (1887). /.ollncr : Pogg. Ann., 143, 88 (1871). BRIGHTNESS OF THE LIGHT 409 placement of the center of D 2 on widening was to be observed. From these observations, it follows that a change in the opti- cal center of gravity is indeed possible with a change in bright- ness in the sodium light. On what does the widening of a spectrum line depend ? As even the most homogeneous spec- trum line is formed by a series of elementary rays whose wave- lengths are infinitely close together, it must follow that the intensities of the elementary rays are a continuous function of the wave-lengths. The brightness of the spectrum lines can- not, therefore, change suddenly on the sides but must grad- ually decrease to zero. If such a line is broadened, it is the intensities of just those wave-lengths on the edges of the spec- tral lines which are so much increased that they become per- ceptible to the eye. We must, of course, distinguish between this broadening and the displacement of the mean wave-length (the optical center of gravity) of the spectrum line in question ; this displacement may depend on a broadening of the line as well as on a change in the form of its intensity curve. The observations of Zollner were fully confirmed by work of Ebert. 1 As the latter, in these investigations, made quan- titative measurements by the method of high interferences of the changes in mean wave-lengths of several lines of the spec- trum, so that we have full data concerning the amount of pos- sible displacements, this work of Ebert 's must be regarded as of great importance in polarimetry and will, therefore, re- ceive here closer attention. In his experiments, Ebert employed a Terquem burner, and modified the emission of light by moving the salt globule to a greater or less depth in the flame. As long as the salt just touches the edge of the flame, the vaporization rind brightness are small, but these increase as the globule is pushed further into the flame until a point is reached near the inner cooler zone where the vaporization again decreases. With such alterations in brightness, Ebert found displacements of the mean wave-length of sodium light amounting to 0.044 f.iu, and increasing toward the less refran- gible end of the spectrum, with increase in brightness. At the came time Ebert showed for sodium light as well as jor sev- eral other spectral lines, that primarily neither the thickness of i Ebert : Wied. Ann., 34, 39 (1888). 410 SODIUM LIGHT the luminous layer nor the temperature of the source of light or the chemical changes taking place in it cause any changes in the wave-length, but that the mean wave-length depends simply and alone on the density of the vapor ; that is, on the amount of vapor of the luminous substance in the unit of volume, and varies with alterations in the density of this vapor. It is also true that the amount of chauge for the different sodium salts under the same conditions of vaporization is, in general, not the same ; but the variations are not large. If now a change in bright- ness obtained by aid of a Terquem burner and salt bead pro- duces a displacement of the optical center of gravity of sodium light amounting to 0.044 W, one may not be mistaken in as- suming that with much greater changes in illumination the displacement of the center will probably be still more marked. This view has recently been completely verified by Schonrock, 1 and by aid of polarization apparatus which for such investiga- tions is doubtless more sensitive than Ebert's method of high interferences. Schonrock worked with a Lippich half-shadow instrument, and employed as source of light the Linnemann oxygen blast- lamp described in 142, in which sticks of fused sodium car- bonate were vaporized ; the sodium light obtained was purified perfectly by a flint glass prism with a ray path of about 3 meters. It could then be shown easily that the optical center of gravity varied considerably with the brightness of the flame, since the amount of an angle of rotation must vary corre- spondingly. If by use of a perfect quartz-plate or a cane-sugar so- lution an angle of rotation of about 100 is obtained, this is found to decrease about 1 40 seconds of arc when the stick of sodium car- bonate is moved more and more into the hottest part of the flame. The zero point of the instrument is in no wise changed in the op- erations. Therefore, according to the formula developed in 148, - 266 the optical center of gravity of sodium light must be moved about o. 1 1 A*/* with increasing brightness, that is, about one- fifth of the distance between the two D lines and toward the 1 Schonrock : "DieThatigkeitder Physikalisch-technischen Reichsanstalt," Ztschr. fur Instrum. 17 (1897). BRIGHTNESS OF THE UGHT 411 red end of the spectrum. This displacement is in the same direction as shown by Kbert, but in consequence of the much greater variation in brightness is more than twice as large as he found. Without doubt by working with the du Bois soda pencils described in 146 still greater displacements of the optical center could be demonstrated. At the same time Schonrock found in complete agreement with the results of Lippich and Ebert, by aid of a Nicol prism placed in front of the illumination lens of the instrument, that with constant emission from the source of light the wave-length of the optical center of gravity does not change with the intensity of the light. The displacement of the center for sodium light is doubtless due not only to the unsymmetrical widening of the line D lf but also to a change in the relation of the lines D l and D 2 to each other as regards brightness. But this point remains to be cleared up by future work. In 148 the optical center of gravity of perfectly purified sodium light is calculated as 589.25 fiif* ; we know now that this value is probably correct for a certain mean brightness of the sodium light, but that in addition it may vary by o. 1 1 yu^u by changes in the emission from the source of light. As with sodium light the optical centers for all other lines or perfectly purified homogeneous lights change more or less strongly with variations in the amount of vapor in the unit volume of the source of light. In his work referred to above Ebert made quantitative measurements of the displacement for some lines of thallium, lithium, potassium, and strontium. Without exception he found that the widening of the lines was stronger toward the end of less refrangibility than toward the other, so that with increasing brightness the optical centers of gravity moved toward the red end of the spectrum. The dis- placements observed by him are given in the following short table, for which it must be remembered that the changes of brightness were produced by aid of a salt bead and Terquem burner only. 4 I2 SODIUM LIGHT DISPLACEMENT OF THE OPTICAL CENTERS OF GRAVITY OF SOME SPEC- TRUM LINES, ACCORDING TO EBERT. Source of light. Wave-lengths in MM- Displacement in MM- T iCI T i CO 670.8 535-1 768.0 404.6 460.8 0.06 0.026 0.046 . . - 1 0.019 X1C1 KC1 KC1 SrOl In their paper " Ueber die Spectren der Alkalien," Kayser and Runge make the following statement : 2 " The majority of the lines of the alkalies are not, however, sharply denned, they broaden by increase in the amount of vapor, and either toward both sides or, more commonly, toward the red end of the spectrum, occasionally however, toward the violet end only. Such broadened lines often reach a width of two to three yw/**." The general conclusion is therefore warranted that the optical center of gravity of a spectrum line is not a constant but is a function of the emission from the source of light. Since the Arons mercury light, to be described later, has recently been made available for polarimetric measurements, it may be in order to discuss briefly the spectrum lines of mercury. It is well known that the different spectra of one and the same chem- ical element show variations depending on whether the} 7 are flame, spark, or arc light spectra. In the case of mercury these differences are very large, 3 so that the strongest lines in one spectrum may be in part wholly wanting in the other. It would appear probable, therefore, that the lines of mercury must change strongly with changes in the light arc, which in turn changes with the intensity of the producing current ; Ebert states that the bright green mercury line, 546 /*/*, in- creases on one side only, which, in all probability, would have a displacement of its optical center of gravity as a consequence. What conclusion must be drawn now for polarimetry from this variability of the optical center of gravity of purified homo- geneous light ? 1 A displacement was found but not measured. * Kayser and Runge : Wied. Ann., 41, 302 (1890). Kayser and Runge : Ibid., 43, 385 (1891). ROTATION OF SODIUM LIGHT BY QUARTZ 413 In the first place in all polarimetric measurements, whether of absolute amounts of rotation, or of differences of rotations, the emission from the source of light must be kept constant, and this may be reached in a satisfactory manner even in investigations of length. In the second place it is necessary that along u'ilh the rotations the optical center of gravity of the light used must be determined and defined with corresponding accuracy, as otherwise the measurements of different observers are not comparable with each other. It is not sufficient to merely state that one has used sodium light after perfect spectral purification, as even then its optical center of gravity remains uncertain to about o. i pi*. This uncertainty corresponds to a difference in rota- tion of 25" in an angle amounting to 20; such an angle may often be measured without difficulty to within 8". If then the expected accuracy of the method is not to be illusory the wave- length of the corresponding optical center of gravity must be given to within 0.03 /^. From this it is seen that in general an angle of rotation may be found with a good polariscope with a degree of accuracy which is much greater than the accuracy with which the corresponding optical centers of gravity may be measured and expressed. Whether it will ever be possible to carry the determination of the optical centers of homogeneous lights so far that they will correspond in accuracy to the delicacy of the modern polarimetric apparatus, remains for the future to show. 1 151. Absolute Determination of the Rotation of Sodium Light for Quartz As we have now seen the several variations to which the optical center of gravity of sodium light from different sources is subject, we shall next attempt a review of the work done on the absolute rotation of quartz for sodium light, as already explained in 44, taking these disturbing variations into consideration. Passing over the oldest and very inaccurate measurements we come at once to the work of v. Lang. A value is often quoted from work of v. Lang done in 1875* 1 This will be possible as soon as the rotation dispersion of quartz is sufficiently well established. Such an accurate determination of the rotation dispersion of quartz may be made, as is sufficiently clear from what has been said, only by aid of the Fraun- hofer lines, and after improvements in methods discussed in the chapter on "Determi- nation of Rotation Dispersion." -' v. I^ang : Wiener Sitzungsber., II, 71, 707 (1875). 414 SODIUM LIGHT which is misleading ; in this work v. Lang investigated the dependence of the circular polarization of quartz on the tem- perature, but did not determine, as he himself explains, the absolute value of the angle of rotation, assuming the rotating power per millimeter as known. But in a second investiga- tion he made an absolute determination. 1 By the aid of the Broch method to be described later, with sunlight, and using a double prism of right and left quartz about 33 mm. thick, he found at 20 C. a rotation of 21.724 per millimeter for the line D. We must take then as the optical center of gravity the mean between the two D lines, that is, 589.3 yw/^. In 1878 Jouberf' made absolute rotation determinations ; but as he worked with the Laurent instrument and sodium light, the optical center of gravity being quite indefinite, his results do not call for consideration. Following the Broch method with sunlight, Soret and Sarasin 3 determined the rotation of two quartz plates, about 30 and 60 mm. in thickness, for each of the two D lines. For the middle of the two lines (589.3 //^) we find as the mean value from these figures 21.708 per millimeter, for 20 C. Soret and Guye 4 have made determinations with the same plate about 60 mm. thick, used by Soret and Sarasin. As they worked with a Cornu instrument (110) and sodium bromide light, purified through the spectrum, the wave-length of the optical center may be taken as before, as 589.3 WJL. The value of 21.723 per millimeter at 20 C. may be calculated as the mean value from their experiments. More recently absolute rotation determinations have been made by Gumlich. 5 The optical center may be again taken as 589.3 w*, as he worked with the Lippich instrument and spectrum sodium light. He found, with four quartz plates of approximately 5, 6, 8, and 10 mm. thickness, values which varied between 21.717 and 21.731 per millimeter at 20 C. The mean value was 21.724. Finally, the author is in a position to give his own results for 1 T. Lang: Wiener Sitzungsber., II, 74, 209 (1876). 4 Joubert : Compt. rend., 87, 497 (1878). 8 Soret and Sarasin : Ibid., 95, 635 (1882). 4 Soret and Guye : Ibid., 115, 1295 (1892). 6 Gumlich : Ztschr. fur Instrum., 16, 97 (1896). RELATION OF THE ANGLES OF ROTATION 415 the absolute rotation of quartz. He (Schb'nrock) made his determinations with a perfect quartz plate, about 5 mm. thick, by aid of a Lippich instrument and sodium light which was absolutely purified through the spectrum, so that the optical center may be taken as 589.3 /*/*. From many measurements made at different times the value 21.722 per millimeter at 20 C., and accurate to 0.003, i calculated, corresponding to the true optical center of gravity of the sodium light used ; but this latter has not yet been found with a satisfactory degree of accuracy. As, therefore, all these observers have made their determina- tions very nearly for the same center, 589.3 nn, their measure- ments must be comparable with each other. The results are given in the following short table : Observer. Rotation of quartz per millimeter at 20 C. v. Lang 2i.724 ; Soret and Sarasin 21.708 Soret and Guye 21.723 Gumlich 21.724 Schonrock 21.722 With exception of the value of Soret and Sarasin the agree- ment is quite a remarkable one. It may be said with certainty that, using perfectly pure sprectrum sodium light, i mm. of quartz, with a certain mean brightness of the light, will rotate the plane of polarization exactly 21.723 at a temperature of 20 C. But on account of the variation in the optical center of gravity with the brightness of the source of light the rotation may vary by about 0.004 P er millimeter. There may accord- ingly be some reason for a re determination of the absolute rotation of quartz only when the corresponding optical center is found with an equal degree of accuracy. 152. Relation of the Angles of Rotation, ctj and a D . In addition to the illumination of saccharimeters a white light is employed in the determination of the angle of rotation, or,-, for mean yel- low rays by aid of the Robiquet polariscope (100). As seen by the expression itself, ' ' mean yellow rays," the angle of rota- 4l6 SODIUM LIGHT tion, a h is not accurately defined; the wave-length correspond- ing to it may be taken as about 556 /'A'. 1 The rotation referred to this, following Biot's suggestion, is represented by oij (jaune moyen). As the wave-length of mean yellow light is less than that of D (589 ///*) the values for otj are always much larger than those for a D . For quartz, as an illustration, we have per millimeter, a D = 21.72 and otj = 24.5; these formulas for conversion are therefore used : (*,= <* D = i . 1 2&a D , and U D = - - ' ' a = o. 887 a-. 21.72 24.5 But on account of unequal rotation dispersion for different sub- stances the relation ofa D to a, is a variable one. The deviations from the above values may amount to 10 per cent, or more. As the rotation a } does not correspond to any accurately definable ray, the determination of or,- is not satisfactory and at the present time is scarcely made. It must be pointed out here that when a rotation is measured in a half-shadow appa- ratus with white light the angle obtained is not accurately atj ; see 153. There are in the literature a large number of observations made by Biot a with red light obtained by aid of glass colored with cuprous oxide, and corresponding approximately in refran- gibility to the Fraunhofer line C (656 /*/<. ) From the state- ment that this light is rotated 18.41 by i mm. of quartz its wave-length is calculated as about 637 nf*. For quartz, there- fore, the rotation of this red light is related to that for D as i to 1.18. 153- Optical Center of Gravity of White Light. The chemist is sometimes in the position where, instead of using yellow light, he is obliged to employ white light. We shall therefore con- sider the optical center of gravity of white light more closely, under the assumption that the rotations are measured with a Lip- pick half -shadow instrument. An optical center of white light can not be directly defined, as it is subject to considerable va- riations according to the source of light employed. Sunlight, gas light, petroleum light, Welsbach light, electric light, zir- 1 L,andolt : Sitzungsber. der. Akad. Berlin, 1896, p. 790. 2 Biot : Mm. de 1' Acad., 3, 177 (1820). OPTICAL CENTER OF GRAVITY 417 cona light, lime light, and so on, have all a different distribution of brightness in the spectrum and, therefore, according to 147, different optical centers of gravity. With white light illumination, however, only very small angles , at most j , may be measured, because with larger rotations, the field of view shows considerable color variations. And, above all, the investigated substances must be colorless and as clear as water. If the active substance is colored, it will have an absorption spectrum, and will, therefore, alter more or less strongly the composition of the white light; the white light will, at the same time, be filtered by the colored body, and in this way the relative brightness of different parts of the spectrum of the original light will be totally altered. For one and the same source of white light the optical center varies, therefore, from substance to substance, if they are colored. 1 That these changes in the optical center are very considerable is shown by the obser- vations of Holzer,- the results of which could be predicted from what has been said. In the investigation of colored sub- stances, therefore, even when the color is slight, the use of white light must be given up and an intense sodium light pro- vided. The following experiments were made to find the optical center of one white light, the Welsbach light, with some degree of accuracy. The small angles of rotation were secured by combination of positive and negative quartz-plates ; a posi- tive plate about 1.49 mm. thick and a negative plate about 1.46 mm. thick, gave together a rotation of about -f- 0.6, while the same negative plate with a positive plate about 1.6 mm. thick gave a positive rotation of about 2.9. These two angles of rotation were measured in the Landolt apparatus (117) with the following three sources of illumination : First, with sodium light made by the Landolt sodium lamp and purified by the Lippich sodium light filter, then with the Welsbach light filtered through a layer of 6 per cent, potassium dichro- mate solution 1.5 cm. thick, and finally with pure Welsbach light. In using the last sources of light, the field of view was colored, especially in measuring the larger angle of rotation ; 1 The I^andolt ray filters depend on this. - Holzer: Ber. d. chera. Ges., 15, 1932 (1882). 27 418 WHITE LIGHT notwithstanding this, the adjustment was always made to secure uniform shade as nearly as possible, the error of adjust- ment being in the mean about 1.5 minutes of arc, with a half-shadow of 3. An)' effects of temperature changes may be eliminated in such long experiments by regularly changing the sources of light and employing them, for example, in this order, I, II, III, I, III, II, I, so that the mean values for each source of light correspond to one and the same mean tempera- ture. The results obtained are given in the following table : ! By quartz-plates. Sodium light purified by the Lippich filter. Welsbach light through layer of 6 per cent, solu- tion of KoCroO 7 1.5 cm. thick. " Simple Wels- bach light. 0.60 2-95 589.3 0.61 2-94 539 0.68 3-39 55i Optical center in /*/* To within about 0.01, the same angles are found with the pure sodium light and the filtered Welsbach light, so that the optical center of the latter may be taken also at 589 /U./M. On the other hand, the results found with the simple Wels- bach light, a n ,, are much larger than the corresponding a D . If we recall now that the rotation of quartz for pure sodium light is 21.72, we may calculate by aid of the known dispersion formulas for quartz that the optical center of gravity of the Welsbach light is about 55 1 /A/A. We have then the conver- sion formulas : a w = 1.1490^, and a n = 0.8700^. But as the relation of t* D to at w varies in different substances because of unequal rotation dispersions, it must always be re- membered that for other bodies than quartz, the values of ct n calculated by the above factor from observations of a u may be in error to the extent of 10 per cent or more. It is, therefore, recommended not to use the simple Welsbach light at all, but when necessary, to employ this light purified by the potassium dichromate solution as above defined, since in this case a re- duction is not required. If one is obliged to measure angles of more than 3 with 1 Without doubt, other results would be obtained by a color-blind eye. DETERMINATION OF ROTATION DISPERSION 419 white light, this should not be done under any circumstances with a half-shadow instrument with circular graduation, but the rotation may be found with a half-shadow saccharimeter ', with which, using the double wedge-compensation, angles of 34 may be measured. The conditions here are very much more favorable, because the rotation dispersion is very largely compensated by the action of the compensating quartz-plate in the wedge, so that its influence on the measured angle of rota- tion is small ; colored substances also, may therefore be in- vestigated in saccharimeters. If then the rotation in Ventzke degrees be multiplied by the factor 0.347, as given in 127, it may be at least said that the product obtained gives the angle of rotation in circular degrees for sodium light with an error of 2 to 3 per cent, at most. 1 But there is no object in trying to find a more accurate reduction factor for each individual substance, because, on account of rotation dispersion, temper- ature effects, and variations in rotation with the kind of illu- mination and color sense of the observer, it is never possible to make accurate determinations, free from appreciable systematic errors, in a saccharimeter. Whenever possible, therefore, avoid the use of a saccharimeter and secure for illumination of a polariscope the most intense sodium light available, the pro- duction of which may be, moreover, accomplished without great difficulty. d. Determination of Rotation Dispersion 154. Method of Broch. To find the rotation dispersion of a substance, it is necessary to determine the angles of rotation for light of different wave-lengths. This may be done by aid of one of the polarization instruments described, using differ- ent kinds of homogeneous light, which method will be later explained. But first, several other methods must be con- sidered. A procedure which permits the determination for a whole series of rays of known wave-lengths was given by Broch, 2 and simultaneously by Fizeau and Foucault. 3 In this process, illumination is furnished by sunlight which, by aid of 1 This follows from the data of L,andolt : Sitzungsber. der Akad., Berlin, p. 959 (1887). 2 Broch : Dove's Repert. d. Phys., 7, 113 (1846). 3 Fizeau and Foucault : Compt. rend., 21, 1155 (1845). 420 DETERMINATION OF ROTATION DISPERSION a heliostat, is thrown horizontally into a darkened room. The rays pass in the following order : the vertical slit A (Fig. 70), the polarizer B, the analyzer C, the prism D placed in position of minimum deviation, and the telescope E F which must be F Fig. 70. provided with cross hairs. In order to adjust the apparatus, the principal section of the movable analyzer C is placed paral- lel to the principal section of the polarizer B, the slit A is illu- minated \vith sodium light and the ocular F of the telescope is focused sharply on the image of the slit A formed by the achromatic objective E. On removing the sodium light and admitting sunlight, a pure spectrum with the Fraunhofer lines is seen with the ocular F. In order to bring any desired part of the spectrum into the center of the field of view, the tele- scope must be movable horizontally around a fixed axis pass- ing through the center of the prism. First, the anal) 7 zer C is placed in the position of greatest darkness, without an active substance following ; this is the zero point. Then when the active body is placed between C and B, the spectrum with the Fraunhofer lines appears again in the telescope. If now the analyzer C is turned, a position is found at which a ver- tical dark band appears in the field of view and, with further rotation of C, moves across it ; an essential condition for this, however, is that the rotation dispersion of the substance must be rather large as compared with the dispersion of the prism. The phenomenon of the dark bands depends on the fact, that in the rotation of the analyzer those rays are extinguished, one after the other, whose planes of polarization coincide with the principal section of the analyzer. If the cross hairs of the telescope are focused on one of the Fraunhofer lines at the start, and then, by movement of the analyzer, the dark band is brought to center with the cross hairs, the rotation read off on the circle shows the amount of rotation for the line in METHOD OF BROCH 4 2I question. In this manner the rotation for each one of the Fraunhofer lines may be found. In place of cross hairs, it is preferable to employ parallel fibers. The dark band is narrower and sharper, the larger the angle of rotation, and consequently, the rotation dispersion. The Fraunhofer lines and the corresponding wave-lengths are given in the table below : Fraunhofer lines. Wave-lengths in iJ.fi. Fraunhofer lines. Wave-lengths in /U./K. A 759-4 b 517.5 a 718.6 P 486.1 B 686.7 G 430.8 C 656.3 h 410.2 D 589.3 H 396.9 E 527.0 Soret and Sarasin 1 in their determination of the rotation dis- persion of quartz have avoided the rather inaccurate zero-point adjustment in this manner that they placed a negative quartz- plate, for example, in position first and brought the dark band into coincidence with a line of the spectrum ; then, after re- moving this plate, a positive one w r as substituted and by move- ment of the analyzer the dark band was again brought into coincidence with the same line of the spectrum. From the rotation of the analyzer there is found, therefore, for the line in question, an angle which corresponds to that of a positive or negative quartz-plate, the thickness of which is equal to the combined thicknesses of the positive and negative plates used. Through several improvements, Lippich 1 has made the Broch method a tolerably accurate one, but for the details, his original paper must be consulted. Finally, by use of a quartz double plate of variable thick- ness, G. Wiedemann 3 has essentially improved the Broch method. The double plate consists of two quartz wedges whose principal surfaces are ground vertically to the optical axis, and each one of which consists of an upper right-rotating 1 Soret and Sarasin : Compt. rend., 95, 635 (1882). Their method has been given above by the author in more practical form. - I,ippich : Wiener Sitzungsber. II, 85, 307 (1882). 3 G. Wiedemann : L,ehre von der Elektricitat, 3, 914 (1883). 422 DETERMINATION OF ROTATION DISPERSION and a lower left-rotating half. By aid of a micrometer screw one of these wedges may be moved in front of the other. The vertical slit A (Fig. 70) is brought between B and C, and at a slight distance from B, and then between A and B, and immediately behind the slit, the double plate is so placed that its junction edges lie horizontally. The double plate must stand vertically to the direction of the light rays, so that they will pass through in the direction of the axis. Using sunlight, the spectrum with the Fraunhofer lines is found as before in the telescope, these lines being broken into upper and lower halves by the horizontal juncture surface of the double plate. As the plane of polarization of each color is turned to the right by the upper plate, and through the same angle to the left by the lower plate, it follows that for any position of the analyzer the spectrum does not show the same colors above and below, but at any given time those are extinguished for which, above and below, the angles of rotation differ by some multiple of 1 80. If then, the same color is to be extinguished above and below, the rotation for this color must be a multiple of 90 ; in this case, the polarizer and analyzer must stand in crossed or parallel position. Just which color is extinguished depends on the thickness of the quartz double plate ; if this thickness may be varied within sufficiently wide limits, it will be possible to follow the bands shut out from one end of the spectrum to the other and to bring them to coincide with any Fraunhofer lines desired. The determination of an angle of rotation for a Fraunhofer line is then made as follows : The line in question is brought between the parallel threads, the analyzer is then turned until at some point of the spectrum the two dark bands stand with one exactly above the other, and the movable wedge is then shoved until the dark band lies between the parallel threads ; this gives the zero-point position of the analyzer. If the active substance is now placed between the analyzer and the slit, the bands no longer stand one above the other, but to secure this the analyzer must be turned through the same angle by which the line chosen is rotated by the active body. It is assumed that the tempera- ture of the double plate does not appreciably change during the measurements. The spectropolarimeter constructed much THE METHOD OF V. LANG 423 later by v. Fleischl 1 is nothing but a bad copy of the Wiede- mann apparatus, the variable double plate of which is replaced by a simple quartz double plate of constant thickness, with which the rotation for a single definite color only may be measured. 155. The Method of v. Lang. The main drawback in the Broch method is that it depends on the not always available sunlight. But v. Lang has shown' 2 that the Broch procedure may be easily so changed that instead of sunlight with the Fraunhofer line? ordinary white light with artificial spectral lines may be employed. The arrangement of the apparatus is preferably that shown in the illustration (Fig. 71). By aid of the lens B, a sharp image of the source of white light A is thrown on the slit C. The light passing the slit is led first through a simple Mitscherlich polariscope consisting of the illumination lens D, the polarizer E, and the analyzer F, which may be rotated, and then through an ordinary spectroscope G to L. The I B D no L Fig: 71. focal distance of the illumination lens D is equal to half the distance between D and G, and the slit C is so placed that a sharp image of it may be formed by D at the objective slit G of the spectroscope. G is situated in the focus of the achro- matic lens H, and the prism J in the position of minimum deviation. The spectrum produced in the achromatic telescope K is seen at the ocular L along with two parallel vertical lines or fibers. The telescope K L is movable horizontally. The adjustment of the spectroscope is accomplished in the usual well known manner. The following operations are necessary to determine the angle of rotation of a substance for a given spectrum line : i . The zero point is found by turning the analyzer F to the position of greatest darkness. 2. The white i v. Fleischl : Repert. d. Exper.-Phys., 21, 323 (1885); Wied. Ann. Beibl., 9, 634 083 5 ). - v.I^ang: Pogg. Ann., 156, 422 (1875). 424 DETERMINATION OF ROTATION DISPERSION light is shut off and between B and C, and near the slit the apparatus is placed which furnishes the lines of the homo- geneous light in question, and the analyzer F is so turned as to transmit the largest possible amount of light. A continuous spectrum is no longer seen in the telescope, but only the sharp and bright spectrum lines. The telescope is then turned horizontally until the desired bright line comes between the parallel hairs. The source of homogeneous light is then re- moved, the strong white light A admitted, the active substance placed in position and the analyzer F turned until a black band appears in the telescope. By further turning the analyzer, this band is brought exactly between the parallel hairs, and the analyzer graduation is then read off. By again admitting the line of the homogeneous light it may be determined whether or not the position of the telescope has suffered any change. If it is desired to work with the Wiedemann double quartz plate of variable thickness, described in the preceding paragraph, F must be placed between Gand H, and the double plate directly in front of G and turned toward the polarizer E. In regard to making the determinations of rotation the direc- tions of the last paragraph obtain. In the production of the spectrum lines various sources of homogeneous light may be used. Salts of metals may be in- troduced into the Bunsen flame by a platinum loop, and in particular the chlorides or carbonates of sodium, lithium, thal- lium, potassium, and strontium. The light of a Geissler hy- drogen tube may be employed also. If a little mercury or some bits of cadmium are introduced into a Geissler tube, the light emitted on warming these metals may be used. In the following table all these artificial lines are given with their wave-lengths, and as seen, they are sufficient for the deter- mination of the rotation dispersion of a substance. LIPPICH' S METHOD 425 Color of the lines. Source of light. Wave-lengths in M/UL. Ka 768.0 Red Li a 670.8 Ha 656.3 Cd 643.8 Na 589.3 Yellow Hg 579-0 Hg 576.9 Hg 546.1 Green Tl Cd 4 535-1 508.6 H0 486.1 Cd 5 480.0 Blue Cd 6 467.8 Sr 8 460.8 Hg 435-9 H7 434-0 Violet Hg K 407.8 404.6 156. Lippich's Method. The methods of Broch and v. Lang give more accurate results the larger the rotation dispersion. But if this is small the dark band is broad and not sharp on the edges so that the adjustment is very inaccurate. In this case it is better to employ the method of Lippich 1 in the meas- urement of rotation dispersion. In this method the light passes into the spectroscope first and then into the polarization ap- paratus, which should be, preferably, the sensitive Lippich half-shadow instrument with double or triple field. The ar- rangement of the parts is shown in Fig. 72. By aid of the lens B a sharp image of the bright white light A is thrown on the vertical slit C of the spectrometer. C is situated at the focus of the achromatic lens D. The prism E is placed in the position of minimum deviation, and the light rays passing through it are thrown as a bright spectrum on the slit screen G, by means of the achromatic lens F, and in such a manner : Wiener Sitzungsber., II, 91, 1070 (1885). 426 DETERMINATION OF ROTATION DISPERSION that only one color of the spectrum may pass through the slit and serve for the illumination of the polarization apparatus. The slit G, therefore, represents the source of homogeneous light, and must be given such a position that a sharp image of the slit may be formed by the illumination lens J, of the polari- Fig. 72. zation instrument, on the analyzer diaphragm. In order to center the rays more completely it is recommended to add the lens H, immediately in front of G, which throws an image of D on the illumination lens J ; but A and B must be then so centered that D appears uniformly illuminated. By turning the slit tube C D any desired part of the spectrum may be thrown into the polarization apparatus, but A and B must be attached in fixed position with reference to this tube ; on ac- count of its small dimensions and slight weight the Linnemann zircona light burner (142) is recommended for this work. If one employs sunlight a small movable mirror is attached in place of A, by aid of which, with the different positions of the slit tube, the heliostat light may be thrown directly into the tube. According to Lippich the direct vision spectrometer of Hilger, in London, with the Christie half-prism 1 is also very satisfactory, as with this the simple turning of the prism is sufficient to bring any desired part of the spectrum to the slit G. The whole spectroscope, C to G, must first be graduated according to wave-lengths, in order to be able to give directly, for any position of C D, the mean wave-length of the light passing the slit G. For this purpose the slit C is illuminated 1 Proc. Roy. Soc., 6, 8 (1878) ; Pogg. Ann. Beibl., i, 556 (1877). LOMMEL'S METHOD 427 with the several homogeneous lights described in the last para- graph, one after the other, and the tube C D is turned so that the image of the illuminated slit C coincides exactly with G, which may be easily verified by aid of a reading glass. The different positions of the tube C D are read off on the gradu- ated circle, and are tabulated along with the corresponding wave-lengths of the homogeneous light. In this way a large number of definite positions of the tube C D are determined for which accurately characterized homogeneous colors pass through the slit G, and the angles of rotation of these may then be measured with the polarimeter in the usual manner. In the graduation of the spectrum apparatus the Fraunhofer lines of sunlight may of course be used. The greater the dis- persion the smaller must be the slits C and G, in order that the light emerging from G may have the necessary homogene- ity. If, finally, the rotation dispersion is very large, the Lip- pich method is then no longer applicable, since, in this case, a slit so narrow would be required that the necessary brightness of the field could not be secured ; it is then better to apply the procedure of Broch or v. Lang. Agreeing in principle with the Lippich method is that of Seyffart, 1 patented in 1886, for the determination of rotation dispersion. But as the latter is very complicated as regards the optical as well as the mechanical construction, without, at the same time, reaching the accuracy of the Lippich method, it need not be discussed here at length. The spectrosaccharim- eter introduced recently by Glan 2 resembles also the Lippich apparatus in principle, but is much inferior to it as regards convenience or accuracy in manipulation. 157. Lommel's Method. Lommel" has so modified the Broch method by addition of a quartz wedge that the final reading is made, as in the case of the Wild polaristrobometer (103), on the disappearance of interference bands. Sunlight is polarized by the prism A (Fig. 73), whose principal section makes an angle of 45 with the horizontal plane, and then reaches the vertical slit D. Just in front of this slit is a quartz wedge C, i Seyffart: Wied. Ann., 41, 113 (1890). - Glan: Ibid., 43, 441 (1891). 5 Lommel : Ibid., 36, 731 (1889). 428 DETERMINATION OF ROTATION DISPERSION with an angle of 7 or 8, whose edge, parallel to the optical axis, is vertical with reference to the slit, and immediately fol- lowing the latter is the analyzer E, the principal section of which either crosses or stands parallel with the principal section of the polarizer and, therefore, makes an angle of 45 with the horizontal plane. The refractive action of the wedge is corrected by that of a glass wedge, B, placed in reversed position. The light leaving the analyzer is decomposed by the prism F, which is placed in the position of minimum H Fig. 73- deviation, and is then collected by the achromatic lens G to form a spectrum that may be observed by the ocular H. As readily understood, this spectrum is filled with numerous dark, somewhat curved, interference bands which stand inclined with reference to the Fraunhofer lines, and it is also shaded by fine dark slanting lines. If the polarizer A is turned through 45, the shading disappears through the whole spec- trum (the zero-point). If now an active body is placed be- tween A and B the bands appear again. If, next, the polar- izer be rotated, a position is found at which a vertical bright band, free from shade, enters the field of view and with further turning travels through the spectrum. By bringing this bright band to coincide with the various Fraunhofer lines, one after the other, so that each line bisects the band exactly, and the corresponding angle is then read off on the graduated circle of the polarizer, the angle of rotation of the particular Fraunhofer line is obtained. The Lommel method may, of course, be so arranged that, as in v. Lang's method, artificial sources of light may be used. Reference only can be made here to Lommel' s interesting determination of the rotation dispersion of quartz by aid of the interference phenomena on a quart/ prism. 1 1 Wied. Ann.. 36, 733 (1889). LAXDOLT'S METHOD 429 158. Landolt's Method with Use of Ray Filters. While the methods thus far described for the measurement of rotation dispersion, necessitate the use of apparatus which is compli- cated for the chemist if not for the physicist, since the com- bination of a spectrometer with a polarimeter is always called for, that of Landolt 1 with ray filters requires the use of a polarimeter only. It may always be recommended, therefore, because of its convenience where the greatest accuracy is not required, and where the angles of rotation remain below 100. In the Landolt method, ordinary white light is used, and by aid of proper absorbing media all colors are removed, within rather narrow limits, except the one desired, which depends, of course, on the nature of the absorbing substance employed. If now, rotation measurements are made with these rather homogeneous colors, the rotations found must correspond, according to 147, with some definite optical center of gravity, which depends simply and alone on the relative distribution of brightness in the spectrum of the color in question, assuming that the rotations are measured by a Lippich instrument, and that the active substance absorbs all wave-lengths of this color uniformly. Therefore, since the optical center of gravity is determined by the source of white light, and the absorbing media, and changes with both, the directions given by Landolt for pre- paring the ray filters must be followed with the greatest exactness if the measurements of different observers are expected to be com- parable with each other. If modifications are adopted, the new optical center must certainly be defined and measured. The Welsbach lamp serves as a source of white light for these ray filters, and as absorbing media solutions of only such sub- stances are employed as are found in commerce in sufficiently pure condition. The solutions are filled into cylindrical glass cells of about 4 cm. diameter which consist of rings with plane glass plates cemented to them. One style of cell contains two compartments, each having an internal thickness of 20 mm., while another style has three divisions of 20, 15, and 15 mm. thickness. Each compartment is supplied with an opening, which may be closed with a glass stopper, and which permits 1 Landolt : Sitzungsber. d. Akad., Berlin, 1894, p. 923 ; Her. d. chem. Ges., 27, 2S 7 2 (1894). 430 DETERMINATION OF ROTATION DISPERSION the filling with the solutions. The cells may be shoved into a metallic framework having square plates on the corners to prevent rolling. These ray filters 1 are always to be placed be- tween the source of light and the illumination lens of the appa- ratus, and the zero point must always be found after they are placed in position. It should be determined anew for each ray filter, even when the half-shadow of the instrument has not been changed (see 106). The five filters permit red, yellow, green, light blue and dark blue to pass. Red. In the production of this color, the hydrochloride of hexamethylpararosaniline is used, which comes into commerce under the name of crystal violet 5 B O, and the anhydrous can- tharides green crystals must be selected. If 0.05 gram of this be dissolved first in a little alcohol and then diluted with water to one liter, this solution filled into a cell 20 mm. deep gives a spectrum which consists of a red band and a broad blue violet part. The latter may be completely absorbed by adding a layer of yellow potassium chromate solution 20 mm. thick, and con- taining 10 grams in 100 cc. The red band now remaining begins with the wave-length about 718 w, and ends sharply at 639 w*. The half-shadow of the instrument may be as small as about 3. Yellow. A solution of 30 grams of crystallized nickel sul- phate in loo cc., in a layer of 20 mm. thickness, absorbs only the red rays and permits all others to pass. If a cell 15 mm. deep is added, containing potassium monochromate solution with 10 grams in 100 cc., the blue and violet are taken out, leaving only orange-yellow and green. The last of these colors may be absorbed by means of a potassium permanganate solu- tion containing 0.025 gram in 100 cc., and used in a 15 mm. layer. The spectrum is now reduced to a narrow orange- yellow band which still shows a little red light and embraces the wave-lengths from 614 nv to 574 W. As the three absorption solutions weaken the light materially, it is neces- sary to employ a half -shadow of 8 to 10. Green. For this, a combination of potassium monochromate with cupric chloride is used. A solution of 60 grams of CuCl, -f- 2H,O to 100 cc. in a 20 mm. layer allows, practically, i Obtainable from Schmidt and Haensch, Berlin. LANDOLT'S METHOD 431 only green and blue rays to pass. The last may be absorbed by a 20 mm. layer of a potassium monochromate solution con- taining i o grams in ioocc., and there remains then abroad green band on the edge of which there is still a little blue. This band embraces the wave-lengths from 540^^ to 505^. The half-shadow must amount to 3 or more. Light Blue. In the production of this color there is used the compound known in commerce as double green S F, which is a combination of chlormethylhexamethylpararosaniline hy- drochloride with zinc chloride, and which appears as a glit- tering bronze-colored powder. An aqueous solution of 0.02 gram of the color to 100 cc. gives in a 20 mm. layer a spec- trum consisting of a narrow red band with a broad green and a light blue part ; the dark blue is absorbed. With a blue vitriol solution containing 15 grams in 100 cc. in a 20 mm. layer, the red band may be absorbed, but it is not possible to so remove the green light that light blue of sufficient intensity alone remains. The light left is, therefore, a combination of green rays of wave-lengths 526 yu/u to about 494/1^, and of light blue rays from 494.;^ to 458^. Asa result, no uniform color is found in the field of view of the polarization instru- ment, but the change of shade on turning the analyzer to and fro may still be followed. The half-shadow may be reduced to about 3. Dark Blue. This color is obtained by a combination of solutions of crystal violet 5 B O, with 0.005 gram in 100 cc. and blue vitriol with 15 grams in 100 cc. , both used in cells 20 mm. deep. The last solution absorbs the red rays which the aniline color passes a,nd there remains only dark blue light of wave-lengths 478 to 410^^. Because of the low intensity of the light, a half -shadow of about 8 must be taken. The absorption solutions, with the exception of the per- manganate, are permanent, but it is advisable to keep the supply of the aniline colors in the dark, and to renew the fill- ing in the cells holding them from time to time. But the per- manganate solution must be freshly prepared, since it easily suffers decomposition. As the rotation dispersion of quartz is pretty accurately known, the optical centers of gravity of the five colors from 432 DETERMINATION OF ROTATION DISPERSION the ray filters may be found by the aid of quartz plates, according to 148. Landolt has done this and has found that the five optical centers are near the Fraunhofer lines, C, D, E, F, and G. The following table shows the exact values : Color filter. Optical center of gravity in /K/U.. Fraunhofer's lines. Wave-lengths in /u/u. Red = nf Yellow =yl Green gr Light blue = Ib Dark blue = db 665.9 591-9 533-0 488.5 448.2 C D E F G 656.3 589.3 527.0 486.1 430.8 These relations then obtain for the rotations a with quartz plates: .032 = ,: .010 otyf = a D .026 OL sr Ot E .Oil Ot lt> = Otp -091 a db = a G By aid of these equations it is, therefore, possible to reduce rotations found by the ray filter method to those which obtain for the corresponding Fraunhofer lines, and this may always be done if the active substance has about the same rotation dispersion as quartz. The following table contains the data showing the prepara- tion of the filter solutions, along with their optical centers of gravity and the corresponding rotations for i mm. of quartz at 20 C. Color. lug I** JV Aqueous solutions of Grams of substance in 100 cc. of sol. Sfci '5,s =*- o8.2 %i\ 3H o" * Red 20 20 Crystal violet 5 B O Potassium monochromate 0.005 10 665.9 16.78 Yellow 20 15 *5 Nickel sulphate, NiSO 4 -f 7 aq Potassium monochromate Potassium permanganate 30 10 0.025 591-9 21.49 Green 20 20 Copper chloride, CuCl. 2 2 aq Potassium monochromate 60 10 533-0 26.85 Light blue 20 20 Double green S F Copper sulphate, CuSO 4 -f 5 aq 0.02 15 4885 32.39 Dark blue 20 20 Crystal violet 5 B O Copper sulphate, CuSO 4 -f 5 aq 0.005 15 448.2 39-05 ARONS-LUMMER MERCURY LAMP 433 159. The Arons-Lummer Mercury Lamp. The simplest method of determining rotation dispersion would be by aid of the Lippich half-shadow apparatus, if one had the means of producing a sufficiently large number of homogeneous lights of different wave-lengths. In 155 w r e have referred to a number of sources of homogeneous light besides sodium light, but they all have this disadvantage that their luminosity is too slight to make accurate observations possible. But, as a matter of fact, since there is need of homogeneous lights of great intensity for polarimetric work it may be safely assumed that before long this want will certainly be supplied. This end might be reached, for example, just as du Bois produced an intense so- dium light (146), by forming similar pencils of salts of other metals as lithium, thallium, potassium, strontium, etc. It would be a great advance to secure a sufficiently bright cadmium light, since accord- ing to Michelson 1 the four lines found in the red, green, and blue portions of the spec- trum (643.8, 508.6, 480.0, 467.8 /^) have an extremely constant optical center of gravity. At the present time we have, in addition to the sodium light, only one other source of homogeneous light of great intensity, and this is the Arons mercury light. The mercury lamp constructed by Arons 2 and improved later by Lummer, 3 is shown in Fig. 74.* The cylindrical tube B, with plane ends s, is connected near the middle with the two short vertical tubes A and C, which are closed below and furnished with platinum wires m and n, fused into the glass ; a small tube is fused to B at r. The tubes ABC are cleaned and dried, A and C are filled with mercury, r is drawn out, 1 Traveaux et mgmoires du bureau international des poids et mesures, Tome, XI, Paris (1895). Michelson, Jour, de Phys., [3], 3, 5 (1894). 2 Arons: Wied. Ann., 47, 767 (1892). ' I P> dO tnat is > M is a function of ex, /, /, and d. With the exception of p, the quantities ex, /, and d are variable with the temperature /, so that [or] also must vary with /. If we express this dependence on the temperature by the index /, then \_a~\ f =f(cr tj l t , p, d t ). If we give / the definite value / then the specific rotation for the temperature t is expressed properly only by the equation, [or], = /(/y, , /,,/, d tQ ) ; the quantities , /, and d must all be determined for the tempera- ture, / . It must, therefore, be considered as absolute!}' im- proper, as is unfortunately often the case to find a and /, for example, for 20, and d for 17.5, and to calculate an [a] with these values, because the [or] so found corresponds naturally to no definite temperature. Now, the coefficients of expansion corresponding to / and d are easily determined (in fact are mostly already known), so that / and d may be readily found for any temperature. Therefore, the temperature at which a is measured, is the one which must be taken as the basis for the whole calculation of [or] . If now the rotation is found at the temperature /,, (,), then /,, and d /t must be calculated to correspond, and finally [or]^ /(/ a , /^, p, d tl }\ in this way the specific rotation for the temperature /, is determined. If, next, a^ is observed in the same manner, we obtain [ar]., 2 f( a '-2i tf*> P> ^2)- ^ e obtain in this way a clear idea of the dependence of [or] on /, and finally after calculating the tem- perature coefficient of \_ai] , we are able to reduce all the ob- served values of [] to one and the same temperature, usually to 20. It must further be mentioned that the specific gravity is always referred to water at # ; the choice of any other tem- perature (o or 17.5) is certainly to be condemned, since water at 4 is the basis of the metric system of iv eight. 163. Schmidt and Haensch Control Tube. The Schmidt and Haensch control observation tube, 1 the use of which was re- 1 Schmidt and Haensch : Ztschr. fur Jnstrum., 4, 169 (1884^. 442 POLARIZATION TUBES ferred to in 128, is illustrated by Fig. 79. The tube A may be moved, in telescope fashion, into the tube B which it fits very accurately, a leather washer being added at C to make the joint perfectly tight. The ends D and E are closed by the ordinary screw caps. The motion of the tube A is accomplished by pinion and rack mechanism F ; on the latter there is a milli- meter scale which works over the vernier G, attached to B, so that tenths of a millimeter may be read off directly. The funnel H, which is detachable, serves to take up the excess of liquid expelled on contraction of the tubes. Instead of closing Fig. 79- the tube A at D it may be closed at the other end at J by screw- ing in a diaphragm with a plane parallel glass plate ; in this arrangement the whole tube is shorter by the length D J, than when the end is taken at D. On this principle (with dia- phragm at J) tubes may be constructed whose length may be decreased to zero. For saccharimeters which are made to take in very long tubes, such an arrangement would be very ad- vantageous as it would permit a determination of the whole error of the quartz wedge by use of a single sugar solution. For the ordinary short saccharimeters, the use of such a tube, which may be shortened to zero, is not to be recommended, since the determination of errors is less accurate, the shorter the tube in fully extended condition. It may be remarked regarding the practical use of the con- trol observation tube, that the solution cannot be introduced through the funnel H. This should be removed and the ori- fice closed by the metallic stopper furnished with the tube. The cap E is removed, the tube fully extended, and then the MEASUREMENT OF THE TUBE LENGTH 443 solution is poured in. After filling and closing E the tube is brought into horizontal position, H is attached, and then the tube is shortened a little and a further small amount of liquid poured into the funnel. To prevent too much evaporation it is recommended to cover the latter with a screw top which has a very small opening to permit the escape of the air. 164. Measurement of the Tube Length. The statement of the manufacturer is commonly taken as regards the length of the tube. But for all accurate work this length must be known with certainty to within o. i mm. , and it is, therefore, de- sirable to have means of measuring it one's self. A measuring instrument de- signed by lyandolt for this purpose is shown in Fig. So. 1 It consists of a me- tallic bar, A, divided into millimeters, on the lower end of which, at a, there is a sharp knife edge. The handle c, of wood, may be fixed at any point desired. The vernier b, provided with two knife edges, may be shoved along the bar with a little friction, and permits a reading to o. i mm. If the bar A is placed vertically with the edge a on a glass plate and then the ver- nier is shoved down until its two edges rest on the plate, the zero point on the bar and the zero on the vernier must agree, so that the reading zero is thus given. If the length of the tube B is to be found, one end is closed by ^a glass plate and screw cap, and then, in vertical direction, the measuring bar A is shoved down into the tube until the edge a touches the glass at the bottom. Then the vernier b is brought down until its two edges touch the glass tube above, taking care that the bar A does not stand 1 From Schmidt and Haensch, Berlin. Fig. 80. A 444 PERCENTAGE STRENGTH OF SOLUTIONS inclined in the tube. The length is then read off. This de- termination is repeated several times, the tube being turned through 90, 1 80, and 270, and the mean of all taken. If the length found is expected to be actually accurate within o. i mm. , the correctness of the measuring bar must be previously determined, and above all the temperature of the bar must be considered. 1 In all cases where it is desired to know the tube length more accurately than to o. i mm. , the simplest plan is to have the measurement made directly in the Physikalisch-Technischen Reichsanstalt at Charlottenburg. The length of polarization tubes may be there accurately found by aid of a comparator to within o.oi to 0.02 mm. If the angle of rotation of a liquid is found at different tem- peratures the change in length of the tube must be taken into consideration. It is sufficient for this purpose to consider the linear coefficient of expansion of glass as ft = 0.000008 for i, and for brass ft 0.000019. If the length of the tube at 20 be taken as 4,o, then the length at / is found from the equation, (I) **> ==*> + L* ft (t 20). If, for example, a glass tube measures exactly 200 mm. at 20, its length, according to this, at 30 is 200.016 mm. The cor- rection need, therefore, be considered only with long tubes and marked temperature changes. C. DETERMINATION OF THE PERCENTAGE STRENGTH OF SOLUTIONS 165. Reductions of Weighings to Vacuo. The determination of the percentage strength by weighing will be considered in what follows in such a manner as to make it as accurate as possible. If for special purposes, less accuracy is called for in the work, the procedure may naturally be easily simplified. As all weighings must be reduced to vacuo, the necessary for- mulas will first be stated. It must be assumed that these weighings are made on chemical balances which permit a mass determination which is accurate to about dr o. i mg. A few remarks only need be made about the weighings them- 1 The Physikalisch-Technischen Keichsatistalt at Charlottenburg makes tests of measuring rods and determinations of their errors in division. REDUCTIONS OF WEIGHINGS TO VACUO 445 selves. The temperature of the weights, of the body to be weighed and of the air in the balance case, should differ but little from the air temperature of the room. Do not place a vessel with calcium chloride in the balance case, as it is not possible to keep the air in the latter dry during the time of weighing ; at most the calcium chloride may be used as a protection against rust when the balance is not in use. All weighings should be made by the method of vibrations, with a determination each time of the zero point and the sensitiveness of the balance ; aside from its greater accuracy, weighings by the vibration method, when one is once accustomed to it, re- quire less time than is necessary to accurately adjust the rider. It is necessary to determine always three or five reversal points. In regard to a determination of the errors in the set of weights it is only necessary to compare one of the pieces with a nor- mal set of weights j 1 the errors of all the others may then be found from this with sufficient accuracy. The list of correc- tions should be pasted on the balance table, and at each weigh- ing the sum of the errors in the weights used should be noted at once. Further directions may be found in Kohlrausch's "Praktische Physik," 1896, 41 to 53. In the reduction of the weighings to vacuo, the density of the air A. (the mass of i cc. in grams) at the time of weighing must be known. The density of the air is a function of the temperature t in the balance case, of the barometric pressure, l\ in millimeters to be reduced to o, and of the tension e of the aqueous vapor in the air : ^ 0.001293 0-375 e i -f- 0.00367 t 760 As for North Germany e is usually between 3 and 15 mm., it is sufficient to take for the purpose e 9, so that we have then, (I) \= 0.001293 3-4 i -f- 0.00367 / 760 The error in A. occasioned by variations in e amounts in the maximum to 0.000004. If tne barometer is not measured on the same floor where the weighing is made, a correction must be made on b, since b decreases o. i mm. for an elevation 1 The Normal-Aichungscommission, at Berlin, makes such comparisons. 446 PERCENTAGE STRENGTH OF SOLUTIONS of i meter. If / is observed accurately to 0.5 and bio i mm., then by aid of equation (I) the air density A. is obtained cor- rectly to within about 0.000008. In the mean, A. is about 0.00119 and varies between the limits 0.00113 and 0.00125. To eliminate the error due to inequality in the arms of the balance, the body whose absolute mass is to be found is weighed first on the one pan and then on the other. If m^ and m., are the weights used in the two cases to hold the body in equilibrium, then the right weight of the body in the air is : (II) m=^^. The mass so found is then to be reduced to the vacuum weight as above. If s is the density of the body and 6 the density of the weights, then the desired mass M oi the body in vacuo is M m -f- m\ { ) , \ s 0" / or in better form for computation (III) M=m + mX ( *^. Ordinarily the weights are of brass, for which a = 8.4 may be taken ; the fact that the smaller weights are of platinum or other foil has no appreciable effect on the value of M. With reference to the densities, s, consult Landolt and Boernstein's Tables, 1894, P- JI 4 to 234, from which \vith the numbers on ff o p. 10 the values for A. - - may be taken. ffs If m is the apparent mass of the body in the air of density X, and m' the apparent mass of the body in air of density A.', then follows from equation (III), (IV) m' = m + m(\ A') ^ -. (TS By means of this equation it is possible to reduce the mass found for any density of the air A. to any other density A/. 166. Preparation of Solutions by Weighing. For this purpose it is advisable to use small flasks with wide neck and glass stopper, having a capacity of 20 to 150 cc. (Fig. 81). After cleaning the flask with water and rinsing with alcohol, it is dried by REDUCTIONS OF WEIGHINGS TO VACUO 447 forcing air for a long time through it. Then, according to the last paragraph (II), its apparent mass m in the air of density A. will be found. Next, one of the components A of the solution is introduced into the flask ; if a definite quantity of A is to be taken it is more convenient to use a coarse balance at first, and then by means of an analyt- ical balance to find the exact weight. According to (II) we have now the apparent weight n of the flask and sub- stance in the air, with the density equal j to A.'. After obtaining, according to (IV), 1 by aid of m, A. and X', the appa- rent mass of the flask m' at the air Fig - 8l - density A', then we have as the apparent mass ?t l of the sub- stance A at the air density A/, n l = n m ] n^ is reduced, according to (III), to the vacuum weight and we obtain then as the real mass M l of A, Then the second component B of the solution is brought into the flask and care is taken to make a homogeneous solution by shaking. Just as M l was found, we find now the true mass M of the solution composed of A and B, but we must substi- tute now in (III) in making the reduction, the specific gravity s of the solution. The real mass, M z , of the component B is then M=M M,. We proceed in the same manner if still other components are to be added. If the solution contains the components A and B only, of which B is the solvent, then the percentage strength, p, of active substance A in the solution is 100 M l (V) P = M It is to be observed that/, also the per cent, of the solvent, 1 The density of the glass may be taken as 2.5. PERCENTAGE STRENGTH OF SOLUTIONS q -.- j = 100 /, are quite independent of the tempera- ture ; the percentage amount p gives clearly the composition of the solution ; the solution is completely defined through p only. It is, therefore, advisable in following changes in specific rotation with increasing dilution of the solution, to represent the specific rotation as a function of p or q as is done on page 5, rather than as a function of the concentration. If w r ater is used as the solvent, it should be thoroughly boiled so as to expel all the air. In this manner, when necessary, the percentage strength p may be found with very great accuracy. If, for example, the normal sugar solution so commonly used in saccharimetry, in which the amount of sugar is about 23.7 per cent., is accu- rately prepared by weighing and is then filled into the polari- zation tube, in which operation a little water is lost by evapo- ration, the actual percentage strength of the solution in the tube, after allowing for all systematic errors, may be given to within Vjo.000 of. its value; that is, accurately to 2 or 3 units of the third decimal place. But if the weighings, are not re- duced to vacuo, the percentage strength would be too large by one unit in the second decimal place. 167. Change in the Percentage Strengthen Filtration of the Solution. But this great accuracy in the percentage strength no longer obtains if, on account of turbidity, one is obliged to filter the solution, because then by partial evaporation of the solvent, the percentage amount of the non-volatile constituent is in- creased. In order to obtain an estimate of the extent of the error so caused some experiments were made, and partly in this way that the whole filtering apparatus was placed on the balance, and partly by determining the strength before and after filtration. Plaited filters of Swedish paper were used and the funnels and vessels were kept covered as well as pos- sible. Aqueous solutions : a. 43.131 grams of water filtered in four minutes and lost 0.019 gram by evaporation. Air tempera- ture, 18. 99.614 grams of water filtered in 1 1 minutes with a loss of 0.041 gram. Temperature, 20. According to this, in the filtration of 40 to 100 grams of a 10 per cent, solution, DETERMINATION OF SPECIFIC GRAVITY 449 the percentage strength would increase by 0.004. b. In the filtration of 50 cc. of an aqueous solution of silver nitrate, which was completed in three minutes, the percentage strength increased from 9.708 to 9.713. The temperature was 20. The increase is 0.005, which agrees with the first determinations. Alcoholic solutions : a. 31.007 grams of 94 per cent, alcohol filtered in four minutes and lost 0.067 gram by evaporation, at a temperature of 18. 71.494 grams of the same alcohol filtered in ten minutes, at 19, and lost 0.114 gram. If the above alcohols had contained 10 per cent, of active substance, this amount would have been increased in the first filtration to 10.019 and in the second to 10.014 per cent. b. 50 cc. of a solution of silver nitrate in 78 per cent, alcohol filtered in ten minutes, at 23, and increased in strength from 9.686 to 9.714 per cent, in one experiment, and to 9.736 per cent, in another ; that is, by 0.03 and 0.05. If we may assume that the evaporation is independent of the percentage strength and proportional to the amount of liquid filtered, which is only approximately true, it may be estimated that in aqueous solutions after filtration, the amount is in- creased by about 0.005 for each 10 per cent, of substance, and for alcoholic solutions by about 0.03 to 0.05. These increases in percentage strength may, therefore, be quite considerable for concentrated solutions, and if alcohol serves as the inactive solvent the first decimal place even may be wrong by several units. Filtration, therefore, should be avoided as far as possible. D. DETERMINATION OF SPECIFIC GRAVITY 168. Construction and Use of the Pycnometer. The specific grav- ity or density of a body is the relation of its mass to the mass of the same volume of water at 4. The choice of any other temperature is inadmissible. If the body is homogeneous the specific gravity is at the same time the mass of the unit of volume, the gram and cubic centimeter being naturally used together. The densities of solid bodies are important in polarimetry only so far as employed in the reduction of weights to vacuo, and the methods of determination need not be discussed here. 29 450 DETERMINATION OF SPECIFIC GRAVITY Reference is made to Kohlrausch's "Praktische Physik," 1896, p. 57 to 60, and it may be remarked further that for the cases coming into consideration here the floating method devised by Retgers gives the best results. The densities of liquids may be found by aid of the pycnom- eter, the Mohr balance or the aerometer. But the last two methods require larger quantities of liquid and may be applied to solutions only when the evaporation of the sol vent is slight; but as they are easily subject to many systematic errors and do not possess at all the accuracy which is reached without difficulty with the pycuometer, the Mohr balance and the aerometer are only used in technical work for approximate determinations, while for scientific investigations the pycnom- eter alone may be employed. With the pycnometer the mass of a definite volume is de- termined. Among the many kinds of pycnometers that of Sprengel gives unquestionably the most accurate results. The most practical form is shown in Fig. 82. The vessel A holds about 15 cc. and is made of rather thin glass. The two capil- lary tubes B and C, bent in the middle, are fused to the upper end of A. The extremities of these are ground. The inter- nal diameter of the capillaries is not the same ; B, which has a mark at D is about 0.9 mm, while C is smaller and about 0.4 mm. in diameter. In order to be able to hang the pycnometer on the balance it is provided with a platinum loop E, as shown in the figure. To prevent evaporation during weighing, and possible loss of liquid by expansion with increase of tempera- ture, the ends B and C may be closed by the two ground glass caps F and G. To distinguish it from G the cap F is marked by a blue glass bead at its end since this one is to be put on the capillary marked at D. To facilitate drawing in the liquid the bent glass tube H with ground end may be attached to the wider capillary B, and to the narrow capillary C, the bulb J, with ground neck, to which a bit of rubber tubing is attached. The definite volume of the pycnometer, which is always used filled, reaches from the point of C to the mark D and changes only by expansion of the glass with the tem- perature. The pycnometer is cleaned by washing with distilled water CONSTRUCTION AND USE OF THE PYCNOMETER 451 Fig. 82. and alcohol, the latter being vaporized finally by a current of air. The apparent mass of the empty pycnometer with the two glass caps F and G in the air, is then a constant to within about 0.2 mg. , the changes on account of variations in the density of the air taken into consideration. In the determina- tions of specific gravity it is not necessary to make allowance for variable air density as may be seen in the following calcu- lation of errors for the specific gravity i . i . The greatest dev- iation in the density of the atmosphere from its mean value of 0.00119, or 0.0012, is about 0.00005. As the water weight of the pycnometer is not commonly redetermined with every new specific gravity test the density of the air can cause a maximum error of about 9 units in the fifth decimal place in the determination of liquid densities ; but it is assumed here that the empty pycnometer is reweighed before each determi- nation, and that, therefore, its apparent mass in the air can be brought into the calculation with an error of at most 0.05 452 DETERMINATION OF SPECIFIC GRAVITY mg. Leaving out of consideration rare exceptional cases, the error in the liquid density by variations in the air density is, as a rule, less than 4 units in the fifth decimal place. As will be further seen below, all other uncertainties taken together may produce an error of about 2 units in the fifth decimal place. By adding the error of four units on account of uncertain air density the final error is 6 units, that is, the error in a specific gravity determination is ordinarily only 6 units in the fifth place, even when the variable air density is left out of considera- tion ; this degree of accuracy in specific gravity determinations is sufficient for all polarimetric work. It is likewise unneces- sary, in specific gravity tests, to make double weighings in de- termining the apparent masses in the air ; if the weighing is always made on the same side the inequality of the balance arms is completely eliminated, since specific gravity is the re- lation of two masses ; it is assumed, of course, that the rela- tion of the two arms is sufficiently constant, which, for the small loads here taken, should always be the case in a good balance. The pycnometer is filled with distilled water, reboiled to free it from air, and brought into a bath of constant temperature. A cylindrical glass dish with a capacity of several liters may be used as a water-bath ; it should be placed in a round wooden vessel, the bottom of which is lined with a thick layer of cotton, and the space between the sides of the glass and wooden ves- sels should be also packed with cotton. For temperatures be- tween 15 and 25 the temperature of the water-bath will then remain constant to within a few hundredths of a degree. A standard with two movable arms is attached to the wooden vessel. One supports the holder K (Fig. 82), which maybe made of a bent glass rod or of a thick piece of covered copper wire, while the other supports a thermometer. The pycuo- meter is hung in the holder K, and is immersed in the water of the bath to the level of the mark I). The thermometer, graduated in tenths, the accuracy of which has been previously tested (see 161) is hung as close as possible to the pycnometer and care is taken to have the mercury in the bulb about at the level of the middle of A. If the water-bath is colder than the pycnometer, the liquid in the latter contracts on immersion in CONSTRUCTION AND USE OF THE PYCNOMETER 453 the bath, but only in the wider capillary B, while the nar- rower one C remains always full. If the water retreats be- yond the mark D, a drop of water on a glass rod is held to the point C. This is immediately drawn into the narrow capillary and the mass of water moves up in the wider tube B. Then the outside ends of B and C are carefully cleaned by bits of filter-paper. After about ten minutes, the pycnometer and contents will have come exactly^to the bath temperature, which may be recognized by the fact that the liquid meniscus in the capillary B remains constant in position. After the tempera- ture on the thermometer is read as accurately as possible, C is touched with a bit of filter- paper which causes the meniscus in B to move toward D ; at the moment in which the middle of the meniscus just reaches D the filter-paper is withdrawn and the adjustment is completed ; in the adjustment the eye is held so that the mark D appears as a line. Finally, the nar- row capillary C is first closed with the cap G and then B with the cap F. When this is done the pycnometer is taken from the bath and dried with a soft linen cloth, and then it is placed on the balance where its apparent mass in air is deter- mined. If from this the apparent mass of the empty pycnom- eter be subtracted, the difference gives the apparent mass in air of the water which fills the pycnometer at the temperature of the bath. This apparent mass, which depends on the tem- perature only, is a constant for the pycnometer used and need not be found anew for each specific gravity determination. After some practice such a degree of accuracy may be reached that with repeated adjustment and reductions to the same temperature, the apparent water masses will not vary by more than o.i to 0.2 mg. But it must be remembered that 15 grams of water weighed in the air may vary about 0.6 mg. in maximum by reason of changes in the air density. The above calculation of errors is based on these relations. After cleaning the pycnometer with alcohol and drying it, it is filled with the liquid whose specific gravity is to be found. All manipulations are now the same as before with water. The adjustment and weighings are made at least twice to avoid possible errors. By subtracting the apparent mass of the empty pycnometer from that of the filled, the apparent 454 DETERMINATION OF SPECIFIC GRAVITY mass of the liquid in the air is obtained. By repeated weigh- ings, and reductions to the same temperature, this should vary at most by o. i to 0.2 mg. 169. Calculation of the Specific Gravity. The following is the calculation of the specific gravity. We represent by W ot the apparent mass of the water in the air at the tem- perature / ; F, the apparent mass of the liquid in the air at the tem- perature / ; <2 , the specific gravity of the water at the temperature / ; 3/3, = 0.000024, the coefficient of cubical expansion of glass ; A., =0.0012, the air density; d t} the density of the liquid at the temperature /, referred to water at 4. Then \ ~ In this, the first factor is the rough uncorrected specific gravity, the second and third factors are corrections. The second factor corrects the specific gravity on account of tempera- ture changes, and the third comes from the reduction of the weights to vacuo. As already explained in the last paragraph, we obtain the specific gravity with the pycnometer to within about 6 units in the fifth decimal place. The specific gravity may be given, therefore, to the fifth decimal, but the first factor must be calculated with seven-place logarithms. If, in each weighing, the air density, A, is found according to equation (I) (165), then F' O F'O (II) d,-- ^ + ^ 3 ft ((,-/), in which W and P are the true masses of the water and the liquid, reduced to vacuo, by aid of equations (I) and (III) (165). The values of the specific gravity, Q , of water at different temperatures of the hydrogen thermometer are found from the following table, which contains the mean values of Thiesen, Scheel and Marek. The numbers are accurate to about five CALCULATION OF THE SPECIFIC GRAVITY 455 units in the sixth decimal place, as may be seen by com- parison with the latest observations. 1 A fuller table is found in L,andolt and Boernstein's Tables, 1894, p. 37. /. A. 4>. 4> 0+ 0-999 874 18.0 0.998 628 21.0 0.998023 I 930 i 609 i ooi 2 970 2 590 2 0.997 979 3 993 3 57i 3 957 4 I.OOO OOO 4 552 4 935 5 0.999 992 5 533 5 913 6 969 6 5H 6 890 7 931 7 495 7 868 8 878 8 476 8 846 9 812 9 456 9 823 JO 731 19.0 437 22.0 800 ii 637 i 417 I 778 12 530 2 397 2 755 13 410 3 377 3 732 14 277 4 357 4 709 15 132 5 337 5 685 16 0.998 976 6 317 6 662 7 296 7 639 17.0 808 8 276 8 615 i 790 9 255 9 592 2 772 3 755 20.0 235 23 568 4 737 I 214 24 326 2 193 5 719 3 172 25 073 6 701 4 151 26 0.996 811 7 683 27 540 8 664 5 130 28 260 9 646 6 109 29 0.995 971 7 087 Q 066 30 674 9 044 3i 368 If, in the equations (I) and (II), (/ i) amounts to but a 1 Thiesen, Scheel, and Diesselhorst : Wied. Ann., 60, 340 (1897) ; 'Die Thatigkjeit der Physikalisch-Technischen Reichsanstalt," Ztschr. fur Instrum., 17, (1897). 456 DETERMINATION OF SPECIFIC GRAVITY few degrees, it is sufficient to take for 3 ft the mean value 0.000024. But it is always better to find the coefficient of ex- pansion, 3/J, of the glass of the pycnometer by direct experi- ment. For this purpose, the pycnometer is weighed, filled with air-free water at two different temperatures. If, for ex- ample, several determinations are made at about 10 and 30, 3/S may be found accurately within about 2 per cent. If the two temperatures are /, and t v w r ith / 2 > /,, Q } and Q., the cor- responding specific gravities of water, W^ and W^ the masses of the water contained in the pycnometer at temperatures / t and /.,, then " W, 0, (',->,) Although the numerator may be simplified to ^l\Q l W\ Oi it is still more convenient in working with definite numbers to use the above form of the equation. If the density of the air remains constant throughout the experiment, it is not neces- sary to reduce the weights to vacuo, as 3 ft is dependent only on the ratio W^ : W,. 170. Variations in Specific Gravity with the Temperature. In all more accurate polarimetric work, as shown in 162, changes of specific gravity with the temperature must be known and, therefore, the coefficient of cubical expansion of the solution must be found. This is best done by aid of the pycnometer. Let this contain the mass of liquid F l at the tem- perature /,, and at the higher temperature f. 2 the mass F. r If 3 ft is the coefficient of cubical expansion of the glass, then the mean coefficient of expansion of the liquid, y, between /j and /.,, or also (with sufficient accuracy) the true coefficient for the temperature - L - , is given in form suitable for calculation by y ,,+,.== *i l \ As y is dependent on the ratio /* : F., only, it is not necessary, constant air density assumed, to reduce the weighings to vacuo. If the temperature difference, / 2 /,, is taken as about 15, the coefficient of expansion may be easily determined to within about 3 per cent., which is sufficient for all polarimetric work. VARIATIONS IN SPECIFIC GRAVITY 457 As far as the variation in y with the temperature is con- cerned, we may assume this relation with sufficient accuracy for simple liquids as well as for solutions of definite percentage strength, (V) y t = y^ + a(t-2o}, in which a is a constant peculiar to the liquid. If, therefore, we find y for two different temperatures, then we may calcu- late first the two constants / 20 , and a, and next the coefficient of expansion for any desired temperature ; it is required, there- fore, to find the mass of the liquid in the pycnometer for at least three different temperatures. For solutions, y varies also with the percentage strength p of dissolved substance, so that y must be written as a function of p and / in the form (VI) y = f + gp + h(t 20) + ip(t - 20), in which y, g, h, and i are four constants. If then the equation (V) is established first for the pure solvent, and then for a sin- gle solution of percentage strength p, the four constants may be found in a manner easily seen, and y may then be calculated for any desired/ and /. In case the equation (V) has been found for several solutions of different strengths, p, the four constants may be calculated by aid of the method of least squares. l As an example, we may take the expansion of a pure sugar solution in water. According to Schonrock 2 the dilation coefficient, y, between 11 and 26, of a sugar solution with percentage strength between p = o and p = 30, is given by the equation y = 0.000208 --{- 0.0000037^ -f 0.0000108 (/ 20) 0.000000 1 9 p(t 20). This formula gives the true coefficient of expansion accu- rately to within 0.000008. If the equation (VI) is found for y, and the specific gravity d at the definite temperature /' has been determined, the specific gravity for any other temperature / may be found from the equation (VII) d t = d, + d, YiL (/' ~ t). 2 1 See Kohlrausch : "Praktische Physik," 1896, p. 9. - Schonrock : Ztschr. fiir lustrum., 16, 243 (1896). E. DETERMINATION OF THE CONCENTRATION OF SOLUTIONS 171. Calculation of the Concentration from the Specific Gravity and Percentage Strength. The concentration c, by which is under- stood the number of grams of active substance in 100 cc. of solution, is found by taking the product of the percentage strength p, and the specific gravity d tt found as explained above : (I) c t =pd t . While, now, the percentage strength p is perfectly indepen- dent of the temperature, the concentration c varies with it. We have, in analogy with equation (VII) (170), (II) c t = Ce+CtYt+At t). 2 Asp and d may be determined with great accuracy, it fol- lows from (I) that c may be also exactly found. As we have seen in 166, it is possible, for example, to determine the per- centage strength of a 24 per cent, sugar solution to about three units in the third decimal place. As, further, accord- ing to i 68, its specific gravity may be found accurately to about _ six units in the fifth decimal place, it follows that the concentration (about 26) may be found for the same temper- ature to within about Veooo f * ts amount; that is, to 4 units in the third decimal place accurately. But equation (/) gives the exact concentration only when the specific gravity d is referred to water at 4. Of course, in this case, we understand by i cc. the volume which i gram of water occupies at 4 weighed in vacua. This cubic centimeter, almost universally employed in scientific work, is usually designated as the true cubic centimeter, 1 and is always understood in what follows, unless something else is mentioned. This true or prac- tical cubic centimeter is, of course, to be distinguished from the theoretical cubic centimeter; that is, the volume of a cube whose edge is i cm. long. From the investigations of Men- deleeff and Mace de Lepinay, it follows that 100 practical cubic centimeters = 100.010 theoretical cubic centimeters. If, for example, 100 practical cubic centimeters contain 26 grams 1 A distinguished from the Mohr cubic centimeter. PREPARATION OF SOLUTIONS IN MEASURING FLASKS 459 of substance, then, with reference to theoretical cubic centi- meters, c = 25.9974 ; in scientific work calculations are never based on theoretical cubic centimeters. Concerning the Mohr cubic centimeter, no longer used in science, see 126. 172. Preparation of Solutions in Measuring Flasks. The concen- tration may also be found directly by dissolving a weighed amount 1 of the active substance in a measuring flask of definite volume. But this method has the disadvantage that the deter- mination of volume in the measuring flask is by no means as accurate as in the pycnometer, and further, that larger vol- umes of liquid must be brought to a constant temperature, which requires a correspondingly longer time. For exact experiments the method with the pycnometer is decidedly pref- erable. The measuring flasks to be used are illustrated in Fig. 83, and have a volume of 20 to 200 cc. , with a neck about 10 mm. wide ; the circular mark on the neck should be down near the body so as to dimin- ish inequality in the solution as far as possible. Measuring flasks with a long narrow neck divided into tenths of cubic centimeters, and closed with ground glass stoppers, and which have still sufficient mix- ing space above the upper mark so that solutions with any volume be- tween loo cc. and no cc. may be made, are very practical. Before the solution, made either in the flask itself or in another vessel, is diluted to the mark, a thermometer is dipped in it and the normal tem- perature (20) secured by help of a water-bath. Finally, the thermometer is lifted, rinsed off with a little of the solvent, more is filled in to the mark and the whole is well shaken after inserting the glass stopper. The contents of the measuring flask must be accurately de- 1 The weight must, of course, be reduced to vacuo. 460 CONCENTRATION OF SOLUTIONS termined by weighing before use. For this purpose, it is filled nearly to the mark with air-free water, a thermometer is in- serted and by warming or cooling in a water-bath it is brought to the normal temperature /, for which it is to be used. After withdrawing the thermometer, enough water is added to bring the lower edge of the concave liquid surface exactly tangent to the mark when viewed horizontally. Then, all drops hanging in the neck of the flask are removed and the mass P of the water, in vacuo, found according to 165 and 166. If the specific gravity of water at the temperature t of experiment is Q (see table, 169), then the volume V t in cubic centimeters, which the flask contains at t, is given by the equation : It is possible in this manner, with several weighings, to find the volume of the flask within about 0.03 cc. When a measuring flask is to be used at any other tempera- ture /' than the temperature t for which it was graduated, the cubical expansion of the glass must be taken into consideration. If 3 ft represents this coefficient (in the mean 0.000024), and V t the volume of the flask at the temperature of graduation, then this formula may be used in finding the volume at /' : (IV) V t ,--~- V t + V (Z ft(t'-t). To obtain an idea of the errors connected with the determi- nation of concentration of solutions made in a measuring flask, we can take as an illustration the preparation of the normal sugar solution commonly employed in saccharimetry. Since at 17.5, 26.003 grams of sugar (true mass) must be dissolved to make 100 cc. (126), the concentration of a correct normal solution is f, 7 . s = 26.003. If we now assume that the sugar is correctly weighed out and that errors in the flask in con- junction with inaccuracies in filling to the mark amount to only 1 0.02 cc., the concentration is already wrong by =F 0.0052, an error which is larger than that found in 171. On the other hand, assuming that at the time of preparing the solution the water has a temperature of 18.5 instead of 17.5, according to (IV) the volume of the measuring flask is no longer icocc., but 100.0024, and the 26.003 grams of sugar are contained in this. Accordingly the concentration of the pre- CALCULATION OF ERRORS 461 pared solution is c lB .. = 26.0024 ; as further, the coefficient of expansion of a normal sugar solution at 18 is 0.000283, then, according to (II), r 17 . 3 = 26.0098. In this case, therefore, a temperature difference of i in the preparation of the solution makes a difference of 0.007 in the concentration. 1 // is seen from this how closely the temperature must be controlled when solutions are prepared in the measuring flask. F. EFFECT OF THE DIFFERENT ERRORS OF OBSERVATION ON THE SPECIFIC ROTATION 173. Calculation of Errors. Each of the factors entering the i oo or 100 at . formulas \ci\ -= or - , is attended by a certain error of observation. If we represent by /() the error in or, by f(a) the error in [or] caused by /(a), and if/( /),/(/). . -F(l), F(6} have the same meanings with reference to/, /, then, since [a] is proportional to a j /,..., the following simple relations exist : , In the calculation of the errors, /% it is only necessary to use approximate values and two places of figures will be suffi- cient. To secure a fairly close idea of the influence which each error has on the value of [a] , we shall carry through the com- putation for the case of the determination of the specific rota- tion of cane-sugar in water where [a] is about 66. Let a = 35, / = 2 dm., p = 24, d == i.i ; these factors may be found with about this degree of accuracy :/() = dz 0.004, /(/) i= d= 0.0002 dm., f(p} = o.oo3,/(y) = dz 0.00006. From these the following errors are calculated for \_ot~] \ F(a) 0.008, F(l} == q= 0.007, F(p) = =p 0.008, F(d) = =F 0.004. As seen, the errors in [ar] are of the same order, so that the final accuracy in [] is about 0.014, or 0.02 per cent. ; that is, the specific rotation in this case is found to within about Vsooo f ^ s value. In all investigations one must take 1 Corresponding to 0.03 Yentzke. 462 CALCULATION OF ERRORS into account, in this manner, the degree of accuracy with which the different measurements may be carried out, in order that an estimate of the value of the final result is possible. The chemist may reply that such great accuracy in the specific rotation is not to be reached because of the chemical impurity of the substance investigated. While this is actually true in very many cases, this very fact should call for more accu- rate work in the determination of specific rotation. // is only after preparations, which have been made or purified by different methods, have in turn been examined with great accuracy, with the result that differences in the specific rotation arc found which lie quite outside the errors of observation, that it may be stated with positiveness that these differences are due to impurities in the material investigated. At the present time, in the great majority of cases, and especially in respect to the commonest and most frequently studied bodies, 1 we are still quite uncertain whether the different values in the specific rotation, found by different observers, are due to errors of observation or to im- purities in the substances. Along with accidental errors, which follow from uncertainties in the observations, there are the systematic errors which are due in the main to the pecu- liarities in individual instruments, and to these the greatest attention must be paid. Considering the delicacy of instru- ments and methods to-day, certainly no great skill is required to obtain results which agree perfectly with each other after repeating whole series of observations by one and the same method ; but the skill of the observer is to be judged rather by his success in eliminating systematic errors by variations in methods and accurate investigation of the apparatus employed. Bessel's statement cannot be too highly appreciated, that every piece of apparatus must be twice constructed, first by the instru- ment-maker, then by the observer ; which is to say, that be- fore use every measuring instrument must be accurately investi- gated as to its errors. For fuller details concerning calculation of errors, see Kohl- rausch's "Praktische Physik," 1896, p. i to 27, or Ostwald's "Physiko-chemische Messungen," 1893, p. i to 18 (consider especially page 9). 1 See, for example, the constants of rotation of cane-sugar, tartaric acid, alkaloids, etc., given in Part VI. PART FIFTH Practical Applications of Optical Rotation i. Determination of Cane-Sugar. Saccharimetry A. Determination of Sugar with Instruments having a Circular Graduation 174. By aid of the polaristrobometers described in 98 to 121, the concentration c or the number of grams of sugar in a solution, may be found by the following formula, where we measure the angle of rotation a D for a layer / decimeters in length : 100 Of = 7T^r As shown by Table II on page 465, to follow, we can take for the specific rotation of cane-sugar the constant value [] D =- + 66.5, for all concentrations below c = 30. This is sufficiently close for practical work, as c may be found from it with accu- racy to o.oi or 0.02. If this number is substituted in the above equation, there follows, c~ 1.504 -j-. For a 2 dm, tube C= 0.752 a D . The percentage amount of pure sugar in a solid saccharine body, of which P grams have been dissolved to make 100 cc., and which is examined in an instrument with circular gradu- ation, is given by the proportion, P ' : 0.752 a : : 100 \x ; 75-2 /, ~~P~ 175. If we have to analyze very strong sugar solutions, or if the greatest degree of accuracy is desired, then the change in 464 PRACTICAL APPLICATIONS OF OPTICAL ROTATION the specific rotation \_ae~] with the concentration must be taken into consideration. The following formulas are given to show the dependence of the specific rotation of cane-sugar on the percentage strength / : I. [] =66.386+ 0.015035 / o.ooo39S6/>-( Tollens.) 1 II. [#] = 66.438 - o.oio3i2/> o. 0003545 /> 2 (Nasini and Villavecchia) 1 The following table (Table I) contains in columns d and e the specific rotations calculated according to the above formulas, and corresponding to percentage amounts of sugar (column a), increasing from 5 to 5 per cent : TABLE I a. b. c. d. e. f. - Per Sp.gr. rf-- 4 Interpolated Con- centration Specific rotation [tf] . cent from the (c=fi.d amount. nearest values of Tollens. according to Tollens). Calculated by Formula I (Tollens). re- ferred to Calculated by Formula II (Nasini), re- ferred to Calculated by Formula III.- referred to P- d. c. P- P- c. 5 I.OI786 5-0893 66.451 66.480 66.473 10 1.03819 10.3819 66.496 66.506 66.500 15 1.05926 15.8889 66.522 66.513 66.514 20 .08109 2I.62I8 66.527 66.502 66.513 25 .10375 27.5938 66.513 66.474 66.496 30 .12721 33-8163 66.479 66.428 66.460 35 15*53 40.3036 66.424 66.365 66.404 40 .17676 47-0704 66.350 66.283 66.324 45 .20288 54.1296 66.256 66.184 66.217 50 .22995 61.4975 66.142 66.067 66.081 The values in columns d and e may be used when the per- centage amount of sugar, p, in 100 parts by weight of a solu- tion is to be found, but the specific gravity must also be known. We find p from looar = Id \a\ ' We proceed in this way. An approximate value for \_a\ is substituted and/> calculated ; then the exact value of [] is 1 See Part VI, Constants of Rotation. '-' See next page. SACCHARIMETRY 465 taken from the table or is interpolated, and on substitution of this in the formula the exact value of p may be found. But the case is much more common in which we desire to find, not the percentage amount, but the concentration of a solution. The formula given above, 100 Of c =- r .. , possesses this advantage that the specific gravity of the sugar solution investigated need not be known, and that at the same time the latter may contain inactive substances also along with the sugar. As up to the present time we have had no formula which presented the specific rotation of cane-sugar as depend- ent on the concentration, the following new one has been cal- culated from the observations of Tollens and Nasini : III. [**]/? = 66.435 0.00870 r o.ooo 235 c- (holds for r -o to 65). The values of [<*] , according to this formula, are given in column f of the above table opposite the corresponding values from Formulas I and II. It will be recognized that all the values from Formula III lie within those from the formulas of Tollens and Nasini. As the latter differ from each other only by amounts which cor- respond to the unavoidable errors of observation, Formula III, for the concentration, possesses a degree of accuracy which satisfies all practical requirements. The specific rotation of sugar solutions with from i to 65 grams of sugar in 100 cc. is then given by the following : TABUS II. c. Mi Diff. for c = i. C. MS- Diff. for c = i. I 5 10 15 20 25 30 66.443 66.473 66-499 66.513 66.515 66.506 66.485 0.0075 0.0052 0.0028 -f 0.0004 0.0018 0.0042 -- 0.0066 35 40 45 50 55 60 65 + 66.452 66.407 66.351 66.283 66.203 66.ni 66.007 0.0090 O.OII2 0.0136 0.0160 0.0184 0.0208 The following example shows that the change in the specific rotation is marked enough to appreciably affect the results of optical analysis with solutions of considerable concentration. 30 466 PRACTICAL APPLICATIONS OF OPTICAL ROTATION Let a rotation of 83. 1 1 be found with a 2 dm. tube. Accord- ingly, the concentration would be 0.752 X 83. 1 1 = 62.5 grams of sugar in 100 cc. But, according to the table, the specific rotation of sugar for this concentration is 66.06. The true concentration will then be found from the equation : loo X 83.11 C - 2 X 66.06 " 62 ' 9 ' gramS " B. Determination of Cane-Sugar with Application of Wedge- Compensation Instruments and the Ventzke Scale. 176. These instruments which have been described in detail in 122 to 137 are the only ones practically employed in the sugar industry. In the course of time many different directions have been given for the manner of using them, as well as for the preparation of solutions of different saccharine substances, but at the same time experience has shown that in the results of different observers differences are often found, the cause of which must lie in the lack of uniformity in methods of pro- cedure. In consequence of this, it has become necessary from the side of the sugar chemist, as well as from that of revenue administration to establish definite methods, and this has been recently done by the Rules of Procedure provided by the German Sugar Tax Law of May 27, 1896, appendices A, B, C, and E. The provisions in the last are based in part on many in- vestigations carried out in the laboratory of the Society for Promotion of the German Beet-Sugar Industry, and particu- larly those of Herzfeld, 1 and partly on decisions reached in the meetings of societies of commercial chemists. 2 1 See especially the following : Herzfeld : Ztschr. Riihenzucker-Ind., 40, 167 to 214 (1890), "Die Bestimmung des Zuckergehaltes der Handelswaare ;" 41, 685 (1891), "Best, des Invertzuckers in Melassen ;" 43,14710 259(1892), "Ueberdie zweckmassigste Art der Werthschatzung des Rohzuckers ;" 43, (1893), "Die Wasserbestimmung ira Roh- zucker." Also, Hammerschmidt : Ztschr. Riibenzucker-Ind., 40, 465 (1890), "Verall- gemeinerung der Clerget'schen Methode." 41, 157 (1891), "Bestimmung der Saccharose mittelst der Inversionsmethode." Besides these, many other papers. 2 See Ztschr. Riibenzucker-Ind., 36, 6 (1886), "Bericht iiberdie Sitzung der Invert - zucker-Commission in Magdeburg vom 5 Dec., 1885," and page 11 appendix to this ; 40, 439 (1890), "Rundschreiben vom 6 Juli, 1890, an die Handelschemiker, betr. die Bestim- mung der Ram nose und des Invertzuckers;" page 443, "Anleitung zur Bestimmtmg des Gehaltes an Ramnose und Invertzucker in den Producten der deutschen Ruben zuckerfabrikation ; " page 446, "Arbeitsvorschrift fur die Invertzuckerbestimmung ; " 4S.73(i895). Allg. Theil, "Bericht iiber die Versammlung der Handelschemiker vom 12 Marz, 1895, in Berlin; "46, 180 (1896), Allg. Theil, "Sit/.ung (k-r Commission der Handelschemiker behufs Priifung von Normalquarzplatten xur Controle der Sac. charimeter. " SACCHARIMETRY 467 177. For the work in hand, it appears most practical to give below an exact reprint of the four appendices of the sugar tax law referred to (in the order, C, A, B, and E). 1 [Note by Translator. Although the directions given below apply in some cases to German conditions only, it was thought best to allow them to stand as written, inasmuch as they are suggestive and have been pretty generally followed in the prac- tice of other countries. Permission was given by the author to modify this section at the discretion of the translator, but instead of doing this, attention will be called to the following books and pamphlets where other details may be obtained : Wiley's " Agricultural Analysis." Vol. III. Parts 2 and 3. Allen's "Commercial Organic Analysis." Vol. I, 3rd ed. p. 243-379. 1 ' Methods of Analysis adopted by the Association of Official Agricultural Chemists." 1898. p. 27-40. " Revised Regulations Governing the Sampling and Classi- fication of Imported Sugar and Molasses. U. S. Treasury Department." Document No. 2113.] Appendix C DIRECTIONS FOR MAKING THE POLARIZATIONS Polariscope. In making polarizations for the purpose of revenue assessment, theVentzke-Soleil color apparatus or a half-shadow saccha- rimeter only may be employed. For both instruments, one degree of rota- tion in a 200 mm. tube, at 17.5, corresponds to a strength of 0.26048 gram of sugar in 100 cc. of liquid ; 2 a sugar solution which contains 26.048 grams in 100 cc. the so-called normal weight produces accordingly a rotation of ioo c . Therefore, when a solution of a substance is examined in a 200 mm. tube, and it contains 26.048 grams dissolved to make 100 cc., the degrees of the scale indicate the percentage amount of sugar present. If only the half of this normal weight is dissolved, the number of degrees read off must be doubled to obtain the correct per cent, of sugar. The same is true for those cases in which the examination is made in a 100 1 Taken from Ztschr. Riibenzucker-Ind., 46, 410 to 427, and 435 to 439(1896). The description of all other methods employed in the laboratories of sugar facto- ries may be found in the work of Friihling andSchulz : "Anleitung zur Untersuchung ecause on longer standing it becomes valueless for analysis. SACCHARIMETRY 473 The liquid mixed with Fehling solution is heated in a flask on gauze over a Bunsen burner or a good alcohol lamp, brought to boiling and kept in ebullition two minutes. This time of boiling must not be shortened. Then the lamp is removed and the liquid is allowed to stand at rest a few minutes to permit a precipitate to settle ; the flask is held toward the light to see whether or not a blue color remains. If there is still copper in the solution, which is shown by a blue color, the solution contained less than 2 per cent, of invert sugar. The color is more easily recognized by holding a sheet of white paper back of the flask and examining it in reflected light. If, after boiling, the liquid appears yellowish green or brownish it is possible that undecomposed copper solution is still present, but masked in color by the yellowish brown color of the sirup. In such cases, pro- ceed as follows : A small filter is made of good thick filter-paper, placed in a glass funnel and moistened with a little water so that it may be pressed against the edge of the funnel. The funnel is placed over a test-tube ; then, about 10 cc. of the boiled liquid is filtered, and to the filtrate about the same volume of acetic acid and one or two drops of an aquequs solution of potassium ferrocyanide are added. If an intense red color appears in the filtrate, copper is still in solution and it is so shown that the sirup con- tains less than 2 per cent of invert sugar. 2. Determination of the Quotient for Sugar Sirup Containing Less than 2 Per Cent, of Invert Sugar The quotient (or coefficient of purity) is that per cent, of sugar in the solids of the sirup which may be calculated from the polarization and the specific gravity on the Brix scale. a. To Find the Specific Gravity in Brix Degrees. In a tared beaker, weigh off 200 to 300 grams of the sirup to be tested, loo to 200 cc. of warm distilled water is added, the mixture is carefully stirred (to avoid breaking the glass) until the whole is brought into solu- tion, and then the beaker is placed in cold water until the contents have cooled to the room temperature. Then the beaker is placed on a balance, and water is carefully added from a wash-bottle until the whole weight of added water is just equal to that of the sirup taken. For example, if 251 grams of sirup were taken for testing, then water must be added until the liquid weighs 502 grams. After adding the water the liquid is stirred and filled then into the specific gravity cylinder so far that, after immer- sion of the Brix spindle, it does not reach quite to the upper edge. The cylinder must be placed in a vertical position so that the spindle will float freely without touching the sides. The spindle is immersed slowly, and care must be taken not to moisten that part of the stem which remains above the liquid after the spindle has come to rest. When this condition is reached, the saccharometer degrees are read off at the point where the liquid meniscus cuts the stem. 474 PRACTICAL APPLICATIONS OF OPTICAL ROTATION The number of degrees read off on the spindle holds for the normal temperature of 17.5 C. If the liquid does not happen to have this normal temperature, the degrees read off must be corrected by aid of the following table, after the true temperature is found by means of a ther- mometer attached to the body of the spindle. After the correction, the Brix degrees are to be rounded off in tenths, by considering five or more hundredths as a full tenth, and smaller fractions neglected. The number of degrees read off are to be multiplied by 2, because the liquid employed in the test had been diluted with an equal weight of water. TABLE FOR THE CORRECTION OF BRIX DEGREES FOR TEMPERATURES DIFFERENT FROM THE NORMAL TEMPERATURE (17.5 C.) At a centigrade temperature of and at 25 30 35 40 50 60 70 75 Brix degrees. There must be taken from the saccharometer readings : Degrees. 0.72 0.82 0.92 0.98 I. II 1.22 1.25 1.2 9 5 0.59 0.65 0.72 0-75 0.80 0.88 0.91 0.94 10 0-39 0.42 0-45 0.48 0.50 0.54 0.58 0.61 II 0-34 0.36 0-39 0.41 0.43 0.47 0.50 0-53 12 0.29 0.31 0-33 0-34 0.36 0.40 0.42 0.46 13 0.24 0.26 0.27 0.28 0.29 0.33 0.35 o-39 H 0.19 0.21 0.22 0.22 0.23 0.26 0.28 0.32 15 0.15 o. 16 0.17 o. 16 0.17 0.19 0.21 0.25 16 O.IO O.I I 0.12 0.12 0.12 0.14 o. 16 0.18 17 0.04 0.04 0.04 0.04 0.04 0.05 0.05 0.06 There must be added to the saccharometer readings : 18 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.02 19 0.10 O. IO O.IO O.IO O.IO O.IO 0.08 O.O6 20 0.18 0.18 0.18 0.19 0.19 0.18 0.15 O.I I 21 0.25 0.25 0.25 0.26 0.26 0.25 O.22 0.18 22 0.32 0.32 0.32 0.33 0-34 0.32 0.29 0.25 23 0-39 0-39 0-39 0.40 0.42 0-39 0.36 o.33 24 0.46 0.46 0.47 0.47 0.50 0.46 0.43 0.40 25 0.53 0-54 0-55 0-55 0.58 o.54 0.51 0.48 26 0.60 0.61 0.62 0.62 0.66 0.62 0.58 0-55 27 0.68 0.68 0.69 0.70 0.74 0.70 0.65 0.62 28 0.76 0.76 0.78 0.78 0.82 0.78 0.72 0.70 29 0.84 0.84 0.86 0.86 0.90 0.86 0.80 0.78 30 0.92 0.92 0.94 0.94 0.98 0.94 0.88 0.86 SACCHARIMETRY 475 b. Polarization In the polarization of sugar sirups, because of the dark color, the direc- tions given in Appendix C for determination must be modified in these respects : Only the half weight, 13.024 grams, is taken in testing sugar sirup. This is weighed into a porcelain dish and treated with 40 to 50 cc. of lukewarm distilled water, and stirred with a glass rod until it has dis- solved completely. Then the liquid is washed into the flask and before filling to the mark is clarified. For the clarification about 5 cc. of basic lead acetate solution is run first into the flask. If after the precipitate has settled, which follows in a few minutes, the liquid is still too dark the addition of the lead acetate is continued until the desired brightness is secured. As much as 12 cc. of the basic acetate is often required for this. But it must be observed that the basic acetate must be added in sufficient, but not in excessive, quantity ; each new drop added must produce a precipitate in the liquid. If it is found that the latter cannot be sufficiently clarified by the addi- tion of basic lead acetate to be polarized in the 200 mm. tube, an effort should be made to polarize it in the 100 mm. tube. If this is likewise impossible a new sample should be prepared, which is treated with about 10 cc. of an alum or tannic acid solution before the addition of the basic lead acetate ; these solutions produce heavy precipitates with basic lead acetate, which have a clarifying effect and permit the use of larger quan- tities of the lead solution. After the polarization is made the number of degrees read off must be multiplied by 2, since only the half-normal weight was taken for the test. If a zoo mm. tube was employed in place of the 200 mm. tube the num- ber of degrees read off must be multiplied by 4. c. Calculation of the Quotient If the observed number of Brix degrees be designated by B, and the degree of polarization by P, then the quotient is calculated by the for- mula, Q = ^ . In stating the final result smaller fractions than full D tenths are omitted. Illustration of the Determination of the Quotient. 200 grams of a sugar sirup are diluted with 200 grams of water. The Brix spindle indi- cates 35.2 at a temperature of 21 C ; from the above table 0.25 must be added ; this gives then 35.45, or rounded off 35.5, and after multiplying by 2, 71 Brix. The polarization of the half-normal weight in a 200 mm. tube gave 25.2; the true polarization is then 25.2 X 2 = 50.4. The quotient calculated is therefore - = 70.9. Final Provision Revision reports must contain the following data : the result of the test for invert sugar, the number of degrees read off on the areometer, 476 PRACTICAL APPLICATIONS OF OPTICAL ROTATION the temperature of the solution, the calculated areometer degrees for the undiluted sirup, the polarization for the whole normal weight and the quotient. Appendix B DIRECTIONS FOR CHEMISTS I. In determining the quotient for sirups containing 2 per cent, or more of invert sugar, and the quotient of sirups to be examined for raffi- nose, also, II. In determining the amount of sugar in crystal sugar supposed to con- tain raffinose. I. The Quotient for Sirups According to the regulations provided by the sugar tax law the deter- mination of the quotient for a sirup shall be left to a chemist when : a. There is no official properly qualified to determine the quotient at the point of declaration or at the office to which the sample is sent; b. The sirup contains 2 per cent, or more of invert sugar; c. The one presenting the sample asks for the calculation of the quo- tient from the amount of pure sugar chemically determined. When samples are sent from the revenue office to a chemist he must be informed as to which one of the above grounds the investigation is called for, and besides in cases coming under r, whether or not the application of the raffinose formula is allowable, according to the directions of #2, section 5 of the last part of the Rules of Procedure, where 2 or more per cent, of invert sugar may be present. In cases under a the chemist must proceed as in Appendix A of the Rules of Procedure, but with the condition that the Brix degrees are to be found as given in the following section i . In cases under b the quotient must be found as explained in the follow- ing section i. In cases under c, as far as the use of the raffinose formula is allowable, the method of section 2 below must be followed, otherwise the method is according to the provisions of section i . If the propriety of using the raffinose formula depends on whether or not the sirup contains less than 2 per cent, of invert sugar, then the sirup must be tested according to the method of section i in Appendix A. /. Determination of the Quotient in Sirups Containing 2 Per Cent, or More of Invert Sugar In the investigation of sirups containing 2 per cent, or more of invert sugar the Brix degrees must be calculated from the specific gravity of the undiluted sirup found by aid of a pycnometer. If a quotient of 70 or more is found from the number of Brix degrees and the direct polarization to be always made in connection, then any further investigation is to be dropped, as this would only lead to an in- crease of the quotient. SACCHARIMETRY 477 But if a quotient below 70 is found in this preliminary test then the exact determination of the amount of sugar is called for. In this it is not the saccharose alone which is to be calculated as sugar, as in factory work, but the invert sugar present, which is calculated to cane-sugar by the subtraction of l K , is to be added to the latter and the sum then taken as the basis of the calculation. In sirups the invert sugar is often inactive, but it may have the normal left rotation and, therefore, make the polarization of the cane-sugar pres- ent appear too low. For this reason it is not permissible in the exami- nation of sirups, to proceed as was suggested by Meissl for solid sugar- cane sugars, to multiply the invert sugar by 0.34, and to add the prod- uct obtained to the polarization. If one should proceed in this way, the sugar content of a sirup would, in many cases, be made to appear too high. But the possibility must always be kept in mind that, in conse- quence of the left-hand rotation of invert sugar, in presence of much of the latter the cane-sugar content will be found much too low. In con sideration of these conditions, it appears in general that the calculation of the total sugar from the polarization and the invert sugar found is allowable only in those cases w y here the amount of invert sugar does not exceed a certain limit. As an illustration, in presence of 6 per cent, of invert sugar the polarization of beet-sugar could be 6 X -34 = 2 -4 percent, too low. It is then advisable, in general, to abandon the optical method for sugar determination in sirups and to apply a gravimetric estimation for which a method that can be quickly carried out is given below under a. But an exception must be made when starch sugar is added to the sirup. As we are unable to determine accurately the amount of starch sugar present, and as, in addition, the reducing power of this sugar, which in the commercial product corresponds to a content of 40 to 60 per cent, of dextrose, remains practically constant under the conditions which are applied in the inversion of sirup for carrying out the gravimetric method, it follows that in cases where this sugar is added, the gravimetric method for determination of the total sugar content, or the quotient, can no longer be applied. It would lead, on the contrary, to gross errors, and sirups with a quotient above 70, with a certain amount of starch sugar added, would be made to appear, when tested in this way, as having a quotient below 70. With starch sugar present, the left-hand rotation of the in- vert sugar no longer affects the polarization as in the case of unmixed sirup, because the starch sugar has a much greater right-hand rotating power than the other kinds of sugars which may be there. To guard against mistakes which are easily possible with the mixing of starch sugar and sirups having a quotient over 70, the total svigar content, in all cases when starch sugar is added, must be calculated from the polari- zation and the invert sugar determined directly, as explained below, under b. Every sirup w r hich contains 2 per cent, or more of invert sugar must, therefore, be tested to find whether or not it contains starch sugar. 478 PRACTICAL APPLICATIONS OF OPTICAL ROTATION In sugar factories starch sirup is seldom added to cane-sugar sirups. As a rule, molasses, which are to be sent to distilleries or to factories for the extraction of sugar, do not contain starch sugar, because such sirups could be worked only with difficulty in these places. If the chemist making the tests has reason to believe from his knowledge of the origin or destination of the sugar sirup in question, and after proper considera- tion, can assume with sufficient certainty that it does not contain starch sugar, then he may omit the chemical tests which would be called for. But, in other cases, the chemical examination for starch sugar must be made in the following manner : The half normal weight is dissolved in the 100 cc. flask in 75 cc. of water, and inverted at 67 to 70 C. by addition of 5 cc. of hydrochloric acid of 1.19 specific gravity. Then the flask is filled to 100 cc., and the solution is decolorized by addition of ^ to i gram of blood- or bone-char- coal which has been washed with hydrochloric acid, or, with dark sirups, with even 2 or 3 giams, added directly to the flask in dry condition. If blood-charcoal is used, its absorption factor for invert sugar must be de- termined, as it is not the same for all kinds, and a corresponding cor- rection made on the polarimeter reading. Unadulterated sirups, as found by experience, show often less than the normal left-hand rotation, which at 20 is 0.327 of the original right-hand rotation, but the amount is always at least the fifth part of the original. Therefore, only such sirups shall be considered as mixed with starch sugar whose left-hand rotation after inversion is less than one-fifth of the right rotation before inversion. For example, a sirup of 55 polarization which, after inversion, shows a rotation of less than 11, or even a right-hand rotation, must be con- sidered as mixed with starch sugar. a. Sirups free from starch sugar In sirups free from starch sugar the determination of total sugar may be made in a single operation. The half-normal weight ( 13.024 grams) is taken and dissolved in a 100 cc. flask in 75 cc. of water, 5 cc. of hydrochloric acid of 1.19 specific gravity is added, and the whole is warmed to 67 or 70 in a water-bath. The flask is kept five minutes longer at this temperature of 67 to 70 and is frequently shaken. As the heating requires two and one-half to five minutes, the whole operation will consume seven and one-half to ten minutes ; in any event it should be completed in ten minutes. The flask is filled to the mark and 50 cc. of the 100 cc. is then diluted to i liter, and 25 cc. of this dilute solution (corresponding to 0.1628 gram of substance) is taken in an Erlenmeyer flask and neutralized by the addi- tion of 25 cc. of a sodium carbonate solution containing 1.7 grams of the anhydrous salt to the liter. Then 50 cc. of Fehling's solution is added and the solution is heated to the boiling-point in the same manner as in the invert sugar determination, and then kept three minutes in ebullition. The liquid should be heated as quickly as possible by means of a good triple burner, using wire gauze and sheet asbestos with a ring cut out of SACCHARIMETRY 479 it, and should require three and one-half to four minutes ; when the liquid begins to boil rapidly a single burner is exchanged for the triple burner. When the boiling is complete the liquid in the flask is diluted with an equal volume of air-free distilled water and the process is con- ducted in general as in the determination of invert sugar. The tables found in the literature cannot be used in the calculation of the result be- cause they do not obtain for invert sugar, but only for dextrose, or for mixtures of invert sugar and saccharose ; the cane-sugar content of the sirup corresponding to the copper obtained must be found by use of the following table only, which gives it directly in per cent. The calculation of invert sugar into cane-sugar is then avoided by use of the table. TABLE FOR THE CALCULATION OF CANE-SUGAR IN PER CENT., CORRE- SPONDING TO INVERT SUGAR PRESENT, FROM THE AMOUNT OF COPPER WEIGHED, AFTER THREE MINUTES' BOILING, WITH o. 1628 GRAM OF SUBSTANCE TAKEN Copper, ing. Cane- sugar. Per cent. Copper, mg. Cane- sugar. Per cent. Copper, mg. Cane- sugar. Per cent. Copper, mg. Cane- sugar. Per cent. 79 24-57 106 32.76 133 41.04 160 49.50 80 24.87 107 33-06 134 41-35 161 49.82 8l 25.17 108 33.36 135 41.66 162 50.13 82 25-47 109 33.67 136 41.98 163 50.45 83 25.78 no 33-97 137 42.29 164 50.76 8 4 26.08 III 34.27 138 42.60 I6 5 5T.08 85 26.38 112 34.58 139 42.91 166 51.40 86 26.68 113 34-88 140 43-22 167 51.72 87 26.98 114 35.19 141 43-53 168 52.04 88 27.29 US 35-49 142 43.85 169 52.35 89 27.59 116 35.8o 143 44.16 170 52.6/ 90 27.89 117 36.10 144 44.48 171 52.99 9i 28.19 118 36.41 145 44.70 172 53-31 92 28.50 119 36-71 I 4 6 45-10 173 53.63 93 28.80 120 37-01 147 45-42 174 53-95 94 29.10 121 37-32 148 45-73 175 54-27 95 29.40 122 37-63 149 46.05 176 54-59 96 29.71 I2 3 37-94 150 46.36 177 54-91 97 30.02 124 38-25 151 46.68 178 55-23 98 30.32 125 38.56 152 46.99 179 55-55 99 30.63 126 38.87 153 47-30 180 55-87 100 30.93 I2 7 39-iS 154 47.62 181 56.19 101 31.24 128 39-49 155 47-93 182 56-51 102 31-54 J2 9 39.80 156 48.25 183 56-83 103 31.85 130 40.11 157 48.56 184 57-15 104 32.15 131 40.42 158 48.88 185 57-47 105 32.45 132 40.73 159 49-19 186 57-79 480 PRACTICAL APPLICATIONS OF OPTICAL ROTATION Copper, mg. Cane- sugar. Per cent. Copper mg. Cane- sugar. Per cent. Copper, mg. Cane- sugar. Per cent. Copper, mg. Cane- sugar. Per cent. lS 7 58.11 207 64.58 227 71.19 247 77.85 188 58.43 ! 208 64.91 228 71.53 248 78.18 189 58.75 209 65-23 22 9 71.86 249 78.52 190 59.07 2TO 65.56. 230 72.19 250 78.85 191 59.39 211 65.89 2 3 I 72.52 251 79.19 192 59-72 212 66.22 232 72.85 252 79-53 193 60.04 213 66.55 233 73-18 253 79.88 194 60.36 214 66.88 234 73-51 254 80.22 195 60.69 215 67.21 235 73.85 255 80.56 196 61.01 216 67.55 236 74.18 256 80.90 197 6i.33 217 67.88 237 74-51 257 81.24 198 61.65 218 68. 2 T 238 74.84 2 5 8 81.59 199 61.98 2I 9 68.54 239 75- r 7 259 8L93 200 62.30 220 68.87 240 75-50 260 82.27 201 62.63 221 69.20 241 75.83 261 82.61 202 62.95 222 69-53 242 76.17 262 82.95 20 3 63.28 223 69.87 243 76.51 263 83.30 204 63.60 224 70.20 244 76.84 264 . 83-64 205 63.93 225 70.53 245 77.18 265 83.98 206 64.26 226 70.86 246 77-51 266 84.32 In the calculation of the quotient, fractions below whole tenths are neglected. Example: 25 cc. of the inverted sugar sirup 0.1628 gram of sub- stance, gave on reduction 171 mg. of copper ; this corresponds to 52.99 or rounded off 52.9 per cent, of sugar. Assuming that the sirup showed 75.6 Brix, its quotient is 69.97, or rounded, 69.9. b. Sirups containing starch sugar With sirups containing starch sugar in order to obtain the total sugar content, the plan must be adopted, as mentioned above, of adding to the polarization the invert sugar, which is to be calculated from the reducing action of the sirup on Fehling's solution. In the determination of the invert sugar in this case, a preliminary test must be made to learn how much substance may be weighed out, as the Fehling solution would not be sufficient for the 10 grams usually taken. This is most conveniently done by dissolving 10 grams of sirup to make 100 cc., and adding different amounts to several portions of Fehling's solution of 5 cc. each in a* many test-tubes, to one 8 cc., to another 6 cc., to another 4 cc., and to the last 2 cc. On boiling now the first test-tube which is not decolorized shows the amount to be taken. If, for example, reduction is not complete in the tube with 6 cc. of the solu- tion, then 6 grams is the amount of sirup to be weighed out for the SACCHARIMETRY 481 analysis. The right amount of substance is dissolved in 50 cc. of water, mixed with 50 cc. of Fehling's solution without previous clarification with lead acetate, boiled two minutes and then treated in the usual manner for the determination of invert sugar in solid sugar. The amount of invert sugar is calculated as follows : Let Pol, = the polarization of the substance, p the amount of substance taken for determination of invert sugar, which yields Cu grams of copper. The amount of invert sugar may be taken approximately as , and may be represented by A. We find then from the proportion, for R the amount of invert sugar which is present in 100 parts of cane- sugar invert sugar. The percentage amount of invert sugar in the substance is given by the formula - X F= per cent, of invert sugar, in which p is the amount of substance taken, and Fa. factor from the table below. In this table the columns and lines are used, the designations of which come the nearest to the values found for A and B ; at the intersecting point, the factor Fis given. TABLE OF FACTORS TO BE TAKEN FOR THE CALCULATION OF INVERT SUGAR IN PRESENCE OF CANE SUGAR Invert sugar in 100 parts of total sugar = B. Milligrams or invert sugar = A. 200 175 150 125 ioo 75 5 100 56.4 55.4 54-5 53-8 53-2 53-o 53- 9 56.3 55-3 54-4 53-8 53-2 52.9 52.9 80 56.2 55.2 54-3 53-7 53- 2 52.7 52.7 70 56.1 55-1 54-2 53-7 S3- 2 52-6 52.6 60 55-9 55-0 54.i 53- 6 53- 1 52.5 52.4 50 55-7 54-9 54-0 53-5 S3- 1 52.3 52.2 40 55-6 54-7 53.8 53-2 52.8 52.1 51-9 30 55-5 54-5 53-5 52.9 52.5 51-9 51-6 20 55-4 54-3 53-3 52.7 52-2 51-7 5 r -3 10 54-6 53-6 53-1 52.6 52.1 51-6 51-2 9 54-1 53-6 52.6 52.1 51-6 51-2 50.7 8 53-6 53-i 52.1 51.6 51-2 50.7 50.3 7 53-6 53-1 52.1 5L2 50.7 50.3 49-8 6 53-1 52.6 51-6 50.7 50-3 49-8 48.9 5 52-6 52.1 51-2 50.3 49-4 48.9 48.5 4 52.1 51-2 50.7 49.8 48.9 47-7 46.9 3 50.7 50.3 49-8 48.9 47-7 46.2 45-1 2 49-9 48.9 48.5 47-3 45-8 43-3 40.0 I 47-7 47.3 46.5 45-i 43-3 41.2 38.1 482 PRACTICAL APPLICATIONS OF OPTICAL ROTATION Example : Assume that the polarization of the sirup is 86.4, and tha for 3.256 grams of substance taken (/>), the amount of copper found (Cu) is 0.290, then : p X Pol \ / 3.256 ' 0.145 + ' L - / . p X Pol \ / (A + '- T5 ^-j : A = ( 2.958 : 0.145 = ioo : 4.9; therefore, B = 4.9. The nearest value in the table to A = 0.145 * s T 5 m g : the number 5 is the nearest to 4.9, the invert sugar in 100 parts of total sugar ; at the point of intersection of the line 5 with the column headed 150 mg. we find the factor 51.2. If this is substituted in the formula X F weob- P tain - X 5L2 = 4.56 per cent, of invert sugar. Then the invert sugar is calculated to cane-sugar by subtraction of Vao. an( l tlle result ob- tained (4-56 0.23^=4.33) added to that for the polarization. From the sum and the Brix degrees the quotient is found in the usual way. 2. Determination of the Quotient in Sirups to be Examined for Raffinose After the value in Brix degrees for the sirup in question has been found by the method of section I, the sugar content in the same is found from the direct polarization (/>), and the polarization at 20, or at a tempera- ture very close to this and properly corrected, after inversion (/), by aid of the following formula : 0.5124 PI 5(Sugar) = 0.839 If the amount of raffinose is to be found in addition, this formula is used : R( Raffinose) =^-g^. The inversion is to be made in the manner described in section I, under a. Example : For a sirup showing 85.6 Brix, 76.6^ direct polarization and 3.0 polarization after inversion (for the whole normal weight), the amount of sugar is found as 0.5124 x 76.6 4- 3 0.839 50.4 per cent., and the quotient is 58.8. II. Determination of the Amount of Saccharose in Crystal Sugar Sup- posed to Contain Raffinose The determination of the saccharose content of crystal sugar contain- ing raffinose is made as for sirups containing raffinose, according to the directions in I, 2. Only such sugars shall be considered as containing raffinose in which the difference between the saccharose content by direct polarization and SACCHARIMETRY 483 that found by application of the raffinose formula is more than i per cent, for sugars of class a, or more than 0.6 per cent, for sugars of classes b and c, because smaller differences may be found in raffinose free sugars at times, and possibly may be results of errors of observation. With differences of i per cent, or 0.6 per cent, or less, in the two classes, the result of the direct polarization is to be taken then as show- ing the real saccharose content of the sugar tested. If the polarization is below 90, a further test is unnecessary. In the statement of the final result, fractions below whole tenths are to be dropped. For example, a sugar content of 97.19 is to be rounded off to 97. i. Final Provision A written certificate must be made out for each investigation and filed with the office sending the sample in question. Besides an accurate description of the sample this certificate must contain : I. In determining the quotient of sirups : 1. In the cases described under a at the beginning : the specific gravity, the Brix degrees calculated from this, the direct polarization, and the quotient calculated. 2. In the cases given under b : the result of the test for invert sugar, the specific gravity, the Brix degrees calculated from this, the direct polarization ; further, in case a quotient below 70 is found from these data, either a statement of why a test has not been made for starch sugar, or the result of such a test with figures for the polarization found after inversion ; further, with reference to sirups free from starch sugar the amount of copper and the calculated sugar con- tent, and for sirups containing starch sugar the amount of copper found, the invert sugar content corresponding to this, the total sugar content (polarization invert sugar), and finally the calculated quotient. 3. In the cases falling under c above : the result of the test for invert sugar, as far as this is necessary, and then, in case the application of the raffinose formula is per- missible, the specific gravity, the Brix degrees calculated from this, the direct polarization, the polarization after inversion, the sugar content calculated from these data by aid of the raffinose formula, and the quotient ; otherwise, the data given under 2 above. II. In determining the saccharose content of crystal sugar supposed to contain raffinose : in case the polarization falls below 90, this only, but otherwise, in addition the polarization after inversion, the sugar content calculated by the raffinose formula, and then the resultant saccharose in per cent, as required by the regulations. 484 PRACTICAL APPLICATIONS OF OPTICAL ROTATIQ Appendix E DIRECTIONS ~ for finding the amount of sugar in saccharine products According to $3 of the rules for carrying out the provisions of $6 of the sugar tax law, a rebate of the sugar tax for saccharine manufactured products, except in the case of caramels containing starch sugar, can be allowed only when they are made without the use of honey or starch sugar. While the fact of not using honey may be established by the factory control and the factory production books, the absence of starch sugar is to be determined by chemical tests of the products of the factory. These investigations are to be made according to the directions in section i of Appendix C of the Rules of Procedure, but with this provision, that in saccharine factory products the presence of starch sugar is to be as- sumed when the left-hand rotation after inversion of the solution is 28 or less, for every 100 parts found in the direct polarization. The saccharose content of starch sugar-free saccharine factory prod- ucts is to be established by different means, according as they contain less than 2 per cent., or 2 per cent, or more of invert sugar. In conse- quence, the test of the product for invert sugar is to be made according to section i, of Appendix B, but with the variation that the sugar solu- tion to be boiled with the Fehling solution shall correspond, not to 10 grams of substance, but to 10 per cent, polarization. Of saccharine, products which contain less than 2 per cent, of invert sugar, the saccharose content will be found according to the Clerget method, in which the inversion is to be made exactly as given in the directions of section i under a in Appendix C, and from the sum of the two polarizations (before and after inversion) the saccharose content is to be found by the formula -_ 100 S ~ 142.66 yt t ' in which S is the amount of sugar, s the sum of the two polarizations for the normal weight, and / the temperature at which the polarizations were made. The constant (C) 142.66 assumes the use of the half-nor- mal weight ( 13.024 grams) of sugar in the observation, and is to be re- placed by different numbers corresponding to the amount of substance taken for inversion. These numbers are given by the following table : For Krains sugar in 100 c c. For C to be taken. For grains sugar in 100 cc. For Cto be taken. I 141.85 II 142.52 2 141.91 12 142.59 3 141.98 13 142.66 4 142.05 14 M2.73 5 142.12 15 142.79 6 142.18 16 142.86 7 142.25 17 142.93 8 142.32 18 M3-00 9 142.39 19 143.07 10 142.46 20 143.13 SACCHARIMETRY 485 If there is found, for example, a direct polarization of -j- 30 in a 200 mm. tube for a solution of the normal weight dissolved to 200 cc. , the calculated direct rotation of the inverted solution containing 75 cc. of the original must be 22.5. As TOO polarization corresponds to 26.048 grams of substance, 5.86 grams, or rounded off, 6 grams of substance would cor- respond to the -f> 22.5 ; according to the table, then, the constant 142.18 is to be applied. Assuming then, that a left-hand rotation of 7.1 is observed at 20, this corresponds Tor the half normal weight to 7.1 X IPO 75 As the direct polarization for the whole normal weight is content is calculated as 100 9.47, and for the whole normal weight to 18.94. 60, the sugar : =59.72 or, rounded, 59.7 per cent., lower fractions than whole tenths being disregarded. The sugar content of such products as contain 2 per cent, or more of invert sugar is to be determined by the copper method given in section i of Appendix C of the Rules of Procedure. A portion of the sugar solu- tion is inverted as there explained and the amount of substance to be em- ployed determined as in the case of finding the invert sugar in products containing starch sugar, and then the properly made solution is boiled three minutes with Fehling's solution. The amount of saccharose cor- responding to the copper found is given in the following table : TABLE FOR THE CALCULATION OF CANE SUGAR CORRESPONDING TO INVERT SUGAR FROM AMOUNT OF REDUCED COPPER AFTER THREE MINUTES' BOILING Copper, mg. Cane- sugar, mg. Copper, mg.' Cane- sugar, mg. Copper, mg. Cane- sugar, mg. Copper, mg. Cane- sugar, mg. 79 40.0 96 48.3 H3 56.8 130 65.3 80 40.5 97 48.8 114 57-3 131 65-8 8l 41.0 98 49-3 H5 57.8 I 3 2 66.3 82 41-5 99 49.8 1 16 58.3 133 66.8 83 42.0 100 50.3 117 58.8 134 67-3 84 42-5 101 50.8 118 59-3 135 67.8 85 42.9 102 51-3 119 59-8 136 68.3 86 434 103 51.8 120 60.2 137 68.8 87 43-9 104 52.3 121 60.7 138 69.4 88 44-4 105 52.8 122 61.2 139 69.9 89 44-9 106 53-3 I2 3 61.7 140 70.4 90 45-4 107 53-8 124 62.2 141 70.9 9i 45-9 108 54.3 125 62.8 142 71.4 92 46.4 109 54-8 126 63-3 143 71.9 93 46.8 no 55-3 I2 7 63.8 144 72.4 94 47-3 III 55-8 128 64.3 145 72.9 95 47-8 112 56.3 I2 9 64.8 146 73-4 486 PRACTICAL APPLICATIONS OF OPTICAL ROTATION Copper, rag. Cane- sugar, mg. Copper, mg. Cane- sugar, mg. Copper, mg. Cane- sugar, mg. Copper, mg. Cane- sugar, mg. 147 73-9 I 7 6 88.9 20 5 104.1 234 II9.7 I 4 8 74-5 177 89.4 206 104.6 235 120.3 149 75-0 I 7 8 89.9 207 105.2 236 120.8 150 75-5 179 90.4 208 105-7 237 121.3 151 76.0 180 91.0 209 106.2 238 I2I.8 152 76.5 181 91-5 210 106.7 1 239 122.4 153 77.0 182 ^2.0 211 107.3 240 122.9 154 77-5 183 92.5 212 107.8 241 123.5 155 78.0 184 93-1 2I 3 108.4 242 124.0 156 78.5 185 93-6 214 108.9 243 124.6 157 79.0 [86 94-1 215 109.4 244 125.1 158 79.6 187 94.6 216 109.9 245 125.7 159 80. i 188 95-1 217 110.5 246 126.2 160 80.6 189 95-7 218 III. I 247 126.8 161 81.1 190 96.2 219 in. 6 248 127.3 162 81.6 191 96.7 220 112. 2 249 127.9 163 82.1 192 97-2 221 112.7 2 5 128.4 164 82.6 193 97-7 222 113.2 251 128.9 165 83.2 194 98.3 22 3 113-7 2 5 2 129.4 166 83.7 195 98.8 224 114.3 253 130.0 167 84.2 196 99-3 225 114.8 254 130.6 168 84.7 197 99-8 226 115-4 255 131.1 169 85-2 198 100.4 227 115.9 256 131.7 170 85.7 199 100.9 228 116.4 257 132.2 171 86.3 200 101.4 22 9 117.0 258 132.8 172 86.8 201 101.9 230 j II7-5 259 133-3 173 87.3 202 102.5 231 118.1 260 133.9 174 87.8 203 103.1 232 ; 118.6 175 88.3 204 103.6 233 ! "9-2 The percentage amount of saccharose is calculated from this, and then the total sugar content is expressed as saccharose and given in terms of per cent, of the substance. With reference to the preparation of solutions of the substance it may be remarked that, as in the case of digestion methods of beet testing, it is in general not allowable to fill up a flask with the solid substance (chocolates, etc.) and water to the mark, because the error caused by the insoluble parts of the solid would be too great. As a rule, therefore, the solution is to be made up to a definite volume only after filtration and washing out of the residue. With reference to the investigation of saccharine products, on which rebate is allowable, the following details may be pointed out : SACCHARIMETRY 487 A. Chocolates It is convenient to moisten the normal weight with alcohol to facilitate the subsequent wetting with water, and then to add about 30 cc. of water and warm ten to fifteen minutes on the water-bath. The liquid is next filtered hot, and may run through turbid without harm ; the residue is washed with hot water. After treatment with about 10 cc. of basic lead acetate the filtrate is allowed to stand a quarter of an hour, then clarified with alum and a few drops of alumina cream, and finally made up to a proper volume, about 200 cc. B. Confectioner's Wares a. Caramels (bonbons, boltjes} with exception of gum drops, which are not rebatable With regard to such caramels as are declared by the manufacturer to contain starch sugar, it must be determined by tests that they show at least 80 of -f rotation and 50 per cent, of saccharose by the Clerget pro- cess. Otherwise they must be considered as not entitled to rebate. Caramels which are declared as free from starch sugar, must be tested foiS this. If no starch sugar is found the further investigation is made as with white sugar candies. b. Dragees (sugar-coated seeds and nuts with addition of flour} Dragees are extracted as are chocolates. They nearly always contain invert sugar. c. White sugar candies (sugar with addition of ethereal oils or coloring- matter} The solid residue may be neglected. The normal weight is, therefore, filled directly into a 100 cc. flask, water added to the mark after solution, and the filtration performed afterwards. d. Porous products (mixtures of sugar with some binding substance as white of egg, with addition of a flavor or remedial agent} The usually very small amount of binding material (white of egg, gela- tin, gumarabic, tragacanth or glue) is to be removed by basic lead acetate or alumina. The santonin lozenges, which are classed among the porous products, contain sodium santoninate. The addition of basic lead acetate is neces- sary to remove the santoninic acid. e. Dessert bonbons (creams, etc,, made of sugar and enclosed fruits or marmalade, etc. } The sample is dissolved in water. If but little residue remains it may be made up to the mark directly ; otherwise it is necessary to filter first. /. Marchpane mass and marchpane cakes (sugar with crushed almonds} The material is conveniently rubbed up with cold water in a porcelain dish and clarified before filtration with much alumina cream. As a rule marchpane is free from invert sugar. 488 PRACTICAL APPLICATIONS OF OPTICAL ROTATION g. Cakes and similar bakers' 1 wares The sugar is extracted with alcohol of 85 to 90 per cent. After evap- oration of the alcohol the filtrate is tested. h. Sugar-coated tropical or native fruits \ glace or candied ; fruits pre- served in sugar solutions (marmalades, pastes, compotes, jellies} If the material is solid, special pains must be taken in the preparation of an averge sample of homogeneous composition ; as for example, by warming and stirring. The sugar is extracted as for g above. As a rule, invert sugar is present. C. Alcoholic Liquors Containing Sugar The alcohol does not interfere with the direct polarization ; but it must be evaporated before the inversion polarization. D. Liquid Refined Sugar Liquid refined sugar contains invert sugar as a rule. The test may be limited to determining that there is a total sugar content of at least 75 per cent. Final Provision A written certificate fur each investigation must be handed to the office which submitted the sample, and this must contain, besides an exact description of the sample, data on the methods and results of the tests carried out, and the percentage amount of sugar calculated from them. II. Determination of Milk- Sugar 178. The specific rotation of crystallized milk-sugar, C 12 H W O U -f- H 2 O, was found by Schmoeger 1 for solutions con- taining from o to 36 per cent, to be, at 20, [or]}? = 52.53 constant. Exactly the same value was fotmd by Parcus and Tollens* at 20 for solutions having a concentration of 4.8 to 7.1 grams in 100 cc. Schmoeger found also that in the neighborhood of 20 the above value of the specific rotation is decreased 0.075 for each degree of increase in temperature. As already shown in 72, crystallized milk-sugar exhibits birotation immediately after solution which, however, may be rapidly changed to the constant rotation by heating to 100. On the other hand, the sugar dehydrated at 100 exhibits, after solution in cold water, at the outset a lower rotation than the normal. This is easily changed also to the normal rotation by heating. 1 Her. d. chem. Ges., 13, 1922 (1880). 5 Parcus and Tollens : Ann. Chem. (l,iebig), 357, 160 (1890). DETERMINATION OF MILK-SUGAR 489 For the determination of milk-sugar with instruments gradu- ated in angular degrees we have therefore : a *-= * -9037 -y, and by use of a 200 mm. tube, with sodium light, at 20, c = 0.9518 a, or with the use of a tube 190.37 mm. in length, c = of. If the problem is to test a substance as to its content of milk- sugar in an instrument with the Ventzke scale, it is best to dis- solve that weight in a 100 cc. flask, which, if it were pure milk-sugar, would polarize 100. This weight is found from the proportion : x : 26.048 : : 66.50 : 52.72* ^ = 32.856. By taking 32.856 grams of substance, each degree would then correspond to i per cent, of milk-sugar. If a solution is to be examined and it is required to find the concentration of the milk-sugar in it, it is to be observed that a polarization of i V. corresponds to a concentration of 0.32856 gram of milk- sugar in loo Mohr units of volume at 17.5. Using a 200 mm. tube, c = 0.32856 Pol. Exactly the same value is reached from the basis of the observation 2 that one Ventzke degree for milk-sugar is equal to 0.3452 circular degrees, with sodium light. As c = 0.9518 <*, we have also c o-QS 1 ^ X 0.3452 Pol, or c = 0.32856 Pol. 179. The Determination of Sugar in Milk is carried out, accord- ing to Schmoeger, 3 in this way : 1 This is the specific rotation of milk-sugar at 17.5. - I^andolt : " Ueber polarimetrisch-cheraische Analyse," Ber. d. chem. Ges., ai, 191 (1888). : Ber. d. milchwirthschaftl. Instituts zu Proskau, 1883-4. 490 PRACTICAL APPLICATIONS OF OPTICAL ROTATION 1. According to Hoppe-Seyler, 50 cc. of milk is boiled with 25 cc. of a 20 to 25 per cent, lead acetate solution, to the still warm liquid, 5 cc. of a 10 per cent, alum solution is added, and then the mixture is cooled, filled up to 100 cc. and filtered. The volume of the precipitate is in the mean 3 cc. , and is to be taken into consideration. 2. 100 cc. of milk is coagulated by addition of 6 cc. of 10 to 15 per cent, acetic acid and after standing half an hour is filtered. A slight turbidity from fat globules does no harm. 50 cc. of the filtrate is heated to boiling with 3 to 4 cc. of basic lead acetate solution (sp. gr. 1.2), and, after cooling, water is added to make up for loss on evaporation. The liquid is then filtered. 3. 100 cc. of milk is coagulated as before or by addition of 6 cc. of 10 to 15 per cent, sulphuric acid, but instead of sepa- rating the proteids by basic lead acetate, 50 cc. of the filtrate is treated in the cold with 5 cc. of commercial phosphotungstic acid, then filtered and polarized. The result must be multi- plied by i.i. As in methods 2 and 3, the volume of the precipitate is taken as 6 cc., in the mean ; it is advisable to add just 6 cc. of acid for coagulation, because then the concentration of the milk-sugar in the filtrate (whey) will be exactly the same as in the original milk. Besides, in methods 2 and 3, the dis- advantageous dilution of the liquid containing the milk-sugar to the double volume is avoided. A comparison of the three processes led to the conclusion that method 3 gives about 0.15 per cent, higher values than method 2, and that this in turn gives again 0.15 per cent, higher results than method i . Schmoeger traces {hese differ- ences to this, that in mixing the milk with the lead solution, milk-sugar, or possibly some other right-rotating substance not clearly known, is thrown out of solution. This is certainly the case when, after using an exce'ss of lead solution, the fil- trate is alkaline. Schmoeger is, therefore, of the opinion that the third method gives correct results, possibly a few hun- dredths too high. But, on the other hand, the results obtained by method i agree very closely with the gravimetric analyses according to Tollens, and those by method 2 with the gravimetric analyses DETERMINATION OF DEXTROSE 49* according to Soxhlet with fair accuracy. It is not possible, therefore, to give a final decision as to which is the more accu- rate procedure. An advantage in the polarimetric method over the gravimetric is found in the greater rapidity and con- venience with which it may be carried out. III. Determination of Glucose (Dextrose, Grape-Sugar) (Crystal- lized C 6 H 12 6 + H 2 0) 180. Tollens 1 has given this formula showing the dependence of the specific rotation of dextrose anhydride on the per- centage strength of the solution : \_a] = 52.50 -f 0.0188 p -f 0.000517 /'-'. From this we have for : P = 5 10 15 20 25 30 [or] =52.61 52.74 52.90 53.08 53-29 53-53 /== 35 40 45 50 55 60 [] =53-79 54.o8 54-39 54-73 55-10 55-49 The specific rotation increases then appreciably with the concentration. But for solutions up to 15 per cent, strength, without very great error, [] can be taken as equal to 52.80. If we substitute this value in the equation [or] = , this for- mula follows for calculating the percentage strength from the observed angle a, /= 1-894 7^, and using a 2 dm. tube, (0 ^ = 0.947-^-- If not the percentage strength but the concentration of the solu- tion is to be found, that is, the number of grams of sugar in 100 cc. of solution, then this formula may be transformed into the simpler one, (2) c = 0.947 a - The error made by neglecting the variation of the specific rotation with the strength of solution reaches then in the most unfavorable case 0.03 per cent., assuming that the observed angle of rotation a is measured at 20 with use of sodium light. 1 Tollens : Ber. d. chem. Ges., 17, 2238 (1884). 492 PRACTICAL APPLICATIONS OF OPTICAL ROTATION For greater concentrations (15 to 50 per cent.) Landolt 1 has calculated the following formula, for finding the percentage strength, from the observations of Tollens : (3) p = 0.948 a 0.0032 of' 2 , where a is the rotation for a 2 dm. tube. Dextrose may be determined by use of the Ventzke sac- charimeter also, since, according to the observations of Hoppe- Seyler,* its rotation dispersion is very nearly the same as that of quartz. It has been found by direct experiments 3 that for dextrose, i V. = 0.3448 0.0008 circular degrees (Na light). If we represent by Pol. the number of Ventzke degrees read off for a 2 dm. tube at 20, the formulas just developed may be transformed into these : (I) p = 0.3265 -~ O = o to 15] (II) c= 0.3265 Pol [r=otoi6] III) p = 0.3269 Pol o.ooo 381 Pol 2 lp = o to 50] Thus, for example, a grape-sugar solution which shows a rotation of 34.48 in an instrument with circular degrees must polarize exactly 100 in the Ventzke saccharimeter. We have by Formula (3), p = 0.948 X 34.48 0.0032 X 34.48' = 28.88 per cent., and likewise by Formula (III), p = 0.3269 X loo 0.000381 X loo 2 = 28.88 per cent. But, nevertheless, one cannot expect as accurate results from the saccharimeter as from the instrument with circular degrees, because the value of a circular degree in saccharimeter degrees might not be the same for all instruments and all solu- tions. For this reason, and because the specific rotation of dextrose is dependent on the temperature and concentration, a normal weight cannot be definitely fixed, which may be dissolved to make 100 Mohr cc. and give directly, in the saccharimeter, the percentage strength of dextrose in the dissolved substance. But, however, for most practical needs, sufficiently accurate values may be derived from the following considerations : 1 I^andolt : Her. d. chem. Ges., ai, 199 (1888). - See I,andolt : Ibid., ai, 194 (1888). /tschr. anal. Chem., 5, 413 (1866). DETERMINATION OF DEXTROSE 493 For dilute solutions we have above, c = 0.3265 X/W, Accordingly, a liquid which contains 32.65 grams of dextrose in 100 cc. must be able to polarize 100 V. But as the Mohr unit of volume is related to the cubic centimeter as 1.00234 : i, then to secure the same concentration in a Mohr 100 cc. flask there must be weighed out, in vacuo, 32.65 X 1.00234 = 32.73 grams of the substance containing dextrose or 32.71 grams in air with brass weights. Also, we obtain the normal weight of dextrose, -/V, when we multiply the normal weight of saccharose, 26.048, by the relation of their specific rotations. We obtain in this way : For 5 per cent, solutions : ^V= 26.048 X - 32.91 grams. For 15 per cent, solutions : N = 26.048 X 3 2 -75 grams. For 25 per cent, solutions : N = 26.048 X = 32.50 grams. The normal weight varies, therefore, for solutions practically the most used, containing o to 25 per cent, of dextrose, be- tween 32.9 and 32.5 grams. In weighing out a dextrose sub- stance then, one must take into consideration whether it con- tains much or little of the sugar. The normal weight derived above for dilute solutions, 32.71 grams, would hold, accord- ing to the last calculation, for solutions containing from 17 to 1 8 per cent, of dextrose. If the amount of dextrose hydrate, instead of that of the anhydride, is desired the result must be multiplied by the re- lation of the two molecular weights = i . i . 1 80 Finally, it must be remembered that solid dextrose dissolved in water exhibits birotation, which is destroyed by allowing the solution to stand twenty-four hours, or by warming. 181. The Determination of Dextrose in Diabetic Urine may be ad- vantageously made when the amount present is more than about 0.2 gram in 100 cc. With smaller amounts, or where the greatest accuracy is desired, as in normal urines or in phys- 494 PRACTICAL APPLICATION OF OPTICAL ROTATION iological investigations, the chemical methods of determination yield more reliable results. It must first be seen whether or not the color of the urine will permit a direct polarization, using if necessary, a tube only 100 mm. long, or after diluting to the double volume. If the urine is not perfectly clear it should be filtered as quickly as possible through soft filter- paper. If the urine is too dark, icocc. should be precipitated by 10 cc. of basic lead acetate solution, and the filtrate tested, or it may be shaken in a flask with some blood-charcoal and then filtered. In the first case, the result of the polarization must be multiplied by i.i on account of the dilution. But in both cases, a part of the grape-sugar may be removed from the urine ; at least this has been shown after application of basic lead acetate, and it may be assumed for the charcoal from the experience gathered in the clarification of dark sugar sirups. These errors may be eliminated by making a parallel experi- ment with like quantities of clarification agents and normal urines whose sugar content is brought to that of the urine under investigation, and which is polarized before and after application of the clearing agent. If the diabetic urine contains albuminous substances they may, on account of their left rotation, make the sugar content appear much too low, and it is, therefore, necessary to remove them. loo cc. of urine is heated in a dish to boiling and then enough dilute acetic acid is added to give an acid reaction and throw down the albumin as a flocculent precipitate. Then the liquid is filtered, the filter washed, and the filtrate made up to 100 cc. Or a measured volume of urine is acidified with acetic acid and then enough concentrated sodium sulphate solution added to bring the volume to double the original. If the liquid is now heated, the albumin separates completely and may be filtered off. Bile acids, which have a right-hand rotation, are not present in urine in amount sufficient to cause an error in the above process. The fact that the albumin in urine rotates the plane of polari- zation to the left very nearly as much as grape-sugar does to the right, furnishes us with a very convenient means of deter- mining the amount of albumin in urine polarimetrically. If DETERMINATION OF MALTOSE 495 the polarization of the urine be observed at the same tempera- ture (17.5) before and after precipitation of the albumin, and at the same degree of concentration, we obtain from the dif- ference, D, of the two readings, with a 2 dm. tube, the amount of albumin, Al, in the liquid, from the formula de- rived above for the determination of the grape-sugar content of dilute solutions : Al = 0.947 D. The firm of Schmidt and Haensch makes a half-shadow in- strument with circular degrees on the Laurent system ( 1 1 1 to 113) for urine analysis. A sodium flame serves for illumina- tion. In order to avoid the necessity of calculating the sugar or albumin content from the rotation read off in circular degrees, according to the above equation, tubes of 188.6 and 94.3 mm. in length (better 189.4 and 94.7) are furnished with the instru- ment, the shorter one for dark liquids, which lengths are so chosen that i or 2 of polarization corresponds exactly to i gram of grape-sugar in 100 cc. of the liquid analyzed. The same firm makes also half-shadow instruments with wedge-compensation for urine analysis, permitting the use of white light (i33~i34), the scale of which is so arranged that by employing a 2 dm. tube, the amount of grape-sugar in 100 cc. may be read off directly. The vernier reads to l l w per cent. IV. Determination of Maltose (Crystallized C 12 H 22 O n -f H 2 O. Right-rotating) 182. Meissl 1 gives the following formula for the dependence of the specific rotation on the percentage strength and the tem- perature /. [a\ l D == 140.37 0.0184 p 0.095 *, which holds good for/ = 5 to 35 and t 15 to 35. We have from this, when t= 20, for p= 5 10 15 20 25 30 35 [a] * = 138. 38 138.29 138.20 138.11 138.02 137.92 137.82 Parcus and Tollens 2 found a somewhat lower value for the specific rotation from a concentration of 10 grams in 100 cc. at 20, MS = I36.85 to 136.96, 1 J. prakt. Chem., [2] 25, 114 (1882). - Parcus and Tollen : Ann. Chem. (I^iebig), 357, 160 (1890). 496 PRACTICAL APPLICATIONS OF OPTICAL ROTATION while for this concentration the specific rotation according to Meissl is 138.3. For the practical determination of maltose by the optical method a mean value of []*= 137-5 may be considered as sufficiently exact. Then at 20, 100 Cf I37 ' 5 ~ = 7x7' a c= 0.7273, and by the use of a 2 dm. tube at 20, c = 0.3636 tf. Meissl gives this formula for the temperature of 17.5 : c = 0.362 a. As freshly prepared solutions show a rotation which is too low, they must be warmed before polarization or allowed to stand some hours. V. Determination of Galactose (C 6 H 12 O 6 . Right-rotating.) 183. According to Meissl 1 the change in the specific rotation with the percentage strength and the temperature is given by the formula, M/>= 83-88 + 0.0785 p 0.209 /, in which p 5 to 35 per cent., / = 10 to 30. According to Rindell, 2 [or]',, = 83.04 -f- o.iggp (0.276 0.0025 /)/, for p = 12 to 20 per cent., t = 4 to 40. If we take / = 20, we have then for P- 5 10 15 20 25 30 35 Meissl 80.10 80.49 So- 88 81.27 81.66 82.06 82.45 Rindell 80.01 81.25 82 -5Q , ~ . f 80 7 81 7 S Kent and Tolled { g j Parcus and Tollens 4 80.33 1 J. prakt. Chem., [2], aa, 97 (1880). 8 Rindell : Ztschr. Riibenzucker-Ind., 30, 163 (1880). */**., 35, 36 (1885). 4 Parcus and Tollens: Ann. Chem. (I^iebig), 357, 160 (1890). DETERMINATION OF CAMPHOR 497 For solutions with o to 1 5 per cent. , and possibly even to 20 percent., we may take, therefore, the Meissl value of 80.88 forp= 15 as the mean specific rotation of galactose at 20. From this there follows : c = 1.236-^-,. and for a 2 dm. tube at 20, c = 0.618 of. Freshly dissolved galactose also exhibits birotation, which at the ordinary temperature reverts to the normal rotation after a lapse of six hours. VI. Determination of Camphor, C io H j6 184. The easy determination of camphor in the optical way has become of greater importance since the introduction of articles made of celluloid, a mixture of nitrocellulose and cam- phor. According to Foerster 1 the determination of camphor in celluloid is made best as follows : About 10 grams of celluloid, containing 2 to 3 grams of camphor, is saponified with four times its weight of 10 per cent, sodium hydroxide solution until all has dissolved, and the mixture is then diluted to 250 cc. Of this, 120 to 150 cc. is distilled off, the camphor, in vapor, passing over completely with the steam. To the distillate, collected in a graduated receiver, 25 to 30 cc. of benzene is added and the mixture well shaken ; the volume of the benzene, which dissolves the cam- phor, is read off and a part is then taken for polarization at 20. The author carried out special tests to determine the rota- tion of camphor in benzene, and as a standard he used pure camphor with melting-point at 178.7, which had been re- crystallized several times from 50 per cent, alcohol. Solutions up to the concentration of 40 grams of camphor in 100 cc. (d M / 4 ) were tested at 20 and with sodium light. The specific rotation ar- dependent on the concentration may be expressed by this formula : (I) M" = 39-755 + 0.1725'. From this the concentration c may be calculated as a function of the angle of rotation a \ 1 Ber. d. chem. Ges., 23, 2981 (1890). 32 498 PRACTICAL APPLICATIONS OF OPTICAL ROTATION (II) - 1 15.205 - I + ^/ ! _|_ 0.04367 " From the observations themselves, the following direct re- lation between concentration and rotation may be derived : (III) c == 2.4683 - 0.01747 -"- This formula was calculated from the earlier determinations of Landolt, 1 showing the specific rotation of camphor in ben- zene, []"= 39- 19 + 0. 17084 r, and from experiments by Rimbach, 2 the following for concen- trations between 10 grams and 53 grams : [a]}? = 40.21 + o. 1309*: -f- 0.000269 r'. These two formulas agree closely with that of Foerster. Test experiments showed that from 99 to 99.3 per cent, of the camphor taken could be found, which, considering the method of separation used, is a satisfactory result. In fats and oils also, camphor may be determined by the optical process. According to Foerster it is best, in such cases, to first distil the camphor from the substance under investi- gation by aid of a current of steam. When about 250 cc. of distillate has been collected in the receiver, this is then used as the distillation flask and the rest of the operation is carried out as with celluloid. If, in place of benzene, alcohol is chosen as the solvent for the camphor, the concentration of the solution may be found by the following formulas, according to Landolt : :{ c = 2.3614 -- 0.01158 or - 177.53 +^/ 31516.45 + 845.74^- These obtain for concentrations between o and 50 grams in loo cc., and for a temperature of 20. VII. Determination of Cinchona Alkaloids 185. The specific rotation of the cinchona alkaloids and their 1 Iandolt : Ann. Chem. (Uebig), i89, 334 (1877). - Kimhach : Ztschr. phys. Chem., 9, 698 (1892). * I^andolt : Her. d. chem. r.es., ai, 204 (1888). DETERMINATION OF CINCHONA ALKALOIDS 499 most important salts has frequently been the subject of ex- tended investigations ; numerous observations have been made especially by Hesse, 1 Oudemans, 2 and Lenz, 3 by which the con- stants of rotation for quinine, hydroquinine, cinchonine, quini- dine, and cinchonidine have been determined with such accu- racy that they may be used in testing other preparations as to their purity, or in finding the composition of mixtures. With all these alkaloids, the specific rotation varies in marked degree with the nature of the solvent, and moreover it is smaller, the greater the concentration and the higher the temperature. Hesse measured the rotation of solutions which contained from i to 10 grams of substance in 100 cc., accord- ing to the degree of solubility. As solvents, alcohol of 97 volume per cent, was used for the pure alkaloids, and either pure water or dilute hydrochloric or sulphuric acid of known strength for the salts. L,enz employed as a solvent a mixture of 2 volumes of chloroform and i volume of 97 per cent, alcohol, and determined the specific rotation in solutions of i to 3 per cent, strength. Notwithstanding these fundamental investigations, no method is yet known by which the alkaloids in extracts of cinchona bark or in the quinine of commerce may be found by the optical process. This is due partly to the fact, already re- ferred to, that the specific rotation is in a large measure de- pendent on the external conditions under which the solutions in question must be tested, and partly to this, that the optical analysis of a mixture of several active substances cannot, in general, be made with accuracy, and even when only two or three are in solution, while in any case the qualitative compo- sition of the mixture must be known, which can be determined only by the methods of chemical analysis. Finally, this diffi- culty is met with in the optical determination of the alkaloids in the extracts from cinchona bark, that these extracts contain a yellow coloring-matter which cannot be separated alone, and the presence of which makes the observation in the polarimeter uncertain. 1 Hesse : Ann. Chem. (I,iebig), 176, 203 ; 182, 128. 2 Oudemans : Ibid., 182, 33. 3 I,enz : Ztschr. anal. Chem., 37, 549 (1888). 500 PRACTICAL APPLICATIONS OF OPTICAL ROTATION In such investigations, therefore, it is customary to effect the extraction and separation of the alkaloids by chemical methods, and then to resort to the optical observations to con- trol the results of the chemical analysis or to test the separated alkaloids as to their purity. For the determination of the quantitative composition of a mixture of alkaloids, the specific rotation may be employed in all those cases where the analysis of a mixture of two known alkaloids only is involved. The following conditions then obtain : Of the mixture of two alkaloids, c grams is weighed off, dissolved to make 100 cc. , and then the angle of rotation, a y is found in a tube of / dm. length, from which may be cal- culated the specific rotation of the mixture, a = - . If the i /\ c mixture contains x per cent, of one alkaloid, whose specific rotation is [<*]> andjy 100 x per cent, of the other con- stituent with the specific rotation [tf] v , then *X OL+ (100 x) [L= 100 [] and consequently, [] - Mv X = 100 ~f-4 f-1 ,[]* I>L O] r - M y loo ~-\ =-* . ixu- 01 In this way, it is possible to analyze mixtures of any active substances, provided the specific rotations of the pure sub- stances are known and are not subject to too great variations within the limits of the concentrations employed. 1 But, in any event, it is advisable to take the concentration of the mixture only as great as appears necessary for the accurate calculation of the specific rotation. Hesse 2 has already employed this general method to deter- mine the amount of cinchonidine sulphate in the commercial quinine sulphate, nearly free from other alkaloids. He proceeded in this way, by taking first, of the sulphate in question, an amount corresponding to 2 grams of the anhydrous salt, dissolving in a 25 cc. flask in 10 cc. of normal hydro- 1 See Hesse : Ann. Chem. (Uebij?), i8, 146 and 152 ; Oudemans : Ibid., i8a, 63, 65. Hesse : Ibid., 205, 217 (1880). C^^fflC DETERMINATION OF COCAINE 5OI acid, and filling to the mark with water at 15 C. After complete solution and mixing, the liquid was filtered into a 220 mm. jacketed tube and polarized at 15 in a Wild polaristrobometer. If a represents the angle of rotation of the anhydrous quinine sulphate under these conditions (a = 40.309 was found), and ft the rotation of the anhydrous cin- chonidine sulphate (ft = -26.598 was observed), and y, finally, the angle of rotation of the mixture taken for the analysis, then the amount of cinchonidine sulphate, y, in the unit of weight of the mixture is given by i = a y^ -40.309 y a ft -13.711 while the amount of the quinine sulphate is ^Y <* _Y + 26.598 X u * a ft 13.711 VIII. Determination of Cocaine 186. The specific rotation of cocaine, C,.H 21 NO 4 , in chloro- form, and of the hydrochloride, C 17 H 21 NO 4 .HC1, in a mixture of 60 parts of absolute alcohol and 90 parts of water, has been determined by O. Antrick. 1 He found for a preparation of the base of the greatest possible purity : []" = - 15-827 0.00585 q, or, [or] = 16.412 + 0.00585 p. This gives then for p = 5 10 15 20 25 30 [a] = 16.38 -16.35 -16.32 16.29 --16.26 -16.24 For solutions containing up to 30 per cent, of cocaine, the value, []" = 16.32, may be taken as the basis of the cal- culation of the percentage strength. We have 100 Of and with use of a 2 dm. tube : c= - 3.o6nr. 1 O. Antrick : Ber. d. chem. Ges., 30, 310 (1887). 502 PRACTICAL APPLICATIONS OF OPTICAL ROTATION le^ras The specific rotation of the hydrochloride of cocaine been even more fully investigated by Antrick. The following formula expresses the results obtained from observations on four preparations which in their properties differed but little from each other : W"= -67.982 + 0.1583^; this holds for c = o to 25, and for true cubic centimeters (a^). It follows then for <* c= 5 10 15 20 25 [a]^ = 67.19 66.40 65.61 64.82 64.02 The change in the specific rotation with the concentration is here so considerable that it is not possible to make a mean value of this rotation the basis of a calculation of the concen- tration from the observed angle of rotation. In the formula loo a = M^/' we have to take for \oi\ that value from the above series which comes the nearest to the expected concentration, or we may make use of the following equation, derived directly from the formula for the specific rotation : (1) c= 214.72 ^46106.8 + 315.86 a which holds for the 2 dm. tube, t = 20, and c o to 25. Finally, the angles of rotation, a= -13.280 and a - 2 5-9 2 7 given by Antrick for the concentration c = 10 and c = 20, may be employed to express c as directly related to the observed angle a. The formula for this reads (2) - 0.7337 a + 0.001454 *, which holds for the 2 dm. tube within the given limits. In using either of these formulas, care must be taken to see that a is introduced with the proper sign, that is the negative sign. In using polarization tubes of other length than 2 dm. the value of the angle read off must be corrected before it is substituted in either of the two formulas, which are based on observations with 2 dm. tubes. The agreement in the results which may be obtained by the two formulas is shown in the following table : DETERMINATION OF NICOTINE 503 c c Concentration Concentration Difference 01. calculated by calculated by c \ f- Formula (i). Formula (2). 5 3.710 3.7 5 0.005 10 7485 7.482 4- 0.003 15 11-331 H.332 O.OOI 20 I5.252 15.256 0.004 25 19.250 19.251 O.OOI 30 23-333 23.320 0.013 IX. Determination of Nicotine, C IO H I4 N, 187. A new method for the quantitative determination of nicotine by the polariscope has been devised by M. Popo- vici. 1 The extraction of the nicotine from tobacco is best ac- complished by Kissling's process : 20 to 40 grams of homo- geneous dry tobacco powder is moistened with 10 cc. of a dilute alcoholic sodium hydroxide solution (6 grams of NaOH dissolved in 100 cc. of 57 per cent, alcohol) and extracted three to four hours with ether in the Soxhlet apparatus. The ether extract is treated with 10 cc. of a rather strong solution of phosphomolybdic acid in nitric acid and shaken, by which means the nicotine is thrown down with other bases (mainly ammonia) in the form of a quickly subsiding precipitate. Then the supernatant ether is poured off and enough water is added to the residue to make a total volume of 50 cc ; and finally 8 grams of finely powdered barium hydroxide is added. In this way the nicotine is obtained as a free base in alkaline solution, which, after some hours with frequent shaking, is poured off from the yellow precipitate, and polarized. The following table was obtained from experiments with known amounts of nicotine : Grams of nicotine in 50 cc. of solution. Rotation in 2 dm. i Minute of rotation tube. corresponds to Minutes. nicotine in grams. 2.00 337 0.00594 1-75 298 0.00588 1.50 258 0.00582 1-25 217 0.00576 1. 00 175 0.00572 0-75 0.50 "8 0.00564 0.00562 0.25 45 0.00556 Popovici : Ztschr. physiol. Chem., 13, 445 (1889). 504 PRACTICAL APPLICATIONS OF OPTICAL ROTATION If an alcoholic solution of nicotine, containing no other ac- tive substances, is to be examined the following formula by Landolt 1 may be employed to find the nicotine content : (1) p = 311.58 -^97082.5 - 449.64-^- , which holds for a temperature of 20, the density d~, and be- tween 10 and 90 per cent, of nicotine. Further, with reference to the nicotine concentration : a (2) c = 0.704 ~ 0.000525 which obtains for 20 and c 10 to 90. 1 I^an dolt: Her. d. chem. Ges., ai, 203 (1888). PART SIXTH Constants of Rotation of Active Bodies In the following tabulation the data on the specific rotation of all active bodies in any degree important have been included and the literature has been fully considered to the middle of 1896. Only a few of .the still later observations could be given a place. 1 Only such specific rotations have received considera- tion for which the data necessary for calculation (density, con- centration, temperature) were given in the original papers. In explanation of the signs employed, see Part First, i and 2 of this book. I. Hydrocarbons (See also Terpenes and Camphor.) Ethylamyl : b. p. 91, d'*> = 0.6895, [<*]/, = ~ 3-93 " /=i 7 , [ci] D = + 6.23)' = 60, = + 6.09 / Propylamyl: / = 16, [<*] D = -f 6.44 ) 4 = 54, = 4- 6.25 / Isobutylamyl: t 20, [a] D = 5.88 = 52, 5-66 -65,' = -5.20} Diamyl : b. p. i59-i62, d T - - 0.7463, [a] D = -f 8.69 T / = 21, [a\ D = ~ 12.08 ) 8 = 78, = 4- 12.06 J = 19, = 4- 10.01 These bodies were all made from active amyl iodide. 1 [In the translation the most important observations to the middle of 1900 have been included. Tr.J Just: Ann.Chem. (lyiebig), 220, 154. Welt : Compt. rend., 119, 743. Welt : Loc. cit. Welt : Loc. cit. Guye and Amaral : Arch. sc. phys. Geneve, [3], 33, 4<>9- Just : Loc. cit. Welt : Loc. at. Guye and Amaral : Loc. cit. 506 CONSTANTS OF ROTATION OF ACTIVE BODIES 2 . Alcohols with One Atom of Oxygen /-AMYL ALCOHOL C,H 5 .CH 3 .CH.CH 2 OH. Commercial fermentation alcohol : a D = - 2 to 4 for i dm. Preparations purified from inactive isomers by L,e Bel's method (repeated treatment with gaseous hydrochloric acid, through which the inactive part is first converted into amyl chloride and may be separated by distillation) gave : /> = 4-53 to 4-63 for i dm. [>]/>= - 5-7 when d = o.Si. 1 []/>= -5- 2, ft' 22 =0.8 1 8, boiling-point 128.5 to 129 (768 mm). 2 According to Schiitz and Marckwald were not yet pure. Derivatives of /-Amyl Alcohol these preparations Amyl chloride :b. p. 97 to 99 = + 1-24 ) = + 3-5 \ ~- + 5-41 > = 0.886, [^/,= Amyl bromide: " 117 " 120, " = 1.225, Amyl iodide : " 144 " 145, " = 1.54, Amyl iodide: " 144 " 146, vy 188 _L 574 Diamyl chlormaleate 176 68 I 74 4 60 Diamyl 0-phthalate- 1/0 23 5 Amvl acetic acid . . 232 2T2-2I7 43 O.QIA6 4- S.;7 I^c Bel : Bull. soc. chim., [2], ai, 542 ; Compt. rend., 77, 1021. Rogers : J. Chem. Soc., 63, 1131. Schiitz and Marckwald : Ber. d. chem. Oes.. 39, 59. I*e Bel : Bull. soc. chim., [2], 25, 545. Walden : Ztschr. phys.Chem., 15, 647. /bid., 15, 642 fT. DERIVATIVES OF /-AMYL ALCOHOL 507 Boiling-point. mm. d. []" 180-182 228-230 262-263 235-238 268-270 (m.p. 90-9i 244-246 280 228-232 265 (m.p. 51 185-186 7 60 775 760 760 775 c , acet( 760 768 aceto 0.8765 0.8631 0.8894 0.8701 0.8594 >ne c 6 67 ) + 6.56 4- 7-oi + 18.27 17 OQ i * / fy 4- 13-96 + 5-25 + 10.14 4- 5-82 + 11.48 inactive 4- 1.25 4- 7-94 0.9665 0-9445 0-9455 0.9120 tie c 20) Diethyl diamylmalonate Ethyl diamylacetoacetate Diethyl a m y 1-^-nitrobenzy] 0.8459 2. \Y.\LDEN. 1 Rotation of the amyl alcohol used, \_a] D = 4.7 Boiling-point- mm. d. []. \tnvl /-lactate 105 22 o 0672 + 2.64 1W O 17 o 0667 -I Q7 I 7O 171 A / 2O I 0520 -j- 2.76 ififi 167 QA O2 Amyl /-phenylchloracetate - . Amyl=: 4.8' Boiling-point. mm. d. Ml: Diamvl ina.lc3.te 170 2Q O Q747 -i- /i 62 i/u T(- 2C T OC^^ i 4 01 Diamyl brommaleate * 10 O 17^177 'D I ^ 1 -* J C>OO I I ^61 1 4- u o 4_ 4 eg ygc 187 A O T C i 168^ + r QQ 10 10 / 1 O 2C o 0661 o-yy 1 d TJ 1 /:7 1 H- 1 4 Diamyl antidimethylsuccin- a ^e i/j-i/^ 168-169 T r u vo/ 4-97 1 7 A2 Diamyl paradimethylsuccin- i8< 1 O in OQJ.^2 T^ 3-4 2 4_ , 66 Diamvl mesotartrate 10 o o w 17 I 06^8 1- /I 77 A / 16 I 064 1 4.// + ? 77 3-37 1 Ztschr. phys. Chem., 17, 705 ff. 2 y*/rf., 20, 378. 508 CONSTANTS OF ROTATION OF ACTIVE BODIES 4. WALDEN. 1 Rotation of the amyl alcohol not given Boiling-point. mm. rf. []??. T78-I 7Q 76* o 8690 _(_ 2 8l i/oi /y 7 CQ o 8018 I A -J* I 7O 171 /D u 761 ""Vo o 8619 \ 4-^4 + 7 TQ 7e 20 o 8781 I 1 ZI /O T7 C i Of, or O O1O2 T ,5-D 1 1 i 76 rfc O IO "yoy^ o 0606 T O'/ _L C Q7 XVJ O 187 188 22 I O1 1 A o-yo 1 1 7C 187 14 I 0560 1 O'/O _L c 78 Diamyl methylsuccinate 172 181 I&A 18 20 0.9529 o 9698 + 3.67 or o-9v) 241 2/11 O 26 9973 I OO2Q + 6 16 Amyl hydrocinnamate Amyl cinnamate Z 4 1 "43 . 172 IQ2 28 2O 0.9721 O QQQ2 + 2.26 I 7 rj iy^ 2IO *y \j. yyy- i /'O 1 ! r c8 Amyl a-naphthoate 222 2 Ac oo 25 IOO 100 l.UUj^ 1.0605 I Oil I 1 J'O + 5-28 + Q 1A J0 5 I ' tj o6 l V-o4 5. GUYE AND CHAVANNE.' 2 Rotation of the common amyl alcohol used, = 4.4. Rotation of the sec-amyl alcohol not given r/v -|20 22 \. a \D [^ + 2 OI -4- I 08 Atjiyi acetate 1 2 Cl i *^p> 4-2 cr i -'OO + 2 77 I *.5i 4- 2 68 * // 4- 2 60 + 2 r,i \.myl palmitate + T AZ. + T l6 1 '4o d []/. o 962 A 06 sec- Amyl propionate o 8oc 8 cc: Sfc- \tsi\\ butvrate u -95 n 8So -oo 8 2? fifi w-Amyl benzoate o 988 O.2 3 4- A 06 Afj#f-Amyl phenylacetate n 08.2 1 i 8.1 o 076 1 O'* + 2 11 w.y/u * *o 6. GOLDSCHMIDT AND FREUND. 3 Rotation of the amyl alcohol used, O]z> - 4.29. The following solid esters were used in chloroform solution p [>]/. Amyl phenylcarbaminate C 0/l8 Amyl 0-tolylcarbaminate 5728 4. iy t ^ fifi .320 5106 4- i 81 Amyl ^-tolylcarbaminate . . .yju 1.277 T 5'j 4- 4.d7 1 /tschr. phys.Chem.,ao,573. Compt. rnd., 120,452. 3 Ztschr. phys. Cheni., 14,394. MIXED AMYL ETHERS 7. GUYE AND GOUDET. 1 509 []/> A l6 4-o u 1 T C^ 6 10 i 48 O-H- 8. GUYE AND AM ARAL 2 investigated the following derivatives of /-amyl alcohol at different temperatures t [3z t [Oz> 16 15 20 18 18 18 - 4-52 -r 14-09 - 11.13 + 2.51 + 3^7 4- 2.54 76 72 5i 62 60 57 5i 62 57 ' 412 + H.I4 9-97 + 2.07 + 3-13 + 2.51 -f 2.97 - 0.99 - 4.71 27 18 18 3.00 - 0.87 -r 5-59 Amylamine hydrochloride. 3 Water, p = 7.84, = -f 12.7' Mixed Amyl Ethers GUYE AND CHAYANNE* give the following data : Boiling-point. rf WD 87 ;- 88 * O 7\A -\- o 10 IO7 ^ TOO O 7^Q o 61 1U /'O 1W 7 o 781 O QO 1^5 */ Txr TX.7 U. /Oj O 771 , \J.^J 4- o 06 T.JP T47 O 774. + O 7O Cctvl-Eunyl ether . . . . 1 4D X 4/ T^C 147 o 805 O ^1 211 212 O QI I 1 i 81 '.S 1 -'o-' 1 Compt. rend., 121, 827. - Arch. Sc. phys. Genfcve [3], 33, 409; Compt. rend.. 120, 1345- 3 Plimpton: J. Chem. Soc., 39, 332 ; Compt. rend., 92, 531, 883. 4 Compt. rend., 120, 452. 510 CONSTANTS OF ROTATION OF ACTIVE BODIES WEI.T 1 gives for other ethers : / i*]* T7 + A rO - 1 / TQ 4.0 1 A ~{. X 7 i 4- 2 1 i 8 22 : 3' 4,5-Methyl-propyl-phenol-atnyl ether. . 4,6-Methyl-propyl-phenol-amyl ether. . 18 19 3-93 4-4.17 + 4.01 /-\TT /-* TT 2. C< CH CHOH' From Ro man camomile oil : Boiling-point 154 (at 758 mm.), d u = 0.829, /= 17, [>]=: + 8.2.' 3. METHYL-HEXYL CARBINOL, Obtained by reduction of the following compound : Methyl-hexyl ketone. tate : /=2i '=57 By saponification of amyl acetoace- = + 5-06 = ' + 4*1 3. Alcohols with Two to Four Atoms of Oxygen A few optically active compounds of this kind are known, such as /-propyleneglycol, but the optical constants of none of them have been accurately determined. As derivatives of such alcohols we may consider : Diphenyl-ethylene-diamine, H 5>CH ~~ CH< NH 5 ' " Obtained by reduction of benzyl dioxime. Melting-point 90 to 92. Split as bitartrate. Ether: /= 15 rf-Base. 0],=: + I34-8 1 /-Base. \fi\D- - 128 1 Ann. chim. phys., [7], 6, 115. 2 van Romburgh: Rec. trav. chim., 5, 220. 3 Welt: Compt. rend., 119, 855. Welt : Loc. cit. 6 Feist and Arnstein: Ber. d. chem. Ges., a8, 3167. ALCOHOLS WITH SIX ATOMS OF OXYGEN 511 /, 5-Tetrahydronaphthylene-diamine. Split as bitartrate. The hydrochlorides in aqueous solution gave at 17.5 : = -h 4o. 10 Hexachlorhydrin, C 6 H 6 C1 6 : Benzene [^] D = + 18.5." 1 Bamberger: Ber. d. chem. Ges., 23, 292. - Ibid... 20, 1234. 3 Fischer, Piloty : Ibid., 24, 521. * Bull. soc. chim., [3], 5, 554. 5 Fischer: Ber. d. chem. Ges., 26, 633. 15 Fischer, Piloty : Ibid., 23, 3102. ~ Miintz, Aubin : Ann. chim. phys., [5], 10, 566. 8 Krecke: Arch. Neerl., VII, 1872. Krecke: Ibid., VII, 1872. 10 Krusemann: Ber. d. chem. Ges., 9, 1468. 11 Mourges: Compt. rend., HI, 112. 512 CONSTANTS OF ROTATION OF ACTIVE BODIES /-MANNITOL. In borax solution strongly left rotating. 1 Isomannitol. Water, p = 6, [<*] = -f 91.36. Alcohol, P = 3, []/> = + 94-66V DULCITOL. Inactive, also in borax solution/ 5 Bouchardat's statement 4 that diacetyl and tetraacetyl dulcitol are slightly right rotating has been contradicted by Crossley. 5 SORBITOL. See statement in 70. = + 65.7 10 Matezttol, identical with pinitol. Melting-point, 187. Water .................... [a] D = + 66.0 Water, c = 3.625, [a]/, = + 64.7 Water, r = 11.867, [a] 7J = -f 65.2 1S /-INOSITOL. Water, []/,= - 65. u Methyl-l-inositol, quebrachitol . Water, [a]^ = -- 80. 1 15 6. Alcohols with Seven and More Atoms of Oxygen Heptitols, CH,OH.(CH.OH) S .CH,OH. ^-MANNOHEPTITOI,, C 7 H 16 O 7 , perseitol. Perseitol from Laurus I Fischer: Ber. d. chem. Ges., 33, 375. - Fauconnier: Bull. soc. chim., [2] 41, 119. Fischer, Hertz: Ber. d. chem. Ges., 25, 1247. Ann. chim. phys., [4], 37, 68, 145. Ber. d. chem. Ges., 35, 2564. Fischer: Ibid., 37, 1524. Fischer, Piloty: Ibid., 33, 3827. Maquenne: Compt. rend., 109, 968. Maquenne: Ibid., 109, 812. 10 Combes: Ibid., no, 46. II Combes: Ibid., no, 46. '* Girard: Ibid., no, 84. 13 Girard: Ibid., no, 84. 14 Tanret: Ibid., 109, 908. Tanret: Loc. tit. ACIDS WITH TWO ATOMS OF OXYGEN 513 persea L., is inactive according to Miintz and Marcano, 1 but according to Gernez, 2 slightly left rotating, and in aqueous so- lution for c - 7.36, [tf]/1 = 1.22. Variety from ^/-mannoheptose : for cold saturated bornx so- lution, c=S, right rotating, [>] -f- 4-75- 3 On the action of acid sodium and ammonium molybdate, see ^70, p. 256. VOLEMITOL, C.H 16 O 7 . Water, p 10, [>]=: -f 1.92.* GALAHEPTITOL, C T H 16 H : . Water -f- borax, p = 8.8, [ci]* D = -4-35 - 5 C 8 H 1S O,.. Water, p= 10.24, & = 1-038, [a] 5? = -f- 2." After addition of borax three times as strong. 7. Acids with Two Atoms of Oxygen d- VALERIC ACID, methyl-ethyl acetic acid, C,H 5 .CH 3 .CH. CO,H. From /-amyl alcohol. [From amyl alcohol of [a]/, = 4.4. Boiling-point, 173 to 174, (730 mm.).] d" 0.938, [a\ D = + 13.64. T [From amyl alcohol of [<*]}5 = 5-2. Boiling-point, 174.5 (768 mm.).] ]g= + io.8 4 . 9 According to Schiitz and Marckwald, 10 valeric acids with the rotations given must contain still about 20 to 25 per cent, of impurities. For the pure substance [oi\ D must be -f 17 to 1 8 ; see /-valeric acid. Derivatives VALKRALDEHYDE, Boiling-point, 92.5, d =0.8209, []/, = -f- i.7 011 (maximum value). \a] l % = -\- 14.09 ; [<*]% = -h 11-14." I Ann. chiin. phys., [6], 3, 279. Compt. rend., 114, 480. Fischer and Passmore: Her. d. chem. Ges., 23, 2226. Fischer: Ibid., 28, 1973. Fischer: Ann. Chem. (Liebig), 288, 147. Fischer : Ibid., 270, 99. Guye and Chavanne: Compt. rend., 116, 1455. Roger: J. Chem. Soc., 63, 1134. Guye and Aniaral: Arch. sc. phys. Geneve, [3], 33, 409. Ber. d. chem. Ges., 29, 59. II Erlenmeyer and Hell: Ann. Chem. il^iebig), 160, 257. 1- - Guye and Amaral: Arch. sc. phys. Geneve, [3], 33, 409. 33 5H CONSTANTS OF ROTATION OF ACTIVE BODIES YALERALDOXIME, |>]~ = -f 11.13; M s /> = ESTERS OF For other esters see Guye and Guerchgorine/ i Guye and Amaral: Loc.cil. - Guye and Chavanne : Compt. rend., 116, 1455. :t Ber. d. chem. Ges., 39, 52. 4 v. Romburgh: Rec. trav. chim.Pays Ha> , 5, 222. Komburgh. " Compt. rend., 124, 130. ACIDS WITH THREE ATOMS OF OXYGEN 515 ^-AMYLACETIC ACID, C,H 5 .CH 3 .CH.(CH,.CH,CO,H). By saponification of amyl acetoacetate. Acid / 20, [] n = - 8.44, /- 54 J []/> = + 7-64 l Methyl ester / 25. " = + 6.71, * = 75. " =+5-92 Ethyl ester / = 21, " - + 6.66, / = 72, = + 5.87 For amyl esters see under amyl alcohol. PARASORBIC ACID, sorbinoil, C 6 H 8 O 2 . \_a~\j = + 40. 8. 2 = 8 5- 9 - 1 If but a few per cent, of the anhydrides are present in lactic acid solutions, they exhibit left rotation. d-Lactic Acid Salts. These all show left rotation. Zinc Salt, Zn(C 3 H 5 O 3 ) 2 + 2H 2 O. The rotation of aqueous solutions increases slightly on dilution. There was found for the hydrated salt : i. c = i6.o 5 , 2 [*]/,= - 6.36 2. p = 9.08, \ci\ D = - 6.56 li n.oi, 2 6.36 8.29, 6.64 7.47, 6.83 653, 6.84 6.13, 7.41 5.89, 6.83 5.26, 7.60 4.18, 7.55 3. =7-49, [oi\ D = -6.83* Zinc Ammonium Salt, Zn.NH 4 (C 3 H 5 O 3 ), + 2H 2 O. Water, c = 8.00, [#] D - - 6.06 .* Calcium Salt, 2Ca(C 3 H 5 O 3 ) 2 + gH,O. Water, c = 7.23, [d] D = - 3-87 5 Water, /> .- 6.25, [] . -3.85" Lithium Salt, Li.C 3 H 5 O 3 - Water, p ^ 5 to 12, [] /( - 10.95 to 12.28 (i d-Lactic Acid Esters. Methyl ester rff i.ioo, [a] 7 , - 11.1 \ 7 Propyl ester ^1.004, []/,= -17.06 < For further determinations see Frankland and Henderson. 8 /-LACTIC ACID. By resolution of fermentation lactic acid by means of strychnine (33), or by the action of Bacillus acidi laevo-lactici on cane-sugar. 1 Wislicenus: Ann. Chem. (i,iebig), 167,321. 2 Supersaturated. Wislicenus: Ibid., 167, 332. Hoppe-Seyler and Araki: Ztschr- physiol. Chem., ao, 371. 4 Purdie : J. Chem. Soc., 63, 1154. 6 Wislicenus : IMC. cit. Hoppe-Seyler and Araki : Lot: cit. ' Walker: J. Chem. Soc., 67, 916. * Proc. Chem. Soc., n, 54 (1895). ACIDS WITH THREE ATOMS OF OXYGEN 517 Water, c -64.8, [ci\ D = - 4-3 ' Water, p = 12.43, d = 1.0348, t = 23.3, [] - 4.72 ! Water, /> 6.57, = 6.751, ],> = + 6.32 . 5 Zinc Ammonium Salt, Zn.NH 4 (C 3 H 5 O 3 ) 3 + 2H 2 O. / - 8.63, df = 1-035, [or]g - -f 6.49 I 6 / = 5.87, " ==1.024, - + 7.7 ' Lithium Salt, LiC 3 H 5 O 3 . p = 3 . 7 to 9.1, [tf]^ = -r 13.5 to 12.7 7 Sodium Salt. See 57, p. 202. Ester. d?> =1.030, [flf]^ = + 14-52 8 Chlor-propionic Acid, CH 3 .CHC1.CO 2 H. Methyl ester of the rf-acid- d' -= 1.1520, [a] D = + 19.01 | u Ethyl ester of the flT-acid ... " =1.0888, " =+12.86 / Methyl ester of the /-acid d\ =1.158, " = 26.83 ^ 12 Ethyl ester of the = 71-09 Sodium salt, water ---- c = 16.530 49.43 Calcium salt, water ---- c = 9.53 " 38.09 d- Ethoxypropionic acid, water ---- c= 29.374 -f- 56.96 Sodium salt. water ---- c= 17.965 + 48.09 Calcium salt, water ---- i 26.87^ -f 38.40 d- Propoxypropionic acid, water.... -=11.450 +55-63 Sodium salt, water. . . . = 30.750 -f- 48.94 Calcium salt, water ---- c = 12.010 -f 48.54 Purdie and Irvine" have determined the optical behavior of methoxypropionic and ethoxypropionic acids and esters pre- pared from active lactic acid. /S-OXYBUTYRIC ACID, CH 8 .CH.OH.CH,.CO,H The /-form has been found in diabetic urine. 3 Acid. Water, [a\ D = 20.6. 4 []/> = 23-4.' Sodium Salt. Water, p = 32.1, [oi] D = 15. 4 p = 20.9, ^=1.0900, \a\ D I3-93 . 6 Stiver Salt. [oi\ D = 10.1 ; 4 c = 1.4, []/,= 8.64. 7 A series of esters of d- and /-oxybutyric acid has been in- vestigated by Guye and Jordan. 8 In hydrochloric acid solution, left rotating. Kef J. Chem..Soc., 73, 862 (1898). /*"/., 75, 483 (1899). Minkowski : Her. d. chem. Ges., 17, Ref. 334, 535. Kiilz : Ber. d. chem G8., 17, 534 ; 18, Ref. 451. Minkowski. Kiilz : Ber. d. chem. Ges., ao, Ref. 591. Deichmiiller, Szymanaki and Tollens : Ann. Chem. (Iiebig), aa8, 94. Kiilz. Compt. rend., lao, 1274 and tao, 630. ACIDS WITH THREE ATOMS OF OXYGEN 519 From conglutin : 20 per cent. HC1 c -- 4-73, M/> ~ l 7-5 From active (conglutin) leucin : 20 per cent. HC1 c 5.00, []/,= - T 7-3 From inactive leucin : 20 per cent. HC1 c = 4-37. [^]z? = ~'7-4 01 By resolution of the or-amido acid from fermentation caproic acid by means of Penicillium glaucum, Schulze obtained an active leucin, different from the ordinary leucin, and which was left rotating in hydrochloric acid : ~ Acid ... c = 4 to 5, [a] D = - 26.0 to 26.5 /-LEUCIN. In aqueous solution, left rotating ; 3 in acid or alkaline solution, right rotating. From casein : 10 per cent. HC1 c = 6.4 [>]/> = + J7-54 V Alkaline <: = 5-6 = + 6.65 / From beet molasses : 4 per cent. XaOH =2.371, / = 20, [a~] D = -f- 8.05* From conglutin : 19 per cent. HC1 c 5.00, [a~\ D = -f 17.31 6 20 per cent. HC1 c= 4-73, = + I7-3 7 RICINELAIDIC ACID, CH 3 (CH 2 ) 5 CHOH.CH:CH.(CH. 2 X- COOH. Melting-point, 53. Acetone c 5 to 15, [-] D = + 4.8 to 5.4 ) 8 Alcohol =12, ^4-6.67 / RICINOLEIC ACID. Isomeric with ricinelaidic acid. Acetone c = 4.8 to 21, /== 22, [] D = -f 6.27107.5* RICINSTEAROLEIC ACID, Q.H^OH.COOH. Melting-point, Acetone c = 6.4, [a]^ = + 13.67 9 1 Schulze and Bosshard: Ztschr. physiol. Chem., 10, 143; Schulze and I,ikiernik : Ber. d. chem. Ges., 24, 472. '- Ber. d. chem. Ges., 26, 56. 3 lyewkowitsch : Ibid., 17, 1439. Mauthner : Ztschr. physiol. Chem., 7, 222. Landolt : Ber. d. chem. Ges., 27, 2838. Schulze and Bosshard ; Ztschr. physiol. Chem., 9, 100. Schulze : Ber. d. chem. Ges., 26, 56. Walden : Ibid., 27, 3471. Walden. 520 CONSTANTS OF ROTATION OF ACTIVE BODIES ]/> = -128.4 (m. p. 35 ) " 3-33, - I26 -4 Acetone 5.81, - 90.62 " 1.16, - 87.1 Carbon disulphide ... 5.00, -180.0 " " ... 2.50, 180.0 " ... 0.88, -180.5 Isobutyl ester Superfused d 1.0870, - 100.73 (b. p. 159, 19 mm. solid) Carbon disulphide... = 5.0, 146.6 "... 2.5, -144.0 j-Amyl ester b.p. i66-i67, 17 mm. d = 1.0531, [a~\ D - - 9 6 -46 /-Amyl ester " i66-i67, 17 " 1.0530, " - 94.02 Acetylmandelic acid, methyl ester " 177, 45 " -.1546, " 146.37 Propionylmandelic acid, methyl ester " 184, 45 " 1.1261, " 135-5 Propionylmandelic acid, ethyl ester, Superfused d =. 1.0936, 113.7 (b. p. 1 77, 30 mm. Chloroform c 10.0 -no.8 m. p. 33 C ) " 5.0 -109.4 Carbon disulphide.. 5.0 131.5 " 2.5 " -126.8 Valerylmandelic acid, ethyl ester, b.p. 1 73-i74, 18 mm. rf =: 1.0544, " - 97.06 Carbon disulphide.. c = 10.0, 117.25 " 5.0, 116.9 Acetylmandelic acid, m. p. 56. Acetone-. 3.33, 156.4 In addition Walden 1 gives the following observations : ^-PHENYLCHLORACETIC ACID, C 6 H 5 .CHC1.COOH. From /-mandelic acid. Melting-point, 56 to 58. Benzene c = 3.33, [orj^^-f 132.13 " 5-33, 4-I3I.6 Carbon disulphide 4.00, -f 131 .3 Chloroform 5.33, 4-107.9 Chloride, liquid. b. p. I20 J (23mm.) Carbon disulphide c 6.0, \_<*\D == ~^ I 5^-33 Methyl ester, liquid b. p. i35-i36(22mm.) d = 1.2087 +107.55 Ethyl ester, liquid b. p. 162 (45 mm.) rff = i.i594, + 25.19 In carbon disulphide, p = 4.96, d = 1.2527, -}- 26.39 -Propyl ester, b. p. 180, 60 mm. 1.1278, + 23.94 1 Ztschr. phys. Chem., 17, 714 ; Ber. d. chem. Ges., 28, 1295. 522 CONSTANTS OF ROTATION OF ACTIVE BODIES i-Amyl ester, " i67-i68, 20 mm. rf= 1.0828, [>]/> ^. + 23.31 /-Amyl ester, " i69-i7O, 24 " 1.0826, -f 26.79 - -135. Quinine salt. m. p. 204 -205. 2 Alcohol = 0.3800, /=I3, -118.4 Cinchonine salt. m. p. 167.'- Alcohol =1.3308, t --- 24, + 83.4 ^/-TROPIC ACID, C,H 5 .CH(CH 2 OH).CO 2 H. Melting-point, 127 to 128, [=: + 66.5. Methyl ester : Alcohol [a] D = + 61.9 Ethyl ester : Alcohol +64.1 J. Chem. Soc., 71, 253. 2 Ibid., 63, 1419. 3 Frankland and MacGregor: /bid., 69, 104 (1896). Prankland and Patterson : Ibid., 73, 181 (1898). Ibid.,., 75, 493 (1899). ACIDS WITH FOUR ATOMS OF OXYGEN 525 /-Acid, [] = - 65.9. ' The following compounds may also find a place here : (/?)-TYROSIN. C 6 H i (OH)CH.NH,.CO 2 H. Melting-point, 235 (Lippmantr) ; 290 (Erlenmeyer and Lippmann 3 ). Rotation in aqueous solution in presence of acid or alkali. From silk : 21 per cent. HC1 = 4-5*, t --= 16.2, [] D = - 7.98 * 1 1. 6 per cent. KOH ......... 5.8, 20.5, 9.01 1 1. 6 per cent. KOH ......... 11.51, 16.1, -8.86 From beet molasses : 21 per cent HC1 ............. C : 3.92, / = 20, []/> = - 8.oy c 5 From conglutin : 21 per cent. HC1 4 per cent. HC1 A right-rotating tyrosin is known, [a~] D = -f- 6.85, ; but whether or not this is a position isomer has not been deter- mined. CYSTIN (SC(CH 3 )(NH,)(COOH)) 2 . Dissolved in ammonia ........ P ~~ J-OS 1 ) \_^\j ~ 141.1 ! Dilute hydrochloric acid ..... c 0.8 to 2, r] Z j= 205.86 9 Dilute hydrochloric acid ..... c = 2.13, 214 10 Derivatives : Bromphenylmercapturic Acid, CH,.CONH.C(SC 6 H 4 Br)(CH 3 ).COOH. Alcohol .................. p = 12 to 15, \_a~\ D = ^6.7 Dilute NaOH ............. 25, -f 6.4 PhenylmercapturicAdd, CH 3 CONH.C(SC 6 H 5 ) (CH 3 )COOH. Na Salt, CH 3 CONH.C(SC 6 H 5 )(CH 3 )COONa. Alkaline solution .......... p = 8, [o-]^ = + 4 Finkenbeiuer : Ber. d. chem. Ges., 27, 889. Ibid., 17, 2837. Ann. Chem. (Mebig), 219, 173. Mauthner : Monatsh. Chem., 3, 343. I^andolt : Ber. d. chem. Ges., 17, 2838. Schulze and Bosshard : Ztschr. physiol. Chem., 9, 98. lyippmann : Ber. d. chem. Ges., 17, 2839. Kiilz : Ztschr. fur Biologic by Kiihne and Voit, 20 [N. F. 2], 9. Mauthner : Ztschr. physiol. Chem., 7, 225. 1C Baumann : Ibid., 8, 305. 5*6 CONSTANTS OF ROTATION OF ACTIVE BODIES Bromphenylcy stein, (NH,) ( CH :J )C(SC 6 H 4 Br)COOH. Dilute NaOH p = 9, [<]/>= ~ 3-7 ! 10. Acids with Five Atoms of Oxygen Malic Adds, CO,H.CH.OH.CH.,.CO 2 H. NATURAL MALIC ACID, commonly known as /-malic acid. This was first investigated by Pasteur in aqueous solution and found to be left rotating. Later investigations by Schneider' showed that the left rotation decreases with increasing con- centration, that inactivity is reached with/> 34.24, and that then right rotation follows. The dependence of the specific rotation on the percentage amount of water in the solution q is shown by the formula, O]# ~ 5-891 0.08959 q (q =a 30 to 92), which gives, for example, the following values : Q- P- Ms- 9- P- Ms. 30 7 4- 3-203 70 30 - o. 3 8o c 40 60 i 2.307 80 20 1.276 50 50 + 1.412 90 10 - 2.172 60 40 + 0.516 95 5 - 2.620 According to the above formula, the anhydrous malic acid must have the rotation, [a] D -. -\- 5.89, and the substance must, therefore, be considered as right rotating. As shown in 63, however, the right rotation may possibly depend on the formation of crystal molecules, and as the ordinary acid shows left rotation in other liquids besides water, acetone and methyl alcohol for example, it is advisable to retain the old designation of /-malic acid to avoid misunderstanding. The right rotation of concentrated solutions is diminished by increase of temperature, but the left rotation of dilute solu- tions, on the other hand, is increased. A solution with per- centage strength/ 28.67 is ri g nt rotating below 15, but left rotating above. For further details see 60. On the rotation dispersion of malic acid see 46. 1 Baumann : Ber. d. chera. Ges., 15, 1731. - Ann. chim. phys., [3], 31, 81. * Ann. Chem. (I y iehig), 307, 261. ACIDS WITH FIVE ATOMS OF OXYGEN 527 For the rotation of solutions in acetone and methyl alcohol, Walden 1 found : Acetone Methyl alcohol ^=13-3, [].,, -5-oto 5.34' 30, - 2.78 Influence of Sulphuric Acid and Acetic Acid on the Rotating Power of I- Malic Acid. According to the experiments of Schneider, 1 ' the left rotation of dilute solutions of malic acid is decreased by addition of increasing amounts of the acids named, and after passing a point of inactivity, an increasing right rotation follows. The following figures were obtained, which are given in connection with the value of []/> for the pure aqueous solutions : I. SULPHURIC ACID. Composition of the solutions in ' 1 per cent, by weight. Mol. H 2 SO 4 to i Found, mol. C 4 H 6 O 5 r/vl 20 -r loo mol. HoO. L"J^- M* without H 2 S0 4 . C 4 H 6 5 . H0. H 2 S0 4 . ' 6.76 90.77 2-47 \ - 1-33 2.46 6.59 88.58 4 .82 I 0.76 ; - 2.48 6.44 86.50 7.06 \\ O.2O - 2.49 6.29 84.51 9.20 2 + 0.21 2.50 6.15 82.61 11.24 A 0.84 2.52 II. ACETIC ACID. Composition of the solutions in per cent, by weight Mol. CH 4 Oo to i mol. C 4 H 6 O 5 + 50 mol. H 2 O. Found Ms. [> without C 2 H 4 2 . C 4 H 6 5 . H,( ) QjH 4 O 2 . 10.04 67.47 22.49 5 -1.35- -2.17 8.20 55.08 36.72 10 -0.57 -2.33 6.00 40.29 53-71 20 0.13 -2.53 5-29 35.52 59-19 25 + 0.14 -2.59 Rotation without Addition of a Solvent. From the formula of Schneider, above, which shows the dependence of the specific rotation of aqueous solutions of malic acid on the con- centration, it follows that the acid must be right rotating in 1 Ber. d. chem. Ges., 29, 137. - Ann. Chem. (Liebig), 207, 279. 528 CONSTANTS OF ROTATION OF ACTIVE BODIES anhydrous condition, and show []p = : + 5.89 at a tem- perature of 20. Walden 1 has been successful in observing this right rotation by rapidly melting the acid at a temperature of 100 to 110 and pouring it into a glass trough with parallel end surfaces. In the hot condition the liquid mass rotated to the left, but after cooling, the rotation was to the right. The following rotations, referred to a layer i dm. in thickness, were found by the Landolt ray filter method. Temperature. 90-95 17 "^ a for Red 4-4 +3-5 Yellow (D) 5-5 +5-2 Green 6. i + 6.8 Light blue 7.0 -f 7.7 The fused mass when dissolved in acetone exhibits left rotation, and the more strongly, the longer the heating was continued ([>]/> = - 8 to 16). This shows the begin- ning of anhydride formation. Alteration in the Acid on Heating. As Walden- found, malic acid, when heated in the dry condition to 100 for twenty-four hours, or when heated through a shorter time to 165 in vacuo, is converted into a tribasic anhydro acid, C 8 H 10 O 9 , while at a higher temperature (i8o),a dibasic acid, C 8 H^O 8 , is formed. The two anhydrides show strong left rotation when dis- solved in acetone, amounting to [<*]/> - 19 to 25, while the original acid shows [#]/> = 6 to 7. Rotation of Malic Acid in Different Solvents. Experiments on this point have been made by Nasini and Geunari 3 and also, especially, by Walden. 4 By help of ray filters, the latter found the following specific rotations, from which the corre- sponding dispersion coefficients, DC, referred to [] re d= i were derived. All the values obtain for a temperature of 18. As usual, c is the number of grams of malic acid in 100 cc. of solution. 1 Ber. d. chetn. Ges., 33, 2849 (1899). - Ibid., Ja, 2706 (1899). * Ztechr. phys. Chein., 19, 117. 4 Ber. d. chem. Ges., 33, 2856 (1899). ACIDS WITH FIVK ATOMS OF OXYGEN 529 It was found that the rotating power of /-malic acid is sub- ject to extremely great variations. Solvent. Benxyl alcohol. i vol. benzyl alcohol, i vol. benzene. 3 vols. benzyl alco- hol, 2 vols. carbon disulphide. * 12. 6. 4.8. [a] Red 2. 7 He I 1-3 >c = i ^ 3 Dc= i Yellow /;. -- 4.0 i-5 2.5 + 3-4 1.9 Green 5.5 2.0 + 4-75 3-7 + 4-4 2.5 Light blue- 7-7 2.8 5 + 6.75 5-2 - 6.5 3-6 Dark blue. - II. 4.1 + 9-7 7-5 4- 8.5 4-8 In the above cases, the /-malic acid exhibits right rotation with strong rotation dispersion. Solvent. Cry st. formic acid. Acetone. c. 7.66. 38.3- 23-7 / = 18. t 23 o [a] Red . , -0 o 8 c n Dr \ fi Dr Yellow D- 0-1 -4-6 -0.75 6.0 1.2 -5-5 1.2 Green 5-2 -0.4 -7-1 1.42 -6.7 1-45 Light blue -5-9 -r 0.6 -7-5 1.50 7.0 1-52 In these solvents, as in the following, the acid is left-rotating. With formic acid the rotation decreases with the concentra- tion and becomes even positive for the blue rays. Solvent. 3 vols. acetone, 2 vols. benzene. Phenyl-methyl ketone. ;i vol. phenyl-methyl ketone, i vol. paraldehyde. c. 9-44 5- 5- [a] Red . 1 r> i 6 1 Yellow D. 4.0 - 4.1 3- -4.0 J- J - 3-o Green Light blue -4.i -4.2 4.0 -4-6 -3.0 3-0 In the above liquids we notice the very peculiar phenom- enon that the rotation is nearly the same for all rays. Th same behavior was observed by Nasini and Gennari for 34 530 CONSTANTS OF ROTATION OF ACTIVE BODIES aqueous solutions of malic acid having a concentration of c = 8.48.' Solvent. i vol. acetone, i vol. isobutyl alcohol. Isobutyl alcohol. i vol. formic acid, i vol. ethyl acetate. r. 11.8 5. 10. 9.58. -5-8 -6.6 -7-4 -8.9 DC. I 1.14 1.28 1.50 6.2 -6.3 -8.4 -8.5 -8.6 DC. 35 37 38 -3-2 -3-7 -4.1 -4-4 5.0 DC: I 1.16 1.30 1-37 1.56 DC. - 7.5 i - 8.9 I.I9 - 10.4 1.39 - 12.0 1. 60 Yellow D. Green Light blue Dark blue - In the above solutions the with the refrangibility of the consequence is but weak. rotation increases but slightly rays ; the rotation dispersion in Solvent. i vol. acetone, i vol. paraldehyde. Acetaldehyde. Pyridine. c. 1 1. 8. 4- 5- Red T 1 T Dr T ?7 7 Dr T Yellow D. Green Light blue -17-6 1.35 -22. 9 1.75 -27.5 2.10 28.7 1. 21 -32.5 1-33 -38.7 1.63 30.0 36.0 1.30 1-57 These three liquids which are able to act chemically on the malic acid produce a very strong increase in the left rotation, while the dispersion is not greatly increased. A substance which has the power of enormously increasing the activity of malic acid is, as Walden 2 observed, uranyl nitrate with simultaneous addition of alkali. The specific rotation with this addition may be increased to 158 times that shown by a pure aqueous solution of the acid of the same con- centration. It is probable that the high rotation corresponds to the alkali salt of a complex uranyl malic acid. Walden has investigated the following ten mixtures of different compo- sitions, in which it is seen that the maximum of rotation is found in solutions Nos. 6 and 7. These correspond to a relation i Loc. cit. 1 Ber. d. chcm. Gs., 30, 2889 (1897). ACIDS WITH FIVE ATOMS OF OXYGEN 531 of i mol. of malic acid, 4 mols. of potas^tum hydroxide, and i to 4 mols. of uranyl nitrate, UO 2 (NO 3 ) 2 -f 6H,O. No. In 100 cc. aqueous solution. <*D 2 dm. M* Malic acid. Grams. Potassium hydroxide. Grams. Uranyl nitrate. Grams. . I 0.65 I. 08 0.04 - 3 2 0.65 10 0.14 ii 3 0.65 0.27 2 -I.8o 139 4 0.65 0-54 2 - 3.60 | 277 5 0.65 0.54 10 -3-13 - 241 6 0.65 i.c8 3 -6.1 7 -475 7 0.65 i. 08 10 6.09 -470 8 0.65 1. 08 20 -5.81 -447 9 0.65 1. 08 34 -5-43 -415 10 0.65 1.62 10 -5-86 -451 By uranyl acetate, with addition of alkali, the rotation not only of malic acid, but also that of o 70 - 4-53 -5.38 -4-53 -4.04 -7-29 - 6.93 - 5.97 -6.17 2o so -5.08 -6.13 -5-59 6.14 -8.06 - 9-43 - 6.26 -6.99 10 90 -5.64 -6. 7 8 .... 8.07 -8.45 -11.36 -6.55 On the inactivity point in the concentration, see 57.* For dilute solutions of salts of malic acid see 61 . A series of observations on the specific rotation of neutral sodium malate has been carried out by Th. Thomsen.' 2 Accord- ing to these experiments the point of inactivity is found for a strength of 45.63 per cent, of the salt. Through the presence of free alkalies, the rotation of the malates is displaced in the direction, left to right, that is, the action follows in the same direction as with addition of free acids. The following experiments by Thomsen 3 show the ex- tent of the change : p. 28.93 19.23 10.94 Ms Amount of change. Of the pure Na 2 C 4 H 4 5 . After addition of i mol. NaOH to i mol. salt. 4.02 -6. 3 I 7.90* + 0.37 -4-05 -6. 5 6 4-39 2.26 1-34 1 In using the interpolation formula with three constants, the composition of the solution with which inactivity must appear, is most easily found by bringing the Ft n A equation A -f Bq -f Cq* o into the form g* -f ? = , two trigonometric formulas : C I-A and then applying the tan 1 J. prakt. Chem., [2], 35, 153. Lof. til. 4 Calculated from Schneider's formula. -v : ACIDS WITH FIVE ATOMS OF OXYGEN 533 Of the pure Na,C 4 H,0, Amouut of change. 2 7 .2 3 -4-84 i8.7i -6.39 9.38 -8.19! f 10.74 - L i-99 - 5-09 15.58 8.38 3-10 The alteration increases, therefore, with the amount of free alkali added, and is the greater the stronger the solution is. Solutions which are originally left rotating become in this way, easily, right rotating. The influence of temperature on the specific rotation of the malates is shown, as with the free acid, in displacing the rota- tion in the direction right to left for increase in temperature. Th. Thomsen 2 has noted the change for the following salts : K 2 C 4 H 4 5 . Na 2 C 4 H 4 5 . M.at [>]z,at p. 10. 20. 30. A 10. 20 ' 30. 33-86 - 4.48 - 5.22 - 5-85 42.75 - 0.38 - 0.89 - 2.04 23-25 -5.18 -5-90 -6.57 28.60 - 3.41 - 4.52 -5.58 16.29 5.62 -6.35 - 7-09 19.51 - 5-30 - 6.36 -7.41 14 46 _ ^o _ o _ 7 ofi ' ' 4.994 4.69 1.965 - 2.58' Barium Malate, BaC 4 H 4 O 5 . p - 9-383 8.505 |>]-= -f-8.i8 +8.05 This salt is, therefore, right-rotating down to a strength of the solutions of about 3 per cent. 3 Ammonium- Antimony I Malate, NH 4 SbOC 4 H 4 O 5 , shows a very high specific rotation to the right : Water., p = 6.845, t = 17, []./ = 4- "5.47, [rm, c water H 4 0, i > i < t 1.2337 1.1294 1.4380 1.0760 1.0418 1.0176 1.0179 0.9761 1.1975 1.1168 1.0724 1.0362 1.1317 1.0736 1.0417 1.0146 1.1255 1.0688 1.1034 1.0605 i .0263 1.0045 1.3062 1.1566 1.5072 1.3936 1.3150 1.2022 1.3325 1.3059 1.2850 = 4-0) - 6.85 TO iH - 11.62 - 10.41 - 11.14 - 6.88 - 9.92 - 6.92 - 22.92 - 22.52 22.85 - 21.88 - 22.94 - 22.20 - 22.44 22.22 22.40 ?T 68 Diiu~>VHitvl \\\_l flTTlvl Dipthvl 4< I Methyl ** Diethyl 4i Dimethyl isobutyrylmalate - 22.36 Diethyl tk Dimethyl isovalcrylmalate 21.99 Diethvl " 22.39 iMpropyl ** -21.68 - 19.91 Diisobutyl Dimethyl chloracetylmalate Dipropyl 23.30 Dimethyl bromacetvlma.la.te *5-o* Diethyl *yy 1 *N Dipropvl 22.24 on i^ Diethyl a-bronipropionylmalate .. Diethyl "^ w^.w 66 r C= 1. 00 < 1. 00 00.5 70.0 " " /^-naphtiniide 5 X .5 Walden /tschr. phys. Chem.. 17, 248. ACIDS WITH FIVE ATOMS OF OXYGEN Anschiitz and Reitter 1 find : 535 Boiling-point at 12 mm. d?. M* Methyl malate 122 211J. 688 Ethvl ' ' . I2Q 2 J -I2Q 6 I28o TO *\5 I ^O O7^6 1 1 60 WHntvl 1 60 J. I 7O A. 0-182 IO 72 ToSt 22 86 Fthyl jj j 2141 4 *Vo 1 1 60 22 60 T ^8 6 I ^Q 2 O72Q 22 68 0/170 IQ Q"% Purdie and Williamson 2 have obtained nearly the same re- sults for some of these esters. Frankland and Wharton 3 have made the following determi- nations : Bp. mm. d' v t. Mi. Methyl benzoyl ma late 210-223 12 1.2121 21 - 5.62 Ethyl 210-220 12 I.I56I 21 - 3.87 Methyl orthotoluyl malate.. 214-225 12 I.I909 23 - 8.94 Ethyl it 215-225 12 I.I39I 21 6.25 Methyl metatoluyl malate 215-225 12 ' I.I925 20 6.34 Ethyl " ..; 212-220 13 I.I37I 21 4.67 Methyl paratoluyl malate 200-225 I 3 I.I957 18.5 -3.14 Ethyl I.I382 20 0.22 The densities and rotations were determined for a wide range of temperatures and the latter were found to vary greatly with the temperature. d- MALIC ACID. Antimalic acid. By reduction of c. cit. 536 CONSTANTS OF ROTATION OF ACTIVE BODIES ]/>-= + 33-3 )' 11 f= 5.586, /-=i8, =4-33.04) ' c--~ 24.65, t--~ 15, " = + 32.59 11 ^^16.08, / 14, ^- + 32.79 " C 8.76, t^ 15, " r -]- 32.70 Acetone c ----- 24.96, / 11, "' = 57.10 ' ^18.77, t -11, " = + 58.29 M c 10.30, /=J4, " - 58.03 < 4-12, t 14, " =r + 59.49 c- 1.65, / 14, " = + 60.09 Ethyl acetate c 20.54, / -- 11, " =- --63.48 " r= 15.87, / 12, M -= + 64.45 " C 8.92, /=I2, " - - 64.64 Salts. In the following data by Ptirdie and Marshall, the concentrations refer to the anhydrous salts : Acid Potassium Salt, crystalline. 6 Water c = 4.010, / -.= 18.5, [],, -. t 23.46 " c-^ 8.150, / : 18, " ^ + 23.26 Normal Potassium Salt, crystalline/ Water == 5.019, / 14.5, []/>=+ 9.36 *== 12.162, / -= 15.5, " -+9-54 Acid Ammonium Salt, crystalline.* Water c -- 6.064, t M , []/> ----- 25.86 Normal A miuon htm Salt, crystalline.' 1 Water c 2.823, t - 14, [a]/, 12.22 " C ' 5-762, / 14, = + 12.32 Her. d. chem. Ges., 39, 137. Purdie and Marshall : ]. Chem. Soc.. 6j, 217. I'urdie and Bolain : Ibid., 67, 946. I'tirdieand Marshall. Purdie and Itolam. I'urdic and Marshall. ACIDS WITH FIVE ATOMS OF OXYGEN 537 Calcium Salt, crystalline. 1 Water c 5.308, / 18, |>]/> -10.10 Barium Salt, C,H 6 O,.Ba + 4H 2 O, crystalline. 1 Water, c 26.125 (anhydrous), / = 18, [or]"/) -14.27 " -=12.416 /=i8, " r - 7.36 " c - 5.746 / 18, " = - 2.21 c=- 1.149 " * 18, " - 3.16 Cinchonine Salt, C-H > O-.C 19 H.,,N. ) O. Crystals, melting-point 171 to 173. 1 Water c -4, ' 17, []/>= -^- 154.89 Methyl Ester, d\* 1.1498, \a] D = -f 52.51.'-' AMETHOXYSUCCINIC ACID. Crystals, melting-point 89. Water = 10.806, / = 18, [a~\ D = 32.94 J " =s 32.70 ~)* Acetone ' c 25.588, /=n, -56.25 " r= 15.614, / = 13, -58.18 Ethyl acetate .... ^^=25.511, / 11, -61.90 " ...r= 19.077, / - 13, " = -62.93 Salts. 1 Acid Potassium Salt, crystalline. Water c = 4.046, / = 18, [a]i,= - 23.59 " ^^4.083, / = i7-5, -23.49 Acid Ammonium Salt, crystalline. Water c = 8.774, / = 18, [0-1^^-25.85 Calcium Salt, crystalline. Water r ^ 5.482, / --= 13, [a]^^ -f- 10.03 " c = 2.210, /=I4.5, " = -|- 4.30 Some esters have been investigated by Purdie and William- n. 4 Ethyl ester a? J 8 = 1.0705, [ 50.11 Propyl ester ^^5^1.0419, 45.21 jV-Butyl ester d lt> ~ 1.0419, 41.63 1 Purdie and Marshall. '-' Purdie and Williamson : J. Chein. Soc., 67, 971. 1 Purdie and Bolam. 4 J. Chem. Soc., 67, 971. 538 CONSTANTS OF ROTATION OF ACTIVE BODIES ACID, CO,H.CH (O.C 2 H 5 ) .CH 2 .C(XH. By resolution of the r-acid by Penicillium glaucum? or with strychnine. 8 Crystals, melting-point 76 to 80. Water ....... r=u.i7, /=I7, []z>= 4- 33-02 ....... c=-- 5.56, /=I9> " = + 32.54 " ....... c=- 8-22.5, / = 12-17, " = + 34.4-34.7 Chloroform .. t = 11.61, / 12, " = + 47.75 .. *=: 4.80, /=I2, "=- + 45-55 .. f s: 4.52, /= II, f 44-17 .. <:=: i .60, * = ii, " = + 39-4o Ethyl alcohol <: = 11.81, / -- n, " = + 60.57 Ethyl acetate r = 5-20, t 18, " = + 69.9-70.5 Acetone ..... c -- 9.57, / = 14, " = 63.39 " ..... c= 3.83, /==i4, -64.87 " ..... r=: 1.53, ^= M, - 66.48 y4W Potassium Salt, KC 6 H 9 O 5 + H,O, crystalline. Melting- point 1 60. Water ......... c - 3.93 (contains water) , t = 19, [a\ D + 26.49 Add Ammonium Salt, NH t C 6 H 9 O 3 + H 2 O, crystalline. 4 Water ......... c - 8. 13 (contains water) , t = 15, [or]/? = + 28.65 11 ......... ^-4-59 " * = i6, = + 29.08 " ......... 6- = 2.58 " /=i7, = + 28.48 Normal Ammonium Salt, (NH 4 ).,C fi H 8 O 5 , crystalline. Water ........... 5.22 (anhydrous), / = 14.5, []/>=+ 18.29 '' ........... <:=.-: i. 48 /^i2, =+18.93 Calcium Salt, CaC 4 H 8 O 5 , crystalline. Water .............. ^ = 3.04, / = 15, [a]/?~ -*- 8.39 Barium Salt, BaC 6 H 8 O 5 (at 160), crystalline. Water .............. ^-4.56, / = 18, []/>- + 6.37 " .............. c~ 10.77, / 18, = + 2.46 " .............. ^^25.08, is 19, -4-37 Estcn.* Methyl ester ............. rf3 = 1.1055, [>]/> = + 59-86 Ethyl ester .............. ^ = 1.0418, = + 55-29 Purdie and Walker : J. Chem. Soc., 63, 229. Purdie and Williamson : Ibid., 67, 963. Purdie and Walker : IMC. cit. Purdie and Walker. Purdie and Williamson : J. Chem. Soc., 07, 971. ACIDS WITH FIVE ATOMS OF OXYGEN 539 /-ETHOXYSUCCINIC ACID. Solutions in water. 1 Acid ammonium salt p = 2.5 \_OL\D 26.05- Methyl ester d 1.0996, " 60.92 ] - 1 Propyl ester d\ --1.0226, " 51.20 ^V-Butyl ester d\ =1.0045, " 46.43 Ethyl ester ^ = 1:1045, " '-44 * Ethyl Ethoxysuccinate. See Purdie and Pitkeathly." PROPOXYSUCCINIC ACIDS, CO,H.CH(O.C 3 H 7 ).CH 2 .CO,H.' rf-Acid. Water c 7.56, t = 12, |>]/> = - 36.04 rf-Acid, potassium salt. . = 3.815, / = 18, " = 4- 32.30 ^/-Normal salt == 3.665, / = 18, " == -f 17.26 " ^1.833, t 18, " = -[-18.69 /-Acid. Water = 7.760, / = 12, - 36.40 " = 3.104, / = 12, 36-24 /- Acid. Acetone r= 5.688, / = 12, 63.29 " =2.275, / = 12, -64.39 /-Barium salt. Water.- = 3.649, / = 18, 10.00 " " " =1.460, / = 18, -10.45 Propyl Propoxysuccinate. For the preparation and optical behavior of this ester see Purdie and Lander. 7 ^-CHLORSUCCINIC ACID, CO 2 H.CHC1.CH 2 .CO,H. From ordinary malic acid by action of phosphorus penta- chloride. Melting-point 174. 1. Water t = 21, c= 16 6.4 3.2 [ct] D ^= -f 20.6 20.8 -+- 21. 3 8 2. Water c 6.66 " = + 20.2^ Ethj-1 acetate c = 10 " = + 52.70 = 6.66 " =-^52.85 Derivatives Anhydride, CO CHC1.CH 2 .CO. Melting-point 80. Ethyl acetate c = 10, [ = -f 30.85 ) 11 c= 5, " " = + 33-6oj Purdie and Walker : J. Chem. Soc., 63, 238. Purdie and Walker. Purdie and Williamson : J. Chem. Soc., 67, 972. Walden : Ztschr. phys. Chem., 17, 252. J. Chem. Soc., 75, 157. Purdie and Bolam : J. Chem. Soc., 67, 949. J. Chem. Soc., 73, 288 (1898). Walden : Ber. d. chem. Ges., 26, 225. Walden : Ztschr. phys. Chem., 17, 253. 540 CONSTANTS OF ROTATION OF ACTIVE BODIES Chloride, COC1.CHC1.CH 2 .COC1. Boiling-point 91 to 93' (iimm.), d*' -- 1.5002, |>]/,=: + 29.53.' Dimethyl ester.. B. p. 114-115(170101.), d?f = Diethyl ester " 130-131(20 " ), ' = Dipropyl ester ... " 148 (20 " ), " Diisobutyl ester. " 162-164(17 " ), " = Diamyl ester " 190 (25 " ), " = 2555, []/>= 4- 41- 42 2 1493, ^7-5 .0925, " =+25.63 .0524, " = + 21.57 .0319, " = + 21.56 /-CHLORSUCCINIC ACID. By action of nitrosyl chloride on /-asparagin. Melting-point i74 . n Water ^ = 9-3, / = 19, M/' = -19.67. ] = + 4.36. Stated earlier by Becker 1 to be left rotating. ' Walden : Loc. cit. 2 Walden : Loc. cit.; and Ber. d. chem. Ges., 28, 1289. Tilden and Marshall : J. Chem. Soc., 67, 494. * Walden : Ber. d. chem. (ies., 28, 1290. i Walden : IMC. cit. 6 Walden : Ztschr. phys. Chem., 17, 254. ' Walden : Ber. d. chem. Ges., a8, 2769. ' Cook : Ibid... 30, 204. b Ibtd., 14, 1035. ACIDS WITH FIVE ATOMS OF OXYGEN 541 2. Solutions in aqueous acids exhibit right rotation. * Becker investigated the following solutions which contained : I FOR i MOL. OF ASPARTIC ACID. i. 54 to 64 mol. water + mol. HC1 []/>. 2. 302 mol. water + n mol. H 2 SO<. M*. n i 30.0 = 0-5 + 21.8 1-5 > 2 - 6 0.6 + 24.2 2 ^3-4 0-75 -f 28.6 3 -f- 34-0 i + 28.8 \i + 33-9 -34-0 3 5 -f 31.5 -f 32.0 .... 10 33-5 3. Solutions in aqueous alkalies are left rotating. 3 Becker* found for solutions which for one molecule of aspartic acid con- tained : b. 302 mol. water -f- i " " " + 3 14 " " -f 5 " " " + 15-1 " " " 20 - 2 mol. NH 3 : [V]=o 9.2 -- 9-4 " " = 12.1 , CONH 2 .CH 2 .CHNH 2 .CO 2 H. For formation see Bischoff and Walden. ' Water ........... [a] D = -f 5.41 (/-Asparagin = 5.43)- 6 /-^8-AsPARAGiN. Ordinary asparagin. In aqueous and alkaline solutions, left rotating, in acid solu tions right rotating. 7 The following determinations were all made by Becker. 1 a. Solutions in water : p = 0.705 d^ 1 = i.ooio 1.049 1.0025 1.409 1.0043 Pasteur: Ann. chim. phys., [3], 31, 81 ; 34, 30. Ber. d. chem. Ges., 14, 1038. Pasteur : Loc. cit. Ber. d. chem. Ges., 14, 1037. "Handb. d. Stereochemie." 1894, p. 220. Piutti : Ber. d. chem. Ges., 19, 1693. Pasteur : Ann. chim. phys., [3], 31, 75. Ber. d. chem. Ges., 14, 1030. - 5-95 -5-42 5-30 542 CONSTANTS OF ROTATION OF ACTIVE BODIES b. Solutions in dilute acids : TO I MOI,. ASPARAGIN -f- 300 MOI,. WATER Mol. HC1. Ms- Mol. HjS0 4 . Ms- Mol. C 2 H4O 2 . MS. I + 26.4 0-5 + 23.1 I -3.49 1-5 ! + 30.4 o.75 + 27.3 2 -3-10 2 ; + 31-5 i + 29.5 5 -i.45 3 4-31-9 3 -f 32.0 7 -0.59 5 -f-32-3 5 + 34.3 10 0.00 10 + 33-3 10 + 35-5 15 + I. II 15 + 33-7 .. 20 + 2.63 20 4-34-3 .... .... While hydrochloric and sulphuric acids produce strong right rotation, small amounts of acetic acid produce left rota- tion. This last decreases with added acid and is finally changed to right rotation. That the addition of acetic acid under certain conditions of concentration may cause the rota- ting power of asparagin to disappear was noticed by Champion and Pellet. 1 c. Solutions in dilute sodium hydroxide. In 100 parts of solution. Mol. proportions. dy. Ms. Asparagin. NaOH. H 2 0. Asparagin. NaOH. H 2 0. 10 3 87.0 I I 63-8 1.0584 -8.6 4 10 6.1 83.9 I 2 6l-5 1.0915 -6.6 9 10 9-1 80.9 I 3 593 1.1232 -6.35 ACID, C ^ 2 >CH.CH 2 .CH 2 .CO 2 H. Water c = 2, / = 21, [a]/= + 10.2 " <: 4 (supersaturated), / 23, " = + 10.6 Dil. nitric acid c 4, t = 22, " = -f- 29.9 Calcium Salt, CaC 5 H 7 NO 4 . Left rotating. Water c = 5.03, / == 20, [a~\ D = -3-7 Hydrochloride, C 6 H,NO 4 .HCL Water c 4, / = 21, (= + 20.4- 1 Compt rend., 8a, 819. * Scheibler: Ber. d. chem, Ges., 17, 1728. ACIDS WITH FIVE ATOMS OF OXYGEN 543 /-GLUTAMINIC ACID. From the racemic acid by means of Penidllium glaucum. In 100 cc. 9 grams HC1 + 5 grams glutaminic acid, [<*]/> - -31.1.* flT-GLUTAMINE, CQI > CH ' CH r CH 2' CO ' In 100 cc. 0.45 grams H 2 SO 4 -f 5 grams glutamine, [or]> = -f 30.0 2 14 loo " 0.3 " H 2 C 2 4 + 2.7 " " " == -f 18.3 ^-PYROGLUTAMINIC ACID, c ^ >CH.CH 2 .CH 2 .CO. Water ............... c = 2.665, * = 25, [a]/, = -f y 03 l-Acid ................. \a\ D = - 6.1 + SHIKIMIC ACID, C 7 H 10 O 5 . Cyclic. Needles. Melting- point 184. The following observations were made by Eyk- man. 5 Water ...... = 36.26, [or]g - - 204.4-, *= 11.19, [a\%= -187.9 30-73, - 201.5, 7-53 - 186.7 21.71, - 195-5, 5-93 - 184.7 13.12, - 190.0, 4.03 - 183.8 From which is calculated : [a],y = 183.3 - 6 5 c Solutions in acetic acid, and especially in 50 per cent, sul- phuric acid, exhibit increasing rotation ; but tellurous acid de- creases the rotation in marked degree, while selenous acid leaves it unchanged. Ammonium Salt, (NHjC.H 9 O 5 . Rhombic prisms. Water ...................... = 32.00, [<*]/>= -189.7 5.23, 172.1 Triacetylshikimic Add, C.H 7 O 5 (C 2 H S O) 3 . Amorphous. Absolute alcohol ............ c = 5.496, [<*]/> = 170.0 " ............ 3-482, - 169.6 " ............ I-45I, -170.2 Benzene .................... 7.255, 191.1 " .................... 4-230, - 191-7 .................... 2.392, -192.1 Chloroform ................. 3.482, 189.7 ................. 4-222, -189.2 1 Schulze and Bosshard : Ztschr. physiol. Chera., 10, 143. 2 Schulze and Bosshard : Ber. d. chem. Ges., 18, 390. 3 Menozzi, Appiani : Rend. Ace. I,inc., 1893, II, 421. * Menozzi, Appiani : Gaz. chim. ital., aa, II, 105. 5 Ber. d. chem. Ges., 34, 1278. 544 CONSTANTS OF ROTATION OF ACTIVE BODIES Tripropionylshikimic Acid, C 7 H 7 O 5 (C 3 H-O) 3 . Amorphous. Absolute alcohol c= 7.361, []/>= 159.1 " 3-680, -159-0 Benzene 7.125, -172.8 " 5.36o, -173-3 Triisobutyrylshikimic Acid, C 7 H 7 O 5 (C 4 H 7 O) 3 . Amorphous. Absolute alcohol c 9.314, []/> = -146.1 Benzene 7.247, -157-9 Shikimic Acid Bromide, C-H 9 BrO 5 . Hemimorphous, hex- agonal needles. Melting-point 235 (uncorr.). Water c 8, [a]/, = 4 22 Hydroshikimic Add, C 7 H la O :> . Monosymmetric prisms. Melting-point 175 (uncorr.). Water .. p = 16.515, d** = 1.054, / = 23, [or]/, -- 35.8 After dilution with an equal volume of water, [<*]/> - 18.2. Hydroshikimic Acid Dibromide, C 7 H 10 Br 2 O 5 . Rhombic sphenoids. Water c -.= 14.263, t ^ 16, [a] D -- 58 * "-ISOTRIOXYSTEARIC ACID, C 17 H 32 (OH) 3 COOH. Glacial acetic acid- . . . c --= 10 to 15, [a]z> -- -- 6.25 to 6.0 a ii. Acids with Six Atoms of Oxygen For the aldehydes, see group 16 (oxyaldehydes, sugars). ; ARABONIC ACID, C 5 H 10 O 6 . Left rotating. In solution passes gradually into the lactone. On the change in rotation, see 75. Strontium Salt, Sr(C 5 H 9 O 6 ) a + 5H,O. Crystals. Water . . c = 4-353 (at 100), / = 20, [or]z>^ -f- 1.96 3 Anhydride, C 5 H 8 O 5 . Crystals. Melting-point 89. Water... [or]i>= -63.37* Water.. . p = 9.45, d = 1.0316, / 20, [a\ D -- - 73.9* Kykman. Walden : Her. d. chem. Ges., 37, 3471. Allen and Tollens : Ann. Chem. (I^iebig), a6o, 313. Bauer : J. prakt. Chem., [a], 34, 47. Fischer, Piloty : Ber. d. chem. Ges., 34, 4215. Fischer and Piloty. ACIDS WITH vSIX ATOMS OF OXYGEN 545 RIBONIC ACID, C 5 H 10 O, ( . Cadmium Salt, Cd(C 6 H 9 O s ) s (at 100). Fine needles. Water / -20, [a]/, ; 0.6 [ Anhydride, C 5 H 8 O 5 . Crystals. Melting-point 72 to 76. Water c ----9.34, t = 20, []/> = -i8.o 02 The rotation remained twelve hours unchanged. XYLONIC ACID, C 5 H 10 O 8 . See 75. Strontium Salt, Sr(C 5 H 9 O 6 ) 2 + 6H 2 O. Crystals. Water. ... c -^ 4.305, (at 100), t = 20, []/, = + 12.14 :t The rotation remained constant. LYXONIC ACID I^ACTONE, C 5 H 8 O 5 . Water . . p 9,783, d = 1.035, / = 20, []/, = -f 82.4 * GLUCOSACCHARINIC ACID ANHYDRIDE, SACCHARiN,C 6 H 10 O 3 . Made from dextrose by heating with lime. 5 Rhombic prisms. Melting-point 160 to 161. Water.-.. []/>= + 9 3.i 0(i Water..-. / 12.077, = - 17.2, and for the calcium salt [a~\ D = 5.7. MALTOSACCHARINIC ACID ANHYDRIDE, ISOSACCHARIN, C C H 10 O 5 . Monoclinic crystals. Melting-point 95. Water., p 9.471 (air dry), d*' -- 1.0302, t 20, []/)------ 61.88 10 Water., c 10 (air dry), t = 10, [a]/> = -j- 62.97 ll Shows no multirotation. 1 Fischer and Piloty. - Fischer and Piloty : Loc. cit. Allen and Tollens : Loc. cit. 4 Fischer, Bromberg : Ber. d. chein. r.es.. 29, 583. Kiliani : Ibid., 15, 2954. 6 Pel i got : Conipt. rend., 90, 1141. ' Scheibler : Ber. d. chera. Ges., 13, 2216. - Herrmann, Tollens : Ibid., 18, 1333. ' Schnelle and Tollens : Ann. Chem. (I.iebig), 271, 66. 111 Wehmer. Tollens : Ib id., 243, 323 " Schnelle and Tollens. 35 546 CONSTANTS OF ROTATION OF ACTIVE BODIES METASACCHARINIC ACID ANHYDRIDE, METASACCHARIN. C S H, O 5 . Rhombic crystals. Melting-point 141 to 142. Water p 7.846, d - 1.026, / 14, []/> - 48.4 ' 7.o, " ~ 46.96 '-' Shows no multirotation. The parasaccharinic acid of Kiliani 3 is left rotating. RHAMNONIC ACID (Isodulcitonic acid), C 6 H 12 O K . See $75. ISORHAMNONIC ACID LACTONE, C B H 10 O-. Melting-point 150 to 152. Water p 8.903, d 1.032, []-;; 62.02 4 After twenty-four hours by change into the acid, [<*] 5.21. DIGITALONIC ACID, C n H 14 O (i . From the mixture of sugars which result from the resolution of pure digitalin. Anhydride, rhombic crystals. Melting-point 138 to 139. Water p 3.3327, d 1.0084, / 28, [a]/, - 79.4 * Tartaric Acids, CO.H.CHOH.CHOH.CO.H. = + 14.154 0.1644 p p 20 to 50) - 2.286 J 0.1644 q ^ = soto8o) 14.615 0.1588 p p = 20 to 50 ) 1.265 0.1588^ ^ = 50 to 80) 15.050 - 0.1535 P P = 20 tO 50 - 0.300 -f o. 1535 q q = 50 to 80 -f 15.429 0.1480 / /= 20 tO 50) 0.629 4- o. 1480 q q = 50 to 80 \ + 15.784 0.1429 p p = 20 to 50) 1.494 0.1429^ ^ = 50 to 80 j From these constants we have the general formula: []/> = ( l 3-9 6 + o.ii39/ 0.00081 f 2 ) (0.1756 0.001135 t)P With reference to the concentration c we have [or]/) == 13.436 0.1187 c ; for c =- 22 to 63 and / = 20 The value of the specific rotation of tartaric acid at a number of different temperatures and percentage strengths according to the Thomsen formula, is shown in the following table. /= 10 15 20 25 30 /> 5= +5-93 6.69 4-7-38 ^-8.08 +8.64 40 7.58 8.28 8.91 9.55 10.07 30 9.22 9.86 10.45 11.02 11.50 20 10.87 11.45 11.98 12.49 I2 -93 On the variation of the specific rotation of aqueous solutions of tartaric acid with the temperature, see 60. 1 Ann. Chem. (I^iebig), 176, 120. - Ber. d. chem. Ges., 6, 1073. a Math, naturw. Ahh. d. Berl. Akad., 1887, p. 248. J. prakt. Chem., [2]. 32, 213. 548 CONSTANTS OF ROTATION OF ACTIVE BODIES Salts of Dextrotartaric Add On the rotation of tartrates in dilute solutions, see 61. Potassium Add Tartrate, KC 4 H 5 O, ; . = 0.615, t=^ 20, [a]/, 22.61 ' In the following observations by Sonnenthal,' T is the time between the preparation of the solution and the observation. p. d-. T. [*}* 0.4116 1.0044 At once and after 48 hours 22.119 0.3001 1.0041 At once 20. 192 0.2407 I.OOII At once 22.241 t < " 48 hours 22.379 (( it 80 hours 23.025 120 hours 23.025 Sodium Add Tartrate, NaC,H 5 O 6 -f H 2 O. i. According to Thomsen 3 the specific rotation decreases with decreasing concentration, but increases slightly with the temperature. Hydrate. L"J r> . L"J/> I L"J /> L J '/' p. c. 12.70 13.50 .... .... .... 22.47 10. 16 10.65 .... .... 22.12 22.19 8.89 9.26 21.84 21.85 22.07 7.62 7.89 21.85 21.88 22. IO 22.29 6-35 6.54 21.56 21.84 21.77 21.88 2. c== 4.409, [or] 2 /,' == 23.95 anhydrous (21.67 hydrate)/ Lithium Add Tartrate, LiC,H,(),. + H 2 O. Anhydrous / 7.998. [= 28.48 (hydrated 27.39). 3 3. Experiments by Schiitt 4 gave : Hydrated salt. Anhydrous salt. ^ = 40 [] ~ = 28.46 c = 38.47 [ aft = 29-59 30 28.08 28.85 29.20 22 27.76 21. l6 28.87 20 27.51 19.23 28.61 10 26.94 9.62 28.01 Tliis formula follows for the anhydrous salt : [<*]*" = 27.14 -f 0.0992 c 0.000938 c* On the effect of addition of KC1 and NaCl on the rotation of potassium tartrate, 70, p. 245. 4. Th. Thomsen 5 found these values referred to the anhy- drous salt : p. c. M* Ma Ms- 54-54 79.24 30.70 30.67 30.57 36.39 46.55 30.07 30.06 30.01 18.09 20.38 29.02 29.19 29.26 9.07 9.62 28.34 28.49 28.65 From these observations we have the following interpolation formulas : [or] % = 27.56 + 0.0925 p 0.00065 p [a]^ = 27.62 + 0.1064^ o. 00108 f- [flf] a j = 27.86 -f 0.0951 p o. 00099 /> 1 Long : Am. J. Sci., [3], 38, 267. 2 Arch. Nerl., VII, 1872. 3 I^andolt. 4 Ber. d. chem. Ges., ai, 2586. ^ J. prakt. Chem., [2], 34, 89. 550 CONSTANTS OF ROTATION OK ACTIVE BODIES Sodium lartrate, Na 2 C 4 H 4 O 6 + 2H.O. 1. Hydrated c == 5 to 15, /= 22.5, \ot] D = 27.85 - 0.17 c l [a]c []/> []/ [or]* | 2. Hydrated^^ 20, /= 25. 20.82 25.79 31.67 32.70 38.49" 3. Anhydrous c= 9. 946, [or] = 30. 85 (hydrated = 26.O2). 3 4. Th. Thomsen* investigated the following solutions: HYDRATED SAI/T. p- c. W3- \\- M* [ 0.00061 7 p- r1" 26 -3 0.02020 ^ 0.000963 /^ []^f 26.65 "" - 03686 p 0.000693 p' 1 Potassium Sodium Tartrate, KNaC 4 H 4 O 8 + 4H,O (Rochelle salt). i. Th. Thomsen 5 has experimented with solutions made by saturating sodium acid tartrate with i mol. KOH and addition of V,oo mol. NaOH : KNaC 4 H 4 6 . p. c. 33-68 25.27 16.84 8.41 42.35 29.99 18.86 8.89 29.31 29-46 29-59 29.47 29-33 29.51 29.77 29.52 29.41 29-55 29.80 29.46 I 1 Hesse: Ann. Chem. (Mebig), 176, 122. ! Krecke : Arch. Neerl., VII, 1872. * I^andolt. J.prakt. Chem., [2] 34.79- * Ibid., [2], 34, 90. TARTARIC ACID 551 From this, calculated by Schiitt, [>] = 29.73 0.0078 c for c =- 8 to 43. 2. J. H. Long. 1 5 22.14 29.73 15 22.16 29.76 25 22.12 29.70 35 22.13 29.72 45 22.06 29.62 [a]* 27.67 32.08 hydrated ^ anhydrous From this, calculated by Schiitt, [a] = 29.77 0.0026 c (anhydrous) for c = 5 to 45. 3. Krecke 2 determined the specific rotation of a solution with c = 20 for different rays ; / = 25. []c [>]/> WE Hydrated c 20 18.52 22.42 26.49 4. Anhydrous c= 10.77, [] 2 ^ = 29.67. Lithium Tartrate, Li 2 C 4 H 4 O 6 . Anhydrous c= 8.305, [or] 3 ='35.84.' Ammonium Tartrate, (NH 4 ) t C 4 H 4 O t . 1. r =9.433, []Sf = 34.26.' 2. Krecke 4 found for different rays : Anhydrous ........ []? []-5 [^pj ^=20 ............ 31.08 37.09 43.05 Potassium Ammonium Tartrate, KNH 4 C 4 H 4 O 6 . c= 10.515, [ci]^ = -f 31- " 3 Addition of NH 4 C1 and of NaCl decreases the rotation, while that of KC1 increases it, but apparently irregularly and notjin proportion to the amount of salt added. 5 * = 2o, = 30.85. 45-27 53-76 In 100 cc. along with 20 grams of tartrate. C]s. A. "\ crams NH Cl . 70 66 c grams KC1 10 80 0.25 + O OA 10 " " o u -y o 1 ^^ 2Q 06 tag 10 " " ^9-9 /jQ rr u.oy 20.51 2.34 i Am. J. Sci., 36, 353- - Arch. Nerl., VII. J I^andolt. 4 Arch. Nerl., VII. Am. J. Sci., [3], 40, 282. 552 CONSTANTS OF ROTATION OF ACTIVE BODIES Sodium Ammonium Tartrate, NaNH 4 C t H 4 O,. 4- 4H,O. Rhombic columns. <- = 9.690 (anhydrous), []^ 32.65 ! Magnesium Jartrate, MgC 4 H 4 O 6 . * = 8.8i8, []-;; 35.86 ' Borohydrotartratc , Boryltartrate, Borotartaric Acid, BO.H.C 4 H 4 6 See 70. Potassium Borotartrate, KBOC 4 H 4 O 6 . Obtained in aqueous solution by addition of i mol. of boric acid to i mol. of potassium acid tartrate. 1. c== 2.744, [] = 5I-48) 1 r= 5.488, [*] = 58.35) 2. c = 5 (dried at 100) ' / = 20, [a]/, = 58.10' f== 20 ( " " ";.) " ;"-.- 7=5 68.29 f == 10 (dried over H 8 SO 4 ) - 59.06 r = 10 ( " " 4< ) /= 29 " -57-29 From the observations of Long there follows according to Schiitt : [a] D = 50.67 -f 1.688 t: 0.04036 <*, for r = 5 to 20 Addition of alkali salts (especially potassium acetate) in- creases the specific rotation. 3 Sodium Borotartrate, NaBOC 4 H 4 O 6 . From equal molecules of H S BO S and NaHC 4 H 4 O 6 in aqueous solution : c= 2.538, []> ^ 55.02 1 ' c=- 5.075, -63.48 c= 10.151, = 71.47 j Arsenyl Tartaric Acid, AsOC 4 H 5 O 6 . Made in solution by heating two molecules of tartaric acid with one molecule of arsenous oxide. c= 12.304, [a] = 16.91 ' 1 J.andolt. 1 Ix>ng : Am. J. Sci., 38, 264. a Ix>ng: Ibid.. [3], 38, 271. TARTARIC ACID 553 Potassium Arsenyl Tartrate, KAsOC 4 H 4 O 6 . Made by heating two molecules of cream of tartar with one molecule of arsenous oxide to complete solution. r = 0.563, [] = 21.13 * Sodium Arsenyl Tartrate, NaAsOC 4 H 4 O H . Made in solution from As 2 O 3 and NaHC 4 H 4 O 6 . * = 3-358, O] = 20.64 T Potassium Antimonyl Tartrate, Tartar Emetic, 2(KSbO C 4 H 4 6 ) + H 2 0. 1. c = 5t t=2$ [>]<; [a],, (XU M* [ng : Sill. Am. J. Sci., [3], 38, 264. /Aid., [3]. 38, 264 ; 40, 275. See 70, p. 245. 6 lyong : Am. J. of Sci., [3] 38, 266. 6 From this according to Schiitt : Ot\ = 1 1.672 0.3788 c 0.01025 c-, c 5 to 20. " Z,ong : Loc. cit. 554 CONSTANTS OF ROTATION OF ACTIVE BODIES Thallium Sodium Tartrate, TlNaC 4 H 4 O 6 + 4H 2 O. r 5 (hydrated) / = 20 [<*]/?-- 9.07 C 20 " /= 20 " = 6.49 c = 10 (anhydrous) / = 20 " = 8.60 r = 10 " t 28 " = 9.49 1 Sodium sulphate increases and thallium sulphate diminishes the specific rotation. Thallium Ammonium Tartrate, T1NH 4 C 4 H 4 O 6 . \a]% = 7.S6 l Increase of temperature and addition of potassium salts in- crease the specific rotation. Thallium Lithium Tartrate, TlLiC 4 H 4 O 6 + H 2 O. = 5 (hydrated) [a] ^9.46 =20 " " ^6.69' Lithium salts increase and thallium sulphate diminishes the specific rotation. Thallium Antimonyl Tartrate, TlSbOC 4 H 4 O 6 -f H 2 O. = 2 /=20 []/>= 100.44 = 2 t = 28 = 99.64 T Acetates produce a decrease in the specific rotation. Ethylene Diamine Ditartrate, C 2 H 8 N r 2C 4 H 6 O 6 . Water ........... ^ = 0.36 " - 1 70.83 2 On the rotation of certain tartrates in glycerol solution, see experiments by Long. 3 rf-Tartaric Acid Esters Monomethyl Tartrate (Methyl Tartaric Add}, HCH 3 C 4 H 4 O 6 . Sirup. Water ............... =2.073 [r]z> 18.1 Alcohol ............. ^=1.037 " 3,22 The salts of methyl tartaric acid are crystalline. : Loc cit. - Colson : Compt. rend., 115, 729. '* ). Am. Cheir. Soc., 33, 813. TARTARIC ACID ESTERS 555 Anhv- [#]/> f. /. ar/?. 1 drous. . . Water 2 IA8 26 ; Alcohol o 1612 O T>4 O J-/ O<* VJTT 4 t 4 i _ OflT 28 o 4 4 V * , * * * * K C C*7 22 7 " o 0088 S- j ^^ *OO / ** / Na " .. (1 2 151 21 O fi rim n o/l Monoethyl Tartrate (Ethyl Tartaric Add}, HC,H 5 C 4 H 4 O. Sirup. Water.. Alcohol c = 2.252 [a} D = 21.8 r 1.126 ' rr= 7.10 Tartaric Acid Salts. Crystalline. 1 c. Anhy- drous. []/>. c. /. or/). Li salt . Water 2 72Q 28 8 Alrohol n 87 Nfl " " *$&} 0.7052 K " " 'OO 1 27 1O '/J 21 6 4 , Ca 4> . " 2 /1CX3 24 ^ B a " 4 -^fV VJ 3IO7 2O "? 1 . AVJ/ "-' J Dimethyl Tartrate, (CH 3 ),C 4 H 4 O 6 . Crystals. Melting- point 48; boiling-point 158.5 (12 mm.), 280 (760 mm.)." Liquid, d* = 1.3403 []^ = 1.83 3 rf 20 =1.3284 [o']^ =2.14* f/ 100 = 1.2500 [o:]^ =6.00 Diethyl Tartrate, (C 2 H 3 ) 2 C 4 H 4 O 6 . 280 (760 mm.). Liquid, boiling-point == 7.47* 1 Fayollat : Compt. rend., 117, 630. '-' Anschiitz : Ber. d. chetn., Ges., 18, 1399. ' Anschutz, Pictet : Ibid., 13, 1117, 1538. 4 Pictet : Arch. sc. phys. nat. [3], 7, 82; Jahresbericht, 1882, p. 856. 5 Anschiitz. Pictet : l*oc. cit. e Pictet : Jsb. Chem.. 1882, p. 856, 556 CONSTANTS OF ROTATION OF ACTIVE BODIES in-n-propyl Tartrate, (C 3 H.) 2 C 4 H 4 O 6 . Liquid, boiling-point 303 (760 mm.). ^^.1300 [] = 14.89 d = 1.0537 [a] = 18.82 * Diisobutyl Tartrate, [(CH 3 ) 2 CH.CHJ 2 C 4 H 4 O 6 . Solid, melt- ing-point 68, boiling-point 323 to 325 (760 mm.). arioo . I-OI45 [ a ]^ = 19.87 2 Potassium Ethyl Tartrate, KC 2 H 5 C 4 H 4 O 6 - Barium Ethyl Tartrate, Ba(C 2 H 5 .C 4 H 4 O 6 ) 2 . Water ..... =12.586 [] 25.68 3 d-Diacetyl Tartaric Acid Compounds d-Diacetyl Tartaric Acid, COOH.CH(OC 2 H 3 O).CH(OC 2 H 3 O).COOH. (Also with 3H 2 O. ) Left rotating in water, methyl alcohol and ethyl alcohol (Pictet), also in ether and benzene. 4 Water c= 17-947 U-357 11.486 9.189 7.351 4.705 3-7&4 " [**] = 23.04 22.4822.1621.5021.3320.07 19.32 Alcohol .................... c= 7.367 4.911 3.274 11 .................... []"= ~ 2 3- 6 3 -23.14 -21.52 Methyl alcohol (d = 0.824) .. ^=4.681 []^ =- 23.74 5 The sodium and barium salts are also left rotating. 6 d-Diacetyl Tartaric Acid Anhydride, (C 2 H 3 O.O)CH.CO) a O. Right rotating. Prismatic crystals. Melting-point 125 to 129. Anschiitz, Pictet : Loc. cit. Pictet : Loc. cit. I^andolt. Colson : Compt. rend., 114, 175. Pictet : Jsb. Chem., 1882. p. 856, 857. Anschiitz, Pictet : Ber d. chem. Ges., 13, 1178. TARTARIC ACID ESTERS 557 In benzene. In acetone. f 2.091 1.045 11.656 4.403 [a]/, 58.69 63.05 59-7o 62.04 ! The esters of diacetyl tartaric acid may be distilled at the ordinary temperature without suffering a change in their rota- ting power." Dimethyl Ester, (CH 3 ) 2 (C 2 H 3 O),C 4 H A- Rhomboidal prisms. Melting-point 103. According to Pictet* left rotating. Alcohol (d = 0.826) = 3.566 [a~\$= - 14-23 ' ^ = 3-254 [<*] = -14-29 Diethyl Ester, (C 2 H 5 ) 2 (C 2 H 3 O) 2 C 4 H,O (i - Triclinic prisms. Melting-point 66.5, boiling-point 291 to 292. [of] y) for the superfused ester = : -f 5.o. 4 Alcohol (d = 0.826) c = 23.644 []$ = + 1-02 ' On the rotation in chloroform solution see 57. N-Dipropyl Ester, (C 3 H.) 2 (C 2 H 3 O),.C 4 H,O 6 . Crystals, melt- ing-point 31, boiling-point 313. Alcohol (d = 0.826) c= 7-855 [a]# = + 7-04 * c = 3-253 [or]]!? = - 6.52 According to Freundler, 5 the ester is liquid at the ordinary temperature, [or]/, = 13.5. Its specific rotation changes greatly with solution in different liquids. See 59. N-Dibutyl Ester, (C 4 H 9 ) 2 (C 2 H 3 O) 2 C 4 H 2 O 6 . Liquid. f = 2o, lai] D =: + 17.8. 4 Diisobutyl Ester, (C 4 H 9 ) 2 (C 2 H 3 O) 2 .C 4 H 2 O fi . Liquid. Boiling- point 322 to 326. / = about 20, [<*] D = + 11.3. 4 Alcohol (d = 0.826) .... c -= 13-559 [0]% = + 10.51 ! c= 7-953 |>]'J = ~ IO - 2 9 Compounds of diacetyl tartaric acid with ethylene diamiue. Neutral Salt, C 2 H S N 2 .2C 8 H 10 O 8 . Crystals. Water ^=11.5 \CC\D --12.74 1 Pictet. 2 Freundler; Compt. rend., 115, 509. 3 Loc. cit. 4 Freundler. 5 Compt. rend.. 117, 556. on : [a] % = - 110.91 []/?= -112.05 [#] = - 1 16.30 558 CONSTANTS OF ROTATION OF ACTIVE BODIES Add Salt, C,H 6 N.,.C 8 H 10 O 8 . Crystals. Water .......... r^u.5 []/> = -17.05* Numerous further determinations of rotation of esters of dipropionyl, dibutyryl, di-w-valeryl and di-rc-caproyl tartaric acid have been made by Freundler. 2 For the rotation of the esters of di-monochloracetyl tartaric acid, see Franklin and Turnbull. 3 See McCrae and Patterson' for other derivatives of diacetyl tartaric acid. Compounds ofd-Dibenzoyl Tartaric Acid. Dibenzoyl Tartaric Acid, COOH.CH.O(C T H 5 O) CH.O(C T H 5 O).COOH -f H a O. Crystals. Melting-point 90. With Water of Crystallization : Alcohol (d 0.818) ......... ......... c - -4-755 [or] J? -117.68 Methyl alcohol ............. --4.625 []$-- -122.14 Dibenzoyl Tartaric Acid Anhydride, (C 7 H 5 O.O.CH.CO) 2 O. Crystals. Melting-point 174. Acetone .......... c = 4.644 [#] Jf = H- 142.94 .......... c 1.572 [] * -f 143-22 G []/ IJ 4 7 Dibenzoyl Tartaric Acid Dimethyl Ester, (CH 3 ) 2 (C 7 H 5 O) 2 - C 4 H,O fi . Crystals. Melting-point 132. Alcohol .......... c 0.245 \_df\ 7,' 96.61 Chloroform ...... c 11.612 [a]J_? 88.24 ...... C*= 8.598 [a]^= - 88.78 6 Dibenzoyl Tartaric Acid Diethyl Ester, (C H 5 ).XC.H 5 O) 2 C 4 H 2 6 . Liquid. 1 Colson : Compt. rend., 115, 729. -' Compt. rend., 115, 509, 556; Bull. soc. chim., [3], 9, 680; n, 366, 470; 13, 1055. Ann. Chim. Phys.. [7], 3, 433 ; 4, 244. * ]. Chem. Soc., 73, 203 (1899). . 4 Ibid., 77, 1096 (1900). I'ictet : Jsh. Chem., 1882, p. 857. Pictet. ' Freundler : Hull. soc. chim.. [3] 7, 804. TARTARIC ACID ESTERS 559 Alcohol (d 0.815) C 9- J 75 = 5-733 c = 2.693 Dibenzoyl Tar taric Acid Diisobutyl Ester, (CJIJ.XCH.O).,. C 4 H.,O K . Liquid. Alcohol (d r.= 0.818) c = 14.085 [#] = - 48.86 *== 4.922 [or];?- -42.94 <:== 2.880 [a] jj = -41-95 ! Diphenyl Acetyl Tartaric Acid Anhydride, (C 7 H..CO.,.CH. C0) 2 0. []== + 58.-' Diphenyl Propionyl Tartaric Add Anhydride, (C,H f( .CO 9 .CH. C0) 2 0. ' []* = + 38V Frankland and Wharton 4 and Frankland and McCrae 5 have given the following determinations : Kb. Dicthvl Tnonofocnzovl tHFtratc -7/1 2 4 20.71 38 4- 20.18 63 4- 19.02 79 4- 18.43 99-5 -f 17.69 135.0 + 16.36 - IOO 137 - 66.84 183 - 58.94 Diethyl dibenzoyl tartratc T Q _ ,* 10 59-36 38 - 61.70 44 - 62.05 53-5 - 62.28 60 - 62.28 IOO - 60.77 ^T/* f^Q I 4 12. OO . 20 + 11.82 32.5 11.74 65 11.20 IOO -f- 10.88 136.5 4- 10.62 ' Pictet. - Freundler : Bull. soc. chim., [3], 7, 804. 3 Freundler. * J. Chem. Soc., 69, 1309, 1583 (1896). 5 Ibid., 73, 307 (1898). 560 CONSTANTS OF ROTATION OF ACTIVE BODIES t. M . 1 14.5 20 54 100 136 100 109 138 180 12 19 33-5 54-5 70 100 136 100 136 183 100 135.5 183 II 30 49 70 100 135 20.5 24.5 44-5 50 100 136 100 137 183.5 + 13-63 -f 13-59 + 13.28 + 12.57 -f 11.92 + 15.85 4- 15-44 + H-59 + I3-38 - 77 82 //.CM 78.42 - 77.00 - 74-23 - 72.02 - 68.10 - 61.28 - 79.02 - 70.58 > 60.96 IO2 82 76.90 60.37 - 60.33 - 59-53 - 57-96 - 54-73 - 50.37 9-3 I - 68.87 69.16 69.00 - 63.74 - 58.71 - 89.98 - 81.46 - 69.50 The observations were made in tubes 44 to 50 mm. in length, and specific gravity determinations were made over a suffi- ciently wide range of temperature to allow the calculation of the specific rotation. TARTAKIC ACID ESTERS 561 I m ides of d- Tartaric Add and Benzoyl Tartaric Acid The following data are from Ladenburg. 1 Methyl Tartrimide. Made by heating methylamine bitar- trate, by which partial racemization followed. Melting-point 178. CO Water, / 7.31 = 196.30 0.2877 p for / 7 to 13. By heating methyl tartrimide with 2 mols. of benzoyl chlo- ride there result : a-Dibenzoyl Methyl Tartrimide. Melting-point 56. Ethyl acetate ................ p = 7.93 [a]/>= + 183.9 " ................ />- 15-83 - + 185.7 ft-Dibenzoyl Methyl Tartrimide. Melting-point 106 to 108. Ethyl acetate ................ p 7.93 \a\ D = + 188.8 " ................ />:-- 15.84 "==+189.8 Ethyl Tartrimide, C 4 H 4 O 4 .NC.,H 5 . Melting-point 171 to 174. Water ....................... p 5.76 []/>= + 164.9 " ....................... p -7-32 " =-+165.6 >( ....................... /> 8.57 =.f 166.2 Changes in the Specific Rotation of d- Tartaric Acid in Presence of Inactive Substances If different bodies are added to aqueous solutions of tartaric acid, the degree of its electrolytic dissociation, as explained in 61, is altered, and as the latter is reduced, the specific rota- tion is lowered. This follows on addition of acids, alkalies, alcohols and other bodies and the decrease in the right rotation, so caused, may extend in certain cases to inactivity, or even to a change to left rotation. When an increase in activity is ob- served, as by addition of boric acid, molybdic acid, or alkalies, this depends on the formation of complex compounds of tar- taric acid, as referred to also in 61. 1 Bcr. d. chem. Ges., 29, 2710. 36 562 CONSTANTS OF ROTATION OF ACTIVE BODIES Of the many observations bearing on this, a number have been referred to in the general part of the work (see 59, 61, 70) ; in regard to others, it is sufficient here to refer to the original papers. Most of these have a limited value only, be- cause they are based on determinations with but few concen- trations, and the corresponding data concerning the degree of dissociation are lacking. Inorganic and Organic Acids (see6i). These produce a decrease in the right rotation. 1 Amido-acetic acid and amido- propionic acid increase the activity. Alkalies bring about a decrease in the right rotation of alkali tartrates, which may extend to left rotation.- Alkali Salts produce sometimes an increase, sometimes a de- crease in the rotation of the tartrates. 70.' Molybdates and Tung states. Increase in rotation to a maxi- mum point. 70. Alcohols. Decrease in rotation . 4 Acetone. Decrease. 5 Benzene and Homologues, mixed with alcohol, produce left rotation. 59." Organic Haloid and Nitro Compounds. Decrease in the right rotation or change to left rotation. 5Q. 7 Inactive Organic Bases (aniline, pyridine). The right rota- tion increases to a maximum and then decreases. 8 Amido compounds (urea, glycocoll, alanin) produce increase in the rotation. Urea, 9 glycocoll, alanin. 10 /-TARTARIC ACID. i. Water p = 35.7, t = 17 ; Biot'sred ray[] r = -8.43. 1 Biot : Mem. del'Acad., 16, 229. l,andolt : Ber. d. chem. Ges., 13,2331. Th. Thom- cn : J. prakt. Chem., [2], 32, 219; Pribram : Sitzber. Wien. Akad., 97, II, 13. s Th. Thorosen : J. prakt. Chem., [2]. 35, 145. Aignan : Compt. rend., 112, 1009. F. Schutt : Ber. d. chem. Ges.. ai. 2586 (KC1 and NaCl). Long : Sill. Am. J. Sci., [3]. 36, 351 ; 38, 264 ; 40, 275. Th. Thomsen : J. prakt. Chem., [2], 34, 83. 4 Biot : Mem. de 1'Acad., 15, 240. Iandolt : Ber. d. chem. Ges., 13, 2332. Pribram: Sitzber. Wien. Akad. 97, II, 468. Landolt : Ber. d. chem. Ges., 13, 2332. Pribram : Sitzber. Wien. Akad., 97, II, 463. Pribram : Ber. d. chem. Ges., aa, 6. ' Pribram: Ibid., aa, 7. Pribram : Ibid., aa, 9. ' Pribram : Ibid., aa, 8. 10 Pribrnni : Sitzber. Wien. Akad., 97, II, 479. TARTARIC ACID ESTERS 563 For right tartaric acid under the same conditions, [a~\ r = + 8-58 2. The review of Pasteur's 1 observations by a committee con- sisting of Biot, Dumas, Regnault, and Balard furnished the following values :' A rff. /. < a r . [<*] r . Levotartaric acid . . Dextrotartaric acid 42.06 41-97 1.21785 5.198 1.21765 5.198 20.5 20.5 -21.485 + 21.452 > i 8.070 + 8.082 Addition of boric acid increases the left rotation in the same degree that it increases the right rotation of dextrotartaric acid. The committee named cite the following parallel experi- ments : 3 Tartaric acid. Boric aci, Water. rf /. /. a r . Wr* Levotar- taric acid 23.89 4.76 71.35 1.13181 5.198 23.2 52.12 - 37.08 Dextro- tartaric acid..., 23.78 4.80 ; 71.42 1.13158 5-1935 23-2 -f 53-07 + 37-97 Salts of I- Tartaric Acid l-Ammonium Tartrate, (NH 4 ) 2 C 4 H 4 O 6 . Water p= 12.16, /= 18.2, [a]j = 38.20* from which by multiplication with --we have [oi] r = 29.29. For the ^/-salt Biot found [<*] r = -f 29.0. /-Sodium Ammonium Tartrate, NaNH 4 C 4 H 4 O 6 -f 4H 2 O. Water/ =33.33, /= 16.5, [flf] y = 26.0 * The solubility is the same as with the d- salt. Levotartar Emetic, KSbOC 4 H 4 O 6 + ^H 2 O. Water p= 6.80, / = 19, []y = 156.2. Asimilarsolu- tion of the af-salt gave [a] y = + 156. 2. 6 Ann. chim. phys., [3] 28, 77. Ibid., [3], 28, 101 to 105. Ibid., [3], 28, no to 112 Pasteur: /bid., [$, 28, 84. Pasteur : Ibid., [3], 28, 90. Pasteur : Ibid., [3], 28, 87. 564 CONSTANTS OF ROTATION OF ACTIVE BODIES /-Calcium Tartrate, CaC.H.O, + 4H 2 O, dissolved in hydro- chloric acid shows right rotation. A solution of 20 grams of the salt in 63 cc. of hydrochloric acid (containing 7.09 grams HC1) showed, for /= 3.9 dm., or, = = -f- 6.7. If ^/-calcium tartrate is dissolved in hydrochloric acid the solution shows left rotation. 1 Combinations of d- and I- Tartrafes with d- and l-Malates Right ammonium acid tartrate and left ammonium malate form a crystallizable combination which, when dissolved in aqueous ammonia, shows the same rotation as a mixture of equal molecules of the two salts. < 4, [},. = + i33-9- I- Tartramide, \oi\j - - 134. 15, for c = = 1.305. l-Malamide, []_,- = 47-5. d- Tartramide and I- Mai amide > dissolved in equal molecular proportions, yield an easily crystallizable compound : d- Tartro-l- malamide, \a\ j -f- 43 . 02 . I- Tartramide and I- Malamide, dissolved in equal molecular proportions, yield also a crystallizable compound which is more soluble than the last : /- Ta rtro-l- mala m ide , [a]j = - 9 5 . 7 1 . :t ]B= -44-09 -43-84 -43-75 5 2. Water, c == 8.9 to 53.03, [>]', = - 43.8 to 43.9." 3. Water, c = = 1.57 to 12.71, [ar]* = - 45.5 to 45.7. 7 Pasteur : Ann. chim. phys., [3], a8, 78. Pasteur: Ibid., [3], 38, 464. Pasteur : Lac. /., p. 465. Pasteur : Loc. cii., p. 437. Hesse: Ann. Chem. (I.iehig), 176, 124. Eykman : Ber. d. chetn. Ges., 34, 1297. Oudemans: Rec. trav. chim. Pays.-Bas., 4, 166. ACIDS WITH SEVEN ATOMS OF OXYGEN 565 4. Water,/ 9.931029.50, \_a]= -43-47 - 0.0230^. ' Quinates. The rotation decreases somewhat with increasing dilution, and finally reaches nearly the same value for all. 2 See $6 1. Addition of free alkali produces an increase in the specific rotation in consequence of a decrease in the extent of dissociation/ 1 12. Acids with Seven Atoms of Oxygen See the aldehydes under group 16 (oxyaldehydes, sugars). = -f- 7.92 Water c - i hour 5.94 to 4.8 Exhibits multirotation. ' Ca(C 6 H n O 7 ) 2 + 2H..O (from water). Water c - 10 (anhydrous), [<*]}? 6.66 K ' c 10 " " 7 ' Shows no multirotation/ Anhydride, C 6 H 10 O 6 . Crystals. Melting-point 130 to 135. Water p 8.32, d v> = 1.032, / 20, [<*]/> = + 68.2 At the end of twenty-four hours the rotation had fallen to 64.2, but the solution had an acid reaction, evidently because free acid had been formed." i Thomstn : J. prakt. Chem., [2], 35, 156. - Oudemans : Loc. cit. ' Thonisen : IMC. cit. 4 Fischer : Ber. d. chem. Ges., 23, 2625. " Herzfeld : Ann. Chem. (L,iebig), 220, 345. 1-i-cher : Ber. d. chem. Ges., 23, 2614. 7 Schnelle, Tollens. - Fischer. - her : Ber. d. chem. Ges.. 23, 2626. 566 CONSTANTS OF ROTATION OF ACTIVE BODIES /-GLUCONIC ACID. The mixture of acid and lactone rotates- strongly to the left. 1 Calcium Salt, Ca(C 6 H H O 7 ), (dried over H 2 SO 4 ). Needles. Water p= 10.298, rfj= 1.049, t=2o, []/?= - 6.64 - Anhydride. The rotation was determined in a hydrochloric acid solution of calcium gluconate. c = 6.9 (anhydride), t = 20, []/>, = 22.0 T ]/>--= -f- 53.81 :; /-MANNONIC ACID (Arabinose carboxylic acid). Anhydride, C 6 H 10 O 6 . From arabinose and hydrocyanic acid. Rhombic crystals which soften between 145 and 150. Water p 9.1807, rf= 1.0329, []/>= 54.8 * Nic ACID, C 6 H 12 O 7 . Anhydride, C 6 H 10 Q 6 . From ^-saccharic acid by reduction. Trimetric crystals. Melting-point 178 to 180. Water. . . . p = 10.219, d 1.0373, t = 20, [a]/, == -f 55.1 5 From glucoronic acid : Water c= 2.157, t = 19, []--=+ 56.1 Calcium salt [a]z> = 14.45 /-GuLONic ACID (Xylose carboxylic acid). Anhydride, C 6 H 10 O 6 . From xylose and hydrocyanic acid. Trimetric prisms. Melting-point 185 (cor.). Water. . . . p = 9.15, d 1.034, / = 20, [a]/, = - 55.3 ' ^-GALACTONIC ACID, C 6 H 12 O 7 . On the multirotation of the free acid, separated from the cal- cium salt by aid of hydrochloric acid, through change into the lactone, see 75. Fischer. Fischer : Her. d. chem. Ges., 33, 2614. Fischer, Hirschberger Her. d. chem. Ges., 22, 3218. Kiliani : Ibid., 19, 3034. Fischer, Piloty : Ibid., 34, 521. Thierfelder : Ztschr. physiol. Chem., 15, 71. Fischer, Piloty. ACIDS WITH SEVEN ATOMS OF OXYGEN 567 Calcium Salt, Ca(C ti H n O 7 ), -f 5H..O. Monoclinic crystals. Water -j- HC1, c = 0.76, / = 15, [a] /> = about + 2.85 But if the calcium salt is decomposed by oxalic acid, two crystalline substances result : i, C 6 H 12 O 7 (=C 6 H 10 O 6 + H 2 O), with melting-point 65, and 2, C 6 H 10 O 6 , with melting-point 90 to 92. The first is not the true galactonic acid, but the hy- drate of the anhydride. 1 TALONIC ACID, C ti H 12 O 7 . Made by heating ^/-galactonic acid, pyridine and water to 150. The mixture of acid and anhydride is strongly left rotating. 2 = 3.25 * RHAMNOHEXONIC ACID (Isodulcite carboxylic acid), CH 3 . (CH.OH),.COOH =zC 7 H 14 O 7 - Anhydride, C 7 H 12 O B . Crystals. Melting-point 168. Water ^=10.034, t = 20, [a~\ D - - 83.8 Shows no multirotation. 5 /-TRIOXYGLUTARIC ACID, C.H S O 7 . By oxidation of arabinose with nitric acid. Microscopic plates. Melting-point 127. Water p ^=9.59, rf = 1.0441, t = 20, [a - ]/> = 22.7 The rotation remained unchanged after twenty-four hours. 6 Potassium Salt, K 2 C 5 H B O 7 . Monoclinic plates or prisms. i . From rhamnose : Water p 10.863. rt^ = 1.0685, ' = 16, [a} D -f 9.35 " /> 9.192, " r= 1.0569, /=I3, " = 9-50 /-=29.49S, " =1.1935, ^-=14, 1 Schnelle and Tollens : Ann. Chem. (I^iebig), 271, 82. - Fischer : Her. d. chem. Ges., 24, 3623. '' Fischer, Fay : Ibid., 28, 1982. 4 Fischer. Fay : Ibid., 28, 1977. ' Fischer, Piloty : Ibid., 23, 3104. i Fischer: Ibid., 24, 1836. 2686. 568 CONSTANTS OF ROTATION OF ACTIYK BODIES 2. From arabinose. Water /> 3-"74i. -} -9.13 ' GLUCURONIC ACID, COOH(CH.OH) 4 .CHO = C,H 10 O 7 - Potassium Salt, KC 8 H !( O. (at 100), needles. Water p 3.85. / 18, [a]/, ,21.25 /> 1.925. / 18, " 21.82 The specific rotation increases on dilution. The potassium salt rotates as strongly as does the amount of anhydride con- tained in it. Anhydride. From euxanthinic acid. Monoclinic crystals. Melting-point 175 to 178 (with decomposition). Water f> 14.14 ,/ ' 8 1.06201, []#-- 19.15 9-575, - 1.04125, 19.26 7.719, " 1.03307, 19.35 7.07, " = 20.06 4.787, 19-22 ' "=- 3-86, 19.89 - 3-54, 20.93 2.39, 21.80 11 " 1.93, 21.66 Effect of temperature r p. [<|z?at 5 18 28 34 Water j 14.14 +17.61 -f 19.15 j > 20.83 +21.00 9-575 17-63 19-26 20.37 21.10 7.719 17.72 19-35 20.47 21.27 Water c 3, / 21, [cr]/,. = 19.4 :t OXYGLUCONIC ACID, C 6 H 10 O 7 + 2H 2 O. Sirup. Water p 2, [>] -14.5 ' SACCHARONIC ACID, COOH.C.OH.CH 3 (CH.OH X.COOH- C.H 10 7 . Anhydride (Saccharon), C B H K O 6 + H 2 O. Formed by the oxidation of saccharin with nitric acid. Triclinic plates. Melting-point 156. Water .... /> = 12.41, d 1.0451. / 18, [a]/, 6.1' Will, Peters: Ber. d. chem. (ies., 22. 1697. Thierfelder : X.tschr. physiol. Chem i , Kiilz : Ztschr. f. niol., aa, 478. Boutroux : Ann. chim. phys., [6], ai. Kiliani: Bcr. d. chem. O**., 15, 2959. ACIDS WITH EIGHT ATOMS OF OXYGEN 569 13. Acids with Eight Atoms of Oxygen LKVULOSECARBOXVLIC ACID, CH,(OH).C(OH.eOOH)(CH.OH),CH,OH == C : H M O,- AnJiydride, plates or prisms, C-H 12 O 7 . Softens at 126 ; melting-point 130. Rotates strongly to the right in 6 per cent, aqueous solution. 1 ^-GLUCOHEPTOXIC ACID (Dextrose carboxylic acid ),CH.,OH. < CH.OH)..COOH = C 7 H H (X. Anhydride, C 7 H I2 O 7 . Rhombic crystals. Water... p = 3-3 = 55-3 '' /?-GLUCOHEPTONIC ACID. Anhydride^ C.H^O.. Colorless needles. Melting-point 151 to 152 (uncor. ). Water .../>- 10.049. d ^1.0372 / - 20 Multirotation. After 20 minutes [a]/7 79.1 After 24 hours constant 67.7 The multirotation is not caused here, as with the other lac- tones, by transformation into the acid, since at the end of the experiment the solution is found perfectly neutral. :; ^-GALACTOSECARBOXYLIC ACID, CH,OH.rCHOH) 5 CO 2 H. vSmall needles ; melting-point 145. The rotation of aqueous galactose solutions, to which hydro- cyanic acid has been added in excess at the ordinary temper- ature, decreases gradually, and finally becomes o. The amide of galactose carboxylic acid is formed, which on boiling with water is decomposed. The acid is inactive in 5 per cent, aqueous solution (/ = 2 dm.).* Barium Salt, Ba< CH^OJ,. From the amide on boiling with baryta water. Water c 12.01, / = 20, [a~\ D --- 5.50 */-MANNOHEPTON'ic ACID (Mannose carboxylic acid), CH a OH.(CH.OH; 5 .COOH. Melting-point 175. Rotates slightly to the left in aqueous solution. 1 Kiliani : Ber. d. chem (",es., 19, 1915. - Kiliani : Ibid., 19, 770. Fischer : Ann. Chem. (Liebig), 270, 84. 4 Maquenne : Conipt. rend.. 106, 570 CONSTANTS OF ROTATION OF ACTIVE BODIES Anhydride, C 7 H 12 O 7 - Crystals. Melting-point 148 to 150. Water c = 10.009, t 20, [<*]/> 74.23 Slight decrease later.' /-MANNOHEPTONIC ACID. Anhydride, C.H 12 O_. Crystals. Melting-point 153 to 155. From /-mannose by the cyanhydric reaction and saponifi- cation. Water .. p =.-. 5.27, d = 1.02, / = 20, [a]/, *= + 75.15 - RHAMNOHEPTONIC ACID, CH 3 (CHOH),..COOH. From rhamnohexose by hydrocyanic acid. Anhydride, C 8 H 14 O 7 . Crystals. Melting-point 160. Water c= 10.036, / = 20, [or]/, = -f 55.6 After six hours the rotation was still unchanged. Saccharic Adds, CO 2 H.(CHOH) 4 .CO 2 H = C.H 10 O 8 . = -f 5.84 Multi rotation could not be detected. 5 The specific rotation of /-saccharic acid has not yet been 1 Fischer, Passmore : Ber. d. chem. Ges., 23, 2226. Smith : Ann. Chem. (Liebig), 371, 183. ' Compt. rend., 53, 343- 4 Ann. Chem. (lyiebig), 345, 9. * Sohst and Tollens : Ibid., 345, 15. ACIDS WITH EIGHT ATOMS OF OXYGEX 571 determined. Its crystalline potassium salt, KC 6 H,,O rotates slightly to the left. 1 ISOSACCHARIC ACID. Rhombic crystals. Melting-point 185. Water-. /> F? 4.266, = 4.7 after heating, [a]/) + 48.93 :{ Isosaccharic Acid Diethyl Ester, (C 2 H 5 ).,C 6 H,O S . Crystals. Melting-point 73 ; boiling-point 250. Water c =-- 5, [<]z> --= - 35-5 4 Isosaccharic Arid Diamide, C fi H H O-(NH 2 ) 2 , Crystals. Melt- ing-point 226. Water ^ = 5, M/> = ~r 7-i6 4 NORISOSACCHARIC ACID, (C 6 H 10 O 8 ). Calcium salt (p = 5) and HC1. After heating : [a]/, = + 51.73 flf-MANNOSACCHARIC ACID. Anhydride, C fi H 6 O 6 -f- 2H.,O. Crystals. Melting-point 1 80 to 190. In fresh aqueous solution : p = 3.432, d = 1.0176, t = 23, [a]/> = + 201.8 6 Mucic ACID, (CHOH) 4 (CO 2 H) a . Crystals. Melting-point 213. This acid is inactive from the symmetrical arrangement of its molecule, and former observations of rotation must be referred to impurities. 7 For this reason the efforts of Ruhe- mann and Dufton 8 to split up the acid into two active compo- nents by aid of quinine, cinchonine, or strychnine failed. TALOMUCIC ACID, C 6 H 10 O S . Microscopic plates. Melting- point 158 (with decomposition). For freshly prepared aqueous solutions,/ = 3.84, ^= 1 Fischer : Her. d. chem. Ges., 33, 2621. - Tieniann and Haarmann : Ibid., 19, 1260. 3 Tieniann : /bid., 27, 137. 4 Tieinann and Haarmann. 5 Tieniann : Ber. d. chem. Ges., 27, 137. - Fischer : /bid., 24, 539. 7 Fischer, Hertz: /bid., 25, 1247. - J. Chem. Soc.. 59, 750. 5J2 CONSTANTS OF ROTATION OF ACTIVE BODIES / 20, [] />^- about 29.4. On heating the solution, the rotation diminishes on account of lactone formation. 1 14. Acids with Nine Atoms of Oxygen "-GLUCOOCTONIC ACID, CH.,OH(CH.OH) 8 .COOH = C,H 16 O a - Anhydride (From or-glucoheptose), CH 14 O S . Melting- point 145 to 147 (uncor.). Water-. /> =fc 10.405, d 1.0417, / 20, [a]/, == -j 45.9 - yfl-GLUCOOCTONic ACID. Anhydride, By product in the formation of the := -j- 23.6 The rotation remained unchanged after twelve hours. :! rt'-MANNOOCTONIC ACID, C^H ]6 O 9 . Anhydride, C 8 H U O M . From mannoheptose. Crystals ; melt- ing-point 167 to 170. Water-, p 9.8534, J*> = 1.0394, t 20, [a\ D 43-58 4 RHAMNOOCTONIC ACID, CH 3 (CH.OH)..COOH = C,,H lw O y . Anhydride, C,H 16 O 8 . From rhamnoheptose. Needles. Melt- ing-point 171 to 172. Water.. /> = 4.762, d 1.0163, t 1.0163, ' ^ 2O []/> - 5O- 8 s GALAOCTONicAcio, lactone, C 8 H I4 O,. Melting-point 220 to 223. P 4.26, rf = I.OI7, t ~ 20, [O-]/, = - 64.0 (1 "-PENTOXYPIMELIC ACID, COOH(CH.OH)..COOH C.H 12 O 9 . From 'dride C 7 H 10 O M is inactive in 10 per cent aqueous solution* 7 )8-PENTOXYPiMELic ACID. From /^-glucoheptonic acid. 1 Fischer : Bcr. d. chem. Ges., 34, 3622. - Fischer : Ann. Chem. (Liebig), 370. 92. :t Fischer : Ibid., 370, 100. 4 Fischer, Passmore : Her. d. chem. Ges., 33, 2233. } i-.cher, Piloty : Ibid., 33, 3109. Fischer: Ann. Chem. (I v iebig), 388, 149. <-r : Ibid., 27 ACIDS WITH TEN ATOMS OF OXYGEN 573 Anhydride, C T H 1(( O,. Crystals. Melting-point about 177. Water /> 9.972, d = 1.0433, t 20, [a]/) 68.5 l TANNIC ACID, TANNIN, C U H 10 O 9 -f 2H..O. Water . - . p i, t 20, a D (for 2 dm.) 1.50 (?)- This acid was formerly supposed to be inactive, but Giinther and also H. Schiff :< recognized it as strongly right rotating. The latter found for pure commercial preparations in aqueous solutions with c = i, a rotation varying from [<*]/, = ; -f 14 to -f 67 for different preparations. This stands in contra- diction to the constitutional formula proposed by Schiff for tannin, according to which there is no asymmetric carbon atom present, and which is supported by the fact that the synthetic tannin obtained from gallic acid is optically inactive, also by the fact that the products obtained from natural tannin by hydrolysis with weak hydrochloric acid (gallic acids) are like- wise inactive. To clear up this point, Wai den* has under- taken some experiments on the separation of different compo- nents from commercially pure tannin ( (/*J /> = : -f- 67.5, in water, c = i ) by means of dialysis, and also by fractional precipitation from solutions in ethyl acetate by addition of ben- zene, etc. By such methods he found it possible to sepa- rate the tannin into fractions with unequally strong rotations. Walden, therefore, considers it probable that the rotating pow r er of the natural tannin is due to admixture with small amounts of highly active substances of unknown composition. Rosenheim and Schidrowitz 5 have also examined the acid anew and come to conclusions somewhat opposed to Wai- den's. They conclude that the larger portion of the commercial acid is a single homogeneous body of high rotation, and that the formula of Schiff must be wrong. 15. Acids with Ten Atoms of Oxygen <*-GLUCONONONIC ACID, C 9 H, H O 10 . From tf-glucooctose. Anhydride, C 9 H 16 O 9 . Has not been obtained crystalline. Water c - about 10, t 20, [a]/? = -f 33 fi 1 Fischer : Loc. cit. - Giinther : Ber. d. deutsch. Pharm. Ges.. 5, 297, 1895. Chem. Ztg., (1895), p. 1680; (1896), p. 865. 4 Ber. d. chem. Ges., 30, 3151 (1897). * J. Chem. Soc., 73, 878. 15 Fischer : Ann. Chem. (Liebig), 370, 102. 574 CONSTANTS OF ROTATION OF ACTIVE BODIES = 104.1 Compound of d-Arabinose with Acetamide, CH 2 OH. (CHOH) 3 CH(NH.C 2 H 3 0),. Fine white needles; melting- point 187. Water ..... p . 10.03, ^ 20 = i-O455> ' = 2O . [**]/> - 9.5 2 l-Arabinose. From cherry gum, exhausted beet cuttings, exhausted beermash, wheat bran, gumarabic, gum tragacanth, quince mucilage, gedda mucilage, etc. Glittering trimetric prisms ; 3 melting-point 160 ; 4 152 to 153 ; 5 158 to i6o. 6 Water ..... c 10, t 5, [or] D = -f 104.4 7 Water ..... c = 10, t = 18, " = + 104.4' Water ..... =10.369, * = 20, " =+105.4, [a]y= + ii8 09 Water ..... c= 8.740, " = -f- 105.1 10 Water ..... c =- 9.730, / = 20, = -f 104.55 n Water ..... c 10.201, t = 20, " = + 104.64 n Water ..... c= 9.016, / = 20, " = -f- 103.87 12 /-Arabinose exhibits the phenomenon of multirotation. The beginning rotation for c = 9.73 is about 157. 13 See 72. Ammonia immediately destroys the multirotation. 14 The rota- Fischer, Passmore : Her. d. chem. Ges., 33, 2236. Wohl : Ibid., a6, 740. O'Sullivan : J. Chem. Soc., 45, 41. Scheibler : Ber. d. chem. Ges., i, 108 ; v. I^ippmann : Her. d. chem. Ges.. 17, 2239. Frankland, MacGreger : J. Chem. Soc., 61, 737. Conrad, Guthzeit : Ber. d. chem. Ges., 18, 2907. Bauer : Ibid., 33, Ref. 835. Scheibler: Ibid., 17, 1731. v. I^ippmann : Ibid., 17, 2239. 10 Kiliani : Ibid., 19, 3031. 11 Parcus, Tollens : Ann. Chem. (lyiebig), 357, 173. 12 Allen, Tolleus : Ibid. t 360, 300. 18 Parcus and Tollens : Loc. cit. 11 Schulze, Tollens : Ann. Chem. (Uebig), 371, 49. OXYALDEHYDES, ALDOSES, ALDEHYDE SUGARS 575 tion decreases with increasing temperature. For example : [flr]# - + 106.0, [] = - 104.5 l Arabinosazone, C 3 H B O 3 (N.NHC 6 H 5 ).,. Crystals ; melting- point 157 to 158 ; 2 159. Alcohol, 95 per cent. c = 3.40, t = 20, []/> = + 18.90 The rotation gradually disappears.* Compare Fischer. 5 i-Arabinose. An aqueous solution which contained 10 per cent, of each of the active components rotated in a i dm. tube, after fifteen minutes, 4 to the right ; after one hour -f- 0.2, and after two hours it was inactive. Wohl 6 refers this to the birotation of the /-arabinose. XYLOSE, CH,OH(CHOH) 3 CHO = C 6 H 10 O 6 . White needles or orthorhombic prisms ; melting-point 144 to 145. Water . - c = 10.664, t = 20, \_oc] D = -f 19.31 \" Water.. =11.070, / = 20, " = -f 19.22 ( Water.. c= 10.108, / = 22, " =+19.39 Water .. /= 9.940, d = 1.0359, t = 20, !>]/> + 18.99 8 (Mean of about 50 readings.) The effect of concentration on the specific rotation of aqueous solutions of xylose was investigated by Schulze and Tollens.* All the solutions were tested after standing twenty to twenty- four hours, the 61 per cent, solution after a quarter of an hour. > d?. My- 3.115 [.00977 -f 18.425 5.376 .01814 18.547 9.706 .03481 18.773 21.744 .08299 19.610 34.355 13750 20.495 46.395 .19266 21.429 56.229 .24205 22.681 61.747 .27258 23.702 Parcus, Tollens. Scheibler. Allen and Tollens. Allen and Tollens: Ann. Chem. (I^iebig), 260, Fischer : Her. d. chem. Ges., 23, 385, note. Ibid., 26, 740. Parcnsand Tollens : Ann. Chem. (Uebig), 257 Wheeler. Tollens : /bid., 254, 310. /bid., 271, 40. 300. 175. 576 CONSTANTS OF ROTATION OF ACTIVE BODIES From these figures, interpolation formulas for the tempera- ture of 20 may be calculated : I C=; 31034, [ a] /> --- 18.095 + 0.06986 />, II ^-^341061, - 23.089 0.1827;!) + 0.0031 2 A" The temperature has no marked effect on the rotation be- tween 15 and 20, but above 20 a change takes place which must be considered in any exact investigations. Water p =- 10.0829, d 1.0362 -f 18.898 20 18.909 25 19.248 30 19.628 I Xylose exhibits multirotation. 72. Observations of Wheeler and Tollens 1 gave : Water . . c ~ 10.236, t c= 20, after 5 minutes []/> =i -f- 85.86 Water-, c = 10.236, ^=20, " 10 " = + 70.14 Water .. = 10.236, t = 20, " 16 hours constant " - - -f 18.59 Parcus and Tollens 2 found the beginning rotation lower : Water C= 10.664, t = 20 after 5.1 minutes [a] D = -f 77.87 constant " 24 hours = -f 19.31 Water r 11.070, / = 20, " 4^ minutes " - = -f 78.61 constant " 24 hours = -f- 19.22 From this []/> two tninutes after solution about 91 " " " immediately " about 100 Ammonia of about o. i per cent strength brings about the end rotation immediately. If more than o. i per cent, is pres- ent the rotating power is very greatly decreased. See 73." Xylosazone, C 5 H 8 O 3 (N a H.C 6 H 5 ) 2 . Bright yellow silky needles; melting-point 159 to 160. Alcohol, 95 per cent. p 2.815, d -85, [<*]/> ~ 43-3 6 After more than a week the solution showed approximately the same end rotation. 4 1 Ann. Chem. (I,iebig), 354, 311. * Ibid., 357, 175. " Schulze, Tollens : Ibid., 371, 49. * Allen, Tollens: Ibid., 360, 295. OXYALDEHYDES, ALDOSES, ALDEHYDE SUGARS 577 FUCOSE, CH 3 . (CHOH) 4 CHO == C 6 H W O 5 - Fr <> m s ^a weeds. Crystals; melting-point 116 to 140. Water.. = 9.1375 (dried at 65), / ^ 20, [!/>= 75.96 Fucose shows marked multirotation. An aqueous solution clarified with alumina cream gave the following rotations: 1 c = 6.9155, / 20, after n minutes [a~\ D = 111.8 " 14 " 106.8 " 21 " 97.8 31 " 90.0 71 " 79.0 ioi " 77.8 " 146 " 76.8 On the following day constant = - 77.0 RHAMNOSE (Isodulcite, rhamnodulcite), CH 3 .(CHOH)* CHO + H 2 = C 6 H U 5 + H 2 0. Ordinary Rhamnose. Monoclinic crystals ; melting-point 93 to 94. (On the three modifications of the substance, see 72.) The following numbers hold for the final constant rotation and the hydrate : From quercitrine, water c = 18.076, t = 17, [a]^ = -J- 8.04 2 From xanthorhamnine, water c 26.04, t = 17, []z> = -f 8.07 3 From xanthorhamnine, water c = 12.390, / 18, []/> = -j- 8.83 4 From naringenine (citrus decu- mana) water <:=25.i66, t=ij, [a] n = -+- 8.2 5 Effect of concentration : c = 5 9 18 22 40 [a]/, = H- 8.48, + 8.52, + 8.50, + 8.51, - 8.65 The rotation is, therefore, influenced but little by the con- centration. With elevation of temperature, the rotation de- creases and vice versa. For / 6 to 20 this formula holds : [<*]/> =9.18 0.035 / 6 Rhamnose exhibits multirotation as described in 72. Freshly prepared solutions show left rotation which changes gradually to right rotation : ' 1 Giinther. Tollens : Ann. Chem. (I^iebig), 371, 86. '- Berend : Ber. d. chem. Ges., n, 1354. a Liebermann, Hormann : Ibid., n, 956; Will : Ibid., ao, 1186. * Stohmann, Iangbein : J. prakt. Chem., [2], 45, 308. '> Will : Ber. d. chern. Ges., 20, 294. 6 Schnelle, Tollens : Ann. Chem. (l,iebig), 27i, 62. ' Jacobi: Ibid.. 272, 175. 37 578 CONSTANTS OF ROTATION OF ACTIVE BODIES Water p - 10, df = J - O2 3 6 > ' r 2O After i minute O]/ 1 - 5- 6 After 8 minutes " o.o Constant after i hour " 8.3 Water c TO, / 19.51020.5 After 5 i minutes []/>" 3-ii After 9 minutes " o.o Constant after 57 minutes " = + 8.56 In another series of experiments, the beginning rotation was -4-5.' The decrease in rotation takes place regularly with gradually- diminishing rapidity. 2 When rhamnose is heated to the melting-point, the end rota- tion is reached immediately/ A small addition of ammonia destroys the multirotation. 4 Water, c 10.052 / = 20, [a]/) after 20 hours = -+- 7.86 Water with o.i per cent ammonia, t 20, [<*]/.> after 7 min. ==-f 7-95 For further observations on rhamnose see Gerne/. 1 and Tanret." On the rotation of rhamnose in alcoholic solution, see 64. Anhydrous, Amorphous Rhamnose, Isodulcitan, C H H K ,O-. Water p 9.238, d^ 1.0281, / 20, [>]/, =-{-8.7 This shows no multirotation in water solution : ! Water /> 27.982, d'* 1.1002, / 20, (>]/> -j 6.36 Water /> 21.519. " 1.0765, / 20, =+ 9.43 Water /> -14.419, " ^1.0507, / 20, s'+ 9.34 Methyl alcohol (97.94 p. c.), /> 19.06,^" 0.8842, [a ]/> -10.59 In alcoholic solution the rotation may be either -j- or according to the concentration and the percentage of water in the alcohol. Ray man and Kruis 7 obtained the following num- bers : 1 Schnelle and Tollens : IMC. at. * I'arcus and Tollens : Ann. Cheni. (I y iebig), 257, 160. * Jacobi. * Schul/.e, Tollens: Ann. Chem. (I^ebig), 371, 49. ' Compt. rend., 121, 1150. //>/ = J 8.96 " (10.91 "") 16.694 1.0432 -f 8.18 " (29.84 " " ) -17.690 1.0170 4.14 " (43-57 " " ) 18-621 0.9960 3.28 " (38.27 " ") 11.177 0.9767 2.35 " (66.09 " " ) 12.669 0.9482 0.84 (94.00 " " ) 10.445 0.8502 9.23 (96.11 " " ) 7.028 0.8292 - 10.04 " (97-36"") 4-875 0.8159 -10.69 " (99-33 " " ) 9-368 0.8176 - 10.65 Rhamnose Oxime, C 6 H,,O 4 :NOH. Crystals ; melting-point 127 to 128. Does not show multirotation. See 72. Rhamnose Phenylhydrazone , C 6 H 12 O 4 . ( N 2 H . C 6 H 5 ) . Colorless scales ; melting-point 159. Dissolved in water by gentle heat and quickly cooled: p - i. oil, d*' = 1.0091, / 20, [a]/> = 54-3 ' It does not show birotation. The rapidity of hydrazone formation may be followed by the polariscope.' GALACTOSE, (Lactoglucose), CH.,.OH.(CH.OH) 4 .CHO. d-Galactose. Found in three modifications. See 72. Ordinary Galadose. Granular crystals ; melting-point 1 68 ; 3 161 to 162 ; 4 162. :> For the constant form there has been found : Water, c 10 to 15, [a]/> ~- 81. 4 to 81.7 '' Water, p 10, d ts 1.0385, / = 18, []/> 81.2 T Water, c 9-973, t *5< [--- 8i.5,;[ - (0.276 0.0025 p) /, holding for/ = 11 to 20 and / = 4 to 40. The two formulas lead to the following specific rotations : A 20 /= io 20 3 I0 20 3 I0 20 30 (Meissl 82.58 80.49 78-40 82.97 80.88 78.79 83.36 81.27 79.18 D I Rindell 82.52 80.01 77.50 83.64 81.25 78.87 84.76 82.50 80.24 Galactose possesses multi rotation. For the beginning rota- tion Meissl found 130 to 140, v. L,ippmann 134.5, Koch^ 132.5, 133.8, 137.4. Parcus and Tollens 4 give the follow- ing observations : = 11.0810. / = = 20 C = 10.2045. / = 2C ). Time after solution. r]. Time after solution. []/.. 117 27 117 48 8 1 16 14 Q 1 16 47 1 14 27 JQ " 1 14 41 20 " IO7 71 10 " AVJ/./l IO2 87 70 " QQ Qr 3 oS 77 50 " 95 88 1 9. 33 94 26 8 S H 4 " 81 02 7 Si 7.7 "OU 1 80 27 6 ^ 80 70 24 " ) ' 4 i 00.39 ! 1 J. prakt. Chem., [2], 23, 97. Scheibler's N. Z. Rbz.-Ind., 4, 170. Pharm. Ztg. f. Russl., 1886. < Ann. Chem. (I 4 iebig), 357, 168. OXYALDEHYDES, ALDQSES, ALDEHYDE SUGARS 581 From the curve constructed from this we have : [ a] r> after 2 minutes = about 122 " at moment of solution = " 127 o.i per cent ammonia destroys the multirotation. 1 d-Galactose Oxime, C 6 H 12 O 5 :NOH. Crystals ; melting-point, 175 to 176. Multirotation shown. Water p = 5.1056, rf~ = 1.017, / = 20 After 10 minutes [ oC\r> + 20.6 " 20 hours constant " -+14.75 On account of the slight solubility in cold water, solutions up to a strength of 5 per cent, only could be investigated. 2 d-Galactose Phenylhydrazone, C 6 H 12 O 5 :(N 2 H.C 6 H 5 ). Crystals; melting-point, 158. Dissolved by aid of gentle heat in water and quickly cooled : p = 1.980 d = 1.0065, t = 20, [a]/> = - 21.6 Multirotation could not be observed. 3 See also Fischer and Tafel. 4 d-Galactose Anilide, C 6 H n O 5 .NH.C 6 H 5 . Triclinic crystals ; melting-point about 147. Ethyl alcohol (90 vol. per cent.) : p = 2.289, ^- = 0.8366, ^ = 20 to 23, [a] = -31-33 p == 2.099, " == 0.8334, t = 20 to 23, - 31.44 Methyl alcohol (4^ = 0.7907^; ^ = 1.699, d? = 0-7997, / = 20 to 23, [<*]/> = - 33- I2 d-Galactost-p-toluide, C 6 H 11 O 6 .HN(CH 1 ).C 6 H 4 . Crystals ; melting-point, 139. Methyl alcohol (as above) : p = 0.6167, ^4 = 0.7952, t = 20 to 23, [a]/, = 33.99 Ethyl alcohol (50 per cent.) : p = 0.9832, d 0.9316, t = 20 to 23, \oi\D - - 10.91 5 d- Galactose Pentaacetate -, C 6 H.O 6 ( C 2 H 3 O) 5 . Glistening rhom- bic plates ; melting-point, 142. Right-rotating. 6 l-Galactose. Formed in the fermentation of z'-galactose by 1 Schulze, Tollens: Ann. Chem. (Liebig), 271, 49. - Jacobi : Ber. d. chem. Ges., 34, 698. 3 Jacobi : Ann. Chem. (Liebig), 272, 173. 4 Ber. d. chem. Ges., 20, 2568. 5 Sorokin : J. prakt. Chem., [2], 37, 295 and 309. 6 Fudakowsky : Ber. d. chem. Ges., u, 1071 582 CONSTANTS OF ROTATION OF ACTIVE BODIES beer yeast. Crystals ; melting-point, 162 to 163 (not cor.). Exhibits multirotation. Water .......... c 10; [<*]/> after 8 minutes 120 " constant 73-6 These values may be looked upon as approximate only as the preparation used was not sufficiently pure. 1 SORBIN (Sorbinose, sorbose), CH 2 OH.(CH.OH) ;r CO. CH 2 OH. Rhombic crystals. Water .......... c 23, OL 46.9- Shows no birotation. GLUCOSE (Glycose). d-Glucose (Dextrose, grape-sugar, starch sugar), CH 2 OH. (CHOH) 4 CHO + H,O. On the three modifications of ^-glucose see 72, p. 260. Ordinary Glucose. Hydrate d : monoclinic crystals ; anhy- drous : rhombic hemihedral. The constant specific rotation of 52.50 ~j 0.018796 /> r 0.0005 1683 />-' 59.55 0.12216 q -f- 0.0005168 q* II. Hydrated Glucose, C.H W O. + H 2 O : M" 47-73 + 0.015534 p -f 0.0003883 p l According to this, we have the following specific rotations for solutions containing different amounts of anhydrous glu- cose : 1 Fischer, Hertz : Ber. d. chem. Ges., 35, 1247. Berthelot : Ann. chim. phys., [3], 35, 222. * Wehmer. Tollens: Ann. Chem. (I,iebig), 243, 320. 4 Ber. d. chem. Ges., 9, 487, 1531 ; 17, 2234. ' I hid., 17, 2238. OXYAUDEHYDES, ALDOSES, ALDEHYDE SUGARS 583 10 15 20 25 30 52.74 52.90 53-08 53-29 53.53 40 50 60 70 80 54.08 54.73 55-49 56.35 51-31 and the reply of [a]/, 52.61 /> 35 !>]/> 53-79 Compare with reference to this, Ost, Tollens.' Solutions freshly prepared, without heat, show birotation (see 72). Among the many experiments on the change in the rotation of glucose, it will be sufficient to quote those of Parcus and Tollens as the most exact.' These were made with glucose, prepared according to Soxhlet's method from cane-sugar, and dried at 60 to 70. / = 20, C = 9.0970. t = 20, C = 5.5255- Time after solution. ' Time after solution. [ f O^- rr^c T i mimit*c. . ir\A "y(-\ sVi minutes 105. 16 7 minutes . . . 104.26 6 1 , " 104.59 8 " 103.64 10 101.55 9 " ... 103.01 12 " 100.03 I0 " 102.38 U >l 97-94 12 " ... 101.13 20 " 92-42 13 " 100.50 30 83.86 14 " ... 99-88 50 72.26 15 " ... 98-63 I hour 68.27 25 92-35 I 1 4 hours 63.33 3 " 88.61 [' 1 " 59.71 i hour 73.58 6 % ' constant.- 52.49 7 hours, constant 52.60 The end rotation appears immediately when the glucose has been heated to melting/ or when the solution is heated for some time. 1 The addition of o.i per cent, of ammonia destroys the multirotation," but if more ammonia is added the specific rotation is much decreased. c about 10, / = 20 : Directly after solution ......... [a]/> 49.82 ' ' 50.00 '* = 49.65 " 46.36 After 30 minutes After i 1 ., hours After 24 hours 1641. 1 Ber. d. chem. Ges , 24 - Ibid., 24, 2000. 3 Ann. Chem. (Uebig), 257, 164. Schmidt : Ibid., 119, 95. ; ' Hesse : Ibid., 192, 172. ' ; Schulze and Tollens : Ibid., 271, 49. 584 CONSTANTS OF ROTATION OF ACTIVE BODIES Influence of Inactive Substances on the Rotation of Glucose Pribram 1 has determined the effect of addition of the follow- ing substances on the specific rotation of glucose, the end rotation reached in each case after long standing being finally measured : 1. Ammonium Carbonate. Solutions which contained in 100 cc. 16.46 grams of anhy- drous glucose and p grams of the salt gave : /> o 2 46 8 10 [<*]= 52.83 52.40 52.22 51.36 51.11 50-85 There is, therefore, a slight decrease in the specific rotation of the glucose. 2. Urea. The solutions contained in 100 cc., 15.797 grams of anhy- drous glucose and p grams of urea : P -. o 4 8 12 16 [a] - 52.91 52.84 52.61 52.23 51.95 There is also here a decrease in the specific rotation, but it is slight. According to Neumann Wender, 2 neither urea nor any other substance in urine has any effect on the rotation of glucose. 3. Acetone. The solution contained in ioocc., 15. 68 grams of anhydrous glucose, c grams of acetone and water : r o 4 8 12 16 20 24 40 50 [or] 52.89 53.29 53.63 53.94 54.23 54.53 54.81 56.19 57-03 The specific rotation of glucose increases in rather marked degree with increase in the acetone content of the solutions. Alkalies and also lime bring about a gradual decrease in the rotation, and its final disappearance by decomposition and production of salts of saccharinic acid (left-rotating), glucinic acid, and humus-like bodies. Derivatives of ^/-Glucose Glucoseamine 1 Sitzungsber. d. Wien. Akad., p7, II 375. * Ber. d. chera. Ges.. 24, 2200. DERIVATIVES OF after 10 minutes 15.3 " ". 12 to 15 hours, constant. 46.9 The multirotation is not due to the fact that the substance had passed into the isomeric modification, because the crystals again obtained from the solution melt at 114.-' Glucose Anilide, C 6 H n O 5 -HN.C 6 H 5 . Sheaves of needles ; melting-point about 147. Methyl alcohol (^= 0.7907): p = 5.029, d 0.8065, t 20 to 23, []/> = 48.32 p 3.326, " - . 0.8055, / 20 tO 23, r- 49.15 Ethyl alcohol of 90 vol. per cent. (d 2 0.8294) : P 4-697, d 0.8453, t 20 to 23, [>]/> - 44.08) ! p 3- 26 9. 0.8407, t 20 to 23, " 44-15 j Glucose Paratoluide, C 6 H U O,.HN. (CH :! )C 6 H 4 -f ! /,H,O. Crys- tals ; melting-point, 106. Methyl alcohol (as above) : P 7.879. d* 1 0.8243, t 22 to 26, [n:]/,^ -43.88 p= 4.082, " 0.8061, / 22 to 26, " 42.55 p 2.6l3, " 0.8007, t 22 tO 26, " 38.23 \ Ethyl alcohol (as above): P 6.658, d 0.8513, t 22 to 26, [>]/,. -38.80] Glucose Ethylmercaptal, C 8 H 12 O 5 (SC 2 H,),. Crystals. Water p 4.878, d= 1.002, t 50, [ -29.8^ /-GLUCOSE. From /-gluconic acid anhydride. Crystals ; melting-point, 141 to 143. 1 Herzfeld: Ann.Chem. (Iviebitf), 230, 219. 2 Jacohi : I bid., aya, 171. 3 Sorokin : J. prakt. Chem., [2], 37, 295. 4 Sorokin : I hid., [2], 37, 308. 6 Fischer : Ber. d. chem. r.es., 37, 675. MANNOSE 587 Water / 4.114, ^=-.-i.oi6, t=- 20 j 1 [a]/) after 7 minutes - 94-5 / " 7 hours, constant - 51.4! MANNOSE (Seminose), CH,OH.(CH.OH) r CHO. 2 d-Mannose. I. From mannitol by oxidation. Crystals. Water /> about 10, ^ -.- 1.0416, / 20, []/> r 12.96 :! II. From nut shells : Water c about 10, / 20, \oi\o ~ H-36 The small difference is due to the amorphous nature of the latter product. 4 d-Mannose Oxime, C 6 H 12 O 5 :NOH. Crystals; melting-point 176 ; 5 176 to 184 (not constant). 6 Exhibits multirotation. On account of the difficult solu- bility, a 5 per cent, solution only could be investigated. Water p = 4.798, d 1.016, t - 20 []/> after 10 minutes = = 4" 7'5 " " 6 hours, constant =-(-3.2! l-Mannose. From /-mannonic acid anhydride. Sirup. In aqueous solution rotates to the left.* The phenylhydrazone, melting-point 195, is right-rotating in hydrochloric acid solution ; the phenylglucosazone, melting- point 205, is strongly right-rotating in glacial acetic acid solution. 9 RHAMNOHEXOSE, Methylhexose, CH 3 .(CHOH) 5 .CHO = C 7 H U O 6 . From rhamnohexonic acid. Crystals ; melting-point, 1 80 to 181. Shows strong multirotation. Water p = 9.675, d = 1.0347, ' 2O ] 10 After ] 2 hour []/,-- - 82.9 j. After 12 hours, constant " 61.4! 1 Fischer : Ber. d. chem. Ges., 23, 2618 '- Fischer, Hirschberger : Ibid., 23, 1155. ;i Fischer, Hirschberger : Ibid., 22, 365. 4 Fischer and Hirschberger : Ibid.. 32, 3218. "' Reiss : Ibid., 32, 611. 6 Fischer and Hirschberger: Ibid., 32, 1156. ' Jacobi : Ibid., 24, 698. 8 Fischer : Ibid., 33, 373. '-' Fischer. 10 Fischer, Piloty : Ber. d. chem. Ges., 33, 3102. 588 CONSTANTS OF ROTATION OF ACTIVE BODIES CH 2 OH.(CH.OH) 5 .CHO -= C 7 H U O T - From ar-glucoheptonic acid. Rhombic plates ; melting-point, 1 80 to 190. Shows weak birotation. For a solution made by gently warming with water, c = 10, t = 20, [<*]/? = 19.7 at once, and after fifteen minutes with greater dilution [] = - 25.- fi-Glucoheptose, not yet investigated. ^-MANNOHEPTOSE, C.H U O. (at 104). Crystals; melting- point, 134 to 135 (notcorr. ). From ^-inannoheptonic acid. Exhibits multirotation. Water .................. p= 10.807, d* = 1.0397, / = 20 1 - After 10 minutes .......................... [or] /> = -f- 85.05 \ After 24 hours, constant ................... " = -j- 68.64 l-Mannoheptose, from /-mannoheptonic acid, could not be ob- tained in a crystalline condition, but is characterized by its phenylhydrazone, melting at 196, with complete decom- position. 3 -GALAHEPTOSE, C.H U O 7 . Melting-point, 190 to 194. Shows multirotatiou. Water ...... p 9.2, 10 minutes after solution [a]^ = 22.5 After 24 hours []== 54.4 4 RHAMNOHEPTOSE, C 8 H 16 O 7 . From rhamnoheptonic acid. Water ....... c = 9.40, t = 20, [a]/, == -f 8.4 (about) a-GLUCOOCTOSE, C 8 H 16 O b + 2H 2 O. From -glucooctonic acid. Needles ; melting-point, 93 (uncorr.). Shows multi- rotation. Water ......... p 6.496, d = 1.0213, ^ 20 '"' Hydratd. Anhydrous. After a short time .......... [<^]/> 61.5 70.8 After 6 hours, constant ..... 43.9 50.5 ^-MANNOOCTOSE, C 8 H 16 O 8 . From ^/-mannooctonic acid. Colorless sirup. [a](approx.) = - 3.3 (i Fischer: Ann. Chetn. (Uebig), 370, 64. Fischer, Passmore : Ber. d. chem. Ges., 33, 222*. Smith : Ann. Chem. (Liehig), 373, 183. Fischer : Ibid., 288, 155- Fischer : Ibid.. 270, 64. Fiacher, Passmore: Ber. d. chem. Ges., 33, 2234. OXYKETONES 589 GALAOCTOSE, C 8 H 16 O, + H 2 O. From galaoctonic acid lac- tone. Thin plates; melting-point, 109 to m. [a] ; - 40 ' = 9, for]^= 103.92 -[-0.671 /, for t= 13 1040 " p = 23.5, " = 107.65+0.692 t, " /= 9 to 45 : b. For the effect of the percentage amount of water q in the solution, [a] = - 113.96 -r 0.258 q, for q ~ 60 to 95 per cent.. from which follows : For loo q=p= 5 10 20 30 40 [a]- -.= -89.42 -90.72 -93.30 -95.88 -98.47 According to Honig and Jesser, the lower values of Herz- feld depend on this, that in the conversion of inulin into sugar Fischer: Ann. Chem. (Liebig), 388, 150. Fischer, Passmore. Compt. rend., 43, 901. Ber. d. chem. Ges., 33, 2107. Ann. Chem. (lyiebig), 344, 287. Ibid., 344, 300. " Ber. d. chem. Ges., 19, 393. Ztschr. fiir Riibenzucker-Ind., (1888), p. 1028. 590 CONSTANTS OF ROTATION OF ACTIVE BODIES dextrines are always formed which, by the method of puri- fication with ether-alcohol employed by the latter, cannot be completely removed. 2. Jungfleisch and Grimbert 1 combine the following deter- minations, made for different concentrations and temperatures : c = 9-75- c = 48.75- / = 17. /. WD. [. 7 ~ 97.31 QI ^5 o , TOC 76 9nc . . QI 1 T 7 lU^./U 102.20 Q7 62 /o * y^-oo - 92.72 y^-oo - 89.90 / 16 A yo 28 y/.u^ - 90.39 82 r-i oV- u yj-o u - 97.07 02 -53 under the general formula : [a\ l D = - [101.38 0.56 / - 0.108 (c 10)]. Ost' 2 investigated fifteen solutions, the strength of which varied between i and 30 per cent, of levulose, at 20, and cal- culated this formula : M/?= - (91-9 + o.iu/0 4. Parcus and Tollens :i found for/> = 10, / 20, [or]/, 92 to 92.5 5. Wohl found 4 for p -^ 10.17 or c = 10.57, / 20, [a]/, 91.8 According to the above five observers, the specific rotation of a 10 per cent, solution at the temperature of observation varies between []/>- - 90.2 and 93. In alcoholic solution, levulose has about two-thirds the rotation shown in water. With reference to the multirotation, or decrease in the rota- tion of freshly prepared solutions, we have the following ob- servations : i . Parcus and Tollens : ' 1 Corapt. rend., 107, 390. - Her. d. chem. Ges., 34, 1636. 1 Ann. Chem. (Uebig), 257, 160. 4 Her. d. chem. Ges., 33, 2090. ' Ann. Chem. (Liebig). 357, 6. INVERT SUGAR 591 c- 10, / 20, First rotation after 6 mi n. [a]/) -104.0 End rotation "25 " " 92.3 and i hour []/> 92.1 2. Schulze and Tollens :' c = 10, / = 20, First rotation after 15 min. [a]/) 92.3 End rotation " 20 hours " 90.9 Iii o. i per cent, ammonia the constant end rotation is reached in five minutes. 3. Jungfleisch and Grimbert 2 give, among others, the fol- lowing observations : c = 1.779, t = ' c = 9.75, t = 7. After 10 min. [a]o= -106.02 After 35 min. [or]/>= 97.33 " 20 " " = 99.32 " 55 " " = 96.11 45 " - 93-83 75 -95-11 " 90 " - 92.00 105 -94-77 If the levulose solution in water is heated, a further gradual reduction takes place according to Jungfleisch and Grimbert, and to avoid this, the temperature should not go above 40. But Ost 3 did not observe this decrease. According to their strength, acids affect the specific rotation of levulose in different ways (Dubrunfaut, Jungfleisch and Grimbert, Ost). Levulose anilide is characterized by a very high rotation. Sorokin 4 found : In ethyl alcohol p = 0.712 [a~\ D = 215.7 p = 2.016 " -185.5 In methyl alcohol p 1.436 " -181.5 1 8. Invert Sugar The rotating pow ? er of the invert sugar solutions obtained by the action of acids on cane-sugar solutions has been investi- gated mainly by Gubbe/' Ost, B Wohl, 7 and others. AsGubbe, and later Ost found, the inversion is most conveniently accom- plished by heating 100 parts of cane-sugar with i part of oxalic 1 Ann. Chem. (lyiebig), 271, 53. '-' Compt. rend., 107, 390. :i Her. d. chem. Ges., 34, 1643. 4 J. prakt. Chem., [2], 37, 195. "' Ber. d. chem. Ges., 18, 2207. ( - Ibid., 34, 1640. 7 Ibid., 23, 20^7. 592 CONSTANTS OF ROTATION OF ACTIVE BODIES acid, in aqueous solution, for several hours to 50 or 60. No modification of the rotation of the invert sugar takes place here as is the case on heating with hydrochloric or sulphuric acid. The rotation of this sugar mixture is subject to many changes, and the following conditions have especial influence. a. Concentration. From careful experiments, Gubbe de- duces the following formulas, based on a temperature of 20 : (I) [or]* = - 19.447 0.06068 p -\- 0.000221 p- (II) 23.305 0.01649 q -f- 0.000221 g*, holding for/> = 9 to 68, or q = 32 to 91 ; (III) []= - 19-657 - 0.0361 c, holding for c up to 35. Ost gives the expression [a]= - 19-82 0.04 />, holding for/> = 2 to 30. b. Temperature. This formula was given by Tuchschmidt, 1 for c 17.21 and / = 5 to 35, {*}'= - 27.9 -f 0.32 /. Lippmann found nearly the same in a series of experiments extending to / = 8oV More accurate formulas deduced by Gubbe from his experi- ments are these : (IV) For* ^= o to 30: [or]',, = [ + 0.0005 1 7 P* l Levulose " 88.13 0.2583 p Invert sugar " - 19.447 0.06068 p + 0.000221 p* ' A If now the specific rotations of these sugars for different strengths, p, be calculated, we have from the above formulas : p. Dextrose. I^evulose. Arithmetical mean. Invert-sugar. Difference. 20 + 53-08 - 93-30 20.11 - 20.57 + 0.46 25 + 53-29 -94-59 20.65 - 20.83 + 0.18 30 + 53-53 -95-88 - 21. 18 21.07 O.I I 35 + 53-79 -97.17 - 21.69 - 21.30 -0.39 The same calculation has been carried out by Ost. 4 The agreement with observation confirms the above assump- tion fully and an earlier suggestion of Winter 5 that invert sugar may be made up of 4 parts of levulose and 3 parts of dextrose is evidently in error. 19. Disaccharides (Saccharoses, Bioses) CANE-SUGAR, C, 2 H 22 O U . Right-rotating. For the relation of the specific rotation of the sugar in aqueous solution to 1. The percentage amount/ of sugar, 2. The percentage amount q of water, 3. The concentration c, or grams of sugar in 100 cc., the following formulas have been given, based on accurate obser- vations : I. By Tollens. 6 Calculated from 19 solutions. i. Specific gravity of the solutions at 17.5, referred to water at 4. Rotation at 20 : a \P= 4 to 18, \JOL\ D = 66.810 0.015553 p 0.0452462 p 1 7 " \q = 82 to 96, " = 64.730 + 0.026045 q -- 0.0452462 q 1 b (p 18 to 69, " =66.386 + 0.015035^ 0.033986 p 1 \q 31 to 82, " 63.904 + 0.064686 q 0.033986 q 2 Tollens. Honig and Jesser. Gubbe. Her. d. chem. Ges., 34, 1640. Ann. Chem. (Iiebig), 344, 295 and 329. Ber. d. chem. Ges., 10, 1403. 1 0.0452462 ^ 0.000052462, etc. DISACCHARIDES 597 Later, Tollens 1 found that the formulas b satisfy dilute solu- tions also, down to p = i even. 2. Specific gravity of solutions at 17.5 referred to water at 17.5. Rotations at 20. From the same experiments as under i there follows for p. a', p 4 to 18, [a]/> == 66.727 0.015534 / 0.0452396 p b '.p= 1 8 to 69, " - 66.303 0.015016 p 0.033981 /* By another method of calculation, Thomsen 2 obtained from the observations of Tollens these expressions in place of those given under b above, v , J> = 18 to 69, \a\ n = 66.577 -f 0.007466 / 0.0331339^ t? = 3 1 to 82 " 64.190 -f 0.055212 q 0.0.331339 # 2 , which agree well with the Schmitz formulas given below under II., i. II. Schmitz 3 gives these formulas deduced from eight solu- tions (see 54, p. 195). 1. Specific gravity of the solutions at 20 referred to water at 4. Rotations at 20. p = 5 to 65, \ci\ D = 66.510 -f 0.004508 p 0.0328052 p <1 35 to 95, " 64.156 0.051596 q 0.0328052 q* 2. For the concentration at 20 Schmitz gives these formulas : c 10 to 86, [<*]/> 66.453 0.001236 c 0.0311704 & c 2.5 to 28, " = 66.639 0.020820 c -^ 0.0334603 c 1 c 2.5 to 28, k ' 66.541 0.008415 <: L,andolt 4 has calculated these approximate expressions from the observations of Tollens amd Schmitz, showing the relation of the specific rotation and concentration of solutions : c 4.5 to 27.7, [a] - 66.67 0.0095 c (true cc.) c 4.5 to 27.7, [or]*?- 5 -- 66.82 0.0096 c (Mohr's cc.) III. The latest experiments have been carried out by Nasini and Yillavecchia. 5 These formulas follow from twelve series of observations with different sugars : P 3 ^ 65, [a]= ( ; 66.438 -f 0.01031 2 p- 0.0335449/ 2 q 35 to 97, " 63.924 + 0.060586 q 0.0335449 q' 1 Ber. d. chem. Ges., 17, 1751. - Ibid., 14, 1652. 3 Ibid., 10, 1414. 4 Ibid., ai, 196. '> Public, de lab chim. centr. delle gabelle, Roma (1891), p.' 47. CONSTANTS OF ROTATION OF ACTIVE BODIES If we combine now the results which the different formulas give for a number of different percentage amounts of sugar, we find these variations : Formula. / = IO. 20. 30- 40. 50. \tl 66 65 66 48 66 70 66 10 Ac QO TA Toll en ^ 66 ;o 66 57 66 48 66 7=; 66 14 TA// 66 62 66 60 66 ;2 66 77 66 T7 Scliinit/ 66 <;; 66 AQ 66 10 00.31 66 24 66 oi III. Nasini and Villavecchia ""Oo 66.51 66.50 WW.f 66.43 66.28 "" ^o 66.07 It is seen that the values given by the last two formulas fur- nish the closest agreement. With reference to the specific rotation of cane-sugar in very dilute solutions, see the observations given already in 55. For the rotation of different rays of light by cane-sugar, see ''Rotation Dispersion," 45. Influence of temperature. This was formerly held by Mit- scherlich, 1 Hesse, 2 also Tuchschmidt 1 to be extremely small. Andrews* was the first to show that the specific rotation of sugar decreases with elevation of temperature and to the extent of 0.0114 for i C. According to new experiments which Schonrock 5 has carried out, using ten sugar solutions, the tem- perature coefficient lies between 0.0132 and 0.0151. This formula may be used to calculate the specific rotation of solu- tions of about 24 per cent, strength, for a temperature differ- ent from 20, between the limits, 12 and 25. []'/, = MS - 0-0144 (t - 20) Recently Wiley* 5 has reinvestigated the subject with extreme care and has found a value lower than those given above. The coefficient as given by him is 0.00994 as tne mean change in the specific rotation for each degree centigrade. In the paper by Andrews, the value is given in one place as ' Ber. d. Bcrl. Akad., (1842), p. 150. * Ann Chem. (t,iehig), 176, 97. J. prakt. Chem., [2], a, 244. Technology Quarterly, May (1889), p. 367 : Chem. Centrhl., (1890), I, p. 20. * Ber. der. phys.-techn. Keichsanstalt von 1896 ; /tschr. fiir lustrum., 17, iho. * J. Am. Chem. Soc., at, 568 (1899). DISACCHARIDEvS 599 o.oooi 14. But the context shows that this is a typographical error for 0.0114. On the specific rotation of cane-sugar in mixtures of water with acetone, methyl alcohol, and ethyl alcohol, see 59. According to Seyffart, 1 sugar retains the same specific rotation when dissolved to the extent of 5 to 40 grams in 100 cc. of alcohol of 9 to go . 2 The changes shown in the specific rotation of cane-sugar in presence of alkalies and salts have been explained already in 70. MILK-SUGAR. Lactose, Lactobiose. Anhydrous, C^H^C^ ; hydrated, C K H M O 10 + H,O. As already shown in the chapter on multirotation, 72, this sugar is found in three modifications, of which a and y in aqueous solution pass slowly into ft at the ordinary temper- ature, but rapidly on warming. The tr-form exhibits greater rotation (birotation), and the y-form less rotation (half rotation) (see 71). Anhydride. Hydrate. 1. or-Modif. ordinary milk-sugar [a] D = 88 + 84 2. /3- " stable form " ^ +55-3 +52-53 3. 7- " second labile form " '-7-36.2 34-4 Ordinary Milk- Sugar. The beginning rotation of this can- not be accurately determined, because of rapid change into the yS-form, and is always found too small. Schmoger 3 found [a] D = -(-84 (for the hydrate), and Parcus and Tollens 4 + 83. The following series of experiments by Parcus and Tollens gives a picture of the change into the stable /?-form which fol- lows in milk-sugar solutions at the temperature of 20, the time being counted from the moment of adding water to the finely powdered substance : i Ztschr.. fiir Riibenzucker-Ind. (1879), III, p. 130. " Tralles. ; Ber. d. cheni. Ges., 13, 1918. 4 Ann.Chem. (I,iebig), 257, 170. 6oo CONSTANTS OF ROTATION OF ACTIVE BODIES / = 20 C = 4.841. t= 20 c = 7.064. Time. [k Time. []/>. 8 min. 82.91 10 " 82.56 15 " Si.ti 20 " 79.69 25 min. 78.86 30 " 76.48 30 " 77.14 45 " 73.26 40 " 74.94 i hour 70.04 I'/s hours 65.76 i l , hours 68.57 2 62.17 2 61.70 2 1 /, " 58.97 3 " 57-54 iV " 3 hi 55.84 4 1 /. " 54.32 5 V 53-60 5 " 54.25 6 53-43 7 53-25 7 52.90 24 " 52-53 24 52.53 As opposed to the earlier statements of Hesse, 1 Schmoger shows that the rotating power of milk-sugar for c = 2.372 to 41.536 is independent of the concentration. He finds from the mean of 70 polarizations the constant rotation at 20, for C 12 H. 22 O n 4- H . 2 [a]/, - + 52.53 Exactly the same value is found by Deniges and Bonvans. 3 However, according to experiments of Schmoger, the temper- ature influences the rotating power which sinks as the solution becomes warm. This effect is greater at 20 than at higher temperatures. In the neighborhood of / = 20, [#] /, changes about 0.070 for one degree of temperature as shown by the following table : at p. *?. 9. 10. MO. 20". 21". 32". 33, 35>- 37. -8.307 1.0301 52.76 52.36 51.60 15.950 16.664 1. 0661 1.064253.46 53-50 52.56 52.36 . I 5L48 51.40 24.785 1.0992 i52.56 51-59 ' Ann. Chem. (Uebig), 176, 98. * Ber. d. chem. Ges., 13, 1922. 3 J. pharm. chim., [5], 17, 363: Chem. Centrhl. (1888), p. 6o; v DISACCHARIDES 6oi If the rotation, a t , of a milk-sugar solution has been found at the temperature /, it may be reduced to the value at 20 by this formula : 1020 / By the action of dilute acids, milk-sugar is inverted and glucose and ga lactose are formed, which may be shown by polarization. If, for example, c = 11.9734 and / = 20, there may be calculated from the rotation of glucose, [a] D = 52.84, and of galactose, []/> = 79-73, that of the inverted milk- sugar as ^1__79^73 = 6629 Rindell 1 found [ci] D = 67.57 Ior this rotation, and showed that it is influenced by the temperature according to the following formula : [a] D = 70.608 - o. 152 /. Alkalies diminish the rotating power of milk-sugar : i mol. cryst. milk-sugar -f I mol. Na.,O : t = 20 t c = z, [atonce ......... - -'- 45-5V / 20, c = 3, " after 24 hours- .. - -f- 12.57 j Half -Rotating Milk-Sugar (y-form). This is produced as already explained in 72. The beginning rotation is found too large because of the rapid conversion into the more strongly rotating /?-form. Schmoger 3 obtained for c = 7.07, [or] y , = 4- 34-4 : c = 7-72, [a] i, = -f 36.2. The conversion of a- and also ^-milk-sugar into the constant y^-form is hastened by addition of o.i per cent, ammonia/ Milk-Sugar Octoacetate, C 1;! H U O U (C 2 H 3 O) 8 . This is left-rota- ting and exhibits neither birotation nor half-rotation. Chloroform .............. p about 10, \_cc] D = 3.5 '' MALTOSE, Maltobiose, Malt Sugar, Ptyalose,C 12 H 28 O n +H a O. Crystallizes in fine needles with i molecule of water of crystal- lization, which is lost at 100. The earlier statements on the rotation of maltose gave values between [<*]/> = -f 139 and 150. The first reliable obser- 1 Ztschr. fiir Riibenzucker-Ind., 4, 163. - Hesse : Ann. Chem. (I^iebig), 176, 101. " Her. d. chem. Ges., 13, 1918. 4 Schulze and Tollens : Ann. Chem. (Iiebig), 371, 49. ;> Schmoger : Ber. d. chem. Ges., 35, 1452. 602 CONSTANTS OF ROTATION OF ACTIVE BODIES vations were made by Meissl. 1 He found that maltose, like lactose, shows half -rotation. c. 15.61 18.94 19.4 [>]/, aft 7 hours. er 8 hours. 5 min. i hour. 4 hours. 24 hours. 4 days. 138.3 138.3 138.2 122.6 122.5 122.0 126.2 127.4 127.0 132.0 !33.o 135.0 137-0 J 37- 6 138.0 137-2 138.2 138.3 133.3 138.2 138.3 Parcus and Tollens 2 also investigated the half-rotation in two different solutions, and found : Preparation Preparation 2. / = 20. c = 10.040 for hydrate. c = 9.537 for anhydr. t = 20". r = 9.679 for hydrate. c = 9.195 for anhydr. / = 20". c = 10.320 for hydrate. c = 9.804 for anhydr. Time. [a]/> for anhydride. 1 Time. |for anhydride. 1 ' Time. M'j for anhydride. 8 min. 119.36 12 min. 120.97 6 min. 118.75 10 " 120.27 14 " 121.16 8 119.46 15 " 121.01 16 " 122.29 10 120.34 20 " 121.72 20 " 123.23 12 120.69 30 " 123.35 25 " 124-55 15 " 121.22 45 '' 125.17 30 " 125.68 20 122.28 i hour 128.07 40 " 127.56 30 124.04 i l ., hours I3 .97 50 ' 129.44 50 127.04 2 " '32-97 I hour 130.57 I hour 128.81 2', 134.05 i l / 4 hours 132.63 i i/ 2 hours 131.45 3 134.96 !*/ I34JI 2 133.22 5 " 136.52 2 135-27 3 I35.I6 7 136.72 2'/4 M T 35-65 1 4 136.40 9 136.96 const 3 3 /* " 136.40 6 " 136. 75 const. 5 136.87001151. I The constant end rotation, as with milk-sugar, is more quickly reached by heating the solution or by adding ammo- nia to it. 3 ' J. prakt. Chem., ai, 284 and 35, n.j. - Ann. Chem. (Uebig), 357, 172. Schulzeand Tollens : Ibid., 27 I, 53. DISACCHARIDES 60.^ The constant end rotation is dependent on the temperature and concentration, as shown by observations of Meissl, 1 which may be expressed by this formula : [a-]/, = 140.375 0.01837 P .95 / From this we have the following agreement between obser- vation and calculation : p. 4.95 d at 17.5". 15"- 17.5". 25. 35"- Found. Calc. Found. Calc. Found. Calc. Found. Calc. .01961 138.67 138.86 138.46 138.62 137.97 137-91* 137.08 136.96 9.70 .03928 138.79 138.77 138.54 138.53 137.84 137.82 136.95 136.87 18.65 .0/777 138.56 138.61 138.33 138.37 137-57 137.66 136.75 136.71 18.67 .07779 138.68 138.61 138.40 138.37 137.68 137-66 r 36./9 136.71 19.68 .08281 138.70 138.59 138.30 138.35 137-59 I37.64 136.66 136.69 19.91 .08309 138.50 138.58 138.39 138.34 137-55 I37-63 136.61 136.68 33-86 .14873 138.35 138.33 138.12 138.09 137.33 137.38 136.38 136.43 34.72 .15361 138.26138.31 138.00 138.07 137.29 137.36 136-37 136.41 34-95 .15500 138.34 138.31 138.11 138.07 137.40 137.36 136.48 136.41 Herzfeld 2 gives : Water * = 20, / - 11.29, aff = 1.044, MZ> 4- 138.29 (From Meissl' s formula there is calculated : 138.27). Ost 3 finds: Water . c --- 2 to 21, t = 20, [cc] D = + 137.04 Brown, Morris and Millar 4 also obtain values which closely agree with those of Meissl. Octacetyl Maltose, C 12 H U O U (C,H 3 O) 8 . Crystals ; melting- point, 150 to 155. Benzene =1.9956, \_O\D = 8i.i8 c Benzene p 2 " =75.7 Alcohol ^=2 " = 60.0 Benzene p 2 " =76.54 Chloroform / 2 " = 61.01 Alcohol p r " ^60.02 1 J. prakt. Chem. [2], 35, 114. - Ztschr. fiir Riibenzucker-Ind., (1895), p. 236. ; Chem. Ztg., 19, 1727. 4 J. Chem. Soc., 71, 72. Herzfeld : Ann. Chem. (I^iebig), jao, 218. 6 Herzfeld : Ztschr. fiir Riibenxucker-Ind., (1895), p. 235. ' Herzfeld : Ber. d. chem. Ges.. 28, 441. r 604 CONSTANTS OF ROTATION OF ACTIVE BODIES ISOMAI/TOSE Gallisin, C 13 H 21 O U . This is formed in the yeast fermentation of grape-sugar, 1 or in the treatment of this sugar with hydrochloric acid." The specific rotation in aqueous solution increases approxi- mately in proportion to the amount of solvent employed : Water / .= 20, --54.58, OL = + 77-32 =27.29, "--4-80.10 c= 10.60, " = -f- 82.76 Lintner and Dull* give for a preparation of isomaltose from starch, but which probably was not pure, [or],, == 140 in 10 per cent, solution. TREHALOSE, Mycose, C,,H 22 O U -f- 2H,O. Rhombic crystals ; melting-point, 210. Water c 8.4 to 14.8 / 15 [cr]y = -f- 199 (.= 220 for anhydrous subst.) 5 Water. . . . c 10.03, [L = + 173-2 6 MELIBIOSE, Raffinoboise, C ]2 H 22 O n . Amorphous powder. Water c 3.957, / 17, I -126.7 Water <" 3-973, f !7, " 127.9 These are only approximate values, as we have no guarantee of the purity of the amorphous sugar. 7 Water / -= 17.5, [or]/, 137.32 " Octacetyl Melibiose, C 1V H U O U ( C 2 H 3 O ) K . - :t Alcohol-- '.chloroform c^- 1.6516, / 18, [a]/, +94.2 09 CYCLAMOSE, C 12 H 22 O n . From the roots of Cyclamen euro- pacum. The specific rotation is not affected by the temper- ature, and is [>]/>= -15- 15. On heating with dilute hydrochloric acid, the rotation is in- stantly changed to \oi\ n - 66.54, but under the influence of the heat, soon decreases. 10 1 Schmitt and Cobenzl : Her. d. chem. Ges., 17, 1000. Fischer : Ibid., 33, 3688. '' Schmitt and Cobenzl : Ibid., 17, 1007. 4 Ztschr. angew. Chem., 1892, p. 263. Berthelot : Ann. chim. phys., [3], 55, 276. Mitscherlich : J. prakt. Chem., [I], 73, 65. ' Scheiblerand Mittelmeier : Ber. d. chem. Ges., a3, 1438. Bau : Chem. Ztg., ai, 186. 1 Scheibler and Mittelmeier. '" Michaud : Chem. News, 33, 232. TRISACCHARIDES AND POLYSACCHARIDES 605 20. Trisaccharides and Polysaccharides RAFFINOSE, Melitose, Melitriose, Gossypose. C 18 H 32 O U -f 5H 2 O. Melting-point 118 to 119. It is found in various vegetable products. The rotating power was found by several older observers, in pretty close agreement, as [or],, =: + 104.5, for f = 20, and to be but slightly dependent on temperature. Tollens 1 found that these different products, investigated by the earlier chemists, melitose from eucalyptus manna,- raffinose from molasses, :{ gossypose from cottonseed/ are identical and show the following values : I. Raffinose from molasses : Water /> 9.817, t = 20, ^=1.03278, [a] D + 104.9 II. Raffinose from cottonseed : Water p. 11.158, / = 20, ^* 1.03279, []/?--- 104.39 III. Raffinose from eucalyptus manna : Water p 9.777, t 20, d' = 1.03172, [a]/) = 104.44" Raffinose is found also in the sugar-beet. Water p = 2.7616, ^ = 18, \a\ D 104.96 ' Rischbiet and Tollens give : 6 1. Raffinose from molasses : Water c = 10, []/> = -f 102.4 to 104.9 2. Raffinose from cottonseed : Water c = 10, [a[> = 104.4 Creydt 7 found : Water / = 20, c = 16.6 Raffinose from molasses [a]z> = 104.2 " cotton-seed " ^104.5 In the inversion of raffinose, two stages may be distinguished. When its solution is warmed gently with weak acid, the rota- tion sinks to about one-half, but goes no further ; but, on the Ann. Chem. (I,iebig), 232, 169. Berthelot : Ann. chim. phys., [3], 46, 66, I,oiseau : Compt. rend., 82, 1058. Bohm : J. prakt. Chem., [2], 30, 37. I,ippinann : Ber. d. chem. Ges., 18, 3089. Ann. Chem. (L,iebig), 232, 169. Inaug. Diss., Erlangen, 1888. 6o6 CONSTANTS OF ROTATION OF ACTIVE BODIES other hand, if the solution is warmed a long time with some- what stronger acid the rotation is decreased to one-fifth the original value. Thus, a raffinose solution with c = 10, to which 6 cc. of sulphuric acid of sp. gr. 1.0591 had been added, gave : After ' , hour's heating to 100 ........ []/> 49.8 at 20 " i " " 80 ........ " 49.1 at 20 " 14 days' standing in the cold ..... " ~ 53.6 at 20 But when 10 grams of raffinose was dissolved with 5 cc. of sulphuric acid of sp. gr. 1.156 and water to make 100 cc., and was then heated five hours to 100, it showed [ai] D = -\- 20.07 ' Scheibler has shown that the specific rotation of raffinose does not change with the concentration or by the addition of alcohol. Raffinose from beet-sugar : Water ............ c 10, / ^ 17.5, [a]/, = + 103.74 " ........... =15. ' - !7-5 ^+103.97 Alcohol (15 p. c.).. c ro, / :.= 17.5 = -f 103.9 Raffinose from cottonseed : Water ............ c --- 5 / 17.5 [or]/? = -f 104.0 " ............ C=IO 1=17.5 = + 103.9 " ............ c 15 /^i7-5 ^+103.95 Undecylacetylmeletriose, C 1H H 21 O 18 (C 2 H 3 O) n . Crystals with melting-point 99 to 101. Alcohol ......... c :- 8.1988, / 17, [a]/, - + 92.2 3 MELEZITOSE, C I8 H M O 16 -f- 2H,O. Rhombic crystals ; melt- ing-point, 147 to 148. For aqueous solutions : O]/> + 83 + 0.07014 /> 4 Water ...... [a]/, = + 88.15 :> Acetate, C^H^O^CC.!!.^),,. Monoclinic prisms ; melting- point, 117. Benzene .......... c 6.243, t 20, [a]/> == 110.44 G 1 Tollens : I^oc. cit. ; Scheibler : Her. d. chem. Ges., 18, 1782. * Ij>c. cit. * Scheibler and Mittelmeier : Ber. d. chem. Ges., 33, 1443. 1 Villiers : Bull. Soc. Chini.. 37, 98 ; Alechin : Russ. phys.-ch. Cles., ai, 420. '" Bourquelot, H6rissey : J. phiirni. chitn., [6], 4, 385. * Alechin. CARBOHYDRATES 607 LUPEOSE, /f-Galactan, C a4 H M O 21 -f- H,O(?). Amorphous. A preparation dried in hydrogen at 100 gave : Water p 5, / 22, [a]/, -. - 138 For three different preparations which were dried at 110 to 115, these values were found earlier by Steiger. 1 [ - 148.7, []/, = + 149-8, M/J 147-2 If we assume that the preparation dried at 100 contains one molecule of water of crystallization, the last three results would agree very well with that calculated for anhydrous lupeose.- TREHALUM, C 24 H 42 O 21 , from tiehala. p = 0.261 100.365, / 18, [a]/, ^ - 179 :! GENTIANOSE, C :J6 H 66 O 31 (?). Small plates melting at 210. 4 An aqueous solution, prepared hot, showed- / 18, [<*]/> ^ -f- 65.7 cold, " ... / ----- 18 " ^4-33.36 STACK VOSE, C^H^O.,,. From tubers of Stachys tubifera. Crystals. Water c ~g (amorphous substance), [<*]/>"- -p- 146.7 " ^ = 9-5 ( " " ), " :=+ H8.8 " c 9 (crystalline ), " = -f- 148.1 21. Carbohydrates, (C 6 H io O.) ;; AMORPHOUS SOLUBLE STARCH. From potato starch by heat- ing with glycerol, water, dilute acids, etc. 6 Water, c = 2.533, I>L == + 206.8 ' " c = 2.215, t = 17.5, [nr]y= 211.50, from which [a]^=z 189.98 " ^^3.995,^^17.5, " -211.97, " -190.24' Other authors give the rotation as [**]/ = : -f 216 ; 9 [] y + 211 ; 10 \_a\j = -. -f 216, O] /) =: + 202. u I^andw. Vers.-Stat., 36, 423 and Ztschr. physiol. Chem., n, 372. Schulze : Ber. d. chem. Ges., 35, 2218. Scheibler, Mittelmeier : Ibid., 26, 1331. Meyer: Ztschr. physiol. Chem., 6, 135. Planta and Schulze : Her. d. chem. Ges., 33, 1692 : 34, 2705. Zulkowsky : id it/., 13,1398: Bechamp : Compt. rend., 39, 653 ; Musculus, Gruber: I bit , 86, 1459. Zulkowsky : Lvc. cit. Salomon : J. prakt. Chem., [2], 28. Musculus : Jahresber. (1879), p. 835. Bechamp : Compt. rend., 39, 653. Brown, Morris, Millar : J. Chem. Soc., 71, 114, 608 CONSTANTS OF ROTATION OF ACTIVE BODIES DEXTRINES. Crystalline Soluble Starch, Amylodextrine. Formed by the action of cold dilute hydrochloric acid or of hot 90 per cent. acetic acid on starch, and separated in crystalline form by freezing the solution. 1 Brown and Morris 2 have determined the molecular weight as C 12 H 22 O n -f 6C 12 H 20 O 1()) and give the rotation as otj = - -f- 206.5. Achrodextrine. According to the statements of different authors, there are several isomeric dextrines formed under dif- ferent conditions from starch. Musculus and Gruber 3 dis- tinguish three kinds : fr-Modification [a] - -f 210 /S-Modification " : = -j- 190 7-Modification " - = -f 150 According to O' Sullivan, 4 there are four different varieties which, however, have the same rotating power. This is given as [a] y = + 215 to 217. Brown and Heron 5 give [or],- = -f- 216. L. Schulze 6 found the rotation of a carefully purified dex- trine as [a] D = + 1 86. Lustgarten 7 found for a dextrine obtained from dinitro- glycogen,/ == 12.175, t= 27, [>]/,= + 194. The rotating power of dextrine is increased by acids. 8 Water c 3.714, /^i7-5i []./ = + 215.06 Sulphuric acid, 0.4 p. c. = 1.064, t~-- J 7-5 " a t once) = -f- 216.5 [or]y (following day) = -f 220.5 An achrodextrine, (C^H.^OJ. -f H 2 O, []/,==+ 183 is described by Lintner and Dull. 9 Maltodextrine. By treatment of starch paste with diastase above 65, [>1 = + 169.9 to 173.4 ; in [ar] y = + 164.2." 1 Musculus : Ztschr. Chem., (1869), p, 446. 2 Brown and Morris : J. Chem. Soc., 55, 452. 8 Ztschr. physiol. Chem., a, 188. * J.Chem. Soc., 35, 770. *> Ann. Chem. (Mebig), 199, 243. J. prakt. Chem.. [2], a8, 327. 7 Monatsh. Chem., a, 626. 8 Salomon : J. prakt. Chem., [2], a8, 42. Ztschr. f. Brauwes., 17, 339. '" Herzfeld : Ber. d. chem. Ges., ia, 2120. 11 Bondonneau : Bull. Soc. Chim., 35, 5. TKISACCHARIDES A>'D POLYSACCHARIDKS 609 Extended investigations on the action of diastase on starch, and especially on malto dextrine, have been carried out by Brown and Morris, 1 also Brown, Morris and Millar. - Artificial Dextrine. From dextrose, sulphuric acid, and alcohol. [a]/j - 131 to 134 :! OL -i23 01 Fermentation Gum, dextran, viscose. Produced in the lactic fermentation of cane-sugar. [a]/, 230 ;> fi-Galactan. From Lupinus luteus. Water C- 10, |>]/, 148.7 K " f 3.082, / 15, [>]/, -i 4 8.6 07 Shows no birotation. The rotation changes very little with the concentration. y-Galactan. In the sugar-beet. Water c 10, / 20, \ci\,> --= 4- 238 B x-Galactin. In the seeds of leguminous plants. []/> + 84.6 GLYCOGEN, C 8 H 10 O 5 , or C M H 62 O 3l . Found in the liver of man and the herbivora. The older statements on the rotation of glycogen vary within very wide limits. Careful observations were made by Kiilz. 1 ' 1 He states that the concentration is without special influence, and he finds as the mean of eighteen readings [],- -)- 211 (greater concentrations than/ 0.6 cannot be employed). Landwehr 11 found for the glycogen of dog's liver, for t= 18, [],, = -}- 213.3. Further values are: []/, =+ 196. 33 " 200.2 1:J 1 Ann. Cheni. (L,iebig), 231, 72 : J. Chein. Soc., 55, 452. - J. Chem. Soc., 71, 72, 109, 115. Musculus and Meyer : Ztschr. physiol. Chem., 5, 122. * Konig and Schubert : Monatsh. Chem., 6, 746 : 7, 455. ^ Stohmann and r y angbem : J. prakt. Chem., [2], 45, 305. Steiffer : Ber. d. chem. Ges.. 19, 829. : Schulze and Steiger : I.andw. Versuchs-Stationen. 36, 4^3. L,ippmann : Ber. d. chem. Ges., 20, 1001. '' Miintz : Compt. rend., 94, 453. '" Jahresber. f. Thierchemie (iSSo), p. 81 ; Arch. f. ges. Physiologic. 24, 35. 11 Ztschr. physiol. Chem., 8, 171. 12 Huppert : Ber. d. chem. Ges., 27, Kef. 85. 13 Kramer : Ztschr. fur Biol., 24, 100. 39 6io CONSTANTS OF ROTATION OF ACTIVE BODIES CELLULOSE. According to Levallois 1 this is left-rotating in solution of ammoniacal copper hydroxide. But Bechamp 2 gives the direction as variable. GRAMININ. Found in various plants (as Trisctiim alpcstrc} . White powder, melting at 209. Water /> - 5, / 12, []/>-- - 38.89 :i INULIN. Found in the roots of different plants. Spheroidal crystalline grains ; melting-point, 160. The older values for the rotation vary between [#] = 26 and 72.42. Kiliani 4 found : Inulin from Dissolved in p. M* Inula Helenium ( Water at 100 diluted with } ( cold water. ) 1.5122 23 - 36.66 Dahlia variabil l A little potassium hydrox- ) ( ide and then diluted, j 0.9172 1.7456 20 20 - 37.15 - 34.10 Lescoeur and Morelle' found : Water [a]/, 36.94 Haller 8 gives : Water c 7.25, [>]/, -38.8 Temperature and concentration appear to have no influence. 7 Water p 2, [a]/> 37.27 * Inulein. Water c 6.6, []/, - 29.6 !1 Pseudo-hnilin. Water c 6.64, [a]/, - 32.2 "' The last two substances were obtained from Helta ti thus tuberosus. Bull. soc. chim., 43, 85. Ber. d. chem. Ges., 18, Kef. 113. Ekstrand and Johanson : Ibid.,, 21, 594. Ann. Chem. (Uebig), 305, 145. Bull. soc. chim 33, 418. Compt. rend., 116, 514. Haller. Wallach : Ann. Chem. (Uebig), 334, 368. Haller: Compt. rend., 116, 514. Haller : I.oc. < it. GUMS 6ll IRISIN, Phlein. In the tubers of Iris pseud-acorus. Melting- point, 218. Water /> 10, / 16, []/> = 51-54 /> 10, *^i6, 51.15 P= 5, '= !6, -51-55 /> 2, / 22, - 49.90 From Iris p 5, [(*]/> 52.34) 2 From Phleum p 5, 48.12 J LEVOSIX. In varieties of grain. Melting-point, 160. Water p =, 5, [a]/, - 36 3 LEVULAN. In beet-sugar molasses. Melting-point, 250. Water c 5 to 30, / -- 20, [a]/, - 221 * The temperature is without influence. SINISTRIN. In the sea onion. []z> - -4i.4 08 - 34-6 fi TRITICIN. In the roots of Triticum repens. []/> = -43-6 07 / 5. - 41-07 8 22. Gums ARABIN, Arabic Acid, C^H^OgC?). The principal constit- uent of gum arabic which, however, is a mixture of at least two gums, a right- and a left-rotating variety. Pure arabin may be made from sugar-beets. [>]/, = -98-5 09 WOODGUM, Xylan, C 5 H & O 4 (?). In foliage trees, especially in the birch. /> i to 2. [or]/, - 84 "' [a]/> usually = - 70 to 85 " Wallach : Ann. Cheni. (I]/> -f 211.7 l 23. Camphors and Terpenes In the nomenclature and classification of the following com- pounds, the fundamental work of Wallach has been taken as the guide. The terpenes are therefore divided into four groups : A. Aliphatic terpenes, B. Terpan group, C. Camphan group, D. Poly terpenes, and at the beginning of each group the hydrocarbons will be treated, then the alcohols, ketones, and so on. A. ALIPHATIC TERPENES i. Hydrocarbons LICARENE, C 10 H 16 . Boiling-point, 176 to 178. d 0.8445, t 20.2, [>]/, 4- 7.85 s 2. Alcohols ^-LiCAREOL, Coriandrol, C IO H,.OH. Boiling-point, 196 to 198. if" 0.8820, []^" r 15-02 4 /-LiCAREOL, /-Linalool, Aurantiol, Lavendol, Nerolol, C IO H 17 OH. Boiling-point, 199 to 200. d" 0.8819. d" 0.868, [a]^ -19 IManta and Schnlzc : Her. d. chem. r.es., 23, 1692. Harhier : Compt. rend., 116, 993. Barbier: Ibid., 116, 1459. Barbier : Hull. soc. chim., [3], 9, 914. Barbier : Compt. rend., 114, 674. Morin : I hid., pa, 998. CAMPHORS AND TERPENES 613 According to Barbier, licareol and linalool are not identical. He gives : I- Linalool. Boiling-point, 98 to 100 (14 mm.) d" 0.8869, []*?< - 11.91 > Boiling-point, 86 to 87 (14 mm.). d' M 0.8622, []=; - 19.62 - i.9 3 From citronellal. (Formula given : C 10 H 19 OH.) Boiling- point, 117 to 118 (17 mm.). rf'7-s 0.8565, [>].5 = -j- 4.0 Acetate, C^H.,,,0,. Boiling-point, 119 to 121 (15 mm.). z= 0.8731, [a] D .-- - 2.11 ! t/ 20 0.8612, [] = 4-33 2 Formula given as C 10 H 19 OH. From Bulgarian rose oil. d = 0.8785. Boiling-point, 224.7. [*]*= -3.22' From French geranium oil. d = 0.8886. Boiling-point, 124 (14 mm.). []:_- -2.5* 3. Aldehydes CITRONELLAL, Cj H 18 O. From melissa and citronella oils. Boiling-point, 205 to 208 (103 to 105 at 25 mm.). =-. + 4 .8 0(i B. TERPAN GROUP i. Hydrocarbons */-LIMONENE (Citrene, Hesperidene, Carvene), C 10 H 16 . In oils of orange peel, citron, bergamot, cumin, erigeron, and dill. Boiling-point, 175 to 176. d =0.846. Chloroform ... p = 14.38, rf 8 -..= 1.353, [], = -f 106.8 ' Derivatives: Hydrochloride, C 10 H 16 HC1. Boiling-point, 97 to 98 (n to 12 mm.). rf 17 ' 8 = 0.973, [nr]jj- - + 39-5 8 Tetrabromide , C, H 16 Br 4 . Rhombic-hemihedral crystals ; melting-point, 104. Chloroform . . . p 14.24, d 9 = 1.555, [a]^ a= -f 73.27 9 a-Nitrosochloride, C 10 H 16 NOC1. Crystals ; melting-point, 103 to 104. Barbier, Bouveault : Compt. rend., laa, 529. Ticmann, Schmidt : Ber. d. ch^m. Ges., 39, 923. Markownikoff, Reformatzky : J. prakt. Chem., [2], 48, 299. Monnet, Barbier: Compt. rend., 117, 1093. Tiemann, Schmidt : Ber. d. chem. Ges., 39, 905. Dodge : Ibid., 33, Ref. 175 and 34, Ref. 90. Wallach, Conrady : Ann. Chem. (Uebij?), 353, 144. Wallach : Ibid., 270, 189. Wallach, Conrady : Ibid., 353, 145. CAMPHORS AND TERPENES 615 Chloroform ---- p= 21.13, d r> r ^ J-379. []/? = Chloroform.... /> = 13.30, = 5.35, d u = 1.445 M/j == + 102.19 From /?-nitrosochloride : Chloroform ---- p = 7.071, rf 19 ' 2 = 1.439, l<*~\%'' = + 102.25 Nitroso-a-nitrolanilide, C lt H M (NO)[N(NO)C t HJ. Crystals; melting-point, 142. Ether or benzene ? ........ p - 4.208, ^ 19 ' 8 0.805, [tr]^- 8 = -f 46.20 p-Nitrolanilide. Matted needles ; melting-point, 153. From -nitrosochloride : Chloroform ---- p = 5.086, (f 24 =1.447, []* = 88.33 From /?-nitrosochloride : Chloroform- . . . p = 5.133, rf 19 ' 4 = 1.447, []^" = - 89.39 5 a-Nitrolbenzylamine, C 10 H 16 (NO)NH.CH 2 C 6 H 5 . Needles ; melting-point, 93. Chloroform ................ p = 7.027, ^ 9 ' 5 =1.459, C^]^ 5 =+i63.8 6 Hydrochloride in dil. alcohol/* = 3. 975, rf 10 =0.906, [a\ - - 82.26 Nitrate " " " p = 1.034, d n =0.900, [n:] 1 ^ = - 81.0 ^/-Tartrate " />= 1.133, rf 1 *- 5 = 0.900, [<*]*= - 49-93 /-Tartrate " / = 0.968, rf 10 ' 5 = 0.899, [o:]-s= - 69.9 Hydrochlor-nitrolbenzylamine, C 10 II 16 HC1(NO)NHCH 2 C 6 H 5 . Xeedles ; melting-point, 103 to 104. Chloroform ..... p = 2.403, rf 18 ' 5 = 1.47, [nr]g-s = -f 149.6 7 1 Wallach : Ann. Chem. (lyiebig), 246, 224. - Wallach, Conrady : Ibid., 252, 145. 3 Wallach, Conrady: Loc. cit. Macheleidt : Ann. Chem. (Uebig), 270, 176. Wallach : Ibid., 270, 171. 6 Wallach, Conrady : Ibid., 252, 148. ' Wallach : Ibid., 270, 192. 6l6 CONSTANTS OF ROTATION OF ACTIVE BODIES a-Nitrolpipcridinc, C 10 H I6 (NO)NC-H 10 . Rhombic crystals; melting-point, 94. Chloroform ..... /> 3.146, rf" 1.475. [a];; --4-67.75 fi-Nitrolpiperidine. Monosymmetric crystals ; melting- point, 110. From oMiitrosochloride : Chloroform ..... /> 3.107, 1.478, [cjr]*s - 60.48 From /f-nitrosochloride : Chloroform ...... /> 2.1104, d r - 1.478, ["]'/; = 60.37' Carvoximc, Isonitrosoterpene, C 10 H U NOH. (By splitting off HC1 from a- or /f-nitrosochloride. ) Crystals ; melting-point, 72. Alcohol ........ p 4-328, d 0.8025, / 18, [a-]/, -39-34' Benzoyl Compound. Crystals ; melting-point, 96. Chloroform ..... p 5.716, d n 1-4455, O]p - 26.97 :>> Benzoyl-hydrochlorcarvoxime, C IO H H NO.CO.C 6 H,.HC1. Crys- tals ; melting-point, 114 to 115. Acetic ether ...... p 1 1.866, rf 2 " 0.926, [a];; - 9.92 4 /-LiMONENE. In pine needle oil. Boiling-point, 175 to 176. d = 0.846. Alcohol ....... p 6.126, d? 0.795, [a-]/, - 105.0 r> Chloroform.... p 14.3 rf 10 ' 5 1-353, * 10.5, [a] 10 -' - 105 fi Derivatives: Hydrochloridc. Boiling-point, 97 to 98 (11 to 12 mm.). Tetrabromide . Rhombic-hemihedral crystals ; melting- point, 104. Chloroform ...... p 12.85, 0.993, d* 1.496, [a~\J, 314.8 1 Wallach, Conrady : Loc. cit. Wallach : Ann. Chem. (Uebig), 346, 227. 3 Wallach, Conrady : IMC. cit. 4 Wallach, Macheleidt : Ann. Chem. (Uebig), 370, 179. '- Wallach : Ibid., 246, 222. ' Wallach, Conrady : I^oc. cit. ' Wallach : Ann. Chem. (Uebig), 370, 189. * Wallach. Conrady : /r. cit. CAMPHORS AND TERPENES 617 ft-Nitrosochloride. Wooly needles ; melting-point, 100. Chloroform /> 0.99*8, d*- :> 1.495, ["]^, 5 242.2 ' Benzoyl Nitrosochloridc. Crystals; melting-point, 109 to 110. Acetic ether. ../> 4.828. d- :> 0.911, []-s 101.84'- ct-Nitrolanilide. Monosymmetric plates ; melting-point, 112 tO 113. Chloroform p 7-344, d' 4 1-437, [>];? -102.62 Nitroso Compound. Crystals ; melting-point, 142. Chloroform p 4.291, 0.899, [a]-s - 83.06 Nitrate /> 1.019, a" 1 "' 0.898, [a]g* .= .-f 81.0 ^/-Tartrate /> 1.378, 0.902, [n-]jf-s = -f- 69.6 /-Tartrate p 1.119, ^ n 0.901, []'/, = -f 51.0 Hydrochlornitrolbenzylamine. Needles ; melting-point, 103 to 104. Chloroform p 2.431, cf ] ^ :> 1.469, / 18.5, [n:]g-5.= - 147.4 5 u-Nitrolpipcridine. Rhombic crystals ; melting-point, 94. Chloroform / 3-"3, rf 11 ' 7 -- 1-475, [^V/\' 7 "-'- -67.60 fi-Nitrolpipctidine. Monosymmetric crystals; melting-point, 110. Chloroform p 3.051, + 39-62). Solutions in chloroform. . P- ] 66.32 7 Wallach, Conrady : IMC. cit. Wallach: Ann. Chem. (Uebig), 370, 179. Ztschr. phys. Chem., 14, 398. Her. d. chem. Ges., 30, 2069. Atterberg : Ibid., 10, 1206. Tilden : J. Chem. Soc., 33, 80. Wallach. Conrady : Ann. Chem. (L,iebig), 253, 149. CAMPHORS AND TERPENES 619 Derivatives : Dihydrochloride, C 10 H 16 , 2H Cl . Monosymmetric crystals ; melting-point, 72. Chloroform /> 14.20, d % = 1.4235. O]^ = + l8 -99 Dihydrobromide , C 10 H 16 ,2HBr. Monosymmetric crystals; melting-point, 72. Chloroform />= 4-359. ^' 5 J -499, M& s = + 1 7^9 Tetrabromide, C 10 H 16 Br 4 . Monosymmetric crystals ; melting- point, 135. Chloroform p = 4-338, d?' h = 1.517, [a]*? = -f 73.74 Nitrolbenzylamine, C 10 H 16 NONH.C 7 H. Crystals ; melting- point, 71. Chloroform p= 1.908, d*'* = 1.495, []% 5 = - 185.6 Nitrolbenzylamine Hydrochloride, C ]0 H 16 NONHC 7 H.HC1. Crystals. Dil. alcohol p= 1.571, d~* = 0.904, [a]^s = + 79.2 ] /-SYLVESTRENE. From Finns Abies. Liquid ; boiling-point, 170.3. d = 0.8664, [nr]< = - 18.3 - ^/-PHELLAXDRENE, C 10 H 16 . In bitter fennel oil, elemi oil and resin. Liquid ; boiling-point, 171 to 172. 64 o 5 Derivatives : Diphellandrene , C M H 32 . Made by heating phellandrene for twenty hours to 140 to 150. Melting- point, 86. Chloroform ^^=5.65, [>]/> = -f- 82.9 4 Nitrite, C 10 H 16 (NO)NO 2 . Melting-point, 94. Chloroform []/>= - 183.5 5 t/-MENTHENE (hydromenthene) , C 10 H 1S . By splitting off water from /-menthol. Liquid ; boiling-point, 167.4. d = 0.8073, [a] = + 10.66. ([a]y =4- 13-25) 6 Wallach, Conrady : Loc. cit. Kuriloff : J. prakt. Chem., [II], 45, 126. Pesci : Ber. d. chem. Ges., 19, Ref. 874. Pesci : Loc . cit. Pesci : Loc. cit. Atkinson, Yoshida : J. Chem. Soc., 41, 53. 620 CONSTANTS OF ROTATION OF ACTIVE BODIES Boiling-point, 167 to 168. d' M 0.814, []- Jf 26.40 ' Boiling-point, 167 to 168. []/> 53-53 -' /-MENTHENE. By splitting off hydrochloric acid from the right-rotating menthyl chloride. Liquid ; boiling-point, 170 to 171. (T M 0.816, [ -0.8845, [ar]= - -50.59 For menthol of American origin (Michigan), Long gives the following values, holding for q == 30 to 92. fi Melted d^- 6 0.8810, t = 46, [>]/> 49.86 Alcohol / 20, [a]/> 48.247 o.oi 1 108 q 0.00001870?* Benzene / 20, " 49.511 | 0.025634 ? 0.0008403 g* -f- o.oooo 1 1 02 ? 5 Glac. acet. acid / 20, " 47.711 -0.006386? 0.000071 42?' Derivatives : Carbonate, (C IO H I9 ).,CO 3 . Mother-of-pearl-like crystals ; melting-point, 105. Benzene p 2.021, [a];j - 92.52 ~ Urethane, C 10 H 19 O.CO.NH,. Rhombic needles ; melting- point, 165. Chloroform p 0.58, [^J^j 85.11 ' Sucdnic Add Monoester, COOH.C.,H 4 .COOC IO H 19 . Crystals; melting-point, 62. Benzene p ~ 1.375, O]/V 59-63 !l Sicker, Kremers : Am. Chem. J., 14, 291. Slavinsky : J. russ. chem. Ges., 39, 118. Berkenheim : Ber. d. chem. Ges., 35, 690. Arth : Ann. chim. phys., [6], 7, 438. Beckmann : Ann. Chem. (Iiehig), 350, 327. J. Am. Chera. Soc., 14, 149 ; Chem. Centrlhl.. is.^2, II, 525. Arth : Loc. cit., p. 470. Arth : Loc. cit., p. 464. Arth : Ijoc. cit.. p. 4Sj. CAMPHORS AND TERPENES 621 Succitiic Add Dicster, C 2 H 4 (COOC 10 H 1!( ) L ,. Rhombic octa- hedra ; melting-point, 62. Benzene p 1.87, [] 81.52 ' PhthalicAcidMonoester, C fi H 4 (COOH) (COOC 10 H 19 ) . Micro- scopic needles ; melting-point, 110. Benzene p 1.575, Oil? ~ IO 5-55 2 Phthalic Add Diester, C tt H,(COOC 10 H 19 ),. Rhombic crys- tals ; melting-point, 133. Benzene p 2.006, [ b. Solid ; melting-point, 32 ; boiling-point, 220. In melted condition d" 0.9533, \. a ~\D ~ - 80. Alcohol [>]/, = 92.32 r> Melting-point, 32 ; boiling-point, 215 to 218. | = - 117.5 7 Derivatives: Formate, C 10 H,.OCOH. Liquid ; boiling-point, 135 to 138 (40 mm.). d" 0.9986, [or],, = - 69.25 H ISOPULEGOL. Boiling-point, 91 (13 mm.). 2.77, d r > -1.0022, [n:]^ = -f 17.24 Ether p = 1.71, d* -0.73, M, :-= + 8.34 The rotation of the solution does not change on standing. Hydrobromide. Small needles ; melting-point, 224. Water p . .-. 1.30, d" -1.03, [] : 13.83 Ether p =- 1.36, rf 11 * = 0.729, [ 2.75, rf 14 "' = 1.009, O]/?' 5 = + ii-79 Formyl Compound, C U ,H 1M NH.COH. Crystals; melting- point, 116 to 117. Acetic ether. . . p = 1.83, d v - ----- 0.9132, [ - r.459, [] 3 /, ^ + 53.96 Methyl alcohol p = 7.16, d" =0.812, [a] 63.30 Acetyl Compound, C 10 H 19 NH.COCH :r Prisms ; melting- point, 166 to 167. Acetic ether... p = 1.77, d r ' = 0.9124, [cr]^ = -f. 44.71 " ... p -1.42, d"< -0.9132, [a]^ z = T-45-48 Chloroform ..-/> 4.40, fl" 1.468, [o:]^ 50.57 ... p . 1.89, = 1.78, d } "" : ' 0.9114, [ = 33 ' Boiling-point, 205. d~ = 0.860. Without solvent Mj> = - 3 8 -o? 2 Alcohol -^=11.269, [<*]>, 31.90 :1 Derivatives : Hydrochloride. Crystals ; decompose above 270. Water p --= 2.99, rf 1!l = i.oor, [a] = -j- 35.66 / -3-2 5.22, d** -31.4555, M 8 /, 5 - - 82.97 .../>= 1.39, rf*- 3 : 1.486, M 8 /, 5 -z - 82.09 Methyl alcohol/ 7.44, d -: 0.8131, [a]))' ~ 83.43 Acetyl Compound. Crystals ; melting-point, 143 to 144. Acetic ether... p ^2.16, d 0.9118, M/} - 76.27 ... p - 2.06, ^ l:> 0.9117, [ 5.36, d* 1.4525, [tf]9 ; -81.73 />- -5-34, ^ 10 1-4515, My; -81.90 .... > 1.48, f/ IJ = 1.483, [tr]5, 82.29 Methyl alcohol /> 2.52, d* 0.805, [ a l 9 > 83.64 " /> -7.38, rf 10 -0.8133, C3s 85.67 Propionyl Compound. Crystals ; melting-point, 88 to 89 Acetic ether... p 2.13, d" 0.911, []J3 - 67.26 Chloroform /> 5.^, d* r- 1.462, []5, - 67.53 Methyl alcohol /> : 8.93, rf = 0.8148, [>], 78.02 l\thyl alcohol . . p = 2.6, rf 0.8045, [tr]^ 76.02 1 Andrew. Audreeff : Bcr. d.chem. Ges., 35, 620. * Wallach, Bin/ : Ann. Chem. lUebij?), 376, 323. 3 Bin/. : /tuchr. phys. Chem., la, 728. CAMPHORS AND TERPENES 625 Butyryl Compound. Crystals; melting-point, 80. Acetic ether... p ^ 2.22, d vi -^0.9122, [] -63.58 .../ = 2.i9, rf 12 ' 5 0.9119, [^]yf 5 = 6 4-75 Chloroform.... /> -4-47, rf 4 =1.464, []/, -72.10 " ..../> ,2.69, rf 4 -1.479. M 4 ,, -70.87 The above determinations all by Binz. 1 The base, C MI H,,N 2 C1, has been made by Wallach by treat- ment of iso-/-menthonoxime with PC1 5 in chloroform solution. Melting-point, 59 to 60. Alcohol .... p-= 2.17, rf 20 - 0.7975, [*] - - l86 -35 J 4. Ketones = 0.8220, [a]^ = - 40- 7 to - 42 " ..../ = io, ^ = 0.7998, M^ -42.51 1 Ztschr. phys. Chem., 12, 727. Wallach : Ann. Chem. (L,iebig), 278, 306. Beckmann : /did., 250, 338. Beckmann : Loc. cit. Negoworoff : Ber. d. chem. Ges., 25, 620. Beckmann: I.oc. cit. Beckmann, Eickelberg : Ber. d. chem. Ges., 29, 41*. Atkinson, Yoshida : J. Chem. Soc., 41, 50. Beckmann : Loc. cit. Binz : Ztschr. phys. Chem., 12, 727. 40 626 CONSTANTS OF ROTATION OF ACTIVE BODIES Oxime Hydrochloride. Crystals ; melting-point, 1 18 to 119. Alcohol p 10, d M 0.8175, [or]* 61.16 ! Isooxime. Melting-point, 119 to 120 ; boiling-point, 295. Alcohol p = 24, d' 21 ac 0.827, Mp ~ 52.25 - TANACETONE (Thujone?) , C, H 16 O. In the oil of Tanacetum vulgare and other oils. Liquid ; boiling-point, 84.5 (13 mm.). d w = 0.9126, /> for 2 dm. = -f- 38.5, from which [<*]/> ~- -f- 21.1 ! PULEGONE (Puleone), C 10 H 16 O. From Mentha pulegium (polei oil). Liquid with boiling-point, 222 to 223. ]*3 = + 25.35 4 d w = 0.9323, [>]= -f 22.89 '"' Derivatives: Oxime, C 10 H 19 NO.,. Needles ; melting-point, '57. Alcohol p 10, d'* 0.7998, [ Oxime. See under derivatives of /-limonene. /-CARVONE. In curled mint and kuromoji oils. Liquid; boiling-point, 223 to 224. d*> = 0.959, [or] = - 62.46 * Derivatives : Hydrogen Sulphide Carvone. Crystals ; melting- point, 187. Chloroform c JO, [<*]' = 5-5 1 Oxime. See under derivatives of ^/-limonene. , C 13 H 20 O. The odoriferous principle of the violet. Liquid ; boiling-point 144 (16 mm.). d* = 0.939, [a]/, == about -f 42.6 * C. CAMPHAN GROUP i. Hydrocarbons ]./ = + 22) * . From camphor dichloride. For the fused substance at t 99.8, d ~ 0.8345, and for t = 83.5, a D = -f 55.1 in a i dm. tube. 6 If d** be calculated from the data given below for /-camphene, we have : = 0.8482, [a]3o = + 64.84 Melting-point, 57 to 59. 1 Beyer. 2 Tiemann, Kriiger : Her. d. chem. Ges., 26, 2680. * Wallach : Ann. Chem. (Uebig). 330, 234. * Kachler: Ibid., 197, 96. * Berthelot : Jahresber. d. Chem. (1862), p. 441. Spitzer: Ann. Chem. (Uebig), 197, 129. " Montgolfier : Compt. rend.. 85, 286. 628 CONSTANTS OF ROTATION OF ACTIVE BODIES c. From bornyl chloride. a D - -\- 19.94 f r fused sub- stance in i dm. tube at 85. If we take the specific gravity as for 1-camphene, we have : ]/> + 7-2 ' I sobutylcamphene , C 10 H 15 .C 4 H 9 . Liquid ; boiling-point, 228 to 229. d- -0.8614, t 21, [ -= -f 7.4 5 /-CAMPHENE, Terecamphene. a. From /-turpentine hydrochloride and alcoholic potash. Melting-point, 45 ; boiling-point, 160. \a\ D = -50.4, (OL- -63)" Melting-point, 45 to 48 ; boiling-point, 156 to 157. For the fused substance at the temperature /, ]/, - 51.34 - b. From citronella oil. Liquid ; boiling-point, 160. rf i:> 0.864, [a]/, = - 67 s c. From kesso oil. Liquid; boiling-point, 159 to 161. rf 15 = 0.871, \_O\D - 70.4 " Derivatives: Hydrochloride, C 10 H 16 .HC1. Solid; melting- point, 147. Alcohol p - 10.5, [a]/, = -f 30.25 10 1 Kachler : Ann. Chein. (I y iebig), 197, qy. Bouchardat : Compt. rend., 117, 1094. Bouchardat : IAC. cit. Spitzer : Ann. Chem. (I^iebig), 197, 135. Spitzer. Bcrthelot : Jahresber. d. Chem. (1862), p. 457. Kiban : Ann. chim. phys., [5], 6, 357. Bertram. Walbaum : J. prakt. Chem., [2], 49, 17. Bertram, Walbaum : Loc. cit. Riban Ann. chim. phys., [5], 6, 360. CAMPHORS AND TERPEXES 629 Formate, C 10 H 16 .CH 2 O.,. Liquid; boiling-point, 125 (40 mm.). d" 1.0276, [tr]/> = ~ 10.3 ' Acetate, C 10 H 16 .C,H 4 O 2 . Liquid ; boiling-point, 123 to 127 (35 mm.). = - 119 :! /J- Camphene Phosphonic Acid, C 10 H,. . PO 3 H.,. Melting-point, 170. Ether ....... [or]/, = - 71 4 , ^-Terebenthene, Australene, C,,,H 16 . Isobtained: a. From American oil of turpentine {Pin us Australis, P. taeda}. The commercial oil shows extremely variable rotation which is, in many cases, due to the presence of /-pinene from the southern spruce pine. For the common oil not known to con- tain the spruce product, there was found [] D -\- 9 to -f- 29. The rotation 5 decreases regularly on fractionation. d-" = 0.9108, [] = -f 14.15 rt b. From Russian oil of turpentine (Pinus sylvestris, P. Abies} . The commercial oil shows []/, -f- 11.5 to 17 (and some- times higher). For the pure ^/-pinene there is given : Boiling-point 156.5 to 157.5- ^'" -^0.8631, \ci\* =^36.3 OT " 155-5 " I56.5 --- d^ 0.8547, [r]^-5 = + 32.4 OK " 155-5 " I56.5 --- d* -0.8600, [aj- = 32.0^ 11 161 ............. [<*]D= -r I/-I - I9-4 010 156 (corr. ) at 753 mm. d 0.8746, d* = 0.8585 M^ 8 -^-45.04 " From cumin oil, b. p. 157 to 158 ...... d 1 0.8404, [>]/> = 29.46 '- I^afont: Ann. chim. phys , [6], 15, 149. I.afont. Marsh, Gardner: J. Chem. Soc., 65, 36. Marsh, Gardner. Ivong: J. Anal. Appl. Chem.. 7, 99 (1893) ; Chem. Centrbl., 1893, i, 835. I,andolt : Ann. Chem. (l,iebig), 189, 315. Atterberg : Ber. d. chein. Ges., 10, 1203. Flawitzky : Ibid., n, 1846. Flawitzky : Ibid., ao, 1956. Berthelot : Ann. chim. phys., [3], 40, 5. 11 Flawitzky : J. prakt. Chem., [2], 45, 115. ''- Wolpian : Pharm. Ztschr. fiir Russia nd, 3s, 145 630 CONSTANTS OF ROTATION OF ACTIVE BODIES Rimbach 1 found for the influence of different solvents 011 turpentine oil having [] = = -f 34.81: Alcohol ? = 10 to 100, / = 2o, [= 34.851 -39 Ir /^ ]/> = 35.63 Glac. acet. acid. . . q 10 to 100, / = 20, [<*]/> = 34.889 -f 0.0017465 q + 0.00033528 q 1 , from this for c 20, [or]/; = 36.90 These figures are given for the rotations of the oil of tur- pentine in mixtures of alcohol and glacial acetic acid : Comp. of the mixture. Turpentine oil in 100 parts of solution . Per cent. d? of the sol. [] of the turp. oil. Glac. acid. Per cent. Alcohol. Percent. 84.8 15-2 20.2 0.9653 36.79 70 30 20.3 0-9349 36.63 50 50 20. 6 0.8949 36.44 30 70 20.4 0.8578 36.15 15 8 5 20.4 0.8298 35-97 Derivatives: Hydrochloride, C 10 H 1(f .HCl. Crystals; melting- point, 125. Alcohol p - 28.7, d" 0.8496, [>l = + 30.96 " p= 12.24, rf 1 * = 0.8147, []* +31.23 The specific rotation calculated from this, independent of the solvent, is : [= -i 28.79 ' 2 According to Wallach and Conrady' 1 the hydrochloride and the hydrobromide are inactive. Dtbromide, C, H )6 Br 2 . Liquid. = : + 7.17 as the value obtained in experiments with the product from American oil of turpentine. When the Ztschr. phys. Cheni., g, 701. Flawitzky: J. prakt. Chem., [2], 45, 118. Ann. Chem. (Uebig). 353, 156. Flawitzky : IMC . < it. Wolpian : Ijoc cit. J. Am. Chem. Soc., ai, 637 (1899). CAMPHORS AND TERPENES 631 nature and sources of the American oils are considered, it is evident that this value cannot be constant. An explanation is thus given for the discordant results of different observers on this point. /- PINENE, Terebenthene. a. French oil of turpentine {Pinus pinaster, Pinus mari- tima). The commercial oil shows []/> = - 25 to 43. b. Venetian turpentine oil {Pinus larix). For the com- mercial oil [a] D = - 4.2 to 4.8. c. Templin oil, oil of pine cones {Ptnus picea, Pinus Pumilio), [a] y = - 8.2 . 1 d* = 0.856, a-yin i dm. tube = 85-2, |>I - - 98.8. Rectified, ^ = - 92.5, [>] y = 107. 6. 2 d. In American oil of turpentine also, /-pinene has been found by Long/' the specific rotation of which, was in one case found to be \OL\ D - - 40.79. The /-pinene in this case was distilled from fresh oleo resin. It is probable that much of the so-called American oil contains /-pinene. 4 e. In Asarum EuropaeumL. Boiling-point, 162 to 165. ,= 20, []= -2 5 .lV Boiling-point 161 [a]z>= 33.8, (OL 42-3) 6 " 161 = -40.3 8 " 155 rf w = 0.8587, [ar] = -43-4 09 [or]/, = - 44-95 lfl [a\ D = -49-I 011 The discordant results are probably explained by the fact that the rotating power of pinene is diminished b} 7 oxidation on standing in the air. For example, Landolt found, with a preparation having originally [ai] D = " 37 , a rotation of 1 Jolly, Buchner : Ann. Chem. (I^iebig), 116, 328. 2 Fliickiger, Berthelot : Jahresber., 1855, p. 643. 3 J. Anal. Appl. Chem., 7, 99 (1893) ; Chem. Centrbl., I, 835 (1893) ; J. Am. Chem. Soc., 16, 844 (1894). 4 J. Am. Chem. Soc., 21, 637. '> Petersen : Inaug.-Diss. (Breslaui, Berlin, 1888. '' Berthelot: Ann. chim. phys., [3], 4o, 5. : I^andolt: Ann. Chem. (Liebig), 189, 311. ? Riban : Ann. chim. phys., [5], 6, 15. 9 Flawitzky : Ber. d. chem. Ges.. 12, 2357. 10 Bouchardat, L,afont : Compt. rend., 102, 320. 11 Bouchardat, I,afout : Ann. chim. phys.. [6], 16, 242. 632 CONSTANTS OF ROTATION OF ACTIVE BODIES [a] 7 , - 35.7 after the same oil had stood four weeks in a flask with air. 1 With increasing temperature up to 150 the rotation of turpentine oil is decreased according to the following formula :' [<*]/-> = 36.61 - 0.00444 / Derivatives: Hydrochloride. Crystals; melting-point, 125. [a]/, = - 26.3 ' [a\ D .-- - 30.69 ' Ifvdrobromidc. Crystals ; melting-point, 92. [or]/, - 24.6 > [a]/, = - 27.802 Phthalimide Compound, C 6 H 4 (CO),N.C ln H i: ,. Rectangular plates; melting-point, 90 to 100. [>]/> -3538 /T l-Isoterebenthcnc, C 1(I H 1( .. Made by heating /-turpentine oil to 300. Liquid ; boiling-point, 175. d~* = 0.8416. [tf]/> = - 9-45, OL - 10.87 Boiling-point, 177.5. / 20, [a]/, - 8.38 ig), 253, 156. Gazz. chim. ital., 18, 223. Wallach, Conrady : IMC. at. Pesci . IMC. ctt. Pcsci : Gazz. chim. ital., ai, i to 4. Ki.:iii Ann. chim. phys., [5], 6, 2is. Barbier : Compt. rend., 108, 519. '" Flawitzky : Ber. d. chrin O 12. 11 Kuriloff: J. prakt. Chem., [2], 43, i.ii. 11 Flawitxky : Ber. d. chem. C.es.. ia, 2357. CAMPHORS AND TERPENES 633 /-FENCHENE, C 10 H 16 . From fenchyl chloride and aniline. Boiling-point 150- 154, cf^ 0.8667. []* = -6. 4 6 01 2. Alcohols rt'- BORNEOL, Borneo Camphor, rf-nr-Camphol, C, H 17 OH. From Dry obcdanops camphor a. Crystals ; melting-point, 198 ; boiling-point, 212. Gives*/- camphor on oxidation. With two different preparations : a. Acetic ether c = 15.4, / = 20, [a]/, 38.83)- . " " /> 17.54- rf 1 " 1 0.8876, / 20, " = + 38.45 ( Melting-point, 203. Alcohol p :=: 20, --- - 37-33 04 Toluene /> 20, [a-]/, - 38 I ' Alcohol /> - 20, " 37 j (On artificial borneols, isocamphols, and on the influence of different solvents on the constants of rotation, see below.) Derivatives: Bornyl Chloride, C 10 H 17 C1. Melting-point, 157. Acetic ether. .. c 17.2, t = 20, []/> - about 23 r> Ethyl Borneol, C 1( ,H 1T .OC,H 3 . Boiling-point, 205 to 208. d" = 0.9490, []/> = + 26.3 ' Chloral Borneol, CCl 3 CH(OH)OC 10 H r . Melting-point, 55 to 56. Benzene, c = 15.07 (\. 2 mol. in i liter), t~ 15 to 16, [a] D = -j- 30.13^ Bromal Borneol, CBr 3 CH(OH)OC 1( ,H 17 . Crystals ; melting- point, 105 to 109. Toluene, c = 21.7 C;., mol. in I liter), / = 15 to 16, [= -f 42.05 Add Suctinate, C 2 H 4 .COOH.COOC 10 H 17 . Crystals ; melt- ing-point, 58. Alcohol c = 25.4, / 15 to 16, [or]/) = -f 35.59 :! Neutral Phthalate, C 6 H 4 (COOC IO H 17 ) 2 . Crystals; melting- point, 101.1. Alcohol .... c = 43-8, t - 15 to 16, \a\ D + 79-54 a AcidPhthalate, C 6 H 4 .COOH.COOC 10 H 17 . Crystals ; melting- point, 164.48. Alcohol .... c = 30.2, /=i5toi6, (>]/> = -f- s8.38 3 Carbonate, CO(OC 10 H 17 ),. Melting-point, 220.6. [a]/, = -f 14-37 4 Borneol Phenyl Urethane, C e H 5 NH . COOC 10 H 17 . Crystals ; Melting-point, i37-75- Toluene .... c = 5.46, / = 15 to 16, [nr]/, + 34.22 5 /- BORNEOL, Valerian Camphor, /-tf-Camphol. From valerian oil, n'gai, bang-phien and madder fusel oil. Crystals; melt- ing-point, 204 ; boiling-point, 210. Yields matricaria camphor on oxidation. Alcohol, c = 15.4, / 15 to 1 6 From valerian oil in. p. 208.8, [<*]/> =- 37-77 ] r> n'ga'i " " 209.0, " = -37-77 I " bang-phien " " 208.0, 38.20 " madder " <( 208.1, "= -37.8 J From valerian oil, alcohol,. . . p = 20, df 20 = 0.828, []^ = 37-74 7 Toluene f> 20, []/?= 38 Alcohol p 20, " 37 Haller : IMC. fit. Haller: Compt. rend., 108, 410. Haller. Haller: Compt. rend., 105, 230. Haller: Ibid , no, 149. Haller: Ann. chim. phys., [6], 37, 395. 7 Recktnann : Ann. Chem. (I,iebig), 180, 353. 1 Beckmann : J. ]>rakt. Chem., |z], 55, 31. CAMPHORS AND TERPENES 635 On the rotation in different solvents, see below under Derivatives: Chloral BorneoL Crystals; melting-point, 55 to 56. Benzene ....... c- 15.07, / = 15 to 16, [a]/? = 30.13 > Bromal BorneoL Crystals ; melting-point, 105 to 109. Toluene ..... c ~- 21.7, / 15 to 16, []/> = 52.4-' Acetate. Crystals; melting-point, 24. Alcohol ..... c= 19.6, / = 15 to 1 6, \_O\D = - 44.02 8 Benzoate. Crystals ; melting-point, 25.5. Alcohol .... <: = 25.8, / = 15 to 16, [a\ D = - 44.18 * Neutral Succinate. Crystals ; melting-point, 83.7. Alcohol .... c --. 39, t 15 to 16, [a]/, - 42.39 5 Acid Succinate. Crystals ; melting-point, 58. Alcohol ..... c = 25.4, t == 15 to 16, \CL~\D = - 35-94 4 Neutral Phthalate. Crystals ; melting-point, 101 . i . Alcohol ---- ^ = 43.8, /=i5toi6, \CL\D -- - 79-14 4 Acid Phthalate. Crystals ; melting-point, 164.48. Alcohol ---- c= 30.2, t 15 to 16, [a] D = 58.27 4 Carbonate. Crystals ; melting-point, 219.4. \a\ D = -44-I 06 Borneol Pheny I Ur ethane. Crystals ; melting-point, 137.25 Toluene.... ^ = 5.46, t = 15 to 16, [a]/> = - 34-79 " The following determinations are by L. Tschugaeff : 8 1 Haller: Compt. rend., 112, 143. - Minguin : Ibid, 116, J5go. '' Haller : Ibid., 109, 29. Haller : Loc. cit. '-> Haller : Compt. rend., 108, 410.' 6 Haller : Ibid., 105, 230. ' Haller: Ibid., no, 149. f P.er. d. chem Ges., 31, 1775. 636 CONSTANTS OF ROTATION OF ACTIVE BODIES /-Borneol formate i .0058 acetate 0.9855 propionate 0.9717 w-butyrate 0.961 1 ;/-valerate -9533 ;/-caprylate ! 0.9343 - 40.46 - 44-40 42.06 39-15 37.08 - 3L45 ft- BORNEOL, Isocamphol, Isoborneol. Montgolfier 1 and Kachler obtained mixtures of two borneols by action of alcoholic potash on the camphors, of which one is stable, the other instable. The first rotates the plane of polarized light in the same direction as does the camphor em- ployed, the second in the opposite direction from that of the camphor, which is reproduced also by oxidation of the mix- ture. Haller" finds that the stable borneols which he desig- nates as ar-borneols, are identical with the natural borneols, while the instable products, or /^-borneols, are isomeric. Further peculiarites are shown in the following table : Description. [], By oxidation there is formed. a-borneols ! right-rotating 37 to 38 ^-Camphor ( left-rotating a 37 to 38 /- " Active -f P-borneols ( right-rotating ft 4 34 I- (isoborneols) ( left- rotating ~p -34 d- f 4- tt and a Inactive borneols ^ " j .... r _L a " ^T rf- a 4< ft /. Haller investigated, further, the influence of different sol- vents on at- and /f-borneol and found that the rotating power 1 Ann. chitn. phys., [5], 14, 13; Compt. rend., 84, 90, and 89, 101. Ann. Chem. (I,iebig), 197, 102. 4 Ann. chim. phys., [6], 37, 414. CAMPHORS AND TERPENES 637 of the latter is changed by the solvent, while that of the first is not altered except by methyl alcohol. [a]/) for / 13 to 15 and c = 7.7. Solvent. Methyl alcohol. Alcohol. '3S3H 1 Isobutyl alcohol. Acetone. Ligroin. cr-Borneol " 35-93 - 37-33 - 37.23 - 37-23 - 37-87 - -37-12 0-Borneol 30.00 -32.90 33-33 - 33-54 22.94 - 22.72 Solvent. Acetic ether. Benzene. Toluene. 1 Xylene. ^-Methyl - propylben'zene. ct-Borneol - 37-55 - 37-66 - 37.87 - 37-66 - 37-66 /T-Borneol - 19.18 - 18.93 - 18.95 - 18-97 d- or l-hoborneol : Toluene Alcohol = 20, =_ 20, a]/> = q: 19 V l( ip 33 j Derivatives : Chloral Borneol. Not crystalline ; melting-point, 55 to 56 Benzene ---- / = 15 to 16, C- 15-07, [<*]/> ^ 56.40 Borneol Phenyl Urethane. Crystals ; melting-point, 130.05. Toluene or alcohol ....... t = 15 to 16, c 5.46, |>P = -56.77' = ; + 10.4.' o ROTQ (\9P\ T 93- 1 Q Q 4- 72 O 5r*o 1 Q 7 T tso-Propyl bornyl aminie- Butyl bornyl amine 0.0919 ^10 ; o.886l (14) o 8002 ( iz\ Oy.O + 84.0 1 QT 7 + 63.3 -i 6 A 8 + 8l.I 4- 80 7 Dimethyl bornyl amine u.oyw* \ 1 O ) 01. / + 62.5' + 48.7 4- V) ^ + 59-6 -f 62 6 Kormyl bornyl amine A? T 1 J U 'O 42.1 A2 Benzoyl bornyl amine 42.9 3T 8 1 Bouchardat, Lafont : Ann. chin, phys., [6], 9, 529. Bouchardat, Lafont : Compt. rend., 113,553. Armstrong. Pope : J. Chem. Soc., 59, 315. 4 I,euckart, Bach : Ber. d. chem. Ges., ao, 104. J. Chem. Soc., 73, 386 ; 73, 934, "49- CAMPHORS AND TER PENES 639 Numerical values for other compounds are also given. ]2, = - 36.17 :> Acetyl Compound, C 10 H 17 NH.COCH,. Crystals; melting- point, 93 to 94. Chloroform ........ p 4.59, d"' = 1.475, [ 1.80, d 4 1.489, [tr]*, = 53.08 | 7 ...... ^ = 1-793, " 1-488, ~ 53-14 J Benzylidene Compound, C 10 H 1T N:CHC 6 H 5 . Crystals ; melt- ing-point, 42. Chloroform ...... p 5.77, d* 1.453, [ tr ] 8 /, = -f 73-23 I T ...... / = 5-7i, " 1-455. - 73.05)" i Wallach : Ann. Chem. (Uebig), 273, 106. - Wallach : Ibid., 263, 142. - 1 Wallach, Binz ; Ibid., 276, 318. 4 Wallach, Binz; Ibid., 276, 318. Binz : Ztschr. phys. Chem., 12, 726. " Binz : IMC. cit. ' Binz. 640 CONSTANTS OF ROTATION OF ACTIVE BODIES o-Oxybenzylidene Compound, C 10 H 17 N:CHC 6 H ( OH. Crystals; melting-point, 94. Chloroform p ^ 4.97, d 1.471, [a~\% = + 66.59 j ' p 2.49, & 1.486, [a\ D = -f 65.99 ( p-Oxybenzylidene Compound, C 10 H 17 N:CH.C 6 H 4 OH. Crys- tals ; melting-point, 175. Shows birotation which disappears in eighteen hours. Chloroform p = 1.28, d 1.4905, [cr] = -j- 72.00 l o-Methoxybenzylidene Compound, C 10 H 1; N : CH.C 6 H 4 .OCH :1 . Crystals; melting-point, 56. Chloroform .... p = 5.56, d == 1.4605, [or]*, = -f 58.98 ) l ....^ = 5.09, fl rio = i.46o, [or]~ =- + 59.42 J p-Methoxybenzylidene Compound, C 10 H 17 N:CH.C 6 H 4 .OCH,. Crystals ; melting-point, 54 to 55. Chloroform.... p = 4.97, rf 11 = 1.4585, [a]^ 1 = + 78.10) ] .... p^ 4.89, ^ 5 = 1.468, [a]^, == + 78.01 f Aminoterebenthene Hydrochloride , C 10 H 15 NH 2 .HC1. \_a-] D = - 48.508 '-' 4. Ketones ^/-CAMPHOR. Ordinary camphor, Japan or laurel camphor, C 10 H ]6 O. Melting-point, 178.6 ; r> boiling-point, 204 ; 4 209.1 (corr. at 759 mm. ) . 5 Camphor is active in the fused condition, in solution, and in vapor (see 9) but not in crystalline form. Landolt 6 investigated the rotatory power of camphor in dif- ferent solvents (see 53). The specific rotation of the pure camphor calculated from the values obtained was found to be, in the mean, \_oi] D - = + 55.4. The following table exhibits the effect of different solvents and the mean value just given is the basis of the calculation. The data refer to t = 20 and q 40 to 90. 1 Binz. - Pesci : Gazz. chim. ital., 18, 219; Ber. d.cheni. Ges., 22, Kef. 108. :t Haller : Compt. rend., 105, 229. 4 I^andolt. 5 Forster: Ber. d. chem. Ges., 23, 2981. Ann.Chem. (I y iebig), 189, 333. CAMPHORS AND TERPEXES 64! Solvent. O]/>for Benzene 55.40 0.1664? -(-42.3 Ethyl alcohol 55-4 0.1780 q 0.00037 ? 2 43.5 Dimethylaniline 55.40 0. 1428 q 44.0 Acetic acid 55.40 - o. 1360 q 44.5 Methyl alcohol 55.40 0. 1630 q -f- 0.00066 q 1 46.6 Monochloracetic ether 55-4 0.0620 q 50.4 Acetic ether 55.40 0.0480? 51.6 We have also the following additional observations : Ethyl alcohol c 7 to 50, t = 20, [a] D = 41.982 -f 0.11824 c l /> = 2o, / = 20, d-" = 0.8255, O]/>= + 44-22 - q = 50 to 95, / = 22.9, [tr]^ =51.945 0.0964?' Alcohol of 80 vol. per cent., c - 2 6 10 ^ * \oi\n 40.9 39- 2 5 Chloroform ^ = 5. [or]z> = 44.2 Acetic ether- . q = 48 to 90, t= 20, \_a\ D - 56.543 0.09065 q + 0.0004005 g 2 5 Benzene q = 47 to 90, / = 20, [a]/> == 55.99, 0.1847 ? -f 0.00026902 ?- 75 per cent, acetic ether -f- 25 per cent, benzene : p = 20, d** = 0.8907, / = 20, [= 50.12 50.5 per cent, acetic ether - 49.5 per cent, benzene : / = 20.3, -cT M = 0.9016, ^ = 20, [a]/)= 48.1 25.7 per cent, acetic ether -f- 74.3 per cent, benzene : /> = 20, ^ = 0.8979, ^ = 20, [a]/) = + 45-89 06 Benzene.. ^^51040, t = 20 [a] />= 39-755 -r o.i 7254 ^ T On the quantitative determination of camphor in solutions from the angle of rotation observed, see 184. The specific rotation of benzene camphor solutions increases with the temperature, but the values bear no simple relations to each other. Forster* found : 1 I^atulolt : Ber. d. chem. Ges., 21, 191. - Beckmann : Ann. Chem. (I y iebig). 250, 352. :1 Arndtsen : Ann. chim. phys., [3], 54, 418. 4 Hesse : Ann. Chem. (L,iebig), 176, 119. " Rimbach : Ztschr. phys. Chem.. 9, 698. 6 Rimbach : IJQC. fit. ~ Forster : Ber. d. chem. Ges.. 23, 2981. " IMC. fif. 41 6 4 2 CONSTANTS OF ROTATION OF ACTIVE BODIES Temperature. 12'. U [ = + 45-6) 2 y-Monochlorcamphor. Crystals; melting-point, 124 to 125 ; boiling-point, 220. \ci\ D = -f 40 ;! DlCHLORCAMPHOR, C ]0 H U C1 2 O. a-Compound. Orthorhombic prisms; melting-point, 96. Alcohol or chloroform . . []/ = -f 57.3, ( []z> = + 45-8) fi-Compound. Mass of crystals ; melting-point, 77. Chloroform [tr] y = - 60.6, ([or] = + 48.5)) * Alcohol " = = + 57-4, ( " : = 4-45-9)i TRICHLORCAMPHOR, C 10 H 13 C1 3 O. Crystals; melting-point, 54 Alcohol ..... />--=4-57, My=-f64, (M/>= + SL 2) 5 MONOBROMCAMPHOR, C 10 H 15 BrO. ct-Compound. Monoclinic prisms ; melting-point, 76. Alcohol ....... \a\ D -|- 139. 6 =+ 127.7 Kipping, Pope : J. Chein. Soc., 63, 593- Cazeneuve : Compt. rend., 94, 1530. Cazeneuve : Ibid., io9, 229. Cazeneuve : Bull. soc. chim., [2], 37, 454 ; 38, Cazeneuve: Corapt. rend., 99, 609. Montgolfier : Ann. chim. phys., [5], 14, no. Marsh, Cousins : J. Chem. Soc., 59, 969. Haller; Compt. rend., 104, 66. 644 CONSTANTS OF ROTATION OK ACTIVE BODIES Derivatives-. Sulphonic Arid, C 10 H,.BrOSO 3 H. Crystals; melting-point, 195 to 196. Water ......... c = 2.577, [a]}} - + 88.27 Potassium Sulphonate, C 10 H u BrOSO ;s K + i'/ 2 H 2 O. Crystals. Water ......... r 4.921, [a] 71.44 (hydrated) = -f- 76.96 (anhydrous) Sodium Sulphonate, C 10 H 14 BrOSO :t Na -f 5H..O. Crystals. Water ........ c -= 4.130, [], == -f 63. i (hydrated) ) l " = 4- 80.2 (anhydrous) ( Water ........ c = 4.305 ^ anhydrous), [ : 29.4 5 Derivatives : Sodium Sulphonate, C 10 H 14 BrOSO H Na -f 2H,O. Water c 3.422 (anhydrous), [c., 63, 586. " Marsh, Cousins : IMC. fit. Kipping I'ope. 4 Mar-.li. Cousins. Marsh, Cousins: J. Chtm. Soc.. 57, .xjs. CAMPHORS AND TERPEXES 645 Ammonium Sulphonate, C H1 H H BrOSO 3 NH 4 . Water ......... [a]/> 82 l y-Monobromcamphor. Crystals ; melting-point, 144 to 145. Alcohol ............ -"5-5, [>]/> + 40 * MONOIODOCAMPHOR, C 1(I H 15 IO. Prisms ; melting-point, 43 to 44. [a~\ D = -f 160.42 3 CHLORBROMCAMPHOR, C 10 H 14 ClBrO. a-Compound. Prisms; melting-point, 98. Chloroform ............ \oi\j = -f 78, ( [ -f 62.4)* fi-Conipound. Crystals ; melting-point, 51.5. []y +51, ([]/> - 4-40.8) * For determinations on the halogen derivatives of camphor, see Marsh and Gardner. 6 XlTROCAMPHOR, C 10 H, 5 NO,O. a- Compound. Monoclinic prisms ; melting-point, 100 to 101. Alcohol ..... p 3.33, [a], 7.5, [ar]^ = 6.o c Benzene ---- p.- 0.5, -140, -112.6 Chloroform. ./= 0.676, 140, 112.6 p= 5.206, - 102, - 96.5 p= 19.978, " =- 98, = - 78.8 For a lengthy study of nitrocamphor and derivatives, see Lowry. 8 Derivatives : Sodium Compound, NaC, H u NO 2 O. Water ....... [or] / -f 289, (\a\ D = -f 232.5) Zinc Compound, Zn(C 10 H u NO 2 O) 2 H 2 O. Alcohol ...... [a:],-- -275, ([or],, = + 221.2) Quinine Compound, C, H 24 N 2 O 2 (C 10 H U NO 2 O) 2 H 2 O. 1 Marsh, Cousins : J. Chem. Soc., 59, 976. - Cazeneuve : Compt. rend.. 109, 439- ; Halle : Dissertation, Xancy, 1879. 4 Cazeneuve: Bull. soc. chim., [2], 44, 116. " Cazeneuve : Ibid.. [2]. 44, 120. " Chem. News, 74, 279. ' Cazeneuve: Compt. rend.. 103, 275 : 104, 1522. - J. Chem. Soc., 73, 986 and 75, 216. -' Cazeneuve : Bull. soc. chim., 49, 92. 646 CONSTANTS OF ROTATION OF ACTIVE BODIES P-Compound. Microscopic plates ; melting-point, 83 to 84. Alcohol /-=3-33, OL = -t- 7-5, (I Benzene p _- 3.33, - 75, ( Alcohol /^3-33, [= + 6.O )") 1 11 --60.3)] OT-CHLORNITROCAMPHOR, C 10 H 14 NO 2 OC1. Prismatic needles; melting-point, 95. Alcohol ...... []y= -6.3, ([a]/, - 5)' BROMNITROCAMPHOR, C 10 H 14 NO 2 OBr. Prisms; melting- point, 104 to 105. Alcohol ...... c=i, LL- -27, ([]/>- - 21. 7) 3 CAMPHONITROPHENOL, C 10 H U NO 2 .OH. Crystals ; melting- point, 220, with decomposition. Alcohol .......... c^i.S, []/> r 10 4 ^/y Derivative, C 10 H 14 NO 2 O.C 2 H ;t O. Crystals : melting- point, 115, with decomposition. Alcohol ............ c = 2, []/, = + 4.25 5 NITROSOCAMPHOR, C 10 H 15 (NO)O. Crystals ; melting-point, about 1 80, with decomposition. Ben/ene ......... c == 0.81 , \_O\D = + 195 6 CAMPHORSULPHOCHLORIDE, C 10 H 15 OSO 2 C1. Tetrahedra ; melting-point, 137.5. Chloroform ....... = 5-349, * = 14, [<*]/> = + 128.7 CAMPHORSULPHONAMIDE, C 10 H 15 OSO 2 NH 2 . Crystals ; melt- ing-point, 136 to 137. Alcohol .......... c 2.252, t = 13, [a]/, = + 93.6 ' CYANCAMPHOR, C 10 H 15 OCN. Crystals; melting-point, 127 to 128. Toluene ...... [a]/? = -f- 44.68 8 1 Cazeneuve : Bull. soc. chim., 47, 922 ; Compt. rend., 104, 1522. 8 Cazeneuve : Compt. rend., 96, 589. Cazneuve : Bull. soc. chim., 43, 69. Cazeneuve : Compt. rend., 108, 302. Cazeneuve : I^oc. cit. ' Cazeneuve : Compt. rend., 108, 857. Kipping. Pope : J. Chem. Soc., 63, 564- Haller : Dissertation, Nancy 1879. Haller : Compt. rend., 115, 98. CAMPHORS AND TERPENES 647 Derivatives: ^f ethyl Compound, C 10 H U OCN.CH 3 . Oil; boil- ing-point, 170 to 1 80 (36 mm.). Toluene c -\- 9.55, \_a\ D = -f- 107.69 ! Haller and Minguin" have recently isolated two isomeric cyanmethylcamphors with the following characteristics : ^-Compound. Melting-point, 63 ; [#],,=: -j- 150.8. it-Compound. Oil, from which crystals with melting-point 38 to 45 separate. \a\ D = = -f- 90.1. Ethyl Compound, C 10 H 14 OCN.C 2 H 5 . Oil ; boiling-point, 163 to 165 (21 mm.). Toluene c= 10.25, M# == + 120.71 Propyl Compound, C 10 H U OCN.C 3 H 7 . Crystals ; melting- point, 46 ; boiling-point, 140 to 150 (20 mm.). Toluene c 10.95, [a]z> = -f 126.16 Benzyl Compound, C 10 H 14 OCN.C 7 H 7 . Crystals; melting- point, 58 to 59. Toluene c = 13.35. M/> = -f- 93-62 o-Nitrobenzyl Compound, C 10 H 14 OCN.C 7 H 6 NO 2 . Needles; melting-point, 104 to 105. Toluene =15.6, \a\ D = + 68.37 l CAMPHOR PINACONE, C 20 H 34 O 2 . Melting-point, 157 to 158. Benzene p = 23, d = 0.9089, [<*]^ = 27.03 " pii. 62, " = 0.8949, -26.13 Derivatives: Chlor Pinaconan, C 20 H 31 C1. Melting-point, 75. Benzene p = 25.47, a?; 8 = 0.9105, []^ 8 = -f 44-17 j 4 " p 74.08, "=0.9436, -4-46.50 a- Methyl Ether, C 20 H 33 O 2 CH 3 . Melting-point, 47. Benzene / = 14.15, ]= - i33-5o' 1 Haller: Compt. rend., 113, 55. 2 Compt. rend., 118, 690. 3 Beckmann: Ann. Chem. (I^iebig), 293, i. * Beckmann, p. 7. 6 Beckmann, p. n. 648 CONSTANTS OF ROTATION OF ACTIVK BODIES CAMPHOROXIME, C 10 H IH :NOH. Monosymmetric prisms; melting-point, 115. Alcohol /> -= 20, d = 0.835, [<*] ?= 42.40 p~ 8.3, "=-.0.812, -41.38 Hydrochloride, C 10 H lt .:NOH.HCl. Crystals; melting-point, 162, with decomposition. Alcohol /> 8.3, if 1 " 0.8185, O]~- -43-98 ' d-Camphorsulphonate, C 10 H lfi X.OH.C 10 H 15 O.SO :i H -f H 2 O. Long needles. Alcohol c 1.7508, []*', ; 4.3 - See F6rster :t on the esters of camphoroxime. CAMPHORDICHLORIDE, C 10 H )6 C1 2 . Needles ; melting-point, 155 to 155-5. Acetic ether c 22.34, [a] - 16 ' CHLORALCAMPHOR, C IO H 16 O.CC1 S CHO. []/, = + 143 " CHLORALHYDRATE CAMPHOR, C 10 H ](i O.CCl,CHOH,O. Thick liquid. d-~- 1.2512, [a]/, == : - 33-45 CHLORALALCOHOLATE CAMPHOR, C 10 H 1( .O.CC1,CHO. C,H,OH. d =1.1777, []/> = + 36.9 (i BENZYLCAMPHOR BROMIDE, C 17 H,,BrO. Melting-point, 82. [a]/, = r 37.7 ' BENZYLCAMPHOR DIBROMIDE, C r H., Br.,O. Melting-point, 92. [a} n = - 6i- MONOCAMPHOR PHENOL, C t; H 6 O.C 10 H lf> O. Liquid ; at 23, crystals. rf = 1.0205, []/' =H 20 !) Beckmann : Ann. Chem. (I.iebijf), 250, 352. Tope: J. Chem. Soc., 73, 1107. J. Chem. Soc.. 71, 1030. Spitzer : Her. d. chein. C.t-s.. n, 1819. I'a-rhkis. Ohtrrniayer : IMiarm. Post., 21, 741. Zeidler : Wien. Akad. Her., 2 Abth., 76, 2^. llaller, Mingum : Hull. soc. chim., [3], 15, 988. Mailer. Mini-uni. : Compt. rend., in, 109. CAMPHORS AND TER PENES 649 HEMICAMPHOR PHENOL, 2C 6 H e O.C 10 H, 6 O. Liquid ; at 50, crystals. if 1.040, [a]/> 10.1 l MONOCAMPHOR RESORCiN, C 6 H 6 O,. C 10 H 16 O. Crystals ; melt- ing-point, 29. Alcohol ...... =25.1, [tf]z>- 22.1 ' DICAMPHOR RESORCIN, C fi H t .O,.2C 1(1 H 1( .O. Sirup; at o, crystals. d i:> = 1.0366, [ = - 25.15 l , C 10 H 8 O.C 1( ,H 16 O. Sirup ; at 16, crystals. .2C ln H 1( .O. Crystals; melting-point, 60. Alcohol ........ c 20.8, \_oc\n : 27.05 ] CAMPHAXIC ACID, C 10 H U O 4 . {<*]j -MS 01 CAMPHINIC ACID, C 9 H 15 COOH. Tough mass. [a]/, == + 15-75 :! CAMPHOLIC ACID, C 7 H 1T COOH. Melting-point, 105. [a], +49 8" ISOCAMPHOLIC ACID ETHYL ESTER, C,,H JT COOC.,H 5 . Boil- ing-point, 228 to 229. ct" 0.9477, [a]/, -21.5 5 CAMPHOCARBOXYLIC ACID, C 1(1 H 16 O 3 . Melting-point, 128.7. [a-] D --.-{- 66.75 - Aschan : Act. soc. scient. fennice, 21, Nr. 5, p. 3 Montgolfier : Ann. chim. phys., [5], 14, 70. 4 Montgolfier : Loc. cit. " Gaerbet : Bull. soc. chim.. [3], 13, 769. Haller : Compt. rend.. 105, 229. 650 CONSTANTS OF ROTATION OF ACTIVE BODIES Derivatives : Methyl Ester, C U H I5 O 3 CH V Boiling-point, 155 to 160 (15 mm.). Alcohol .......... c- 21, [ar]z> - + 61.9 T Methylcamphocarboxylic Add Methyl Ester, C 12 H 17 O 3 CH 3 . Melting-point, 85. Alcohol ........ =11.2, [a~\ D -f 17.25 2 Ethyl Ester, C 12 H 17 O 3 C 2 H 5 . Melting-point, 60 to 61. Alcohol ........ c= 11.9, [a]/> === -f 13.8 2 ACID, C 8 H ]4 COOH.CH 2 COOH. Melting-point, 234. Derivatives : Methyl Compound, C,H 14 COOH. CH ( CH 3 ) . COOH. Melting-point, 175. \a\ D = + 26.31 3 Monethyl Ester, C 8 H 14 COOC 2 H 5 .CH 2 COOH. Liquid; boil- ing-point, 228 to 230. []/> - + 5i. i 04 Diethyl Ester, C 8 H 14 COOC 2 H 5 .CH 2 COOC 2 H 5 . Liquid; boil- ing-point, 220 to 230 ( 1 60 mm.). Alcohol ............ c 27.0, [a]z> == -f- 49-6 to -f 50.6 5 Monobenzyl Ester, C 8 H 14 COOC T H 7 .CH,COOH. Oil ; boiling- point, 250 to 275 at 10 mm. Alcohol ......... c 7.6, [or]/, = - 52.62 G Dibenzyl Ester, C b H 14 COOC 7 H 7 .CH 2 COOC 7 H 7 . Thickliquid ; boiling-point, 260 to 269 at 10 mm. Alcohol ......... c 9.85, [ = -f 35.5 7 Mononitrile (Cyancampholic Acid), C 8 H 14 COOH.CH 2 CN. Crystals ; melting-point, 164. Alcohol ......... c - 19.5, \_ci\D = + 64.61 8 Minguin : Compt. rend., ua, 1369. Minguin. Haller and Minguin : Compt. rend., 118, 691. Haller, Minguin : Compt. rend., no, 410. Haller, Mins^uin. Minguin : Compt. rend., na, 1454. Minguin. Minguin : Compt. rend., 112, 51. CAMPHORS AND TERPENES 651 Mononitrile, Ethyl Ester, C S H U COOC 2 H 5 .CN. Rhombic crystals; melting-point, 57 to 58. Alcohol ......... c = 22.3, [or]/, = -j- 57.7 * Mononitrile, Benzyl Ester, C S H U COOC 7 H..CN. Crystals; melting-point, 70 to 71. Toluene ......... =^28.5, [a~\ D = -j- 42.8 2 Mononitrile, Phenyl Ester, C 8 H U COOC 6 H 5 .CN. Boiling- point, 267 to 270 at 40 mm. Alcohol ........ c = 2j.i t [ a] D = + 26.66 3 Mononitrile, fi-NaphthylEster, C S H U COOC 10 H..CN. Crystals; melting-point, 117. Toluene ......... 32.1, []/? = + 17.1 4 Monamide, C 8 H U COOH.CH 2 CONH 2 . Melting-point, 205 to 206. Alcohol .......... =9.18, [a] D = 63.5 ], for t = 20. limits. [a]/? for / = 15- Abs. alcohol . . % .178 352 -f O.OII74 p 0.01174 q P q = 17 = 57 to to 43 83 }- 47-35 Acet '{ 50 5i .689 .524 0.00835 p 0.00835 q p q = 8 = 84 to 15.5 5 to 92 1 + 50.8i Glac . acetic acid- 45 50 .921 825 -- 0.04904 p \ p - 0.04904 q q = 6 = 84 to 16 \ to 94 j 46.66 Derivatives : Salts. Hartmann made the following deter- minations with aqueous solutions at t = 20. The values for p refer to the anhydrous salts : i Haller : Compt. rend., 109, 68. - Minguin : Ibid., 112, 51. Minguin : Ibid., iia, 101. Minguin : Ibid., 112, 102. Compt. rend., 108, 984. Ibid., 80, 1381. Ber. d. chem. Ges., ai, 223. 652 CONSTANTS OF ROTATION OF ACTIVE BODIES Salt. Mi'. limits. \ci\D for ( 17-750 4- 0.23257 p p = 13 to 25 I 4- 20.08 ( 41.007 0.23257 q q = 75 to 87 j NX, " ..".-j 14.778 4- 0.21288 / 36.066 0.21288 q /= ii to 37 q = 63 to 89 j 4- 16.91 K " .. ^ 13.081 4- o. 13994 p p 19 to 43 1 4- 14.48 ( 27.075 0.13994? q 57 to 81 f (NH,),- [ 16.447 + o. 14242 p 30.689 0.14242 q P " to 37 q 63 to 89 \ + 17-87 Mg " ....-[ 17.824 4 o.i 8779 p 36.653 0.18779 <1 p 8 to 16 q = 84 to 92 > 4- 19.70 c. - :...( 16.457 + o.i 2286 p 28.733 0.12286? P 3 to 6 q = 94 to 97 | 4- 17.69 Ba " ....{ 10.908 4- o.i 2980 p 23.888 0.12980? p = 18 to 36 q = 64 to 82 I 4- 12.21 Observations by Thomsen 1 agree very well with these data. He determined the specific rotation of sodium camphorate also, and the effect of addition of excess of sodium hydroxide solution. Landolt" gives the following figures : Water. . . / = 20, C 10 H, 4 O 4 K 2 c - 4 to 16, [a] D = 14.39 + o -6 c ... / = 20, Na, 2 c=2to 9, " = 16.62 4- 0.06 r " .../-= 20, " (NH 4 ) 2 c = 4 to 17, " = 16.98 4- 0.13 c Esters. On the nomenclature of the camphoric acid esters, see Briihl and Braunschweig. :! al-Methyl Ester, C h H 14 .COOH.COOCH 3 . Melting-point, 85 to 86 ; boiling-point, 193 (15 mm.). []/> = + 43- 55 4 o- Methyl Ester, C,H U .COOCH 3 .COOH. Melting-point, 75 to 76 ; boiling-point, 199 (15 mm.). a/, = 4- 51.52 Dimethyl Ester, C 8 H 14 .COOCH,.COOCH :i . 264 (738 mm.) ; 149.5 C 11 mm.). d 1,0747, [a]"---. + 48.i6 0(1 1 J. prakt. Chem.. [a], 35, 157. '-' "Optisches DrehuiiRsvcrinoxen," ist. t-d., p. 225. 1 Ber. d. chem. Cies., aj, 1796. 4 Haller : Compt. rend., 114, 1516. Haller : l.m . , it. " Briihl : LOC, /. Boiling-point, CAMPHORS AND TERPEXES 653 d\~ 1.075, Mtf : -r 48.32 ' [*]/>= -f 44.4 J al- Ethyl Ester, C S H 14 .COOH.COOC,H V Melting-point, 57; boiling-point, 207 to 208 (21 mm.)- rf" --= 1.1004, []/>-= -f 23.9 '' o- Ethyl Ester, C S H U .COOC,H,.COOH. Boiling-point, 216 to 2 19 (30 mm.)- d" 1.1133, [or]/> -39-i8 * al-Methyl-o-ethyl Ester, C 8 H I4 . COOCH :; . COOCXH 5 . Boiling- point, 277 (746 mm.) ; 169.5 (33 mm.). d 1.0528, [a]-- 2 = -f- 38.43 5 o-Methyl-al-ethyl Ester, C 8 H 14 .COOC 2 H 5 .COOCH V Boiling- point, 278 (747 mm.); 175 (38mm.). d = 1.0448, [ []/> 7.12- Benzene..' M/> 3-7 3 Benzene [<*ly = 3.68) 4 Chloroform " = 0.0 j Amide, C 10 H U O 2 .N 2 H 4 . Melting-point, 241 to 242. Chloroform [<*]/> = - 10.6 :> Chloride, C 1() H 14 O 2 C1 2 . Liquid ; boiling-point, 140 (15 mm.). [a]z> = -3.0 to -3.6) :< Benzene " = -7.110 8.3 j a-AminicAdd, C 8 H 14 (CONH 2 )COOH. Melting-point, 176 to 177. Alcohol [(*]/> = -f- 45 fl-AminicAcid, C H H 14 (CONH 2 )COOH. Melting-point, 180 to 181. Alcohol [a] D = -f 60 Nitrilic Add, Cyanolauronic Add, C S H 14 (CN)COOH. Melt- ing-point, 151 to 152. [a]* =; -f 67.5 6 Bromcamphoric Anhydride. Chloroform [or] / -21.1 Chlorca mphoric A n hydride. \oi\j - 16.3 ' Campholide, C 10 H 16 O 2 . Melting-point, 210 to 212. ISOCAMPHORIC ACID, C, H 16 O 4 . Melting-point, 170; boil- ing-point, 294. Alcohol [cr] /, 48.09 :t Alcohol [a]/j - 46 H Alcohol []/== - 48.3 !) Bcr. d. chera. Ges., ai, 223. Moutgolfier : Ann. chim. phys., [5], 14, 86. Marsh : Chein. News, 60, 307. Aschan : Act. soc. scient. fennice, ai, Nr. 5, p. i. (iuareschi : Centralbl., 1887, p. 1355. Hoogewerff, Dorp: Rec. trav. chim. Pays- Has., 14, 252. Haller : Bull. soc. chim.. [3], 13, 984. Friedel : Compt. rend., 108, 980. Aachan : l.nc. cit. CAMPHORS AND TERPEXES 655 Derivatives: o- Ethyl Ester, C IO H,.O t .C,H.. Melting-point, 195 to 197 (18 mm.). d 1.1159, []/' 49-51 ' Diethyl Ester, C 10 H 14 O/C 2 H 5 ).,. Boiling-point, 165 (2510 28 mm.). d 1.0473, \.<*\D - -48.53 - CHOLECAMPHORIC ACID (Choloidanic acid), C 1( ,H, 6 O 4 . From cholic acid by action of nitric acid. Crystals, which turn brown at 270 without melting. Absolute alcohol ............. c -6.42, [<*]$= -7- 56.17 Glacial acetic acid ............ c 1.44, 57-83 The specific rotation in alcoholic solution is independent of the concentration. 3 CAMFHORONIC ACID, C 6 H n (COOH),. Melting-point, 158 to 159. Water ...... p = 10, [a]j-5 = - 26.9, ([a]^-s == - 23.91 ) 4 CAMPHOLYTIC ACID, C S H 13 COOH. Oil ; boiling-point, 240 to 242. d* = 1.017, []- -5 5 Derivatives'. Ethyl Ester. Oil ; boiling-point, 212 to 213. df = 0.962, [a] = - 5.04 CAMPHOTHETIC ACID, C 16 H 28 (COOH) r Derivatives: Diethyl Ester. Liquid ; boiling-point, 135 to 140 (15 mm.). df 1.019, []]? == + 30.6 T DlHYDROXYCYANCAMPHOLYTlC ACID, C S H U COOH.CN. Melting-point, 109 to m. Alcohol ....... [or]/, = 18.20 " DICAMPHOR : Melting-point, 165 to 166. Benzene ......... p 5, M^ ; -28.07 1 Friedel : Compt. rend., 113, 831. - Friedel. ; lyatschinoff : Ber. d. chem. Ges., 13, 1052. 4 Ossian, Aschan : Ibid., 28, 16. Walker : J. Chem. Soc., 63, 499. Walker : Loc. cit., 498. 7 Walker : J. Chem. Soc., 63, 504. Hoogewerff, Dorp : Rec. trav. chira. Pays.-Bas., 14, 252. 656 CONSTANTS OF ROTATION OF ACTIVE BODIES Dica mpha ndihydropyridazine : Melting-point, 155 10156. Benzene f> -..-- 5, [or]" = 4 118.13 Dicamphanhexandion : Melting-point, 192 to 193. Benzene p = 3.5, [a]g = - r 33 1 Alcohol p 2.9, [orjg r 381 Dicamphanhexanasine : Melting-point, 201 to 202. Benzene p 5, [a]*] -\S*-i Dicamphanhexadienperoxide. Benzene p =; 3.5, []# = 4- 296 Alcohol p 2.9, [ a] g : -}- 345 . j j o Camp haucamphoric Acid. Melting-point, 224 to 225. Alcohol p -.= : 4.75, [] = 4- 93.6 ot-Dicamphandiacidanhydride : Melting-point, 143 to 144. Alcohol /-=i.r, O])j - 142 ' /-CAMPHOR. Matricaria camphor. Found in the oil of Matricaria Parthenium, and may be made by oxidation of /-borneol. Melting-point, 178.6 ; 2 boiling-point, 204 ; d^ - 0.9853. The specific rotation in alcoholic solution for red light (A. = 635), and/ 10 was found as \oi\ r = 33 ; that is, in agreement with the rotation of laurel camphor. The formula for the specific rotation of the latter, :! [or] r = 45.25 0-1369 3 gives for/ == 10 (q = 90), [>],. = + 32,9. 4 Alcohol P = 2o, d% 0.8255, M = - 44-22 ' Transformation Products of I- Camphor /-CAMPHORPINACONE. Right-rotating. Benzene p 23.74, d 0.9075, [a];; 26.52 /-CAMPHOROXIME. Melting-point, 115. Alcohol p : 20, d = 0.835, [a-]-;; _ -f- 42.5 1 P 8.3, ' 2 '^o.8i2, = + 41-38 Hydrochloride, C IO H,.NOH.HC1. Melting-point, 162. Alcohol p = 8.33, d = 0.8185, [cr]--; r 42.52 1 Oddo : Oazz. chini. ital., ay, X, 149. - Haller : Compt. rend., 105, 229. 3 Biot : Ann. chini. phys., [3], 36, 301. * Chantard : J. pharm. Chem., [3], 44, 13 ; Jahresber., iS6;,, p. 555. 5 Beckmann : Ann. Chem. (I.iebijf), 250, 253. ' Beckmann : /hid., 292, 25. 7 Beckmann : I.n< . cit. Beckmann. o i; o - CAMPHORS AND TERPENES 657 /-CAMPHANIC ACID. [l/= + 7 01 /-CAMPHOCARBOXYLIC ACID. Melting-point, 128.7. [a]/, = 66.86 - /-CAMPHORIC ACID. Resembles ^/-camphoric acid and rotates as strongly to the left as the latter does to the right. 3 Absolute alcohol [<*]./ 49-5 * l-Bromcamphoric Acid, Chloroform [a] y = + 21.6 * /-IsocAMPHORic ACID. Resembles the left-rotating iso- camphoric acid from ^-camphoric acid and rotates as strongly to the right. OL-- + 48.6 04 /-CAMPHORONIC ACID. . ;, C 10 H 16 O. From oil of fennel. Crystals; melting-point, 5 to 6 ; boiling-point, 192 to 193, d 19 - 0.9465. a. Direct from fennel oil. Alcohol p 8.333, d* s =0.8045, L a 3S = + 7L97 b. From fenchyl alcohol : Alcohol p = 12.93, d = 0.8090, []g = ~\- 71.7 6 Derivatives : Oxime, C 10 H 16 NOH. Crystals ; melting-point, 161. Alcohol. . . . p = 1. 14, d 19 = 0.793, [a] = - + 65.94 6 Acetic ether p = 2.72, rf 14 ' 5 = 0.911, []^ >s = -f 52.61 " " p = 2.24, d u =0.9115, [a] = = -f 52.28 41 p i. 60, ff 12 ' 5 = 0.9121, [ Ossian, Aschan : Her. d. chem. Ges., 38, 16. Wallach : Ann. Chem. (^iebig), 363, 132. " Binz : Ztschr. phys. Chem., 13, 725. 42 658 CONSTANTS OF ROTATION OF ACTIVE BODIES Oxime Anhydride, C 10 H,.N. Oil; boiling-point, 217 to 218; d = 0.898. Alcohol / = 6.8i, ^-=0.7985, [a] * -j 43-31 01 /-FENCHONE. From thuja oil. Crystals ; melting-point, 5 ; boiling-point, 192 to 194; d = 0.948. Alcohol /> 14.36. d'* - 8l 5i []$ - - 66.94 - Derivatives : Oxime. Resembles closely the ^-fenchone oxime, and in equivalent concentration rotates as strongly to the left as the latter does to the right." (from flf-carvone, \a\ D =-. -f 62), C 10 H lfi O. Boil- ing-point, 210, with decomposition ; d = 0.9567. [or],, 173-8 :1 MATICO CAMPHOR, C^H^O. In matico oil (Piper angusti- folium). Hexagonal crystals. Melting-point, 94. Fused See further, 7. D. POLYTERPENES. /. Sesquiierpenes, C,.H 2l . CADINENE (true sesquiterpene). Found in many ethereal oils. Liquid; boiling-point, 274 to 275; d w = O.QIS. Chlorofonn p= 13.05, rf 9 '* = 1.385, [a]^s _ 98.56 nihydrochloride, C 15 H, r 2HCl. Rhombohedral-hemihedral prisms. Melting-point, 118. Chlorofonn ft - -. 7.212, d 9 '' 9 = 1.460, [a]*? 36.82 Wallach : I.nc. cil. Wallach : Ann. Chem. (Uebig), 373, 102. Baeycr, Kriihl : Her. d. chem. Ges., 38, 639. H. Traut* : Ztschr. f. Kryst., 33, 47. Montgolfier : Bull. soc. chitn., [2], 38, 414. CAMPHOR AND TERPENES. 659 Dihydrobromide, C 15 H. H .2HBr. Needles ; melting-point, 124. Chloroform ..... P 7.227, d 9 ^ = 1.490, [tt]/> 3 36.13 Dihydroiodide, C 15 H M .2HI. Needles ; melting-point, 105 to 106. Chloroform ..... p = 5.568, rf 9 ' 5 = 1.507, [a]fcs = - 48 tf-PARACOTOL. Is probably a hydrate of cadinene, C 15 H M O. Boiling-point, 220 to 222. = - 11.87 a HEMP OIL. From Cannabis sativa. Boiling-point, 120 to 121 (9 mm.). [or]/, = - 10.81 3 PATCHOULENE. From patchouli camphor. Liquid ; boiling- point, 254 to 256. 2. Diterpenes, /-DiTERPiLEXE. From left turpentine oil. d n = 0.9446, [a]/j = - 14.25 5 j. Triterpene, C^H^. ^-O'-AMYRILENE. From -amyrin. C^stals ; melting- point, 134 to 135. Benzene ......... c = 4, [a]/> = -f- 109.48 6 Derivatives : ct-Amyrin, C 30 H 49 OH. Fine needles ; melting- point, 181 to 181.5. Benzene ...... c = 3.839, [tf]^ = -f 9 I -59 ~ Oxy-a-amyrin, C.^H^O.OH -f 2H,O. Needles; melting- point, 207 to 208. Benzene ..... c = 1.653 (anhydrous), [>]^- 5 = + 108.6 8 ; Wallach, Conrady : Ann. Chem. (Liebig), 252, 150. 2 Jobst, Hesse : Ibid, 199, 75. ; Valente : Gaxz. chim. ital., n, 196. 4 Montgolfier : Bull. soc. chim., 28, 415. 6 i,afont: Corapt. reud., 106, 140. Vesterberg, Backstrom : Ber. d. chem. Ges., 20, 1245. Vesterberg, Svensson : Ibid., 23, 3186. Vesterberg, Svensson : Loc. cit. 660 CONSTANTS OF ROTATION OF ACTIVE BODIES Brom-a-amyrin, C^H^Br.OH. Crystals ; melting-point, 177 to 178. Benzene ^2.590, O]# 3 -f 72.8 l a-Amyrin Acetate, C :w H 4fl .C 2 H s O 2 . Plates; melting-point, 221. Benzene c = 4.074, [a]^- 6 = -f 77-o ] /-a-AMYRiLENE. From. the ^/-compound and P 2 O 5 . Pris- matic crystals ; melting-point, 193 to 194. Benzene c = 0.8709, []^ 104.9'- :. From y#-amyrin. Crystals ; melting-point, 175 to 178. Benzene ^ = 1.515, [a] D = -f 112.19 I 3 " c = 0.8 [<*]/> -- -f 110.42 / Derivatives: fi-Amyrin, C S(I H 4!) .OH. Crystals; melting- point, 193 to 194. Benzene --1.9055, [a]^ = -f 99.81 ' 11 =5 [a^ = + 94.2* P-Amyrin Acetate, C^H^.C^HjO.,. Prisms ; melting-point, 236. Benzene r--4.i5r, [a] 1 *-? = + 78.6 l ft-Amyrin Palmitate, C^H^O.,. In coca wax. Crystals ; melting-point, 75. Benzene c = 2, [crpj = -f 54.5 < 24. Ethereal Oils The rotation of the active ethereal oils varies in general, more or less, according to the part of the plant from which they are obtained and the method of preparation (pressure or distillation). Besides this, oils of known purity, in many cases, have not been obtainable in quantity through a suffi- ciently long period to permit the determination of certain data. 1 Vesterbcrg, Svensson. 1 Vesterberg, Koch : Ber. d. chem. Ges., 34, 3834. Vesterberg, Backstrom : Ibid., ao, 1246. Hetfte : Ann. Chem. (lyiebig), 371, 216. ETHEREAL OILS 66 1 But for a number of oils of the aurantium group, on the other hand, it has been possible in the laboratory of Schimmel & Co., of Leipzig, to establish, after years of investigation, the limiting values for the rotation, so that oils which give results outside these may, with considerable certainty, be looked upon as adulterated. In what follows the data furnished by the reports of this firm will be given ; the rotations, a^, stated are the angles read off directly in degrees and minutes, with a tube length of 100 mm. and sodium light. Oil of Berg am ot (Citrus bergamia}. d\\ 0.883 to 0.886, not below 0.881. a D = -f 9 to -j- 15, not above -f 20 (principal constituents limonene, dipentene, linalool about 38 per cent., linalyl acetate). Oil of Lemon (Citrus limonum). d[\ = 0.858 to 0.861. a % 59 to 6 7, not below 59. For observations made at temperatures below 20, under the same conditions, 9' must be subtracted for each degree C, and for observations made at temperatures above 20, 8.2' must be added for each degree. This correction holds for the interval, 10 to 30 C. The rotation does not change on long keeping, even under unfavor- able conditions (principal constituents limonene, citral, no pinene). Oil of Orange : a. Sweet (Citrus aurantium}. d\\ =0.848 to 0.852. oi^ + 96 to 98, not below 96. Temperature correction for 10 to 30 C : for each degree under 20, - - 14.5' ; above 20, -f~ 13.2' (principal constituents limonene, citral). b. Bitter (Citrus bigaradia} . d\\ = 0.848 to 0.852. a% = -f 92 to 98, not below 92 (limouene). For most of the other oils the limits of rotation have not yet been established with the same certainty. The data in the following list, which likewise have been taken from the Schimmel reports up to April, 1897, may, nevertheless, be found useful for comparison in some cases. For each oil, the first figures refer to the specific gravity at 15, and the second give the mean rotation for the D line in a 100 mm. tube, as read off directly. 662 CONSTANTS OF ROTATION OF ACTIVE BODIES Andropogon Oil (Andropogon lanigcr}. 0.915 to 0.919, - 4 to -f- 3438'. Andropogon odoratus, 0.945 to -95 - 22 to 23 (phellandrene). Angelica Oil (Archangelica officinalis}. a. Fruit: 0.856 to 0.89, -f 11 to -f- 12. b. Fresh plant : 0.869 to 0.886, -f- 8 to -f 21. c. Root: 0.855 to 0.905, -f 25 to + 32 (phel- landrene, methylethylacetic acid). Anise Oil (Pimpinella anisum). 0.980 to 0.990. Slightly left-rotating, a D to i5o' (anethol, methyl chavicol, anis- ketone). Asafetida Oil {Ferula asafetida}. 0.975 to 0.990. t* D = 9 1 5' (pinene? bodies containing sulphur, and compounds (C,.H, S 0),,). Basilicum Oil ( Ocimum basilicum}. 0.909 to 0-990. a D = - 22 to -f- 16 (pinene, cineol, camphor, linalool, methyl chavicol). Betel Oil {Piper betle}. Java: 0.958 to 1.04. (* = -f- 253' (betelphenol, cadinene, chavicol). Cajiput Oil (Melaleuca, spec.). 0.92 to 0.93 according to province. ct n -oio'to 2 (cineol, terpineol, terpenyl acetate). Calamus Oil (Acorus calamus}. Fresh roots : 0.960100.970. a D 20 to 31. Dry roots: 0.960 to 0.970. a D = 13 to 21. Commercial oil : 0.962 to 0.970. <*/, never below + 10. Camphor- Leaf Oil (Camphora officinalis}. 0.932. /, - 452' (camphor). Caraway Oil {Carum carvi). 0.905 to 0.915. t* J} -f- 70 to H- 85 (limonene, carvone). Cardamom Oil {Elettaria cardamom it ;;/). Ceylon. 0.895 to 0.910. /, = : -f 12 to -f 13 (terpenine, dipentene? ter- piueol?). Cascarilla Oil {Croton elutcria). 0.890^0.930. ot D - -(- 5. Cedar- Leaf Oil (Jiniipcrus virginiand). 0.884 to 0.886. /= + 59-5. Cedar- Wood Oil ( Juniperus virginiana}. 0.940 to 0.960. ETHEREAL OILS 663 <*D - - 30 to 40 (cedrene, cedar camphor). Oils of dif- ferent countries and unknown botanical origin, 0.906 to 0.928. /> = - 5 to -f 1 8 6' (cadinene, cedar camphor). Celery Oil (Apiumgraveolens). Fresh leaves : 0.848 to 0.850. a D =: -f 48 to -f 52. vSeed : 0.870 to 0.895. <* = : Hh 67 to -f- 79 (limonene). Conifer Oils : Turpentine Oil (Finns spec.). 0.855 to 0.876. American oil, right- and left-rotating. French oil, left-rotating ; other European oils are right-rotating (pinene, dipentene in Russian and Swedish oils, also sylvestrene). Pine Needle Oil (Pi?ms silvestris}. Swedish, 0.872. tx D = io4o'. German, 0.884 to 0.886. CK D =:-\--J to -f 10 (pinene, sylvestrene, bornyl acetate, cadinene). Scottish, 0.885 to 0.889. <* - 7 45' to 19. Pine Needle Oil (Picea vnlgaris). 0.888. a D = ~ 21 40' (pinene, phellandrene, dipentene, bornyl acetate, cadinene); from two-year cones, 0.892. a D = - 2oi2 r . Dwarf Pine Oil (Latschenkiefer) (Finns pumilio} . 0.865 to 0.875. a D - 5 to 9 (pinene, phellandrene, sylves- trene, 4 to 7 per cent, of bornyl acetate, sesquiterpene). Silver Fir Oil (Edeltannen) (Abies pectinata). Needles: 0.865 to 0.875. a n - - 20 to 60 (pinene, limonene, 4.5 to 7 per cent, of bornyl acetate, sesquiterpene). Young cones, 0.855 to 0.870. a D - - 60 to 80 (pinene, limonene, 0.5 to 3 percent, of ester, C ]0 H 17 OCH 3 CO). Hemlock Oil (Abies canadensis}. 0.907. ot n - ~ 20 to - 26 (pinene, camphene, bornyl acetate, sesquiterpene). Fir Needle Oil, Siberian (Abies siberica}. 0.91 to 0.92. a n = - 40 to 45 (bornyl acetate). Coriander Oil (Coriandritm sativnni}. 0.87 to 0.88. a D = -f 8 to -f 13 (pinene, linalool). Creeping Thyme Oil ( Thymns serpylliuni) . 0.905 100.930. a D - - i to 11 (thymol, carvacrol, cymol). Curled Mint Oil ( Mentha crispa}. 0.92 to 0.98. a D = 43 or lower (carvone). 664 CONSTANTS OF ROTATION OF ACTIVE BODIES Dill Oil (Anethum graveolens). German, Russian, Rouma- nian fruit : 0.895 to 0.915. <*/, + 70 to + 80 (limouene, carvone). East Indian fruit : a == + 40. Dog Fennel Oil (Eupatorium focniculaceum}. 0.935. a n - -f 1 7 50' (phellandrene). Elemi Oil ( Canarium spec. ) . 0.87 to 0.91. fr y , - + 45 ( phellandrene dipentene ) . Eucalyptus Oil {Eucalyptus % spec.). 0.86 to 0.95. ot n = 71 to -f- 20 according to source (citronellal, cineol, citronel- lone, pinene, phellandrene, alcohols, and aldehydes of the fatty series). Fennel Oil (Foeniculum vulgar e and varieties) . 0.920 to 0.987. <*D = : + 7 to -f 22. French, to -f 48 (pinene, phellan- drene, dipentene, limonene, fenchone, anethol). I-'rankhicense Oil (Bosu'ellea, spec.). 0.875 to 0.885. a D - 11 35' (pinene, phellandrene, dipentene). Galbanum Oil (Peucedanumgalbanifluum and varieties) .0.91 to 0.94. r y , = - 5 to -j- 20 (pinene, cadinene). Geranium Oil {Pelargonium, spec.). According to origin : 0.886 to 0.906. ot n - 7 to 16 (gerauiol, citronellol, and 1 9 to 33 per cent, of tiglic acid esters of the latter). Ginger Oil (Zingiber ojficinalc) . 0.875 to 0.885. a n = 26 to 44 (camphene, phellandrene). Gurjon Balsam Oil (Diptcrocarpus, spec.). 0.92 to 0.93. *D = - 35 to 106. Hop Oil (Humuljis lupulus}. 0.855 to 0.88. ct n = + 44o' and lower (terpenes, humulene). Juniper Oil (Junipcrus communis}. 0.865 to 0.885. <*/, = o to 18 (pinene, cadinene, juniper camphor). Lavender Oil (Lavcndula irra}. French: 0.885 to 0.895. <*t, - - 3 to 9 (linalool, linalyl acetate 30 to 45 per cent., geraniol, cineol). English: 0.885 to 0.900. a,, 7 to 10 (much cineol, linalyl acetate 7 to 10 per cent.). Lime Oil, Italian (Citrus limetta}. 0.872. a,,- 58 15'. ETHEREAL OILS 665 West Indian ( Citrus mcdica , var . acida ) . o. 880 to 0.885. a D = -h 35 to -f 40 (limonene, citral). Mace Oil (Myristica officinalis). 0.91 to 0.93. ct n -f 10 (pinene, myristicene). Mandarin Oil (Citrus madurensis). 0.854 to 0.858. ot n = -f 65 to -f- 75 (citral, limonene). Mastic Oil (Pistacia lentiscus). 0.855 to 0.870. a D = -f 22 tO -f 72. Onion Oil (Allium cepa} . 1.040. y , = - 5 (C 6 H 12 S,). O range- Floiver Oil, sweet {Citrus aurantiitm). 0.87 to 0.890. a D = -f- 16 to -f 29. Bitter, Citrus bigaradia, 0.87 to 0.885. a D = -j- 5 to -f 10 (limonene, linalool, linalyl acetate ) . Palma Rosa Oil {Andropogon SchoenantJius}. 0.888 to 0.896. a D - - i4o' to + i45' (dipentine, geraniol, ger- anyl esters). Pennyroyal Oil (Mentha pulegium, Hcdeoma pulegioides} . 0.93 to 0.96. flr/, = + 17 to -f 23 (pulegone). Peppermint Oil ( Mentha piper ita ) . German : o. 902 to o. 9 1 5. = - 37-49 CONVOLVULINIC ACID. Melting-point, 175. [a] D = - 45.30 > 26. Alkaloids Alkaloids of the Aconite Species ACONITINE, C. !3 H 45 NO 12 . Rhombic crystals ; melting-point, 188.5 (corr.). Alcohol p = 3.726, [a]^ = ~f 11.01 V-' =2.746, [a]~=+n.io j Formerly described as left-rotating. See Jiirgens. 3 Hydrobromide, C 33 H 45 NO 12 .HBr -f 2 1 /,H 1 O. Crystals. Water ^ = 5.183, [or] = -3'-3]. 4 Recrystallized / = 1.95, 30.47 / Is ACONITINE (Napelline), CjgH^NOj.,. Crystals; melting- point, 125 (corr.). Alcohol c= 7.86, [a]g = + 4-48 Hydrochloride, C 33 H 45 NO 12 .HC1 + H 2 O. White needles; melting-point, 268 (corr.). Water tf=t, [or]^ = 28.74 Hydrobromide, C 33 H 45 NO 12 .HBr. Needles; melting-point, 282 (corr.). Water ];s = - 26.94 3 PYROACONITINE, C 31 H 41 NO 10 . Water c = 1.121, [or]^ = 9Q-99 Hydrobromide, C 31 H 41 NO 10 .HBr. Water r = 2.136, [or] - -46.8 PYROACONINE, Hydrochloride, C 24 H 37 NO 9 .HC1 + H 2 O. Water ^ = 1.96, [or]g = - 102.07 " Kromer : Z. Oesterr. Apothek.-V., 49, 520. Dunstan, luce : J. Chem. Soc., 59, 281. Jahrcsb., 1885, 1722. Dunstan, Ince : Loc. cit. Dunstan, Harrison : J. Chem. Soc., 63, 445- Dunstan. Carr : Ibid., 65, 176. 668 CONSTANTS OF ROTATION OF ACTIVE BODIES ACONINE, C^H^NO,,. Amorphous; melting-point, 132 (corr. ). By hydrolysis from isaconitine. Water p 3-534, rfjs = i.oi, [a]>$ -r -f 23 Hydrochloride, C, G H 41 NO n .HCl -f 2H.,O. Crystals ; melting- point, 175.5 (corr.). Water, /> ^ 5.75 (anhydrous), ig), an, 254. ALKALOIDS 669 ASPIDOSPERMATINE, C 22 H 2(J N 2 O 2 . Wart-like lumps ; melting- point, 162. Alcohol (97 vol. p. c.) .......... c=2, [a]^5 _= _ 72.3 i QUEBRACHINE, C 21 H 26 N 2 O 8 . Prisms ; melting-point, 214 to 216 (uncorr.).- Alcohol (97 vol. p. c. ) ---- c == 2, [or] js = 62.5 | 3 - Chloroform .............. <: 2, = -f 18.6 PAYTINE, C 21 H, 4 N,O -f- H 2 O. Crystals ; melting-point, 156. Alcohol (96(?) vol. p. c.) ....... <:-= 0.454, []g - - 49-5 * ATROPINE, C 17 H,,NO,. Crystals ; melting-point, 1 15 to 1 16. Absolute alcohol. . c = 3.22, [<*]^f = 0.4 5 Alcohol .......... c = 6.67, [a] = - i .89 fi According to Ladenburg 7 and Gadamer 8 atropine is optically inactive. The Atropinum naturale is a mixture of atropine and hyoscyamine. Sulphate, (C 1T H 23 N0 3 ) 2 H 2 S0 4 + H,O. Crystals. Water ...... c 2 (anhydrous), [a]^ = - 8.8 9 SCOPOLAMINE, C ]T H. ;1 NO 4 + H 2 O. Melting-point, 59. Absolute alcohol ---- p = 2.65, [apj = - 13.7 10 Melting-point, 56. Water ........ [or]/? - - 14.97 n HYOSCYAMINE, C n H 23 NO 3 . White needles ; melting-point, 108 to 109. Absolute alcohol ......... c = 17.2, [a~\ 2 = 2i.6o c ......... C 12.4, - 21.76 ......... c= 6.2, " = 21.25 " ......... c= 3.1, " = 20.26 Alcohol (50 per cent) ..... c= 12.4, 20.27 1 Hesse : Loc. cit. 259. Hesse : Her. d. chem. Ges., 13, 2308. 3 Hesse : Ann. Chem. (I^iebig). ail, 265. 4 Hesse : Ibid., 166, 272. '-> Hesse : Ibid., 271, 101. Will, Bredig : Ber. d. chera. Ges., ai, 2777. 7 Ber. d. chem. Ges., ai, 3065. Arch. Pharm., a34, 543. Hesse : I^c. cit. 10 Hesse: Ann. Chem. (Liebig), ayi, in. 11 Zuboldt : Diss., Marburg, 1895. 12 Will : Ber. d. chem. Ges., ai, 1717. 670 CONSTANTS OF ROTATION OF ACTIVE BODIES Alcohol c i to 12, [tt] - - 21.016 0.0154 c l Absolute alcohol, c = 3.22, [o-]^ = - 20.3 - By action of bases, hyoscyamine is converted into atropine. Sulphate, (C 17 H 23 NO 3 ),H,SO, + 2H,O. White needles; melting-point. 201. Water c 2 (anhydrous), [tfpj - 28.6 4 Water p = 2.9, [a]/, =i - 26.8 to 27.3 5 Oxalate, anhydrous. Water /> = 1.6, f - - 24.07 6 PSEUDOHYOSCYAMINE, C 17 H 23 NO 3 . Yellow needles ; melting- point, 133 to 134. Absolute alcohol p - 5.26, [>]],= 21.15 7 HYOSCINE, C, 7 H. a NO 4 (?). Hard, transparent, resin-like masses ; melting-point, 55. Absolute alcohol c= 2.65, [arpj = - 13.7 " The rotation is greatly decreased by addition of a small amount of sodium hydroxide solution. Hydrobromide, C 1T H M NO 4 .HBr + 3H,O. Rhombic crystals. Water c = 4, M'J = - 22.5 4 BEBIRINE, C, H H. 21 NO 3 . Melting-point, 214. Absolute alcohol p = 1.6, [or] 5" - 298 !) BERBERINE, C 20 H 17 NO 4 + 4H 2 O. Crystals ; melting-point, 120. Inactive. OXYACANTHIN, C lh H 19 NO 3 . Crystals; melting-point, 138 to 150 (from water) ; 208 to 214 (from alcohol), i vol. alcohol (97 vol. p.c) -}- 2 vol. chloroform, c = 4, [<*]$ = -}- 131.6 1(( Hydrochloride C 1H H J9 NO 3 .HCI -f 2H. 2 O. Needles. Water c 2, [or]$ + 163.6 4 1 Hammerschmidt, Will and Bredig : Ber. d. chem. Ges., 21, 2784. 2 Hesse : Ann. Chem. (Iiebig), 271, 103. Will and Bredig : Ber. d. chem. Ges., 21, 2777. Hesse : IMC. oil. Gadamer : Arch. I'harm., 234,543. Gadamer : Luc. cit. Merck : Arch. Pharm.. 231, 117. * Hesse: Ann. Chem. (I.iebijo, 271, no. v Scholtz : Ber. d. chem. Ges.. 29, 2058. 10 Hesse : Ibid., 19, 3192. o -i ALKALOIDS 671 HYDRASTINE, C 21 H. n NO H . Rhombic crystals ; melting-point, 132. Chloroform c = 2.552, [or]^ . 67.8 V Water - 2 inol. HC1-. c -- 4.050, - 127.3 Chloroform c 3.042, [<*]/> = - 57. Methylhydrastine Hydrochloride , C 2 ,H 23 NO 6 .HC1. Crystals; melting-point, 241. Inactive in aqueous solution/' HYDRASTININE, C H H n NO, -f H 2 O. Crystals ; melting-point, 1 1 6 to 117. Inactive in aqueous solution. 1 CHELERYTHRINE (Sanguinarine), C 17 H 15 NO 4 . Crystals; melting-point, 160. Inactive. Cinchona Alkaloids ' CUPREINE, C 19 H 22 N 2 O 2 + 2H 2 O. Prisms ; melting-point, 197 to 198. The following determinations on the alkaloid and its salts were made by Oudemans. 15 Absolute alcohol. .. c --- 0.69, 1.24, 1.78 " O] - -175-4, -175-5, -173-3 Alcohol (97 p. c.).. c= 1.50, [*]* = -175-3 Neutral Hydrochloride, C ]C( H.,. ; N 2 O r HCl -h H 2 O. Colorless needles. Water r ^ 0.567, []^ - 157.1; alcohol, [or]$ = -184.7 " c = 0.871, " : -154.8; " = -182.0 Absolute alcohol c 0.927, " = 169.7; -199.8 " c 1.421, '' -167.3; -196-7 Acid Hydrochloride, C w H M N 2 O r 2HCl. Hard crystals; -f 2H 2 O, rhombic crystals. Water, c= 1.194 (hydrated), [nr]}j- 211.0 ; alkaloid, [<*]^ 7 = 283.8 " c= 2.508 " 210.6; 282.3 " c 4.687 " -206.0; 276.2 " c= 8.589 " " = -200.4; " -268.6 41 f- 17.278 '' " = - 191.1; " 256.2 1 Freund, Will : Ber. d. chem. Ges., 19, 2797. - Eijkman : Rec. trav. chim. Pays-Has, 5, 290. ' Freund and Rosenberg : Ber. d. cheni. Ges., 23, 404. * Freund, Will : Loc. cit. Numerous earlier observations on the rotation of the alkaloids were made by Bouchardat : Ann. chim. phys., [3], 9, 213: Bouchardat and Boudet : J. pharm. Chim., [3], 23, 288 : Buignet : Ibid., [3], 40, 268 ; DeVrij and Alluard : Compt. rend., 59, 201. The numerical data refer to Biot's red ray, but they cannot be used, as the nature of the solvent and the concentration are, in most cases, not given ' Rec. trav. chim. Pays-Has. 8, 153. 672 CONSTANTS OF ROTATION OF ACTIVE BODIES Water, ; 5.26 cc. norm. HC1 c =-- 2.088 (hydrated), [a]^ = 210.8 alkaloid, " = -282.6 -f- 10.53 cc. norm. HC1 c = - 2.070 (hydrated), " = 210.2 alkaloid, " = 281.7 -[- 22.73 cc. norm. HC1 c 21.987 (hydrated), " = 205.5 alkaloid, " = 276.4 -(- 52.63 cc. norm. HC1 c - 2.085 (hydrated), " = 199.5 alkaloid, " = 267.8 -f- 90.90 cc. norm. HC1 c =-. 2.056 (hydrated), " = 194.0 alkaloid, " = 260.9 Neutral Hydrobromide , C 19 H 22 N 2 O,.HBr -f H 2 O. White needles. Water c = 0.491, [a]$ = - 145.8 ; alkaloid, [<*]$ = - 192.7 " = 191.1 M =-183-7 " = 181.1 =I.I&>, = 144.8 Absol. alcohol c 1.434, " = - 139.2 c= 1.687, " = 137.3 Acid Hydrobromide, C lf H lt N f O,.2HBr. Crystals. Water =1.564, [a]g = - 189.0 ; alkaloid, [or]^ = -287.7 " 2.991, " - -184.6; 281.0 " C= 7.012, " = -176.8; " *' : -268.6. Neutral Hydroiodide, C 19 H 22 N 2 O r HI. Crystals. Water c 0.802, [a]g = -126.3; alkaloid, [or]^ ^ -178.4 Absol. alcohol c = 0.982, " = 128.3; " " -180.9 Acid Hydroiodide, C 19 H 22 N 2 O 2 .2HI (at 150). Yellow crystals. Water =1.504, [a]g = - 151.2 ; alkaloid, [a]^ ~ -283.2 " =1.491, " 5 -147-6; -276.4 Neutral Nitrate, C la H 22 N 2 O 2 .HNO 3 + 2H 2 O. White needles. Water c 1.129, [ a l^ = 138.4 ; alkaloid, [a]$ = 182.5 Add Nitrate, C 19 H 22 N 2 O 2 .2HNO 3 + H a O. Light yellow crystals. Water =1.283, Olg = -1794; alkaloid, [a] ^ = -289.1 " =2.347, " = -193-4; -283.2 -=3.815, " = -188.9; -276.7 " 5-035, " -189.8; " " -278.0 " = 6.551, " .. -188.9; " -276.7 Chlorate, C 19 H 22 N 2 O 2 .HC1O 3 . Fine white needles. Water c ----- 1.033, [or]j = - 144.9 \ alkaloid, [a]% - 184.4 ALKALOIDS 673 Neutral Sulphate, C 19 H M N,O r H a SO 4 + 2H 2 O. Crystals. Water 1 = 0.905, [<*]% = - 202.4 ; alkaloid, [a] g = -289.9 = 1.087, l>]g=- I97-I " IXlg =-282.3 [a]g-s= -281.7 [a]g = -282.8 [a] = 286.6 [a] # = 285.3 .................... -167.5 169.7 170.4 171.9 174.3 176.1 176.5 f Benzene. Toluene. Chloroform. *r = o.6i 0.39 0.775 1.465 [a] = -136 -127 -126 -117 Chloroform-alcohol mixture- . . p = 0.7 to 2.3, [a]^ 8 - 1 ^ = - 164.4 ' Hydrochloride, C 20 H., 4 N 2 O,- H C1 -f- 2H,Q- Long asbestos- like prisms. a. Water ......... c = 1.54 to 1.62 (calc. as alkaloid), [cr]^ = - 133.7 (anhydrous) ; alkaloid, [a] 1 ^ = - 163.6. Absolute alcohol c = 1.54 to 1.62 (as alkaloid), [or] 15 = - 138.0 (anhydrous) ; alkaloid, [or])? = - 169.0. The following data, b to e, are by Hesse. 3 b. Water .............. c = i to 3, [a]'j = - 144.98 -f 3.15 r, alkaloid, " = - 167.41 -f- 4.71 c, 2mol. HC1 + water, c = i to 7, "a - 229.46 -f 2.21 r, alkaloid, '* = - 280.78 + 3.31 r, Alcohol (97 vol. p.c.) c = i to 10, [a]/> = - 147.30 + 1.958 c 0.1039^' -f 0.0021 1 c\ c. When aqueous alcohol was used, the specific rotation showed a maximum with 60 volume per cent, of alcohol. For c = 2 and t = 15, the following figures were obtained. Alcoholic strength: Vol. p. c. 97 90 85 80 70 O]z? - -143.86 -160.75 -168.25 -174.75 -182.27 Alcoholic strength : Vol. p.c. 60 50 40 20 o( water) [>]/>:= -187.75 -187.50 -182.82 -166.59 -I38.75 Chloroform-alcohol mixture, c 2, []J| = - 126.25 d. Chloroform. Anhydrous salt, c = 0.9 to 9 : [] = - 81.81 + 23.756 c - 3.9556 c 1 + 0.2198 r\ e. Dilute hydrochloric acid. To i mol. of salt, n mol. HC1, C = 2. n o i 2 4 jo 16 = -138.75 -223.2 -225.7 -223.6 -213.9 -209.5 Ztschr. anal. Chem., 27, 549. a Oudeinans : Ann. Chem. (Uebig), i8a, 46. 1 Ibid., 176, 210. ALKALOIDS 675 Fuming hydrochloric acid, c 2, [>]'j = - 158.8. LHquinine Sulphate, (C 20 H., 4 N 2 O 2 ) 2 H 2 SO 4 + 8H 2 O. Crystals. Alcohol (80 vol. p. c. ) r == 2, M/5 = 162.95 " (60 " ) C=2, " =166.36 Water + 4 mol. HC1 r = 2, " = 239.2 Chloroform-alcohol mixture. . . r = i to 5, " = 157.5 -f- 0.27 c 40 cc. normal HC1 -j- water c = 8 (anhydrous), [nr]^ = 229.03 * (C 20 H 24 N 2 O a ) 2 H 2 S0 4 + 7 V 2 H 2 0. Absol. alcohol, c = 1.54 to 1.62 (calculated as alkaloid) ; salt anhydrous, f a] 17 = 157.4 ; alkaloid [<*]^ = 214.9 3 Absol. alcohol, c = 1.3, [a] 1 /, = - 155.2 ; c 2.0, [ar]^ = - 158.4 * Quinine Sulphate, C 20 H 24 N 2 O 2 .H,SO 4 -f 7H 8 O. Rhombic crys- tals. Water c i to 6, salt, [a]^ = -- 164.85 -f 0.31 c ; alkaloid, [<*]^ = 278.71 + 0.89 c, Alcohol (97 vol. p. c.) c = 2, [a]g = 134.75 5 (80 ) c=z, " = 142.75 (60 " ) f=a, " = - I55-9 1 Chloroform-alcohol mixture- c = 2, " = 138.75 Water c 1.54 to 1.62 (calc. as alkaloid), 1 s [a]g 213.7 (anhydrous) ; alkaloid, [tr] 1 ^ = 278.1 Absol. alcohol c = 1.54 to 1.62 (calc. as alkaloid), [a]^7 = 134.5 (anhydrous) ; alkaloid, [cr]^ = 227.6 The specific rotation of alcoholic solutions, with c = 2, de- creases, with elevation of temperature, 0.65 for iC. 6 Quinine Disulphate, C 20 H 24 N 2 O,.2H 2 SO 4 -f 7H 2 O. Prisms. Water c = 2 to 10, [a] T J = - 170.3 -f 0.94 c\ 7 Alcohol (80 vol. p. c. ) f=i, " = 154.5 (80 j C^H^NA^H.SO, + 4 H 2 0. Water ............ c = 2 to 10, [a] js = _ 155.69 + 1.14 c Alkaloid " = - 284.48 -f 3.79 c 1 Hesse : Ann. Chem. (Liebig), 176, 213. 2 Hesse: Ibid., 205, 219. :! Oudemans : Ibid., i8a, 46. Oudemans: Rec. trav. chim. Pays-Has, i, 27. 5 Hesse : Ann. Chem. (I,iebig), 176, 215. e Draper: Am. J. Sci.. [3], n, 42- J Hesse: Ann. Chem. (Liebig), 176, 217. 676 CONSTANTS OF ROTATION OF ACTIVE BODIES i mol. quinine hydrate -j- 3 mol. H.,SO 4 -f water to 100 cc., c = i to 5. [o-pj (calc. as hydrate) = - 246.63 -f- 3.08 c [a]g (alkaloid) - 287.72 + 4.19* i mol. sulphate -f 2 mol. H 2 SO 4 -f- water to 100 cc. c = i to 10. Salt [a]s = - 171.68 -f 0.78^; alkaloid, [a] J j = - 290.36 + 2.23 c i mol. disulphate + i mol. H 2 SO 4 + water to 100 cc. c = 2 to 6. Disulphate [a]g= - 153.87 -f 0.92 r; alkaloid, [a] ^ = 281. 15 + 3.1 \c * Oxalate, (C M H J4 N 2 O 2 ) 2 C 2 H 2 O 4 -f 6H 2 O. Long prisms. Chloroform-alcohol mixture ..... c = i to 3, [orj^s = - 141.58 -j- 0.58 c 2 .Sa/l w/M j J/]/>= - 158 * Sulphate, (C 19 H 22 N 2 2 .C 20 H 24 N 2 0. 2 ) 2 -H 2 S0 4 + 6H,O. Color- less prisms. 40 cc. norm. HC1 -j- 60 cc. water, c = 5 (anhydrous), [a~\ i s = 235.6 * Water with 0.5 p. c. H 2 SO 4 , c = 5, [a] D = - 209 ) 3 " " I.O " " f=5, " = 220 j Nitrocamphor Quinine, C W H 24 N 2 O 2 -f- 2C 10 H U (NO 2 )O + H a O. Needles; melting-point, about 131 (decomposes). Alcohol p = 2.72, [or] y = + 45-9 * Acetyl Quinine, C 20 H 23 (C 2 H 3 O)N 2 O 2 . Prisms; melting- point, 108. Alcohol (97 vol. p. c.) c = 2, [a]'J = - 54-3 j 5 3 mol. HC1 + water c = 2, - 114.8 j Propionyl Quinine, C, H 23 (C 3 H 5 O)N,O 2 . Rhombic prisms; melting-point, 129. 3 mol. HC1 + water c = 2, [a]'J = - 108.8 6 APOQUININE, C 19 H 82 N 2 O 2 -f 2H 2 O. Amorphous ; melting- point, 1 60 (turns brown). Alcohol (97 vol. p. c.) c = 2 (anhydrous), [a]g = - 178.1) 7 3 mol. HC1 + water c = 2 - 246.6 j Alcohol (97 p. c.) c = 0.7877, Mg= - 217-1 } Diacetylapoquinine, C 19 H 20 (C 2 H 3 O) 2 N 2 O 2 . Amorphous. Alcohol (97 vol. p. c.) f=2, \a\^= - 6i.8| 9 3 mol. HC1 -f water c = 2, -107.5 j Hydrochlorapoquinine, , C 19 H 23 C1N 2 O 2 -f 2H 2 O. Flakes ; melt- ing-point, 1 60. Alcohol (97 vol. p. c. ) c = 2 (anhydrous), [or]g = - 149- J } w 3 mol. HC1 + water c=2 - 245-7 1 Howard, Hodgkin : J. Chem. Soc., 41, 66 : Ber. d. chem. Ges., 15, Ref. 379. 2 Hesse : Ann. Chem. (I^iebig), 325, 104. 3 Howard, Hodgkin : J. Chem. Soc., 41, 66; Ber. d. chem. Ges., 15, Ref. 734. Cazeneuve : Bull. soc. chim., 49, 97. Hesse r Ann. Chem. (I,iebig), 205, 3'7- Hesse : Loc. cit., p. 358. Hesse : I^oc. cit., p. 323. I^ippmann : Ber. d. chem. Ges., 28, 1972. Hesse, p. 337. Hesse, p. 341. 678 CONSTANTS OF ROTATION OF ACTIVE BODIES CHITENINE, C W HN 2 O 4 + 4H 2 O. Crystals. Alcohol (d 0.958) p = 0.1093, d = 0.9595, []/> = 142.7 * QUINICINE, C 20 H 24 N 2 O 2 . Amorphous; melting-point, 60. Chloroform c=2, [a]% = + 44.1 2 Oxalate, (C 20 H a4 N 2 O 2 ) 2 C 2 H 2 O 4 + 9H 2 O. Crystals. Chlorofonn-alcohol mixture c I to 3, [<*]$ = -f 20.68 1.14^-1 s Water c = 2, = -f 9.54 Water -f 2 mol. H 2 SO 4 c = 2, = + 15.54 QUINIDINE (Conquinine), C 20 H 24 N 2 O 2 . Crystals (from ben- zene) ; melting-point, 171.5 (corr.). i vol. alcohol (97 vol. p. c. ) (/> = 2.1, [<*]^' 9 = + 269.7 )* 4- 2 vol. chloroform (p = 1.06, [<*]$ - - -j- 274.7 Benzene r 1.62, =+195.2 Toluene ^7=1.62, " =+206.6 Chloroform ^=1.62, " = + 228.8 Alcohol ( wt. p. c. ; loo. o 95.3 90.5 85.0 80.0 75.0 \* W%= +255.4 257.6 259.0 259.4 259.3 259.4} Alcohol (97 vol. p. c.) c = i to 3, [tf]^f = + 269.57 3.60^! T Chloroform c= 1.756, " == + 230.35 J Alcohol, \a]j = 268.6 ; methyl alcohol, [] 7 =257.5. Wyrouboff explains this difference by formation of alco- holates : (C W H 24 N 2 O 2 )CH 4 O, [] y = + 236.1 and (C 20 H 24 N 2 O 2 ) C 2 H 6 0, [XL = + 235.3. CwHj.NA + 2 ! / 2 H 2 O (from water) : Alcohol (97 vol. p. c.) ] = + i?9-54 1 ' alkaloid, " == + 215.55 I 4 mol. HC1 + water, c = 2 (anhydrous), = + 286.4 alkaloid, " = + 329.8 5 mol. H.,SO t + water, r= 2 (anhydrous), " == + 281 alkaloid, " == + 323 Absolute alcohol c = 2.8 (anhydrous), [a]$ = + 211.5 alkaloid, " + 255.2 Alcohol (80 vol. p.c.) c=2, [or]^s = + 218.2 (60 " ) c= 2, " = + 227.0 Chlorof. -alcohol mix. c 2, = + 209.25 Chloroform c = 3 (anhydrous), = + 184.2 f =5. r + l8 - 1 Acid Sulphate, C 20 H 24 N 2 O 2 .H 2 SO 4 + 4H 2 O. Asbestos-like needles. Water c = 2 to 8, [a]g = + 212.0 0.8 c 3 alkaloid, = + 3 2 3- 2 3 1.86 c 2 mol. H 2 SO 4 water c = i to 10, " = = + 215.49 1.41 c alkaloid, " - = + 328.55 3.27 c Alcohol (97 vol. p. c.) c = 2, = + 183 Nitrate, C 20 H 24 N 2 O 2 .HNO 3 . Short thick prisms. Absol. alcohol, c = 2.17, [a] 1 ^ = 199.3 ; alkaloid, [a]g = + 232.6 1 Hesse: Ann. Chem. (I^iebig), 176, 225. '-' Oudemans : Ibid., 182, 49. 3 Hesse : Loc. cit. * Hesse. 5 Oudemans. *> Oudemans : Loc. cit. 680 CONSTANTS OF ROTATION OF ACTIVE BODIES Oxalate, (C 20 H 24 N 2 O 2 ) 2 C 2 H 2 O 4 -f H,O. Very small crystals. Chloroform-alcohol mixture c i to 3, [<*]$ = + 189.0 2.18 c 1 Acetyl Quinidine, C, H 23 (C 2 H,O)N 2 O,. Amorphous. Alcohol (97 vol. p. c.) c = 2, [a]js = + 127.6 | 2 3 mol. HC1 + water c 2, = + 158.6 J APOQUINIDINE, C 19 H 2a N a O 2 + 2H 2 O. Amorphous ; melt- ing-point, 137. Alcohol (97 vol. p. c.) c 2 (anhydrous), [a] == + 155.3 1 3 3 mol. HC1 + water c = 2 " = + 216.5 J Diacetylapoquinidine , C 19 H 20 (C 2 H 3 O),N 2 O 2 . Resin ; melting- point, 60. Alcohol (97 vol. p. c. ) c = 2, [or] }$ = + 40.4 ! 4 3 mol. HC1 + water r = 2, " == + 78.4 j Hydrochlorapoquinidine, C 19 H 23 C1N 2 O 2 -f 2H 2 O. Crystals ; melting-point, 164. Alcohol (97 vol. p. c.) c = 2 (anhydrous), [tf]$ + 203.7 i 5 3 mol. HC1 + water c=2 " = + 258.4 j Diacetylhydrochlo rapoquin idine, C 19 H 21 (C 2 H 3 O) 2 C1N 2 O,. Rhombic leaves (from ether) ; melting-point, 168. 3 mol. HC1 + water c 2, [or] jf = + 94.6 6 CINCHONINE, C 19 H 2a N 2 O. Crystals; melting-point, 255.4 (corr.). 7 Alcohol (97 vol. p. c.) r = o.5, [a] y> = + 226.26 -j 8 t=I, " = + 225.96 j- Chloroform-alcohol mixture c= i to 5, " == + 238.8 - i.46rJ Absolute alcohol c = 0.5 to 0.75, [or] % + 223.3 1 9 Chloroform = 0.455, = + 214.8 = 0.535, " -= + 212.3 ^=0.560, = + 209.6 J Chlorof. -alcohol mixture p = i .061, rfj- 8 = 1.2508, [or] Jf- 8 = + 239.40 10 ^=2.123, fllj- a =1.2497, []^- 2 = + 234.55 i Absolute alcohol 0.4715, []], = + 222.92 u Hesse. Hesse : Ann. Chem. (Liebig), 305, 318. Hesse : Ibid., p. 326. Hesse : Ibid., p. 327. Hesse : Ibid., p. 343. Hesse Ibid., p. 352. I,enz : Ztschr. anal. Chem., 37, 572. Hesse : Ann. Chem. (lyiebig), 176, 228. Oudemans : Ibid.., 182,44. 10 lyenz : Loc . nt. 11 I'um : Wien. Monatsh. Chem., 13, 683. ALKALOIDS 68l From benzoyl cinchonine. Absolute alcohol c = 0.75, []^ 7 -f- 233.1 l Neutral Hydrochloride, C 19 H 2 ,N,O.HC1 -f 2H,O. Rhombic crystals. Water c = 0.5 to 3, [a]^ = + 165.50 2.425 c^ * alkaloid, = -j- 204.46 3.7 c 2 mol. HCI + water c = I to 7, = + 214.0 1.72 c alkaloid, " = + 264.37 2.625 c Alcohol (97 vol. p. c.) c i to 10, [tfpj = 179.81 6.314*: + 0.8406 0.0371^ | (80 " ) c= 2, " == + 188.9 (60 " ) C=2, ""=? + 195-5 Chlorof. -alcohol mixt. c= 2, - + 152.0 j Add Hydrochloride, C 19 H 22 N 2 O.2HC1. Prisms. Water t = 3, [or] = -f 206.1 :i Neutral Sulphate, (C 19 H 2 .,N 2 O) 2 .H.,SO 4 + 2H 2 O. Monoclinic crystals. Water c = i to 2, [ar]*s = -f 170.3 0.855 c - * alkaloid, " = -j- 206.79 ~~ I - 2 6 <" 2 1 2 mol. H 2 SO 4 -j- water c = 0.5 to 6, " == + 219.10 1.85 c alkaloid, = + 266.07 2.69 c ^ Alcohol (97 vol. p. c.) r = 3 to 10, " ==+ 193.29 0.374 r (80 " ) C=2, " = + 202.95 " (60 " ) <~ = 2, = + 204.14 Chloroform-alcohol mixture- c = 2, = + 185.25 According to Wyrouboif, 5 the sulphate and selenate of cin- chonine ([<*]y == 234) take up one molecule of crystallization alcohol from alcoholic solution. (C 19 H 29 N 2 0) 2 H 2 S0 4 + C 2 H 6 0, [or], - + 185 ; (C 19 H 22 N 2 0) 2 H 2 Se0 4 -|- C 2 H 6 O, ; = + 182.5. Oxalate, (C 19 H 22 N 2 O) 2 C 2 H 2 O 4 + 2H 2 O- Prisms. Chloroform-alcohol mixture c = i to 3, [r]^ = 165.46 0.763 r 6 Wyrouboff 7 gives, also, the constants of rotation of cincho- 1 I^eger: Compt. rend., 117, no. - Hesse : Ann. Chem. (I^iebig), 176, 230. 3 Hesse : Ibid., 376, 91. 4 Hesse : Ibid., 176, 231. "' Compt. rend., 115, 832. c Hesse: Ann. Chem. (Liebig), 176, 232. ~ Ann. chim. phys., [7], i, 5. 682 CONSTANTS OF ROTATION OF ACTIVE BODIES nine salts which have separated from different solutions com- bined in crystalline form with part of the solvent. Diisobutylcinchoninehydrobromide, C 19 H 22 N 2 O.C 4 H,,Br-f H..O. Melting-point, 176. Water p = i, [a]}? = -f- 125 ' Acetylcinchonine , C 19 H 21 (C 2 H 3 O)N 2 O. Amorphous. Alcohol (97 vol. p. c.) c 2, [r]^f = + 114.1 > 2 3010!. HC1 + water c = 2, " ^+139.5 J Hydrochlorcinchoninedihydrochloride, C 19 H 2S C1N 2 O . 2HC1. Monoclinic crystals. Water c = 3, [or]g = -f 185.9 j 3 i mol. HC1 -r water r = 2.4 = -f 187 i /?-CINCHONINE. C 19 H 22 N 2 O. Probably identical with one of the bases made by Jungfleisch and Leger (see below ) . Crystals ; melting-point, 250 to 252. Absolute alcohol c = 0.4715, [#]/> = -f 195.77 4 tf-CiNCHONiNE, C 19 H 22 N 2 O. Prisms; melting-point, 150. Alcohol (97 p. c. ) c = i, [or] $ = + 125.2 5 2 mol. HC1 -f water... f=i, =+176.9 4 " "-f '* ... c=i t " = -f 178.2 a-IsociNCHONiNE (Cinchoniline) , C 19 H 22 N 2 O. Anhydrous monoclinic crystals ; melting-point, 126;" 125 to i27; 7 130.4 (corr.). 1 Absolute alcohol c = 3, [a]^ = -f- 5 1 .6 !( Alcohol (97 p. c) c=i, = -f 53.22 ) 10 f = 0.5, Mg = + 50.3 i According to Hesse, this last value is wrong, because, as he finds, the rotation increases with increasing dilution. 2 mol. HC1 + water c=i t [a]g = + 59.15 4 " " + " ci t = + 63.10 Vial : J. pharm. Chim., [5], 30, 52. Hesse : Ann. Chem. (lyiebig), 205, 321. Hesse. Ibid., 376, no. Pum : Wien. Monatsh. Chem., 13, 683. Jungfleisch, Leger : Compt. rend., 118, 31. Hesse : Ann. Chem. (Uebig), 376, 93. Comstock, Konigs : Ber. d. chem. Ges., 30, 2521. Jungfleisch, Llger : Compt. rend., 106, 658. Hesse. Jungfleisch, I^eger. ALKALOIDS 683 Hydrochloride, C 19 H 2 ,N,O.HC1 + 2 or 3H 2 O. Prisms ; melt- ing-point, 226. In pure aqueous solution, the salt with 2 molecules of water, for c = 2.5, and the salt with 3 molecules, for c = 4, do not rotate the plane of polarized light. Salt with 2 mol. H,O. 3 mol. HC1 + water, c=4, [a]g = + 40.6 ; alkaloid, [a]$ = -f 50.6 ! Salt with 3 mol. H 2 O. Water r=i, [<*]=+ 5-o 2 mol. HC1 -f water c = i, = + 59-3 4 " " + " | 5 ^7 = 0.5, [ajg = 61.16 I 2 mol. HC1 -f water c=i, 40 4 " " -r " c=i, -38.21 - 1 Hydrochloride, C 19 H, 2 N 2 O.HC1 -f H 2 O. Prisms; melting- point, 201. Water c=I, [a]^ = - 68.10 -j 6 44 f = 2 - 71. < 2 mol. HC1 -f water. . . c = 2 2 " "-r " ^=5 5 -f ... c=2 Chloroform c = 2 Hesse. Jungfleisch, L,ger. Hesse : Ann. Chem. (Liebig), 276, 97. Hesse : Ibid., a6o, 215 Jungfleisch, I,6ger : Compt. rend., 106, 358. Hesse : Ann. Chem. (Uebig), a6o, 216. 684 CONSTANTS OF ROTATION OF ACTIVE BODIES As distinguished from all other di-acid cinchona alkaloids, the rotating power of this alkaloid increases with the concen- tration, but decreases with increase of acid. C 19 H 22 N 2 O.HC1 +.2H 2 O. Colorless, prismatic needles; melting-point, 213 (corr.). Water c = i, [a] 55 = -65.41' CINCHONIDINE, C 19 H 22 N 2 O. Needles. Alcohol (97 p. c. ) c 0.75, [a]- D = + 195.0 2 Alcohol 1=0.75, [a]^ = + 201.4 -\ 3 2 HC1 + water p = i, =+228.9 I 2HC1+ " p = 1.5 "== + 225.13! 4HC1+ " p=i t =+ 226.3 J APOCINCHONINE, C 19 H 22 N 2 O. Prisms ; melting-point, 209 ; 4 228. 5 Alcohol (97 p. c.) r=i, [ a ]$ = + 160.0 ^ 6 2 mol. HC1 + water- r 2, " =-(-212.5 3 " " + " .... c= 2, " == -f 212.3 Chloroform mixture.... c = 3, =+ 197.5 1 Absolute alcohol c = 1.56, [or] + 159.7 8 Hydrochloride, C 19 H 22 N 2 O.HC1 + 2H 8 O. Needles. Water c 0.006, [or]g = + 139.0 ; alkaloid, [a] 1 /; = -f- 171.9 c = o.oio, = + 138.5 ; =+171.3 f = 0.015, = + 138.5; =+171.3 Hydrobromide, C 19 H 22 N 2 O.HBr + H 2 O. Needles. Water c = 0.0076, [a] = + 126.2 ; alkaloid, [a]g =i + 168.7 Hydroiodide, C 19 H 22 N 2 O.HI + H 2 O. Needles. Water c 0.006, [a]g = + 117.2 ; alkaloid, [a] 1 ^ = + 175.5 Sulphate, (C 19 H 22 N 2 O) 2 .H 2 SO 4 + 3H 2 O. Needles. Water c 0.0048, [or]^ = + 130.0 ; alkaloid, [a]'J = + 164.0 Chlorate, C 19 H 22 N 2 O.HC1O 3 . Needles. Water == 0.0069, [a]^ = + 129.0 ; alkaloid, [or]jf = + 166.2 Junjffleisch, L6ger : Compt. rend., 106, 358. Jungfleisch, I^eger : Ibid., 105, 1257 ; Bull. soc. chim., 49, 747. JungfleiMh, Mger: Compt. rend., 118, 536. Hesse : Ann. Chem. (I^iebig), 205, 330. Hesse: /hid., 276, 115. Hesse : /hid., 205. Hesse : /Wrf., 276. iiians : Rec. trav. chim. Pays- Has, i, 175. ALKALOIDS 685 Perchlorate, C 19 H,,N 2 O.HC1O, + H,O. Needles. Water = 0.0052, [<*]= + 124.9 \ alkaloid, O] = -|- I75-3 01 Oudemans 1 studied, further, the effect of acids (HC1, HBr, HN0 3 , HC10 3 , HC10 4 , CH.A, C 2 H 4 O,, H 2 SO 4 , C 2 H 2 O 4 , H 3 PO 4 , C 6 H 8 O.) on the rotation of apocinchonine. Acetylapocinchonine, C 19 H 21 (C 2 H 3 O)N 2 O. Crystals. Alcohol (97 vol. p. c.) c = i, M'J = + 71.4 > - 3 mol. HC1 -f- water ** = 2, = + 97.9 j Chlorapotinchonine , C 19 H, 3 C1N.,O. Needles ; melting-point, i 97 . Alcohol (97 vol. p. c.) ... r = o.5, M^ = + 205.4 3 mol. HC1 water c = 2, = + 208.0 Alcohol (97 vol. p. c.) c 0.4745, []^ = + 2i .... C = 0.2655, [] X J = + 2I Neutral Hydrochloride, C 19 H, 3 C1N 2 O.HC1 -f H 2 O. Needles. Water c = 0.0045, []^ = + l6 5-9 ; alkaloid, [cr]^ = -f 193.2 Add Hydrochloride, C 19 H 23 C1N 2 0. 2HC1. Prisms. Water ^ = 0.0197, [or] 1 ^ =+ 185.0 ; alkaloid, [a] = + 226 Sulphate, (C 19 H 23 C1N 2 0),H 2 S0 4 + 3 H 2 O. Needles. Water c= 0.005, C^Pj = + 156.6 ; alkaloid, [a]^ = -f 192.5 Nitrate, C 19 H 2S C1N 2 O.HNO 3 . Needles. Water c= 0.005, |>] = + 173.5 ; alkaloid, [a]^ = + 194.8 Chlorate, C 19 H 33 C1N 2 O.HC1O 3 . Crystals. Water c = 0.005, [or] = + 155.3 ; alkaloid, [or] = -f I94-9 04 On the effect of acids (HC1, HBr, HNO 3 , HC1O 3 , HC1O 4 , CH 2 2 , C 2 H 4 2 , H 2 S0 4 , C 2 H 2 4 , H 3 PO 4 , C 6 H 8 O 7 ) on the specific rotation of chlorapocinchonine, see paper of Oudemans. 4 AcetylChlorapocinchonine, C 19 H 22 (C 2 H S O)C1N 2 O. Amorphous varnish. Alcohol (97 vol. p. c.) c 2, [a]g = + 108.0 3 mol. HC1 -f- water c = 2, = f 118.8 Oudemans : Loc. cit. Hesse : Ann. Chem. (Liebig), 305, 338. Hesse : Ibid., p. 349. Oudemans: Rec. trav. chim. Pays-Bas, i, 182. Hesse : Ann. Chem. (Liebig), 205, 354. 686 CONSTANTS OF ROTATION OF ACTIVE BODIES DIAPOCINCHONINE, C M H 44 N 4 O 2 . Amorphous. Alcohol (97 vol. p. c. ) . . . . c = 2, [] g = + 20.0 > ' 3mol. HC1 + water ..... c = 2, = + 23.6 J Jungfleisch and Leger 2 look upon this body as a mixture of several bases, Hesse, 5 however, not. Diacetyl&pocinchonine, C^H^ ( C 2 H 8 O ) 2 N 4 O 2 . Yellow varnish- like mass. Inactive in 2 per cent, alcoholic solution. 3 mol. HC1 -f water ............ c = 2, []}f = -f 26.1 4 ISOAPOCINCHONINE, C 19 H 22 N 2 O. Anhydrous prisms ; melt- ing-point, 223 to 224. Absolute alcohol ....... c = 3, [crj^s = -f 186.2 5 APOISOCINCHONINE, C 19 H, 2 N. 2 O. Anhydrous white needles ; melting-point, 216. Absolute alcohol ....... r = 3, []$= + 166.8 6 Chlorapoisocinchonine , C 19 H 23 C1N 2 O. Anhydrous white needles ; melting-point, 203. Absolute alcohol ...... c = 3, [or]$ = + 189.8 Dihydrochloride, C^H^ClNjO^HCl. Anhydrous crystals. Water ............ c = 3, [ar]^5 = + 172.5 7 HOMOCINCHONINE, C 19 H 2 ,N 2 O. Prisms ; melting-point, 251 . 2 vol. chloroform + i vol. absolute alcohol. . . c = 3, []^f = -f 208.9 Hydrochloride, C 19 H 22 N 2 O.HC1 + 2H 2 O. Needles. Alcohol (97 vol. p. c.) ........... ^ = 3, []}? = + 159-7 Dihydrochlori.de, C 19 H 22 N,O.2HC1. Prisms. Water ......... =2.528, [a]'J = -f 198.5 PSEUDOCINCHONINE, C^H.^NjO. Anhydrous white needles > melting-point, 252. 2 vol. chloroform -(- i vol. absolute alcohol. .. c = 3, ["])f = -|- 198.4 1 Hesse: Ann. Chem. (Uebig), 305, 333. * Compt. rend., 114, 1192. Ann. Chem. (I^iebig), 376, 118. Hesse : Ibid., 203, 339- Hesse: Ibid., 376, 117 Hesse : Ibid., 376, 100. Hesse : /*/']}= = -f 1 15.5 - 5 2 mol. H 2 SO 4 -f water c = 2, = + 175.5 J CINCHOTENICINE, C 18 H 20 N 2 O 3 . Dark brown amorphous mass; melting-point, 153 (uncorr.). Water c = 2.614, M^ = + 0.9 6 CINCHOTENIDINE, C^H^N.O., + 3H 2 O. Monoclinic prisms ; melting-point, 256 (corr.). Water p 0.212 []/>= - 189 7 3 mol. HC1 -f water. . . c = 5 (anhydrous), [tt]g = - 201.4 8 CINCHONICINE, C W H 22 N 2 O. Tough yellowish mass. Alcohol (95 vol. p. c. ) c = i, [<*] Chloroform c = 2, 1 Hesse : Ann. Chem. (I,iebig), 276, 107 and 108. - Jungfleisch, l,ger : Compt. rend., 108, 952. 3 Jungfleisch, I,ger : Compt. rend., 105, 1257 ; Bull. soc. chim., 49, 747. 4 Jungfleisch, I,ger : Compt. rend., 119, 1264. 9 Hesse : Ann. Chem. (Liebig), 176, 233. 6 Hesse : Ber. d. chem. Ges., 11, 1983. " Skraup, Vortmann : Ann. Chem. (Liebig), 197, 240. 8 Heese : Ber. d. chem. Ges., 14, 1893 (note). Hesse : Ann. Chem. (Liebig), 178, 262. 688 CONSTANTS OF ROTATION OF ACTIVE BODIES Roques obtained it in crystalline condition, and found : Alcohol [a]/> = + 48.25 P 2 mol. HC1 + water " s = + 28.72 ) Oxalate, (C 19 H 22 N,O) 2 C 2 H 2 O 4 + 3(?)H 2 O. Crystals. Alcohol (97 vol. p. c) c 2, [rtr]^s = + 23.5 -j a Chloroform-alcohol mixture c = i to 3, " + 23.1 Water c - 2, " = = + 22.6 2 mol. H 2 SO 4 + water c 2, " = = + 25.75 CINCHONIDINE, C 19 H M N 2 O. Crystals ; melting-point, 207.2 (corr.) ; 3 200 to 201 ;* 210.5." Alcohol (97 vol. p. c.) c i to 5, []^f 107.48 + 0.297 c (95 " " ) r = 2, " = 113.53 + 0-426 c (80 " " ) r =2, - 119.5 Chloroform-alcohol mixture <: 2, " = -108.9 Chloroform c = 2, " = 83.9 Absolute alcohol c -- 1.5 2 2.5 3 3.5 4 / = 15 -110.0 109.6 109.2 108.8 108.4 108.0 20 - 109.0 108.6 108.2 107.8 107.4 107.0 Alcohol (wt. p. c.) loo.o 90.5 80.2 70.8 60.0 For r 1.54, [a] 1 ^ -109.6 115.0 117.8 120.4 121. i Chloroform c= 1.545, [<*]$ = -77-3 <: = 3.4i, - 74.0 Chloroform-alcohol mixture... p = i.i to 2.1, [a]^- 2 - 18 -* = 107.9 8 Chloroform c 4, [ a ]/? " 70.0 ) 9 3 mol. HC1 + water = 5, -174.6] Hydrochloride, C 19 H 22 N 2 O.HC1 + H 2 O. Triclinic crystals. Water c i to 3, (>] = - 105.34 + 0.76 c alkaloid, - 123.98 + 1.05 c Water + 2 mol. HC1-- c i to 10, " = 154.07 + 1.39 c alkaloid, " = 181.32 + 1.925 c Alcohol (97 vol. p. c. ) c = 3, [<*]}5 = - 108.0 (80 " " ) .... c= 2, - 135.25 Chloroform c = 2.85 (anhydrous), " = 24.2 2 mol. HC1 + water c = 10, []^J -- 142.1 * 1 Compt. rend., 120, 1170. * Hesse : Ann. Chem. (I^iebig), 178, 263. l^enz : Ztschr. anal. Chem.. 27, 565. Hesse: Ber. d. chem. Ges., 14, 1891. Skraup, Vortmann : Ann. Chem. (I,iebig), 197, 229. Hesse : Ibid., 176, 219. Oudemans : Ibid., i8a, 44. I^nz : Loc. cit. Hesse: Ann. Chem. (I^iebig), 205, 196. > Hesse: Ibid., 176, 220. ALKALOIDS 689 Salt, C 19 H 22 N 2 O.HC1 -f 2H 2 O. Water c = 1.54 to 1.62 (calculated as alkaloid), ") 1 [a]j7 104.6 (anhydrous) ; alkaloid, [<*]$ = - 129.2 Absolute alcohol c = 1.54 to 1.62 (calculated as alkaloid), [or]^ = 99-9 (anhydrous) ; alkaloid, [a]jj = 123.5 Alcohol (89 wt. p. c.) . . c = 1.54 to 1.62 (calculated as alkaloid), [a]}] = 119.6 (anhydrous) ; alkaloid, [a]^ == 147.7 Alcohol (80 wt. p. c.) c = 1.54 to 1.62 (calculated as alkaloid), [tt] 1 ^ = - 128.7 (anhydrous) ; alkaloid, [(*]% = ~ I 59- Neutral Sulphate, (C 19 H 22 N 2 O) 2 H 2 SO 4 -f- 6H 2 O. Glistening prisms. Water <: 1.06, [orpj = - 106.77 , alkaloid, []}f -142.31 Salt with j Mol. Water. Alcohol (80 vol. p. c.) =*, O]^s = - 144.5 2 Absolute alcohol c 1.54 to 1.62 (calculated as alkaloid), ^ L [a]j7 - 118.7 (anhydrous) ; alkaloid, [or]^ = - 157.5 Alcohol (89 wt. p. c.).- c = 1.54 to 1.62 (calculated as alkaloid), [] == - 128.7 (anhydrous) ; alkaloid, [ar]j7 - 171.8 j Alcohol (80 wt. p. c. ) . . c = 1.54 to 1.62 (calculated as alkaloid), [or]'/ 131.2 (anhydrous) ; alkaloid, []^ = - 175.1 J Acid Sulphate, C 19 H 2 ,N 2 O.H 2 SO 4 + 5H 2 O. Large prisms. Water c = 2, [a]^ = -110.5; ^i 5 alkaloid, [a]g = - 1 77-95 I Alcohol (80 vol. p. c. ) c = 2, [a]^ = - 109.0 Chloroform-alcohol mixture r = 2, - 101.0 Disulphate, C 19 H, 2 N 2 O.2H 2 SO, -f 2H 2 O. Small prisms. Water, c= i to 7, [])= 105.96 1.0267 c 0.03376 c 1 -f 0.00104^") 4 Alkaloid, 185. 77 -f- 3. 1557 c o. 18158 c~ -f 0.00981 r* f Nitrate, C 19 H 22 N 2 O.HNO :{ + H 2 O. Large prisms. Water <: 1.54 to 1.62 (calculated as alkaloid), "j 1 []^ = - 99-9 (anhydrous) ; alkaloid, [a]g = 126.3 Absolute alcohol c= 1.54 to 1.62 (calculated as alkaloid), [o:]^ = - 103.2 (anhydrous) ; alkaloid, [a] 1 ^ = 130.4 Alcohol (89 wt. p. c. ) .. r= 1.54 to 1.62 (calculated as alkaloid), [a]^7 = - 119.0 (anhydrous) ; alkaloid, [a]^ = - 150.4 Alcohol (80 wt. p. c.)... c= 1.54 to 1.62 (calculated as alkaloid), [a] 1 ^ = - 127.0 (anhydrous) ; alkaloid, [>]$ = - 160.4 } 1 Oudemans : Ann. Chem. (I^ebig), 182, 46. '-' Hesse : Ibid., 176, 221. 3 Hesse : Ibid., 176, 222. 4 Hesse : I^oc. cit. 44 r 2, [<*]^ = 38.4 ^ 5 c = 2, " 66.6 | c = 2 t " 81.3 ) 690 CONSTANTS OF ROTATION OF ACTIVE BODIES Oxalatc, (C 1!( H 2 ,N,O),C 2 H.,O 4 + 2H 2 O. Prisms. Chloroform-alcohol mixture, c = i to 3, [a]$ = 98.7 ' Oudemans 2 has made observations on the changes which take place in the specific rotation of quinine, quinidine, cin- chonine and cinchonidine in presence of variable amounts of HC1, HNO 3 , HC1O 3 , HC1O 4 , H.,SO 4 , H 3 PO 4 , HCHO,, and H 2 C,,O 4 . Also similar observations on quinamine and con- quinamine. 3 Wyrouboff 4 has published numerous observations on cinchon- idine salts, which, on crystallizing from different solvents, com- bine with varying amounts of the latter. Acetylcinchonidine, C 1 , J H.,,(C,H 3 O)N 2 O. Brittle mass ; melt- ing-point, 42. Alcohol (97 vol. p. c.) i mol. HC1 + water 3 " " + " Cinchonidine Sulphonic Acid, C 19 H 21 N 2 O.HSO 3 + H,O. Needles ; melting-point, 225. 3 mol. HC1 + water c 2 (anhydrous), [<*]^ == I 4 (i /^-CINCHONIDINE, C 19 H 22 N 2 O. Crystals ; melting-point, 206 to 207. 3 mol. HC1 + water c= 1.25, |>]^ 181.4 7 Melting-point, 244. Alcohol (d 0.7944) r 0.5, [a]= - 171.5 * y-CiNCHONiDiNE, C 19 H 22 N 2 O. Crystals; melting-point, 238. Alcohol ( -f 37. 75\ 3 Chitenine " f = 4, ^ = 23 = + 34.0 j Cinchonidine ... " c = 4, Y = 2o = + 40.2* Quinidine " =4, t = 2o = -f 39.6 6 Cinchonicine . . " <: 4, / 20 = + 35-6| 7 Quinicine " 4, t = 2o = + 35-6^ ARICINE, C 23 H, 6 N 2 O 4 . Crystals ; melting-point, 188. Ether (^=0.72) r^= 1102.5, []^^ - 94-7i 8 Alcohol (97 vol. p. c.) ... r=i, 54.1 -I No rotation in hydrochloric acid solution. CUSCONINE, C, 3 H 26 N 2 O 4 -f- 2H 2 O. Crystals; melting-point, 110. Ether 0/^0.72) r=i, [or]g = - 27.1 ] 9 r = 2, " = 26.8 I Alcohol (97 vol. p. c.) r = 2, " 54-3 j 3 mol. HC1 -f water ^0.5, " -71.8 J CONCUSCONINE, C 23 H 26 N 2 O 4 + H 2 O. Monoclinic crystals ; melts at 144, then solidifies and melts again at 206 to 208. Alcohol (97 vol. p. c.) c=2, [>]$= -f 36.8 '" r--- 2, " : = + 40.8 " Skraup : Monatsh. Chem., 10, 46. Schniderschitsch : Wien. Monatsh. Chem., 10, 60. Skraup. Schniderschitsch. Wiirstl : Wien. Monatsh. Chem., 10, 70. WML Skraup, Wiirstl : Wien. Monatsh. Chem., 10, 226. Ann. Chem. (Uebig), 185, 313. Hesse : /*/]^ = + 73 fi-Concusconinemethylsulphate, ( C 23 H 26 N 2 O 4 . CH 3 ),. SO 4 ( at 120). A 2 per cent, aqueous solution was optically inactive. 1 QUINAMINE, C 19 H 24 N,O 2 . Long prisms ; melting-point, 172. Alcohol (96 (?) vol. p. c. ) c 0.8378, O]^s = + 106.8 "- Alcohol (97 vol. p. c.) c = 2, [a~\% = + 104.5 1 3 i mol. HC1 + water =2, = -}- 116.0 ] 3 " "+ " ^=2, " =+117.2 J Chloroform c = 2, [ a l^ = + 93-5 * Absolute alcohol = 0.502, [<*]^ = + 104.6 5 " " r=i.oi6, " =4-103.9 " r =1.494, ^+102.8 =1.774, .+ 100.7 Alcohol (90 wt. p. c.) =1.648, =+101.5 Absolute ether c = 0.458, = + 121.4 14 =1.024, =+119.9 Chloroform = 0.722, =+ 94.9 =1.512, =+ 94.0 =2.235, " = + 93.3 Benzene = 0.056, =+ 99.3 " c = 1.489, " =+100.9 The influence of acids (HC1, HNO 3 , HC1O 3 , HC 2 H 3 O,, HCHO 2 , H 2 SO 4 , H 2 C 2 O 4 , H 3 PO 4 ) on the specific rotation is discussed in these papers. Hydrochloride, C 19 H 24 N 2 O 2 .HC1 -f H 2 O. Prisms. Alcohol (97 vol. p. c.) = 2, [a] = + 118.1-* 3 Water = 2, = + 100.0 } Hydrobromide, C 19 H 24 N 2 O 2 .HBr + H 2 O. Prisms. Water = 4, OPj = + 88.2 6 Hydroiodide, C 19 H 24 N 2 O 2 .HI. Crystals. Absolute alcohol = 1.068, [a]^ = + 92.5 ^ 7 " =1.644, = + 94-4 " =2.310, " = + 95-8 ) 1 Hesse : Ann. Chem. (I^iebig), 325, 241. 2 Hesse : Ibid., 166, 2^2. 3 Hesse : Ibid., 307, 307. Hesse : Ibid., 199, 337. '> Oudemans : Ibid., 197, 54 ; Rec. trav. chini. Pays-Bas, I, 22 1024. G Hesse : Loc. cit. 7 Oudemans : Ann. Chem. (I^iebig), 197, 60. 694 CONSTANTS OF ROTATION OF ACTIVE BODIES Nitrate, C 19 H 24 N 2 O 2 .HNO 3 . Monoclinic crystals. Water c = 0.997, []* = + 96.8 ^ l " =1.934, =4- 97.0 Absolute alcohol c = 0.9945, ' ' - = + 109. 2 " c= 2.036, " == + 109.6 Perchlorate, C 19 H 24 N 2 O 2 .HC1O 4 . Anhydrous crystals. Absolute alcohol c = 0.709, []$=+ 99-3 ^ 2 " =2.1335, " =+ 101.8 f QUINAMIDINE, C 19 H 24 N 2 O 2 . Wart-like bunches ; melting- point, 93 (not corr.). Alcohol (97 vol. p. c.) c=2, [a] J = + 4.5 * Hydrochloride, C 19 H 24 N 2 O 2 .HC1 + H 2 O. Prisms. Inactive in 2 per cent, aqueous solution. 4 QUINAMICINE, C 19 H 24 N 2 O 2 . Crystals ; melting-point, 109 (uncorr.). Alcohol (97 vol. p. c. ) c = 2, [a] 'J = + 38. i ) * 3 mol. HC1 + water c = 2, = + 47.0 / APOQUINAMINE, C 19 H 22 N 2 O(at 100). Crystals; melting- point, 114 (uncorr.). Inactive in 2 per cent, alcoholic solution. i.i mol. HC1 + water c = 2, [apj = 28.4 ^* 3 " " + " 10 " " + " .. c = 2, [a]'j = - 28.4 ^ ^ = 2, -29-1 C=2, " = 30.0 J Acetylapoquinamine, C 19 H 21 (C 2 H 3 O)N :! O (at 100). Amor- phous. Alcohol (97 vol. p. c.) ^=2, [a]jf = o 10 mol. HC1 + water c ^= 2, " = 31.2 HYDROCINCHONIDINE, C, 9 H 24 N 2 O, according to Forst and Boehringer, 5 is identical with Hesse's cinchamidine. Crystals ; melting-point, 229 ; 6 230. ' Alcohol (97 vol. p. c.) c=2, Mjf = 98.4 8 Oudemans : Loc. cit., p. 58. Oudemans : Loc. cit., p. 59. Hesse: Ann. Chem. (lyiebig), 207, 307. Hesse : Loc. cit. Her. d. chem. Ges., 15, 520. Forst, Boehringer. Hesse. Hesse : Ber. d. chem. Ges., 14, 1683. ALKALOIDS 695 The specific rotation is greater in acid solution. On fusing the acid sulphate, an amorphous modification of hydrocinchonidine is formed, which melts below 100. 3 mol. HC1 + water c 2, [or]^ - 12 ' Hydrochloride, C 19 H 24 N 2 O.HC1 -f- 2H 2 O. Prisms. Water =2, [a]** = - 80.4 ] 2 " * = 5, " = 66.0 I " f = 8, - 60.4 2 mol. HC1 + water = 5, -109.4 Alcohol (97 vol. p. c.) = 5. - 7 2 -4 J Neutral Sulphate, (C 19 H 24 N 2 O) 2 .H 2 SO 4 + 7H 2 O. Needles. Water c = 2, [a]^ = - 75.2 - 3 Alcohol (97 vol. p. c.).. =2, 93-8 j Add Sulphate, C 19 H 24 N,,O.H 2 SO 4 -f- 4H 2 O. Prisms. Water c = 4, . [] - - 92-7 3 Acetylhydrodnchonidine, C 19 H 23 N 2 O.C 2 H 3 O. Amorphous; melting-point, 42. Alcohol (97 vol. p. c. ) = 2, [a]^ = - 29.5 ~) 3 3 mol. HC1 water r = 2, []-- 50.9 j CONQUINAMINE, C 19 H, 4 N 2 O,. Triclinic crystals; melting- point, 123 (corr. ). Alcohol (97 vol. p. c. ) c = 2, [or]]* = + 204.6 ] * Chloroform c=i, M-5 J I mol. HC1 water =2, = 229.1 3 " " + " ........ =2 , = + 230.0 | 3 " " + " c = 4, -- + 230.0 J Absolute alcohol c= 0.8025, [a] = + 205.1 ^ * " " = 0.8195, -- 204.2 c= 1.531, 203.5 " " = 2.7II5, =+202.6 =3.154, " = + 203.0 " " = 4.0l8, = + 204.1 =4.986, "=- + 203.5 Alcohol (91 wt. p. c. ) C= 1.7595, r + 2 4.3 (80 Wt. p. C.) = I.8I3, = + 205.5 Absolute ether = 0.7655, = + 192.7 " =1.1515, " 4 190.6 " =1.522, " ==-fl88.1 1 Hesse: Ann. Chem. (Liebig), 314, i. - Hesse : Loc. cit 3 Hesse. * Hesse : Ann. Chera. (I,iebig), 209, 68. 5 Oudemans : Ibid., aop, 46 ; Rec. trav. chim. Pays- Bag, i, 23 to 25. 696 CONSTANTS OF ROTATION OF ACTIVE BODIES Absolute ether c 1.6155, []g = -f 189.0 " = 3-052, =+190.7 " c 3-0585, "==+190.5 " c= 4.6465, " -+190.3 Chloroform = 0.7945, " ==+176.1 ^=1.531, "=+173-8 , " = 3. 5 , = + 171.2 Benzene c 0.8955, " = =+ 180.1 " f= 1.540, " ==+179.1 " c = 2.1285, " --= + 178.6 " c = 3.028, "= + 178.2 " = 3-477, "== + 178.0 The influence of different acids on the specific rotation of conquinamine is discussed in the same papers. Hydrochloride, C 19 H 24 N 2 O 2 .HC1. Octahedral crystals. Water c = 4, []'j = + 205.3) 2 Alcohol (97 vol. p. c.) = 4, " == + 206.4 ) Hydrobromide, C 19 H 24 N 2 O 2 .HBr. Monoclinic crystals. Absol. alcohol, c = 1.162, [a]g = + 182.7; alkaloid, [a]g= + 230.0 " " =1.9935, " = + 181.0; " =- + 228.1 Hydroiodide, C 12 H 24 N 2 O 2 -HI. Crystals. Absol. alcohol, t = i.oii, [<*]g = + 162.8; alkaloid, [or]g= + 229.6 ^=2.213, " = + 162.2; " = + 229.5 Nitrate, C 19 H 2 ,N 2 O 2 .HNO 3 . Rhombic crystals. Absol. alcohol, c = 1.2685, [] = + 190.0; alkaloid, []=+ 228.6 Chlorate, C 19 H a4 N 2 O 2 .HClO :r Monoclinic needles. Absol. alcohol, c 0.915, [a]g = + 184.0; alkaloid, [] f j = + 234.0 Perchlorate, C 19 H 84 N 2 O 2 .HC1O 4 . Monoclinic needles. Absol. alcohol, c 0.710, [or] = + 175.4; alkaloid, [<*]$--+ 231.8 =1.4755, =+i75.o; " " = + 231.4 Formate, C 19 H 24 N 2 O r CH 2 O 2 Monoclinic crystals. Absol. alcohol, c = 0.884, [] = + 195.8; alkaloid, [or] = + 224.7 r - 1.785, " ==+193.0; " " = + 222.6 Acetate, C 19 H 24 N 2 O 2 .C 2 H 4 O 2 . Tetragonal crystals. Absol. alcohol, c = 0.921, [or]g = + 181.0; alkaloid, [a]g = + 215.8 " = 0.8395, =+179.0; " - + 213.5 1 Oudemans: Ann. Chem. (Uebig), aop, 46; Rec. trav. chim. Pays-Bas, i, 231025. 8 Hesse : Ann. Chem. (Uebig), 209, 68. ALKALOIDS 697 Oxalate, C 19 H 24 N 2 O 2 .C 2 H 2 O 4 + 3H,O. Rhombic crystals; melting-point, about 115. Absol. alcohol, c = 1.0315, [a] = -f 163.0; alkaloid, O]'J = + 200.6 " " c=i. 525, " =+162.6; " " = + 200.6 HYDROQUININE, C 20 H 26 N 2 O 2 -f 2H 2 O. Crystals ; melting- point, 172.3 (corr.). Chloroform-alcohol mixture =2.49, [a]$- 8 160.25 ' Melting-point, 168. Alcohol (95 vol. p. c. ) c= 2.4, [<*] = 142.2 In loo cc., 40 cc. of normal HC1 : r=2. 4 , O]- = 227.1* Neutral Sulphate, (C 20 H 26 N 2 O 2 ) 2 H 2 SO 4 -f- 6 or 8H 2 O. Color- less needles. 4 mol. HC1 + water c = 4 (anhydrous), [a]$ = 222.5 3 4 " " + " c = 4 " -I93.4 ) 4 alkaloid, - 255.9 j ^/y Hydroquinine, C 20 H 25 N 2 O 2 (C 2 H 3 O) . Varnish-like mass ; melting-point, about 40. 3 mol. HC1 -f water c = 3, []g = - 73-9 5 HYDROQUINICINE, C 20 H 26 N 2 O 2 . Yellowish varnish. 3 mol. HC1 + water c = 3, [] = - 17 6 DICINCHONINE, C 19 H 22 N 2 O. 7 Amorphous ; melting-point, 40. Alcohol (97 vol. p. c.) c =1.516, [<*]$ = + 91. 7 3 mol. HC1 -j- water c = I, = + 80.4 Hydrochloride, C 19 H 22 N 2 O.HC1. Prisms. Water c = 5, M = + 58-7 PARICINE, C 16 H 18 N 2 O + iV a H 2 O. Yellow powder ; melting- point, 130. The alcoholic solution is inactive. 9 GEISSOSPERMINE, C 19 H M N 2 O 2 -h H 2 O. Small prisms ; melt- ing-point, 1 60, with decomposition. Alcohol (97 vol. p. c.) c= 1.5 (anhydrous), [a]^ = 93.4 10 I,enz : Ztschr. anal. Chem., 27, 561. Hesse : Ann. Chem. (I^iebig), 341, 259. Hesse : Ber. d. chem. Ges., 15, 856. Hesse: Ann. Chem. (Liebig), 241, 262. Hesse: Loc. cit., p. 278. Hesse : Loc. cit., p. 274 Hesse : Ann. Chem. (I^iebig), 276, 119. Hesse : Ibid., a7, 153. Hesse : Ibid., 166, 263. 11 Hesse: Ber. d. chem. Ges., 10, 2164. 698 CONSTANTS OF ROTATION OF ACTIVE BODIES CINCHONAMINE, C 19 H 24 N 2 O. Glistening needles ; melting- point, 184 to 185. Alcohol (97 vol. p. c.) c=2, [a]^s = -f- 121.1 l According to Arnaud, 2 [<*]/?== -f- H7-9 i n alcoholic solu- tion (93 per cent.). Melting-point, 195. Neutral Sulphate, (C 19 H 24 N 2 O) 2 .H 2 SO 4 . Prisms. Water c = 2, [or]'j = -f- 36.7 " *=6, " = + 39-8 1 mol. H 2 SO 4 + water. . . c = 2, " = = -f 39.6 2 " " + " ... c = 2, " = = + 35.7* i " + "... t = 3, = + 43-S 5 Acid Sulphate, C 19 H 24 N 2 O.H 2 SO 4 . Prisms. Water f = 2, [a]g = \ 34-9) 6 ^ = 6, - + 37-4 f CHAIRAMINE, C 22 H 26 N 2 O 4 -f- H 2 O. Crystals ; melting-point, 233. [a~\ D = about -f 100 ' CONCHAIRAMINE, C 22 H 26 N 2 O 4 + H 2 O + C,H 6 O (from alco- hol). Prisms; melting-point, 82 to 86 ; 120 (anhydrous). Alcohol (97 vol. p. c.), c= 2, (alcohol and water-free), [<*]$ = + 68.4 8 CHAIRAMIDINE, C 22 H 26 N 2 O 4 + H,O. Amorphous ; melting- point, 126 to 128. Alcohol (97 vol. p. c.) c = 3 (anhydrous), [tt]js = -f 7.3 9 CONCHAIRAMIDINE, C 22 H 26 N 2 O 4 + H 2 O. Crystals ; melting- point, 114 to 115. Alcohol (97 vol. p. c. ) c = 3 (anhydrous), C^]^ ~ 60 Alkaloids of Coca Leaves fl?-EcGONiNE, C 9 H 15 NO 3 . Crystals ; melting-point, 254. The constitutional, formula of Einhorn and Tahara 11 contains 1 Hesse : Ann. Chem. (L,iebig), 335, 220. 2 Compt. rend., 93, 593. Hesse : Loc. cit., p. 224. Hesse : Ber. d. chem. Ges., 16, 62. Arnaud : Compt. rend., 97, 174. Hesse : Loc. cit. Hesse. Hesse : Loc. cit., p. 248. Hesse : Loc. cit., p. 254. Hesse : Loc. cit., p. 256. 11 Ber. d. chem. Ges., a6, 324. ALKALOIDS 699 three asymmetric carbon atoms (compare Einhorn 1 ), while that of Merling 2 has four such atoms. Hydrochloride, C 9 H 15 NO 3 .HC1. Monoclinic hemimorphous crystals. Water p = 4.4, [or]^ = - 18.2 :i Methyl Ester, C 9 H 14 NO 3 .CH 3 . Prismatic crystals; melting- point, 115. Dilute alcohol p = 6.22, [a]^ = -f 22.5 ) * " / = 6.25, " = 4-22.4 J Isovaleryl-d-Ecgonine Methyl Ester Hydrochloride, C 5 H 9 O C 9 H I3 NO 3 .CH S .HC1. Thin leaves; melting-point, 192. Alcohol =2.01, [a]- D = + 25.4 5 Cinnamyl-d-Ecgonine Methyl Ester Hydrochloride, C 9 H 7 O C 9 H 13 NO 3 .CH 3 .HC1. Needles ; melting-point, 186 to 188. Alcohol c 2. 1 1, [a] 9 ^ = + 47.4 6 Deriva fives of Benzoyl- d- Ecgon in e Methyl Ester (^-Cocaine). Constitution : Einhorn, 7 two asymmetric C atoms ; Einhorn, Tahara, s three asymmetric C atoms, C 5 H 7 N.CH 3 .CH 3 .CHO(CO.C 6 H 5 )CH 2 .COOCH 3 . Hydrochloride, C n H 21 NO 4 HCl. Monoclinic plates ; melting- point, 205. Alcohol (d = 0.9353) ^=1.9, O? = + 39-47 9 Ethyl Ester Hydrochloride , C 18 H 23 NO 4 .HC1 -f H 2 O. Plates ; melting-point, 215. Water c 2, [a] , = + 40 Propyl Ester Hydrochloride, C 19 H 25 NO 4 .HC1 + H 2 O. Prisms ; melting-point, 220. Water c= 2.6, []= + 46.2 Ber. d. chem. Ges., aa, 1495. Ibid., 24, 3116. Einhorn, Marquardt, Koch : Ibid., 23, 468. L,iebermann, Giesel : Ibid., 23, 926. Deckers, Einhorn : Ibid., 24, 7. Deckers, Einhorn. Ber. d. chem. Ges., 21, 3029. Ibid., 26, 324. Einhorn, Marquardt : Ibid., 33, 468. yoo CONSTANTS OF ROTATION OF ACTIVE BODIES Butyl Ester Hydrochloride, C^H^NO^HCl -f H,O. Fine matted needles ; melting-point, 201. Water c = 2.5, [a] -,= + 46 Amyl Ester Hydrochloride, C 21 H W NO 4 .HC1. Fine matted needles; melting-point, 217. Water C= 2.2, [a]], = -f 38.6 ! /-EcooNiNE HYDROCHLORIDE, C 9 H 15 NO 3 .HC1. Triclinic plates ; melting-point, 246. Water [a]/, - - 57 2 Cinnamyl-l-ecgonine Methyl Ester, C 9 H.O.C 9 H 13 NO 3 .CH,. Crystals ; melting-point, 121. Chloroform c 10, [a]g = - 4.7 :! Hydrochloride, C 19 H 23 NO 4 .HC1 + 2H 2 O. Crystals ; melting- point, 176. Water r = 66 (anhydrous), [apj = - 104.1 * Benzoyl-l-ecgonine Methyl Ester (/-Cocaine ordinary cocaine) , C 5 H 7 N. CH 3 . CHO ( CO. C 6 H 5 ) CH 2 . COOCH S . Mono- clinic prisms ; melting-point, 98. Chloroform p = 10 to 20, [tf]^ = - 16.412 + 0.00585 />, from the following observations : MS; p. rf- Found. Calculated. 9-925 1.4480 - 16.356 -16.354 25.484 1.3971 16.280 -16.263 15.643 1.4293 -16.319 - 16.320 18.793 1.4190 -16.299 16.302 20.242 1.4126 16.283 -16.293 J The specific rotation is, therefore, almost constant. Hydrochloride, C 17 H 21 NO 4 .HC1. Crystals; melting-point, 181.5.* Five preparations from different factories, when dis- Kinhorn, Marquardt : Ber. d. chem. Ges., a3, 986-988. Einhorn : Ibid., aa, 1495. Hesse : Ann. Chem. (lyiebig), 371, 185. Hesse : I = - 61.5 8 Melting-point, 238 to 240. Water, c = 3 (anhydrous) . . [<*]^f = 62.7 4 /-EcGONiNic ACID, C.H n NO 3 . From d- and /-ecgonine. Crystals; melting-point, 117. Water c = 12.37, [or]], = 43.2 5 ^-TROPINIC ACID, C 6 H H N(COOH) 2 . Crystals; melting- point, 247 to 248. Water p = 11.76, = -100.4]* Alcohol (d = 0.865) ....... ^ = 0-97, " - 98.9 Water .................... ^=2.5, " 77 " .................... r = 0.996, " = 72 The effect of acids (HC1, HNO 3 , HCHO,, HC 2 H 3 O,, H 2 SO H,C,O 4 , H 3 PO 4 , H,AsO 4 , C 6 H 8 O.) on the rotation of morphine has been investigated by Tykociner. 5 CODEINE (Morphine methyl ether), C 17 H 17 (OCH 3 )(OH)NO -f- H 2 O. Rhombic crystals ; melting-point, 155 ; (1 153. ~ Alcohol (97 vol. p. c. ) c = 2 to 8, [a]^f - 135.8 ~\ 8 " (80 " " ) C=2 t - 137.8 | Chloroform ........... ^=2, " = 111.5 J Absolute alcohol ............ c = 2.32 (anhydrous), [a~\ = 134.3) * " ............ ^=2.97 " = -141.1 j Alcohol .................... 1=4.1, []^ = 103.13 5 LAUDANINE, C 20 H., 4 NO 4 . Rhombic crystals ; melting-point, 166. Chloroform c -. 2, []^' 5 = - i3-5\ 6 2 mol. Na-jO -j- water ^ = i, 11.4 j Chloroform ^ = 3-35 [ a ]/> H~ -57 This number lies within the limits of experimental error, and Goldschmidt, 7 therefore, considers laudanine inactive. Hydrochloride, C W H 25 NO 4 .HC1 + 6H 2 O. Inactive/ LAUDANIDINE, C 17 H 15 N(OH)(OCH 3 ) 3 . Melting-point, 177. [a] D = 87.8 9 ' LAUDANOSINE, C 21 H 27 NO 4 . Crystals ; melting-point, 89. Alcohol (97 vol. p. c. ) c = 2.79, [tfJ'J : = + 103.2 " (97 (< " ) c 2, [<*]%* -= + 105.0 Chloroform c 2, " = + 56.0 2 mol. HC1 + water c i, ^ + 108.4 J Hesse : Ann. Chem. (l,iebig), 176, 196. Hesse : Loc. '/., p. 197. Hesse: Ann. Chem. (I^iebig), 335, 229. Hesse : Ibid., 176, 195. Donath : J. prakt. Chem., [a], 33, 562. Hesse : Ann. Chem. (I,iebig), 176, 201. Wiener Monatsh. Chem., 13, 693. Hesse : Loc. i it. Hesse : Ann. Chem. (Uebig), a8a, 208. " Hesse : Ibid., 176, 202. ALKALOIDS 705 PAPAVERINE, C 20 H, M NO 4 . Triclinic prisms; melting-point, 147. The earlier figures of Hesse 1 Alcohol (97 vol. p. c. ) i- :-.: 2, [or] '* 4.0 Chloroform c = 2. " = 5.7 are wrong, according to Goldschmidt ;-' the papaverine formula contains no asymmetric carbon atoms, in consequence of which the alkaloid must be inactive, as Hesse gives for the hydrochlo- ride. Goldschmidt found : Chloroform c = 17.8, [or]# = -f- 0.11 Tetrahydropapaverine, C 20 H.,.NO 4 . The dextro and levo bases have been obtained by Pope and Peachey 3 by resolution of the racemic compound by dextro- o'-bromocamphorsulphonic acid. /-Base r = 4.0032, [<*]/>= -143.4 rf-Base ^=4-337 6 > [] = + 153-7 Because of difficulties in the purification of the separated bases, the rotations are probably not quite accurate. CRYPTOPINE, C 21 H 23 NO 5 . Crystals; melting-point, 217. The alkaloid is inactive, dissolved in chloroform or hydrochlo- ric acid. 4 NARCOTINE (Opianine), C,,H., 3 NO 7 . Crystals; melting-point, 176. Alcohol (97 vol. p. c.) c = 0.74, [a]-s : - 185.0 ^ 5 Chloroform mixture c = 2, 191.5 Chloroform c = 2 to 5, " -207.35 2 mol. HC1 -j- vrater c = 2, = + 47- 2 " " + " ^ = 5, =+ 46.4 10 " " -f " C = 2, = -f 50.0 Alcohol (80 vol. p. c.) + 2 mol. HC1-. c = 2, = + 145.5 Benzene, p = 1.59, []/> = - 229 ; dilute oxalic acid, [or]/? =-f62 06 PSEUDONARCEINE, C 23 H 27 NO 8 .+3H 2 O. Crystals ; melting- point, about 175. Is inactive in acetic acid solution. 7 Narceine, inactive in neutral or acid solution. 8 Ann. Chem. (Liebig), 176, 198. Wiener Monatsh. Chem., 13, 691. J. Chem. Soc., 73, 893. Hesse : Ann. Chem. (L,iebig), 176, 200. Hesse : Ibid., 176, 192. Dott : Ber. d. chem. Ges., 17, Ref. 77. ' Roser : Ann. Chem. (I,iebig), 347, 169. * Hesse : Ibid., 176, 198. 45 706 CONSTANTS OF ROTATION OF ACTIVE BODIES Strychnos Alkaloids STRYCHNINE, C 21 H 22 N 2 O._,. Rhombic columns ; melting- point, 284. Alcohol (d = 0.865) ..... = 0.91, [CK]D= -128 I 1 Chloroform ............. c = 4, - 130 ............. ^=2.25, " =137.7' ............. c= 1.5, 140.7 Amyl alcohol ............ c = 0.53 " = 235 Alcohol (d = 0.8543).... r = 0.25, []*> = - 114.7) 2 " ( H 2 SO 4 , H 2 CrO t , H 3 PO 4 , H 3 AsO 4 , H 3 C 6 H 5 O 7 ) on the specific rotation of strych- nine was investigated by Tykociner. 4 Desoxystrychnine Hydrochloride, C 21 H 26 N 2 O.HC1 (at 100). Crystals. Water ............ c = 10, \_a] D = - 16.0 5 BRUCINE, C 23 H 26 N 2 O 4 -f 4H 2 O. Monoclinic crystals; melt- ing-point, 178. Alcohol (d = 0.865) ........ c = 5.4, (anhydrous), [<*]/>= - 85 i 6 Chloroform ................ =1.9, " " = 127 j- ................ ^ = 4-9, " - H9 J Absolute alcohol ............ =2.129, " M/? -= -8o.i 07 Effect of acids, see under strychnine/ Oxyethylbrudne Chloride, C 23 H 26 NO 4 . N<' CHa< OH Colorless, columnar crystals;; melting-point, 185. Water ............ = 4-5, M Oudemans : Ann. Chera. (lyiebig), 166, 76. Tykociner: Rec. trav. chim. Pays-Bas, i, 146. Compt. rend., 115, 832. Loc. cil. Tafel : Ann. Chem. (I^iebig), a68, 245. Oudemans : Ibid., 166, 69. Tykociner: Rec. trav. chim. Pays-Has, i, 148. Tykociner : Loc. cit. MeulenhofT: Centrbl., 1893, II, 761. ALKALOIDS 707 Other Alkaloids and Bases CORYDALINE, C 22 H 2 .NO 4 . From Corydalis cava. Crystals ; melting-point, 134. Chloroform c = 6.55, [a]^ + 300.1 J BULBOCAPNINE, C 19 H 19 NO 4 . From Corydalis cava. Mono- clinic columns ; melting-point, 199. Chloroform c 4.48, [a] = -f- 237. i - CYTISINE (Ulexine, Sophorine), C U H U N 2 O. Colorless crystals; melting-point, 150 to 151.5 (uncorr.). Water c = 2, [a]% = -120 i 3 Alcohol (90 vol. p. c.) c=z, 100.42 | Chloroform c 2, - 65 .42 J Nitrate, C U H 16 N 2 O. NO 3 H + H 2 O. Crystals. Water c = 5, M = - 90.1?) * " c=2.5 -89.33 i DELPHININE, C 22 H 35 NO 6 (at 100). Rhombic crystals. DELPHINOIDINE, C^H^NjO,. Crystals ; melting-point, 110 tO 120. STAPHISAGRINE, C 22 H 33 NO 5 . Amorphous. The last three alkaloids are inactive in alcoholic solution. 5 ECHITAMINE, C 22 H 28 N 2 O 4 + 4H 2 O. Crystals. Alcohol (97 vol. p. c.) - c 2, [ar]^f = -288 According to Hesse, 6 it is identical with Harnack's ditaine. 7 IMPERI ALINE, C^H^NO^?). Needles ; melting-point, 254. Chloroform p = 5.262, [<*]/>= 35-4 8 PILOCARPINE, C 23 H 34 N 4 O 4 -f 4H 2 O. Tough colorless mass. (Solvent?) c = 7.24, []/> = -f 101.6 ; r = 25.89, [_a] D + 87.77 9 A review of recent determinations on pilocarpine is given by Jowett, 10 and the optical rotation of the base and salts has been measured. The formula of the base is taken as C U H 16 O 2 N 2 . Freund, Josephi : Ann. Chem. (I^iebig), 277, 7. Freund, Josephi : Ibid., 277, 12. van de Moer : Arch. d. Pharra., 229, 57. van de Moer : Loc. cit. Marquis : Jahresbericht, 1877, p. 896. Ann. Chem. (I,iebig), 203, 144- Ber. d. chem. Ges., u, 2004. 8 Fragner : Ibid., 21, 3284. Poehl : Jahresbericht, 1880, p. 993, 1075. 10 J. Chem. Soc., 77, 473- 708 CONSTANTS OF ROTATION OF ACTIVE BODIES For the salts, the solvent used appears to be alcohol or water. Nitrate, C n H 16 O 2 N 2 .HNO 3 , =9.572, \ci\ n = + 82.90 Hydrochloride, c 9.924, = + 91.74 Hydrobromide, c = 10.058, = + 77.05 Sulphate, c = 7.318, ^ + 84.72 ISOPILOCARPINE, C H H 1$ N 2 O 2 . Base =11.652, [<*]/> = + 42.91 1 J Nitrate = 6.586, " = + 35.68 | Hydrochloride c= 4.974, [or] = + 38.8 Hydrobromide c= 2.288, " - + 32.8 J PlLOCARPIDINE, C 10 H 14 O 2 N 2 . Base ................ ^=1-5374, Nitrate ............. = 1-5374, []/> = + 81.3 "I 2 7.104, = + 73-2 \ PiPERiNE. Inactive. SPARTEINE. C 15 H 26 N 2 . Thick oil, colorless and odorless, Boiling-point, 180 to 181 (20 mm.). Alcohol (96 p. (:.) ..... r=2 3 .88, [a]g = - 14.6 3 ^--PiPECOLiNE (or-Methylpiperidine), C.H 10 N.CH 3 . By resolution of the racemic synthetic a-pipecoline by means of rf-tartaric acid. Boiling-point, 118 to ngV do = 0.860, [a\ D = + 21.73 5 Purest preparation : []/> = -f 37- 2 9 \ 3 6 -9 ', 37-2 6 ( = -f 13-79 8 d i= 0.845, b. p. 166 to 167, \_a] D is + 15.6 9 ^ = 0.8438, b. p. 165.710 165.9 (759 mm-). O]# - + 15-7 1(> Jowett : Loc. cit. Solvent not given. Jowett : Loc. cit. Solvent not clearly given. Bernheimet : (ia/.z. chini. ital., 13, 451. I^uletiliurg : Ann. Chem. (lyiebig), 347, 65. I^adenburg : Loc. cit. (1888). Iadenburg ; Her. d. chem. Ges., 37, 856, 3063 (1894). H. Schiff: Ann. Chem. (Uebig), 166, 94 (1873). I^adenburg: Ibid., 347, 86 (1888). I,.id< nburg : Ber. d. chem. Ges., 37, 858 (1894). Wolffenstein : /hid., 37, 2612. ALKALOIDS 709 2. Artificial Conine, by resolution of or-A^-propylpiperidine by means of = -f 13-87 2 rf == 0.8438, b. p. 167.7, [>]/> = + 18.3 ' The following data on the specific rotation of conine and its salts are given by Zecchini.* Name. Solvent. c. /. M*. ' Conine Benzene 13.094 24.2 + 9-54 11 20.464 22.1 + 9-77 " " 33-29 23-9 + 11.14 11 Alcohol 10.841 24.4 -f 8.12 1 1 < i 15.172 22.7 + 8.70 " 44.687 26 + 9-98 < < Water 1.071 25-7 + I.2I Acetate Benzene 22.854 25-7 + 3.63 " Alcohol 21.904 25.0 -f 2 -35 .< Water 31-944 26.6 -|- 1.16 Hydrochloride Alcohol 6.722 25 + 4.56 ( 4 Water 11.458 26 + 0.27 Hvdrobromide Alcohol 6.053 23-4 -f 4.28 1 Water 11.890 25.6 -j- 0.27 Acetylconine, C 8 H lt N(C 8 H 8 O). Liquid; boiling-point, 125 (14 mm.). rf 16 = 0.9616, []* = + 34-2 9 Benzoylconine, C S H 16 N(COC 6 H 5 ), Liquid. ^ = 1.0623, [a]g= + 2 9 .i I- Conine. Liquid. Absolute alcohol ...... c = 50, [a]*> = + 15.2 7 N-Methylconine, C 8 H 16 (CH 3 )N. Boiling-point, 173 to 174' (757mm.). rf 2 *- 3 = 0.8318, t = 24, [a]z> = + 81.33 8 ISOCONINE (?), C 8 H 17 N. Ladenburg: See 33, p. 112. Ladenburg: Ann. Chem. (I^iebig), 247, 86 (1888). Ladenburg : Ber. d. chem. Ges., 37, 3066 (1894). Esperienze sul potere rotatorio della coniina e dei suvi sali. Roma, 1893. I^adenburg : Ber. d. chem. Ges., 26, 854- I^adenburg : Loc. cit. I^denburg: Ann. Chem. (Uebig), 247, 86. Wolffensteiti : Ber. d. chem. Ges.. 27, 2611. I^adenburg : Ibid., 26, 854. 710 CONSTANTS OF ROTATION OF ACTIVE BODIES According to Wolff enstein, 1 this is a mixture of ^/-conine with inactive conine. Benzoylisoconine, C 8 H ]6 N(COC 6 H 5 ). Liquid. = + 29.1 - /S-PROPYL PIPERIDINE, C 5 H 10 N.C 3 H.. The synthetic base has been resolved by means of tartaricacid by J. D. Granger. 3 d-Base, d 0.8517, |>] = -f 6.39 I -Base, d = 0.8517, - 6.39 a-IsoBUT YLPIPERIDINE (Homoconine), C 5 H 10 N.CH 2 . CH(CH 3 ) r Liquid ; boiling-point, i8ito 182 ; d\ = 0.8583. Inactive.* Homoconic Acid, C^H 17 NO 2 , and Amido Valeric Acid, C 5 H n NO 2 , formed by the oxidation of conine, are inactive in 5 per cent, solution. 5 PARACONINE, C 8 H 15 N. Liquid ; boiling-point, 168 to 170. Inactive. 6 E, C 8 H 15 N. Liquid ; boiling-point, 168. d*i = 0.8976, [a]/, = - 7.8 T -CONICEINE, C 8 H 15 N. Liquid ; boiling-point, 150 to 151. [a]/? = about -f 42 s The statement of Hesekiel 9 that /?-picoline is active is an error, according to Landolt. 10 NICOTINE, C 10 H U N.,. Boiling-point, 246.7 (745 mm.). d - - i.oiioi. - 164.0 '- O]g = 161.09 13 [a] = - 162.84, Ber. d. chem. Ges., 39, 1956. leaden burg: Loc. cit. Ber. d. chem. Ges., 30, 1060. Jacobi, Stoehr : Ibid., a6, 949. Schotten : Ibid., 19, 503, 507. Michael : Ibid., 14, 2105. I.t-llmann : Ann. Chem. (Uebig), 359, 199. Iellmaiin, p. 203. Ber. d. chem. Ges., 18, 3091. Ibid.. 19, 157. " Mndolt, 52: Ann. Chem. (Uebig), 189. 241. 11 Hein : Inaug. Diss., Berlin, 1896. 13 N.-isini and Pezzolato : Ztschr. phys. Chem., 12, 501. M Gennari : Ibid., 19, 130, 46. ALKALOIDS yil Rotation dispersion of nicotine. 1 Effect of temperature between 10 and 30 . 2 Solutions in water. 3 Solutions in water, great dilution. 4 Solutions in water, increase of rotation on standing. 3 Solutions in ethyl alcohol. 6 Solutions in propyl alcohol, ether, acetone. 7 Solutions in benzene/ NICOTINE SALTS. Right-rotating. Hydrochloride, C 10 H 14 X,.HC1. Water, [a]^ = -f- 51.50 0.7931 q 0.004238 q', for q 57 to 90, from which for p = 10 : [a] = - 14.44 ; p = 3p: [a]~ == + 16.75 9 Neutral Sulphate, (C 10 H 14 N,) 2 .H 2 SO 4 . Water, [a] = 19.77 0.05911 q, for q = 30 to 90, from which for p'= 10 : [or]- = + 14-52 ; p = 30 : [] - - 15.64 lft /4/afc, C 10 H M N 2 .C 2 H 4 = 10 : [or] = + 14.57 ; / = 30 : [or] = - 18.81 10 Water M = + 13.204 0.11406 c 0.002073 [>]/> = - i5-4<> 2 2 Water .......... p = 4-93 8 > d = ioi35, [<*]/ = - 62.56 3 Water .......... p = 2.78, \OL\J = - 73-4 * Alcohol ( 50 p. c. ) , [or] /> 50.30 0.05026 q>, t = 22 tO 26, ? = 90 tO 96 5 HELICIN, C 13 H 16 O 7 + :! / 4 H,O. Alcohol (50 p. c. ) ...... / = 20, p =3 to 9, [<*]/>= 47-O3 6 W T ater ...... p = 1.351 (anhydrous), d 1.0084, [ar]* = - 60.43 3 POPULIN, C 20 H 22 O 8 - Water ............... p = I, L<*L' = ~ 53 7 PHLORIDZIN, C 21 H 30 O U + 2H 2 O. Alcohol (97 p. c. ) .... /> = i to 5, [or]-s = - 49.40 2.41 p s Alcohol .............. /> = 4-6, [a]/> = -521 9 Wood alcohol ........ / = 3.9, 52 )* AMYGDALIN, C., H 27 NO U + 3H..O. MD= -35.5 010 APIIN, C.,-H 32 O 16 . Weak alcoholic solution. =4-173" CONIFERIN, C 16 H 22 O 8 + 2H,O. Water ...... />:= 0.621 (anhydrous), ^ = 0.9998, [a]= 66.90 3 GLUCOVANILLIN, C U H 1S O, -f- 2H 2 O. Water ........ / = 0.896 (anhydrous), rf =r i.oon, [a]^= 88.63 o :t 1 Ber. d. chem. Ges., 27, 75. 2 Hesse: Ann. Chem. (Uebig), i76, 116. 3 Wegscheider : Ber. d. chem. Ges., 18, 1600. 4 Biot, Pasteur : Compt. rend., 34, 607. 5 Sorokin : J. prakt. Chem., [2], 37, 320. 6 Sorokin : Ibid., [2], 37, 291. " Biot, Pasteur : Loc. ctt. 8 Hesse: Ann. Chem. (I^iebig), 176, 117. 9 Oudemans : Ibid., 166, 69. 10 Bouchardat : Compt. rend., 19, 1174. 11 I,indenborn, Gerichten : Ber. d. chem. Ges., 9, 1123. CONSTANTS OF ROTATION OF ACTIVE BODIES CONVALLAMARIN, C 2S H 44 O 12 . Alcohol [<*]/> = - 55 e ' DIGITONIN, C, 7 H 46 O 14 + 5H..O. Acetic acid (75 p. c.) .-/> 2.8, [a]/, 50 2 IVY LEAF GLUCOSIDE, C 3a H 54 O n . Alcohol t=22, [>]/> _ 47.5 ISOHESPERIDIN, C 22 H 2 gO I2 + 2H 2 O. Alcohol....".' []/)- -89* NARINGIN (Aurantiin, Hesperidin ) , C 21 H 26 O n . Water c = 20.46, / = T7, [a]z>= 84.5 j 5 Alcohol = 7-299, * = 17, [a]z>=: 87.6 j OUBAIN, C^H^O,, H- 7H,O- Water / 0.65, [a]g = -34 06 PICEIN, C 14 H 18 O 7 -f H 2 O. Water p = 2.5, [>]z>- -84" CONVOLVULIN, C 61 H 1W ,O 27 . Melting-point, 140 to 148. Alcohol [<*]z> 36.9 s In addition, E. Fischer has produced a series of simple glu- cosides synthetically and has determined their constants of ro- tation : OOIVCI1V. V " 20' L"J/>. rv.nst.ner.' a-MethyW-glucoside Water 5 4- 157.6 28, 1152 a- " -/- 5 156.9 28, II 5 2 a-Ethyl-rf- " 9.002 1.025 -f 150.6 28, 1154 " 8.897 f 150-3 28, 1154 a-Methyl galactoside 9.119 1.026 f 179-0 28, 1155 ft. " " 10 Borax sol. 8-5 + 2.6 28, 1155 Ethyl Water 9-47 1.0273 4- 178.75 27, 2481 Benzyl arabinoside . . ' 1.03 I.OOI3 -f 215.2 27, 2482 Methyl glucoheptoside nr-Methyl xyloside . . 10.06 9-32 1.0338 1.026 - 74-7 -f 153-2 28, 1156 28, 1158 ft. " < c 9.14 1.024 65-9 28, 1157 Methyl rhamnoside. . " 9.68 1.024 62.2 28, 1159 " sorboside " 9.12 1.028 - 88.5 28, 1160 " 8.19 1.026 88.9 28, 1160 Mandelnitrile glu- \ cosiclc i H 8.25 I.OI8 - 26.9 28, 1509 Acetone rhamnoside . . U 9.16 I.OI7 + 17-4 28, 1163 Diacetone arabinoside " 2.41 1.003 H- 5-4 28, 1164 glucoside " 4-93 I.OOS - 18.5 28, 1167 fructoside- ' 7-29 I.OI4 - 161.3 28, 1165 Triacetone mannitol . . Alcohol 9.58 o.Sir 4- 12.5 28, 1168 Tanret : Jahresbericht, 1882, p. 1130. Vernet : Jahresbericht, 1881, p. 991. Will : Her. d. chem. Ges., ao, 294. Tanret : Bull. soc. chim., [3], n, 944. Her. d. chem. Ges. - Kiliani : Her. d. chem. Ges., 34, 339. 4 Tanret : Bull. soc. chim., [2], 49, ai. r> Arnaud : Compt. rend., 107, 1162. * Kromer : Chem. Centibl., (1894), I, 635. 10 In aqueous solution inactive. BITTER PRINCIPLES AND INDIFFERENT BODIES 715 Also : df-and /-Methylmarmoside. Water ....... /> - S, [a] D = 79.2 to 79.4 d-lcrm : Water . Alcohol i, r 0^.5- ) = , " 79-2 =!, " =4-87.3 ' 28. Bitter Principles and Indifferent Bodies Santonin Group The following data, unless otherwise stated, are from Carne- lutti and Nasini. 3 According to these authors the concentra- tion is without effect on the rotation : Substance. Observer. Santonin Alcohol 97 " 90 15 11 " 80 15 " Chloroform 15 ' " 26 Metasantonin 26 Santonide 26.5 " " 20 lletasantonide 26 Parasantonide 26 " " 20 Santonic acid 26.5 Methyl ester 26.5 Ethyl ester 26.5 -Propyl ester 26.5 Allyl ester " 26.5 z-Butyl ester 27 Parasantonic acid 26 Methyl ester 26 Ethyl ester 26 Propyl ester 26 Santonyl chloride 26.5 Santonyl bromide 26 Santonyl iodide 26 2 2 2 2-IO 6 i-5 1.5 3.1-30.5 i-5 i-5 2.6-50 6 6 6 6 6 6 6 6 6 6 6 6 6 175.4 176.5 I7I.5 I7L37 292.15 744.6i 754 223.46 897-25 891.7 70.31 52.33 45-35 Hesse. < xasini.-- Nasini. 39-54 41.63 98-51 108.91 99.98 91.27 I3-I4 100.53 99.21 1 Fischer, Beensch : Ber. d. chem. Ges., 29, 2927. - Ekenstein : Rec. trav. chim. Pays- Has, 15, 221. 3 Ber. d. chem. Ges., 13, 2208. 4 Ann. Chem. (L,iebig), 176, 125. 5 R. Accad. L,incei, 13, 1892. 6 Loc. cit. 7 i6 CONSTANTS OF ROTATION OF ACTIVE BODIES Substance. Solvent. Per cent. t c []/> Observer. Alcohol ij. 2 121 6 ii 1 i O 7 4- 76 77 Dehydrophotosantonic > arid . 1 " 1 o 20 1.4 + 31-9 Villavec- chia.i <( 2 8 1 20 A. Isophotosantonic acid Monoacet ylisophoto- < i II '7 2.4 0.9 + 124.17 + 58.6 v Canniz- l zaro, Fa- \ bris.s ' ' Q7 % OC S t " 80 % 22.5 3 2 1 *o- 26 t; Hesse.^ Na salt Water -"O * O ^"D J Hyposantoninic acid Dihydrosantinic acid 22.5 J 9-3 D 4.62 f 62.07 + ft A 11 I Gucci and ! G r a s si- Santoninaminesulphate. . Santoninamine hydro- ^ chloride . . .: .... 4-37 - 103.67 - 136.83 f Cristal- di.< The following data are given by Nasmi 5 for the specific rota- tion of some bodies of the santonin group for light of different colors : B C I) E 1 Santonin. Meta- s.-uitonin. Santonide. Para- santonide. Chlorof. Santonic acid. Chlorof. c 27 192 Chloroform. cmoro- Alcohol. form - Alcohol. Chlorof. 1 q = 75 to 96.5 / = 20. sa^p^ /"t$* c = 3 to 30 t=20. / = 20. / = 20. 686.7 I40.I +0.2085? 110.4' + 92 + 442 484 + 580.5 - 49 656.2 -149.3 +0.1555? 118.8 + 104 + 504 + 549 + 655-6 - 57 589-2 202.7 +0.3086? -161.0, +124 + 693 :- 754 + 891.7 - 74 526.9 285.6 ; 0.5820? 222.6 + 167 + 991 + 1088 + 1264 - 105 5^8.3 -302.38+0.6557? -237.1 + 182 + 1053 + 1148 + 1334 112 486.1 365.55 +0.8284? -261.7 +217 + 1323 + 1444 + 1666 -137 438-3 534.98 r 1.5240? 380.0 4-257 + 201 1 + 2201 + 2510 -197 422.6 2 3 8l + 26lO + 2963 230 1 Her. d. chem. Ges., 18, 2859. Ibid., 19, 2^60. * Ann. Chem. (L,iebig), 176, 125. 4 Gazz. chim. ital., aa, i. < R. Accad. I^incei, [3], 13, 1882. BITTER PRINCIPLES AND INDIFFERENT BODIES Jl Andreocci 1 gives : ^/-Disantonious acid ............. [<*]/> = " 85.9 /- " " .............. " =-85.8 ^/-Santonious " .............. =4-74.6 /- " " .............. " = 74-3 Desmotropodisantonious acid.--. 64.5 Desmotroposantonious acid ...... ' = 53.3 Other Vegetable Substances ASEBOTOXIN (Andromedotoxin), C U H 51 O 10 . Water .................. c = 2.8, [<*] = - 9 . 7 \ 2 Chloroform ............. c = 0.41, = -f 10.1 j PlCROTtoxiN, C,,H 14 O 5 . Alcohol .............. ^ = 3-125, [<*]/ = - 28.1 3 ECHICERIN, CgoH^Oj. Ether ................ c = 2, \_a]% = + 63.75 * Chloroform .......... c 2, = + 65.75 ECHITIN, C S2 H 5 ,O,. []^ = -f 72-5l = -f 75-3 ) Ether ................ c = 2, - 5 Chloroform c = 2, ECHITEIN, C 4 .,H TO O. Ether * = 2, []g = + 88 Chloroform t = 2, " = -f 85 ECHIRETIN, CsjH^O^,. Ether c = 2, [a]g = + 54-8 5 EUPHORBON, C 15 H 24 O. Ether * = 4, [a]g = + -7 Chloroform T = 4, ' = + l8 - 8 ANTIARONIC ACID, C 6 H 10 O 5 (from Antiaris toxicaria). Water []/> = + 3 7 LUPEOL, C M H 42 O?. Ether <: = 9-97, []^ = + 2 7 * Atti. R. Accad. I*incei, [5], 4. 164. Zaayer : Rec. trav. chim. Pays- Has, 5, 313. Bouchardat and Boudet : J. pharm. chim., [3], 33, 288. Jobst and Hesse : Ann. Chem. (Liebig), 178, 49- Jobst and Hesse : Loc. cit. Hesse : Ann. Chem. (Liebig), 193, 195. : Kiliani : Arch. der. Pharm., 334, 438. Likiernik : Ztschr. physiol. Chem., 15, 415 7 i8 CONSTANTS OF ROTATION OF ACTIVE BODI1 PHASOL, C 15 H, 4 O. Chloroform . . . [a\ D = + 30.6 r [a]3 = + 76.2 MS- + 38.2 " f>]'/5 = -29.3* p = 3.156, M^ = -37-5* ClNCHOL, C ao H 34 O. Chloroform /> == 6, [a]^f = 34.6 5 29. Biliary Substances CHOLESTERIN, C, 6 H 44 O(?) (melting-point, 145) and C 26 H 44 O OT-LACTACEROL, C^H^O. Chloroform p 2.372, /?-L,ACTACEROL, C^H^O. Chloroform p = 4, QUEBRACHOL, C^H^O. Chloroform p = 4, CUPREOL, C 20 H 34 O. Chloroform Anhydrous ether chloroform c = 2, ^/> = 31.12 > = 2 5 8 . _ > J^ 5 = - 37-02, - 37.8i, - 38.63 f from which ..... [<*]$= 36.61 0.249 r Solutions of the anhydrous substance in ether (c = 7.941), and in petroleum spirit (c = 10) gave, in agreement with each other, the following numbers : 7 tight. B. C. D. E. fe F. . [*] = 20.63 -25-54 -31-59 39.91 -41.92 -48.65 62.37 = ~ l8 ' 8 Cholesterin Ester of Oleic Add, C M H 43 O.(C W H M O). From dog serum. Melting-point, 41 to 45. Equal parts of ) alcohol 4 chloroform j c " 7 ' 4 ' Ukiernik : Ztschr. physiol. Chem., 15, 430. Hesse : Ann. Chem. (Uebig), 234, 248. Hesse : Ibid., an, 272. Hesse: /bid., aa8, 291. Hesse : Ibid., 228, 294. Hesse : /< = 44- l ' ISOCHOLESTERIX, C, 6 H 44 O. Melting-point, 138 to 138.5. Ether c= 7-344, [<*]/> = + 60 ' 11 ^ = 6.435, " = + 59.1* PARACHOLESTERIX, C 26 H U O. Melting-point, 134 to 134.5. Chloroform p 2.7, d 1.4717, [a] = - 28.88 5 CAULOSTERIN, C 26 H 44 O. Melting-point, 158 to 159. Chloroform c 5.0905, [<*]z> = 49.6 6 ERGOSTERIN, C.^H^O. Chloroform c 3.33, [a]/? = 114 7 KOPROSTERIX, C., 5 H 44 O. From human feces. Melting- point, 95 to 96. Ether f= 13.2 (1.581 gram subst. in 12 cc. ether), [or]/) = 4- 24 8 GLYCQCHOLIC ACID, C, 6 H 43 NO 6 . Alcohol c 9.504. Rotation independent of the concentration. Light. C. D. E. t>*. F. G. ! ^ [or] = 21.6 - 29.0 + 37.9 + 40.0 + 48.7 + 56.8 Sodium Salt, NaC, 6 H 4 ,NO 6 . Alcohol c = 20.143, [<*]/? = + 2 5-7) 9 Water c = 24.928, " = + 20.8 } The concentration is without influence. TAUROCHOLIC ACID, C 26 H 45 NSO 7 . Sodium Salt, NaC 26 H 44 NSO.. Alcohol -.= 9.898, \_a-] D = -f- 24.5, [ajyr^ + 39 ) 10 Water c = 8.856, " = = + 21.5 " = + 34 ) 1 Hesse: Ann. Chem. (Liebig), 192, 177. - Ivikiernik : Ztschr. physiol. Chem., 15, 430. 3 Schulze : Ber. d. chem. Ges., 12, 149. * Schulze, Barbieri : J. prakt. Chem. [2], 25, 170. ' Reinke, Rodewald : Ann. Chem. (t,iebig), 207, 229. Schulze, Barbieri: J. prakt. Chem., [2], 25, 166. 7 Tanret : Ann. chim. phys., [6], 20, 289. s Bondzynski and Humnicki : Ztschr. physiol. Chem., 22, 396 ; Bondzynski : Ber. d. chem. Ges., 29, 476. 9 Hoppe-Seyler : J. prakt. Chem., [ij, 89, 261. 1( > Hoppe-Seyler : Loc. cit., p. 263. 720 CONSTANTS OF ROTATION OF ACTIVE BODIES The concentration is without influence. CHOLAUC ACID, C M H 40 O 6 (+ H,O) and ( + 2V,H,O). a. Anhydrous Cholalic Add. From ox gall : Alcohol, c == 3.338, [a]/? = + 50.2 ) J " dogfeces: " 2.942, " -[-47.6 j b. Cholalic Acid with 2 1 /, mols. water of crystallization.- c. Anhydrous. M* Anhydrous. [*]/>. Hydra ted. 6.070 + 35.4 + 31.9 4-433 + 34.8 + 31-4 Solutions 2.707 +35-2 + 31.7 in alcohol 2.659 + 33-9 + 3 .4 2.030 + 34-5 + 3I-I 1.804 + 34-2 + 30.8 Cholalic acid with 2 ! / 8 mols. water. Alcohol, c = 2.962 (2.659 anhydrous) : 3 Une. B. C. D. E. b*. F. G. M For hydrated ) substance ) + 25.3 + 27.0 + 30.4 + 40.1 + 42.2 + 47-3 -f 60.8 -f w For anhydrous ) substance I + 28.2 4-30.1 + 33-9 + 44-7 + 47-0 + 52.7 -f6 7 . 7 + 78.0 c. Cholalic Add Alcoholate, C, 4 H 40 O 5 -f C 2 H r> OH. 4 Solution in alcohol of 97 volume per cent. p lK with crystal alcohol. pure acid. pure acid. 1.9207 1.7262 0.81518 l6.4 15 + 3L30 \ 34.83 16.2 1-3340 1.1989 0.81050 23.4 23.4 + 31.12 + 34.63 23 I.I525 1.0358 0.81006 20.4 20.4 + 3L30 + 34.83 21. 2.9204 2.6246 0.81493 21. 1 21. 1 + 3 r -72 +35.29 21.2 0-9559 0.8591 0.81502 14.2 15 + 32.02 + 35.63 12.2 0-4794 0.4309 O.8l26o 13.4 15 + 31.85 + 35-44 14 1 Hoppc-Seylcr : Loc. cit., p. 266. 1 Hoppe-Scyler : IMC. cit., p. 267. 3 Hoppc-Seyler : J. prakt. Chem., [i], 89, 267. Vahlen : Xtschr. physiol. Chem., ai, 253. BILIARY SUBSTANCES 721 Potassium Salt, KC., 4 H 3 ,,O V a. Aqueous solutions. c = c = c = c C c = 6.004, \_ai\D = 4- 28.2 i J 7.000, " = -f- 27.5 12.562, " == -f 25.9 16.749, -4-24.6 22.332, -24.1 29-775, = + 24.9 } P. d . [a]*,. 1-0433 3-4940 5-4570 i < 6.6825 b. Alcoholic solutio P I.OO2O 15 4~ 29.10 -| -1 0-9993 30 + 30.79 1-00975 5 4- 27.23 1.00923 20 4- 26.89 1.01498 ii 4- 27.69 1.00695 40 4~ 27.06 1.01843 24 -f 26.51 ns. d' 2b = 0.81027, [a]^ = 4- 31.60 i 3 d 1 ' = 0.81875, [ajg = 4- 31-27 j I.ight. C. D. E. b. F. O] = | +23.7 30.8 +38-5 4-40.9 4-47-5 * I i i Sodium Salt, NaC 24 H 39 O 5 . a. Aqueous solutions. 5 c = 19.049. Ught. R. C. D. E. b. F. [a] = - 19.7 21.0 -f 26.0 +33-1 4-34-0 4-42.0 Hoppe-Seyler : J. prakt. Chem., [i], 89, 270. Vahlen : Ztschr. physiol. Chem., 21, 253. Vahlen. Hoppe ; Seyler : Loc. cit., p. 269. Hoppe-Seyler : Loc. cit , p. 271. 46 722 CONSTANTS OF ROTATION OF ACTIVE BODIES p. d<. < wi. 7.5888 1.01969 22.5 + 27.46 1 4.9450 1.01298 24 + 28.19 4.0419 1.00826 20 -f 27.65 2.2920 1.00595 23 + 30.61 From this, the specific rotation appears to be increased by lowering the concentration. b. Alcoholic solution. c = 2.230, [a]/? = + 31.4 - Cholalic Add Methyl Ester, CH 3 .C 24 H 39 O 5 . Crystals. Alcohol c= 4-59, l = + 31-9 Cholalic Add Ethyl Ester, C 2 H 5 . C 24 H 39 O 5 . Crystals. 8 Alcohol c = 18.479. Ivight. B. D. E. b. [a] = + 25.4 + 32.4 + 40.5 -f 42.3 CHOLEINIC ACID, CJH^O 4 (?).--C 25 H 42 Vahlen : Aor . /. * Hoppe-Seyler : Loc. cit., p. 271. * Hoppe-Seyler: Loc. cit., p. 272. I^atschinoff : Ber. d. chem. Ges., 19, 475. GELATINOUS SUBSTANCES 723 LITHOFELLINIC ACID, C 20 H 36 O 4 . Melting-point, 205. Alcohol [or]/, -f 13.76 l 30. Gelatinous Substances a'-GLUTiN (ordinary gltitin, gelatine). Water,, = 6.ia|' = 24t025 ' W = ~ ^^ (t = 35 to 40, ' = 123.0 Water, ,^ 3-06 S^ 24t 25 ' ~ ^ 5 f/ = 35, ' =-125-0 The rotation decreases, therefore, with the temperature ;but, on the other hand, it does not appear to be much influenced by the concentration. The effect of acids and alkalies is shown by the following experiments : Glutin solution with c = 3.06 : Mixed with equal vol. of ammonia [a]/> = 130.5 i * " " a few drops of sodium hydroxide " = 130.5 | " " the same volume of hydroxide " = 112.5 ( " " " " " " acetic acid " = -114.0 j The specific rotation of aqueous glutin solutions is decreased by long boiling. 3 /?-GivUTiN. Obtained by heating i part of gelatine with 2 to 3 parts of water, in a pressure bottle, for several days to 100, until the liquid no longer solidifies on cooling. 4 On account of lack of uniformity, the products showed dif- ferences in the rotating power : Product with 1.40 p. c. ash. Water, c = 5, []^ 5 ' 5 = 130.6 " 1.96 " " " - : Hoppe-Seyler's med.-chem. Untersuch., i, 71. 9 Nasse : Maly's Jahresbericht, 1889, p. 29. 4 Framm : Arch, fur die ges. Physiol., 68, 144 (1897). 724 CONSTANTS OF ROTATION OF ACTIVE BODIES 2. The addition of methyl and ethyl alcohols in amount too small to produce precipitation, decreases the rotation. 3. Alkali chlorides (and also KI), and also alkali nitrates, bring about a decrease in the rotation which is independent of time and temperature, but a chemical change in the glutin is not produced. Alkali sulphates have no similar action. 4. Acids diminish the rotation, and in greater degree the larger the amount used and the higher the temperature. The intensity of the action follows in the order : HC1, H. 2 SO 4 > HC.HgO.,, H 3 PO 4 . The change does not increase with time. On neutralization of the acid the original value of the rotation is not restored. 5. Alkalies and ammonia (the last in concentrated condition only) diminish the rotation. The decrease becomes greater with time, and is not altered by neutralization of the alkali ; a chemical change in the glutin is, therefore, produced. 1 CHONDRIN. c = 0.957. Water with a few drops of sodium hydroxide solution [<*]/ 213.5 1 2 The liquid mixed with an equal volume of sodium hydroxide solution " 552.0 The latter mixed with the same volume of water " = 281.0 31. Protein Bodies SERUM ALBUMIN. 1. Neutral aqueous solution [#1^ 5^ Aqueous solution, saturated with NaCl " = - 64 acetic acid added " = 71 " hydrochloric acid added until pre- cipitate formed redissolves "= 78.7 2. [O]D = - 62.6 to - 64.6 * 3. Serum albumin crystals : Third crystallization p .._ 2.07 \ct\ D 62.6 ) Still further purified /> = 2.345 " = 60. 1 ) 4. Serum albumin crystals : Water p = 3.92, [a]/, -61.2 " P 3-2 -61 Third crystallization : " = 64 1 Framm : Loc. cit. - de Bary : Hoppe-Seyler's med.-chem. Untersnch., i, 71. Hoppe-Seyler : Ztschr. fur Chem. u. Pharm. v. Erlenmeyer, 1864, p. 737. 4 Starke : Jahresbericht fiir Thierchemie, 1881, p. 18. 4 Sebalien : Ztschr. physiol. Chem., 9, 439. Michel : Verh. d. physik.-med. Ges. zu Wiirzburg, 29, No. 3. PROTEIN BODIES 725 LACTALBUMIN. Water, /> 2 20, " p 3-32, / 4.23, " p -3.12, []/> -36.6' -36.4 " = 37-0 [ " -38.0 J Another preparation : EGG ALBUMIN. i . Aqueous solution : Independent of the concentration [<*]/> = 35-5 By addition of hydrochloric acid . 2. Dilute hydrochloric acid : 35-5 ) 2 37-7 J ["]*= -37-8 03 3. White of egg, fractionated by its solubility in ammonium sulphate solutions, gave Fraction i soluble in concentrated salt solution, 3.75 per cent albumin ............................ [a]/, = 42.90 Fraction 2 soluble in half-saturated salt solution, 8.59 per cent, albumin ............................ " 34-30 Fraction 3 soluble in dilute salt solution, 6.48 per cent, albumin ................................ " 25. 13 4. Ash-free egg albumin, wheat albumin and pea albumin made by the process of Harnack.' The specific rotations refer to \a\ D . 6 Ash-free albumin from In ico cc. solution. Gram. In acid solution. In alkaline solution. The acid solution treated with XaOH Not dialyzed. Dialvzed to turbidi- ty and then cleared with HC1. until it became clear. -6 9 . 9 -66.8 to alkaline and fur- reaction ther addi- \vith tion of sec. phenol- cone. NaCl phthalein. solution. Piep. f A Eggs ' B 1 ' \ A Wheat \ B ( 0.1365 0.326 0.265 -57-0 -54-6 - 4 6.2 "77-5 -52.5 -97-6 - 55-9 0.287 0.188 93.0 0.102 -88. 7 - 7 2.6 99-3 54-6 -31.0 60.2 Peas- 0.534 8? 7 .... 62.0 o-/ 1 Sebalien : Ztschr. physiol. Chem., 9, 457, 459- - Hoppe-Seyler : Ztschr. fur Chem. u. Pharm., 18^4, p. 737. 3 Starke: Jahresber. f. Thierchemie, 1881. p. 18. 4 Bondzynski and Zoja . Ztschr. physiol. Chem., 19, n. '' Ber. d. chem. Ges., 22, 3046 ; 23, 3745 : 25, 204. '' Billow : Pfluger's Arch. f. d. ges. Physiol., 58, 219. 7 26 CONSTANTS OF ROTATION OF ACTIVE BODIES ALBUMOSES ( Propeptones ) . 1 Protoalbumose : ! Dissolved in hydrochloric acid of o 04 to 0.08 per cent. Dissolved in sodium carbonate solution of o.i 2 per cent. A c* = 1.588 [<*]/> = - 72.6 (* = 2.198 / = 2o [a]/, -81.2 B " =2.277 " - 79.1 " = 2.223 t = 21 " 70.6 C " = 1.925 - 77.9 " = 1.87.3 i = 23-5 " -80.1 11 11 = 2.361 /=22.5 " -79-2 D = 1.366 " - 73 .2 .... .... E " = I.68I - 71.4 " = 1.904 '=24.5 " -76.3 .... .... 11 = 2.494 ^=24.5 " = 75-3 Deuteroalbumose : Hydrochloric acid with 0.06 p. c. HC1 c* = i .680 (>]z) = - 74.4 " 0.04 " " " = 1.517 " = --79.1 Sodasol. " 0.12 " Na^Oa. Prep. A " = 2.537 " = -74.3(77.6) " " " 0.12 " " B "=1.765 " = 75.3 Salt " " 0.5 " NaCl " A " = 1.287 " = -77-7(77.o) " " " 0.5 " " B " = 1.916 " - 72.0 Heteroalbumose : Hydrochloric acid of 0.07 p. c. HC1 c= 1.748 [or]/>= 68.65 Soda solution c --.-- 1.584 " = 60.6 FlBRINOGEN. i. Dissolved in 2 to 3 per cent. NaCl solution: NaCl + ash. fc = 0.426 i. 362 p. c. [ = - 35-2 \* From ox blood U = o. 4 i4 1-752 -36.2! 1* = 0.318 1.415 -36.5 I [^ = 0.263 1.819 " -37-7 i Mean " = - 36.8 | From horse blood <:= 0.808 3.744 p. c. " 50.5 J Solutions with more than 0.5 per cent, of fibrinogen can not be polarized on account of the marked opalescence. *c ash-free. 1 Kiihne and Chittenden : Z. f. Biologie von Kiihne u. Voit., ao, (n. F. 2). 25 1048. * Cramer : Ztschr. physiol. Chem., 23, 83. PROTEIN BODIES 727 2. Dissolved in NaCl solution : XaCl - ash. (----0.205 i. 045 p. c. []/> 50.6 ] From horse blood -j i = 0.532 2.251 =-53-9 L = 0.291 2.409 " 54-1 Mean ' = 52.5 3. Dissolved in o. i per cent, solution of sodium carbonate : Na 2 C0 3 + ash. C 2= 0.481 1.788 p. C. [a]z> = 46.7 i * = 0.225 1.341 " -44-5 = 0.400 1.611 " 45-7 = 0.592 2.284 " " 45-1 SERUMGLOBULIN. Salt solution WD = 47.8 3 CRYSTALLIN. From the crystalline lens : or-Crystallin : Water, p = 3.29, [<*]/> '-= 46.9 ^i 4 /3-Crystallin : " p = 3.12 " : 43-3 " p=i.Bo " -43-1 Albumoid from the crystalline lens : Water ^ = 2.33, [a] D = - 5-9) 5 " P=2.$l, " = 52.2 j VlTELLIN. " = -43-35 SYNTONIN. From the myosin of muscles by solution in very dilute hydrochloric acid, or by action of strong hydrochloric acid on albumin. Solution in very dilute hydrochloric acid : [or]z> = 72 (Independent of the concentration.) Almost the same rotation is found in weak alkaline solution. 1 Mittelbach : Ztschr. physiol. Chem., 19, 289. * Cramer: Ibid., 23, 86. 3 FrexJericq : Arch, de Biolog., i, 17. * Morner : Ztschr. physiol. Chem., 18, 91, 99. 5 Morner : Ibid ., 18, 77, 88. 6 Chittenden and Mendel : Jahresher. f. Thierchemie von Maly, 1895, pp. 29. 33. 728 CONSTANTS OF ROTATION OF ACTIVE BODIES On heating the hydrochloric acid solution in a closed vessel to about 100, the rotation is increased to : Wi> = - 84.8 i ALBUMINATES. Formed by the action of strong potassium hydroxide solution on albumins. As maxima there were found : Albuminate from serum albumin [&]D ~ 86 1 - " uncoagulated egg albumin " 47 " coagulated " " "= 58.5! " casein (solution in strong KOH. Rotation variable with amount of alkali) " = 91 CASEIN. In magnesium sulphate solution [<*]/> 80 1 :l In water containing 4 cc. of fuming hydrochloric acid, per liter * " = - 87 |- In the smallest possible amount of sodium hydroxide solution " = 76 In aqueous solution, as strong as possible [<*]> = 1*7-7 * HEMIELASTIN. Water p 2.509, [a] D = - 92.7 (approx. ) 5 Elastinpeptone : Water p ^ 6.14, [a]/, = - 87.9 6 I Hoppe-Seyler : Ztsch. f. Chem. u. Pharm. von Erlenmeyer 1864, p. 742. -' Hoppe Seyler : Ibid., 1864, p. 757. * Hoppe-Seyler : Loc. <-/'/. 4 Bechamp : Bull. soc. chim., [3], n, 152. 5 Horbac/ewski : Ztschr. physiol. Chem., 6, 337, 343.. II Horbaczewski : Loc. cit. GENERAL INDEX Absolute rotation for quartz 413 Accuracy of Laurent polarizer ... 348 Action of molybdates and tungstates on malic acid 250 tartaric acid Active bodies, definition of i formed in vegetable cell 137 crystals 13 isomers 130 from active materials 132 transformation of 238 modifications 2 Addition of inactive bodies 243 Alkaloids, determination of 49^ used in resolution 103 Allowance for earth's magnetism 364 Analysis of celluloid 497 chocolates 487 cinchona alkaloids 49 s " confectionery 4^7 grape sugar 49 1 milk sugar 4^8 sugars 463 tobacco 503 Analyzer 3 11 Andrews, observations of 3 S 4 Anomalous rotation dispersion 157 Antimonyl tartrates 22 ^> Antipodes, electrical conductivity 68 physical differences physiological differences 72 refraction 63 taste 76 toxicity ?6 transformation of *3 S Apparatus and methods 36 for exact measurements 361 Applications of optical rotation 463 Arons-L,ummer lamp 433 Arsenyl tartrate 226 Artificial preparation of active compounds 13 Aspergillus glaucus I2 3 Asymmetric atoms, definition 44 carbon ' 47 nitrogen 38 and 52 sulphur 38 and 53 Asymmetry, Pasteur's theory 44 Beet juice saccharimeter 389 Behavior of antipodes a- Benzyl-phenyl-allyl-methyl ammonium iodide, resolution of 115 Biot's determination of specific rotation . . .... 170 formulas J 46 proof of 176 2 730 GENERAL INDEX Birotation 257 Bodies resolved by fungi 127 Boltzmann's formulas 147 Bon-bons, sugar in 487 Boric acid and tartrates 246 Boryl tartrates 225 Brix table for sugar solutions 474 Broch's method for rotation dispersion 419 Bromcamphorsulphonic acid in resolution 113 Calculation of optical modifications 55 to 65 sensitiveness 340 Camphor, determination of 497 oxime, resolution of 114 specific rotation 191 Candy, sugar in 487 Cane sugar and alkalies 251 determination of 463 specific rotation 166 and temperature effect 383 Caramels 487 Causes of multirotation 273 Celluloid, camphor in 497 Change in specific rotation by dilution 203 Changes in specific rotation, various 215 Chlorcamphorsulphonic acid in resolution 113 Cinchona alkaloids '. 498 Cinnabar, crystal rotation 13 Circular polarization, theory of 41 Classification of active bodies 7 Clerget's formula 484 Cocaine, determination of 501 specific rotation 168 Complex polymerized molecules 23^ Compounds with several asymmetric carbon atoms 296 Concentration and specific rotation 203 of solutions 458 effect on rotation 169 Conditions of racemization 96 Confectionery, analysis of 487 Constants of rotation 505 Construction of I^ippich's polarizer 358 polariscopes 316 polarization tubes 436 Copellidine, resolution of 112 Cornu's polarizer 344 Crum-Brown's hypothesis on rotation 299 Crystalline form and rotation 7 mixtures 1 1 Crystals, rotation dispersion of 148 Crystal sugar and raffinose 482 Cultures for resolution 117 Density of racemic compounds ' 79 Dependence of rotary power on masses of radicals 299 Determination of alkaloids 498 camphor 497 cane sugar 463 cocaine 501 concentration 458 GENERAL INDEX 731 Determination of galactose 496 glucose 491 maltose 495 milk sugar 488 nicotine 503 percentage strength 444 rotation dispersion 419 specific gravity 449 specific rotation 165 Dextrorotation i Dextrose and calcium chloride 253 determination of 491 in diabetic urine 493 Diabetic urine, sugar in 493 Dilute solutions, specific rotation of 196 Directions for making polarizations 467 Dispersion formulas of Boltzmann and Trommel 147 Dissociation of active salts 225 Double field instruments * 351 wedge compensation 368 Earth's magnetism, effect of 364 Effects of errors of observation 461 linkage 293 temperature in saccharimetry 383 on specific rotation 441 Electric light 394 Electrolytic dissociation 215 Errors of observation, effect of 461 saccharimeters 380 Esterification, resolution by 115 Esters of lactic acid 284 malic acid 283 Ethyl dextrotartrate. specific rotation 185 Ethyl piperidine, resolution of 112 Filtration of solutions, change of strength by 448 Formation of active bodies in vegetable cell 137 active isomers 130 racemic bodies 90 Formula of Clerget 484 Formulas for rotation 3 to 6 of reduction 175 French saccharimeter scale 372 Fresnel's theory 4 1 Fric's saccharimeter 392 Fungi, resolution by 117 Galactose, determination of 496 Gas lamps 393 General formulas 3 to 6 Glan's prism ... 310 Glucose, determination of 49 1 Grape sugar, determination of 491 Gumlich, rotation of quartz 152 and 415 Guye's hypothesis 299 Half-shadow instruments 335 by Cornu 344 Fric 392 Heele 34 Jellett 342 732 GENERAL INDEX Half-shadow instruments by Landolt 358 Laurent 344 Lippich 351 and 354 Lunimer 356 and 365 saccharimeters 387 Hartnack prism 310 Heele's half-shadow polarizer 348 Hemihedry 46 Kinks' petroleum lamp 394 Historical remarks 6 Homologous series, rotation in 290 Hydrolysis 236 Hydrolytic dissociation 223 Iceland spar prism 308 Illuminating lamps 393 Incandescent light 394 Influence of boric acid 246 source of light 338 Inorganic salts and tartrates ....'. ' 244 Intense sodium light 398 Isocopellidine, resolution of 112 Isomers. formation of 130 rotation of 285 Jellett's polariscope 342 lactic acid esters 284 specific rotation 281 Lactones. multirotation of 275 Lamps 393 electric 394 Kinks' . . 394 Landolt's 397 mercury 433 Pribram's 396 sodium 395 Welsbach 394 Landolt's apparatus 358 lamp , 397 large polariscope 361 method for rotation dispersion 429 v. I .a n ii rotation dispersion 423 of quartz 415 Laurel camphor 15 I v aurent's polarizer 344 Laws of polarization 146 LeBel theory 47 Length of tube, measurement 443 Levorotation i Light, purification of 399 and 405 spectral purification 405 Lindner, data on fungi for resolution 117 Linkage of carbon atoms 293 Lippich's light filters 399 method for rotation dispersion 425 polarizer 351 Liquid racemic compounds 86 Lomniel's method for rotation dispersion 427 Lumim-r s instruments 356 and 365 Magnetism, earth's, effect on rotation 364 GENERAL INDEX 733 Malic acid, esters 283 rotation dispersion 159 Maltose, determination of 495 Matico camphor 14 Measurement of angle of rotation 306 length of tube 443 Measuring flasks 459 Melting-point of compounds 83 Mercury lamp 433 Milk sugar, determination of 488 specific rotation 167 Mitscherlich apparatus 313 and 326 Mixture of two active liquids 240 Modifications of inactive configuration 140 Mohr cubic centimeters 373 Molecular aggregation 227 asymmetry 44 rotation 6 and 39 weight of racemic compounds 77 Molybdates, effect on rotation 248 Multirotation 257 Nature of rotating power 39 Nicol's prism 309 Nicotine, determination of 503 rotation dispersion 161 specific rotation 168 and 181 Nitrogen compounds, asymmetric 38 Normal sugar solution 373 Number of optical modifications 55 to 65 Numerical values for specific rotation 165 Observations, errors in 461 with homogeneous light 327 Optical center and brightness 407 of gravity 399 constitution of active liquids 43 modifications 54 superposition 296 rotation, applications 463 Oxy-acids, multirotation of 275 Parasantonide, specific rotation 167 Pasteur's theory of asymmetry 44 Patchouli camphor 15 Path of rays in polariscope 319 Penicillium glaucum in resolution .121 Percentage strength of solutions .... 444 Peters' saccharimeter 391 Petroleum lamp 394 Phenyl ethyl amine, resolution of 113 Physical and chemical behavior of optical modifications 67 laws of polarization 146 Physiological differences in antipodes 71 Pipecoline, resolution of m Polariscopes, construction of 3 J 6 Polarizer 3" Polarization by reflection 307 refraction 38 in quartz 41 tubes 436 734 GENERAL INDEX Polymerized molecules , ; . 233 Pope and Peachey, asymmetric nitrogen and sulphur 53 Position isomerism 286 Powdered crystal?, rotation of n Practical applications of rotation 463 directions 314 Preparation of sugar scale 369 Pribram's lamp 396 Prisms 308 Production of racemic bodies 97 Proof of Biot's formulas 176 Propylene diamine, resolution of 112 Propyl piperidine, resolution of 112 Pure cultures for resolution 117 Purification of light 399 Pycnometer 451 Quadruple field instrument 356 Quartz, rotation dispersion of 149 specific rotation 151 Quotient for sirups 476 Racemic bodies, production by heat 93 compounds, distinction from active forms 76 Racemically and structurally inactive isomers 144 Racemization by heat 94 Raffinose 482 Rate of change in rotation 266 Ray filters 429 Rays, path of, in polariscope 319 Reciprocal transformation of isomers 140 Reduction formulas 175 Relation of angles Q( D and a j 4 ! 5 Resolution by active compounds 102 alkaloids ' 103 crystallization 99 esterification and saponification 115 fungi 117 stronger organic acids 113 tartaric acid no of racemic compounds 99 Reusch's mica plates 42 Reversal of direction of rotation 201 Rhamnose, specific rotation 167 Ring structure molecules 66 Robiquet's polariscope 328 Rochelle salt, rotation of 243 Rotating power 2 Rotation and chemical constitution 282 crystalline form 7 length of column 146 wave length of light 146 change in 266 dispersion 146 and 419 Broch's method 419 l..-in>s effect on specific rotation 207 of transition 90 Testing the saccharimeter scale 376 Tetrahydronaphthylenediamine, resolution of 113 Tetrahydropapaverine, resolution of 114 Tetrahydroquinaldine, resolution of 113 The 100 point on saccharimeters 374 Theory of Soleil double plate 329 Tobacco, nicotine in 503 Transformation of active isomers 138 Transition temperature 90 Triple field instrument 354 True specific rotation 170 Tubes for polarization 436 Tungstates, effect on rotation of tartaric acid 248 malic acid 250 Turpentine, specific rotation of 177 Van't Hoff, calculation of active modifications 55 Van't Hoff-IveBel theory 47 Variable specific rotation 169 Variations in specific rotation 194 Vegetable cell, formation of active bodies in 137 Ventzke sugar scale 372 Water jacket tubes 438 specific gravity tables 455 Wave length of light and rotation 146 Wedge compensation 366 Weighing solutions 446 Welsbach lamp 394 White light lamps ^ 393 optical center " 416 Wild's polaristrobometer . . . 331 Zirconium light 394 INDEX OF ACTIVE SUBSTANCES Abietinic acid 666 Acetyl cinchonine 682 coiiine 709 quinine 677 Achrodextrine 608 Aconine 668 hydrochloride 668 Aconitine 667 hydrochloride 667 Adonitol 511 structure 61 Albumin 72^ Albuminates 728 Albumoid from crystalline lens 727 Albumoses 726 Alcohol, /-amyl o6 Aldehyde sugars 574 Aldoses 574 Aliphatic terpenes 612 Alkaloids. 667 aconite species 667 cinchona . 671 coca leaves 698 opium 701 strychnos varieties 706 Alkyl oxypropionic acids 518 Allocinnamic acid dibromide, resolution 109 Allomucic acid, structure 63 Aminoterebenthene hydrochloride . . . 640 /-Ammonium antimonyl malate 533 d- hydromalate . 535 /- hydromalate 531 d- hydrotartrate 548 /- malate 531 d- tartrate 551 /- tartrate 563 Amygdalin 713 Amyl acetate 506, 508, 509 Amyl acetate, sec 508 d-Amylacetic acid 515 /-Amyl alcohol 506, 509 derivatives 506 amine 509 hydrochloride 509 amylacetate 297, 507, 509 r-amylmalonate 509 (/-amylmalonate 509 benzoate 508 bromide 506, 509 brom-w-butyrate 506 brom-isobutyrate 506 butyrates 506, 508 47 . /-Amyl chloracetates chloride cinnamate crotonate diamylacetate esters, effect of linkage. . Guye'9 hypothesis, rotation ethers formate hydrocinnamate iodide isobutyrate lactates malonic acid mandelates inethacrylate a- and /3-naphthoates . . . oxybutyrates palmitate phenylacetate phenylcarbaminate. . . . phenylchloracetate . . . phenylpropiolate ..... phenylpropionate .... piperidine propionates tartrate tolylcarbaminates, o- , ;-, valerate Amylodextrine rf-a-Amyrilene derivatives .... /-a-Amyrilene /3-Amyrilene derivatives a-Amyrin acetate Audropogon oil Angelica oil Anhydroecgonine. hydrochloride Anise oil . ( Antiaronic acid . . . .' Antimonyl tartrate, dissociation Apiin Apocinchonidine acetyl derivative chlor derivatives chloracetyl deriv Apocinchonine chlorate . 506,509 . 506 . 508 . . 508 507 293 302 290 509 508 508 506,509 506,508 297, 50? 507 298,507 . - 508 . 508 297 508 . - 508 . - 508 298, 507 - . 508 - - 508 507 - . 508 - 507 .... 297 . ... 608 .... 659 . - 659,660 . ... 660 . ... 660 .660 .... 659 . ... 660 . ... 662 . ... 662 . ... 701 ... 701 . ... 662 . ... 717 ... 225 ... 713 . ... 690 .... 691 ... 691 ative . 691 . ... 684 . ... 684 INDEX OF ACTIVE SUBSTANCES Apocinchonine chlor derivative 685 hydrobromide 684 hydrochloride 684 hydroiodide 684 perchlorate 685 sulphate 684 Apoquinamine 694 acetyl derivative 694 Apoquinidine 680 chloride 680 diacetyl chloride 680 diacetyl derivative 680 Apoquinine 677 chloride 677 Arabic acid 611 Arabin 611 Arabinose, d- and /- 574 multirotation 259 7-Arabinose 575 Arabinosazone 575 Arabitol 511 structure 61 Arabonic acid 544 anhydride 544 multirotation 279- strontium salt 544 Arginine 668 salts. . . 668 Aribine 668 Aricine 692 Arsenyl tartaric acid 552 tartrate, dissociation 225 Artificial conine 709 dextrine 609 Asafetida oil 662 Asebotoxin 717 Asparagin, if- and /-ft 541 resolution 102 Aspartic acid 58, 540 racemization 94 rotation 211 Aspidospermatine 669 Aspidospermine 668 Asymmetric nitrogen compounds .... 52 sulphur compounds 53 Atisine 668 salts 668 Atropine 669 salts 669 Aurantiin 714 Aurantiol 612 Australene 629 Barium ethyl tartrate 556 malate 533 Basil icum oil 662 Bcbirine 670 Benzoyl carvoxime 618 Benzoyl conine 709 rf-ecgonine 699 amylester hydrochlor- ide 700 butylester hydrochlo- ride 700 ethylester hydrochlo- ride 699 hydrochloride .... 609 methyl ester .... 699 propyl ester hydro- chloride 699 /-ecgonine, methyl ester 700 h y d ro- chloride 700 -tartaric acid esters 559 Benzyl amyl ether 509 camphor bromides 648 valerate 514 Berberine 670 Bergamot oil 661 Betel oil 662 Biliary substances 718 Bioses 596 Bitter principles 715 rf- Borneol 633 acetate 633 benzoate 634 carbonate 634 phenylurethanej 634 phthalate 634 succinate . . 634 /-Borneol 634 acetate 635 benzoate 635 carbonate 635 phenylurethane 635 phthalate 635 succinate 635 rf-Bornylamine 638 derivatives 638 chloride 633 Boryl tartrate, dissociation 225 rf-Bromal borneol 633 /-Bromal borneol 635 Brom-a-amyrin 660 Brombenzoyl carvoxime, (>-, m- and p- . . 618 Bromcamphoric acid 657 Bromnitrocamphor 646 Bromphenylcystein 526 Bromphenylmeicapturic acid 525 Brompropionic acid esters 517 Bromsuccinic acid 540 density of active :in 58 chlormaleate 56. 507 Diamyl chlorsuccinates 298, 507, 508 citraconate 507 divaleryl tartrates 298 fumarate 506, 508 itaconate 507 malates 298, 507, 534 maleate 506, 507 mesaconate 508 mesotartrate 507 methylsuccinate 508 oxalate 506 paradimethylsuccinate 507 phthalate 506 racemate 507 succinate 508 tartrate 298, 507 Diapocinchonine 686 diacetyl derivative ... 686 Dibenzoylglyceric acid esters 250 Dibtnzoylmethyl tartrimide 561 Dibenzoyltartaric acid 558 anhydride .... 558 diethyl ester . 558, 559 diisobutyl ester . . 559 dimethyl ester 558, 559 Dibrommenthone 625 Dicaniphor 655 derivatives 656 resorcin 649 Dicaprylmalate 534 Dichlorcamphor 643 Dicinchonine 697 hydrochloride 697 Diethyl acetylmalate 534 amylmalonate 507 amylnitrobenzylmalonate . . . . 507 bromacetylmalate 534 brombut3-n-lmalate 534 bromisobutyrylrnalate 534 brompropionylmalate 534 diamylmalonate 507 dibenzoyltartrate 539 di-o-, m-,p-, toluyltartrates .... 560 isobutyrylmalate 534 isovalerylmalate 534 malate 534 monobenzoyltartrate 559 mono-r>- /-,/>-,toluyltartrates .559, 560 propionylmalate 534 tartrate 555 Digitalonic acid 546 Digitonin 7*4 Dihydroxycyancampholytic acid 655 photosantinic acid 716 esters . . . ^716 Diisobutyl acetylmalate 534 742 INDEX OF ACTIVE SUBSTANCES Diisobutyl broinacetylmalate 534 -butyrylmalate 534 cinchonine hydrobromide . . 682 isovalerylmalate 534 malate 534 tartrate 556 Diisopropyl malate 534 tartrate 556 Dill oil 664 Dimethyl acetylmalate 534 bromacetylmalate 534 -butyrylmalate . .... 534 chloracetylmalate 534 dibenzoyltartrate 559 di-o-.Jw-^-.toluyltartrates. . . . 560 isobutyrylmalate 534 malate 534 nitromalate 534 propionylmalate 534 tartrate 555 Diphellandrene 619 Diphenyl acetyltartaric acid anhydride . 559 ethylene diamine .... 510 propionyltartaricacidanhy'ri'e 559 Dipropyl acetylmalate 534 bromacetylmalate 534 w-butyryl malate 534 chloracetylmalate 534 isovalerylmalate 554 malate 534 tartrate 556 Diquinine sulphate 675 Disaccharides 596 Diterpenes 659 Diterpilene 659 Dog fennel oil 664 Dulcitol 512 Dwarf pine oil 663 rf-Ecgonine 698 derivatives 699 hydrochloride 699 methyl ester 699 /-Ecgonine hydrochloride 700 Ecgoninic acid 701 Echicerin 717 Echiretin 717 Echitein 717 Echitatnine 707 Kchitin 717 Egg albumin 725 Elastin peptone 728 Elemi oil 664 Ergosterin 719 Esters of rf-amylacetic acid 515 brompropionic acid 517 chlorpropionic acid 517 glycericacid 523 Esters of lactic acid 516, 517 /-mandelic acid 520 methoxysuccinic acid 537 oxybutyric acid 518 phenyl bromacetic acid .... 522 chloracetic acid .... 521 dichlorpropionic acid . 524 tartaric acid 554 valeric acid 514 Ethereal oils 660 Ethoxysuccinic acid, resolution 106 rf-Ethoxysuccinic acid 538 acid am'onium s'lt 538 barium salt . . . 538- calcium salt . . . 538 esters of 538 normal ammoni- um salt 538 /-Ethoxysuccinic acid 539 acid ammonium salt 539 esters 539 Ethyl acetylmalate 535 amyl 505 amylacetate 507 amylacetoacetate 507 amylether 509 benzoylmalate 535 borneol 633 camphene 628 dextrotartrate 185, 555 diamylacetate 507 diamylacetoacetate 507 malate 535 o-, /-,/>-, toluylmalates 535 piperidine, resolution 112 tartaric acid 555 barium salt 555 calcium salt 555 lithium salt 555 potassium salt .... 555 sodium salt 555 tartrimide 561 valerate 514 Ethylene diamine ditartrate 554 Eucalyptus oil 664 Euphorbon 717 /-Fenchene 633 Fenchol 637 rf-Fenchone 657 oxime 657 density 80 /-Fenchone 658 oxime 658 rf-Fenchyl alcohol 637 /-Fenchyl alcohol 637 rf-Fenchyl amine 639 l>en/.ylidene compound 659 INDEX OF ACTIVE SUBSTANCES 743 .-Fenchyl amine 639 acetyl compound .... 639 benzylidene compound 639 Imtyryl compound . . . 639 derivatives, change in rotation 292 formyl compound . . . 639 methoxybe nzylidene compound 640 oxybenzylidene c o in - pound 640 propionyl compound . 639 Fennel oil 664 Fermentation gum 609 lactic acid, resolution ... 101 Fibrinogin 726 Fir needle oil 663 Frankincense oil 664 -, ///-, and />-toluylmalate .... 535 valerate 514 Milk sugar 599, 601 birotation 600 constant specific rotation . . . 167 determination 488 multirotation 265 octoacetate 601 a-Monobronicamphor 643 sulphochloride . . 644 sulphonamide . . 644 sulphonic acid . . 644 salts 644 /3-Monobrom camphor 645 y-Monobrom camphor 645 Monocamphor phenol 649 Monocamphor resorcin 649 o-Monochlorcamphor 642 sulphochloride . . 643 sulphonamide . . . 643 sulphonates .... 643 /3-Monochlorcaraphor 643 y-Monochlorcamphor 643 Monoiodo camphor 645 Monomethyl tartrate 554 Morphine 701 acetate 702 hydrochloride 702 sulphate 702 used in resolution 104 Mucic acid . . . 571 My cose 604 Napeliine 667 a- and /3-Naphthol camphor 649 Narceine 705 Narcotine 705 Naringin 7 J 4 Natural conine 708 malic acid 526 Nicotine 710 acetate 7 11 anomalous dispersion 161 constant specific rotation .... 168 determination 503 hydrochloride 711 minimum specific rotation ... 197 neutral sulphate 711 rotation in different solvents . . 181 salts 7" Nitrobenzoylcarvoximes 618 Nitrocamphor 645 derivatives 645, 646 quinine 677 Nitromannitol 511 Xitrosocamphor 646 Norisosaccharic acid 571 Octacetyl diglucose 585 maltose 603 melibiose 604 Oils, ethereal 660 Onion oil 665 Opianine 705 Opium alkaloids 701 Orange flower oil 665 oil 661 Ordinary milk sugar 599 Oubain 714 Oxyacanthine 670 hydrochloride 670 Oxyacids, multirotation 275 Oxyaldehydes 574 Oxy-a-aniyrin 659 a-Oxybutyric acid, resolution of 107 /3-Oxybutyric acid 518 Oxycamphocarboxylic acid 650 esters .... 650 a-Oxycinchonine 687 hydrochloride 687 /3-Oxycinchonine 687 Oxydimorphine 704 Oxyethylbrucine hydrochloride 706 Oxygluconic acid 568 Oxyketones 589 Oxypropionic acids 518 Palma rosa oil 665 Papaverine 705 Paracholesterin 719 i Paraconine 710 Paracotol 659 Paralactic acid 515 Paraphytosterin 719 Parasaccharinic acid 546 Parasantonic acid 715 esters 715 Parasantonide, constant specific rotation 167 Parasorbic acid 515 Paricine 697 Patchoulene 659 Patchouli camphor 658 rotation of solid ... 15 Paytine 669 Pea albumin 725 Pennyroyal oil 665 Pentitols 511 Pentoxypimelic acids, a- and ft- 572 Peppermint oil 665 Phasol 718 Phellandrene 619 nitrate 619 Phenacetylcarvoxime 618 748 INDEX OF ACTIVE SUBSTANCES Phenolamylether 510 Phenoxacrylic acid 49 Phenyl bromacetic acid 522 a-bromlactic acid, resolution . . no chloracetic acid 521 dibrombutyric acid, resolution . 109 a- and 0-dibrompropionic acid, resolution 108 dichlorpropionic acid .524 a- and ^-dichlorpropionic acid. resolution 109 ethylamine, resolution 113 mercapturic acid 525 Phlein 611 Phloridzin 713 Photosantoniu 716 Phytosterin 719 Picien 714 Picrotoxin 717 Pilocarpidine 708 nitrate 708 Pilocarpine 707 hydrobromide . 708 hydrochloride 708 nitrate 708 sulphate 708 Pine needle oil 663 rf-Pinene 629 dibromide 630 hydrochloride 630 /-Pinene 631 hydrobromide 632 hydrochloride 632 Pinitol 512 Pinol hydrates, d- and /- 638 Pipecoline 708 resolution of a- and ft- forms . in, 112 Pipecolinic acids, d- and /- 515 Pipeline 708 Podocarpinic acid 666 Polei oil 626 Polyterpenes 658 Populin 713 Potassium antimonyl tartrate 553 arsenyl tartrate 553 ethyl tartrate 556 hydromalate 531 malate 531 tartrates 548 Propeptones 726 Propionylquinine 677 d-Propoxysuccinic acid 539 salts 539 /-Propoxysucctnic acid 539 salts and esters . . 539 Propyl acetylmalate 535 Propyl amyl 505 ether 509 malate 535 a-piperidine 708 resolution 112 /3-piperidine 710 tartrates in different solvents . . 206 valerate 514 Propylene diamine, resolution 112 glycol 58 oxide 49 Protein bodies 724 Pseudo cinchonine 686 dihydrochloride . . . 687 codeine 703 hyoscyamine 670 inulein 610 morphine 704 hydrochloride 704 narceine 705 Pulegone ; 626 bromide 626 oxime 626 oxime 626 hydrochloride 626 Ptyalose 601 Pyroaconine hydrochloride 667 Pyroaconitine 667 hydrobromide 667 Pyroglutaminic acid, d- and /- 543 Pyrotartaric acid, resolution 106 Quebrachine 669 Quebrachitol 512 Quebrachol 718 Quinamicine 694 Quinamidine 694 hydrochloride 694 Quinamine 693 hydrobromide 693 hydrochloride 693 hydroiodide 693 nitrate 694 perchlorate 694 Quinic acid 565 salts 565 Quinicine 678 oxalate 678 Quinidine 678 acetyl derivative 680 acid hydrochloride 679 acid sulphate 679 neutral hydrochloride .... 678 neutral sulphate 679 nitrate 679 oxalate 680 used in resolution 103 Quinine 673 INDEX OF ACTIVE SUBSTANCES 749 Quinine, acetyl derivative 677 anhydride 673 determination 498 disulphate 675 hydrochloride 674 nitrocamphor derivative .... 677 oxalate 676 propionyl derivative 677 sulphate 675 sulphonic acid 676 used in resolution 103 Racemic acid, resolution 104 Raffhiobiose 604 Raffinose 605 Resin acids 666 Rhamiiitol 511 Rhamnodulcite 577 Rhamnoheptonic acid 570 Rhamnoheptose 588 Rhamnorexitol 512 Rhamnohexonic acid . . . 567 anhydride 567 Rhamnohexose 587 multirotation 265 Rhamnooctonic acid 572 Rhamnonic acid 546 multirotation 278 Rhamnose 577 constant specific rotation . . . 167 multirotation 263 oxime 579 multirotation 264 phenyl hydrazone 579 Rhodinol, d- and /- 613 acetate 613 Ribonic acid 545 anhydride 545 cadmium salt 545 structure 60 Ribose, structure 60 Ricinelaidic acid 519 Ricinoleic acid 519 Ricinstearoleic acid 519 Rochelle salt 550 Rosemary oil 665 Rubidium tartrate, rotation of solid ... 16 Russian oil of turpentine 629 Saccharic acid 570 ammonium salt 570 multirotation 278 structure 63 Saccharimetry 463 Saccharin 45 multirotation 280 Saccharinic acid, multirotation 280 Saccharonic acid and anhydride 568 Saccharose 596, 597, 598 Saccharose, determination 463 Saccharoses 596 to 604 Salicin 713 Salicylic acid camphor 649 Salts of /-glyceric acid 523 d- and /-isopropylphenyl glycolic acid . . 522 lactic acid 516 malic acid 531 d- and /-mandelic acid 520 -oxybutyric acid 518 oxypropionic acid 518 rf-tartaric acid 548 /-tartaric acid 563 Sandalwood oil 665 Santinic acid 716 Santonic acid 715 esters 715 Santonide 715 constant specific rotation . . . 168 Santonin bodies, specific rotation for different colors 716 dispersion coefficient 155 group 715 Santonious acid 717 Santonyl bromide 715 chloride 715 iodide 715 Sarcolactic acid 515 Sassafras oil 665 Savin oil 665 Scopolamine 669 Semmose 587 Serum albumin 724 globulin 727 Sesquiterpenes 658 Shikimic acid 543 ammonium salt 543 bromide 544 derivatives 543 Silver fir oil 663 Sinistrin 611 Sobrerol, density of active and racemic forms 80 Sodium ammonium tartrates .... 552, 563 arsenyl tartrate 553 malate and hydromalate .... 531 tartrates 548 Soluble starch 607 Sorbin 582 Sorbinose 582 Sorbitol 512 structure 63 Sorbose 582 Sparteine 708 Spike oil 666 Staphisagrine 707 750 INDEX OF ACTIVE SUBSTANCES Star anise oil 666 Starch, soluble 607 sugar 582 Storax oil 666 Strychnine 706 salts 706 sulphate, rotation of solid . . 17 used in resolution 103 Strychnos alkaloids 706 Sugar 596 cause of multirotation 273 changes in rotation 194 directions for tests 467 formulas for specific rotation . . . 195 in confectionery 487 rotation in presence of alkalies . . 251 temperature effect .... 598 Sweet marjoram oil 666 rf-Sylvestrene 618 dihydrobromide ...... 618 dihydrochloride 619 nitrolbenzylamine 619 tetrabromide 619 /-Sylvestrene 619 Synthetic conine, resolution . 112 glucosides ...."?. 714 Syntonin 727 Talitol 512 structure 63 Talomucic acid 62, 571 Talon ic acid 567 structure 62 Talose, structure 62 Tanacetone 626 Tannic acid ... 37.; Tannin 573 Tansy oil 666 Tarragon oil 666 Tartar emetic 553, 563 action of alkali salts .... 245 d- Tartaric acid 546, 547 action of molybdates and tungstates 248 density of active and ra- ctrmic forms 80 effect of boric acid on ro- tation 246 esters 554 to 563 melting-point 83 rotation dispersion .... 157 solubility 81 specific rotation of dilute solutions 196 structure 60 /-Tartaric acid 562 ammonium salt 563 calcium salt 564 /-Tartaric acid, potassium antimonyl salt 563 sodium salt 563 Tartrates, acid ammonium 548 boryl 552 lithium 548 potassium 548 sod iu in 548 thallium 549 ammonium 551 potassium . . . . 551 sodium 552 lithium 551 magnesium 552 potassium 549 antimonyl .... 553 boryl 552 sodium 550 potassium 550 boryl 552 thallium 553 ammonium ... 554 antimonyl 554 lithium 554 potassium 553 sodium 554 and malates. combinations . . 564 dissociation 225 formation of racemic bodies . 91 influence of alkali sails on ro- tation 243 transition temperatures .... 91 water of crystallixation of dif- ferent forms 79 ' Tartrimides 561, 564 Tartromalamides 564 Taurocholic acid 719 sodium salt 719 - Temperature, influence cm rotation of sugar 598 Terebenthene 629 Terecamphene 628 acetate 629 formate 629 hydrochloride 628 Terpan group 614 r/-Terpineol 622 ATerpineol 622 formate 622 Tetrahydronaphthylene diamine .... 511 resolution 113 Tetrahydropapaverine, resolution .... 114 Tetrahydroquinaldine 712 resolution . . . 1 13, 114 Tht-baine 704 hydrochloride 704 Thuja oil 666 Thujone ... 626 INDEX OF ACTIVE SUBSTANCES 751 Toluy Icarvoximes, o-, /-, and p- Toluyltartaric acid esters Trehalose Trehalum Triacetylshikimic acid . . Triamylaconitate Triamyltricarballylate . . Trichlorcamphor Triisobutyrylshikimic acid Trioxyglutaric potassium salt structure Tripropionylshikimic ac Triterpene Triticin Tropic acid, d- and /- . . resolution . Tropinic acid Turpentine oil dispersio rotation i: rotation c Turpethinic acid .... Tyrosin Valeraldehyde 509, 513 559 d- Valeric acid y, D 4 604 /-Valeric acid .--. 5M 607 Valeric acid esters ... 514 543 rotation . . . 290 508 resolution of ... .... 107 508 Valerion oil .... 666 643 Venetian turpentine .... 631 544 Viscose .... 609 567 Vitellin 727 t 567 Volemitol 513 61 544 Wheat albumin AiVoocl cru ni .... 725 611 659 Xylan 611 611 Xylitol 5" 522 structure 61 108 Xvlonic acid _^ 701 multirotation . . . 280 629, 631 strontium salt 545 663 Xylosazone 576 efficient. . . 155 Xylose 575 "erent liquids 177 multirotation 260 ixed oils . . 240 structure 60 667 Ylang-ylang oil 666 - 525 Zinc lactate . . . . Sl6. SI7 ERRATA. On page 60, half the structural formula for xylose and xylonic acid is missing. On page 684, read cinchonifine for cinchonidine. UNIVERSITY OF CALIFORNIA LIBRARY BERKELEY Return to desk from which borrowed This book is DUE on the last date stamped below. Mflf 25 V INTER-LI BRIARY LOAM 4'c .MAY 1 - I95t LI LIBRARY USE DEC LD 21-100m-9,'47(A5702sl6)476 REC'D LD DEC 2 7 1956 TER-LIBRARY LOAN MAY 2 5 1964 r 12jg 6SO ; WY 1 '69 -6 pf LOAN DSPT. Y.C VI535 IOZOZ C THE UNIVERSITY OF CALIFORNIA LIBRARY