^BBTffKffi^ 
 
 EX L1BR.IS 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
GRIFFIN'S STANDARD PUBLICATIONS. 
 
 In Two Vols., Large 8vo. With Illustrations. Sold separately. 
 
 CHEMISTRY FOR ENGINEERS AND 
 
 MANUFACTURERS, 
 A PRACTICAL TEXT-BOOK. 
 
 BY 
 
 BERTRAM BLOUNT, AND A. G. BLOXAM, 
 
 F.I.C., F.C.S., Assoc.Inst.C.E., F.I.C., F.C.S., 
 
 Consulting Chemist to the Crown Agents for the Consulting Chemist, Head of the Chemistry 
 
 Colonies. Department, Goldsmiths' Inst., New Cross. 
 
 " The authors have SUCCEEDED beyond all expectations, and have produced a work which should 
 give FRESH POWER to the Engineer and Manufacturer." The Times. 
 
 VOLUME I. Price 10s. 6d. 
 CHEMISTEY OF ENGINEERING, BUILDING, AND METALLURGY. 
 
 GENERAL CONTENTS. INTRODUCTION Chemistry of the Chief Materials of 
 Construction Sources of Energy Chemistry of Steam-raising Chemistry of Lubrica- 
 tion and Lubricants Metallurgical Processes used in the Winning and Manufacture 
 of Metals. 
 
 "PRACTICAL THROUGHOUT .... an ADMIRABLE TEXT-BOOK, useful not only to Students, but to 
 ENGINEERS and MANAGERS OF WORKS in PREVENTING WASTE and IMPROVING PROCESSES." Scotsman. 
 
 "EMINENTLY PRACTICAL." Glasgow Herald. 
 
 "Abook worthy of HIGH RANK .... its merit is great .... treatment of the subject of GASEOUS 
 
 FUEL particularly good WATER GAS and the production clearly worked out Altogether 
 
 a most creditable production. WE WARMLY RECOMMEND IT, and look forward with keen interest to 
 the appearance of Vol. II." Journal of Gas Lighting. 
 
 VOLUME II. Price 16s. 
 THE CHEMISTRY OF MANUFACTURING PROCESSES. 
 
 GENERAL CONTENTS. Sulphuric Acid Manufacture Manufacture of Alkali, etc. 
 Destructive Distillation Artificial Manure Manufacture Petroleum Lime and Cement 
 Clay Industries and Glass Sugar and Starch Brewing and Distilling Oils, Resins, 
 and Varnishes Soap and Candles Textiles and Bleaching Colouring Matters, Dyeing 
 and Printing Paper and Pasteboard Pigments and Paints Leather, Glue, and Size 
 Explosives and Matches Minor Chemical Manufactures. 
 
 " Certainly a GOOD and USEFUL BOOK, constituting a PRACTICAL GUIDE for students by affording a 
 clear conception of the numerous processes as a whole." Chemical Trade Journal. 
 
 " We CONFIDENTLY RECOMMEND this volume as a PRACTICAL, and not overloaded, TEXT-BOOK, of 
 GREAT VALUE to students. "The Builder. 
 
 In Two Volumes, Large 8vo. Strongly bound. 21s. each. 
 
 PHYSICO-CHEMICAL TABLES. 
 
 For the Use of Analysts, Physicists, Chemical Manufacturers, and Scientific Chemists. 
 Volume I. Chemical Engineering, Physical Chemistry. 
 Volume II. Chemical Physics, Pure and Analytical Chemistry. 
 
 * By JOHN CASTELL-EVANS, F.I.C., F.C.S., 
 
 Superintendent of the Chemical Laboratories, and Lecturer on Inorganic Chemistry and 
 Metallurgy at the Finsbury Technical College. 
 
 The Tables may almost claim to be exhaustive, and embody and collate all the :most recent data 
 established by experimentalists at home and abroad. The volumes will be found invaluable to all 
 engaged in research and experimental investigation in Chemistry and Physics. 
 
 The Work comprehends as far as possible ALL RULES AND TABLES required by the Analyst, Brewer, 
 Distiller, Acid and Alkali Manufacturer, etc., etc. ; and also the principal data in THERMO-CHEMISTRY, 
 ELECTRO-CHEMISTRY, and the various branches of CHEMICAL PHYSICS. Every possible care has been 
 taken to ensure perfect accuracy, and to include the results of the most recent investigations. 
 
 LONDON: CHARLES GRIFFIN & CO., LTD., EXETER STREET, STRAND. 
 
Weber 's India Rubber. 
 
 FRONTISPIECE. 
 
 FIG. 1. Structure of Insoluble Constituent of India Rubber. 
 Photo-micrograph : Apochrom. Zeiss 8 mm., Projec. Eyep. 4. x310. 
 
 FIG. 15. Dendritical Crystallisation induced in deposit of Sulphur Globules. 
 Photo-micrograph : Apochrom. Zeiss 4 mm., Projec. Eyep. 4. xGOO. 
 
 Charles Griffin & Co., Ltd. 
 
 To face Title. 
 
THE CHEMISTRY 
 
 OF 
 
 INDIA RUBBER; 
 
 INCLUDING THE OUTLINES OF A THEORY 
 ON VULCANISATION. 
 
 BY 
 
 CARL OTTO WEBER, PH.D. 
 
 ffour plates anD Several Illustrations in tbe Gejt. 
 
 OF THE 
 
 UNIVERSITY 
 
 OF 
 
 LONDON: 
 
 CHARLES GRIFFIN AND COMPANY, LIMITED, 
 
 EXETER STREET, STRAND, W.C. 
 
 1902. 
 
 [All Eights Reserved.] 
 
Defcfcatefc 
 
 TO THE 
 
 MEMORY OF DR. ROBERT HENRIQUES, 
 
 IN HIGH APPRECIATION OF HIS WORK 
 
 ON THE ANALYSIS 
 
 OF 
 
 INDIA RUBBER. 
 
 GENERAL 
 
PREFACE. 
 
 IN undertaking, at the request of the publishers, to write this book, the 
 chief inducement to me was the hope of arousing the interest of chemists 
 and technologists alike in the study of the scientific aspects and problems 
 of the remarkable class of substances known as india rubber, the very name 
 of which would seem to suggest that our present day knowledge of them 
 consisted chiefly in the fact that they may be used for erasing pencil 
 marks. 
 
 The intention in the first instance was to deal, in the following pages, 
 simply with the analytical methods, which are almost exclusively the work 
 of R. Henriques, more particularly as regards the rubber substitutes so- 
 called, and my own researches, chiefly concerning india rubber itself and the 
 vulcanisation problem. This refers to work done within the last ten years. 
 Before that time india rubber analysis, if it existed at all, was unknown to 
 the outside world. But I discovered very soon that a mere compilation of 
 analytical methods could at best only claim a very limited technical interest, 
 and only point the way to obtaining a number of statistical data respecting 
 india rubber and its manufactures, without affording any solid basis for 
 their subsequent interpretation. This might have sufficed for those who 
 simply want to ' do ' india rubber, but would have rendered the work almost 
 worthless to those who hereafter may feel inclined to turn their attention 
 to a field of inquiry equally fascinating in its chemical, physical, physio- 
 logical, and technical aspects. I resolved, therefore, to enlarge the scope of 
 the work by the introduction of a chapter on the "Chemistry of India 
 Rubber." The literature on this subject is extraordinarily scanty, and at the 
 same time, with perhaps two or three exceptions only, untrustworthy to a 
 degree. The writing of this chapter proved, therefore, a very arduous task, 
 and involved an amount of work altogether out of proportion to its 
 volume. 
 
 ^ n 1 "? vi ** 
 
VI PREFACE. 
 
 Incidentally I felt compelled to deal at some length with the nature of 
 the colloidal state itself, the phenomena of which have an important bearing 
 upon the chemical, and still more upon the technical, behaviour of india 
 rubber. Indeed, the fact that india rubber is a colloid may be said to 
 constitute the root difficulty of this subject to the chemical investigator and 
 the technologist alike. All these circumstances together led to the book 
 assuming the form of a monograph rather than of a text-book. 
 
 Of the many unavoidable gaps, of the many points I had to leave un- 
 decided or only half discussed, I am only too painfully aware, and in 
 excuse can only plead the limitations of one largely engaged in technical 
 work. Some of these omissions I hope to supply in the future ; many, I 
 trust, will be made good by the work of others. 
 
 Manufacturing processes as such have not been dealt with, as to have 
 done so would have resulted in the destruction of the unity and aim of the 
 work ; but I shall not be surprised to meet with the criticism from certain 
 quarters that, as a consequence of this restraint, the book is too theoretical. 
 I am content to let it be so. 
 
 The cradle of the india rubber industry, as of so many others, stands in 
 this country, which, from her former paramount position in almost every 
 one of them, has fallen, and is still falling, to the second or third place, or 
 no place at all a result owing in large part to the shortsightedness of the 
 manufacturers who conduct their undertakings simply on what they are 
 pleased to call 'practical lines,' which so often means that all forethought and 
 present research is sacrificed to the reaping of immediate profits, the future 
 being left to take care of itself entirely. The rude awakening which the 
 national industries are experiencing at the present moment has perhaps 
 naturally enough considering the state of mind of their leaders resulted 
 in a gross misconception of the nature of the industrial stimulus operating 
 abroad, precipitating the country into a feverish craze for stuffing the 
 heads of ignorant schoolboys with 'technical education.' 
 
 Salvation is to be found in the * technical education ' of the working 
 classes, but that the { masters ' stand infinitely more in need of education 
 than any other class of the community has scarcely been realised yet. 
 Until the nation, as a whole, recognises that the prosecution of scientific 
 study as a mere means of money-making is a profanation defeating its own 
 ends, the history of industrial development in England will afford the same 
 melancholy spectacle in this as in the last century, so-called technical 
 education notwithstanding. 
 
 As regards the india rubber industry, it is still very largely developing 
 
PREFACE. Vll 
 
 on empirical lines. In noue of the industrial countries have any organised 
 efforts been made to bring to bear upon the problems of this industry the 
 full resources of modern chemical and physical research. This is, no doubt, 
 also the reason why, notwithstanding the occasional claims of 'trading 
 puffs,' the efficiency of the rubber trade in Great Britain is in no way 
 inferior to that of either the United States, Germany, or France. But it 
 is of importance that manufacturers should clearly realise that india rubber, 
 and the industry connected with it, offer wide fields for scientific research, 
 the exploration of which has already commenced ; and the india rubber in- 
 dustry of the future will belong to whatever country may take the lead in 
 the scientific investigation and elaboration of its problems. The coal tar 
 colour industry has been lost to Great Britain through the indifference of 
 her statesmen, the shortsightedness of her manufacturers : is her india 
 rubber industry to suffer the same destiny 1 
 
 At the end of the book will be found a short chapter on the " Sanitary 
 Conditions in India Rubber Works," a matter of not inconsiderable import- 
 ance with which the chemist is much more competent to deal than the 
 engineer. 
 
 To my former assistant Dr. F. Betteridge I am indebted for help in con- 
 nection with some of the researches contained in this book, and to my sou 
 Mr. L. E. Weber for the preparation of the micro-slides and photo-micro- 
 graphs which illustrate certain sections. 
 
 THE AUTHOR. 
 
 CRUMPSALL, November 1, 1902. 
 
CONTENTS. 
 
 CHAPTER I. 
 
 PACK 
 
 THE CHEMISTRY OF INDIA RUBBER, . . . 1-115 
 
 Constituents of India Rubber The Physical Properties of India Rubber- 
 Action of Heat Chemical Behaviour The Vulcanisation of India 
 Rubber The Colloidal State Outlines of a Theory of Vulcanisation. 
 
 CHAPTER II. 
 EXAMINATION AND VALUATION OF GUTTA PERCHA AND INDIA RUBBER, . 116-129 
 
 Points to be Considered Loss on Washing Tables of the Origin, Character- 
 istics, etc., of the Chief Commercial Varieties of India Rubber. 
 
 CHAPTER III. 
 EXAMINATION OF INDIA RUBBER SUBSTITUTES, .... 130-159 
 
 Reclaimed or Recovered India Rubber Fatty and other Organic Constituents 
 Tables Resins, Pitches, and Waxes Chemical Analysis of Organic 
 Substitutes. 
 
 i 
 CHAPTER IV. 
 
 INORGANIC COMPOUNDING MATERIALS, ..... 160-175 
 
 Tables of Chemical Tests Tables of Physical Tests Vulcanising Compounds. 
 
 CHAPTER V. 
 
 VULCANISERS AND SULPHUR CARRIERS, . 176-187 
 
 Sulphur Chloride of Sulphur Lead Thiosulphate Antimony Pentasulphide 
 Zinc Sulphide and Lithopone. 
 
 CHAPTER VI. 
 INDIA RUBBER SOLVENTS, . . . . . 188-200 
 
 Naphtha Petroleum Spirit Shale Naphtha Turpentine Carbon Bisulphide. 
 
 I 
 
CONTENDS. 
 
 CHAPTER VII. 
 
 PAGE 
 
 COLOURING MATTERS, ....... 201-214 
 
 White Pigments Red and Brown Pigments Yellow Pigments Green 
 Pigments Blue and Violet Pigments Black Pigments Metallic Lustres 
 (Bronzes) Dyeing of India Rubber with Soluble Dyes. 
 
 CHAPTER VIII. 
 CONSTRUCTIVE COMPONENTS OF INDIA RUBBER ARTICLES, . . 215-223 
 
 Textile Fabrics - Determination of Shrinkage * Dressing ' Dyes Mordants 
 'Curling.' 
 
 CHAPTER IX. 
 ANALYSIS OF RUBBER ARTICLES, ...... 224-303 
 
 Physical Tests for Manufactured India Rubber Mechanical Tests Electrical 
 Tests Chemical Analysis Preparation of Samples Analytical Methods 
 Notes on the Analysis of Hard Rubber Goods (Ebonite) Interpretation of 
 the Results Examination of Colouring Matters in Hot Cured Goodfe 
 Examination of Colouring Matters in Cold Cured Goods Sulphur and 
 Chlorine Unusual Constituents Defects in India Rubber Goods. 
 
 APPENDIX. 
 
 SANITARY CONDITIONS IN INDIA RUBBER WORKS, . . . . . 304 
 
 INDEX, . . ...:... 309 
 
LIST OF ILLUSTRATIONS. 
 
 FIG. . PAGE 
 
 1. Structure of insoluble constituent of india rubber. Photo-micrograph: 
 
 Apochrom. Zeiss 8 mm., Projec. Eyep. 4. x 310, . . Frontispiece. 
 
 2. Diagram to KRAFFT'S theory of the colloidal state, . . . .67 
 
 3. Disposition of apparatus for the observation of the ' TYNDALL effect,' . . 73 
 
 4. Digestor for vulcanisation experiments, . . . . .84 
 
 5. Vulcanisation curves of Pani, ....... 86 
 
 6. Vulcanisation curves of Para, Congo, Beni River, Ceara, Borneo, at 125 C., . 87 
 
 7. Vulcanisation curves of Para, Congo, Beni River, Ceara, Borneo, at 135 C. , . 88 
 
 8. Micro-section of vulcanite (ebonite). Photo-micrograph : Apochrom. 
 
 Zeiss 4 mm., Projec. Eyep. 4. x 600, .... 
 
 9. Vulcanite powder. Photo-micrograph : Apochrom. Zeiss 4 mm., 
 
 Projec. Eyep. 4. x 600, ...... 
 
 10. Efflorescence upon india rubber goods. Photo-micrograph: Apochrom. 
 
 Zeiss 16 mm., Projec. Eyep. 4. x 150, .... 
 
 11. Globules of supervised (colloidal) sulphur. Photo - micrograph :' 
 
 Apochrom. Zeiss 4 mm. , Projec. Eyep. 4. x 600, . 
 
 12. Spontaneous crystallisation in deposit of sulphur globules. Photo- 
 
 micrograph : Apochrom. Zeiss 4 mm. , Projec. Eyep. 4. x 600, 
 
 13. Absorption of sulphur globules by sulphur crystal. Photo-micro- 
 
 graph : Apochrom. Zeiss 4 mm., Projec. Eyep. 4. x 600, . 
 
 14. Spread of crystallisation from globule to globule in sulphur deposit, . . ill 
 
 15. Dendritical crystallisation induced in deposit of sulphur globules. Photo- 
 
 micrograph : Apochrom. Zeiss 4 mm., Projec. Eyep. 4. x 600, Frontispiece. 
 
 16. Diagram of distilling flask for benzene (naphtha) test, . . . .192 
 
 17. Diagram of distillation apparatus for benzene (naphtha) test, . . .193 
 
 18. Cloth testing machine, . . . . . . . 216 
 
 19. Diagram showing disposition of apparatus for taking the galvanometer 
 
 constant, ..... ... 231 
 
 20. Diagram showing disposition of apparatus for determining insulation re- 
 
 sistances, ......... 232 
 
 21. Micrometer, ........ 234 
 
 22. Mixing machine (Laboratory scale), ...... 238 
 
 23. Apparatus for the extraction of india rubber samples, .... 240 
 
 24. Diagram of air bath for nitro-naphthalene extraction, . . 
 
 249 
 
 25. Diagram of air bath, construction of lid, . . , , / 
 
 26. Diagram : Tensile strength of differently vulcanised specimens of india rubber, 284 
 
UNIVERSITY 
 
 OF 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 CHAPTER I. 
 THE CHEMISTRY OF INDIA RUBBER. 
 
 Xo class of bodies offers such formidable manipulative difficulties to 
 the investigating chemist as the remarkable group of colloids comprised 
 under the collective name of india rubber. These difficulties are physical 
 rather than chemical that is to say, they do not so much consist in the 
 functional complexity of the molecules of india rubber, as in the circum- 
 stance that these molecules are, at any rate at present, known only with 
 the colloidal state superimposed upon them. Matter in this state does 
 not exhibit the abrupt changes of physical condition which, in crystalloids, 
 take the form of melting points, boiling points, and solution. The physical 
 changes induced in colloids by heat, or by solvents, are perfectly continuous 
 changes as long as the underlying chemical molecule or configuration 
 remains intact. As a consequence, the characterisation and identification 
 of the colloids and their derivatives, and their isolation and purification 
 from mixtures of them, offers frequently almost insuperable difficulties. 
 
 India rubber, as is well known, is the product of the coagulation of the 
 milky juices (emulsions) of a large number of trees, creepers, and shrubs, 
 comparable in this respect to the oils, and more particularly to the 
 terpenes. Accordingly, the commercial article can hardly be expected to 
 be a homogeneous, and still less a pure, product in the chemical sense. 
 Indeed, a cursory examination of a few samples at once reveals the presence 
 in them of water, sand, vegetable fragments, fibres, wood, and other more 
 or less accidental impurities. These may be removed by a washing pro- 
 cess, which, however, can only be satisfactorily carried out on a com- 
 mercial scale, strong machines being needed for the purpose. The water 
 also removes from the india rubber certain carbohydrates (sugars), the 
 presence of which in various brands was discovered by AIM GiRARD, 1 
 together with practically the whole of the above named mechanical im- 
 purities present. 
 
 1 Comptes Rcnclus, Ixvii. p. 820. 
 
 1 
 
INDIA RUBBER AND ITS ANALYSIS. 
 
 The carbohydrates thus removed are : inosite, bornesite, matezite, and 
 dambonite. 
 
 Inosite occurs in Gaboon rubber ; it is identical with Girard's dambose 
 (MAQUENNE). 1 According to its constitution, it is one of the numerous 
 stereo-isomeric hexa-oxy-hexa-methylenes : 
 
 H OH 
 
 V 
 C 
 
 H 
 
 HO. 
 
 H 
 
 OH 
 
 H 
 
 OH 
 
 = C 6 H 12 6 . 
 
 HO A H : 
 
 The inosite of Gaboon rubber is i-inosite (inactive). It is soluble in 
 water, alcohol, and acetic acid, and crystallises in fine needles (M.P. 225 C., 
 B.P. (invacuo)319 C.). 
 
 Bornesite, mono-methi/l-a-inosite C 6 H n 6 .CH 8 , occurs in Borneo rubber. 
 It crystallises in white rhombic prisms, readily soluble in water, little 
 soluble in alcohol. Its melting point lies at from 199 to 203 C. ; at 
 205 C. it sublimes almost without decomposition. It is optically active, 
 dextro-rotary (0^= +31 '6). 
 
 Matezite, which occurs in Madagascar rubber, is the mono-methyl 
 ether of i-inosite. M.P. 186 C. ; sublimes without decomposition at 
 200 C. Very soluble in water, sparingly soluble in strong alcohol, 
 insoluble in absolute alcohol. It is dextro-rotary (a D = 65'7). 
 
 Dambonite occurs in Gaboon rubber, and is the di-methyl ether of 
 i-inosite C 6 H 10 6 (CH 3 ) 2 . It crystallises, with three molecules of water, in 
 the form of white hexagonal prisms readily soluble in water and alcohol. 
 M.P. 195 C. ; sublimes without decomposition at 210 C. Dambonite is- 
 optically inactive. 
 
 These sugars seem never to be present in any larger quantity than 
 0'5 per cent., but the average percentage is rather lower. They are, of 
 course, completely removed from the rubber by the washing operation 
 above referred to. 
 
 The drying of the washed rubber is best performed in vacuo, to prevent 
 the otherwise unavoidable oxidation. The substance then forms techni- 
 cally pure rubber, but is by no means chemically pure, as solvents, such as 
 alcohol or acetone, readily extract from it certain oily and resinous 
 matters. The amount of these in the different brands of india rubber 
 varies enormously, from 1 to about 40 per cent. Nor is the quantity 
 1 Comptes Rendus, civ. p. 1853. 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
 constant in the same brand, the less so the greater the average amount 
 of resinous matter usual in any given brand. Thus, while the amount of 
 such matters in Para rubber varies from 1 to 2 per cent., it varies in 
 Lagos rubber from 3 '6 to 7 per cent., and in Borneo rubber from 6 to 21 
 per cent. The removal of these resinous and oily matters from the india 
 rubber is readily effected by extracting a weighed quantity of the india 
 rubber with acetone in a Soxhlet extractor. For this purpose it is, how- 
 ever, essential to use the rubber in the form of the highly porous washed 
 sheets as they are obtained in rubber factories. If cut pieces of crude 
 rubber are used, the complete extraction becomes an extremely tedious 
 operation. The following tables show the percentages of extract obtained 
 from various brands of india rubber :- 
 
 
 PERCENTAGE OF RESINOUS EXTRACT. 
 
 -p T - TJ 
 
 
 .DRAND OF INDIA JxUBBER. 
 
 Terry. 1 
 
 Weber. 2 
 
 Henriques. 3 
 
 Para, fine, .... 
 Ceara Scraps, 
 
 1 '2 per cent. 
 1-3 
 
 1 '3 per cent. 
 2-1 
 
 1 -3 per cent. 
 2-1 
 
 Columbia, .... 
 
 2-5 
 
 
 3-8 
 
 
 5'0 
 
 
 Mozambique, 
 
 3-0 
 
 3-2 
 
 
 3-8 
 
 
 Manoh Twist, 
 
 
 2-9 
 
 
 5-8 
 
 
 
 
 4-4 
 
 
 4 '5 
 
 
 Rio Janeiro, . . . - 5 '8 
 
 
 5-2 
 
 
 2-5 
 
 
 Madagascar, . . . 6 '4 
 
 8-2 
 
 
 7'1 
 
 
 Sierra Leone, red, . . 7*4 
 
 
 97 
 
 
 5'8 
 
 
 Borneo, . . . . j- 7'9 
 
 
 117 
 
 
 10-0 
 
 
 Assam, . . . . j 9*3 
 
 
 11-3 
 
 
 11-3 . 
 
 
 Mangabeira, . . . j 10 '5 
 
 
 13 1 
 
 
 5-8 
 
 
 African Ball, I., . .- . 18 '5 
 
 
 22'8 
 
 
 
 II., . . . 22-8 
 
 
 26-1 
 
 
 . 
 
 African Flake, . . . 41 '2 
 
 
 63-9 
 
 
 
 Balata, .... 
 
 8'4 
 
 
 ... 
 
 Gutta Percha (Penang), 
 
 9-3 
 
 
 ... 
 
 
 
 i 
 
 How far these figures are correct depends, of course, upon the efficiency 
 of the solvent employed i.e., its ability to dissolve all such resinous 
 bodies as may occur in india rubber. As regards this point, it is worth 
 while to draw attention to the fact that, of the large class of resins proper, 
 the great majority are either insoluble, or only partially soluble in acetone, 
 so that the acetone might leave undissolved certain resinous bodies 
 occurring in india rubber. 
 
 Indeed, in Sierra Leone rubber, both red and white, I have ascertained 
 the presence of a considerable quantity of a resin, in its appearance closely 
 resembling crude albane as obtained from gutta percha, which is only 
 very sparingly soluble in even boiling acetone. The complete extraction 
 of this resin from the rubber requires days, and it is not impossible that 
 
 1 Journ. Soc. Chem. Ind., 1892, p. 970. 
 
 3 Der Kautschuk und seine Qutllen, Dresden, 1899. 
 
 2 Ibid., 1894, p. 477. 
 
4 INDIA RUBBER AND ITS ANALYSIS. 
 
 other brands of rubber may contain similar resins, so that the above 
 figures might be open to very considerable correction. 
 
 No such relations as obviously exist between gutta percha and the 
 resins invariably accompanying it have been shown to exist between india 
 rubber and its native resinous impurities. 
 
 The nature of the resins extracted from various brands of india rubber 
 varies very greatly. In some cases a distinctly oily extract is obtained, in 
 others treacly masses, in others resins similar in appearance to colophony, 
 or elemi resin, and in a few cases white powdery products, in appearance 
 resembling the crude albane of gutta percha, but in every case differing 
 from it in composition. Respecting the chemical character of these extracts 
 nothing at all is known. 
 
 These resinous impurities determine to a considerable degree the 
 behaviour, especially of the inferior brands, of india rubber in practical 
 working, and they are also supposed to have a marked influence upon the 
 stability of the finished article. The technical effect due to the presence 
 of these resins must, no doubt, be ascribed to their decided action upon the 
 sulphur in the vulcanising process ; but seeing that the presence of these 
 resins in crude india rubber does not appear to affect its stability in an 
 appreciable degree, it is difficult to understand why they should act 
 detrimentally after vulcanisation a process which is generally admitted, 
 and rightly so, to enhance its chemical indifference. The explanation of 
 this mistaken supposition lies probably in this, that undervulcanised india 
 rubber, or india rubber vulcanised with an insufficient amount of sulphur, 
 is well known to be highly perishable. Now, it is evident that in vulcanis- 
 ing india rubber either without taking into account the amount of resin 
 present, or underrating that amount, more or less of the sulphur intended 
 for the vulcanisation of the india rubber is taken up by the resin. The 
 result is an undervulcanised, and therefore unstable, vulcanisation product. 
 Such a product may still contain free sulphur, but not sufficient to induce 
 vulcanisation at the temperature employed. This, of course, might be 
 rectified by increasing the temperature of vulcanisation, but such increase 
 of temperature would equally favour the combination of the residual 
 sulphur with the resin, so that the final result would remain essentially 
 unaffected. 
 
 A thorough study of these ' resins ' will very likely enable us in the 
 future to determine in manufactured goods the chief kinds of india rubber 
 employed. At the present time this is impossible, although it is quite 
 within our power to determine whether a given article has been manu- 
 factured from a high, medium, or low grade rubber. 
 
 It is only to be expected that crude rubber, besides the above named 
 carbohydrates, should also contain albuminous matters (proteids) which 
 invariably form one of the constituents of the latex of the india rubber 
 tree. These proteids reappear in the india rubber if the latex coagulation 
 is effected by heat, chemical means, or desiccation, the methods which are, 
 
THE CHEMISTRY OF INDIA RUBBER. 5 
 
 at least at present, employed in the preparation of the different brands of 
 India rubber. If BIFFEN'S T centrifugal method were generally adopted, 
 this impurity would be almost completely avoided. 
 
 The percentage of these proteids in the rubber is small, but, as to its 
 exact amount, we possess no reliable data. SEELIGMANN 2 quotes the 
 composition of the latex as : 
 
 India rubber, . . . . . . .32 
 
 Proteid and mineral matter, . .12 
 
 Water, ...... .50 
 
 These data are altogether too vague. FARADAY gives the following 
 analysis of the latex of Hevea braziliensis : 
 
 India rubber, ..... 3770 per cent. 
 
 Albumen, . . . . . . 1'90 ,, 
 
 Bitter nitrogenous substance soluble in water and 
 
 alcohol, ...... 7*13 ,, 
 
 Substance soluble in water, insoluble in alcohol, . 2 '90 ,, 
 
 Wax, ...... 013 ,, 
 
 Water, slightly acid, . . . . 50 '37 
 
 From these figures it would follow that Para rubber would contain 
 about 4*3 per cent, of albuminous matter, but I consider the presence of 
 such an amount highly questionable, although LASCELLES-SCOTT 3 states it 
 even higher than this in a latex of unnamed origin : 
 
 India rubber, ..... 37*13 per cent. 
 
 Albumen, . 271 ,, 
 
 Resins, ...... 3'44 ,, 
 
 Essential oils, ... . Traces. 
 
 Sugar, . 4-17 
 
 Mineral matter, . . . . . 0'23 ,, 
 
 Water, . . . . . 52 '32 
 
 According to these figures the percentage of proteids in the dry rubber 
 would amount to about 5 '6 per cent. The absence of information respecting 
 the origin of the above milk is to be regretted, as its very high percentage 
 of sugar and its quite unusually low percentage of mineral matter alike 
 render it very remarkable. 
 
 On the large scale the removal of these proteids from the commercial 
 india rubber is a practical impossibility. On the small scale they can be 
 eliminated by dissolving the india rubber and allowing the dilute solution 
 to stand, when the insoluble coagulated proteids gradually settle out. A 
 somewhat more expeditious method is the centrifuging of the solutions. 
 From these solutions the india rubber is precipitated by alcohol, and the 
 product thus obtained, when dry, appears to have always been looked 
 upon as chemically pure india rubber. At any rate, all known analyses of 
 india rubber, with the exception of those made by GLADSTONE and 
 
 1 Ann. of Botany, xii., 1898, p. 165. 
 
 2 Le Caoutchouc et la Outta Percha, Paris, 1896, p. 55. 
 
 3 India JKubber Joum. t v. No. 11, 1889. 
 
INDIA RUBBER AND ITS ANALYSIS. 
 
 HiBBERT, 1 have been made with products freed from the various impurities 
 named above. The last named were the first to clearly recognise the 
 presence of comparatively small and varying proportions of oxygen in 
 India rubber, and they also observed that by fractional precipitation the 
 percentage of oxygen can be greatly reduced, although its complete elimina- 
 tion was found impossible. 
 
 The elementary composition of india rubber as determined by the 
 -earlier workers up to 1888 has been compiled in the following tables : 
 
 i 1 
 
 | Faraday, i Ure. \ Williams. 
 
 i ! ! 
 
 Berzelius. 
 
 Adrian! 
 
 Gladstone 
 and 
 Hibbert. 
 
 Calculated 
 for 
 c ioHi 6 - 
 
 i j I 
 Hydrogen, > 12'8 10 '0 121 
 
 12-5 
 
 11-94 
 
 12-00 
 
 11-76 
 
 Carbon, . 87 '2 90 '0 87 '3 
 
 ' :-.;* i 
 
 87'5 
 
 87-91 87-46 
 
 88-24 
 
 It will be seen that these data agree fairly well with the figures 
 calculated for a compound of the terpene type C 10 H 16 , and this formula 
 is now generally held to express the quantitative composition of 
 india rubber, although, in view of the impossibility of producing an 
 absolutely oxygen-free product, HENRIQUES within recent years pointed out 
 the possibility that india rubber might after all be found to be an oxygen 
 compound. 
 
 PAYEN was the first to observe that on dissolving india rubber the 
 greater part of it passed into solution, but that there remained always an 
 insoluble constituent, consisting of very voluminous and gelatinous flakes, 
 which, under the microscope, presented a peculiar reticulated appearance. 
 For some extraordinary reason these two constituents appear never to 
 have been analysed separately. By common consent the insoluble con- 
 stituent was looked upon as an ' insoluble modification ' of india rubber. 
 GLADSTONE and HIBBERT 2 express themselves on this point as follows : 
 
 " We are inclined to believe that the less soluble modification of the 
 hydrocarbon which has been frequently observed is produced during the 
 drying of the juice, for we find that if the more soluble part be heated, it 
 is more or less changed, and less susceptible of subsequent solution. The 
 change increases as the temperature is raised, and also with the length of 
 time during which the heat continues. This will account for the very 
 varied results obtained by different experimenters as to the proportions of 
 soluble and insoluble caoutchouc." 
 
 Even more definite are SEELIGMANN'S 3 statements respecting the nature 
 and relationship of these two constituents : 
 
 "Le caoutchouc est, en effet, forme de deux substances isomeriques, 
 dont 1'une, solide et elastique, resiste a presque tons les agents ; tandis 
 
 1 Journ. Chem. Soc., 1888, p. 680. 2 Ibid., 1888, p. 679. 
 
 3 Le Caoutchouc et la Gutta Percha, Paris, 1896. 
 
THE CHEMISTRY OF INDIA RUBBER. 7 
 
 que 1'autre, demi liquide et poisseuse, est beaucoup plus facilement attaquee 
 et dissoute. C'est & cette seconde matiere que la gomme doit la propriete 
 de se souder a elle-ra^me lorsqu'on souinet a une forte compression ses 
 surfaces recemment coupees. On donne a la premiere matiere le nom de 
 nerveuse et a la seconde le nom d'adhesive. Si, par une dissolution approprie, 
 on separe ces deux principes immediats isomeres, aucun d'eux ne garde 
 les propriete's elastiques et extensibles au mime degre ; il semblerait, dit 
 Payen, que 1'adherence entre les filaments, lubrifies par une matiere grasse 
 et assouplie par la portion soluble et molle, ait ete partiellement detruite." 
 
 It will be observed that SEELIGMANN describes the two constituents in 
 question as isomers, and that he, following PAYEN in this respect, ascribes 
 to the insoluble constituent important physical functions. In this he 
 appears to have been influenced by the statements of various authors, who 
 found the percentage of the insoluble constituent or modification of india 
 rubber to vary between 30 and 70 per cent., or an average of 50 per cent. 
 GLADSTONE and HIBBERT, however, found only 4 per cent, of this insoluble 
 body in Para rubber. 
 
 In view of this very large and important discrepancy, I reinvestigated 
 this subject. 1 The inner parts of a loaf of crude Para 2 were split into 
 very fine films. As these contain a considerable proportion of water, they 
 were, to begin with, subjected to an often repeated treatment with acetone, 
 during which they (originally cream-coloured films) assumed a pronounced 
 tan colour. Finally, they were completely extracted with acetone in a 
 Soxhlet extractor, and then, without drying, transferred into chloroform. 
 After a little over a fortnight, during which time the bottle containing 
 the solution was frequently shaken, a solution was obtained consisting of 
 two layers. The lower of these was a bright clear solution, while the 
 upper consisted of small clots or shreds of the insoluble part of the rubber. 
 This upper layer, which appeared to represent a very substantial pro- 
 portion of the rubber used, was removed, and a number of times treated 
 with fresh quantities of chloroform, two or three days intervening between 
 each of these changes of solvent. This treatment was continued until the 
 chloroform extracted no longer any perceptible quantity of soluble matter 
 from the flocculent jelly, the apparent bulk of which very greatly decreased 
 during this treatment. 3 Finally, the product so purified was dried in 
 
 1 Journ. Soc. Chem. Ind., 1900. 
 
 2 Washed rubber which has been sheeted in a ' rubber washer ' cannot be used, as in 
 this operation the flakes of the insoluble constituent are so broken up and comminuted, 
 that the whole afterwards forms an apparently complete, homogeneous solution (rubber 
 varnish). 
 
 3 On a rough estimation this decrease amounts to quite five-sixths of the original 
 apparent bulk. It might, perhaps, be thought to be due to a gradual slow solution of the 
 supposed insoluble constituent, but this explanation is certainly incorrect. I suggest 
 that the phenomenon in question is due to the india rubber solution acting like a solvent 
 upon the insoluble constituent, the dissolving action of which is the greater the higher 
 the concentration of the india rubber solution. As by successive washings with chloro- 
 form the concentration of the rubber solution rapidly diminishes, the insoluble constituent, 
 
8 INDIA RUBBER AND ITS ANALYSIS. 
 
 vacuo, when it was obtained as a light-brown stringy mass, the bulk of 
 which was almost infinitesimal compared with the volume it occupied 
 during its preparation. In various operations the weight of the insoluble 
 constituent obtained varied from 3 to 6'5 per cent., the average being 
 about 3 '5 per cent. 
 
 In india rubber solvents this substance swells to an enormous extent 
 without any appreciable quantity passing into solution. The fibrous 
 structure which thin sections of it exhibit under the microscope is not, as 
 has been suggested, comparable with that of certain elementary animal or 
 vegetable tissues, but appears to be the result of a peculiar process of shrink- 
 ing of the drying substance, producing a reticulated appearance closely 
 resembling that of some of the artificial colloidal structures described 
 by BuTSCHLi. 1 (See Frontispiece.) 
 
 This insoluble body, even when dilated by solvents, is entirely free from 
 stickiness, and possesses no trace of that adhesiveness, or capability of 
 interfusion, so characteristic of india rubber. When dry, it shows a 
 remarkable degree of toughness, but very moderate distensibility. On 
 analysis, its composition was found to correspond very closely to that of a 
 body expressed by the formula C 30 H 68 10 2 . 
 
 On the other hand, the analysis of the soluble part of india rubber 
 without any further purification led to figures which revealed the presence 
 of still nearly 2 per cent, of oxygen : 
 
 Found. Calculated for C 10 H 16 . 
 C, . . . 86-51 88-23 
 
 H, . . . 11-59 11-76 
 
 0, . . . 1-89 
 
 By redissolving this product in chloroform, diluting the solution so 
 that it contains about O'l per cent, of india rubber, submitting this solution 
 to a very careful fractional precipitation with alcohol, and rejecting the 
 precipitate first obtained, a product of a high degree of purity resulted. 
 The analysis of this product led to the figures : 
 
 Found. Calculated for C 10 H ]6 . 
 
 C, . . 87 '91 88-23 
 
 H, . . . 11-86 11-76 
 
 These agree very satisfactorily with the composition of a compound 
 
 C 10 H 16' 
 
 The two results are of extreme interest, and, to begin with, enable us 
 to rectify the erroneous statement that the soluble and the insoluble con- 
 being deprived of its proper solvent, or swelling agent, contracts and becomes increasingly 
 less in bulk, though not in weight. All those who have had a fair amount of experience 
 regarding the behaviour of colloids in solution will have observed cases where a colloid, 
 insoluble in a certain solvent, becomes soluble in the solution of another colloid in the 
 same solvent. 
 
 1 BUTSCHLI, j Ueber Strutturen, Leipzig, 1898. This book should be in the hands of 
 every student of the problems of india rubber and colloids generally. 
 
 2 Journ. Soc. Chem. Ind., 1900. 
 
THE CHEMISTRY OF INDIA RUBBER. 9 
 
 stituents of India rubber are identical in their composition (isomeric). Our 
 data further enable us to refute as incorrect the exaggerated figures quoted 
 by various writers respecting the proportions of the soluble, and of the 
 insoluble, constituents of india rubber ; and it is, at the same time, quite 
 evident that the proportion of the latter is on the average so small that it 
 would be utterly futile to ascribe to it any important, or even distinctly 
 traceable, function as regards the technical aspect of the physical properties 
 of india rubber. We further see that GLADSTONE and HIBBERT'S attempt 
 to explain the presence of the insoluble part of india rubber as due to the 
 action of heat upon the latter can no longer be accepted, although their 
 statement regarding the general effect of heat upon india rubber is, no 
 doubt, true. 
 
 The most remarkable fact concerning this insoluble portion is its high 
 percentage of hydrogen, and the very large amount of oxygen it contains. 
 By some kind of common consent the oxygen which is invariably present 
 in technically pure india rubber has been ascribed to the oxidation of the 
 india rubber by atmospheric oxygen. This notion is no longer tenable, as 
 our above figures conclusively demonstrate that at any rate the bulk of the 
 oxygen observed in india rubber is present in the form of the compound 
 30 H 68 10 , which, whatever it be, can scarcely be described as an oxidation 
 product of india rubber. It appears to me that there exists an important 
 genetic relationship between this compound and india rubber on the one 
 hand, and the inosite, or inosite derivatives so commonly accompanying 
 india rubber, on the other hand. This becomes, perhaps, even clearer by 
 observing that 
 
 OwHwO,. = 3(C 10 H 1C ) + 10H 2 0. 
 
 Ill brief, I suggest that this insoluble compound is a link between india 
 rubber and the complex carbohydrates, the celluloses in particular, which 
 I am inclined to consider the raw material from which the plant produces 
 all the terpenes, including india rubber. 
 
 There is, on the other hand, no doubt whatever that india rubber 
 'absorbs' oxygen when exposed to the atmosphere, and this oxygen 
 absorption finally always results in the conversion of the india rubber into 
 a brittle resinous body, generally described as SPILLER'S resin. SPILLER 
 has published an analysis of this body, 1 and gives the following figures: 
 
 C, 64-00 
 
 H, , . . . . 8-46 
 
 0, ... -. 27'54 
 
 It is interesting to note that these figures very exactly agree with the 
 composition of a body of the formula C 30 H 48 10 , the relation of which to 
 our insoluble constituent C 30 H 68 10 is obviously significant. Equally 
 interesting to observe is the fact, immediately deducible from the above 
 given percentage composition of SPILLER'S resin, that this oxidation of india 
 
 1 Journ. Chem* Soc., vol. iii. (new series), 1865, p. 44. 
 
IO INDIA RUBBER AND ITS ANALYSIS, 
 
 rubber consists purely and simply in the addition of oxygen to the 
 unsaturated india rubber molecule, and that consequently the carbon- 
 hydrogen ratio of india rubber 10 : 16 remains unaffected in this 
 process. 
 
 If we now examine the analytical figures, above given, of the soluble 
 constituent of india rubber, before its further purification by fractional 
 precipitation, we find at once that in this case also the carbon-hydrogen 
 ratio is that of 10 : 16. From this we may infer that the oxygen 
 percentage of india rubber in its resin free state of purity is due to 
 two factors, one of which is the presence of the insoluble constituent 
 C 30 H 68 10 , the other the formation of oxygen addition products of india 
 rubber. 
 
 The insoluble constituent does not seem to be present in all rubbers, 
 but it appears that whenever it occurs it does so in rather small or in- 
 finitesimal quantities only. Whether its composition in the various brands 
 of india rubber is the same as in Para has not yet been ascertained, and will 
 be a highly interesting subject for further enquiry. The statement I made, 
 in the paper in which I first announced the composition of the insoluble 
 constituent of Para rubber, 1 that no corresponding body was to be found 
 in any of the other varieties of india rubber, I can no longer maintain. 
 The failure to find it in such rubbers as Lagos, Congo ball, Borneo, and 
 Assam rubber was simply due to the fact that the examination was made 
 on solutions obtained from these rubbers after they had been subjected to 
 the process of washing in the roller mills generally used for this process. 
 In this operation the insoluble body is disintegrated to such a degree that 
 it no longer separates out from the solutions containing it, nor can it be 
 isolated from them by filtration. It is there, nevertheless, but can only be 
 obtained from solutions of the crude rubber. A preliminary examination 
 of a large variety of commercial brands of india rubber in the crude state 
 revealed the fact that nearly all appear to contain small but appreciable 
 quantities of similar bodies. As showing again how comparatively 
 insignificant is the amount of this insoluble body, I may already now call 
 the attention of the reader to the tables below, which demonstrate the fact 
 that the carbon-hydrogen ratio is for all varieties of india rubber practically 
 a constant, very closely approximating the ratio 10 : 16. Considering 
 the composition of the insoluble body, this is only possible if its quantity 
 in every case is very small. 
 
 Already in the discussion of the composition of SPILLER'S resin it has 
 been pointed out that this product is formed by the direct addition of 
 oxygen on india rubber. We have seen further that the oxygen present 
 in the soluble part of india rubber is not due to the presence, in a very 
 finely subdivided state, of some of this insoluble constituent, but that the 
 oxygen is in combination with the rubber itself. In this case likewise 
 the constancy of the carbon-hydrogen ratio justifies the conclusion that 
 1 " On the Nature of India Rubber," Journ. Soc. Chem. Ind., 1900. 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
 II 
 
 this oxidation product is merely an addition product of india rubber and 
 oxygen. 
 
 The following table shows that this is not only true of Para rubber 
 itself, but of all varieties of india rubber. 
 
 VARIETY. 
 
 C. 
 Per Cent. 
 
 H. 
 
 Per Cent. 
 
 0. 
 Per Cent. 
 100 -(C + H). 
 
 i 
 C. : H. 
 
 Para, 1 . 
 
 87-87 
 
 11-51 
 
 0-61 
 
 10 : 157 
 
 Congo Ball, 2 . 
 
 85-15 
 
 10-82 
 
 4-02 
 
 10 : 15-2 
 
 
 
 86-41 
 
 11-07 
 
 2-51 
 
 10 : 15-3 
 
 Borneo, 2 
 
 86-63 
 
 10-83 
 
 2-53 
 
 10 : 15-0 
 
 i 
 
 8776 
 
 11-23 
 
 1-00 
 
 10 :15'3 
 
 Ceara,- . 
 
 83-21 
 
 11-41 
 
 5-37 
 
 10 : 16-4 
 
 
 
 86-20 
 
 11-56 
 
 223 
 
 10 : 16-0 
 
 Lagos,- . 
 
 86-11 
 
 11-54 
 
 2-34 
 
 10 : 16-0 
 
 Sierra Leone, 2 
 
 86-38 
 
 11-34 
 
 2-27 
 
 10 : 157 
 
 Assam,- . 
 
 87-17 
 
 11-47 
 
 1-35 
 
 10 : 15-8 
 
 Mangabeira, 2 . 
 
 87-08 
 
 11-39 
 
 1-52 
 
 10 : 157 
 
 Cameta, 2 
 
 85-41 
 
 11-09 
 
 3-49 
 
 10 : 15-5 
 
 Santos Sheet, 2 
 
 83-56 
 
 10 90 
 
 5-35 
 
 10 : 15-6 
 
 Bahia. 1 . 
 
 
 
 
 
 Beni River, 2 . 
 
 85 ; 55 
 
 11-46 
 
 2-98 
 
 10 V 16-1 
 
 Madagascar (Pinky), 2 
 Balata, 2 . 
 
 85-29 
 85-69 
 
 11-05 
 11-45 
 
 3-65 
 
 2-85 
 
 10 : 15-5 
 10 : 16-2 
 
 j 
 
 86-34 
 
 11-69 
 
 1-97 
 
 10 : 16-2 
 
 Gutta Percha, 2 
 
 84-20 
 
 11-35 
 
 4-44 
 
 10 : 16 
 
 i 
 
 88-01 
 
 11-78 
 
 0-21 
 
 10 : 16 
 
 1 Purified by extracting with acetone, dissolving in large volume of chloroform, and 
 fractionately precipitating with absolute alcohol ; first fraction rejected. 
 
 2 Purified only by thorough extraction with acetone. 
 
 These figures show clearly that a hydrocarbon of the formula C 10 H 1(3 
 is the greatly preponderating constituent of all varieties of india rubber. 
 The impossibility of completely removing all the oxygen from india rubber 
 is of very little weight as an objection against this conclusion, seeing that 
 every kind of india rubber subjected to a number of successive purifying 
 treatments is obtained with an increasingly smaller percentage of oxygen 
 until it reaches practically a vanishing point. 1 
 
 The oxygen addition products contained in india rubber do not appear 
 to sensibly affect the chemical behaviour of the whole, at least not so 
 long as the percentages of oxygen so present remain within normal limits. 
 This applies as much to the vulcanisation process of india rubber, which, 
 as a chemical reaction, is rather badly defined, as to the derivatives we 
 shall subsequently discuss. But this, of course, does not refer to the 
 technically pure india rubber, which still contains the insoluble constituent 
 and the ' oily and resinous matters.' 
 
 Considering the chemically inert nature of these impurities, it is not to 
 be wondered at that their influence upon the chemical properties of india 
 
 1 It is rather remarkable to observe that there appears not to exist any recognisable 
 relation between the amount of resinous extract yielded by a sample of india rubber 
 and its percentage of oxygen. 
 
12 INDIA RUBBER AND ITS ANALYSIS. 
 
 rubber in its technically pure form is but slight, unless, indeed, the percent- 
 ages of resins present become high when their influence upon the process 
 of vulcanisation becomes correspondingly marked. There is, however, 
 scarcely any doubt that the presence of what, so far, in the absence of any 
 more definite knowledge, we had to term the 'insoluble constituent' of 
 india rubber, must, to some extent, affect the behaviour of india rubber 
 against certain chemical agents; and I am inclined to believe that 
 certain irregularities observed in the processes known as the vulcanisa- 
 tion of india rubber are largely due to the presence of this insoluble 
 constituent. 
 
 There remains only one impurity of india rubber to be mentioned, the 
 complete removal of which is j ust as difficult, if not even more so, than that 
 of the oxygen addition products. This impurity in the technically pure 
 india rubber is generally described as ash, and is the mineral residue 
 remaining after incineration. This impurity probably may in small part 
 represent mineral matter in an extremely fine state of subdivision which 
 the washing process has failed to remove, but a large part of it is un- 
 doubtedly present in the form of organic salts. This conclusion follows 
 from the fact that on extracting india rubber with acetone, filtering the 
 solution so obtained, evaporating to dryness, and incinerating the residue, 
 a certain amount of ' ash ' (in some cases rather considerable quantities) 
 is invariably obtained. It is obvious that the filtered acetone solution can 
 contain this mineral matter only in the form of organic salts. Indeed, I 
 have occasionally obtained extracts from which small quantities of crystals 
 could be obtained which consisted evidently of such salts, but I never could 
 procure them in quantities sufficient for analysis. According to HENRiQUES 1 
 such salts are obtained as sediments in solutions of technically pure india 
 rubber, in chloroform, and similar solvents. That these salts possess prop- 
 erties similar to india rubber, as that author suggests, appears to me 
 highly improbable. The fact of their settling out from india rubber solutions 
 points entirely against this assumption. They are most probably salts of 
 resin acids, and I observed their melting points to be very low, generally 
 much less than 100 C. As all india rubber, before being washed on the 
 rollers, is subjected to treatment with boiling water, these salts would 
 naturally interfuse with the india rubber and the resinous constituents in 
 it ; and as they are, moreover, insoluble in water, they naturally remain 
 behind in the washed india rubber. 
 
 The mineral bases present consist of lime, magnesia, alumina, and 
 iron, the latter, according to HBNBIQUES, in the African varieties of india 
 rubber being present in the form of ferrous salts. The amount in which 
 these bases occur in india rubber is remarkably constant in all brands. 
 Indeed, the percentage of ash to be observed in the various qualities 
 of india rubber is subject to much less variation than any of its other 
 constituents. 
 
 1 Der KautscliuTc und seine Quellen, Dresden, 1899. 
 
Para, fine, 
 
 0'30 per cent. 
 
 ; Peruvian Ball, 
 
 Mollendo, . 
 
 . 0-50 
 
 Congo Ball, 
 
 Santos Sheet, . 
 
 . 0-16 
 
 Mangabeira, 
 
 Cameta, 
 
 . 0-40 
 
 Lagos, 
 
 Ceara (Machado), 
 
 . 2-80 
 
 Assam, 
 
 Beni River, 
 
 . 2-50 
 
 Borneo, fine, 
 
 THE CHEMISTRY OF INDIA RUBBER. 13 
 
 The following table gives the percentages of ash observed in a number 
 of commercial brands of india rubber after washing : 
 
 1 '40 per cent. 
 . 0-82 
 . 2-42 
 . 0-54 
 . 072 
 . 0'80 
 
 All the figures above quoted are averages calculated in every case from 
 a large number of individual determinations. Of course, the presence of 
 these impurities cannot at present be employed for the chemical identifica- 
 tion of the various brands, although this might become possible as the 
 result of an exhaustive investigation of the quantitative composition of 
 the ash furnished by different brands. On this point our information is at 
 present extremely scanty. HENRIQUES l states that lime predominates in 
 Para rubber, ferrous iron in African rubbers, while the presence of 
 magnesia appears to be characteristic of Ceara rubber. 
 
 \ 
 
 The Physical Properties of India Rubber. 
 
 India rubber in the highest state of chemical purity is a practically 
 colourless substance, the specific gravity of which is 0'911 at 17C., but 
 the technically pure article shows a specific gravity varying from - 915 
 (Santos sheet) to 0'931 (Machado). Much higher figures 0*942, 0'966 
 have been quoted, but I have never been able to obtain specimens of such 
 products. 
 
 In water india rubber is insoluble, but it absorbs about 25 per cent. 
 on prolonged immersion, swelling considerably during this absorption. 
 HANCOCK already observed that water diffuses through india rubber, and 
 the experiment by which he demonstrated this fact is characteristic of 
 the laborious perseverance and patience of this worker. His experiment 
 is best given in his own words : " I suspected rubber was not entirely 
 impervious to water, and I had the curiosity to make a bag of canvas 
 lined with sheet rubber, which I filled with water and sealed hermetically." 
 The following are the figures obtained in this test : 
 
 Ibs. ozs. drs. 
 
 The weight on October 21, 1826, was 114 
 25, 1827, ,112 
 
 2, 1835, 
 
 November 1844, 
 October 1849, 
 
 February 1851, 
 
 May 1854, 
 
 100 
 14 12 
 13 4 
 7 8 
 3 14 
 
 Two years later i.e., after the lapse of thirty years the bag was cut 
 open, when it was found to be dry, having originally contained a little 
 over 12 ozs. of water. 
 
 1 Der Kaidschuk utid seine Quellen, Dresden, 1899. 
 
14 INDIA RUBBER AND ITS ANALYSIS. 
 
 The amount of water absorbed by India rubber, although generally 
 accepted to amount to approximately 25 per cent., has been found by 
 various experimenters to vary considerably without, it is stated, any 
 apparent reason. In this relation it should be pointed out that the 
 absorption of water by india rubber depends very greatly upon the pro- 
 portion of * oily and resinous ' matter contained in it. Indeed, by mixing 
 india rubber with oils, resins, or waxes, its absorptive capability for water 
 at once falls very considerably. Conversely, it increases by freeing the 
 rubber of its resinous impurities by extraction with acetone. 
 
 The water absorption of vulcanised rubber is extremely small, certainly 
 not large enough to appreciably affect the insulation of a rubber cable 
 after five years' continuous immersion. 
 
 With alcohol india rubber behaves much as with water, the quantity 
 absorbed averaging about 20 per cent. 
 
 Acetone swells the india rubber even more than alcohol, but without 
 dissolving any trace of it. 
 
 All the organic acids slightly swell the india rubber, but are otherwise 
 without any perceptible action. 
 
 Esters and ethers produce a swelling of the india rubber very much in 
 excess of the similar action produced by any of the previously named 
 solvents. This applies more particularly to ethyl ether, which is often,, 
 though erroneously, stated to be a solvent for india rubber. 
 
 The most important solvents of india rubber, approximately in the 
 order of their rate of efficiency, are turpentine (pinene), dipentene, petroleum 
 spirit, carbon bisulphide, benzene and its homologues, and chloroform. Of 
 these, petroleum spirit and benzene and its homologues are especially notable 
 on account of their technical importance. For the purposes of research 
 a mixture of carbon bisulphide with 10 per cent, of absolute alcohol, and, 
 more generally, benzene and chloroform, are especially suitable. 
 
 According to HANAUSEK, 1 the solubility of india rubber in various, 
 solvents is as follows : 
 
 Ceara. Negrohead. Sierra Leone. 
 
 100 parts of ethyl ether dissolve 2 '6 3 -6 4 '5 
 
 turpentine , 4 '5 5'0 4'6 
 
 chloroform 
 petroleum spirit 
 solvent naphtha 
 
 3-0 37 3-0 
 
 1-5 4-5 4-0 
 
 4'4 5'0 47 
 0-4 
 
 ,, carbon bisulphide 
 
 The solubility in benzene of various brands of india rubber is given 
 by HsEREN 2 as under: 
 
 Guayaquil, . . 20 percent. 
 
 Para, . . . 17 ,, 
 
 Carthagena, . . 16'1 ,, 
 
 Borneo, . . . 13 '8 ,, 
 
 African, . . . 127 ,, 
 
 Ceara, . . . 12 "0 
 
 1 Jahresber. Chem. Tech., 1885, p. 1143. 2 Polyt. Journ., 221, p. 391. 
 
 Mozambique, . . 11*5 per cent. 
 
 Rangoon, . . . 9'1 
 
 Quiesembo, . . 9'0 
 
 Thimbles, . . . 8 '6 
 
 Negrohead, . . 7 '8 
 
 ,r, 57 
 
THE CHEMISTRY OF INDIA RUBBER. I 5. 
 
 How the solubilities in the above two tables, which are everlastingly 
 quoted, have been arrived at, I have not been able to ascertain, not having 
 seen the papers of the two authors in extenso. I have, however, no hesita- 
 tion in describing these figures as misleading and altogether erroneous, 
 for the simple reason that, as far as my very numerous observations go,. 
 india rubber possesses no solubility at all in the generally accepted sense 
 of this term. India rubber is either insoluble in a given medium, or, if 
 ' soluble,' it is so in absolutely all proportions. Nor do any of these 
 authors seem to have observed the fact that the 'solubility' of india 
 rubber, which in this case simply means the degree of ease with which it 
 passes into solution, depends very greatly upon the condition under which 
 it is subjected to the action of the solvent. The 'solubility' of crude 
 (though dry) india rubber in its original state is always very much slower 
 than that of the same rubber after it has been washed, and the longer 
 the india rubber has been subjected to the operations known as masticating,, 
 rolling, or kneading, the more rapidly does it pass into solution. 
 
 But under whatever conditions india rubber is subjected to solvents 
 there is no definite solubility, for the simple reason that such a thing as 
 saturation does not exist in india rubber solutions, only concentration or 
 dilation. Indeed, the actual fact is that such a thing as an india rubber 
 solution does not, strictly speaking, exist. India rubber 'solutions'* are not 
 solutions of india rubber in a solvent, but solutions of a so-called solvent in 
 india rubber a distinction with a considerable difference. 1 
 
 If we place a piece of india rubber for the purpose of this experi- 
 ment preferably masticated india rubber in one of its 'solvents,' the 
 rubber does not pass into the solvent, but the latter passes into the former, 
 swelling it very greatly, permeating every part of it, and forming a 
 colloidal jelly. On adding more of the solvent, the same process repeats 
 itself until, very gradually, we reach a point when the jelly begins to flow. 
 On the addition of fresh quantities of ' solvent ' to this jelly, the latter tends 
 to separate from the liquid employed as oil from water, and only very 
 gradually does the latter enter into and further distend the jelly ; whereas, if 
 normal solution took place, we should expect to see the jelly melting away 
 into the liquid. But the whole process is of an exactly opposite character. 
 This becomes strikingly apparent if we add to such a jelly a further quantity 
 of solvent and stir vigorously. In this case we obtain at first an apparently 
 thin solution, which, however, on close examination, will be found to possess 
 all the characteristics of a gross emulsion of the jelly in the solvent. On 
 continuing the agitation, the apparent thin solution rapidly begins to- 
 thicken, often giving one the impression that the addition of more solvent 
 had actually resulted in producing a more viscous 'solution' than we 
 possessed before this dilution. By the time the jelly has been distended 
 with solvent to such an extent as to represent the analogue of about a 
 4 per cent, solution, we have reached such a degree of fluidity that further 
 
 1 The same, indeed, applies, pari passu, to all the so-called solutions of colloids. 
 
1 6 INDIA RUBBER AND ITS ANALYSIS. 
 
 additions of the solvent appear to mix at once homogeneously with the 
 whole solution. 
 
 GLADSTONE and HIBBERT 1 observed that heating reduces the solubility ' 2 
 of India rubber. If, however, the heat be applied to it in conjunction 
 with a mechanical kneading treatment, its solubility vastly increases. 
 Thoroughly 'masticated' india rubber does not take as many hours to 
 dissolve as the crude product requires days. 
 
 The cause of this extraordinary physical behaviour is not at present 
 understood, although we know that neither the application of heat by 
 itself to india rubber, nor the combined influence of heat and mechanical 
 working, induce any perceptible chemical change. 3 The physical change 
 is, however, marked enough. It finds expression, as already remarked, 
 in the much increased solubility of the product, and also in the increase 
 of its plasticity at the expense of its elasticity. This change becomes the 
 more marked the longer the rubber has been mechanically worked. But 
 these observations, upon which rest important technical applications, are 
 
 1 Journ. Chem. Soc., 1888, p. 679. 
 
 2 By this term, here and in future, as applied to india rubber, I mean simply to 
 designate the relative ease of forming a pseudo- solution. 
 
 3 This statement might be called in question in view of the fact, well known to 
 practical workers, that the longer india rubber has been worked mechanically, the more 
 sulphur it requires for vulcanisation. It must be borne in mind, however, that to the 
 manufacturer the term ' vulcanisation ' denotes merely the physical aspect of the 
 change produced in india rubber by sulphur at higher temperatures. The discovery that 
 this physical change is due to the formation of an addition product of india rubber and 
 sulphur is of quite recent date, and even at present there are still some workers who 
 either altogether deny the formation of such compounds, or, if they admit it, consider it 
 as without influence upon the physical result of the vulcanisation. Therefore, if the 
 manufacturer describes a certain article as vulcanised, lie does so merely upon the evidence 
 of its physical properties, without any regard as to whether this article contains any free 
 sulphur, and if so how much, or whether it does not. Now, while in the author's 
 opinion this book furnishes conclusive proof that the vulcanisation is the physical effect 
 of the chemical combination of india rubber and sulphur, the degree of the physical 
 effect thus produced is, however, a rather complicated function of the amount of sulphur 
 combined with the india rubber (sulphur of vulcanisation), and of the physical state of 
 the india rubber when subjected to vulcanisation. While, therefore, for india rubber in 
 the same physical condition the physical vulcanisation effect increases proportionately 
 with the amount of combined sulphur, equal quantities of combined sulphur in different 
 samples do not necessarily imply physical equality of vulcanisation. In the latter 
 respect two chemically equivalent samples differ according to the difference of the 
 physical conditions of the india rubber from which respectively the two vulcanised 
 samples were obtained. A highly interesting application of this fact is to be found in 
 the manufacture of ebonite (hard rubber). For the manufacture of this product the 
 India rubber is mechanically worked into a state of complete plasticity, almost entirely 
 destroying its elasticity. Vulcanisation is then effected with a very large quantity of 
 sulphur, yielding a tough, highly elastic product. If the india rubber were as little 
 worked as is customary for the manufacture of soft rubber goods, the high degree of 
 vulcanisation would result in the production of a very hard, but extremely brittle, 
 product. In other words, the great physical difference between rubber worked for 
 varying lengths of time does not appreciably affeet its chemical behaviour towards 
 sulphur, although it greatly influences the physical effect of the vulcanisation. 
 
THE CHEMISTRY OF INDIA RUBBER. 1 7 
 
 as yet quite unaccounted for. So is also the well-known fact that the 
 mechanical strength of india rubber which has once been dissolved is 
 greatly inferior to that of the crude rubber, a difference which even 
 persists after vulcanisation, and which finds expression in the universal 
 commercial distinction of cut sheet rubber as against 'spread sheet 
 rubber,' and between the goods manufactured from either of them. 
 
 It is not surprising that these facts, in conjunction with the observation 
 that rolled (calendered) sheets of india rubber show a considerable 
 difference in tensile strength in the directions respectively of the length 
 and width of the sheet, should have given rise to the opinion that india 
 rubber possesses something akin to the warp and weft structure of a 
 textile fabric. Indeed, this crude analogue may yet be found to have 
 some foundation in an ultimate definite disposition of the molecules of all 
 colloids in the plastic state at right angles to any strain acting upon the 
 colloidal aggregate. It is certainly very curious that all colloids, which 
 in the plastic state are formed into films, should possess considerably 
 greater strength across the sheet or film than along it. 
 
 To the unaided eye india rubber, as all other colloids, appears quite 
 structureless. The reticulated structure attributed to it by PAYBN is 
 perhaps due, not so much to its porosity, as to a particular form of 
 distribution throughout its mass of the above mentioned insoluble con- 
 stituent. WIESNBR l has also observed this reticulated structure in some 
 kinds of india rubber, but certain varieties he found to be entirely without. 
 
 All varieties of india rubber, viewed under the polarising microscope, 
 exhibit brilliant interference colours, particularly if thin sections are 
 strongly compressed between two slides. 
 
 According to EXNER, india rubber possesses 'perfect elasticity.' But 
 this statement is, to say the least, certainly misleading. It is neither true 
 of crude nor of manufactured (vulcanised) india rubber, and is probably 
 only another instance of the incorrect way in which the word * elastic ' is 
 so often used, by none more so than by rubber manufacturers themselves. 
 In scientific terminology that substance is regarded as having the highest 
 coefficient of elasticity which requires the greatest stress to produce a 
 given deformation or strain. Thus, the hardest steel is more perfectly 
 elastic than a piece of India rubber, but it is certainly not so stretchable 
 or distensible. 
 
 The distensibility of india rubber increases considerably with the 
 temperature, and is, in the crude product, the same in every direction. It 
 appears so far to have escaped observation, that, on distension, india rubber 
 acquires a strong negative electric charge i.e., loses electric potential. This 
 interesting phenomenon can easily be observed by stretching over an 
 ordinary gold-leaf electroscope a small sheet, preferably unvulcanised, of 
 pure india rubber. On removing the sheet while still in the stretched 
 condition, the leaves will be found to diverge widely. On releasing the 
 
 1 WIESNER, Rohstoffe, 2nd ed., 1900, 386. 
 
 2 
 
1 8 INDIA RUBBER AND ITS ANALYSIS. 
 
 stretched sheet, it returns, of course, to its original condition, and the 
 potential becomes normal. On exposing distended india rubber to low 
 temperatures, the distension remains permanent at ordinary temperature, 
 but gradually disappears on heating. URB found the specific gravity of 
 india rubber in the normal and in the distended condition to be 0*9259 
 and 0*9487 respectively. On first heating india rubber in the distended 
 condition to 115 C. and then cooling it, no contraction occurs, and the 
 distension is rendered permanent. 
 
 Not unconnected with the above described electrical phenomenon 
 accompanying the distension of india rubber is probably the fact, observed 
 by JOULE, that distension is accompanied by the evolution, and contraction 
 by the absorption, of heat. In agreement with this is the observation due 
 to TYNDALL that india rubber, distended to three times its length, contracts 
 on warming. On closer examination of this phenomenon, SCHMULEWITSCH 
 showed that this contraction occurs only if a considerable load is attached 
 to the india rubber : under a light load further distension takes place on the 
 application of heat. Accordingly, there exists an intermediate load, under 
 which the degree of distension of india rubber remains constant at various 
 temperatures. 
 
 On submitting un vulcanised india rubber to heat, its softness and 
 stickiness increases, it becomes more plastic. Subjected to low tempera- 
 tures, it grows very stiff, almost horny. This behaviour is considerably 
 altered by vulcanisation, the effect of which is to render the physical 
 properties of india rubber almost indifferent to temperature changes within 
 the interval of from C., or even somewhat lower, to 100 C. On 
 immersion, however, into boiling liquid air, even vulcanised india rubber 
 becomes hard and as brittle as glass. 
 
 The rate of diffusion of different gases through india rubber septa was 
 found as follows by GRAHAM l : 
 
 Nitrogen, . . ;. . 1 j Oxygen, 2'556 
 
 Carbon monoxide, . . . 1-113 | Hydrogen, .... 5 '500 
 
 Air, . . f ,.; . . . 1'149 
 Methane, 2 '148 
 
 Carbonic acid, .... 13 '585 
 
 GRAHAM, whose experiments in this direction were based upon 
 corresponding observations made by Dn MITCHELL of Philadelphia, sug- 
 gested that " those gases penetrate most readily which are easily liquefied 
 by pressure, and which are also generally highly soluble in water or other 
 liquids," and he assumes that the gases entering into the rubber septum 
 are liquefied in it, permeate it, and are again evolved on the other side. 
 
 WROBLEWSKI arrived at the following conclusions in an investigation on 
 the absorption of gases : 
 
 1. Henry Dal ton's law for the absorption of liquids and gases applies 
 also to india rubber and gases. 
 
 2. The absorption coefficients of vulcanised india rubber are linear 
 
 1 Phil. Trans., 1866, 399. 
 
THE CHEMISTRY OF INDIA RUBBER. 19 
 
 functions of the temperature ; they decrease with increasing temperature 
 for nitrous oxide and carbonic acid, but the absorption coefficient for 
 hydrogen increases with the temperature. 
 
 3. The absorption of nitrous oxide, carbonic acid, and hydrogen is a 
 purely physical phenomenon involving no chemical reaction. 
 
 4. Graham's hypothesis assuming the liquid condensation of the gases 
 within the india rubber is untenable. 
 
 5. The absorbed gases persist in the gaseous state and retain all the 
 properties characteristic of the latter. 
 
 6. The gases distribute themselves through the india rubber, according 
 to the laws of thermal conductivity of solid bodies, which laws apply also 
 to the absorption phenomena of all substances. 
 
 7. The constant D, of nitrous oxide, carbonic acid, and hydrogen is 
 independent of the chemical nature of the gases, and also of their co- 
 efficients of absorption and saturation. 
 
 8. It depends only upon the physical properties of the gases, their 
 specific gravities more particularly. Approximately it varies inversely as 
 the square roots of their specific gravities, with this exception, however, 
 that the specifically very light gases possess a much greater constant than 
 what would follow from the above relation. 
 
 9. In the case of nitrous oxide and carbonic acid the constant Z), 
 increases with the temperature. 
 
 10. At 10 C. it amounts for these gases to about 0'02 of the 
 corresponding value for carbonic acid and water. 
 
 11. An india rubber septum acts like a porous plate, possessing the 
 property of condensing gases. The porosity of india rubber is of the 
 same order as that of solid graphite. 
 
 12. From GRAHAM'S experiments it may be inferred that also in 
 heated metals the constant D, is smaller for the gases of lower specific 
 gravity. 
 
 In a series of experiments on grey vulcanised india rubber, HtJFNER 1 
 arrived at the following conclusions : 
 
 1. Within a range of temperatures from 5 to 25 C. there is no definite 
 absorption coefficient of india rubber for air. The apparently existing 
 absorption is an absorption of oxygen only, causing probably the oxidation 
 of the india rubber. 
 
 2. Within a range of temperatures from 15 to 25 C. india rubber 
 absorbs no measurable quantities of nitrogen, and the same is true of 
 
 3. Hydrogen within a range of temperatures from - 2 to + 13 C. 
 
 4. On the other hand, at a temperature of - 2 C. india rubber absorbs 
 its own volume of carbonic acid. 
 
 ZULKOWSKI has shown that the absorption of ethylene and of the higher 
 aromatic hydrocarbons of coal gas is sufficiently marked to interfere with 
 the illuminating power of the gas. India rubber is able to absorb 8 '64 
 1 Ann. Phys. Cliem., N.F. [34], 1. 
 
20 INDIA RUBBER AND ITS ANALYSIS. 
 
 per cent, of its weight of these gases. It must, however, be pointed out 
 that these gases, when liquefied, are all energetic solvents of india rubber, 
 and this absorption is therefore equivalent to the absorption of water vapour 
 from the air by hygroscopic substances. 
 
 Regarding the molecular weight of india rubber nothing of any definite 
 nature is known at present. Its statistical formula, on the basis of the 
 best analyses made with the most highly purified products obtainable, may 
 safely be accepted as C 10 H 16 , but is very probably many times larger. It 
 cannot be smaller, for, although C 5 H 8 would express its percentage com- 
 position equally well, GLADSTONE and HIBBERT l have shown (see page 23) 
 that for C 10 H 16 india rubber contains three ethylene bonds. From the 
 behaviour of india rubber against sulphur monochloride, the author 2 has 
 deducted the formula C 50 H 80 as a possible minimum, and certain other 
 chemical facts since discovered have rendered this formula, or, as an alter- 
 native, C 60 H 96 , as well within the range of possibilities, without, however, 
 settling this important point. 
 
 GLADSTONE and HiBBEiiT 3 attempted to ascertain it by determining the 
 depression of the freezing point of a 10 per cent, solution of india rubber 
 in benzene, but the observed depression was so exceedingly small, that if 
 RAOULT'S law holds good in this case, the molecular weight of india rubber 
 would appear to be of enormous magnitude. 
 
 Action of Heat upon India Rubber. 
 
 Under the influence of comparatively moderate degrees of dry heat the 
 tackiness of all brands of india rubber increases more or less rapidly, but 
 no comparative investigation of the softening effect of heat upon the differ- 
 ent brands has been made. It is, however, well known in rubber works 
 that certain brands of india rubber cannot be hung up to dry in the form 
 of sheets after the washing process, as they become so soft as to fall to pieces. 
 It must not be supposed, however, that this softening of technically pure 
 rubber is a property of the india rubber hydrocarbon itself absolutely. In 
 commercial qualities of india rubber the resinous and oily substances all 
 possess very low melting points ; in many cases, indeed, they are liquid or 
 semi-liquid at ordinary temperatures. At higher temperatures they act 
 upon the india rubber as solvents, and in this manner often greatly 
 exaggerate what might seem to be the softening action of heat upon india 
 rubber. For this reason the temperature at which any given brand of 
 india rubber begins to soften or melt can only be stated within very wide 
 limits, as it varies for the same brand according to the percentage of resins 
 present. Indeed, india rubber has no true melting point. With the rising 
 temperature it grows softer and stickier, reaching sooner or later a con- 
 dition in which it can no longer be handled without adhering to everything 
 
 1 Journ. Chem. Soc., 1888, p. 682. 
 
 2 Journ. Soc. Chem. Ind. [xv.], 1894, p. 16. 
 
 3 Journ. Chem. Soc., 1888, p. 691. 
 
THE CHEMISTRY OF INDIA RUBBER. 21 
 
 it touches, and eventually it passes into a more or less fluid state. The 
 point at which this occurs is generally stated to lie at a temperature of 
 about 170 to 180 C. This, however, refers probably to india rubber (Para) 
 containing its usual amount 1*5 per cent. of oily and resinous matter. 
 The melting point of india rubber purified by extraction with acetone certainly 
 lies much higher. GLADSTONE and HIBBERT l found that on heating a solu- 
 tion of pure Para in toluene to 200 C. for two hours, and then distilling off 
 the solvent at 112 C., there was left a residue of absolutely unaltered india 
 rubber. Even when heated at 210 C. in a Sprengel vacuum, the india 
 rubber was found to be only superficially melted, though after that treat- 
 ment it dissolved very slowly in benzene. 
 
 The combined action of heat and mechanical working upon india 
 rubber is much more marked, and it is of considerable technical importance. 
 It has, generally speaking, the effect of gradually increasing its plasticity 
 and adhesiveness at the expense of its elasticity and cohesion. This effect 
 is very curious and cannot be explained by the absorption of oxygen by the 
 india rubber during the treatment, as this amounts to only from 0'5 to 1 
 per cent. If this oxidation is prevented by carrying out the whole operation 
 in an atmosphere of carbonic acid, the result is still the same. More 
 remarkable still is the fact that the longer the above treatment is continued, 
 the greater is the amount of sulphur required to obtain a certain mechanical 
 standard of vulcanisation, or, in chemical terms, the higher is the amount 
 of combined sulphur required in the india rubber in order to produce a 
 solid and stable vulcanisation product. For this reason it is obvious that 
 in the production of soft rubber goods the india rubber should be worked 
 as little as possible ; whereas, for the production of hard rubber, the longer 
 it is worked, the better for the purpose. MINDER'S statement 2 that articles 
 produced from ' long worked ' india rubber are liable to decay rapidly is 
 therefore entirely misleading ; they are so only if the degree of vulcanisation 
 has not been adapted to the altered physical conditions produced in the 
 india rubber. But this is equally true of india rubber in any other con- 
 dition. 
 
 Although no perceptible decomposition takes place at the temperature 
 given above, india rubber liquefied at this temperature is found to have 
 undergone a great change. On cooling, it remains very soft and sticky, 
 and although on long standing it may return to a more solid condition, the 
 former elasticity of the india rubber has disappeared, and it exhibits now a 
 brittleness much resembling that of certain kinds of bitumen. To account 
 for this remarkable change, HEINZERLING 3 suggests that the conversion of 
 the india rubber into a tarry mass by heat may be a process similar to the 
 conversion of starch into starch paste. This analogy is, however, too 
 superficial ; it barely expresses the physical aspect of the observed facts, 
 and it is either inadequate or misleading if applied to the explanation of 
 
 1 Journ. Cliem. Soc., 1888, p. 686. 2 Gummi Zeitung, 14, 687. 
 
 3 Fabrik. dcr Kautschuk- und GuMa-Pcrcha-Waaren, Braunschweig, 1883. 
 
22 INDIA RUBBER AND ITS ANALYSIS. 
 
 the chemical aspect of the change in question. For whereas the process of 
 the liquefaction of starch is a hydration process dependent upon the inter- 
 action of starch and water under the influence of higher temperature, the 
 liquefaction of India rubber under the influence of heat involves no 
 chemical reaction ; it is most likely due to a physico-chemical or molecular 
 change comparable to the conversion of paraldehyde into aldehyde. In 
 other words, the liquefaction of india rubber is probably due to the breaking 
 up of the no doubt very large india rubber molecule into smaller molecules 
 of the same empirical formula. This is shown by the fact, which will be 
 dealt with in detail later on, that india rubber in its primary state, as well 
 as after liquefaction, at any temperature up to 200 C., and even the 
 products of the dry distillation of india rubber, all possess the empirical 
 formula C 10 H 16 , and form a bromine addition product C 10 H 16 Br 4 . But there 
 is this difference between these conversion products and india rubber, that 
 the higher the temperature at which they have been formed, the more 
 soluble and the more reactive they are, and this also holds good for the 
 bromides of these conversion products as compared with the bromide 
 obtained direct from india rubber. Now this break up of the india rubber 
 molecule is carried to its extreme limits in the dry distillation of india 
 rubber. This has been investigated, with highly interesting results, by a 
 number of observers. Nevertheless, the subject is one still inviting research, 
 and, even as far as it has been dealt with, one that requires revision in a 
 number of points. 
 
 The dry distillation of india rubber was first investigated by HiMLY, 1 
 at almost the same time as by BoucHARDAT. 2 But no more recent investi- 
 gation than that of WILLIAMS 3 in 1860 appears to have been made. From 
 the distillation products three definite compounds were isolated : isoprene 
 C 5 H 8 (B.P. 37-38 C.), caoutchene C 10 H 16 (B.P. 171 C.), and heveene 
 (B.P. 315 C). According to BOUCHARDAT, the crude distillate also contains 
 a butylene C 4 H S boiling at between - 1 8 and C. Five kilos of fresh 
 Para rubber yield, besides a small quantity of gases, 250 grms. of isoprene, 
 2000 grms. of caoutchene, and 600 grms. of heveene. The rest consists of 
 very heavy oils possessing very high boiling points. There is, however, no 
 doubt that these yields depend largely upon the conditions under which 
 the distillation is carried out, although the qualitative products remain 
 the same in every case. Now, while this shows that the production of the 
 above named bodies in the dry distillation of india rubber does not indicate 
 three or more different lines of cleavage in the molecule of the latter, it 
 points very distinctly towards the assumption that these three or more 
 compounds stand in close genetic relation to each other. If we disregard 
 the observation of BOUCHARDAT that a small quantity of butylene is formed 
 in the dry distillation of india rubber, an observation which, so far, has not 
 been confirmed, we find that all the characteristic products formed in this 
 
 1 Ann. Chem., 27, 41. 2 Bullet. Soc. chim., 24, 108. 
 
 3 Proc. Roy. Soc., 10, 517. 
 
THE CHEMISTRY OF INDIA RUBBER. 23 
 
 process are characterised by the same percentage composition, and their 
 respective molecular complexes by the general formula (C 5 H 8 ) W , which 
 also includes the mother substance, india rubber itself. This consideration 
 renders it highly improbable that any of the above named compounds are 
 immediate products of the destructive distillation of india rubber," but 
 rather suggests that they are all due to the disintegration of larger 
 molecules. With regard to isoprene, this is scarcely open to doubt, inas- 
 much as TILDEN l showed it to be one of the products of the destructive 
 distillation of various terpenes. The fact that of the above named bodies 
 only isoprene, and neither caoutchene nor heveene, is capable of polymerisa- 
 tion into india rubber 2 renders the existence of the hexane ring, in the 
 molecule of the latter at least, very doubtful. But this doubt becomes 
 a certainty in the light of the results obtained by GLADSTONE and HIBBERT 3 
 in the optical examination of india rubber. In investigating the optical 
 properties of india rubber, taking C 10 H 16 as the molecule of this compound, 
 these authors desired to ascertain whether the C 10 H 16 molecule of india 
 rubber contains one, two, or more pairs of doubly linked carbon atoms. 
 
 " The theoretical refraction and dispersion equivalents of C 10 H 16 in the 
 three conditions above mentioned would be as follows, assuming that each 
 pair of doubly linked carbon atoms would produce the same increment as 
 in the aromatic series " : 
 
 p IT T>MA-! 
 
 ^lo^ie * ^ 
 
 1 pair of doubly linked carbons, . . . 73 '0 
 
 2 ... 75-2 
 
 3 ... 77-4 
 
 GLADSTONE and HIBBERT found that the optical constants of iudia 
 rubber exhibit considerable variations in different specimens. These 
 variations appear to be in direct proportion of the amount of oxygen 
 present, which, it is safe to assume, attaches itself to the doubly linked 
 carbon atoms present, thereby suspending the double bond and lowering 
 both the refraction and dispersion equivalents. However, two especially 
 carefully prepared samples yielded the following figures on examination : 
 
 Sample. P^ P^-I^ 
 
 No. 10, 4 76-87 5-67 
 
 No. 11, 77-59 5-34 
 
 These figures leave little doubt that india rubber, for C 10 H 16 , contains 
 three pairs of doubly linked carbon atoms. 
 
 "If this be the case, the molecular formula cannot be C 5 H 8 like 
 isoprene, or C 15 H 24 like citrene, as these would give respectively one and 
 a half, and four and a half, pairs of carbon atoms united by double linking. 
 
 1 Trans. Chem. Soc., 45, 410. 2 TILDEN, Chem. News, 1892, p. 265. 
 
 3 Journ. Chem. Soc., 1888, p. 680. 
 
 4 These numbers are taken from the experimental series given in the authors' 
 repeatedly quoted paper. 
 
24 INDIA RUBBER AND ITS ANALYSIS. 
 
 It cannot contain the hexagonal ring, 1 but must be expressed graphically 
 by a chain formula. This may account for the wide difference of properties 
 between caoutchouc and the various essential oils. 
 
 " The general results of the destructive action of heat upon caoutchouc 
 may therefore be considered as involving no change in the proportion of 
 carbon to H, but changes of structure in the hydrocarbon which are best 
 represented by the following optical constitutional formulae. In this table 
 C" is used for C, having the refraction equivalent 6'1 and the dispersion 
 equivalent 0'66 " : 
 
 Substance P tical Constitutional 
 
 Formula. 
 
 India rubber (polyprene), n C" 6 C 4 H 16 
 
 Isoprene, ..... C" 4 CH 8 
 
 Caoutchene (dipentene), . . . C" 4 C 6 H 16 
 
 Heveene, ..... nC" 2 C 3 H 8 
 
 One is forcibly reminded by these facts of the very similar type of 
 reactions exhibited by a well-known camphor, geraniol, and geranial, its 
 aldehyde. Geraniol has been shown to possess the formula C 10 H 18 
 
 (CH 3 ) 2 CH.CH 2 .CH : CH.C(CH 3 ) : CH.CH 2 (OH). 
 
 On oxidation with bichromate it yields an aldehyde, geranial, C 10 H 16 O 
 (CH 3 ) 2 CH.CH 2 .CH : CH.C(CH 3 ) : CH.CHO, 
 
 and this, on treatment with dehydrating agents, splits off water with 
 formation of cymene. This transformation is an exact parallel to the 
 formation of dipentene and heveene in the destructive distillation of india 
 rubber. We have at any rate here the same transformation of an olefinic 
 carbon chain (geranial) into a derivative of a cyclohexane. This process, in 
 its application to the india rubber problem, is so interesting that it is worth 
 while illustrating more clearly the formation of the hexane ring from the 
 open chain. This is best done by the following graphical representation : 
 
 CH 3 CH 3 CH 3 CH 3 
 
 Cxi Cxi 
 
 CH 2 C 
 
 CH.OCH 
 
 I II 
 
 CH CH CH CH 
 
 V 
 
 CH 3 CH 3 
 
 Geranial. Cymene. 
 
 In its results upon the optical constants of the two compounds this process 
 is, however, exactly the reverse to what we observed in the case of india 
 rubber. It will, of course, be observed that this is simply due to the fact 
 
 1 The exactly contrary statement in the article " India Rubber," in vol. i. p. 311 of 
 Thorpe's Diet, of Appl. C?tem., is obviously a misquotation by the compiler of that 
 article. 
 
THE CHEMISTRY OF INDIA RUBBER. 25 
 
 that in the case of geranial the ring formation is not due, as in the case of 
 india rubber, to a physical rearrangement of carbon bonds, but to a 
 dehydration process which results in the formation of a benzene derivative 
 as distinguished from a terpene. 
 
 Now, if we apply to geraniol the same process of dehydration which 
 from geranial led to cymene, it is obvious that we should obtain a com- 
 pound C 10 H 16 , a terpene. This important experiment was carried out by 
 SEMMLER, 1 and resulted in the formation of an undoubted terpene. The 
 optical analysis of this terpene, geraniene, indicated the existence in it of 
 three ethylene bonds, full proof of which was furnished by the formation 
 of an addition product C 10 H 16 Br 6 . This shows that in geraniene we have a 
 terpene which stands evidently in the same relation to india rubber as 
 the monocyclical terpenes to the polyterpenes. Just as there is a homo- 
 logous series of cycloterpenes, the terms of which differ from each other 
 by an increment C 5 H 8 , and all of which answer the general formula (C 5 H 8 ) n , 
 so also there is, no doubt, a series of olefinic or open chain terpenes, 2 of 
 which geraniene and india rubber are extreme terms. 
 
 The connecting link between these two series is the hemiterpene isoprene 
 C 5 H 8 , which on polymerisation may, and indeed does, form simultaneously 
 representatives of both series ; for we know already that in the polymerisa- 
 tion of isoprene, india rubber, the olefinic polyterpene, is formed at the 
 same time as the cycloterpene dipentene. 
 
 The constitution of this connecting link, isoprene, has long been 
 doubtful, but has recently been elucidated by IPATIEFF and WiTTORF. 3 
 Already TILDEN,* however, came to the conclusion that the constitution of 
 isoprene would most probably have to be expressed by the graphical formula 
 CH 2 = C(CH 3 )-CH = CH.,, which compound he described as -methyl- 
 crotonylene, although he did not succeed to furnish experimental proof 
 for this assumption. WALLACE'S 5 investigations thoroughly confirmed 
 BOUCHARDAT'S 6 and TILDEN'S results, and he further established the perfect 
 identity of the terpene C 10 H 16 , formed by the polymerisation of isoprene 
 with cymene (dipentene). 
 
 The first approach towards experimental confirmation of TILDEN'S 
 formula of isoprene was, however, furnished by GADZIATZKY, 7 who 
 from the monochlorhydrate of isoprene obtained a tertiary alcohol 
 
 CH 3 , 
 
 >COH-CH=CH 2 
 CH/ 
 
 dimethyl-vinylcarbinol. Soon afterwards the same author 8 found that, 
 
 1 er., 24, 682. 
 
 2 With the exception of india rubber, no higher term of the olefinic terpene series is 
 known at present than C 10 H 16 , but several isomers of this formula and series are already 
 known. 
 
 3 Journ. prakt. Chem. [2], 55, p. 1, 1896. 4 Chem. News, 46, 129. 
 
 5 Ann. Chem., 227, 292. 6 loc. tit. 7 Bull. Soc. chim. [2], 47,168. 
 
 8 Journ. russ. chem. Ges., 20, i. 706. 
 
26 INDIA RUBBER AND ITS ANALYSIS. 
 
 on heating with weak alcoholic hydrochloric acid, isoprene is transformed 
 into an isomeric dimethyl-allene, which transformation he expressed graphi- 
 cally thus : 
 
 /CH 3 /CH 3 
 
 pTT p / j_ TTP1 PTT PTT P/ 
 
 i un v^ v \ T xii^i ojio L/i u < 
 
 I X CH 3 
 
 Cl 
 
 /CHo 
 
 TTpl PTT P P / 
 
 -LlV^l \_>Jtlo v> %*?X 
 
 Cl 
 
 So far all the addition products of isoprene obtained by the above named 
 authors are liquids possessing somewhat indefinite physical characteristics, 
 and more or less prone to decomposition on distillation. But in 1895 
 MOKIEWSKY l succeeded to obtain a solid crystalline body (M.P. 81 C.) by the 
 action of hypochlorous acid upon isoprene. The empirical formula of this 
 substance was found to be C 5 H 10 2 C1 2 . Besides this solid body, which he 
 regarded as an addition product of two molecules of hypochlorous acid with 
 one molecule of the hydrocarbon C 5 H 8 , he obtained a liquid (B.P. 141 C.), 
 the analysis of which led to the formula C 5 H n OCl, and which, on treat- 
 ment with alkali, yielded trimethyl-ethylene oxide and a glycol. From 
 this MOKIEWSKY inferred that in the distillation of pinene through red- 
 hot tubes, by which process he had obtained the raw material for his 
 investigation, a mixture of trimethyl ethylene (CH 3 ) 2 C = CH.CH 3 and of 
 a hydrocarbon C 5 H 8 is obtained, the latter constituting the bulk of the 
 distillate. 
 
 This fact must appear particularly interesting since IPATIEW and 
 WITTOJRF 2 succeeded in isolating trimethyl-ethylene also from the products 
 of the dry distillation of india rubber. The hydrocarbon C 5 H 8 accom- 
 panying it, they isolated in the shape of a dihydrobromide of the formula 
 C 5 H 10 Br 2 , and this substance was found to be identical with the dihydro- 
 bromide obtainable from the above mentioned /5-dimethyl-allene. This led 
 the authors to conclude that the hydrocarbon C 5 H 8 , isoprene, must 
 possess the constitution of a /3-methyl-divinyl, as only such a compound 
 could be expected to yield the same dihydrobromide as the unsymmetrical 
 dimethyl-allene. This will become perfectly clear by a perusal of the 
 following graphic representation : 
 
 ^wjTI OTT f^TT i OTTT3-.* >^r^1> OTT OTT T>- 
 
 \\j ui ^-tig " ^-H-JtH >L/r>r 0x12 ~~ uxigDr 
 
 CH/ CH/ 
 
 Isoprene (8-methyl-dmnyl) Dim ethyl-trimethylene -bromide. 
 
 /ITT /^TT 
 
 O-Llov V^ilo^ 
 
 /C = C CHo + 2HBr = yCBr CHo ~~ CH<>J3r 
 
 CH/ CH/ 
 
 j8-dimethyl-allene. Dimethyl-trimethylene-broniide. 
 
 In completing his proof of the constitution of isoprene, IPATIEW 3 shows 
 
 1 Journ. russ. chem. Ges., 27, 516. 
 
 2 Journ. praJct. Chem. [2], 55, 1. 3 Ibid. [2], 55, 4. 
 
THE CHEMISTRY OF INDIA RUBBER. 2 7 
 
 that the action of alcoholic potash upon dimethyl-trimethylene-bromide, 
 whether prepared from isoprene or /?-dimethyl-allene, always yields the 
 same hydrocarbon, boiling at from 32 to 33 C. This remarkably low 
 boiling point of this hydrocarbon (C 5 H S ), as well as its other properties, 
 altogether preclude the assumption of its constitution being that of either 
 a secondary or di-substituted acetylene, nor of /2-dimethyl-allene, so that 
 there remains for isoprene only the formula of /3-methyl-di vinyl, 
 
 2 %C-CH = CH 2 . 
 CH/ 
 
 That the peculiar distribution of the ethylene bonds in methyl-divinyl 
 must actually exist in isoprene, he proves, not only by the identity of the 
 dihydrobromide of isoprene with the dihydrobromide of allene, but more 
 particularly by the fact of it producing with hypochlorous acid a dichlor- 
 hydrine, a property very characteristic of hydrocarbons of the diallyl type. 
 The final proof of the identity of isoprene with /2-methyl-divinyl was 
 subsequently furnished by EuLER, 1 who succeeded in preparing )S-methyl- 
 divinyl synthetically, and found it in all its remarkable properties identical 
 with isoprene. He used as his starting-point the observation of CIAMICIAN 
 and MAGNAGHi, 2 that dimethyl-pyrollidyl-ammonium-iodide on treatment 
 with caustic potash yields divinyl. There being a /8-methyl-pyrollidine 
 known, it was to be expected that this substance could be similarly con- 
 verted into /8-methyl-di vinyl. This he accomplished by the following 
 reactions : 
 
 I. 
 CH 3 .CH.CH 2 \ CH,.CH.CH 2 \ /CH 3 
 
 >NH + 2CH 3 I + KOH = >N ACH 3 + KI + H 2 
 
 CHfrCHa/ CH 2 .CH 2 X \I 
 
 j8-methyl-pyrollidine. j8-methyl-dimethyl-pyrollidyl-ammonium- 
 
 iodide. 
 
 II. 
 
 CH 3 .CH.CH 2 \ /CH 3 CH 3 .C = CH 2 
 
 >N4-CH 3 + KOH = 
 
 p-p- prr / \ T ' PTT PTT 
 
 Ul 2 .Orl 2 x \i V^xl 2 .Uxl 2 
 
 /8-metliyl-dimethyl-pyrollidine. 
 III. 
 
 piTi p PTT PTT P PTI 
 
 L/X1 3 .O VI1 2 OX1 3 .O U1 2 /PJT 
 
 PTT PTI "NT '/ PTI PTI XT/ ^^3 
 
 UH>\/ZM>XIC UXl^UJtlo. JW y. PTT 
 
 \CH 3 \\CH 3 
 
 ^8-raethyl-trimetliyl-pyrollidyl- 
 ammonium-iodide. 
 
 IV. 
 CH 3 .C = CH 2 X HTI (^TI Q CH 
 
 3 V 
 
 3 + KOH = 
 
 - methyl -divinyl. 
 
 Journ.prakt. Chem. [2], 57, 131. 2 Ber., 18, 2080. 
 
28 INDIA RUBBER AND ITS ANALYSIS. 
 
 The /3-methyl-divinyl so obtained is in every respect identical with 
 isoprene, the constitution of which is therefore definitely established. 
 
 I have dwelt upon this matter of the constitution of isoprene to an 
 extent which might, perhaps, appear rather beyond the scope of this book. 
 But I would point out that this book, although purporting to be an ana- 
 lytical treatise, is also intended for the use of those to whom india rubber 
 is more than a raw material for technical purposes, and to those an account 
 such as the above will appear deeply interesting and even helpful. 
 
 It has been already repeatedly stated that isoprene under various con- 
 ditions may be polymerised again into india rubber, which process is, 
 however, invariably accompanied by the polymerisation of a considerable 
 part of it into a cycloterpene, sesquiterpene, and even higher members of 
 the cycloterpene series. The process is no doubt a very complicated one, 
 and is entirely unknown how, and what description of, an open chain is 
 formed representing india rubber. All we can say on this point is that no 
 configuration containing cyclical rings of any sort would result in a product, 
 the configuration of which, as regards the number of ethylene bonds in 
 the C 10 H 16 unit, could be brought into accordance with the above quoted 
 results of the refraction equivalent of india rubber. The manner in which 
 isoprene can polymerise into derivatives of the cycloterpene series is, how- 
 ever, quite obvious, and will readily be seen from the two following diagrams, 
 
 CH 3 HC CH 
 
 I //\ H /X 
 
 ATT // : ^ 
 
 HC C CH 
 
 CH 2 | | 
 
 HC C CH 
 
 HC CH H / tf'|Y\ H , 
 
 CH CH 3 CH 
 
 Sesquiterpene 
 
 > 
 
 Terpene C 10 H 16 . 
 
 in which the dotted bonds show the junctions of one molecule to the other. 
 It will be observed that even polymerisations such as these are very 
 complicated processes indeed, involving a peculiar process of migration 
 of hydrogen atoms. 
 
 By far the bulk of the products of the destructive distillation of india 
 rubber consists of the well-known cycloterpene known as dipentene, the 
 constitution of which may be expressed graphically : 
 
 CH 2 CH CH 2 
 
 CH 3 C = CH 2 A" 3 Terpadiene. 
 
 CH 2 C(CH 3 ):CH J 
 
 Besides occurring naturally in various essential oils, it may be obtained in 
 
THE CHEMISTRY OF INDIA RUBBER. 29 
 
 a variety of ways from either certain terpenes or terpene derivatives. The 
 boiling point of dipentene, the caoutchene of the older literature, was 
 observed by WALLACH at 178 C., 1 its specific gravity as 0*845. Dipentene 
 is optically inactive, and as all its derivatives can also be prepared by mix- 
 ing together equal quantities by weight of the corresponding derivatives of 
 leavo- and dextro-limonene, it follows that dipentene is the racemic modifica- 
 tion of limonene. 
 
 A rather considerable proportion of distillate yielded by the dry 
 distillation of india rubber consists of heveene C 15 H 24 , a sesquiterpene , 
 B.P. 255-265 C. Its constitution, like that of the sesquiterpenes in 
 general, is still quite unknown. It certainly cannot be that of the 
 sesquiterpene figured on page 28, since GLADSTONE and HIBBEBT showed 
 heveene to contain one pair of doubly linked carbon atoms for C 5 H 8 . 
 
 The very considerable remaining part of the india rubber distillate 
 consists of bodies boiling above 300 C. These bodies are not only, like the 
 lower boiling fractions, hydrocarbons, but their composition still conforms 
 to the formula (C 10 H 16 ) n . They are therefore, no doubt, diterpenes, and 
 perhaps even in part still higher polyterpenes. For the disentanglement 
 of mixtures of this description our present methods are ineffective. 
 
 The amount of carbon remaining behind on the destructive distillation 
 of india rubber is surprisingly small indeed, so small as to suggest that it 
 is more due to its albuminous and other impurities than to the india rubber 
 itself. From 3 kgms. of carefully washed and vacuum-dried india rubber 
 (Para) I obtained: 
 
 Isoprene, .... 186 grras. = 6 '2 per cent. 2 
 Dipentene, .... 1380 = 46'0 
 
 Heveene, .... 510 
 Polyterpenes, . . .806 
 Carbon residue, ... 59 
 Mineral residue, . . . 16 
 Loss (water and gases), . . 43 
 
 = 17-0 
 
 = 1-9 
 
 = 0-5 
 
 =: 14 
 
 To draw inferences from these figures as to the chemical constitution 
 of india rubber is extremely difficult, if not altogether premature. The 
 formation of these, compounds may be accounted for in three ways, and 
 I believe that a careful study of the behaviour of india rubber at 
 temperatures below the point at which dry distillation begins would 
 throw much light upon the mechanism of this splitting-up process. The 
 split might, in the first instance, occur in its totality at a given temperature, 
 about red heat, resulting in the total resolution of the india rubber 
 molecule into isoprene. In this case the terpenes and polyterpenes 
 present in the distillate would hav<* to be looked upon as secondary 
 
 1 Sometimes the boiling point is given as 174 '5 C., but this was most likely observed 
 on a still impure product. 
 
 2 It will be observed that this yield of 6 per cent, of isoprene is considerably larger 
 than the yield obtained by BOUCHARD AT (see page 22). This is no doubt due to the 
 difference of the conditions under which the dry distillation was carried out. 
 
3O INDIA RUBBER AND ITS ANALYSIS. 
 
 products i.e., as polymerisation products of the isoprene. If this is so, 
 it would follow that the yield of isoprene should increase with the rate 
 at which the distillation is carried out, and should also be increased by 
 carrying the distillation products rapidly from the hot retort by means of 
 a current of some inert gas. Or the action of the heat might produce a 
 gradual desintegration of the rubber molecule, accompanied by rearrange- 
 ment of C 6 -chains, into cyclical form and products of the colophene type, 
 the break-down of which might then produce dipentene and isoprene, with 
 the possibility of the formation of the latter being due to the further 
 splitting up of the dipentene ring. 
 
 In this case . valuable information would be obtained by heating the 
 india rubber for various lengths of time to temperatures somewhat below 
 the point at which actual distillation begins, and examining the changes 
 occurring both physically 1 and chemically. It might be further 
 ascertained whether dipentine, when distilled through red-hot tubes, 
 yields appreciable quantities of isoprene. , Or, lastly, the india rubber 
 might, in the first instance, without any change in the size of its molecule, 
 be converted into its corresponding cyclical modification, which then would 
 probably possess considerable stability, and therefore, when breaking up, 
 yield at once all the above characterised products side by side. Although 
 the data furnished by an exhaustive investigation would perhaps scarcely 
 carry us very far towards the solution of the question of the constitution 
 of india rubber, they could not fail to be of the highest scientific and 
 technical value. 
 
 As above stated, the behaviour of the various brands of india rubber 
 against temperatures upwards of 100 C. varies considerably as regards 
 the temperature at which the india rubber becomes softened or liquefied. 
 But there appears to be no difference whatever in the products of the 
 break-down of india rubber by heat as it occurs in the process of dry 
 distillation. All brands of- india rubber, including Balata and gutta 
 percha, seem to yield the same series of hydrocarbons as Pard itself. It 
 can, however, scarcely be said that this point has been closely studied, and 
 it is quite possible that further research in this direction might reveal differ- 
 ences in these distillation products which would in some degree explain the 
 very marked physical and chemical differences of the various brands of 
 india rubber. 
 
 Chemical Behaviour of India Rubber. 
 
 The chemical behaviour of india rubber has received less attention than 
 that of any other colloid. The explanation of this fact is no doubt to be 
 found in the circumstance that india rubber is a colloidal hydrocarbon. 
 The absence of reactive carboxyl, hydroxyl, or ami do groups deprives the 
 india rubber molecule of those peculiar lines of cleavage, the existence of 
 which determines the nature and constitution of the break-down products 
 1 Especially refractrometrically. 
 
THE CHEMISTRY OF INDIA RUBBER. 31 
 
 'of complex molecules. And this chemical indifference of india rubber 
 becomes still more accentuated owing to its colloidal state. Indeed, but 
 for the existence of a number of doubly linked carbon atoms in the 
 molecule of india rubber, this substance would in chemical indifference 
 equal, if not expeed, the higher members of the paraffin series. As it is, 
 all the more definite chemical reactions of india rubber are reactions 
 resolving one or more of its double carbon bonds reactions generally 
 leading to the formation of addition products. Very few only of these 
 have been described, and none of them thoroughly investigated. 
 
 Alkalies. Neither aqueous nor alcoholic solutions of the alkalies 
 appear to have an appreciable action upon india rubber at any rate, not 
 at temperatures below 100 0. It is, however, stated l that on prolonged 
 digestion in these solutions the india rubber becomes sticky and ' greasy.' 
 This change is probably an isomerisation change similar to the one which 
 occurs when india rubber is heated by itself to higher temperatures. 
 
 Rather interesting is the observation 2 that on prolonged digestion with 
 ammonia the india rubber passes into the state of an emulsion, in 
 appearance closely resembling india rubber milk. 
 
 Halogens. Chlorine. WI)RTZ and SCHI)TZENBERGER 3 were the first 
 to observe the action of chlorine upon india rubber, but neither of these 
 authors appears to have recognised the formation of any definite compound. 
 This is, no doubt, due to the fact that they allowed the chlorine to act 
 upon solid rubber, a procedure which might be expected to give unsatis- 
 factory results. 
 
 By passing chlorine gas through a solution of india rubber in chloro- 
 form, GLADSTONE and HIBBERT succeeded in obtaining a body G 10 H 14 C1 8 
 in the form of a white powder. This formula shows that the reaction 
 taking place was not merely an addition process, but also that substitution 
 of chlorine for hydrogen occurred. Indeed, the authors state that they 
 found it impossible to obtain an addition product C 10 H 16 C1 6 , as the 
 reaction proceeds so energetically that addition and substitution take 
 place concurrently. 
 
 Bromine. The same observers obtained, by the action of bromine upon 
 india rubber dissolved in chloroform, the products C 10 H 16 Br 4 and C 10 H 15 Br 5 , 
 although they did not succeed in isolating the former. It was found 
 impossible to obtain the hexabromide C 10 H 16 Br 6 , although the existence 
 of three ethylene bonds in the molecule of india rubber would appear to 
 render such a compound possible. In this relation it may, however, be 
 pointed out that some law of chemical statics seems to prevent the 
 accumulation of iodine and bromine in a molecule, irrespective of the 
 constitution of the latter. Thus the interaction of acetylene and bromine 
 always produces some tribrom-ethylene CHBr = CBr 2 , besides CHBr 2 : 
 
 1 Of. CHAPEL, Caoutchouc et Gutta Percha, 319. 
 
 2 HEINZERLING, Fdbr. d. Kautschuk- iitid Outta-Percha- Waaren, p. 29. 
 
 3 CHAPEL, loc. cit., 322. 
 
32 INDIA RUBBER AND ITS ANALYSIS. 
 
 CHBr 2 , which forms the chief product of the reaction. This is 
 exactly analogous to the formation of the above named pentabromide. 
 Iodine only yields a di-iodide with acetylene. 
 
 The simplest way to prepare the above named body, polyprene tetra- 
 brornide C 10 H 16 Br 4 , is to run bromine in slight excess into a solution of 
 pure india rubber in chloroform, cooling the solution well all the time. 
 Only a very slight amount of hydrobromic acid is evolved during this 
 process. The solution of the tetrabromide thus obtained is transferred to 
 a separating funnel, from which it is run in a very fine jet into alcohol. 
 In order to avoid the tetrabromide separating out in the form of clots, its 
 solution should be very dilute, and a large volume of alcohol should be 
 used for precipitation. The tetrabromide is then obtained in the form of 
 very fine white flakes, which are collected on a filter and washed with warm 
 alcohol until free from every trace of chlorine. The product is then 
 pressed and dried at a temperature not exceeding 40 C. Its composition 
 approximates more and more closely to the formula C 10 H 16 Br 4 , the greater 
 the care taken to free the india rubber from which it was prepared as 
 much as possible from its oxygen. 
 
 In the cold, polyprene tetrabromide is fairly easily soluble in chloro- 
 form and thiophenol, but quite insoluble in all other solvents tried. It 
 dissolves in aniline, pyridine, and piperidine on moderately heating it in 
 these solvents, but solution always appears to be attended by the loss of 
 more or less hydrobromic acid. 
 
 On heating the polyprene tetrabromide to 60 C., it begins to decompose 
 with evolution of hydrobromic acrd ; a yellow discolouration, which increases 
 in intensity as the decomposition proceeds, becomes at the same time 
 noticeable. At higher temperatures this decomposition proceeds very 
 rapidly, resulting finally in the formation of a very uninviting-looking dark- 
 brown residue, which, however, still contains a more or less considerable 
 quantity of bromine. 
 
 The action of boiling alcoholic caustic soda upon the tetrabromide is 
 surprisingly slight, but sodium ethylate has a more pronounced effect, 
 especially at temperatures ranging from 100 C. upwards. 
 
 0'624 grms. of C 10 H 16 Br 4 were heated with sodium ethylate to 100 C. 
 in a sealed tube for 6 hours : 
 
 Solid substance recovered : 0'3623 grms. =58*06 per cent. 
 Br in recovered portion : 0'1134 ,, =18'28 ,, l 
 Br in solution : 0'3210 ,, =51 '44 ,, 
 
 0-6154 grms. of C 10 H 16 Br 4 were heated with sodium ethylate to 120 C. 
 for 10 hours : 
 
 Solid substance recovered : 0'3438 grms. =55 '87 per cent. 
 Br in recovered portion : 0'0823 ,, =13 '38 ,, 1 
 Br in solution : 0'3478 ,, =56 '51 ,, 
 
 1 The amount of Br is calculated upon the weight of the original substance employed, 
 and not upon the weight of recovered solid substance. 
 
THE CHEMISTRY OF INDIA RUBBER. 33 
 
 Attention must be called to the fact that the weight of substance 
 recovered plus the weight of bromine in solution in both experiments is 
 considerably in excess of the weight of tetrabromide used. This must be 
 accepted as pointing to the fact that the removal of the bromine is a 
 substitution process, probably of ethoxyl groups for bromine. 
 
 Against acid and oxidising reagents the tetrabromide exhibits consider- 
 able indifference. Concentrated nitric and sulphuric acid affect it very 
 slowly. Chromic acid, or mixtures of chromic and sulphuric acid, are with- 
 out any perceptible action, even at 100 C. 
 
 Very remarkable is the reaction which takes place on heating the 
 tetrabromide with phenol. The operation is best conducted on the water- 
 bath by adding the tetrabromide to the melted phenol. Already at 
 60 C. a rapid change of colour is noticeable, the tetrabromide assuming 
 first a dull slate colour which rapidly darkens to a blackish-brown. The 
 mass at the same time becomes gummy and begins to form a brownish- 
 purple solution with the phenol. At this stage a steady evolution of 
 hydrobromic acid gas commences. As the reaction proceeds, the mass 
 grows thinner until, after about thirty minutes' heating, a thin purple- 
 coloured solution is obtained. This solution is allowed to cool, and, after 
 dilution with twice its volume of ether, it is filtered. If the operation 
 has been properly conducted, there should remain no insoluble residue on 
 the filter. The filtrate is then distilled on the water-bath until no more 
 ether passes over, and the residual solution is poured into a sufficient quantity 
 of a 6 per cent, solution of caustic soda, in order to dissolve all the phenol 
 present. A dark-brown solution is obtained, which is heated until all the 
 residual ether is driven off. Acetic acid is then slowly added until a 
 permanent, greyish-brown, and flocculent precipitate has been formed. This 
 is separated from the phenol solution by filtration, and, without previous 
 washing, again suspended in warm water and acidulated with acetic acid* 
 The suspended precipitate changes its colour into a more or less pronounced 
 brown, much resembling the colour of ferric hydroxide. It is heated to 
 boiling, filtered and washed free from acid. When dry, it forms an ochre- 
 coloured mass, easily reducible to an impalpable powder. 
 
 This substance is readily soluble in aqueous and alcoholic caustic soda, 
 alcohol, acetone, ethyl ether, ethyl acetate ; insoluble in benzene, chloroform, 
 carbon bisulphide, and petroleum ether. Its analysis gave the following 
 figures : 
 
 0-1622 grm. substance, . . . '4785 grm. CO. 2 , '1184 grm. H 2 0. 
 Calculated for C^H^, . . . C 80 '16, H 7 '14. 
 Found, . . . . 7 ... . C 80-15, H 7'42. 
 
 The analysis agrees very well with the figures calculated for a body 
 C 10 H 16 (O.C 6 H 5 ) 4 , tetroxyphenyl-polyprene, and its formation therefore would 
 appear to be due to the very remarkable reaction 
 
 C fo H 16 Br 4 + 4C 6 H 5 OH = C U H.,(C.O H 5 ) 4 + 4HBr. 
 
34 INDIA RUBBER AND ITS ANALYSIS. 
 
 The yield of this body was found in the great number of preparations 
 made to amount very uniformly to from 60 to 60*5 per cent, of the weight 
 of tetrabromide acted upon. This accounts almost exactly for one half of 
 the tetrabromide taken. What happens to the other half, I have not yet 
 been able to ascertain. Considering, however, the fact demonstrated by 
 GLADSTONE and HiBBERT, 1 that a solution of india rubber in toluene can be 
 heated to 200 C. without being thereby in the least affected, it seems 
 highly improbable that the above described reaction which can be affected 
 at water-bath temperature should lead to a break-down of the polyprene 
 molecule. Nevertheless, the inference to be drawn from the yield in the 
 above process must be that the reaction in question is not so simple as it 
 looks. This is fully borne out by the observation that, by confining the 
 heating in the above process to from 90 to 95 C. on the water-bath, and 
 treating the resulting solution in the same manner as before described, a 
 body is obtained which in appearance scarcely differs from the first, and 
 which behaves much in the same way against solvents. On analysis this 
 body gave the following figures : 
 
 0-1395 gnu. substance, . . . '3860 grm. C0 2 , 0846 '0 grm. H 2 0. 
 Calculated for C 84 H 36 6> . C, 75 -55 ; H, 6 '66. 
 Found, C, 75-41; H, 673. 
 
 According to this analysis, this body would appear to be tetroxyphenyl- 
 dioxypolyprene. 
 
 The formation of this body is very difficult to explain, and this difficulty 
 increases if we precipitate the phenol melts as above described by running 
 them into a large volume (4000 c.c.) of benzene. In this case precipitates 
 are obtained, the colour of which, according to the conditions under which 
 the reaction was carried out, varies from a somewhat dull crimson to a rich 
 deep shade of indigo. 
 
 In carrying out the above described reaction by heating on the water- 
 bath only, and precipitating the melt by running it into 4000 c.c. of benzene, 
 a dark-purple precipitate is obtained, separating out in very fine flakes. 
 This is allowed to settle, the bulk of the supernatant liquid decanted off, 
 the rest filtered, the filter-residue washed with benzene, and dried at from 
 40 to 50 C. The product formed is of a very fine deep indigo colour, 
 and is easily crushed into an impalpable powder. It is soluble in alcohol, 
 acetone, glacial acetic acid, and acetic anhydride. On standing, the purple 
 colour of these solutions soon changes into a pale brown, but the colour 
 of the solution in glacial acetic acid persists for a long time. The product 
 appears to be very, slightly soluble in chloroform. It is insoluble in ether, 
 benzene, and carbon bisulphide. Its analysis gave the following figures : 
 
 01742 grm. substance, . ,..- . '4472 grm. C0 2 , '1054 grm. H 2 0. 
 Calculated for C 40 H 41 10) . C, 70 -48 ; H, 6 -02. 
 
 Found, . . . . . . C, 70-09; H, 678. 
 
 1 Journ. Chem. Soc., 1888, p 686. 
 
THE CHEMISTRY OF INDIA RUBBER. 35 
 
 It will be seen that this analysis agrees tolerably well with the com- 
 position calculated for a body C 10 H 16 5 (OC 6 H 5 ) 5 . I have, however, strong 
 reasons for believing that this substance contains some slight impurity, to 
 which, probably, its colour is due, but which, so far, I have not succeeded 
 in removing. It will also be observed that this body would appear to be a 
 still more complicated compound of the type of bodies represented by the 
 above described compound C 34 H 36 6 . 
 
 I possess, at the present moment, no conclusive proof of the purity of 
 this substance, or, rather, respecting the amount and nature of its impurity. 
 But it is perhaps worth mentioning that the temperature at which it is 
 prepared may be considerably varied from 100 C. downwards without any 
 sensible alteration of the composition of the product. Thus, 5 grms. of 
 polyprene tetrabromide were heated with a solution of 50 grms. of phenol 
 in 50 grms. of benzene, this solution boiling at 82 C. The reaction pro- 
 ceeded exactly as with phenol alone, but rather slower. The product, 
 which was isolated by running the melt into a large volume of benzene as 
 before described, gave, on analysis, the following figures : 
 
 0-2031 grm. substance, . . . 0'5240 grra. C0 2 , 0'1201 grin. H 2 0. 
 Calculated for C4oH 41 10 , . C, 70 "48 ; H, 6 '02. 
 Found C, 70-40; H, 6-54. 
 
 On prolonged boiling with aqueous caustic soda, or very rapidly on 
 heating with alcoholic caustic soda, all these bodies split off phenol. On 
 acidulating these solutions with acetic acid, the alcoholic solution being 
 first diluted with much water, a flocculent brown precipitate is obtained, 
 which is collected on a filter and washed neutral with water. After drying, 
 it forms a light-brown, readily friable mass, which is easily soluble in dilute 
 aqueous caustic soda, alcohol, ether, acetone, acetic acid, acetic anhydride ; 
 insoluble in benzene, chloroform, and carbon bisulphide. The analysis of 
 this substance gave the following figures : 
 
 0-2000 grm. substance, . . . 0'4481 grm. C0 2 , 0'1218 grm. H 2 0. 
 Calculated for C^H^O^, . . . C, 61 '22 ; H, 6 "08. 
 Found, C, 61-11; H, 6 76. 
 
 This body may again be subjected to the above named treatment with 
 caustic soda, with the result that bodies with a still greater number of 
 hydroxyl groups are formed. These bodies, like those before described, 
 are precipitated from their solutions by means of acetic acid. Such a 
 preparation, which, after drying, formed a dark-brown, somewhat gritty 
 powder, the solubility of which in the solvents of this class of bodies is 
 greatly diminished, gave the following results on analysis : 
 
 0-1661 grm. substance, . 0-2501 grm. C0 2 , '1030 grm. H 2 0. 
 
 Calculated for C^H^O^, . . . C, 40 -00 ; H, 6 '66. 
 Found, C, 41-06; H, 6 '88. 
 
 On repeating the above hydrolytic treatment, successively darker 
 products are obtained in the form of more and more colloidal precipitates, 
 in form and colour greatly resembling colloidal precipitates of ferric 
 
36 INDIA RUBBER AND ITS ANALYSIS. 
 
 hydroxide. The thorough washing of these precipitates is extremely tedious. 
 After drying, they form very hard, almost black, resinous masses. I refrain 
 from quoting analyses of these bodies, but may mention incidentally that, 
 on drying, these colloidal products suffer very extensive dehydration. 
 
 I am continuing the investigation of these curious bodies, which, I 
 believe, will furnish excellent material for the prosecution of the enquiry 
 into the nature and constitution of india rubber. The reaction to which 
 we owe their formation is by no means confined to phenol. The cresols 
 carvacrol, thymol, as well as the naphthols, react in quite an analogous 
 manner. Resorcinol, under the conditions above stated, reacts most 
 energetically, forming an eosin-coloured mass. Pyrocatechol and hydro- 
 quinone react much more slowly. Of substituted phenols the action of the 
 three chlorophenols and of the three nitrophenols was examined. The 
 first named yield chlorinated reaction products ; the last named appear to 
 complicate the reaction by oxidation. Thiophenol acts in an entirely 
 different manner, by simply eliminating the bromine of the tetrabromide as 
 hydrobromic acid. Phenol-ethers, such as anisol or phenetol, do not react 
 at all. This is almost conclusive evidence that it is the hydroxyl hydrogen 
 of the phenol which forms the point of attack in the reaction in question, 
 and this fact is at the same time the reason why, in the above formulae, 
 the oxygen appears as ether oxygen, and not as hydroxyl oxygen. 
 
 Iodine. ADRIANI 1 states that the action of iodine on the globules of the 
 latex is more marked than that of bromine, but he does not appear to have 
 investigated the nature of this action. On the other hand, however, 
 GLADSTONE and HIBBERT found iodine to have little or no action upon 
 india rubber dissolved in chloroform. 
 
 My own observations on this subject tend to show that iodine does 
 indeed act upon india rubber dissolved in chloroform. A 3 per cent, 
 solution was mixed with a large excess of iodine (3 mols. of I to 1 mol. of 
 india rubber) dissolved in chloroform. After two days' standing, the 
 mixture had become gelatinous. It was thoroughly stirred and transferred 
 into absolute alcohol. An insoluble precipitate formed, which was 
 completely freed from iodine by washing with warm alcohol. In this 
 manner a bright straw-coloured product was obtained in the form of a 
 fine powder, which was found to be absolutely insoluble in all ordinary 
 solvents. This preparation on analysis gave the following figures : 
 
 0-5470 grms. gave by CARIUS' method 07273 grins. Agl. 
 
 0*2008 grrns. gave 0'1689 grms. C0 2 and '0560 grms. H 2 0. 
 
 Calculated for (C 10 H 16 I 3 ),, 
 
 Carbon, . . . 22 '35 per cent. 23-21 
 
 Hydrogen, . . 3 '07 3 '09 
 
 Iodine, . . . 72 "03 ,, 73 '69 
 
 97-45 
 The original sample of india rubber from which this preparation was 
 
 1 Chem. News, 2, 278. 
 
THE CHEMISTRY OF INDIA RUBBER. 37 
 
 made contained 3 '94 per cent, of oxygen, which is obviously responsible 
 for the deficiency in the above analysis. Nevertheless, these figures agree 
 fairly with the ratio C 10 H 16 I 3 . Of course an addition product of this 
 formula is not possible, and we must therefore assume that this tri-iodide 
 has been formed by a process analogous to that which leads to the 
 formation of the pentabromide, 
 
 Ci H 16 I 4 = C 10 H 15 I 3 + HI. 
 
 The difference of one atom of hydrogen is so small as to fall within 
 the limits of the experimental error. The splitting off of hydroiodic acid 
 in this manner is, however, perhaps not very probable, and the difficulty 
 of explaining the composition of the above addition product might very 
 likely be met by the fact that india rubber is not C 10 H 16 but (C l0 H 16 ) n , 
 so that if n is an even number, the number of iodine atoms in our 
 substitution product necessarily becomes even too. 
 
 This iodide is a far less stable substance than the bromide. Its colour 
 changes to a dull brown on prolonged exposure to light, but this change 
 of colour does not seem to be accompanied by loss of iodine. At 80 C. 
 the compound undergoes slow but complete decomposition. A dark-brown 
 resinous body remains behind, most likely an oxidation product of india 
 rubber. The iodide was found to be insoluble in all solvents. 
 
 Action of Acids. Hydrochloric Acid. Although the destructive action 
 of hydrochloric acid gas upon india rubber stoppers and tubing must have 
 been observed by thousands of chemists, the cause of this action has 
 never been investigated until quite recently by myself. 
 
 The action of hydrochloric acid gas upon india rubber dissolved in 
 chloroform differs materially according to whether moisture be present or 
 not. In the latter case I have not been able to obtain any definite result. 
 The evidence at hand appears to show that a product C 10 H 16 .HC1 
 polyprene hydrochloride is formed, but I have never been able to obtain 
 a product containing the requisite amount of chlorine. 1 Moreover, all 
 preparations obtained under these conditions form very sticky masses 
 which are very difficult to handle. Once precipitated they are very 
 insoluble, and therefore practically impossible to purify. 
 
 If the hydrochloric acid gas is passed in the undried state into the 
 india rubber solution at ordinary temperature, the result is different. The 
 absorption proceeds at first fairly rapidly, but gradually becomes slower, 
 and it is therefore necessary to allow the gas to pass through the solution 
 for a whole day. At the end of this time the solution has perceptibly 
 darkened in colour, but appears otherwise unchanged. On pouring this 
 solution slowly into absolute alcohol, it is precipitated as a mass of whitish 
 clots which are still sensibly elastic to the touch. On standing for a 
 short time under alcohol, these clots become hard and brittle, often 
 
 1 The products obtained under various conditions of temperature were found to 
 contain from 11 to 18 per cent, of Cl. The body C 10 H ]7 C1 would contain 20 '22 per 
 cent, of Cl. 
 
38 INDIA RUBBER AND ITS ANALYSIS. 
 
 breaking up spontaneously into a mass of small irregularly-shaped 
 fragments. This product is washed free from acid with warm alcohol, 
 and dried in a desiccator. It forms a perfectly white, easily friable 
 substance, which is readily soluble in chloroform, but insoluble in alcohol, 
 ether, acetone, benzene, and carbon bisulphide. On gently warming it 
 with organic bases such as aniline, pyridine, or piperidine, it dissolves 
 gradually, but this solution appears to be accompanied by loss of hydro- 
 chloric acid. The analysis of this compound led to the following 
 figures : 
 
 0*3837 grms. gave by CARIUS' method 0*5223 grms. AgCl. 
 
 0"2188 grms. gave 0*4588 grms. C0 2 and 0'1615 grms. H 2 0. 
 
 Calculated for C 10 H 18 G1 2 . 
 
 C, 57'12 per cent. 57'41 percent. 
 
 H, 8*13 8*62 
 
 01, ... 33*67 ,, 33*97 
 
 These figures agree well with those of a polyprene dihydrochloride. 
 The india rubber from which the specimen analysed was prepared con- 
 tained 2-10 per cent, of oxygen. This hydrochloride is rather difficult 
 to prepare, as it gives off part of its hydrochloric acid very readily, so 
 that in many preparations the percentage of chlorine fell as low as 31 
 per cent. On the other hand, it is worth mentioning that as soon as the 
 decomposition of the dihydrochloride has reached a point that the amount 
 of residual chlorine is about 18 per cent., even a very considerable increase 
 of temperature will cause only a very insignificant further reduction in 
 the percentage of chlorine. Seeing that 18 per cent, of Cl comes very 
 near the amount of chlorine required for polyprene monohydrochloride, 
 we may perhaps infer that this forms the chief product of the decomposition 
 of the dihydrochloride at temperatures not exceeding 100 C., and that, 
 compared with the latter, it is a rather stable body. 
 
 Hydrobromic Acid. The action of hydrobromic acid upon india rubber 
 is much the same as that of hydrochloric acid, although rather slower. 
 It has been found impossible to prepare a hydrobromide corresponding to 
 the above described dihydrochloride. On passing hydrobromic acid gas 
 into a solution of india rubber in chloroform, and subsequently running 
 this solution into alcohol, a product is obtained which in appearance greatly 
 resembles the dihydrochloride, but already on washing with alcohol it 
 begins to lose hydrobromic acid. I have never succeeded in preparing a 
 compound with more than 21 per cent, of bromine, and the appearance 
 of all preparations I obtained clearly indicated their heterogeneous 
 nature. 
 
 Hydriodic Acid. This acid seems to have no action upon india rubber, 
 
 or, if formation of a hydro-iodide does take place, this compound must 
 
 be extremely unstable. According to BERTHELOT, 1 treatment of india rubber 
 
 with hydriodic acid at 280 C. results in the formation of viscous hydro- 
 
 1 Bull. Soc. chim., 11, p. 33. 
 
THE CHEMISTRY OF INDIA RUBBER. 39 
 
 carbons of the paraffin series, boiling without decomposition above 350 C. 
 This statement requires confirmation. 
 
 Sulphuric Acid. The concentrated acid acts very energetically upon 
 india rubber, charring and oxidising it. Nothing definite is, however, 
 known respecting the ultimate products of this destructive action. The 
 action of dilute sulphuric acid has never been investigated. 
 
 Nitric Acid. Strong nitric acid attacks india rubber very vigorously, 
 forming at first a yellow body which is subsequently decomposed into 
 nitrogen, carbonic acid, oxalic acid, and a substance of the character of a 
 fat. 1 On prolonged boiling, this latter body is dissolved and converted 
 into camphresic acid, which was shown by KACHLBR 2 to be a mixture of 
 camphoric and camphoronic acid. 
 
 Nitrous Acid. In treating a dilute solution of india rubber with 
 nitrous acid, produced by very slowly running concentrated sulphuric acid 
 into a 20 per cent, solution of sodium nitrite, an amorphous yellow body is 
 'formed, insoluble, or very little soluble, in most of the commonly employed 
 solvents, but fairly soluble in dimethyl oxalate. The composition of this 
 body, as obtained in different preparations, varies very greatly, but by 
 passing the nitrous acid gas, prepared as described above, through a series 
 of drying tubes charged with phosphoric anhydride, and by using absolutely 
 dry india rubber dissolved in benzene, which has been freshly distilled 
 over metallic sodium, a product is obtained showing the following 
 composition : 
 
 C 55'5, H 7-0, N 13-1. 
 Calculated for C 10 H ]6 N 2 3 : C 56-6, H 7'5, N 13 '2. 
 
 The product of the reaction is evidently polyprene nitrosite. The 
 full investigation of this highly interesting body is not yet concluded. 
 
 Nitrogen Dioxide. In treating a solution of one part of india rubber 
 in a hundred parts of absolutely dry benzene with nitrogen dioxide, a 
 considerable rise of temperature takes place, and the whole of the india 
 rubber present separates out in the form of a coherent amorphous mass. 
 The nitrogen dioxide is most conveniently produced by heating well dried 
 lead nitrate in a retort. The gas evolved must be dried by passing it 
 through a series of tubes containing phosphoric anhydride. 
 
 The above described operation being completed, the benzene, which 
 will be found to contain mere traces only of solid matter, is poured off the 
 product of the reaction. The latter forms a very friable mass, which, 
 after washing with alcohol, is obtained as a dark -yellow powder. This 
 product is, however, not yet pure. It dissolves in acetone with the 
 greatest ease, forming a deep sherry-coloured solution containing in 
 suspension a small quantity of an insoluble substance. This is filtered off, 
 and the perfectly clear solution run into water. The very fine precipitate 
 thus obtained is collected on a filter, washed with water, pressed, and dried, 
 
 1 SCHWANERT, Ann., 128, p. 123. 2 Ann., 191, p. 143. 
 
40 INDIA RUBBER AND ITS ANALYSIS. 
 
 when it is obtained as a straw-coloured impalpable powder, which, on 
 analysis, gave the following figures : 
 
 C 52-5, H 7-1, N 12-3. 
 Calculated for C 10 H 16 N 2 4 : C 52 -6, H 7'0, N 127. 
 
 Whether this body represents a dinitro-polyprene C 10 H 16 (NO 2 ) 2 , or a 
 nitroso-nitrate (nitrosate) similar in constitution to WALLACE'S l amylene- 
 
 nitrosate, 
 
 ,NO 
 
 O.N0 2 
 
 further investigation will have to show. 
 
 The product purified as above described is insoluble, or only very 
 slightly soluble, in ethyl alcohol, methyl alcohol, ether, chloroform, and 
 carbon bisulphide. It dissolves almost instantly in acetone, and is also fairly 
 soluble in ethyl acetate, ethyl formate, dimethyl oxalate, epichlorhydrine, 
 and nitrobenzene. Ammonia dissolves it very readily even in the cold, but 
 it is almost insoluble in either aqueous or alcoholic soda. 
 
 This product, as already mentioned, is insoluble in methyl alcohol, but, 
 on adding to the latter a few drops of very concentrated caustic soda, or, 
 better still, a small quantity of sodium methylate, solution takes place 
 instantly. Curiously enough, ethyl alcohol cannot be used in this reaction 
 instead of methyl alcohol. The action of the alkali upon the polyprene 
 derivative appears to be the same in both cases, but the product of the 
 reaction is soluble only in methyl alcohol, and not in ethyl alcohol. Indeed, 
 on adding to its solution in methyl alcohol an equal volume of ethyl 
 alcohol, a flocculent precipitate is at once formed. The investigation of these 
 bodies is still proceeding. 
 
 Nitrosyl Chloride. On treating a solution of 13'6 grms. of polyprene in 
 300 c.c. of benzene with 12 grms. of amyl nitrite and 8 grms. of acetyl 
 chloride, a reaction takes place, resulting in the formation of a very stiff jelly. 
 After about 20 minutes' standing gas bubbles appear in the jelly, which 
 rapidly grow in size and number. In the course of a few hours the jelly is 
 reconverted into a solution distinctly thinner than the original rubber 
 solution. On precipitating this solution, a clot of a substance is obtained, 
 which, after purification, by again dissolving and precipitating it, was found 
 to be apparently unaltered polyprene containing traces only of chlorine and 
 nitrogen. A polyprene nitrosyl chloride could in no way be obtained. 
 
 Oxygen. The action of oxidising agents, such as chromic acid, potassium 
 permanganate, persulphates, and percarbonates, upon india rubber has 
 never been systematically investigated. Ethereal solutions of hydrogen 
 peroxide are stated to convert india rubber into products soluble in alcohol, 
 but I have never been able to confirm this result. 
 
 The liability of both crude and vulcanised india rubber to deteriorate 
 under the influence of atmospheric oxygen has been known for some con- 
 1 Ann. Chem., 241, 288 [1887]. 
 
THE CHEMISTRY OF INDIA RUBBER, 41 
 
 siderable time, and was investigated by SriLLER, 1 who isolated a product 
 soluble in alcohol and chloroform, melting below 100, and which possessed 
 the following percentage composition : 
 
 C, 64-00 per cent. 
 
 H, 8-46 
 
 0, ... 27-54 
 
 It is perhaps worth pointing out that this composition fairly closely 
 corresponds with C 30 H 48 10 . This may be accidental, but as this body still 
 shows the ratio C 10 : H 16 , this formula, at any rate, appears to prove that this 
 oxidation process is purely an addition process. 
 
 SPILLER'S observations were made on a thin film of manufactured but 
 unvulcanised india rubber. Articles of this description are, under normal 
 conditions, only very slowly affected in this manner, owing to their non- 
 porosity and smoothness of surface. Crude india rubber, particularly in 
 the state of the highly porous and reticulated sheets in which it leaves the 
 washing-machine for the drying-room, is apt to suffer considerably from 
 oxidation during the drying process. 
 
 The Vulcanisation of India Rubber. 
 
 Historical Introduction. From a technical point of view, the action 
 of sulphur and of sulphur monochloride upon india rubber is of funda- 
 mental importance. It may well be said that to the action of these two 
 compounds the india rubber industry of to-day chiefly owes its existence 
 and extent. 
 
 1. Vulcanisation with Sulphur. In 1832 LI)DERSDORF made the 
 observation that sulphur removes the natural stickiness of india rubber, 
 and at about the same time an American, HAYWARD, employed flowers of 
 sulphur to counteract this stickiness of india rubber sheets. It was, 
 however, unquestionably NELSON GOODYEAR who, in 1839, first definitely 
 established the fact that the treatment of india rubber with sulphur at 
 high temperatures has the effect of maintaining the elastic properties 
 constant within a comparatively very wide range of temperatures, and who 
 subsequently also discovered the manufacture of ebonite, the final product 
 of the action of sulphur upon india rubber. In England HANCOCK appears 
 to have independently made the discovery of the vulcanising action of 
 sulphur, but as, according to his own statement, 2 he had seen samples 
 vulcanised by Goodyear, the priority of this invention clearly belongs to 
 the latter. HANCOCK'S narrative of his countless attempts to produce such 
 a product, and the manner in which he finally succeeded, is so fascinating, 
 that an extract of it from his above quoted book, which is by no means 
 readily obtainable, may be acceptable. 
 
 " Some time in the early part of the autumn of 1842, Mr. Brockedon 
 
 1 Journ. prdkt. Chem., 94, 502. 
 
 - HANCOCK, Personal Narrative of the Origin and Progress of the Caoutchouc or India 
 Rubber Manufacture in England, London, 1857, p. 94. 
 
42 INDIA RUBBER AND ITS ANALYSIS. 
 
 showed me some small bits of rubber that he told me had been brought 
 by a person from America, who represented himself as the agent of the 
 inventor : it was said that they would not stiffen by cold, and were not 
 much affected by solvents, heat, or oils. Mr. Brockedon told me that the 
 mode of manufacturing this rubber was a secret, and that the agent who 
 had shown them to him, declared himself totally ignorant of it; all he 
 knew was, that it had been done by some new solvent which was very cheap 
 in America, and that he wished to find parties who would purchase the 
 secret in this country, and asked Mr. Brockedon to give him any intro- 
 duction he could to the trade. Mr. Brockedon gave him the names of 
 Mackintosh & Co., and the representatives of the Caoutchouc Company 
 (and I think at least one other), with whom the party also left specimens. 
 Mackintosh & Co. told the agent that, as he could give no information, 
 they could not judge of the merits of the invention, as it might be easy to 
 make these small specimens, whilst difficulties might be found in its 
 application upon a large scale ; nor could they judge whether their present 
 appliances, which had been very costly, would be suitable for this manu- 
 facture, or whether any modification of their present plant would answer, 
 or whether the modes were so altogether new as to require a fresh outlay. 
 Under these circumstances, as they could not act in the dark, they 
 recommended the taking out of a patent, when a clear comprehension of 
 the whole could be obtained, and the invention openly dealt with according 
 to its merits. This course was approved of, and the agent said he would 
 immediately advise his principal to take out a patent. 
 
 "Mr. Brockedon cut, from two of the small pieces which had been 
 given him, two very diminutive bits, and gave them to me ; one bit was 
 about one and a half or two inches long, and one inch and a quarter wide, 
 and perhaps the twentieth of an inch thick. This bit, on the exterior, was 
 of a dirty yellowish-grey colour, and a little dusty powder upon it ; when 
 cut across, the cut edge appeared of a dark colour. The other bit was of a 
 dark reddish-brown colour throughout, with a clean surface" without any 
 dusty appearance. The first named had a slight smell of sulphur, and I 
 thought that a little sulphurous powder had been rubbed on to mislead. 
 On stretching them out thin, I observed they were both quite opaque. I 
 found they were, as stated, not affected by cold. 
 
 " Finding now that this object had been somehow or other effected, and 
 therefore demonstrated to be practicable, and as it was my particular 
 department to keep up the quality of our manufactures, and to maintain 
 our standing and the position our goods had attained, I set to work in 
 earnest, resolved, if possible, not to be outdone by any. As I advanced, 
 I spent all my spare time to attain the two objects I had in view. 
 
 "I succeeded in producing beautiful drab- and tan-coloured surfaces, 
 perfectly inadhesive, by combining silicate of magnesia, and compounds of 
 this silicate, with fuller's earth, whiting, ochre, and other earths, in all 
 of which the silicate of magnesia formed the foundation and predominated. 
 
THE CHEMISTRY OF INDIA RUBBER. 43 
 
 For a dark colour I employed asphalt, both the natural mineral kinds as 
 well as that obtained by boiling down coal-tar to the state of asphalt. 
 With this asphalt I also sometimes combined plumbago, and by these 
 means produced an unadhesive black surface. Since that time, a taste for 
 a more glossy article prevailing, varnishes have been added. In carrying 
 out these improvements, no new machinery was needed, the rubber and 
 compounds being made, and the straining and spreading effected, by the 
 means before mentioned. 
 
 " During my progress with these experiments, I gave attention also to 
 my other object, and having nearly brought the former to a close, I could 
 give to the latter more special attention. The little bits given me by Mr. 
 Brockedon certainly showed me for the first time that the desirable change 
 in the condition of rubber of not stiffening by cold had been attained, but 
 they afforded no clue to the mode by which it had been brought about. 
 
 " As the knowledge of this improvement in the qualities of rubber soon 
 spread abroad, several persons began to make experiments in order to 
 discover means to effect the same ; but as there were very few who at that 
 time had any considerable experience in the treatment of rubber beyond 
 the ordinary applications for waterproofing purposes and the construction 
 of pneumatic articles, they were not very likely to succeed in an attempt 
 which required, besides an experimental knowledge of the subject, unwearied 
 application and the closest observation of the effects of every new addition 
 to the compounds, and of the proportions indicative of improved results 
 nor were high temperatures employed by such persons. Mr. A. Parkes, for 
 instance, was not much acquainted with the rubber manufacture, and 
 therefore, after he had seen my specification, and the change that had been 
 effected by sulphur and heat, went immediately to work as a chemist, and 
 tried the effect of the whole round of sulphurous combinations with the 
 rubber, and for a time gave up the pursuit ; but being urged to continue 
 his experiments, he at length succeeded, by purely chemical means, in 
 effecting a change in rubber very closely resembling vulcanisation. 
 
 " It is a singular fact that although sulphur had long been compounded 
 with rubber by myself in this country, and in America by others, yet that 
 its true value for producing such a result had never been even dreamt of 
 by any of us. I made no analysis of these little bits, nor did I procure, 
 either directly or indirectly, any analysis of them. In making my experi- 
 ments, I depended entirely and solely on my own exertions, having some 
 confidence in an experience of upwards of twenty years of unceasing appli- 
 cation to the manipulation with my own hands of the substance I was 
 dealing with ; and it having been shown that rubber could be so made as 
 not to stiffen by cold, I devoted myself to the discovery, if possible, of some 
 mode by which this property could be imparted; and I considered the 
 small specimens given to me simply as a proof that it was practicable. It 
 had been my constant practice to make my experiments alone in my 
 private laboratory at Stoke Newington, into which, on these occasions, no 
 
44 INDIA RUBBER AND ITS ANALYSIS. 
 
 person entered but myself. I consequently lighted my own fires, and did 
 all the labour myself; and having constantly to attend to the general 
 business in London, and to answer innumerable enquiries, both verbally 
 and by letter, besides attending consultations with parties taking out 
 patents, or others requiring information in the application of our materials 
 to their different purposes, I was obliged in a great measure to pursue my 
 experiments after the hours of business, and was frequently employed till 
 midnight in these labours. I think it necessary to mention this, because 
 it has been alleged that nothing could have been less difficult than such 
 a discovery, and that anyone acquainted with the substance might have 
 accomplished it with great ease. After a discovery is once made, it is 
 generally said to be easy of attainment ; anybody might have done it, the 
 thing was so simple. According to Milton, an idea of this kind occurred 
 to some as far back as the council in Pandemonium : 
 
 ' The invention all admired ; and each how he 
 To be th' inventor missed, so easy it seem'd, 
 Once found, which yet unfound most would have thought 
 Impossible. ' 
 
 "But to resume, I knew nothing more of the composition of the 
 small specimens given me by Mr. Brockedon than what I or any other 
 person might know by sight and smell. In such a pursuit I was obliged 
 to grope a good deal in the dark ; for although my former experience, such 
 as heat, masticating, rolling, and the use of solvents, served as landmarks, 
 still my experiments could only be made pretty much at random, not at 
 all knowing what the result of any of them might be. I remember that 
 I had from the first a strong impression that the rubber underwent the 
 change either when in a state of solution or when greatly softened by 
 heat, and these two points I kept pretty much in view. I compounded 
 with the rubber in one or other of these conditions an almost endless 
 round of matters of all kinds. 
 
 "I treated these with the rubber separately and in innumerable 
 combinations. In some of them I included sulphur, and employed heat in 
 almost every case without regarding what degree of heat I used. I have, 
 in an early part of my narrative, stated how useful I found it to employ 
 high temperatures, and attribute my final success in my present pursuit to 
 the habit I had acquired in its use, particularly when I thought it necessary 
 to reduce the rubber to the softest or most plastic state without solvents. 
 When making the compounds with solvents, I first dissolved the rubber, 
 either making the solution of a thin consistence, or of the consistence of 
 dough, and then mixed or worked up the other matters with it. When I 
 wished to expedite their drying, I sometimes laid them on a small metal 
 plate heated over a chamber lamp, or larger plates heated by the fire, and 
 at other times I submitted them to the heat of an oven. As before 
 mentioned, I frequently employed sulphur in the compounds, but they 
 were not at all improved by it, not having yet by any chance used heat 
 
THE CHEMISTRY OF INDIA RUBBER. 45 
 
 sufficiently high when drying them to produce the change ; or, if the heat 
 was sufficient, which it most likely sometimes was, the compound became 
 dry in too short a period to affect it (as I removed the pieces as soon as 
 they were dry) in any way to attract my observation, having nothing to 
 guide me in the object in view. I therefore for a time relinquished the 
 use of sulphur in most of the compounds as useless, and pushed on with 
 other matters, still feeling a conviction, as I think most would naturally 
 have done, that the rubber must necessarily undergo the change in its 
 constitution whilst either very soft and plastic or in a state of solution. 
 However, after trying an endless round of mixtures in this way, with and 
 without heat, I still failed of success. 
 
 " Whilst looking over some of my former experimental scraps, I saw 
 in some of those containing sulphur, variations I could not at the time 
 account for : some portions were different to others in the same specimen, 
 which for the present I could not comprehend, although 1 well knew 
 afterwards. I resolved now to take another course ; I dissolved sulphur 
 in oil of turpentine, and finding the solution proceed slowly, I raised 
 the temperature of the turpentine to the boiling point (316), and then 
 it took up the sulphur freely. Rubber dissolved in this solution was good 
 in some respects, but produced a weak material, and did not come up to 
 my wishes. 
 
 " I spent all my spare time for months with these experiments ; my 
 habit was to despatch them quickly, making them very small in bulk, 
 throwing aside some thousands of trial scraps, and selecting and keeping 
 for inspection any that appeared promising. During the winter months 
 I generally found the weather cold enough to test my scraps, but as the 
 spring and summer came on, I employed ice, purchased from an ice-cart 
 which passed my gate every morning, and many an earnest and careful 
 examination have I made in a morning of the scraps in the ice. 
 
 " My experiments now had become very interesting ; I had certainly 
 produced in some of my scraps, or portions of some of them, that condition 
 of rubber which I afterwards called the * change.' 
 
 " I cannot, of course, now say in which scrap or at what particular 
 moment I struck out the first spark of the 'change,' because I had a great 
 number of scraps under my hand at one time, and I also found that when 
 exposed even to ice, the stiffening effect of cold was not always immediately 
 apparent ; and I was thus, during my comparative ignorance of the nature 
 of my discovery, frequently misled and disappointed from two causes : first, 
 what it was in the successful compounds that produced the effect; and 
 secondly, I was ignorant of the importance of the degrees of temperature 
 employed, as well as of the like importance of the period to which any 
 specific compound required to be submitted to heat. 
 
 " These points could be ascertained with exactness only by time and 
 vigilant watching, and this I began immediately to set about ; and as the 
 law allows a patentee six months to work out his discoveries before he is 
 
46 INDIA RUBBER AND ITS ANALYSIS. 
 
 called upon to enrol his specification, I applied for and obtained a patent 
 for my inventions, which passed the great seal on the 21st November 1843. 
 
 "I found that, when submitting the compounds containing sulphur to 
 heat, it was necessary, after ascertaining the temperature that suited any 
 compound, to find also the period of exposure to the heat that produced the 
 best result. Until I noticed the necessity for this particularity, I was often 
 sadly perplexed, as the same compounds exposed to the same temperature 
 were sometimes good and sometimes bad in practice : the variation is from 
 one to six or seven hours, or more. All the way through these experiments 
 for producing the ' change,' I had no other guide, of course, than to watch 
 for any promising appearance in any of the scraps, and to improve upon 
 them ; but I now know I was frequently thwarted by my want of informa- 
 tion as to what cause the different appearances were due, and particularly 
 in regard to the temperature I employed, which was somewhat at random, 
 knowing how freely I could use it within certain limits without injury. 
 
 ."A thought now occurred to me that in the end proved extremely 
 valuable. Revolving in my mind some of the effects produced by the high 
 degrees of heat I had employed in making solutions of sulphur and rubber, 
 as before stated, in oil of turpentine, it occurred to me that as the melting 
 point of sulphur was only about 240, which I knew would not be injurious 
 to the rubber, it would be well to see what would ensue on immersing a 
 slip of sheet rubber in sulphur at the lowest melting point. I accordingly 
 melted some sulphur in an iron vessel and immersed in it some slips of cut 
 sheet rubber about half an inch wide and about one-sixteenth of an inch 
 thick. After they had remained some time I examined them, and found the 
 surface had assumed a yellowish-tan colour. I immersed them again ; and 
 on withdrawing them the second time, I cut one of them across with a wet 
 knife, and found that the rubber was tinged of this tan colour to a con- 
 siderable depth. I immersed them again ; and on the third examination I 
 found the tan colour had quite penetrated through the slip. This was 
 strong evidence that the rubber had freely absorbed the sulphur, and I 
 fully expected to find that these slips were now * changed,' but in this I 
 was greatly disappointed, for, on applying the tests, I found that not the 
 least 'changing' effect had been produced. I now replaced them and 
 raised the temperature of the sulphur, and allowed them to remain a con- 
 siderable time ; and on withdrawing one of them the fourth time, I found 
 to my great satisfaction that it was perfectly 'changed,' retaining the 
 same tan colour throughout. The other slips remained in the sulphur whilst 
 this examination was going on, and on withdrawing them, I found the lower 
 and nearest the fire turning black and becoming hard and horny, thus at 
 once and indubitably opening to me the true source and process of pro- 
 ducing the * change ' in all its pure and pristine simplicity." 
 
 The action of sulphur upon india rubber depends upon a great number 
 of factors, the most important of which is heat. Whether the action be 
 induced by immersing the india rubber into molten sulphur, or by ex- 
 
THE CHEMISTRY OF INDIA RUBBER. 47 
 
 posing mixtures of india rubber and sulphur to temperatures above the 
 melting point of sulphur (114 C.), no marked action takes place until the 
 temperature has passed 120 C. Even then, what reaction does take 
 place proceeds with extreme slowness. For practical purposes the tempera- 
 ture employed ranges from 125 to 360 C. 
 
 The amount of sulphur employed in practice varies very greatly, from 4 
 up to 50 per cent, j but the amount of sulphur fixed by the india rubber very 
 rarely exceeds 3 per cent., and is generally less. It is therefore obvious that 
 free sulphur in various proportions always forms a constituent of ordinary 
 vulcanised india rubber goods. 
 
 The vulcanisation process is considerably accelerated, and in several 
 respects distinctly facilitated, by the presence of sulphur carriers. In this 
 respect antimony pentasulphide, arsenic pentasulphide, arsenic trisulphide 
 (orpiment), and arsenic disulphide (realgar) prove most effective, and, in a 
 lesser degree, lead thiosulphate, lead sulphide, zinc sulphide, lead oxide 
 (litharge), and lead carbonate (white lead). 
 
 It is, however, not only by means of free sulphur that the vulcanisation 
 of india rubber can be effected. GERARD showed that this can also be done 
 by heating the india rubber with solutions of the alkaline polysulphides. 
 Indeed, this process renders it possible to effect vulcanisation at temperatures 
 very little above the boiling point of water. 
 
 The nature of the vulcanisation process is still very little understood, 
 most of the writers expressing themselves very guardedly upon this point, 
 and attributing vulcanisation to 'absorption' of sulphur. BURGHARDT l 
 suggests that vulcanisation consists " in the alteration of the caoutchouc 
 resin through the production in it of a * sulpho-compound,' a certain amount 
 of hydrogen in the caoutchouc being replaced by sulphur." BURGHARDT 
 evidently overlooks the fact that this assumption would imply the formation 
 of 34 parts of hydrogen sulphide for every 32 parts of sulphur combining 
 with the india rubber. If this view was correct, the amount of hydrogen 
 sulphide produced by an india rubber factory vulcanising only 5 cwt. of 
 india rubber per day would amount to considerably over 100 cubic feet per 
 day. There would be, to say the least of it, strong evidence of the evolu- 
 tion of such a volume of this obnoxious gas in the vulcanising rooms, which, 
 under normal conditions and circumstances, is never the case. Numerous 
 experiments I conducted to detect the evolution of this gas in the vulcanis- 
 ing process under laboratory conditions failed to disclose the formation of 
 even traces of hydrogen sulphide. 
 
 Very likely BURGHARDT'S view was founded upon a statement by PAYEN 
 on this point, which is to the following effect : " With the beginning of 
 this reaction, and during the whole of the time it is in progress that is, at 
 a temperature of from 135 to 145 C. combination of the sulphur with a 
 small quantity of the hydrogen of the organic substance is going on, result- 
 ing in the continuous formation of hydrogen sulphide, of which sulphur can 
 1 Thorpe's Diet, of Appl. Chem., vol. ii. p. 312. 
 
48 INDIA RUBBER AND ITS ANALYSIS. 
 
 absorb nearly its own volume. A curious phenomenon results from this 
 fact if the vulcanisation is carried out in a bath of melted sulphur : as 
 soon as, on completion of the operation, the temperature of the bath is 
 allowed to fall, the crystallising sulphur liberates the hydrogen sulphide, and 
 the evolving gas causes the semi-fluid mass of sulphur to rise." 
 
 This statement is not very conclusive concerning the point in question, 
 and assuming that PAYEN employed in his experiments india rubber which 
 had not first been freed from all resinous and oily matters, the evolution of 
 hydrogen sulphide in small quantities would thereby be explained. How- 
 ever, PAYEN'S experiments led him to consider the vulcanising process as a 
 substitution of sulphur for hydrogen. 
 
 HEINZEKLING expresses himself as follows on the vulcanising question : 
 " It is not known with certainty whether the chemical union of sulphur 
 with india rubber is an addition or a substitution process. That chemical 
 combination does take place seems to follow with necessity from the fact 
 that, whereas before vulcanisation both india rubber and sulphur are soluble 
 in carbon bisulphide, they are rendered completely insoluble by this 
 process." 
 
 DONATH'S views upon this subject are interesting. He seems to consider 
 the vulcanising process, not as a purely chemical phenomenon expressible in 
 definite terms of molecular interaction, but as a phenomenon of molecular 
 aggregation in indefinite proportions, and in many respects comparable to 
 metallic alloys. Against the assumption of definite compounds of sulphur 
 and india rubber, he objects that vulcanisation can be carried out by means 
 of metallic sulphides as readily as with free sulphur, which fact would 
 necessitate the assumption that the sulphides of antimony, mercury, and lead 
 give up their sulphur to the india rubber. 
 
 TH. SEELiGMANN 1 assumes "that the phenomenon which is understood 
 by the name of vulcanisation is only the first of a series of successive 
 transformations occurring at definite temperatures and in the presence of 
 excess of sulphur in the hydrocarbons which constitute india rubber. 
 The terminal stage of this transformation series is reached with ebonite, 
 which, if we adopt the views of C. 0. WEBER, would represent one or more 
 double atoms of polyprene linked together by one or more atoms of 
 sulphur. 
 
 "To us vulcanisation appears simply as an operation by which, at the 
 most opportune moment and at a certain stage, a chemical reaction is 
 arrested, the completion of which would lead to a vastly different product 
 than the one it is desired to obtain." 
 
 The only element of progress in SEELIGMANN'S view, as it appears to me, 
 is his somewhat timid adoption of the view of the nature of vulcanisation 
 first stated by me some years ago. 2 For a satisfactory theory of vulcanisa- 
 tion this is scarcely sufficient. 
 
 1 TH. SEELIGMANN, Le Caoutchouc et la Gutta Percha, Paris, 1896, p. 178. 
 
 2 Soc. Chem. Ind. 
 
THE CHEMISTRY OF INDIA RUBBER. 49 
 
 SBELIGMANN believes that an argument in favour of his views is 
 furnished by an experiment by PAYEN, which I quote in full : 
 
 "Ether and carbon bisulphide, on prolonged standing with vulcanised 
 india rubber, dissolve about 4 to 5 per cent., which can be isolated and 
 separated from free sulphur by evaporation and re-solution in ether. The 
 rubber obtained from the ethereal solution yields to absolute alcohol from 
 1 to 1 '5 per cent, of fatty matter. 
 
 "The india rubber thus extracted can be separated into two parts, 
 one of which is very ductile and soluble in benzene, from which it can 
 be obtained by evaporation, while the other part is tougher, less elastic, 
 and insoluble. These two parts are derived from the inner parts of the 
 vulcanised rubber where the compound is less resistant and poorer in 
 sulphur than the parts near the surface. 
 
 "After its vulcanisation there exist in the india rubber still two parts 
 possessing different degrees of cohesion and solubility. This becomes 
 noticeable on keeping a piece of vulcanised rubber tape immersed during 
 two months in a mixture of carbon bisulphide and absolute alcohol 
 (10 : 1). The solution thus obtained is evaporated to dryness, the free 
 sulphur removed by treatment with caustic soda, and we obtain then the 
 less aggregated part of the organic substance (our * adhesive ' hydrocarbon 
 transformed by sulphuration) possessing little strength, yellowish in colour, 
 and translucent. The insoluble part of the tape is found to have preserved 
 its form, but to have turned brown and less transparent (our more 
 sinewy l hydrocarbon undiluted by the adhesive and soluble part). These 
 constituents, omitting the fatty matter, were obtained in the following 
 proportions : 
 
 Insoluble, sinewy portion, .... 65 per cent. 
 Soluble, adhesive portion, . . . . 25 ,, 
 Free sulphur, . -, . , .__ . 10 ,, " 
 
 I have no reason to doubt that PAYEN effected a separation of vul- 
 canised rubber into a soluble and an insoluble part in the above propor- 
 tions, but I emphatically deny that this separation bears out in the least 
 the important conclusions PAYEN based upon it. The extent to which 
 the fact of the existence of an insoluble constituent in india rubber has 
 inspired the imagination of our French colleagues is truly astonishing. 
 But even more so is the uncritical manner in which the amount of this 
 insoluble matter in india rubber has been exaggerated to about ten times 
 its real value, assuming at the same time for this insoluble part identity 
 of elementary composition with that of the soluble part, and all this in 
 the absence of any reliable analytical data of any sort. 
 
 PAYEN'S above described experiment, and its entirely erroneous inter- 
 pretation, prove nothing for or againt SEELIGMANN'S before stated views 
 regarding the nature of vulcanisation. They appear to me merely as an 
 attempt to demonstrate the persistence in the india rubber even after 
 1 The original has ' nerveux.' 
 
 4 
 
5O INDIA RUBBER AND ITS ANALYSIS. 
 
 vulcanisation of its * sinewy ' and of its insoluble, as well as of its 
 adhesive and soluble, constituents. We have, however, seen 1 that the 
 insoluble part of india rubber (Para) forms a very insignificant portion 
 only of the whole, and we have also disposed of the current assumption 
 that the composition of this insoluble part is identical with that of the 
 rest of the india rubber. 
 
 I have attempted to repeat PAYEN'S above described experiment in 
 order to make an exhaustive examination by analysis of the products of 
 this separation. A specimen of Para rubber vulcanised with 10 per cent, 
 of sulphur at 135 C. during two hours was employed. This specimen 
 contained after vulcanisation 3 '98 per cent, of combined, and 5 '96 per cent, of 
 free, sulphur. It was cut into very fine threads, which were during three 
 months kept submerged in PA YEN'S mixture of carbon bisulphide and 
 alcohol (10 : 1). At the end of this time the threads, although much 
 distended and swelled, showed not the slightest sign of disintegration 
 such as PAYEN observed, and after drying appeared practically unaltered. 
 Assuming that the failure of this experiment might be due to the degree 
 of vulcanisation of the sample being too high, another sample, vulcanised 
 at 125 C. for two hours, was treated in exactly the same manner. This 
 sample contained 1 -9 per cent, of combined sulphur. The result, however, 
 was exactly the same. In order to ascertain finally whether a desintegra- 
 tion of the kind described by this author could be effected by the action 
 of his special solvent, the last named of the above mentioned samples, 
 together with the solvent in which it had been immersed, was for a whole 
 week heated to boiling point on a reflux condenser. This experiment 
 also failed to produce the effect upon the india rubber described by 
 PAYEN. The extracts obtained in every case consisted of the free sulphur 
 contained in the sample, together with the resinous matter of the india 
 rubber amounting to less than 3 per cent, of the mass of the sample 
 operated upon. 
 
 A still more recent attempt to explain the vulcanisation phenomena 
 is due to J. MINDER. 2 His fundamental views are expressed in the 
 following three sentences : 
 
 1. "Vulcanisation is produced by sulphur in the state of a vapour. 
 In this form it penetrates into the rubber molecule, thus vulcanising the 
 india rubber." 
 
 2. "Already at ordinary temperatures sulphur possesses a sensible 
 vapour tension, which explains the fact that thin sheets of india rubber 
 containing sulphur undergo spontaneous vulcanisation at ordinary 
 temperature when left to themselves for a long time." 
 
 3. "In order to accelerate vulcanisation, the tension of the sulphur 
 vapours, and their density, must be increased. This may be attained 
 either by heating the sulphur and immersing the india rubber in it 
 (sulphur bath), or by mixing the india rubber with sulphur and heating 
 
 1 See page 8. 2 Gummi Zeitung [14], 1900, 247. 
 
THE CHEMISTRY OF INDIA RUBBER. 51 
 
 the mixture (dry heat cure, steam cure), or by dissolving in a suitable 
 solvent chemically combined sulphur vapours, and allowing such solutions 
 to act upon india rubber." 
 
 These statements suffer from a certain vagueness of diction and want 
 of definition. This particularly refers to the first of them, and so, by 
 implication, to the second likewise, although they are, in my opinion 
 unquestionably correct as far as they go. The third of these sentences 
 I consider in every respect erroneous ; indeed, it is in part quite 
 unintelligible. It is an everyday experience that to accelerate the 
 vulcanisation process we must increase the temperature. This will 
 incidentally result in an increase of the vapour tension of the sulphur, 
 but we possess no evidence whatever that this vapour tension itself is the 
 determining factor of the rate of vulcanisation. Vapour tension is a 
 purely static condition which is devoid of any influence upon chemical 
 reactions, although, if a change of vapour tension involves a change of the 
 molecular (chemical) state of the vapour, such an influence may become 
 apparent, but is even then due entirely to the latter cause. So that, if we 
 produce the latter change without changing the former, or even changing 
 this in the opposite sense, the accelerating effect of the molecular change 
 would nevertheless be there. 
 
 Entirely erroneous is MINDER'S view as to the necessity of increasing 
 the density of the sulphur vapours in order to increase their vulcanising 
 action, and still worse is his assumption that this increase of density can 
 be achieved by raising the temperature of the sulphur vapours. 
 
 Equally unfortunate is MINDER'S explanation of the curing action of 
 chloride of sulphur as quoted above in the third of his statements. To 
 begin with, his description of chloride of sulphur as * chemically combined 
 sulphur vapours' is rather peculiar, and he altogether omits to explain 
 why 'chemically combined sulphur vapours' should combine with india 
 rubber at all. Indeed, the evidence of the scores of liquid and gaseous 
 sulphur compounds, to all of which his description of the chloride of 
 sulphur would apply likewise, would tend to prove that ' chemically 
 combined sulphur vapours ' do not combine with india rubber. 
 
 2. Vulcanisation with Chloride of Sulphur. The process of vulcanising 
 india rubber with chloride of sulphur sulphur monochloride S 2 C1 2 was 
 discovered by PAKKES in 1846, and is now technically known as the 'cold 
 cure,' to distinguish it from the 'dry heat cure' or 'steam cure,' as the 
 processes of vulcanising with sulphur are variously called. It is carried 
 out by treating the objects to be vulcanised with a from 1'5 to 3 per 
 cent, solution of sulphur monochloride in carbon bisulphide. To a very 
 limited extent, chiefly for the vulcanisation of certain kinds of double- 
 faced sheets and waterproof cloth, a 'vapour cure' is also in use, in 
 which the goods to be cured are hung in, or continuously carried through, 
 a chamber which is either charged with the vapours of chloride of 
 
52 INDIA RUBBER AND ITS ANALYSIS. 
 
 sulphur, or in which chloride of sulphur is slowly evaporated at a definite 
 rate. 
 
 The action of sulphur monochloride upon india rubber is extremely 
 energetic, so that for practical purposes a diluent must be employed. The 
 diluent which is practically universally used is carbon bisulphide, but 
 the extremely inflammable nature of this liquid, and its very serious 
 action upon the health of the work-people, caused numerous attempts to 
 be made to substitute a safer and less obnoxious solvent. As such, 
 petroleum ether and benzene have generally been suggested. But al- 
 though in a few isolated instances these, and especially benzene, may be 
 used with satisfactory results, it cannot be denied that in the large 
 majority of cases carbon bisulphide is vastly superior to all other solvents. 
 The reason for this becomes at once evident if we compare the behaviour 
 of the three above mentioned solvents. We then observe that there 
 exists a very striking difference, not only in the degree, but also in the 
 rate, of distension exhibited by strips of india rubber immersed in these 
 three solvents. It is obvious that the more rapidly a solvent diffuses into 
 and through india rubber, the more homogeneous will be the vulcanisation 
 obtained. Now, the great drawback and technical limitation of the ' cold 
 cure ' process of vulcanisation lies in the fact that only very thin sheets 
 of india rubber can be homogeneously vulcanised by means of this process, 
 and this is due to the fact that vulcanisation to the degree required for 
 practical purposes takes place within a few seconds. Longer immersion 
 would produce fatal overvulcanisation, and consequently the sheets or 
 articles must be withdrawn as soon as their outer surface is fully vul- 
 canised. It is obvious, then, that vulcanisation can only be effected to 
 the extent that the solution was capable of penetrating to during the time 
 required for the full vulcanisation of the surface, and this, in thicknesses 
 of india rubber within this range of penetration, must necessarily imply 
 that the employment of the most rapidly diffusing solvent must produce 
 the most homogeneous vulcanisation. It is, no doubt, the very high rate 
 of diffusion of carbon bisulphide into india rubber which confers its 
 superiority over all other solvents which have been tried for the purpose. 
 
 At one time a process known as * vapour cure J was fairly extensively 
 employed, especially for the curing of waterproof cloth. In this process 
 the articles to be vulcanised were hung in special chambers and there 
 subjected to the action of the vapours of sulphur monochloride. This 
 process suffers in a much more pronounced manner from the drawbacks 
 we have already observed in the previous process. 
 
 As far as the volume of the literature dealing with this subject is 
 concerned, it may be said that the vulcanising action of chloride of sulphur 
 upon india rubber has been much less enquired into than the corresponding 
 action of sulphur alone. Indeed, what statements we can find respecting 
 the nature of the ' cold cure ' process are, in the majority of cases, not 
 more than guesses, often made without any experimental reference, and 
 
THE CHEMISTRY OF INDIA RUBBER. 53 
 
 even without any proper consideration of the chemical nature of India 
 rubber. 
 
 Thus we are told by earlier writers upon this subject that the vulcan- 
 ising action of chloride of sulphur is due to its chlorine exclusively, one 
 author going so far as to assert that chloride of sulphur was merely chlorine 
 in a diluted form. The chemical reaction involved in the action of 
 chloride of sulphur upon india rubber is by a sort of common consent 
 assumed to be one of substitution. 
 
 FAWSITT l expressed the opinion that the action of the chloride of sulphur 
 consisted in the substitution of sulphur for hydrogen, the latter escaping 
 either in the gaseous state, or forming hydrochloric acid with the chlorine 
 of the sulphur monochloride. BURGHARDT'S 2 explanation is much on the 
 same lines. THOMSON expresses the opinion that the vulcanising action 
 of sulphur monochloride is due to the chlorine, and not to the sulphur, of 
 this compound. TERRY 3 favours a very similar opinion, as will be observed 
 from the following statement : 
 
 " With regard to the action of this body (chloride of sulphur), I think 
 there can be no doubt that it is due to the chlorine, and that although 
 some sulphur must necessarily be ' absorbed ' at the same time by the 
 rubber, yet that this latter does not play any part in the reaction. Sulphur 
 in the cold has no effect on rubber, and chloride of sulphur must be looked 
 upon as merely chlorine in a diluted form. The action of chlorine in 
 whatever form applied may be measured by the amount of hydrochloric 
 acid produced." 
 
 Two further views have recently been presented on this subject by 
 HbHx, 4 and subsequently by MiNDER. 5 The former states that for the 
 vulcanisation with sulphur monochloride, a softening of the india rubber, 
 analogous to the softening produced by heat in the ' dry cure ' process, is 
 essential. This softening, according to him, is produced by the solvents 
 employed for the sulphur monochloride. After this introduction he 
 advances as his main proposition that the 'cold cure' process consists 
 merely in an ' absorption ' of sulphur by the india rubber, unaccompanied 
 by any chemical reaction. The proofs he adduces in support of these 
 startling statements are confused. The whole of his reasoning is, indeed, 
 devoid of any trace of experimental evidence. 
 
 MINDER'S views may be summarised as follows : 
 
 1. Chloride of sulphur is the condensed or concentrated vapour of 
 chloride of sulphur, which, by virtue of its very high density, has an 
 action upon india rubber so violent as to be destructive. This action 
 can be moderated by diluting the chloride of sulphur with a solvent, so 
 that the solution represents really a concentration of the sulphur vapours 
 
 1 Journ. Soc. Chem. Ind., 1886, p. 638. 
 
 2 Thorpes Diet, of Appl. Chem., vol. ii. p. 312. 
 
 3 Industries, 9393. 4 Gummi Zeitung [14], 17, 33. 5 Ibid. [14], 247. 
 
54 INDIA RUBBER AND ITS ANALYSIS. 
 
 equal to that of sulphur vapour at the temperature of the sulphur 
 bath. 
 
 2. The chlorine in the cold cure process assumes the function of the 
 heat in the sulphur bath process by maintaining the sulphur vapours in 
 the condition in which alone they are capable of vulcanising the india 
 rubber. 
 
 These statements add in no way to our knowledge of, and our insight 
 into, the nature of the cold cure process, as they draw only an extremely 
 precarious parallel between the ' cold ' and the ' hot ' process of vulcanisa- 
 tion. The suggestion that chlorine in combination with sulphur can 
 assume the function of heat acting upon sulphur may or may not have 
 a substratum of truth in it. If so, it is put in a most unfortunate way, 
 reminding one rather of seventeenth century methods of reasoning from 
 analogy than of modern methods of scientific argument. 
 
 Nor does MINDER'S presentment of the case give the least idea why in 
 the cold cure process it should not be possible to substitute any other 
 sulphur compound known, all of which he will be equally justified in 
 considering as ' condensed or concentrated sulphur vapours.' 
 
 The fact is, MINDER has missed the most important point in his 
 enquiry : the elucidation of the fact, if such it be, that only sulphur 
 vapour is capable of effecting vulcanisation. He was satisfied to accept the 
 first property of sulphur vapour which he found to vary directly as the 
 temperature as a sufficient basis for a theory of vulcanisation, altogether 
 overlooking the fact that vapour tensions are equally affected by pressure 
 as by temperature. The density of a vapour being its most easily variable 
 constant, he at once falls back upon this to explain the, as a matter of fact, 
 unquestionable increase in the rate of vulcanisation observable with a rise 
 of temperature. But here he is led hopelessly astray by his assumption 
 that the density of sulphur vapour increases with temperature, a statement 
 utterly at variance with facts. To do him justice, in a subsequent, largely 
 explanatory communication 1 he attempts to justify that assumption by 
 pointing out that vulcanisation is always conducted under pressure, and 
 that pressure certainly increases the density of gases. But in his first 
 paper he distinctly states that india rubber cannot be cured in a sulphur 
 bath at from 115 to 120 C., because at these temperatures the density of 
 the sulphur vapours is too low, and must be increased by raising the 
 temperature. Of course no information is vouchsafed as to how the 
 density of the sulphur vapours of an open sulphur bath is increased by a 
 rise of temperature. 
 
 MINDER'S hypothesis suffers, in my opinion, from the radical defect 
 
 that he attempts to base it upon the phenomenon of vapour density, a 
 
 purely physical phenomenon which in no way but from the point of mass 
 
 in unit volume exerts a determining influence upon the reactivity of a 
 
 1 Gummi Zeitung> 14, 441. 
 
THE CHEMISTRY OF INDIA RUBBER. 55 
 
 chemical system. He leaves altogether out of consideration the molecular 
 state of the sulphur vapour and its molecular changes due to a rise of 
 temperature. 
 
 3. Vulcanisation with HypoMorites. This process was at one time 
 much more commonly used than at present, although it still forms a very 
 useful adjunct to the before described processes for the production of 
 special effects. It is, moreover, very probable that its apparent disuse 
 lies in the secrecy with which especially the less commonly practised 
 processes are still surrounded in rubber factories. In England and 
 America the process is alluded to as * acidising,' but I have been unable 
 to discover who originated it, or who is responsible for its barbarous 
 name. 
 
 The process consists briefly in the treatment of the india rubber with 
 solutions of either calcium hypochlorite (bleaching powder) or sodium 
 hypochlorite. But more frequently a solution of free hypochlorous acid 
 is employed. This solution is prepared by adding to a solution of bleaching 
 powder sufficient 10 per cent, sulphuric acid to convert into calcium 
 sulphate one half of the total calcium present. Using more than this 
 amount of acid would result in the liberation of hydrochloric acid, and 
 thereby, according to the reaction 
 
 HC10 + HC1 = H 2 + C1 2 , 
 
 the hypochlorous acid would be destroyed. Instead of sulphuric acid, the 
 employment of which involves filtration of the solution from the precipi- 
 tated calcium sulphate, nitric acid is frequently used. I personally prefer 
 boric acid, which I find to yield much the best solutions, perhaps owing to 
 its inability to decompose the calcium chloride present in the solutions. 
 It is, however, very probable that very good results can be obtained by 
 decomposing the solutions of bleach lime with the sulphates of very weak 
 bases, such as the sulphates of zinc, magnesium, or aluminium. 
 
 From the fact that an aqueous solution is employed, it will readily be 
 inferred that this vulcanisation process is strictly confined to the vulcan- 
 isation of surfaces only, the solution being quite unable to penetrate into 
 the india rubber. Its action is, therefore, even much more local than that 
 of chloride of sulphur, and it is evident that, as an independent vulcanising 
 process, this agent could only be employed for the vulcanisation of films of 
 very extreme thinness. What recommends it greatly for this purpose is 
 the fact that, by following the vulcanisation by a washing first with very 
 dilute ammonia, and subsequently with clean water, articles are obtained, 
 which, without coatings of farina or French chalk, are free from adhesiveness, 
 and contain no residue of any kind of the vulcanising agent, any excess of 
 hypochlorous acid being destroyed by the ammonia according to the 
 following equation : 
 
 3HC10 + 5NH :5 = 3STH 4 C1 + 3H 2 + K 2 . 
 
56 INDIA RUBBER AND ITS ANALYSIS. 
 
 The ammonium chloride is, of course, readily removed by the washing 
 with water. 
 
 The most frequent and most useful application of this vulcanisation 
 process is, however, the vulcanisation of ' surfacings ' or ' enamels ' on 
 ordinary rubber goods. 
 
 - The chemical aspect of this vulcanisation process calls for little com- 
 ment. The action of hypochlorous acid upon polyprene has been described 
 above (page 55), and we may therefore take it that vulcanisation by 
 this agent consists simply in the formation of an addition product of 
 the type of the chlorhydrines : 
 
 CH 2 : CH 2 + HC10 = CH 2 (OH).CH 2 C1. 
 
 Whether a reaction of this type really represents a true vulcanisation 
 appears to me somewhat uncertain. At the present moment it still appears 
 to me that the true vulcanising effect consists, not only in the formation of 
 an addition product, but also in a conjugation of india rubber molecules 
 (see also page 92) in this manner : 
 
 And if this assumption be correct, it would follow that only divalent elements 
 such as sulphur, but not monovalent elements such as Cl, or radicals such 
 as HO, are capable of producing true vulcanisation. 
 
 It is also noteworthy that, whereas a surface vulcanisation with the 
 halogens (see below) generally imparts a more or less pronounced harsh 
 feel to the surface, this is not the case with hypochlorous acid, and there 
 is little doubt that the greater softness and suppleness in the latter instance 
 is an effect due to the entrance of the hydroxyl group into the polyprene 
 molecule. Indeed, in discussing the products formed by the interaction of 
 solutions of polyprene and hypochlorous acid, I already called attention to 
 the fact that they preserve to a surprising extent the flexibility and resil- 
 iency of the original india rubber. 
 
 4. Vulcanisation with the Haloids. The vigorous action of chlorine 
 and bromine upon india rubber could not fail to suggest their utilisation for 
 vulcanising purposes. HURTZIG proposed, as far back as 1865, the employ- 
 ment of chlorine for the manufacture of a kind of hard rubber. In 1869 
 and in 1870 BALLON, NBWBROUGH, and TAGAN patented various processes for 
 vulcanising with bromine or iodine, or with mixtures of them, either alone 
 or with sulphur. These attempts were bound to result in failure. The 
 patentees attempted to vulcanise at a temperature of 1 20 C. mixtures of 
 
THE CHEMISTRY OF INDIA RUBBER. 57 
 
 India rubber with from 25 to 30 per cent, of iodine, or equivalent quantities 
 of iodine bromide. The cost of this process, as well as the peculiar chemi- 
 cal properties of the halogen addition and substitution products of poly- 
 prene, are sufficient to establish a priori its practical vvorthlessness, par- 
 ticularly in the absence of proof of any description of a new and useful 
 technical effect. 
 
 Weak solutions of bromine in carbon bisulphide l give fairly satisfactory 
 results in surface vulcanisation, but, as already stated above, the surfaces 
 so vulcanised are characterised by a more or less harsh feel, and if the 
 process be not carried out very carefully, especially as regards the strength 
 of the solutions employed, the surfaces are apt to be non-elastic and 
 brittle. 
 
 Altogether, it is evident that neither from the point of view of cost, 
 ease of application, nor of technical effect, is there any inducement to 
 employ the halogens as vulcanising agents, or even to experiment with 
 them in this direction. 
 
 The Colloidal State. 
 
 The difficulty in the way of a rational explanation of the vulcanising 
 process lies in this, that the vulcanisation effect calls for consideration even 
 more from the physical aspect than from the chemical, and while these 
 two aspects are obviously correlated, our almost complete ignorance of the 
 nature of the colloidal state makes the elucidation of the nature of this 
 correlation a practical impossibility. For this reason, I think, it is reason- 
 ably safe to assert that we cannot hope to arrive at a clear understanding 
 of the vulcanisation process until we are in possession of a satisfactory 
 working theory as to the nature of the colloidal state. At present we are 
 far from this, but a summary of our scanty knowledge of this difficult and 
 important subject may be desirable before we turn our attention towards 
 the chemical side of the vulcanisation problem. 
 
 As is well known, the distinction between colloid and crystalloid sub- 
 stances is the outcome of GRAHAM'S researches on hydro-diffusion. 2 The 
 monumental work which he has left us has not only, up to present time, 
 scarcely been extended in any direction, but, as a matter of fact, it has 
 almost dropped out of sight, and the brilliant appreciation in his master 
 mind of the remarkable physical and chemical facts connected with the 
 colloidal state is, if not indeed forgotten, certainly realised by very few of 
 present day workers. GRAHAM'S views on this matter deserve to be quoted 
 here at some length. He says : 
 
 " The colloids appear singularly inert in the. capacity of acids and bases, 
 
 1 Benzene cannot be used, owing to the action of bromine upon it. 
 
 2 GRAHAM, " On the Diffusion of Liquids," The Bakerian Lecture, Phil. Trans., 1850, 
 1-46, 805-836; 1851, 483-494. Especially "Liquid Diffusion applied to Analysis," 
 Phil. Trans., 1861, 183-224. Jour. Chem. Soc., 1864. 
 
58 INDIA RUBBER AND ITS ANALYSIS. 
 
 and in all the ordinary chemical relations. But, on the other hand, their 
 peculiar physical aggregation with the chemical indifference referred to, 
 appears to be required in substances that can intervene in the organic 
 processes of life. The plastic elements of the body are found in this class. 
 As gelatine appears to be its type, it is proposed to designate substances of 
 the class as colloids, and to speak of their peculiar form of aggregation as 
 the colloidal condition of matter. Opposed to the colloidal is the crystal- 
 line condition. Substances affecting the latter form will be classed as 
 crystalloids. The distinction is, no doubt, one of intimate molecular con- 
 stitution. 
 
 " Although chemically inert in the ordinary sense, colloids possess a 
 compensating activity of their own, arising out of their physical properties- 
 "While the rigidity of the crystalline structure shuts out external im- 
 pressions, the softness of the gelatinous colloid partakes of fluidity, and 
 enables the colloid to become a medium for liquid diffusion, like water^ 
 itself. The same penetrability appears to take the form of cementation 
 in such colloids as can exist at a high temperature. Hence a wide 
 sensibility on the part of colloids to external agents v Another and 
 eminently characteristic quality of colloids is their mutability. Their 
 existence is a continued metastasis. A colloid may be compared in this 
 respect to water while existing liquid at a temperature under its usual 
 freezing point, or to a supersaturated saline solution. 'Fluid colloids 
 appear to have always a pectous modification, and they often pass under 
 the slightest influences from the first into the second condition. The 
 solution of hydrated silicic acid, for instance, is easily obtained iri a state 
 of purity, but it cannot be preserved. It may remain fluid for days or 
 weeks in a sealed tube, but is sure to gelatinise and become insoluble at 
 last. Nor does the change of this colloid appear to stop at that point, 
 for the mineral forms of silicic acid deposited from water, such as flint, 
 are often found to have passed, during the geological ages of their existence, 
 from the vitreous or colloidal into the crystalline condition. The colloidal 
 is, in fact, a dynamical state of matter ; the crystalloidal being the statical 
 condition. The colloid possesses ENERGIA. It may be looked upon as the 
 probable primary source of the force appearing in the phenomena of 
 vitality. To the gradual manner in which colloidal changes take place 
 (for they always demand time as an element), may the characteristic 
 protraction of chemico-organic changes also be referred." 
 
 And further on : 
 
 " I may be allowed to advert again to the radical distinction assumed 
 in this paper to exist between colloids and crystalloids in their intimate 
 molecular constitution. Every physical and chemical property is charac- 
 teristically modified in each class. They appear like different worlds of 
 matter, and give occasion to a corresponding division of chemical science. 
 The distinction between these kinds of matter is that subsisting between 
 the material of a mineral and the material of an organised mass. 
 
THE CHEMISTRY OF INDIA RUBBER. 59 
 
 " The colloidal character is not obliterated by liquefaction, and is 
 therefore more than a modification of the physical condition of solid. 
 Some colloids are soluble in water, as gelatine and gum arabic ; and some 
 are insoluble, like gum tragacanth. Some colloids, again, form solid 
 compounds with water, as gelatine and gum tragacanth ; while others, like 
 tannin, do not. A certain parallelism is maintained between the two 
 classes notwithstanding their differences. 
 
 " The phenomena of the solution of a salt or crystalloid probably all 
 appear in the solution of a colloid, but greatly reduced in degree. The 
 process becomes slow ; time, indeed, appearing essential to all colloidal 
 changes. The change of temperature usually occurring in the act of 
 solution becomes barely perceptible. The liquid is always sensibly 
 gummy or viscous when concentrated. The colloid, though often dissolved 
 in a large proportion by its solvent, is held in solution by a singularly 
 feeble force. Hence colloids are generally displaced and precipitated by 
 the addition to their solution of any substance from the other class. Of 
 all the properties of liquid colloids, their slow diffusion in water, and their 
 arrest by colloidal septa, are the most serviceable in distinguishing them 
 from crystalloids. Colloids have feeble chemical reactions, but they 
 exhibit, at the same time, a very general sensibility to liquid reagents, as 
 has already been explained. 
 
 " While soluble crystalloids are always highly sapid, soluble colloids 
 are singularly insipid. It may be questioned whether a colloid, when 
 tasted, ever reaches the sentient extremities of the nerves of the palate, 
 as the latter are probably protected by a colloidal membrane, impermeable 
 to soluble substances of the same physical constitution. 
 
 "It has been observed that vegetable gum is not digested in the 
 stomach. The coats of that organ dialyse the soluble' food, absorbing 
 crystalloids and rejecting all colloids. This action appears to be aided 
 by the thick coating of mucus which lines the stomach. 
 
 "The secretion of free hydrochloric acid during digestion at times 
 most abundant appears to depend upon processes of which no distinct 
 conception has been formed. But certain colloidal decompositions are 
 equally inexplicable upon ordinary chemical views. To facilitate the 
 separation of hydrochloric acid from the perchloride of iron, for instance, 
 the salt is rendered basic by the addition of peroxide of iron. The com- 
 paratively stable perchloride of iron is transformed by such treatment 
 into a feebly constituted colloidal hydrochloride. The latter compound 
 breaks up under the purely physical agency of diffusion, and divides on the 
 dialyser into colloidal peroxide of iron and free hydrochloric acid. The 
 superinduction of the colloidal condition may possibly form a stage in many 
 analogous organic decompositions. 
 
 " A tendency to spontaneous change, which is observed occasionally in 
 crystalloids, appears to be general in the other class. The fluid colloid 
 J|^^4ics pectous and insoluble by contact with certain other substances, 
 
60 INDIA RUBBER AND ITS ANALYSIS. 
 
 without combining with these substances, and often under the influence of 
 time alone. The pectising substance appears to hasten merely an impend- 
 ing change. Even while fluid, a colloid may alter sensibly, from colourless 
 becoming opalescent; and while pectous, the degree of hydration may 
 become reduced from internal change. The gradual progress of alteration 
 in the colloid effect by the agency of time is an investigation yet to be 
 entered upon. 
 
 " The equivalent of a colloid appears to be always high, although the 
 ratio between the elements appears to be simple. Gummic acid, for 
 instance, may be represented by C 12 H 22 O n , but judging from the small 
 proportions of lime and potash which suffice to neutralise this acid, 
 the true numbers of its formula must be several times greater. It is 
 difficult to avoid associating the inertness of colloids with their high 
 equivalents, particularly where the high number appears to be attained 
 by the repetition of a smaller number. The enquiry suggests itself, 
 whether the colloidal molecule may not be constituted by the grouping 
 together of a number of smaller crystalloid molecules, and whether the 
 basis of colloidality may not really be this composite character of the 
 molecule. 
 
 " With silicic acid which can exist in combination both as a crystalloid 
 and colloid, we have two series of compounds silicates and cosilicates, the 
 acid of the latter appearing to have an equivalent much greater (thirty-six 
 times greater in one salt) than the acid of the former. The apparently 
 small proportion of acids in a variety of metallic salts, such as certain red 
 salts of iron, is accounted for by the high colloidal equivalent of their 
 bases. The effect of such an insoluble colloid as Prussian-blue in carrying 
 down small proportions of the precipitating salts, may admit of a similar 
 explanation. 
 
 "Gelatine appears to hold an important place as a colloidal base. This 
 base unites with colloidal acids, giving a class of stable compounds, of 
 which tanno-gelatine only appears to be hitherto known. Gelatine is pre- 
 cipitated entirely by a solution of meta-phosphoric acid added drop by 
 drop, 100 parts of gelatine uniting with 3 '6 parts of the acid. The com- 
 pound formed is a semi-transparent, soft, elastic, and stringy solid mass, 
 presenting a startling resemblance to animal fibrin. It will be an inter- 
 esting enquiry, whether meta-phosphoric acid is a colloid, and enters into 
 the compound described in that character, or is a crystalloid, as the small 
 proportion and low equivalent would suggest. Gelatine is also precipitated 
 by carbolic acid. 
 
 "The hardness of the crystalloid, with its crystalline planes and angles, 
 is replaced in the colloid by a degree of softness, with a more or less 
 rounded outline. The water of crystallisation is represented by the water 
 of gelatination. The water in gelatinous hydrates is aptly described by 
 M. CHEVREUL as retained by 'capillary affinity ' that is, by an attraction 
 partaking both of the physical and chemical character. While it 
 
THE CHEMISTRY OF INDIA KUBBER. 6 1 
 
 is here admitted that chemical affinity of the lowest degree may shade 
 into capillary attraction, it is believed that the character of gelatinous 
 hydration is as truly chemical as that of crystalline hydration. Combina- 
 tion of a colloid with water is feeble, it is true, but so is combination in 
 general with the colloid. Notwithstanding this, anhydrous colloids can 
 decompose certain crystalline hydrates. The alcohol of greater strength 
 than corresponds with the density 0*926, which represents the definite 
 hydrate C 2 H 6 + 3H 2 O, is certainly in a state of chemical union. But 
 alcohol so high as 0*906, contained in a close vessel, is concentrated in a 
 notable degree by contact with dry mucus, gelatine, and gum, and sensibly 
 even by dry parchment paper. Dilute alcohol divided from the air of 
 the atmosphere by a dry septum of mucus, gelatine, or gum, is also 
 concentrated by evaporation, as in the well-known bladder experiment 
 of SOMMERING. The selective power is here apparent of the colloid for 
 water, that fluid being separated from alcohol, and travelling through 
 the colloidal septum by combination with successive molecules of the 
 latter, till the outer surface is reached and evaporation takes place. 
 The penetration in this manner of a colloid by a foreign substance may 
 be taken as an illustration of the phenomena of cementation. Iron 
 and other substances which soften under heat may be supposed to 
 assume at the same time a colloidal constitution. So it may be supposed 
 does silica when fused into a glass by heat, and every other vitreous 
 substance." 
 
 " Gelatinous hydrates always exhibit a certain tendency to aggregation, 
 as is seen in the jelly of hydrated silicic acid and of alumina. With 
 some the jelly is also adhesive, as in glue and mucus. But unless they 
 be soluble in water, gelatinous hydrates, when once formed, are not in 
 general adhesive. Separate masses do not reunite when brought into 
 contact. This want of adhesiveness is very remarkable in the case of 
 the gelose of PAYEN, which resembles gelatine so closely in other respects. 
 Layers of a gelose solution, allowed to cool and gelatinise in succession 
 in a diffusion jar, do not adhere together." 
 
 I have quoted GRAHAM'S singularly suggestive remarks at some 
 length, as in the main they still represent the greater part of our 
 knowledge of the colloids, and the original papers are not everywhere 
 accessible. 
 
 NERNST 1 considers that the fact of colloidal solutions also showing 
 the phenomenon of diffusion, although in very slight degree, proves that 
 there exists the same tendency in colloidal as in crystalloidal solutions 
 for the dissolved substance to migrate from places of high to places of 
 low concentration, but that the extreme slowness of this migration in 
 the case of the colloids indicates a very small propulsive force, i.e., 
 osmotic pressure both of which conclusions would suggest that the 
 colloids possess an extremely high molecular weight, but in their 
 1 NERNST, Theoretical Chemistry, chap. x. 
 
62 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 fundamental physical constitution, in the state of solids, as also in 
 solution, do not differ from the bodies described as crystalloids. 
 
 Indeed, PFEFFER* has shown that the osmotic pressures of solutions 
 of the colloidal gum arabic vary exactly as the concentration, just as is 
 the case with the solutions of the crystalloids : 
 
 Concentration. 
 
 Gum Arabic. 
 
 Cane Sugar. 
 
 M. 
 
 1 per cent. 
 6 ,, 
 14 ,, 
 18 
 
 6 '9 cm. Hg 
 25-9 ,, 
 70-0 
 119-2 
 
 51-8 cm. Hg 
 310-8 
 725-2 ,, 
 932-4 
 
 2570 
 4110 
 3540 
 2680 
 
 The molecular weight M in the last column has been calculated in 
 the well-known manner by multiplying the known molecular weight of 
 cane sugar with the ratio of the, respective osmotic pressures, which latter 
 are expressed in centimetres of mercury. 
 
 In the following table are recorded a few further observations of the 
 molecular weights of colloids by the osmotic pressure method. These 
 determinations are PFEFFER'S, with the exception of the last two, which are 
 due to LINEBARGER : 
 
 
 
 
 
 
 Substance. 
 
 
 Membrane. 
 
 Cone. ' Temp. 
 
 Pressure. 
 
 M. 
 
 Dextrin 
 
 Conglutin 
 
 Cupric ferrocyanide 
 Calcium phosphate 
 
 1 per cent. 16 C. 
 2 , 16 
 
 16 '6 cm. Hg 
 
 3-8 ,, 
 
 1080 
 9500 
 
 Glue 
 
 Cupric ferrocyanide 
 
 6 , 23-3 
 
 23-7 ,, 
 
 4900 
 
 Glue 
 
 Parchment 
 
 6 , 23 -3 
 
 21-3 
 
 5200 
 
 Tungstic acid 
 
 Parchment paper 
 
 2-467 , ! 17 
 
 25-2 ,, 
 
 1700 
 
 Tungstic acid 
 
 Parchment paper 
 
 i-o , ! ... 
 
 ... 
 
 1720 
 
 In perfect agreement with the above small values for the osmotic 
 pressure is the fact that the boiling and freezing points of colloidal 
 solutions differ only very little indeed, in some cases scarcely perceptibly 
 from the respective boiling and freezing points of their solvents. For 
 this reason the determinations of the molecular weights of the colloids 
 by these methods 2 are far less reliable than the determinations made by 
 the osmotic pressure method, where the observations are based upon a 
 phenomenon producing values of a much higher order of magnitude. 
 However, the molecular weights of a number of colloids were calculated 
 from the observed depressions of the freezing points of their solutions. 
 
 1 Osmotische Untersuchungen, Leipzig, 1877. 
 
 2 It is practically the freezing point method only which is available for these 
 determinations, as the coagulation throughout of some colloids at higher temperatures, 
 and generally the formation of incrustation deposits or coagula on the sides of the 
 boiling vessels, renders the boiling'point method useless. 
 
THE CHEMISTRY OF INDIA RUBBER. 63 
 
 SUBSTANCE. M. OBSERVEK. 
 
 Albumen, j . . 14000 Sdbanejew and Alexandrow. 
 
 Inuline, .... 2200 "i 
 
 Starch, . . . about 25000 } Brown and Morris. 
 
 India rubber, . . . .. 6504 \ 
 
 Gum arabic, . . . 1612 I 
 
 Caramel, .... 1745 l^ladstone and Hibbert. 
 
 Ferric hydroxide, . . 5452 J 
 
 Tungstic acid, . . . 800 \ 
 
 Glycogen, . . . 1625 I 
 
 Silicic acid, . . at least 49000 [Sabanejew. 
 
 Tannicacid, . . . 1100 J 
 
 These figures, no doubt, are open to the objection that they involve 
 more or less considerable experimental errors ; but even so, they still afford 
 ample proof of the enormous magnitudes of colloidal molecules, and they 
 further support the assumption that the difference between colloids and 
 crystalloids Is one of degree rather than of fundamental chemical constitu- 
 tion or function. This view is unavoidable in face of the fact that we 
 know colloidal elements sulphur, mercury, cadmium, tin, antimony, silver, 
 gold , we know colloidal sulphides, acids, bases, salts ; organic substances 
 with alcoholic, aldehydic, carboxylic, aminic functions ; carbohydrates and 
 hydrocarbons, both saturated and unsaturated. Moreover, we have seen, 
 in the case of india rubber, that we can subject it to considerable chemical 
 changes, both functional and constitutional, without affecting the colloidal 
 state. Particularly striking in this relation appears to me the fact that 
 the colloidal state of a colloid generally even persists in the combination of 
 a colloid with a crystalloid. 
 
 That all colloids possess extremely high molecular weights has already been 
 amply demonstrated by GRAHAM, and the same fact has been insisted upon 
 by every one of the numerous authorities who have investigated the osmotic 
 phenomena of crystalloids and colloids. The only exception in this respect 
 is KRAFFT, and as his theory of colloidal solutions is the only systematic 
 attempt to bring into a comprehensive system the facts in relation to the 
 colloidal state, his papers on this subject must be more closely examined. 
 
 KRAFFT l repeated CHEVREUL'S experiments on the hydrolysis of the salts 
 of the higher fatty acids (soaps) in aqueous solution. He shows that at 
 lower temperatures the salts in these solutions behave as crystalloids, but 
 that at certain higher temperatures these same solutions exhibit the 
 characteristics of colloidal solutions. Thus dilute soap solutions possess a 
 Tx)iling point not exceeding that of water, and although occasionally a slight 
 rise is observable, this is, no doubt, due to the solutions containing caustic 
 soda, owing to the hydrolysis soaps undergo in aqueous solution. He next 
 states that the alkali salts of acetic, propionic, and caproic acid are also 
 hydrolysed in aqueous solutions, because their molecular weights, as 
 estimated by either the cryoscopic or the boiling point method, are only 
 one half of the recognised theoretical values. A similar series of salts 
 1 er., xxvii. (1894), pp. 1747, 1755. 
 
64 INDIA RUBBER AND ITS ANALYSIS. 
 
 KRAFFT believes he has found in the chlorides of the aliphatic amines, for 
 he finds l that hexadecylamine hydrochloride C 16 H 33 .NH .HC1 behaves in 
 aqueous solution like a soap, in that the former boils without a temperature 
 rise, whereas the solutions of the lower amines e.g., methylamine show a 
 considerable temperature rise, leading, like the salts of the low fatty acids, 
 to values for their molecular weights of only half their theoretical amounts. 
 Consequently KRAFFT comes to the conclusion that in aqueous solution 
 the salts of the fatty acids, and those of the amines, are all, like soap 
 solutions, more or less completely hydrolysed. This hydrolysis would, of 
 course, double the number of molecules in solution, and thus account for 
 the molecular weight being found too low by one half. 
 
 KRAFFT does not appear to admit that the phenomenon in question is 
 perfectly satisfactorily accounted for by the theory of ionisation, nor does 
 he recognise that his alternative explanation is hopelessly opposed to the 
 fact that acetic acid can be titrated with caustic soda, as well as methylamine 
 with hydrochloric acid. Moreover, if sodium acetate or methylamine 
 hydrochloride are hydrolysed in their aqueous solutions, the former ought 
 to possess saponifying, the latter inverting (to cane sugar), properties. 
 Adopting KRAFFT'S interpretation, we should have to assume also caustic 
 soda to be * hydrolysed ' in aqueous solutions, which in the boiling test also 
 yield values for the molecular weight of that base amounting to only one 
 half of the real value a fact entirely in agreement with the theory of 
 ionisation, but utterly unintelligible in the light of KRAFFT'S assumption. 
 So much for KRAFFT'S curious basis of his theory of the colloidal state. 
 
 Observing next that alcoholic solutions of sodium oleate exhibit a 
 marked increase of their boiling point above that of alcohol in other words, 
 that the oleate in these solutions behaves as a crystalloid KRAFFT infers 
 that " water, at least under ordinary pressure, is as necessary for the 
 formation of colloidal soap solutions as soap itself." With this statement 
 one cannot but agree, but upon the real significance of this observation 
 KRAFFT has nothing further to say. He then proceeds 2 : 
 
 " The observation that such well-known organic salts, the composition 
 of which is known beyond doubt, are capable of forming colloidal solutions, 
 and therefore belong to the class of colloidal bodies, is of considerable 
 interest in view of the fact that the nature of all colloid substances is 
 still more or less wrapped in mystery. Particularly the observations 
 made on hexadecylamine chlorhydrate C 16 H 33 .NH 2 .HC1 and methyl- 
 ammonium palmitate C 16 H 31 2 (NH 3 .CH 3 ) seem to confirm that although 
 a not inconsiderable size of molecule seems to be essential for the forma- 
 tion of a colloidal substance, the chemical composition of the latter may, 
 nevertheless, be a very simple one." 
 
 In the first instance, I would point out that we have no right to call 
 the soaps colloidal substances ; they are very well defined crystalloids, which 
 may or may not form colloidal solutions, according to the nature of the 
 1 Eer., xxix. (1896), p. 1331. 2 Ibid., xxix. (1896), p. 1335. 
 
THE CHEMISTRY OF INDIA RUBBER. 65 
 
 solvent employed.-^ If we call the crystallisabje sodium salts of the fatty 
 acids colloidal bodies, why not also aluminium chloride and ferric chloride^ 
 seeing that aluminium hydrate and ferric hydrate can also exist in colloidal 
 ' modifications ' ? The very fact that an aqueous solution of sodium 
 palmitate is a colloidal solution, but that an alcoholic solution of the same 
 salt is a crystalloid solution, is proof conclusive that sodium palmitate itself 
 is not a colloid, but a body which may undergo in aqueous solution a change 
 leading to a colloidal solution. KKAFFT'S expression of surprise that such 
 simple salts as the ones he mentions are capable of forming colloidal solu- 
 tions is peculiarly inappropriate, considering for how long a time colloidal 
 alumina, ferric hydrate, and arsenious sulphide have been known, and 
 considering further that we know quite a number of colloidal elements. It 
 is, moreover, incorrect to say that to be able to assume the colloidal state, 
 a substance must possess a high molecular weight. The very existence of 
 all the inorganic colloids disproves this assumption. The colloidal mole- 
 cule, no doubt, is always extremely complex, but it often, if not generally, 
 consists of very small molecular units. 
 
 KRAPFT now advances the proposition : " Colloidal liquids-or solutions 
 contain the liquefied substances in the molecular state." The proof of 
 this proposition he finds in the fact, as presented by him, that the salts 
 of the low fatty acids in aqueous solution are hydrolytically dissociated, 
 and consequently, in the determination of their molecular weights, yield 
 figures amounting to only one half of their known molecular weights. The 
 sodium salts of the high fatty acids are, in aqueous solution, likewise 
 hydrolysed ; but in spite of this, the boiling points of these solutions are 
 identical with the boiling point of water, or, in other words, these salts 
 exhibit the properties of colloidal substances. On the other hand, KRAFFT 
 observed that the melting and solidification points of colloidal soap jellies 
 are identical with the same constants of the respective fatty acids contained 
 in these salts, He further observed that the molecules of the liquid fatty 
 acids in question are exactly normal, and from these observations he con- 
 cludes that the soap jellies also contain the free fatty acids, and so, by 
 implication, also the soaps themselves, in the unimolecular state. 
 
 ' KRAFFT, of course, realises that his conclusions bring him into conflict 
 with the laws and phenomena in relation to the osmotic pressure of solutions, 
 and this difficulty he attempts to bridge over by a second proposition ; 
 " Molecules in the state of colloidal liquefaction rotate in small closed curves 
 or surfaces." This curious conception would, of course, fairly account for 
 the fact that colloidal molecules no longer appear to obey Avogadro's law, 
 but this hypothesis calls just as much for explanation as do the facts it was 
 invented to explain. The need for it simply arises out of KRAFFT'S first 
 proposition, with the truth of which it stands or falls. But the fact 
 that the melting and solidification points of colloidal soap jellies are 
 approximately identical with the melting and solidification points of their 
 respective fatty acids, although a very interesting observation, scarcely 
 
 5 
 
66 INDIA RUBBER AND ITS ANALYSIS. 
 
 justifies KRAFFT'S revolutionary conclusions. It is, indeed, by no means 
 so certain that in these concentrated soap jellies the soaps are, as KRAFFT 
 assumes, completely hydrolysed at any rate, this assumption should not 
 be made without further experimental evidence, which might have been 
 obtained by the determination of the conductivities of these jellies. Nor 
 should it be overlooked that it is even doubtful whether soaps, simple as 
 is their composition, are really the best starting-point for a study of the 
 colloidal state. For the same reason the salts of aluminium, iron, and 
 chromium, from which colloidal hydrates are readily obtainable, or the 
 silicates, or arsenious acid and its conversion into colloidal arsenious- 
 sulphide, would appear far more suitable objects. But there is, I conceive,, 
 one point which appears to me to call for careful consideration in enquiring 
 into the causes and nature of the change of crystalloid into colloidal solu- 
 tions a point which, I think, should not be so summarily dismissed as has 
 been done by KRAFFT. This is that in all these cases the colloidal state 
 is induced by hydrolysis in aqueous solution. In this fact KRAFFT sees no 
 more than what is expressed by the bare equation by which the chemical 
 effect of hydrolysis is recorded : 
 
 C 16 H 31 2 Na + H 2 = C^EgA + NaOH. 
 
 But the facts in connection with aqueous soap solutions, the conditions, 
 under which all the inorganic colloidal solutions and colloids are formed, 
 the behaviour of the colloidal carbohydrates, and many similar considera- 
 tions, all strongly suggest that the water plays a far more important and 
 determining part than is expressed by a mere hydrolytic equation. Indeed, 
 I am inclined to think that simple as the composition of the soaps and of the 
 above named inorganic compounds appears, this is really a simplicity of a 
 pragmatical order only, the simplicity of our ignorance ; and that, in reality, 
 the colloidal change induced in the above named substances by hydrolysis 
 is an extremely complex phenomenon, and for this reason much more diffi- 
 cult to investigate than the nature and cause of the colloidal condition of 
 a hydrocarbon substance like polyprene. 
 
 KRAFFT has shown 1 that by treating aqueous soap solutions with 
 toluene, the whole of the fatty acid contained in the solution can be re- 
 moved by successive extractions with that solvent. But from this it can- 
 not be concluded that aqueous soap solutions contain the whole of the 
 fatty acid in the free state. Indeed, according to his own showing, 2 this, 
 is not the case. An equilibrium, determined by the quantity of water 
 present, and probably also by the temperature, is produced in these 
 solutions : 
 
 C 16 H 31 2 Na + H 2 ^~> C 16 H 32 2 + NaOH. 
 
 But even at enormous dilutions this hydrolytic action is never complete, 
 as is shown by a glance at the following curve (fig. 2) showing the per- 
 
 1 Ber., xxvii. (1894), p. 1752. 2 Loc. cit., p. 1751. 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
 6 7 
 
 centages of sodium contained in crystallisations obtained from solutions 
 of various strengths of sodium palmitate. 
 
 These facts KRAFFT utilises for his theory, laying great stress upon 
 the free acid and the free alkali in his colloidal soap solutions, but giving 
 no further consideration to those portions of acid and alkali which remain 
 in combination. He seems, on the strength of his observations of the 
 result of the extraction of soap solutions with toluol, to assume that practi- 
 cally the whole of the soap in the solution is split up into free acid 
 and alkali. But from his experiments it is perfectly obvious that this 
 extraction simply disturbs the above mentioned equilibrium by the 
 gradual abstraction of one of the constituents. He therefore entirely 
 overlooks the highly significant fact that it is impossible to obtain his 
 colloidal soap solution in its ideal state of purity by trying to prepare 
 an aqueous and colloidal solution of pure palmitic acid. 1 He ignores that 
 this is only possible if there be any alkali present, and from this fact I 
 prefer to conclude that the colloidal state of a solution of sodium palmitate 
 is induced in the first instance by the hydrolysis of that salt, which process, 
 
 Percentage of Ha tnHydrolisedDepo. 
 
 N5* Ol Oi ^| 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -j 
 
 ^-^. 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^J 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 x^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 "^^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ > 
 
 "x 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^x 
 
 00 300 400 450 500 
 
 90 
 
 Solutions of I gram of Sodium Palmitate in Water 
 FIG. 2. Diagram to KRAFFT'S theory of the colloidal state. 
 
 even at very great dilutions, does not appear to proceed appreciably beyond 
 the formation, besides free alkali, of the dipalmitate C 16 H 31 O 2 Na + C 16 H 32 O 2 . 
 With the formation of this latter salt the conditions for the colloidal state 
 are obtained, but whether this salt is eo ipto a colloid, or forms a colloidal 
 aggregation product with water, is not known, although I strongly incline 
 to the latter conception. 
 
 Exactly the same is to be observed in the case of the salts of aluminium 
 and iron, which in absolute alcoholic solution can be dialysed without the 
 slightest decomposition. The dialysis of aqueous solutions of these salts, 
 on the other hand, always yields the solutions containing what is known 
 as colloidal alumina and colloidal ferric hydrate. To the well-known 
 fact that these solutions always retain slight quantities of acid I have 
 
 1 This may be attempted by dissolving palmitic acid in acetone, and diluting this 
 solution with a large volume of water. 
 
68 INDIA RUBBER AND ITS ANALYSIS. 
 
 already drawn attention, and I strongly believe that the presence of this 
 acid residuum should be looked upon, not as incidental, but as very 
 probably playing a determining part in the subsistence of the colloidal 
 state in the above named bases. Aluminium hydrate and ferric hydrate 
 in the pure state no more form colloidal solutions than does pure palmitic 
 acid: the presence of a very small proportion of acid in the former, or of 
 alkali in the latter, case appears to be essential. If this be correct, we arrive 
 at the conclusion that neither the salts of aluminium, nor of iron, nor the 
 pure hydrates of the oxides of these, bases, nor the soaps, nor their free fatty 
 acids, are colloids, but that they are all capable of forming colloidal aqueous 
 solutions, the colloidal state being induced by the hydrolysis of these salts, 
 and becoming the more pronounced, the nearer this hydrolysis approaches its 
 chemical limits. Corresponding to the chemical lability oj these salts is tl>e 
 peculiar physical instability of their aqueous solutions. 
 
 How far these conclusions apply in regard to the silicates and colloidal 
 silicic acid, to the tungstates and colloidal tungstic acid, and to colloidal 
 arsenious sulphide, no observations are at hand which would enable us to 
 form an opinion. It is, however, known 1 that on dialysing potassium 
 silicate, gelatinous silicic acid is deposited on the septum, which fact 
 suggests that alkaline silicates, like soaps, undergo hydrolysis in aqueous 
 solution. On the other hand, colloidal solutions of silicic acid persist even 
 in the presence of excess of acid ; and silicic acid obtained by dialysis is, 
 to the extent of 14 per cent., soluble in water 2 facts which distinguish 
 colloidal silicic acid rather sharply from the before discussed colloidal 
 solutions. The facts in respect of tungstic acid are somewhat similar. 
 Both the silicates and the tungstates are characterised by the enormous 
 molecular complexity of their salts polysilicates, poly tungstates ; and 
 this fact must appear highly suggestive in view of the evidence we have 
 been quoting in support of the assumption of high molecular complexity 
 and distension of the colloids as a class. Ii^ is, indeed, remarkable that 
 stannic acid, which possesses, though in a somewhat lesser degree, the 
 above mentioned characteristics of silicic and tungstic acid, is also known 
 to exist in a colloidal form. All three acids are rather weak acids, and 
 in this relation it is important to observe that amongst the many colloids 
 there is neither a strong acid nor a strong base. As a matter of fact, 
 a considerable number of organic colloids, including some of the most 
 typical members of this class, are either essentially polyalcoholic bodies, or 
 simply hydrocarbons. 
 
 No doubt the unquestionable difference in the chemical conditions con- 
 ducive to the induction in inorganic compounds of the colloidal state, as 
 well as of its persistence when established, render it difficult to discover the 
 determinant cause of the colloidal state ; and this difficulty would seem to- 
 be increased by any attempt to explain the remarkable difference of the 
 stability of the colloidal condition in the above named substances. The 
 1 BECQUEREL, C. R., 67, 1081. 2 CHURCH, Journ. Chem. Soc., 15, 107. 
 
THE CHEMISTRY OF INDIA RUBBER. 69 
 
 very facts in relation to these questions render it, however, obvious that 
 the colloidal state is a purely physical condition induced in different sub- 
 stances by a number of rather varying chemical reactions. TJie presence of 
 water is the only condition common to all the cases of colloidal induction we 
 quoted or might have quoted, and in every one of these cases we are there- 
 fore justified in assuming hydrolysis or hydration as a contributing factor. 
 Nevertheless, we are not justified in concluding from this that hydroxyl 
 groups must be present in all colloids, for one of the most pronounced 
 colloids, india rubber, is a hydrocarbon. Indeed, the induction of the 
 colloidal state in the above discussed inorganic compounds, and probably in 
 all ionisable compounds, does not appear to be determined by any definite 
 chemical composition, but rather seems to represent the physical corollary of 
 a more or less limited phase in the decomposition of a reversible system. 
 With this conception of colloidal induction as the physical concomitant of 
 an unstable chemical equilibrium, the fact that this induced colloidal state 
 ceases to exist as soon as the chemical equilibrium to which it is due is 
 disturbed by the intervention of some definite chemical influence, is in 
 thorough agreement. Thus we know that as soon as we disturb the 
 chemical equilibrium expressed by 
 
 C 16 H 31 2 Na + H. 2 < ~~^ C 16 H 32 2 + NaOH, 
 
 by abstracting gradually the palmitic acid by means of toluol, the colloidal 
 state rapidly disappears in the realisation of the final condition of 
 CwHjnOaNa + H 2 = C 16 H 32 2 + NaOH. 
 
 In the great majority of other cases, the addition of small quantities of 
 indifferent salts, or of acids or alkalies, disturbs the colloidal state by dis- 
 turbing the equilibrium to which it is due, and at which only it persists. 
 For this reason I hold that it is inadmissible to apply the term ' colloid ' to 
 substances which only within a more or less limited phase in a continuous 
 series of physical changes induced by chemical differentiation can assume 
 the colloidal state. Or, in other words, it is utterly misleading to describe 
 as colloids compounds which in themselves are unquestionably crystalloids, 
 and which exhibit colloidal characteristics only when forming part of 
 certain chemical systems. Obviously in such cases the colloidal nature is 
 due to the system as a whole, and not to any one of its components. The 
 term colloid should therefore only be applied to compounds, the solutions of 
 which, under all conditions, and in whatever solvents, behave as colloidal 
 solutions, and which further maintain this character through all chemical 
 changes which leave the molecule, such as we conceive it, in/act. 
 
 Of course it might be argued that the globomorphous precipitates 
 obtainable from the colloidal solutions of the soaps, of the salts of 
 aluminium, iron, chromium, silicic acid, tungstic acid, stannic acid, and 
 others, are the solid forms of the colloidal states of these compounds. This, 
 however, would be tantamount to the assumption that the amorphous and 
 the solid colloidal states represent essentially identical forms of matter quite 
 
7O INDIA RUBBER AND ITS ANALYSIS. 
 
 an unacceptable proposition. The globomorphous form of precipitate is 
 extremely common amongst the precipitates formed in a solution by a 
 sudden and radical change of the composition of the solvent, but it has 
 never yet been suggested that these globomorphous precipitates in any way 
 represent a colloidal state of the substances precipitated. On the other 
 hand, there is no doubt as to the close relationship between these globomor- 
 phous precipitates and the amorphous condition of matter in general ; but 
 in a vague way it is realised that there is a considerable difference between 
 the amorphous state of matter in general and the equally amorphous solid 
 colloids. Very little is known respecting this highly complex problem, but 
 a careful study of the changes involved in the transformation of a colloidal 
 solution into a jelly, and of such a jelly into a globomorphous or amorphous 
 precipitate, will readily produce the conviction that the latter change in- 
 volves an abrupt physical change such as is scarcely consistent with the 
 assumption of the persistence of the colloidal in the globomorphous or 
 amorphous state of matter. That there exists, nevertheless, a close con- 
 nection between these very obscure states of matter is as certain as the 
 fact that under certain conditions crystalloids, even in the crystalline form, 
 as well as colloids, may assume globomorphous forms and structures. 1 All 
 this, however, merely goes to show that the obscurity enshrouding the 
 nature of colloidal solutions becomes utter darkness as we proceed from 
 such solutions to globomorphous, amorphous, and colloidal solids. This 
 fact is scarcely surprising, seeing that our knowledge of the nature of the 
 gaseous state and of crystalloidal solutions is incomparably in advance of 
 our insight into the constitution of solid crystalloidal matter. 
 
 But there are two states of colloidal solutions which are clearly recog- 
 nisable as continuations of the colloidal state in solutions. These are the 
 gelatinous and pectous states, both of which are familiar to everybody. 
 
 In the gelatinous state we have a colloidal solution assuming a peculiar 
 state of semi-rigidity or quasi-solidity. The degree of rigidity of a jelly 
 appears to be determined by the concentration of the gelatinising colloidal 
 solution, and by the temperature. With the concentration, and with 
 falling temperature, the rigidity increases ; and it decreases, or even entirely 
 disappears, with rising temperature. The change from the liquid colloidal 
 to the gelatinous state is absolutely continuous, the solution losing gradu- 
 ally its fluidity until eventually the individual parts of the solution can no 
 longer change their positions freely and independently, but become more or 
 less rigidly fixed in relation to one another. It is well known that the 
 peculiar physical properties of jellies can be explained by the assumption 
 that every particle of the solution moves in extremely minute closed curves. 
 Indeed, Lord KELVIN has experimentally demonstrated that by setting 
 water into a state of 'fine grained vortex motion,' or 'vortex sponge,' it 
 
 1 This has been abundantly demonstrated by BiixscHLi in his remarkable works 
 Ueber Strukturen (Leipzig, 1898) and Unters. fiber die Mikrostrukturen des erstarrten 
 Schwefels (Leipzig, 1900). 
 
THE CHEMISTRY OP INDIA RUBBER. 71 
 
 assumes all the properties characteristic of a jelly. It will at once be seen, 
 how closely this conception of the gelatinous state, a particular form only 
 of the colloidal state, approximates to KRAFFT'S theory of the nature of the 
 colloidal state in general, and very likely this author drew his inspiration 
 from this vortex theory of the ether. This application of this theory to 
 the colloids as a whole suffers, in the first instance, from the difficulty 
 that while this vortex theory might well be used to explain the phenomena 
 exhibited by colloidal jellies, it would also seeui to suggest that the 
 gelatinous state is the only state in which colloids can and do exist. But 
 this is not at all true indeed, there is a number of undoubted colloids, 
 india rubber being probably the best known of them, which have never been 
 observed to assume the form of a jelly. A second, and at present perhaps 
 even more formidable, difficulty lies in the fact that the application of the 
 vortex atom theory to the explanation of the constitution of matter suffers 
 from as yet insuperable mathematical difficulties. 1 
 
 The property of cementation is one of the most characteristic properties * 
 of the colloids while they are in that peculiar rnetastatic condition which we 
 discussed above as the gelatinous state. This property of cementation, or, 
 in other words, the complete absence of, and the tendency to obliterate, 
 possible cleavage planes, is also -very probably one of the reasons why 
 colloids have no solubility, and instead of dissolving in their so-called 
 solvents in the manner of the crystalloids, gradually resorb them, swelling 
 into jellies which eventually begin to flow, the resorption of the solvent 
 still proceeding. This metastatic condition and property of cementation^ 
 of the colloids is entirely absent in what GRAHAM termed their pectous___ \ 
 modification. The passing of a colloid from either the state of a solid, or a 
 jelly, or a solution into the pectous state at once terminates its metastatic 
 condition, its capability of cementation. The wide range of sensibility of 
 the colloids to physical and chemical influences becomes greatly restricted ; 
 or conversely, the physical and chemical indifference of the colloids is 
 strikingly enhanced in the pectous state. A reversion from the pectous 
 to the metastatic state of the colloid is therefore either impossible, or, when 
 possible, does not reproduce the original colloid condition. Attempts to 
 affect the pectous colloid by physical or chemical agents very frequently 
 result in its physical and chemical break-down. A number of highly 
 
 1 This is also, at any rate, largely the reason why the vortex atom has not already 
 consigned the utterly unthinkable and unphilosopliical atomic theory to the already well- 
 stocked lumber-room of science. Incidentally it may be mentioned here that the con- 
 ception of vortex atoms and of material masses as crowds of vortices is by no means of 
 entirely modern date, but has had, like the theory of the hard material atom, several 
 precursors. Particularly lucid and definite in this respect is MALEBUANCHB (Recherche de 
 la Write", 1712): " Preuve de la supposition que j'ai faite: Que la matiere subtile ou 
 ^theree est necessairement composee de petits turbillons, et qu'ils sont les causes 
 naturelles de tons changements qui arrivent a la matiere; ce que je confinne par 
 Pexplication des effets les plus gene"raux de la physique, tels que sont la durete des corps, 
 leur fluidite, leur pesanteur, leur le"gerete, la lumiere et la refraction et reflexion de ses 
 rayons." 
 
72 INDIA RUBBER AND ITS ANALYSIS. 
 
 important technical processes, such as the tanning of leather, the vulcan- 
 isation of india rubber, the printing of pigment colours in. albumen thick- 
 enings upon calico, are based upon these facts. 
 
 The pectous condition of a colloid or colloidal solutions may be induced 
 
 (a) By heat. Colloidal solutions of alumina, ferric hydrate, chromic 
 
 hydrate, and some of the albuminoid bodies are pectised by 
 heat. The pectisation in this case is probably due to dehydra- 
 tion or intramolecular condensation. The presence of water, at 
 least in the case of the albumens, appears to be essential, since it- 
 has been demonstrated that most carefully dried albumen is not 
 pectised even at temperatures considerably above its coagulation 
 temperature. 
 
 (b) By internal change (spontaneous action). Almost all the inorganic 
 
 colloidal solutions, and also a few of the organic colloids, of the 
 classes of the carbohydrates and albuminoids are able to pectise 
 spontaneously. It is highly probable that pectisation in all 
 these cases is due to a process of condensation and dehy- 
 dration. 
 
 (c) By indifferent salts. That many colloids are precipitated from their 
 
 solutions by the addition of generally very small quantities of so- 
 called indifferent salts is well known, but whether these precipitates 
 are generally, or even frequently, pectous is to me a matter of 
 some doubt. GRAHAM certainly considered them as such. 
 
 (d) By chemical interaction between colloids. The solutions of acid 
 
 colloids on the one hand, and of basic colloids on the other, mu- 
 tually precipitate each other. The products thus formed are often 
 very remarkable, and their systematic study offers a great field 
 to scientific workers and technologists alike. Whether, however, 
 these colloidal salts represent always pectous colloids is a question 
 involving some uncertainty an uncertainty largely due to a 
 certain indefiniteness attaching to the definition of the pectous 
 state. But in all those cases where pectous colloids are produced 
 by either a process of substitution or of addition, we obtain pro- 
 ducts which very clearly exhibit all the above named character- 
 istics of the pectous state. 
 
 The change of a colloidal solution to the gelatinous, from this to the 
 pectous, and subsequently to the solid colloidal state, represents a con- 
 tinuous physical change without any abrupt break, although the appearance 
 of the pectous state, as has already been pointed out above, is probably 
 always due to a chemical change in the colloid. In this series of changes, 
 as described above, the pectous state need not occur at all, its appearance 
 being mostly determined by specific physical and chemical conditions, and 
 in this case we pass gradually from the gelatinous direct to the solid 
 colloidal state. It follows, therefore, that in the latter series of changes 
 liquid, gelatinous, solid the order must be reversible ; and that in the former 
 
THE CHEMISTRY OF INDIA RUBBER. 73 
 
 ti 
 
 series liquid, gelatinous, pectous, solid this order cannot be reversed, 
 owing to the non-reversibility of the pectous state. All these points are 
 much more clearly exhibited by the organic than by the inorganic colloids. 
 Indeed, the only substances which throughout a long series of chemical 
 changes preserve their colloidal characteristics are the organic colloids. 
 What in the inorganic colloids, including the salts of organic acids and 
 organic bases, represents a mere transient phase, becomes in the organic 
 colloids the permanent physical condition of their existence. 
 
 Thus far we find ourselves without any satisfactory working hypothesis 
 as to the nature of the colloidal state, upon the basis of which the further 
 exploration of this exceedingly interesting field might be^ pursued. We 
 have so far only established the fact that the molecular weight of all 
 colloids is extremely high, with the characteristic qualification that in the 
 case of a great number of inorganic colloids it is a very high multiple of 
 the normal molecular weight. This is the unavoidable conclusion to be 
 derived from the behaviour of the colloids in the BECKMANN test, as well 
 as from PFEFFER'S and LINEBA.RGER'S investigations on the osmotic pressure 
 of colloidal solutions. 
 
 There are, however, some other points in connection with the physical 
 behaviour of colloidal solutions which are capable of throwing some further 
 light on their nature. The longest known of these is the fact that a great 
 number of colloidal solutions, though by no means all, scatter polarised 
 light in a direction at right angles to that of a pencil of light passing 
 through them. This phenomenon, which was discovered by TYNDALL, was 
 extensively studied in its application upon colloidal solutions by PICTON 
 and LiNDER. 1 The most convenient way of observing it I found to be by 
 means of the arrangement shown below : 
 
 H< 
 
 FIG. 3. Disposition of apparatus for the observation of the 'Tyndall effect.' 
 
 a is a lime light jet, which is, of course, preferable employed enclosed 
 in a lantern ; b is a condenser provided with an iris diaphragm, 2 the best 
 distance of which from the incandescent lime is easily determined by 
 actual experiment, c is a glass tube of about f inch inner diameter, the two- 
 ends of which have been ground perfectly smooth. One end, the left hand 
 one in the drawing, is closed by cementing upon it a micro cover glass of 
 
 1 Journ. Chem. Soc., 1892, p. 137. 
 
 2 The type of condenser frequently employed for micro-photographic work answers 
 very well. 
 
74 INDIA RUBBER AND ITS ANALYSIS. 
 
 suitable diameter. The other end can be closed with a stopper d, in the 
 perforation of which a piece of glass tubing, bent at right angle, is fixed. 
 This allows the tube to be filled with the solution to be examined without 
 leaving in the tube any air bubbles to obstruct the path of the light. The 
 apparatus is so arranged that the pencil of light coming from the oxy- 
 hydrogen jet passes through the condenser b, and traverses axially the 
 tube e containing the colloidal solution under examination. The radially 
 reflected light is examined by means of a Nichol's prism, which is fixed in 
 some suitable support. 
 
 By means of this arrangement the 'Tyndall effect' may readily be 
 observed on a large number of inorganic colloids in solution, but it is 
 important to state that it is not observable in all inorganic colloidal 
 solutions, and in some only if they have been prepared in some special 
 way. The case of silicic acid is particularly notable in this respect. 
 
 Most colloidal solutions of low concentration can be filtered, but here 
 also we observe some very marked differences respecting the ease with 
 which different solutions of equal concentration pass through the same 
 kind of filter paper. Solutions of colloidal sulphur, inorganic sulphides, 
 some forms of colloidal silicic acid, molybdic acid, and many others pass 
 through the filter almost as readily as water, and leave no sensible residue 
 on the filter. Other solutions filter more or less difficultly, and some few 
 scarcely at all. Exactly similar differences exist between the very numerous 
 organic colloids. 
 
 Filtration through porous cells is a much more severe test, and it 
 appears that, even under pressure, only very few of the colloids are capable 
 of passing through this material. There is, however, a limited number of 
 inorganic, and a somewhat larger number of organic, colloids which are 
 capable of hydrodiffusion, although the rate of diffusion in every case is 
 very small. 
 
 Of extreme interest is the fa^ first observed by PiCTON, 1 that in a 
 few colloidal solutions examination with the microscope reveals the presence 
 of solid particles. The following quotation from PICTON'S paper well 
 describes the microscopic aspect of a colloidal solution of mercuric sulphide : 
 "At first it is impossible to make out anything beyond a dimly granular 
 appearance ; but gradually one becomes aware of exceedingly minute 
 particles in very rapid Brownian movement. The apparently homogeneous 
 solution is really the scene of ceaseless oscillations of innumerable solid 
 particles, crowded together so closely that very little free path is left. 
 Indeed, this last fact is the probable explanation of their being seen at all. 
 In a dilute solution the motion most likely becomes so rapid as to render 
 the particles invisible." 
 
 It will readily be perceived that the very high molecular weight of the 
 colloids as a class is well borne out by their comparative chemical and 
 physical indifference, the low osmotic pressure of their solutions, and 
 1 Journ. Chem. Soc., 1892, p. 137. 
 
THE CHEMISTRY OF INDIA RUBBER. 75 
 
 consequently their low rate of diffusion and the almost immeasurably small 
 depression of their freezing points and the equally small elevation of their 
 boiling points. 
 
 It is, however, quite obvious that although all these constants are all 
 very low as compared with the corresponding constants of the crystalloid 
 bodies, this is, after all, a matter of degree only, and there would not be the 
 slightest difficulty of compiling for every constant a scale, the terms of 
 which, without any appreciable gap, gradually proceed from the most 
 characteristic crystalloids to the most typical colloids. But this being so, 
 it is evidently a most objectionable procedure to erect a purely artificial 
 barrier between crystalloids and colloids by describing the solutions of the 
 latter as pseudo-solutions, or, worse still, as suspensions. 
 
 No doubt this mistaken view is largely based upon the fact that a 
 great many colloidal solutions contain particles large enough to scatter 
 light, and even, in some cases, to become visible under the microscope 
 optical properties which appear to be incompatible with our previous con- 
 ceptions and experience of molecules. But, granted a sufficiently high, 
 molecular weight, the low osmotic pressure of colloidal solutions becomes a 
 necessary corollary, and with it all the rest of those physical properties 
 which at first sight appear to differentiate so clearly colloids from crystal- 
 loids. There remains then only the difficulty of admitting the possibility 
 of the existence of molecules large enough to scatter light and, indeed, to 
 come within actual vision. 
 
 But quite recently LOBRY DE BRUYN l has shown the futility of even 
 this argument. This author calls attention to the fact that the smallest 
 particles capable of polarising light are from fifty to a hundred times smaller 
 than the mean wave length of light. Putting the latter at 0'5/x, the dia- 
 meter of these particles could therefore not be less than from 5 to 10/x/x. 
 
 We are, however, able to determine the molecular diameters of several 
 colloids on the basis of their molecul^ weights as determined by purely 
 chemical methods. 
 
 By the cryoscopic method the following molecular weights were ob- 
 tained : starch, 25,000; albumen, 1 3,000 to 14,000; gelatine, 5000; dex- 
 trine, 3000 to 4000 ; inuline, 2200. But these figures have always been 
 looked upon as rather doubtful, owing to the considerable experimental error 
 involved by the application of the cryoscopic method to the determination 
 of the molecular weight of colloidal substances. It is therefore important 
 to point out that BROWN and MiLLAR, 2 in their valuable paper on starch, dex- 
 trine, and the successive hydrolysis of these bodies, arrived at the result 
 that the molecular weight of soluble starch prepared by Lintner's method 
 is at least 32,500, and that of dextrine 6500. In the course of their in- 
 vestigations on iodine starch, RODEWALD and KATTEIN 3 deduced for starch 
 a molecular weight of 32,700. The remarkably close agreement between 
 
 1 Rec. des Trav. chim. des Paijs Bas et de la, elg., 1900, xix., 236, 251. 
 
 2 Journ. Soc. Chem., 75, 331. 3 ZeiL Phys. Chem., 33, 579. 
 
76 INDIA RUBBER AND ITS ANALYSIS. 
 
 the last named and BROWN and MILLAR'S figure will at once be perceived, 
 as also the fair approximation of both these figures to the one obtained 
 cryoscopically. 
 
 According to VAN DER WAAL'S calculations, the diameter of gas mole- 
 cules is about from O'l to 0'3/x/x, and the diameter of the molecule of 
 chlorine was stated by JAEGER 1 as 0'66/^/x. Now, taking as our basis the 
 molecules of water vapour, oxygen, carbon dioxide, and chlorine, with the 
 respective molecular weights of 18, 32, 44, and 71, it will be seen that the 
 molecular weight of starch is, roughly speaking, a thousand times heavier 
 than the mean of the above. Assuming now, as an approximation, that the 
 specific volumes of these molecules are equal, it follows that the diameter 
 of the starch molecule in solution is 3/1000=10 times larger than that of 
 the above named molecules, and therefore about 5/z//-. But this value is 
 practically identical with the value deduced for the diameter of the 
 particles which, in colloidal solutions, are capable of scattering polarised 
 light, so that by entirely different and absolutely independent methods, we 
 arrive at values of exactly the same order of magnitude. 
 
 From this extremely interesting exposition it follows that colloidal 
 solutions which scatter and polarise a pencil of light passing through them, 
 or which even prove themselves microscopically discontinuous, cannot for 
 this reason be set apart as 'pseudo-solutions' or 'molecular suspensions.' 
 The introduction of these conceptions can only have the result of altogether 
 obscuring the clear fact that there exists at present no recognisable line of 
 demarcation between crystalloidal and colloidal solutions. The dis- 
 continuity between these two so-called classes of solutions is only apparent, 
 and a discontinuity of methods of perception rather than of phenomena. 
 
 LOBRY DE BRUYN 2 quite clearly realises this, but his proposal, based 
 upon PICTON and LINDER'S observation that colloidal solutions are non- 
 electrolytes, to distinguish electrolyte and non- electrolyte solutions will, no 
 doubt, in time share exactly the fate of the older distinction. It is 
 certainly very striking that all the colloidal solutions, organic or inorganic, 
 even solutions of colloidal silver, are non-electrolytes ; but everything is in 
 favour of the assumption that as we proceed from the high molecular 
 colloidal solutions to the crystalloidal solutions, the phenomenon of 
 ionisation, conductivity, will gradually appear and increase. 
 
 But if all this requires any further proof, there is the case of oxyhsemo- 
 globin as given by PICTON and LINDER themselves. PREYER assigns to 
 haemoglobin the empirical formula C 600 H 960 N 154 FeS 3 O 139 , which agrees very 
 satisfactorily with MARSHALL'S result, who, from the amount of carbonic oxide 
 absorbed by haemoglobin, determined its molecular weight as 14,157. 
 Oxyhsemoglobin is a finely crystallisable compound, the scarlet solution of 
 which in water (0'3 per cent.) very markedly exhibited TYNDALL'S phe- 
 nomenon. Nor was it possible to filter this solution through a porous cell r 
 the filtrate obtained was absolutely colourless. We have therefore here a 
 
 1 Zeit. Phys. Chem., 32, 17. 2 Rec. des Trav. chim. t xix., 255. 
 
THE CHEMISTRY OF INDIA RUBBER. 77 
 
 crystalline compound of enormous molecular weight exhibiting all the 
 characteristics of the colloids. 
 
 Of course there is this difficulty, that probably all the inorganic, and 
 perhaps also some of the organic, colloids represent high molecular aggre- 
 gates of crystalloid molecules ; and as the same is undoubtedly true of 
 matter in suspension, this might appear a strong justification for the 
 1 suspension hypothesis' of the colloidal state. But it is by no means 
 certain that the high molecular organic colloids are merely aggregates of 
 the crystalloid molecules as we conceive them. It is by no means certain 
 that colloidal arsenious sulphide is simply (As 9 S 3 ) w , or colloidal silicic acid 
 {Si0 2 ) wl . Indeed, as regards the latter, we have certain indications that 
 this is not the case. GRAHAM'S colloidal silicic acid showed only a faintly 
 acid reaction, and 100 grms. of Si0. 2 contained in it required for complete 
 neutralisation only 1'85 grms. of KOH instead of 93 '33 grms. Thus 
 it would appear that in the colloid only one molecule of Si0 2 in fifty 
 is capable of acting chemically as Si0 2 . This appears to me a point of great 
 importance/ the further study of which is likely to throw considerable light 
 upon the great chemical indifference of colloids. 
 
 There is, however, a further point which seems to afford strong proof 
 of the very wide difference between a suspension and a colloidal solution. 
 This is the remarkable fact that a very considerable number of crystalloidal 
 precipitates formed by double decomposition of various salts cease to be 
 produced when the precipitation is attempted in highly concentrated 
 colloidal solutions, although double decomposition certainly still takes 
 place. Some of the most insoluble precipitates are in this manner 
 obtained, as was recently shown in an admirable paper by LOBRY DE 
 BRUYN, 1 in an apparent state of perfect solution. It is utterly inconceiv- 
 able that mere suspensions of high molecular aggregates could produce 
 such a result. 
 
 Nor is t*he property of the colloids of gelatinising, or pectising, at all 
 intelligible as a function of suspended matter. It is, indeed, when we enter 
 upon the consideration of these two phenomena of the colloidal state that 
 the difficulty of the colloidal problem becomes altogether overwhelming in 
 the present condition of our knowledge. 
 
 The gelatinising of a colloidal solution must be clearly distinguished from 
 the change GRAHAM termed pectisation. The gelatinous state of a col- 
 loidal solution, relative to the liquid state of solution, is characterised by its 
 reversibility. The physical state of a colloidal jelly is essentially a function 
 of concentration and temperature. A solution may be induced to gelatinise 
 either by increasing its concentration or by lowering its temperature ; and 
 conversely, a jelly may be liquefied either by reducing the concentration or 
 raising the temperature. For every colloid capable of gelatinising 2 there 
 
 1 Rec. des Trav. chim. des Pays Bos, xix. (1900), 235. 
 
 2 Only comparatively few of the known colloids are capable of forming jellies in the 
 above sense. 
 
7 INDIA RUBBER AND ITS ANALYSIS. 
 
 seems, however, to exist a critical concentration, below which no reduction 
 of temperature will induce gelatinisation. 
 
 From these facts it may safely be inferred that gelatinisation is an 
 aggregation phenomenon resulting in the cessation of the free mobility 
 of the colloidal molecules, and the fixation of their relative positions. 
 Colloidal matter in the gelatinous state represents therefore a molecular 
 sponge, and it becomes evident that colloidal matter in this state already 
 possesses a definite structure, determined, no doubt, largely by the stereo- 
 metric condition of the individual molecules, and, as I believe, in a minor 
 degree also by the nature of the solvent. 
 
 This highly probable assumption explains also most satisfactorily the 
 extraordinary microscopic reticular structures, of which we owe such 
 admirable accounts to Prof. BUTSCHLI. The conception of BUTSCHLI'S 
 colloidal structures * Schaumstrukturen ' which are of an order of 
 magnitude respecting their elements in most cases approaching the limits 
 of microscopic vision, as molecular sponges likewise cannot be objected to 
 on the ground of their visibility, since we already have to face the fact of 
 actual molecules in solution coming within the limits of optical perception. 
 The pectous state of colloids is in its outward appearance often 
 indistinguishable from the gelatinous state, but the two are sharply 
 distinguished from each other by the circumstance that the pectisation 
 of colloidal solutions is not a reversible process. Colloidal solutions can 
 be pectised at very much lower concentrations than those required for 
 gelatinisation, and in those cases where temperature is of any marked 
 influence upon pectisation, this influence shows itself as the effect of a 
 rise of temperature. 
 
 The influences which induce the pectisation of colloidal solutions 
 are somewhat varied and certainly not easy to classify. Some colloidal 
 solutions pectise spontaneously on standing, some by boiling, a large 
 number of the inorganic colloids by freezing. The majority of the in- 
 organic colloids pectise on adding to their solutions minute quantities 
 of electrolytes, but in a number of cases the precipitates obtained are 
 re-soluble in pure water, and therefore certainly not pectous. 
 
 The most typical cases of pectisation are those in which the colloid 
 in solution forms a pectous jelly on undergoing a chemical change. 
 Cases of this kind are extremely rare amongst inorganic colloids, but 
 rather frequent among the organic colloids. The question of the change 
 involved in the pectisation of a colloid is, on the whole, still a very 
 obscure point, and the probability is, that it will ultimately be found to 
 resolve itself into a group of partly physical, but mostly chemical changes. 
 As regards the physical condition of matter in a pectous jelly, it must 
 obviously be described in exactly similar terms to those in which we have 
 above described the constitution of a * reversible ' jelly. But the ' fixation of 
 the relative positions ' of the colloidal molecules in the case of the pectous 
 jelly being due to a chemical change produced in the molecule, it could 
 
THE CHEMISTRY OF INDIA RUBBER. 79 
 
 only be reversed by reversing the chemical reaction in question, and not 
 by a mere change of physical conditions, such as temperature or solvent. 
 
 Such a chemical change may, of course, be induced by purely physical 
 agents, as in the case of albumen by the application of higher temperatures, 
 in which case pectisation would be due to a process of condensation or 
 dehydration ; but in the majority of cases it is the result of a chemical 
 reaction between the colloid in solution and an added chemical reagent. 
 
 While comparatively few only of the numerous organic and inorganic 
 colloids are capable of forming reversible jellies, all are capable of forming 
 pectous jellies. Indeed, even the solid colloids are pectised, rendered 
 permanently insoluble, whenever we are able to induce in or throughout 
 them a chemical reaction under the influence of some outside reagent. 
 And this brings us into contact with a matter of extreme technical 
 importance in regard to colloids namely, the fact that the ultimate 
 physical condition of every colloid in the pectous state depends to a very 
 considerable extent upon the physical state of the colloid at the time 
 pectisation was produced. In other words, pectisation is the means of 
 perpetuating any physical state of a colloid. The vast majority of the 
 most important technical problems in relation to colloids become intelligible 
 once this fact is clearly realised. 
 
 There is one curious point to be noticed in regard of the pectisation 
 of colloidal solutions. It consists in the fact that while colloidal solutions 
 can be pectised i.e., converted into insoluble colloids however great 
 their dilution, pectous jellies can be obtained only with solutions the 
 concentration of which does not fall below a certain limit, which differs 
 not inconsiderably for different colloids. 
 
 It may at once be stated that these limits are, for the various colloids, 
 very much beyond the limits of concentration determining the formation 
 of reversible jellies in the different colloids ; or, in other words, a colloidal 
 solution will form a pectous jelly at a dilution at which the solution 
 would no longer form a reversible jelly. But the concentration at which 
 a colloidal solution has been converted into a pectous jelly leaves, 
 nevertheless, its mark upon the colloid, so as to render it permanently 
 distinguishable from any pectous jelly of the same colloid, but produced 
 at a different concentration. Nor does this difference disappear if such 
 different pectous jellies are, by desiccation or otherwise, freed from the 
 solvents distending them. In fact, in most cases there remains a difference 
 in their general appearance, their softness, often even their colour; and 
 on re-immersing these different products in their former solvents, which 
 cause them to swell or distend, it will be found that the volumes to 
 which they are distended bear an evident close relation to the volumes 
 they occupied before their pectisation. 
 
 Colloidal solutions subjected to a pectising treatment at concen- 
 trations beyond the limits above referred to, as already stated, no- 
 longer form pectous jellies, but simply amorphous precipitates, and it is 
 
So INDIA RUBBER AND ITS ANALYSIS. 
 
 highly significant and suggestive that at whatever concentrations these pre- 
 cipitates have been produced, they behave physically and chemically 
 exactly alike. One cannot help thinking that in colloidal solutions the 
 individual molecules do not possess free mobility, but that the latter is 
 impeded by a general action attraction 1 of all the molecules upon each 
 other. This would necessarily lead to the displacement of any one 
 molecule, or group of molecules, by an external force resulting in a 
 corresponding displacement, but one decreasing with the distance from 
 the centre of disturbance, of all the surrounding molecules, Such a 
 conception appears to afford a satisfactory explanation of the often very 
 striking viscosity of colloidal solutions, which often exhibits itself in such 
 .a peculiar form as to suggest that in colloidal solutions we are con- 
 fronted with matter in a state of continuous molecular concatenation 
 or aggregation. It likewise seems to me to explain the cause of the 
 often enormous internal friction of colloidal solutions, and would suggest 
 it to consist really in a molecular shearing stress. This peculiar mo- 
 lecular effect, which also exists in many highly concentrated crystalloidal 
 solutions, would, of course, tend to decrease with the concentration, and 
 would entirely disappear beyond a certain limit of dilution. The exist- 
 ence or disappearance of this state of molecular interaction would determine 
 the capability, or the reverse, of colloidal solutions to form pectous jellies. 
 
 A hypothesis of this kind may or may not prove of service, but it 
 suggests several highly interesting lines of experimental research which 
 in any case would be productive of valuable results. The problem in 
 question is, however, considerably complicated by the fact that the nature 
 of the solvent itself by no means plays a passive part in the formation 
 of a colloidal solution. 
 
 This is well exemplified in the case of india rubber, the viscosity of 
 the solutions of which varies markedly with the nature of the solvent 
 employed. And these differences seem to endure even after the separa- 
 tion of the solvent from the colloid. At any rate, India rubber separated, 
 either by precipitation- or by evaporation, from its solution exhibits 
 considerable differences in its physical properties notably in tensile 
 strength according to the nature of the solvent in which it was dissolved ; 
 and this fact I believe to be closely related to the further observation 
 that a similar striking difference exists between the relative volume to 
 which india rubber will swell in different solvents. These facts are 
 matters of considerable technical importance. 
 
 Thus far the condition of matter in the colloidal state would appear to 
 be simply that of matter possessing extremely large molecular weights, 
 but not differing from crystalloidal substances by any specific physical 
 or chemical characteristics. The transition from the crystalloidal to the 
 colloidal state would, accordingly, appear to be perfectly gradual and con- 
 tinuous ; and although the extreme terms of the series exhibit enormous 
 physical and chemical differences, these become quite insignificant between 
 
\ 
 
 THE CHEMISTRY OF INDIA RUBBER. 8 1 
 
 neighbouring terms. There is much in the facts known at present in 
 reference to colloids to commend this view, but there are also certain 
 properties of colloidal solutions, most of which have already been referred 
 to above, which do not become more intelligible on this assumption. 
 
 Perhaps the most striking of these is the observation that colloids 
 in solution, or in the state of a jelly, are capable of altogether preventing 
 the formation of a great number of crystalloidal precipitates. This fact 
 has long been known, but it was first studied more thoroughly by LOBRY 
 DE BmjYN, 1 who showed that insoluble precipitates of a number of salts, 
 which form by double decomposition, are no longer obtained if these solu- 
 tions contain from 5 to 10 per cent, of gelatine. Double decomposition, 
 indeed, does take place under those circumstances, but the products remain 
 in colloidal solution. On cooling such mixtures, absolutely glass-clear jellies 
 are obtained. LOBRY DE BRUYN obtained in this manner perfectly trans- 
 parent colloidal solutions of metals (by reduction), sulphides, hydroxides, 
 ferrocyanides, cuprous oxide, and manganese peroxide. 
 
 These facts are quite inexplicable on the above assumption. They 
 show that colloidal solutions are capable of inducing the colloidal condition 
 in normally crystalloidal precipitates at the moment of their formation. 
 This very curious phenomenon is, of course, not incompatible with our 
 above assumption, nor is it very likely that it conflicts with a property 
 of colloidal matter which is not also shared by matter of a lower molecular 
 weight, and which, indeed, may be a function of that constant ; but even so, 
 its explanation cannot be hoped for until our knowledge of the physics 
 of matter of extremely high molecular weight has considerably advanced 
 beyond its present state. 2 
 
 1 Rec. des Trav. chim., 1900, p. 236. 
 
 2 The further question as to whether the colloidal state is confined to the solid and 
 liquid states of matter, or whether there is such a thing as a colloidal gas, I did not 
 venture upon in the foregoing pages for very obvious reasons. At first sight this sugges- 
 tion appears to be verging on the absurd, as it might seem to imply that colloids could be 
 converted into gases. Of this no instance is known. But, on the other hand, TOWNSEND 
 (Proc. Camb. Phil. Soc., 1897) made the very curious observation that electrified 
 hydrogen diffuses with extreme slowness through a porous pot as Prof. J. J. THOMSON 
 (Discharge of Electricity through Gases, p. 16, London, 1898) puts it, just as if an 
 electrified particle were "the centre of an aggregate of some kind which is very large 
 compared with the dimensions of a molecule." But this, according to the result we 
 arrived at above, is exactly what would constitute a colloidal gaseous molecule. 
 TOWNSEND filled dust-free electrified hydrogen into a porous pot: "The rate at which 
 hydrogen in its normal condition escaped from the pot was determined, and then, by 
 connecting the porous pot with an inductor connected with an electrometer, the rate at 
 which the charge escaped from the pot was determined. If the electrical particles had 
 been simple molecules, the rate at which the charge escaped would have been equal to 
 the rate of escape of the hydrogen ; whereas the experiment showed that in reality the 
 rate at which the charge escaped was only a small fraction of the rate at which the 
 hydrogen escaped." In other words, here we have hydrogen assuming properties which 
 distinguish it from its normal condition, just in the same manner in which a crystalloidal 
 substance is distinguished from iis colloidal form. Whether TOWNSEND was operating 
 with colloidal hydrogen in this case is a question of extreme interest. If so, we should 
 
 6 
 
82 INDIA RUBBER AND ITS ANALYSIS. 
 
 It is quite possible that experiments on the phenomena just discussed 
 applied to india rubber would be capable of yielding highly interesting 
 technical results, and would well repay careful study. 
 
 Outlines of a Theory of Vulcanisation. 
 
 India rubber, within a comparatively very limited range, displays great 
 sensitiveness to changes of temperature. In the early stages of the manu- 
 facture of rubber goods this drawback was very keenly felt, and so was 
 the further difficulty that india rubber surfaces have a great tendency 
 to cling and unite, and that the plasticity of the material was very apt 
 to lead to deformations. The ' tackiness ' of india rubber, it was soon 
 found, could be largely remedied by treating the surfaces with mineral 
 powders, such as French chalk (talcum), precipitated calcium carbonate, 
 or sulphur. But this would, of course, still leave the other difficulties to 
 be felt. It is therefore not surprising that the early practical workers tried 
 to devise a process to convert the crude india rubber into a, physically, 
 more stable product. This was accomplished by treating india rubber with 
 sulphur at temperatures exceeding 120 C., a process jointly credited to 
 GOODYEAR in America, and HANCOCK in England. The process is now 
 known by the not very felicitous description of vulcanisation. 
 
 The recent efforts of chemists to bring to bear upon the numerous 
 operations and problems of the india rubber industry the methods and 
 results of modern chemistry have repeatedly been made the subject of 
 adverse criticism. The smallness of the results so far achieved has been 
 contrasted with the enormous importance of the above named discovery of 
 GOODYEAR and HANCOCK. Such criticism entirely overlooks the fact 
 that this discovery was the result of years of experimenting, and was then, 
 in the end, in the case of both discoverers, due far more to perseverance 
 and accident that to rational method and insight. Of inventions and 
 discoveries made in this more or less haphazard manner we only hear of 
 the successes, but not of the countless failures. On taking possession of 
 the india rubber field, the chemist found himself confronted by an enormous 
 number of empirical, disconnected observations resting upon no other 
 basis than that of daily rule-of-thumb routine. As regards the india rubber 
 itself, quite apart from the phenomena encountered in its manufacture, 
 nothing at all of any reliable, or even tangible, nature was known. In 
 other words, the very basis had to be created for the scientific method 
 to be applied. It is not usual that pioneer work of this kind leads to 
 immediate revolutionary discoveries. 
 
 Often these same critics assure us that vulcanisation being a long ac- 
 complished fact, there is, indeed, nothing left for the chemist to discover 
 concerning india rubber. This statement is made, be it noted, at a time 
 
 expect the vapour density of this electrified hydrogen to be very much higher than the 
 vapour density of this gas in its normal condition. An experimental decision of this 
 highly important point should not be very difficult. 
 
THE CHEMISTRY OF INDIA RUBBER. 83 
 
 when not one manufacturer out of ten has the faintest idea of the intrinsic 
 nature of the process upon which he depends for his whole production, and 
 when the same critical authorities cannot agree between them whether 
 sulphur does or does not chemically combine with the india rubber in the 
 vulcanising process, and even seriously state that india rubber contains 
 liquid hydrocarbons, or that such are formed in the process of roller masti- 
 cation. 
 
 We turn our attention first to the question of the general action of 
 sulphur upon india rubber at higher temperatures, and it will be advisable 
 to complete this enquiry by ascertaining at the same time the nature of 
 the action of PARKES' cold vulcanising agent, sulphur monochloride. 
 
 The sulphur bath method might appear from several points of view the 
 most suitable method of studying this question, but after a number of 
 attempts, I abandoned it in favour of the method of subjecting carefully 
 prepared homogeneous mixtures of Pard rubber with a definite amount of 
 sulphur to the action of heat. Again in this case we have the choice of 
 several methods of heating, but the one of heating pieces of this Para 
 mixture of uniform thickness to vulcanising temperatures when immersed 
 in water appeared to me the most satisfactory, as it involves the minimum 
 of loss of sulphur by evaporation. 
 
 The experiments were carried out with strips cut from a calendered 
 sheet, 3 mm. in thickness, of a mixing of 100 parts of Para rubber with 
 10 parts of pure precipitated sulphur. These strips were vulcanised in a 
 phosphor bronze digester of the construction shown in fig. 4. 
 
 The digester is provided with a thermometer tube (thermometer in 
 mercury), a pressure gauge, and a blow-off valve. In the digester a 
 porcelain beaker is suspended, so that it is clear of the bottom. The 
 digestor is filled to about one quarter of its capacity with water; the 
 beaker is completely charged with water, and a number of the strips to 
 be experimented upon immersed in it. The digestor is then closed, rapidly 
 heated to the required temperature, and maintained thereat, either by 
 carefully adjusting the gas burner, or by means of some form of thermo- 
 regulator. At regular intervals one of the strips is withdrawn after 
 blowing off steam and rapidly opening the digestor, which is then im- 
 mediately closed again to continue the series. The time error caused by 
 these successive withdrawals does not exceed four minutes per sample. Of 
 course the water lost by the blown-off steam is from time to time made 
 up with boiling water. 
 
 The strips thus withdrawn are marked, and subsequently cut into very 
 fine threads, which are freed from every trace of uncombined sulphur by 
 extraction with acetone in a Soxhlet extractor. The greatest care was 
 employed to render this operation perfect, every sample being subjected to 
 a three days' continuous extraction. The extracted samples were dried in 
 a current of carbonic acid in the water oven and, until analysis, preserved 
 in carefully stoppered glass tubes. 
 
8 4 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 About 1 grm. of eacli of these samples was used for analysis. The 
 sulphur determinations were all, without exception, carried out by CARIUS' 
 method, as the results by the much simpler and more expeditious method 
 
 FIG. 4. Digester for vulcanisation experiments. 
 
 proposed by HENRIQUES (see page 253) were found to be liable to an error 
 approaching -01 percent, in magnitude. In this manner the following 
 results were obtained : 
 
 VULCANISATION OF PARA RUBBER. 
 
 
 Temperatures of Vulcanisation. 
 
 Duration of 
 
 
 Vulcanisation. 
 
 
 
 
 
 
 120 C. 
 
 125 C. 
 
 130 C. 
 
 135 C. 
 
 140 C. 
 
 
 s. % 
 
 s. % 
 
 s. % 
 
 s. % 
 
 s. % 
 
 30 mins. 
 
 071 
 
 071 
 
 0-99 
 
 176 
 
 
 60 
 
 1-18 
 
 T32 
 
 1-44 
 
 2-17 
 
 ... 
 
 90 
 
 1-31 
 
 1-67 
 
 2-04 
 
 2-36 
 
 
 120 
 
 1-62 
 
 1-91 
 
 2'32 
 
 3-92 
 
 5-07 
 
 150 
 
 
 
 
 4-02 
 
 ... 
 
 180 
 
 1-78 
 
 2-11 
 
 2-94 
 
 4-18 
 
 6-05 
 
 240 
 
 1-93 
 
 2-22 
 
 5-00 
 
 5'50 
 
 ... 
 
 300 
 
 2-25 
 
 2-35 
 
 5-27 
 
 674 
 
 
 360 
 
 2-60 
 
 3-80 
 
 5-82 
 
 6-88 
 
 ... 
 
 420 
 
 371 
 
 4-04 
 
 6-04 
 
 6-97 
 
 
 480 
 
 3-94 
 
 4'31 
 
 6'33 
 
 7-13 
 
 ... 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
 These figures amply suffice to demonstrate indisputably the fact, even 
 quite recently again denied, that the vulcanisation of india rubber with 
 sulphur involves the chemical combination of these two substances, at any 
 rate as far as the vulcanisation of Para rubber is concerned. 
 
 That different brands of india rubber behave very differently in the 
 vulcanisation process is a well-known fact, but what we know at this 
 moment respecting the composition and chemical relationship of these 
 different brands entitles us to assume that, although their behaviour under 
 vulcanisation may not be identical with that of Para rubber, it will be 
 more or less closely analogous to it. 
 
 Experiment entirely confirms this expectation, as is shown by the 
 following tables, which show the effect of the vulcanisation of Lagos, Beni 
 River, Ceara, and Borneo rubber at the temperatures of 125 C. and 135 C. 
 respectively : 
 
 VULCANISATION OF UPPER CONGO, BENI RIVER, CEARA, AND 
 BORNEO RUBBER. 
 
 
 Upper Congo. 
 
 Beni River. 
 
 Ceara. 
 
 Borneo. 
 
 Duration of 
 
 
 
 
 
 Vulcanisation. 
 
 
 
 
 
 
 
 
 
 
 125 C. 
 
 135 C. 
 
 125 C. 
 
 135 C. 
 
 125C. 
 
 135 C. 
 
 125 C. 
 
 135 C. 
 
 
 s. % 
 
 s. % 
 
 s. % 
 
 s. % 
 
 s. % 
 
 s. % 
 
 s. % 
 
 s. % 
 
 30 mins. 
 
 0-82 
 
 1-93 
 
 0-41 
 
 1-07 
 
 0-92 
 
 1-37 
 
 0-52 
 
 0-48 
 
 60 
 
 1-25 
 
 2-21 
 
 0-78 
 
 1-84 
 
 1-38 
 
 2-01 
 
 0-98 
 
 1-15 
 
 90 
 
 1-36 
 
 272 
 
 1-10 
 
 3-26 
 
 2-01 
 
 2-59 
 
 1-21 
 
 1-76 
 
 120 
 
 1-54 
 
 3-02 
 
 1-58 
 
 375 
 
 2-26 
 
 3-23 
 
 1-34 
 
 2-02 
 
 180 
 
 170 
 
 3-86 
 
 2-18 
 
 4-68 
 
 2-52 
 
 3-83 
 
 1-84 
 
 2-41 
 
 240 
 
 2'06 
 
 4-53 
 
 3-08 
 
 5-67 
 
 3-22 
 
 4 19 
 
 2-06 
 
 2-62 
 
 300 
 
 2-48 
 
 5-26 
 
 4-57 
 
 672 
 
 3-43 
 
 4-55 
 
 2-30 
 
 3-23 
 
 360 
 
 2-94 
 
 6-17 
 
 5-11 
 
 7-38 
 
 3-59 
 
 473 
 
 2-67 
 
 379 
 
 420 
 
 3-39 
 
 7-12 
 
 5-31 
 
 7-92 
 
 4-01 
 
 4-99 
 
 3-19 
 
 4-00 
 
 480 
 
 3-81 
 
 7-53 
 
 6-35 
 
 8-28 
 
 4-31 
 
 5-27 
 
 3-59 
 
 4'21 
 
 The extremely interesting results 1 here tabulated remove all doubt 
 
 1 It is perhaps scarcely necessary to point out the interest and technical value of the 
 above series and their graphic representations. They offer a means of reducing to a 
 rational basis the factory practice in regard to the treatment in the vulcanisation of 
 1 mixings ' containing different brands of india rubber. They show by simple calcula- 
 tions which brands may, with advantage, be combined in a ' mixing* to produce an article 
 possessing any desired physical properties, and they show conversely also which brands 
 will not produce a satisfactory blend. 
 
 In this matter the manufacturer has so far been guided by practical experience alone, 
 which may have been satisfactory in a number of cases, but leaves it an open question 
 whether the best possible result has really been obtained. In other cases, there is no 
 doubt, the blending of different brands of india rubber is made a matter of price merely, 
 which, as the above tables clearly show, is unquestionably the worst procedure 
 imaginable, and one that must often lead to sheer waste of individually good material. 
 
 Of course before these tables become generally useful it will be necessary to work out 
 the same series for all important brands of india rubber, and, with advantage, for a 
 larger range of temperatures ; and it will further be necessary to determine for each series 
 from point to point the mechanical strength of the test pieces. 
 
86 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 that the vulcanisation of india rubber is a chemical process resulting in 
 the formation of a polyprene sulphide. The rate at which the sulphur 
 enters into combination with the india rubber hydrocarbon (polyprene) is- 
 characteristic for each brand of india rubber. Some of the above series 
 were repeatedly investigated, always with the same result. The different 
 behaviour of the various brands is particularly striking in the graphic 
 representations, figs. 5, 6, and 7. 
 
 There arises now, of course, at once the question as to the nature of 
 
 Vulcanisation Curve of Para 
 at Different Temperature 
 
 140 
 
 (35C 
 
 I30?C 
 
 I25C 
 I20C 
 
 30 60 90 120 180 24-0 300 360 
 
 FIG. 5. Vulcanisation curves of Para. 
 
 420 
 
 480 Mm 
 
 the process by which sulphur enters into combination with the polyprene, 
 whether the polyprene sulphide or sulphides formed are addition or 
 substitution products. Certainly what we already know respecting the 
 chemical nature of india rubber leads us to infer that the vulcanisation 
 process consists essentially in the formation of an addition product of 
 sulphur and polyprene. This assumption, however, requires further 
 support in view of the fact that quite a number of writers, from PAYBN to 
 most of the recent authors, declare that vulcanisation is accompanied by 
 evolution of sulphuretted hydrogen, thereby implying that the process is 
 a substitution, and not an addition, process. Indeed, most of the recent 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
 authors on this subject state this in so many words. We shall therefore 
 have to subject this point to a careful examination. 
 
 Assuming the compound of polyprene and sulphur, which indisputably 
 forms in the vulcanisation process, to be a substitution product, it follows 
 with absolute necessity that for each 32 parts of sulphur combining with 
 the polyprene, we must obtain 34 parts of hydrogen sulphide. Now, in the 
 process of vulcanisation as practically carried out, we obtain, on an average, 
 a product containing, say, 2 '5 per cent, of combined sulphur. Consequently 
 the vulcanisation of one ton of india rubber, on the above assumption, would 
 
 Vulcanisation Curves of 
 
 Upper Congo, Beni River,Borneo,Ceara & Para . 
 at I25C. 
 
 f 
 
 j Ben/ 
 River 
 
 Para 
 Ceara 
 
 Upper 
 Congo 
 Borneo 
 
 30 60 90 120 160 240 300 360 420 *60Min. 
 
 FIG. 6. Vulcanisation curves of Para, Congo, Beni River, Ceara, Borneo, at 125 C. 
 
 be bound to yield very nearly 60 pounds of hydrogen sulphide, or approxi- 
 mately 18,000 litres. Considering that in a number of factories the amount 
 of india rubber vulcanised daily largely exceeds one ton weight, we should 
 expect to find the vulcanising rooms of these factories reeking with this gas. 
 As a matter of fact, however, there is scarcely ever a trace of this gas to be 
 discovered in the rubber works' atmosphere, and the very rare cases in 
 which its presence becomes noticeable may always be considered as an 
 indication of something * having gone wrong.' 
 
 In the vulcanisation of * hard rubber ' goods (ebonite, vulcanite) faint 
 but distinct traces of sulphuretted hydrogen are generally, perhaps always, 
 observable, but they could not be ascribed to the vulcanisation process proper 
 
88 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 the combination of polyprene with sulphur which process, if it consisted 
 in the substitution of hydrogen by sulphur, should cause a perfectly 
 torrential evolution of hydrogen sulphide, seeing that ' hard rubber ' contains 
 at least 20 per cent, of combined sulphur. 
 
 It is therefore certain that if hydrogen sulphide forms at all in the 
 vulcanising process, its amount is utterly inadequate to support the assump- 
 tion that the process of vulcanisation is a substitution process. 
 
 Laboratory experiments on this question lead to exactly the same 
 
 // 
 
 Bent 
 River 
 
 Upper 
 Congo 
 
 Para 
 
 Borneo 
 
 30 60 90 120 180 24-0 300 360 420 4-BOM/n. 
 
 FIG. 7. Vulcanisation curves of Para, Congo, Beni River, Ceara, Borneo, at 135 C. 
 
 conclusion. If the experiments are carried out with technically pure Para 
 rubber under conditions absolutely precluding the escape of any gaseous 
 product of the reaction, very minute traces of hydrogen sulphide may some- 
 times be observed ; but in a considerable number of carefully devised 
 experiments with highly purified Para rubber, no hydrogen sulphide at all 
 could be detected. 
 
 If, on the other hand, the ' insoluble ' part of india rubber is mixed with 
 sulphur, and this mixture subjected to vulcanising temperatures say, about 
 135 C. a considerable evolution of hydrogen sulphide takes place due to 
 the formation of a substitution product of this insoluble body, C 30 H 68 O 10 , 
 
THE CHEMISTRY OF INDIA RUBBER. 89 
 
 with sulphur. This substitution process certainly proceeds much slower 
 than the vulcanisation process of india rubber (polyprene). Under the 
 same conditions of temperature and time under which polyprene forms a 
 vulcanisation product containing 4 per cent, of sulphur, the above named 
 insoluble constituent forms a substitution product containing at most 
 0'7 per cent, of sulphur. 
 
 From these facts we are entitled to draw the following conclusions : 
 
 1. The india rubber hydrocarbon, polyprene C 10 H 16 , combines with 
 sulphur without evolution of hydrogen sulphide. The vulcanisation process 
 of india rubber is therefore an addition process. 
 
 2. The insoluble constituent of india rubber, ivhich forms only an in- 
 significant proportion of the technical product not exceeding 5 per cent, of 
 the total, combines with, sulphur under vulcanising conditions at a very slow 
 rate with evolution of hydrogen sulphide, and ivith the formation of a sub- 
 stitution product. 
 
 The above conclusively settles the question regarding the general 
 chemical aspect of the vulcanisation process, but it confronts us with 
 the further question respecting the quantity of sulphur combining with 
 india rubber in this process, as well as the more intimate structure of the 
 compound thus formed. 
 
 The tables on pages 84 and 85 very distinctly show that the reaction 
 taking place between polyprene and sulphur leads to polyprene sulphides 
 with an increasingly larger proportion of sulphur, without at any time any 
 very distinct break in the rate of progression of the reaction being notice- 
 able, which break might be interpreted as an indication of the formation of 
 a definite compound. Indeed, one might be tempted to explain the very 
 curious irregularities in the vulcanisation curves of Para rubber, as given 
 on page 84, in the above suggested sense, but on closer examination it will 
 be found that by carrying out the vulcanisation at different temperatures, 
 these irregularities appear at very different places, and therefore may be 
 taken as probably indicating a particular physical, rather than some definite 
 chemical, condition of the vulcanisation product. In this relation atten- 
 tion may be drawn to the tabulated results and curve plottings in 
 reference to the vulcanisation of Beni River, in which case no sudden rise 
 of the vulcanisation coefficient, and consequently also no irregularities in 
 the corresponding curves, are to be observed. 
 
 In the presence of a very large excess of sulphur, and at temperatures 
 between 135 to 150 C., vulcanisation proceeds at a still more rapid rate, and 
 the amount of sulphur combining with the polyprene increases enormously, 
 leading to the formation of products technically known as 'hard rubber,' 
 ebonite, or vulcanite. The formation of these products is accompanied by 
 the evolution of only very small quantities of hydrogen sulphide, so that 
 obviously the compounds formed are still essentially addition products. 
 Even if it could be shown that these very small quantities of hydrogen 
 sulphide were essentially due to the action of the sulphur upon the polyprene, 
 
QO INDIA RUBBER AND ITS ANALYSIS. 
 
 this would not in the least vitiate my above stated conception of the intrinsic 
 nature of the vulcanisation process, as, in the extreme conditions under 
 which these highly sulphuretted bodies are produced, substitution of sulphur 
 for hydrogen is to be expected as a matter of course. Indeed, it is a matter 
 for surprise that this secondary reaction should take place to such an in- 
 significant extent only as appears to be the case. 
 
 It may be taken, therefore, that these products represent the upper 
 limit of the normal reaction (addition) of polyprene and sulphur, and it is. 
 a matter of considerable interest to ascertain the numerical value of this, 
 limit. For this purpose three intimate mixtures of fine Pard and pure 
 sulphur 1 were prepared : 
 
 1. 100 grms. technically pure fine Para, 
 
 50 ,, sulphur, 
 Vulcanised : 12 hours at 138 C. 
 
 2. 100 grms. technically pure fine Para, 
 
 75 ,, sulphur, 
 
 Vulcanised : 9 hours at 140 C. 
 
 3. 100 grms. technically pure fine Para, 
 100 ,, sulphur, 
 
 Vulcanised : 8 hours at 140 C. 
 
 All three samples were excellent specimens of ebonite, free from porosity,, 
 and of remarkable strength. No. 1 was comparatively the softest of the 
 three, No. 3 being extremely hard without being brittle. 
 
 Under the microscope all three show much the same appearance of an 
 almost structureless mass, exhibiting no recognisable differentiation (fig. 8> 
 Plate L). 
 
 This becomes even more distinct by examining these specimens in the 
 state of an extremely fine dust (fig. 9, Plate I.), which is obtained by working 
 them with a file and passing the dust thus obtained through very fine silk 
 gauze. 
 
 On extracting these products, in the state of an impalpably fine powder, 
 with acetone or carbon bisulphide, an extract is obtained consisting chiefly 
 of sulphur and a very small quantity of organic matter. To completely 
 extract the free sulphur contained in these samples is, however, a sur- 
 prisingly slow process requiring several days' treatment in spite of the 
 extremely fine subdivision of the substance. It is further curious that 
 there is no difference in the rate at which the extraction proceeds in using 
 carbon bisulphide instead of acetone in fact, the latter, on the whole, is the 
 more satisfactory solvent. I take this as proof that the free sulphur in 
 vulcanite is present as amorphous (colloidal) sulphur insoluble in the 
 solvents for sulphur. Under the influence of the boiling solvents a 
 gradual reversion of this insoluble amorphous modification into the soluble 
 crystalline form takes place, so that the rate of extraction does not depend 
 so much upon the nature of the solvent as upon the rate of reversion of the 
 sulphur. 
 
 1 Precipitated with hydrochloric acid from a solution of calcium tetrasulphide. 
 
fiber 's India Rubber. 
 
 PLATE I. 
 
 FIG. 8. Micro-section of vulcanite (ebonite). 
 Photo-micrograph: Apochrom. Zeiss 4 mm., Projec. Eyep. 4. x (500. 
 
 i 
 a \Qb. 
 
 FIG. 9. Vulcanite powder. 
 Photomicrograph: Apochrom. Zeiss 4 mm., Projec. Eyep. 4. x 600. 
 
 Charles Griffin & Co., Ltd. 
 
 To face p. 90. 
 
THE CHEMISTRY OF INDIA RUBBER. 9! 
 
 The amount of organic matter extracted is very slight, amounting to- 
 less than 2 per cent, in all the three samples. 
 
 The samples were extracted until a full day's further extraction pro- 
 duced no appreciable increase in the amount of free sulphur removed. 
 They were then dried, and the combined sulphur in them estimated by 
 CARIUS' process. 
 
 The following are the results obtained : 
 
 Sample I. 33*08 per cent, of combined sulphur. 
 II. 33-11 
 III. 32-64 
 
 These figures quite satisfactorily approximate to the theoretical amount 
 of sulphur in polyprene disulphide, C 10 H 16 S. 2 = 32'00 per cent. S. It is 
 rather curious that the amount of sulphur ascertained approaches the 
 theoretical amount the more closely, the shorter the time the sample has 
 been heated, and the larger the amount of sulphur with which it had 
 been mixed. I am inclined to think that this is not accidental, but points 
 to the fact that substitution takes place after, or towards the end of, the 
 addition process. Whether it is the india rubber constituent itself, or the 
 insoluble constituent, which is thus acted upon, I have not been able to 
 decide. If the vulcanisation treatment of the above samples is consider- 
 ably prolonged, products are obtained in which the combined sulphur 
 gradually, but very slowly, increases. Under these conditions the evolution 
 of hydrogen sulphide becomes more marked as time goes on, but, at the 
 same time, the vulcanisation products deteriorate very rapidly in quality. 
 Their colour changes more and more into a greyish-brown, they lose their 
 strength and hardness, and become very porous, and finally almost friable. 
 Even in products of this description I only once observed 38' 18 per cent, 
 of sulphur. 
 
 It appears, therefore, that the sulphur addition of india rubber has 
 reached its upper limit with the formation of a polyprene disulphide, 
 C 10 H 16 S 2 , containing 3 2 '00 per cent, of sulphur. There must, of course, 
 be a lower limit, below which we cannot go without leaving some of the 
 rubber in the unvulcanised state. What this limit is we cannot ascertain 
 at present, as we possess no means for isolating any of the terms of the 
 series of sulphides lying within these two limits, nor for separating any 
 of them from unvulcanised india rubber. 
 
 In view of the circumstance, however, that the important change pro- 
 duced by vulcanisation in the production of soft rubber goods becomes 
 first clearly apparent when the vulcanised india rubber contains from 2 to 
 2 -5 per cent, of sulphur, we may assume that this represents the lowest 
 limit of the possible polyprene sulphides. Accepting this hypothesis, we 
 find, by a simple calculation, that this corresponds to a compound (C 10 H 16 ) 10 S> 
 or C 100 H 160 S, so that the polyprene disulphide we recognised above in the 
 specimens of hard rubber would be C 100 H 160 S 20 . A compound C 100 H 160 S 
 would contain 2 '35 per cent, of sulphur, and it is rather remarkable that 
 
92 INDIA RUBBER AND ITS ANALYSIS. 
 
 this figure should come so well within the limits observable in well-cured 
 rubber goods possessing the highest degree of elasticity and distensibility 
 combined with the highest degree of tensile strength. India rubber possess- 
 ing a higher coefficient of vulcanisation may and does show even higher 
 tensile strength, but at the expense of the other physical constants. 
 
 HENRIQUES states that he observed perfectly vulcanised specimens of 
 india rubber containing less than 1 per cent, of combined sulphur, but while I 
 do not doubt that he examined such specimens, I feel compelled to controvert 
 most emphatically the statement they were ' perfectly vulcanised.' He gives 
 no data in relation to their physical constants to justify the acceptance of 
 his assurance. Throughout my experience in manufacture and analysis I 
 have never met with a rubber sample containing less than 1 per cent, of 
 combined sulphur which could have been described as perfectly vulcanised. 
 It i., of course, quite possible that the technical effect describable as * per- 
 fect vulcanisation ' does not bear the simple quantitative relation to the 
 chemical aspect of the problem which I tentatively attribute to it. In this 
 case it would follow that the lower limit of the sulphur addition to india 
 rubber would be still lower than I hypothetically assumed it to be, with 
 the result that the above named molecular weight of india rubber would 
 be still higher. But even in this case it could only be a multiple of 
 CjooH^o. We are therefore now in a position to state the process of 
 vulcanisation as follows : 
 
 3. The process of vulcanisation consists in the formation of a continuous 
 series of addition products of polyprene and sulphur, with probably a poly- 
 prene sulplnde C^H^S as the lower, and C 1()Q H 16() S 2() as tie upper, limit of 
 the series. Physically this series is characterised by the decrease of distensi- 
 bility, and the increase of rigidity, from the lower to the upper limit. Which 
 term, of the above series i.e., ivhich degree of vulcanisation is produced is in 
 every case only a function of temperature, time, and proportion of sulphur 
 present. 
 
 Respecting the configuration of any of the terms of this series, all is 
 surmise, but the following possibilities exist : 
 
 H XHHX XH 
 
 V 
 
 G 
 
 C 
 
 | >S II. I I III. I >S : S 
 
 C/ C - S - C C/ 
 
 H Y' H H X Y H 
 
 XH HX XHHX 
 
 C S S C (X /C 
 
 IV. I I V. I >S : S< I 
 
 C S S C <y X C 
 
 YH H Y YHHY 
 
 These diagrams show only one of the ethylene bonds known to exist in 
 
THE CHEMISTRY OF INDIA RUBBER. 93 
 
 the polyprene complex, groups X and Y representing the rest of this com- 
 plex. It will be seen that for a polyprene complex C 100 H 160 , only configura- 
 tions I. and II. would be possible if the lower limit of vulcanisation is 
 expressed by the composition of a compound containing 2*29 per cent, of 
 sulphur that is, by C 100 H 160 S. In this case, but with the configurations- 
 shown in the diagrams III. and IV., the polyprene molecule would require 
 doubling. Whether any, and if so which, of these diagrams represent the 
 constitutional configuration of india rubber, only further investigation of 
 the subject can decide. 
 
 There is one fact, well known to all rubber manufacturers, which seems to 
 suggest rather strongly that there is, indeed, no definite relation at all 
 between the quantitative chemical result and the physical technical effect 
 of the vulcanisation process, inasmuch as the same degree of vulcanisation 
 in the same kind of india rubber need not result in the formation of physi- 
 cally identical vulcanisation products. In other words, the fact that twa 
 different specimens of vulcanised india rubber, produced from the same 
 brand, even from the same batch of crude rubber, possess equal amounts of 
 combined sulphur i.e., equal coefficients of vulcanisation by no means im- 
 plies that these two specimens also possess the same physical constants, the 
 same technical properties. It is well known that not only may two such 
 specimens differ very widely in this respect, but it may further happen that 
 while one of them is a practically permanently sound product, the other very 
 rapidly deteriorates. It is, no doubt, owing to this fact that many practical 
 rubber workers still refuse to look upon vulcanisation as an essentially 
 chemical process, but go as far as to roundly deny it, and attempt to explain 
 the vulcanisation phenomena as due in one part to the action of heat upon 
 india rubber, in the other part to the mere physical penetration of the 
 india rubber by the sulphur. 
 
 This fact is not lightly to be dismissed indeed, it is the direct cause for 
 the great number of conflicting statements put forward at various times 
 by different writers, it is responsible for the very confused state of our 
 present day views on ' the nature ' of vulcanisation, and lies at the bottom 
 of all the well-recognised technical difficulties of the vulcanisation process. 
 Amongst the various observations bearing upon this fact the most 
 important are as follows : 
 
 I. Long continued working between mixing rollers destroys the 
 resiliency of the india rubber, rendering it more and more plastic. In 
 this state the india rubber is known to require much more sulphur for 
 vulcanising than if it is only worked sufficiently long to mix it homo- 
 geneously with the sulphur and such other admixtures as it is desired ta 
 employ. 
 
 II. The different modes of carrying out the process of hot vulcanisation, 
 even if performed in every case at the same temperature, produce very 
 different results in one and the same mixing. Thus it is very well known 
 in practice that the methods known as dry heat curing, steam curing,. 
 
94 INDIA RUBBER AND ITS ANALYSIS. 
 
 press curing, hot- water curing, or curing in a sulphur bath, are by no 
 means interchangeable processes for one particular mixing. 
 
 III. The results of quick vulcanisation at a high temperature with a 
 large excess of sulphur, slow vulcanisation at a moderate temperature, or 
 gradual vulcanisation in a temperature gradually rising to a maximum in 
 the course of a very long time, even if the operations be so conducted as 
 to produce in every case the same degree of vulcanisation, produce physi- 
 cally very different products. 
 
 IV. India rubber which has been rendered plastic ('killed ') by much 
 prolonged working, and exhibits the vulcanising characteristics described 
 above (I.), on long keeping gradually loses these characteristics and reverts 
 towards the character of moderately worked india rubber. As it does so, 
 it also recovers largely its former resiliency, and loses the plasticity 
 characterising its * fatigued ' condition. 
 
 The above instances comprise the most important of the physical facts 
 influencing the physical result of the vulcanisation process, but the most 
 remarkable point is that none of them has the least influence upon the 
 chemical result of the process, or upon the coefficient of vulcanisation. 
 This, as already stated, is only dependent upon the same factors which 
 determine the rate of progress of every chemical reaction temperature, 
 time, and mass. In other words, whether india rubber containing a 
 certain amount of sulphur be only moderately worked, or whether it be 
 overworked (fatigued), or by whichever method it be vulcanised, as long 
 as the time and temperature of vulcanisation are the same, the chemical 
 result, the coefficient of vulcanisation, is in every case also the same, 
 though the physical results may, and do differ, often very widely. Thus, 
 in order to produce a vulcanisation product containing 3 per cent, of 
 combined sulphur in a fatigued rubber mixing requires, as a matter of 
 fact, no more sulphur than what will produce the same result in the same, 
 but only moderately worked, mixing. But, in order to obtain the same 
 physical vulcanisation result in the first as in the second case, a much 
 higher coefficient of vulcanisation, and therefore correspondingly more 
 sulphur, is needed. 
 
 The clear recognition of these facts is a matter of the highest scientific 
 and technical importance. It enables us at once to co-ordinate and reduce 
 to order the whole of the chaotic jumble of empirical facts in relation to 
 vulcanisation which up to now surround the practice of the process with 
 quite a mystic halo. The above instanced relations between the chemical 
 and the physical effects of vulcanisation demonstrate in the most 
 unmistakable manner that the process of vulcanisation i.e., the process of 
 the combination of sulphur with polyprene as a chemical reaction, proceeds 
 independently of the physical condition or configuration of the india 
 rubber colloid, bat the physical constants of the vulcanisation product 
 depend to a large extent upon the particular physical state of the india 
 rubber colloid induced by the conditions under which vulcanisation was 
 
THE CHEMISTRY OF INDIA RUBBER. 95 
 
 carried out. The vulcanising process is therefore a means by which we 
 are in a position to perpetuate any physical state to which the india rubber 
 colloid may be brought within the conditions required for the combination 
 of sulphur and polyprene. Or, somewhat more briefly: 
 
 4. As a chemical reaction the vulcanisation process is not influenced by the 
 physical state of the india rubber colloid ; but the physical state of the india 
 rubber colloid, while under vulcanisation, largely determines the physical con- 
 stants of the vulcanisation product. 
 
 This brief sentence contains the explanation for the often apparently 
 so utterly erratic and unaccountable behaviour of india rubber under 
 vulcanisation, as judged by the vulcanisation products. 
 
 A most striking illustration of the correctness of the above proposition 
 is furnished by the behaviour of recovered india rubber. In the chapter 
 on india rubber substitutes I have pointed out that there is no recovered 
 rubber which could be described as ' devulcanised ' india rubber, that all 
 the recovery processes can aim at and achieve is the restitution of more or 
 less perfect plasticity in the once vulcanised india rubber colloid. I have 
 further pointed out that there is no necessity whatever to aim at the de- 
 vulcanisation of the india rubber, that this needless aim was due to an 
 altogether erroneous conception of the nature of the india rubber colloid, 
 and of the process of vulcanisation. It is the presence of free ethylene 
 bonds in the india rubber which renders vulcanisation possible, and as in 
 all soft rubber goods only one or two at most out of thirty ethylene bonds x 
 can be occupied by sulphur, it is obvious that recovered soft rubber must be 
 capable of revulcanisation, or, rather, of repeated vulcanisation. This is even 
 true of ground soft vulcanised rubber, but in this case, fully in accordance 
 with the above statement (4), as the physical state of the rubber in this 
 case is that of a more or less dry cohesionless powder, revulcanisation, the 
 actual taking place of which can easily be proved analytically, cannot 
 produce a coherent solid unless the vulcanisation conditions be such as to 
 produce a fusing of the powder, and therefore a cementation of its grains 
 into one plastic mass. In the latter case this plastic condition is then the 
 physical state in which vulcanisation takes place, which in this case, there- 
 fore, results in the formation of a coherent solid. This is clearly the reason 
 why the employment of * rubber dust ' in the manufacture of ebonite yields 
 such surprisingly good results, as the much higher vulcanisation tempera- 
 tures applicable in this case are so much more favourable to the formation 
 of a continuous rubber plasma by cementation due to fusion. 
 
 Now, in the manufacture of recovered rubber the object to be aimed at 
 is not simply to leave this cementation to occur at the vulcanisation 
 temperature as best it may, but to predetermine it. This is the reason 
 why fixed solvents, notably mineral oils, must be present in every re- 
 vulcanisable recovered rubber. The effect of these fixed solvents is to 
 
 1 The truth of this statement, as will readily be seen, is independent of the question 
 of the molecular weight of india rubber. 
 
96 INDIA RUBBER AND ITS ANALYSIS. 
 
 produce at very moderate temperatures continuous, and more or less 
 already homogeneous, rubber plasmata. In this physical state, vulcanisation 
 supervening, the result is a coherent, continuous solid which in many cases 
 is indistinguishable from similar products made from new rubber. The 
 vulcanised product will be found the more satisfactory, the more easily the 
 recovered rubber used reassumes, at vulcanisation temperatures, the state 
 of a continuous plastic colloid. If we take any brand of recovered rubber 
 and free it from its oil by extraction with acetone, it forms, after drying, 
 an inelastic, almost friable mass. If we mix this with sulphur and attempt 
 to vulcanise it, the result is invariably disappointing. That vulcanisation 
 has taken place, analysis of the product will readily show ; but the oily 
 solvent having been abstracted, the particles of vulcanised India rubber 
 were unable to assume the physical condition necessary, in accordance 
 with the above fourth proposition, to ensure a physically satisfactory 
 vulcanisation product. 
 
 For theoretical reasons I am inclined to think that this proposition (4) 
 applies, not only to the vulcanisation process of india rubber, but, mutatis 
 mutandis, to all derivatives of all colloidal substances, the formation of 
 which does not involve the cessation of their colloidal state, or an essential 
 alteration of their ultimate structure or constitution. 
 
 On page 32 I have already stated that polyprene tetrabromide is gen- 
 erally obtained as a body forming a colloidal solution in chloroform, but 
 insoluble in benzene and all other solvents ; that in one case a product 
 insoluble even in chloroform was obtained ; in another case, however, a 
 product, not only soluble in chloroform, but also freely soluble in benzene, 
 resulted. The chemical composition of these compounds was exactly the 
 same in every case, so that these physical differences are very difficult to 
 account for. But I think that there is little doubt that they will find 
 their explanation in the generalisation expressed in the above fourth 
 proposition. 
 
 Of course it is quite possible that there may be found eventually a 
 concurrent chemical explanation. It is well known that variations of 
 temperature have a determining influence upon the formation of isomeric 
 substitution derivatives. Scores of examples of this influence may be 
 found amongst the isomeric halogen and nitro derivatives of all the 
 aromatic hydrocarbons. The cases are comparatively rare of such an 
 influence of different physical conditions upon substitution products, but a 
 certain number of such cases have been observed. Considering the large 
 number of ethylene bonds in the polyprene molecule, it is highly probable 
 that, under varying physical conditions, the sulphur addition characterising 
 the vulcanisation process might take effect in different places of the poly- 
 prene molecule, leading thereby to the formation of isomeric polyprene 
 sulphides. But it has yet to be proved that in such enormous molecular 
 aggregates isomerism of this order tangibly affects the physical properties 
 of the compounds. The fact that, in spite of numerous attempts, not the 
 
THE CHEMISTRY OF INDIA RUBBER. 97 
 
 slightest evidence has been obtained that the Para polyprene differs from 
 the polyprene of any of the other numerous brands of india rubber rather 
 points this way, as the great technical difference between them cannot be 
 understood on any other assumption. Further investigation of this im- 
 portant point is urgently required, but it will be noticed that the results 
 of such investigation could not affect our above fourth proposition, the 
 peculiar differences between the polyprene tetrabromides, above alluded to, 
 not being capable of explanation on the supposition of isomerism. 
 
 Vulcanisation with chloride of sulphur in the cold offers an exact 
 parallel to vulcanisation with sulphur at higher temperatures. In an in- 
 vestigation of this subject 1 1 showed, some years ago, that the vulcanisation 
 with chloride of sulphur is an addition process. The action of sulphur 
 monochloride upon unsaturated hydrocarbons was many years ago investi- 
 gated by GuTHRiE, 2 who obtained the addition products ethylene chloro- 
 sulphide (C 2 H 4 ) 2 S. 2 C1 2 and amylene chlorosulphide (C 5 H 10 ) 2 S 2 C1 2 . These 
 chlorosulphides are saturated substances, and, according to these formulae, 
 would have the constitution, 
 
 CH 3X 
 H,C Cl >C-C1 
 
 "I CH/ | 
 
 H 2 C S CH,.HC S 
 
 | and | 
 
 HoC-S CH 3 .HC S 
 
 " 
 
 I 
 H 2 C 
 
 Cl C Cl 
 
 CH 
 
 assuming the constitution of sulphur monochloride to be Cl S S Cl. 
 More recent investigations, however, tend to show that the constitution 
 of sulphur monochloride must be expressed by the formula C1 2 = S:S. 3 
 GUTHRIE'S addition products would have to be formulated thus : 
 
 H 2 C Cl 
 
 V>Xljj\ 
 
 ycc\ 
 
 1 
 
 CH/I 
 
 HA 
 
 \S:S 
 H 2 C/ 
 
 CH 3 .HC\ 
 and \S:S 
 
 1 
 
 CH 3 \ 
 
 H 2 C-C1 
 
 >C-C1 
 
 
 CH/ 
 
 1 " On the Vulcanisation of India Rubber," Journ. Soc. Chem. Tnd., 1894. 
 ' 2 Ann. Chem., 113, 270. 
 
 3 HENRIQUES.* by the action of sulphur monochloride upon j8-naphthol, obtained the 
 following bodies : 
 
 HO.C 10 H 6 S C 10 H 6 .OH dioxy-dinaphtyl-sulphide, . . M.P. 211 C. 
 HO.C 10 H 6 S S C 10 H 6 .OH dioxy-dinaphtyl-disulphide, . M.P. 166 C. 
 S:S.C 10 H 6 .OH 
 
 dithio-dioxy-dinaphtyl-disulphide, M.P. 141 C. 
 S:S.C 10 H 6 .OH 
 
 The first of these is the chief product of the reaction, and its constitution strongly 
 suggests that the formula of sulphur monochloride is to be expressed by S : SC1 2 , and not 
 
 * Ber., xxvii. p. 2993. 
 
 7 
 
98 INDIA RUBBER AND ITS ANALYSIS. 
 
 The difference for our purposes is rather insignificant, the main fact 
 remaining that this addition process results in the formation of saturated 
 compounds. 
 
 Assuming that the above compounds are typical of the interaction of 
 sulphur monochloride with doubly linked carbon atoms, we are justified in 
 assuming that the action of sulphur monochloride upon polyprene proceeds 
 on analogous lines. But the reaction between these two substances is so 
 violent, that comparatively dilute solutions should be employed, and it is 
 also necessary to take care that only india rubber previously freed from 
 its insoluble constituent in the manner described on page 7 is used. 
 
 The benzene serving as a solvent in these investigations should be 
 deprived of every trace of moisture by distillation over metallic sodium. 
 These precautions are necessary, as solutions of sulphur monochloride are 
 very rapidly decomposed by moist solvents, and also because the action of 
 sulphur monochloride upon the insoluble constituent of india rubber gives 
 rise to the formation of by-products, which, if not recognised as such, would 
 almost certainly vitiate our conclusions. 
 
 If, observing the above precautions, we prepare a 5 per cent, solution of 
 india rubber in benzene and add to it an excess of sulphur monochloride 
 also dissolved in benzene, the temperature rapidly rises, and within a few 
 seconds the solution is transformed into a firm, very transparent jelly. 
 The reaction is not accompanied by evolution of hydrochloric acid. The 
 product is allowed to stand for about twenty minutes ; it is then placed in 
 a bag of fine but strong muslin, immersed in pure benzene, and the jelly 
 squeezed through the meshes of the fabric. This is not a particularly 
 agreeable operation, but as the product is absolutely insoluble in all sol- 
 vents, it is the only means of reducing it to a satisfactory condition for 
 further purification. It is now obtained in the state of a mass of extremely 
 fine flakes, which are washed with benzene until every trace of the excess 
 of sulphur monochloride has disappeared. After this it is filtered, pressed 
 between filter paper, without previous drying transferred to a Soxhlet 
 
 by Cl S S Cl. The same observation applies to the third of these substances. The 
 constitution of the second body at first sight appears to suggest a reaction, which could 
 only be satisfactorily expressed by the adoption of the symmetrical formula for sulphur 
 monochloride. But it will readily be seen that the third compound, by elimination of 
 sulphur, must form the second of these bodies. The evidence of this investigation is 
 therefore strongly in favour of the asymmetrical formula for sulphur monochloride. 
 
 The observations recently made by LIPPMANN and POLLAK * on the action of sulphur 
 monochloride upon anthracene lead to the same conclusion. The product of this reaction, 
 anthracene di-thiochloride, appears to possess the constitution, 
 
 CL 
 
 which also supports the asymmetrical formula of sulphur monochloride. 
 
 * Ber. t xxxiv. p. 2767. 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
 99 
 
 extractor, and extracted with acetone until all soluble matter consist- 
 ing entirely of sulphur has disappeared. It is then dried at a low 
 temperature. 
 
 The dried product is obtained in the form of fine cream-coloured flakes 
 which are very friable. The substance is insoluble in all solvents, but the 
 coal tar hydrocarbons, chloroform, carbon bisulphide, nitrobenzene, terpenes, 
 and ethereal oils cause it to become transparent and to swell more or less 
 considerably. On heating it to temperatures above 140 C. with 
 mesitylene, cumene, cymene, or nitrobenzene, solution, accompanied by 
 evolution of hydrochloric acid, takes place, the product being decomposed. 
 In the dry state this decomposition already begins at 80 C. 
 
 On suspending this product in carbon bisulphide, and treating it with 
 a solution of bromine in the same solvent, no absorption of bromine takes 
 place. The product behaves, therefore, as a saturated compound. 
 
 On analysis it gave the following figures : 
 
 
 I. 
 
 II. 
 
 Calculated for 
 CioH 16 S 2 Cl 2 . 
 
 c, 
 
 43-61 
 
 43-51 
 
 44-28 
 
 H, . 
 
 6-21 
 
 6-14 
 
 5-90 
 
 S, . . . 
 
 23 -38 
 
 23-81 
 
 23-62 
 
 01, . . . 
 
 25-97 
 
 26-11 
 
 2619 
 
 
 
 These figures prove conclusively that 
 
 5. The product of the interaction of polyprene and sulphur monochloride 
 residts in the formation of an addition product of these two compounds, 
 possessing the empirical formula C W H 16 S 2 C1 2 . 
 
 The composition of this polyprene chlorosulphide is not affected by 
 preparing it in the presence of an excess of sulphur monochloride of 
 varying amount. In decreasing the proportion of chloride of sulphur, the 
 reaction proceeds much as before, the time within which the beginning of 
 the reaction, indicated by the mixed solutions becoming first stringy, then 
 gelatinous, lengthens quite inappreciably even when so low a proportion as 
 5 parts of sulphur monochloride to 100 parts of polyprene is used. But 
 the time in which the jellies formed reach their respective maximum 
 stiffness gradually increases as the proportion of chloride of sulphur used 
 grows less. 
 
 Provided the strength of the polyprene solution and that of the sulphur 
 monochloride remains constant say, 5 per cent, in both cases the 
 length of time that elapses before the first appearance of gelatinising 
 (pectinising) in the reacting mixtures varies not more than about ten 
 minutes between the limits of 120 and 5 parts of total sulphur mono- 
 chloride reacting upon 100 parts of polyprene. 
 
 With increasing dilution the rate at which the reaction proceeds 
 slackens considerably, though the final result remains thereby unaffected. 
 
IOO INDIA RUBBER AND ITS ANALYSIS. 
 
 If the concentration of the polyprene solution falls to 0*5 per cent., even an 
 excess of sulphur monochloride will only produce a peculiar ' stringy ' jelly. 
 Polyprene solutions of 0*25 per cent, strength remain apparently unaltered ; 
 whether in this case also polyprene chlorosulphide is formed or not, I have 
 not yet been able to ascertain. 
 
 The action of sulphur monochloride upon the insoluble constituent of 
 india rubber is distinct, but not characterised by the formation of any 
 tangible compound. It should, however, be mentioned that it is accom- 
 panied by a very marked evolution of hydrochloric acid. 
 
 It has already been stated that the compound C 10 H 16 S 2 C1 2 , the product 
 formed by the addition of 132 parts of polyprene and 135 parts of sulphur 
 monochloride, forms a brittle, friable mass. As we reduce the proportion 
 of the latter, the products formed lose their friability ; they become tougher, 
 merging by gradual, imperceptible degrees into more and more resilient 
 and distensible masses. Characteristic of the whole series of products 
 thus obtainable is the fact that they all contain sulphur and chlorine in 
 exactly equimolecular proportions. In other words, they are all addition 
 products of polyprene and sulphur monochloride. They also all appear to 
 be perfectly homogeneous products, but it is obvious that whether this, 
 homogeneity is not apparent rather than real must be decided by special 
 experiment. 
 
 For this purpose 1 5 grms. of pure Para rubber were dissolved in 1000 
 c.c. of pure benzene, and 5*1 grms. of sulphur monochloride were added. 
 In consequence of the great dilution, the reaction proceeded but slowly. 
 As soon as the solution had assumed the condition of a semi-liquid stringy 
 jelly, it was divided into two parts, the first of which was proceeded with 
 in the above described manner. To the second part a solution of 20 grms. 
 of Para in 200 grms. of benzene was added, and the whole vigorously 
 stirred, when it assumed almost the condition of an ordinary rubber 
 solution. This solution was poured into absolute alcohol, when a 
 voluminous, somewhat stringy precipitate was obtained. This was filtered 
 off, the filter residue washed, and finally dried in vacuo. 
 
 The product of the first portion was, on analysis, at once found to be 
 identical with our polyprene chlorosulphide C 10 H 16 S 2 C1 2 . The dry second 
 product formed a rather elastic, somewhat tacky mass, which, on analysis, 
 was found to contain 4'88 per cent, of sulphur and 5-32 per cent, of 
 chlorine. It contained, therefore, sulphur and chlorine in equimolecular 
 proportions, and was, in this respect, indistinguishable from the product 
 obtainable by adding to a solution of 100 grms. of Para 11 '42 grms. of 
 sulphur monochloride. 
 
 If now we treat this substance with benzene, we obtain, in the course of 
 a few days' standing, interrupted by frequent shaking, a slightly turbid 
 solution, which, after considerable dilution, may be filtered through fine 
 
 1 This is practically a repetition, quite recently made, of the experiment described in 
 my paper already alluded to in Journ. Chem. Soc., 1894. 
 
THE CHEMISTRY OF INDIA RUBBER. IOI 
 
 cambric on a hot-water funnel. A fine clear nitrate is thus obtained. On 
 the filter remains a finely distributed residue, which, after repeated washings 
 with warm benzene, was dried. It was thus obtained as a whitish non- 
 coherent mass, which, on analysis, was found to contain 23'46 per cent, of 
 sulphur and 25 '62 per cent, of chlorine. This substance was therefore 
 unquestionably our chlorosulphide C 10 H 16 S 2 C1 2 . It follows, therefore, that 
 the substance containing 4 - 88 per cent, of sulphur and 5*32 per cent, of 
 chlorine, out of which it was obtained, was simply a mixture of this chloro- 
 sulphide and free india rubber. 
 
 But if we now take the substance obtained by the direct interaction 
 of 100 grms. of Para and 11 '42 grms. of sulphur monochloride, the 
 chemical constants of which are exactly the same, and treat it with 
 benzene, we find that although the product swells considerably, not a 
 trace of India rubber passes into solution ; so that, while this product 
 may or may not be a homogeneous substance, it certainly does not con- 
 tain any unvulcanised india rubber. Its elasticity is much greater than 
 that of the above described mixture, and its stickiness very much less 
 marked. 
 
 But if a product with only 11*42 grms. of sulphur monochloride pro 
 100 grms. of polyprene does not contain any free polyprene, it is quite 
 obvious that between this product and the product C 10 H 16 S. 2 C1. 2 there must 
 exist a whole series of intermediate chlorosulphides, and that, in reality, the 
 compound C 10 H 16 S 2 C1 2 represents simply the upper limit of this series, 
 exactly corresponding to the upper limit, C 10 H 16 S 2 , in the polyprene 
 sulphide series. 
 
 In my paper, already quoted, on the vulcanisation of india rubber with 
 sulphur monochloride I have shown that a product containing 4*26 per 
 cent, of sulphur arid 4 '71 per cent, of chlorine or, say, 9 per cent, of 
 S 2 C1 2 and therefore closely approximating in its composition to the 
 statistical formula (C 10 H 16 ) 10 S 2 C1 2 , or C 100 H 160 S 2 C1 2 , does not contain any 
 uncombined polyprene ; and from this fact I drew the conclusion that the 
 polyprene molecule could not be less than C 50 H 80 , and that it might possibly 
 be found to be still higher. 
 
 Whether this body, C 100 H 160 S. 2 C1 2 , is really a homogeneous compound 
 remains to be proved, but for the purpose of our argument the question is 
 comparatively irrelevant. The mere fact, demonstrated by experiment, that 
 this body contains no free polyprene is in itself sufficient proof that its 
 molecule cannot be expressed by any smaller formula than C 100 H 160 S 2 C1 2 . 
 But the existence of such a body is entirely sufficient to draw the following 
 conclusion : 
 
 6. The process of vulcanisation with sulphur monochloride consists in the 
 formation of a continuous series of addition products of polyprene and sulphur 
 monochloride, irith a polyprene chlorosulphide W H 16 S 2 C1. 2 as the upper, and 
 a compound C m H l6Q S 2 Cl 2 as the Imcer limit, as far as at present ascer- 
 tained, of the series. Physically this series is characterised by the decrease 
 
IO2 INDIA RUBBER AND ITS ANALYSIS. 
 
 of distensibility and the increase of rigiditij from the lower to the upper 
 limit. 
 
 It will be seen that this conclusion is a perfect parallel to the third of 
 our conclusions arrived at respecting the process of vulcanisation with 
 sulphur. 
 
 The difficulty of locating the lower limit of the chlorosulphide series 
 with a more satisfactory degree of accuracy appears to be rather physical 
 than chemical, as will be seen from the following experiment. 
 
 ^Nine grms of pure Para were dissolved in 200 c.c. of absolutely dry 
 benzene. To this solution was added 0*144 grm. ( = 1*6 %) of SgCl^dissolved 
 in 10 c.c. of carbon bisulphide. The whole was violently agitated and 
 allowed to stand for twelve hours in a dry atmosphere. At the end of 
 this time the mass formed a stringy jelly, which could only with difficulty 
 be distributed in 1000 c.c. of benzene. Part of this jelly passed into 
 solution, the rest remained suspended in the solvent as an exceedingly tender 
 colloidal sponge. The solution was separated by decantation from the latter, 
 which was further treated with benzene. In attempting to repeat this opera- 
 tion until this sponge, the supposed lowest member of the above named chloro- 
 sulphide series, yielded no more india rubber to the solvent, it was found that this 
 colloidal sponge gradually dissolved, and eventually altogether disappeared. 
 It was then found that already the first treatment of the jelly with benzene 
 yielded a solution containing combined chlorine and sulphur very nearly 
 in the proportions in which these, in the form of S 2 C1 2 , had been added to 
 the india rubber. Further, on carrying the washing process with benzene 
 just short of the complete solution of the above mentioned colloidal sponge, 
 removing the latter as carefully as possible from the surrounding solution, 
 and coagulating it by means of absolute alcohol, the india rubber colloid 
 thus obtained contained only mere traces of sulphur and chlorine. We 
 would have to assume, therefore, either that the jelly prepared as stated 
 contained no free india rubber, and that the chlorosulphide it represents is 
 soluble in benzene, or that the lowest possible chlorosulphide of polyprene 
 is soluble in rubber solutions. Of these two conclusions I prefer the last 
 one, as we have no tangible evidence on which to adopt the former,, whereas 
 the latter is not only feasible, but is, indeed, only what we should expect 
 when forming within colloidal solutions bodies normally insoluble. 1 
 
 The solvent action of colloidal solutions upon insoluble colloids depends, 
 however, very much upon the concentration of the former, and ceases 
 below a certain point. What that point is appears to depend largely upon 
 the degree of constitutional similarity subsisting between the two colloids, so 
 that the greater this similarity, the lower the concentration of the soluble 
 colloid 2 necessary to keep the insoluble colloid 3 in solution. I therefore 
 
 1 LOBUY DE BRUYN'S observations on the result of the formation of ordinarily in- 
 soluble bodies in colloidal solutions of gelatine are a case in point. See page 77. 
 
 2 India rubber, polyprene, in this case. 
 
 3 Polyprene chlorosulphide in this case. 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
 103 
 
 undertook a series of experiments, with the view of ascertaining as closely 
 as possible the quantitative conditions under which a polyprene chloro- 
 sulphide is obtained which should be just insoluble in benzene or any of 
 the other india rubber solvents. 
 
 From the result of the optical investigation of polyprene, GLADSTONE and 
 HIBBERT, in their repeatedly quoted paper, arrived at the conclusion that 
 the smallest possible formula of india rubber had to be expressed as C 10 H 1(i . 
 This formula, indeed, agrees very well with the composition of quite a number 
 of the derivatives of polyprene described in the foregoing pages. But in the 
 case of polyprene iodide we were already compelled to double it, and the 
 existence of a chlorosulphide with only 9 per cent, of, S 2 C1 2 rendered it 
 necessary to quintuple it. It should, however, be borne in mind that the 
 proof of the homogeneous nature of this chlorosulphide is wanting, so that 
 all it proves is that the reacting unit of polyprene cannot be less than 
 (C 10 H 16 ) 5 = C 50 H 80 ; it leaves it uncertain whether it is not perhaps C 60 H % 
 or some higher multiple of C 10 H 16 . For this reason I undertook the pre- 
 paration of a series of chlorosulphides of the general formula (C 10 H 16 ) H S 2 CL> 
 rising from the above described chlorosulphide (C 50 H 80 ). 2 S 2 C1 2 in increments 
 of (C 10 H 16 ) 2 = C 20 H 32 . All the experiments were carried out with a 2 per 
 cent, solution of very pure polyprene in benzene, to which the sulphur 
 monochloride was added, the whole of the calculated quantity at once, in 
 the form of a solution in carbon bisulphide containing O'Ol grm. S 2 C1 2 in 
 1 c.c. In every case the mixture was left standing for twelve hours, the 
 gelatinous chlorosulphide vigorously stirred up in 1000 c.c. of benzene 
 and allowed to stand for twelve hours with the colloidal sponge floating 
 in it, the mass being from time to time vigorously agitated. Part of the 
 benzene solution was then filtered off and shaken up with three times its 
 volume of absolute alcohol, so as to precipitate every trace of polyprene 
 present. 
 
 The result of this series of experiments is given in the following table, 
 embodying, of course, only that part of the series approaching in either 
 direction the critical point : 
 
 Statistical Composition. 
 
 Percentage S 2 C1 2 . 
 
 Examination of Benzene Solution. 
 
 Cgoo-HsaO' Sgdg 
 
 p TT Q pi 
 
 9-03 
 
 472 
 4-31 
 3-97 
 
 3-67 
 3-42 
 3-20 
 
 ! Starting term of whole series. 
 No trace of polyprene in solution. 
 No trace of polyprene in solution. 
 Very slight opalescence. 
 Distinct opalescence, but no trace 
 of precipitate. 
 Faint but distinct precipitate. 
 Very marked, stringy precipitate. 
 Copious precipitate. 
 
 These experiments tend to show that the lower limit of the chloro- 
 sulphide series is represented by a body, the composition of which is most 
 
104 INDIA RUBBER AND ITS ANALYSIS. 
 
 probably expressed by C 240 H 384 S 2 C1 2 , and that the reacting unit of poly- 
 prene, its molecular weight, must be taken either as C 240 H 384 or C 120 H 192 , 
 according to whether we choose to represent the mechanism of the 
 addition of sulphur monochloride to polyprene by 
 Cl Ck 
 
 \ \P TT 
 
 \ /*-'i'>o ri :i9 
 
 ) or S:S/ 
 
 The molecular weight expressed by the first of these diagrams would, 
 indeed, be in fair agreement with the result of GLADSTONE and HIBBBRT'S 
 cyroscopic determination, 1 but this is perhaps hardly a sufficient reason to 
 reject the second diagram with a polyprene molecule only one half that of 
 the former, as the question of aggregation in colloidal solutions yet remains 
 to be decided. 
 
 Incidentally also in this case attention must be drawn to the fact of 
 the curious coincidence of the above set forth ' lower limit ' with what is 
 recognised as the best practice in manufacturing i.e., a 'cold cure' re- 
 sulting in a product containing from 3 '5 to 4 per cent, of sulphur mono- 
 chloride in the india rubber. 
 
 We may now amend our former statement (6) by saying : 
 
 7. The series of chlorosulphides formed ~by the interaction of sulphur 
 monochloride upon polyprene ranges from a body (G-^H^S^Cl^ 
 C 2 ^H M8 S 4S C1 4S as the upper limit, to a body C 240 H MB S 2 Cl z as the lower 
 limit. 
 
 The physical behaviour of the above described colloidal polyprene- 
 chlorosulphide ' sponges ' is highly interesting. It was found to depend 
 very largely upon the concentration of the solution in which the ' sponge ' 
 has been formed. Thus, in forming a chlorosulphide of exactly identical 
 composition in a series of solutions of polyprene varying in concentration, 
 we obtain a number of chlorosulphide sponges of identical chemical 
 composition, but markedly differing in their behaviour to solvents, and 
 also in the strength of the filaments obtainable from them on drying. This 
 difference may be stated thus : 
 
 8. The, greater the dilution of a solution of polyprene, the greater the 
 tendency of the chlorosulphide formed therein to break down or disintegrate 
 under the influence of the addition of further quantities of the solvent. 
 
 It will be found in accordance with this that the lowest terms of the 
 series of polyprene chlorosulphides, when formed in 0'5 per cent, solutions 
 of polyprene in benzene, form extremely mobile jellies, which can be diluted 
 or distended by the addition of fresh quantities of solvent, almost as 
 readily as if they were moderately concentrated rubber solutions. More- 
 over, in carrying out this experiment with 0*25 per cent, solutions of 
 polyprene, no jellies are obtained at all, nor is there even a sensible alteration 
 in the viscosity of the solution, although the chlorosulphides are formed as 
 1 Phil. Mag., 1889, p. 39. 
 
THE CHEMISTRY OF INDIA RUBBER. 105 
 
 usual a fact easily demonstrable by their precipitation with absolute 
 alcohol and subsequent analysis. 
 
 Exactly similar observations may be made by forming chlorosulphides 
 of identical compositions in solutions of polyprene of the same strength, 
 but in different solvents. In treating polyprene solutions of equal con- 
 centration in different solvents with the same proportion of sulphur mono- 
 chloride, jellies exhibiting surprising differences are obtained, and on 
 subsequently drying these jellies, the tenacity of the films obtained from 
 the different solutions varies in a remarkable degree. 
 
 That solutions of the same substance, and possessing equal degrees of 
 concentration, may vary very considerably in their structure or constitution, 
 or in the state of aggregation of the dissolved body, is quite a familiar 
 phenomenon in the solutions of crystalloid substances. That it should be 
 much more pronounced in colloidal bodies is, therefore, to be expected. 
 Bearing this in mind, the above described phenomena become at once 
 intelligible in the light of the important conclusion embodied in our pro- 
 position (4) on page 95. l 
 
 9. The difference in the physical state of the above described solutions of 
 polyprene after their conversion into chlorosulphides is determined by the 
 difference of the physical conditions of the polyprene itself in different solvents, 
 or in the same solvents at different concentrations. 
 
 In this statement will be found the clue to all the most perplexing 
 facts observed in the process of vulcanising with sulphur monochloride, and 
 even of those at first sight quite unaccountable. 
 
 The indifference of polyprene to sulphur at ordinary temperatures is 
 one of the chief causes of the difficulty of practically controlling the 
 process, the chief advantage of vulcanisation with sulphur monochloride 
 being the promptness of the reaction. c Practical ' workers have made this 
 certainly striking difference the basis for some extraordinary conclusions. 2 
 In reality this difference is no greater than that which we find in the 
 behaviour of unsaturated hydrocarbons with sulphur and sulphur mono- 
 chloride respectively, and probably less. 
 
 As regards the action of the sulphur upon polyprene, it is perfectly 
 obvious that the slowness of its action is due to the fact that sulphur 
 vapours do not reach the molecular state at which their reactive energy 
 is at its highest until very high temperatures are reached. The sulphur 
 molecule assumes its normal density, S 2 = 64, only at temperatures from 
 about 700 to 1500 C., as will be seen from the following figures : 
 
 Bineau, 714 to 743 C. D = 27 
 
 Deville and Troost, . . . 860 to 1040 C. D = 2'13 
 
 Bineau, 834 to'1162 C. D = 2'36 
 
 Y.Meyer, . . . . 1560 C. D = 2'17 
 
 1 In all these experiments the concentration of the solutions is the gramme weight of 
 substance contained in 100 c.c. of the solvent. 
 
 2 See Gummi Zeitung, 1899, p. 17 (HOHN) ; ibid., p. 247 (MINDER). 
 
io6 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 The calculated density for the S 2 molecule is D = 2-216. The 
 vulcanisation temperature of polyprene is necessarily much lower than the 
 above temperatures which lie between bright red heat to white heat. 
 The highest vulcanising temperature which has ever been proposed and 
 used does not exceed 185 C., and in ordinary practice it is rarely more 
 than 140 C. But at these temperatures the sulphur molecule is in a 
 state of aggregation largely exceeding its normal value of S 2 = 64. Indeed, 
 BLEIER and KOHN l ascertained the following values for temperatures 
 below the boiling point of sulphur (440 C.) : 
 
 In the Yapour of 
 
 Temperature. 
 
 Molecular 
 Weight. 
 
 Molecular 
 Aggregation. 
 
 Diphenylamine, 
 
 310 C. 
 
 237-9 
 
 7-43 
 
 Arayl benzoate, 
 
 262 
 
 240-1 
 
 7-50 
 
 Quinoline, 
 
 236 ,, 
 
 245-2 
 
 7-66 
 
 Ethyl benzoate, 
 
 212 ,, 
 
 249-6 
 
 7'80 
 
 Dimethylaniline, 
 
 193 ,, 
 
 251-1 
 
 7-84 
 
 These figures leave no doubt that at practical vulcanising temperatures 
 the molecule of the sulphur vapour present differs from S 8 only by an 
 extremely minute fraction, and as it has been shown by BECKMANN that 
 the sulphur in solution consists of S 8 molecules only, no doubt exists that 
 the molecule of melted (colloidal) sulphur is at least of the same order of 
 magnitude, not improbably much larger. But as vulcanisation can only 
 be effected by sulphur molecules S 2 , it is obvious that the rate of 
 vulcanisation at any given temperature must be determined by the state 
 of dissociation of the sulphur at that temperature i.e., by the relative 
 proportions of S 8 and S 2 molecules present. At ordinary temperatures 
 dissociation is almost nil, hence the fact that sulphur at these temperatures 
 cannot vulcanise india rubber, or does so to quite an unappreciable extent 
 only. In order, therefore, to induce vulcanisation the combination of 
 sulphur and polyprene we must increase the state of dissociation of the 
 sulphur. This, as the above figures conclusively show, is effected by 
 higher temperatures, and accordingly we find that at increasingly higher 
 temperatures vulcanisation proceeds at a rapidly increasing rate. 
 
 The necessity for continuing the heating throughout the vulcanising 
 operation is to be found in the circumstance that the energy consumed in 
 the dissociation of S 8 into S 2 molecules is very high, and also that the 
 heat of combination of polyprene sulphide is very low. Otherwise we 
 should expect that the vulcanising reaction, once started, would proceed 
 spontaneously without the further application of extraneous heat. 
 
 As this point can readily be proved by experiment, I considered it 
 worth while doing so ; but as the reaction between polyprene and sulphur 
 does not lend itself to calorimetrical determinations, the more circuitous 
 1 Ber., 1900 [33], p. 50. 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
 lO/ 
 
 plan had to be adopted of determining the heat of combustion of polyprene, 
 and of a polyprene sulphide of known composition, and to calculate from 
 these data the heat of formation of polyprene sulphide. 
 
 The polyprene used in these determinations was obtained by most 
 careful purification of specially selected Para rubber. The polyprene 
 sulphide used was prepared from the same lot of Para rubber by mixing it,, 
 after thorough purification, with 10 per cent, of sulphur, vulcanising it 
 immersed in water for two hours at 135 C., extracting all the free sulphur 
 with acetone, and subsequently drying the product in a current of carbon 
 dioxide at 80 C. The following figures were obtained : 
 
 Calories per Gram 
 
 Water Liquid. 
 
 10,669 'c. 
 
 Calories per Gram 
 Water Liquid. 
 
 10,215 2 c. 
 
 Polyprene. 
 
 Molecular Combus- Calculated for 
 tion Heat, C, H W . 10C,0 2 + 8H 2 ,0. 
 1450 C. 1529 C." 
 
 Polyprene Sulphide : 3 '8 per cent. S. 
 
 Molecular Combus- 
 tion Heat. 
 MoL= 141-87. 
 1444 C. 
 
 Calculated for 
 C 10 H 16 S . 17 . 
 
 1474 C. 
 
 Heat of Formation, 
 79 C. 
 
 Heat of Formation. 
 30 C. 
 
 It is well known that, owing to the high combustion heat of organic 
 substances and their low formation heat, the calculation of the latter in 
 the above manner is apt to involve considerable error. These are, however, 
 not of such an order of magnitude as to vitiate the above figure for the heat 
 of formation of polyprene sulphide for the purpose of our discussion. In- 
 deed, this figure amply suffices to demonstrate the above proposition, that 
 the heat of formation of polyprene sulphide is low ; and this, together with 
 the energy required for the gradual transformation of the S 8 molecules into 
 S 2 molecules, is sufficient to satisfactorily account for the indifference of 
 polyprene to sulphur at ordinary temperatures, as also for the necessity 
 of continuous heating during the process of vulcanising at higher 
 temperatures. 
 
 It is incidentally interesting to compare the thermochemical data of 
 some of the cyclical terpenes with those above given for polyprene, which 
 we are regarding as an olefinic compound : 
 
 Citrene. 3 
 
 Calories per Gram. 
 10, 959 c. 
 
 Moiecuiar uomous- 
 tion Heat. 
 1490 C. 
 
 Calculated. 
 1529 C. 
 
 Heat of Formation. 
 
 OQ /Y 
 
 oy c/. 
 
 
 Pinene. 3 
 
 
 
 Calories per Gram. 
 10, 852 c. 
 
 Molecular Combus- 
 tion Heat. 
 1475 C. 
 
 Calculated. 
 1529 C. 
 
 Heat of Formation. 
 54 C. 
 
 1 Mean of five determinations. 
 
 2 Mean of three determinations ; S determined as S0 3 , aq. 
 
 3 FAVRE and SILBERMANX, Ann. chim. pkys. (3), 34, 434. 
 
108 INDIA RUBBER AND ITS ANALYSIS. 
 
 Assuming all these figures to be reliable, they reveal the curious fact 
 that the heat of formation of these terpenes is much lower than that of 
 polyprene. The circumstance that the last named product, on dry 
 distillation, is almost quantitatively converted into cycloterpenes, would 
 lead one to suppose that these, as stabler forms, would possess a higher 
 heat of formation. This, quite apart from the fact that colloidal com- 
 pounds are generally assumed to possess a higher energy content i.e., a 
 lower heat of formation than the corresponding crystalloids. That this, 
 at least as a generalisation, is not correct, appears also to follow from the 
 observation due to BERTHELOT, 1 that the conversion of soluble amorphous 
 (colloidal) sulphur into the rhombic (octahedral) modification is accom- 
 panied by a negative temperature reaction of 160 calories per gram- 
 molecule. 
 
 The important part played in the vulcanising process by the dissociation 
 of the complex sulphur molecules naturally suggests the adoption of means 
 favouring this process, either for the purpose of reducing the time or the 
 temperature required in practical working. One device of this kind has 
 been in practical use almost ever since the discovery of the process of 
 vulcanisation, and consists in the use of 'sulphur carriers.' All substances 
 to which this term applies are sulphides of the heavy metals. This state- 
 ment might appear to be in opposition to the fact that lead compounds, 
 such as litharge, lead carbonate, lead sulphide, lead chloride, lead thio- 
 sulphate, antimony oxide, are also able to act as sulphur carriers. But I 
 contend that these compounds are capable of this action only owing to 
 their more or less rapid conversion into sulphides under the influence of 
 the melted sulphur and its vapours. This is strongly supported by the 
 fact that the best of the above named sulphur carriers is lead thiosulphate, 
 which already at water-bath temperature decomposes into lead sulphate and 
 lead sulphide. All metallic sulphides possess a more or less pronounced 
 tendency to form polysulphides, the vulcanising action of which has already 
 been demonstrated by GIRARD; and the action of these sulphides, or 
 sulphur carriers, would appear, therefore, to consist in the acceleration of 
 the splitting up of the complex sulphur molecules. No explanation of the 
 action of these bodies is to be obtained from the thermochemical data of 
 the reactions in question, as the intervention of any intermediate stages 
 in the reaction leaves the energy change of the reacting system unaffected. 
 It is much more probable that the catalytic action of these sulphur 
 carriers rather depends upon the matter of chemical equilibrium, and 
 that their presence tends to render it unstable, thus hastening progress 
 towards a lower potential, which in this case is represented by polyprene 
 sulphide. 
 
 There is, however, another possibility of increasing the state of 
 dissociation of the sulphur molecules, which consists in the reduction of 
 the pressure under which the reaction (vulcanisation) takes place. We owe 
 1 Essai de Mecaniyue Chimiyue, 1879, i. 553. 
 
THE CHEMISTRY OF INDIA RUBBER. 
 
 IO9 
 
 to H. BiLTZ 1 an exhaustive series of determinations of the density of 
 sulphur vapour under reduced pressure at 440 C., which show to what 
 a considerable extent the dissociation of the sulphur molecules increases 
 under these conditions. I give the results of the observations of the above 
 named author in the following table. I have added a third column, 
 showing at a glance the mean atomic aggregation the sulphur vapour 
 possesses in each case. 
 
 Pressure, 
 mm. 
 Mercury. 
 
 Vapour 
 Density. 
 
 Molecule. 
 
 Pressure, 
 mm. 
 Mercury. 
 
 Vapour 
 Density. 
 
 Molecule. 
 
 Pressure, 
 mm. 
 Mercury. 
 
 Vapour 
 Density. 
 
 Molecule. 
 
 
 
 
 
 
 
 
 i 
 
 14-0 
 
 4'85 
 
 4*37 
 
 48'2 
 
 7-24 
 
 6-52 
 
 203-6 
 
 7-84 
 
 7-07 
 
 14-4 
 
 5-30 
 
 478 
 
 64-2 
 
 7-28 
 
 6-56 
 
 204-2 
 
 7-95 
 
 7-17 
 
 16-1 
 
 5-52 
 
 4-96 
 
 71-4 
 
 7-17 
 
 6-46 
 
 261-2 
 
 7-85 
 
 7-08 
 
 19-1 
 
 5-40 
 
 4-87 
 
 81-6 
 
 7*51 
 
 677 
 
 318'5 
 
 8-13 
 
 7-33 
 
 20-5 
 
 6-08 
 
 5'48 
 
 82-6 
 
 7-58 
 
 6-81 
 
 326-4 
 
 7-87 
 
 7-10 
 
 25-6 
 
 6 04 
 
 5-45 
 
 83-0 
 
 7'51 
 
 677 
 
 411-5 
 
 8-03 
 
 7-24 
 
 26-9 
 
 6 53 
 
 5-89 
 
 90-8 
 
 7-39 
 
 6'66 
 
 458-1 
 
 8-18 
 
 7-38 
 
 31-5 
 
 6-37 
 
 5-74 
 
 95-2 
 
 7-41 
 
 6-68 
 
 458-2 
 
 8-08 
 
 7-29 
 
 31-6 
 
 6-27 
 
 5'65 
 
 104-2 
 
 7'63 
 
 6'88 
 
 536-1 
 
 7-82 
 
 7-05 
 
 39-1 
 
 6-60 
 
 6-02 
 
 119-1 
 
 7-86 
 
 7'09 
 
 539-2 
 
 7-89 
 
 7-12 
 
 481 
 
 7-20 
 
 6'49 
 
 
 
 
 
 
 
 From these figures it would appear that the dissociation of sulphur 
 vapour at constant temperatures increases very considerably under the 
 influence of reduced pressures. The suggestion of utilising this fact for 
 the quickening of the vulcanisation process naturally occurs ; and inversely, 
 the observation of this quickening action would lend powerful support to 
 the above pronounced theory of vulcanisation. The technical application 
 of this suggestion is, however, out of the question, as it would invariably 
 result in the 'blowing ' of the goods treated. For this reason vulcanisation 
 in the vast majority of cases is carried out in presses, or in pans into which 
 live steam at from 2 -5 to 5 atmospheres (38 to 75 Ibs.) is admitted, or in 
 steam-jacketed pans charged with compressed air to a pressure up to 10 
 atmospheres (150 Ibs.). According to the above remarks, we should expect 
 the process of vulcanisation to suffer a retardation, owing to the reduced 
 rate of dissociation of the sulphur vapour. That this is the case there can 
 be no doubt whatever, but this effect can be minimised by raising the 
 temperature, and as with rising temperature the rate of progress of the 
 vulcanisation increases much more rapidly in proportion to the retarding 
 effect due to increased pressure, this latter drawback is not of much 
 practical moment. 
 
 The views on vulcanisation above set forth also explain at once the 
 fact that, in order to effect vulcanisation within a time interval satisfactory 
 for practical purposes, we are compelled to employ what is in all cases a 
 considerable excess of sulphur over and above the quantity entering into 
 
 1 Ber., 34, 2493. 
 
I IO INDIA RUBBER AND ITS ANALYSIS. 
 
 combination with the india rubber. This excess has obviously the effect 
 of increasing the total number of dissociated (S 2 ) sulphur molecules 
 present, thereby accelerating vulcanisation, which, as is well known to all 
 rubber manufacturers, proceeds the quicker, the larger the proportion of 
 sulphur employed. The uncombined excess of sulphur remaining after 
 vulcanisation in the india rubber does not crystallise on cooling, but re- 
 mains distributed through its mass in extremely small globules. This can 
 easily be demonstrated by a vulcanisation experiment under the microscope, 
 "but unfortunately I have found it impossible to produce a satisfactory 
 photograph of this experiment. After a certain length of exposure, the 
 vulcanised rubber is covered with a fine greyish deposit consisting of 
 minute sulphur crystals, which in course of time become, on close 
 examination, visible to the naked eye, and are easily seen by examining the 
 object with a low-power microscope. In fig. 10 (Plate II.) will be seen a low- 
 power photomicrograph of the surface of a piece of drab sheeting, showing 
 the splendid crystallisation of sulphur on its surface. This phenomenon is 
 technically known as ' blooming ' or sulphuring up, and is generally sup- 
 posed to be due to the sulphur ' efflorescing ' in some unaccountable way 
 to the surface. The ignorance of the real state of affairs is the reason that 
 often the most inadequate means are adopted, or the use of some quack 
 preparation, believed to possess some occult virtue, resorted to, to prevent 
 this efflorescence. 
 
 In order to clearly understand the nature of this phenomenon, we must 
 examine the nature of the above mentioned sulphur globules which are 
 present in the india rubber for some time after vulcanisation. On 
 examining these globules under the microscope, it will be found that 
 they have no action upon polarised light. They are therefore structure- 
 less, and do not consist of sulphur in the crystallised state. In time they 
 crystallise, and subsequently form centres from which radiate in all 
 directions beautiful dendritical crystals of sulphur. 
 
 These facts may be conveniently studied by producing these globules 
 outside the india rubber by allowing sulphur vapours to deposit upon a 
 micro cover glass. Successful preparations are easily obtained by placing 
 in the centre of a small porcelain basin one of the metal rings used for 
 making cells for microscopic objects. In the centre of this ring is laid 
 a minute crystal or two of sulphur, and a cover glass deposited upon 
 the ring. The basin so prepared is placed upon an asbestos sheet 
 and heated from underneath with a small flame. An iridescent coating 
 of sulphur deposits upon the under side of the cover glass, and an 
 examination of this coating at once reveals the fact that this deposit 
 consists of small globules of colloidal sulphur. The globules are largest 
 in the centre ; towards the periphery of the cover glass they gradually 
 diminish in size, passing in the outermost regions below the limits of 
 microscopic vision. The appearance of this deposit is shown in fig. 11 
 (Plate II.). 
 
Welter's India Rubber. 
 
 PLATE II. 
 
 FIG. 10. Efflorescence upon India rubber goods. 
 Photo- micrograph : Apochrom. Zeiss 16 mm., Projec. Eyep. 4. x 150. 
 
 FIG. 11. Globules of supervised (colloidal) sulphur. 
 Photo-micrograph : Apochrom. Zeiss 4 mm., Projec. Eyep. 4. x 600. 
 
 Char Jen Griffin < Co., Ltd. 
 
 To face p. 110. 
 
PLATE III. 
 
 Weber's India Rubber. 
 
 
 FIG. 12. Spontaneous crystallisation in deposit of sulphur globules. 
 Photo-micrograph : Apochrom. Zeiss 4 mm., Projec. Eyep. 4. x 600. 
 
 FIG. 13. Absorption of sulphur globules by sulphur crystal. 
 Photo-micrograph : Apochrom. Zeiss 4 mm., Projec. Eyep. 4. x 600. 
 
 Charles Griffin & Co., Ltd. 
 
 To face p. Ill, 
 
THE CHEMISTRY OF INDIA RUBBER. I I I 
 
 . If the deposit has been produced upon a scrupulously clean cover 
 glass, the preparation may be kept unaltered for weeks. Several times 
 I succeeded in keeping them for such a length of time that the de- 
 posit completely volatilised without undergoing any change. As these 
 globules consist of colloidal sulphur, they have no effect upon polarised 
 light. 
 
 As a rule, after a few days' standing crystallisation begins at one point 
 or another of these preparations. The first sign is always one or more of 
 the globules becoming opaque and distinctly yellow. The crystalline 
 structure of these globules is at once shown by the appearance of axial 
 crosses on examining them under the micro-polariscope. The surface of 
 these crystalline globules in course of time becomes covered with incrusta- 
 tions of crystallised sulphur, which often assume comparatively large dimen- 
 sions, evidently at the expense of the still colloidal and brilliantly trans- 
 parent globules surrounding them (fig. 12, Plate III.). This curious 
 phenomenon is undoubtedly due to the vapour tension of colloidal sul- 
 phur being much higher than that of crystallised sulphur, so that each 
 crystallised globule becomes a centre of low pressure, towards which, 
 consequently, the sulphur vapours distil, to condense by crystallising. 
 This leads to the gradual consumption of the colloidal globules by the 
 crystallised globules surrounding them, those nearest suffering most in 
 this respect, so that eventually the crystallised globule becomes sur- 
 rounded by a zone of clear glass. All this is very clearly shown by fig. 13 
 (Plate III.). 
 
 Occasionally the crystalline needles forming upon a crystallised globule 
 extend so rapidly as to make contact with one of the colloidal globules. 
 The instant this occurs the globule becomes turbid, and almost at the same 
 time, but always on the opposite side to that where the contact took place, 
 a crop of extremely fine needles appears, and these making similar contact 
 with adjacent globules, produce eventually very extraordinary dendritic 
 designs consisting of crystallised sulphur. The following rough sketch will 
 muke the phenomena just described clear : 
 
 FIG. 14. Spread of crystallisation from globule to globule in sulphur deposit. 
 
 This phenomenon is not always observable, but I have found that it can 
 at any time be produced with perfect ease by securing the cover glass 
 carrying the sulphur deposit upon a micro slide, so as to prevent disturb- 
 ance of the sulphur deposit by contact with the slide. 1 A few drops of 
 
 1 This is easily done by pasting across the slide two narrow strips of gummed paper 
 at a distance somewhat less than the diameter of the cover glass. The latter, the 
 
I I 2 INDIA RUBBER AND ITS ANALYSIS. 
 
 alcohol are run between the slide and the cover glass, when within a few 
 seconds the above described phenomenon may be observed taking pLice in 
 every direction. Fig. 15 (see Frontispiece) gives an idea of the magnificent 
 dendritic designs produced, the development of which is one of the most 
 fascinating and interesting sights conceivable. This is, no doubt, a matter 
 deserving further careful study from the point of view of molecular physics 
 in general. 1 
 
 Incidentally these phenomena enable us to obtain a perfect insight into 
 the cause and nature of the 'sulphuring up' in rubber goods a subject 
 which up to now has been shrouded in a great deal of mystery. It becomes 
 at once clear that if we adopt means to start the crystallisation of the 
 minute globules of sulphur distributed throughout the mass of the india 
 rubber, in the inner parts of the latter no * sulphuring up ' can take place. 
 As a matter of fact, the vapours of the colloidal sulphur globules 
 lodging on the surface will actually travel into the rubber substance 
 to crystallise there. But, on the other hand, should crystallisation 
 begin upon the surface, the sulphur will pass from the colloidal globules 
 in the inside to the surface, and produce there a frequently quite 
 prodigious crystallisation, which is as a rule looked upon with extreme 
 disfavour. 2 
 
 In ebonite the trouble of ' sulphuring up ' is unknown. This, in the 
 light of the above, must appear as rather surprising, seeing that even high 
 grades of ebonite rarely contain less than 15 per cent, of free sulphur, and 
 often up to 60 or 70 per cent. We must therefore assume that ebonite 
 represents a solid colloidal solution of the colloidal polyprene sulphide 
 C 10 H 16 S in colloidal sulphur. Nor is this assumption without experimental 
 support. Indeed, the fact that ebonite under the microscope presents the 
 appearance, as shown in fig. 8, of a homogeneous mass, although it con- 
 tains always a large amount of free sulphur, appears to me capable of no 
 other explanation. 
 
 The process of vulcanising with sulphur monochloride is, of course, 
 characterised by the fact that it takes place at ordinary temperatures. 
 Indeed, so energetic is the reaction in this case, that in employing this 
 process, and in order to render it controllable, the sulphur monochloride 
 must be used in a state of considerable dilution. These facts have been 
 just as much misunderstood as those in reference to the process of 
 vulcanising with sulphur. 
 
 At first sight it might appear strange that a compound of sulphur should 
 
 sulphur covered face downwards, is then placed upon the slide so that it rests upon these 
 strips, upon which it is secured with two similar strips. 
 
 1 To those interested in molecular physics, BUTSCHLI'S Microstrukturen des erstarrten 
 Schwefels, Leipzig, 1900, may be strongly recommended. 
 
 2 I have obtained similar deposits of colloidal selenium forming magnificent, highly 
 transparent, irregularly shaped (not globular) ruby beads. These do not exhibit any of 
 the above described phenomena in fact, some preparations made over eighteen months 
 ago have been preserved, and show, so far, no tendency at all to crystallise. 
 
THE CHEMISTRY OF INDIA RUBBER. J 1 3 
 
 be such an enormously more active vulcanising agent than free sulphur 
 itself. If polyprene and sulphur monochloride united in the vulcanising 
 process in the manner in which the components of a double salt unite, this 
 would at once render the process intelligible, but this is not the case. The 
 chemical behaviour of the polyprene chlorosulphide clearly shows that the 
 chlorine and sulphur of the monochloride separate and pass to different 
 parts of the polyprene molecule. Now the formation of sulphur mono- 
 chloride is expressed by the equation : 
 
 (1) S 2 , C1 2 = S 2 C1 2 + 14257 cal., 
 
 and therefore its combination with polyprene involves the inverse reaction : 
 
 (2) S 2 C1 2 = S 2 , C1 2 - 14257 cal., 
 
 or, in other words, it requires the expenditure of a considerable amount of 
 energy. 
 
 It will be remembered that in discussing above the chemical indifference 
 of polyprene to free sulphur at ordinary temperatures, I pointed to the 
 fact that the dissociation of S 8 into S 2 molecules would require a consider- 
 able expenditure of energy, and that, moreover, the heat of formation of 
 polyprene sulphide was probably very small an assumption fully sustained 
 by the result of the determination of that constant. It might seem, there- 
 fore, as if the circumstances were exactly the same in the case of vulcanising 
 with sulphur monochloride. This is of course true, as far as the necessity, 
 just pointed out, of splitting up of the S 2 C1 2 molecule is concerned ; nor 
 can the 'heat-toning,' due to the combination of the sulphur of the latter, 
 amount to any more than in the former example. 
 
 The explanation of the energetic reaction between polyprene and sulphur 
 monochloride is obviously to be found in the circumstance that besides the 
 vulcanising action the combination of S 2 with polyprene there is a 
 secondary reaction simultaneously taking place the combination of C1 2 
 with polyprene (addition) and this whole process must be represented by 
 the equation : 
 
 (3) (C 10 H 16 ) n + S 2 C1 2 = (C 10 H 16 ) n S 2 01 2 - 14257 cal. + (14257 + x) cal., 
 
 or, in other words, the energy absorption due to the decomposition of S 2 C1 2 
 into S 2 and C1 2 must be very greatly exceeded by the 'heat-toning' of 
 that phase of the reaction expressible as 
 
 (4) (C 10 H 16 ) n + C1 2 = (C 10 H 16 ). )t Cl 2 , 
 
 i.e., by the formation of a chlorinated addition product. 1 
 
 Now, for the final proof of the above we would require to know the 
 formation heat of the body (C 10 H l6 ) n Cl 2 . This constant I have not been 
 
 1 Of course I do not mean by this that a polyprene chloride and a sulphide form 
 side by side as separate bodies. The above only refers to the two phases in which the 
 chlorosulphide is formed. 
 
 8 
 
114 INDIA RUBBER AND ITS ANALYSIS. 
 
 able to ascertain, for the simple reason that the body in question has, so 
 far, not been obtained. Chlorine acts upon even very dilute solutions of 
 polyprene with such violence as to lead immediately to the body C 10 H 14 C1 8 
 (see page 31). The energy liberated in this reaction is so considerable, that 
 in working with a 2 per cent, solution of polyprene in chloroform the latter 
 soon begins to boil. This fact in itself is almost sufficient to prove the 
 above conclusion namely, that the energy with which sulphur monochloride 
 combines with polyprene is due to the very considerable heat-toning ac- 
 companying the combination of chlorine with polyprene. The sulphur 
 thereby liberated appears in what we have recognised as the most active 
 form of S 2 , the simultaneous combination of which with the polyprene 
 produces a further addition to the total heat evolved. 
 
 Quite in accordance with the foregoing is the fact that the action of 
 sulphur monobromide upon india rubber is very much slower than that 
 of the monochloride. Sulphur iodide is practically without action at 
 ordinary temperatures. The reason is evidently that the formation heat 
 of polyprene bromide is a fraction only of that of the chloride, and the 
 heat-toning due to the combination of iodine with polyprene is almost 
 imperceptible. 
 
 The theory of the action of sulphur monochloride upon polyprene, 
 given above, furnishes, moreover, an explanation of why such bodies as 
 thionyl chloride SOC1 2 , or sulphuryl chloride SOgCl^ 1 which, from the 
 similarity of their structure or constitution with that of sulphur mono- 
 chloride, would be expected to react in a similar manner with polyprene, 
 forming a new type of ' vulcanised ' india rubber, have, as a matter of fact, 
 not the slightest effect upon it. The explanation is as follows : Equation 
 (3) above contains the term (14257 + #) 2 , which represents the heat-toning 
 due to the combination of C1 2 with polyprene. I have given it that 
 particular form merely to denote that this amount is x cals. larger than 
 the energy absorption due to the splitting up of S 2 C1 2 into S 2 and C1 2 , 
 and very little consideration will show that by the introduction into the 
 equation (3) of either thionyl chloride or sulphuryl chloride, this term 
 (14257 + &) would remain quite unaffected. The formation heat of 
 thionyl chloride is + 40,800 cals., that of sulphuryl chloride 82,540 cals. 
 Inserting these values in equation (3), we would obtain respectively for 
 thionyl chloride : 
 
 (5) (C 10 H 16 ) n + SOC1 2 = (C 10 H 16 ) n SOCl 2 - 40800 cal. + (14257 + ) cal., 
 and for sulphuryl chloride : 
 
 (6) (C 10 H 16 ) M + S0 2 C1 2 = (C 10 H 16 ) W SOC1 2 - 82540 cal. + (14257 + x) cal. 
 
 These two reactions to maintain the exothermic character of equation 
 (3) would therefore require an enormous heat-toning for the combination 
 C 10 H 16 C1 j and although this is considerable, approximate thermometrical 
 
 1 Not to mention the large number of organic and inorganic acid chlorides. 
 
THE CHEMISTRY OF INDIA RUBBER. 115 
 
 observations of the reaction according to equation (3) leave no doubt that 
 it is much too small to render equations (5) and (6) exothermic. Hence 
 the reactions assumed in them do not take place at all. 
 
 Inversely, there is hardly any doubt that if we could find compounds of 
 the above named type of a formation heat approximately equal to, or lower 
 than, that of sulphur monochloride, we could obtain by means of them a 
 reaction analogous to that expressed in equation (3). Such a compound 
 would represent a new vulcanising agent, the discovery of which would not 
 only be of great theoretical interest, but most likely also of considerable 
 technical value. 
 
CHAPTEE II. 
 
 EXAMINATION AND VALUATION OF INDIA RUBBER AND 
 GUTTA PERCHA. 
 
 WHETHER we examine India rubber or gutta percha, the first consideration 
 is obviously how much of the pure hydrocarbon (C 10 H 16 ) M a given sample 
 contains. Upon this point depends very largely the value of the sample, 
 and also, though in a less degree, its suitability for a given purpose. 
 Since, however, the pure hydrocarbons contained in the different brands 
 of india rubber are neither physically nor chemically identical, it is 
 evident that they cannot be technically equivalent. This fact is, as the 
 result of technical experience, clearly expressed in the scale of prices of 
 the different brands of india rubber, which is not so much an indication of 
 the relative purity of these brands, as of their manufacturing value based 
 upon their properties and applicability. 
 
 The systematic examination and valuation of india rubber is at present 
 very little practised otherwise than by estimation of the 'loss on washing/ 
 and it certainly must be admitted that under the conditions under which, 
 at the present time, india rubber is bought and sold, the result of such an 
 examination could not influence the price paid, which is determined chiefly 
 by the state of the market, and accepted or refused by the buyer upon the 
 strength of a cursory empirical examination of samples which are often 
 submitted 'bulk not guaranteed.' Under these circumstances the prices 
 paid for different lots of the same brand of india rubber, supposed to be 
 of identical quality, often show the most absurd variations when calculated 
 upon the yield of washed product, and after deducting from this the 
 estimated percentage of resinous matter it invariably contains. These 
 variations are, of course, greatest in medium and low class rubbers. 
 
 This state of affairs is highly anomalous. It could not be said to be 
 due to the conditions under which crude rubber is collected, nor could it 
 be considered to be a deliberate creation of the rubber broker. Indeed, 
 there is little doubt that it is simply an old custom perpetuated by the 
 conservatism, not to say indolence, of the rubber manufacturer, who, as a 
 class, has done next to nothing to bring to bear upon the development of 
 his industry that line of scientific enquiry chemistry which has done 
 
EXAMINATION, ETC., OF INDIA RUBBER AND GUTTA PERCHA. T 17 
 
 more for the industrial development of this century than any other branch 
 of human knowledge. Pig iron, caustic soda, wood pulp, and scores of 
 similar articles, costing, comparatively speaking, a few shillings per ton, 
 are bought and sold on the basis of strict analytical standards ; but india 
 rubber, costing from 150 to 500 per ton, changes hands without either 
 buyer or seller having more than a vague knowledge of its intrinsic 
 value. 
 
 But even more important is the systematic analytical control of the 
 crude rubber before it enters into the manufacturing process. This 
 applies more particularly to the amount of what, in the absence of more 
 definite knowledge, is described as 'oily and resinous matter' contained 
 in the washed rubber. The substances falling under this head exert a 
 very marked influence upon the behaviour of the rubber in working, and 
 also, as might be expected, upon the properties and quality of the manu- 
 factured article. As already stated, the amount of these impurities, which 
 is slight in fine Para, varies often rather considerably in medium and low 
 grade rubbers ; and uniformity in the manufactured article, the lack of 
 which is so frequent a cause of trouble, cannot be expected if the most 
 elementary precautions to attain it are neglected. 
 
 The most important points in the examination and valuation of crude 
 rubber are : 
 
 (a) Loss on washing. 
 
 (b) Oily and resinous matters. 
 
 (c) Percentage of oxygen combined with rubber. 
 
 (d) Ash. 
 
 The loss on washing can only be estimated in the factory by determin- 
 ing the yield of dry washed rubber obtained from an ascertained bulk of 
 crude rubber. This loss consists of water, salts, wood, fibres, and mineral 
 impurities. Of the nature of the oily and resinous substances nothing 
 definite is known. The oily substances always form a very small part only 
 of the total extract. The resinous substances generally form semi- 
 transparent, yellowish-brown or brown substances, which in some cases 
 are semi-resilient and slightly sticky, in other cases hard and brittle, and 
 in a few cases represent white powders in appearance greatly resembling 
 the albane extracted from gutta percha. This kind of resin, which is very 
 difficultly soluble in all solvents hot and cold, occurs chiefly in African 
 rubbers, most plentifully in Sierra Leone. Its composition is expressed by 
 the formula C 10 H 16 2 . 
 
 The estimation of these oily and resinous constituents is best carried out 
 by drying a piece of the washed rubber either in vacuo or in a current of 
 carbonic acid, and extracting from 5 to 10 grms. of the sample in a Soxhlet 
 extractor by means of acetone. 1 The extraction should be continued for 
 
 1 In some cases it may be found preferable to perform the extraction by means of 
 glacial acetic acid, owing to the presence in a few cases, notably Sierra Leone, of resins 
 which in acetone are scarcely soluble at all. 
 
Il8 INDIA RUBBER AND ITS ANALYSIS. 
 
 from six to ten hours. After distilling off the acetone, the flask with the 
 extract is for about twenty minutes placed in a drying oven at 100 C. 
 An air current is then blown through the flask while still hot to expel the 
 last traces of acetone and water vapour, and finally the flask and its 
 contents are weighed. The amount of extract thus found should be cal- 
 culated upon the washed, and not upon the crude, rubber. 
 
 It is at the present time often assumed that the percentage of resinous 
 matter in india rubber may be taken as an indication of the care bestowed 
 upon its collection, coagulation, and preparation for the market in general. 
 I have, however, satisfied myself that this view is not correct. A number 
 of samples of different kinds of india rubber milk, which I had the oppor- 
 tunity to examine, showed that these resinous matters already exist in the 
 milk as it flows from the trees, and much in the same proportion to the 
 rubber present which we find in the corresponding commercial rubbers. 
 Of course, the very fact of the existence of Spiller's resin, which body is 
 certainly an oxidation product of the rubber itself, would seem to suggest 
 and support the above mentioned view; but I am of opinion that the pro- 
 portion of resinous matters in crude rubber due to this cause is in all but 
 very exceptional cases extremely small. My reason for this statement lies 
 in the fact that there must be quite a number of intermediary oxidation 
 stages between india rubber and Spiller's resin, so that commercial india 
 rubber containing throughout its mass a certain proportion of Spiller's 
 resin would almost certainly exhibit such a high degree of general oxidation 
 as to possess the properties of a more or less resilient resin rather than 
 those of india rubber. It might perhaps appear that the above views 
 scarcely offer an explanation of the variation of the amount of resin in the 
 same description of india rubber a variation which is often not inconsider- 
 able, especially in medium and low grades. I am inclined to ascribe this 
 variation, as far as it is bond fide, either to the condition or the age of the 
 tree from which the latex is obtained, or to the very common indiscriminate 
 mixing of the milks or products of good and of indifferent quality; but there 
 is no doubt that not infrequently it may be due to deliberate adulteration 
 practised by the collectors. The sometimes very striking qualitative varia- 
 tion of the resinous extract points rather strongly, I think, to the latter 
 cause. It is, however, highly desirable that this question should be studied 
 in a systematic manner in the various rubber districts. 
 
 The rubber, freed from oily and resinous matters by extraction with 
 acetone, is by no means the pure hydrocarbon, but contains invariably a 
 certain amount of oxygen derivatives of the general formula (C 10 H 16 ) n O m . 
 From the very composition of these products I am inclined to consider 
 them as true oxidation products (by addition) of india rubber, and I further 
 hold that the amount of such oxygen found in india rubber is a good indi- 
 cation of its quality from the point of view of its collection and preparation 
 for the market, also of its condition after shipping and storing. 
 
 The estimation of this oxygen is, of course, carried out by combustion, 
 
EXAMINATION, ETC., OF INDIA RUBBER AND GUTTA PERCHA. I 19 
 
 the rubber obtained after extraction with acetone and drying in an atmos- 
 phere of hydrogen or carbonic acid being used for this purpose. At the 
 end of the combustion, the weight of the ash remaining in the porcelain or 
 platinum boat must be ascertained, and deducted from the weight of the 
 substance used for combustion, before calculating the percentages of carbon 
 and hydrogen in the sample. The amount of ash thus found might be 
 supposed to represent the total amount of ash present in the washed rubber, 
 but this is only true as an approximation. It may be taken as practically 
 correct for all those brands of india rubber containing not more than about 
 4 per cent, of resin. In rubbers in which this quantity is exceeded, a certain 
 proportion of mineral matters invariably passes into the extract this the 
 more, the higher the proportion of resin present. For practical purpose* 
 the amount of ash found in the boat after combustion is sufficiently 
 near the truth ; if greater accuracy is desired, it should be estimated in 
 the washed rubber, before extraction, by incinerating in a platinum 
 capsule. 
 
 The following tables show the mean values for the loss on washing, oily 
 and resinous matters, oxygen, and ash of a number of different brands of 
 india rubber. 
 
 Trade Name. 
 
 Loss on 
 Washing. 
 
 Oils and 
 Resins. 
 
 Oxygen. 
 
 Ash. 
 
 Para, hard cure, . 
 
 15 
 
 21 
 
 12 
 
 05 
 
 soft . 
 
 17 
 
 2'5 
 
 1-4 
 
 0-3 
 
 Negro Head, 
 
 32 
 
 71 
 
 
 1-4 
 
 Matto-Grosso. 
 
 26 
 
 2'5 
 
 
 ... 
 
 Peruvian Ball, . 
 
 22 
 
 3-6 
 
 
 1-4 
 
 Cam eta, 
 
 26 
 
 2-2 
 
 
 
 Ceara (Mani9oba), 
 
 32 
 
 2-0 
 
 5-4 
 
 274 
 
 Mangabeira, 
 
 34 
 
 6'3 
 
 1-5 
 
 
 Nicaragua Scrap, 
 
 31 
 
 5'9 
 
 6-9 
 
 5-3 
 
 Gambia, 
 
 28 
 
 7'4 
 
 
 2'4 
 
 Gaboon, 
 
 28 
 
 ... 
 
 
 
 Manoh Twist, . 
 
 25 
 
 7-4 
 
 4'4 
 
 07 
 
 Lagos, 
 
 22 
 
 4-8 
 
 2-3 
 
 0-5 
 
 Congo Ball, 
 
 28 
 
 8-0 
 
 4-0 
 
 0-8 
 
 Sierra Leone, red, 
 
 26 
 
 ... 
 
 2-3 
 
 1-5 
 
 , , white, . 
 
 23 
 
 
 21 
 
 21 
 
 Lopori, 
 
 12 
 
 2 : 8 
 
 3'9 
 
 
 Kassai, red, 
 
 8 
 
 3-8 
 
 5'5 
 
 
 ,, black, 
 
 10 
 
 4-0 
 
 5'3 
 
 2 : 3 
 
 Angola, black, . 
 
 26 
 
 
 
 
 Mozambique, 
 
 16 
 
 51 
 
 7 : 2 
 
 ... 
 
 Madagascar, pinky, 
 ,, black, . 
 
 20 
 22 
 
 8-4 
 8'8 
 
 4'2 
 
 2-0 
 
 Assam Slabs, 
 
 24 
 
 137 
 
 1*4 
 
 4'3 
 
 Borneo, 
 
 48 
 
 2'2 
 
 2-53 
 
 2-2 
 
 Rangoon, . 
 
 17 
 
 11-0 
 
 7'4 
 
 4-9 
 
 
 
 
 L . 
 
 
 It should be clearly understood that the percentages figuring in the 
 different columns of these tables are not all calculated upon the crude 
 rubber. Indeed, this is only the case for the 'loss on washing.' The oils 
 
120 INDIA RUBBER AND ITS ANALYSIS. 
 
 and resins are calculated upon the washed rubber, the oxygen and the ash 
 upon the washed and extracted rubber. 
 
 The ' loss on washing ' of a given brand of india rubber can hardly be 
 looked upon as an indication of its intrinsic quality. Its importance lies, of 
 course, in the fact that its amount proportionately influences the price of 
 the rubber in its manufacturing condition. Nor is the amount of ash 
 found in a sample of india rubber a matter of much moment from the 
 technological point of view, although its estimation from time to time 
 serves as a useful check of the efficiency of the washing operation in the 
 factory. On the other hand, the percentage of oils and resins in a given 
 sample of india rubber has an important bearing upon its behaviour in 
 manufacturing, and upon the quality of the manufactured goods. Generally 
 speaking, a rubber will prove the * softer ' in working, the higher the per- 
 centage of oily and resinous matter it contains. This is roughly true for 
 rubbers from all sources, and absolutely true for different lots of the 
 same brand. The effect of these oils and resins is therefore invariably to 
 facilitate the compounding and the mechanical working of the rubber in 
 general, and for this reason their presence in rubber would not appear to 
 be objectionable, but only becomes so in those rubbers in which the amount 
 of these resins is liable to vary between wide limits. The objection in the 
 latter case rests upon the fact that the resins exert a very distinct retarding 
 effect upon the vulcanisation, and tend to produce 'shortness' in the 
 manufactured article. The technological effect of the oxygen in the india 
 rubber seems to have been altogether overlooked up to now. Its retarding 
 effect upon the vulcanisation process is quite remarkable, and likewise the 
 considerable reduction of the elasticity of the manufactured article. 
 
 A careful consideration of the above points enables us to judge in a 
 very satisfactory manner the relative commercial value of a given sample 
 of india rubber in reference to a standard sample of the same brand. To 
 a certain point these determinations will also enable us to compare two 
 samples of different brands of india rubber, but it should be well under- 
 stood that they are entirely inadequate for the estimation of the intrinsic 
 value of the pure rubber contained in one brand, as compared with the 
 value of the pure rubber contained in another different brand of india 
 rubber." We v have seen tha't all the different brands of india r libber 
 contain as chief part the same hydrocarbon (see page 11) C 10 H 16 , but it 
 is, on the face of it, most improbable that this similarity should extend 
 beyond the empirical formula. We possess no chemical evidence at the 
 present time of any structural chemical difference between the india rubber 
 hydrocarbons of different brands of commercial india rubber, but it is an 
 absolute certainty, well known to every manufacturer, that the different 
 brands of india rubber vary considerably in their behaviour under manu- 
 facturing conditions. Thus we distinguish between quick and slow 
 vulcanising rubbers ; we know from experience that certain brands of rubber 
 form satisfactory products if vulcanised with chloride of sulphur, others do 
 
EXAMINATION, ETC., OF INDIA RUBBER AND GUTTA PERCHA. 121 
 
 not admit of this kind of vulcanisation. There are ranges of india rubber 
 which experience has proved to be particularly suitable for rubber shoes, 
 others for elastic thread, others for waterproofing, and again, others for the 
 manufacture of ebonite. 
 
 Making full allowance for this fact that the india rubber manufacturer 
 is practically only interested in the physical property, and takes only the 
 slightest cognisance, if any, of the chemical side of his manufacture, it is, 
 nevertheless, remarkable that no reaction is known at present which would 
 allow us to differentiate any of the india rubber hydrocarbons from all the 
 others. 
 
 The following synoptical tables, comprising all important brands of 
 india rubber, their geographical and botanical origin, general appearance, 
 technical and analytical constants, will be found useful for reference. 
 
 [TABLES. 
 
122 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 'QJ O CT 1 
 
 &to 
 
 
 , 
 - 
 
 CO 
 
 1 
 
 I 
 
 o 
 
 CO 
 
 O 
 
 
 
 *|iff ll 
 
 r^ ^ &c" <B ^ g -e 
 
 S 3.S ^ ^a 1 
 
 o 
 
 O'Q 'S I -& 
 
 sts 
 
 >>PH ^ iLa ' QJ m 
 
 ii |iii f-iil 
 
 6 ^ 
 
 1 
 
 
 i 
 
 .2 8 S 
 .1, .1 
 
 
 tfS 
 
 
 t 
 
 *- 
 
EXAMINATION, ETC., OF INDIA RUBBER AND GUTTA PERCHA. 123 
 
 
 " fefl ' 
 
 t- >i- ' ' X 
 
 O '-f 
 
 
 o 
 
 
 
 
 ^ 11 s . 
 
 II 
 
 8-f|J 
 
 ** s 
 
 >-l o 
 o < 
 
 c > 
 
 1 
 
 O! ^3 
 
 .2 
 
 cc p 
 'o 5P 
 
 1l 
 
 <'^ 'b 
 
 4^ O2 
 
 Qj"-^^ ^ f^l Cl 
 
 S,^ 
 
 ft 
 
 +3 ?5 
 
 .^ 
 
 rH S 
 
 rt >, en ^ 
 
 fl c3 
 
 
 
 
 
 
 
 
 ill 
 
 l|lip 
 
 M ^rrJ 
 
 1 
 
 1^ 
 
 "~a o 
 
 ill 
 
 +jj 2o ^ w 
 
 2-3 
 
 ^ .2 12 - >3 ^ 
 
 p ^ ^3 
 
 
 ^ 
 
 o? 1 
 
 g 'd'aj 
 
 P5 02 
 
 ^ 
 
 <j 
 
 
 
 * 
 
 
 
 
 
 C? 
 
 ^ CO 
 
 CO 
 
 i i 
 
 ^ 
 
 , 
 
 CO 
 
 ^ 
 
 (N 
 
 <N 
 
 
 
 CO 
 
 CN 
 
 
 
 <N 
 
 
 
 
 At (N 
 
 O5 
 
 OS 
 
 CO 
 
 I 1 
 
 
 
 CN 
 
 s 
 
 O O 
 
 CO CO 
 
 ^ s 
 
 CO 
 
 I 1 
 
 o 
 
 2 
 
 CO 
 
 
 
 CO 
 1 
 o 
 
 CO 
 
 1 1 
 
 o 
 
 O 
 
 i i 
 
 o 
 
 I 
 
 ot smoked, but often 
 deep black outside, al- 
 most pure white inside. 
 
 tangle of shreds, rib- 
 bons, and ' tears. ' 
 Amber-yellow to pale 
 brown colour. 
 
 11 
 h 
 
 B& 
 
 M 
 
 3Ctangular slabs varying 
 in size. On the outside 
 reddish-brown, on the 
 cut light pink. 
 
 regularly shaped lumps. 
 
 ery similar to Para. 
 
 ft 
 
 CO "^ 
 
 1^ 1 
 
 appearance much like 
 Para. 
 
 most black surface, con - 
 tains much moisture 
 and solid impurities. 
 
 fc *4 
 
 3 
 
 P9 
 
 
 > 
 
 > 
 
 
 < 
 
 i 
 
 
 1" 
 
 
 8 
 
 4 
 
 
 - 
 
 1 
 
 a 
 
 S 
 
 * 
 
 1 
 
 .2 ^ 
 
 1 
 
 'v 
 
 i i .^ 
 
 | 
 
 1 
 
 
 1 
 
 S 4 ! 
 
 a. 
 
 1 
 
 fe ' 
 
 1 
 
 1 
 
 
 O . ^ 
 
 r 
 
 11 
 
 i 
 
 1 
 
 g i 
 
 1 
 
 L 
 
 | 
 
 
 
 i 
 
 
 
 
 CO 
 
 g 
 
 ^ 
 
 
 
 .2 
 
 
 I 1 
 
 -M ^ .Si 
 
 E3 O PP 
 
 M 
 
 Province 
 Brazils. 
 
 ij 
 
 Province I 
 Brazils. 
 
 1 
 
 Columbia. 
 
 o 
 
 I 
 
 Ecuador. 
 Columbia. 
 
 
 
 
 
 if* 
 
 
 
 
 fL 
 
 
 g 
 
 S S3 
 
 
 c 
 
 
 _ 
 
 1 it 
 
 1 
 
 11 
 1$ 
 
 ii 
 
 11 
 
 I S J^ 
 
 sis 
 
 Columbia 
 Carthage 
 
 1 
 
 f 
 
 O 
 
 
 
 
 o 
 
 1 _ 1 
 
 O3 
 
 CO 
 
 ^ 
 
 o t>- 
 
 CO 
 
 05 
 
 i i 
 
 i * 
 
 
 1 1 
 
 *""* 
 
124 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 .2 CD as a 
 
 a HIS 
 
 -< ^ S, 
 
 a -*> 
 
 g fl 
 
 a 
 
 t 
 
 O 
 
 Ill Si 
 
 di O -I- 1 "~! ^ 
 
 l 
 
 02 
 
 id 
 
 a 
 
 
 
 
 
 CO 
 
 
 
 g $ 
 
 .S -3 
 
 3 
 Q 
 
 - -o 
 
 co 
 SOW 
 
 
 if 
 
 
EXAMINATION, ETC., OF INDIA RUBBER AND GUTTA PERCHA. 125 
 
 ,Q 5 
 
 2 1 
 
 u 
 
 ry fair q 
 
 11 
 
 S 13 
 
 >rO 
 
 42 g d 
 
 -s a o 
 
 ^ ao* 
 p 2 
 
 |ll 
 
 cr" 5 
 
 ^ ' 
 
 r2 00 
 
 fr ^' 
 
 o> o> 
 
 oo o 
 1^-1 O 
 
 
 
 Balls containing much 
 water and bark. Pink 
 and white colour. 
 
 ps wound 
 
 !4 
 
 p^r=! 
 
 a or 
 
 
 rrow 
 balls. 
 
 it I 
 
 arge balls, dark outside, 
 grey on the cut. 
 
 Jfr 
 
 Is 
 
 l 
 
 1 
 i 1 
 
 all m 
 balls. 
 
 HI 
 
 !B 
 
 1 
 
 
 s i ^ 
 
 ist. 
 h T 
 
 . 
 
 C 
 
 g'Sg 11 Igi 
 
 SH I 3 lid 
 
 n Strip. 
 
126 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 III. India Rubber Brands of Africa continued. 
 
 Appreciation. 
 
 2 ~ fcOtJd w kC fi 
 
 HH CO ^ C3 C 
 
 "S ^. o -g^ .2 
 
 1^ g l ^ a 1 ^11 1 
 -g.g ,s s. -1 ^ rt ^. 
 
 - & ft-^S > ^ * 
 
 !-s r l 31 1 -stl P 
 
 g B w^cr 1 ^ .*p 2 JsC-p g cr 1 
 i-3 <J > W H > 
 
 g ID 13 
 2 fc*f Jl 
 
 o o AH o o o 
 
 Hi! 
 
 iO ^ CO CO iO ^O 
 
 if! 
 
 1-H )g 
 rH T-H 
 
 oi 3 o!> 5 ^ 
 
 Characteristics. 
 
 1 If III Jl?l 1^1 
 
 o r< a O4J S2'^d ^,~,fe 
 
 ss 1 1 1^ sil ^- 
 
 T 
 
 O 
 
 m 
 
 i 
 
 ".!.'" "" ' . " 
 
 Geographical 
 Origin. 
 
 02 
 
 i D 
 
 1 " = s a 
 
 fe 
 
 05 
 
 'i ^ a" 2 
 
 iitf 
 
 t= i^ H <1MWW^gf^^ t> 
 
 O i-H C^l CO "^ iO 
 
 rH i-H j-H T-H i-H y-H 
 
EXAMINATION, ETC., OF INDIA RUBBER AND GUTTA PERCHA. 127 
 
 fe b42 
 
 _OTJ & 
 
 
 g 
 
 5 
 
 >> 
 
 d 
 
 AH 
 
 ^D 0} O * 
 
 2 f3 
 
 .p 
 
 ,cj 
 
 ^ 
 
 
 
 
 'g ^-g 
 
 o 4 i 
 
 1 
 
 * 
 
 S 
 
 1 
 
 -3 
 
 -1^ 
 
 tJ3_co "|2 
 
 'S'S M 
 
 ll 
 
 
 
 
 cr" 
 
 >> 
 
 111 
 
 2- '^ 
 
 g * o 
 
 1 
 
 i 
 
 -2 
 ,2-d 
 
 1 
 
 1 
 
 i1 
 
 loll 
 
 *ll 
 g5fg 
 
 1 
 
 !* 
 
 |1 
 
 1 
 
 ,5 
 
 ^ 
 
 -11 
 
 ^^ 
 
 * 
 
 w 
 
 < 
 
 B 
 
 5 
 
 
 
 
 s 
 
 (N 
 
 . 
 
 10 
 
 o 
 
 CO 
 
 
 
 (M 
 
 t^ 
 
 i 1 
 
 
 
 
 
 
 1-1 
 
 C, 
 
 
 
 
 N 
 
 (D 
 
 
 
 
 
 T* 
 
 
 
 
 ^ 
 
 00 
 
 
 OS 
 
 
 
 
 
 
 
 1-4 
 
 CO 
 
 
 
 
 S 
 
 o 
 
 
 
 
 
 3 
 
 1 
 
 00 
 I 1 
 
 1 
 
 <N 
 1 1 
 
 S 
 
 
 
 O 
 
 CO 
 
 >mall cubes, brown or 
 black outside, generally 
 brown or grey inside. 
 
 Aggregates of small balls, 
 grey outside, light grey 
 inside. 
 
 imall balls, brown out- 
 side, pinky-brown in- 
 side, almost transparent 
 in centre. 
 
 >mall balls or sausages, 
 reddish-brown outside 
 and inside. Contains 
 much bark. 
 
 Inger-shaped pieces. 
 Colours yellow, pink, 
 brown, or black. 
 
 Jlack or light brown 
 small balls, reddish - 
 brown, sometimes white 
 inside. 
 
 labs and balls, brown 
 outside, flesh-coloured 
 inside. 
 
 jarge round cakes, black 
 outside, white, pink, or 
 buff-coloured inside. 
 
 02 
 
 Lower Congo 
 (Thimbles). 
 Luvituku. 
 
 Loanda 
 Thimb 
 
 Mozambiq 
 Marbles. 
 
 
128 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 'S S-S 
 
 Ml 
 
 d 
 
 1 
 
 -fo 
 
 
 ^-S 
 
 M 
 
 G 1 
 
 ity 
 tte 
 
 Q 
 
 . 
 
 O 
 
 " 
 
 d fl'c es 
 
 i-i C .9 r: 
 
 I ft .a 
 
 " 
 
 Nit Ij 
 
 *H M P PH Sj+3 
 
 3 3 
 
 O . 
 
 'o S 
 
 o 
 
 
 fill 
 
 . 
 
 r 
 
 li 
 It 
 
 ii 
 
 
EXAMINATION, ETC., OF INDIA RUBBER AND GUTTA PERCHA. 129 
 
 _ 
 
 . 
 
 >> 
 
 2 
 
 
 1 
 
 1 
 
 s 
 
 1 
 
 1 
 
 45 
 
 6C 
 
 ,0 ^ 
 
 O ,J 
 
 3 
 
 o . 
 
 s 
 
 
 
 CC 
 
 ll 
 
 P ""^ 
 
 | 
 
 F 
 
 o 
 
 11 
 
 ff 
 
 I s 
 
 ^ 
 
 O 
 
 P 
 
 >" 
 
 s> 
 
 
 
 
 
 00 
 
 
 
 <N 
 
 
 
 
 
 o 
 
 N 
 
 CO 
 
 
 O5 
 | 
 
 
 
 c^ 
 
 05 
 
 rh 
 
 
 CO 
 
 
 
 
 00 
 
 
 o 
 
 
 
 
 o 
 
 
 
 
 
 
 (M 
 
 .44 s 
 
 li 
 
 1 
 
 'o 
 O 
 
 f| 
 
 
 
 2 
 
 S 
 
 I 
 
 cf 
 
 |f 
 
 ll| 
 
 .- 
 
 a 
 
 &j 
 
 || 
 
 Sj'i* 
 
 II 
 
 3 
 
 
 
 iS 
 
 02 
 
 o 
 
 B 
 i i 
 
 i , 
 
 g 
 
 
 
 i 
 
 * ^ 
 
 I 
 
 
 
 i 
 
 . 5 1 
 
 1 
 
 1 
 
 " 
 
 1 
 
 w 
 
 1 
 
 1 
 
 
 i* 
 
 
 
 1 
 
 8 
 
 c8 
 fl 
 
 8 
 
 fi - 
 
 1 
 
 I 
 
 Q 
 
 1 
 
 I 
 
 1 
 
 CO 
 0) 
 
 i 
 
 
 
 O 
 
 1 
 
 .1 
 
 6 
 
 o 
 
 c 
 
 r3 
 
 PQ w 
 
 a 
 
 01 
 
 1 
 
 o 
 
 1 
 
 1" 
 
 11 
 
 1 
 
 
 
 to 
 
 t>: 
 
 00 
 
 OS 
 
CHAPTEE III. 
 EXAMINATION OF INDIA RUBBER SUBSTITUTES. 
 
 (A.) Reclaimed or Recovered India Rubber. 
 
 OF late years the use of reclaimed or recovered india rubber, and, as a 
 natural consequence, the manufacture of this article, has assumed very 
 large proportions indeed. The raw material is always either factory scraps 
 and cuttings, or old and worn-out rubber articles of every description. 
 
 Generally speaking, the recovery consists in such a treatment as to 
 restore to the india rubber contained in those scraps and old articles its 
 original plasticity and solubility in the technically used solvents, together 
 with its capability of vulcanisation. It is generally considered that the 
 last named requirement is the most difficult to attain, the current idea 
 evidently being that this desirable result has the more or less complete 
 devulcanisation of the waste rubber as its necessary condition. But it is 
 easily shown that the capability of recovered india rubber to vulcanise is 
 necessarily consequent upon the restoration of plasticity in the vulcanised 
 india rubber. As we have seen in the discussion of the vulcanisation 
 problem, the amount of sulphur entering into actual combination with 
 india rubber is not limited to the small proportion of combined sulphur 
 found in ordinary india rubber articles, a proportion rarely exceeding 
 3 per cent. Vulcanised india rubber, as far as the process of vulcanisation 
 can be chemically interpreted at the present time, is an addition pro- 
 duct of the unsaturated polyprene molecule and sulphur. But this 
 addition product is still a highly unsaturated body so much so, that 
 out of thirty individual ethylene bonds in vulcanised india rubber, on 
 an average, and under ordinary circumstances, not more than two 
 can have become saturated by sulphur addition. 1 This holds true, 
 
 1 This calculation is based on the consideration that, according to Gladstone and 
 Hibbert, india rubber for each C 10 H ]6 complex contains three ethylene bonds, and 
 further, that the process of sulphur addition to the india rubber must take the form of 
 either 
 
 v v \/ \/ \/ \/ 
 
 p p r 1 p r* Q n 
 
 \j ^ \ V V ^> >J \J 
 
 || + S = i >S or || || + 2S = I 
 
 C C/ C C C-S C 
 
 /\ /\ ' /\ /\ /\ /\ 
 
 In the first case, for the formation of a vulcanisation product containing about 2 '5 per 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 131 
 
 whatever may eventually be found to be the size of the india rubber 
 molecule. 
 
 Now, it is well known that vulcanisation products, with any amount 
 of combined sulphur up to, and most probably exceeding, 20 per cent, 
 may be produced under suitable conditions ; and, from the composition of 
 the addition products polyprene forms with bromine or with sulphur mono- 
 chloride, we may infer that at least twenty out of the thirty ethylene bonds 
 occurring in any given complex of iudia rubber molecules can simultane- 
 ously enter into the process of forming an addition product. Consequently 
 there is no theoretical objection to a true ' revulcanisation ' of vulcanised 
 india rubber if we succeed in removing the physical effects of the original 
 vulcanisation in other words, if we restore to the vulcanised india rubber 
 the plasticity which characterises it in its primitive condition. In the 
 process of revulcanisation, some of those ethylene bonds which remained 
 intact in the first vulcanisation then come into play and take up the 
 sulphur present. This process of recovery and revulcanisation can, 
 moreover, be repeated as long as the recovered india rubber contains any 
 free ethylene bonds, although there is no doubt that the quality of the 
 india rubber deteriorates rapidly with each successive revulcanisation. 
 This is only what is to be expected, considering that with each successive 
 vulcanisation we obtain a product, the composition and constitution of 
 which exhibits an increasing difference from the composition and con- 
 stitution of the primary vulcanisation product. For whereas the latter 
 is a very highly unsaturated body, with, at the outside, 3 per cent, of 
 sulphur, each successive revulcanisation increases the constitutional 
 saturation, and likewise the percentage of combined sulphur in the product. 
 This, in my opinion, sufficiently explains the fact of the already alluded 
 to deterioration of the product with each revulcanisation a fact which is 
 certainly no more surprising than the extraordinary change of properties 
 following upon the successive saturation of radicals in the benzene 
 ring. 
 
 In the above it is, of course, taken for granted that the sulphur of 
 vulcanisation, the combined sulphur, is not removed in the process of 
 recovery. I am well aware that this assumption is not shared by the 
 manufacturers, who expressly declare their recovered india rubber to be 
 * devulcanised ' or 'desulphurised.' This declaration is, no doubt, made on 
 the strength of the fact that recovered rubber can be revulcanised, the idea 
 being that to produce this result is only possible by removing the chemical 
 cause of the first vulcanisation. Analysis of any sample of recovered india 
 rubber will, however, at once show this conclusion to be erroneous, as in 
 no case is the combined sulphur found to have been eliminated, although 
 most processes, patented or otherwise, for the recovery of india rubber aim 
 at this result. It will, indeed, be generally found that the apparently most 
 
 cent, of S, 10C 10 H 16 containing thirty ethylene bonds would combine with IS ; in the 
 second case. 20C ]0 H 16 with 2S. 
 
132 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 promising of these processes are just those that fail most conspicuously to 
 yield a satisfactory product. The most satisfactory processes are invariably 
 those in which no account is taken, chemically, of the combined sulphur, 
 but where such means and agencies are employed as have been found to 
 result merely in the restoration of the original plasticity of the india 
 rubber. 
 
 Analysis of a number of representative commercial brands of recovered 
 india rubber, and some others that I prepared myself, gave the following 
 figures : 
 
 
 I. 
 
 II. 
 
 III. 
 
 IV. 
 
 V. 
 
 VI. 
 
 Moisture, 
 
 50 
 
 81 
 
 32 
 
 
 14 
 
 
 Oils and resins, 
 
 10-23 
 
 10-62 
 
 9-85 
 
 18-53 
 
 19-44 
 
 19-04 
 
 Fe 2 3 and A1 2 0,, 
 
 77 
 
 1-01 
 
 88 
 
 79 
 
 38 
 
 19 
 
 PbO, 
 
 14-20 
 
 
 15-30 
 
 
 
 
 PbS0 4 , 
 
 3-29 
 
 18 V 88 
 
 10-81 
 
 ... 
 
 
 ... 
 
 ZnO, 
 
 ... 
 
 
 
 
 23-89 
 
 21-03 
 
 CaC0 3 , 
 
 10-41 
 
 
 
 
 
 
 CaS0 4 , 
 
 21-93 
 
 26-20 
 
 2277 
 
 
 
 
 BaS0 4 , 
 
 ... 
 
 
 
 
 
 20-85 
 
 Carbon, 
 
 7-54 
 
 8-27 
 
 8 V 49 
 
 
 
 
 S free, 
 
 88 
 
 62 
 
 61 
 
 "20 
 
 "15 
 
 trace 
 
 S combined, 
 
 3-92 
 
 2-46 
 
 2-71 
 
 4-84 
 
 4-16 
 
 2-28 
 
 I. R. substitute, 
 
 
 
 
 
 ... 
 
 9-26 
 
 India rubber, . 
 
 26-32 
 
 31-12 
 
 28-25 
 
 75-63 
 
 5o'-83 
 
 27-34 
 
 Price per Ib. , . 
 
 5d. 
 
 lljd. 
 
 l/5Jd. 
 
 
 
 
 
 
 
 
 
 
 
 These figures, to begin with, show that in the recovered india rubber of 
 commerce the price bears no relation to the quantity of india rubber actually 
 present. It may be, and certainly is, influenced by the quality of the india 
 rubber present, which is a point respecting which we are unable, at any rate 
 at present, to give an account in our analyses. But it is perhaps worth 
 pointing out that in the above commercial samples (I. to III.) the price 
 rises in a ratio which cannot possibly represent the ratios of the values of 
 the original india rubbers contained in the scrap from which respectively 
 these three qualities of recovered india rubber were obtained. Likewise 
 these analyses show that commercial recovered india rubber is loaded with 
 a very considerable amount of mineral matter. Sample III. is considered 
 a very high class article of its kind, but it contains as much mineral matter 
 as sample I., which is sold as a low quality. As to the advantage to the 
 manufacturer of this commercial recovered india rubber, a simple calcula- 
 tion of the price of 1 Ib. of india rubber in those samples will be found 
 highly instructive. 
 
 From the chemical point of view, the respective figures for free sulphur, 
 combined sulphur, and india rubber first attract attention, especially if we 
 calculate in every case the percentage of the two first named items in rela- 
 tion to the india rubber. We then obtain : 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 
 
 133 
 
 
 I. 
 
 II. 
 
 III. 
 
 IV. 
 
 V. 
 
 VI. 
 
 Sfree, . 
 
 3-34 
 
 1-99 
 
 2-15 
 
 26 
 
 29 
 
 
 S combined, 
 
 14-89 
 
 7-90 
 
 9-59 
 
 6-38 
 
 8-17 
 
 8-33 
 
 India rubber, . 
 
 100-00 
 
 100-00 
 
 100-00 
 
 100-00 
 
 100-00 
 
 100-00 
 
 What strikes one very forcibly in these figures is the enormous propor- 
 tion of sulphur of vulcanisation which in every case far exceeds the propor- 
 tion ever occurring in any commercial vulcanised india rubber manufactured 
 from india rubber in its native state. The first three samples of commercial 
 recovered india rubber are particularly remarkable in this respect, as, to 
 to begin with, the total sulphur (free + combined) they contain is, in the 
 case of I., about double the usual amount, and, even in the case of II. and 
 III., either close to or distinctly above the highest limit, which is about 10 
 per cent. The samples IV. to VI. show an amount of total sulphur which 
 is well within the practically adopted limit, but, nevertheless, even in these 
 three samples the amount of combined sulphur is extraordinarily high, 
 showing percentages which in ordinary vulcanisation are never obtained, 
 and perhaps even not obtainable. 
 
 For the elucidation of this surprising fact, as far as the samples I. to 
 III. are concerned, we have no available data. As regards the samples I"V. 
 to VI., these were ' recovered ' from vulcanised india rubber (Para), which 
 before recovery showed, on analysis, the following ratios between sulphur 
 and india rubber : 
 
 
 IV. 
 
 V. 
 
 VI. 
 
 Sfree, . 
 
 4-88 
 
 6'49 
 
 6-42 
 
 S combined, 
 
 2-12 
 
 2-30 
 
 2-47 
 
 India rubber, . 
 
 100-00 
 
 100-00 
 
 100-00 
 
 To begin with, we see from these figures that in the recovery process the 
 percentage of total sulphur is decreased, partly, as will be shown later on, 
 through some of the sulphur volatilising as such, partly through the evolu- 
 tion of sulphuretted hydrogen. But still more important is the fact that 
 these figures show very clearly that the bulk of the free sulphur present 
 enters into combination with the india rubber, so that the recovery process 
 not only effects no devulcanisation, but actually results in a higher degree 
 of vulcanisation of the already vulcanised india rubber. This at once shows 
 conclusively that the efficiency of the recovery process depends entirely 
 upon the solvent or softening actions of the oils or hydrocarbons, which, as 
 far as my experience goes, are at present invariably employed in all india 
 rubber recovery processes. It is also obvious that the great inferiority 
 of even the best recovered india rubber to the native article is due to this 
 further vulcanisation during the process of recovery. 
 
134 INDIA RUBBER AND ITS ANALYSIS. 
 
 It would be exceedingly interesting to examine the effect of this 
 recovery upon an india rubber article largely or entirely composed of once 
 recovered india rubber. It is almost certain that such successive recoveries 
 will result in a progressive increase of the combined sulphur of vulcanisa- 
 tion, and a consequent rapid deterioration in value of the recovered 
 product ; and I suggest that this theory might be found to account for the 
 enormous percentages of combined sulphur in the above samples I. to III. 
 Whether such successive recoveries, independent of the quality of the 
 recovered product, can be effected a number of times, leading eventually to 
 a saturated polyprene polysulphide, or whether there is a limit to this 
 number, and by what that limit is determined, cannot at present, in the 
 absence of all experimental data, be decided. There can be, however, no 
 doubt that the recovery processes at present conducted must, after a few 
 repetitions, yield products no longer suitable for manufacturing purposes. 
 
 Information as to the actual recovery processes employed is very scarce, 
 the methods employed being kept jealously secret. From this it should 
 not be inferred that these processes are of any particular merit indeed, the 
 analysis of any sample of commercial recovered india rubber amply proves 
 them to be of a highly primitive and empirical description. The fact 
 that every sample of recovered india rubber contains from 8 per cent, 
 upwards of mineral oils, sometimes even fatty oils, or a mixture of the two, 
 shows that the recovery process, generally speaking, consists in treating 
 the vulcanised india rubber with oil under certain conditions. The oils 
 I have so far found in samples of recovered india rubber are either 
 ordinary mineral oils, such as vaseline oil, petroleum jelly, or rosin oil ; or 
 vegetable and fatty oils, such as castor oil or colza oil. Experiment shows 
 that the action of any of these oils upon vulcanised india rubber at 
 ordinary temperatures is extremely slow, but quickens as the temperature 
 rises. The best results appear to be obtained at temperatures lying below 
 the upper limit of the range of temperatures at which vulcanisation is 
 carried out. The heat may be applied to the mixture of vulcanised india 
 rubber and oil, either externally, or by treating the mixture in closed 
 vessels with superheated steam at the requisite temperature, or the two 
 modes of heating may be combined. The latter is most probably the way 
 by which the best results are obtained. 
 
 This recovery process can scarcely be expected to effect devulcanisation 
 of the vulcanised india rubber, and, indeed, as I have shown above, it 
 actually does just the reverse. Of processes devised to accomplish de- 
 vulcanisation, as distinguished from mere recovery as above outlined, 
 only two or three have been proposed. The first is due to A. F. B. 
 GORNESS (Germ. Pat., 62,738/1892), and consists in the treatment of the 
 vulcanised substance first with iron filings and dilute sulphuric acid, 
 followed by treatment with some alkaline solution. The less said about 
 this process the better. Another process proposed by MICHELIN & Co. 
 (Germ. Pat., 75,063/1894) looks more rational. It is based upon the well- 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 135 
 
 known affinity of finely subdivided metals, such as copper, tin, lead, 
 mercury, and others, for sulphur. It is proposed to mix these metals 
 intimately with the india rubber to be devulcanised, then to treat this 
 mixture with an indifferent india rubber solvent, such as benzene, toluene, 
 chloroform, or carbon bisulphide. It is stated that under these conditions 
 devulcanisation takes place, the regenerated india rubber passing into 
 solution. Very similar to this is GERBBR'S Patent (Eng. Pat., 19,284/1893), 
 according to which the vulcanised india rubber, containing from 5 to 8 
 per cent, of sulphur, is treated, under a pressure of three or four at- 
 mospheres, and with constant agitation, with three and a half times its 
 weight of toluene and one fourth its weight of tin, or any other sulphide 
 forming metal. The heating is continued for eight hours, when the resulting 
 solution is poured off and allowed to settle. 
 
 It is hardly probable that any of these processes will answer its 
 purpose: certainly not, unless the combined sulphur in vulcanised india 
 rubber proves to be far more active than we know it to be in other organic 
 sulphides an assumption not warranted by facts. 
 
 There is, however, yet another kind of recovered india rubber in the 
 market, mention of which should not be omitted. This is the india rubber 
 recovered from the cuttings of waterproof cloth vulcanised by the cold 
 process (PARKE'S). In this case the aim is to free the cuttings from all 
 textile materials. The remaining india rubber scraps represent only semi- 
 vulcanised india rubber, owing to the cold vulcanising process leaving the 
 deeper layers or coats of india rubber on the cloth almost untouched, and 
 consequently these scraps can again be worked up into coherent masses 
 between hot rollers. This kind of recovered india rubber, of course, still 
 contains the chlorine derived from the sulphur monochloride, and con- 
 sequently it cannot be used for ' dry heat ' mixings, but should again be 
 used for the manufacture of ' cold cured ' articles. 
 
 A great number of purely empirical processes for the recovery of india 
 rubber that is, processes not based upon a well-understood chemical 
 reaction have been suggested and patented. None of them solves the 
 question of devulcanising the india rubber. At most, an attempt has 
 been made to remove the free sulphur by boiling the ground scraps with 
 aqueous solutions of alkalies or alkaline carbonates. It is scarcely 
 doubtful that, with a view to ultimate devulcanisation, the removal of the 
 free sulphur is of great importance, and this is also easily and cheaply 
 accomplishable ; but the attempt to effect final devulcanisation by treat- 
 ment of this partly desulphurised india rubber with ordinary rubber 
 solvents is certainly inadequate. What is evidently required is a treat- 
 ment with reducing agents, and from that point of view some of the above 
 quoted patents, employing metals in the molecular state, appear rational. 
 Experiments I made in this direction gave, however, most discouraging 
 results. Nevertheless, further research on these lines should be made. It 
 must be borne in mind that the conditions of these reduction experiments, 
 
136 INDIA RUBBER AND ITS ANALYSIS. 
 
 in which, as long as metallic powders are employed as reducing agents, 
 both the reducing agent and the body to be reduced are present and 
 remain throughout the process in the solid state, are not very conducive 
 to success. The vulcanised india rubber being unavoidably solid, the 
 reducing agent should certainly be liquid, or, if a solid, it should be used 
 in solution. If this is so, the use of metallic reducing agents will have 
 to be abandoned, and liquid or soluble reducers should be tried instead. 
 
 The analysis of recovered india rubber is, of course, carried out on 
 exactly the same lines as that of ordinary india rubber articles. The 
 various points to be determined are : 
 
 1. Specific gravity. 
 
 2. Acetone extract. 
 
 3. Nitrobenzene (iiitro-naphthalene) extracts. 
 
 4. India rubber substitutes (fatty)-/ ^ in substitute - 
 
 \ Cl in ,, 
 
 5. India rubber combined S. 
 
 6. Carbonaceous matter. 
 
 7. Mineral constituents. 
 
 The working details of these analyses will be found in a later chapter. 
 
 A few words are, however, necessary concerning the interpretation of 
 the analytical results, for the simple reason that data which would be con- 
 sidered satisfactory in an ordinary india rubber article would at once 
 condemn a sample of recovered rubber, and vice versa. 
 
 1. Specific Gravity Determination. This is carried out in the usual 
 manner by means of the pyknometer. The test is of comparatively little 
 importance. 
 
 2. Acetone Extract. The ratio between the percentage of acetone 
 extract and that of india rubber present in the sample is of considerable 
 moment in judging the intrinsic quality of a sample of recovered india 
 rubber. It may safely be assumed that the higher this ratio, the poorer 
 is the quality of the recovered india rubber. The amount of acetone 
 extract is indeed a reliable index of the efficiency of the process, which- 
 ever it be, by which the recovery has been effected, especially if we calcu- 
 late the percentage of acetone extract upon the actual amount of india 
 rubber present. In doing this, it will be found that the vast majority of 
 commercial brands of recovered india rubber rarely yield less than 20 per 
 cent, of extract. The extracted residue will then frequently be found to 
 consist of a dry, crumbly mass, possessing very little elasticity and still 
 less plasticity. The value of such a recovered india rubber is scarcely 
 higher than that of a mixture of ground rubber scraps (vulcanised), with a 
 corresponding proportion of oil. In a good recovered india rubber the 
 percentage of oils or hydrocarbons should be below 10 per cent, (the lower 
 the better), and the residue after the extraction should itself possess an 
 appreciable amount of plasticity. When this is not the case, the manu- 
 facturing value of the recovered sample is really only that of a mere 
 'filling' i.e., bulk-giving substitute. 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 137 
 
 In many cases it may also be advisable to examine the nature of the 
 acetone extract specifically, by estimating the amount of saponifiable matter 
 it contains by KOETTSTORFER'S process. The saponification number of the 
 extract should be very low, as the presence of an appreciable quantity of 
 fatty oil in recovered india rubber is apt to cause serious trouble. It 
 becomes, therefore, rarely necessary to attempt the identification of any 
 fatty oil present, but it is generally desirable to convert the observed 
 saponification number into a figure expressing the percentage of fatty oil 
 in the extract. The kind of fatty oil present being generally unknown, the 
 simplest way towards an approximation is to ascribe to the fatty oil a mean 
 saponification number of 190. The percentage of fatty oil in the extract 
 may then be calculated by means of the formula, 
 
 0/ _ 100 n 
 /0 190 w 
 
 in which n is the number of milligrammes of potassium hydroxide that 
 disappeared in the saponification process, and w the weight of the acetone 
 extract used for sapouification. 
 
 3. Nitrobenzene Extract (cold). This process, suggested by HENRIQUES 
 (see page 242) for the detection of asphaltum in india rubber articles, may 
 be employed for the same purpose in the analysis of recovered india rubber. 
 The process, is, however, scarcely a very satisfactory one, as, on the one 
 hand, not every specimen of asphaltum is readily or completely soluble in 
 slightly warm nitrobenzene, and, on the other hand, coal tar and coal tar 
 preparations, which are freely employed in the production of a number of 
 india rubber manufactures, will also dissolve to greater or less extent in 
 the same solvent. Its usefulness is, however, unquestionable, as it permits 
 of the removal of certain organic non-rubber components, which could not 
 be eliminated either by the acetone extraction or by the later treatment 
 with alcoholic caustic soda. 
 
 4. India Rubber Substitutes (fatty). The detection and estimation of 
 these very largely vised so-called substitutes is based upon the saponifying 
 action of alcoholic caustic soda upon them. This valuable process, which 
 is also due to HENRIQUKS, is fully described in a later portion of 
 this book. To obtain the correct figure of the sulphur of vulcanisation 
 contained in the india rubber of a sample of recovered rubber, the de- 
 termination of the sulphur in the soda extract is indispensable. The test 
 for chlorine and its determination in this extract should likewise never be 
 omitted. 
 
 5. India Rubber and Combined Sulphur. The sample which has served 
 for the three last named tests is carefully dried under conditions preventing 
 its oxidation, and accurately weighed out into two exactly equal portions 
 one of which serves for the estimation of the india rubber, the other for the 
 determination of the combined sulphur. The processes adopted are, of 
 course, exactly the same as those used in the analysis of ordinary india 
 rubber articles. 
 
130 INDIA RUBBER AND ITS ANALYSIS. 
 
 6. Carbonaceous Matter, and 7. Mineral Constituents. The insoluble 
 residue remaining after the resolution of the india rubber by means of boil- 
 ing nitrobenzene retains all the mineral and carbonaceous matter contained 
 in the original sample. It may also contain vegetable and animal fibres. 
 The separate estimation of these in the presence of carbonaceous matter is 
 scarcely feasible, but their presence should be noted as objectionable. The 
 determination of the carbon, or carbonaceous matter, in cases where an 
 absolutely correct analysis is required, is rather troublesome. It is carried 
 out as described in Chap. IX. (page 252). Where no carbonaceous matter 
 is present, the usual lines of inorganic analysis are followed. 
 
 The most important criteria as to the quality of a sample of recovered 
 rubber are the amounts of india rubber, free sulphur, and sulphur of vul- 
 canisation present. As regards free sulphur, I consider its presence in any 
 appreciable amount as a sure indication that not even an attempt at de- 
 vulcanisation, the only satisfactory aim of recovery, has been made. Any 
 treatment calculated to eliminate the sulphur of vulcanisation would be 
 certain to effect, in the first instance, the removal of some of the free 
 sulphur. No doubt it is quite possible to remove the free sulphur without 
 producing de vulcanisation, so that the absence of sulphur in this form could 
 not be held to prove that even a partial actual devulcanisation has been 
 effected ; but I have satisfied myself by numerous experiments that all 
 known processes for the production of recovered rubber, when carried out 
 in the presence of free sulphur, invariably affect the result detrimentally, so 
 that, as I stated above, the presence of free sulphur may safely be taken to 
 prove the employment of an irrational recovery process. 
 
 All recovering processes at present in use, as far as is known, are 
 carried out at temperatures above 150 C. At this temperature free 
 sulphur contained in the india rubber under operation may partly sub- 
 lime, or enter into combination with metallic oxides contained in the india 
 rubber, or interact with the oils or other substances employed in the 
 process. It is therefore even possible for a sample of recovered rubber to 
 have passed through the recovery process with all the free sulphur in it, 
 and in the end to be altogether free from this element. Whether this can 
 be accomplished in this manner without injury to the rubber it is impossible 
 to say : as far as my experiments go, they show that this result can only be 
 obtained at an inordinately high temperature. Such a sample may then 
 contain no free sulphur and still be objectionable, as in this case freedom 
 from sulphur has been purchased at the expense of the india rubber, which 
 at the high temperature at which recovery was effected underwent partial 
 decomposition. It is therefore important, for the purposes both of practical 
 assay and of experimental research on the question of recovery, to possess 
 means to enable us to detect the presence of decomposition products of 
 india rubber in the recovered article. 
 
 Now, the fact that such decomposition products, if present, were formed 
 in the recovering process, not so much through chemical action as through 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 
 
 139 
 
 the action of heat, at once suggests that we may expect them to be sub- 
 stances of the kind we encounter amongst the products of the dry distillation 
 of india rubber, or, what is even more likely, similar to those obtained in 
 boiling india rubber in nitrobenzene. But in either case we may expect 
 them to be soluble in acetone in other words, they will form part of the 
 acetone extract, the importance of ascertaining which we have already 
 discussed. The detection and estimation of these decomposition products 
 in the presence of the oils, resins, and paraffin wax almost invariably 
 occurring in recovered india rubber is, however, no easy matter. A separa- 
 tion of these substances from the decomposition products of the india 
 rubber by solvents appears to be impossible. Fairly satisfactory results may, 
 however, be obtained by utilising the fact that the decomposition products 
 of india rubber are terpenoid bodies, 1 which form bromine addition products 
 containing upwards of 60 per cent, of bromine. These are insoluble in 
 alcohol, and very difficultly soluble in benzene ; whereas any bromine addition 
 or substitution products of the oils occurring in recovered india rubber are 
 readily soluble in both alcohol and benzene, and also in a mixture of the 
 two. Therefore, if we dissolve the acetone extract, after distilling off the 
 acetone, in just sufficient of a mixture of two volumes of benzene with one 
 volume of absolute alcohol to dissolve the whole of the extract, and then 
 add bromine to slight excess, we obtain, if decomposition products of india 
 rubber are present, a precipitate consisting of the bromine addition products 
 of the latter. This can be filtered off, and should be washed first with the 
 mixture of benzene and alcohol, and finally with absolute alcohol. It is 
 then dried at a temperature not exceeding 50 C. The composition of this 
 bromo-derivative, although constant for the same sample of vulcanised 
 india rubber or its decomposition products, is not so for different samples of 
 vulcanised india rubber. The reason for this has been already fully dealt 
 with,'- but for our present purpose has to be borne in mind. It necessitates an 
 estimation of the bromine in the dried and weighed bromo-derivative, and from 
 the weight of the latter must be deducted the weight of the total bromine 
 contained in it. The result gives us, then, very approximately the quantity 
 of soluble decomposition products of the india rubber of a given sample of 
 recovered rubber. The following figures were obtained in this manner : 
 
 Description of Sample. 
 
 5 grms. yield 
 Acetone Extract. 
 
 Bromo- 
 Deriva- 
 tive. 
 
 Bromine. 
 
 Decomposition 
 Product. 
 
 
 percent. 
 
 grms. 
 
 . 
 grms. 
 
 percent. 
 
 grms. 
 
 grms. 
 
 percent. 
 
 f No. 1 black, 
 
 9-94 
 
 497 
 
 117 
 
 66-66 
 
 078 
 
 039 
 
 78 
 
 American] No. 2 ,, 
 
 9-38 
 
 469 
 
 128 
 
 67-10 
 
 086 
 
 042 
 
 84 
 
 I No. 3 grey, 
 ( No. 1 black, 
 
 10-11 
 
 12-00 
 
 505 
 600 
 
 151 
 314 
 
 65-56 
 64-96 
 
 099 
 204 
 
 052 
 
 no 
 
 1 04 
 2'34 
 
 English \ No. 2 ,, 
 
 10-40 
 
 520 
 
 233 
 
 65-66 
 
 153 
 
 080 
 
 1-60 
 
 (No. 3 ,, 
 
 14-27 
 
 713 
 
 581 
 
 63'85 
 
 371 
 
 210 
 
 4-21 
 
 n i \ No. 1 grey, 
 
 11-67 
 
 583 
 
 661 
 
 64-29 
 
 425 
 
 236 
 
 473 
 
 ./No. 2 
 
 10-35 
 
 517 
 
 654 
 
 62-53 
 
 409 
 
 245 
 
 4-92 
 
 1 Sec page 22 ff. 
 
 See page 11. 
 
140 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 All the above percentages are calculated upon the India rubber only 
 contained in each sample, irrespective of anything else present. The above 
 acetone extracts consisted, to much the greater part, of mineral oil. No. 1 
 of the American samples contained vaseline ; the two French samples 
 paraffin wax. 
 
 The percentage of ' decomposition product ' contained in a sample of 
 recovered rubber is well illustrated by the following analyses of three 
 samples prepared from the same ground rubber scrap with the same amount 
 of mineral oil, but treated at different temperatures. 
 
 Sample. 
 
 5 grnis. yield 
 Acetone Extract. 
 
 Bromo- 
 Derivative. 
 
 Bromine. 
 
 Decomposition 
 Product. 
 
 No. 1 treated at 1 40 C. 
 
 p.c. 
 10-44 
 
 grms. 
 522 
 
 grms. 
 149 
 
 p.c. 
 6174 
 
 grms. 
 92 
 
 grms. 
 057 
 
 p.c. 
 1-16 
 
 No. 2 ,,160C. 
 
 10-84 
 
 542 
 
 239 
 
 61-08 
 
 146 
 
 093 
 
 1-86 
 
 No. 3 185 C. 
 
 11-62 
 
 581 
 
 325 
 
 61-53 
 
 199 
 
 126 
 
 2-53 
 
 The above figures show that much valuable information is to be gained 
 from the estimation of the decomposition products in recovered india 
 rubber. Of course I do not suggest that this determination should be 
 carried out with every sample of recovered india rubber which is in any 
 way analytically examined. But the fact that such an examination is at all 
 possible is well worth recording, and, moreover, the method is not only useful 
 for the purpose for which we have been discussing it, but will most likely 
 also render useful service in the analysis of decomposed india rubber articles, 
 and investigations into the cause of such decomposition. It is, moreover, 
 noteworthy that the bromine percentages given above fall far short of 
 what is required for the polyprene tetrabromide, and, moreover, exhibit 
 considerable variations in this respect. It is highly probable that the 
 degree of vulcanisation of a sample of india rubber is, to a large extent, if 
 not entirely, responsible for this difference and its variation ; and if this should 
 be so, the above method could be modified into a chemical test for the 
 degree of vulcanisation of vulcanised india rubber. On the other hand, the 
 effect of oxidation upon india rubber would likewise tend to reduce its 
 bromine absorption, so that the above method can no doubt also be em- 
 ployed to throw light on this still very obscure subject. Altogether, I 
 think there are enough reasons to make this method one of peculiar interest 
 to the chemist engaged in the analysis of india rubber, and it offers in itself 
 a wide field for promising research in a number of almost unexplored 
 directions. 
 
 B. Fatty and other Organic Substitutes. 
 
 The number of ' Rubber Substitutes ' which have been proposed is legion, 
 and the annual patent registers of all manufacturing countries invariably 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 141 
 
 record a number of more or less new preparations intended to serve the 
 same purpose, but scarcely any one of which seems even to pass into the 
 advanced experimental stage, and I am not aware that any single one of 
 these preparations suggested during the last ten or fifteen years has found 
 practical employment to any appreciable extent. Indeed, although certain 
 preparations, which we are about to discuss in the following pages, have 
 been for a long time, and still are at the present moment, largely employed 
 under the name of rubber substitutes, it must at once be said that this 
 description as applied to them is a misnomer. We know at the present 
 time of no manufactured product which in any way would deserve the 
 name of rubber substitute. All the substances so described are pure and 
 simply organic adulterants, as distinguished from inorganic or mineral 
 bulking or filling agents. They reduce the quality of an india rubber 
 article exactly in the proportion in which they are present in the article, 
 and the justification of their use lies purely in the popular demand for 
 cheapness. 
 
 All the rubber substitutes which at the present time are of any 
 commercial importance are obtained from one or other of the * unsaturated ' 
 fatty oils, either by a process of oxidation, or by treatment at higher 
 temperatures with sulphur, or by treatment with sulphur monochloride 
 in the cold. The process of manufacture of these products will be found 
 below under the various headings descriptive of each of them. 
 
 All these so-called ' substitutes ' have one and the same fault in 
 common, which lies in the fact that their typical chemical character is still 
 that of a triglyceride. They are, therefore, all readily saponifiable by any 
 of the ordinary means of saponification. In this respect more particularly 
 the action of alkalies and high pressure steam deserves consideration. 
 The latter especially represents a condition to which a very large and 
 important class of india rubber manufactures is continually exposed, and 
 therefore the use of these substitutes in the manufacture of such articles 
 must be strongly condemned. It might perhaps be objected that the 
 saponifying action of high pressure steam is of no practical importance 
 owing to the imperviousness of india rubber, but, in reality, this counts 
 for little. The saponification begins at the surface, but proceeds apace, 
 chiefly owing to the corrosive (dissolving) action of the free fatty acids 
 formed upon the india rubber. As far as the author is aware, the ' moist 
 heat ' English Admiralty Test l for india rubber goods is the outcome of 
 purely empirical experiments and considerations, but it finds its entire 
 justification and rational explanation in the above noted behaviour of the 
 india rubber substitutes to high pressure steam. The Admiralty test 
 is simply a rough and ready test what is euphoniously called a 
 practical test for the presence of * substitute ' in an india rubber 
 article. 
 
 In articles which under ordinary conditions of use are exposed neither 
 
 1 See page 229. 
 
142 INDIA RUBBER AND ITS ANALYSIS. 
 
 to the action of alkalies nor to the influence of high pressure steam, the 
 presence of these substitutes is less objectionable, and the justification for 
 their employment in these cases rests upon the price of the articles in 
 question. It should, nevertheless, be pointed out that admixture with 
 substitutes invariably tends to shorten the life of an india rubber article, 
 generally increases its liability to decay, and greatly diminishes its 
 resistance to friction and to torsional or longitudinal stresses. 
 
 The above remarks concerning the use of substitutes in india rubber 
 goods apply equally to all the substitutes which are in the market at the 
 present time. It is therefore of considerable importance to examine, in 
 the first instance, the considerations upon which the quality and 
 suitability of an india rubber substitute may be determined with a 
 view to its employment in manufacturing, and also what methods we 
 possess for the detection and quantitative determination of rubber 
 substitutes in rubber articles. But while doing so, a rough outline of the 
 manufacture of the most important substitutes in use at the present time 
 may also be found helpful. 
 
 The Manufacture of Rubber Substitutes from Fatty Oils. The origin 
 of the manufacture of rubber substitutes from fatty oils probably dates 
 back to a much earlier time than the india rubber industry itself. 
 Preparations obtained by heating linseed oil with sulphur were probably 
 known to the alchemists. At any rate, under the description of ' oleum lini 
 sulphuratum,' or 'balsam sulphuris,' they occupied an important place 
 amongst the remedies of our forefathers, and, indeed, are even yet retained 
 in the pharmacopoeias of most countries. The fact, that under certain 
 conditions, as regards degree of temperature and time during which the 
 heat was applied, elastic solids are obtained by this process, would soon be 
 observed, and when, towards the middle of this century, the manufacture 
 of india rubber articles became established, the remarkable similarity 
 between india rubber and these sulphuretted oils could not fail to attract 
 attention and suggest the use of such products in conjunction with 
 india rubber. Accordingly, we find that the first rubber substitute which 
 appeared in the market, and which to this day still holds the field, is a 
 product obtained by the interaction of linseed oil and sulphur. 
 
 The action of sulphur upon unsaturated fatty oils is by no means 
 completely understood. ALTSCHUL l showed that two different compounds 
 are formed according to the temperature at which the reaction takes place. 
 At temperatures not exceeding 150 C., up to 16 per cent, of sulphur are 
 at first dissolved, and after some time absorbed by the oil. At this point 
 it will be found that a drop of the oil, on being brought upon a glass 
 plate, no longer turns milky or opalescent, owing to merely dissolved 
 sulphur crystallising out, nor does precipitation of sulphur occur on 
 dissolving the oil in ether. As, moreover, this point can be reached without 
 any appreciable quantities of hydrogen sulphide being evolved, we may 
 1 Zeitschr.f. angew. Chem., 1895, p. 535. 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 143 
 
 infer that this absorption of the sulphur by the oil under the above stated 
 conditions is an addition process pure and simple. 
 
 If the interaction between the unsaturated oil and sulphur is carried on 
 at temperatures from 150 C. upwards, the evolution of hydrogen sulphide 
 becomes more marked as the temperature increases. This is accompanied by 
 a very considerable thickening of the oil, which eventually becomes perfectly 
 solid. Substances solidified in this manner are prepared at present from 
 a great variety of oils, such as linseed, colza, castor, cotton seed, and maize 
 oils, and these products are largely used as so-called rubber substitutes, 
 although in reality they are nothing but organic filling or bulking 
 material. 
 
 In preparing such substitutes, the quantity of sulphur employed is, of 
 course, a matter of some importance. Nevertheless will it be found that 
 the commercial products vary very greatly in this respect, the percentage 
 of sulphur in them being often as low as 5 per cent., and, on the other hand, 
 exceeding in many cases 17 per cent. This is, in the first instance, due to 
 the fact that, instead of employing for the manufacture of these substitutes 
 the above named oils in their natural state, 'blown' oils are used i.e., oils 
 which have been treated with air or oxygen at higher temperatures. 
 In this treatment oxygen addition products are formed, the oxygen com- 
 bining with the unsaturated fatty acid of the glyceride. This is accom- 
 panied by a thickening of the oil, which becomes more pronounced the 
 greater the amount of oxygen taken up by the oil, although there is no 
 doubt that this treatment, at the temperature at which it is effected, leads 
 to further effects which tend to increase the viscosity of these oils. The 
 treatment may be carried so far, that eventually the oil is converted into 
 a more or less ' dry ' and resilient solid (see below) ; but for the purpose of 
 the manufacture of the sulphuretted substitutes, the 'blowing' is only 
 carried far enough to reach a specific gravity of about 0'960. The con- 
 stitution of the addition products thus formed is entirely conjectural, but 
 the effect of this oxygen addition is, of course, a reduction of the state of 
 non -saturation of the oil, and consequently a smaller percentage of sulphur 
 is necessary to convert one of these ' blown ' oils into a completely saturated 
 compound than would be required for the natural oil. In the trade the sub- 
 stitutes made from blown oils are described as ' floating substitutes ' owing 
 to their property to float on water, which, of course, is due to their specific 
 gravity being slightly less than that of water ; while the substitutes made 
 from ordinary oils possess a specific gravity higher than that of water, and 
 so are non-floating. 
 
 There are two points as regards substitutes of the above description 
 upon the careful control of which depend their satisfactory working in the 
 factory, as well as the life of the articles manufactured with their aid. These 
 two points are (a) the amount of soluble fatty oil they contain, and (b) the 
 amount of free sulphur. Both these quantities should be as low as possible. 
 A high percentage of soluble fatty matter invariably shortens the life of 
 
'.44 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 the India rubber, and a high percentage of free sulphur would either require a 
 diminution in the proportion of the substitute used, or, if unnoticed, might 
 lead to serious mishaps in the vulcanisation process. This is an obvious 
 point, which, however, is often neglected, not infrequently with disastrous 
 results. In calculating the amount of sulphur to be added to a rubber 
 mixing containing substitute, the amount of sulphur present in the latter 
 should evidently be deducted, and for this reason it is most important that 
 this amount should be accurately known. 
 
 Some typical samples of brown substitute gave the following results on 
 analysis : 
 
 
 I. French Make. 
 
 II. English Make. 
 
 III. American Make. 
 
 FreeS, 
 Combined S, 
 Soluble oil, 
 
 2-31 p.c. 
 6-58 
 2-20 ,, 
 
 2-41 p.c. 
 6-84 
 2'40 
 
 5 '22 p.c. 
 19-81 
 978 
 
 Samples I. and II. are practically identical. Sample III. I consider a 
 rather objectionable specimen, not so much on account of its rather high 
 percentage of free sulphur, as on account of the large proportion of soluble 
 oil it contains. This sample has been made from maize oil. 
 
 Now it might appear that samples I. and II., which are curiously similar 
 in composition, would also in their working produce the same result. This, 
 however, these analytical data do not enable us to infer. Indeed, there is 
 no analytical method which would allow us to decide upon the quality of 
 two different brands of brown substitute which are, like the above, practically 
 similar in their chemical constants. Nevertheless, we are able to decide 
 this question by examining mixtures of india rubber and these substitutes 
 microscopically. Mixtures of these two substances, of course, do not repre- 
 sent intermolecular penetration of the constituents as do true solutions, 
 but are a sort of solid emulsions of the substitute in the india rubber ; and 
 we may therefore judge the quality of two substitutes by determining 
 which of them forms the finest emulsion. The difference in this respect is 
 often very great even in samples which are apparently chemically identical. 
 In bad cases no granular emulsion is obtained at all, but the substitute 
 breaks up into a mass of streaky or scaly flakes. Practical experience has 
 shown this microscopical test to be invariably reliable and trustworthy. 
 Details regarding the working of this test will be found in the analytical 
 portion of this chapter. 
 
 There is, however, another class of substitutes which is distinguished 
 from the 'brown substitutes' above described by the name of white 
 substitutes, and which is formed by substances produced by the interaction 
 of sulphur monochloride and unsaturated oils. ULZER and HoRN, 1 who 
 were the first to investigate the products of this reaction, found them to 
 
 1 Jahresber. d. chem. TechnoL, 1890, p. 1177. 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 145 
 
 consist of glycerides, the fatty acids of which contain chlorine and sulphur. 
 They further found that these products can be saponified with alcoholic 
 caustic soda, during which process all the chlorine is eliminated from the 
 fatty acid, which, however, retains its sulphur intact. HENRIQUES l has very 
 considerably extended ULZER and HORN'S work, and established the fact 
 that these white substitutes are simply addition products of the fatty oils 
 and chloride of sulphur. In accordance with this fact, we find that the 
 iodine absorption of these substitutes is very much smaller than that of the 
 oils from which they were produced. 
 
 The simplest way to produce these substitutes is by mixing the fatty 
 oil with a certain quantity of chloride of sulphur and agitating the mixture 
 well. In a very short time a vigorous reaction sets in, the mixture getting 
 very hot, evolving vapours of sulphur monochloride, and also small 
 quantities of hydrochloric and sulphurous acid. Simultaneously the mass 
 thickens, and is rapidly converted into a solid, semi-transparent, elastic 
 substance, the colour of which varies from a pale amber colour to a more 
 or less pronounced brown. This product generally feels the * dryer,' the 
 darker its colour. It is easily ground into a mass of fine flakes of much 
 lighter colour than the original substance. As is to be expected, the result 
 depends very much upon the proportions of oil and sulphur monochloride 
 employed, and also upon the kind of oil used. HENRIQUES 2 determined the 
 proportion of sulphur monochloride required to produce dry solid substitutes- 
 with various kinds of oils, and obtained the following figures : 
 
 100 parts of linseed oil require 30 parts of S 2 C1 2 . 
 poppy seed oil 35 
 
 , colza oil 
 
 , cotton seed oil 
 
 , olive oil 
 
 , castor oil 
 
 25 
 45 
 25 
 20 
 
 It must, however, not be assumed that the products obtained from 
 these oils by their interaction with sulphur monochloride form all equally 
 satisfactory * rubber substitutes.' This is, indeed, not the case. Especially 
 the product obtained with castor oil has been found wanting in stability, 
 but also the substitutes from olive oil and linseed oil proved defective in 
 this respect. At the present time a well-refined German seed, colza oil, is 
 used for this purpose, to the exclusion practically of any other oil. 
 HENRIQUES 3 patented the use of oxidised (blown) oils for the manufacture 
 of these substitutes, and he claims that in this manner he obtains sub- 
 stitutes which are not only cheaper than substitutes prepared from natural 
 oils, owing to the considerably smaller quantities of sulphur monochloride 
 required for their conversion, but he also states them to be superior to the 
 ordinary substitutes. The latter claim is certainly not in accordance with 
 the experience obtained with such substitutes in English factories, where 
 
 1 Chem. Zeitg., 1893, p. 707. 2 Ibid. 
 
 3 Germ. Pat., No. 73,045, 1893. 
 
 10 
 
146 INDIA RUBBER AND ITS ANALYSIS. 
 
 they were tried fairly and extensively years before the date of HENRIQUES' 
 patent. 
 
 In preparing these substitutes on the large scale, the most important 
 point is to control the reaction so as to prevent a rise of temperature of 
 above 70 C. in any part. For this reason it is generally recommended to 
 prepare them in small quantities at a time. SCHULZE l states that not more 
 than 20 Ibs. of oil should be treated at a time. As a matter of fact, much 
 better results are obtained by working with 3 to 5 cwt. batches, but in 
 order to check the reaction, the addition of a suitable quantity of some 
 hydrocarbon, the boiling point of which should not exceed 70 C., is 
 necessary. Also SCHULZE'S statement, that the operation should not be 
 carried out in wooden vessels, is without justification. Of course, when 
 preparing these substitutes on such a large scale, open vessels can no longer 
 be employed, and a closed apparatus with an effective vapour-condensing 
 plant attached must be used. 
 
 The relative nature of both the ' brown ' and ' white ' class of substitutes 
 above described is very clearly brought out by the analytical results 
 published by HENRIQUES, 2 and given on the following page. 
 
 It is obvious from these figures that there can hardly ever arise any 
 doubt as to whether a fatty substitute contained in an india rubber article 
 is a brown or a white substitute. The readiness with which these 
 substitutes are saponified by alcoholic caustic soda renders their complete 
 extraction from the finely ground india rubber sample an easy matter. 
 The amount of substitute present in any given sample cannot, of course, be 
 estimated by direct methods at any rate, in most cases without liability 
 of considerable error. It is therefore estimated by the loss in weight of 
 the sample after treatment with alcoholic caustic soda. The amount 
 of chlorine extracted by the alcoholic caustic soda decides, of course, the 
 nature of the substitute present. 
 
 The general practice is to use the brown substitutes for 'dry heat 
 cured 7 goods only i.e., goods vulcanised with sulphur; and the white 
 substitutes for 'cold cure' goods i.e., goods vulcanised with chloride of 
 sulphur (S 2 C1 2 ). There is, however, a certain class of ' dry heat cured ' 
 goods, for the manufacture of which white substitute is employed. This at 
 first sight appears surprising, in view of the fact that white substitute 
 already at a temperature of 90 C. loses a considerable part of its chlorine 
 in the shape of hydrochloric acid ; and at vulcanising temperature 130 C. 
 this effect of heat upon white substitute becomes considerably more 
 noticeable. It might, therefore, be thought that on hot vulcanising a 
 mixing containing white substitute, a considerable risk of ' blowing ' was 
 incurred owing to the evolution of hydrochloric acid gas by the substitute ; 
 but in practice, provided the mixture does not contain any inorganic 
 carbonates or sulphides which are acted upon by sulphuretted hydrogen, 
 the curing of such a mixture causes no difficulty whatever, and 
 
 1 Gummi Zeitung., 1898, p. 311. 2 Chem. Zeitg., 1893, p. 707. 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 
 
 147 
 
 Substitute from 
 Oxidised 
 Cotton Oil. 
 
 CO 
 
 CO 
 
 CO 
 
 : co 
 
 o 
 53 
 
 CO 
 
 o 
 
 CO 
 
 rH 
 Oi 
 
 CO 
 rH 
 
 Substitutes with 
 maximum S 2 C1 2 
 
 I 
 
 O5 
 
 oo I : 
 
 I ! 
 
 1 
 
 05 
 !M 
 
 CO 
 rH 
 
 CO 
 
 O 
 rH 
 
 Castor Oil with 
 minimum S 2 C1 2 . 
 
 oo 
 
 o 
 
 CO CO 
 
 (M 
 
 CO 
 
 J* 
 
 
 *f 
 
 co : : 
 
 oo 
 
 o 
 
 Substitute from 
 Oxidised Lin- 
 seed Oil. 
 
 
 
 : : 
 
 
 00 
 <M 
 
 O5 
 
 : 
 
 Substitute from 
 Oxidised Poppy- 
 seed Oil. 
 
 oo 
 
 CO 
 
 rfl 
 
 O (M 
 
 05 CO 
 ^f OO 
 
 
 CO 
 
 CO 
 
 CO 
 
 CO 
 
 
 Linseed Oil Substitutes. Colza Oil Substitutes. 
 
 S* g 03 ' 
 
 O5 
 
 O5 
 
 Ci o 
 
 fe 
 
 1 
 
 O5 
 
 s 
 
 oo 
 
 i 
 
 : 
 
 Prepared Prepared 
 from frum 
 Oxidised Natural 
 Oil. Oil. 
 
 oo 
 
 00 
 CO 
 
 , 
 
 CO 
 
 CO 00 
 
 oo 
 
 1 
 
 CO 
 
 
 rH 
 
 rH 
 CO 
 
 00 OO 
 
 *** o ' 
 
 t^ CO 
 
 to p 
 
 00 
 
 o 
 
 CO 
 
 o 
 
 CO 
 
 C<J O 
 
 rH rH 
 
 05 
 
 Prepared 
 from 
 Natural 
 Oil. 
 
 CO 
 05 
 
 00 O 
 
 OO DO ' 
 
 oo 
 
 CO OO 
 
 05 05 
 
 1 
 
 CO 
 
 5 
 
 CO 
 
 O 
 CO 
 
 i 1 
 
 s 
 
 Commercial Products. 
 
 ( Z \ 
 
 TH 
 
 CO 
 CO 
 
 o 
 
 
 p 
 
 rH 
 
 
 rH 
 
 o : : 
 
 : A 
 
 
 
 !L 
 
 00 
 
 ^ 
 
 rH 
 
 
 
 
 <N 
 
 
 O5 
 (M 
 
 
 " 
 
 o : : 
 
 rH 
 
 
 
 i f d 
 
 I 
 
 oo 
 
 00 
 
 ob : : 
 
 rH 
 
 : oo 
 
 
 CO 
 
 CO 
 
 o 
 
 rH 
 
 : 
 
 ~s 
 1 1 
 
 -2 
 
 'i 
 
 CO rH 
 00 
 
 00 O 
 CO CO 
 
 CO 
 
 T* 1 
 o 
 
 CO 
 
 rH 
 
 O5 
 
 ; 
 
 
 CO 
 
 kfl 
 
 O OO 00 
 000 
 
 Jf i 1 
 
 CO 
 
 oo 
 o 
 
 05 
 
 o 
 
 CO 
 
 CO 
 rH 
 O5 
 
 
 
 
 
 
 
 
 1 
 
 O 
 
 <3 . 
 
 'a 
 
 si 
 i 
 
 1 
 
 i 
 
 Sulphur in substitute, 
 
 o 
 
 1 
 
 III 
 
 ' 'i 
 
 3" !g 
 
 1 9 
 
 fi 5 ^ 
 
 r5 
 
 Chlorine ,, ,, 
 
 Iodine number of substi 
 tute, 
 
148 INDIA RUBBER AND ITS ANALYSIS. 
 
 there is certainly no escape of hydrochloric acid noticeable. This apparent 
 anomaly, no doubt, finds its explanation in the existence of a polyprene 
 dihydrochloride, in the fact that india rubber forms an addition product 
 with hydrochloric acid. Although this addition product under the 
 influence of heat gives off hydrochloric acid, this process (see page 146) 
 is very far from complete; and as the white substitute at vulcanising 
 temperature also parts with a certain fraction only of its chlorine in the 
 form of hydrochloric acid, it follows that under practical conditions the 
 amount of hydrochloric acid evolved by the substitute always falls con- 
 siderably short of the quantity india rubber can absorb and retain at 
 vulcanising temperature. This is all the more so, as the quantity of white 
 substitute which can be used in admixture with india rubber for the 
 manufacture of ' dry heat ' goods is limited by the fact that such mixtures, 
 in order to ensure satisfactory vulcanisation of the india rubber, require 
 quantities of sulphur increasing rapidly with the proportion of substitute 
 present, so that, for the vulcanisation of mixtures in which white substitute 
 predominates, quite enormous proportions of sulphur (over 30 per cent.) 
 would be required. It cannot be said that goods of this description are 
 not made, but that they should be must be regretted, the more so as the 
 only reason for the employment of white substitute in ' dry heat ' goods i& 
 the colour of the finished substance. Brown substitute, of course, 
 produces goods of a somewhat dark-yellowish or brownish-grey colour, but 
 the trade insists upon having certain articles, gas-tubing especially, of a 
 light greyish-white colour, which can only be obtained by means of white 
 substitute. There is nothing, apart from the question of colour, to 
 commend the use of white substitute in 'dry heat' goods indeed, the 
 practice is unquestionably detrimental to the quality of these goods. 
 
 The above described ' brown ' and ' white ' substitute represent probably 
 fully 90 per cent, of all the so-called rubber substitutes used. How little 
 they really deserve this name has already been shown in the introduction 
 to this chapter. Nevertheless, every year unfailingly produces its regular 
 crop of 'inventions,' which in their early stages are generally stated to 
 realise the production of artificial india rubber, but which are simply the 
 application of the above described reaction of sulphur, or chloride of 
 sulphur, upon fatty oils, upon either some new kind of oil, or upon an oil 
 subjected to some preliminary treatment (blowing), or upon such oil 
 under conditions more or less novel generally less. Of all these new 
 substitutes there is only one which has been at all able to obtain some 
 foothold in the market, though it could by no means be said to be firmly 
 established at the present. This is the curious product named, after its in- 
 ventor, Fenton rubber. This substitute is manufactured l by mixing oxidisable 
 vegetable oils, either in the raw or oxidised state, with certain proportions 
 of tar, pitch, creosote, or * other form of pyroligneous acid,' and placing " the 
 mixture, either cold or previously heated together, into a bath of diluted 
 1 Eng. Pat, No. 26,548, 1896. 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 149 
 
 nitric acid, and allowing it to remain for maceration cold, or quickening 
 the action by boiling until the whole is coagulated into a tough elastic 
 magma ' having the appearance ' and nature of caoutchouc-like substance." 
 This product is poured upon pans and roasted for a short time, "when it 
 coagulates into a highly elastic mass in appearance like india rubber, and 
 can be vulcanised .exactly as that substance is." 
 
 This is simply a re-invention of a process first employed in 1848 by 
 SACHS and JONAS, with the very questionable ' improvement ' of the addition 
 of tar, pitch, creosote, or * other form ' of pyroligneous acid. 
 
 The rubber substitutes we have been considering so far are all intended 
 to replace part of the india rubber in rubber mixtures. There is, however, 
 a very numerous class of substances which are actually intended to replace 
 india rubber entirely for both soft and hard rubber (ebonite) articles. 
 Attempts in this direction up to now have always failed, and probably 
 always will, if these substitutes are used for articles in which high 
 tensile strength, great elasticity, resistance against abrasive forces, and 
 indifference to chemical, especially alkaline, reagents, are chiefly required. 
 Where waterproof qualities, electrical insulation resistance, sound-deadening, 
 and similar qualities only are required, such substitutes may be more or less 
 serviceable. In the past they have been greatly discredited, owing to the 
 immoderate claims put forward on their behalf claims which it was in- 
 variably found could not be sustained. 
 
 The number of patents which have been taken out for rubber sub- 
 stitutes of this description is truly appalling, and becomes still more so 
 when we discover that, with only an occasional exception, all these patents 
 are based upon two reactions : either the action of sulphur upon fatty oils, 
 or the action of oxygen upon oxidisable oils. And herein lies the weak 
 point of all these substitutes : they are all saponifiable, and they all lack 
 the resistance possessed by india rubber to temperature changes within 
 comparatively wide limits. To call these substances india rubber sub- 
 stitutes, or artificial india rubber, is grossly misleading. They can in no 
 case compete with india rubber, and their use begins only where that of 
 india rubber ceases i.e., uses where such highly compounded (mineralised) 
 mixings are required, that the india rubber acts more or less exclusively as 
 a cement for the mineralic powders only, as is the case in the very lowest 
 classes of india rubber goods. 
 
 It is impossible to give even an outline of all the patents that have 
 been taken for the manufacture of such substitutes, nor would such an 
 outline be at all worth the trouble. Even the latest patents in this 
 direction represent no progress over the methods which were employed 
 for the same purpose thirty years ago. What little actual improvement 
 there is, is simply one of manipulation. We may therefore confine our- 
 selves to a brief exposition of the mode of manufacture of one of the more 
 recent of these compounds, known as * Oxolin.' 
 
 This is one of the innumerable patents based upon the formation of 
 
I5O INDIA RUBBER AND ITS ANALYSIS. 
 
 solid products by the oxidation of linseed oil. In this case this is per- 
 formed by impregnating vegetable fibre, such as flax, hemp, or jute, with 
 linseed oil in a hydro-extractor. The impregnated material is then spread 
 on trays and oxidised by exposure to the air at a temperature of about 
 45 C. This operation is repeated a number of times, when the material 
 obtained by working between rollers is transferred into an apparently homo- 
 geneous mass in appearance much like india rubber, and which, by mixing 
 it with sulphur, is capable of a certain degree of vulcanisation on heating. 
 
 The elasticity of this product is very small, its tensile strength not to 
 be compared to that of india rubber; alkalies and high pressure steam 
 saponify it. But, no doubt, for a certain number of uses this product 
 answers as well as very low qualities of india rubber. 
 
 A very great number of preparations have been patented which are 
 intended to serve as india rubber substitutes by themselves, and not in 
 admixture with india rubber. So far not one of these preparations has 
 been more than partially successful, their applicability being mostly con- 
 fined to the insulation of electric conductors. 
 
 The oldest of these preparations is CALLENDER'S bitumen, which has 
 been, and is still, largely employed for the manufacture of the larger sizes 
 of insulated electric cables. It is a preparation very similar to the ordinary 
 brown substitute of the rubber manufacturer, and obtained by heating a 
 mixture of stearine pitch and sulphur under certain conditions. The 
 electric insulation obtainable with this preparation is very low as compared 
 with a good india rubber insulation, but its behaviour against high voltage 
 currents is nevertheless very satisfactory. For telephone or telegraph 
 cables this material is unsuitable, owing to its rather high inductive 
 capacity. 
 
 Highly interesting are NOBEL'S attempts to produce 'artificial india 
 rubber ' from nitrocellulose l by dissolving this substance in nitro or bromo- 
 nitro derivatives of certain aromatic hydrocarbons camphor or resin oil. 
 That ' resilient solutions ' of this description could at all be employed in 
 place of india rubber is highly improbable, but NOBEL'S idea of arriving 
 at a rubber substitute by preparing what I termed above a 'resilient 
 solution ' of a colloid is well worth further consideration. Very probably 
 the recent elaboration of 'Velvril,'a new substitute for iudia rubber and 
 gutta percha, by W. F. RsiD, 2 was suggested by NOBEL'S experiments in 
 this direction. 
 
 REID'S ' Velvril ' is a solution of a nitrocellulose of a ' low degree of 
 nitration ' in a nitrated oil, preferably linseed or castor oil. The progress 
 in this case lies in the substitution of a nitrated oil for NOBEL'S aromatic 
 nitre-compounds, and would no doubt result in the production of a much 
 less combustible, and probably also more 'structureless' or 'colloidal, 7 
 product than could be obtained by NOBEL'S formulae. But whether 
 
 1 Eng. Pat, No. "20,234, 1893 ; and No. 15,964, 1894. 
 
 2 Journ. Chem. Soc., 1899, p. 972. 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 151 
 
 ' Velvril,' for which its inventor claims a great deal, is a really satisfactory 
 substitute for india rubber and gutta percha requires more proof than we 
 possess at present. A priori it would appear very doubtful whether a 
 mixture of nitrated oils and nitrocellulose is a sufficiently stable product 
 to be generally used in place of india rubber and gutta percha. It would 
 certainly be surprising if such a mixture or solution was not greatly 
 affected by any considerable rise or fall of temperature over and below 
 mean atmospheric temperature. 
 
 Resins, Pitches, and Waxes. 1 
 
 The number of different resins and waxes, or waxlike substances, and 
 pitches which have been used in connection with rubber, or which are 
 supposed to impart new and valuable properties to rubber goods, is almost 
 endless. That in many cases, by the addition of one of these substances to 
 india rubber mixings, a certain technical effect can be obtained is un- 
 questionable, but it may, nevertheless, be safely said that the less any of 
 these substances are used, the better for the article produced. In soft 
 rubber goods only a very moderate use is and should be made of these 
 substances, but hard rubber goods often contain very large proportions of 
 resins, chiefly shellac and amber. 
 
 Perhaps the most interesting of these substances are the pitches, and 
 more especially those which are more of a colloidal or elastic description 
 rather than brittle. Used in moderate proportions in india rubber mixings, 
 they certainly seem to tend to increase the resiliency of the india rubber, 
 and to decrease its liability to deterioration by oxidation. Very valuable 
 results may be expected from a thorough methodical study of this interest- 
 ing subject, which, so far, has received scant attention only. 
 
 The analytical detection of those bodies in india rubber mixtures offers 
 very little difficulty as far as the resins are concerned, owing to their 
 insolubility in acetone and their solubility in alcohol. The latter offers 
 also a ready means for their separation from waxes and pitches which are 
 quite insoluble in alcohol. The waxes are all fairly freely soluble in 
 acetone, but insoluble in alcohol. The pitches boiled tar, stearine pitch 
 are partly soluble in acetone, and also in cold nitrobenzene, but the 
 proportion of soluble constituents in them is by no means constant. 
 Asphaltum and sulphuretted stearine pitch (CALLBNDER'S compound) are 
 scarcely, if at all, soluble in acetone. They dissolve more or less completely 
 in warm nitrobenzene. But it will be readily inferred from these that a 
 satisfactory estimation of these pitches, and especially if they should occur 
 together with resins or waxes, may occasionally be a very difficult matter. 
 Our inabilities in this respect are, however, to a considerable extent counter- 
 
 1 By this terra only the mineral waxes paraffin wax, ceresin, and ozokerite are 
 understood, but not bee's wax, the use of which offers no tangible advantage in conjunction 
 with india rubber, although it is supposed to be used in a certain kind of ' hard rubber ' 
 (Kiel compound). 
 
152 INDIA RUBBER AND ITS ANALYSIS. 
 
 balanced by the fact that mixtures of these substances are very rarely, if 
 ever, employed in rubber mixings ; and further, it is generally quite sufficient 
 for practical purposes to establish the presence and determine the quantities 
 of resinous or waxy substances in an india rubber sample without finally 
 identifying them. This, in fact, can always be done without much 
 difficulty. 
 
 The Chemical Analysis of Organic India Rubber Substitutes. 
 
 1. Recovered Rubber. The chemical analysis of recovered india rubber 
 is carried out in exactly the same manner as the analysis of manufactured 
 india rubber. A complete analysis, therefore, would have to account for the 
 following items : 
 
 (a} Free oils, or fats. 
 
 (&) Resins, waxes, pitches. 
 
 ( S in substitute. 
 
 (c) Fatty rubber substances | ^ - n 
 
 (d) Free sulphur. 
 
 (e) India rubber. 
 
 (/) Sulphur of vulcanisation. 
 
 (g) Insoluble organic matter other than above named. 
 
 (h) Carbonaceous matter. 
 
 (i) Mineral constituents. 
 
 Indeed, to some extent this subject has already been dealt with in the 
 foregoing pages. For the determination of the suitability of a given sample 
 of recovered rubber, such a detailed analysis is generally unnecessary, and 
 the determination of the above items (a), (6), and (c) under one head, (d), 
 (e), (/), and again (g), (h), and (i) in the aggregate, will be found, in the 
 vast majority of cases, to supply all the information required and which 
 is obtainable by chemical analysis. 
 
 The value of a given sample of recovered india rubber depends upon two 
 factors, of which the first represents the ' waste rubber value ' of the sample, 
 and which is, with a very fair approximation, ascertainable by analysis. 
 The second factor consists in the ' technical effect ' produced by the recover- 
 ing process which is demonstrated by the plasticity of the sample, and its 
 elasticity and tensile strength after revulcanisation. 
 
 By estimating the amount of pure rubber in a sample of recovered 
 rubber, we are at once able to calculate the price paid for the rubber 
 actually present. The value of the other constituents substitute, oil, 
 sulphur, mineral matter is so small as to be hardly worth allowing for. 
 It is, of course, obvious that the ' waste rubber value ' should also depend 
 upon the quality or brand of rubber present, but as yet no sufficiently 
 reliable constants are known which would enable us to distinguish 
 analytically between different brands of india rubber ; and if there were 
 such, the fact that rubber goods are in the vast majority of cases manu- 
 factured from a mixture of at least two different brands of rubber would go 
 a long way to render such an estimation practically illusory. In practice, 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 153 
 
 however, this objection proves much less important as it might perhaps 
 appear, owing to the fact that the 'technical effect' produced by the 
 recovering process is determined not only by the efficiency of the 
 particular process employed, but very largely also by the brand of rubber 
 upon which it is exercised. We are not able to determine the ratio in 
 which these two items contribute to the * technical effect ' achieved, but it 
 is for practical purposes sufficient to determine it in its totality by 
 examining the elasticity and tensile strength of the sample after 
 re vulcanisation. 
 
 All the commercial samples of recovered rubber, many of which are 
 offered under very misleading descriptions, issue very badly from this 
 efficiency test. Indeed, it may safely be said that most of them are very 
 little more than mere bulking agents. Their chief advantage against the 
 fatty substitutes lies in the fact that they are very much more stable than 
 the latter, and therefore less liable to cause trouble in the manufacturing 
 process, or when in use in the form of india rubber articles of one kind or 
 another. That the foregoing statement is none too severe will become at 
 once apparent if a number of commercial samples of recovered rubber are 
 subjected to the extraction test with acetone. It will be found that all 
 these samples yield from 6 to 14 per cent, of an extract, the greatest part of 
 which consists of either a fatty oil, a hydrocarbon oil (both mineral and 
 resin oils are used), or a mixture of them. The extracted rubber will now 
 be found to have lost almost all its former plasticity, elasticity, and tensile 
 strength, and to consist of a more or less cohering aggregate of ' crumbs ' 
 of vulcanised rubber possessing neither plasticity nor tensile strength. 
 This shows that this kind of * recovered ' rubber is at best merely a kind of 
 semi-solid solution of the vulcanised rubber in oil, and in the worst cases 
 only a mass of crumbs, the cohesion amongst which is due by the stickiness 
 produced by the combined action of heat and oil upon india rubber. In 
 neither case could it be said that a true ' recovery ' of the india rubber had 
 been effected. All that can be claimed for such products is that they con- 
 tain vulcanised india rubber in a form in which it can more satisfactorily 
 be mixed with fresh rubber than if it was merely in the state of dry crumbs. 
 
 On submitting samples of recovered rubber of this description to the 
 vulcanisation test, products are obtained which compare very unfavourably 
 indeed with even the very low grades of crude rubber, and I am certainly 
 strongly of opinion that the prices asked and paid for recovered rubber of 
 this description are altogether out of proportion to its intrinsic value. 
 They are, unfortunately, used at present in a very indiscriminate and un- 
 critical manner, producing the semblance, but not giving the substance of 
 quality, in the articles produced by their aid. 
 
 The vulcanisation test is carried out by mixing the sample to be tested 
 in the usual way between mixing rollers with enough sulphur to make the 
 mixed product contain 6 per cent, of free sulphur. This mixture should 
 be calendered into a sheet of standard thickness, say J or T T F of an inch. 
 
154 INDIA RUBBER AND ITS ANALYSIS. 
 
 The sheets are then vulcanised under standard conditions. What these 
 conditions are is not very essential as long as all the tests are made under 
 the same conditions. The vulcanisation is effected between steam-heated 
 plates, but personally I prefer to vulcanise in high pressure steam, 
 as this method is certainly handier in the laboratory. In the latter case, 
 a digestor of the construction of fig. 1 is employed. It should be capable 
 of withstanding a working pressure of at least 150 Ibs., and should be fitted 
 with pressure gauge, thermometer tube, safety valve, and blow-off tap. 
 Inside this digestor a cage can be inserted, the bottom of which reaches to 
 within 2 '5 inches of the bottom of the digestor. This cage receives the 
 sample to be vulcanised, which is either secured upon a strip of glass plate 
 or wound helically, the convolutions not touching, upon a glass cylinder of 
 suitable size. The digestor is served with water to a height of 1*5 to 2 
 inches from the bottom, and the cage should certainly be clear of the sur- 
 face of this water. The digestor is heated by means of a Bunsen burner. 
 The height of the flame requires careful regulation. Even with this 
 arrangement it will, however, still be necessary to employ a gas governor 
 (pressure regulator) in order to keep the temperature reasonably constant. 
 ELSTER'S or STOTT'S gas governors prove most satisfactory, and the first 
 named is also the least expensive. 
 
 With regard to the temperature at which the vulcanisation is carried 
 out, and its duration, no hard and fast rules can be suggested. These 
 conditions may with advantage vary in different laboratories, according to 
 the varying vulcanising conditions which are dictated by the nature of the 
 articles manufactured. As a good average I recommend a temperature 
 of 130 C. for from two to two and a half hours. 
 
 The vulcanised strip is cut to a standard width, say 1 inch, and its 
 breaking strength estimated in a machine constructed upon the principle 
 of the well-known yarn testing machine. 
 
 It is very frequently found that the amount of mechanical work re- 
 quired to mix certain descriptions of recovered rubber with fresh rubber, 
 so as to produce as nearly homogeneous a * compound ' as possible, varies 
 very considerably with different brands and grades of recovered rubber. 
 Indeed, with some of them this result is unattainable, as the ' compound r 
 would require an amount of working which would prove fatal to the fresh 
 rubber employed. As a consequence, it is frequently found that 'com- 
 pounds' of this description are simply colloidal agglomerates exactly as 
 c compounds ' of rubber and oil substitutes, but generally much coarser. 
 It will generally be observed that this phenomenon almost invariably is 
 exhibited by those kinds of recovered rubber which pass the vulcanising 
 test badly, or which, in other words, have been inefficiently 'recovered.' It 
 is therefore generally advisable to put samples of recovered rubber to a 
 mixing test, and examine the appearance of a surface cut produced with a 
 sharp knife. A granular, spotted, or streaky section indicates a more or 
 less heterogeneous mixture. Very characteristic for the recognition of this 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 155 
 
 defect is also the appearance of such a section after wetting it with one of 
 the india rubber solvents : a section of a homogeneous mixture swells into 
 a smooth surface, a section of a heterogeneous mixture becomes granular, 
 or irregularly furrowed and indented. 
 
 2. Fatty Substitutes. The fatty substitutes discussed in the earlier 
 part of this section, which are offered in a great number of variations, are 
 commercially divided into two classes viz., brown and white substitute 
 and to this division corresponds also an essential difference in their com- 
 position, colour, mode of preparation, and manner of employment. 
 
 The brown substitutes are prepared by the action of sulphur upon the 
 triglycerides of unsaturated fatty acids at high temperatures, the white 
 substitutes by the action of sulphur monochloride upon similar triglycerides 
 in the cold. Accordingly, the brown substitutes represent sulphides, the 
 white substitutes chlorosulphides of fatty oils. 
 
 The value of these substitutes lies in their chemical stability and their 
 insolubility in solvents, which renders it possible to use them in more or 
 less considerable quantities in rubber mixings as bulking agents, without 
 increasing the specific gravity or reducing the suppleness of the manu- 
 factured articles. 
 
 The solubility of these substitutes can be ascertained by experiment, 
 and is carried out by extraction with ether or acetone. The extract, after 
 distilling off the ether, is dried for one hour at 100 C. It must, however, 
 not be assumed that this extract, as a whole, represents dissolved substitute. 
 Indeed, this is only true of a very small portion of it. The bulk of this 
 extract consists of triglycerides which have not taken part in the reaction 
 with either sulphur, or chloride of sulphur, and besides this, it always con- 
 tains a sensible quantity of free sulphur. 
 
 It is obvious that the smaller the quantity of this extract, the less is a 
 substitute likely to cause trouble in the manufacture. THOMSON l pointed 
 out that the oily extracts obtained from rubber substitute do not affect 
 india rubber detrimentally. This, no doubt, is to a large extent correct, 
 as the oils which escape the reactions leading to these rubber substitutes 
 are bodies which have no perceptible action upon india rubber. Moreover, 
 their presence in the substitutes cannot altogether be avoided. But this 
 point, nevertheless, should be made the subject of careful control in manu- 
 facturing, first with a view of keeping the percentage of soluble fatty 
 matter in the substitute as low as possible, and further to keep this per- 
 centage as near as possible constant. That for certain purposes fatty 
 substances are designedly added to india rubber mixings does not minimise 
 the importance of this control, one of the chief purposes of which is to 
 assist towards that uniformity of working conditions, without which 
 serious break-downs in the uniformity of the manufactured goods are 
 unavoidable. 
 
 All these substitutes contain, as has already been stated, free sulphur, 
 1 Journ. Soe. Chem. Ind., 1885, p. 710. 
 
156 INDIA RUBBER AND ITS ANALYSIS. 
 
 the presence of which in white substitutes is always rather small and 
 practically irrelevant. Not so in brown substitutes, which not only 
 reaches often rather high figures, 8 per cent, and more, but is, moreover, 
 apt to vary considerably. As brown substitutes are used exclusively for 
 dry heat cured goods, the knowledge of the exact amount of free sulphur 
 in the substitute becomes an important matter, and the neglect of this 
 obvious precaution is a fruitful cause of many vulcanising troubles. 
 
 The determination of the free sulphur is best carried out in conjunction 
 with the determination of the soluble fatty matter. The extraction of the 
 substitute may be carried out with ether, but acetone is to be preferred as 
 being a better solvent for sulphur. Carbon bisulphide might perhaps 
 appear the most suitable solvent to employ, but, as a matter of fact, it can- 
 not be used, as it always dissolves from the substitute a small quantity of 
 organic matter which cannot be readily separated from the free sulphur. 
 Accordingly we proceed in this manner, that from 3 to 5 grms. of the 
 finely ground substitute are weighed out in one of the paper tubes, closed 
 at one end, which are much used in milk analysis. The open end of this 
 tube is then lightly closed by inserting into it a small tuft of fat-free 
 cotton wool, transferred to a Soxhlet extractor, and extracted for at least 
 four hours with acetone, the extract being received in a weighed flask. 
 After extraction the acetone is completely distilled off. The extract thus 
 obtained is now treated with 20 c.c. of acetone to redissolve the organic 
 matter when the free sulphur separates out as a crystalline mass. The 
 fatty matter being completely dissolved, the solution is filtered through a 
 small filter into another weighed flask, care being taken to retain the 
 precipitated sulphur as much as possible in the first flask. The latter is 
 then washed three times with 5 c.c. of cold acetone, all the washings being 
 successively filtered into the second flask. Finally, the small filter is 
 carefully washed with another 5 c.c. of acetone. The filter is then 
 removed to the first flask, and 10 to 15 c.c. of carbon bisulphide are 
 filtered through it into the first flask. Any sulphur which was poured 
 into this filter in the first washings will thus be dissolved. The contents 
 of both flasks are now evaporated to dryness, and dried for one hour at 
 100 C. If the operation has been properly conducted, the first flask will 
 be found to contain a number of globules of pure sulphur, 1 the second a 
 clear, semi-fluid, fatty substance. The amount of each is determined by 
 weighing. 
 
 The figures so obtained are satisfactorily accurate for ordinary purposes, 
 but occasionally it may appear desirable to apply to them a correction on 
 account of the solubility of sulphur in cold acetone, which amounts to 
 0'06 per cent, at a temperature of 18 C. Accordingly, the 40 c.c. of 
 acetone used in separating the sulphur and fatty matter would dissolve 
 0'020 grms. of sulphur. To correct the figures we would therefore have 
 
 1 These remain liquid for a long time, but crystallise instantly on being touched with 
 a glass rod. 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 157 
 
 to deduct this amount L from the amount of fatty matter present, and add 
 it to the quantity of sulphur as ascertained by weight. 
 
 These figures already afford a very satisfactory criterion by which to 
 judge the identity of the quality of successive deliveries. In doubtful 
 cases, however, or when it becomes necessary to decide as to the identity 
 or otherwise of samples from different sources, the determination of the 
 combined sulphur, or of combined sulphur and chlorine, in the case of 
 white substitutes should not be omitted. To employ for this purpose the 
 substitute in its ordinary condition i.e., to determine the total sulphur in 
 it, and then put the combined sulphur as the difference between this total 
 sulphur and the amount of free sulphur previously ascertained is, however, 
 not permissible. The reason for this objection lies in the fact that the 
 fatty extract always also contains a certain amount of combined sulphur 
 which cannot properly be considered to represent sulphur present in the 
 form of insoluble substitute. For this reason the combined sulphur should 
 be determined in the extracted sample of substitute after drying. 
 
 For this purpose 1 grm. of the extracted substitute is treated according 
 to HENRIQUBS' method for the determination of the total sulphur in india 
 rubber fully described on page 253. The amount of free sulphur should be 
 calculated, not upon the extracted, but upon the unextracted, substitute. 
 
 The above data soluble fatty matter, free sulphur, combined sulphur 
 suffice, in the vast majority of cases, to decide the question of conformity 
 to standard of samples of substitute, white or brown, and also to 
 judge their quality in general as far as chemical analysis can decide 
 this point. Regarding this, it may be taken that a white substitute 
 containing not more than 10 per cent, of soluble fatty matter and not less 
 than 8 per cent, of combined sulphur, or a brown (black) substitute 
 containing not more than 5 per cent, of soluble fatty matter, and the 
 amount of free sulphur in which does not exceed 40 per cent, of the 
 amount of the combined sulphur, are, from the chemical point of view, safe 
 to use. It should, however, always be remembered that these figures 
 represent higher limits at any rate, as far as the percentages of soluble 
 fatty matter are concerned. 
 
 There are also, however, some technological points in connection with 
 these substitutes of which the analyst should take note. It is quite 
 feasible that a sample of substitute may in every respect come up to our 
 analytical standard and nevertheless prove unsatisfactory with regard to 
 the mixture it forms with the india rubber between the mixing rollers. 
 This mixture should be so fine, that on pulling part of it into a transparent 
 skin, no appearance of structure, due to microscopic particles or flakes 
 of substitute, should be observable. The skin should have the appear- 
 ance of an opalescent or turbid medium. Under the microscope a thin 
 film should present the image of fig. 8, and not of fig. 9. And this result 
 
 1 It is scarcely necessary to point out that it would be useless to carry this correction 
 beyond the first decimal. 
 
158 INDIA RUBBER AND ITS ANALYSIS. 
 
 should be arrived at without the mixture requiring an amount of working 
 detrimental to the india rubber. 
 
 The microscopic preparations of such mixings are obtained by dissolving 
 a small quantity of the latter in sufficient benzene to produce a solution of 
 the consistency of thin honey. A very small drop of this solution is placed 
 upon one end of a micro slide, and by drawing another micro slide rapidly 
 across this drop, an extremely thin rubber-substitute film is produced, 
 which, after a few moments' drying, is ready for observation. In permanent 
 preparations no mounting media should be used, or only such as possess a 
 very low refractive index. 
 
 It is further advisable to test the behaviour of both white and brown 
 substitutes against high pressure steam at vulcanising temperatures applied 
 for three hours. Substitutes of apparently closely similar chemical 
 composition not infrequently behave very differently in this test, some 
 standing it very satisfactorily, others suffering more or less pronounced 
 decomposition. 
 
 Substitutes, whether white or brown, manufactured from oxidised 
 (blown) oils contain considerably less chlorine and sulphur, or sulphur, 
 than substitutes manufactured from the natural oils. Thus, the brown 
 substitutes I. and II. quoted on page 144 have been manufactured from 
 blown colza oil, and No. III. from ordinary refined maize oil. Likewise, 
 the white substitutes A and B in HENRIQUES' table (page 147) have been 
 manufactured from blown or oxidised oils, and C from a natural oil. But 
 while brown substitutes produced from blown oils are at least as satis- 
 factory as those obtained from natural oils, and for some purposes distinctly 
 superior, the same is not true for white substitutes prepared from blown 
 oils. These exhibit a distinct lack of stability as compared with the 
 normal products. 
 
 3. Resins, Pitches, and Waxes. The analysis of these substances is a 
 very simple operation. 
 
 The resins which are employed in the manufacture of india rubber 
 goods are very few. The most important are colophony and shellac. The 
 low price of colophony renders an adulteration of it a practical impossibility. 
 Shellac, owing to its much higher price, is much more liable to be tampered 
 with. It is stated that adulterations of this resin with colophony have 
 been observed. These can readily be discovered by the solubility of pure 
 shellac in ether, in which colophony is practically insoluble. 
 
 The pitches used in the rubber manufacture are coal tar pitch, stearine 
 pitch, and asphaltum. Coal tar pitch is prepared by boiling the liquid 
 coal tar until it has reached such a consistency that it sets just hard, but 
 not brittle, on cooling. At least 90 per cent, of it should be soluble in 
 boiling benzene. 
 
 Stearine pitch, the origin of which is sufficiently indicated by its name, 
 varies very greatly in its composition and appearance. It occurs some- 
 times as a very sticky, almost treacly mass ; sometimes it forms hard, 
 
EXAMINATION OF INDIA RUBBER SUBSTITUTES. 159 
 
 brittle pieces ; but the only form in which it is satisfactory to use is in the 
 form of a coherent semiplastic mass. It is by no means a homogeneous 
 body, but consists of a part soluble in ether, acetone, and benzene, and of 
 another part which is insoluble in all solvents. The soluble part consists 
 practically entirely of fluorescent hydrocarbons, and should contain no 
 saponifiable matter. The insoluble part is a black carbonaceous mass. Its 
 amount should not exceed 20 per cent., but often forms as much as 60 per 
 cent, of the product. 
 
 Asplialtum is not extensively used, and the products in the market 
 under this name differ very greatly in composition and properties. The 
 quality generally used in rubber works is that known as Syrian or 
 Judea Asphaltum. Only from 4 to 5 per cent, of it should be soluble in 
 alcohol. From 40 to 50 per cent, should be soluble in ether, and the 
 whole in chloroform, turpentine, and nitrobenzene. 
 
 Practically the only waxes used in connection with india rubber are 
 paraffin wax, and ceresin or ozokerite. The quality of these products is 
 expressed by their melting points, which for paraffin wax ranges from 
 about 28 to 53 C., and for ceresin from 70 to 76 C. Adulterations of 
 paraffin wax are scarcely feasible, but adulterations of ceresin with resin 
 (colophony) are by no means uncommon. This is, however, easily detected 
 by the solubility of resin in 60 per cent, strong alcohol, in which ceresin is 
 entirely insoluble. The quantitative determination of this adulteration is 
 carried out by boiling 2 grms. of the ceresin with 20 c.c. of alcoholic caustic 
 soda (6 per cent.) for twenty minutes on a reflux condenser. The ceresin 
 hydrocarbon is then allowed to solidify by cooling, when the alcoholic 
 solution is poured into a dish, the flask with the ceresin rinsed several 
 times with small quantities of alcohol, and the washings added to the 
 alcoholic solution. This solution is then evaporated to dryness, the residue 
 again taken up with boiling water, slightly acidulated with hydrochloric 
 acid, and again evaporated to dryness. The free resin is now separated 
 from the sodium chloride present by dissolving it in alcohol, evaporating 
 to dryness, and weighing the residue. Or the weight of the first obtained 
 mixture of resin and sodium chloride is ascertained, the whole then ignited, 
 and the weight of the residual sodium chloride determined. The calcula- 
 tion of the amount of resin present is obvious. 
 
CHAPTER IV. 
 INORGANIC COMPOUNDING MATERIALS. 
 
 THE inorganic compounds used in india rubber mixings serve a number 
 of purposes, and their suitability must be considered, therefore, from the 
 point of view of the particular effect aimed at by their employment. 
 
 Foremost stand the compounds used as filling or bulking agents, the 
 chief purpose of their employment being to reduce the price of the manu- 
 factured article. Next in importance stand the substances acting as 
 * sulphur carriers,' and the effect of which is either to hasten, or else to 
 modify, the vulcanisation in certain ways. Lastly, there are a number of 
 compounds which are employed to give the manufactured article some 
 definite colour. It must not, however, be assumed that there exists a sharp 
 line of division between the compounds of one group and those of the other 
 two. Most of the filling agents more or less distinctly influence the vul- 
 canisation ; some of the sulphur carriers might perhaps be employed as much 
 for bulk-giving purposes as on account of their colouring properties. The 
 above division into three groups is therefore in some cases somewhat 
 arbitrary, but this objection is scarcely of sufficient weight to sacrifice to 
 it the obvious advantages of the classification. In the present section we 
 shall only deal with those inorganic compounds which are chiefly used as 
 bulking or filling agents. 
 
 It is almost needless to say that all filling materials to be used in india 
 rubber should be chemically absolutely inert substances ; they should 
 not be appreciably affected by temperatures of from 180 to 200 C., and 
 their solubility in water should be very slight practically nil. That they 
 should be free from impurities deleterious to india rubber goes without 
 saying, but any danger in this respect is very remote. This makes it 
 evident that the chemical points in relation to inorganic compounding 
 materials are of a very simple description, calling rarely, if ever, for more 
 than the most elementary chemical tests. The selection of the material to 
 be employed for some particular mixing is, indeed, in the first instance more 
 a matter of certain physical considerations than of chemical ones in other 
 words, the inorganic material to be * compounded ' with the india rubber is, 
 in the vast majority of cases, selected simply on account of the physical 
 
INORGANIC COMPOUNDING MATERIALS. 
 
 161 
 
 properties it confers upon the manufactured article. This is exactly as we 
 should expect, considering that the vast majority of india rubber goods are 
 subjected to physical wear only ; but it is, at the same time, obvious that in 
 the manufacture of india rubber goods intended for use in chemical 
 operations, the selection of the particular inorganic compounding material 
 to be employed must primarily be influenced by the nature of the chemical 
 influences to which the goods would be subjected when in use, and only 
 such compounding materials should be used which behave as entirely 
 indifferent substances under the chemical influences in question. 
 
 Perhaps the most obvious effect of mineral ' compounding materials ' in 
 india rubber is their influence upon the specific gravity of the manufactured 
 article. Indeed, this is so pronounced that the specific gravity of such an 
 article is in certain circles still looked upon as a reliable criterion by which 
 to judge its quality. Of course, I need hardly point out that the extensive 
 employment of * substitutes ' has rendered this criterion largely illusory. 
 Nevertheless, a high specific gravity is scarcely, if ever, looked upon as a 
 recommendation in manufactured india rubber, and for this reason the 
 choice of a particular compounding material is often dictated by the con- 
 sideration of this point. Since, however, the addition of these inorganic 
 bodies is chiefly made with a view of cheapening the manufactured product 
 in other words, to gain weight it should be pointed out that the dete- 
 rioration of quality caused in manufactured rubber by such addition is 
 not merely a matter of their absolute weight, but also of their bulk. The 
 greater this latter, the greater the deterioration of quality in the product, 
 so that on this account it would appear preferable to produce a desired 
 increase in weight by using a compounding material of high specific gravity 
 and small bulk, rather than one of low gravity and therefore large bulk, 
 although it could not be said that this is the general practice in rubber 
 factories. It must, however, be admitted that in cases where the article to 
 be manufactured is intended to resist shocks (stability) rather than tension, 
 a compounding material of smaller specific gravity is often to be preferred. 
 
 The specific gravities of the more generally used inorganic compounding 
 materials are as follows : 
 
 Material. 
 
 Whiting, CaC0 3 , . 
 Plaster of Paris, CaS0 4 , 
 Gypsum, CaS0 4 ,2H 2 0, . 
 Barytes, BaS0 4 , . 
 Magnesia, MgO, . 
 Magnesium carbonate, MgC0 3 , 
 Zinc oxide, ZnO, . 
 Asbestos (Magnesium Calcium 
 silicate), . . . ,. : 
 
 Specific 
 Ore --"- 
 
 27-2-9 
 
 2'9 
 
 3-2 
 
 4-3-4-6 
 3-2-3-6 
 2-8-31 
 
 5-6 
 
 Material. 
 
 Litharge, PbO, . 
 White lead, 3PbO. 2C0 2 . H 2 0, 
 Lead sulphate, PbS0 4 , 
 Silica, Si0 2 ( Infusorial earth), 
 Lithopone, ZnS + BaS0 4 , . 
 French chalk, Mg 3 Si 4 12 H 2 , 
 China clay, Al 2 SijO^H 4 , . 
 
 Specific 
 Gravity. 
 
 9-3 
 
 6-1-6-2 
 
 6-1-6-2 
 
 1-66-1-95 
 
 3-6-4-1 
 
 It should, however, be pointed out that the physical constants shown by 
 manufactured india rubber containing any of the above materials are by no 
 
 11 
 
1 62 INDIA RUBBER AND ITS ANALYSIS. 
 
 means simply a function of the specific gravity of the latter. This, indeed, 
 is only true of the specific gravity of the manufactured article. All the 
 other constants vary in a manner which bears no relation to the specific 
 gravity of the inorganic compounding materials used, and the cause of these 
 variations is entirely unknown. It is most probably closely connected 
 with the physical state or structure of these compounding materials. 
 
 This matter is one of very considerable importance. A most valuable 
 contribution on this point we owe to HEINZBRLING and PAHL. 1 These 
 authors prepared a large number of rubber mixings with a considerable 
 variety of inorganic and organic admixtures, and submitted these samples 
 after vulcanisation to a series of chemical tests, the tabulated results of 
 which are reproduced in extenso below. 
 
 All the samples were subjected to the following tests : 
 
 Sulphuric Acid. The weighed samples were immersed in sulphuric acid 
 of sp. gr. 1*1562 at ordinary temperature. After immersion they were 
 dried, weighed again, and the changes which had taken place during 
 immersion noted. For samples I. to XVIII., the time of immersion was 
 20 days ; for samples XIX. to XXXII., 10 days. 
 
 Acetic Acid. An acid of sp. gr. 1*0584 was used. The immersion was 
 only continued for 3 days, the action of this acid being much more rapid. 
 
 Caustic Soda. A solution of sp. gr. 1'3084 was employed and allowed 
 to act during 3 days. The weights and specific gravities of the samples 
 before and after treatment were observed, as also in the case of the acetic 
 acid and ammonia. 
 
 Ammonia. A solution of sp. gr. 0'9775 was allowed to act during 
 4 days. 
 
 Samples I. to XVIII. were washed after treatment with the above 
 solutions, and dried for 6 hours at 100 C. ; samples XIX. to 
 XXXII. were, without washing, exposed to the air for 48 hours, and 
 subsequently dried at 100 C. for 30 minutes. 
 
 Colza Oil of sp. gr. 0'9102 was used, and the samples immersed in it 
 during 5 days. In these, as also in the two following tests, the specific 
 gravities of the samples before and after immersion were determined. 
 
 Mineral Oil. A lubricating oil of sp. gr. 0'8991 was employed. Time 
 of immersion 5 days. 
 
 A mixture of 90 per cent, of mineral oil and 10 per cent, of tallow was 
 allowed to act upon the samples during 5 days at 100 C. 
 
 Verhandl. d. Vereins z, B(ford. d. Gewerbefl., Berlin, 1891-92. 
 
INORGANIC COMPOUNDING MATERIALS. 
 
 To test the action of illuminating gas upon the samples, the latter were 
 during 20 days inserted in a tube through which the gas was passing to a 
 burner. 
 
 Changes of volume were determined by means of a micrometer capable 
 of indicating O'Ol mm. 
 
 The tensile strength was determined by means of a machine similar to 
 those used for the determination of the strength of textile fabrics. The 
 samples used for this purpose were 2 cm. long, 0*5 cm. wide, and 2 mm. in 
 thickness. 
 
 The changes of form were determined by means of a machine similar 
 to those used for making crushing tests on building materials, and the 
 samples were subjected to a compression of 4'9 kilos per square centimetre 
 during one minute. The change in thickness of the samples under this 
 test was determined by means of the micrometer. 
 
 The changes of form under the influence of repeated shocks were 
 determined by subjecting the samples to 50 blows of a hammer of 2 kgms. 
 weight, falling each time through a height of 0*25 m. 
 
 The nature of the deformations caused by steady pressure and by 
 hammering respectively may be gleaned from the following figures A 
 (crashed) and B (hammered) : 
 
 FIG. A. 
 
 FIG. B. 
 
 The tests of the behaviour of the samples against heat were carried out 
 in an oven provided with a glass door, so that they were under observation 
 all the time. The temperature was raised 10 C. every half hour until 
 150 C. had been reached. 
 
 The electric insulation resistance was ascertained by means of WEBER'S 
 electrometer. The samples were all heated for half an hour before being 
 tested. 1 
 
 1 The above particulars of tests, as well as the following tables and conclusions, have 
 been translated from SEELIGMANN'S French translation of the German original, which, in 
 spite of many efforts, was inaccessible to me. For this reason I can only vouch for the 
 conformity of my own with Seeligmann's translation. 
 
 [TABLES. 
 
164 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 Chemical Tests. 
 
 A. Samples prepared to Author's Instructions. 
 
 
 
 Action of C 2 H 4 2 . 
 
 
 3 
 
 Action of 
 
 Modifications per cent. 
 
 
 | 
 
 Modifica- 
 
 
 
 
 
 
 
 
 
 
 ?, 
 
 tions in 
 
 
 
 - if 
 
 Sulphur 
 per cent 
 
 Quantity and Nature of 
 Admixtures. 
 
 it 
 
 8,1 
 
 QfJ 
 
 1 
 
 Weights 
 per cent. 
 
 In 
 Volumes. 
 
 In 
 Weights. 
 
 I. 90 
 
 10 
 
 
 
 
 0-999 
 
 
 -4-259 
 
 + 2-912 
 
 + 0-932 
 
 II. 85 
 
 5 
 
 10 
 
 Golden sulphide, . 
 
 1-070 
 
 o 
 
 - 3-099 
 
 + 2-985 
 
 + 2-097 
 
 III. 75 
 
 5 
 
 20 
 
 Golden sulphide, . 
 
 1-101 
 
 2 
 
 -3-521 
 
 + 3-846 
 
 + 3-048 
 
 IV. 80 
 
 10 
 
 10 
 
 Zinc oxide, . 
 
 1-087 
 
 | 
 
 -3-274 
 
 + 4-854 
 
 + 29-132 
 
 V. 50 
 
 10 
 
 40 
 
 Zinc oxide, . 
 
 1-283 
 
 | 
 
 -2-074 
 
 + 28-155 
 
 + 153-035 
 
 VI. 10 
 
 10 
 
 80 
 
 Zinc oxide, . 
 
 1-514 
 
 ^ 
 
 -1-810 
 
 + 27-868 
 
 + 116-950 
 
 VII. 80 
 
 10 
 
 \l 
 
 Zinc oxide, . 
 Chalk, . 
 
 1-184 
 
 ee 
 
 TJ 
 
 e 
 
 -2-930 
 
 + 5-357 
 
 + 30-887 
 
 VIII. 
 
 50 
 
 10 
 
 (20 
 1 20 
 
 Zinc oxide, . 
 Chalk, 
 
 1-255 
 
 o3 
 
 o 
 
 -2-280 
 
 + 28-828 
 
 + 131-899 
 
 IX. 
 
 10 
 
 10 
 
 /40 
 140 
 
 Zinc oxide, . 
 Chalk, . 
 
 1-452 
 
 00 
 
 CO 
 
 - 1 -802 
 
 + 29-464 
 
 + 160-097 
 
 X. 
 
 50 
 
 10 
 
 (20 
 \20 
 
 Zinc oxide, . 
 Fluor spar, . 
 
 1-295 
 
 3 
 vn 
 
 -1-464 
 
 + 17-475 
 
 + 62*616 
 
 XI. 
 
 10 
 
 10 
 
 f40 
 1 40 
 
 Zinc oxide, . . \ 
 Fluor spar, . . j 
 
 1-502 
 
 i 1 
 
 1 
 
 -1-527 
 
 + 31-730 
 
 + 124-539 
 
 XII. 
 
 50 
 
 10 
 
 J 20 
 1-20 
 
 Zinc oxide, . . \ 
 Litharge, . . / 
 
 1-340 
 
 <*-. CO 
 
 <c^ 
 
 -1-864 
 
 + 12765 
 
 + 78-542 
 
 XIII. 
 XV. 
 
 10 
 50 
 
 10 
 10 
 
 J40 
 \40 
 40 
 
 Zinc oxide, . . \ 
 Litharge, . . J" 
 Brown substitute, . 
 
 1-569 
 1-053 
 
 
 
 5 * 
 
 rt co 
 
 i* O 
 
 -1-406 
 -4-081 
 
 + 23-711 
 n.c. 
 
 + 77-411 
 + 0-560 
 
 XVI. 
 
 30 
 
 10 
 
 60 
 
 Brown substitute, . 
 
 1-052 
 
 |o| -2-473 
 
 n.c. 
 
 + 2-180 
 
 XVII. 
 
 50 
 
 10 
 
 40 
 
 White substitute, . 
 
 1-054 
 
 ^^ -6-076 
 
 + 1-709 
 
 + 6-373 
 
 XVIII. 
 
 30 
 
 10 
 
 60 
 
 White substitute, . 
 
 1-031 
 
 21 
 
 -6-306 
 
 + 2-970 
 
 + 8-695 
 
 XIX. 
 
 87 
 
 10 
 
 3 
 
 Magnesia, 
 
 1-130 
 
 * 
 
 -0-932 
 
 + 4-231 
 
 + 18-718 
 
 XX. 
 
 87 
 
 10 
 
 3 
 
 Slaked lime, . 
 
 1-027 
 
 
 -0-924 
 
 + 4-273 
 
 + 6-297 
 
 XXI. 
 
 75 
 
 10 
 
 15 
 
 Paraffin wax, 
 
 1-075 
 
 2 
 
 -4-806 
 
 + 5-063 
 
 - 0-354 
 
 XXII. 
 
 65 
 
 10 
 
 25 
 
 Paraffin wax, 
 
 1-041 
 
 (H 
 
 2 
 
 -3-273 
 
 + 2-197 
 
 - 0-674 
 
 XXV. 
 XXVI. 
 
 75 
 65 
 
 10 
 10 
 
 15 
 25 
 
 Turpentine, . 
 Turpentine, . 
 
 1-030 
 1-065 
 
 1 
 
 -3-367 
 -3-582 
 
 n.c. 
 n.c. 
 
 + 2-732 
 + 6-442 
 
 XXVII. 
 
 85 
 
 10 
 
 5 
 
 Glycerine, 
 
 1-062 
 
 
 
 -5-921 
 
 n.c. 
 
 - 0-420 
 
 XXVIII. 
 
 80 
 
 10 
 
 10 
 
 Glycerine, 
 
 1-060 
 
 
 
 S> 
 
 -6-571 
 
 n.c. 
 
 + 2-400 
 
 XXIX. 
 
 80 
 
 10 
 
 10 
 
 Bitumen, 
 
 1-080 
 
 f 
 
 & 
 
 -3-686 
 
 + 1-980 
 
 + 0-650 
 
 XXX. 
 
 70 
 
 10 
 
 20 
 
 Bitumen, 
 
 1-085 
 
 | 
 
 -2-421 
 
 n.c. 
 
 - 0-607 
 
 XXXI. 
 
 60 
 
 10 
 
 30 
 
 Ground rubber waste, 
 
 1-101 
 
 
 -2-336 
 
 n.c. 
 
 + 0-332 
 
 XXXII. 
 
 30 
 
 10 
 
 60 
 
 Ground rubber waste, 
 
 1-101 
 
 a 
 <n 
 
 -3-889 
 
 + 1-769 
 
 + 0-729 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 0> 
 
 
 
 
 
 aSS 
 
 
 
 
 
 ,} 
 43 
 
 
 
 
 lA. 
 
 r r -S? 
 
 5 
 
 
 
 0-925 
 
 H 
 
 -1-344 
 
 n.c. 
 
 n.c. 
 
 
 
 
 
 
 
 <1 
 
 
 
 
 
 " d8 
 
 
 
 
 
 
 
 
 
 n.c. No change. 
 
INORGANIC COMPOUNDING MATERIALS. 
 
 165 
 
 Chemical Tests. 
 
 
 No. 
 
 Action of NaOH. 
 
 Modifications 
 per cent. 
 
 Action of NH 3 . 
 
 Modifications 
 per cent. 
 
 Action of Colza Oil. 
 Modifications in 
 Volumes per cent. 
 
 "tj 
 
 jjjjl 
 
 + 53-846 
 + 34-615 
 + 21-782 
 
 + 58-878 
 
 Action of Mineral Oils. 
 Modifications in 
 Volumes per cent. 
 
 Action of Coal Gas. 
 Modifications in 
 Weights per cent. 
 
 In 
 
 Volumes. 
 
 In 
 
 Weights. 
 
 In In 
 Volumes. I Weights. 
 
 I. 
 II. 
 III. 
 IV. 
 V. 
 
 n.c. 
 n.c. 
 n.c. 
 n.c. 
 n.c. 
 
 -4-354 
 -4-354 
 -4-271 
 -2-871 
 -0-084 
 
 n.c. 
 n.c. 
 n.c. 
 n.c. 
 n.c. 
 
 -0727 
 -0-974 
 -0-896 
 -0-326 
 -0-314 
 
 + 23-214 
 + 20-617 
 + 28-431 
 + 25-242 
 + 32-110 
 
 + 42-718 
 + 39-694 
 + 51-485 
 + 39-423 
 + 43-519 
 
 + 9765 
 + 9-156 
 + 7-561 
 + 11-079 
 + 10-085 
 
 VI. 
 
 n.c. 
 
 -0-087 
 
 n.c. -0-779 
 
 + 47-272 +73-947 
 
 + 36-440 
 
 + 7-674 
 
 
 VII. 
 
 n.c. 
 
 -2-824 
 
 n.c. 
 
 -0-629 
 
 + 24-137 +65-765 
 
 + 40-178 
 
 + 10-863 
 
 
 VIII. 
 
 n.c. 
 
 -2-059 
 
 n.c. 
 
 -0-856 
 
 + 37-247 
 
 + 77-570 
 
 + 46-296 
 
 + 7-370 
 
 
 IX. 
 
 - 0-892 
 
 -1-422 
 
 n.c. 
 
 -0-707 
 
 + 27-193 
 
 + 65-853 
 
 + 35-593 
 
 + 5-821 
 
 
 X. 
 
 - i-ooo 
 
 -3-348 
 
 n.c. 
 
 -0-455 
 
 + 28-853 
 
 + 51-923 
 
 + 40-384 
 
 + 8-549 
 
 XI. 
 
 n.c. 
 
 -2-160 
 
 n.c. -0-512 
 
 + 28-571 
 
 + 65-048 
 
 + 41-121 
 
 + 1-452 
 
 
 XII. 
 
 n.c. 
 
 -1-158 
 
 n.c. -O'Oll 
 
 + 37-894 
 
 + 53-763 
 
 + 38-613 
 
 + 10-043 
 
 
 XIII. 
 
 n.c. 
 
 -0-282 
 
 71. C. 
 
 -0-388 
 
 + 28-712 
 
 + 57-916 
 
 + 30000 
 
 + 8-695 
 
 * 
 
 XV. 
 XVI. 
 
 - 1-388 
 n.c. 
 
 -5-745 
 -4-965 
 
 n.c. 
 - 3-571 
 
 -0-787 
 -0744 
 
 + 21-428 
 + 20-339 
 
 * 
 
 + 50714 
 + 32-773 
 
 + 11-793 
 + 14-047 
 
 XVII. 
 XVIII. 
 XIX. 
 XX. 
 XXI. 
 
 - 7-017 
 - 9-708 
 n.c. 
 - 0-847 
 -10-588 
 
 -9-723 
 -9-658 
 -1-321 
 -6-569 
 -5-602 
 
 -1-724 
 -1-941 
 n.c. 
 n.c. 
 n. c. 
 
 -2-147 
 -1-649 
 -0-247 
 -0-327 
 -1-722 
 
 + 11-111 
 + 14-423 
 + 6-250 
 + 18-487 
 + 23-595 
 
 * 
 * 
 
 + 44-545 
 + 55-445 
 + 36-440 
 + 29752 
 + 43-750 
 
 + 7-574 
 + 7-987 
 + 10-143 
 + 10-043 
 + 12758 
 
 XXII. 
 XXV. 
 XXVI. 
 XXVII. 
 XXVIII. 
 
 - 7-692 
 n.c. 
 n.c. 
 - 5-940 
 - 3-846 
 
 -4-847 
 -5-210 
 -5-961 
 -7-628 
 -7-992 
 
 n.c. 
 n.c. 
 n.c. 
 n.c. 
 n.c. 
 
 -2-882 
 + 0-239 
 + 0-496 
 + 0-309 
 + 3-909 
 
 + 10-000 
 
 + 20-491 
 + 26-732 
 + 23-904 
 + 21-551 
 
 * 
 * 
 * 
 
 * 
 
 + 35-781 
 + 39-603 
 + 31-460 
 + 48-214 
 + 45-794 
 
 + 13-581 
 + 10-363 
 + 11-382 
 + 11-618 
 + 8-905 
 
 XXIX. 
 XXX. 
 XXXI. 
 XXXII. 
 
 n.c. 
 n.c. 
 - 2-849 
 - 0-833 
 
 -4-382 
 -3-425 
 n.c. 
 -3-473 
 
 n.c. 
 n.c. 
 n.c. 
 + 0-869 
 
 + 0-121 
 + 0-225 
 + 0-718 
 -2-031 
 
 + 10-434 
 + 17-647 
 + 14-150 
 + 15781 
 
 + 73-684 
 + 75757 
 
 * 
 
 + 45-714 
 + 40740 
 + 53-846 
 + 47-000 
 
 + 12-514 
 + 11-248 
 + 10-810 
 + 5-807 
 
 
 lA. 
 
 n.c. 
 
 n.c. 
 
 n.c. 
 
 + 1-592 
 
 + 28-461 
 
 * 
 
 + 54-198 
 
 + 15-121 
 
 n.c. No change. 
 
 These samples dissolved entirely. 
 
1 66 
 
 INDIA KUBBER AND ITS ANALYSIS. 
 
 Physical Tests. 
 
 
 j 
 
 
 A. Samples prepared to Author's Instructions. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 .2 
 ts 
 
 of 
 Pressure 
 
 Modulus 
 of Resist- 
 
 bility 
 in mm. 
 
 
 No. 
 
 *l 
 
 I! 
 
 Quantity and Nature of 
 Admixtures. 
 
 I! 
 
 1 
 
 per 
 I 2 mm. 
 
 ance per 
 I 2 mm. 
 
 per 
 
 100 mm. 
 
 
 I. 
 
 II. 
 
 90 
 
 85 
 
 10 
 5 
 
 
 10 
 
 Golden sulphide, 
 
 0-999 
 1-070 
 
 
 grins. 
 48-5 
 38-0 
 
 grms. 
 213 
 213 
 
 1100 
 790 
 
 
 III. 
 
 75 
 
 5 
 
 20 
 
 Golden sulphide, . 
 
 1-101 
 
 
 47-5 
 
 259 
 
 850 
 
 
 IV. 
 
 80 
 
 10 
 
 10 
 
 Zinc oxide, 
 
 1-087 
 
 
 48-5 
 
 427 
 
 825 
 
 
 V. 
 
 50 
 
 10 
 
 40 
 
 Zinc oxide, 
 
 1-283 
 
 2 
 g 
 
 47-5 
 
 480 
 
 725 
 
 
 VI. 
 
 10 
 
 10 
 
 80 
 
 Zinc oxide, 
 
 1-514 
 
 E 
 
 43-5 
 
 543 
 
 660 
 
 
 VII. 
 
 80 
 
 Hi i 
 
 Zinc oxide, . . 1 
 Chalk, 
 
 1-184 
 
 p. 
 
 03 
 
 46-0 
 
 440 
 
 760 
 
 
 VIII. 
 
 50 
 
 MS 
 
 Zinc oxide, 
 Chalk, . 
 
 1-225 
 
 a 
 & 
 
 65-0 
 
 542 
 
 710 
 
 
 IX. 
 
 10 
 
 10 IIS 
 
 Zinc oxide, 
 Chalk, . 
 
 1-452 
 
 C3 
 Q 
 
 69-0 
 
 653 
 
 665 
 
 
 X. 
 
 50 
 
 10 {20 
 
 Zinc oxide, 
 Fluor spar, 
 
 1-295 
 
 i l 
 
 48-0 
 
 404 
 
 660 
 
 
 XL 
 
 10 
 
 10 
 
 J40 
 
 140 
 
 Zinc oxide, . . \ 
 Fluor spar, . . J 
 
 1-502 
 
 a 
 
 iO 
 CO 
 
 77-5 
 
 563 
 
 640 
 
 
 XII. 
 
 50 10 
 
 20 
 (20 
 
 Zinc oxide, . . \ 
 Litharge, . . / 
 
 1-340 
 
 g 
 
 31-5 
 
 404 
 
 490 
 
 
 XIII. 
 
 10 10 
 
 J40 
 1 40 
 
 Zinc oxide, 
 Litharge, 
 
 1-569 
 
 & 
 
 31-0 
 
 352 
 
 305 
 
 
 XV. 
 
 50 
 
 10 
 
 40 
 
 Brown substitute, . 
 
 1-053 
 
 2 g 
 
 22-5 
 
 220 
 
 750 
 
 
 XVI. 
 
 30 
 
 10 
 
 60 
 
 Brown substitute, . 
 
 1-052 
 
 c3 co 
 
 P,XO 
 
 22-5 
 
 77 
 
 360 
 
 
 
 
 
 
 
 
 So 
 
 
 
 
 
 XVII. 
 
 50 
 
 10 
 
 40 
 
 White substitute, . 
 
 1-051 
 
 fjs 
 
 16-0 
 
 66 
 
 600 
 
 
 XVIII. 
 
 30 
 
 10 
 
 60 
 
 White substitute, . 
 
 1-031 
 
 8 | 
 
 15-0 
 
 55 
 
 455 
 
 
 XIX. 
 
 87 
 
 10 
 
 3 
 
 Magnesia, 
 
 1-130 
 
 
 70-5 
 
 427 
 
 465 
 
 
 XX. 
 
 87 
 
 10 
 
 3 
 
 Slaked lime, . 
 
 1-027 
 
 i-3 O 
 
 45-0 
 
 225 
 
 665 
 
 
 XXL 
 
 75 
 
 10 
 
 15 
 
 Paraffin wax, . 
 
 1-075 
 
 n 
 
 15-0 
 
 109 
 
 830 
 
 
 XXII. 
 
 65 
 
 10 
 
 25 
 
 Paraffin wax, . 
 
 1-041 | 
 
 14-0 
 
 115 
 
 1025 
 
 
 XXV. 
 
 75 
 
 10 
 
 15 
 
 Turpentine, 
 
 1-030 S 
 
 ; 
 
 15-0 
 
 96 
 
 825 
 
 
 XXVI. 
 
 65 
 
 10 
 
 25 
 
 Turpentine, 
 
 1-065 
 
 0) 
 
 12-0 
 
 51 
 
 630 
 
 
 XXVII. 
 
 85 
 
 10 
 
 5 
 
 Glycerine, .'".- 
 
 1-062 
 
 g 
 
 29-9 
 
 376 
 
 975 
 
 
 XXVIII. 
 
 80 
 
 10 
 
 10 
 
 Glycerine, . . 
 
 1-060 
 
 a* - 
 
 25-5 
 
 365 
 
 1050 
 
 
 XXIX. 
 
 80 
 
 10 
 
 10 
 
 Asphaltum, . . 
 
 1-080 
 
 1 
 
 45-0 
 
 342 
 
 950 
 
 
 XXX. 
 
 70 
 
 10 
 
 20 
 
 Asphaltum, 
 
 1-085 
 
 i 
 
 67-5 
 
 407 
 
 875 
 
 
 XXXI. 
 
 60 
 
 10 
 
 30 
 
 Ground rubber waste, 
 
 1-101 
 
 ~ 
 
 29-0 
 
 178 
 
 650 
 
 
 XXXII. 
 
 30 
 
 10 
 
 60 
 
 Ground rubber waste, 
 
 1-101 
 
 5j 
 
 25-5 
 
 156 
 
 600 
 
 
 
 ill 
 
 
 
 
 
 
 
 
 
 
 IA. 
 
 12 J 
 
 5 ... 
 
 
 0-925 
 
 
 38-5 
 
 230 
 
 772 
 
 
INORGANIC COMPOUNDING MATERIALS. 
 
 I6 7 
 
 Physical Tests. 
 
 
 
 Change of Form under the ; Change of Form under the 
 Influence of Strong Com- Influence of a Succession 
 
 . a 
 
 
 
 
 press'\on. of Blows. 
 
 > an 
 
 
 
 
 
 8 
 
 00 
 
 
 s * 
 
 
 
 So. 
 
 
 Cgj 05 
 
 *J 
 
 03 
 
 *C ; 
 
 H 
 
 M > 
 
 c'o 
 
 Action of Heat. 
 
 
 
 | [3 v 
 
 ' zz o 
 
 Thickness 
 Swelled Ed 
 
 Thickness 
 Centre. 
 
 Stretch c 
 Extreme E 
 
 ~g 
 
 || 
 
 Thickness 
 Centre. 
 
 Stretch t 
 Extreme EC 
 
 Insulati 
 Observed ] 
 
 
 
 
 1 mm. 
 
 mm. 
 
 mm. 
 
 mm. mm. 
 
 mm. ! mm. 
 
 mm. 
 
 
 
 I. 
 
 ! 0-95 
 
 1-18 
 
 0-93 
 
 9-00 0-95 
 
 1-54 i 071 
 
 275 
 
 117 
 
 Normal. 
 
 
 II. 
 
 1-36 
 
 1-48 1-27 
 
 9-00 1-31 
 
 1-68 T03 
 
 4-50 12 
 
 
 
 III. 
 
 1-08 
 
 i-ie roi 
 
 8-25 j 1-10 
 
 1-40 0-09 
 
 5-00 6 
 
 
 
 IV. 
 
 1-01 
 
 1-07 
 
 0-98 
 
 9-50 i I'OO 
 
 1-31 0-82 
 
 4-50 171 
 
 
 
 V. 
 
 1-12 
 
 1-15 
 
 1-10 
 
 9-50 | 1-09 
 
 1-40 j 0'91 
 
 4-75 
 
 123 
 
 
 
 VI. 
 
 1-17 
 
 1-20 
 
 I'll 
 
 9-50 I'll 
 
 1-35 099 
 
 5'50 4 
 
 |f 
 
 
 VII. 
 
 1-05 
 
 1-15 
 
 I'Ol 
 
 9-00 
 
 1-05 
 
 1-30 0'91 
 
 4-25 4 
 
 ,, 
 
 
 VIII. 
 
 1-09 
 
 1-20 
 
 1-02 
 
 9-00 
 
 1-02 
 
 1-28 
 
 0-90 
 
 475 4 
 
 ,, 
 
 
 IX. 
 
 1-14 
 
 1-21 
 
 1-03 
 
 9-00 
 
 ri3 
 
 1-56 0'95 
 
 5-00 3-5 
 
 ,, 
 
 
 X. 
 
 0-98 
 
 1-02 
 
 0-91 
 
 9-00 
 
 i-oo 
 
 1-42 0-82 
 
 4-75 2 
 
 
 
 
 XI. 
 
 1-03 
 
 1-07 
 
 0-95 
 
 9-00 
 
 1-09 
 
 1-26 
 
 0-98 
 
 4-50 3-5 i 
 
 
 XII. 
 
 0-94 
 
 0-89 
 
 0-85 
 
 8-25 
 
 0-93 
 
 T15 
 
 0-82 
 
 3-50 2257 
 
 ,, 
 
 
 XIII. 
 
 0-91 
 
 0-84 
 
 0-86 
 
 8-00 
 
 0-91 
 
 119 
 
 078 
 
 4-00 220-5 
 
 >5 
 
 
 XV. 
 
 0-85 
 
 0-96 
 
 0-75 
 
 6-50 
 
 078 
 
 1-22 
 
 0-67 
 
 475 9 
 
 Vapours on raising 
 
 
 
 
 incom- 
 
 
 
 
 
 
 
 temperature. 
 
 
 XVI. 
 
 T19 
 
 mensur- 
 able 
 
 0-92 
 
 4-25 
 
 1-15 
 
 1-21 
 
 0-91 
 
 5-00 2 
 
 " 
 
 
 XVII. 
 
 0-96 
 
 1-20 
 
 0-84 
 
 5-50 
 
 1-02 
 
 1-05 
 
 0-73 
 
 375 
 
 50 
 
 
 
 XVIII. 
 
 i-oo 
 
 1-29 
 
 0-91 
 
 ncom- 
 mensur- 
 able 
 
 0-94 
 
 0-96 
 
 0-64 
 
 4-50 
 
 222 
 
 it 
 
 
 XIX. 
 
 T04 
 
 1-08 
 
 1'03 
 
 10-00 
 
 1-01 
 
 1-28 
 
 0-91 
 
 4'50 
 
 7-5 
 
 Normal. 
 
 
 XX. 
 
 1-23 
 
 1-35 
 
 1-06 
 
 8-50 
 
 1-21 
 
 1-55 
 
 0-96 
 
 3-25 
 
 9-5 
 
 
 
 XXI. 
 
 0-91 
 
 1-18 
 
 078 
 
 7-50 
 
 1-21 
 
 1-55 
 
 0-96 
 
 3-25 
 
 1 
 
 Vapours on raising 
 
 
 
 
 
 
 
 
 
 
 
 
 temperature. 
 
 
 XXII. 
 
 1-15 
 
 1-20 
 
 1-01 
 
 7-25 
 
 1-11 
 
 1-45 
 
 077 
 
 575 
 
 
 
 
 
 XXV. 
 
 1-06 
 
 1-25 
 
 0-97 
 
 8-50 
 
 1-16 
 
 1-54 
 
 1-01 
 
 3-25 
 
 1-5 
 
 Vapours at higher 
 
 
 XXVI. 
 
 1-02 
 
 1-21 
 
 0-91 
 
 8-50 
 
 1-09 
 
 1-26 
 
 0-82 
 
 4-00 
 
 1-5 
 
 temperatures. 
 
 
 XXVII. 
 
 ro7 
 
 1-41 
 
 1-00 
 
 775 
 
 1-03 
 
 1-64 
 
 0-93 
 
 4-25 
 
 3 
 
 
 XXVIII. 
 
 1-14 
 
 1-35 
 
 1-03 
 
 8-00 
 
 1-16 
 
 1-92 
 
 0-92 
 
 2-50 
 
 5 
 
 \\ 
 
 XXIX. 
 
 1-25 
 
 1-92 
 
 1-15 4-50 
 
 1-28 
 
 1-06 
 
 1-06 
 
 3-75 
 
 2 
 
 Normal. 
 
 
 XXX. 
 
 1-07 
 
 1-42 
 
 0-95 3-50 
 
 1-08 
 
 0-90 
 
 0-90 
 
 2-00 
 
 2 
 
 
 
 XXXI. 
 
 1-04 
 
 1-35 
 
 0-91 
 
 2-50 
 
 1-03 
 
 0-91 
 
 0-91 
 
 275 
 
 3-5 
 
 
 
 XXXII. 
 
 1-25 
 
 1-44 
 
 1-17 
 
 4-50 
 
 1-42 
 
 1-08 
 
 1'08 
 
 4-00 
 
 1-5 
 
 
 
 
 lA. 
 
 T38 
 
 1-88 
 
 1-25 
 
 10-00 
 
 1-38 
 
 1-29 
 
 1-29 
 
 775 
 
 2-22 
 
 M 
 
i68 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 Chemical Tests. 
 
 B. Commercial India Rubber Articles. 
 
 
 
 ~A; , * 
 
 Action of C 2 H 4 2 . 
 
 
 No. 
 
 l 
 
 ( 
 
 e - 
 
 , 
 
 Quantity and Nature of 
 Admixtures. 
 
 > P 
 
 i ^ 5 oJ 
 5|| 
 
 CJiM 
 
 i AWUfUlJ UA 
 
 H 2 S0 4 . 
 Modifica- 
 tions in 
 Weights 
 
 Modifications 
 per cent. 
 
 
 
 Is 
 
 02 a> 
 
 
 oj_g^ 
 
 l3 
 
 per cent. 
 
 
 
 
 1 
 
 
 
 CC*cS 
 
 
 
 Volumes. Weights. 
 
 
 1 
 
 91 
 
 9 
 
 
 
 
 j 
 
 0-990 
 
 0-999 
 
 + 0-583 
 
 + 0714 ! + 2-540 
 
 
 2 
 
 82 
 
 8 
 
 10 
 
 Chalk, ."' 
 
 1-100 
 
 1-111 + 0-998 
 
 + 19-083 ; +37-688 
 
 
 3 
 
 43 
 
 5 
 
 U 
 
 Brown substitute, .) i. Ann 
 Zinc oxide, . . j 1 40 
 
 1-490 
 
 + 0-402 
 
 + 34-259 +51-319 
 
 
 4 
 
 47 
 
 7 i 27 
 7 it 16 
 
 Chalk, . . -I 1 1.400 
 Zinc oxide, . . / ' 
 
 1-500 
 
 - 1-427 
 
 + 46-216 | +81790 
 
 
 
 
 ( 1H 
 
 Golden sulphide, . } 
 
 
 
 
 
 5 
 
 66 
 
 
 
 < 9 
 
 Chalk, . . . [ 1-200 
 
 1-304 
 
 - 0-665 
 
 + 42-975 +90-209 
 
 
 
 
 
 9 
 
 Zinc oxide, . . J 
 
 
 
 
 
 6 
 
 70 
 
 7 
 
 23 
 
 Chalk, . . . 1-165 
 
 1-222 
 
 - 0-830 
 
 + 37-190 
 
 + 72-947 
 
 
 7 
 
 92 
 
 
 
 8 
 
 Metal, 1 . . . 0-999 
 
 1-111 
 
 - 1-168 
 
 + 2-205 
 
 + 26-263 
 
 
 8 
 
 71 
 
 7 
 
 fie 
 
 1 6 
 
 Chalk, . . . } 
 Zinc oxide, . . / 
 
 1-660 
 
 - 0-832 
 
 + 9-278 
 
 + 53-321 
 
 
 
 
 
 f 6 
 
 Brown substitute, . \ 
 
 
 
 
 
 
 9 
 
 58 
 
 
 
 J 30 
 1 5 
 
 Chalk, . 
 Metal, 1 . . . f 
 
 1-215 
 
 1-330 
 
 - 0-532 
 
 + 7-627 
 
 + 78-623 
 
 
 
 
 
 I 1 
 
 Paste, 1 . . . J 
 
 
 
 
 
 
 
 
 
 
 f 4 
 
 Brown substitute, . ) 
 
 
 
 
 
 
 
 10 
 
 36 
 
 6 
 
 I 27 
 
 Chalk, . . . [ 
 
 1-550 
 
 1 720 
 
 - 0-159 
 
 + 19-841 
 
 + 11-298 
 
 
 
 
 
 (27 
 
 Zinc oxide, . . ] 
 
 
 
 
 
 
 
 
 ' 6 
 
 Brown substitute, . ^ 
 
 
 
 
 
 
 11 
 
 31 
 
 6 
 
 115 
 
 1 40 
 
 Chalk, . 
 Zinc oxide, 
 
 1-600 
 
 1-806 
 
 - 0-859 
 
 + 21-481 
 
 + 13-589 
 
 
 
 
 
 I 2 
 
 Paste, 1 . . . J 
 
 
 
 
 
 
 
 
 
 
 f 4 
 
 Brown substitute, . "| 
 
 
 
 
 
 
 
 
 
 
 10 
 
 Chalk, . 
 
 
 
 
 
 
 
 
 
 
 3 
 
 Zinc oxide, 
 
 
 
 
 
 
 
 12 
 
 39 
 
 
 
 -{ 3 
 
 Metal, 1 . 
 
 1-650 
 
 1-920 
 
 + 1-303 
 
 + 15-126 
 
 + 31-422 
 
 
 
 
 1 0-5 
 
 Paraffin wax, . 
 
 
 
 
 
 
 
 
 
 
 32 
 
 Fluor spar, 
 
 
 
 
 
 
 
 
 
 
 I 8 
 
 Mixture, 1 . . J 
 
 
 
 
 
 
 
 
 
 
 f 8 
 
 Golden sulphide, . ) 
 
 
 
 
 
 
 
 
 
 
 52 
 
 Chalk, . 
 
 
 
 
 
 
 
 13 
 
 22 
 
 
 
 0-5 
 1 0-5 
 
 Paraffin wax, . 
 Paste, 1 . . . f 
 
 n.c. 
 
 1-760 
 
 + 23-443 
 
 + 17-355 
 
 + 21-078 
 
 
 
 
 
 11 
 
 Recovered rubber, . 
 
 
 
 
 
 
 
 
 
 
 I 6 
 
 Brown substitute, . J 
 
 
 
 
 
 
 
 
 
 
 f40 
 
 Chalk, . . o 
 
 
 
 
 
 
 
 14 
 
 30 
 
 
 
 !25 
 1 3 
 
 Zinc oxide, . . ! 
 Metal, 1 . . . j 
 
 n.c. 
 
 1-903 
 
 + 5-824 
 
 + 21-705 
 
 + 15-703 
 
 
 
 
 
 I 2 
 
 Paste, 1 . . .J 
 
 
 
 
 
 
 
 
 
 
 CIS 
 
 Chalk, . . ^ 
 
 
 
 
 
 
 
 
 
 
 3 
 
 Metal, 1 . . .| 
 
 
 
 
 
 
 
 15 
 
 20 
 
 
 
 { 2 
 
 Paste, 1 . . . } 
 
 n.c. 
 
 2-041 
 
 + 41-666 
 
 + 21-538 
 
 + 15-228 
 
 
 
 
 
 35 
 
 Fluor spar, . . | 
 
 
 
 
 
 
 
 
 
 112 
 
 Lead oxide, . . J 
 
 
 
 
 
 
 1 For further information on these compounds, see page 174. n.c. No change. 
 
INORGANIC COMPOUNDING MATERIALS. 
 
 169 
 
 Chemical Tests. 
 
 
 No. 
 
 Action of NaOH. 
 
 Modifications 
 per cent. 
 
 Action of NH 3 . 
 
 Modifications 
 per cent. 
 
 Action of Action of 
 Colza Oil. Mineral Oils. 
 Modifications Modifications 
 in Volumes in Volumes 
 per cent. per cent. 
 
 Action of 
 Mixture of 
 Mineral Oil, 
 90 per cent. ; 
 Tallow, 
 10 per cent. 
 Modifications 
 in Volumes 
 per cent. 
 
 Action of 
 Coal Gas. 
 Modifications 
 in Weights 
 per cent. 
 
 Volumes. Weights. 
 
 Volumes. 
 
 Weights. 
 
 1 
 2 
 
 n. c. - 1-934 
 7i. c. - 2-332 
 
 n.c. 
 + 3-937 
 
 + 1-503 
 + 3-357 
 
 + 18-045 
 + 6-666 
 
 + 36-641 
 + 23-664 
 
 + 89-312 
 + 70-992 
 
 + 12-431 
 + 12-139 
 
 
 3 
 
 n.c. - 3-058 
 
 + 0-482 
 
 + 4-309 
 
 + 47-663 
 
 + 31-428 
 
 + 84-955 
 
 + 6-027 
 
 
 4 
 
 n.c. 
 
 - 1-847 
 
 + 4-001 
 
 + 4-722 
 
 + 3-703 
 
 + 10-077 
 
 + 55-118 
 
 + 5-596 
 
 
 5 
 
 n.c. 
 
 - 0-564 
 
 + 16-071 + 2-877 
 
 + 6-1*0 
 
 + 12-931 
 
 + 7-692 
 
 + 8-271 
 
 6 
 
 7 
 
 8 
 
 n.c. 
 -1-470 
 
 71. C. 
 
 - l-i-34 
 - 1-266 
 
 - 1-600 
 
 + 5-660 ! + 6-600 
 + 11-450 +23550 
 
 + 15-217 ; +17-652 
 
 + 5-737 
 n.c. 
 
 + 6-382 
 
 + 12-068 
 + 8-450 
 
 + 18-556 
 
 + 18-333 + 8-059 
 + 34-677 + 8-413 
 
 + 54-347 + 4-836 
 
 
 9 
 
 n.c. 
 
 - 4-650 
 
 + 10-250 
 
 + 22-405 
 
 n.c. 
 
 + 5-024 
 
 + 13-559 
 
 + 4-133 
 
 
 10 
 
 n.c. 
 
 -19-083 
 
 + 6-299 
 
 + 6-863 
 
 + 27-480 
 
 + 30-708 
 
 + 81-679 
 
 + 4-494 
 
 11 
 
 -0-699 
 
 -20-551 
 
 + 4-347 
 
 + 6-238 
 
 + 20-595 
 
 + 21-428 
 
 + 64-492 
 
 + 5-069 
 
 
 12 
 
 7i. e. 
 
 -14724 
 
 + 7-438 
 
 + 6-009 
 
 + 27168 
 
 + 22-413 
 
 incommen- 
 surable 
 
 + 5-843 
 
 
 13 
 
 n.c. 
 
 - 1-601 
 
 + 9-090 
 
 + 8-391 
 
 + 15-125 
 
 + 14-876 
 
 + 47-580 
 
 + 4-020 
 
 
 14 
 
 n.c. 
 
 - 1-655 
 
 + 10-156 
 
 + 8-665 
 
 + 7-031 
 
 + 8-064 
 
 + 33-343 
 
 + 3-837 
 
 
 15 
 
 n.c. 
 
 - 0-163 
 
 + 6-666 
 
 + 6-474 
 
 + 10-687 
 
 + 11-627 
 
 + 27-480 
 
 + 3-122 
 
 n.c. No change. 
 
170 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 Physical Tests. 
 
 B. Commercial India Rubber Articles. 
 
 
 
 
 
 
 Modul us 
 of Pressure 
 per I 2 mm. 
 
 Modulus 
 of Resist- 
 ance per 
 l a mm. 
 
 Distensi- 
 bility 
 in mm. 
 per 
 100 mm. 
 
 
 No. 
 
 India Rubber 
 per cent. 
 
 Sulphur 
 per cent. 
 
 Quantity and Nature of 
 Admixtures. 
 
 Is 
 
 
 
 2C ^3 
 
 Spec. Grav. 
 after 2i Years. 
 
 
 
 
 
 
 
 
 grms. 
 
 grms. 
 
 
 
 1 
 
 91 
 
 9 
 
 
 
 
 0-990 
 
 0-999 
 
 33-5 
 
 201 
 
 675 
 
 
 2 
 
 82 
 
 8 
 
 10 
 
 Chalk, .'" . 
 
 1-100 
 
 1-111 
 
 40 
 
 230 
 
 600 
 
 
 3 
 
 43 
 
 5 
 
 \45 
 
 Brown substitute, . \ 
 Zinc oxide, . . J 
 
 1-400 
 
 1-490 
 
 21 
 
 349 
 
 580 
 
 
 4 
 
 47 
 
 7 
 
 (27 
 116 
 
 Chalk, . . . \ 
 Zinc oxide, . . ] 
 
 1-400 
 
 1-500 
 
 38-5 
 
 124 
 
 160 
 
 
 6 
 
 66 
 
 
 
 OS OS 
 
 T 1 
 
 Golden sulphide, . \ 
 Chalk, . . . L 
 Zinc oxide, . . J 
 
 1-200 
 
 1-304 
 
 43 
 
 173 
 
 235 
 
 
 6 
 
 70 
 
 7 
 
 23 
 
 Chalk, 
 
 1-165 ! 1-222 
 
 35 
 
 86 
 
 190 
 
 
 7 
 
 92 
 
 
 
 8 
 
 Metal, 1 . 
 
 0-999 1-111 
 
 16 
 
 25 
 
 110 
 
 
 8 
 
 71 
 
 7 
 
 f!6 
 I 6 
 
 Chalk, . . . \ 
 Zinc oxide, . . J 
 
 1-450 
 
 1-660 
 
 31-5 
 
 63 
 
 145 
 
 
 
 
 
 I 6 
 
 Brown substitute, . \ 
 
 
 
 
 
 
 
 9 
 
 58 
 
 
 
 J 30 
 1 5 
 
 Chalk, . . . 1 
 Metal,i . . . f 
 
 1-215 
 
 1-330 
 
 12 
 
 33 
 
 110 
 
 
 
 
 
 I 1 
 
 Paste, 1 . . . J 
 
 
 
 
 
 
 
 
 
 
 f 4 
 
 Brown substitute, . "| 
 
 
 
 
 
 
 
 10 
 
 36 
 
 6 
 
 427 
 
 Chalk, . V 
 
 1-550 
 
 1-720 
 
 37-5 
 
 183 
 
 550 
 
 
 
 
 
 ^ 27 Zinc oxide, . . J 
 
 
 
 
 
 
 
 
 f 6 
 
 Brown substitute, . ^ 
 
 
 
 
 
 
 
 11 
 
 31 
 
 6 
 
 115 
 J40 
 
 Chalk, . . . 1 
 Zinc oxide, . . f 
 
 1-600 
 
 1-806 
 
 36 
 
 145 
 
 310 
 
 
 
 
 
 I 2 
 
 Paste, 1 . . J 
 
 
 
 
 
 
 
 
 
 
 f 4 
 
 Brown substitute, . ~) 
 
 
 
 
 
 
 
 
 
 
 110 
 
 Chalk, 
 
 
 
 
 
 
 
 
 
 
 3 
 
 Zinc oxide, 
 
 
 
 
 
 
 
 12 
 
 39 
 
 
 
 1 3 
 
 Metal, 1 . . .\ 
 
 1-650 
 
 1-920 
 
 24 
 
 81 
 
 270 
 
 
 
 
 
 1 0-5 
 
 Paraffin wax, . 
 
 
 
 
 
 
 
 
 
 
 32 
 
 Fluor spar, 
 
 
 
 
 
 
 
 
 
 
 L 8 
 
 Mixture, 1 . . J 
 
 
 
 
 
 
 
 
 
 
 ( 8 
 
 Golden sulphide, 
 
 
 
 
 
 
 
 
 
 
 62 
 
 Chalk, . 
 
 
 
 
 
 
 
 13 
 
 22 
 
 
 
 J 0-5 
 1 0-5 
 
 Paraffin wax, . 
 Paste, 1 
 
 n.c. 
 
 1-760 
 
 41-5 
 
 116 
 
 140 
 
 
 
 
 
 1 11 
 
 Recovered rubber, . | 
 
 
 
 
 
 
 
 
 
 
 I 6 
 
 Brown substitute, . J 
 
 
 
 
 
 
 
 
 
 
 f40 
 
 Chalk, . . .^ 
 
 
 
 
 
 
 
 14 
 
 30 
 
 
 
 i25 
 1 3 
 
 Zinc oxide, . . ! 
 Metal, 1 . . . f 
 
 n.c. 
 
 1-903 
 
 38-5 
 
 161 
 
 125 
 
 
 
 
 
 I 2 
 
 Paste, 1 . . .J 
 
 
 
 
 
 
 
 
 
 
 (18 
 
 Chalk, . . o 
 
 
 
 
 
 
 
 
 
 
 I 3 
 
 Metal, 1 . . . | 
 
 
 
 
 
 
 
 15 
 
 30 
 
 
 
 -I 2 
 
 Paste, 1 . . . } 
 
 n.c. 
 
 2-041 
 
 22-5 
 
 61 
 
 110 
 
 
 
 
 
 | 35 
 
 Fluorspar, . . | 
 
 
 
 
 
 
 
 
 
 
 112 
 
 Lead oxide, . . J 
 
 
 
 
 
 
 
 1 For further information on these compounds, see page 174. 
 
 n.c. No change. 
 
INORGANIC COMPOUNDING MATERIALS. 
 
 171 
 
 Physical Tests. 
 
 
 No. 
 
 Change of Form under the 
 Influence of Strong Compression. 
 
 Change of Form under the 
 Influence of a Succession of Blows. 
 
 Action of Heat. 
 
 Initial 
 Thickness. 
 
 Thickness of 
 Swelled 
 Edges. 
 
 Thickness at 
 Centre. 
 
 - Stretch of 
 *? * g Extreme 
 - 3 Edges. 
 
 Initial 
 Thickness. 
 
 Thickness of 
 Swelled 
 Edges. 
 
 Thickness at 
 Centre. 
 
 Stretch of 
 Extreme 
 Hdges. 
 
 1 
 2 
 
 mm. 
 1-36 
 1-26 
 
 mm. 
 T62 
 175 
 
 mm. 
 1-29 
 1-24 
 
 mm. 
 T42 
 1-32 
 
 mm. 
 2-04 
 1-72 
 
 mm. 
 1-18 
 
 1-08 
 
 mm. 
 275 
 1-50 
 
 Normal. 
 
 > > 
 
 
 3 
 4 
 
 1-10 
 1-18 
 
 1-27 
 
 1-25 
 
 1-04 
 1'04 
 
 7-50 
 5-25 
 
 114 
 
 1-28 
 
 1-48 
 1-42 
 
 1-02 
 1-07 
 
 4-75 
 4-50 
 
 Evolution of vapours on 
 raising temperature ; 
 yellow discoloration 
 due to substitute. 
 Develops discoloration 
 and brittleness. 
 
 
 5 
 
 1-10 
 
 1-20 
 
 1-03 
 
 6-50 
 
 1-09 
 
 1-16 
 
 1-01 
 
 5 
 
 At 150 becomes soft and 
 
 
 
 
 
 
 
 
 
 
 
 spongy. 
 
 6 
 
 7 
 
 1-25 
 T21 
 
 1-32 
 1-06 
 
 1-16 
 0-91 
 
 7-25 
 7'50 
 (12-50) 
 
 1-29 
 1-24 
 
 1-36 
 1-01 
 
 1-10 
 
 0-90 
 
 5-25 
 6-25 
 
 Becomes hard and brittle. 
 > 
 
 
 8 
 
 0-39 
 
 0-76 
 
 0-82 
 
 7-50 
 (13-25) 
 
 0-92 
 
 0-87 
 
 0-75 
 
 7 
 
 Blisters and turns hard. 
 
 
 9 
 
 1-12 
 
 0-82 
 
 0-87 
 
 9-25 
 
 (14) 
 
 1-15 
 
 0-86 
 
 0-82 
 
 9-24 
 
 Turns still harder than 
 preceding sample. 
 
 
 10 
 
 1-29 
 
 1-35 
 
 1-20 
 
 775 
 
 cz 
 
 1-29 
 
 1-64 
 
 1-12 
 
 4 
 
 At from 40 to 50 curls 
 and breaks. 
 
 11 
 
 1-34 
 
 1-41 
 
 1-27 
 
 6'50 
 
 1-38 
 
 1-46 
 
 1-24 
 
 5-25 
 
 Curls. 
 
 
 12 
 
 1-25 
 
 1-21 
 
 1-04 
 
 6-50 
 
 (12-50) 
 
 1-24 
 
 1-08 
 
 1-05 
 
 7 
 
 Turns very soft. 
 
 
 13 M6 
 
 1-91 
 
 1-18 
 
 6-25 
 
 1-23 
 
 T07 
 
 1-03 
 
 9 
 (12) 
 
 Turns spongy, and above 
 130 breaks. 
 
 
 14 I 1'27 
 
 0-78 
 
 1-07 
 
 8-25 
 (14-50) 
 
 1-24 
 
 I'OG 
 
 1-03 
 
 9-25 
 (12-50) 
 
 Becomes brittle. 
 
 
 15 1-31 
 
 0-8.8 
 
 0-90 
 
 8 
 (15-50) 
 
 1-32 
 
 1'22 
 
 1-04 
 
 8'25 
 (12) 
 
 Becomes spongy and 
 brittle. 
 
172 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 Chemical Tests. 
 
 C. Samples of Ebonite Manufactures. 
 
 
 Action of C2H 4 2 . 
 
 
 
 Action of 
 H 2 S0 4 . 
 
 Modifications per cent. 
 
 
 1 * 
 
 
 
 Modifica- 
 
 
 
 
 Sj 
 
 
 
 
 tions in 
 
 
 
 No. 
 
 1! 
 
 11 
 
 Quantity and Nature o 
 Admixtures. 
 
 11 
 
 Weights 
 per cent. 
 
 Volumes. 
 
 Weights 
 
 
 
 r 
 
 & 
 
 
 ** 
 
 
 
 
 
 XXXIII. 
 
 80 
 
 20 
 
 
 
 
 1-095 
 
 - 2-028 
 
 n.c. 
 
 + 0-746 
 
 
 XXXIV. 
 
 60 
 
 20 
 
 (10 
 110 
 
 Turpentine, . . \ 
 Colophony, . . / 
 
 n.c. 
 
 + 15-133 
 
 n.c. 
 
 - 3-501 
 
 
 XXXV. 
 
 50 
 
 20 
 
 ? 16 
 
 U5 
 
 Turpentine, . . \ 
 Colophony, . . / 
 
 n.c. 
 
 - 2-018 
 
 + 2-608 
 
 - 4-989 
 
 
 XXXVI. 
 
 60 
 
 20 
 
 20 
 
 Asphaltum, . 
 
 n.c. 
 
 + 28-531 
 
 n.c. 
 
 - 7-345 
 
 
 XXXVII. 
 
 40 
 
 20 
 
 40 
 
 Asphaltum, . 
 
 n.c. 
 
 - 9-504 
 
 n.c. 
 
 - 9-127 
 
 
 XXXVIII. 
 
 70 
 
 20 
 
 10 
 
 Vermilion, 
 
 1-177 
 
 - 3-225 
 
 + 2-000 - 0114 
 
 
 XXXIX. 
 
 60 
 
 20 
 
 20 
 
 Vermilion, 
 
 1-384 
 
 + 17-551 
 
 n.c. n.c. 
 
 
 XL. 
 
 40 
 
 20 
 
 MO 
 1 30 
 
 Turpentine, . . \ 
 Slaked lime, . . / 
 
 1-275 
 
 - 2-603 
 
 + 3-200 +25-454 
 
 
 XLI. 
 
 10 
 
 20 {JJ 
 
 Turpentine, . . \ 
 Slaked lime, . . J 
 
 1-311 
 
 - 7-880 
 
 + 4-964 +61-148 
 
 
 XLII. 
 
 40 
 
 20 |{30 
 
 Turpentine, . . \ 
 Magnesia, . . j" 
 
 1-203 
 
 - 4-203 
 
 n.c. - 1-296 
 
 
 XLIII. 
 
 10 
 
 o ' / 10 
 
 Turpentine, . . \ 
 
 1-428 
 
 - 8-333 
 
 n.c. - 2-886 
 
 
 
 
 ^ 60 
 
 Magnesia, . . j 
 
 
 
 
 
 XLIV. 
 
 65 
 
 20 , ... 
 
 
 n.c. 
 
 n.c. 
 
 n.c. n.c. 
 
 
 Physical Tests. 
 
 C. Samples of Ebonite Manufactures. 
 
 
 
 
 
 
 
 Modulus 
 
 Distensi- 
 
 
 No. 
 
 ia Rubber 
 er cent. 
 
 phur per 
 cent. 
 
 Quantity and Nature of 
 Admixtures. 
 
 Is 
 
 Modulus 
 of Pressure 
 per I 2 mm. 
 
 of Resist- 
 tance per 
 I 2 mm. 
 
 bility 
 in mm. 
 perlOOmm. 
 
 
 
 I" 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 grms. 
 
 grms. 
 
 
 
 XXXIII. 
 
 83 
 
 20 
 
 
 
 
 1-095 
 
 43-5 
 
 105 
 
 145 
 
 
 XXXIV. 
 
 60 
 
 20 
 
 rio 
 
 115 
 
 Turpentine, . . \ 
 Colophony, . . f 
 
 n.c. 
 
 15 
 
 142 
 
 175 
 
 
 XXXV. 
 
 50 
 
 20 
 
 115 
 
 Turpentine, . . "1 
 Colophony, . . J 
 
 n.c. 
 
 12 
 
 46 
 
 135 
 
 
 XXXVI. 
 
 60 
 
 20 
 
 20 
 
 Asphaltum, . 
 
 n.c. 
 
 17 
 
 69 
 
 145 
 
 
 XXXVII. 
 
 40 
 
 20 
 
 40 
 
 Asphaltum, . 
 
 n.c. 
 
 24 
 
 244 
 
 175 
 
 
 XXXVIII. 
 
 70 
 
 20 
 
 10 
 
 Vermilion, 
 
 1-177 
 
 32-5 
 
 197 
 
 185 
 
 
 XXXIX. 
 
 60 
 
 20 
 
 20 
 
 Vermilion, 
 
 1-384 
 
 61 
 
 341 
 
 190 
 
 
 XL. 
 
 40 
 
 20 
 
 10 
 30 
 
 Turpentine, . . "1 
 Slaked lime, . . J 
 
 1-275 
 
 22-5 
 
 122 
 
 275 
 
 
 XLI. 
 
 10 
 
 20 
 
 10 
 60 
 
 Turpentine, . . 1 
 Slaked lime, . . J 
 
 1-311 
 
 10 
 
 55 
 
 285 
 
 
 XLII. 
 
 40 
 
 20 
 
 10 
 30 
 
 Turpentine, . . \ 
 Magnesia, . . / 
 
 1-203 
 
 31-7 
 
 390 
 
 105 
 
 
 XLIII. 10 
 
 20 
 
 rio 
 
 \60 
 
 Turpentine, . . \ 
 Magnesia, . . / 
 
 1-428 
 
 25-9 
 
 272 
 
 101-25 
 
 
 XLIV. ! 65 
 
 35 
 
 
 
 n.c. 
 
 excellent 
 
 excellent 
 
 incommen- 
 
 
 
 
 
 
 
 
 surable 
 
 
 n.c. No change. 
 
INORGANIC COMPOUNDING MATERIALS. 
 Chemical Tests. 
 
 173 
 
 
 No. 
 
 Action of NaOH. 
 
 Modifications 
 per cent. 
 
 Action of NH 3 . 
 
 Modifications per 
 cent. 
 
 Action of Colza Oil. 
 Modifications of 
 Volumes per cent. 
 
 Action of Mineral Oils. 
 Modifications of 
 Volumes per cent. 
 
 e g o| 
 
 Action of Coal Gas. 
 Modifications of 
 Volumes per cent. 
 
 Volumes. 
 
 Weights. 
 
 Volumes. 
 
 Weights. 
 
 XXXIII. n.c. 
 
 - 2-692 n.c. 
 
 + 0-490 
 
 + 12-068 
 
 + 7-874 
 
 + 20-000 
 
 + 11-629 
 
 
 XXXIV. n.c. 
 
 - 1-426 
 
 n. c. 
 
 + 0-611 
 
 + 2-728 
 
 n.c. 
 
 + 2-932 
 
 + 11-339 
 
 
 XXXV. 
 
 n.c. 
 
 - 0-187 
 
 n.c. 
 
 + 0-428 
 
 + 1-428 
 
 n.c. 
 
 + 6-293 
 
 + 8-053 
 
 
 XXXVI. 
 XXXVII. 
 
 n.c. 
 n.c. 
 
 - 0-112 
 - 3-192 
 
 n.c. 
 n.c. 
 
 + 0-344 
 + 0-488 
 
 + 6-440 
 + 4-918 
 
 + 1-418 
 + 1-515 
 
 + 26-086 
 + 16-883 
 
 + 11-804 
 + 10-062 
 
 XXXVIII. 
 XXXIX. 
 
 XL. 
 
 n.c. 
 n.c. 
 
 n.c. 
 
 - 3-182 
 - 2-361 
 
 - 1-404 
 
 n.c. 
 n.c. 
 
 n.c. 
 
 + 0-406 
 + 0-281 
 
 -0-689 
 
 n.c. 
 n.c. 
 
 + 0-862 
 
 + 11-956 
 + 7-777 
 
 + 0-862 
 
 + 41-758 
 + 33-707 
 
 + 20-000 
 
 + 8-571 
 + 0-581 
 
 + 9-311 
 
 XLI. 
 
 -2-054 
 
 -14-145 
 
 -1-351 
 
 -8-198 
 
 n.c. 
 
 n.c. 
 
 + 7-284 
 
 + 7-296 
 
 
 XLII. 
 XLIII. 
 
 n.c. 
 
 - 1-278 
 
 n.c. 
 
 + 0'396 n.c. 
 
 n.c. 
 
 n.c. 
 
 + 7-472 
 
 n.c. 
 
 - 0-539 
 
 n.c. 
 
 + 0-396 
 
 n.c. 
 
 n.c. 
 
 n.c. 
 
 + 4-510 
 
 XLIV. 
 
 n.c. 
 
 - 0-646 
 
 n.c. 
 
 n.c. 
 
 n.c. 
 
 n.c. 
 
 n.c. 
 
 + 3-569 
 
 Physical Tests. 
 
 
 No. 
 
 Change of Form under the In- 
 fluence of Strong Compression. 
 
 Change of Form under the 
 Influence of a Succession of Blows. 
 
 Insulating Power. 
 Observed Depressions. 
 
 Action of 
 Heat. 
 
 Initial 
 Thickness. 
 
 Thickness of 
 Swelled 
 Edges. 
 
 Thickness at 
 Centre. 
 
 Stretch of 
 Extreme 
 Edges. 
 
 *** 
 's.w 
 H 
 
 Thickness of 
 Swelled 
 Edges. 
 
 Thickness at 
 Centre. 
 
 Stretch of 
 Extreme 
 Edges. 
 
 XXXIII. 
 XXXIV. 
 
 mm. 
 1-25 
 
 2-30 
 
 mm. 
 1-44 
 
 171 
 
 mm. 
 1-17 
 
 n.c. 
 
 mm. 
 4-50 
 
 10-50 
 
 mm. 
 1-26 
 
 2-38 
 
 mm. 
 1-23 
 
 2-25 
 
 mm. 
 1-20 
 
 2-25 
 
 mm. 
 n.c. 
 
 10-00 
 
 103 
 575 I 
 
 Normal. 
 Evolution of 
 vapour. 
 
 
 XXXV. 
 
 1-04 
 
 0-89 
 
 0-89 
 
 11-50 
 
 2-07 
 
 1-57 
 
 1-57 
 
 n.c. 
 
 8 
 
 > 
 
 
 XXXVI. 
 XXXVII. 
 
 1-67 
 1-72 
 
 0-96 
 0-91 
 
 1-65 
 1-54 
 
 1-66 
 1-58 
 
 11-00 
 12-25 
 
 1-92 
 2-81 
 
 1-67 
 2-24 
 
 1-67 
 2-25 
 
 11-25 
 11-00 
 
 60 
 2 
 
 Becomes soft. 
 
 XXXVIII. 
 XXXIX. 
 
 0-86 
 0-89 
 
 0-87 
 0-89 
 
 10-00 
 10-00 
 
 0-93 
 0-95 
 
 0-87 
 0-93 
 
 0-87 
 0-93 
 
 11-25 
 
 10-00 
 
 111 
 46 
 
 Normal. 
 
 
 
 XL. 
 
 1-20 
 
 1-42 
 
 1-02 
 
 7-50 
 
 1-22 
 
 1-41 
 
 0-98 
 
 6-50 
 
 105-5 
 
 Evolves H 2 S. 
 
 
 XLI. 
 
 1-56 
 
 1-67 
 
 1-21 
 
 3-50 
 
 1-36 
 
 broken 
 
 n.c. 
 
 n.c. 
 
 3 
 
 M 
 
 
 XLII. 
 
 1-32 
 0-61 
 
 0-79 
 
 0-89 
 
 12-50 
 
 0-97 
 
 0-76 
 
 0-89 
 
 11-50 
 
 220 
 
 Normal. 
 
 XLIII. 
 
 0-50 
 
 0-50 
 
 11-25 
 
 0-62 
 
 0-55 
 
 0-57 
 
 11-00 
 
 218 
 
 
 
 
 XLIV. 
 
 1-27 
 
 1-27 
 
 1-27 
 
 10-00 
 
 1-25 
 
 1-29 
 
 1-29 
 
 splintered 
 
 68 
 
 " 
 
 n.c. No change. 
 
174 INDIA RUBBER AND ITS ANALYSIS. 
 
 The vulcanising compounds repeatedly alluded to in the foregoing 
 tables are ' secret ' preparations, the employment of which, as is evidenced 
 by the behaviour of the samples prepared with them, generally results in 
 the production of defective goods. PAHL and HEINZBRLING have analysed 
 these products and found their composition as follows : 
 
 Metal. A mixture of sulphur with the constituents of ultramarine. 
 
 Paste. Paraffin wax, or a mixture of resin with certain organic 
 residues: the whole dissolved in turpentine, and mixed with plaster of 
 Paris, lime, and magnesia. 
 
 Mixture. A solution of regenerated rubber in turpentine. 
 
 The following are the conclusions arrived at by PAHL and HEINZERLING 
 at the end of their numerous and laborious experiments : 
 
 1. All organic or mineral additions diminish the elasticity of soft rubber 
 goods; vulcanisation with 10 per cent, of sulphur produces the best results 
 in point of elasticity. On the other hand, certain organic substances such 
 as turpentine, colophony, asphaltum, and likewise certain inorganic 
 compounds such as vermilion and slaked lime, not only do not diminish 
 the elasticity, but often even increase it. 
 
 2. The resistance to pressure of soft rubber goods is, to a certain degree, 
 increased by the addition of asphaltum, chalk, fluor spar, zinc oxide, and 
 especially by calcined magnesia. The resistance of hard rubber is increased 
 by the addition of vermilion and of magnesia, likewise by prolonged 
 vulcanisation and a higher proportion of sulphur. All other additions of 
 inorganic and organic compounds, with the exception of asphaltum in 
 small quantity, decrease the resistance. 
 
 3. Additions of mineral substance such as zinc oxide, litharge, chalk, 
 magnesia, lime, also asphaltum and glycerine, improve the mechanical 
 strength of soft rubber. Provided certain limits are not exceeded, this is 
 also true for hard rubber containing asphaltum, turpentine, and colophony, 
 together with chalk or magnesia. On the other hand, the mechanical 
 resistance of india rubber is sensibly diminished by the addition of 
 vulcanised oils (substitutes), turpentine, and paraffin wax. 
 
 4. The opinion that pure india rubber mixed with sulphur is the best 
 insulator is erroneous. Mixtures of india rubber with zinc oxide, lime, 
 magnesia in small quantities, antimony pentasulphide, brown substitute, 
 white substitute in very small quantities, turpentine, colophony, and 
 paraffin wax all give superior results. Especially mixtures of india rubber 
 and paraffin wax stand out very prominently in this respect. Vermilion 
 and litharge, on the other hand, reduce the dielectric properties of india 
 rubber, and magnesia in large proportions has a similar effect. The 
 worst results under all circumstances are obtained with cold cured india 
 rubber. 
 
 5. All mixtures of india rubber with organic compounds give off 
 vapours at temperatures from 130 to 150 C. For articles which in use 
 .are exposed to higher temperatures, such admixtures should therefore be 
 
INORGANIC COMPOUNDING MATERIALS. 175 
 
 avoided. The other compounding materials generally behave well in this 
 respect. 
 
 6. All mixtures containing chalk, fluor spar, litharge, and zinc oxide are 
 strongly acted upon by sulphuric and acetic acid. The action of these 
 acids always causes a considerable increase in weight due to the formation 
 of either insoluble or basic salts. 
 
 The destructive action of all oils upon india rubber is more particularly 
 checked by the addition of litharge and zinc oxide ; other inorganic ad- 
 mixtures are much less effective. The mixtures examined in the foregoing 
 tables are not sufficiently modified by either the alkalies or coal gas to 
 allow any definite conclusions to be drawn. All organic admixtures 
 protect the compound more or less from the corrosive action of the acids. 
 
 7. There is a general tendency in india rubber containing inorganic 
 admixtures to harden and become brittle on prolonged storing. They also 
 offer less resistance to mechanical influences, but their insulating power 
 appears to remain unimpaired. 
 
 All those who have any practical acquaintance with the manufacture 
 of india rubber goods will entirely agree with the above conclusions, 
 considered merely as general statements. But while I respect the 
 enormous amount of painstaking work accomplished in these tables by the 
 authors, I nevertheless think it should be pointed out that the value of 
 the tables (A) is much diminished by the fact that all the compoundings 
 were vulcanised under identical conditions of temperature and pressure and 
 time. But, at the same time, the ratio of sulphur to rubber varies 
 enormously, from 10 to 100 per cent., so that practically in each test the 
 most important of all vulcanising conditions, the ratio (sulphur : rubber), 
 will not bear comparison with any of the other tests. Therefore the 
 difference in the degree of vulcanisation of the different samples must have 
 been enormous, even if we leave altogether out of account that, for the 
 same ratio of S/K, the degree of vulcanisation in samples vulcanised under 
 identical physical conditions varies very considerably with the nature and 
 quantity of even such inorganic admixtures which we know to be indifferent 
 to sulphur. For these reasons I fear that these tables do not give us any 
 very reliable indication of the influence of the inorganic and organic 
 admixtures experimented with upon the manufactured india rubber. 
 This uncertainty, no doubt, affects the chemical tests to a very considerable 
 degree, but even more so the physical tests. 
 
 As regards the tables (B), I must confess to being altogether unable to 
 comprehend the meaning of these compoundings, the greatest part of 
 which appear to me as extraordinary specimens of the * rubber mixer's ' 
 fancy and imagination. Perhaps my experience of rubber manufactures 
 is not sufficiently extensive, but, excepting the first nine specimens, I 
 certainly must confess that I could not name any article which might 
 be manufactured from the compoundings specified from No. 10 onwards. 
 
CHAPTER V. 
 VULCANISERS AND SULPHITE CARRIERS. 
 
 THE chemical and physical aspects of the vulcanising process have been 
 fully dealt with in the foregoing chapters, not only as regards the vul- 
 canisation with sulphur and chloride of sulphur, but also as regards the 
 various other processes proposed at different times, and perhaps also 
 practised, though, no doubt, often experimentally only. The question of 
 vulcanisers and sulphur carriers from the manufacturing point of view is, 
 however, well worth a separate discussion, for not only are commercial 
 brands of sulphur, chloride of sulphur, and the various recognised sulphur 
 carriers apt to vary in composition, but also to contain impurities 
 capable of affecting the india rubber detrimentally. A constant control of 
 the vulcanising materials used in the factory is therefore one of the most 
 important tasks of the rubber works' chemist. 
 
 Sulphur. As is well known, sulphur exists in a number of modifica- 
 tions, which, apart from their morphological distinctions, differ from each 
 other, in some cases considerably, in their physical properties, and show 
 also unmistakable, if slight, differences in their chemical behaviour. We 
 possess at present no positive information which would enable us to decide 
 whether these differences have any effect upon the vulcanising properties 
 of these various modifications. Since the vulcanising temperatures lie 
 more or less considerably above the melting point of sulphur, it might, 
 perhaps, a priori be assumed that no such differences in the vulcanising 
 properties of the various modifications of sulphur can exist, but it is not 
 at all improbable that differences of molecular aggregation and configura- 
 tion of at least some of these modifications actually persist in the fused 
 sulphur, and some little support for this proposition is perhaps to be found 
 in the empirically established fact that the different kinds of sulphur not 
 infrequently produce rather different vulcanisation results. This applies 
 more particularly to the crystalline, as distinguished from the amorphous, 
 kinds of sulphur. 
 
 The modifications of sulphur are very numerous, and the subject is by 
 no means exhausted from the scientific point of view. We need not, how- 
 ever, deal here with the rare and somewhat doubtful modifications of this 
 
VULCANISERS AND SULPHUR CARRIERS. 
 
 177 
 
 element, but confine ourselves to the commonly occurring ones. These 
 are : 
 
 I. Ordinary Rhombic (Octahedral] Sulphur. M.P. 1 11-1 14 C., B.P. 
 448 C., Sp. Gr. 2-060 to 2-070. 
 
 II. Monoclinic (Prismatic) Sulphur. M.P. 117-4-120 0., Sp. Gr. 1'958. 
 This modification is rather unstable, and readily reverts to the first and 
 denser modification. 
 
 These crystalline modifications are very soluble in carbon bisulphide 
 and in a large number of organic solvents. It is, indeed, questionable 
 whether any organic body exists which does not dissolve sulphur. Some 
 of the determined solubilities are : 
 
 100 parts of Carbon bisulphide at 15 C. dissolve 37 parts of S. 
 
 
 1 
 
 46 
 
 73 
 
 , 
 
 
 , Benzene 
 
 26 
 
 0-965 
 
 i 
 
 
 > > 
 
 71 
 
 4-377 
 
 j 
 
 
 , Toluene 
 
 23 
 
 1-479 
 
 
 
 , Ether 
 
 23-5 
 
 0-972 
 
 i 
 
 
 , Chloroform 
 
 22 
 
 1-250 
 
 j 
 
 
 , Phenol 
 
 174 
 
 16-350 
 
 
 
 , Aniline 
 
 130 
 
 85-270 
 
 t 
 
 The solubility of sulphur in acetone I found to be as follows : 
 
 AtlOC., 
 22-5 
 56 
 
 '04 per cent. 
 
 0-08 
 
 0-25 
 
 III. Amorphous Sulphur. This seems to exist in a great number of 
 varieties. Most of the amorphous varieties are insoluble in carbon bi- 
 sulphide, but, according to DEVILLED these are to a considerable extent 
 soluble in chloroform, ether, and alcohol. The conditions determining the 
 reversion of insoluble to soluble sulphur were investigated by KtfSTER, 2 who 
 showed that the two modifications not only differ in their molecular 
 weights, but that they must be regarded not only as physical, but also as 
 "chemical isomers, comparable to the allotropic modifications of oxygen. 
 
 Very finely subdivided sulphur is at ordinary temperatures slowly 
 oxidised by atmospheric oxygen. This is, no doubt, the reason why the 
 variety known as ' flowers of sulphur ' invariably contains a small amount 
 of sulphuric acid, occasionally as much as 0'5 per cent. In this respect 
 it is, however, remarkable that the still much more finely subdivided ' pre- 
 cipitated sulphur ' very rarely shows any trace of acidity. 
 
 The chiefly used commercial varieties of sulphur are : 
 
 1. Sulphur sticks. 
 
 2. Flowers of sulphur. 
 
 3. Milk of sulphur. 
 
 4. Precipitated sulphur. 
 
 Journ. pmkt. Chem., 56, 359. 
 
 2 Ztschr.f. anorg. Chem., 18, 365. 
 
 12 
 
178 INDIA RUBBER AND ITS ANALYSIS. 
 
 1. For the analysis of sulphur in sticks or lumps, in which form it is 
 generally used for ' sulphur baths,' a determination of the ash present is 
 all that is required. This is carried out by careful ignition of a few grammes 
 of sulphur in a porcelain crucible, and weighing the residue. 
 
 Arsenic occurs only in very minute quantities in Sicilian sulphur, but 
 not inconsiderable quantities are often to be found in ' recovered ' sulphur. 
 It may be present in the form of arsenious acid or as arsenious sulphide. 
 The presence of these compounds is detected by treating the carefully 
 ground sample at a temperature of 80 C. with a 10 per cent, solution of 
 ammonia, filtering, and evaporating the filtrate. Or the filtrate is acidu- 
 lated with hydrochloric acid, the arsenious sulphide which separates out 
 filtered off and determined separately, while the arsenious acid in the 
 filtrate is determined by the well-known methods. 
 
 Selenium, which occurs not infrequently, is of 110 account for the 
 purpose of vulcanisation ; moreover, it is only present, if at all, in extremely 
 small quantities. 
 
 2. Flowers of Sulphur call for rather more careful analysis. The ash 
 is determined as above described. The acidity is ascertained by suspend- 
 ing 5 to 6 grms. of the powder in from 80 to 100 c.c. of water and titrating 
 with a ~jj standard solution of caustic soda. As the 'flowers' are only 
 with great difficulty wetted by the water, it is to be recommended to 
 weigh the quantity to be used for titration in a dry wide-necked flask, 
 and moisten the sulphur with a small quantity of alcohol. If now the 
 water is added, a good suspension is readily obtained. The total acidity 
 ascertained is calculated as sulphuric acid. How much acid is admissible 
 must, of course, somewhat depend upon the quality of the products for 
 the vulcanisation of which the sulphur is to be employed ; but considering 
 how greatly even a very small amount of sulphuric acid may affect the 
 resiliency of india rubber articles, not more than 0'2 per cent, should ever 
 be allowed, and for all those goods in which either high elasticity or 
 lightness of colour are important considerations, sulphur containing acid 
 should be altogether avoided. 
 
 In many cases flowers of sulphur contain a somewhat large amount of 
 moisture. This should be determined by drying in a desiccator at ordinary 
 temperature. Drying in a stove, owing to the not inconsiderable vapour 
 tension of sulphur at 100 C., is apt to give entirely erroneous figures. 
 
 I consider it further desirable to determine the proportions of soluble 
 .and of insoluble sulphur by treating for an hour 5 grms. of the sample 
 with 50 c.o. of carbon bisulphide, shaking the mixture at frequent intervals. 
 The solution is separated from the insoluble portion by filtration, and the 
 filter residue washed with a small volume of carbon bisulphide. The 
 filtrate and washings are collected in a weighed flask. After distilling off 
 the bisulphide and drying the flask while still hot by a moderate air blast, 
 the amount of soluble sulphur is ascertained by weighing. The insoluble 
 portion, of course, represents insoluble sulphur, ash, acids, and moisture. 
 
VULCANISERS AND SULPHUR CARRIERS. 179 
 
 It is surprising how tenaciously rubber manufacturers cling to the use 
 of flowers of sulphur for vulcanising purposes, considering that this 
 product is one of the most variable and impure forms of commercial 
 sulphur. 
 
 3. Milk of Sulphur. Under this name a product is offered which is 
 supposed to offer advantages over flowers of sulphur. All the samples I 
 have met with contain a more or less considerable amount 20 to 30 per 
 cent. of calcium sulphate. Very often this product contains also soluble 
 salts, in which case it is invariably strongly hygroscopic. As often as not, 
 free acid is present. Altogether, there is nothing in this preparation to 
 recommend it to the rubber manufacturer. 
 
 4. Precipitated Sulphur. This, undoubtedly the best of all forms of 
 sulphur for vulcanising purposes, is prepared by the precipitation of 
 sulphur from the solutions of polysulphides with hydrochloric acid under 
 closely defined conditions. It forms an impalpable powder presenting, 
 under the microscope, the appearance of very small globules of amorphous 
 sulphur (see fig. 11), but is completely soluble in carbon bisulphide. It 
 should not contain more than 3 per cent, of moisture, nor more than a 
 trace of ash. Neither should it contain any matter soluble in either water, 
 dilute hydrochloric acid, or dilute ammonia. 
 
 PAUL and HEINZ KRLING (see page 174) describe, under the names of 
 ' Paste, Metal, and Mixture,' certain vulcanising compounds which are 
 more or less complicated mixtures of quite a variety of heterogeneous 
 substances. These preparations are at best utterly worthless ; the prices 
 at which they are sold bear no relation to their intrinsic value, and, un- 
 fortunately, their continued existence is an undeniable proof that the india 
 rubber industry still harbours and encourages the technical superstitions of 
 the dark ages. 
 
 Chloride of Sulphur. This product, sulphur monochloride S.,C1 2 , is 
 extensively employed for the process of vulcanisation in the cold. The 
 closer the composition of the commercial product corresponds to the 
 theoretical figures calculated for S 2 C1 2 Cl, 52 -51 per cent. ; S, 47 '48 the 
 better. Commercial sulphur monochloride may contain either free sulphur, 
 of which it is an excellent solvent, or an excess of chlorine in the form of 
 the dichloride SC1 2 . Both impurities are known to occur not uncommonly. 
 While free sulphur, however, only becomes objectionable if its amount 
 exceeds certain limits, the presence of sulphur dichloride, in however small 
 a quantity, should not be tolerated. Excess of sulphur may cause an 
 efflorescence, technically termed 'sulphuring up,' of sulphur upon the 
 vulcanised surfaces, which, although objectionable from the point of view 
 of appearance, does not affect the soundness of the vulcanisation ; but the 
 presence of comparatively small quantities of the dichloride produces a 
 disagreeable and more or less pronounced hardness in the india rubber. 
 The cause of this is, of course, perfectly obvious, and lies in the excessive 
 chlorination of the india rubber. 
 
l8o INDIA RUBBER AND ITS ANALYSIS. 
 
 The commercial article is by no means very constant in its composition 
 and properties, and the products of different makers differ widely. Con- 
 sidering what was said above respecting the presence of either sulphur or 
 chlorine in sulphur monochloride, and further, bearing in mind the serious 
 difficulties which are likely to arise through unnoticed variations of the 
 chloride of sulphur used for either the vulcanisation of india rubber or 
 for the manufacture of white india rubber substitutes, the urgent necessity 
 for the continuous analytical control of the chloride of sulphur used in a 
 rubber factory is very obvious. It may safely be asserted that the 
 mistrust with which the cold vulcanising process is frequently regarded is 
 in a large measure due to the indiscriminate use of chloride of sulphur of 
 varying composition. 
 
 Sulphur monochloride in the pure state forms a reddish-yellow liquid. 
 Its specific gravity at 16-7 C. is 1'685 according to KOPP, 1-6826 according 
 to FAWSITT. There appears to be some uncertainty about the boiling 
 point, the temperatures stated by various authors ranging from 136 to 
 139 C. 
 
 The colour of the commercial sulphur monochloride varies 1 from a 
 bright amber to an almost ruby-coloured liquid, but none of these varieties 
 possesses a constant boiling point. The following are the boiling points of 
 two typical specimens of the commercial product, which are respectively 
 described as ' yellow product ' and ' red product/ : 
 
 100 c.c. yield Yellow Product. Red Product. 
 5c.c at 131 C. 128 C. 
 
 10 
 20 
 60 
 70 
 80 
 85 
 90 
 95 
 
 132 129 
 
 133 136 
 135 135 
 135 135 
 135-5 135-5 
 135-5 135-5 
 135-5 136 
 135-5 136 
 
 On analysis these two products were found to be composed as follows : 
 
 Yellow Product. Red Product. S 2 C1 2 (calcul. ) 
 
 Sulphur, . . 49*39 per cent. 46 '02 per cent. 47 "41 per cent. 
 Chlorine, . . 50 "60 53 '97 ,, 52 '52 
 
 ' These data show that an even slight excess of chlorin'e causes a con- 
 siderable depression in the boiling points of the lower fractions. But it 
 will also be observed that even the 'yellow product,' containing a slight 
 excess of sulphur, possesses a range of boiling points extending not in- 
 considerably below the true boiling point of sulphur monochloride. The 
 cause of this is by no means certain, but probably finds its explanation in 
 
 1 It will be clearly understood that the boiling of S 2 C1 2 is perfectly constant i.e., not 
 accompanied by any decomposition. HOHN'S contradicting statement (Oummi Zeitung 
 [14], 1899, p. 34) is entirely erroneous. The irregularity of the boiling point of the 
 commercial article is, of course, due to its impurity. 
 
VULCANISERS AND SULPHUR CARRIERS. l8l 
 
 the law of partial vapour tensions as exhibited by solutions of two volatile 
 bodies S 2 C1 2 and S. 
 
 It is scarcely ever necessary to make a complete analysis of sulphur 
 chloride. For practical purposes it is sufficient to determine the quantity 
 of one of its constituents, and determine the amount of the other by the 
 difference. This calculation is carried out by estimating the amount of 
 chloride of sulphur corresponding to the estimated percentage of that 
 constituent, the amount of which falls short of the theoretical figure. The 
 amount of chloride of sulphur thus ascertained is deducted from 99*99, and 
 the difference then shows the absolute excess of the constituent over the 
 theoretical figure. Thus we find, on comparison of the analytical figures of 
 the yellow product, as above quoted, with the theoretical figures given, that 
 it contains less than the theoretical amount of chlorine 50 '60 instead of 
 52*52 per cent. But 50-60 per cent, of chlorine represents 96*34 per cent, 
 of chloride of sulphur S 2 C1 2 , and deducting the latter amount from 99*99, 
 we obtain the absolute excess of free sulphur present as 3*65 per cent. In 
 exactly analogous manner the absolute excess of free chlorine present in the 
 above quoted 'red product' is ascertained as 3*05 per cent. 
 
 The analysis of chloride of sulphur is, according to FAWSITT/ carried 
 out in the following manner : 
 
 1. Sulphur. Small, weighted, thin-walled glass bulbs are filled with the 
 chloride of sulphur ; the points are then sealed in a gas flame for the pur- 
 pose of weighing. After ascertaining the weight of the chloride of sulphur, 
 the point is broken off, and the bulb immersed in 25 c.c. of concentrated 
 nitric acid in a small flask. If necessary, the rate at which the reaction 
 proceeds is moderated by cooling the flask. If sulphur separates out, this 
 is oxidised by theaddition of potassium chlorate. Oxidation being com- 
 pleted, the contents of the flask are emptied into an evaporating dish, the 
 flask washed with warm water, the washings added to the acid in the dish, 
 and the whole evaporated until all the nitric acid has disappeared. This 
 operation may be hastened by the addition of hydrochloric acid. The 
 residue is finally dissolved in water, and the amount of sulphuric acid 
 formed determined by weighing as barium sulphate. 
 
 2. Chlorine. One of the bulbs containing a weighed quantity of 
 chloride of sulphur is allowed to drop into a 250 c.c. glass-stoppered flask 
 containing wai^r or weak caustic soda solution. The contents of the flask 
 are repeatedly and violently shaken. The decomposition being completed, 
 the flask is filled up to the mark with water. An aliquot part of the 
 solution is then taken, neutralised with caustic soda, boiled for a few 
 minutes, and subsequently acidulated with nitric acid. Any sulphur 
 separating out is filtered off, and the hydrochloric acid present in the 
 nitrate determined in the well-known manner. 
 
 Of these two processes the first gives very satisfactory results, but the 
 operation is by no means a pleasant one. It becomes rather troublesome if 
 1 Journ. Soc. Chem.'Ind., 1886, p. 638. 
 
182 INDIA RUBBER AND ITS ANALYSIS. 
 
 sulphur, the oxidation of which is sometimes very tedious, separates out. 
 The method for the determination of the chlorine is very unsatisfactory. 
 Should the small bulb fail to break on shaking the flask, the completion of 
 the decomposition may require days. The process is very slow under all 
 circumstances, and cannot safely be accelerated by heating. Of course, as 
 I have shown above, it is quite sufficient to determine only the percentage 
 of sulphur contained in the chloride of sulphur, and to calculate from this 
 figure the chlorine. In this case FAWSITT'S above described method may 
 be used. 
 
 I prefer,, however, the determination of the chlorine by the following 
 method as a quicker and even more accurate process : Place in a weighed 
 200 c.c. flask (glass-stoppered) about 10 c.c. of the chloride of sulphur to be 
 examined, and ascertain the weight of this volume ; then fill up to the 
 graduation mark with pure benzene. Of this solution run 5 c.c. into 20 c.c. 
 of a solution of sodium ethylate contained in a wide-necked flask of about 
 150 c.c. capacity. The solution of sodium ethylate contains 5 grms. of 
 metallic sodium in 100 c.c. of absolute alcohol. The mixture is then very 
 gently shaken, allowed to stand for an hour on the water bath, evaporated 
 to dryness, and subsequently heated for one hour in a water oven. The 
 contents of the flask are then dissolved in about 75 c.c. of boiling water 
 faintly acidulated with nitric acid, 10 c.c. of a 10 per cent, solution of cupric 
 sulphate added, the whole heated for a few minutes on the water bath, 
 filtrated, and the filter residue washed with hot water. In the filtrate the 
 chlorine is then determined as silver chloride. The results obtained by 
 this method are very accurate. 
 
 The treatments of rubber with either sulphur or chloride of sulphur are 
 the only methods at present employed for vulcanisation. It is true other 
 methods have been suggested and often also practically tested, but none 
 have gained a foothold in manufacturing. GERARD'S process of treating 
 the india rubber for three hours in a solution of calcium pentasulphide 
 (1'205 sp. gr.) under a pressure of 60 pounds (140 C.) yields very satis- 
 factory results, but the method offers no advantage over the ordinary 
 processes. It is, nevertheless, worth remembering that the principle 
 involved in this process vulcanisation with molecular sulphur in a 
 nascent state would, if a substance could be found by means of which it 
 could be employed successfully, lead to very considerable and highly desir- 
 able improvements of the vulcanising process (cf. page 106). FAWSITT'S 
 process 1 is not so much a new vulcanising process, as one based upon 
 the employment of a new sulphur carrier, antimony iodide. For a variety 
 of reasons also, this process proved not more than a partial success, and 
 though much was expected of it, it never came into general use. My own 
 process 2 is a much simpler application of the same principle, and is also 
 rather cheaper. It allows the shortening of the time of vulcanisation by 
 almost one half, or else a reduction in the same proportion of the amount 
 1 Eng. Pat., No. 17,197, 1890. 2 Ibid., No. 16,937, 1900. 
 
VULCANISERS AND SULPHUR CARRIERS. 183 
 
 of sulphur employed. Either of these alternatives represents a very 
 important advantage, consisting either of a substantial saving of time i.e., 
 increased power of production or in a very large reduction of the amount 
 of free sulphur contained in the india rubber, thus entirely preventing the 
 objectionable appearance of the rubber 'blooming' or 'sulphuring up.' It 
 is, however, impossible at the present moment to say whether the process 
 will bear the test of time. 
 
 Lead Thiosulphate and other Lead Compounds. Various prepara- 
 tions of lead thiosulphate occur in commerce under the trade designation 
 of ' Hypo ' and ' Black Hypo.' A compound may be manufactured in the 
 wet way by precipitating a solution of lead nitrate with an equivalent 
 quantity of sodium thiosulphate. The product, if carefully prepared, 
 forms an almost white, micro-crystalline powder ; or a plumbago-coloured 
 product may be obtained by heating a mixture of white lead and sulphur. 
 The first of these two preparations is technically pure lead thiosulphate 
 (PbS 2 3 ) ; the second contains varying amounts of this salt, lead sulphate, 
 lead sulphide, and considerable quantities of unaltered white lead. To 
 the lead sulphide it owes, of course, its greyish- black colour. 
 
 ' Hypo ' is frequently advertised as an infallible agent for the prevention 
 of c sulphuring up ' of rubber goods, and as a vulcaniser, the use of which 
 does away with any risk of overcuring. Both statements are more or less 
 obvious exaggerations, and probably every experienced rubber manufacturer 
 has seen goods produced with * Hypo ' damaged by ' sulphuring up ' or by 
 overvulcanisation, or by both. 
 
 The truth is that * Hypo,' like all lead compounds, acts as a sulphur 
 carrier, and thus facilitates the vulcanisation ; and there is also no doubt that 
 the presence of lead sulphide has a restraining influence upon the further 
 addition of sulphur to rubber as soon as a certain degree of vulcanisation 
 has been reached. In this sense 'Hypo,' or, rather, the lead sulphide into 
 which it is converted during vulcanisation, has the property, within certain 
 limiting conditions, of reducing the risk of overvulcanisation. From 
 experiments I have made, it appears, moreover, that * Hypo,' or, at least, 
 that part of it which may be described as lead thiosulphate, is by itself 
 capable of producing vulcanisation. A series of experiments which I made 
 leave no doubt on this point, and the explanation of this fact is probably 
 to be found in the circumstance that one of the decomposition products of 
 lead thiosulphate under the influence of high temperature is sulphur : 
 
 4PbS 2 3 = PbS + 3PbS0 4 + 2S 2 . 
 
 Lead thiosulphate is, therefore, more properly described as a vulcaniser 
 than as a sulphur carrier, but in the exertion of its vulcanising properties 
 it forms at the same time, according to its above suggested decomposition, 
 two sulphur carriers lead sulphide and lead sulphate. Of these, as might 
 be expected, lead sulphide is much the more active. 
 
 The most generally used sulphur carrier is, however, neither lead 
 
184 INDIA RUBBER AND ITS ANALYSIS. 
 
 thiosulphate nor lead sulphide, but lead oxide, commonly known as 
 litharge. 
 
 While there is no doubt respecting the accelerating effect of these lead 
 compounds upon the rate and the amount of the sulphur addition in the 
 vulcanising process, the nature of this action is very little understood. It 
 is generally assumed that in the first instance lead sulphide is formed, and 
 that this exerts the above named accelerating effect by the intermediate for- 
 mation of unstable polysulphides. But even taking this view as correct, this 
 assumption cannot be said to be an explanation. Indeed, the very difficulty 
 of an explanation of this ' catalytic ' action lies in the fact that the energy 
 transformation due to a certain reaction remains absolutely the same, how- 
 ever many intermediate phases or reactions we interpolate between the 
 limiting terms of the reacting system. 
 
 To the rubber manufacturer these sulphur carriers are of very con- 
 siderable importance, their use enabling him to increase the degree of 
 vulcanisation within considerable limits without recourse to either an 
 objectionable excess of sulphur, undesirably high temperatures, or very 
 undue prolongation of the vulcanising process. The beneficial effect of 
 these sulphur carriers in these respects is so marked that at the present 
 time the majority of rubber goods in which a white or drab colour is not 
 essential contain more or less of one of the above named substances. 
 Litharge especially is very freely used for this purpose. 
 
 Whether litharge acts in the india rubber as a sulphur carrier only is 
 by no means certain. That it may act also in other ways is to be inferred 
 from the observation, which is quite familiar to most india rubber workers, 
 that mixtures of india rubber with a large proportion of litharge are apt to 
 become very hot during their treatment between the mixing rollers. In 
 extreme cases the mixture, in spite of its high temperature, grows con- 
 tinually harder, until eventually it forms a mass of crumbs or flakes 
 possessing little cohesion and adhesiveness, and which can no longer be 
 rolled into a sheet. The cause of this behaviour is unknown, but will 
 probably be found in the incontestible fact that india rubber absorbs a not 
 inconsiderable amount of oxygen while it is being worked upon the mixing 
 rollers. This oxidation process is intensified and accelerated by the litharge 
 presentj which is known to act as an oxygen carrier. Certain it is that none 
 of the other lead compounds 1 used in conjunction with india rubber 
 exhibit this action in the slightest degree. 
 
 A complete analysis of the above named lead compounds may, of course, 
 under certain circumstances become necessary, but is not necessary for 
 ordinary purposes. The following tests will usually suffice : 
 
 Lead Thiosulphate. Composition PbS 2 3 , containing 64 '89 per cent, 
 of lead. Commercial samples should not contain less than 60 per cent, of 
 Pb. This is ascertained by boiling about 1 grm. of the sample with dilute 
 
 1 Not even white lead, which, according to its approximate formula (PbC0 3 ) 2 Pb(OH).;, 
 contains about 31 per cent, of lead hydroxide. 
 
VULCANISERS AND SULPHUR CARRIERS. 185 
 
 nitric acid (10 per cent. HN0 3 ) until the precipitate is perfectly white, and 
 the supernatant liquor free from lead. The precipitate is then filtered, 
 washed neutral, dried, removed from the filter, placed in a crucible for 
 about fifteen minutes heated to a temperature not exceeding 130 C., and 
 subsequently weighed. The yield of PbS0 4 thus obtained should not be 
 less than 88 per cent. Lead thiosulphate should be almost completely 
 soluble in a 20 per cent, solution of sodium thiosulphate at a temperature 
 not exceeding 50 C. 
 
 Black Hypo. This product is extremely variable in its composition, 
 representing generally a mixture of lead carbonate, sulphide, thiosulphate, 
 sulphate, and free sulphur. Occasionally the product bearing this name is 
 simply a mixture of lead sulphide with sulphur. For the purpose of ana- 
 lytical control, a determination of the total lead and of the free sulphur 
 present is sufficient, but what there should be in amount it is impossible to 
 say. The analyses will therefore have to be made in reference to an 
 adopted standard. 
 
 Litharge, PbO. If it is thought desirable, the total lead present may 
 be determined by dissolving a weighed quantity in nitric acid, with 
 addition of a crystal of oxalic acid, precipitating the filtered solution with 
 sulphuric acid, and weighing the lead sulphate precipitated. Litharge 
 should always be tested for the absence of lead peroxide and copper. 
 Dissolve a small quantity of the sample in as little nitric acid as possible. 
 A dark, blackish-brown residue dissolving after the addition of a small crystal 
 of oxalic acid indicates the presence of lead peroxide. The solution obtained 
 is now treated with sulphuric acid in slight excess, the lead sulphate pre- 
 cipitated is filtered off, and an excess of ammonia added to the filtrate, 
 when the presence of copper is shown by the appearance of the well-known 
 blue coloration. 
 
 Red Lead is frequently used in small quantities in rubber mixings, with 
 a view to * harden ' the rubber. I do not consider this as a satisfactory 
 addition, as this hardening effect is obviously due to a partial oxidation of 
 the india rubber, which appears to me to be a particularly undesirable 
 manner of obtaining such an effect, which can, moreover, be produced by a 
 variety of other means in quite an unobjectionable manner. 
 
 Antimony Pentasulphide. Antimony pentasulphide, Sb 2 S 5 , com- 
 mercially known as 'Golden Sulphide,' is the most important and the most 
 efficient sulphur carrier used in conjunction with india rubber. It is 
 looked upon by many, not only as a sulphur carrier, but as a vulcaniser 
 itself. This view is, however, based upon undoubtedly erroneous observa- 
 tions. The pure pentasulphide is quite incapable of effecting the vulcan- 
 isation of india rubber. Commercial pentasulphides will indeed vulcanise 
 india rubber without special addition of sulphur, but this is simply due to 
 the fact that they all contain free sulphur in quantities ranging from 5 to 
 as much as 30 per cent. This free sulphur, at any rate as far as the 
 lower percentages are concerned, cannot be looked upon as a mere 
 
1 86 INDIA RUBBER AND ITS ANALYSIS. 
 
 adulteration, for, without entering into any details of its manufacture with 
 which in this place we are not concerned, there is no doubt that the amount 
 of free sulphur in a given sample of pentasulphide plays a determining 
 part in the production of a pentasulphide of a certain colour. 
 
 The colour of golden sulphide varies from a full orange-tan colour to 
 almost a purplish-scarlet, German rubber manufacturers preferring the 
 former, their French colleagues the latter variety, whilst in English 
 factories an intermediate shade is almost exclusively used. 
 
 The properties of antimony pentasulphide are rather interesting. 
 Heated to the boiling point of sulphur, 448 C., with exclusion of the 
 air, it decomposes into antimony trisulphide, Sb 2 S 3 , and sulphur. Treat- 
 ment with hydrochloric acid in the cold produces exactly the same 
 decomposition. Ammonia dissolves it on gentle heating, any antimonious 
 acid, Sb 2 3 , and free sulphur present remaining undissolved. The be- 
 haviour of different samples of pentasulphide with boiling carbon bisulphide 
 varies very considerably, in some cases reduction to Sb. 2 S 3 taking place. 
 Equally marked is the behaviour of different samples on drying them at 
 110 C. 
 
 Considering the fact that even in products of the same colour the per- 
 centage of free sulphur present may exhibit considerable variations, it is 
 evident that, apart from the question of colour, all samples and deliveries 
 of antimony pentasulphide should invariably be analysed. The neglect of 
 this precaution is likely to cause most objectionable variations and difficulties, 
 in the vulcanising process. 
 
 The most obvious method for the determination of the amount of free 
 sulphur present by the extraction of a weighed quantity of the sample with 
 bisulphide of carbon cannot be employed in this case, owing to the more 
 or less pronounced decomposing action of this solvent upon the penta- 
 sulphide. Short of the rather tedious complete analysis, the most satis- 
 factory test consists in treating a weighed quantity of the sulphide with 
 strong ammonia, and heating very gently on the water bath. The penta- 
 sulphide rapidly dissolves any antimonious acid, the sulphur present 
 remaining behind. The solution is filtered, the filter residue washed with cold 
 dilute ammonia until the washings, 011 acidulation, no longer yield a deposit 
 of pentasulphide. The filter and its contents are dried at 60 C. and treated 
 with carbon bisulphide in a Soxhlet extractor, the extract being collected 
 in a weighed flask. The residue remaining on the filter after extraction is 
 antimonious acid, which is removed from the filter, dried, and weighed. Of 
 course any adulterants would likewise remain undissolved on the filter, but 
 it is very rarely that such adulterants occur. 
 
 Zinc Sulphide and Lithopone. Zinc sulphide is an excellent material 
 to compound with india rubber, but its high price on the one hand, and its 
 yellowish colour on the other hand, restrict its use very considerably. It 
 is, however, very largely used in the form of the brilliantly white pigment 
 known as ' Lithopone.' This product is formed by the simultaneous pre- 
 
VULCANISERS AND SULPHUR CARRIERS. 187 
 
 cipitation of barium sulphate and zinc sulphate by the interaction of 
 barium sulphide and zinc sulphate : 
 
 Ba(SH) 2 + ZnS0 4 = BaS0 4 + ZnS + H 2 S. 
 
 Accordingly, lithopone should contain at least 29 per cent, of zinc 
 sulphide, and no product containing less should be accepted, as this would 
 involve a decrease both in covering power and efficiency in the vulcanising 
 process. 
 
 Lithopone is by far and away the best material to use in ' drab ' 
 rubber goods which are subjected to the action of alkalies, but should be 
 rigorously excluded from all goods subjected to the influence of acid 
 solutions. 
 
CHAPTER VI. 
 INDIA RUBBER SOLVENTS. 
 
 THE importance of solvents for India rubber in manufacturing has been well 
 put forward by HANCOCK J himself, who notes that "it is a singular 
 fact that, although this substance had attracted the notice of chemists 
 from the earliest date of its importation into Europe, and some of the 
 ablest had employed themselves upon it, they failed to discover any means 
 of manufacturing it into solid masses or to facilitate its solution. I was 
 at first imbued with the notion that, to make it useful, I must find a good 
 solvent ; and I think my first experiments were directed to some attempts 
 to dissolve it in oil of turpentine, but I found I could only make very thin 
 solutions, and these dried so badly, or, rather, not at all, that they were 
 useless. The oil of turpentine then procurable was no doubt of inferior 
 quality; when pure, it dries perfectly. This was about the year 1819." 
 
 As the result of this failure to discover a satisfactory solvent for india 
 rubber, which would have enabled HANCOCK to utilise the rubber scraps 
 which were fast accumulating at his works, he attempted to find a mechan- 
 ical treatment to attain this end, and already, a year later, in 1820, he 
 accomplished this important task by the construction of his 'masticator.' 
 It was, however, not until two years later that he discovered that this 
 masticated rubber dissolved with the greatest facility in * good oil of 
 turpentine.' 
 
 These earliest experiences show already very clearly the fact that the 
 solubility of india rubber depends as much upon its physical condition as 
 upon the nature of the solvent. It may, indeed, be stated that a homo- 
 geneous rubber solution cannot be produced without either first changing 
 the physical condition of the india rubber by mechanical working, or by 
 mechanically working the gelatinous magma produced by the treatment of 
 unmasticated india rubber with solvents ; and to obtain the best results, 
 it is necessary to work the india rubber mechanically prior to its immer- 
 
 1 Personal Narrative of the Origin and Progress of the Caoutchouc or India Rubber 
 Manufacture in England. By Thomas Hancock, of the firm of Charles Macintosh & Co., 
 London and Manchester. London: Longman, Brown, Green, Longmans, & Roberts, 
 1857 (p. 2). 
 
INDIA RUBBER SOLVENTS. 189 
 
 sion in solvents, and to subject this crude solution to a farther process 
 in machines constructed upon the familiar principle of the paint-grinding 
 mill. The necessity for these mechanical operations lies, of course, in the 
 fact that all varieties of india rubber appear to contain, besides the soluble 
 india rubber, a further product (see page 8), which, although insoluble, 
 swells to enormous dimensions in the rubber solvents. As long as this 
 substance is present in the form of particles of appreciable size, the ' rubber 
 solution ' will appear lumpy or thready, and the effect of the mechanical 
 working consists simply in reducing these insoluble masses to particles of 
 imperceptible size. This is done far more effectively by the mastication of 
 the rubber than could be accomplished by the working of it after solution. 
 It will be observed that this insoluble compound behaves in this respect 
 much as the 'fatty substitutes,' which cannot be satisfactorily introduced 
 into a rubber solution or dough after the rubber has once been dissolved^ 
 but have to be reduced to imperceptibly small particles by being 'dry 
 mixed ' into the rubber in a roller mill. At the same time, there can, 
 however, be no doubt that the trituration of the insoluble constituent i& 
 not the only effect of the mastication of india rubber. The India rubber 
 hydrocarbon itself undergoes some peculiar modification in this process 
 which certainly renders it easier soluble, and which is also capable of 
 profoundly altering its behaviour in the process of vulcanisation. Nothing 
 is known respecting the cause of this remarkable phenomenon, and equally 
 striking is the fact that the rubber films resulting from the evaporation of 
 the solvents from coatings spread with solutions of india rubber in differ- 
 ent solvents exhibit striking differences in their chemical properties. Thia 
 applies even to solvents which may be considered to be closely related 
 chemically to each other, such as the various terms of the hydrocarbons 
 of the aromatic and aliphatic series respectively. In practice, this would 
 mean that we cannot indiscriminately substitute one solvent for another, 
 and leave the other manufacturing operations out of consideration, without 
 some change in the state and properties of the manufactured products. 
 This is a point which at the present time is very little appreciated. 
 
 1. Coal Tar Naphtha. This is by far and away the most generally 
 used india rubber solvent, and it is unquestionably also the best. Its 
 composition is subject to very great variations, which chiefly depend upon 
 the demand for benzene and toluene on the part of the coal tar colour in- 
 dustry, and as far as benzene is concerned, also on the demand of that article 
 by gas works. It goes without saying that the price of the 'solvent 
 naphtha ' also closely follows the benzene market indeed, this extends even 
 to the prices of petroleum spirit and Scotch shale naphtha. It is therefore 
 obvious that not only will the price of coal tar naphtha, but also its com- 
 position, vary considerably according to the demand for various specific 
 fractions of the coal tar hydrocarbons. 1 Not infrequently are the varia- 
 
 1 The effect of the excessively low prices, which, during the greater part of the last 
 two years, were quoted for benzol, furnish an instructive example of this. At the time 
 
INDIA RUBBER AND ITS ANALYSIS. 
 
 tions due to higher prices to the advantage of the rubber manufacturer, 
 those due to lower prices being to his detriment. Tn either case it is desir- 
 able, in the interest of uniform production, to test the boiling points of 
 solvent naphtha deliveries. 
 
 Of the aromatic hydrocarbons occurring in coal tar, of course, a limited 
 number only can, or should be, present in a good solvent naphtha, which 
 should not contain more than 5 volumes per cent, of substances boiling 
 above 150 or 160 C. preferably the lower limit. Of aromatic hydro- 
 carbons, only the following can therefore occur : 
 
 Boiling Point. 
 
 Benzene, C 6 H 6 , .... 81 C. 
 
 Toluene, C 7 H 8 , . . . . Ill 
 
 Orthoxylene, ^ 1 141 
 
 Metaxylene, lc 8 H 10 , . . . . | 140 
 
 Paraxylene, J 1 138 
 
 Styrene, C 8 H 8 , .... 146 
 
 Mesitylene, C 9 H 12 , . . . . 163 
 
 Pseudocumene, C 9 H 12 , . . . . 166 
 
 It is, of course, also possible for aliphatic and olefinic hydrocarbons to be 
 present in solvent naphtha, but their quantity will always be very small. 
 The customary methods of refining solvent naphtha, if properly carried out, 
 would, at any rate, tend to completely eliminate the olefinic hydrocarbons. 
 
 The above named series of aromatic hydrocarbons represents a range of 
 boiling points from 81 to 166 C., and this would fairly correspond to a 
 solvent, 95 per cent, of which boils at 160 C., some part of the higher 
 boiling fraction being carried over at temperatures at and below 160 C. in 
 the ratio of their vapour tensions at these temperatures. The evaporation 
 of benzene and toluene at ordinary temperatures is, however, so considerable 
 that the presence of large volumes of these hydrocarbons in the solvent 
 naphtha would cause considerable inconvenience and loss at the early stages 
 of manufacturing, and for this reason good solvent naphtha should contain 
 no benzene and only very little toluene. In short, the initial boiling point 
 of solvent naphtha should, in my opinion, be not lower than 110 C., and it 
 is highly desirable that 95 per cent, of such a naphtha should distil only a 
 few degrees above 150 C. the fewer, the better. Accordingly, the range 
 of boiling points of a good naphtha should not appreciably exceed 40 C. It 
 will be found that a solvent of this description answers admirably for the 
 most important purposes of the rubber manufacturer. For the production 
 of a few specialities, such as quick drying 'solution' or 'varnish,' it is 
 scarcely volatile enough. Nor should it for this purpose be mixed with 
 benzene, as is often done under the mistaken impression that this renders 
 
 benzol was 7d. per gallon, 'solvent naphtha' had fallen to lid., and it was therefore 
 greatly to the interest of the distiller, though not of the rubber manufacturer, to send 
 out in the solvent as much benzol as possible. The result of this was that solvents 
 which formerly used to show an initial boiling point of 107 C., now began to boil at 
 85 C., or lower. 
 
INDIA RUBBER SOLVENTS. IQI 
 
 the solvent quick drying, but, as a matter of fact, only renders the drying of 
 the solution more erratic. The only satisfactory procedure is in this case 
 to employ a differently constituted solvent say, one boiling from 110 to 
 140 C. In other words, if for some purpose or other we require a 
 ' lighter ' solvent, the usually adopted method for procuring it by mixing it 
 with fractions reducing only the initial boiling point of the original solvent 
 is, at best, only a rather inefficient make-shift. To ' lighten ' a solvent, its 
 terminal boiling point should be lowered by keeping out the fractions of 
 highest boiling point. The constriction of the range of boiling points thus 
 effected may then be compensated by the addition of fractions of a lower 
 boiling point. 
 
 The use of almost pure benzene as a rubber solvent for manufacturing 
 purposes has sometimes been suggested, but apart from the fact that its 
 boiling point is altogether too low, and its volatility far too high, for most 
 practical purposes, the use of a homogeneous solvent possessing almost a 
 constant boiling point results in solutions or ' doughs ' drying from the 
 surface and yielding harsh and rough coatings. Waterproof cloth produced 
 with such solvents often exhibits the very objectionable defect of curling 
 i.e., the cloth curls towards the proofing surface. The cause of this 
 phenomenon lies in too ' light ' a solvent being employed. The coating of 
 india rubber dough on the cloth dries from the surface, the latter therefore 
 hardening and contracting, while the lower strata, still containing the 
 solvent, cannot contract proportionately. This differentiation in the 
 stresses of the different strata of the rubber coating becomes permanently 
 established if the goods are cold cured, and the inevitable result is a 
 curling or cockling waterproof cloth. This is the simple explanation of a 
 defect which has numberless times caused great annoyance and loss to 
 manufacturers, and the occurrence of which is generally looked upon as a 
 great mystery. 
 
 The specific gravity of solvent naphtha is, on the whole, a rather im- 
 material point bearing little or no relation to its boiling point. Beside the 
 latter, it is of no importance whatever, unless, indeed, the solvent be bought 
 by weight, which is very unusual. As a knowledge of the specific gravity 
 of a solvent gives no tangible information respecting its quality, this 
 constant calls for no special observations. 
 
 The usual method observed in rubber factories for ascertaining the 
 boiling point of a given sample of solvent naphtha consists in distilling 
 100 c.c. in a fractionating flask over a naked flame at the rate of two drops 
 per second. The distillate is collected in a graduated cylinder (100 c.c.), 
 and the successive temperatures, and the length of time during which each 
 fraction distils over, are noted. The following table contains a number of 
 such tests selected from many hundreds of records. There is, however, a 
 second method, which, as a commercial test, is much used. In this method 
 a'tubulated retort of about 250 c.c. capacity is used charged with 100 c.c. of 
 the naphtha, and a thermometer is so fixed in the tubulus that its bulb is 
 
INDIA RUBBER AND ITS ANALYSIS. 
 
 exactly within three-eighths of an inch from the bottom of the retort. 
 Accordingly, the thermometer is immersed in the liquid, and the tempera- 
 ture noted at which the first drop of the distillate passes over into the 
 receiver. The first method is certainly preferable from the rubber 
 manufacturer's point of view. The second method is altogether antiquated ; 
 it is open to a number of grave objections, although the English benzol 
 trade adheres to it with its characteristic conservatism. 
 
 The most satisfactory method for ascertaining the boiling point of coal 
 tar naphtha is one which was first proposed by BANNOW in 1886, and 
 which has been very generally adopted. The apparatus used consists of 
 the following parts : 
 
 1. An almost spherical flask of 150 c.c. capacity made of copper 0*6 
 to 0*7 mm. in thickness. Its diameter amounts to about 66 mm. The form 
 and dimensions of the neck are clearly shown by the accompanying fig. 16. 
 
 FIG. 16. Diagram of distilling flask for benzene (naphtha) test. . 
 
 2. A dephlegmator made of glass, and consisting essentially of a glass tube 
 of about 14 mm. diameter and 150 mm. in length, the middle part of which is 
 expanded into a bulb of moderate size. Near its upper end this tube carries 
 a side tube of about 8 mm. diameter, the two joining at nearly a right angle, 
 as shown in fig. 17. 
 
 The distilling flask is mounted on a draught-proof stand, the construc- 
 tion of which is clearly indicated in the diagram. The top of this stand is 
 formed by a stout asbestos board in which a circular central opening 50 mm. 
 in diameter is provided. This distilling apparatus is joined to a condenser 
 80 cm. in length, and so inclined that the difference of level between its two 
 ends is 1 cm. 
 
 The flask is filled with 100 c.c. of the naphtha, and the distillation is so 
 conducted that about 5 c.c. of distillate are obtained per minute, or approxi- 
 mately two drops per second. Distillation is continued until 95 c.c. have 
 passed over, the distillate being collected in a graduated 100 c.c. cylinder. 
 
 The thermometer used should be made of thin glass, and its diameter 
 should not exceed 7 mm., although it may be less. The bulb of the thermo- 
 meter should occupy the centre of the bulb of the dephlegmator. 
 
INDIA RUBBER SOLVENTS. 
 
 193 
 
 By means of this apparatus the boiling points of a number of rubber 
 solvents recorded on the following page were obtained. 
 
 Of these samples, No. 1 is a little too high in boiling point, but other- 
 wise satisfactory. No. 2 is, however, distinctly preferable. Nos. 3, 4, 6 r 
 and 7 refer to tests of a special brand of solvent naphtha, which is 
 extensively used for waterproofing purposes. No. 3 is a thoroughly good 
 type of this solvent, but, of course, there is, in employing it, a rather consider- 
 able loss amounting, on the average, to about 10 per cent, before it reaches 
 the spreading machine. With the solvents 1 and 2 this loss is very slight 
 indeed. No. 4 shows what the solvent No. 3 degenerated into at the time 
 benzol had reached an unprecedentedly low price. The test shows that at 
 least 50 per cent, of the solvent, and probably a good deal more, consisted 
 simply of benzol. Seeing that, at the same time, the price of this ' special 
 
 FIG. 17. Diagram of distillation apparatus for benzene (naphtha) test. 
 
 solvent' was more than twice that of the benzol, it is evident that this 
 adulteration was highly profitable to the maker, and proportionately 
 detrimental to the user. No. 6 has throughout rather too low a boiling 
 point, and this applies even more to No. 7, the range of boiling points of 
 . which is far too short, and much too low at the higher end. 
 
 A solvent naphtha furnishing a satisfactory boiling point test cannot, 
 however, for that reason at once be passed as meeting all the requirements 
 of the rubber manufacturer. Such a solvent may yet contain a number of 
 impurities, the presence of which might not, and, indeed, in the majority of 
 cases would not, appreciably affect its boiling point, but would, neverthe- 
 less, greatly prejudice its adoption. 
 
 The most important impurities are : carbon bisulphide, naphthalene, 
 pyridiue and its homologues. Adulteration of solvent naphtha occurs very 
 rarely, and the only possible adulterant is petroleum spirit. 
 
 The detection of carbon bisulphide may be carried out by means of the 
 xanthate test. For this purpose 100 c.c. of the naphtha are distilled, and 
 
 13 
 
194 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 d 
 
 ?0 
 
 os 
 
 d 
 10 
 
 CO 
 
 iO iO 
 
 CO CO CO CO 
 
 IO CO ^ O 
 
 COt>-l>OOrHi lOSrHCOrH 
 COrHlOlOCOCOlOr^^lO 
 
 rt< OS 
 iO iO 
 
 
 
 
 
 
 d 
 
 d 
 
 <* CO CO 
 
 iO CO * OS 
 
 ip 
 
 COlOCOCO^fCOt~COiOCO 
 CO rH rfl "^ ^ ^1 ^ CO CO "^ 
 
 J* ^ 
 
 
 
 
 
 
 o" 
 
 d 
 
 iO iO 
 
 
 
 00 
 
 in 
 
 IO CO CO OS 
 
 COrHCOCOCOCOCOCOCOCO 
 
 CO CO 
 
 
 
 
 
 
 d 
 b 
 
 0* 
 p 
 
 o 
 
 ^ co co *> 
 
 rt< CO CO 00 
 
 COCOCOCOOSOOSiOOCO 
 rHrHCOrHrHCOrHCOCOCO 
 
 
 
 
 
 
 
 
 d 
 
 d 
 
 
 
 OS OS rH CO 
 CO rH CO CO 
 
 -*icooo-*cococooscoo 
 
 rHr-HOSOSOSOSOSrHrHCO 
 
 CO IO 
 rH rH 
 
 
 
 
 
 
 d 
 
 o 
 
 CO 
 
 d 
 
 i>. OS rH CO 
 CO O OS CO 
 rH rH 
 
 lOCOCOrHrHlOCOrHfr-O 
 O O -t*** ^^ t** ^* ^* OS CO O 
 
 rH rH rH 
 
 ip 
 
 CO CO 
 
 OS OS 
 
 d 
 
 o 
 
 1 1 
 
 d 
 
 IO 
 
 ^ i>. !> CO 
 CO O CO CO 
 rH rH 
 
 COiOGOrHO^^^rHCO 
 OOCOVO1O1O1OCOCOOS 
 rH rH 
 
 !> rH 
 
 CO 00 
 
 d 
 
 o" 
 
 
 rH 
 CO 
 rH 
 
 V s 
 
 rH IO CO CO 
 CO O CO CO 
 
 rH rH 
 
 rHCOCO-^rHCOOOOSO 
 OOlOCOCO-*-*COJ>.OO 
 rH rH 
 
 s s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 | 
 
 
 
 rtf 
 
 
 1 
 
 
 
 1 
 
 
 I 
 
 .- 
 
 1 
 
 ef -*^ 
 
 | -c 
 
 
 
 1 
 
 | 
 
 rt 
 
 IH 
 
 a 
 
 5 -'o, 
 
 w ' "" 
 
 5 g ^ 
 
 ^ i 
 
 
 1 
 
 = 1 = 8, 
 
 OQ OS 
 
 O2 PH O2 
 
 - = 
 
 
 rH 
 
 CO CO * iO 
 
 COt^OOO4OrHOicO-*lO 
 
 S r^ 
 
 
 
 
 
 
INDIA RUBBER SOLVENTS. 1 95 
 
 the first 10 c.c. of the distillate are shaken with 10 c.c. of standard alcoholic 
 potash, slightly warmed, and allowed to stand for about an hour. Part of 
 the alcoholic solution is then moderately diluted with water, and a solution 
 of lead acetate added. A white precipitate is obtained in the absence of 
 carbon bisulphide, its presence being indicated by the formation of a yellow 
 or yellowish-brown precipitate. This test is, however, not very sensitive. 
 More satisfactory is the sulphocyanide test. It is carried out by treating 
 the first 10 c.c. of the distillate as above mentioned with 10 c.c. of alcoholic 
 ammonia, shaking the mixture, and heating gently. The whole is then 
 evaporated to dryness, the residue taken up with a little water, and with 
 ferric chloride, tested for the presence of sulphocyanides. This test is much 
 more delicate than the first named, but, of course, scarcely suitable for the 
 determination of the quantity of carbon bisulphide present. This is best 
 carried out by means of the hydrazine test, which is also the most satis- 
 factory test for the ready detection of even minute traces of carbon 
 bisulphide. 1 In carrying out this test, a few drops of phenylhydrazine 
 (free base) are added to about 10 c.c. of the solvent, and the mixture is 
 allowed to stand for about two hours. As long as the quantity of carbon 
 bisulphide present is not less than 0*02 per cent., a more or less abundant 
 crystallisation of phenylhydrazine phenylsulphocarbazinate 
 
 N 2 H 4 .C 6 H 6 
 
 is obtained. The crystals are filtered off into a weighed Gooch crucible by 
 means of a vacuum pump, 25 c.c. of pure benzol being used for the purpose 
 of washing, and the crucible is then dried for one hour in a vacuum 
 desiccator. The weight of the precipitate may then be ascertained by 
 weighing crucible and contents. By multiplying the quantity of the 
 phenylhydrazine compound thus ascertained by the factor 0'26, the 
 quantity of carbon bisulphide contained in 10 c.c. of the solvent is obtained. 
 This test is much more sensitive than the disagreeable test with triethyl- 
 phosphine. 
 
 The presence of carbon bisulphide in solvent naphtha is objectionable 
 only from the hygienic point of view. As small a quantity as O'l per cent. 
 is quite sufficient to make itself noticeable in the atmosphere of even a well- 
 ventilated spreading room, and the baneful effects of carbon bisulphide 
 upon the workers render the absence of this substance in the solvent 
 naphtha absolutely imperative. There is, moreover, no excuse for its 
 presence in solvent naphtha, possessing, as it should do, an initial boiling 
 point considerably above 100 C. 
 
 Naphthalene, which was at one time a very frequent impurity in 
 solvent naphtha, now very rarely occurs. Its presence in any appreci- 
 able quantity is highly objectionable on account of the very persistent 
 1 LIEBERMANN and SEYEWETZ, Ber., 24, 788. 
 
196 INDIA RUBBER AND ITS ANALYSIS. 
 
 strong smell it imparts to the manufactured goods. The methods for its 
 detection are obvious. 
 
 Pyridine and its higher homologues occur with surprising frequency in 
 solvent naphtha, apparently often escaping the refining process with 
 sulphuric acid, They are objectionable, in the first instance, on account of 
 the extremely acrid and irritating odour they impart to the atmosphere of 
 the spreading rooms, and also on account of their undoubtedly detrimental 
 action upon india rubber. For hygienic and technical reasons it is there- 
 fore equally desirable to possess a ready means of detecting their presence 
 in solvent naphtha. This is much more difficult than would at first appear, 
 chiefly owing to the fact that it is practically impossible to completely 
 extract pyridine and its higher homologues from solvent naphtha by means 
 of dilute acids, 1 and the difficulty is further increased by the circumstance 
 that the evaporation of dilute aqueous solutions is impossible without total 
 loss of the pyridine. The following technical test gives, however, entirely 
 satisfactory results, and is of quite sufficient sensitiveness for the purpose. 
 
 100 c.c. of the solvent are energetically shaken with 100 c.c. of water in 
 a separating funnel, and the liquids are then allowed to separate. About 
 15 c.c. of the water are then filtered through a dry filter into a test tube, 
 a few drops of a 10 per cent, solution of cupric sulphate added, and the 
 contents of the test tube heated to boiling point. If the liquid remains 
 perfectly clear, the solvent is either free from pyridine, or its quantity does 
 not reach 1 part in 1500 parts of naphtha (0*06 per cent.). Anything 
 from the faintest to a pronounced bluish-white turbidity, but not leading 
 to an appreciable precipitate of the same colour, after thirty minutes' 
 standing, indicates an amount of pyridine ranging in proportion from 
 1 : 1500 to 1 : 1000 (0'06 to O'l per cent.). If the amount of pyridine 
 present at all exceeds O'l per cent., a more or less copious precipitate is 
 obtained. If the amount of pyridine present exceeds 0'3 per cent., 
 turbidity is at once produced in the aqueous extract on adding to it the 
 cupric sulphide. The sensitiveness of this test is exactly coterminous with 
 that of the normal sense of smell for pyridine i.e., a solvent naphtha, 
 which, according to this test, would be passed as free from pyridine, may, 
 as a matter of fact, contain up to 0*04 or 0*05 per cent, this body ; but 
 if so, it cannot be detected either by this test or by the physiological 
 properties of the naphtha as experienced in the manufacturing departments. 
 The presence of so small a quantity as 0'06 per cent, is, however, clearly 
 recognisable by this test, and equally so in the atmosphere of the spreading 
 room. 
 
 Adulterations of solvent naphtha by either petroleum spirit or shale 
 spirit are, no doubt, best avoided by purchasing direct from the producers. 
 They betray themselves in most cases already by an impossibly low initial 
 boiling point, and likewise by the lower specific gravity. If a sample of 
 
 1 This, no doubt, finds its explanation in the fact that dilute aqueous solutions of the 
 pyridine salts are almost completely hydrolysed. 
 
INDIA RUBBER SOLVENTS. 197 
 
 solvent naphtha is suspected in this respect, about 50 c.c. of it should be 
 placed in a flask of about 200 c.c. capacity, and connected with a reflux 
 condenser. The flask is well cooled, and a mixture of 50 c.c. of concentrated 
 nitric acid with 50 c.c. of concentrated sulphuric acid slowly added, gently 
 shaking from time to time. As soon as the, at first rather brisk, reaction 
 begins to slacken, water-bath heat is applied for fifteen minutes. The mass 
 is then allowed to cool, carefully diluted with cold water, transferred as 
 completely as possible into a separating funnel, and allowed to settle. The 
 whole of the lower layer is run into a fractionating flask of about 250 c.c. 
 capacity, an excess of aqueous caustic soda is added, and the mass distilled 
 in a current of steam until about one-third of the nitrated bodies have 
 passed over. This operation is repeated with the whole of the distillate 
 for a second time, using a small separating funnel as the receiver, when 
 the unnitrifiable hydrocarbons will be found floating upon the water. 
 The latter is drawn off as completely as possible, and the volume of the 
 unnitrifiable hydrocarbons ascertained by running them into a small 
 graduated cylinder. It must, however, be remembered that even 90's 
 benzol contains invariably from 1 to 2 per cent, of such hydrocarbons, 
 and solvent naphtha from 2 to 6 per cent. 
 
 2. Petroleum Spirit. The fractions boiling below 160 C. which are 
 obtained under the name of ' Deodorised Petroleum Spirit ' come into the 
 market from the distillation of American crude petroleum. This solvent, in 
 its generally very high state of purity, is used extensively as an India rubber 
 solvent, generally in conjunction with a greater or smaller proportion of 
 coal tar naphtha. Its use without the latter is not satisfactory, as its 
 solvent properties are greatly inferior to that of coal tar naphtha, and also 
 on account of its far greater volatility. Against coal tar naphtha, petroleum 
 spirit has the advantage of being almost odourless, or practically so a 
 point which, especially in the manufacture of waterproof cloth, is rather 
 important. 
 
 Deodorised petroleum spirit consists almost entirely of hydrocarbons 
 of the aliphatic series C,,H 2n+2 . On submitting it to the boiling point, 
 test figures are obtained, of which samples 9, 10, 11, and 12 in the tables 
 already given are good types. As a matter of fact, however, it will be 
 found that petroleum spirit contains not inconsiderable quantities of much 
 lower boiling hydrocarbons. Indeed, in fractionating large quantities of 
 petroleum spirit (500 gallons at a time), a very large volume of vapours of 
 extremely low boiling point is given off at first, followed by a distillate 
 amounting to about 10 per cent, of the whole, the boiling point of 
 which ranges from about 18 C. to about 40 C., and which, therefore, is 
 evidently a mixture of butane with much pentane. 
 
 Of course petroleum spirit of as low a boiling point as the samples 
 above quoted is rather the reverse of advantageous to the manufacturer, 
 and for this reason products like No. 8, and even more so No. 13 and 14, 
 are much preferable. 
 
198 INDIA RUBBER AND ITS ANALYSIS. 
 
 Besides the boiling point test, petroleum spirit should be tested for the 
 presence of pyridine, and it is also advisable to ascertain whether its 
 purification has been carried far enough. The test for pyridine has 
 already been described above; the purification test is carried out by 
 shaking violently in a test tube 10 c.c. of the spirit with 5 c.c. of sulphuric 
 acid of 1'55 specific gravity. There should be no perceptible amount of 
 discoloration either of the acid or of the hydrocarbon liquid. 
 
 There is a general concensus of opinion among india rubber manu- 
 facturers that there is greater risk of fire by ' spontaneous ignition ' in the 
 use of petroleum spirit, than when coal tar naphtha is employed. This 
 opinion appears to be confirmed by the fire statistics available, which 
 unquestionably show that during the ruling of high coal tar solvent prices, 
 when there is naturally a tendency to use more petroleum spirit in its- 
 place, the frequency of fires in rubber factories invariably rises. 
 
 3. Shale Naphtha. This is a product produced in large quantities by 
 the Scotch shale distilleries. In its chemical properties this solvent is very 
 similar to petroleum spirit, but while the latter consists practically entirely 
 of hydrocarbons of the aliphatic series, the hydrocarbons of shale naphtha 
 consist almost entirely of defines. The specific gravity and boiling points 
 of shale naphtha approximate closely to those of coal tar naphtha. Against 
 petroleum spirit it has, however, the disadvantage of a very pronounced 
 and persistent odour. This odour, though highly characteristic, is by no 
 means offensive. Indeed, it is by many considered preferable to the odour 
 of coat tar naphtha, and the preference often expressed for the latter is 
 probably chiefly due to old custom. Wherever the question of odour is 
 of not much account, shale naphtha will be found to be a very effective 
 and cheap solvent for many purposes in connection with the manufacture 
 of india rubber. 
 
 4. Turpentine. Turpentine was the first organic solvent which 
 HANCOCK applied to india rubber in the hope of producing india rubber 
 solutions. In this he failed, ascribing his failure to the impure description 
 of his turpentine. Shortly after his discovery of the effects of mastication 
 upon india rubber, he found that one of the most valuable results of 
 this operation was the readiness with which the rubber so treated dis- 
 solved in turpentine an effect which is, however, just as pronounced 
 with any other solvent. To-day the use of turpentine as a rubber solvent 
 has become almost obsolete save for a few specific purposes. This is, on 
 the one hand, due to its high price (two shillings per gallon) ; and on the 
 other, to its high boiling point. 
 
 It is rather remarkable that all dextro-rotary brands of commercial 
 turpentine are better solvents than the laevo-rotary kinds. For this no 
 doubt very complex phenomenon, no explanation can at present be 
 suggested. Somewhat more intelligible is the fact that the more dipentene 
 a turpentine contains, the better a solvent it is. This in some measure 
 appears to be a contradiction to the first statement, as dipentene is optically 
 
INDIA RUBBER SOLVENTS. 199 
 
 inactive. It is, however, not impossible that the dextro-rotary turpentines 
 contain more dipentene than their optical antagonists. Certain it is that 
 Russian turpentine, the occurrence of dipentene in which is definitely 
 ascertained, and which is also dextro-rotary, surpasses in dissolving power 
 all the other brands of turpentine. The curious old statement that 
 turpentine, after having been distilled over india rubber, acquires thereby 
 much greater dissolving power than before, finds its explanation, perhaps, in 
 the same fact, as it is well known that the distillation of turpentine from 
 retorts heated by direct flame always leads to the formation of more or 
 less dipentene. But in this case the presence of india rubber in the 
 distilling liquid would appear to be quite unnecessary. 
 
 5. Carbon Bisulphide. This substance is employed in rubber factories, 
 to a very slight extent only, as a solvent, its volatility, inflammability, and 
 very generally its bad odour being equally objectionable for this purpose. 
 It is, however, employed in enormous quantities as a solvent for chloride 
 of sulphur in the process of cold vulcanisation. 
 
 Carbon bisulphide boils at 46 C., and it will generally happen that 
 whatever be the degree of purity of the commercial product, the variations 
 of the boiling point are very insignificant so much so, that the boiling 
 point in this case practically ceases to be a criterion of the purity of the 
 sample. 
 
 The impurities occurring in carbon bisulphide are very little known, 
 especially those that cause the very offensive odour of many of the com- 
 mercial samples. Free sulphur is almost invariably present, and as long 
 as its amount does not exceed 0'15 grins, in 100 c.c., is quite unobjection- 
 able. For practical purposes the whole of the residue remaining after dis- 
 tillation may be taken to represent sulphur. This, strictly speaking, is not 
 quite correct, and this residual sulphur always contains more or less minute 
 proportions of at least two crystallisable compounds which appear to 
 be to a large extent responsible for the bad odour of the great majority 
 of the commercial products. Besides these, carbon bisulphide often con- 
 tains sulphuretted hydrogen in solution, and it is scarcely necessary to point 
 out that this occurrence is proof of a very perfunctorily performed purifi- 
 cation. Such products are always characterised by a more or less marked 
 yellow colour. The purer a sample of bisulphide of carbon, the more its 
 odour approximates that of chloroform. An acrid or otherwise badly smell- 
 ing product should unhesitatingly be rejected, as there is no doubt what- 
 ever that an article perfectly 'waterwhite' and free from any offensive 
 odour can be and is produced on a commercial scale. 
 
 Many attempts have been made to substitute less obnoxious solvents for 
 carbon bisulphide, but although in a few applications pure benzol has 
 proved fairly satisfactory, its general use as the solvent for chloride of 
 sulphur in the cold cure process is barred by the fact that it produces 
 surfaces having a 'starved,' leathery feel, although the vulcanisation, as 
 such, is perfect. This matter has already been discussed (page 80), and it 
 
200 INDIA RUBBER AND ITS ANALYSIS. 
 
 will suffice here to state that in my own opinion this is due to the fact 
 that the swelling of the rubber is very much less with benzol than with 
 carbon bisulphide. Consequently, in curing a thin rubber film e.g., water- 
 proof cloth with benzol as the solvent, only a very slight swelling takes 
 place before the * setting ' i.e., the pectisation of the rubber ; but under 
 the same conditions, using, however, carbon bisulphide as the solvent, the 
 pectisation of the rubber in other words, the curing effect takes place 
 while the film is in a state of considerable distension as regards thickness. 
 I have shown that vulcanisation results in the permanent fixation of the 
 physical condition as to structure of the rubber at the moment at which 
 it occurs, and it is therefore evident that any difference in the physical 
 effect produced by different solvents used in the vulcanisation process must 
 necessarily persist after, and be rendered permanent by, the process of 
 vulcanisation. 
 
 This is, in niy opinion, the only feasible explanation of the otherwise 
 quite unintelligible fact that, chemically, the process of cold vulcanisation 
 varies considerably in its results according to the nature of the solvent 
 employed. It shows also that the real difficulty of the problem of finding 
 a satisfactory substitute for carbon bisulphide lies not so much, as has been 
 assumed up to the present, in the discovery of some cheap solvent which 
 is fairly volatile and indifferent to india rubber, but rather in finding a 
 solvent, the physical effect of which upon india rubber shall be identical 
 with that of carbon bisulphide. 
 
CHAPTER VII. 
 COLOURING MATTERS. 
 
 ALTHOUGH a great number of India rubber articles are produced having a 
 colour more or less pronounced and differing from that of pure vulcanised 
 india rubber, yet colouring matters do not play a very important part in 
 the manufacture of these goods. The obvious reason is that the uses to 
 which these articles are put are, with very few exceptions only, so highly 
 technical and purely utilitarian as to render their colour a matter of more 
 or less complete indifference. The vast majority of india rubber goods are 
 either grey, black, or red in colour. The grey colour is that of an india 
 rubber article containing no colouring matter of any sort ; the black 
 colour is either the colour of desulphurised 'cut sheet ' articles, or of goods 
 containing lamp black, pitches, and similar compounds ; while the red colour 
 is due to the rubber containing antimony pentasulphide, an excellent sulphur 
 carrier, or it may also be due to the presence in the rubber of nothing 
 more than ' red iron oxide ' used for the purpose of imitating the colour of 
 sulphide of antimony, which colour many people still connect with the idea 
 of a very high quality rubber article. 
 
 In one branch particularly of the india rubber trade that of ladies' 
 waterproofs colours are used extensively for decorative and ornamental 
 purposes, either by producing self-coloured rubber surfaces, or by printing 
 upon variously coloured rubber surfaces textile designs either with 
 ordinary pigment colours, with pigment lakes, or with metallic powders 
 (bronzes). 
 
 There is, perhaps, only one other article in which colour is an essential 
 feature. This is in the hard rubber compounds for dental purposes (gums), 
 although obviously in this case the colour serves a utilitarian purpose like- 
 wise, being intended to match the colour of the tissues of the mouth. In 
 the production of this 'compound,' Continental and American factories 
 have altogether outstripped their English rivals. The colouring matter 
 employed in this case is invariably either vermilion, the colour of which is 
 attenuated by the admixture of zinc white, or vermilion with a trace of 
 cobalt blue. 
 
202 INDIA RUBBER AND ITS ANALYSIS. 
 
 In examining any colouring matter intended for use in conjunction 
 with india rubber, the following points call for particular attention : 
 
 1. The colouring matter should contain no substance or compound 
 detrimental to india rubber. 
 
 2. The colouring matter should be able to withstand the chemical and 
 physical influences of vulcanisation. 
 
 3. The colour in the vulcanised article should be fairly fast to light ; 
 it should not bleed on immersion of the article in boiling water rendered 
 faintly alkaline by ammonia, or in water rendered faintly acid by acetic 
 acid. 
 
 4. The colour used should not affect the durability of the rubber 
 articles. 
 
 White Pigments. 
 
 White Lead. This product is used fairly extensively, owing to its- 
 enormous covering power. Its purity may be ascertained by the well- 
 known chemical methods. Seeing, however, that the chief recommenda- 
 tion of white lead lies in its covering power, which is greatly affected by 
 impurities and adulterations, and that even different samples of pure 
 white lead exhibit very considerable variations in this respect, it appears 
 preferable to dispense with the chemical test of this product, and to ascer- 
 tain its covering power instead, in comparison with some known product of 
 standard excellence. The test is carried out by placing upon a glass plate 
 of convenient size 2 grms. of the standard white lead, and at a distance of 
 from 4 to 6 inches another 2 grms. of the white lead to be examined. To- 
 each of these samples 0*2 grm. of lamp black and 20 drops of linseed oil 
 are then added. With a palette knife of suitable size the separate samples 
 are then successively worked by pressing and rubbing, until in each case 
 an absolutely uniformly grey paint is produced, which, by means of the 
 palette knife, is spread in an even coating over the glass. Corresponding 
 to the two samples of white lead so treated, two separate coatings of grey 
 are thus obtained almost side by side, and it is very easy to judge whether 
 they are of identical shade or not. The grey shade produced is, of course, 
 the lighter, the higher the covering power of the white lead. It is now 
 very easy to reduce the darker of the two shades of grey to the shade 
 of the lighter one by gradually working into the darker shade small 
 portions of the same white lead until the two shades are exactly equaL 
 The covering power of the two samples then varies inversely as the 
 quantities of white lead required for the production of identical shades- 
 of grey. 
 
 White lead is an excellent material for the production of 'cold cured ' 
 rubber goods, but it cannot be employed for the manufacture of hot vul- 
 canised articles, the free sulphur at the vulcanising temperature entirely 
 discolouring the white lead. Cold cured goods may also develop this dis- 
 coloration in a slight degree if the chloride of sulphur employed contains 
 
COLOURING MATTERS. 203 
 
 too much free sulphur, or if the carbon bisulphide used be contaminated 
 with H 9 S, or contain too much free sulphur. 
 
 Zinc White. This compound is very extensively employed in the 
 manufacture of white rubber goods. In testing samples it should be 
 remembered that inferior brands often contain sensible quantities of lead 
 or metallic zinc, both of which impurities very detrimentally affect the 
 purity of the colour as it appears after vulcanisation. The covering power 
 of different brands of zinc white is subject to not inconsiderable variations, 
 and should therefore be ascertained by the test above described for the 
 examination of white lead. 
 
 Zinc white is equally suitable for the manufacture of cold as of hot 
 cured goods. 
 
 The general analysis and valuation of zinc white calls for no special 
 remark, and is carried out by well-known methods. The quantity of 
 metallic zinc or lead contained in it is often so small as to escape detection. 
 It will, however, frequently be found that different brands of zinc white, 
 owing to differences in the quantities of the above named impurities con- 
 tained therein, exhibit surprising variations in the colour of the rubber 
 goods manufactured by their aid. This is due to the fact that variations 
 in the quantities of these impurities will necessarily also cause variations 
 in the colour of the zinc white as it appears in the iiidia rubber articles 
 after vulcanisation, in which process these impurities are converted into 
 their respective sulphides. The effect of this discoloration can readily be 
 ascertained in any given sample of zinc white by shaking about 2 grms. of 
 it in a test tube with 15 c.c. of water, adding to this suspension one or two 
 drops of ammonium sulphide, and, after two or three minutes' standing, 
 comparing the colour of the zinc white so treated with a suspension of some 
 of the same sample in pure water. Or the zinc white is mixed in a mortar 
 with a small quantity of sulphur, the mixture transferred to a crucible, the 
 sulphur driven off at a gentle heat, and the change of colour noted. 
 
 Lithopone. The theoretical composition of this pigment is expressed by 
 the formula ZnS, BaS0 4 , and it should therefore contain 29'4 per cent, 
 of zinc sulphide and 70'6 barium sulphate, but most of the commercial 
 products of this description contain much less zinc sulphide. It is a 
 matter of general practical experience that the value of the product as a 
 white pigment rapidly diminishes with the quantity of zinc sulphide 
 present, so that its analysis and valuation resolves itself into the deter- 
 mination of the respective proportions of zinc sulphide and barium sulphate 
 present. This is carried out with sufficient accuracy by dissolving a 
 weighed quantity of the lithopone in dilute hydrochloric acid, adding a 
 moderate quantity of a solution of ammonium chloride, and boiling for 
 from ten to fifteen minutes. The insoluble barium sulphate is filtered off, 
 washed neutral, dried on the filter, the latter subsequently incinerated, 
 and the residue weighed. Deducting the weight thus obtained from the 
 weight of the lithopone used, yields the weight of the zinc sulphide present. 
 
204 INDIA RUBBER AND ITS ANALYSIS. 
 
 Barium Sulphate. This is used either in the form of the finely ground 
 mineral 'barytes,' or in the form of artificially precipitated barium sulphate. 
 The latter preparation has a somewhat greater covering power than the 
 natural product, however finely ground. All these, and the following 
 adjuncts, can scarcely be described as white pigments, their whiteness 
 being simply due to the diffused reflexion of extremely fine crystalline, but 
 individually transparent, particles. By embedding these powders in a 
 medium of high refractive index, they become, of course, more or less 
 transparent i.e., the * whiteness ' disappears. They behave in this manner 
 when incorporated with india rubber, the refractive index of which is 
 comparatively high. They must, therefore, be described as filling agents 
 rather than as white pigments. 
 
 Paris White. This product, also known as whiting, is a more or less 
 impure natural calcium carbonate (chalk). The impurities consist, besides 
 traces of organic matter, of silica (diatoms) ; in exceptional cases it may 
 contain magnesium carbonate (magnesite), and it invariably contains traces 
 of ferric oxide. It is the most extensively used filling agent ; its value as a 
 white pigment is of the slightest. 
 
 Red and Brown Pigments. 
 
 Golden Sulphide. This product consists of antimony pentasulphide 
 containing up to 40 per cent, of free sulphur. It is one of the most active 
 ' sulphur carriers ' known, but there is little doubt that it was originally 
 employed, and has been extensively used up to the present day, perhaps 
 even more on account of its colour, there being very little choice in red or 
 scarlet pigments suitable for the manufacture of hot cured * red rubber ' 
 goods. The shade of this product varies rather considerably according to 
 the mode of its preparation, and it is very curious to observe that different 
 countries favour different shades, to the practical exclusion of any other. 
 Thus, in France a bright, deep red is preferred. Germany favours a bright 
 orange or tan colour, whilst English manufacturers select a medium shade. 
 For the analysis of this product, see page 185. 1 
 
 Vermilion. This is the most brilliant of all the scarlet pigments avail- 
 able for the purposes of the india rubber manufacturer. It is perhaps the 
 only one which is equally suitable for soft rubber, however cured, as for 
 ebonite. The colour of vermilion varies from a fine but pale bluish- 
 crimson to a full bright scarlet. The shade depends essentially upon the 
 size of the crystals. The smaller these, the paler and bluer the shade ; the 
 larger the crystals, the darker and brighter the colour. Adulterations of 
 vermilion are nowadays practically unknown, and it is therefore scarcely 
 ever necessary to proceed to an analysis of it. Under certain circumstances 
 
 1 The frequently made statement that antimony pentasulphide is an excellent 
 vulcanising agent is a pure myth, one of the many to be found in the rubber trade, and 
 as hard to kill as the sea-serpent. The fact of the case is that most of the antimony 
 sulphide in the market contains from 15 to 20 per cent., and more, of free sulphur, to 
 which the supposed vulcanising action of this pigment is to be ascribed. 
 
COLOURING MATTERS. 205 
 
 it may be desirable to ascertain the covering power of a given sample. 
 This is carried out exactly as before described. 
 
 Both red antimony sulphide and vermilion proper can be employed for 
 the manufacture of hot or cold vulcanised goods. Their use in the latter 
 case would, however, amount to sheer waste of money, since as good and 
 even much better effects can be obtained with the enormous range of 
 orange, scarlet, and crimson pigment lakes (aniline lakes), which are quite 
 as permanent, and, if competently manufactured for the purpose, perfectly 
 indifferent to india rubber. 
 
 Red and Brown Iron Oxides. These products appear on the market 
 under countless fancy names. Some of the so-called ( Venetian reds ' of this- 
 class rival in brilliancy and character of shades good qualities of antimony 
 sulphide, but the various products exhibit very great differences in cover- 
 ing power. This is ascertained in the already described manner, and the 
 strongest products should always be given the preference. The reason for 
 this is that ferric oxides, especially the more or less hydrated varieties, are 
 by no means so indifferent to india rubber as is generally supposed. Cold 
 cured india rubber goods containing this article are often found to undergo- 
 oxidation rather quickly. This is much less the case in hot cured goods, 
 but there is no doubt that the iron oxide pigments greatly reduce their 
 resiliency. 
 
 A complete analysis of these pigments is generally of not much use, as 
 the percentage of iron oxide in them bears only a very remote relation to 
 their intrinsic value. Some of the very brightest ' red oxides ; owe, how- 
 ever, the richness of their shade very largely to the presence in them of 
 small quantities of apparently free sulphuric acid. Such products are 
 extremely detrimental to india rubber, and for this reason the neutrality of 
 any given sample should always be ascertained by a special test. 
 
 Occasionally brown * oxides ' are offered containing manganese peroxide. 
 The presence of this compound in hot cured goods is highly undesirable and 
 altogether fatal to cold cured goods. This same obnoxious compound is 
 also occasionally encountered in the brown shades bearing the names of 
 umber, sienna, or vegetable brown. Such products call for unhesitating 
 rejection on the part of the rubber chemist. 
 
 Red and Brown Lake Pigments. The very large range of brilliant 
 pigments is almost exclusively prepared from coal tar dyes ; a few of them 
 from vegetable dyes. They contain from 5 to 20 per cent, of dye-stuff 
 fixed upon some suitable mineral base. The pigments used for the purposes 
 of the india rubber manufacturer contain these dyes fixed upon either 
 alumina or a mixture of alumina and barium sulphate, or barytes, or white 
 lead, and in some cases even red lead. It is scarcely necessary to remark 
 that pigments containing the latter base should be strictly avoided. 
 
 To test for the presence of red lead, the lake is repeatedly boiled with 
 dilute nitric acid, when the appearance of a more or less dark-brown 
 insoluble residue generally sufficiently clearly indicates its occurrence. In 
 
2O6 INDIA RUBBER AND ITS ANALYSIS. 
 
 cases of doubt, the residue is boiled with a fresh quantity of nitric acid and 
 a few crystals of oxalic acid or sugar. The ensuing solution is filtered. A 
 white precipitate in the nitrate, on addition of dilute sulphuric acid, 
 indicates the presence of red lead in the pigment. 
 
 Aniline Lake Pigments, as these pigments are commonly called, are 
 manufactured from either basic or acid coal tar dyes. Lakes produced 
 from basic dyes are generally brighter and also faster than the lakes obtained 
 from acid dyes, but for scarlet shades the latter yield a far more brilliant 
 series than the former. There is, however, a class of coal tar dyes which 
 may be described as sulphonated basic dyes, and which yield even more 
 brilliant shades than either of the before mentioned classes, although they 
 are perhaps a trifle inferior in fastness to the lakes obtainable from basic 
 dyes. The fastness of all these products both to water and light depends 
 very materially upon the methods by which they have been prepared, but 
 this is scarcely a subject which can be gone into in this place. 1 
 
 Of considerable importance is the question of the action of these lakes 
 upon india rubber. Of course, what action they possess is entirely due to 
 the coal tar dyes contained in them. There is no doubt whatever that 
 some of these have a very distinctly unfavourable action, Bismarck brown 
 having long enjoyed a particularly bad reputation in this respect. This 
 action is certainly greatly reduced in the lake pigments, but it calls, never- 
 theless, for careful investigation before any particular lake is generally 
 employed in manufacturing the more so, as it is a practical impossibility to 
 say off-hand what dyes are to be avoided in the form of lake pigments, so 
 much depending upon the manner in which they have been produced. As 
 .a general truth, it may, however, be stated that lakes produced from acid 
 dyes are, so far as is known at present, very much safer to use than 
 those manufactured by the use of basic or sulphonated basic dyes. The 
 value of this information is unfortunately largely discounted by the fact 
 that the lakes produced from acid dyes represent only a comparatively 
 limited range of scarlet shades ; for all other shades, lakes manufactured 
 from the other classes of dye-stuffs have to be drawn upon. There are, of 
 course, numerous blue, green, and violet acid dyes, but the lakes they form 
 are far too dull for the purpose of the rubber manufacturer. 
 
 The valuation of any of these products is best carried out by the 
 covering power test i.e., by mixing 0'2 grms. of the lake and also of the 
 standard with which it is to be compared with 2 grms. of white lead and 
 20 drops of linseed oil each upon a glass plate in the manner described for 
 testing the covering power of white lead. This test also gives very trust- 
 worthy information respecting the brilliancy of the shade, and renders 
 differences, too slight to be with certainty discernible in the full strength 
 lakes, at once apparent. 
 
 1 Full information on this important subject is to be found in a paper by myself : 
 "Researches on the Formation of Pigment Lakes," Journ. Soc. Chem. Ind., 1891, 
 p. 896 ; and a further paper under the same title in Journ. Soc. Chem. Ind., 1893, p. 650. 
 
COLOURING MATTERS. 207 
 
 Very frequently dyes soluble in benzene or coal tar naphtha are 
 recommended for the manufacture of coloured india rubber goods. These 
 products, known as aniline resinates, are the resinates of basic dyes pre- 
 cipitated upon a basis of aluminium resinate. They may be used for the 
 manufacture of coloured toy balloons or similar articles, but are useless 
 for general purposes. 
 
 All these lake colours can be used for the manufacture of cold cured 
 goods only. In the hot cure process they undergo more or less complete 
 destruction. 
 
 Yellow Pigments. 
 
 Yellow pigments are only of subordinate importance in the rubber 
 industry. They are occasionally used for toys, and have also been 
 suggested as a means to prevent the ' suncracking ' of rubber goods. 
 Their efficiency in this last respect appears, however, doubtful. 
 
 Cadmium Yellow is undoubtedly the safest, and also, from the hygienic 
 point of view, the most satisfactory, product to use. It consists of cadmium 
 sulphide, and its shades vary from a deep orange to a rich golden-yellow, 
 according to the conditions under which it was produced. It is undoubtedly 
 the safest yellow pigment to use in connection with india rubber for both 
 cold and hot vulcanising. Whether in the latter process it acts as a sulphur 
 carrier is at least uncertain. 
 
 For the valuation of this product it is quite sufficient to ascertain 
 qualitatively its freedom from other metals than cadmium ; zinc notably 
 lends itself best to its adulteration. For the rest, the covering power test 
 will give all the necessary information. 
 
 Chrome Yellow. Under this name a wide range of products is known 
 varying in shade from a pale primrose-yellow to a full and almost bright 
 red lead colour. The essential colouring constituent of all these products 
 is lead chromate, and the variations in the shade are due to the presence in 
 them of varying quantities of lead sulphate in the yellower shades, and of lead 
 oxide in the orange and red shades. The presence of lead sulphate is often 
 considered to be an adulteration of the chrome yellow, but if the lead 
 sulphate has been precipitated simultaneously with the chromate, the result- 
 ing precipitate has colour qualities which cannot be obtained in any other 
 way. This should be borne in mind when chrome yellows are subjected 
 to analysis. The best test of these, as of all pigments, is to be found, 
 not in their qualitative or quantitative composition, but in their covering 
 power. 
 
 Chrome yellows cannot be used for hot cured goods, but behave very 
 well in the cold cure process. 
 
 Zinc Yellow, a double salt of zinc chromate and potassium chromate, 
 is distinguished from all chrome yellows by its superior greenish tint, but 
 also by its greatly inferior covering power. It cannot, however, be recom- 
 mended for use in rubber goods. 
 
2O8 INDIA RUBBER AND ITS ANALYSIS. 
 
 Yellow Ochre. This is a natural product consisting of ferric hydrate 
 deposited upon alumina, silica, or aluminium silicates generally all three. 
 To the rubber chemist its analysis is without interest, but as some ochres 
 contain the dreaded manganese peroxide, the absence of this compound 
 should be ascertained. In addition to this, an examination of its covering 
 power is all that is required. 
 
 All products of this class are equally suitable for hot, as for cold, 
 vulcanised goods. 
 
 Yellow Lake Pigments. There are a number of very brilliant lake pig- 
 ments derived from coal tar colours, but, like yellow pigments in general, 
 they are little used. The same is the case with the various yellow lakes 
 obtainable from fustic, Persian berries, and other yellow vegetable dyes. 
 
 Green Pigments. 
 
 Chrome Oxide Green. This fine green, which should consist of pure 
 chromic oxide, is perhaps the fastest of all mineral pigments. It is not in 
 the least affected in either hot or cold vulcanisation, nor has it the slightest 
 chemical action upon india rubber. Perhaps its only drawback is its 
 comparatively high price. Adulterations of this product are readily 
 detected by ascertaining the percentage of chromic oxide it contains. Its 
 covering power varies according to the care with which it has been prepared. 
 
 Brunswick Green (Chrome Green), The greens which are offered under 
 a variety of other names consist of chrome yellow and Paris blue, and 
 contain generally more or less considerable quantities of mineralic fillings. 
 They are of only limited applicability for india rubber, as they lose 
 considerably in brilliancy in rubber mixings. They stand the cold curing 
 process satisfactorily, but cannot be used for the manufacture of hot cured 
 goods. There is, moreover, a suspicion that they exert some detrimental 
 action upon india rubber. Their analysis is scarcely ever called for, the 
 covering power test readily giving all the required information. 
 
 Green Lake Pigments. A number of very fine green lake pigments are 
 obtainable from the coal tar dyes. The fastness of all of them is very 
 satisfactory, but they can, of course, only be used for the manufacture of 
 cold cured goods. 
 
 A large number of green pigment colours exist containing either copper, 
 or copper and arsenic. It is scarcely necessary to say that all such products 
 are from every point of view highly dangerous. Also the use of the products 
 described as zinc green, mixtures of zinc yellow with Prussian blue, are 
 likely to cause trouble in india rubber. 
 
 Blue and Violet Pigments. 
 
 Ultramarine. This blue is on the market in a variety of shades, from 
 ultramarine violet to ultramarine green. It is one of the safest colouring 
 agents to use in india rubber, and will withstand both the hot and the cold 
 
COLOURING MATTERS. 209 
 
 curing process. It is, however, advisable to use only those ultramarine 
 blues which are described as 'alum proof,' otherwise the shade may 
 undergo considerable deterioration, especially in the cold cure process. 
 Its analysis could serve no purpose of the rubber chemist ; all the informa- 
 tion required concerning any given sample is obtained by ascertaining its 
 covering power, and its behaviour against a 5 per cent, solution of alum. 
 The last named test is carried out by filling two 100 c.c. cylinders with 
 a 5 per cent, alum solution and water respectively, and adding about 
 2 grms. of the ultramarine to both these liquids. After five hours' standing 
 the cylinders are well shaken and the ultramarine suspensions examined. 
 There should be no change at all observable, or only a very slight one, in 
 the suspension in alum. 
 
 Prussian Blue. This product, which also passes under the name of 
 Chinese blue, is a ferric ferrocyanide, a pigment of a covering power and 
 intensity unsurpassed by any mineral pigment. It has frequently been 
 recommended for the colouring of india rubber goods, but has, so far, not 
 proved of much service in this application. It is produced in a variety of 
 shades ranging from a deep, almost violet blue to a magnificent clear blue 
 of almost the character of an ultramarine (Milori blue). Very little is 
 known as to how these blues behave in the process of vulcanisation, and 
 there has always been some sort of a suspicion that they affect india rubber 
 detrimentally, which would rather suggest that vulcanising agents and 
 conditions are not without effect upon them. 
 
 These blues are very commonly adulterated. The quickest way to 
 ascertain their purity, or otherwise, is to digest a sample, which must have 
 been ground into an impalpable powder, for several hours, with a 5 per cent, 
 solution of ammonia until the decomposition is complete. The solution of 
 ferrocyanide is now separated from the insoluble residue by filtration. 
 The filter residue, when washed neutral, should consist of pure ferric 
 hydrate only. 
 
 Cobalt Blue (Smalt). This blue has repeatedly been recommended for 
 colouring india rubber, for which purpose its great indifference to chemical 
 influences would appear to fit it. Its low covering power and very high 
 price have, however, always prevented its general employment. 
 
 Blue and Violet Lake Pigments. There are an abundance of fine 
 blue and violet lakes, which, together with ultramarine, amply and 
 satisfactorily supply all the requirements of the rubber manufacturer in 
 the way of blue pigments. Some care is, nevertheless, required in the 
 choice of the lakes employed, as some of them are rather fugitive, and 
 others have a slight, but none the less distinct, deteriorating effect upon 
 india rubber. It is, however, difficult to single out any particular lakes 
 in this respect, as their behaviour depends very greatly upon the process 
 by which they have been manufactured. It is therefore necessary to 
 conduct a series of experiments with such of these lakes as are selected for 
 use with india rubber. 
 
 14 
 
210 INDIA RUBBER AND ITS ANALYSIS. 
 
 Black Pigments. 
 
 Black pigments are very largely used in the manufacture of rubber goods. 
 The vast majority of these preparations consist of carbon, or contain, at 
 least, carbon as their chief constituent. There are, indeed, some black lake 
 pigments, but these are, however, without any practical importance for the 
 purpose of the rubber manufacturer. 
 
 Lamp Black or Vegetable Black. These products are produced by 
 the imperfect combustion of all kinds of oily materials, and they are 
 always distinguished by their enormous bulkiness, or lightness as it is 
 termed. The fine state of subdivision, which is the chief cause of this 
 property, is, of course, largely responsible for the great covering power of 
 these blacks, but is, however, not an unmixed blessing, as it renders the 
 incorporation of even a moderate proportion of them with india rubber not 
 only a very laborious process, but also one which can only be accomplished 
 by subjecting the india rubber to more or less prolonged mechanical 
 treatment, which is often detrimental to the quality of the articles manu- 
 factured from it, and may also cause serious trouble in the process of hot 
 vulcanising such goods. There is further to be considered the fact that 
 each particle of black, as of any other solid substance incorporated with 
 india rubber, carries with it a certain quantity of air into the india rubber, 
 which is only partly eliminated during the mechanical working of the 
 india rubber. Hence it follows that the bulkier the material incorporated, 
 the greater this volume of air, which is distributed throughout the india 
 rubber in bubbles of microscopic minuteness. The porosity thus produced 
 in india rubber articles interferes seriously with the quality of the goods. 
 It is therefore obvious that, as far as pigments are concerned, the denser 
 kinds are much preferable to those possessing great lightness, or bulk. 
 This is especially true of the lamp blacks used in connection with india 
 rubber, and for this reason blacks only should be used, the density of which 
 has been increased by subjecting them, packed into closed crucibles, for 
 several hours to a temperature somewhat above red heat. This treatment 
 has the further advantage of freeing the black from the oily or greasy 
 matters, of which it always contains a more or less considerable proportion. 
 It should also be noted that the colour and depth of shade of the black is 
 vastly improved by this second burning. Altogether, the improvement in 
 every respect produced by this additional process is so great that most of 
 those india rubber works which consume a particularly large quantity of 
 lamp black carry out themselves this second burning or calcining operation. 
 
 Bone Black. This black pigment, which is produced by the charring in 
 closed vessels of bones, is much inferior to the lamp blacks, and its use 
 offers little or no advantage to the rubber manufacturer. 
 
 Black pigments should always be analytically controlled in every india 
 rubber works, as the appearance and quality of the goods produced with 
 them depends very much upon their composition and colour qualities. It 
 
COLOURING MATTERS. 211 
 
 has already been stated that lamp black generally contains a certain 
 proportion of oily or greasy matter, not infrequently as much as 12 per 
 cent, and over. This is estimated by extracting a weighed quantity of the 
 black in a Soxhlet extractor. To prevent any of the black passing into the 
 extract, the simplest device is to fill the black in one of the cylindrical 
 filter paper tubes, which can be purchased ready made. Before placing 
 this tube containing the black in the Soxhlet extractor, the open end of 
 the tube is plugged by inserting into it a wad of cotton wool. The extract 
 is collected in a weighed flask. The extraction is effected with either 
 acetone or ether, and after the extraction the solvent is distilled off the 
 flask, the contents of which are then dried in an oven at a temperature 
 not exceeding 80 C., and subsequently weighed. 
 
 The only satisfactory way to ascertain the quality of the shade is to 
 observe the shades of grey yielded by the black in admixture with white 
 lead. This test is, of course, included in the ordinary covering power test. 
 The black in this test should produce with the white lead an absolutely 
 neutral grey tint without the least suggestion of a brownish or reddish 
 tint (foxiness). Not infrequently it will be found that a black of an 
 inferior shade has been corrected by adding to it a small quantity of some 
 blue pigment. It is almost invariably Prussian blue which is thus used, 
 and to the experienced eye this sophistication is almost invariably betrayed 
 by the strikingly bluish tint of such blacks, especially after dilution with 
 white lead. Whether such a correction is satisfactory for the more common 
 uses of black pigments we need not discuss here ; it is certainly not so to 
 the rubber manufacturer, and blacks of this kind should be rejected. The 
 presence of Prussian blue is readily demonstrated by boiling some of the 
 suspected black with a 10 per cent, solution of sodium dioxalate, when a 
 more or less strongly coloured blue solution is obtained. Should this test 
 fail, the correction may have been effected with either a blue lake pigment, 
 or with ultramarine. The former may be detected by boiling a sample of 
 black with alcohol (85 per cent, strong) acidulated with a few drops of 
 acetic acid, the latter by treating the black with dilute sulphuric acid, 
 when the development of sulphuretted hydrogen can be demonstrated by 
 means of filter paper impregnated with a solution of lead acetate. Should 
 these tests all fail, some of the black embedded in a highly refractive 
 medium, such as Canada ba-lsam or cedar wood oil, is examined under the 
 microscope. Its uniformity, or otherwise, can thus be readily ascertained. 
 
 Metallic Lustres (Bronzes). 
 
 Impalpable metallic powders, generally misnamed 'bronzes,' have 
 within recent years been extensively employed for the decoration of india 
 rubber goods, more particularly waterproof cloth. They are produced in a 
 very great variety of colours, which are either the colours of the metals or 
 of the alloys of which these powders consist, or which have been obtained 
 by subjecting the finished powders to a dyeing process with aniline colours. 
 
212 INDIA RUBBER AND ITS ANALYSIS. 
 
 These metallic lustres consist either of copper, tin, or aluminium, or 
 alloys of these with tin, zinc, or occasionally lead ; and considering the well- 
 known sensitiveness of india rubber to various metals, a careful selection 
 of the lustre colours to be used in connection with india rubber is of 
 considerable importance. In this respect it is obvious, and, indeed, well 
 recognised, that all those containing copper are strictly to be avoided. There 
 is still some doubt concerning the effects of colours consisting of tin, or of 
 an alloy of this metal, but there is no doubt that under certain conditions 
 even tin may injuriously affect india rubber. On the other hand, it is 
 certain that the metallic powders consisting of pure aluminium are without 
 any action upon india rubber, and this class of lustres, many of them dyed 
 with aniline dyes, are now practically exclusively employed. The avoidance 
 of ' decomposition ' effects ensured by the exclusive use of aluminium was 
 first pointed out in FRANKENBURG'S specification (No. 160, 1896), the date 
 of which, indeed, marks the starting-point of the successful and now very 
 extensively practised application of metallic powders, or lustres, to the 
 ornamentation of india rubber goods. 
 
 The question of the nature of the action of the metals upon india rubber 
 will be fully dealt with in a subsequent chapter. Incidentally I may state 
 that the metals, as such, have in no case the slightest action upon india 
 rubber, but a large number of their compounds are more or less injurious. 
 
 BURGHARDT l was the first to draw attention to the destructive effect of 
 copper upon india rubber. THOMSON 2 arrived at the same conclusion, and 
 made a similar statement with regard to the salts (nitrates) of copper, 
 manganese, and silver ; but he found that gold, silver, bismuth, antimony, 
 arsenic, tin, chromium, iron, nickel, cobalt, zinc, and cadmium had no effect, 
 and that lead, aluminium, palladium, and platinum acted only to a very 
 slight degree. This statement of the case is somewhat misleading. The 
 fact of the matter is, that in no case do the metals, as such, affect india 
 rubber; but already the sulphides of some of them notably cupric sulphide, 
 the oxides of gold, silver, tin (stannic), arsenic, nickel, cobalt, manganese, 
 palladium, and platinum have a more or less marked effect, which is 
 greatly intensified in their salts, particularly such of them as are soluble 
 in water and which act as oxygen carriers. It is, indeed, highly probable 
 that the supposed action of the sulphides, as of some of the oxides, only 
 begins with their transformation into salts under the influence of an 
 impure atmosphere. This, indeed, would account for the fact that copper 
 in waterproof cloth acts far more energetically in cold cured goods than in 
 those cured by the ' dry heat ' process. The obvious reason appears to be 
 that the hydrochloric acid vapours, which unavoidably accompany the first 
 named vulcanising process, convert the copper, or rather its oxide, into 
 the chloride, one of the most energetic oxygen carriers. In dry cured 
 goods cupric sulphide would be formed, which would only very slowly 
 undergo oxidation into the much more active sulphate. Accordingly, we 
 l Jourti. Sec. Chem. Ind. 2 Proc. Lit. Phil. Sec. of Manchester, 1890-91. 
 
COLOURING MATTERS. 213 
 
 would have to assume, as a general rule, that metallic contaminations are 
 much more to be guarded against in cold cured than in hot cured goods. 
 This is in perfect agreement with all our practical experience. 
 
 It is therefore evident that the employment of even tin or aluminium 
 lustres for the ornamentation of cold cured goods calls for great care in 
 the working if thoroughly satisfactory results are to be ensured. For hot 
 cured goods even these metals cannot be used, as they are considerably 
 tarnished by the sulphur vapours. This is self-evident as regards tin 
 lustres, but might appear unintelligible in the case of aluminium. The 
 explanation is to be found in the fact that commercial aluminium always 
 contains traces of iron, the conversion of which into ferrous sulphide causes 
 the dullening of the aluminium lustre. 
 
 The brilliance of these metallic lustres varies appreciably with the fine- 
 ness of the metallic flakes, and is the greater, the larger these flakes. It is for 
 this reason necessary to examine these colours microscopically in order to 
 ascertain whether they are of the fineness requisite for the various methods 
 employed for their application to india rubber. 
 
 The Dyeing of India Rubber with Soluble Dyes. 
 
 It is perhaps scarcely surprising that the very large number of organic 
 natural and artificial dye-stuffs should have stimulated would-be inventors 
 to apply these to india rubber. A great many of the splendid so-called aniline 
 dyes have been tried, their intensity and brilliancy appearing highly 
 suitable for the purpose. 
 
 It may at once be stated that the idea of utilising these products for 
 the dyeing of india rubber, as distinguished from staining, suffers from a 
 fundamental misconception, or, rather, complete ignorance, of the nature of 
 the dyeing process at any rate with the above named dyes. 
 
 The dyeing of animal or vegetable fibres with these dye-stuffs is a 
 function of the chemical constitution of both the fibre to be dyed and the 
 dye, and the constitution of either determines the nature of the dyeing 
 process. We find, on the one hand, that any fibre to be capable of being 
 dyed must contain either hydroxyl groups OH, carboxyl groups COOH, 
 or amido groups NH 2 , or a combination of all three. On the other hand, 
 we also know that a substance not containing any of these groups cannot be 
 employed as a dye-stuff, although it may possess a strong bright colour. 
 
 The case of diazo-benzene C 6 H 5 N : NC 6 H 5 is a very familiar illustration 
 of these facts. This body is a fine scarlet colour, but, in spite of this, has 
 no dyeing properties whatever. By introducing into one of two benzene 
 rings either the amido group or the hydroxyl group C 6 H 5 N : NC 6 H 4 NH 9 , 
 or C 6 H 5 N : NC 6 H 4 OH we obtain at once pronounced colouring matters. 1 
 
 1 A detailed account of the chemical theory of dyeing is to be found in the author's 
 paper "On Substantive Dyes and Substantive Dyeing: a Contribution to the Theory 
 of Dyeing, " Journ. Soc. Chem. Ind., 1894, p. 120. 
 
214 INDIA RUBBER AND ITS ANALYSIS. 
 
 From these few remarks it follows conclusively that a dyeing of the 
 India rubber i.e., the formation of a salt-like compound of india rubber and 
 either a dye-acid or a dye-base is an impossibility, india rubber containing 
 neither hydroxyl, nor carboxyl oxygen, nor amido groups. It might, of 
 course, be objected that this would still leave open the question of the 
 possibility of dyeing india rubber with the mordant dyes basic dyes on 
 tannin mordant, alizarine on alumina mordant but the fixing of a mor- 
 dant upon the textile fibres involves, on the part of the fibre substance, just 
 the same chemical conditions as the fixing of the dyes direct without 
 mordants. The best result which could be obtained on rubber with this 
 class of dyes would be by distributing a suitable mordant uniformly 
 throughout the india rubber, and then dyeing the latter with the corre- 
 sponding mordant dye. But the result, assuming such a process to be 
 practicable, would be none other than the one far easier and incomparably 
 better attainable by simply compounding the rubber with a pigment lake 
 consisting of the particular mordant and dye. 
 
 Even if we would waive all these weighty chemical objections, there 
 remains the serious difficulty of rendering the india rubber permeable to 
 the dyeing solution, in which water would have to be employed. Dyeing 
 in any other (non-ionisable) solvent is clearly impossible, and not only dis- 
 advantageous as MARKFELDT l suggests. 
 
 There are on the market a considerable number of dye-stuffs which are 
 soluble in turpentine, benzene, and petroleum hydrocarbons. They are 
 offered under a variety of names, and consist either of the stearates of the 
 basic dyes, or of the aluminium resinate lakes of basic or acid dyes. For 
 any process of dyeing, in the legitimate sense of this word, these products 
 are quite unsuitable indeed, they are no longer dyes, but simply lake 
 pigments soluble in certain organic solvents, and for this reason they do- 
 not dye, but only stain. They impart to rubber a characteristic colour 
 only if it is heavily charged with white pigments, or when it is manu- 
 factured into extremely thin, and therefore highly transparent, films. 
 Indeed, these products are satisfactory only for the manufacture of stained 
 toy balloons and similar articles. 
 
 MARKFELDT actually advocates the use of these products for the 
 colouring of india rubber for waterproofing. The adoption of this suggestion 
 would most certainly prove disastrous to the manufacturer, as all these 
 colours after vulcanisation 2 are more or less liable to 'bleed.' Altogether 
 it may safely be said that the aniline dyes and the above named ' soluble 
 lakes ' are useless for dyeing india rubber, and only of very limited use for 
 staining it. 
 
 1 Gummi Zeitung [15], 1900, p. 244. 
 
 2 Only the cold cure process could at all be used. 
 
CHAPTER VIII. 
 CONSTRUCTIVE COMPONENTS OF INDIA RUBBER ARTICLES. 
 
 THE applications of India rubber are so extraordinarily numerous and 
 varied that it is scarcely possible to name a single substance used in the 
 arts and industries which, in some form or other, does not find employment 
 as a constructive component of one of the almost innumerable india rubber 
 articles. I do not propose to attempt in this chapter to discuss all the 
 numerous substances which are associated in manufacturing processes with 
 india rubber, but to deal with a few only of these most extensively employed. 
 Undoubtedly the most important of these substances are the textile fabrics 
 which are very generally used either as ' insertions ' to give strength to, 
 or limit the distensibility of, sheets, hose, boots and shoes, etc., or simply 
 to produce the various kinds of waterproof goods in which the cloth is the 
 really important part, the india rubber coating being merely an accessory 
 imparting to the cloth the waterproof quality. 
 
 The textile fabrics used for constructive purposes proper are fabrics 
 consisting of cotton, jute, or hemp, very rarely wool, and these fabrics are 
 used almost exclusively in the 'grey' state. For waterproofing, cotton, 
 wool, silk, and mixed fabrics are used, which, according to the purpose for 
 which the cloth is intended, are employed in the ' grey ' state, but in most 
 cases dyed or printed. 
 
 The tests to which these fabrics should be subjected before use vary, 
 therefore, considerably according to the use to which they are to be put. 
 Generally speaking, the textile materials employed for insertions and 
 stiffenings need only be tested for strength before and after exposure to 
 vulcanising temperatures ; whereas the fabrics intended for the manufacture 
 of waterproof cloths require also examining for fastness of colour, as well 
 as for the absence of ingredients which are likely to affect the india rubber 
 coating. 
 
 Examination of Textile Fabrics. Insertion cloths are invariably used 
 with a view to strengthen the article to be manufactured, and to reduce its- 
 distensibility without destroying its flexibility. The employment of 
 ' insertions ' in steam sheeting, pressure, and suction tubing, etc., sufficiently 
 illustrate this. The range of fabrics employed for this purpose varies from 
 very tine * Cantons ' to the coarsest ' ducks,' * twills,' and ' Hessians.' 
 
2l6 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 The strength of these fabrics is tested by means of a cloth testing- 
 machine, of which fig. 18 is an excellent type. It consists of a bed a, in 
 which slide the two jaws b and e. Of these, b is connected with the weight- 
 ing lever d and registering dial e. Through the lower part of c passes the 
 travelling screw /, which can be operated by the wheel g. The weighting 
 lever is provided with catches h, which at any point engage in the notches 
 of the semicircle i Any pull on b sets the weight, and with it the lever 
 d, into motion ; but on releasing the pull i.e., on the material under test 
 breaking the weight is prevented from falling back by the catches of the 
 lever engaging in the notches of the semicircle, and as the lever actuates 
 
 FIG. 18. Cloth testing machine. 
 
 the indicating dial, the index finger of the latter remains stationary at the 
 point where the break occurred. 
 
 For the purpose of determining the distension of the material under test 
 at any point, or at its break, the scale k is attached to the jaw b, which 
 slides upon the bed a. Jaw c carries a pointer I, which determines the 
 exact length of the material under test between the jaws. 
 
 It is advisable to carry out all tests with test pieces of the same 
 dimension, so as to obtain all round comparable results. For English 
 practice it will generally be found most satisfactory to adopt the dimensions 
 prescribed in the War Office and Admiralty specifications issued with their 
 forms for tender, viz. : 
 
 The test should always be applied both in the direction of the warp 
 
CONSTRUCTIVE COMPONENTS OF INDIA RUBBER ARTICLES. 217 
 
 and in that of the weft of each cloth. It is further very important to 
 repeat this test with two pieces of the same sample which have been 
 exposed to vulcanising conditions, which, as regards temperature and time, 
 are as nearly as possible alike to those subsequently to be observed in the 
 factory. In a very large number of cases a treatment of the samples 
 at 135 C. for two hours will be quite sufficient. The following table 
 gives the results of a few tests obtained with various descriptions of cloths, 
 the tests being tabulated in the manner adopted by the author : 
 
 Description of Cloth. 
 
 Air Dry. 
 
 After Heating. 
 
 Tendering. 
 
 
 
 
 
 
 
 Warp. 
 
 Weft. 
 
 Warp. 
 
 Weft. 
 
 Warp. 
 
 Weft. 
 
 
 Ibs. 
 
 Ibs. 
 
 Ibs. 
 
 Ibs. 
 
 per cent 
 
 per cent. 
 
 Hessian Insertion, 
 
 430 
 
 397 
 
 360 
 
 322 
 
 16 
 
 9 
 
 Wigan, . 
 
 280 
 
 278 271 
 
 272 
 
 3 
 
 2 
 
 Heavy Duck, 
 
 320 
 
 311 301 
 
 303 
 
 6 
 
 2 
 
 Light Cambric, . ; 146 
 
 146 
 
 132 
 
 128 
 
 9 
 
 12 
 
 Cotton Paramatta, 
 
 94 
 
 86 
 
 71 
 
 62 
 
 24 
 
 28 
 
 Union ,, ; 108 
 
 80 
 
 97 
 
 71 
 
 10 
 
 9 
 
 Wool ,, 96 
 
 91 
 
 90 
 
 89 
 
 9 
 
 2 
 
 Gloria Silk, 
 
 130 
 
 82 
 
 111 
 
 74 
 
 14 
 
 9 
 
 Determination of Shrinkage. The determination of the shrinkage of 
 the fabrics used both for insertions and for the manufacture of waterproof 
 tissues will not infrequently be found of value. It is carried out by pour- 
 ing hot water over a piece of cloth, the length and width of which have 
 been accurately measured, leaving it immersed overnight, and drying it 
 afterwards at a moderate temperature without stretching. The resulting 
 decrease in length and width gives the shrinkage, which is usually expressed 
 in percentages. Shrinkages up to 6 per cent, in either direction are 
 generally found to be unobjectionable. The shrinkage of cotton goods is 
 generally below, that of unions or woollens above, this figure. In very low 
 class woollen fabrics the shrinkage is frequently very considerable. 
 
 Determination of the Dressing. It is quite obvious that the presence of 
 a large amount of dressing or sizing materials in fabrics to be used either 
 as in insertions or for the manufacture of waterproof goods is objectionable, 
 as these materials, without exception, tend to interfere with the adhesion of 
 the india rubber to the fabrics. 
 
 In fabrics to be used for insertions, no dressing, sizing, or filling material 
 of any description should be allowed. In the various descriptions of fabrics 
 used for the manufacture of waterproof cloth, the * finish ' of the latter is 
 often a matter of some importance. In many cases it can be obtained only 
 by the aid of foreign materials, and no hard and fast rules can be laid down 
 as to what materials and what quantities should be permitted. 
 
 Cotton goods are generally dressed with flour, starch, and similar 
 materials, often with additions of glue, gelatine, tallow, soap, paraffin wax, 
 
2l8 INDIA RUBBER AND ITS ANALYSIS. 
 
 zinc chloride, and magnesium chloride. China clay, glycerine, glucose, and 
 Epsom salt are used for weighting. 
 
 Woollen goods are dressed with Carraghen moss, gelatine, glue, dextrin,, 
 starch, albumen, and sodium silicate. 
 
 Silk finishes very commonly contain flea-wort, tragacanth, shellac, and 
 gelatine. 
 
 In examining a given fabric for the presence of dressing or filling 
 materials, no detailed analysis is required. For the purpose of the rubber 
 manufacturer it is therefore quite sufficient to first dry a weighed piece of 
 the cloth to be examined in a water oven for a period of at least two hours, 
 and ascertain the loss of moisture by weighing. It is, perhaps, needless to> 
 point out that the weighings must be performed in tightly stoppered glass 
 vessels. 
 
 The dried fabric is then thoroughly well washed in boiling water, the 
 fingers being used to rub and knead every part of the sample. This wash- 
 ing process is repeated in several changes of hot water. The washed piece 
 is then wrung out and dried in the water oven until the weight ceases to 
 sensibly decrease. The loss in weight is due to the presence of extraneous 
 matter. 
 
 There is only one constituent, very commonly occurring in textile 
 fabrics of every description, the amount of which is in every case 
 determined by a special test. This constituent is talloiv, or fatty substances 
 in general, an excessive amount of which is considered to affect the india 
 rubber very injuriously, although, in my own opinion, this danger has 
 been much exaggerated. 
 
 The determination of the percentage of fatty matter contained in a given 
 sample of cloth is carried out by first drying it in a water oven for at least 
 two hours. The weight of the dry cloth being ascertained, the latter is 
 transferred to a Soxhlet extractor and extracted with ether or acetone, 
 the extract being collected in a weighed flask. After extraction the 
 solvent is distilled off in a water bath, the flask containing the fatty extract 
 dried for thirty minutes at 90 C., and subsequently weighed. 
 
 An amount of ' fatty extract ' not exceeding 2 per cent, of the weight 
 of the dried cloth is accepted by the rubber manufacturer as involving 
 no risk to his production. This, however, should not be relied upon 
 too much by the chemist, for while on the one hand, given favourable 
 conditions, this allowance might be, without risk, considerably increased, 
 there is, on the other hand, no doubt that much less than this amount may 
 seriously affect the india rubber, should the cloth at the same time contain, 
 slight quantities of copper or manganese. Even certain dyes, notably 
 Bismarck brown, will, in the presence of very small quantities of ' grease,' 
 affect the india rubber very seriously, although in the absence of any fatty 
 matter no perceptible action would have taken place. NOT should the fact 
 be overlooked that different kinds of ( fatty matter ' differ very markedly 
 in their action upon iudia rubber. In this respect it may be stated that 
 
CONSTRUCTIVE COMPONENTS OF INDIA RUBBER ARTICLES. 2 19 
 
 the more pronounced the tendency of a fatty substance towards spon- 
 taneous hydrolysis in other words, to turn rancid the more energetic is 
 its action upon india rubber. 
 
 Examination of the Dyes. These textile fabrics which are employed in 
 the dyed state, as is the case with the vast majority of the cloths used for 
 the manufacture of waterproof cloth, must always be subjected to an 
 examination respecting the fastness of the colour and of the mordants used 
 in conjunction with the latter. 
 
 According to the conditions to which waterproof cloth is subjected ira 
 its ultimate use, it is evident that the fastness of the dye to rain, street 
 mud, and dust is essential. 
 
 Fastness to Rain. The obvious test might appear to consist in examin- 
 ing the behaviour of the dye in the course of several hours' immersion in 
 cold distilled water, when the latter should only be very slightly dis- 
 coloured. Xo doubt, if all waterproof cloth were worn in the open country 
 only, this test might be considered fairly satisfactory, but the rain falling 
 in large towns differs greatly from distilled water. Amongst the impurities 
 sulphurous acid stands prominent, and although rain water contaminated 
 with this acid does not appear to perceptibly enhance the tendency of dyes 
 to 'bleed,' its effect upon a considerable number of dyes is, nevertheless, 
 very pronounced. 
 
 I prefer, therefore, to test the cloth by immersing part of a strip of it 
 in a 0*05 per cent, solution of sulphurous acid in water, wringing out, and 
 drying at a moderate temperature. By comparing the change thereby 
 produced on the immersed part of the strip with the untreated part, 
 much more satisfactory information is gained respecting the behaviour of 
 the cloth in rain than by the distilled water test. 
 
 It is impossible to state definitely what degree of discoloration pro- 
 duced in the above test should be obtained to justify the rejection of any 
 given sample for waterproofing purposes. The judgment required for this 
 purpose can only be obtained by practical experience. As a general state- 
 ment, it may be said that any stain developed in this test becomes the 
 more objectionable, the greater the difference between the shade of tha 
 stain and that of the cloth. Reduction of the intensity of the shade, at 
 any rate within certain limits, is therefore much less objectionable than 
 the production of a stain of a different colour. 
 
 Resistance to Street Mud and Dust. The colours of waterproof cloaks are 
 often very badly affected by street mud, more especially that of towns, which, 
 for many reasons, is always distinctly alkaline (ammonia), and invariably 
 contains, besides more or less indifferent mineral detritus, a great deal of 
 organic matter, generally in a decomposed state, and also surprisingly large 
 quantities of iron oxides. The latter are, no doubt, chiefly derived from; 
 horse-shoes, human footwear, vehicle tires, and tram rails. It will be- 
 readily understood that a complex mixture of this description affects many 
 colours very energetically. 
 
220 INDIA RUBBER AND ITS ANALYSIS. 
 
 Undoubtedly the most reliable, if not the most elegant, method of 
 testing the fastness of the dye against street mud would be to treat the 
 cloth to be tested with some typical specimen of street mud, allowing it 
 to dry on the cloth, subsequently brushing it off and observing the effect. 
 It is preferable, however, to have a test solution which can always be 
 obtained of the same efficiency, and in this respect a 2 per cent, solution 
 of ammonium sulphide answers very satisfactorily. The best way to 
 employ it is to saturate a small part of the cloth with the solution, and to 
 allow it to dry, when its effect can immediately be judged. It must, 
 however, be pointed out that although in a great many cases the effect of 
 this reagent very closely simulates the action of street mud, it does not do 
 this without some notable exceptions. I have particularly observed that 
 with all cloths mordanted or saddened with tannic acid materials, the 
 difference is very considerable, the street mud showing a much greater 
 effect. I believe this to be due to the presence in the latter of iron oxides, 
 or, rather, of soluble salts of iron. 
 
 Examination of Mordanting Materials. Most of the mordanting 
 materials used in the dyeing of textile fabrics are entirely unobjectionable 
 from the point of view of the rubber manufacturer. There are, however, 
 two metals, the presence of which must be carefully guarded against : 
 manganese and copper. Of these two, manganese is now very little used. 
 Its presence is readily detected by boiling a small strip of the suspected 
 cloth with dilute hydrochloric acid, filtering the resulting solution, adding 
 to it excess of ammonium chloride and ammonia, and again filtering. On 
 adding to the filtrate hydrogen peroxide, the presence of manganese is 
 indicated by the immediate formation of hydrated manganic peroxide. 
 
 Copper occurs very frequently in dyed fabrics of almost every 
 description, and as extremely small quantities of this metal, or, rather, its 
 oxides and salts, are capable of detrimentally affecting india rubber, the 
 examination of all dyed fabrics for copper is a matter of considerable 
 importance. Enormous losses have frequently been caused by the neglect 
 of this very simple precaution. 
 
 With rare exceptions the quantities of copper to be tested for are very 
 small, but although, generally speaking, almost any quantity of copper in 
 cloth intended for proofing is objectionable, it is, nevertheless, desirable to 
 ascertain always the amount present. 
 
 This is carried out most expeditiously by means of a colorimetric test, 
 in which the colour of the ammoniacal solution of the copper obtained 
 from a weighed quantity of the cloth is compared with an ammoniacal 
 copper solution of known strength. It is at the outset, therefore, 
 of considerable importance to prepare such an ammoniacal standard 
 solution, the colour of which is certain to keep unaltered for at least 
 twelve months; otherwise, one of the various colorimeters, or LOVIBOND'S 
 tintometer, may be employed. 
 
 The colour scale is prepared as follows : About 0*3 grms. of pure 
 
CONSTRUCTIVE COMPONENTS OF INDIA RUBBER ARTICLES. 2 2 I 
 
 copper foil are dissolved in 5 c.c. of nitric acid (1*4 sp. gr.) and 5 c.c. of 
 concentrated sulphuric acid. This solution is evaporated until the heavy 
 vapours of sulphuric acid begin to appear. The mass is allowed to cool, 
 diluted with 25 c.c. of water and enough ammonia to produce a clear 
 blue solution, which is then further diluted with a mixture of one volume 
 of ammonia with six volumes of water until a solution is obtained contain- 
 ing 0*0005 grms. of copper in 1 c.c. 
 
 Next, a set of eight comparison tubes is selected holding 20 c.c. of water 
 without being filled to more than one half, and it is essential that the 
 bores of these tubes should be exactly similar. This is ascertained by 
 placing them in a test tube stand, charging each of them with 20 c.c. of 
 water, and selecting a set of eight of them exhibiting the same height of water 
 column. These tubes are then carefully washed, dried, and charged with 
 1, 2, 3, 4, 6, 8, 10, 15, and 20 c.c. of the above respectively cupr- 
 ammonium solution, the volumes being made up to 20 c.c. in every case 
 with dilute ammonia (1 : 6), and the tubes finally sealed in the blow-pipe 
 as near the top of the solution as possible. We obtain thus a scale 
 corresponding respectively to O'OOa, O'OIO, 0-015, 0'020, 0*030, 0'040, 
 0-050, 0-075, O'lOO per cent, of copper in 10 grms. of substance assayed. 
 These tubes should, of course, be labelled correspondingly. 
 
 For the subsequent colour test, 10 grms. of the cloth are ignited in a 
 porcelain basin in a muffle furnace. The basin containing the ash is then 
 allowed to cool, the ash carefully and completely transferred to a small 
 dish, and boiled for a few minutes with a few cubic centimetres of dilute 
 sulphuric acid (5 per cent. H.,S0 4 ). After addition of an excess of strong 
 ammonia, the liquid is again boiled up and filtered into a test tube 
 of as nearly the same size and calibre as those of the colour scale. The 
 filter residue is washed with dilute ammonia (1:6) until the total volume 
 of filtrate, when cold, is exactly 20 c.c. The presence of copper is in- 
 dicated by a blue coloration, and on comparing its intensity with that of 
 the tubes of the scale, a fair idea of the approximate quantity of copper 
 present is readily obtained. Of course no great accuracy can be claimed 
 for this method, but considering the smallness of the percentages to be 
 tested for, even a considerable percentage error on the result is of little 
 moment. 
 
 It is very rare that the percentage of copper exceeds O'l per cent., and 
 if it does, it will only be so much more reason for rejecting the particular 
 cloth. Should it be desired, however, to approximately estimate per- 
 centages above O'l per cent., this can easily be done by taking aliquot 
 parts of the contents of the test tube diluting to 20 c.c. with dilute 
 ammonia, ascertaining the percentage in this diluted solution by means of 
 the colour scale, and multiplying the result by the reciprocal of the aliquot 
 part taken. 
 
 The experimental results quoted on page 300 leave no doubt that any 
 cloth containing above 0'005 per cent, of copper should be rejected if the 
 
222 INDIA RUBBER AND ITS ANALYSIS. 
 
 cloth is to be used for the manufacture of 'cold cured' textures. In 
 cloths used for the manufacture of 'dry heat cured' goods O'Ol per cent, 
 of copper may safely be admitted. 
 
 Not infrequently it is stated that fabrics containing chromic oxide 
 Cr 2 3 are liable to deleteriously affect india rubber. This statement is, 
 however, devoid of any foundation. The oxides, or mordants, capable of 
 destroying india rubber are oxides which, either as such, or particularly in 
 the state of chlorides, act as strong oxidising agents towards unsaturated 
 hydrocarbons, being thereby transformed into subchlorides which are 
 readily reoxidisable by atmospheric oxygen. 
 
 Accordingly ferric oxide might be suspected of a similar detrimental 
 action, and, indeed, statements to this effect have sometimes been made, but 
 I have never been able to observe such an influence. Of course it is not 
 : at all impossible that under favourable conditions ferric oxide might prove 
 capable of such an action, and the fact of this possibility should certainly 
 be borne in mind. Ferric oxide mixed and vulcanised together with 
 rubber is obviously likely to behave rather differently than when the 
 oxide is only present as a dyeing mordant in the cloth. 
 
 Aniline black dyed goods, in the production of which vanadium salts 
 have been used, should be carefully avoided, the merest trace of vanadium 
 being capable of destroying the rubber. 
 
 For the detection of vanadium, the cloth is incinerated in the muffle 
 furnace, the ash transferred to a porcelain crucible, covered with a mixture 
 of equal parts of calcined sodium carbonate and sulphur, the crucible 
 covered with a lid and heated until all excess of sulphur has disappeared. 
 The melt is dissolved in a small quantity of water, the solution filtered 
 and decomposed by the addition of a slight excess of hydrochloric acid. 
 A brownish coloration of the ensuing sulphur precipitate indicates the 
 presence of vanadium. The suspended sulphur is precipitated by shaking 
 the liquid with a small quantity of amorphous silica, the precipitate 
 filtered off, dried, burnt off in a porcelain crucible, fused with a few 
 crystals of potassium nitrate, dissolved in water, acidulated with dilute 
 nitric acid, and filtered. If, on addition of a few drops of hydrogen per- 
 oxide to the filtrate, a distinct pink coloration appears, the presence of 
 vanadium is proved to a certainty. 
 
 Tannic Acid Mordants, which are much employed in dyeing, have no 
 action upon india rubber ; but if employed for the production of a range 
 of light fawn or slate shades, they are very apt to cause trouble in another 
 direction. Such cloths, when in wear, often develop numerous spotty or 
 streaky stains of a peculiar dull maroon colour. As spots, they occur 
 particularly at the hem of the garments ; as streaks, more particularly on 
 the upper parts. They are caused by almost infinitesimal traces of iron 
 coming in contact with the cloth. The spots round the hem are therefore, 
 no doubt, due to the iron in the street mud ; the streaky spots are caused 
 by water, which may be contaminated with iron in a variety of ways. 
 
CONSTRUCTIVE COMPONENTS OF INDIA RUBBER ARTICLES. 223 
 
 These stains are extremely unsightly, and once they appear cannot be 
 removed. 
 
 The only way to avoid these spots is to reject all fabrics of the above 
 description containing a tannic acid mordant, and since the discoloration 
 is due to the action of iron salts upon the tannic acid mordant, the most 
 satisfactory test is to bring upon the cloth a small drop of a solution of 
 1 grm. of ferrous sulphate and 1 grm. of ferric chloride in 1000 c.c. of 
 water, and allow it to dry upon a water oven. Any serious discoloration 
 occurring is sufficient justification for the rejection of the cloth. 
 
 Amongst the more common defects of india rubber waterproof cloth, 
 the curling of the proofed fabrics is probably the most disagreeable. It is 
 often considered to be due to a shrinkage of the cloth, but this produces 
 the defect known as cockling, also a very unsightly defect. As a matter 
 of fact, the curling defect is always due to either a faulty composition of 
 the rubber mixing employed for the proofing, to a ' curing ' fault, or to 
 some omission or commission in the process of proofing (see page 303). 
 Quite a number of causes may produce this defect, but the enumeration 
 and discussion of these is a subject for a treatise on the manufacture of 
 india rubber waterproof cloth, which lies outside the scope of the present 
 work. 
 
CHAPTER IX. 
 ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 INDIA RUBBER articles may require investigation for a number of reasons : 
 we may desire simply to ascertain whether successively manufactured 
 batches of goods, of known composition and produced under known con- 
 ditions, are of uniform quality, or whether they are in accordance with a 
 specified standard j or we may desire to determine the composition of an 
 article in order to reproduce a similar product ; or we may desire to ascer- 
 tain whether an article, the composition of and conditions of manufacture 
 being unknown, fulfils certain specified conditions ; or, finally, we may re- 
 quire to ascertain the causes responsible for an article differing from a 
 given standard, or showing a variety of manufacturing defects, or under- 
 going one or other of the various changes collectively described as ' decom- 
 position.' In all these cases an analysis and general physical examination 
 will be found of the utmost value. 
 
 In a number of cases it will be observed that physical tests are the 
 quickest means of gaining the desired information. This applies especially 
 to the control of the production of a factory, or to the comparison of a 
 delivery with a standard sample, and we infer from equal physical constants 
 the chemical equality of the sample tested with the standard. This infer- 
 ence is only admissible with very considerable limitation; as a general 
 statement it is without any truth, and grossly misleading. It is all the 
 more necessary to point this out, as there are still some people who declare 
 the chemical analysis of rubber goods as valueless, and advocate reliance 
 upon the testimony of the physical tests alone. This attitude is due, no 
 doubt, partly to ignorance and inexperience a still greater part, however, 
 to the fact that an india rubber analysis is at the present moment capable 
 of satisfactory interpretation by those only possessing a long and exten- 
 sive practical manufacturing experience. The cause of this lies not only 
 in the fact that our chemical knowledge of india rubber is still very im- 
 perfect, but also in the circumstance that the physical conditions under 
 which an india rubber article is produced affect its properties quite as 
 much as does its chemical composition. We find, therefore, that although 
 articles produced with the same materials and under identical physical 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 225 
 
 conditions possess the same physical and chemical constants, it is, as a 
 general principle, equally inadmissible to infer from the identity of the 
 physical tests of different samples their chemical identity, as to pro- 
 nounce different samples of identical chemical composition to be physically 
 identical. 
 
 The course chosen should therefore be adapted to current practice, as, 
 for example, in the iron and steel industry, where the physical tests and 
 the chemical analysis mutually complement each other. This course 
 should recommend itself all the more for general adoption, as it is equally 
 satisfactory in the case of rubber articles chiefly subjected to mechanical 
 wear, as to those subjected to chemical action, or to mechanical and chemical 
 action combined. 
 
 Of course the necessity of taking a sample through all the physical 
 tests, and subjecting it at the same time to a full chemical analysis, will 
 hardly ever arise, and it is very easy to decide in the case of any given 
 sample what physical tests to apply, and to what length to carry the 
 chemical analysis. In the vast majority of cases, one or two physical tests, 
 and the determination of two or three of the chemical constituents, are 
 amply sufficient. A full analysis is practically only required when an 
 article has to be produced equal to a sample of unknown composition, or in 
 cases where the causes of the premature break-down or decay of rubber 
 goods have to be investigated. Under such circumstances the complete 
 analysis may become a rather protracted piece of work, but in the former 
 of the above mentioned cases it is a far more rational proceeding to expend 
 a few days of analytical work, instead of taking empirical and random 
 shots in the factory, which are far more expensive and generally much more 
 tedious, and which may or may not be successful. In the investigation of 
 the causes of decay of rubber goods, all physical tests are useless. 
 
 (a) PHYSICAL TESTS FOR MANUFACTURED INDIA EUBBER. 
 
 1. Specific Gravity. Time was when the 'specific gravity test' was 
 looked upon as the most reliable guide for judging the quality of a rubber 
 article. The advent of fatty substitutes, the increasing use of resinous 
 and hydrocarbon compounds, of * ground waste ' and reclaimed rubber, has 
 rendered this constant at present almost valueless, and the only inference 
 the low specific gravity of a sample enables us to draw to-day is that it 
 contains little or no mineral matter. Of course this conclusion does not 
 imply that such an article is one of high quality, although there is still a 
 large number of consumers who regard it in this light, and who insist upon 
 having the goods delivered to them of 'floating quality.' 
 
 All india rubber articles contain more or less considerable quantities of 
 gases, chiefly nitrogen. It is doubtful whether these gases are dissolved 
 in the india rubber, the probability being that they simply fill the pores, 
 which unavoidably exist in manufactured india rubber, but which are, as a 
 
 15 
 
226 INDIA RUBBER AND ITS ANALYSIS. 
 
 rule, so minute as to be invisible to the unaided eye. On account of this 
 occlusion of gases, a rubber article possesses a real and an apparent specific 
 gravity according to the conditions under which this constant is determined. 
 It will further be obvious that the apparent specific gravity is always the 
 lower of the two, and also that all admixtures which tend to increase 
 porosity will increase the difference between the real and the apparent 
 specific gravity. It is therefore very commonly to be observed that a 
 'floating' rubber article, on careful examination, is found to possess a 
 specific gravity considerably above that of water, so that what keeps it 
 afloat must be evidently the buoyancy due to the presence of occluded 
 
 These very simple facts, while showing, on the one hand, the comparative 
 insignificance, or even worthlessness, of the specific gravity test, show us, 
 at the same time, the ratio between the real and the apparent specific 
 gravity to be a very convenient and exact numerical expression of the 
 porosity of a rubber article. This fact has not yet found the 
 practical recognition it deserves, but is thoroughly well worth the careful 
 attention of every rubber worker. The bad behaviour of many india 
 rubber articles in use will be found to be closely connected with their 
 porosity, and I may also state that the addition of small proportions of 
 various (chiefly organic) substances to india rubber mixings additions 
 which frequently appear to be utterly worthless factory superstitions in 
 reality serve the purpose of reducing porosity. 
 
 I find it convenient for practical purposes to express this ratio p in the 
 following terms : 
 
 = / real specific gravity _ A 
 \apparent specific gravity / 
 
 which, for an absolutely non-porous rubber, would give p = 0. The factor 
 1000 is, of course, introduced simply in order to obtain whole numbers 
 instead of fractions for p. 
 
 It will be readily understood that in determining the apparent and the 
 real specific gravity respectively, the samples must be used in such a way, 
 and the tests carried out under such conditions, as to ensure in the former 
 case the retention, and in the latter case the complete expulsion, of all 
 occluded air. 
 
 The apparent specific gravity should be determined with pieces of the 
 sample as large as may conveniently be used. These pieces are floated in 
 water in a beaker of suitable dimensions. If the piece floats on the sur- 
 face, alcohol is gradually added to the water until the test piece remains in 
 suspension at any point in the liquid. The apparent specific gravity of the 
 sample is then identical with the specific gravity of the liquid, which should 
 be ascertained by means of a Mohr's balance. If, on the other hand, the 
 test piece sinks in the water, a concentrated solution of calcium chloride is 
 added until it remains permanently suspended at any part of the liquid 
 mixture. It is scarcely necessary to mention that after every addition of 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 either alcohol or calcium chloride solution to the water, the mixture should 
 be vigorously stirred. 
 
 The real specific gravity should always be determined with the sample 
 in as fine a state of comminution as it is possible to obtain. To attain this, 
 the sample is ground between a pair of steel rollers working at different 
 speeds : a pair of small scale mixing rollers (see fig. 22) answer best for the 
 purpose. Where these are not available, working the sample with a very 
 coarse file will often supply suitable material : mere cutting into small 
 strips will not do. 
 
 If the sample consists chiefly of rubber with but little foreign admixtures, 
 the sample cannot be * crumbed ' by filing, and even on the above mixing 
 rollers it may occasionally refuse to break up, forming sheets instead. 
 These are, however, always very porous, and if they are only thin enough, 
 cutting them into narrow strips may bring them into a satisfactory 
 condition for testing. 
 
 In carrying out the test, an accurately weighed quantity of the sample, 
 prepared as above described, is placed in a pyknometer. the latter filled to 
 about three quarters with distilled water and connected to an air pump. 
 The vacuum treatment is continued until all the air has been expelled. 
 The pyknometer is then completely filled with distilled water previously 
 boiled and brought to standard temperature, and after the insertion of 
 the stopper it is weighed, and the specific gravity of the rubber calculated 
 according to the well-known equation : 
 
 d = 
 
 P W +P-P, 
 
 P = Weight of the substance (rubber) taken. 
 
 P w = Weight of pyknometer filled with water, at temperature t. 
 
 Pj = Weight of pyknometer filled with water and substance, at temperature t. 
 
 It is advisable to use, whenever possible, a quantity of substance amount- 
 ing to at least one quarter of the weight of the water the pyknometer will 
 hold. Perhaps it is scarcely necessary to mention that in determining the 
 1 porosity ' as above described, both the apparent and the real specific 
 gravity must be determined at exactly the same temperature, otherwise 
 serious errors may be committed. 
 
 It will be found that the porosity of manufactured india rubber is 
 influenced, not only by the quantity of the admixtures this is obvious 
 but very much also by their nature. The effects of the various inorganic 
 filling and colouring materials show very considerable variations, and are 
 in some cases very marked. The highest figures for p are to be found in 
 rubber charged with considerable quantities of rubber substitutes, and 
 recovered rubber. 
 
 It must, however, not be assumed that the porosity is a kind of constant 
 of the various admixtures. This is by no means true, as the same mixing 
 proportions give very different 'porosities' with different kinds of india 
 
228 INDIA EUBBER AND ITS ANALYSIS. 
 
 rubber. The highest figures are always obtained with the ' driest ' rubber 
 qualities, the soft varieties giving proportionately lower figures. 
 
 The experienced rubber technologist will therefore also readily perceive 
 that the porosity for the same mixing may vary considerably according to 
 the time during which the mixing is worked between the rollers. He will 
 further see that some of the substances frequently used to shorten the 
 time required for mixing by softening the rubber also reduce the porosity 
 of the mixing. Altogether, the determination of p is a test of very wide 
 applicability, and of great utility in the control of operations in the factory. 
 
 2. Mechanical Tests. The extremely useful tests known as the ' tensile 
 strength ' and the ' elongation ' test are carried out on a machine of exactly 
 the same construction as the one shown on page 216. The indicator scale 
 should, however, be so arranged that it stands at zero, when the two jaws 
 are exactly one inch distant from each other, so as to allow sufficient space 
 for the often very considerable elongation of the test pieces. 
 
 For the controlling of the production of a rubber factory by means of 
 these tests, it is advisable to use a machine giving readings from four to 
 five pounds up to three or four hundred pounds of tensile strength. On 
 such a machine readings from pound to pound are easily taken, and this 
 is quite accurate enough, providing the test pieces employed are of such 
 dimensions as to give a total minimum reading of not less than a hundred 
 pounds. There is, in the vast majority of cases, scarcely ever any difficulty 
 in fulfilling this condition. This applies likewise to series of test pieces 
 prepared for the study, either of the influence of different vulcanising 
 conditions upon the physical constants of a given mixing, or of the influence 
 of different admixtures under identical vulcanising conditions. 
 
 HBINZERLING and PAHL, in their elaborate tables (see pages 164-173), 
 express the tensile strength in reference to test pieces of one square milli- 
 metre section and hundred millimetres' length. I take it that their data 
 were not obtained from the examination of test pieces of this dimension, 
 ljut were subsequently arrived at by calculation. This, of course, can be 
 done on the assumption that the tensile strength varies as the sectional 
 area of the test pieces, but that the elongation at break is independent of 
 the dimensions of the test piece, and stands for the same mixing, cured 
 under identical conditions, in a constant ratio to the length actually under 
 test. As a general statement, these assumptions are approximately true, 
 but the approximation in the case of india rubber is none too close. As a 
 matter of fa'ct, according to my own observations, the tensile strength 
 increases in the case of india rubber at a slightly higher rate than the 
 sectional area, and likewise I have found that the elongation or distensibility 
 slightly decreases with an increasing cross section. 
 
 Nor is the mere statement of the tensile strength and elongation at 
 break altogether a full statement of the physical behaviour of a test piece 
 under an increasing stress. A far more instructive way of showing the 
 comportment of the test pieces is to measure the stress applied for a number 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 229 
 
 of successive and increasing elongations, and plot the data thus obtained in 
 the manner shown by the curve tracings in fig. 26. It will be seen at a 
 glance how much more expressive are these curves than merely the two 
 figures above referred to. 
 
 3. Dry and Moist Heat Test. These tests are made by exposing the 
 samples for one hour to a 'dry heat' temperature of 270 F., and for three 
 hours to a 'moist heat' temperature of 320 C. 1 The dry heat test is 
 simply carried out in an air bath ; for the moist heat test a digester is 
 required, which should be provided with a pressure gauge and thermometer 
 tube (fig. 4). 
 
 This test is obviously of some importance in the examination of rubber 
 goods which are to be exposed to heat or high pressure steam, and it is 
 generally supposed to furnish clear indications as to the presence of fatty 
 substitutes in the india rubber. This, within certain limitations, is true 
 of the moist heat test, but a deterioration of a sample in the dry heat test 
 may be due to quite a variety of causes, amongst which the presence of a 
 considerable excess of sulphur should particularly be mentioned. 
 
 While it is not to be denied that tests of this description may furnish 
 indications as to the behaviour of an india rubber sample under working 
 conditions, I strongly hold that the indiscriminate application of these tests 
 to all classes of rubber goods is often likely to be mischievous and mis- 
 leading when, as is often the case, it imposes upon rubber goods test 
 conditions which bear no relation to the conditions under which the goods 
 are used. 
 
 4. Suncracking Test. This test does not, strictly speaking, fall under 
 the heading of physical tests, but it finds, nevertheless, a suitable place in 
 this connection. 
 
 The liability of india rubber goods of various descriptions, but more 
 particularly tire covers, to develop, often to a very disagreeable degree, the 
 defect known as suncracking, is very generally recognised. It is the 
 result of the oxidation of the india rubber by the atmosphere. The term 
 ' suncracking ' might therefore appear to be a misnomer, but, as a matter 
 of fact, it emphasises very expressively the marked hastening of this 
 oxidation under the influence of direct sunlight. 
 
 To test the liability of any given sample of india rubber to suncracking, 
 the most satisfactory way would appear to be to expose the sample to the 
 sun's rays for a certain period, perhaps together with some standard 
 sample. This, of course, renders the test dependent upon the weather and 
 season, and at best is a very tedious process, so that it is of some importance 
 to have a more controllable test for this purpose. 
 
 Suncracking being, as above stated, due to oxidation, the use of some 
 
 form of oxygen more active than atmospheric oxygen at once suggests 
 
 itself. Hence, no doubt, the repeatedly made suggestion to expose the 
 
 india rubber to the action of ozone. The energetic action of ozone upon 
 
 1 These two tests are collectively known as the ' Admiralty Test.' 
 
230 INDIA RUBBER AND ITS ANALYSIS. 
 
 India rubber is well known, but it is by no means easy to regularly obtain 
 a gas of the same concentration, and without this the results obtained 
 would be almost worthless. 
 
 Of the numerous oxidising agents, nitric acid could not be used, as it 
 not only oxidises india rubber, but also forms nitro and nitroso compounds 
 with it. There is no such objection in the case of chromic acid and the 
 alkaline chromates and permanganates, but the aqueous solutions in 
 which these have to be employed preclude any but the merest surface 
 action. This also applies to aqueous solutions of hydrogen peroxide. It 
 has been stated that an ethereal solution of hydrogen peroxide has a strong 
 action upon india rubber, converting it completely into a resinous mass 
 soluble in alcohol. Unfortunately, however, this statement appears to be 
 devoid of foundation at least, I have never observed any action whatever 
 of either dilute or concentrated ethereal solutions of hydrogen peroxide 
 upon india rubber. 
 
 Some years ago WOLFFENSTEIN 1 described an acetone peroxide 
 [(CH 3 ) 2 .C0 2 ] 3 obtained by the action of hydrogen peroxide upon acetone. 
 This compound dissolved in benzene readily attacks india rubber. In the 
 dry state it is, however, highly explosive, and I prefer, therefore, not to 
 isolate it, but to employ it for the test as follows : 
 
 20 grms. of acetone are mixed with 60 c.c. of 20 per cent, solution of 
 hydrogen peroxide, and the mixture, which will keep unaltered for a long 
 time, is allowed to stand for at least a fortnight before using. The test 
 is then carried out by immersing in this mixture, for a period of at least 
 two days, weighed strips of equal area of both the sample to be tested 
 and the standard with which it is to be compared. After two days' 
 immersion the samples are withdrawn, washed with a little acetone and 
 water, and dried at 100 C. The increase in weight denotes the amount 
 of oxygen absorbed, and is a fairly reliable measure of the relative sun- 
 cracking liability of the sample tested. 
 
 I have so far applied this test to the examinations of tire covers only, 
 and the results have proved very trustworthy and in close agreement with 
 the observations on strips exposed to sunlight for prolonged periods. 
 
 5. Electrical Tests. The two most important electrical tests in 
 reference to india rubber, and with which every chemist engaged in 
 practical india rubber work ought to be sufficiently familiar to dispense 
 with the assistance of an electrician, are the insulation test and the 
 electrostatic capacity test. These are carried out in the following 
 manner : 
 
 Measurement of Insulation Resistance. 
 
 The apparatus required consists of a galvanometer of high sensitiveness. 
 A reflecting galvanometer of the THOMSON type with a resistance of about 
 
 1 er., xxviii. p. 2265. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 2 3 I 
 
 20,000 ohms is preferable, but for many purposes the much less expensive 
 D'ARSONVAL galvanometer will suffice. Either of these galvanometers is 
 worked with any of the well-known forms of transparent scale. In parallel 
 with any of these galvanometers, some convenient shunt is employed, by 
 means of which any desired fraction of the current may be sent through 
 the galvanometer. To determine the constant of the latter, the most 
 convenient arrangement is a megohm resistance. The further appliances 
 required are a battery of at least a hundred cells, a battery key, and a 
 short circuit key, together with two strong, perfectly plane brass plates 
 provided with binding screws. 
 
 The first operation is to ascertain the constant of the galvanometer. 
 For this purpose the battery, short circuit key, megohm resistance, and 
 galvanometer are connected in series, the shunt being connected in parallel, 
 with the galvanometer in the manner indicated in the diagram fig. 19. 
 
 FIG. 19. Diagram showing disposition of apparatus for taking the 
 galvanometer constant. 
 
 In this diagram, g represents the reflecting galvanometer, s the uni- 
 versal shunt, b the battery, m the megohm resistance, and k the short 
 circuit key. For ordinary practice a reflecting galvanometer of the 
 D'ARSONVAL type possessing a resistance of about 800 ohms is sufficient. 
 The employment of a key in the galvanometer circuit, which is closed only 
 when the short circuit k (battery circuit) has been closed, is indispensable. 
 
 In taking the constant of the galvanometer, the centre line of the light 
 spot on the screen is brought to zero, the shunt switch turned on the knob 
 marked 1/1000, the battery key closed first, then the galvanometer key. 
 The light spot will move up on the scale, and as soon as it has become 
 stationary, its position is read off. By taking the number of degrees thus 
 read, and multiplying it by 1000 (the dividing value of the shunt), we 
 obtain at once the constant of the galvanometer i.e., the number of the 
 scale deflection which would have been produced by the battery current 
 acting upon the galvanometer through a resistance of 1 megohm with 
 the shunt at 1, when the whole of the current would pass through the 
 galvanometer. 
 
232 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 The constant being ascertained, we can now proceed to measure the 
 insulation resistance of any material. India rubber should be used in the 
 shape of a sheet or film, by substituting it for the megohm resistance in 
 the manner shown in the diagram fig. 20. 
 
 In this diagram, g again represents the galvanometer, s the shunt, 
 k the short circuit key, and b the battery. The sheet to be tested (i) 
 is placed upon the earth plate e, which is connected to the wire coming 
 from the short circuit key. Upon the sheet under test is placed the 
 circular brass plate p, which is connected to shunt and galvanometer. 
 Surrounding this brass plate is the guard ring r, made of brass, which is of 
 larger diameter than the brass plate p. This ring is connected to the 
 galvanometer side of the battery through the wire /, and it serves to 
 annul the effect of any surface leakage of current over the surface of the 
 sheet i. 
 
 FIG. 20. Diagram showing disposition of apparatus for determining 
 insulation resistances. 
 
 In practice it is also necessary in this case to employ a separate key in 
 the galvanometer circuit. Great care must also be taken to have the 
 shunt switch turned on the 1/1000 knob before closing the battery key, 
 and then to close, in its turn, the galvanometer key. 
 
 On depressing the galvanometer key, the light spot on the scale rapidly 
 moves up, and after a few seconds begins to move again slowly backwards. 
 The slower and steadier this backward movement proceeds, the better the 
 insulating material under test, apart from the actual insulation resistance 
 ultimately ascertained. After one minute's electrification, the reading of 
 the position of the spot on the scale is taken. 
 
 If the amount of deflection produced is very slight, the shunt switch is 
 shifted to the knobs bearing successively lower dividing values until a 
 distinctly readable deflection is obtained. 
 
 The insulation resistance is then calculated in the following manner : 
 Assuming that in taking our constant we obtain, with the shunt switch at 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 233 
 
 1/1000, a deflection of 15 degrees on the scale, then the galvanometer con- 
 stant for a resistance of one megohm is 15x1000=15000. We then 
 remove the megohm resistance, inserting in its place the two brass plates, 
 between which the dielectric material to be tested has been placed. We 
 then put the guard ring r in position, concentrically to the round brass 
 plate p, and taking care that it is in contact with the dielectric only, and 
 not, at any point, with the lower brass plate. Then we connect the guard 
 ring to the battery arm, as shown in the above illustration. 
 
 We place the shunt switch at 1/1000, close the battery key, and sub- 
 sequently the galvanometer key. If no readable deflection be obtained, 
 the shunt switch is turned on to the next following knob 1/300. Assuming 
 now we obtain a deflection of 9 degrees, the deflection without the shunt 
 would have been 9x300 = 2700 scale divisions. Now, as the insulation 
 resistance is inversely proportionate to the deflections, the following pro- 
 
 portions obtain : 
 
 15000 : 2700 ::*:!, 
 or 15000 
 
 2700 
 x = 5-5 n. 
 
 This figure is, of course, only relative i.e., it is inversely proportional to 
 the area of the brass plate p t but directly proportionate to the thickness of 
 the dielectric. Consequently, in stating the result, the area of that plate 
 and the thickness of the dielectric must be stated at the same time. Thus, 
 if the area of the test plate is 200 sq. cm., the thickness of the sheet 3 mm., 
 the result would be recorded thus : 
 
 Resistance of ' Insulating Compound ' per 200 sq. cm. area, 3 mm. thick = 5 '5 fl. 
 
 Or, if it should be desired to reduce the result to a smaller unit area of 
 the same thickness, this can be done by multiplying the result by the area 
 of the test plate, and dividing the product by the unit area selected. 
 
 To ascertain the thickness of such plates or sheets of dielectric materials, 
 the screw micrometer is the most convenient instrument, and is sufficiently 
 accurate. In working with india rubber sheets which have frequently to 
 be tested in the unvulcanised plastic state, the usual form of these instru- 
 ments with very narrow contact pins is, however, rather unsatisfactory, the 
 readings, owing to the yielding nature of the material to be measured, 
 being liable to be too low. For this reason, that form of micrometer with 
 special broad contact plates, as made for paper makers' use, is much to be 
 preferred. This instrument (see fig. 21) should also be provided with a rack 
 work head to prevent the application of undue or variable screw pressure. 
 
 The electromotive force of the battery to be employed in the above tests 
 depends, of course, materially upon the nature of the dielectric substance 
 to be tested. On the whole, it may be said that the results are the more 
 trustworthy, the higher, within certain limits, the E. M. F. of the testing 
 battery. For tests on india rubber and gutta percha, a battery of from 
 3 to 400 cells (5 to 600 volts) is generally used ; but in many cases it may 
 
234 INDIA RUBBER AND ITS ANALYSIS. 
 
 be desirable to employ a lower voltage, and when experimenting with some 
 new material, it is generally advisable to ascertain in a few preliminary 
 tests the most suitable number of cells to employ. 
 
 The above described insulation test is, of course, the most important 
 test applied to technical dielectrics, but it is not the only important test. 
 The electrostatic capacity, also the coefficient of self-induction of a 
 dielectric material, are in many cases of almost equal importance, especially 
 in submarine and telephone cables. But as these tests require a very 
 fully equipped electrical laboratory or test room, which is generally under 
 
 FIG. 21. Micrometer. 
 
 the control of a specialised electrician, their detailed description here 
 would be out of place, and for full information on this subject I refer the 
 student to the larger works on electrical testing. 1 It is certainly indis- 
 pensable that every chemist interested in the electrical properties of india 
 rubber and its congeners should possess not only a sound theoretical 
 training in electricity, but should also be thoroughly familiar with the 
 practical methods of electrical testing. 
 
 (6) CHEMICAL ANALYSIS OF MANUFACTURED INDIA RUBBER. 
 
 The great value of the physical tests applied to india rubber articles 
 has long been recognised, but, until comparatively recently, chemical tests 
 were only used in a somewhat random fashion. The analytical investiga- 
 tion of manufactured india rubber, and the analytical control of the works' 
 production, is even at present practised by a few only, and is by some still 
 looked upon merely as a gratuitous attempt on the part of the chemist to 
 supersede the 'practical man,' and to intrude where he is neither com- 
 petent nor required. This attitude of the 'trade,' unfortunately, does not 
 1 KEMPE'S Handbook of Electrical Testing may be particularly recommended. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 235 
 
 even exclude that of a considerable number of the heads of India rubber 
 works. There is no doubt that at the present moment the chemical 
 analysis of india rubber has not yet reached the importance to which it is 
 certain ultimately to attain, but this is largely, if not entirely, due to the 
 fact that our means for the interpretation of the already numerous 
 analytical data which are available are very scanty and inadequate. But 
 this defect is clearly remediable, and its force will become increasingly less 
 as the chemistry of india rubber progresses. Already, at the present 
 moment, chemical analysis is capable of elucidating many of the most 
 difficult and abstruse points with which we are confronted in the 
 manufacture, of often furnishing an explanation of the causes of premature 
 decay in rubber goods, and, in conjunction with the physical tests, of 
 supplying in the fullest manner all the necessary data for the satisfactory 
 control of the production of the factory. 
 
 Thus, although at the present moment the advantages of chemical 
 analysis in reference to india rubber are necessarily restricted by our very 
 fragmentary knowledge of india rubber from the purely chemical point of 
 view, this, so far from being an argument for the exclusion of the 
 resources of chemistry from the india rubber industry, should rather act 
 as a stimulant to bring to bear upon the working methods of this great 
 branch of manufacture the possibilities lying in chemical analysis and 
 research, which, in the century just closed, have proved themselves the 
 most powerful agents for quite unexampled developments of old, as for the 
 creation of new, industries. 
 
 General Scheme of India Rubber Analysis. 
 
 The fundamental difficulty of the analysis of india rubber, as, indeed, 
 of most technical products, lies in the fact that what is required is not an 
 elementary analysis, but a determination of the independent composite 
 substances contained in the manufactured goods. Seeing the almost 
 countless number of substances, both organic and inorganic, which may, 
 with or without some clearly conceived reason, be incorporated in a 
 ' rubber mixing,' it is obvious that in many cases the complete qualitative 
 and quantitative analysis of a rubber article may be a very formidable 
 task, taxing in the highest degree the resources and perseverance of the 
 chemist. Indeed, in many cases the definite qualitative identification, of 
 some of the minor constituents especially, may be quite impossible, and in 
 other cases where we may succeed in identifying a number of these 
 constituents, their quantitative determination may be found impossible. 
 I may, however, at once point out that cases where these difficulties render 
 the analyst's efforts altogether nugatory are of the rarest occurrence. As 
 a matter of fact, the very detection of the presence of such substances 
 which defy exact identification, and consequently also quantitative 
 determination, very frequently gives sufficient indirect information respect- 
 ing their nature as to lead to their ultimate recognition. 
 
236 INDIA RUBBER AND ITS ANALYSIS. 
 
 It will be evident that in a very large number of cases there is 
 absolutely no necessity for a full and detailed analysis, and it is frequently 
 quite sufficient to determine the various constituents in groups, or to 
 determine the components of one only of these groups. The objection that 
 india rubber analysis can be accomplished only with an inordinate 
 expenditure of time is therefore quite inadmissible. On the other hand, 
 over and over again it can be observed that weeks of time and considerable 
 expenditure are wasted in ' practical experiments ' in the factory, where an 
 analysis would have at once solved the whole difficulty. 
 
 In its broadest sense the analysis of india rubber consists in the deter- 
 mination of the following five main groups : 
 
 I. India Rubber. 
 
 II. Organic Constituents not Rubber. 
 
 III. Sulphur. 
 
 IV. Chlorine. 
 
 V. Inorganic Constituents other than Chlorine and Sulphur. 
 
 In actual practice, even in those cases where, for the purpose in view, 
 a division into these broad groups would suffice, the analysis cannot be 
 carried out so as to yield simply the respective proportions in which either 
 of these four groups is present in any given sample. This could only be 
 done in respect of groups III., IV., and V., but the result in respect of 
 the former would be almost worthless unless the absence of sulphuretted 
 substitutes were known with certainty. 
 
 The above five groups comprise within them the following sub- 
 divisions : 
 
 I. a. India Rubber. 
 
 b. Recovered India Rubber. 
 
 c. Balata. 
 
 d. Gutta Percha. 
 
 II. a. Fatty Substitutes (Sulphides, or Chlorosulphides). 
 
 b. Fatty Substitutes (Oxidised or ' Blown ' Oils). 
 
 c. Free Fatty Oils. 
 
 d. Mineral and Resin Oils. 
 
 e. Solid Hydrocarbons (Paraffin Wax, Ceresin, Ozokerite). 
 /. Resins. 
 
 g. Pitches (Coal Tar Pitch, Stearine Pitch). 
 
 h. Bituminous Bodies (Elastikon, Asphaltum). 
 
 i. Carbohydrates (Starch, Cellulose). 
 
 k. Organic Bodies not above named. 
 
 III. a. Sulphur of Vulcanisation. 
 
 b. Free Sulphur. 
 
 c. Sulphur in Organic Constituents other than Rubber. 
 
 d. Sulphur in Inorganic Constituents. 
 
 IV. a. Chlorine in India Rubber. 
 
 b. Chlorine in Fatty Substitutes. 
 
 V. a. Filling or Weighting Agents. 
 
 b. Sulphur Carriers. 
 
 c. Colouring Agents Pigments, Lake Pigments. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 237 
 
 This list appears formidable, but, as a matter of fact, it is no more so 
 than the list of constituents often ascertained in inorganic analysis. Just 
 as ultimate analysis is simplified through the use of 'group reagents,' so 
 in the same way the constituents of india rubber articles can be separated 
 into groups. We obtain in this manner, in the first instance, a very general 
 view of the composition of the article under analysis, which is then further 
 developed by subdividing these different groups into their several con- 
 stituents. 
 
 The Preparation of the Samples for Analysis. The procuring of a 
 representative average sample for analysis is one of the most important 
 operations, but certainly also frequently one of the most difficult. The 
 proportion of the various constituents is scarcely ever the same throughout 
 the sample indeed, it varies often very considerably, alike to the dis- 
 advantage of the article as of the analysis. The reason, of course, lies in 
 the difficulty of mixing colloidal masses homogeneously with each other 
 and with the various pulverulent substances employed in rubber com- 
 pounds. 
 
 It is therefore important, in order to obtain a fair average sample, to 
 procure as large a sample of the article to be analysed as possible, and to 
 grind it between a pair of 'mixing rollers.' With rare exceptions all 
 vulcanised rubber goods are thus converted into a mass of fine flakes. 
 This mass, after being well mixed, is again passed through the rollers, an 
 operation which is repeated a number of times, and in this manner a 
 material sufficiently homogeneous for analysis is obtained. 
 
 Of course, cutting a strip from end to end from a sample to be analysed, 
 and cutting this into very fine pieces, may serve as an approximation, 
 but it is a very tedious process, and the material is not in such a porous 
 and permeable condition as the above described grinding produces in it. 
 Every laboratory in which much rubber work is done should therefore 
 possess a model mixing roller. This may be worked by hand as long as 
 the width of the rollers does not exceed 6 inches ; beyond this, mechanical 
 power is required. Model machines of this kind in various sizes are made 
 by Messrs. J. Robinson, Salford, Manchester, and fig. 22 illustrates a 
 laboratory 'dry-mixer' of their design. 
 
 It is scarcely necessary to point out that before passing any rubber 
 articles through this machine, constructive materials, such as cloth inser- 
 tions, etc., must be carefully removed. This offers little or no difficulty 
 in all those cases where the fabric is completely exposed on one side of 
 the article, as, for instance, in waterproof fabrics or bicycle tire covers. 
 Thoroughly saturating the fabric with benzene is generally quite sufficient 
 to enable its being stripped off the india rubber. 
 
 In waterproof cloth it is, in spite of this treatment, often difficult to 
 strip the rubber films, either owing to their extreme thinness, or on 
 account of the films being too weak in consequence of the large amounts 
 of non-rubber constituents they may contain. In this case the simplest 
 
238 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 procedure is to saturate the cloth side with benzene, and, as soon as the 
 rubber film begins to shrivel, to place the fabric, rubber face upwards, 
 upon a glass plate, when the ' proofing ' can be completely scraped off by 
 means of a good sized palette knife. The rubber obtained from any 
 articles thus treated must, of course, be dried in a water oven before it is 
 further proceeded with. 
 
 ' Built up ' articles i.e., articles containing rubber mixings of different 
 composition in superposed layers (electric cables, with their * pure rubber 
 lapping,' 'inner separator,' and 'outer jacket/ are good examples of this 
 class of goods) should be separated into these respective layers, and each of 
 
 FIG. 22. Mixing machine (Laboratory scale). 
 
 these be analysed separately. In this case it is almost always desirable 
 to determine the ratios of the weights of these different layers to the 
 weight of the whole article. The separation of such layers of different 
 composition may often be accomplished by making a slight starting split 
 with a knife along the plane of union of the two layers, when frequently 
 a practically complete separation may be effected by the exercise of a 
 little patience and some manual dexterity. Occasionally this separation 
 takes place with quite surprising ease, but quite as often it may be found 
 utterly impracticable, and in this case a tedious dissection by means 
 of a sharp knife is the only means available. This, of course, renders 
 a determination of the weight ratios of the different layers by the above 
 suggested process practically impossible, but a very satisfactory and 
 accurate way of obtaining these data consists in the preparation of a 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 239 
 
 cross section of the sample, measuring the exact thickness of each layer 
 (either macroscopically or microscopically, according to the thickness of 
 the object), and multiplying the thicknesses measured by the specific 
 gravities of the respective layers. The figures thus obtained are exactly 
 the desired ratios. 
 
 The samples prepared for analysis should be kept in air-tight bottles, 
 which should not be exposed unnecessarily to daylight, and never to direct 
 sunlight. Samples which are to be preserved for future reference, par- 
 ticularly those which are or may become objects of legal proceedings, must 
 be preserved in sealed glass tubes. 
 
 Analytical Methods. A brief glance at the list of the chief constitu- 
 ents to be met with in manufactured india rubber shows that they may 
 contain quite a number of substances soluble in neutral organic solvents 
 of low boiling points. Some, indeed, contain substances (inorganic salts, 1 
 dextrine) soluble in water, but these cases are rare. It may therefore 
 occasionally be of importance to boil a small quantity of the sample with 
 water, determine the reaction of the solution, and ascertain whether, after 
 having been filtered, it leaves a residue which can be further examined. 
 If the water shows an acid reaction, this is generally due to the sample 
 having been vulcanised with sulphur monochloride. Very frequent also is 
 the presence of sodium chloride in the aqueous extract, as this is a very 
 frequent constituent of ' white substitute,' in the manufacture of which 
 sodium bicarbonate has been employed. 
 
 The treatment of india rubber with organic solvent requires some 
 consideration. Carbon bisulphide and benzene are excellent solvents for 
 free fatty and mineral oils, solid hydrocarbons, and similar substances. 
 At the same time, they are, however, also solvents for india rubber and its 
 congeners, and although vulcanised india rubber is practically insoluble in V 
 all solvents, this is, strictly speaking, only true in those cases where the ^ 
 india rubber is fully vulcanised. Even more serious is the fact that many 
 of the ' brown substitutes ' yield to these solvents considerably more than 
 the free fatty oils they contain. This is, of course, a serious objection, as 
 their employment would result in the appearance of these bodies in two 
 different analytical groups. 
 
 The alcoholic solvents are free from these objections, but they are very 
 bad solvents for the hydrocarbon oils, although, on the other hand, they are 
 on the whole good solvents for the resins, or, at least, of those which are 
 used in the manufacture of india rubber goods. 
 
 Acetone is, however, the solvent which is free from any of the objec- 
 tions applying to the above named solvents. It will, particularly at higher 
 temperatures, easily dissolve all the oily and resinous constituents of india 
 rubber, gutta percha, and Balata without dissolving a trace of any of these 
 last named bodies themselves. It also readily dissolves free fatty oils, 
 
 1 I have found goods containing as much as 6 per cent, sodium carbonate, though 
 what the merit of this addition is supposed to be I have never been able to ascertain. 
 
240 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 mineral oils, solid hydrocarbons, resins, and the free sulphur. For this 
 reason I have made use of this solvent for years, and always found it to 
 answer most satisfactorily. 
 
 In carrying out the extraction of an india rubber sample with acetone, 
 from 1*5 to 2 grnis. of the sample are placed in one of Schleicher and 
 Schiill's extraction thimbles, which is inserted in the extractor. The type 
 of extractor to be used is a matter of some importance, as upon this depends 
 to a considerable degree the rapidity with which the extraction may be 
 completed. The most generally used form of extraction apparatus is the 
 one devised by Soxhlet, and as long as its capacity does not exceed from 
 
 \Torefbixamdenser 
 
 FIG. 23. Apparatus for the extraction of india rubber samples. 
 
 50 to 60 c.c., it answers fairly well, but for our purpose I very much 
 prefer a form of extractor in which an inner tube, which receives the 
 thimble, fits into an outer jacket, so that there remains a free annular 
 space between them through which the vapour of the extracting liquid 
 may pass up into the reflux condenser, thus keeping the material to be 
 extracted always at the temperature of the boiling liquid. Care need 
 only be taken to arrange the tube leading from the reflux condenser to 
 the inner extractor tube in such a way that the condensed solvent falls 
 straight into the thimble containing the material to be extracted. In this 
 apparatus even the most refractory sample can be extracted in less than 
 five hours, whereas about double that time is required if a Soxhlet appar- 
 atus be used. 
 
 The whole of the apparatus is shown in working order at fig. 23. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 241 
 
 The separation effected by this extraction in reference to our list of 
 constituents on page 236 is as follows : 
 
 A. Extract. 
 
 B. -Residue. 
 
 I. Oily and Resinous Constituents of 
 India Rubber. 
 
 II. Free Oils from Substitutes. 
 Free Fatty Oils. 
 Mineral and Resin Oils. 
 Solid Hydrocarbons. 
 Resins. 
 Organic Bodies not named above (?). 
 
 III. Free Sulphur. 
 
 I. India Rubber, Gutta Percha, Balata. 
 
 II. Chlorosulphide Substitutes. 
 Sulphide Substitutes. 
 Oxidised or ' Blown ' Oils. 
 Pitches, Tar. 
 Bituminous Bodies. 
 Carbohydrates. 
 Organic Bodies not named above (?). 
 
 III. Sulphur of Vulcanisation. 
 Sulphur in Organic Constituents 
 
 other than Rubber. 
 Sulphur in Inorganic Constituents. 
 
 IV. Chlorine in India Rubber. 
 Chlorine in Fatty Substitutes. 
 
 V. All Inorganic Constituents. 
 
 It will at once be seen that the complete analysis of this extract may 
 be a very difficult matter indeed, it may safely be said that our present 
 day methods would be utterly inadequate for the analytical disentangling 
 of such a mixture, this being particularly so in view of the fact that the 
 quantities of substance on which this would have to be done rarely exceed 
 10 per cent, of the rubber sample. 
 
 There is, however, scarcely ever any need for a complete analysis of 
 this extract, an investigation of its general characteristics yielding sufficient 
 indication respecting its nature. 
 
 If the extract forms a more or less viscous liquid, it may be taken for 
 certain that it does not contain any solid hydrocarbons or resins, both of 
 which, even in small quantities, deprive fatty and mineral oils of their 
 fluidity. 
 
 If, on warming the extract with about 5 c.c. of absolute alcohol, there 
 remains a liquid residue, this consists of mineral oils. If the warm alcohol 
 on cooling deposits voluminous flakes, the presence of solid hydrocarbons 
 (paraffin wax, ceresin, ozokerite) is a practical certainty. 
 
 A rough estimation of the fatty oils present may be effected by extract- 
 ing at least 10 grins, of the sample, saponifying the extract with 25 c.c. of 
 standard alcoholic caustic potash, and titrating back with standard hydro- 
 chloric acid. 1 As the fatty oils present, in nine out of ten cases, consist of 
 colza oil, we may assume that every 172 mgrms. of caustic potash which has 
 disappeared in the saponification represents 1 grm. of the fatty oil present. 
 
 In the presence of a composite extract, the determination of the amount 
 of extract derived from the india rubber is an absolute impossibility, largely 
 owing to the very indifferent character of the soluble matter contained in india 
 1 This is practically the KOETTSTOHFER test. 
 
 16 
 
242 INDIA RUBBER AND ITS ANALYSIS. 
 
 rubber. A very little experience will suffice to enable the worker to judge 
 from the appearance of an extract as to whether it is derived from india 
 rubber solely, or whether it is contaminated with one of the other sub- 
 stances named in schedule A. 
 
 Of course the acetone extract also contains the whole of the free sulphur 
 present, the amount of which in the various india rubber articles may vary 
 from nothing to 20 per cent, and more. In rubber analyses pretending to 
 the highest attainable degree of accuracy, the weight of the free sulphur 
 must, of course, be deducted from the weight of the total acetone extract. 
 If the free sulphur separates out in the extract in the form of well-shaped 
 crystals, this is an unmistakable sign of the absence of oily or resinous 
 ingredients other than those derived from india rubber, or of solid hydro- 
 carbons, in the extract. In this case the acetone is distilled off, the flask 
 dried in a water oven, and the weight of the total dry extract ascertained. 
 The dry extract is then treated five times in succession with increments of 
 2 c.c. of acetone, and these washings are filtered through a small filter into 
 a weighed flask. As this filter may contain a few crystals of sulphur, it is 
 placed in the original flask and washed with 5 c.c. of carbon bisulphide. 
 After the solvents have been driven off on the water bath, the two flasks 
 are dried, and their contents of organic extract and sulphur ascertained 
 respectively by weighing. 
 
 This method involves an error due to the solubility of sulphur in 
 acetone; This error is represented by from 5 to 6 mgrms. of sulphur dis- 
 solving in the 10 c.c. of acetone used for the above separation, and it is 
 therefore necessary to increase the amount of sulphur, and to reduce the 
 organic extract by this amount in the subsequent calculations. 
 
 If the extract is of a highly composite nature, and especially if solid 
 hydrocarbons are present, the above method for the estimation of the free 
 sulphur cannot be employed. If, nevertheless, its direct estimation be 
 deemed important, it may be effected by oxidising the whole of the acetone 
 extract with fuming nitric acid on the water bath, diluting with water, 
 filtering and washing, and finally precipitating the sulphuric acid formed 
 as barium sulphate. 
 
 For the detection and determination of tar, pitches, and bituminous 
 bodies (asphaltum) in india rubber, HENRIQUBS proposed to treat the 
 sample in the cold on a filter with nitrobenzene. If freshly distilled, and 
 therefore very pale-coloured, nitrobenzene is used, this method allows us to 
 detect the presence of the above bodies by the dark colour of the nitro- 
 benzene passing through the filter. In the presence of tar or pitches the 
 colour of the filtrate is a brownish-black ; asphaltum dissolves with a 
 characteristic rich mahogany colour. The direct gravimetric estimation of 
 the substances present in the extract is, however, impossible owing to the 
 very high boiling point of nitrobenzene. 
 
 HBNRIQUBS' process consists in treating 1 grm. of the sample with 30 c.c. 
 of nitrobenzene at the ordinary temperature in a small beaker for about 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 243 
 
 one hour. The mass is then transferred to a filter, and the solvent squeezed 
 out as much as possible by pressing it with a small pestle. It is then 
 washed with another 30 c.c. of nitrobenzene. The residue is then washed 
 with water into a porcelain basin and boiled with water until all the nitro- 
 benzene has disappeared. The washed mass is then dried and weighed, the 
 loss consisting of the above named bodies. 
 
 According to HENRIQUES, tar, pitch, and asphaltum are completely 
 soluble in nitrobenzene. As a general statement, this is certainly not 
 correct. Very frequently these bodies contain a not inconsiderable amount 
 of insoluble matter which is not extracted in the above treatment. This 
 is not of much account, as the amount of these substances used in rubber 
 mixings is, as far as my experience goes, always very small, scarcely ever 
 exceeding 2 per cent. Samples like those quoted by HEN&IQUES, containing 
 respectively 8, 21, and 14 per cent, of asphaltum, I have never met with, 
 and I doubt very much whether they represent bond fide manufactures. 
 
 I look, nevertheless, upon the direct estimation of these bodies as a 
 matter of some importance, as, in the first instance, their employment in a 
 'mixing' must always be regarded as a matter of not inconsiderable 
 technical significance. Moreover, as all asphaltum contains from 3 to 10 
 per cent, of combined sulphur, the estimation of the latter can rarely be 
 neglected, and this estimation is only possible if we get the actual extract, 
 free from its solvent, to deal with. 
 
 This I carry out by following up the extraction with acetone by an 
 extraction with pyridine without any intervening drying of the paper cone 
 and its contents. After removing from the extractor the flask containing 
 the acetone extract, I therefore at once connect the extractor with a 
 weighed flask containing about 60 c.c. of pyridine (B.P. 109 to 111 C.). 1 
 This flask I heat in an oil bath, the temperature of which is kept between 
 116 and 120 C. In the presence of tar, pitch, or asphaltum, the paper at 
 once assumes a more or less pronounced brown colour, and I continue 
 extraction until it has assumed its normal tint. I then distil the pyridine 
 off on an oil bath, the temperature of which is again kept at from 
 116 to 120 C. The employment of a sand bath is inadvisable on account 
 of the risk of overheating towards the end of the distillation. 
 
 On removing the flask from the oil bath, it is carefully cleaned with 
 filter paper, and finally with filter paper moistened with benzene. The 
 flask is then heated in a water oven until the smell of pyridine has dis- 
 appeared. Subsequently the weight of the extract is ascertained. 
 
 Whether the extract consists of tar, pitch, or asphaltum may generally 
 be decided by an examination of the consistency and colour of the extract, 
 tar being a black viscid liquid, pitch a ductile black mass, and asphaltum 
 a brittle, very dark brownish-maroon substance. A mixture of any two of 
 these substances would be extremely difficult to detect. But this is 
 scarcely of any account, as the employment of such a mixture would be 
 1 Pure pyridine boils at 116 C. 
 
244 INDIA RUBBER AND ITS ANALYSIS. 
 
 due to the individual fancy of the manufacturer rather than to any 
 particular virtue of the blend. 
 
 The addition of all these substances to rubber mixings only serves the 
 purpose of softening the india rubber without subjecting it to an undue 
 and detrimental amount of mechanical working. This softening may be 
 desirable, either for the purpose of enabling the rubber to take up readily 
 large quantities of mineral matter, or to assist in any subsequent calender- 
 ing of the rubber mixing. 
 
 The separation we have effected by this pyridine extraction is shown in 
 the following schedules : 
 
 C. Extract. 
 
 D, Residue. 
 
 II. Tar. 
 Pitch. 
 
 Bituminous Bodies. 
 
 III. Sulphur in Extract. 
 
 I. India Rubber, Gutta Percha, Balata. 
 
 II. Chlorosulphide Substitutes. 
 Sulphide Substitutes. 
 Oxidised or ' Blown ' Oils. 
 Carbohydrates. 
 
 III. Sulphur of Vulcanisation. 
 Sulphur in Organic Constituents 
 
 other than Rubber. 
 Sulphur in Inorganic Constituents. 
 
 IV. Chlorine in India Rubber. 
 Chlorine in Fatty Substitutes. 
 
 V. Inorganic Constituents. 
 
 The estimation of the sulphur in all the various forms in which it may 
 appear in manufactured india rubber is one of the most important points 
 in the complete analysis of a rubber article. We must, therefore, also test 
 for sulphur in the pyridine extract (C), and if any be present, determine 
 its amount quantitatively. The sulphur in this case cannot be present in 
 the free state, but only in organic combination, for it cannot be 
 estimated by the method adopted for its determination in the acetone 
 extract, owing to the frequent conversion of these organic sulphides, 
 partially at least, into sulpho-acids, the barium salts of which are soluble 
 in dilute hydrochloric acid. The following procedure is therefore 
 adopted : 
 
 The pyridine extract is oxidised on the water bath with fuming nitric 
 acid in the flask in which it was weighed. As soon as complete solution 
 is obtained, it is transferred to a porcelain incinerating dish of about 6 cm. 
 diameter. The flask is then washed with hot fuming nitric acid, the 
 washings being added to the contents of the dish. The whole is now 
 concentrated to a syrupy consistency, about half a grm. of pure sodium 
 acetate is added, and evaporation continued to dryness. The dry residue 
 is ignited over a Bunsen burner, the carbonaceous mass completely oxidised 
 by slowly adding the requisite quantity of sodium nitrate, and the mass 
 allowed to cool. It is then dissolved in hot water, the solution filtered, 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 245 
 
 and the filter washed neutral with water. After acidulating the filtrate 
 with hydrochloric acid, the sulphuric acid is precipitated, and its amount 
 ascertained in the usual manner. 
 
 We proceed now to determine in the residue from the pyridine ex- 
 traction (D) the amount of fatty substitutes present. This is carried out 
 by the very simple and satisfactory method due to HEXRIQUES, and which 
 is based upon the fact that all these substitutes are still essentially 
 triglycerides, and, as such, saponifiable with caustic soda or potash. 
 
 Before applying this saponification test, the residue D is freed from all 
 adhering pyridine by washing it in the extraction thimble with acetone. 
 It is then dried, weighed, completely transferred to a wide-necked flask, 
 and after addition of 25 c.c. of alcoholic caustic potash or soda, heated 
 upon the water bath on a reflux condenser for from three to four hours. 
 In important cases it is advisable to repeat this treatment with a fresh lot 
 of caustic potash. All the alcoholic liquors are carefully collected, the 
 insoluble residue washed upon a filter with boiling water, and completely 
 freed from alkali by washing with water. All these washings are united 
 with the alcoholic saponification liquors. 
 
 The filter residue is dried in a water oven in a current of coal gas. 
 This is best carried out by placing the residue in a weighed wide-necked 
 flask provided with a doubly perforated cork stopper. Through these 
 perforations pass two short pieces of glass tubing. The one of these is 
 connected to the gas supply, the other to the Bunsen burner, heating the 
 bath by means of india rubber tubing which is passed into the oven 
 through its aperture at the top. The residue is thus dried in a current of 
 coal gas to avoid oxidation, which, without this precaution, might affect 
 the final result to a very considerable extent. The dry residue is finally 
 weighed in the flask, and the loss of weight very accurately represents the 
 weight of the fatty substitutes removed by saponification. The percentage 
 represented by this amount must, of course, just as all the previously 
 ascertained data, be calculated upon the original weight of the sample 
 taken, and not upon the weight of the residue from the pyridine 
 extraction. 
 
 According to HENRIQUES, the above described method for the determina- 
 tion of the fatty substitutes is liable to be vitiated by the fact that the 
 india rubber retains as much as 6 per cent, of its weight of caustic potash 
 in an insoluble condition, and " he states that this alkali can only be 
 removed by boiling the india rubber with dilute hydrochloric acid. Of 
 course there is no objection to this treatment in the case of india rubber 
 free from mineral matter, or of india rubber free from mineral matter 
 soluble in hydrochloric acid. But, as a rule, the greater part of the 
 mineral matter used in rubber mixings is soluble in hydrochloric acid ; 
 others are partially soluble, undergoing decomposition under the influence 
 of the acid. It will readily be perceived that in such cases this treatment 
 would very considerably complicate the determination of the mineral 
 
246 INDIA RUBBER AND ITS ANALYSIS. 
 
 constituents of the rubber articles, all the more so as the complete removal 
 from india rubber even of those inorganic bodies which are easily soluble 
 in hydrochloric acid is an extremely tedious operation. I have found, 
 however, that the absorption of alkali by the india rubber is not what 
 would appear to be a somewhat obscure function of the india rubber itself, 
 but is entirely due to the presence of resinous bodies in the latter. If these 
 are removed before the application of the above described saponification 
 process, the absorption of alkali by the india rubber is too insignificant 
 as to call either for HENRIQUES' hydrochloric acid treatment, or for any 
 other correction. But this is exactly what our first operation, the 
 extraction with acetone, accomplishes. 
 
 The fatty substitutes extracted are either chlorosulphides, sulphides, 
 or oxides of triglycerides, and it is generally of importance to ascertain 
 which of these is present. Not infrequently two of these substitutes are 
 used in rubber mixings, but which of them can be discovered fairly 
 approximately only by a quantitative estimation of the amount of chlorine 
 and sulphur present in the saponification liquors ; and as these figures are 
 also of importance in respect to the question of the vulcanisation of the 
 rubber sample under examination, their determination becomes almost 
 invariably necessary. 
 
 This I carry out by evaporating the whole of the washings which have 
 been collected from the saponification process, in a porcelain basin, to a 
 volume of about 50 c.c. This is then made up to 100 c.c. with water in a 
 graduated cylinder, the water which has to be added being used to wash 
 the basin. I then take 50 c.c. of this solution, evaporate it to dryness in 
 an incinerating dish, ignite over a Bunsen burner, oxidise the carbonaceous 
 matter by the careful addition of sodium nitrate, dissolve the subsequently 
 resulting mass in water, acidulate with nitric acid, and add silver 
 nitrate. Any precipitate of silver chloride thus obtained is filtered off, 
 treated in the usual manner, and weighed, The amount of chlorine thus 
 found must, of course, be doubled. 
 
 The other 50 c.c. of the above solution are first oxidised on the water 
 bath with bromine, or bromine water, in order to oxidise the alkaline sul- 
 phides present, and are then concentrated by evaporation to the state of a 
 syrupy liquid, which is oxidised with fuming nitric acid, evaporated to 
 dryness and ignited, finally also being fused with sodium nitrate. The 
 resulting mass is dissolved in water, and the sulphuric acid present deter- 
 mined in the usual manner. It is obvious that the amount of sulphur thus 
 ascertained must also be doubled. 
 
 If there are chlorine and sulphur present, it will generally be found 
 that their respective amounts appear approximately in the ratio of their 
 molecular weights. In this case the fatty substitute present is either the 
 chlorosulphide of one of the unsaturated fatty oils (colza oil) i.e., ' white 
 substitute' or a mixture of the latter with one of the 'blown' oil 
 substitutes. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 247 
 
 If there are both chlorine and sulphur present, but the latter consider- 
 ably in excess of the above stated ratio, this is evidence that the fatty sub- 
 stitutes employed are a mixture of * white ' and ' brown ' substitute, or 
 perhaps even of these two with a ' blown ' oil substitute. 
 
 The presence of sulphur only indicates the presence of either * brown ' 
 substitute alone, or a mixture of it with ' blown ' oil substitute. 
 
 The absence of both chlorine and sulphur proves the substitute present 
 to consist entirely of ' blown ' oil substitute. 
 
 Further information on this point will be found in the next chapter, 
 but attention may already here be drawn to the fact that if * white ' sub- 
 stitute, as is frequently the case, is employed in the manufacture of hot 
 vulcanised goods, it loses a more or less considerable proportion of its chlorine 
 in the form of hydrochloric acid. HENRIQUES assumed that in this case 
 the hydrochloric acid escapes altogether during the process of vulcanising. 
 This assumption is, however, entirely erroneous. The hydrochloric acid 
 eliminated from the substitute is retained entirely by the india rubber, 
 with which it combines to form polyprene hydrochloride, and it is also 
 probable that some part of it, in the presence notably of litharge or 
 calcium carbonate, combines with these. But as the de-chlorination of the 
 4 white ' substitute under the conditions mentioned is far from complete, 
 the presence of chlorine in the substitute cannot escape detection. 
 
 The saponification of the residue D with caustic alcoholic potash has 
 therefore resulted in the following further separation : 
 
 E. Extract. 
 
 II. Chlorosulphide Substitutes. 
 Sulphide Substitutes. 
 Oxidised or ' Blown' Oils. 
 
 III. Sulphur in Fatty Substitutes. 
 Chlorine in Fatty Substitutes. 
 
 F. Residue. 
 
 I. India Rubber, Gutta Percha, Balata. 
 II. Carbohydrates. 
 
 III. Sulphur of Vulcanisation. 
 Sulphur in Inorganic Constituents. 
 
 IV. Chlorine in India Rubber. 
 V. Inorganic Constituents. 
 
 For the separation of the india rubber from the mineral matter, HEN- 
 RIQUES proposed to employ the solvent action of boiling petroleum, or, 
 rather, of those fractions of petroleum spirit boiling above 150 C. This 
 is, however, an extremely unsatisfactory process, as the solvent named will 
 attack very lightly vulcanised india rubber only. I subsequently suggested, 
 and until quite recently practised, the use of boiling nitrobenzene, which 
 is a far more powerful solvent, and yields solutions which are rather easier 
 to deal with than those obtained by the aid of petroleum. But even this 
 solvent is not capable of resolving satisfactorily samples possessing a vul- 
 canisation coefficient exceeding 4 per cent, of sulphur, and the number of 
 
248 INDIA RUBBER AND ITS ANALYSIS. 
 
 such being rather large, the limitations of this process were often very 
 restrictive. 
 
 There remained, then, only the determination of the mineral matter by 
 the most obvious process namely, incineration. But to this process 
 attaches the very serious objection that it interferes with the composition 
 of the inorganic substances present in such a manner as to render al- 
 together impossible the determination of the exact nature of the mineral 
 substances added to the rubber mixing. Any mercury compounds volatilise 
 altogether ; sulphides are converted into oxides or sulphates ; carbonates 
 lose their carbonic acid ; oxides are converted into sulphides, which may then 
 undergo further oxidation; carbon pigments, carbohydrates, animal and 
 vegetable fibres are burnt off, and thus often unavoidably accounted for 
 as * india rubber.' As a rapid control for works' purposes, the incineration 
 test answers in the majority of cases well enough, but in the analysis of 
 samples of unknown composition it is entirely unsatisfactory. 
 
 Quite recently I have succeeded in finding a far more effective resolvent 
 for even highly vulcanised india rubber than nitrobenzene. This substance 
 is a-nitro-naphthalene. The readiness with which this body resolves india 
 rubber possessing a vulcanisation coefficient of from 2 to 8 per cent, is 
 perfectly surprising ; even samples showing vulcanisation coefficients up to 
 15 per cent, are quickly and completely decomposed at a temperature of 
 from 170 to 180 C. The solutions obtained in the nitrobenzene treatment 
 are often extremely difficult to deal with, the mineral matter absolutely 
 refusing to settle out, 1 and forming often practically unfilterable solutions. 
 On the other hand, the solutions obtained in the nitro-naphthalene treatment, 
 after dilution with benzene, and after a few hours' standing, filter very 
 readily, and yield clear filtrates. 
 
 The only peculiarity the nitro-naphthalene treatment presents is one it 
 shares with the nitrobenzene treatment viz., that if it is applied to 
 substances containing either red lead or ferric oxide, these bodies at 
 a temperature exceeding 180 C. gradually oxidise the india rubber, con- 
 verting it into black carbonaceous masses, which might afterwards be taken 
 to represent carbon black present in the mixing. This difficulty may be 
 avoided in the case of the nitrobenzene treatment, by adding to the nitro- 
 benzene T ^th or Y^th volume of chloroform, whereby the boiling point is 
 reduced from 208 C. to about 170 C. A mixture of 100 parts of nitro- 
 naphthalene and 5 parts of chloroform also appears to act even better than 
 nitro-naphthalene alone, but there is, of course, only an insignificant lowering 
 of the boiling point in this case, and I prefer, therefore, to carry out the 
 nitro-naphthalene treatment in an air bath which can without difficulty be 
 kept at the safe temperature of 180 C. 
 
 The treatment named, applied to residue F, for the separation of the 
 india rubber from the mineral matter is therefore carried out as follows : 
 The residue F, the weight of which is known, is placed in a weighed wide- 
 1 This is especially the case if the mineral matter consists of Sb.,S s , or of ZuO. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 249 
 
 necked flask together with from 60 to 70 grms. of a-nitro-naphthalene, 
 M.P. 58 C. The flask is then placed in an air bath, almost completely 
 enclosing it. The air bath carries a thermometer, and care should be taken 
 that the temperature is kept as near 180 C. as possible. The form and 
 construction of this air bath is represented in figs. 24 and 25. 
 
 The cylindrical body (a) of the air bath is made of stout copper. It is 
 provided with a covering of stout asbestos paper, or thin asbestos sheet. 
 The bottom (b) of the bath should iiot be soldered to the body, but secured 
 on the flanges c by means of screws, joints being made with asbestos. 
 
 i=tc 
 
 FIG. 24. Diagram of air bath for 
 nitro-naphthalene extraction. 
 
 FIG. 25. Diagram of air bath, construction 
 of lid. 
 
 This avoids much trouble in repairs, as the bottom of the bath has to be 
 frequently renewed. On the top the body has a collar d provided, over 
 which fits the cap e. The latter is slotted in the manner shown in fig. 25, 
 just wide enough to allow the neck of the flasks to pass. The flask being 
 passed in the slot, it is secured in its position by the slide, and the grooves g 
 in the edges of the two semicircular sections of the slot and the slide. 
 Asbestos cording is fixed so as to secure a tight joint round the neck of 
 the flask. The opening h is for the thermometer. 1 This is an extremely 
 serviceable air bath. The desired temperature is reached very quickly, and 
 as easily regulated. It may be mounted on tripod legs, or simply placed 
 on a round tripod of suitable dimensions. To prevent too direct a radiation 
 
 1 This air bath maybe obtained from Messrs. Fred. Jackson & Co., 11 Half Moon 
 Street, Manchester. 
 
250 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 of heat from the bottom of the bath to the flask, I cover the former on the 
 inside of the bath with a thin layer of fluffy asbestos fibre. 
 
 During treatment the contents of the flask are at frequent intervals 
 gently stirred with a small glass rod. The completion of the solution 
 requires from fifteen to sixty minutes according to the description of the 
 sample. If the solution is completed, the flask is taken from the bath, 
 allowed to cool, diluted with about 150 c.c. of benzene, and left to stand for 
 the mineral matter to settle out. The solution is then filtered, flask and 
 filter washed thoroughly with warm benzene, and subsequently dried for one 
 hour at 110 C. We obtain in this manner the weight of the mineral 
 matter, pigment colours, carbohydrates, and organic fibrous materials which 
 may be present in the sample. The loss of weight is put down as india 
 rubber, including sulphur of vulcanisation and chlorine in india rubber 
 (chlorine of vulcanisation). The separation effected is therefore as follows : 
 
 G. Solution. 
 
 I. India Rubber, Gutta Percha, Balata. 
 
 Dissolved but not determined. 
 
 III. Sulphur of Vulcanisation. 
 
 IV. Chlorine in India Rubber. 
 
 H. Residue. 
 
 II. Carbohydrates. 
 III. Sulphur in Inorganic Constituents. 
 
 V. Inorganic Constituents. 
 Fibrous Materials. 
 
 The india rubber cannot be recovered from its solution in the above 
 filtrate, for the simple reason that it is no longer there in the form of india 
 rubber, but of decomposition products of the latter. For this reason it is 
 also impracticable to determine the sulphur of vulcanisation and the 
 chlorine in the india rubber by means of this solution. 
 
 It is also impossible, therefore, to detect admixture of the rubber with 
 gutta percha or Balata as soon as the india rubber is at all fully vulcanised. 
 In only slightly vulcanised articles it is possible to extract the gutta percha 
 and Balata by repeated treatment of the original sample with cold benzene, 
 and precipitation of the benzene extract with absolute alcohol. It is, how- 
 ever, very rare that a sample of manufactured india rubber contains more 
 than 1 or 2 per cent, of either gutta percha or Balata, and there is amongst 
 india rubber technologists very considerable divergence of opinion respect- 
 ing the practical value of these admixtures. 
 
 Much more serious is the fact that, as yet, we are practically unable to 
 detect, with any degree of certainty, the admixture of 'recovered' india 
 rubber; still less, of course, can we estimate its amount. Nor do the 
 physical tests in particular, the much vaunted tensile strength test in the 
 least assist us to arrive at any conclusion in this respect. This is all the 
 more regrettable as ' recovered ' india rubber is now very commonly used 
 in goods in which its employment must appear in the highest degree in- 
 judicious. This applies notably to bicycle and carriage tires. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 251 
 
 In an earlier chapter (III.) I have shown that all recovered India 
 rubber, in spite of its trade description as ' devulcanised ' india rubber, 
 still contains the whole of the sulphur of vulcanisation of the waste rubber 
 from which it has been obtained. I have further shown that all this 
 recovered rubber again takes part in the vulcanisation of the article 
 manufactured with its aid. Indeed, this is even true of the ground rubber 
 waste (rubber dust), which is also largely used in the manufacture of rubber 
 goods. This must not infrequently lead to an abnormally high coefficient 
 of vulcanisation, 1 which may therefore occasionally serve as a criterion as 
 to the proportion of recovered rubber present. 
 
 From the above it will be seen that there lies still a large unexplored 
 field before us, and a great deal of patient research and experiment will 
 be required before we shall be able to differentiate analytically the con- 
 stituents of the rubber portion of a sample of manufactured india rubber. 
 
 The filter with the residue h upon it is replaced in a funnel, and 
 washed with water heated to boiling in the flask in which the nitrobenzene 
 treatment was carried out. Washing the filter with 150 to 200 c.c. of 
 boiling water, added in small quantities at a time, is quite sufficient to 
 remove from the residue all dextrine or soluble starch present. The filtrate 
 is tested for the presence of dextrine or soluble starch by the iodine test. 
 
 If either of these two substances has been detected, filter and flask are 
 again dried as before, and the loss of weight noted. This loss may repre- 
 sent either dextrine, or farina, or other starchy matter, which in the nitro- 
 benzene treatment is partially converted into soluble starch, but chiefly 
 into dextrine. To decide as to which form the carbohydrate is present in 
 the rubber, some of the Original sample is boiled with water for a few 
 minutes. The water is then filtered off, and the filtrate tested for dextrine 
 as above. According to the result of this test, the carbohydrate observed 
 is put down either as dextrine or as farina. 
 
 The result of this determination leaves our analysis as follows : 
 
 I. Solution. 
 
 K. Residue. 
 
 II. Carbohydrates. 
 
 V. Inorganic Constituents (total). 
 
 III. Sulphur in Inorganic Constituents. 
 
 IV. Inorganic Constituents (detail). 
 
 Not determined in solution G. 
 
 III. Sulphur of Vulcanisation. 
 
 IV. Chlorine in India Rubber. 
 
 The items which have still to be determined are those appearing in the 
 right column of the above table. 
 
 Before proceeding with the residue K, it should be closely examined for 
 the presence of vegetable or animal fibres. If any are present, and the 
 1 Respecting the significance of this term, the chapter following should be consulted. 
 
252 INDIA RUBBER AND ITS ANALYSIS. 
 
 determination of their amount is of importance, a fairly approximate separa- 
 tion from the mineral matter may be effected "by shaking a weighed 
 quantity of the residue with chloroform. This operation is best carried 
 out by placing the residue in a test tube, pouring about 20 c.c. of chloro- 
 form upon it, very slightly shaking the tube, and allowing it to stand for 
 some time, during which it is repeatedly tapped. The mineral matter 
 settles at the bottom, while the vegetable and animal fibres rise to the top. 
 They are floated off into a dish by holding the test tube in an inclined position 
 and overflowing it with chloroform which is run in from a pipette, the out- 
 flow end of which is all the time kept in the chloroform in the test tube. 
 The chloroform is evaporated from the dish, and the fibres remaining behind 
 are transferred by the aid of a camel-hair brush to a watch-glass and 
 weighed. From the weight of fibre thus found, the total amount of fibres 
 present in residue K may be readily calculated. 
 
 If the residue K contains any carbon black, the presence of which is 
 always easily recognisable, the only way to estimate it with a fair degree 
 of approximation consists in burning off part of the residue K with lead 
 chromate in the manner adopted for the elementary analysis of organic 
 bodies containing sulphur. 1 
 
 The detailed qualitative examination, followed by the quantitative 
 estimation of the inorganic constituents of the mineral residue K, is, of 
 course, conducted on the well-known lines of qualitative inorganic analysis, 
 a description of which in this place is unnecessary. The variety of 
 inorganic constituents to be found in manufactured india rubber is at times 
 astonishing, often defying explanation. Nevertheless will it be found that 
 in the large majority of cases a comparatively small number of inorganic 
 compounds are used. A list of those compounds I have observed in a very 
 large variety of rubber goods I give below. They are arranged approxi- 
 mately in the order of their technical importance and the frequency with 
 which they occur. 
 
 Litharge, PbO. 
 
 Paris White, Whiting, CaC0 3 . 
 
 Zinc White, ZnO. 
 
 Lime, CaO. 
 
 Barytes, BaS0 4 . 
 
 Lithopone, ZnS,BaS0 4 . 
 
 Golden Sulphide, Sb 2 S 5 . 
 
 Lamp Black, C (chiefly). 
 
 Magnesia, MgO. 
 
 Red Lead, Pb 3 4 . 
 
 Red Iron Oxide, 
 
 French Chalk. 
 
 China Clay. 
 
 Vermilion, HgS. 
 
 Infusorial Earth, Kieselgulir, Si0 2 . 
 Zinc Sulphide, ZnS. 
 Lead Sulphide, PbS. 
 Lead Thiosulphate, PbS 2 3 . 
 Lead Sulphate, PbS0 4 . 
 Graphite, C (chiefly). 
 Magnesium Carbonate, MgC0 3 . 
 Pumice Stone Powder. 
 Glass Powder. 
 Mica. 
 
 - Arsenious Sulphide, As 2 S 3 . 
 Zinc Dust. 
 Aluminium Powder. 
 Iron Filings. 
 
 Plaster of Paris, CaS0 4 . Brass Filings. 
 
 Gypsum, Pearl White, CaS0 4 , 2 aq. Asbestos. 
 
 White Lead. 
 
 3 If the residue contains graphite (plumbago), this would, of course, also be found as 
 ' carbon ' in this process. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 253 
 
 It may safely be said that fully 80 per cent., probably even more, of 
 the manufactured india rubbers are produced with one or more of the 
 first twelve of the above named substances present, and these are also the 
 ones which are the easiest identified. In respect of some of the others, 
 identification in the purely chemical way is often extremely difficult 
 indeed, frequently the easiest and quickest way consists in a microscopic 
 examination. Gypsum, as distinguished from plaster of Paris, infusorial 
 earth, lead sulphate, pumice stone, glass, mica, and asbestos, can readily be 
 identified in this manner, and it is advisable to have a set of slides of all 
 these products, simply mounted in air, always ready for comparison. In 
 doubtful cases examination in polarised light often proves useful. 
 
 For the quantitative determination of the inorganic compounds, the 
 simplest way frequently is to dissolve a fresh part of the sample in nitric 
 acid, evaporate to dryness, add a few drops of strong sulphuric acid, drive 
 off the excess on the sand bath, and proceed with the residue in the well- 
 known manner. But, if it is preferred, some of the original sample may, 
 after extraction with acetone, be immediately subjected to the nitrobenzene 
 treatment, and the quantitative determination of the mineral constituents 
 completed by the analysis of the residue obtained. This course is 
 particularly to be recommended with samples yielding easily to the nitro- 
 benzene treatment. 
 
 The determination of the sulphur in the inorganic constituents is 
 carried out by placing a small incineration dish of about 6 cm. diameter 
 upon the water bath, charging it with about 10 c.c. of fuming nitric acid, 
 and adding a weighed quantity of the residue K in small portions. It is 
 advisable during this operation to keep the dish covered with a watch-glass 
 to avoid loss through spirting. After the whole of the substance has been 
 added, the watch-glass is thoroughly wiped with moistened filter-paper 
 cuttings, which are then also added to the acid. The whole is then 
 evaporated to a syrupy consistency, when about 2 grms. of pure sodium 
 acetate are added, and the whole evaporated to dryness. The dry mass is 
 carefully ignited over a Bunsen burner, the carbonaceous matter oxidised by 
 cautiously adding potassium nitrate, preferably in the form of small crystals. 
 The whole is then kept in quiet fusion for a short time, the dish allowed 
 to cool, the melt dissolved in boiling water, filtered, and the filter thoroughly 
 washed with hot water. The filtrate is then acidulated with hydrochloric 
 acid, heated to boiling, the sulphuric acid precipitated with a hot 
 solution of barium chloride, and the weight of the barium sulphate obtained 
 ascertained in the usual manner. The amount of sulphur thus found is, of 
 course, in the first instance calculated upon the total mineral matter 
 ascertained in the sample ; its percentage, however, being calculated upon 
 the whole of the rubber sample. 
 
 We now determine the total sulphur in the sample by weighing out 
 from 1*5 to 2 grms. of the original sample, and adding this quantity in 
 small portions to about 15 c.c. of nitric acid (about 64 per cent, strong, 
 
254 INDIA RUBBER AND ITS ANALYSIS. 
 
 not 'fuming') contained in an incincerating dish of the already named 
 description. The dish is placed on a water bath and covered with a watch- 
 glass, the whole operation being carried out exactly in the manner already 
 described for the determination of the sulphur in the mineral matter. 
 
 We now have determined the following forms of sulphur in the rubber 
 
 sample : 
 
 1. Free sulphur. 
 
 2. Sulphur in pitch, tar, and bituminous bodies. 
 
 3. Sulphur in fatty substitutes. 
 
 4. Sulphur in inorganic compounds. 
 
 5. Total sulphur in sample. 
 
 It is therefore obvious that by adding the amounts of sulphur of the 
 first four determinations, and deducting the sum from the total sulphur, 
 the difference of the latter over the former must represent the sulphur in 
 combination with the india rubber in other words, the sulphur of vul- 
 canisation. 
 
 It might, however, be objected, and not unjustly, that considering the 
 great importance of the sulphur of vulcanisation, its determination by 
 calculation from possibly five separate determinations, the errors of which 
 under unfavourable conditions would all accumulate in this final calcula- 
 tion, would tend to render the result of doubtful value. For this reason 
 it is much more satisfactory to treat from 1 to 1 - 5 grms. of the original 
 sample, according to its now ascertained components, with the above 
 mentioned extraction solvents, so as to reproduce a residue K, 1 the total 
 sulphur of which is ascertained exactly as the total sulphur in the sample. 
 By deducting from the figure thus obtained the sulphur contained in the 
 mineral matter, we obtain at once the sulphur of vulcanisation. 
 
 There remains now only the determination of the chlorine in the india 
 rubber, which is carried out as follows. About 0'3 grm. of the rubber 
 sample 2 are sealed up in a bomb tube with 5 c.c. of fuming nitric acid and 
 1 grm. of silver nitrate. 3 It is generally advisable to heat the tube for 
 two days in the ' bomb ' furnace, relieving it, of course, of its pressure by 
 unsealing the capillary, and re-sealing it, as soon as the gases have escaped, 
 prior to resuming the heating on the second day. When the cold tube is 
 opened and cut, its contents are washed with hot water into a small glass 
 beaker, and the precipitate filtered off. If the india rubber sample contains 
 no mineral matter insoluble in nitric acid, the washed filter residue consists 
 simply of silver chloride, the weight of which is ascertained in the usual 
 manner. But in most cases other insoluble matter is present. In this 
 case the filter residue is washed neutral and dried. It is then placed in a 
 porcelain crucible, the filter burned, and its ash added to the precipitate. 
 
 1 This to remove the free sulphur present, which may greatly retard the completion 
 of the nitrobenzene treatment. 
 
 ' 2 If the sample contains chlorosulphide substitutes, extraction with alcoholic caustic 
 potash must precede the above determination. 
 
 3 For particulars consult GATTERMANN, Practical Methods of Organic Chemistry. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 2 55 
 
 The whole is then carefully mixed with from 6 to 10 times of its weight of 
 sodium carbonate, and the mixture fused over a powerful Bunsen burner. The 
 fused mass is dissolved in water, filtered, and the filter residue thoroughly 
 washed. The latter is then washed off the filter into a porcelain dish, the 
 filter washed with about 50 c.c. of hot 5 per cent, strong nitric acid, which 
 is allowed to run from the filter into the basin. The contents of the latter 
 are then heated for about ten minutes to near boiling point, the solution 
 then filtered, and the filter residue carefully washed neutral. To the 
 filtrate, the total volume of which should not exceed 100 c.c., hydrochloric 
 acid is added in very slight excess, the precipitated silver chloride filtered 
 off and treated in the usual manner for weighing. We obtain in this 
 manner the amount of chlorine contained in the rubber of the sample. 
 
 The whole of the analytical data obtainable, and the successive pro- 
 cesses by which they are obtained, are embodied in the following table : 
 
 I. EXTRACTION WITH ACETONE. 
 
 
 
 VI. SEPARATE 
 ESTIMATIONS. 
 
 A. SOLUTION. 
 
 Resinous con- 
 stituents of 
 india rubber. 
 Fatty oils. 
 Mineral oils. 
 Resin oils. 
 Solid hydro- 
 carbons. 
 Resins. 
 Free sulphur. 
 
 B. RESIDUE. 
 
 II. EXTRACTION WITH 
 PYRIDINE. 
 
 Total sulphur. 
 Chlorine in 
 rubber. 
 
 C. EXTRACT. 
 
 Tar. 
 Pitch. 
 Bituminous 
 bodies. 
 Sulphur in 
 above. 
 
 D. RESIDUE. 
 
 III. EXTRACTION WITH 
 ALCOHOLIC POTASH. 
 
 E. EXTRACT. 
 
 Chlorosulphide 
 substitutes. 
 Sulphide sub- 
 stitutes. 
 Oxidised 
 (blown) oils. 
 Sulphur in 
 substitutes. 
 Chlorine in 
 substitutes. 
 
 F. RESIDUE. 
 
 IV. EXTRACTION WITH 
 NITRO-NAPHTHALKNE. 
 
 
 G. EXTRACT. H. RESIDUE. 
 
 India rubber, i V. EXTRACTION WITH 
 Sulphur in } BOILING WATER. 
 india rubber. 
 
 Chlorine in 
 india rubber, i -EXTRACT 
 The total of 
 above three starch 
 estimated by , (Farina), 
 loss. | Dextrine. 
 
 j 
 
 K. RESIDUE. 
 
 Mineral matter. 
 Free carbon, 
 fibrous 
 materials. 
 Sulphur in in- 
 organic com- 
 pounds. 
 
 Tiie number and nature of the analytical operations required for the 
 complete analysis of an india rubber sample may appear alarming to many, 
 but it may be at once pointed out that the number of individual deter- 
 minations recorded in the above scheme are probably never required in the 
 analysis of any one sample. Indeed, in many cases a complete analysis is 
 not required at all, one or two of the above operations furnishing all the 
 information required. It must therefore be decided from case to case 
 which are the analytical data required to furnish any particular informa- 
 tion asked for in respect of a certain sample. Not only is it, then, often 
 
256 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 possible to abbreviate the above course, but also to substitute more 
 expeditious, if somewhat less accurate, working methods in the place of 
 those we have been recommending. 1 
 
 The most usual shortening of the above scheme is rendered possible in 
 the analysis of samples, the mineral matter of which contains neither 
 sulphides, nor sulphates, nor any other sulphur compounds, and the 
 detailed composition of which is a matter of secondary importance. If, 
 moreover, the rubber sample is of the 'grey' or 'red' variety, the 
 analytical course may be further shortened, as none of these rubbers ever 
 contain any substances of the group C. of our above scheme. The mode 
 of procedure then assumes the following comparatively simple form : 
 
 I. EXTRACTION WITH ACETONE. 
 
 SEPARATE 
 ESTIMATIONS. 
 
 A. EXTRACT. 
 
 B. RESIDUE, 
 
 Mineral matter by 
 
 
 
 ignition. 
 
 Resinous constituents 
 
 III. EXTRACTION WITH ALCOHOLIC 
 
 India rubber by 
 
 of india rubber. 
 
 POTASH 
 
 difference. 
 
 Fatty oils. 
 
 
 
 Mineral oils. 
 
 
 
 
 Resin oils. 
 
 E. EXTRACT. 
 
 F. RESIDUE. 
 
 
 Solid hydrocarbons. 
 
 
 
 
 Resins. 
 Free sulphur. 
 
 Chlorosulphide sub- 
 stitutes. 
 
 S in rubber. 
 
 
 
 Sulphide substitutes. 
 Oxidised (blown) 
 
 (Sulphur of Vul- 
 canisation. ) 
 
 
 
 oils. 
 Sulphur in sub- 
 
 01 in rubber. 
 
 
 
 stitutes. 
 
 
 
 
 Chlorine in sub- 
 
 
 
 
 stitutes. 
 
 
 
 It will be seen that, in working according to this scheme, the extraction 
 with acetone and the treatment with alcoholic potash have remained, and 
 extracts A. and E. are therefore identical with the corresponding extracts 
 in the preceding table. Residue F. is, however, here used for the deter- 
 mination of the ' sulphur of vulcanisation ; and of the chlorine. 
 
 The estimation of the sulphur in residue F. is, of course, carried out in 
 exactly the same manner as the determination of the total sulphur. The 
 chlorine in this residue is determined by CARIUS' method, as already 
 described. A rather quicker, although not quite so accurate method con- 
 sists in mixing the rubber intimately with a fusion mixture consisting of 
 2 parts of sodium carbonate and 1 part of potassium nitrate, spreading this 
 mixture evenly on the bottom of a platinum dish which has previously 
 been covered with a thin layer of the same fusion mixture, and finally 
 covering the rubber particles completely with the same. The mass in the 
 dish is then very gently and gradually ignited. When the mass no longer 
 gives off vapours, the heat is increased until the whole is in quiet fusion 
 
 1 There is, of course, no necessity for a re- estimation of the respective extracts. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 257 
 
 and all carbonaceous matter burned off. The fused mass, after cooling, is 
 dissolved in water, filtered, the filter washed neutral, the filtrate acidulated 
 with nitric acid, and the chlorine precipitated with silver nitrate. This 
 operation should be performed in an ERLENMEYER flask, and not in a dish 
 or beaker, in order to avoid loss through the effervescence taking place on 
 acidulating the solution. 
 
 This same method has also been recommended for the determination 
 of the total sulphur in the india rubber, but it is quite worthless for this 
 purpose, the results obtained being much too low, often by over 50 per 
 cent, of their actual value. The cause of this is obvious, and may easily be 
 discovered by subjecting vulcanised india rubber to dry distillation, when 
 it will be found that almost exactly one half of the sulphur present is 
 given off in the shape of hydrogen sulphide, the rest in the form of organic 
 sulphides. The former, no doubt, is readily oxidised by the fusing 
 nitrate, but the greater part of the latter most probably are volatilised 
 before any action takes place. The chlorine of the india rubber, on the 
 other hand, is very readily given off at compamtively low temperatures, 
 and it is, moreover, evolved entirely as hydrochloric acid, which is rapidly 
 and completely absorbed by the sodium carbonate present. 
 
 In the above abbreviated scheme we determine the mineral matter by 
 ignition. As is well known, this method is by no means accurate indeed, 
 if not very carefully carried out, it may be very faulty. From 1 to 2 
 grms. of the india rubber are thinly spread out in a weighed porcelain 
 dish of about 6 cm. diameter. The latter is then heated over a small 
 flame, so regulated that the vapours evolved never burst into flame : should 
 this occur, it is a sign that the flame of the burner is too high. If the 
 evolution of vapours has ceased, the flame is moderately increased until 
 all the carbonaceous matter is burned off. The results are almost 
 invariably too high, owing to part of the organic sulphur being retained 
 either as sulphide in the presence of zinc oxide, or as sulphate in the 
 presence of lime, magnesia, or litharge. The error thus incurred may 
 amount to as much as 2 per cent., but the fact is that as far as the figures- 
 affected by it are concerned mineral matter and india rubber this fault is, 
 in the majority of cases, of little, if any, practical account. 
 
 (c) NOTES ON THE ANALYSIS OP HARD RUBBER GOODS (EBONITE). 
 
 The analysis of hard india rubber goods suffers from even greater dis- 
 advantages than that of soft goods, as the india rubber constituent, which, 
 in the former, only dissolves with difficulty, is, in the latter, totally in- 
 soluble. We have already incidentally alluded to the fact that the readiness 
 with which a sample of soft rubber is resolved in the nitrobenzene treat- 
 ment depends very materially upon the degree of vulcanisation i.e., 
 coefficient of vulcanisation of the sample, and decreases as the latter in- 
 creases. Long before we reach products of a coefficient of vulcanisation 
 as high as that of the typical hard rubbers, we arrive at products which 
 
 17 
 
258 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 are irresolvable by any known solvent. This fact adds, of course, very 
 considerably to the difficulty of the analysis of hard rubber goods. 
 Some compensation arises, however, from the fact that hard rubber never 
 contains any of the fatty substitutes. The organic non-rubber constituents 
 of hard rubber consist entirely of resins and pitches. The inorganic con- 
 stituents of hard rubber are much the same as those of soft rubber, but 
 in smaller variety. 
 
 The preparation of a sample for analysis is again a point of import- 
 ance. The only satisfactory way consists in reducing the sample, or as 
 large a part of it as possible, to a fairly fine powder by means of a file. 
 This powder is passed through a sieve of from 80 to 100 mesh per 
 square inch. What does not pass through the sieve is ground in a well- 
 adjusted grinding mill until the whole of the filings have passed the 
 sieve ; and after the fine powder thus obtained has been thoroughly mixed 
 in a mortar, it may be taken to represent a fair average sample. 
 
 In analysing hard rubber, an estimation of the following constituents is 
 desired : 
 
 (a) India rubber. 
 
 (6) Resins, or resinous matters. 
 
 (c) Bituminous bodies, pitches. 
 
 (d) Mineral matter. 
 
 f Sulphur of vulcanisation. 
 
 (e) Total sulphur, -| Free sulphur. 
 
 I Sulphur in mineral matter. 
 
 The analytical course to be followed may now be easily arrived at from 
 the scheme given on page 255. It is as follows : 
 
 I. EXTRACTION WITH ACETONE. 
 
 A. EXTRACT. 
 
 India rubber 
 
 resins. 
 Added resinous 
 
 bodies. 
 Free sulphur. 
 
 B. RESIDUE. 
 
 II. EXTRACTION WITH 
 EPICHLORHYDRINE. 
 
 C. EXTRACT. 
 
 All resins in- 
 soluble in 
 acetone. 
 
 Sulphur in 
 above. 
 
 D. RESIDUE. 
 
 III. EXTRACTION WITH 
 PYRIDINE. 
 
 E. EXTRACT. F. RESIDUE. 
 
 Bituminous bodies. 
 Pitches. 
 Sulphur in above. 
 
 Sulphur in 
 residue. 
 
 IV. SEPARATE 
 ESTIMATIONS. 
 
 Mineral matter. 
 Sulphur in mineral 
 matter. 
 
 The extraction with acetone requires much more time in the case of 
 hard rubber than with soft rubber goods indeed, nothing less than a two 
 days' extraction will be found satisfactory. This appears to be due to 
 the fact that hard rubber contains the free sulphur in a peculiarly in- 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 259 
 
 soluble form probably in an amorphous variety which only gradually 
 passes into solution. This seems to be the reason why the substitution 
 of carbon bisulphide l for acetone in this process proves no more expedi- 
 tious, as the solubility of amorphous sulphur very largely depends on the 
 temperature, and would therefore take place much less rapidly at the 
 boiling point of carbon bisulphide (46 C.) than at that of acetone (56 C.), 
 although the latter is a much inferior solvent for sulphur than the former. 
 
 The general conditions for carrying out the acetone extraction of hard 
 rubber are exactly the same as that for soft rubber that is, the powder of 
 hard rubber, about 2 grms., is placed in an extraction thimble and treated 
 for two days in the apparatus figured upon page 240 of this book. 
 
 As a rule, the organic extract obtained in this manner is very small 
 in quantity, consisting chiefly of sulphur, the exact amount of which is 
 ascertained by oxidation with fuming nitric acid, and determination of the 
 sulphuric acid in the usual way. 
 
 If the amount of organic matter present, as ascertained by the differ- 
 ence between the amount of the total extract and the amount of free 
 sulphur, exceeds 4 per cent., it may be assumed that resinous bodies or 
 resins other than those naturally occurring in india rubber are present. 
 A great number of the resins are soluble in acetone, others partially 
 soluble, but these solubilities may undergo considerable changes in the 
 course of the action upon them of high temperatures in the presence 
 of a large amount of sulphur. For this reason the examination of the 
 extracted resins can only deal with their general features, and the identi- 
 fication of any particular resin is always a matter of very great difficulty 
 and uncertainty, especially as the chemistry of the resins is still in almost 
 as chaotic a state as the chemistry of india rubber. 
 
 While acetone dissolves a considerable number of resins completely, 
 it dissolves a number of them only partially, and others not at all, as will 
 be seen from the following table, which contains those more generally used 
 in india rubber goods, especially in the manufacture of hard rubber. 
 
 Name of Resin. 
 
 Acetone. 
 
 Epichlorhydrine or 
 Dichlorhydrine. 
 
 
 soluble. 
 
 soluble 
 
 Amber, 
 Gum Benzoin, ...... 
 Colophony (Rosin), 
 Copal, 
 
 insoluble, 
 soluble. 
 
 partially soluble. 
 
 
 Elemi, . 
 
 soluble, 
 partially soluble 
 
 
 Sandarac, . . * . . 
 
 Shellac . .-'.'"'. 
 
 soluble, 
 insoluble 
 
 partially soluble, 
 soluble 
 
 
 
 
 1 Owing to the almost imperceptible swelling action of carbon bisulphide upon hard 
 rubber, the employment of this solvent in the extraction in place of acetone would 
 otherwise be wholly unobjectionable. 
 
26O INDIA RUBBER AND ITS ANALYSIS. 
 
 From this table it is evident that in a number of cases the acetone 
 extraction fails to remove the whole, and in two rather important instances 
 to remove even part, of the resins present. We require, therefore, a 
 further treatment with a solvent capable of extracting those resins which 
 are insoluble in acetone. 
 
 Such a solvent we possess in epichlorhydrine, a and therefore the acetone 
 extraction is immediately followed by extraction with this solvent. For 
 this purpose it is quite unnecessary to dry the extraction thimble after the 
 acetone extraction. As soon as the flask containing the acetone extract 
 is taken off, a similar weighed flask containing about 75 c.c. of epichlor- 
 hydrine is connected with the extractor. Owing, however, to the higher 
 boiling point of epichlorhydrine (117 C.), this flask must now be heated 
 upon the sand bath, or upon an asbestos-coated piece of wire netting. 
 Three hours' extraction fully suffice to remove all the residual resins 
 present. The flask is then removed, and the epichlorhydrine distilled off. 
 To avoid overheating and decomposition of the resins extracted, the flask 
 is heated in a paraffin bath, the temperature of which is kept at from 120 
 to 122 C. Finally, the flask is dried for two hours at 110 C. after its 
 outside has been carefully cleaned by wiping it first with filter paper, and 
 then with filter paper moistened with benzene. 
 
 This extract not infrequently contains sulphur in organic combination, 
 the amount of which may readily be estimated in the same manner as the 
 sulphur in extract C. (see page 244). ^ 
 
 In proceeding next to the extraction with pyridine, it is advisable 
 first to dry the paper thimble and its contents, as pyridine and epichlor- 
 hydrine interact with each other. The dry thimble is then replaced in 
 the extractor, to which a weighed flask containing about 60 c.c. of pyridine 
 (B.P. 109 to 111 C.) is fitted. It is heated upon a sand bath or in an oil 
 bath, the temperature of which is kept at from 116 to 120 C. Extraction 
 being complete, which is easily judged by the extraction thimble having 
 resumed its normal colour, the flask is taken off and the pyridine distilled 
 off in an oil bath at a temperature not exceeding 120 C. The flask is 
 then carefully cleaned and dried until the smell of pyridine has completely 
 disappeared, when the weight of the extract may be ascertained. 
 
 Also in this extract the sulphur should be determined whenever a fully 
 detailed analysis is required. The determination of the sulphur is carried 
 out in the manner described on page 244. 
 
 In the residue from the extraction with pyridine we determine the 
 sulphur present in the same manner as the ' total sulphur in rubber ' is 
 determined. The amount of sulphur thus found represents the sum of 
 the sulphur of vulcanisation and of the sulphur in the mineral matter. 
 In order to ascertain the sulphur of vulcanisation, we must therefore 
 estimate the sulphur contained in the mineral matter in the form of sul- 
 phides and sulphates. 
 
 1 Dichlorhydrine is not so suitable on account of its high boiling point (177 C.). 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 26 1 
 
 To this end we first determine the mineral matter in the sample by 
 incinerating a weighed quantity and making a complete analysis of the ash 
 in respect of the bases present. All the lead found is calculated as 
 litharge, the antimony as the pentasulphide, and the mercury as mercuric 
 sulphide. Any zinc found may be present either as oxide or sulphide ; 
 calcium may be present either as carbonate or sulphate l \ barium is present 
 either as carbonate, but much more frequently as sulphate ; magnesium 
 is always present as oxide. Information upon this point is obtained by 
 treating a sample of the ebonite powder with warm hydrochloric acid 
 (1 : 3). Any hydrogen sulphide evolved will be due to the presence of 
 zinc sulphide, evolution of carbonic acid to calcium carbonate. If no 
 barium passes into solution, this may be taken to prove that any present 
 is in the form of barium sulphate. 
 
 According to the results thus obtained, the quantities of sulphur corre- 
 sponding to the amount of zinc present, calculated as zinc sulphide, and of 
 the amount of barium present, calculated as barium sulphide, together 
 with any sulphur present in the form of antimony pentasulphide and 
 mercuric sulphide, are deducted from the amount of sulphur determined 
 in the residue F., and the difference taken as sulphur of vulcanisation. 
 It is scarcely necessary to point out that in every case the respective per- 
 centages must be calculated upon the whole of the sample from which the 
 residue F. is derived. 
 
 It could not be said that this method for the determination of the 
 sulphur of vulcanisation in hard rubber is particularly accurate. Tangible 
 errors must be caused if zinc or barium should be present in two forms at 
 the same time. That this ever occurs in respect of barium is highly im- 
 probable, but is perhaps not so very uncommon in respect to zinc, as, for 
 instance, the use of both zinc oxide and lithopone (ZnS, BaS0 4 ) in the 
 same mixing has several points of practical import to recommend it. 
 
 By way of a further, but by no means very accurate, control of the 
 amount of mineral matter present, a sample of the hard rubber is incin- 
 erated with the same precautions as are observed in the case of soft rubber. 
 The loss of weight is, of course, organic matter, together with the whole of 
 the sulphur not present in inorganic combination. In the interest of 
 further control of the total of the sulphur determinations, a .determination 
 of the total sulphur in the original sample is also desirable. 
 
 We have thus estimated : 
 
 1. Extract from acetone. 
 
 2. Extract from epichlorhydrine. 
 
 3. Extract from pyridine. 
 
 , (a} Sulphur in 1 . 
 
 j (6) Sulphur in 2. 
 
 -, \ (0 
 
 4. Total sulphur, \ (c) Sulphur in 3. 
 
 I (4) Sulphur in residue from pyridine. 
 ^ (5) Sulphur in mineral matter. 
 
 5. Mineral matter. 
 
 1 Quite negligible amounts also, as hydroxide. 
 
262 INDIA RUBBER AND ITS ANALYSIS. 
 
 The difference between all these determinations, expressed in percent- 
 ages from 100, represents the percentages of india rubber present in the 
 sample. Like all indirect estimations, this is also open to a number of 
 objections and errors. With normal samples these are hardly serious, but 
 in exceptional cases they may become rather troublesome. This simply 
 amounts to saying that there are more limitations to our present power of 
 analytical discrimination of the organic constituents of hard rubber goods 
 than in the case of soft rubber goods. Imperfect as are these analytical 
 methods at the present moment, they are, however, already a valuable aid 
 in gaining an insight, otherwise unobtainable, into the rationale of many of 
 the crudely empirical processes of the india rubber factory, and in throwing 
 light upon an ever-increasing number of the many obscure phenomena 
 observed, as well as on the behaviour, in practical work, of manufactured 
 india rubber. Considering the years of work required to bring the analysis 
 of many infinitely less refractory industrial products to a satisfactory con- 
 clusion, the results at present obtainable in rubber analysis are highly 
 gratifying, but it is neither a very suitable field for the half-educated chemist, 
 nor for the chemist who merely wants ' to do rubber,' to engage in. 
 
 (d) INTERPRETATION OF ANALYTICAL RESULTS. 
 
 Once the analytical figures have been obtained, we have to interpret them. 
 This may be a very simple operation, according to the question the analysis 
 is intended to solve ; or it may be a task of extreme difficulty, one at any 
 rate involving a thorough acquaintance with every known point of the 
 chemistry of india rubber, as well as extensive practical experience in its 
 manufacture. It is an utter impossibility to discuss exhaustively the in- 
 terpretation to be placed upon every one of the available analytical data, 
 or every combination of them, so as to embrace every possible case ; but 
 there are, with regard to every one of these data, certain features of 
 permanent significance, and these I now propose to discuss in the order 
 in which they appear in the various consecutive operations of the analytical 
 scheme above set out. 
 
 1. Extraction with Acetone. To the extraction with acetone I attach 
 considerable importance. It gives us, at the very outset, a very fair idea, 
 both from the point of view of its quantity and of its quality, of the general 
 character of the sample under examination. 
 
 A sample consisting only of india rubber and sulphur, or india rubber, 
 sulphur, and mineral matter, always yields a very clean-looking extract, 
 whatever kind of india rubber it contains. The extracts in this case are 
 rather light in colour, and do not prevent the separation of the sulphur in 
 well-formed crystals. By washing such extracts about five times in suc- 
 cession, each time with 2 c.c. of acetone in the cold, and filtering the 
 successive washings into a weighed flask through a very small filter, a very 
 satisfactory separation of the organic extract, due to india rubber, from the 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 263 
 
 free sulphur present may be effected. In doubtful cases it is not quite 
 easy to decide whether the sulphur is actually free from all organic 
 impurities. A good test in this case is to add to the sulphur residue in 
 the flask about 2 c.c. of carbon bisulphide, this preferably filtered through 
 the filter which served for the acetone washings. The sulphur almost 
 instantly dissolves, and the carbon bisulphide is now evaporated on the 
 water bath, and the flask then for a few minutes dried in a water oven. In 
 the absence of all organic matter the sulphur in the flask is thus obtained 
 in the form of globular masses of liquid (colloidal) sulphur adhering to the 
 sides. These globules often persist for days without crystallising. In the 
 presence of mere traces, even of organic matter, these globules are not 
 obtained, the sulphur at once forming somewhat indistinctly crystalline 
 scales on the sides of the glass. 
 
 The amount of organic extract obtained from rubber samples as above 
 described depends, of course, entirely upon the brand of india rubber used 
 in their manufacture, and for this reason the organic extract obtained from 
 such samples enables us to draw a general conclusion of the class of rubber 
 used for their manufacture, though not of the particular brand. This, even 
 with the limitation stated, is always a valuable point in rubber analysis. 
 Now, from the table given on page 3 of the first section of this book, it 
 appears that the percentage of resinous extract yielded by the fine grades 
 of Para rubber is very regular indeed, amounting almost uniformly to 1 *3 
 per cent., but from this it must not be assumed that a sample of manu- 
 factured Para must, on analysis, yield no more than 1 '3 per cent, of organic 
 extract. It is a common property of all kinds of rubber to show a 
 considerable increase of resinous matter after vulcanisation. But in order 
 to recognise the relation of the resinous extract obtained to the amount of 
 india rubber present, the percentage figures for these two constituents must 
 always be calculated separately, so as to show the absolute percentage of 
 resin contained in the india rubber of the sample alone. Thus, a sample l 
 might have yielded the following figures on analysis : 
 
 India rubber, 24 '49 percent. 
 
 Organic extract, I'lO ,, 
 
 Free sulphur, 1'23 ,, 
 
 Sulphur of vulcanisation, . . . . 0'84 ,, 
 
 Mineral matter, 72 '33 ,, 
 
 At first sight the amount of organic extract found might appear to be 
 well within the limits of the figure to be expected for pure Para, but very- 
 little reflection will show that the percentage of organic extract is stated 
 here in relation to the whole mixing instead of in relation to the amount of 
 india rubber alone. For this reason it is advisable to append to the per- 
 centage figures of every analysis a synoptical figure organic extract in 
 india rubber stating the percentage of this constituent in the india 
 rubber itself. In this way we obtain for the 'organic extract in india 
 i The sample was a roller covering manufactured from fine Para, 
 
264 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 rubber' the figure 4-29. 1 This curious increase of the 'resinous' matter 
 of india rubber in the course of vulcanisation, which I have already remarked 
 upon in an earlier paper, 2 not only occurs in Para rubber, but in all brands 
 of india rubber generally, and it amounts in all cases I have examined to 
 an increase of from 3 to 5 per cent. The following table contains the 
 figures of a few typical varieties of india rubber after vulcanisation with 
 1 per cent, of sulphur : 
 
 Brand of Rubber. 
 
 Resin in "Washed 
 Rubber. 
 
 Resin in Vulcanised 
 Rubber. 
 
 Para, fine, 
 Geara, 
 
 1 '2 per cent. 
 2'1 
 
 4*04 per cent. 
 5'12 
 
 Upper Congo, 
 Lagos 
 Sierra Leone, ..... 
 Borneo, ...... 
 
 37 
 4-5 
 6-1 
 10 '3 
 
 7'60 
 7-13 
 9-97 
 14-44 
 
 
 
 
 This fact discounts to a considerable extent the value of the conclusions 
 drawn in respect of the india rubber contained in an unknown sample 
 upon the basis of the resinous extract obtained. On the other hand, 
 however, on what is always the most important question in this relation 
 namely, whether any one sample has been manufactured from Para 
 the percentage of resinous extract obtained may always be used as a 
 very safe guide. The highest percentage of extract I ever obtained 
 from vulcanised Para never reached 4 '5 per cent., and always exceeded 
 3 per cent. 
 
 Of course this increase in the resinous extract only occurs in warm 
 vulcanised rubber; cold vulcanised samples invariably show the normal 
 percentages of resinous matter. The above mentioned increase is, as I 
 have satisfied myself by special experiments, chiefly due to the heating of 
 the rubber, and in a lesser degree to the action of the sulphur, probably 
 upon the oxygenated bodies which are contained in the crude india rubber, 
 and which are considerably larger in amount than the resins which can be 
 extracted from it by means of acetone. 
 
 India rubber goods which have undergone oxidation always show a 
 
 1 It is obvious that this figure is not strictly accurate. The technically pure india 
 rubber, from which the above article was manufactured, contains the resinous matter 
 subsequently extracted as a natural constituent. The proper course would be, therefore, 
 to add the percentages of rubber and resin found, and calculate the percentage of resin 
 observed upon this sum. But this would lead to considerable errors in the case of 
 samples yielding large percentages of organic extract due in small part only to the rubber 
 present, but chiefly derived from free oils, fatty substitutes, and the like contained in 
 the sample. It is therefore all round preferable to calculate, as we have done above, the 
 ratio between the percentage of actual rubber and the percentage of organic extract 
 observed. The error thus introduced is smaller than the variations in the percentage of 
 resin in crude india rubber. 
 
 2 "On the Analysis of India Rubber Articles," Journ. Soc. Chem. Ind., 1894, 
 pp. 3650-3658. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 265 
 
 more or less considerable increase of the resinous extract, and by the time 
 a sample has suffered from oxidation to such a degree as to render it 
 clearly defective, it will show an amount of resinous matter very largely in 
 excess of the normally occurring amounts. 
 
 The presence of mineral oils is always clearly recognisable by the 
 strong colour, and particularly by the characteristic fluorescence, they 
 impart to the acetone solution. They are easily separated from the 
 resinous constituents of the extract by means of absolute alcohol, in which 
 they are quite insoluble. In fact, this is a very satisfactory method for 
 their separation and estimation. 
 
 Not infrequently the organic extract contains resin oils instead of 
 mineral oils. Their presence generally betrays itself by the blue fluor- 
 escence of the acetone solution. They may be more definitely identified 
 by their very characteristic odour, and still better by RENARD'S reaction 
 with anhydrous stannic chloride, with which they develop a very fine and 
 intense purple coloration. 1 
 
 Mineral or resin oils are frequently added to india rubber mixings. 
 They also form one of the constant constituents of recovered rubber. As 
 soon, therefore, as the presence of any of these oils is observed, this point 
 should be remembered, as, in conjunction with other criteria to be mentioned 
 later, it often enables us to pronounce upon the presence of recovered 
 rubber in a sample under examination. 
 
 Free fatty oils in the extract occur just as often as mineral oils. Either 
 they have been added as such to the mixing tallow being also occasionally 
 used in this manner or they are derived from fatty substitutes present 
 which contain from 2 to 14 per cent, of soluble glycerides. Of course it 
 is also possible that substitutes and free fatty oils are at the same time 
 added to a rubber mixing. 
 
 The insolubility of these oils renders their separation from the resinous 
 extract a very simple matter. It is effected by repeated treatment with 
 absolute alcohol, in which the fatty oils, with the exception of castor oil, 
 are practically insoluble. 
 
 If in these fatty oils the presence of mineral oils is suspected, they 
 should be saponified with a few cubic centimetres of alcoholic caustic soda. 
 This solution is evaporated to dryness, the residue taken up with boiling 
 water, transferred to a small separating funnel, and extracted with ether. 
 On evaporating the ether, any mineral oil present remains behind and 
 may be further identified. 
 
 If no fatty substitutes are present, any oils found in the extract must 
 have been added to the rubber mixing either as such or in the form of 
 recovered rubber, which very uniformly contains about 8 per cent, of them. 
 
 1 ALLEN recommends in place of stannic chloride the use of stannic bromide, which 
 is easily prepared by dehydrating bromine with sulphuric monohydrate in a small 
 separating funnel, and allowing it to flow, drop by drop, upon tin shavings contained in a 
 well-cooled flask. 
 
266 INDIA RUBBER AND ITS ANALYSIS. 
 
 If fatty oils alone are present, the inference is obvious : in the presence of 
 fatty substitutes, at least part of them will be derived from these. If the 
 amount of fatty oils present does not exceed 10 per cent, of the amount of 
 fatty substitutes, the addition of free oils as such to the rubber is very 
 improbable. 
 
 It not infrequently happens that the correctness of an analysis is 
 challenged if mineral oils or fatty oils are reported as present. This, of 
 course, can only happen if the sample in question contains recovered 
 rubber or fatty substitutes respectively, of the composition of which most 
 rubber manufacturers are blissfully ignorant. It is therefore advisable, in 
 preparing a report, always to point out that mineral oil, if any be present, 
 may or may not have been added as such. Similarly, attention should always 
 be called to the relation of fatty oils to fatty substitutes. 
 
 For manufacturing purposes it is comparatively irrelevant whether any 
 fatty oil found in a sample to be matched has been added as such, or 
 whether it is derived from the fatty substitutes present, but it is, of course, 
 necessary to ascertain the percentage of free fatty oil in the substitutes 
 which are to be employed, and to make up the deficiency of free fatty oil, 
 so as to bring it to the proportion of the free oil in the sample. If the 
 total of free fatty oil in the substitute is, however, more than the corre- 
 sponding proportion in the sample, it is obvious that a substitute 
 containing less free oil should be employed. 
 
 The same principle must also be observed in respect of the amount of 
 resinous extract ascertained in the sample, and it is obvious that the 
 amount of resin contained in the rubber to be employed must be taken into 
 consideration. 
 
 Whether any mineral oil present has been added as such in the 
 manufacture of the sample, or whether it is due to the presence of 
 recovered rubber, is a question which at present cannot be decided by 
 chemical means. Often the tensile strength test furnishes indications on 
 this point. 
 
 The presence of paraffin wax or ozokerite, of which some samples 
 contain as much as 15 per cent., always betrays itself by the peculiar 
 waxy character of the extract, and also by its comparative insolubility in 
 cold acetone. Not infrequently these hydrocarbons are added to the 
 mixing in the form of a preparation which is now in the market under the 
 name of ' Nantusi,' but which for a number of years has been employed by 
 several firms with a great display of secrecy. It is a quack ' medicine ' 
 against suncracking, consisting of about 30 per cent, of sulphur and 
 about 70 per cent, of a mixture of paraffin wax with a small quantity of 
 bee's- wax and tallow. If it is desired to determine quantitatively the 
 amount of paraffin wax or ozokerite present, the extraction residue is 
 warmed with about fifty times its weight of glacial acetic acid. All the 
 resinous and fatty extract, together with mineral oils present, is dissolved. 
 The paraffin wax, on cooling, separates out almost quantitatively, 1 part 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 267 
 
 of it requiring 1668 parts of acetic acid for solution. 1 The solution is 
 filtered off through a small filter, flask and filter are washed with a little 
 acetic acid, the filter placed upon the flask containing the paraffin wax, and 
 any particles of the latter which are upon the filter washed back with about 
 50 c.c. of warm carbon bisulphide. The latter is then distilled off, and the 
 flask dried and weighed. 
 
 Incidentally it may be of interest to remark upon the effect of paraffin 
 wax in rubber mixings. It lies, briefly, in the fact that india rubber and 
 paraffin wax will not intermix homogeneously to form a sort of colloidal 
 solution. Indeed, india rubber is quite insoluble in paraffin, and therefore 
 does not absorb it as it absorbs a solvent. As the result of this, there is a 
 distinct tendency in compounding india rubber and paraffin wax for the 
 latter to be squeezed out of the mixing. This is, indeed, what happens if 
 an excess of paraffin wax is used. What represents an excessive proportion 
 depends, however, largely upon the other constituents and their proportions 
 in the mixing. Now, under the influence of the heating of the mixing upon 
 the rollers, the paraffin wax melts, and by capillary action fills the pores of 
 the mixing, thereby greatly reducing the absorptive power of the manu- 
 factured india rubber for water and oxygen. The effect of this is, 
 of course, to correspondingly reduce the liability of such india rubber to 
 oxidation in general, and to suncracking in particular, which, as I have 
 already pointed out, is simply oxidation accelerated by the influence of the 
 actinic rays of the sun. There is, however, this disadvantage, that this 
 immunity against oxidation is obtained at the expense of tensile strength. 
 This result, which any physicist might at once predict for a mixture 
 of two bodies of the nature of paraffin wax and india rubber intermingled, 
 is iu some cases quite immaterial, although objectionable in the highest 
 degree in others. In bicycle tyres, I have, in a large number of cases, 
 observed a reduction of tensile strength amounting to more than 50 per 
 cent. Under the most favourable conditions of manufacture, this may be 
 reduced to about 30 per cent., which is still quite prohibitive in an article 
 requiring to withstand so much wear and tear. 
 
 II. Extraction with Pyridine. Additions of tar both coal tar and that 
 obtained in the destructive distillation of wood (Stockholm tar) and pitch 
 are frequently made to rubber mixings. Their action consists in the 
 facilitating of the mixing operation, and also in the reduction of the porosity 
 of the manufactured india rubber. Asphaltum is also occasionally used for 
 this purpose, but its advantage is not very apparent. 
 
 The tar is often added in its crude commercial state, but more 
 frequently the plan is adopted of thickening it by continued heating, 
 generally with addition of resin, until it forms, when cold, a scarcely tacky, 
 semi-elastic, but not brittle, mass. In using asphaltum, no such operation 
 is required, as the asphaltum of commerce already forms a hard, very brittle 
 body which can easily be powdered. According to HENRIQUES, asphaltum 
 1 PAWLEWSKI and FILEMONOWICZ, Ber., xxi. p. 2973. 
 
268 INDIA RUBBER AND ITS ANALYSIS. 
 
 is much used in large proportions in india rubber mixings ; but as far 
 as the author's experience goes, it tends to show that, at any rate in soft 
 rubber goods, its use is of the rarest occurrence. It may possibly find a 
 somewhat more extensive employment in hard rubber goods, but its 
 advantage in such mixings over many other organic ' filling agents ' is by 
 no means clear. 
 
 It is in most cases impossible to decide whether crude or boiled tar has 
 been added to a sample, but it is safe to assume, in the absence of contrary 
 evidence, that in ' spread ' goods the former, and in moulded, calendered, 
 or ' built up ' goods the latter, has been employed. 
 
 Not infrequently coal tar contains a considerable amount of car- 
 bonaceous matter which, of course, is not extracted by the pyridine. 
 This carbonaceous matter is encountered in the subsequent estimations, 
 and there is no possibility of distinguishing it from lamp black, which may, 
 or may not, be present at the same time. Of this carbonaceous matter 
 crude gas tar contains generally about 15 per cent., but as much as 35 
 per cent, have been found occasionally. Boiled tar contains about 20 per 
 cent, of it. Considering that the proportion of tar added to rubber mixing- 
 varies in the vast majority of cases from 0'5 to 1 per cent, of the total 
 mixing, and only in quite exceptional cases amounts to 2 per cent., it is 
 evident that the possible error due to this carbonaceous matter is so small 
 that it is without any sensible influence upon our determinations. 
 
 Incidentally it may be mentioned here that it is sometimes desirable, 
 for manufacturing purposes, to know the amount of carbonaceous matter 
 contained in either crude coal tar, or in the boiled tar, or in coal tar pitch, 
 which is also occasionally used in rubber mixings. This can be ascertained 
 by mixing 5 grms. of the tar with 1 5 grms. of aniline, or by boiling 5 
 grms. of the boiled tar or pitch for a few minutes with 25 grms. of aniline, 
 and pouring the solution in a small dish of unglazed porcelain. The 
 solvent, together with the soluble constituents of the tar, is absorbed by 
 the porous dish, while the carbonaceous matter remains behind as a 
 coherent, laminated mass, which, by means of a horn spatula, can be trans- 
 ferred without any loss to a weighed watch-glass, and after several hours' 
 drying in a water oven, weighed until constant. 
 
 III. Extraction with Alcoholic Potash. This extract contains what is 
 at present the most important class of organic filling materials for rubber 
 goods, and which comprises the triglycerides of a variety of unsaturated 
 fatty acids transformed into fairly resilient and insoluble compounds, either 
 by the addition of oxygen, of sulphur, or of sulphur monochloride. 
 
 Of these three classes of fatty substitutes, the oxidised oils and the 
 chlorosulphides (white substitute) may occur in either warm or cold cured 
 goods; the sulphides (brown substitute) are found in hot cured goods 
 only. Not infrequently mixtures of two of these substitutes are used. 
 
 The identification of these substitutes in the extract depends upon the 
 fatty acids obtainable from the saponification products still possessing the 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 269 
 
 essential chemical characteristics of the original substitutes. We therefore 
 find that : 
 
 1. The fatty acids obtained from oxidised oils are either entirely free 
 from sulphur or chlorine, or contain only very small quantities of these 
 elements, the presence of which is due to the influence of the sulphur, or of 
 the chloride of sulphur, upon the oxidised oil during the process of vul- 
 canisation. In oxidised oil substitutes the percentages of chlorine or 
 sulphur present, therefore, rarely exceed 0'5 per cent. 
 
 2. The fatty acids from the class of substitutes which I described 
 as sulphide substitutes (brown substitutes) contain essentially the same 
 quantity of sulphur as the substitutes themselves i.e., from about 4 
 to 9 per cent. but they are entirely free from chlorine, or, at least, 
 practically so. 
 
 3. The fatty acids of the chlorosulphide substitutes (white substitutes) 
 still contain practically the whole of the sulphur of the original substitute, 
 but the greatest part of the chlorine has been eliminated by the saponi- 
 fication process, and can be determined in the saponification liquors by the 
 usual means. 
 
 It will be seen, therefore, that the identification of the oxidised oil 
 substitutes and of the sulphide substitutes offers not the least difficulty if 
 they are present by themselves only. Equally simple is the detection of 
 chlorosulphide substitutes in cold cured articles, but their use in hot cured 
 goods is apt to lead to not inconsiderable analytical complications. 
 
 It has been shown (see page 146) that the chlorosulphide substitutes at 
 vulcanising temperatures evolve a more or less considerable proportion of 
 their chlorine in the form of hydrochloric acid. 1 HENRIQUES 2 surmised, 
 erroneously however, that the chlorine thus evolved is altogether lost. In 
 fact, scarcely a trace of it disappears, it being almost completely taken up 
 by the india rubber forming a polyprene hydrochloride. If such a sample 
 should contain at the same time substances like zinc white or litharge, a 
 small but distinct proportion is absorbed by these compounds. Anyhow, 
 it will be observed that in cases like these only the sulphur of the 
 substitute is found where we should expect it, the chlorine having 
 migrated to the india rubber to such an extent that its presence is either 
 not recognised at all, or misinterpreted as being due to the vulcanisation 
 of the sample by the cold process. The consequence of this mistake is 
 the misinterpretation of the chloride substitute as a sulphide substitute. 
 
 It is therefore evident that for the purpose of distinguishing between a 
 sulphide substitute and a chlorosulphide substitute, some information 
 beyond that obtainable from the components of the extract derived from 
 
 1 Chlorosulphide substitutes prepared from different oils show rather considerable 
 variations respecting the proportion of their total chlorine given off by them at 
 vulcanising temperatures. Thus, colza oil substitutes lose about 80 per cent.; cotton 
 seed substitutes, about 94 per cent. ; castor oil substitutes, practically the whole of their 
 chlorine. 
 
 2 Chew. Zeit., 1893, No. 40. 
 
270 INDIA RUBBER AND ITS ANALYSIS. 
 
 the treatment with alcoholic caustic potash is required. It will be seen 
 that this information is yielded partly by the estimation of the sulphur and 
 chlorine in the india rubber (section VI. of the analytical scheme on page 
 255), partly also by purely technical considerations. 
 
 In this relation it is clear that if there is no chlorine in the india 
 rubber in a quantity about equal to the amount of sulphur in the saponi- 
 fication extract, the substitute indicated by the latter cannot be a chloro- 
 sulphide substitute, but must be a sulphide substitute. Of course it 
 might be suggested that a rubber article containing a sulphide substitute 
 might be vulcanised with chloride of sulphur, in which case, according to 
 the above proposition, the substitute would be misinterpreted as a chloro- 
 sulphide substitute. I have, however, already stated that sulphide 
 substitutes are unsuitable for, and therefore never used in, mixings to be 
 vulcanised by the cold process. As there is, however, no telling what 
 might or might not be attempted in rubber mixings, attention should be 
 drawn to the fact that cold cured rubber goods yield either no free sulphur 
 at all, or only quite infinitesimal quantities of it, in the acetone extract. 
 On the other hand, both sulphide substitutes and hot vulcanised india 
 rubber goods contain substantial amounts of free sulphur, so that in such 
 a doubtful case the amount of sulphur in the acetone extract would not 
 fail to attract attention. 
 
 Of course where there is no chlorine to be found in the india rubber, 
 the identification of the substitute from the indications furnished by the 
 saponifi cation extract is perfectly simple and obvious. In a very large 
 number of cases the interpretation is at once simplified by the fact that 
 the article under examination is of a thickness which would not admit of 
 its vulcanisation by the cold process, which is applicable only to very thin 
 rubber structures, such as waterproof cloth, tobacco pouches, toy baloons, 
 and the like. Articles of greater thickness can only have been produced 
 by the process of hot vulcanisation, and any chlorine they contain must 
 be derived from chlorosulphide substitutes in them. 
 
 The presence of a chlorosulphide substitute in a cold cured sample of 
 india rubber is, of course, equally simple to detect. It is characterised by 
 the fact that chlorine and sulphur (not including the sulphur in inorganic 
 compounds) are present in approximately equal amounts, the total of 
 which is, however, considerably in excess of the corresponding total 
 which could be due to the chlorosulphide substitute alone. Nevertheless, 
 it must be stated that this circumstance alone is not invariably sufficient 
 to arrive at a certain conclusion, for the simple reason that the sum of the 
 percentages of chlorine and sulphur in chlorosulphide substitutes may 
 vary, according to the nature of the oil employed, from about 8 to 18 per 
 cent., although the bulk of the chlorosulphide substitutes give figures 
 varying within a narrow margin around an average of about 15 per cent. 
 Conclusive evidence is, however, to be found in the fact that in a sample 
 constituted as above, the ratio of chlorine to sulphur in both saponification 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 271 
 
 extract and in the residual india rubber is approximately the same as in 
 sulphur monochloride i.e., sulphur is present in about equal parts. 
 
 If a sample contains mixtures of these various fatty substitutes, their 
 identification becomes, of course, somewhat more difficult, and, indeed, in 
 certain limiting cases practically impossible. 
 
 The easiest to detect are mixtures of the chlorosulphide substitutes 
 with either the oxidised oil substitutes or the sulphide substitutes. 
 Mixtures of both these descriptions are frequently used in hot cured 
 goods. 
 
 The presence of oxidised oil substitutes, together with chlorosulphide 
 substitutes, has, of course, simply the effect of making the ratio between 
 the total fatty substitute present, inclusive of the chlorine in the india 
 rubber, and the sum of the sulphur and chlorine in them, appear pro- 
 portionately low. We obviously obtain the quantity in which each of 
 these substitutes is present by calculating the amount of chlorosulphide 
 substitute corresponding to the sum of the sulphur and chlorine in the 
 substitute. In doing this, the sum of these constituents in a normal sub- 
 stitute is taken to amount to 15 per cent., and by multiplying the sum of 
 these constituents, as analytically ascertained, by the factor 6 '6, we obtain 
 the quantity of normal chlorosulphide present. The difference between 
 this product and the total amount of substitute ascertained is assumed to 
 be oxidised oil substitute. 
 
 This might appear a very arbitrary proceeding, seeing that chloro- 
 sulphide substitutes themselves may contain very much less than 15 per 
 cent, of chlorine and sulphur collectively. But, in reality, this procedure is 
 not so arbitrary as it looks. It is only the chlorosulphides of the tri- 
 glycerides, of oxy-acids, or of partially oxidised (blown) oils which contain 
 less than 15 per cent, of sulphur and chlorine combined; and although, 
 from the point of view of ultimate chemical constitution, this calculation 
 would perhaps not be permissible, I have in very numerous practical 
 experiments never been able to discover the slightest difference between 
 the effect in a rubber mixing of a certain weight of (supposed) homogene- 
 ous chlorosulphide substitute containing only 9 per cent, of chlorine and 
 sulphur combined, or an equal weight of a mixture of 60 parts of normal 
 substitute (containing 15 per cent, of chlorine and sulphur combined) with 
 40 parts of oxidised oil substitute. In fact, the above calculation may be 
 taken in the sense of either of these two possibilities from the purely 
 analytical, as well as from the technical, points of view. 
 
 The very frequently occurring presence of a mixture of sulphide sub- 
 stitute and chlorosulphide substitute in rubber samples is equally easy 
 to establish. As the sulphide substitute also contains sulphur, it is evident 
 that we must calculate the amount of chlorosulphide substitute present by 
 calculating the amount of chlorosulphide substitute corresponding to the 
 amount of chlorine present, taking in this case, of course, the normal 
 chlorosulphide substitute as containing 7*5 per cent, of chlorine. The 
 
272 INDIA RUBBER AND ITS ANALYSIS. 
 
 difference between the amount of chlorosulphide substitute thus calculated 
 and the total amount of substitute as estimated represents the amount of 
 sulphide substitute present. 
 
 India rubber articles containing mixtures of oxidised oil substitutes and 
 sulphide substitutes are always hot vulcanised. The characteristic feature 
 of the saponification extracts lies, in this case, in the abnormally low per- 
 centage of sulphur in the fatty acids. But, in the chapter on india rubber 
 substitutes, I have shown that the amount of sulphur in the sulphide 
 substitutes (brown substitutes) is apt to display as great variations in the 
 percentage of sulphur as do the chlorosulphide substitutes, in respect of 
 the percentage of the sulphur monochloride they contain, and for the same 
 reasons. Sulphide substitutes prepared from the unsaturated oils in their 
 natural condition contain much more sulphur than those prepared from 
 partially oxidised (blown) oils. If we take, therefore, the percentage of 
 sulphur of a normal sulphide substitute as 15 per cent., and calculate 
 upon this basis, and upon the basis of the amount of sulphur ascertained 
 in the saponification extract, the amount of normal sulphide substitute 
 present, the figure thus arrived at, deducted from the total saponifica- 
 tion extract, gives us the amount of oxidised oil substitute present. 
 In this case also such an interpretation may be chemically correct, 
 although not so technically. In other words, in the manufacture of the 
 sample, a sulphide substitute prepared from a partially oxidised oil may 
 have been used, and not a mixture of a normal sulphide substitute and an 
 oxidised oil substitute. On the point of fact this ambiguity is quite 
 irrelevant, as the same technical effect is obtained in either case. 
 
 Occasionally it may be desirable to characterise somewhat more 
 definitely the nature of a substitute contained in manufactured india 
 rubber. This is, however, only worth attempting in cases where the 
 analysis has shown one only of these substitutes to be present. To identify 
 the oils from which fatty substitutes have been prepared in mixtures of the 
 free acids of the latter, our present methods of oil analysis are, however, 
 quite inadequate. 
 
 Where an examination in this direction appears desirable, the most 
 satisfactory method consists in submitting the free fatty acids isolated from 
 the saponification extract to HUBL'S iodine absorption test. The pre- 
 liminary operation is carried out as follows : 
 
 The saponification extract is evaporated to dryness, redissolved in 
 water, the solution acidulated with dilute sulphuric acid, and extracted in 
 a separating funnel with ether. The ether solution is dried by adding to 
 it some granulated calcium chloride, and allowing it to stand for a few 
 hours. It is then filtered through a dry filter, the ether evaporated, and 
 the residue, consisting of the fatty acid of the substitute, then dried for a 
 short time in a water oven until all smell of ether has completely dis- 
 appeared. 
 
 A part of this extract is weighed in a flask and subjected to the iodine 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 273 
 
 absorption test, for the details of which some standard work on oil analysis 
 should be consulted. 1 The figures thus obtained should be compared with 
 the iodine absorption of the fatty acids of substitutes of known composition. 
 A few such determinations are given in the table on page 147. 
 
 Incidentally it may be remarked here that before saponifying a fatty 
 substitute for the purposes of the above determination, it is absolutely 
 essential to first submit the substitute to an extraction with acetone so as 
 to remove free fatty oils and sulphur. Unless this is done, the iodine 
 absorption standards are useless for comparison with the corresponding 
 figures obtained with the free fatty acids extracted from manufactured 
 india rubber by means of alcoholic potash. 
 
 IV. Extraction with Nitro-naphthalene. The extraction with nitro- 
 naphthalene, or boiling nitrobenzene, results in the solution of the 
 vulcanised india rubber in the solvent named. The process is not 
 attended by the evolution of sensible quantities of hydrogen sulphide or 
 hydrochloric acid. If the sample dissolves without leaving an appreciable 
 amount of mineral residue, the mineral matter present cannot exceed 1 per 
 cent., and this result should be confirmed by determining, on another 
 quantity of the sample, the amount of ash obtained by incineration. As 
 a matter of fact, seeing that this method of separating india rubber and 
 mineral matter still suffers from various difficulties and drawbacks, it is 
 advisable to check the result of this extraction in every case by an in- 
 cineration test. Indeed, in all cases not calling for a detailed analysis of 
 the mineral matter, the incineration method may be substituted for this 
 extraction method. 
 
 Of far more importance than the investigation of the mineral residue 
 would be a method enabling us to differentiate between the various brands 
 of india rubber which may be present in the nitrobenzene solution. At 
 present we possess no such method. In the first part of this book I have 
 shown that the chemical behaviour of the various kinds of india rubber, 
 apart from the results of vulcanising process, yields not the slightest clue 
 to indicate anything like a chemical difference between the hydrocarbons 
 forming the chief constituent of the numerous brands of india rubber. 
 
 Nor does this difficulty become less if we consider the question of 
 differentiating between the proportions of new rubber, and recovered, or 
 simply ground up waste rubber (rubber dust) in a given sample. But 
 while the former problem, the identification of the brand or brands of 
 india rubber in a manufactured sample, must at present be confined to 
 attempting to decide as to whether the india rubber is Para rubber or not, 
 the probability is that the second problem, the detection of recovered 
 rubber in manufactured articles, will before very long become amenable ta 
 analytical treatment. What is particularly required for this purpose is a 
 careful study of the relation of the coefficient of vulcanisation to the 
 tensile strength of the various brands of india rubber, likewise the 
 1 LEWKOWITSCH'S Analysis of Oils, Fats, and Waxes, 2nd ed., 1895. 
 
 18 
 
274 INDIA RUBBER AND ITS ANALYSIS. 
 
 influence upon the same constant of admixtures of fatty and other 
 substitutes, and of mineral matter. HEINZERLING and PAHL'S tables l are 
 a first step in this direction, though not sufficiently methodical for the 
 purpose. 
 
 Particularly instructive is the determination of the tensile strength 
 {breaking strain) in conjunction with the elongation or stretch t ol the 
 sample under test at the moment the break occurs. 2 In this respect I 
 have observed that while the addition of recovered rubber does not particu- 
 larly affect the tensile strength, it reduces the corresponding elongation 
 the more markedly, the higher the proportion of recovered rubber present. 
 
 The appearance of the surface of the rupture is also often highly 
 characteristic. The peculiarly smooth, often almost greasy-looking rupture 
 of articles containing fresh rubber only, rapidly disappears and becomes 
 characteristically rough as the admixture of recovered rubber increases. 
 Admixture of ' rubber dust ' is always easily recognisable by a % micro- 
 scopical examination of a freshly ruptured surface. A low magnification 
 of about 30 diameters is quite sufficient. This is obtained with an 
 objective of about 8 mm. (J inch) focus. 
 
 All the samples of recovered rubber I have examined contain oil of 
 one kind or another, generally mineral oil or petroleum jelly, not un- 
 commonly resin oil, comparatively rarely fatty oils, in quantities of from 8 
 to 10 per cent, and more. But it is, of course, rarely possible to decide 
 with certainty whether any such oil discovered in the course of analysis is 
 due to the presence of recovered rubber. Considering, moreover, that an 
 .admixture of 10 per cent, of recovered rubber would in the analysis yield 
 1 per cent, of mineral oil only, it is obvious that no definite conclusions can 
 be based upon the presence of the above named oils in a sample. 
 
 Already, in the chapter on rubber substitutes, the fact has been men- 
 tioned that recovered rubber undergoes vulcanisation just like fresh 
 rubber. In the same place I have pointed out that the description, often 
 to be met with, of recovered rubber as * devulcanised ' rubber is a 
 fraudulent misrepresentation, as the ' recovery ' process leaves the sulphur 
 -of vulcanisation quite unaffected. It is therefore obvious that in many 
 cases rubber goods manufactured with a large percentage of recovered 
 material exhibit an abnormally high percentage of sulphur of vulcanisa- 
 tion (coefficient of vulcanisation). This is, therefore, a fairly reliable in- 
 dication of the presence of recovered rubber, provided the result of the 
 tensile strength test does not indicate over vulcanisation. As a very large 
 proportion of the commercial recovered rubber is manufactured from worn 
 goloshes, the india rubber of which always possesses a high coefficient of 
 vulcanisation, the result indicated above will frequently be observed; it 
 fails, however, if the recovered rubber has been produced from rubber 
 articles possessing a low coefficient of vulcanisation. 
 
 2 The machine described on page 216 furnishes both these data in one test. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 275 
 
 Cold vulcanised rubber waste is also submitted to various ' recovery ' 
 processes, and the products are again used in the manufacture of cold 
 vulcanised goods. In this case, however, the detection of the admixture 
 offers no difficulty. Cold vulcanised rubber in the recovery process loses 
 practically the whole of its chlorine, and consequently in cold vulcanised 
 rubber goods containing this kind of recovered rubber the normal ratio 
 between sulphur and chlorine no longer exists, the sulphur considerably 
 preponderating. Of course this result would be much the same if ordinary 
 recovered rubber were used in cold cured rubber goods. 
 
 V. Extraction with Boiling Water. The extract obtained in this 
 operation scarcely calls for any comment. I am not aware that either 
 farina, or dextrine, which is only very rarely present, are more than mere 
 filling materials in a rubber mixing, with nothing to recommend them for 
 the purpose. 
 
 VI. Mineral Matter. This group comprises filling materials, inorganic 
 substances acting as filling materials and at the same time as sulphur 
 carriers, sulphur carriers proper, and colouring matters. 
 
 The first of these three groups calls for no special remarks, and such 
 points as refer to the influence of these filling materials upon the be- 
 haviour of the manufactured goods under the various conditions of their 
 practical uses will be better discussed in the next chapter. 
 
 It may be taken that all the sulphides of the heavy metals notably 
 those of antimony, lead, and zinc act as sulphur carriers. So do, to a 
 lesser degree, all lead compounds, owing to their partial conversion into 
 lead sulphide during the process of vulcanisation. Lead oxide (litharge) 
 is, in particular, very largely used for this purpose. Red lead, the presence 
 of which is not uncommon in ' mechanical ' goods, acts, of course, similarly, 
 and at the same time has a hardening action upon india rubber, probably 
 in consequence of its oxidising power. Indeed, the rationale of its employ- 
 ment seems to lie in this oxidising action, which, in articles produced from 
 a ' soft ' brand of india rubber, produces a firmness which can be obtained 
 by vulcanisation alone, and with the better brands of india rubber only. 
 The presence of red lead in the mineral matter may therefore be generally 
 taken as an indication that the rubber used in the manufacture of the 
 sample is almost to a certainty of somewhat low quality. 
 
 Very remarkable, likewise, is the hardening action of barium oxide, 
 calcium oxide, and magnesium oxide upon india rubber. The two first 
 named substances are always employed in the state of hydrates, the latter 
 in the form of the anhydrous oxide. The quantities used are always very 
 small, rarely exceeding 0'5 per cent., although magnesia is sometimes used 
 in somewhat larger proportions. 
 
 The cause of the hardening action of these substances is very obscure. 
 Baryta and lime might be assumed to form sulphides, and then act as 
 sulphur carriers, but in mixin'g either barium or calcium sulphide with the 
 india rubber, no such hardening effect is noticeable after vulcanisation, 
 
276 INDIA RUBBER AND ITS ANALYSIS. 
 
 and this would not, moreover, explain the analogous action of magnesia, 
 which does not even form a stable sulphide. 
 
 In the presence of calcium carbonate it is often rather difficult to detect 
 such small quantities of caustic lime. In this case, either the residue from 
 the nitro-naphthalene treatment, or a fresh part of the finely ground original 
 sample, is boiled for a short time with a 10 per cent, solution of sugar. 
 Any caustic lime present dissolves in this solution, from which it can be 
 precipitated by the addition of a moderate quantity of a solution of 
 ammonium carbonate. The precipitate, however, is never obtained in the 
 cold solution, but forms only after boiling it for a short time. 
 
 In coloured samples of rubber goods l it is often desirable to ascertain 
 the nature of the colouring matter present. In ' rubber enamelled ' goods 
 the colours of the enamel (suspensions of pigments in thin rubber solutions) 
 may be due either to mineral or lake pigments (aniline lakes), as these 
 ' enamel ' coatings are either lightly vulcanised by the cold process, or else 
 * acidised.' 2 The same is true in regard to coloured cold vulcanised articles 
 of every description. In rubber goods which are coloured throughout 
 their mass, and which have been vulcanised by the hot process, the colours 
 can only be mineral pigments, as even the fastest of lake pigments are 
 more or less seriously affected in the hot vulcanising process, although 
 they stand the cold process very well, almost without exception. 
 
 (a) Examination of Colouring Materials in Hot Cured Goods. 
 
 The colour imparted to rubber goods by antimony pentasulphide 
 (golden sulphide) or vermilion, or by a mixture of the two, is well known, 
 and the presence of either, or both, of these bodies can always be specially 
 confirmed by making a test with a fresh part of the sample. 
 
 Brown shades are most usually obtained with ferric oxide, which can- 
 not escape detection in the analysis of the mineral residue. It must, how- 
 ever, be borne in mind that these colours, as used, represent always 
 partially hydrated products, and, moreover, very frequently contain 
 considerable quantities of alumina, aluminium silicate, and similar bodies. 
 
 Yellow shades are somewhat unusual, and are almost invariably 
 produced with cadmium yellow. This should also be confirmed by special 
 test. Very occasionally zinc yellow, a double salt of zinc chromate and 
 potassium chromate, is used. Its presence is readily detected by fusing a 
 small quantity of the rubber sample with an intimate mixture of 1 part 
 of potassium nitrate and 2 parts of sodium carbonate. The fused mass 
 is dissolved and filtered, when the presence of a chromate in the rubber 
 sample is at once demonstrated by the characteristic yellow colour of the 
 filtrate. Lead chromates cannot be used in hot cured goods, owing to the 
 action of the free sulphur upon lead compounds. 
 
 1 This has, of course, no reference to goods with surfaces decorated with paints or 
 lacquers. 
 
 3 See page 55. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 277 
 
 Green shades are almost invariably produced with chromic oxide, 1 
 very rarely with zinc green, a mixture of zinc yellow (see above) with 
 Prussian blue. Which of these two is present may be decided by boiling 
 a quantity of the finely ground sample with caustic soda, when, in the 
 presence of zinc green, a strongly yellow solution is obtained. If this test 
 yields a negative result, another part of the sample is fused with the 
 above mentioned fusion mixture, the melt dissolved in water, and filtered. 
 A bright yellow filtrate is proof of the green present being chromic oxide. 
 
 Blue shades are produced with ultramarine only, the very fine cobalt 
 blues being far too expensive. Any blue coloration of rubber may easily 
 be recognised as due to ultramarine by the decoloration produced by 
 moistening a piece of the original sample with dilute hydrochloric acid. 
 
 (b) Examination of Colouring Materials in Cold Cured Goods. 
 
 Cold cured goods may contain the same colouring agents as are 
 employed in the coloration of hot cured goods, but in the former a number 
 of mineral pigments may be employed which would not stand the hot 
 vulcanising process satisfactorily. This refers particularly to zinc yellow, 
 zinc green (mixtures of zinc yellow with Prussian blue), chrome yellow, 
 and chrome green (mixtures of chrome yellow with Prussian blue), and 
 Prussian blue. The detection of these pigments offers no difficulty what- 
 ever, and follows on the lines already mentioned above. The presence of 
 Prussian blue, whether by itself or in admixture with zinc chromate or 
 lead chromate, is discovered by boiling the finely ground sample with 
 dilute caustic soda, filtering, acidulating the filtrate with dilute sulphuric 
 acid, and adding a few drops of a solution of ferric chloride. Formation 
 of the well-known blue precipitate of potassium-ferri-ferrocyanide proves 
 the presence of Prussian blue in the sample. 
 
 It should be noted that in cold cured goods antimony pentasulphide, 
 vermilion, and cadmium sulphide are rarely used. The obvious reason is 
 that, as mere pigments, they are much more expensive than cheaper 
 colouring matters which are available for use in cold cured goods in which 
 there is no need for any sulphur carriers. 
 
 Of recent years lake pigments prepared from coal tar dyes have been 
 very extensively employed in the manufacture of cold cured rubber goods. 
 These pigments, which contain the dye fixed upon some inorganic basis, 
 such as alumina, barium sulphate, barium carbonate, or china clay, contain 
 varying amounts of the dyes named from 1 to 30 per cent., and both acid 
 and basic dye stuffs are employed in their manufacture. The presence of 
 such lake pigments is readily discoverable by boiling the ground rubber 
 sample in a test tube with 80 per cent, alcohol, and filtering. Any marked 
 
 1 Or, rather, with GUIGNET'S green, a chromic sesquihydrate 2Cr 2 3 ,3H a O. SHIPTON, 
 in A. W. HOFFMAN'S Report of the Juries, 1863, p. 75, gives its composition as 
 3Cr 2 3 .B 2 3 .4H 2 0. 
 
278 INDIA RUBBER AND ITS ANALYSIS. 
 
 coloration of the filtrate indicates the presence of a lake pigment (aniline 
 lake). 1 If the lake is one containing a basic dye, this alcohol extract is 
 generally very strongly tinted, whereas lakes prepared from acid dyes give 
 only a lightly coloured extract, the coloration being, nevertheless, always 
 unmistakable. 
 
 If it is desirable to obtain some more definite information on the nature 
 of the pigment present, the above test should be repeated, and two separate 
 quantities of the sample, the one being boiled with a 1 per cent, alcoholic 
 (80 per cent, strong alcohol) solution of ammonia, the other with a 1 per 
 cent, alcoholic solution of acetic acid. In the first of these solutions the 
 acid dyes dissolve copiously from their lakes, whereas basic dyes yield 
 only more or less discoloured solutions. In the second solution the basic 
 dyes in pigment lakes dissolve very freely, the acid dyes very slightly only, 
 if at all. 
 
 By evaporating these solutions on a watch-glass on the water bath, the 
 extracted dyes are obtained in a convenient form for further examination, 
 with a view to the further identification of the coal tar dyes present. 
 This is an operation, requiring, however, much experience of coal tar 
 colours and of their chemical behaviour. 
 
 The determination of carbon black (lamp black) or graphite (plumbago) 
 in rubber mixings, whether hot or cold vulcanised, is a matter of consider- 
 able difficulty. It is even more complicated when there should at the 
 same time happen to be present one of those peculiar brown pigments 
 known as vegetable brown (Van Dyke brown). That this latter pigment 
 should ever occur together with graphite is most unlikely, although there 
 is no telling what extraordinary combinations may here and there be used 
 in rubber mixings; but the above named class of brown pigments are 
 frequently used in combination with lamp black. But if, to crown all, 
 fibrous organic materials should be added to this complication, the possi- 
 bilities of chemical analysis are altogether at an end, and we have to fall 
 back largely upon optical methods. The conclusions drawn from the 
 latter require, above all, much experience gained more particularly in the 
 microscopic investigation of the mineral residue derived from rubber 
 samples specially produced with the constituents in question. 
 
 Even where lamp black is the only carbonaceous pigment present, its 
 detection is often a matter of no small difficulty, as in a large number of 
 rubber goods not more than 1 per cent, of it is used. Of course in these 
 cases the failure to detect it may be due to various causes connected with 
 the separation of the India rubber from the mineral matter. Where larger 
 quantities of carbon black are present, its detection offers no difficulty, pro- 
 vided the degree of vulcanisation of the sample is such as to render it 
 reasonably amenable to the nitrobenzene treatment. 
 
 1 Very commonly the presence of such lake pigments, especially of such prepared 
 with the basic dyes, already shows itself in the very pronounced coloration of the 
 acetone extract. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 279 
 
 In the majority of cases it will suffice to detect its presence, and to 
 form by microscopic examination a rough idea of its quantity. As by 
 this time the other mineral constituents are known, the less experienced 
 worker is advised to prepare an intimate mixture of the mineral con- 
 stituents present in their ascertained proportions, blend this mixture with 
 varying proportions of lamp black, and compare these blends with the 
 mineral residue under examination. Surprisingly accurate results are 
 obtainable by this method, the error in careful working amounting 
 generally to less than 0*5 per cent. This is, of course, largely due to the 
 enormously fine distribution (covering power) of carbon black, and its 
 very low specific gravity. 1 
 
 Attention must here be drawn to the fact that large quantities of car- 
 bonaceous matter are sometimes observed in the mineral residue of samples 
 which, in fact, do not contain a trace of carbon black (lamp black). To 
 this peculiar point I drew attention some years ago, and showed that all 
 rubber samples containing red lead, or highly vulcanised samples contain- 
 ing iron oxide in larger quantities, are apt to show this phenomenon, which 
 is due to the circumstance that at the high temperature of the nitrobenzene 
 or nitro-naphthalene treatment, red lead oxidises, or, rather, carbonises, the 
 india rubber. Ferric oxide does not act in this way in samples which are 
 easily resolved in the nitrobenzene treatment, but in highly vulcanised 
 samples requiring a more prolonged treatment the same action takes place, 
 though in a less pronounced degree than in the presence of red lead. 
 
 This matter is of such interest it shows so clearly, on the one hand, 
 the necessity for the extension of our knowledge of the chemical nature 
 and behaviour of india rubber, and, on the other, the care and circum- 
 spection required in the analysis of an india rubber article of unknown 
 composition as to warrant a discussion of one of the first cases in reference 
 to the above named phenomenon. 
 
 Two samples of the same make of steam packing, though not of the 
 same batch, had to be analysed. The first sample (a) had undergone vis- 
 cous decomposition, the second sample (b) represented the same article in 
 a perfectly sound condition. The analysis of these two samples produced 
 the following figures : 
 
 India rubber, . 
 Organic extract, 
 Mineral matter, 
 Carbonaceous matter, 
 Total sulphur, 
 
 a. 
 
 9 '85 per cent. 
 
 8-35 ,, 
 63-11 ,, 
 1679 
 
 1-89 
 
 12 '73 per cent. 
 
 2-82 
 62-81 ,, 
 19-53 ,, 
 
 2-10 
 
 1 HENRIQUES' suggestion to determine carbon black in rubber goods by combustion, 
 weighing the C0 2 obtained, is entirely unreliable, and quite worthless in the case of 
 combinations such as those above named. 
 
280 INDIA RUBBER AND ITS ANALYSIS. 
 
 On the face of it, this analysis appears extremely dubitable, if not clearly 
 impossible, for, looking at the normal sample (5), the compounding of 
 12*73 parts of rubber with nearly 63 parts of mineral matter and 20 parts 
 of lamp black is a practically impossible operation. Its accomplishment 
 would, at least, require the addition of a substantial proportion of mineral 
 oil or fatty oil, but of such the organic extract did not contain a trace. 
 On the other hand, the organic extract in b would correspond to a rubber 
 containing 18 per cent, of resin. Such a percentage only occurs in one or 
 two of the lowest brands of india rubber, and the preliminary physical 
 examination of the sample plainly indicated that a good medium grade 
 rubber had been employed in its manufacture. Such a rubber would not 
 contain above 6, or at most 7, per cent, of resin. This incongruity seemed 
 strongly to point to something abnormal in the analysis of the sample. 
 
 The analysis of the decomposed sample (a) showed a minimum of india 
 rubber, a considerable maximum of organic extract obviously due to the 
 decomposition of the india rubber, but also a minimum of carbonaceous 
 matter to an amount much too considerable to be accepted as falling within 
 the limits of probable experimental error. There is, moreover, a peculiar 
 parallelism between the amounts of rubber and carbonaceous matter in the 
 two analyses. 
 
 All these facts seemed to point to an identical cause of error in both 
 analyses, which only became intelligible on the assumption either that 
 there was no carbon present, or that rubber had disappeared in one place 
 to reappear in another as carbonaceous matter. This latter supposition 
 seemed perfectly feasible in view of the circumstance that nitrobenzene at 
 the temperature of its boiling point (210 C.) would exhibit a not incon- 
 siderable tendency to oxidise so highly oxidisable a body as india rubber. But 
 in this case reduction products of nitrobenzene, azobenzene, oxyazobenzene, 
 possibly even aniline, would have to be traceable in the nitrobenzene solution. 
 These were, however, entirely absent. There then seemed to remain only 
 the assumption of the rubber sample itself, containing an oxidising agent 
 which became active at the high temperature of boiling nitrobenzene. 
 It appeared, therefore, desirable to reduce the boiling point of the 
 nitrobenzene. This could be done by boiling under a partial vacuum, but 
 a simpler way suggested itself in the employment of a mixture of nitro- 
 benzene with about 5 per cent, of chloroform. Such a mixture shows on a 
 reflux condenser a practically constant boiling point of about 173 C. 
 
 With such a mixture the samples were re-examined, with the following 
 
 result : 
 
 a. &. 
 
 Mineral Matter, . . . 63 '33 per cent. 63 '02 per cent. 
 
 Not a trace of carbonaceous matter was now to be found. As it appeared 
 likely that the detailed analysis of the mineral matter obtained in both 
 these treatments respectively would furnish further information, it was 
 carried out on both samples by subjecting them to the pure nitrobenzene 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 28l 
 
 treatment, another part of the same samples to the corresponding treat- 
 ment with the above mentioned mixture with nitrobenzene and chloroform. 
 The result in all cases was the same as before : the nitrobenzene treatment 
 yielded a mixture of mineral and carbonaceous matter, the nitrobenzene 
 chloroform treatment yielded a residue of mineral matter only. The 
 determination of the mineral constituents in these four residues gave the 
 following results : 
 
 Treatment with Pure Nitrobenzene. 
 
 Sample a. 
 
 Sample b. 
 
 Carbonaceous Matter, .... 
 f CaC0 3 
 PbC0 3 \ 
 Mineral Matter, . . .4 PbO / 
 |PbS \ 
 LPbS0 4 j 
 
 Total Mineral Matter, .... 
 
 17 '22 per cent. 
 2078 
 
 40-27 
 2-10 
 
 19-93 percent. 
 21-04 
 
 41-83 
 1-00 
 
 3'15 per cent. 
 
 63-87 percent. 
 
 Treatment with a Mixture of Nitro- 
 benzene and Chloroform. 
 
 Sample a. 
 
 Sample b. 
 
 Carbonaceous Matter, .... 
 fCaC0 3 
 j PbC0 3 \ 
 Mineral Matter, . . ^ PbO 
 PbS 2 \ 
 I PbS0 4 / 
 
 Total Mineral Matter, : 
 
 Trace. 
 21 -02 per cent. 
 
 32-22 
 
 9-97 
 1-05 
 
 Nil. 
 21-11 percent. 
 
 31-56 
 10*05 ,, 
 0-88 
 
 64-26 per cent. 
 
 63 -60 per cent. 
 
 At a glance it will be observed that there is a considerable percentage 
 of lead peroxide in the mineral matter isolated by means of the nitro- 
 benzene-chloroform treatment. There is no trace of this peroxide to be 
 found in the mineral matter isolated by the nitrobenzene treatment, but 
 we have here a large amount of carbonaceous matter. The causal con- 
 nection between the carbonaceous matter observed in the first series, and 
 the lead dioxide found in the second series, is obvious, and justifies the 
 conclusion that this carbonaceous matter is a product of the oxidation of 
 the india rubber by the lead dioxide. The two analyses quoted above, 
 therefore, now assume this quite intelligible form : 
 
 
 a. 
 
 b. 
 
 India rubber, . 
 Organic extract, 
 Mineral matter, 
 Total sulphur, 
 
 26 '64 per cent. 
 8-35 ,, 
 63-11 
 1'89 
 
 32 '26 per cent. 
 2-82 
 62-81 ,, 
 2-10 
 
 Of course the lead peroxide found in these samples was not added to 
 the mixings as such, but, as is customary, in the form of red lead. 
 
282 INDIA RUBBER AND ITS ANALYSIS. 
 
 Quite recently I have observed an effect very similar to the one above 
 described in the analysis of highly vulcanised brake blocks heavily com- 
 pounded with ferric oxide. The nitrobenzene treatment only resolved the 
 sample very slowly, and the operation had to be greatly prolonged. At 
 first the treatment took its normal course, a bright brick-red solution 
 (suspension) forming. After about an hour's treatment a change of colour 
 was observed first into brown, which rapidly darkened into a deep black. 
 Finally, the mineral matter, together with a large quantity of carbonaceous 
 matter, was obtained ; the colour of the original sample at once precluded 
 the acceptance of this result as in accordance with the composition of the 
 sample. 
 
 These cases clearly show that the nitrobenzene treatment requires 
 careful and critical examination, especially if its completion demands 
 prolonged application. 
 
 VII. Sulphur and Chlorine in India Eubber. One of the most important 
 points in rubber analysis is the determination of the combined sulphur, or 
 of the combined sulphur and chlorine in the india rubber. The properties 
 of manufactured india rubber, its suitability for various uses, are much 
 more dependent upon a suitable vulcanisation than upon anything else at 
 any rate, a badly, insufficiently, or too highly vulcanised india rubber 
 article always represents a bad article. Here, however, we touch at once 
 the main difficulty of the vulcanisation problem, from the technical point 
 of view : a degree of vulcanisation quite unsuitable for a certain article 
 may represent the ideal vulcanisation condition of another article. Such a 
 thing as a standard degree of vulcanisation applying to all kinds, or even 
 to one range, of india rubber articles does not exist. Before long we shall 
 probably be in a position to define the minimum degree of vulcanisation 
 which should be aimed at, in order to obtain the most favourable physical 
 constants for each brand of india rubber combined with the highest degree 
 of chemical stability and inertia. But even these data will not reach their 
 full importance for the purpose of the interpretation of analytical results,, 
 until we may be able to discriminate analytically between the different 
 brands of india rubber. Not until then shall we be in a position to define 
 analytically the technical value of manufactured india rubber. 
 
 DoNATH 1 made an utterly premature attempt in this direction. He- 
 suggests that the value of an india rubber article depends upon the quantity 
 of india rubber it contains, and its ratio to the sulphur of vulcanisation, or 
 to the vulcanising metallic sulphides present. The larger this ratio, the- 
 larger he assumes the value of the article to be. Indeed, he finds it possible 
 to express the value W of every india rubber article by the simple formula r 
 
 W = 100 -(S + B) 
 
 S 
 
 in which S represents the sulphur present in the article, B the mineral and 
 
 organic admixtures. This would render the work of analytical interpreta- 
 
 1 Zeitschr.f. d. Chem. Ind., i. [1887], pp. 77, 110, 127. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 283 
 
 tion delightfully simple, but, unfortunately, this formula is entirely 
 worthless for the purpose for which it was devised. It is perfectly obvious 
 that, according to this formula, the value of an india rubber mixing before 
 and after vulcanisation would be exactly the same, and quite irrespective 
 of the degree of vulcanisation. It is only necessary to state this, to 
 demonstrate the absurdity of the formula. 
 
 The commercial value of manufactured india rubber is simply the sum 
 of the values of its constituents increased by the amount of the factory 
 charges labour, power, capital charges, rates, depreciation, and adminis- 
 tration expenses. The technical value of manufactured india rubber may 
 considerably exceed the commercial value according to the utility or 
 serviceability of the articles concerned. 
 
 Of all the constituents of manufactured india rubber, none possesses 
 such a paramount influence upon the general excellence of the product as 
 the sulphur of vulcanisation. We may be absolutely compelled, by the 
 exigencies of practical requirements or appearance, to use filling or 
 colouring materials i.e., to reduce the commercial value of certain articles 
 but their technical value remains thereby largely unaffected, provided the 
 degree of vulcanisation produced in them is such as to give them the 
 maximum utility or serviceability for which they are in demand. 
 
 The percentage of sulphur combined with the india rubber is, however, 
 no measure of its degree of vulcanisation. The following instance of the 
 analysis of the outer cover of a tire will make this clear : 
 
 India rubber, . . . . . .54*70 per cent. 
 
 Organic extract, . . . . . 1'34 ,, 
 
 Sulphur of vulcanisation, . . . . 1'99 ,, 
 
 Free sulphur, 2'88 ,, 
 
 Total mineral matter, 41 '08 ,, 
 
 99'99 percent. 
 
 Organic extract in rubber, . . . . 2*45 ,, 
 Total sulphur in rubber, . . . 8 '80 ,, 
 
 Coefficient of vulcanisation, . . . 3*63 ,, 
 
 Here the sulphur of vulcanisation amounting to 1'99 per cent, gives 
 us no direct idea of the degree of vulcanisation of the india rubber. This 
 we only obtain by calculating the percentage ratio between the amount of 
 india rubber (54 '70) and sulphur of vulcanisation (1'99) present, and we 
 obtain in this manner the degree or coefficient of vulcanisation (3'63 per 
 cent.). To obtain a comprehensive view of the distribution of the whole 
 of the sulphur in regard to the rubber, I find it of much value to calculate 
 similarly the percentage ratio between rubber and total sulphur (sulphur 
 of vulcanisation + free sulphur) present. By further applying the same 
 treatment to the organic extract, we obtain another very important ratio 
 organic extract in rubber which, in the case of the above sample, enables 
 us to draw the conclusion that the india rubber present is pure Para, as 
 no other rubber would yield such a low percentage of organic extract. 
 
284 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 In samples containing no mineral matter, or in samples the mineral 
 matter of which is free from sulphur, the sulphur of vulcanisation can be 
 estimated directly. If the mineral matter contains sulphur, the indirect 
 estimation described on page 254 has to be resorted to. 
 
 In the absence of positive vulcanisation standards, the following con- 
 siderations must guide us in judging the coefficient of vulcanisation : As 
 has been pointed out in the first section of this book, with increase of 
 the coefficient of vulcanisation come increase tensile strength, elasticity, 
 and distensibility. This, however, is only true up to a certain point, 
 beyond which there is, first, a rather rapid decrease in elasticity and 
 distensibility, followed soon by a decrease of tensile strength. The follow- 
 ing diagram referring to four tire covers of absolutely identical composi- 
 tion as regards the quality and proportions of Para rubber, total sulphur, and 
 mineral matter, but cured under different conditions, renders this evident. 
 
 . inches Elongation 
 
 6, 1 U , , , , 
 
 m 
 
 10 20 30 4-0 50 60 70 80/bS 
 
 FIG. 26. Diagram : Tensile strength of differently vulcanised specimens of india rubber. 
 
 The vulcanisation coefficients of these four samples are as follows : 
 
 Sample I. , 
 ,, II., 
 HI., 
 
 1*78 percent. 
 2-14 
 2-87 ,, 
 4-44 
 
 Of these, I. and II. are obviously undervulcanised, though II. but 
 slightly; III. is very satisfactorily vulcanised; IV. clearly overvulcanised. 
 It will be seen in every case that the higher the coefficient of vulcanisa- 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 285 
 
 tion, the higher also the tensile strength for the same distension. This is 
 still true of the overvulcanised sample IV., which, however, already breaks 
 at a distension of 1'25 inches, whereas the two undervulcanised samples, 
 more particularly I., show an inordinate amount of distension under very 
 low stresses. These diagrams may be considered as quite typical for 
 articles known or alleged to be manufactured from Para rubber, but, of 
 course, other brands of rubber would show different results, although with 
 every brand of rubber the curves of over- or under-vulcanisation would 
 show analogous differences from the curve of correct vulcanisation. Much 
 work remains to be done in this direction. 
 
 Overvulcanisation must, however, be clearly distinguished from a high 
 degree, or high coefficient, of vulcanisation, which, of course, is in many 
 articles quite a necessity. Overvulcanisation, in accordance with the 
 results I arrived at regarding the nature of vulcanisation in the first part 
 of this book, does not consist simply in a high coefficient of vulcanisation, 
 but in one obtained under unsuitable conditions, either of temperature, or 
 of time, or of quantity of sulphur. Overvulcanisation may therefore be due 
 to too high a temperature, too long a time, or an unsuitable proportion of 
 sulphur. Of these causes, the one first named is by far the most common, 
 while too short a time is the most usual cause of undervulcanisation. 
 
 A high degree of vulcanisation combined with low tensile strength, but 
 fair distensibility, is an almost certain indication that the rubber has been 
 ' overworked ' upon the rollers. If, at the same time, the distensibility is 
 low, it is, as far as my experience goes, invariably due to the presence of 
 larger proportions of recovered rubber. 
 
 It is evident from the above that undervulcanisation is much more 
 clearly defined than Overvulcanisation. This is rather important, as under- 
 vulcanised articles are much more prone to 'decomposition' than over- 
 vulcanised ones. Indeed, I have never yet seen a case of ' decomposition ' 
 due with certainty to Overvulcanisation. 
 
 Both these faults are much accentuated in goods containing large 
 amounts of mineral matter. The latter always greatly increases the 
 porosity of rubber goods, and therefore also their liability to decay. 
 
 Undervulcanised goods containing large proportions of fatty substi- 
 tutes very rapidly undergo decomposition ; overvulcanised goods under the 
 same circumstances suffer very much less. 
 
 If there is chlorine in the india rubber, its presence may be due either 
 to the india rubber having been cold vulcanised with chloride of sulphur, 
 or to the mixing containing fatty chlorosulphide substitutes. In the case 
 of articles the thickness of which exceeds a few millimetres, cold vulcanisa- 
 tion is out of the question ; but under all circumstances the satisfactory 
 interpretation of the significance of the presence of chlorine in the india 
 rubber is possible on the following considerations. 
 
 In articles free from fatty substitutes, the presence of chlorine in the 
 india rubber is a certain proof of the vulcanisation having been effected 
 
286 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 by the cold process. ' Rubber enamelled ' goods should, however, never be 
 analysed in this connection simultaneously with their enamel coating, but 
 the latter should be carefully pared off. The reason for this is that 
 rubber goods (enemas, etc.), the body of which has been hot vulcanised, 
 often receive a ' rubber enamel ' surface coating, generally coloured, which 
 is then cured by the ordinary cold process, or by the ' acidising ' process 
 (hypochlorous acid). For this reason an article should not be pronounced 
 as c cold cured ' unless the * sulphur of vulcanisation ' present approximates 
 very closely in amount to the 'chlorine of vulcanisation.' 
 
 In articles containing fatty substitutes, the correct interpretation of 
 the analytical results obtained is not so simple. The presence in the 
 rubber of sulphur and chlorine in approximately equimolecular proportions 
 is conclusive evidence of cold vulcanisation only if the alcoholic extract is 
 free from chlorine, or, in other words, if there are no chlorosulphide sub- 
 stitutes present, as it is quite conceivable that in the presence of the latter 
 the amount of hydrochloric acid evolved during hot vulcanisation, and 
 absorbed by the india rubber, might coincidentally be equivalent to the 
 coefficient of vulcanisation of the rubber. This point must not be lost 
 sight of when pronouncing a specimen of rubber as cold vulcanised. The 
 mere fact that it contains chlorine is no proof to that effect, and the over- 
 looking of this point has often led to gross mistakes being made. 
 
 Indeed, the only satisfactory criterion which enables us always to 
 decide with certainty upon the question under discussion is that obtained 
 from the amount of free sulphur as ascertained in the 'acetone extract.' In 
 cold vulcanised goods free sulphur is rarely absent, but its quantity is 
 always very small, rarely exceeding 1 per cent. On the other hand, hot 
 vulcanised goods containing chlorosulphide substitutes always contain 
 more or less considerable quantities of free sulphur, for reasons connected 
 with the composition of the mixings required to ensure a thorough vul- 
 canisation. If, therefore, there is any doubt respecting the vulcanisation 
 of a given sample of manufactured rubber, the amount of free sulphur 
 ascertained in the sample readily supplies the required answer. 
 
 The following typical analyses will make this clear : 
 
 
 Tobacco Pouch. 
 
 . Garden Hose. 
 
 India rubber, 
 
 
 
 50 '22 per cent. 
 
 31'29 per cent. 
 
 Organic extract, 
 
 
 
 4-88 
 
 
 7-34 
 
 
 Fatty substitute, 
 
 
 
 37-21 
 
 
 28-90 
 
 
 Mineral matter, 
 
 
 
 219 
 
 
 26-28 
 
 
 Free sulphur, 
 
 
 
 0-27 
 
 
 1-83 
 
 
 Sulphur of vulcanisation, 
 
 
 
 272 
 
 
 2-15 
 
 
 Chlorine in rubber, 
 
 
 
 2-50 
 
 
 2-20 
 
 
 Free sulphur in rubber, 
 Coefficient of vulcanisation, 
 
 
 
 0'53 per 
 5-41 
 
 cent. 
 
 5 -84 per 
 6-87 
 
 cent. 
 
 ,, chlorination, 
 
 
 
 4-97 
 
 
 7-03 
 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 287 
 
 The analytical data l above quoted at first sight appear to refer, if not 
 to two similar qualities of rubber goods, at any rate to two similar types, 
 and the inexperienced analyst might well feel tempted to interpret both of 
 them as cold vulcanised articles. Even the coefficients of vulcanisation 
 and chlorination taken conjointly would appear to be well in keeping with 
 this interpretation. But the 'free sulphur in india rubber,' while support- 
 ing this assumption entirely in the case of the tobacco pouch, renders it clearly 
 untenable in the case of the garden hose. Such an amount of free 
 sulphur can be present only if sulphur, as such, has been added to the 
 mixing for the purpose of hot vulcanising. The garden hose must there- 
 fore be pronounced a hot vulcanised article, and from the presence of 
 chlorine in the india rubber, it follows that the fatty substitute is a chloro- 
 sulphide substitute. This conclusion, it may be remarked, in the above 
 case was fully confirmed by the detailed analysis of the 'saponification 
 extract.' 
 
 There remains to be mentioned the not very common case of rather 
 small quantities of chlorine, rarely exceeding 1*5 per cent., occurring in 
 samples which either contain no substitutes at all, or no chlorosulphide 
 substitutes, while, at the same time, the combined sulphur in the rubber 
 greatly exceeds the amount of chlorine present. The interpretation of 
 this observation is that the vulcanisation of such a sample is essentially a 
 hot vulcanisation, but that, for reasons connected with the technical 
 conditions of the vulcanising process, the articles have been subjected to 
 a ' surface hardening ' produced by 'semi-curing' them i.e., treating them 
 with a weak solution of sulphur monochloride. The surfaces thus 
 hardened are in this manner protected against mechanical injury. 
 
 Of course in hot vulcanised goods containing chlorosulphide substitutes 
 this treatment could not be discovered. This inability is, however, of 
 little moment, as the treatment in question is merely a technical artifice, 
 and has no perceptible influence upon the physical constants of the articles 
 in question. 
 
 UNUSUAL CONSTITUENTS OP RUBBER GOODS. 
 
 An exhaustive description of all the substances which are, or have 
 occasionally been, introduced into rubber mixings is almost an impossibility. 
 There is no substance which lends itself so readily to admixture with the 
 most heterogeneous substances imaginable, and this quality leads occasion- 
 ally to absurdities in this respect which surpass similar abuses in any other 
 technical material known. 
 
 In a steam packing which enjoys a not inconsiderable popularity, I found 
 considerable quantities of brass filings and coke powder. A specially 
 'cheap' bicycle tire contained, besides a very generous allowance of fatty 
 
 1 I have purposely omitted the analytical composition of the fatty substitutes, so as 
 to render more clear the point at issue. 
 
288 INDIA RUBBER AND ITS ANALYSIS. 
 
 substitute, 10 per cent, of cork dust. Quite a number of supposed high 
 class tires on the market at present contain from 10 to 15 per cent, of 
 paraffin wax. This, it is alleged, will prevent ' suncracking,' but however 
 that may be, it certainly produces a poor tire. A 'metallic' rubber 
 packing contained 35 per cent, of metallic zinc (zinc dust). In various 
 other goods I have detected caseine, caseiue lime, cork dust, large quanti- 
 ties of tannic acid, glass powder, alum, cholesterine (wool grease), bone 
 ash, mica, and orpiment. The last named substance, arsenic trisulphide, 
 was found to the amount of 6*5 per cent, in a specimen of waterproof 
 cloth for ladies' wear. It is impossible to assume that any manufacturer 
 would knowingly employ such a substance, and I surmise that he was in- 
 duced to buy it under the guise of some innocently sounding name. 
 Indeed, within my knowledge this substance has been, and is still, offered 
 under the very inoffensive description of ' yellow antimony.' 
 
 (e) Defects in India Rubber Goods. The most usual defects occurring 
 in manufactured india rubber may be arranged under the following 
 
 heads : 
 
 a. Defects in the rubber mixings. 
 
 b. Defects due to faulty vulcanisation. 
 
 c. Defects due to faults of the constructive materials. 
 
 d. Defects arising in use. 
 
 e. Defects in appearance. 
 
 a. Defects in the Rubber Mixings. A rubber mixing must be con- 
 sidered as defective or faulty if either its composition or its physical con- 
 dition are such as to preclude its vulcanisation into a physically and 
 chemically sound article. Most commonly this defect arises from the mix- 
 ing having, as a mechanical operation, been carried out negligently. It is, 
 of course, often extremely difficult to produce a homogeneous mixture of 
 india rubber with pulverulent admixtures. In many cases this could only 
 be done by working the mixture between the rollers for such an inordinate 
 length of time as to produce in the india rubber the often fatal phenomenon 
 of fatigue (see page 93) generally described in the trade as 'killing the 
 rubber.' But if we find in different parts of a sample of manufactured 
 rubber a considerable variation in the several mineral, or, more generally 
 speaking, non-rubber, constituents, it is obvious that the mixing in respect 
 of these non-rubber admixtures has been imperfectly carried out. For 
 this defect there is no excuse, and, at the same time, it will be readily under- 
 stood it is highly detrimental to the production of a sound article. Very 
 often this defect is the cause of premature decay in rubber goods, and the 
 fact that it is so very common tends to show how little its significance and 
 consequences are realised. 
 
 The evils arising from a too prolonged working of a rubber mixing, 
 alluded to above as fatigue, although their true nature has up to now 
 been entirely misunderstood, are well known to every experienced rubber 
 worker, and consequently this defect, at least in a pronounced form, is 
 comparatively rarely met with. From what has been said in reference to 
 
ANALYSIS OP INDIA RUBBER ARTICLES. 289 
 
 this subject on page 94, it will be readily understood that this defect is. 
 characterised by well-marked physical signs of under vulcanisation in the 
 presence of a percentage of combined sulphur in the india rubber suggest- 
 ing a thorough degree of vulcanisation. Manufactured india rubber 
 showing these characteristics is extremely prone to premature decay. 
 
 Not very common is the fault arising from the use of slightly damp 
 powders in india rubber mixings. Under the influence of the heat and 
 pressure of the rollers, they form minute hard flakes, which remain 
 embedded in this state in the india rubber. In articles in which a perfect 
 surface is required, these flakes are, of course, highly objectionable, even 
 more so in sheets or strips for rubber insulated cables, in which case they 
 very generally result in ' weak spots,' often in short circuits. 
 
 b. Defects due to Faulty Vulcanisation. As the result of the vulcanisa- 
 tion process is determined by the proportions of sulphur and rubber used^ 
 the temperature employed, and the time allowed, but further also by the 
 physical condition of the india rubber, it is evident that with such a 
 number of co-determinants, the number of possible defects must be very 
 large the more so, as very few brands of india rubber behave similarly, or 
 yield similar results, even when treated under equal vulcanising conditions. 
 It is therefore not surprising to find that probably more than one half of 
 all the defects observed in india rubber goods arise from some defect in 
 its vulcanisation. A very large number of these defects may, moreover^ 
 be described as either over- or under- vulcanisation, and be identified, on 
 the one hand, by the vulcanisation coefficient ; on the other hand, by the 
 tensile strength and elongation tests (see diagrams, fig. 284). To exactly 
 describe the more particular cause of either of the two above named 
 defects is often very difficult, but in many cases a careful study of the 
 coefficient of vulcanisation, the total sulphur in rubber, and of the 
 mechanical tests (tensile strength and elongation) will give valuable 
 indications. In this respect it must be remembered that overvulcanisation 
 in the presence of a large amount of total sulphur is due either to too 
 high a temperature or to too long a time, but in the presence of a 
 moderate or small amount of total sulphur, the fault almost invariably 
 lies in the employment of too long a time. Undervulcanisation in the 
 presence of a large amount of sulphur is due either to too low a tempera- 
 ture or to too short a time ; in the presence of a moderate or small amount 
 of total sulphur, either this amount of sulphur itself is too small, or the 
 two before named causes concur. Somewhat similar in its effect to over- 
 vulcanisation is the very curious defect arising from vulcanisation at too 
 high a temperature or too long a time, in the presence of too little 
 sulphur. We obtain, then, the peculiar result of a ' rotten ' vulcanisation 
 product possessing all the characteristics of overvulcanisation, but differing 
 from it by a low coefficient of vulcanisation. On the other hand, under- 
 vulcanisation is often suggested by the physical constants of samples, 
 which, on analysis, are found to be fully vulcanised with a full allowance of 
 
 19 
 
290 INDIA RUBBER AND ITS ANALYSIS. 
 
 total sulphur. This is very frequently an indication of the phenomenon 
 of 'fatigue,' but by no means always. 1 The physical effect due to the 
 same vulcanisation coefficient is apt to vary considerably in different brands 
 of india rubber, and a percentage of combined sulphur which in Para 
 would represent a very full degree of vulcanisation, would leave Beni 
 River or Manao rubber rather imdervulcanised. 
 
 Turning to the defects arising from faults or faulty conditions in the 
 cold process of vulcanisation, we find that the diagnosis of over- or under- 
 vulcanisation is very much simpler in some respects, but complicated by 
 the fact that some brands of india rubber form very unstable vulcanisa- 
 tion products when vulcanised with sulphur monochloride ; and further, 
 that sulphur monochloride being a very reactive body, is apt to cause 
 secondary reactions, and act in various ways not intended in its employ- 
 ment. 
 
 Cold cured goods waterproof fabrics, tobacco pouches, gas tubing, 
 and similar articles are in less need of tensile strength than of softness, 
 pliability, and chemical stability. The behaviour of cold cured goods 
 exposed to the heat of a water oven for from 8 to 10 hours furnished much 
 information on this point. In the case of undervulcanisation, the articles 
 become more or less rapidly very limp, and occasionally almost plastic 
 under the influence of the heat. Moistening them with benzene, and 
 continuing the heating for a few minutes, renders them distinctly sticky. 
 In this test overvulcanisation results in the article becoming either 
 very stiff, sometimes almost brittle, or else very tender and lifeless. The 
 latter is generally observed in the case of goods manufactured from india 
 rubber unsuited for cold vulcanisation. 
 
 If effective precautions are not taken to prevent the formation of dew 
 upon the articles during the process of cold vulcanisation, part of the 
 sulphur monochloride is decomposed before it is able to combine with the 
 india rubber. 2 The hydrochloric acid thus liberated is at the root of many 
 of the defects in cold cured goods. If these contain zinc oxide or ferric 
 oxide, any hydrochloric acid produced may lead to the formation of zinc 
 chloride or ferric chloride respectively, both of which are absolutely fatal 
 to cold cured goods. 
 
 But even if no formation of dew and consequent decomposition of 
 sulphur monochloride takes place, the interaction of the latter with 
 certain other substances which may form part of the mixing will often 
 produce disastrous results. In this respect copper and manganese 
 compounds and red lead must be mentioned. The first two occur not 
 infrequently as impurities in certain pigments and compounding materials 
 
 1 This phenomenon can easily be studied by exposing a rubber article containing a 
 moderate amount of total sulphur for one or two days to a temperature of about 90 C. 
 in a water oven. The outer covers of bicycle tires, stripped of their fabrics, are very 
 suitable for this experiment. 
 
 2 The formation of this dew is caused by the considerable fall of temperature produced 
 by the rapid evaporation of the carbon bisulphide. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 291 
 
 used in rubber mixings, while red lead is often deliberately added; under 
 the assumption, no doubt, that what is, under certain circumstances, a 
 useful constituent in a mixing for hot vulcanisation, must also act 
 beneficially in the cold process of vulcanising. 
 
 c. Defects due to Faults of the Constructive Materials. The only kinds 
 of constructive materials to be considered are the various textile materials 
 used either for the strengthening of rubber structures or for the pro- 
 duction of waterproof fabrics. 
 
 In undyed textile materials the only defect likely to occur is loss of 
 strength during the vulcanising process. This in hot cured goods is un- 
 avoidable, especially if the fabrics are cotton fabrics. The tendering in 
 this case is, of course, due to the action of the vulcanising heat (say 130 C,) 
 for a more or less considerable time upon the cotton, which at tempera- 
 tures above 80 C. becomes increasingly sensitive to heat. If the loss in 
 tensile strength caused by the vulcanisation process exceeds 15 per cent., 
 the textile material in question may be considered to be at fa-ult. On the^ 
 other hand, there should be no tendering of the cloth in cold cured goods, 
 and its occurrence is in such articles due simply to some fault in the 
 vulcanisation. Either the chloride of sulphur is of unsuitable quality, or 
 the vulcanisation has been carried out under conditions favouring the 
 deposition of dew upon the rubber coating. The hydrochloric acid thus 
 formed, coming subsequently in contact with the face of the fabric, then 
 attacks the cotton fibre. The defect in this case is therefore entirely the 
 fault of the rubber manufacturer. If, however, the selvage of the rubber- 
 coated fabric is essentially stouter than its body, it often happens that this 
 selvage is not covered by the india rubber coating. On cold curing such 
 cloth, the selvages are liable to take up some of the vulcanising solution, 
 with the result that the cloth along the edges contains chloride of sulphur, 
 the decomposition of which results in the formation of hydrochloric acid, 
 which then destroys the fibre. This effect often spreads for a considerable 
 distance from the edges to the centre of the cloth, so that, on superficial 
 examination, it may sometimes appear as if the cloth were tendered all 
 over. But whenever this tendering is due to an absorption of the vulcan- 
 ising solution by the selvage, it will be found that the central part of the 
 fabric shows either no tendering at all, or a very slight degree of it only, 
 with a very marked increase of it towards the edges. 
 
 In dyed waterproof fabrics (macintosh cloth) tendering is much more 
 frequent. It occurs but rarely in cold cured goods, and then always 
 through one of the two above stated causes most generally through badly 
 carried out vulcanisation. In hot cured goods it is rather common. It is 
 seldom due to the inferior quality of the yarns used in the manufacture of 
 the textile fabrics, but arises almost invariably from the presence in the 
 cloth of small quantities of sulphuric acid, or from the employment of 
 mordants which at vulcanising temperatures attack the textile fibres. The 
 presence of sulphuric acid in these cases is obviously due to the imperfect 
 
INDIA RUBBER AND ITS ANALYSIS. 
 
 washing of the cloth after dyeing. This is confirmed by the fact that the 
 dyed fabrics showing excessive tendering in the vulcanisation process, due 
 to the presence of free sulphuric acid, are woollens or unions. 1 The 
 amount of sulphuric acid thus present very rarely exceeds 0*25 per cent. r 
 an amount which appears to be quite unobjectionable while the fabrics 
 contain their normal proportion of atmospheric moisture. In the process 
 of vulcanisation, however, this moisture, which serves as a diluent for the 
 sulphuric acid, disappears, and the acid rapidly attains a concentration 
 fatal to the textile fibre. This is naturally, therefore, most pronounced in 
 the case of unions, and results in extreme cases in the almost complete 
 destruction (carbonisation) of the cotton warps. 
 
 Among mordants having such a tendering effect, ferric oxide is to be 
 particularly mentioned. At vulcanising temperature this substance affects 
 cotton fibre materials very energetically. Wool seems to suffer much less. 
 Whether the enormous tendering of heavily loaded black silk is to be 
 ascribed to the tin and iron mordants used, I have not sufficient experi- 
 mental data to decide, but the probabilities are in favour of the 
 assumption. 
 
 d. Defects arising in Use. The discussion of all the defects coming 
 under this head would, in itself, fill a fair-sized volume. No doubt the- 
 composition of india rubber articles of every description is, at times, such 
 as to render them more or less unsuited for their purpose, or, at least, to 
 favour the development of certain defects ; but I am not sure whether 
 cases, in which the appearance of defects is not so much the fault of the 
 rubber manufacturer as of the consumer, are not much more numerous. 
 There is, to begin with, the fatal tendency on the part of the consumer 
 to buy increasingly cheaper grades without the slightest consideration of 
 the market prices of crude india rubber, and there is, moreover, the un- 
 conscionable treatment to which rubber goods are often subjected. Indeed, 
 it is very doubtful whether there is any other class of goods which is so 
 habitually maltreated as rubber goods, from which a great number of 
 people appear to expect the properties of almost all the other technical 
 raw materials combined. 
 
 Defects arising from use alone may be due to sins of omission or 
 commission in respect of the constituents of the mixings, or to unsuitable 
 vulcanisation. That is to say, considered from general points of view, a 
 sample of manufactured rubber may in every respect appear a thoroughly 
 satisfactory production, and nevertheless prove unsuited to the special 
 requirements and conditions of its practical use. In such cases the defect 
 obviously arises from the circumstance that in its composition and manu- 
 facturing treatment insufficient attention was paid to the chemical and 
 physical influences to which it would be exposed in practical use, or else 
 
 1 In the dyeing of cotton fabrics, sulphuric acid is only very rarely used. If used, it 
 is easily removed by washing, whereas it clings pertinaciously to wool. This, no doubt, 
 is due to -functional chemical differences between vegetable and animal fibres. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 293 
 
 the consumer himself uses an article manufactured to meet other and 
 different requirements. 
 
 The defects in question may be caused by the following chemical 
 agencies : 
 
 1. Acids. The action of the various acids upon manufactured india 
 rubber varies considerably according to the nature of the acid, and, for 
 the same acid, with its concentration, and with the temperature at which 
 it is applied. The latter point is rather important, as will be seen on 
 examining the first of HEINZBRLING and PAHL'S tables given on page 165. 
 From the figures there given for the action of sulphuric and acetic acid 
 respectively, one might feel justified in drawing the conclusion that the 
 action of sulphuric acid upon manufactured india rubber invariably pro- 
 duced a decrease in weight, the action of acetic acid almost as invariably 
 an increase in weight, in the test piece. But this conclusion would be 
 entirely erroneous. The above named authors, to begin with, operated with 
 concentrations of these acids (27 '5 per cent. H S0 4 , and 46 '5 per cent. 
 C 2 H 4 2 ) which cannot be said to be fairly representative of the conditions 
 most frequently encountered in practice. They altogether omitted the 
 investigation of the simultaneous effect of acids and temperatures up to 
 the boiling point of water. In my opinion, this work should be repeated 
 with acids not exceeding 10 per cent, in strength at the temperatures of 
 say 20, 60, and 90 C. The results of such a series would bear much more 
 clearly and surely upon the behaviour of manufactured india rubber under 
 practical conditions. My own observations leave no doubt that the action 
 of a 10 per cent, sulphuric acid upon india rubber, both at ordinary and at 
 higher temperatures, is much more marked in the vast majority of cases 
 than the prolonged action of 27 per cent, acid in the cold. Treatment of 
 rubber samples, similar in composition to those experimented upon by the 
 above named authors, with a 10 per cent, strong acetic acid has very 
 little effect in the cold, but acts most energetically at temperatures above 
 50 C., the effect, however, being not an increase in the weight of the 
 sample, 1 but a more or less considerable decrease. 
 
 Articles manufactured from sound india rubber and sulphur alone 
 withstand the action of 10 per cent, sulphuric acid even at boiling 
 temperature for very long periods, provided they are reasonably treated as 
 regards mechanical wear. But goods of this high quality are not the rule, 
 the majority of consumers preferring those of a low first cost, and these 
 can only be supplied by resorting to 'mixings' containing more or less 
 considerable quantities of mineral matter, and it is in this class of goods 
 that various defects make their appearance on exposing them under 
 practical conditions to the action of acids. 
 
 It is therefore these mineral matters which are primarily responsible 
 'for the appearance of defects in rubber articles exposed to the action of 
 dilute acids, such as sulphuric, hydrochloric, or acetic acid ; and it is obvious 
 1 See the extraordinary figures obtained by HEINZERLING and PAHL. 
 
294 INDIA RUBBER AND ITS ANALYSIS. 
 
 that if the mixings contain any substances easily acted upon by these acids, 
 a more or less rapid disintegration is the unavoidable result. It might be 
 thought, however, that these acids could at most affect merely the surface 
 of the rubber articles, as this substance is generally considered the water- 
 proof substance par excellence. This is another of those deep-rooted errors 
 which it is almost impossible to destroy. There are, indeed, scores of 
 substances which, as mere waterproofing agents, are infinitely superior to 
 india rubber, and which absorb scarcely appreciable quantities of water on 
 prolonged immersion, whereas india rubber under the same conditions may 
 absorb as much as 25 per cent, of its weight. This absorption is greatly 
 facilitated by an increasing porosity of the india rubber, and as the porosity 
 of india rubber unavoidably increases with the proportion of mineral matter 
 which is worked into it, the action of the acids is by no means confined to 
 the surface only of the articles. If, then, this mineral matter is readily 
 acted upon by acids, the primary nature of their action upon rubber 
 articles is perfectly obvious. This action in its simplest form is simply a 
 dissolving action, but in many cases the products formed by it may exert a 
 secondary action upon the india rubber often more destructive than the 
 first. 
 
 Zinc white, Paris white (CaC0 3 ), and lithopone are the greatest 
 offenders in this respect, and they are also used more frequently in rubber 
 mixings, and in greater quantities, than any other mineral substance. 
 
 Dilute sulphuric acid will dissolve out the zinc white and the zinc 
 sulphide of the lithopone in the form of zinc sulphate, thus converting the 
 rubber into a finely porous mass, and I look upon this porosity (sponginess) 
 as the immediate cause of the mischief. It not only greatly reduces the 
 resistance of the remaining india rubber to friction and mechanical wear 
 and tear generally, but it lays it open to the influence of oxidation in a 
 peculiarly effective manner. Whether it is merely the enormous increase 
 of surface which renders india rubber in this state such a certain prey to 
 oxidation, I am not prepared to insist, but it is an everyday observation that 
 india rubber rendered porous in this manner undergoes oxidation at an 
 extraordinarily rapid rate. 1 
 
 1 The rate at which a sample of manufactured india rubber, when deprived of the free 
 sulphur which is distributed throughout its mass, will undergo oxidation is perfectly 
 astounding. This observation, that manufactured india rubber, after extraction with 
 acetone or carbon bisulphide, is extremely oxidisable, led HEXRIQUES to assume that 
 free sulphur was not only not objectionable in manufactured india rubber, but acted, 
 indeed, as a preservative of it. This assumption I consider entirely erroneous, and I 
 hold that this phenomenon is simply due to the finely porous condition of the india 
 rubber caused by the removal of the free sulphur. This view is confirmed by the 
 circumstance that porosity produced by the removal of any other body present in a state 
 of fine distribution has exactly the same effect. It is also confirmed by the fact that 
 goods, which have been vulcanised in such a way as to contain only just appreciable 
 traces of free sulphur, show for that reason no particular inclination to oxidation ; and it 
 is further supported by the observation, much utilised in practical working, that the 
 admixture to india rubber of substances counteracting incipient porosity by filling up the 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 295 
 
 The action of dilute sulphuric acid (10 per cent.) upon india rubber 
 containing Paris white leads to the gradual transformation of this 
 substance into calcium sulphate CaS0 4 ,2H 2 0. This reaction involves a 
 considerable increase on the bulk of the original calcium carbonate present, 
 and it produces, therefore, a correspondingly considerable internal strain in 
 the india rubber, which renders it apparently hard and inelastic, and 
 reduces considerably its resistance to wear and tear. 
 
 Dilute hydrochloric acid (5 to 10 per cent.) dissolves the zinc white as 
 well as the Paris white and the zinc sulphide of the lithopone, and its 
 action is distinctly more rapid than that of sulphuric acid. It is the zinc 
 oxide and zinc sulphide which, in the case of this acid, are most objection- 
 able on account of the formation of zinc chloride, which has a curiously 
 destructive action upon india rubber. 1 Concentrated hydrochloric acid 
 (from 12 per cent, upwards) induces a more or less rapid destruction of the 
 india rubber by converting it into polyprene hydrochloride. 
 
 Dilute acetic acid dissolves the zinc white and Paris white, but is with- 
 out effect upon lithopone. The latter may therefore safely be used in the 
 manufacture of rubber goods required to withstand the action of this acid. 
 Concentrated acetic acid dissolves the zinc white and Paris white rather 
 more slowly than the dilute acid, but it shows the curious effect of inducing 
 a considerable swelling of the india rubber, especially if the latter happens 
 to possess a low coefficient of vulcanisation. This is curious in view of the 
 fact that acetic acid has neither a solvent action nor even a perceptible 
 swelling action upon unvulcanised india rubber. 
 
 The above named acids are those to the action of which manufactured 
 india rubber in its technical use is most frequently exposed, and it will have 
 been noticed that any such detrimental action of these acids is due not so 
 much to their direct action upon the india rubber, as to their action on 
 one or other of the mineral matters with which it has been compounded. 
 There are, however, some acids, the action of which is characterised by the 
 circumstance that it affects the india rubber itself rather than the mineral 
 matter present. The acids of this description most likely to be encountered 
 by india rubber articles are chromic acid (including bichromates) and 
 nitric acid. Both rapidly destroy soft rubber goods of every description, 
 nor are any means known of reducing such action to a satisfactory 
 
 pores as soon as formed invariably reduces its liability to oxidation. I look upon the 
 admixture of tar, mineral oils, paraffin wax, and similar bodies as substances performing 
 this valuable service. 
 
 In this connection it is also worth pointing out that finely ground (vulcanised) 
 rubber waste, which has been boiled with some alkaline solution so as to deprive it of 
 its free sulphur, and which is subsequently dried in a vacuum stove, on being removed 
 from the latter, suffers oxidation at such a rate that, after a few minutes' exposure to the 
 atmosphere, the whole mass reaches a temperature often exceeding 250 C. If this 
 reaction is not checked in time, the rubber bursts into flames. 
 
 1 I am inclined to believe that this action consists in an intramolecular rearrange- 
 ment of the polyprene complex, more particularly in the commutation of the olefinic 
 polyprene into a cyclical polyterpene. 
 
296 INDIA RUBBER AND ITS ANALYSIS. 
 
 minimum, still less to prevent it. Hard rubber goods are extremely 
 slightly affected by these acids, evidently owing to the fact that vulcanite 
 constitutes a saturated compound. 
 
 It is only to be expected that solutions of oxidising agents in general 
 affect india rubber in much the same way as the two above named acids. 
 This action is very marked in the case of solutions of bleaching powder, 
 ferric salts, nickel salts, cupric salts, and vanadium salts. 
 
 2. Alkalies. Dilute solutions of the alkalies have very little action 
 upon manufactured india rubber in the cold. At higher temperatures and 
 concentrations they have a tendency to considerably soften india rubber 
 possessing a low coefficient of vulcanisation, and still more so india rubber 
 containing much free sulphur. India rubber thus softened tends to decay 
 very rapidly. 
 
 If the india rubber contains mineral matters which are chemically 
 affected by alkalies, such as zinc oxide and, to a lesser degree, litharge, the 
 articles very generally suffer disintegration, which rapidly proceeds from the 
 surface inwards. 
 
 As all the fatty substitutes are saponifiable by alkalies, ammonia 
 excepted, it is obvious that rubber goods coming into contact, especially 
 with hot alkalies, should not contain that admixture. For the same 
 reason such goods should contain no saponifiable resins. In this respect it 
 should be remembered that some brands of india rubber contain saponifiable 
 resins as a natural impurity. 
 
 3. Organic Compounds. Generally speaking, all organic substances 
 which are readily absorbed by india rubber tend to deteriorate it, 
 especially at higher temperatures. Fatty, mineral, and resin oils are 
 especially conspicuous in this respect. That these exert any chemical 
 action upon the india rubber is very improbable, but they all soften it, 
 and thereby render it more liable to oxidation. Aniline and similar 
 organic bases act in the same manner, but the surprisingly destructive 
 action upon india rubber of some of them notably hydrazine and its 
 organic homologues strongly suggests a direct reaction between them 
 and the rubber substance. The action of all these bodies decreases with 
 an increase of the coefficient of vulcanisation, and the latter, therefore, 
 should be carefully considered in manufacturing goods known to be 
 subsequently exposed to the influence of the above named classes of 
 compounds. 
 
 4. Influence of High Temperatures. India rubber articles exposed to 
 high temperatures may develop quite a number of defects. The most 
 characteristic of these are as follows : 
 
 (a) The articles, while apparently remaining unaltered, are found, on 
 
 closer examination, to have lost more or less completely their 
 strength or cohesion. 
 
 (b) The articles assume a much darker colour, often look as if their 
 
 surface had been singed, and the latter is more or less sticky. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 
 
 297 
 
 (c) The articles become hard, frequently cracking on sharply bend- 
 ing or doubling them. 
 
 (A) Articles possessing a comparatively low coefficient of vulcanisa- 
 tion are particularly liable to develop this defect, especially if the time 
 allowed for vulcanisation has been short. Much the same phenomenon 
 occurs also in articles containing a considerable amount of paraffin wax 
 (5 per cent, and upwards), or similar hydrocarbons. 
 
 (B) This defect is generally associated with goods containing mineral 
 oils, large quantities of recovered rubber, or large proportions of * sulphide ' 
 substitutes. 
 
 (C) This last defect invariably appears in goods containing 'white' 
 substitutes (chlorosulphides), but more commonly it is due to the presence 
 of a considerable amount of free sulphur. 
 
 The first two defects are fatal to articles which are to be habitually 
 exposed to high temperatures, particularly to high pressure steam. The 
 'blowing out* of joints is generally the result of the use of this class of 
 goods. Goods showing the second defect require, moreover, a frequent 
 tightening up of the joints, which in this case often results in the packing 
 being pressed some distance into the free lumen of the pipes, thus 
 throttling the flow of steam. 
 
 If the defect described above under C is not accompanied by loss of 
 strength, it is not in itself objectionable in steam packing. Indeed, in 
 modern practice the production of goods, which in use form a hard 
 vulcanised joint, is aimed at. These remain absolutely steam proof as 
 long as they remain undisturbed. A typical American product of this 
 description gave, on analysis, the following results 
 
 India rubber, 
 Balata, 
 
 Organic extract, 
 Free sulphur, . . 2 '24 
 Sulphur of vulcanisa- 
 tion, . . .1-42 
 French chalk, . . 26'62 
 
 8 '28 per cent. 
 10-13 
 1-85 ,, ' 
 
 Paris white, . . . 19*68 per cent. 
 Plumbago, . . .20'62 ,, 
 Brass filings, . . . 9 '15 
 Total sulphur in rubber, .19-87 ,, 2 
 Coefficient of vulcanisa- 
 tion, .... 771 2 
 
 The low percentage of resilient substances (rubber and Balata) con- 
 tained in this packing renders it rather rigid to begin with. This 
 rigidity is subsequently very much increased by the high coefficient of 
 vulcanisation, and in use this steam packing will become harder still as 
 the result of the after-vulcanisation produced by the large amount of free 
 sulphur present. 
 
 There are, of course, other classes of rubber goods, the hardening of 
 which under the influence of heat would be highly objectionable, as 
 steam hose, motor car tires. 
 
 5. The Oxidation of Rubber Goods. Under this heading are classed a 
 
 1 This evidently consisted of some soft resin such as colophony. 
 
 2 Calculated upon the sum of the percentages of india rubber and Balata. 
 
298 
 
 INDIA KUBBER AND ITS ANALYSIS. 
 
 number of defects which all possess the common characteristic of the 
 rubber, becoming stiff and hard, or else being converted into a more or 
 less sticky and plastic substance. 
 
 Whether all, or, indeed, any, of these defects are to be attributed in 
 their final stages to oxidation or not, is uncertain, but they undoubtedly 
 arise in the first instance from a great variety of causes. 
 
 That india rubber is extremely prone to oxidation is a matter of every- 
 day observation. Indeed, this is only what the peculiar chemical nature 
 of india rubber would lead one to expect. The more or less pronounced 
 tendency of all terpenes to * resinify ' i.e., to form resinous bodies by 
 absorption of oxygen goes a long way to explain the analogous tendency 
 of india rubber. The spontaneous oxidation of the terpenes has been 
 very little studied, but it appears to proceed on lines similar to a great 
 number of other cases in which the formation of hydroxyl plays an im- 
 portant part. The same is, no doubt, true of many of the cases of oxida- 
 tion in india rubber. Perhaps this explains the very curious phenomenon 
 that india rubber in intimate contact with oxidising bodies invariably 
 becomes oxidised itself. Herein lies one of the most potent reasons against 
 the use of unstable or oxidisable substances in india rubber mixings. 
 
 The formation of resinous bodies from india rubber (un vulcanised) 
 under the influence of atmospheric oxygen was first observed by SPILLBR 
 (see page 9). Subsequently BuRGHARDT 1 paid some attention to the 
 technical side of this subject, and he arrived at the conclusion that "the 
 amount of oxygen taken up or combined with the caoutchouc is an index 
 of the amount of deterioration which it has undergone." In support of 
 this statement, he quotes the following analyses : 
 
 
 No. 1. 
 
 No. 2. 
 
 No. 3. 
 
 No. 4. No. 5. 
 
 
 87'27 
 
 87-50 
 
 77-91 
 
 72-53 64-00 
 
 Hydrogen, . . . 
 Sulphur, ..... 
 
 12-73 
 Nil. 
 
 10-00 
 
 2-50 
 
 10-33 
 5-15 
 
 11-31 9-26 
 1-97 2-28 
 
 Oxygen 
 
 Nil. 
 
 Nil. 
 
 6-61 
 
 14-19 24-46 
 
 1 
 
 No. 1. Pure caoutchouc from Para. 
 
 No. 2. Theoretical vulcanised elastic thread. 
 
 No. 3. Vulcanised elastic thread, perfectly sound. 
 
 No. 4. ,, ,, damaged, but still elastic. 
 
 No. 5. ,, ,, ,, brittle and very hard. 
 
 This subject was further investigated by THOMSON, 2 who also observed 
 the marked oxidation by the oxygen of the air especially at higher 
 temperatures. The detrimental effect he ascribes to the air occluded in 
 the pores of rubber goods is, however, greatly exaggerated, as the amount 
 of oxygen thus present, even if it is all taken up by the india rubber, is 
 
 1 Thorpe's Diet, of Appl. Chem., vol. ii. p. 320. 
 
 2 Journ. Soc. Chem. Ind., 1885, p. 710. 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 299 
 
 far too small to sensibly affect the latter. His suggestion, therefore, to 
 vulcanise rubber goods in vacuo is entirely unnecessary, and it is, moreover, 
 a practical certainty that it would result in the production of an enormous 
 proportion of faulty (blown) goods. THOMSON was also the first to observe 
 that the extraction of the free sulphur renders the rubber liable to very 
 rapid deterioration, and his explanation that this is due to the porosity 
 caused by the removal of the sulphur particles is undoubtedly correct. 
 This is equally true in regard to his statement that the presence of oils, 
 especially fatty oils, in india rubber much increases its tendency to 
 oxidation ; and his statement that in this respect palm oil shows the most, 
 and colza oil the least, activity is fairly in accordance with our practical 
 experience at any rate, so far as the last named of these oils is concerned. 
 
 We owe to HUFNER l a quantitative observation of the absorption of 
 oxygen by india rubber. He ascertained that an india rubber stopper of 
 40-8 grms. weight showed during six months a daily oxygen absorption of 
 0'074 c.c., or a total of about 14 c.c. = 20 mgrms. This result bears 
 obviously no appreciable relation to the weight of the rubber stopper 
 experimented upon, but only to its surface, which is not stated. 
 
 The disastrous effect of copper upon india rubber was first pointed out 
 by BURGHARDT, but it must be remarked that different copper compounds 
 act very differently in this respect, insoluble compounds scarcely acting at 
 all. This is the reason why ' dry heat cured ' waterproof fabrics, the cloth 
 of which contains copper, very rarely give trouble on that score, the cupric 
 sulphide formed in the vulcanisation process being practically innocuous. 
 Under the same conditions cupric chloride, which acts as a most energetic 
 oxygen carrier, and rapidly destroys the proofing, would form in 'cold 
 cured ' cloth. 
 
 In the series of observations on the following page are given the 
 results of a number of tests in this direction made by impregnating 
 scoured, bleached, and carefully washed cotton cloth with solutions of 
 various strengths of cupric sulphate. The fents were then dried, passed 
 through a dilute solution of sodium carbonate, washed and again dried. 
 In the various fents so prepared, the percentage of copper present was 
 estimated. Fents 1 to 9 were spread with pure Para rubber, and * cold 
 cured ' ; fents 10 to 17 were spread with a mixture of pure Para with 6 per 
 cent, of sulphur, and 'dry heat cured.' These samples were then put by 
 and examined upon from time to time. 
 
 The remarkable influence of the mode of vulcanisation upon the 
 activity of the copper will readily be noted ; so will the curious fact that so 
 very minute quantities of copper are capable apparently to destroy 
 unlimited quantities of india rubber. Respecting this point, I suspect that 
 the cupric salts in the textile fabrics enter into organic combination 
 probably with acid products of the oxidation of india rubber, and these 
 compounds gradually penetrate into the india rubber, carrying their effect 
 1 Ann. Phyr. Chern., N.F. xxxiv. 1. 
 
300 
 
 INDIA RUBBER AND ITS ANALYSIS. 
 
 through its whole mass. The action of the cupric salts is, of course, that 
 of the oxygen carrier, and may be expressed by the equations : 
 
 2CuO = Cu. 2 + 0, 
 2CuO, 
 
 which explain why mere traces of copper suffice to destroy an almost 
 unlimited quantity of india rubber. This action of copper in the cloth is 
 enormously accelerated by the presence of oils or greases in the cloth, 
 which no doubt act as an excellent vehicle for the carrying of the cupric 
 salts in the india rubber. 
 
 No. of Test. 
 
 Per cent. 
 Cu. 
 
 Appearance after 
 
 1 
 
 0-004 
 
 6 months perfectly sound. 
 
 
 
 12 
 
 
 
 18 
 
 
 
 21 very slight signs of decomposition. 
 
 2 
 
 O'OOS 
 
 6 sound 
 
 
 
 9 doubtful. 
 
 
 
 12 distinctly decomposing. 
 
 3 
 
 0010 
 
 6 very doubtful. 
 
 4 
 
 0-025 
 
 y Q6COII1J)OS111. 
 
 6 very doubtful. 
 
 5 
 
 0-03 
 
 9 badly decomposed, hard. 
 6 decomposition very marked. 
 
 
 
 9 quite hard and brittle. 
 
 6 
 
 0-045 
 
 3 distinctly decomposing. 
 
 
 
 6 badly decomposed. 
 
 7 
 
 0-065 
 
 1 month decomposing. 
 
 
 
 3 months quite hard. 
 
 8 
 9 
 
 o-ioo 
 
 Nil. 
 
 1 month badly decomposed. 
 Sound after 6 years. 
 
 10 
 
 0-004 
 
 > ? 
 
 11 
 
 0-008 
 
 6 years : distinctly stiffened, but still saleable. 
 
 12 
 13 
 
 o-oio 
 
 0-025 
 
 Getting stiff after 3 years ; after 6 years decomposed. 
 21 months : signs of change. 
 
 14 
 
 0-030 
 
 >> 5) 
 
 15 
 
 0-045 
 
 15 months : distinctly decomposing. 
 
 16 
 
 0-065 
 
 9 ,, : already hard. 
 
 17 
 
 o-ioo 
 
 4 , , : distinctly decomposing. 
 
 The salts of nickel, manganese, and tin act similarly to those of copper, 
 though in a much less degree. Whether the salts of iron are capable of 
 such action is still doubtful ; theoretically one would expect it. Chromic 
 salts are without any action. 
 
 Rubber goods containing excessive quantities of heavy coal tar 
 hydrocarbons, mineral oils, fatty oils, or greases, are very liable to rapid 
 destruction by oxidation. Goods of this description will generally be 
 found to be of very low quality, and as the above named oily bodies tend 
 to swell the india rubber, they thereby impart to it a certain softness and 
 resiliency suggesting a better quality. In this swelled state india rubber 
 is particularly sensitive to oxidation. 
 
 Fatty substitutes, if- present in large quantities, are liable to affect 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 30 1 
 
 India rubber in the same way if they contain high percentages of free oils, 
 or if they are chemically unstable. 
 
 In a very large number of cases we find, however, india rubber goods 
 suffering rapid destruction from oxidation without there being any of the 
 above named contributive causes present. I believe that in all these cases 
 a faulty vulcanisation is at the bottom of the mischief. Unfortunately 
 neither a faulty, nor, as a matter of fact, a correct, vulcanisation can be 
 denned in simple terms, such as, for instance, may be used respecting the 
 coefficient of vulcanisation. The latter certainly may be evidence of faulty 
 vulcanisation when it is obviously unusually low or unreasonably high, 
 but it is quite conceivable that a piece of rubber may be badly vulcanised 
 though it possesses a satisfactory coefficient of vulcanisation. The physical 
 conditions under which vulcanisation has been carried out exert enormous 
 influence upon the stability of the product, more particularly upon its 
 liability to oxidation, and at the present time we are not yet in a position 
 to ascertain in numerical terms the effect of any possible combination of 
 vulcanising conditions in relation to any given coefficient of vulcanisation. 
 
 e. Defects in Appearance. Defects of this description may be due to 
 mechanical causes, to unsuitable physical conditions (apart from those 
 which may possibly exist in the vulcanising process) during manufacture, 
 or to certain faults in the mixings or curing. We need only concern our- 
 selves here with defects due to the last two causes indeed, it will be found 
 that as far as blemishes in the appearance of india rubber goods are con- 
 cerned, the cause lies almost invariably in the vulcanising, and consists 
 generally of some kind of discoloration. These defects appear, therefore, 
 most frequently in white, drab, and coloured rubber goods, and they are 
 caused either by the action of sulphur vapour or by the action of chloride 
 of sulphur, and not uncommonly by the combined action of chloride of 
 sulphur and moisture, upon the pigments and other colouring matters con- 
 tained in the india rubber. Either these are not able to withstand the 
 influence of these substances, in which case the colouring materials used 
 have been injudiciously selected, or they contain some impurity which is 
 responsible for their discoloration, or perhaps the ' cure ' has been too pro- 
 longed or carried out at too high a temperature, or the ' cold curing ' 
 solution has been too strong or carried out under conditions favouring the 
 deposition of moisture upon the surface of the goods. A careful examina- 
 tion of the goods and of the materials used in the manufacture will in 
 most cases readily furnish the clue to the explanation of the defect. 
 
 The appearance of the free sulphur contained in manufactured india 
 rubber upon its surface ('sulphuring up') is in the case of a large number 
 of rubber articles looked upon as an effective finish, in every way to be 
 preferred to the rather crude appearance of the goods fresh from the 
 vulcanising room, as it imparts to them a soft, velvety surface. In many 
 other articles, notably coloured and black goods, this ' sulphuring up ' is, 
 however, regarded as a serious defect. 
 
302 INDIA RUBBER AND ITS ANALYSIS. 
 
 This phenomenon is, of course, due to the excess of sulphur in the 
 vulcanised goods, which, after vulcanisation, gradually changes from the 
 colloidal to the crystalloidal form (see page 111), and it is evident that it 
 is closely connected with this change of physical condition, because an 
 unvulcanised mixture of india rubber with crystalloidal sulphur flowers 
 of sulphur, ground rock sulphur never sulphurs up. A vulcanised 
 rubber article which has a tendency to ' sulphur up ' will not do so for 
 months if it be kept at a fairly uniform temperature in a perfectly dust- 
 free atmosphere, but the moment it is touched by the hand or by a brush, 
 or as soon as it is exposed to dust, the process of ' sulphuring up ' begins, 
 and in a few hours will spread all over the surface. At first a very fine 
 film is formed, composed of microscopically fine sulphur crystals, which in 
 course of time (and provided there be a sufficient amount of free sulphur 
 in the article) grow in size until eventually they become plainly visible to 
 the naked eye. 
 
 The current notion is that this occasionally very troublesome defect is 
 simply due to the fact that sulphur is slightly volatile at ordinary tempera- 
 tures. This view I consider entirely erroneous. Indeed, if the extremely 
 slight volatility of ordinary sulphur were the cause of the ' sulphuring 
 up,' we should expect this defect to disappear even at a quicker rate than 
 at which it appears, or, in other words, it could not exist at all. From what 
 has been shown respecting the curious phenomena accompanying the 
 crystallisation of colloidal sulphur (page 110 /.), it is evident that this 
 process of ' sulphuring up ' would much more aptly be described as a 
 process of 'sulphuring down' i.e., the process does not, as is generally 
 suggested, begin below the surface, gradually working its way outward, but 
 it begins at the very surface itself. 
 
 The sulphur remaining free after vulcanisation is fairly evenly dis- 
 tributed throughout the india rubber in the form of extremely small 
 globules of colloidal sulphur, and these retain their colloidal state until 
 one or more of them lying exposed on the surface begin to crystallise 
 under the influence of extraneous agencies. Crystallisation once having 
 started on the surface, rapidly spreads over it at any rate, under the condi- 
 tions of repeated handlings to which rubber goods are subject in the 
 factory. The action of all these crystallised globules is now exactly 
 identical to those described on page 111 and figured on Plate III. i.e., the 
 vapour tension of crystallised sulphur being considerably lower than that 
 of colloidal sulphur, every one of them represents the centre of a mini-, 
 mum of vapour pressure, to which, consequently, all the colloidal sulphur 
 from the inside of the rubber article gradually distills, crystallising around 
 it in the well-known dendritical forms. It is therefore evident that the 
 'sulphuring up' starts entirely from the surface, the colloidal sulphur 
 inside the rubber then, so to speak, condensing on this outer crystalline 
 surface. 
 
 Once formed, the permanent removal of this sulphur condensation is an 
 
ANALYSIS OF INDIA RUBBER ARTICLES. 303 
 
 extremely difficult matter. The remedy therefore lies in the prevention of 
 its formation. This must obviously be possible in one of the following 
 ways : 
 
 (a) Vulcanising with as small an excess of sulphur as possible. 
 
 (b) Vulcanising under such conditions that at the end of the operation 
 
 all the free sulphur has disappeared. 
 
 (c) Using rubber mixings containing substances which will chemically 
 
 combine with the free sulphur. 
 
 (d) Coating the vulcanised articles with some finish or polish imper- 
 
 meable to sulphur vapours. 
 
 (e) Employing, in the mixing, substances possessing the property of 
 
 rendering, after vulcanisation, the free colloidal sulphur crystalline 
 throughout the mass of the articles. 
 
 In practice method (a) is the one most generally followed, but to be 
 efficacious it requires an extremely nice adjustment of the proportions of 
 india rubber, sulphur carriers, rubber substitutes, and sulphur employed, 
 and of the time and temperature to be allowed for vulcanisation. 
 
 Method (b) can only be used with goods vulcanised by the 'open cure,' 
 in which the sulphur vapours may freely pass off the vulcanising goods. 
 On a very large scale this method is adopted for the curing of rubber 
 shoes and goloshes. 
 
 Method (c) would be perhaps the most satisfactory method to employ 
 all round. There are many preparations in the market for the purpose, 
 but, as yet, none of them can be relied upon. 
 
 Whether method (d) is satisfactory or not depends much upon the 
 amount of free sulphur present, as, at the present moment, we do not 
 know of any absolutely sulphur-proof coating for the purpose. 
 
 I am not aware that this method (d) has so far designedly been resorted 
 to, but the action it involves does by no means appear impossible of 
 realisation. Its achievement would represent a most valuable technical 
 acquisition. 
 
 The defect known as ' rolling / or ' cockling,' which not uncommonly 
 occurs in waterproof fabrics, must also be mentioned here. It is most 
 usually due to shrinkage of the cloth on exposure to rain, in which 
 shrinkage, of course, the rubber coating does not take- part. It must, 
 however, be borne in mind that a very similar phenomenon may be pro- 
 duced in the cold curing of proofed cloth, the rubber coating of which has 
 not been dried effectually, and consequently still retains some naphtha, 
 Whether the cause of the rolling lies in the cloth itself or in the proofing, 
 may readily be decided by observing in which direction the cloth rolls. 
 Rolling towards the cloth side is certainly due to shrinkage of the cloth, 
 rolling towards the rubber side to the cold curing of an imperfectly dried 
 proofing. 
 
APPENDIX. 
 
 SANITARY CONDITIONS IN INDIA RUBBER WORKS. 
 
 IN the mind of the general public, an india rubber works is always 
 associated with all sorts of evil smells, and it must, unfortunately, be 
 admitted that there is substantial ground for this indictment, although 
 within the last ten years much has been attempted, and not without 
 success, in mitigation of this state of things. Still much remains to be 
 done. Fortunately, in this respect most manufacturers to-day clearly 
 realise that the existence of insanitary conditions in their workshops is 
 not only reprehensible on ethical grounds, but also a very potent cause of 
 'slackness,' or loss of power of production, in the work people, though 
 there is still a good deal of indifference with regard to insanitary con- 
 ditions, the remedy for which is suspected of having ' no money in it.' 
 
 In the following pages I propose to deal with the insanitary conditions 
 in india rubber works in the order in which they are likely to arise in the 
 progress of manufacture. 
 
 1. Rubber Washing. The first operation the crude rubber undergoes 
 on entering the factory consists in a washing process. As a preliminary, 
 the rubber, if necessary, after being reduced by cutting to pieces of a 
 suitable size, is placed in tanks, where it is boiled with water for a certain 
 time. From these tanks it passes to the well-known washing mills, the 
 washed sheets emerging from which are subsequently dried either in hot-air 
 stoves or in vacuum pans. 
 
 The nuisance likely to arise in this process is in the boiling of the 
 rubber. No trouble in this respect is likely to arise in the boiling of very 
 high grades of rubber, but a few of the medium and low grade brands give 
 off an almost .unbearable stench. The obvious remedy for this is to carry 
 out the boiling in closed and ventilated vessels. The alternative of fixing 
 a hood and ventilating shaft above the tank would, no doubt, be equally 
 efficacious in abating the nuisance inside the workshops, but its trans- 
 ference to the open air in densely inhabited districts would not long be 
 tolerated. 
 
 2. Rubber Mixing. A large variety of organic and mineral compounds 
 are mixed with india rubber, and amongst the latter the compounds of lead, 
 zinc, and antimony occupy a very prominent place. In many rubber 
 
SANITARY CONDITIONS IN INDIA RUBBER WORKS. 305 
 
 factories these materials are sifted before being mixed with the india 
 rubber, and it is still customary in many places to perform this operation 
 by hand. It is quite needless to dwell upon the danger likely to arise from 
 this reprehensible practice, and the sifting by machinery of all lead and 
 zinc compounds should be rigorously insisted upon. 
 
 The mixing rollers employed in rubber factories must also be mentioned 
 as a source of danger. These machines, as is well known, consist of two 
 heavy rollers placed horizontally, between which the india rubber and 
 other ingredients are worked into a homogeneous mixture. Considering 
 the dangerous character of this kind of machine, it is really surprising that 
 the number of accidents occurring is so small. Those that get their fingers 
 caught somehow always manage to tear them from the grip of the rollers, 
 though they may leave their nails behind and get their finger ends more or 
 less severely crushed. There are, however, a number of cases on record of 
 men not having succeeded to save themselves in this manner, but being 
 drawn into the machine and having their arm or arms torn from their 
 bodies. This, of course, means death in its most appalling form. 
 
 For starting and stopping, these machines possess on their driving 
 wheels a clutch and striking gear ; but it is a well-known fact, that when 
 the machine has its load on, it is a physical impossibility to open the clutch 
 and stop the machine. There is no reason whatever why such an in- 
 efficient type of clutch should be tolerated, as there are quite a number of 
 clutches known which do not possess this grave defect. 1 
 
 3. Coal Tar Naphtha, Petroleum Spirit, Shale Oil. These well-known 
 solvents are the cause of a good deal of trouble in india rubber works. In 
 those workrooms in which these solvents are not deliberately evaporated, 
 where, therefore, they are only exposed to normal workshop temperatures, 
 no inconvenience is experienced, as it is obviously in the interest of the 
 manufacturer to adopt all possible means to avoid a premature evaporation 
 t'.e., waste of solvents. The case is different in those rooms in which these 
 solvents are purposely evaporated so as to leave a dry coat of rubber on a 
 textile fabric. Evaporation of the solvents is carried out on the machines 
 known as ' spreading machines,' and used in the manufacture of waterproof 
 cloth. This industry has assumed enormous proportions. The yearly 
 production of waterproof fabrics in Great Britain alone amounts, on a 
 moderate computation, to at least 125,000 pieces of 60 yards each. 
 Taking the very moderate average of 8 oz. dry weight of rubber mixing 
 per yard of cloth, and taking the ' naphtha moisture ' of the rubber doughs 
 used as amounting to 50 per cent., also a moderate estimate, this would 
 give us the quantity of solvents evaporated in British factories alone as 
 amounting to 2000 tons, or say 600,000 gallons, per annum. It is there- 
 fore not to be wondered at that the atmosphere, even in the most modern 
 
 1 A very ingenious device, enabling the worker to stop his machine practically 
 instantly even if both his hands should be caught, has been described by Mr F. CLOUTH 
 in his book on Gfummi, Guttapercha und Balata, Leipzig, 1899. 
 
 20 
 
306 INDIA RUBBER AND ITS ANALYSIS. 
 
 and spacious of spreading rooms, is quite unbearable for anybody not 
 inured to it, and becomes at times even unbearable to the workmen, who 
 then resort to the simple British plan, not unknown in respect of less 
 objectionable occupations, of 'walking out of it.' 
 
 Upon the question whether the vapours of the above named rubber 
 solvents are dangerous to health, I do not feel competent to express an 
 opinion. They certainly irritate the eyes and the respiratory organs, but 
 this effect is not observable in those accustomed to work in an atmosphere 
 charged with these vapours. On the other hand, there is no doubt what- 
 ever that working in the atmosphere of a spreading room very frequently 
 produces strongly marked effects on the men, who often plainly show all the 
 symptoms of intoxication. Under specially unfavourable conditions cases 
 of actual stupor and markedly dilated pupils not infrequently occur. Even 
 where these symptoms are absent there is generally a decided increase of. 
 the frequency of the pulse. The latter I have repeatedly observed as high 
 as 86 per minute. In young boys who are new to the work, this naphtha 
 intoxication produces a curious hysterical condition followed by stupor. 
 The effect passes off after a few hours' sleep, and gradually the boys appear 
 to become less subject to the influence of these vapours. As a curious fact, 
 I may mention that the number of abstainers amongst india rubber 
 spreaders appears to be singularly small in fact, I have never met 
 one yet. 
 
 Whether the above conditions are matters for Government interference 
 must be left to the medical advisers of the Home Office to decide, but 
 there can be no doubt that it is inimical to all the interests of manu- 
 facturers to have a large body of men in their factories working all the 
 year round in a state bordering on intoxication. 
 
 The obvious remedy for this nuisance, to apply a very lenient term to 
 it, is not to allow these solvents to pass into the atmosphere, but to con- 
 dense them as soon as they are formed. This problem has attracted the 
 attention of quite a number of manufacturers and inventors, many firms 
 putting up installations for the purpose. Most of these were abandoned 
 after a period of experimenting ; some of them are working still, but, from 
 all the information I could gather, with rather indifferent results. This r 
 in my opinion, is entirely due to the wrong principles upon which the 
 plants in question were or are constructed. 
 
 It appears that in every case the essential features of the plant em- 
 ployed were practically the same : a low cowl or hood is fixed over the 
 heating chests upon which the rubber-coated fabrics are dried. By 
 connecting this cowl through some form of condenser with a suction pump 
 or fan, the naphtha vapours rising from the fabrics are ' carried through a 
 condenser. But it will be seen that in order to lift the naphtha vapours 
 from the cloth and to carry them to the condenser, a current or volume 
 of air is required greatly exceeding the volume of the naphtha vapours to 
 be condensed. This subsequently impedes considerably the efficiency of 
 
SANITARY CONDITIONS IN INDIA RUBBER WORKS. 307 
 
 the condenser, owing to the low heat conductivity of air ; and as, moreover, 
 the hydrocarbons in question possess a very considerable vapour tension 
 even at ordinary temperatures, it is obvious that a considerable proportion 
 of the vapours thus aspirated must unavoidably remain uncondensed. I 
 am informed that the best results obtained with this description of plant 
 varied from 20 to 25 per cent, of the quantity of naphtha carried by the 
 rubber-coated fabrics. Against this there was the cost of the suction 
 current and of the working of the condenser, which, together with the 
 charges for interest on capital and depreciation, rendered the process 
 unremunerative. To this must be added the danger arising from fires, 
 the latter tending at once to spread from one machine to the other when 
 they were all connected by the above named suction tubes, and the possi- 
 bility of serious explosions of the mixtures of air and naphtha vapours. 
 
 To my mind all these attempts were either doomed to failure at the 
 very outset, or could, at best, attain to very mediocre results only, owing to 
 the application of an air current for the carrying of the vapours to a con- 
 denser ; and I am of opinion that the simplest, the most natural, and, at 
 the same time, the most efficient process will be found to consist in allowing 
 the naphtha vapours to fall by their own gravity 1 upon a suitably 
 arranged condenser surface. Such a plant has been devised by WKBER and 
 FRANKENBURG (Eng. Pat, No. 16,919, 1901), and it is working in a most 
 satisfactory manner, from 60 to 70 per cent, of the naphtha carried by 
 the rubber-coated fabric being recovered. In case of fire there is no risk 
 of its spreading to other machines or of causing explosions, every machine 
 being entirely self-contained. 
 
 The above named solvents are also apt to render the atmospheres of 
 the ' making up rooms ' rather oppressive, owing to the large quantities of 
 rubber cement used in them. A ' solutioning ' machine patented by 
 J. FRANKENBURG (Eng. Pat., No. 3568, 1899) effects a great improvement 
 in this respect at a considerable saving of the expensive rubber cement. 
 
 4. The Vulcanising Process. There is nothing in any of the various 
 methods of vulcanising india rubber by heat (in conjunction with sulphur) 
 which could detrimentally affect the workers. As the temperature is 
 rather a high one, in some of the methods small quantities of sulphur 
 vapour being given off, the rooms in which these processes are carried out 
 should be spacious and effectually ventilated. 
 
 The ' cold ' process of vulcanisation, with a solution of sulphur mono- 
 chloride in carbon bisulphide, has already, on the part of the Home Office, 
 been made the subject of a number of regulations and rules respecting the 
 men employed. The strict administration of these regulations under the 
 control of the factory inspectors has now almost completely stopped the 
 
 1 It is curious to see how persistently rubber manufacturers and engineers treat the 
 vapours of these solvents, and of carbon bisulphide, as ' light vapours,' meaning thereby 
 lighter than air ; whereas, as a matter of fact, the vapours of solvent naphtha are 
 several times as heavy as air i.e., about as heavy as is zinc compared with water. 
 
308 INDIA RUBBER AND ITS ANALYSIS. 
 
 occurrence of acute and chronic cases of poisoning with carbon bisulphide 
 which were formerly so common. 
 
 As regards the machinery by which this process is at present carried 
 out, this is still capable of considerable improvement. Its present form is 
 very primitive, and only too plainly shows that it was designed with the 
 simple view to its purpose, but without the least regard of some of the 
 physical peculiarities and of the dangers to life and limb likely to arise 
 from these. Of course I do not suggest that these omissions were 
 intentional in fact, I am quite certain that they are simply due to the 
 entire lack of physical and chemical knowledge on the part of those who 
 designed them. A rationally constructed ' sincaloring ' machine will not 
 only produce superior goods, but it will also entirely prevent those 
 disastrous explosions which have been uncomfortably numerous of recent 
 years. It even appears to me perfectly feasible to prevent altogether the 
 discharge, now practised, of the carbon bisulphide vapours into the open 
 atmosphere, and to recover practically the whole of them. This would 
 finally remove one of the last remaining of the serious objections against 
 the use of* carbon bisulphide in rubber works. The discharge of enormous 
 quantities of the vapours of this obnoxious liquid within more or less 
 densely inhabited districts is a piece of unmitigated barbarism. 
 
 The substitution of other solvents benzene, petroleum spirit, carbon 
 tetrachloride for carbon bisulphide has been attempted times without 
 number, but without any satisfactory result. The reason of this is, of 
 course, that the carbon bisulphide employed in the process does not act as 
 an indifferent solvent merely for the sulphur monochloride, but has a 
 specific physical effect upon the india rubber, which is absent in any of the 
 above named proposed substitutes (see page 199). 
 
 From the above it will be seen that the sources of danger to health 
 peculiar to the india rubber industry are few, and, such as they are, 
 unquestionably capable of being worked in such a manner as to preclude 
 entirely their affecting the workpeople to any appreciable degree. This 
 desirable state has not been realised at the present moment, or very im- 
 perfectly only, but we are gradually getting nearer to it, thanks partly to 
 the common-sense of the manufacturers, partly to the pressure exerted in 
 this direction by the Home Office. 
 
 THE END. 
 
INDEX. 
 
 A 
 
 ACCRA rubber, 125. 
 
 Acetone extract, 241, 258, 262. 
 
 Acidising process, 55. 
 
 Acids, Action of, upon manufactured 
 
 rubber, 295. 
 Admiralty test, 229. 
 African ball, 3. 
 
 ,, flake, 3. 
 
 Alcoholic potash extract, 245, 268. 
 Alkalies, Action of, upon manufactured 
 
 rubber, 296. 
 
 Analytical results, Interpretation of, 262. 
 Angola rubber (niggers), 119, 127. 
 Aniline lake pigments, 206. 
 Antimony pentasulphide, 185. 
 Asbestos, 161. 
 Assam rubber, 3, 13, 119, 128. 
 
 B 
 
 BAHFA rubber, 123. 
 Balata, 3. 
 
 Banqui rubber, 126. 
 Barium sulphate, 204. 
 Barytes, 161. 
 Benguela niggers, 127. 
 Beni River rubber, 13. 
 
 ,, ,, ., vulcanisation constants, 
 
 85, 87. 
 
 Bituminous bodies, Determination of, 244. 
 Black pigments, 210. 
 Blown oils, Determination of, 245, 269. 
 Blue lake pigments, 209. 
 
 ,, pigments, 208. 
 Bolivian rubber, 1 123. 
 Bone black, 210. 
 Borneo rubber, 2, 3, 13, 14, 119, 129. 
 
 ,, ,, Vulcanisation constants of , 
 
 85, 87. 
 Bornesite, 2. 
 
 Bronzes and bronze colours, 211. 
 Brown pigments, 204. 
 Brunswick green, 208. 
 Bumba rubber, 1 26. 
 Bussira rubber, 126. 
 Butylene, 22. 
 
 CARBON black in rubber, Determination of, 
 
 252, 276. 
 Chlorine in rubber, 282. 
 
 Chlorine in rubber, Determination of, 254. 
 ,, substitutes, Determination of, 
 
 246. 
 Chlorosulphide substitutes, Determination 
 
 of, 245, 269. 
 Chrome green, 208. 
 Cobalt blue, 209. 
 Coefficient of vulcanisation, 283. 
 Colouring matters in rubber goods, 276. 
 Constructive components of india rubber 
 
 articles, 215. 
 Copper, Action of, upon manufactured 
 
 rubber, 220, 299. 
 
 DAMBONITE (dambose), 2. 
 
 Dead Borneo, 129. 
 
 Defects in rubber goods, 288. 
 
 Dextrine in rubber, Determination of, 251. 
 
 Dimethyl-allene, 25. 
 
 Dimethyl-inosite, 2. 
 
 Dimethyl - pyrollidyl - ammonium iodide,. 
 
 27. 
 
 Dimethyl-trimethylene-bromide, 27. 
 Dimethyl-vinyl-carbinol, 25. 
 Dinitro-polyprene, 40. 
 Dipentene, 29. 
 Distension of rubber, Electrical effect of,. 
 
 17. 
 
 Divinyl, 27. 
 Dry and moist heat tests, 229. 
 
 E 
 
 ELECTRICAL tests, 230. 
 Epichlorhydrine as a solvent, 260. 
 Equateur rubber, 126. 
 
 FATTY substitutes, Determination of, 245^ 
 
 247, 269. 
 
 Fenton rubber, 148. 
 Fibrous materials in manufactured rubber,. 
 
 Determination of, 251. 
 Free fatty oils, Determination of, 241. 
 
 ,, sulphur, Determination of, 241. 
 French chalk, 161. 
 
 G 
 
 GABOON rubber, 2, 119, 125. 
 Gambia rubber, 119, 125. 
 Geranial, 24. 
 
3 io 
 
 Geraniene, 25. 
 
 Geraniol, 24. 
 
 Glacial acetic acid test for paraffin wax, 
 
 266. 
 
 Golden sulphide, 204. 
 Grand Bassam rubber, 125. 
 Green lake pigments, 208. 
 
 ,, pigments, 208. 
 Guayaquil rubber, 14, 123. 
 Gutta percha, 3. 
 
 H 
 
 HARD rubber, Analysis of, 257. 
 ,, ,, Composition of, 90. 
 
 Heveene, 22. 
 
 Hexa-oxy-he-xa-methylene, 2. 
 
 High temperatures, Action of, upon manu- 
 factured rubber, 296. 
 
 Honduras rubber, 124. 
 
 Hydrocarbons, Determination of, 241. 
 
 Hypo, 183. 
 
 I 
 
 INDIA rubber, Action of heat upon, 20. 
 ,, ,, Analysis, Acetone extrac- 
 tion, 240. 
 
 ,, ,, ,, Determination of car- 
 
 bohydrates, 251. 
 ,, ,, ,, ,, chlorine, 254. 
 
 ,, ,, fatty substitutes, 
 
 245. 
 ,, ,, ,, ,, mineral matter, 
 
 248. 
 
 ,, ,, ,, ,, sulphur of vulcan- 
 
 isation, 254. 
 
 ,, ,, ,, ,, total sulphur, 253. 
 
 ,, ,, ,, General scheme of, 
 
 236. 
 
 Methods of, 239. 
 ,, ,, ,, Nitrobenzene method 
 
 in, 248. 
 ,, ,, ,, Nitro-naphthalene 
 
 method in, 248. 
 ,, ,, ,, Preparation of samples 
 
 for, 257. 
 
 ,, ,, \, Separation effected by 
 
 acetone extraction, 
 241. 
 
 ,, ,, ,, ,, alcoholic potash 
 
 extraction, 245. 
 
 ,, ,, ,, ,, pyridine extrac- 
 
 tion, 243. 
 ,, ,, ,, Synoptical tables on, 
 
 255, 256. 
 , , of recovered, 
 Artificial, Nobel's, 150. 
 Albuminous matter in, 5. 
 Ash in, 12. 
 Behaviour of, against alkalies, 
 
 31. 
 ,, ,, the halogens, 
 
 31. 
 ,, ., ,, hydrochloric 
 
 acid, 37. 
 
 ,, ,, ,, hydrobromic 
 
 acid, 38. 
 
 INDEX. 
 
 India rubber, Behaviour of, 
 
 gainst hydri- 
 odic acid, 38. 
 ,, ,, ,, nitric acid, 
 
 39. 
 
 nitrogen di- 
 oxide, 39. 
 
 nitrosy 1 
 chloride, 
 40. 
 ,, ,, ,, nitrous acid, 
 
 39. 
 
 ,, oxygen, 40. 
 ,, ,, ,, solvents, 14. 
 
 Chemical behaviour of, 30. 
 Chemistry of, 1. 
 Determination of oxygen in, 
 
 118. 
 Diffusion of gases through, 
 
 18. 
 
 Distensibility of, 17. 
 Drying of, 2. 
 Dry distillation of, 22. 
 Dyeing of, with soluble 
 
 dyes, 213. 
 
 Effect of heating upon, 16. 
 Elasticity of, 17. 
 Elementary composition of, 
 
 5, 11. 
 
 Examination of, 116. 
 Impurities in, 1, 3, 5. 
 Insoluble constituent of, 6. 
 latex, Composition of, 5. 
 Loss on washing, 117. 
 Mastication of, 16. 
 Methods of the analysis of 
 
 recovered, 136, 152. 
 Mineral impurities of, 12. 
 Molecular weight of, 20. 
 Optical properties of, 21. 
 Oxygen absorption of, 6. 
 
 ,, in crude, 118. 
 Permeability of, by water, 
 
 13. 
 Products of dry distillation 
 
 of, 29. 
 Reclaimed or recovered, 
 
 130. 
 
 Resins in, 3, 117. 
 separator, Biffen and 
 
 Howard's, 5. 
 Solubility of, 14. 
 solvents, 188. 
 
 Boiling-points of, 
 
 194. 
 
 Impurities in, 193. 
 Specific gravity of, 13. 
 substitutes, 130. 
 
 ,, Analysis of, 155. 
 
 ,, Analytical 
 
 tables on 
 fatty, 147. 
 
 ,, Fatty and or- 
 
 ganic, 140. 
 ,, Free oil in fatty, 
 
 144. 
 ,, Manufacture of, 
 
 142. 
 Vulcanisation of, 41. 
 
INDEX. 
 
 Inosite, 2. 
 
 Insulation resistance, Measurement of, 
 
 230. 
 
 Iron oxides, Red and brown, 205. 
 Isoprene, 22, 25. 
 
 ,,, monochlorhydrate, 25. 
 
 ,, Polymerisation of, 28. 
 
 K 
 
 KASSAI rubber, 119, 125, 126. 
 
 LAC LEOPOLD rubber, 126. 
 
 Lagos rubber, 3, 13, 119. 
 
 Lake pigments, Red and brown, 205. 
 
 Lamp black, 210. 
 
 Lead sulphate, 161. 
 
 ,, thiosulphate, 183. 
 
 ,, vulcanisers, 183. 
 Liberia rubber, 125. 
 Litharge, 161. 
 Lithopone, 161, 186, 203. 
 Loanda rubber, 127. 
 Lopori rubber, 119, 126. 
 Lower Congo rubber, 127. 
 
 M 
 
 MADAGASCAR rubber, 2, 3, 14, 119, 127, 
 
 128. 
 
 Magnesia, 161. 
 Magnesium carbonate, 161. 
 Mangabeira rubber, 3, 13, 119, 123. 
 Manicoba rubber, 119, 123. 
 Manoh twist, 3, 119, 125. 
 Manufactured india rubber, Apparent spe- 
 cific gravity 
 of, 226. 
 
 ,, ,, ,, Chemical ana- 
 
 lysis of, 234. 
 
 ,, ,, ,, Dry and moist 
 
 heat tests, 
 229. 
 ,, ,, ,, Mechanical 
 
 tests, 228. 
 ,, ,, ,, Physical tests, 
 
 225. 
 Porosity of, 
 
 226. 
 
 ,, ,, Real specific 
 
 gravity of, 
 226. 
 
 ,, ,, ,, Specific grav- 
 
 ity test for, 
 225. 
 
 Massai rubber, 125. 
 Matezite, 2. 
 
 Matto-Grosso rubber, 119, 122. 
 Metallic lustres, 211. 
 
 Metals, Action of upon india rubber, 212. 
 -Methyl-divinyl, 27. 
 j8-Methyl-pyrollidine, 27. 
 Mexican strips, 124. 
 
 Mineral and resin oils, Determination of, 
 241. 
 
 Mineral matter in rubber, Determination 
 
 of, 247, 275. 
 Mollendo rubber, 13. 
 Mongala rubber, 126. 
 Mono-methyl-inosite, 2. 
 Mozambique rubber, 3, 14, 119, 127. 
 
 NANTUSI, 266. 
 Naphtha, 189. 
 Negrohead, 14, 119, 122. 
 New Guinea rubber, 129. 
 Nicaragua scrap, 119, 124. 
 Nitro-naphthalene extract, 247, 273. 
 
 , , treatment, Air bath for, 
 
 249. 
 
 ORGANIC compounds, Action of, upon 
 
 manufactured rubber, 296. 
 Overvulcanisation, 285. 
 Oxidation of rubber goods, 297. 
 Oxidised oils, Determination of, 245, 269. 
 Oxolin, 149. 
 
 Oxygen-absorption of india rubber, 299. 
 Ozokerite, 266. 
 
 PADANG rubber, 128. 
 Panama rubber, 124. 
 Para rubber, 3, 13, 14, 119, 122. 
 ,, ,, Vulcanisation constants of, 
 
 84, 86. 
 
 Paraffin wax, 266. 
 Paris white, 204. 
 Penang rubber, 128. 
 Pernambuco rubber, 123. 
 Peruvian ball, 13, 119, 124. 
 Petroleum spirit, 194, 197. 
 Physical tests, 225. 
 
 Pigments. See under respective colours. 
 Pinene, Combustion heat of, 107. 
 Pitches, 151. 
 
 ,, Determination of, 244. 
 Plaster of Paris, 161. 
 Polyprene chlorosulphide, 99. 
 
 Combustion heat of, 107. 
 
 dihydrochloride, 38. 
 
 iodide, 36. 
 
 nitrosite, 39. 
 
 sulphide, 107. 
 
 tetrabromide, Preparation of, 
 
 32. G""i 
 ,, ,, Action of phenol 
 
 upon, 33. 
 Pontianac, 129. 
 Proteids in rubber, 5. 
 Prussian blue, 209. 
 Pyridine extract, 244, 267. 
 
 Q 
 QUIESEMBO rubber, 14. 
 
3 I2 
 
 INDEX. 
 
 R 
 
 RANGOON rubber, 119, 128. 
 
 Recovered rubber, 95. 
 
 Red lead, 185. 
 pigments, 204. 
 
 Resin, Spiller's, 9. 
 
 ,, oils, Identification of, 265. 
 
 Resinous and oily constituents, Determina- 
 tion of, 241. 
 
 Resins, 151. 
 
 ,, Determination of, 241. 
 
 ,, Influence of, upon vulcanisation, 
 
 ,, Solubilities of, 259. 
 Rio Janeiro rubber, 3. 
 Ruki rubber, 126. 
 
 SANITARY conditions in rubber works, 
 
 304. 
 
 Sankuru rubber, 126. 
 Santos sheet, 13. 
 Senegal rubber, 125. 
 Sernamby, 122. 
 Shale naphtha, 194, 198. 
 Sierra Leone rubber, 3, 13, 117, 119, 125. 
 Silica, 161. 
 Solvents, 188. 
 Specific gravity, 225. 
 
 ,, ,, Apparent and real, 226. 
 
 Spiller's resin, 9. 
 Stannic bromide test, 265. 
 Starch in rubber, Determination of, 251. 
 Sulphide substitutes, Determination of, 
 
 245, 269. 
 Sulphur, 177. 
 
 ,, carriers, 108, 176. 
 
 Chloride of, 179. 
 ,, Combined, in pyridine extract, 
 
 244. 
 
 ,, Flowers of, 178. 
 ,, free, Determination of, 242. 
 ,, in india rubber, 282. 
 ,, ,, ,, Determination of, 
 
 254. 
 ,, ,, substitutes, Determination of, 
 
 246. 
 
 Milk of, 179. 
 
 ,, of vulcanisation, 283. 
 , , , , Determination 
 
 of, 254. 
 
 Precipitated, 179. 
 Solubilities of, 177. 
 ,, Solubility of, in acetone, 177. 
 ,, Vapour density of, 105, 108. 
 ,, ,, ,, under reduced 
 
 pressure, 109. 
 
 Sulphuring up in rubber goods, 110. 
 Suncracking test, 229. 
 
 TAR, carbonaceous matter in, Determina- 
 tion of, 268. 
 ,, Determination of, 244. 
 
 Tar in rubber mixings, 267. 
 Terpadiene, 28. 
 
 Tetroxyphenyl-dioxypolyprene, 34. 
 Tetroxyphenyl-polyprene, 33. 
 Textile fabrics, Dressing or sizing in, 217. 
 
 ,, ,, Examination of, 215. 
 
 ,, ,, Fastness to rain, 219. 
 
 ,, ,, Fatty matter in, 218. 
 
 ,, ,, Mordanting materials in, 
 
 220. 
 
 ,, ,, Resistance to street mud 
 and dust, 219. 
 
 ,, Shrinkage of, 217. 
 
 ,, ,, Tensile strength test of, 
 
 216. 
 
 Thimbles, 14, 127. 
 Trimethyl-ethylene-oxide, 26. 
 Turpentine, 198. 
 Tyndall effect, 73. 
 
 U 
 
 UELLE rubber, 126. 
 Ultramarine, 208. 
 Undervulcanisation, 285. 
 Upper Congo rubber, 126. 
 
 VAPOUR, cure, 52. 
 Vegetable black, 210. 
 Velvril, 150. 
 Vermilion, 204. 
 Violet lake pigments, 209. 
 
 ,, pigments, 208. 
 Virgin rubber, 122. 
 
 Vulcanisation of india rubber, 4, 12, 41. 
 Theory of, 82. 
 with alkaline polysulphides, 
 
 47. 
 
 hypochlorites, 55. 
 sulphur, 41. 
 sulphur - monochloride, 
 
 51, 97, 112. 
 the haloids, 56. 
 
 Vulcanisers, 174 
 
 176. 
 
 W 
 
 WAXES, 151. 
 White lead, 161, 202. 
 ,, pigments, 202. 
 Whiting, 161. 
 
 YAKOMA rubber, 126. 
 Yellow lake pigments, 208. 
 
 ,, ochre, 208. 
 
 ,, pigments, 207. 
 
 ZINC oxide, 161. 
 ,, sulphide, 186. 
 ,, white, 161, 203. 
 ,, yellow, 207. 
 
INDEX TO AUTHORS' NAMES. 
 
 ADRIANI, 6, 36. 
 Allen, 36. 
 
 B 
 
 BANNOW, 192. 
 
 Berthelot, 38. 
 
 Berzelius, 6. 
 
 Biffen, 5. 
 
 Biltz, H., 109. 
 
 Bineau, 105. 
 
 Bleier and Kohn, 106. 
 
 Bouchardat, 22. 
 
 Brown and Millar, 75. 
 
 Biitschli, 78. 
 
 Burghardt, C., 47, 53, 212, 298, 299. 
 
 DEVILLK and Troost, 105. 
 Donath, E., 48, 28:2. . 
 
 EULEK, W., 27. 
 Exner, F., 17. 
 
 FARADAY, 5, 6. 
 Fawsitt, C. A., 53. 
 
 Q 
 
 GADZIATZKY, 25. 
 
 Gerard, 47. 
 
 Girard, 2. 
 
 Gladstone and Hibbert, 5, 6, 9, 16, 20, 
 
 21, 29, 31. 
 Goodyear, 41. 
 Graham, 18, 57. 
 Guthrie, 97. 
 
 II 
 
 HANAUSEK, 14. 
 Hancock, 41. 
 Hay ward, 41. 
 Heeren, 14. 
 Heinzerling, 21, 48. 
 
 and Pahl, 162, 179, 274. 
 
 Henriques, R , 3, 6, 13, 84, 97, 145, 146, 
 
 243, 245, 247, 267, 269. 
 Himly, 22. 
 Hbhn, C., 53. 
 Hiifner, G., 19, 299. 
 
 JAEGI R, W., 76. 
 Joule, 18. 
 
 KACHLER, I., 39. 
 Kelvin, 70. 
 Krafft, F., 63. 
 
 LASCKLLES-SCOTT, 5. 
 Liebermann and Seyewetz, 195. 
 Linder, S. E., 76. 
 Linebarger, 62. 
 Lobry de Bruyn, 75, 76, 81. 
 Liidersdorf, 41. 
 
 M 
 
 MAGNAGHI, 27. 
 Markfeld, 14. 
 Marshall, 76. 
 Maquenne, 2. 
 Meyer, V., 105. 
 Minder, I., 21, 50, 53. 
 Mitchell, 18. 
 Mokiewsky, 26. 
 
 N 
 
 NERNST, 61. 
 Nobel, 150. 
 
 PARKES, 51. 
 
 Pawlewski and Filemonowicz, 267. 
 
 Payen, 6, 17, 47. 
 
 Pfeffer, 61. 
 
 Picton, H., 74. 
 
 REID, W. F., 150. 
 
 Renard, 265. 
 
 Rodewald and Kattein, 75. 
 
 21 
 
3'4 
 
 INDEX. 
 
 S 
 
 SACHS and Jonas, 149. 
 Schmulewitsch, 18. 
 Schulze, E., 146. 
 Schwanert, 39. 
 Seeligmann, 5, 6, 48. 
 Spiller, 9, 10, 41. 
 
 TEKBY, H., 3, 53. 
 
 Thomson, W., 53, 155, 212, 298. 
 
 Tilden, 23, 25. 
 
 Townsend, 81. 
 
 Tyndall, 18, 73. 
 
 U 
 
 ULZER and Horn, 145. 
 Ure, 6, 18. 
 
 VAN DER WAALS, 76. 
 
 W 
 
 WALLACE, 25, 29. 
 Wiesner, 17. 
 Williams, 6. 
 Wittorf, N., 25, 26. 
 Wolffenstein, R., 230. 
 Wroblewski, S., 18, 22. 
 Wiirtz and Schiitzenberger, 31 
 
 PRINTED BY NEILL AND CO., LTD., EDINBURGH. 
 
YC 18544 
 
 THE UNIVERSITY OF CALIFORNIA LIBRARY