Class. 
 Book. 
 
 
 Copyright^ .. 
 
 CflEffilGHT DKFOSm 
 
THE ANALYSIS OF 
 
 RUBBER 
 
 BY 
 
 JOHN B. TUTTLE 
 
 American Chemical Society- 
 Monograph Series 
 
 BOOK DEPARTMENT 
 The CHEMICAL CATALOG COMPANY, Inc. 
 
 19 EAST 24th STEEET, NEW YORK, U. S. A. 
 
 1922 
 Monograph 
 
Copyright, 1922, By 
 
 The CHEMICAL CATALOG COMPANY, Inc. 
 
 All Rights Reserved 
 
 ^ qa 
 
 CI.A6U004 4 
 
 Press of 
 
 J. J. Little & Ives Company 
 
 New York, U. S. A. 
 
 NOV -7 .922 
 
 

 GENERAL INTRODUCTION 
 
 American Chemical Society Series of 
 Scientific and Technologic Monographs 
 
 By arrangement with the Interallied Conference of Pure and 
 Applied Chemistry, which met in London and Brussels in July, 
 1919, the American Chemical Society was to undertake the pro- 
 duction and publication of Scientific and Technologic Mono- 
 graphs on chemical subjects. At the same time it was agreed 
 that the National Research Council, in cooperation with the 
 American Chemical Society and the American Physical Society, 
 should undertake the production and publication of Critical 
 Tables of Chemical and Physical Constants. The American 
 Chemical Society and the National Research Council mutually 
 agreed to care for these two fields of chemical development. 
 The American Chemical Society named as Trustees, to make the 
 necessary arrangements for the publication of the monographs, 
 Charles L. Parsons, Secretary of the American Chemical So- 
 ciety, Washington, D. C; John E. Teeple, Treasurer of the 
 American Chemical Society, New York City; and Professor 
 Gellert Alleman of Swarthmore College. The Trustees have ar- 
 ranged for the publication of the American Chemical Society 
 series of (a) Scientific and (b) Technologic Monographs by the 
 Chemical Catalog Company of New York City. 
 
 The Council, acting through the Committee on National 
 Policy of the American Chemical Society, appointed the editors, 
 named at the close of this introduction, to have charge of secur- 
 ing authors, and of considering critically the manuscripts pre- 
 pared. The editors of each series will endeavor to select topics 
 which are of current interest and authors who are recognized as 
 authorities in their respective fields. The list of monographs thus 
 far secured appears in the publisher's own announcement else- 
 where in this volume. 
 
 The development of knowledge in all branches of science, and 
 
 3 
 
4 GENERAL INTRODUCTION 
 
 especially in chemistry, has been so rapid during the last fifty 
 years and the fields covered by this development have been so 
 varied that it is difficult for any individual to keep in touch with 
 the progress in branches of science outside his own specialty. In 
 spite of the facilities for the examination of the literature given 
 by Chemical Abstracts and such compendia as Beilstein's Hand- 
 buch der Organischen Chemie, Richter's Lexikon, Ostwald's Lehr- 
 buch der Allgemeinen Chemie, Abegg's and Gmelin-Kraut's 
 Handbuch der Anorganischen Chemie and the English and 
 French Dictionaries of Chemistry, it often takes a great deal of 
 time to coordinate the knowledge available upon a single topic. 
 Consequently when men who have spent years in the study of 
 important subjects are willing to coordinate their knowledge and 
 present it in concise, readable form, they perform a service of 
 the highest value to their fellow chemists. 
 
 It was with a clear recognition of the usefulness of reviews 
 of this character that a Committee of the American Chemical 
 Society recommended the publication of the two series of mono- 
 graphs under the auspices of the Society. 
 
 Two rather distinct purposes are to be served by these mono- 
 graphs. The first purpose, whose fulfilment will probably ren- 
 der to chemists in general the most important service, is to pre- 
 sent the knowledge available upon the chosen topic in a readable 
 form, intelligible to those whose activities may be along a wholly 
 different line. Many chemists fail to realize how closely their 
 investigations may be connected with other work which on the 
 surface appears far afield from their own. These monographs 
 will enable such men to form closer contact with the work of 
 chemists in other lines of research. The second purpose is to pro- 
 mote research in the branch of science covered by the mono- 
 graph, by furnishing a well digested survey of the progress 
 already made in that field and by pointing out directions in which 
 investigation needs to be extended. To facilitate the attainment 
 of this purpose, it is intended to include extended references to 
 the literature, which will enable anyone interested to follow up 
 the subject in more detail. If the literature is so voluminous that 
 a complete bibliography is impracticable, a critical selection will 
 be made of those papers which are most important. 
 
 The publication of these books marks a distinct departure in 
 the policy of the American Chemical Society inasmuch as it is 
 
GENERAL INTRODUCTION 5 
 
 a serious attempt to found an American chemical literature with- 
 out primary regard to commercial considerations. The success 
 of the venture will depend in large part upon the measure of 
 cooperation which can be secured in the preparation of books 
 dealing adequately with topics of general interest; it is earnestly 
 hoped, therefore, that every member of the various organizations 
 in the chemical and allied industries will recognize the importance 
 of the enterprise and take sufficient interest to justify it. 
 
 AMERICAN CHEMICAL SOCIETY 
 
 BOARD OP EDITORS 
 
 Scientific Series: — Technologic Series: — 
 
 William A. Noyes, Editor, John Johnston, Editor, 
 
 Gilbert N. Lewis, C. G. Derick, 
 
 Lafayette B. Mendel, William Hoskins, 
 
 Arthur A. Noyes, F. A. Lidbury, 
 
 Julius Stieglitz. Arthur D. Little, 
 
 C. L. Reese, 
 C. P. Townsend. 
 
American Chemical Society 
 
 MONOGRAPH SERIES 
 
 Other monographs in the series of which this book is a part now 
 ready or in process of being- printed or written. 
 
 Organic Compounds of Mercury. 
 
 By Frank C. Whitmore. 397 pages. Price $4.50. 
 Industrial Hydrogen. 
 
 By Hugh S. Taylor. 210 pages. Price $3.50. 
 The Chemistry of Enzyme Actions. 
 
 By K. George Falk. 140 pages. Price $2.50. 
 Tri(Z VztciwviThs 
 
 By H. C. Sherman and S. L. Smith. 273 pages. Price $4.00. 
 The Chemical Effects of Alpha Particles and Electrons. 
 
 By Samuel C. Lind. 180 pages. Price $3.00. 
 Zirconium and Its Compounds. 
 
 By F. P. Venable. Price $2.50. 
 The Properties of Electrically Conducting Systems. 
 
 By Charles A. Kraus. Price $4.50. 
 The Origin of Spectra. 
 
 By Paul D. Foote and F. L. Mohler. 250 pages. Price $4.50. 
 Carotinoids and Related Pigments: The Chromolipoids. 
 
 By Leroy S. Palmer. 305 pages. Price $4.50. 
 Thyroxin. By E. C. Kendall. 
 
 The Properties of Silica and the Silicates. By Robert B. Sosman. 
 Coal Carbonization. By Horace C. Porter. 
 The Corrosion of Alloys. By C. G. Fink. 
 Piezo-Chemistry. By L. H. Adams. 
 Cyanamide. By Joseph M. Braham. 
 Liquid Ammonia as a Solvent. By E. C. Franklin. 
 Wood Distillation. By L. F. Hawley. 
 Solubility. By Joel H. Hildebrand. 
 The Animal as a Converter. By H. P. Armsby and C. Robert 
 
 Moulton. 
 Glue and Gelatin. By Jerome Alexander. 
 Organic Arsenical Compounds. By George W. Raiziss and Jos. 
 
 L. Gavron. 
 
 Valence, and the Structure of Atoms and Molecules. By Gil- 
 bert N. Lewis. 
 Shale Oil. By Ralph H. McKee. 
 
 Aluminothermic Reduction of Metals. By B. D. Saklatwalla. 
 The Chemistry of Leather Manufacture. By John A. Wilson. 
 Absorptive Carbon. By N. K. Chaney. 
 Refining Petroleum. By George A. Burrell, et al. 
 Extraction of Gasoline from Natural Gas. By George A. Burrell. 
 Chemistry of Cellulose. By Harold Hibbert. 
 
 The CHEMICAL CATALOG COMPANY, Inc. 
 19 EAST 24th STREET, NEW YORK, U. S. A. 
 
PREFACE 
 
 The tendency of the industry is towards simplification of 
 methods and materials. The readjustment of conditions to the 
 basis of an adequate supply of crude rubber — a condition which 
 did not obtain twenty years ago — has by the operation of natural 
 economic laws eliminated from general use many pigments, rub- 
 ber substitutes, and low grades of rubber. These are not likely to 
 return, and we may dismiss them from consideration, confining 
 ourselves to the materials found in the every day life of the 
 industry as it exists today. 
 
 In any work on analysis, and especially on industrial sub- 
 stances, it is impossible to avoid the presentation of the subject 
 from a very personal point of view. Many methods, and modifi- 
 cations of methods, are written on a single phase of the analysis, 
 with a great variety of purposes back of them. In the analysis 
 of rubber, methods have been published because they were shorter 
 than existing ones; some used less expensive materials, or more 
 simple equipment; and some because they really were an improve- 
 ment. Few of these methods were thoroughly developed before 
 publication; the user must discover for himself the limits of error 
 and applicability. It is usually safer to hold fast to such methods 
 as have stood the test of time, and whenever there may be any 
 methods for any part of a rubber analysis which are not included 
 herewith, it is because data are lacking as to their ability to 
 accomplish the desired purpose. The omission does not imply 
 lack of merit, but merely that sufficient experimental evidence is 
 not yet forthcoming to warrant an unqualified approval. 
 
 Primarily, this monograph is addressed to the chemists in the 
 consumers' laboratories, and to those who, without any previous 
 experience in the technology or analysis of rubber, may be called 
 upon to deal with a problem in which the composition of rubber 
 may play a more or less important part. Nevertheless, it is the 
 author's hope that it may not come amiss to those colleagues 
 comprizing the technical staff of the laboratories of the manu- 
 facturing plants, who may find it desirable to study a competi- 
 
 7 
 
8 PREFACE 
 
 tor's product, or who may be required to produce materials to 
 accord with the consumers' specifications. 
 
 In view of the probability of this monograph reaching chemists 
 of limited experience in the technology of rubber, Appendix A, 
 on the methods of preparation of rubber compounds, and Ap- 
 pendix B, on the physical testing of rubber, have been added. 
 These appendices are necessarily elemental in character, but they 
 may serve as connecting links between these subjects, and the 
 chemistry of the analysis of rubber. 
 
 J. B. T. 
 
TABLE OF CONTENTS 
 
 PAGE 
 
 Preface 7 
 
 CHAPTER 
 
 I The Purpose of Rubber Analysis; What Is Rub- 
 ber; the Need for Chemical Analysis of 
 Rubber 11 
 
 II The Composition of Crude Rubber; Constitu- 
 ents of Crude Rubber Other Than the 
 Rubber Hydrocarbons 16 
 
 III The Preparation of Rubber Compounds; Crude 
 
 Rubber; Reclaimed Rubber; Oil Substitutes; 
 Mineral Rubber ; Mineral Hydrocarbons ; Oils 
 and Waxes; Vulcanizing Materials; Organic 
 Accelerators; Inorganic Accelerators; Inor- 
 ganic Fillers 26 
 
 IV Theory of Vulcanization; Cold Vulcanization; 
 
 Vulcanization With Mixed Gases; Ostro- 
 muislenskii's Theories of Vulcanization . 57 
 
 V Sampling 64 
 
 VI Extractions; Acetone Extract; Chloroform Ex- 
 tract; Alcoholic Potash Extract; Analysis 
 of the Acetone Extract 68 
 
 VII The Determination of Rubber; the Tetrabro- 
 mide Method; the Nitrosite Method; the 
 Indirect Methods; Difference Methods . . 76 
 
 VIII Sulfur Determinations; Total Sulfur; Free 
 Sulfur; Sulfur of Vulcanization; Sulfur in 
 Fillers 84 
 
 IX Detection of Organic Accelerators .... 94 
 
10 TABLE OF CONTENTS 
 
 CHAPTER PAGE 
 
 X Mineral Analysis; Special Determinations; 
 
 Specific Gravity 97 
 
 XI Micro-sectioning and Microphotography . . . 114 
 
 XII Calculation to Approximate Formulas . . . 117 
 
 Bibliography 121 
 
 Appendix A Preparation of Materials for Rubber 
 
 Manufacture 139 
 
 " B Physical Tests 143 
 
 " C Table of Specific Gravities 149 
 
 Index 153 
 
THE ANALYSIS OF RUBBER 
 
 ■ 
 
 Chapter I. 
 
 The Purpose of Rubber Analysis. 
 
 The growth of the rubber industry has been tremendous, espe- 
 cially, so far as volume is concerned, since the advent of the 
 pneumatic tire. More than any one other cause, the resiliency 
 afforded by the pneumatic bicycle tire was responsible for the 
 wide spread popularity of the bicycle, and the rubber automobile 
 tire has played an equally, if not more important role in the 
 development of motor driven vehicles. In the production of the 
 various rubber articles, besides the essential rubber and sulfur 
 which make up the vulcanized rubber, we need a vast volume of 
 pigments or fillers, because by their use we may modify the 
 properties of the vulcanized rubber so as to attain a degree of 
 service which would otherwise be impossible. We must not, 
 therefore, look upon these added substances as adulterants, or 
 even mere diluents, but as integral parts of the whole, for by 
 their service they have earned the right to due consideration. 
 Some, it is true, are largely valuable, owing to the fact that they 
 lower the cost of the products in which they are included. The 
 rubber industry makes big demands upon the producers of raw 
 materials, such as zinc oxide and sulfide, lead compounds, carbon 
 black, magnesium oxide, and talc, and were we forced to depend 
 upon these pigments alone, the costs would soon rise to prohibi- 
 tive heights, with concomitant injury not merely to the rubber 
 industry, but to others as well, such as paints and inks, which 
 depend for their existence upon an adequate supply of these same 
 pigments. We are thus doubly obliged to seek far and wide for 
 new materials which will accomplish one of two things: produce 
 the same or even better quality at a lower cost, or a better 
 quality at the same cost. 
 
 In every line, there is a more or less clearly defined standard 
 
 11 
 
12 THE ANALYSIS OF RUBBER 
 
 of service, and the future of the industry is quite definitely tied 
 up with the results obtained in the present by the attainment of 
 that standard. In order to use correctly any material, it is 
 important that we know the degree of purity of the commercial 
 grades, the influence of possible impurities upon the quality of the 
 products, and, most important of all, the degree of uniformity 
 obtainable from one period to another. These data can be 
 secured only through careful and persistent testing of the raw and 
 finished materials. 
 
 It is not our purpose here to undertake a description of the 
 functions and usage of the various materials to be mentioned 
 later, but merely to discuss them from the point of view of their 
 chemical properties. For the various other phases of the subject, 
 the reader is referred to the bibliography which is included here- 
 with. 
 
 What Is "Rubber"? 
 
 Probably few words in general usage are applied as generally 
 as the word rubber. Strictly speaking, the word belongs to the 
 polyterpene having the formula (C 10 H 16 ) X . We know, however, 
 that there is an homologous series of these polymerized products 
 differing from each other by the constant quantity 2 CH 2 , and 
 these products have so many of the qualities peculiar to 
 (C 10 H 16 ) X that they too are called rubber. Thus we may say that 
 there is a rubber series analagous to the paraffin series, etc. 
 The planters who cultivate the plantations call their product rub- 
 ber, or crude rubber, although, in addition to the polyterpene, 
 there is 2% and upwards of acetone-soluble substances called 
 resins; 2 to 6% of nitrogenous substances, akin to the proteins; 
 and small amounts of substances possessing the properties of 
 catalysts of the vulcanization process. The manufacturer pro- 
 duces rubber products, although in addition to the rubber as 
 obtained from the plantations, many other substances are added, 
 both organic and inorganic, because of certain qualities which 
 such additions produce in the finished article. Moreover, chemi- 
 cally speaking, we have in the hot vulcanized articles an entirely 
 new series of products, viz., the sulfur addition products of the 
 polyterpene, a series which passes from the extreme of pure 
 rubber at one end to a constant composition of (C 10 H 16 S 2 )x at 
 the other end. 
 
THE PURPOSE OF RUBBER ANALYSIS 13 
 
 In order to avoid confusion, for our own purposes we will use 
 the term "rubber" to mean any mixture of (C 10 H 16 ) X (its homo- 
 logues are negligible commercially, at present) with any other, 
 substances, in either the vulcanized or unvulcanized state. 
 "Crude rubber" will apply only to the materials as obtained from 
 the rubber trees; and where the polyterpene itself is indicated, 
 we will use the term "rubber hydrocarbon." "Rubber compound" 
 will be used to indicate the formula of a commercial mixing. 
 
 From an analytical point of view, it is of little consequence 
 whether the unit in the molecule of rubber is C 5 H 8 or C 10 H 16 . 
 There seems to be a preponderance of evidence in favor of the 
 latter; in any event, we do know that while rubber can be 
 synthesized from isoprene, CH 2 :C(CH 3 ) .CH:CH 2 , the rubber 
 molecule itself contains only one double bond for each group of 
 C B H 8 . By vulcanizing rubber with a large excess of sulfur, C. 0. 
 Weber obtained a hard rubber corresponding to C 10 H ltt S 2 ) x . 
 With bromine, rubber has been found to form (C 10 H 16 Br 4 ) x . 
 These experiments have been repeated so many times, that there 
 seems to be no necessity for further argument as to the existence 
 of the two double bonds. The importance of this fact is seen 
 when we realize that this fact is the basis upon which have been 
 built the two classes of direct determinations of rubber, the 
 tetrabromide, and the nitrosite, both of which will be discussed 
 in their proper place. 
 
 The Need for Chemical Analysis of Rubber. 
 
 Any scheme which may be suggested for the analysis of rubber, 
 either vulcanized or unvulcanized, must take into consideration 
 the fact that there are two groups, with widely differing points 
 of view, which are interested in the subject. We have first the 
 manufacturers' chemists, who test their own products to deter- 
 mine what changes have taken place during the process of manu- 
 facture; and their competitors, products to ascertain what the 
 latter are using. Analysts of this group may use methods which 
 their knowledge of the subject tells them will give accurate 
 results, even though it is known that such methods are not uni- 
 versally applicable. The second group embraces the consumers, 
 who, as a rule, are endeavoring to learn whether or not the article 
 complies with certain stipulated requirements, or specifications, 
 
14 THE ANALYSIS OF RUBBER 
 
 Here a definite procedure is obligatory, for in order to avoid dis- 
 putes, the specifications usually (and if they do not, they 
 should), 1 contain a more or less detailed description of the analyt- 
 ical methods. Since the composition is unknown, it is clear that 
 the methods in use by the consumers should be as nearly uni- 
 versally applicable as it is possible to make them. From time to 
 time, there have appeared suggestions for making analyses of 
 rubber compounds, but too frequently the authors have neglected 
 to take into consideration these different points of view, and this 
 omission has materially reduced the value of the suggestions. 
 
 Rubber is a hydrocarbon of the terpene family, existing in a 
 polymerized form, and having the composition (C 10 H 16 ) X . The 
 size of the molecule is unknown, although it is believed to be quite 
 large, but we do know that each group of C 10 H 16 contains two 
 double bonds. Double bonds are unstable, and there is always a 
 tendency for such double bonds to add various elements, or 
 group of elements, which will tend to produce a more stable 
 form. Thus we find that the double bonds of rubber may take 
 up oxygen, ozone, sulfur, selenium, sulfur chloride, chlorine, 
 bromine, etc., producing new chemical substances with distinctly 
 new properties, many of which are more useful in a commercial 
 sense than the original substance. Industrially, the most im- 
 portant of these compounds are those formed by the addition of 
 sulfur, and sulfur monochloride, the chemical process being 
 termed "vulcanization," or "curing." Crude rubber is a soft, 
 plastic substance, soluble in naphtha, benzene, chloroform, carbon 
 bisulfide, from which, by the simple process of evaporation, it 
 may be recovered in its original form. The addition of compara- 
 tively small amounts of sulfur is sufficient to destroy the solu- 
 bility in these solvents. Such vulcanized compounds can, by pro- 
 longed heating, be brought into solution in various solvents, but 
 there is this distinction, that, in the latter case, the solution is 
 accompanied by a depolymerization, and evaporation of the sol- 
 
 1 Ttrere is such a diversity of opinion concerning the best method for any 
 single determination, and since the interpretation of the analysis, rather than 
 the absolute results obtained, is the more important of the two, it must neces- 
 sarily follow that the results of the analysis are inseparable from the method 
 by which they were obtained. It is not sufficient merely to say that a sample 
 has 3.00% of sulfur, it must be stated that it has 3.00% of sulfur when deter- 
 mined by a certain method. This has been one of the glaring weaknesses of 
 the average specification in this country, and has been the cause of a great 
 deal of controversy and actual financial loss. 
 
THE PURPOSE OF RUBBER ANALYSIS 15 
 
 vent will not give us the rubber in the same condition in which it 
 existed before solution. Rubber containing only a small quantity 
 of combined sulfur is tough and elastic, but as the percentage of 
 combined sulfur increases, the degree of extensibility becomes less 
 and less, the rubber becomes harder until we obtain the familiar 
 substance, vulcanite, or hard rubber, and the limit of sulfur addi- 
 tion is found at (C 10 H 16 S 2 ) x . To effect the combination between 
 the rubber and sulfur, catalysts are employed, both organic and 
 inorganic, while to produce the desired properties in the finished 
 article, various oils, waxes, gums and pigments are added. 
 
 It would seem, therefore, to be quite apparent that in order to 
 understand the analysis of rubber, one must be familiar with the 
 materials which enter into the rubber compounds, the chemical 
 changes which take place during vulcanization, as well as merely 
 the analytical methods. In this way, the analysis may be 
 directed towards bringing out the really important points in the 
 rubber compound. The general scheme which has been adopted, 
 is to first present a description of the raw materials, the methods, 
 and the theories of vulcanization. This will be followed by direc- 
 tions for sampling, general and specific methods of analysis, and 
 finally some suggestions for interpreting the results of the analy- 
 sis, with the view to reconstructing the formula of the rubber 
 compound. 
 
Chapter II. 
 The Composition of Crude Rubber. 
 
 Crude rubber is obtained from various trees, shrubs, or vines. 
 Some of these grow wild, and others are cultivated for the sake of 
 their yield of rubber. Twenty years ago cultivated or plantation 
 rubber was practically unknown; the crude rubber of that time 
 was obtained from all quarters of the tropical world, Brazil fur- 
 nishing the greater portion of the wild rubber. Not only did 
 Brazil furnish the major part of the rubber, but it was also the 
 best in quality, largely because of the care taken in preparation, 
 and the uniformity achieved in spite of the rather crude methods 
 which were employed. One reason for this uniformity was that 
 most of the rubber was obtained from a single species, the Hevea 
 Braziliensis, which today is not only the source of the best wild 
 rubber, but, through transplanting, is also the chief, one might 
 almost say the entire, source of the plantation rubber as well. 
 The Hevea rubber became known commercially as Para rubber, 
 from the port from which shipments were made. 
 
 Para Rubber. Two main subdivisions are made in Para rub- 
 ber, the Up-river, and Islands. The former comprises rubber 
 collected in the inland section, along the Amazon river and its 
 branches. The Islands rubber is so named because it is largely 
 collected in the islands of the delta of the Amazon, and the ad- 
 jacent country. There are subdivisions of the main group, the 
 Up-river including Acre, Bolivian, Madeira, Manaos, etc., and 
 the Islands rubber is similarly subdivided. 
 
 Rubber comes from the latex of the trees, and the latex is 
 gathered by making a number of small cuts extending just below 
 the bark. The latex flows from these cuts, and is caught in small 
 cups. The rubber gatherer collects the latex daily, takes it to his 
 hut, and prepares it for the market by the process of coagulation 
 and smoking. A paddle is dipped into the latex, and the thin 
 film which remains on the surface is coagulated by holding it over 
 
 16 
 
THE COMPOSITION OF CRUDE RUBBER 17 
 
 the smoke from burning uri-curi nuts. The heat and smoke break 
 down the emulsion, separating the rubber from the so-called 
 serum of the latex. Much of the serum drips out, but a consider- 
 able portion is retained, and the solids contained therein become 
 a part of the crude rubber, profoundly influencing the vulcaniza- 
 tion and the physical properties. The operation of dipping (or 
 the latex may be poured over the paddle) and smoking is con- 
 tinued until a fair sized ball is obtained. The rubber so prepared 
 is called Fine Para, but if for any reason fermentation or oxida- 
 tion should set in, and the rubber become sticky, it is classed with 
 the lower grades. The scrap from the cups, buckets, and from the 
 bark of the trees, is gathered together, and called "Coarse 
 Para." The shape and general appearance of these "balls" varies 
 widely, but the method of preparation is the same throughout, so 
 that there actually exists a remarkably uniform method of prep- 
 aration throughout the entire territory where the Para rubber is 
 gathered. 
 
 Castilloa. Second among the wild rubbers is that obtained 
 from the Castilloa Ulei or Castilloa Elastica which produce the 
 kinds known as Caucho, Centrals, etc. This rubber is coagulated 
 in bulk, is not smoked, and appears on the market as balls, 
 sheets, strips, or slabs. It is subdivided into grades, but, even in 
 the best, there is nothing like the uniformity of quality which 
 one finds in the Para rubber. 
 
 African Rubbers. African rubbers are largely gathered from 
 vines, chiefly the Landolphia, with innumerable sorts and grades 
 many of which are quite indistinguishable, even to the expert, and 
 certainly cannot be identified after vulcanization. Some African 
 rubbers are prepared according to methods peculiar to the place, 
 by means of which they can be identified, or they may gather 
 from the means by which they are coagulated an odor peculiarly 
 their own, but the differences from one lot of the same name to 
 the next is frequently greater than that between two entirely 
 different sorts. During the past year there has been a decided 
 diminution in the quantity of African rubber produced, and many 
 sorts have entirely disappeared from the market as their quality 
 is so poor that the price they will bring on the present day 
 markets is not sufficient to pay the cost of collecting. If the 
 Plantation rubber continues to increase at anywhere near the 
 rate it has for the past eight or ten years, it will mean such a low 
 
18 THE ANALYSIS OF RUBBER 
 
 standard market price for the best grades of rubber that the 
 poorer African sorts will disappear altogether. From the experi- 
 ences which the manufacturers have had in trying to produce 
 uniform quality material with such stuff, we may surmise that 
 no tears will be shed at the loss. 
 
 Guayule. Guayule is the rubber obtained from the shrub 
 Parthenium Argentatum, found extensively in Mexico and Texas. 
 This rubber is not obtained in the form of a latex, but the 
 plants are cut down, and the rubber which exists in the stems, 
 leaves, and branches of the plant, is separated by mechanical or 
 chemical means, or both. The crude Guayule thus obtained runs 
 very high in resins and other impurities ; indeed, these form about 
 two thirds of the crude rubber. It usually undergoes a process 
 of purification, or deresinification in order to prepare it for the 
 market, whereby the rubber hydrocarbon content is raised to 
 somewhere around 75%, or even higher. Guayule is a soft, 
 sticky, stretchy rubber, retaining these properties to a high 
 degree even after vulcanization, and it finds its chief use as a con- 
 stituent of frictions. 
 
 Pontianak. Java, Borneo, and the neighboring countries, pro- 
 duce a tree, the Dyera Costulata, which yields a product contain- 
 ing about 90% of resins and similar substances, and about 10% 
 of rubber. This mixture is known chiefly as Pontianak, or Jelu- 
 tong rubber. In the crude form, it is quite hard, owing to the 
 high resin content, and particularly to the nature of the resin. 
 In the process of purification of crude Pontianak, a large part of 
 this resin is removed, and is marketed separately. Pontianak 
 resin finds some use in rubber mixings; it is hard, brittle resin, 
 with a conchoidal fracture, very much resembling our ordinary 
 rosin. It is soluble in acetone, chloroform, benzene, and other 
 organic solvents, and consists largely of unsaponifiable hydrocar- 
 bons. Ellis and Wells 1 find that on heating, the solubility of the 
 resin and the percentage of unsaturated compounds increase. 
 While there is some demand commercially for this resin, it does 
 not appear to be sufficiently extensive and remunerative to permit 
 much Pontianak rubber to come to this country. At the present 
 prevailing market prices, it seems obvious that the rubber portion 
 must be handled as a by-product only. 2 
 
 1 J. Ind. Eng. Chem. 7, 747-50 (1915). 
 
 2 As an indication of the disappearance of Pontianak rubber from the market, 
 it is only necessary to note that according to reasonably reliable statistics, only 
 
THE COMPOSITION OF CRUDE RUBBER 19 
 
 When the resin content is materially reduced, Pontianak rub- 
 ber is very tacky, and plastic, making it difficult to store, as it 
 has the tendency to flow together to form one huge, unmanageable 
 mass. 
 
 Plantation Rubbers. The development of the Hevea on the 
 plantations of the Far East, has reached such proportions as to 
 make it the dominating feature of the rubber market. Fifteen 
 years ago, plantation rubber was of small commercial importance, 
 very little of it being produced. Today, the plantations furnish 
 fully 80% of the world's supply. The rapidity of the growth is 
 well illustrated in the following figures, which while they may not 
 be absolutely accurate, are sufficiently so to show the rapidity of 
 the growth of this phase of the industry : 
 
 Production of Plantation Rubber. 
 
 Tons 
 
 1903 25 
 
 1904 50 
 
 1905 150 
 
 1906 500 
 
 1907 1,000 
 
 1908 2,000 
 
 1909 4,000 
 
 1910 8,000 
 
 1911 15,000 
 
 1912 30,000 
 
 1913 50,000 
 
 1914 75,000 
 
 1915 110,000 
 
 1916 160,000 
 
 1917 225,000 
 
 1918 190,000 
 
 1919 360,000 
 
 The time has arrived when cultivated rubber can be produced 
 so cheaply that the poorer grades of wild rubber have been 
 forced out of the market, and even the better grades have suffered 
 severely. The analyst may therefore expect less and less to be 
 confronted with samples for analysis which have been made up 
 wholly, or in great part, of wild rubbers. Only in the Para grades 
 does there seem to be any sort of adherence to the old grades of 
 wild rubber. There are still some specifications for various ma- 
 terials, which insist upon the use of Fine Para rubber (although 
 unless some representative of the purchaser actually sees the 
 
 1000 tons were imported during 1921. During the same period, crude rubber 
 imports were estimated to be between 275,000 and 300,000 tons. In 1905-6, the 
 ratio of imports was 2 tons of crude rubber to one ton of Pontianak. 
 
20 THE ANALYSIS OF RUBBER 
 
 material made, how they are going to distinguish good smoked 
 sheets from Fine Para is more than one can say), and they are 
 unwilling to change over to plantations because they do not know 
 what the effect of such a change would make on the life of the 
 articles. Some rubber specialties have been made from the same 
 formulas, calling for Para grades, for a number of years, and still 
 continue to be made in this fashion, although at times it is 
 difficult to get just the grades of wild rubber needed. 
 
 Smoked Sheet. Although at times it does not command the 
 highest price, it is the standard grade of plantation rubber. 3 The 
 rubber should be clean, dry, firm, of a good color and free from 
 more than traces of mold or rust. The moisture content will vary 
 between 0.3% and 1.0%. The acetone extract will usually be 
 between 2.5% and 3.0%, and almost always will be below 4%. 
 The ash should be negligible. 
 
 Pale Crepe. Pale crepe is frequently called first latex, al- 
 though the same latex may, at the choice of the plantations, be 
 made into either ribbed smoked sheet, or pale crepe. The latter 
 is usually cleaner than smoked sheet; chemically, they are very 
 much alike. The moisture content will average lower than 
 smoked sheet, the ash is negligible, and the resin content between 
 2.5% and 4.0%. 
 
 Smoked Crepe. Smoked crepe is usually cleaner than smoked 
 sheet (the latter frequently contains bark, etc.), with a lower 
 moisture content, approaching that of pale crepe. The resins 
 seem to run about the same; if anything, a bit higher than the 
 average of smoked sheets. No other particular differences have 
 been noted. 
 
 Amber Crepe. Amber crepe comes in several grades, according 
 to color. There is no sharp dividing line between these grades 
 and the pale crepe, or even amongst themselves. Some of the 
 lighter amber crepes are very much like the poorer lots of pale 
 crepe. The resins, moisture, and ash in the paler colored amber 
 crepes is about the same as for pale crepe or smoked sheet; the 
 lower grades are apt to be sticky, run high in dirt and moisture, 
 and by reason of surface oxidation, they may be tacky and show 
 a higher acetone soluble figure. 
 
 Roll Brown Crepe. Roll brown crepe comes into the market 
 
 3 For the methods of preparation of Smoked Sheet, and Crepe, cf. Whitby, 
 "Plantation Rubber, and the Testing of Rubber." 
 
THE COMPOSITION OF CRUDE RUBBER 21 
 
 in the form of sheets of crepe which have been rolled up into 
 small bundles about 5 to 6 inches in diameter, and about 10 to 15 
 inches in length. It is the lowest grade of plantation rubber on 
 the market, is very tacky, and dirty, and must always be washed 
 in the factory before it can be used. When washed clean, and 
 dried, it replaces acceptably the wild rubbers which have been 
 used in friction stocks, such as Guayule, etc. 
 
 Constituents of Crude Rubber, Other Than the 
 Rubber Hydrocarbons. 
 
 We have already drawn attention to that portion of the crude 
 rubber which is soluble in acetone, and which is known com- 
 mercially as rubber resins. Apart from the dirt, bark, and water, 
 which may be included in crude rubber, but which we cannot 
 consider as anything but contamination, there are some other sub- 
 stances, which are not rubber, but are nevertheless found in all 
 crude rubbers. 
 
 Resins. Hevea rubber contains, in addition to the rubber 
 hydrocarbons from 2% to 4% of resins. These resins are about 
 80% saponifiable, and 20% unsaponifiable. They are soluble in 
 acetone, alcohol, chloroform, and many other organic solvents. 
 The solution is usually a pale yellow color, and the residue, when 
 the solvent has been driven off, is light colored with the consist- 
 ency of butter. In the unsaponifiable portion, Whitby* has 
 identified some five substances from the unsaponifiable portion, 
 some of which show optical activity, and some give sterol reac- 
 tions. The acetone extract of Hevea rubber may go higher than 
 4%, but this does not necessarily mean that the resin content is 
 high, but rather that there has been oxidation and depolymeriza- 
 tion of the rubber, producing by-products which also are soluble 
 in acetone. 
 
 Insoluble Matter. If we take a sheet of pale crepe, smoked 
 sheet, etc., and dissolve it in gasoline, being careful not to shake 
 too much, we will find flakes of the crude rubber which will not 
 dissolve. This is what is known as the "insoluble matter." The 
 amount will vary with the method of preparation; analyses have 
 run between 2% and 6%. Rubber prepared by the total evapora- 
 
 * Paper read at the Spring meeting of the American Chemical Society at 
 Rochester, April 1921. "Contribution to the knowledge of the resins of Hevea 
 rubber," by G. Stafford Whitby and J. Doolid. 
 
22 THE ANALYSIS OF RUBBER 
 
 tion of the latex will have the highest figure, whereas the ordinary- 
 methods of coagulation with acetic acid, washing, etc., reduce this 
 figure considerably. The insoluble matter resembles the proteins, 
 and, according to Eaton, its fermentation will permit the forma- 
 tion of nitrogenous decomposition products which act as acceler- 
 ators of vulcanization. Such reactions take place in the so-called 
 slab rubber, in which the coagulum is only slightly pressed, and 
 which retains a large amount of the non-soluble substances in 
 the latex. 
 
 While the insoluble matter may be shown by treating the orig- 
 inal sheet with gasoline as described above, it is next to impossible 
 to wash out all of the rubber, so that we cannot depend upon 
 this separation as a means of a quantitative separation. The 
 nitrogen factor is obtained by dividing the weight of the nitrogen- 
 containing substance by the nitrogen it contains; one determines 
 the nitrogen and multiplies by this factor to arrive at the total 
 amount of nitrogen-substance present. This factor varies con- 
 siderably, but 6.25 is a fair average, and will give results near 
 enough to the truth to be acceptable for all practical purposes. 
 In the determination of glue by the Kjeldahl method, this in- 
 soluble matter appears as a conflicting element in the determina- 
 tion, and must be taken into account. 
 
 The best rubbers are clean and dry, and have practically no 
 ash. A high ash indicates a rubber which has been poorly 
 washed or which has since picked up dirt, sand, etc. 
 
 There are usually small amounts of substances, whose composi- 
 tion we do not know, but which we recognize by the fact that they 
 act as accelerator of the vulcanization process. In amount, they 
 are negligible, except in the case of compounds composed entirely 
 of rubber and sulfur, when their presence or absence may bear an 
 important part in securing the proper degree of vulcanization. 
 
 Tests for Crude Rubber. 
 
 Crude rubber may contain dirt, bark, moisture, resins, proteins, 
 and oxidized or depolymerized rubber. Bark, dirt, moisture, and 
 any water-soluble substances, are grouped together as "loss on 
 washing." 
 
 Loss on Washing. For plantation rubbers, in which the mois- 
 ture and dirt is usually very low, a 5 lb. to 10 lb. sample will suf- 
 
THE COMPOSITION OF CRUDE RUBBER 23 
 
 fice. The sample should be taken in small pieces from different 
 parts of the lot, and at least every five cases should be sampled. 
 If the sample thus taken proves to be too large to handle, it can 
 be weighed, broken down on the mill, and a smaller sample taken 
 from this broken down rubber. The latter should be weighed 
 when cool, in order to ascertain whether or not any loss in weight 
 has taken place. For wild rubbers, not less than 50 lbs., and pref- 
 erably 100 lbs. should be taken for the loss in washing test; 
 afterwards, for the other determinations, a smaller sample may 
 be drawn from the washed and dried rubber. Even greater care 
 must be exercised in sampling wild rubber, because of the uneven- 
 ness in size, cleanliness, moisture, etc., of the various balls or lots 
 of wild rubber. Fine Para, for example, may be sampled by cut- 
 ting the balls into quarters, until about 50 lbs. are obtained. 
 Dirtier rubbers, or those which will vary more from lot to lot, 
 should be sampled up to 100 lbs. In a later chapter, we propose 
 to deal more at length with this subject of sampling, but suffice 
 it to say here that unless the proper care is exerted to make the 
 sample drawn for this test one which is of the same average 
 quality as the lot, the entire work of testing is worse than if it 
 were not done at all, for it may lead to totally false results. The 
 rubber should be washed immediately after the sample has been 
 drawn and weighed. 
 
 Plantation rubber may be washed directly, without any pre- 
 vious treatment; wild rubbers should be heated in hot water to 
 soften them, and render them more plastic, so as to facilitate the 
 operation. The rubber is washed in the usual factory manner, 
 and then dried in a vacuum dryer. After removal from the 
 vacuum dryer, the rubber is cooled, and weighed, and the loss 
 noted. 
 
 A new sample of about 1000 grams is taken from different parts 
 of the washed and dried sample, and united by passing several 
 times through a laboratory mill. Five grams are weighed out, 
 sheeted thin on the laboratory mill (care must be taken to see 
 that no mechanical loss occurs), and dried to constant weight at 
 100C. A laboratory vacuum oven may be used, but the tempera- 
 ture should be less than 100C, since with the reduced pressure 
 the higher temperature is not necessary, and there is less likeli- 
 hood of damage to the rubber at the lower temperature. The loss 
 on drying the 5 gr. sample, plus the shrinkage during washing, 
 
24 THE ANALYSIS OF RUBBER 
 
 gives the total loss in weight, and should be calculated to per- 
 centage, based upon the original weight of the sample. 
 
 Resins. Sheet out thin, 5 gr. of rubber, 5 calculated to the 
 dry basis, and wrap in filter paper which has previously been 
 extracted with acetone, place in the extraction flask, 6 and extract 
 continuously with acetone for eight hours. Remove the solvent, 
 dry the flask and contents to constant weight at 90C and calcu- 
 late to percentage. The color, hardness, and odor of the extract 
 should be noted. 
 
 Moisture. It is sometimes desirable to know simply the mois- 
 ture in the original sample. This is not practicable with most 
 wild rubbers, where the moisture is very unevenly distributed, 
 but with plantation rubbers it is quite feasible, and often a 
 valuable figure. 
 
 Cut up 5 grams into small pieces, dry to constant weight in an 
 inert atmosphere at 90C. Calculate to percentage. 
 
 Nitrogen. A 1 gr. sample is placed in a Kjeldahl flask, with 
 10 gr. of potassium sulfate, 50 cc. of cone, sulfuric acid and 1 gr. 
 of copper sulfate. Heat for three to four hours (it is not neces- 
 sary for the solution to become clear), transfer to a distilling 
 flask, make the solution alkaline with caustic soda, and distil the 
 ammonia into standard sulfuric or hydrochloric acid. Titrate the 
 excess of acid with standard sodium carbonate, using methyl 
 orange or methyl red as indicator. 
 
 Various determinations on the amount of nitrogen in the in- 
 soluble matter, have given figures ranging between 12% to 16%. 7 
 The usual factor of 6.25 will give a conservative figure for the 
 proteins, but it is likely that 8.0 or even higher, may frequently 
 be the more correct value. It will be seen from these figures that 
 the determination of nitrogen does not signify very much. 
 
 Curing Tests. It is desirable not merely to know the chemical 
 composition and the loss on washing of crude rubber, but also to 
 know something of its vulcanizing properties. For this purpose, 
 a standard formula should be employed, a series of cures made 
 from this mix, and stress-strain curves drawn for each cure. 
 
 B It is convenient, if not pressed for time, to take the dried rubber from the 
 moisture determinatibn in loss on washing. This simplifies the correction, but 
 in so doing, it must be seen that the sample has not been altered during the 
 drying, by oxidation, or depolymerization. 
 
 6 Cf . Acetone extraction, under methods of analysis, page 68. 
 
 » Cf. Schmitz, Gummi Ztg. 27, 1085. 1131 ; Spence and Kratz, Koll. Zeit. 1J,, 
 262-77 (1914). 
 
THE COMPOSITION OF CRUDE RUBBER 25 
 
 The question of a standard formula is one which may nut be 
 dismissed lightly. At present, many of the plantation and factory 
 chemists are using a mixture of rubber and sulfur. This, how- 
 ever, is open to serious objection, 8 and a less objectionable pro- 
 cedure, even granting that the formula itself may not be the 
 best one, or most suited for all work, is to use a formula contain- 
 ing a small amount of zinc oxide, and sufficient accelerator and 
 sulfur to produce satisfactory cures. One such formula would 
 be: hexamethylenetetramine 0.5%, sulfur 4.5%, zinc oxide 5%, 
 rubber 90%. This mixture contains enough sulfur for a 
 coefficient of 5.0, 9 which is higher than one would ordinarily go, 
 and zinc oxide in excess of that required to neutralize any organic 
 acids in the rubber, and provide a basic mix for vulcanization, 
 since practically all organic accelerators seem to work better 
 under such conditions. 10 Particular pains should be taken regard- 
 ing the quality of the zinc oxide, sulfur, and accelerator; they 
 should be of C. P. grade, and not just the commercial stuff used 
 in the factory. Such grades are to be found in the market, and 
 are worth the extra cost. It is not without the bounds of reason 
 that much of our unexplainable vagaries in rubber testing is 
 really traceable to the impurities in the pigments, and not to the 
 rubber itself. 
 
 Needless to say, perhaps, the results depend largely on the 
 cleanliness and technique in mixing and curing, the accuracy of 
 the thermometers, and the accuracy of the testing machines. No 
 tests should be made until at least 48 hours after vulcanization. 11 
 
 8 Cf. J. B. Tuttle, Variability of Crude Rubber, J. Ind. Eng. Chem. is, 519-22 
 (1921). 
 
 •The sulfur coefficient, sometimes called the coefficient of vulcanization, is 
 the ratio of combined sulfur to the rubber. It is calculated by dividing the 
 percentage of rubber by the percentage of combined sulfur. 
 
 It may be mentioned that the coefficient of vulcanization necessary to pro- 
 duce identical physical properties in two or more compounds, is not a constant, 
 but varies with the amount and nature of the accelerator employed, and to a 
 lesser extent on the other constituents of the compound. 
 
 10 The real purpose of the use of the added organic accelerator and the zinc 
 oxide should not be lost sight of in any discussion of the advisability of using 
 this or any similar formula. It has been shown that crude rubber contains 
 varying amounts of natural organic accelerators, and we must eliminate their 
 effect if we are to study the actual variation of the rubber itself. 
 
 u There are some who believe that 24 hours is sufficient to permit the rubber 
 samples to reach equilibrium. At times, we have taken samples from the vul- 
 canizing press, and after cooling in running water, tested them immediately. 
 But where results of today are to be compared with those of the past, or with 
 those to be obtained in the future, the only safe procedure is to allow the full 
 48 hours, so that such comparisons as may be made will be made under identical 
 circumstances, and any differences noted will be real ones, and not those caused 
 by the fact that at times samples had not yet reached equilibrium. 
 
Chapter III. 
 Tlie Preparation of Rubber Compounds. 
 
 The art and science of preparing rubber compounds is some- 
 thing which may well deserve treatment of its own. It is not 
 the intention to explore the whys and wherefores of the matter, 
 for many of the commercial compounds just "grew up" as time 
 went on, a little of a new material here, and a little less of an old 
 one there, until at present they are so complicated that even the 
 owners of the formulas are afraid to make any further alterations. 
 On the other hand, we have a very large number of formulas 
 which have been constructed on the basis of the definite physical 
 and chemical properties of such a mixture as determined by 
 years of research. Irrespective of why it was used, the analyst is 
 primarily interested only in what materials are likely to be 
 used. 1 Moreover, it is utterly impossible to include every article 
 which has ever been used in rubber manufacture, but only those 
 which have really attained some commercial importance, and 
 hence are likely to be encountered in an analysis. 
 
 Crude Rubbers. In the preceding chapters, the general proper- 
 ties of the most important crude rubbers were given. This is 
 probably as good a time as any to draw attention to the fact that 
 seldom will one find a single kind of crude rubber in a rubber 
 compound. Coarse Para will be mixed with Fine Para, or amber 
 crepes will be mixed with smoked sheets or pale crepe. It may 
 
 'At the time of writing, the situation with respect to crude rubber is such 
 that the preparation of a new compound is a more than usually serious prob- 
 lem. With the best grades of plantation rubber selling around 15 cents a pound, 
 the saving in the use of reclaimed rubbers and substitutes is questionable, if 
 we consider that such materials are to replace the rubber. Some reclaimed 
 rubbers may have an added value on account of the active fillers, such as zinc 
 oxide and gas black or lamp black, which they may contain ; or we may use 
 reclaims and substitutes in special cases on account of special properties which 
 they impart. However, it is incredible that such conditions as now prevail are 
 to continue indefinitely, and hence we are proceeding on tbe basis that the 
 normal price for crude rubber will be from 25 to 30 cents (if not higher), and 
 at this price the use of certain grades of reclaims and substitutes will effect 
 savings in costs, and bence the analyst may expect to find them in examining 
 manufactured articles. 
 
 26 
 
THE PREPARATION OF RUBBER COMPOUNDS 27 
 
 seem superfluous, but it is safer to call attention to the fact that 
 replacing Fine Para with Coarse Para, or smoked sheet with am- 
 ber crepe, is merely a matter of economy ; the rubbers used are not 
 as good as those they replace, and the quality of the compound is 
 lowered. It is purely a question of deciding whether or not the 
 properties of the compound are sufficient to meet the demands 
 of the service. On the other hand, rubber such as Pontianak, 
 Guayule, roll brown crepe, etc., when used as softeners, are used 
 independently of their cost, and their use has continued in many 
 cases when they cost practically as much, or even more than the 
 so-called better rubbers. These points are worth bearing in mind 
 in figuring out the probable formula from the analysis of a 
 rubber compound. 
 
 Reclaimed Rubber. We have seen that the rubber hydrocarbon 
 can combine with sulfur until the compound (C 10 H 16 S 2 ) x is 
 reached, when the ratio of rubber to sulfur is 136:64. In the 
 ordinary soft vulcanized articles, the sulfur coefficient is between 
 1.5 and 5.0, depending upon the type of accelerator, and the 
 degree of vulcanization. Such material is able to take up fur- 
 ther quantities of sulfur to form a new compound with a higher 
 coefficient, which, while somewhat harder than the material from 
 which it was made, may still be of service. Each addition of 
 sulfur, other conditions being equal, produces a harder product 
 than before, until, with the maximum amount of sulfur which 
 may be added, we reach the product ebonite. The hardness of 
 the rubber itself is frequently lessened by the admixture of soft- 
 ening oils, and the partial depolymerization which takes place 
 produces a soft and tacky substance, which also helps to counter- 
 act the hardening effect of the additional sulfur. 
 
 Before vulcanized rubber can be used a second time, it must 
 be put into condition to be mixed in a homogeneous manner with 
 new rubber. There are two general processes employed, (a) the 
 acid reclaiming process; and (b) the alkali reclaiming process. 
 These processes serve to remove any fabric which may be present, 
 the free sulfur, and, of course, some of the fillers, both organic 
 and inorganic. In the latter case, the amount and nature of the 
 fillers removed will depend largely upon the process which is used 
 and the chemical nature of the fillers. Zinc oxide and whiting 
 are largely removed in the acid process, zinc oxide to a slight 
 extent in the alkali process, while gas black and lamp black are 
 
28 THE ANALYSIS OF RUBBER 
 
 unaffected by either. Oil substitutes are not attacked in the 
 acid process, but are almost completely removed by the alkali 
 process. 
 
 These processes of reclaiming do not reverse the vulcanization 
 process; on the contrary, if there be any quantity of free sulfur 
 present, part of it will combine with the rubber during the re- 
 claiming, the sulfur coefficient being higher afterward than before. 
 Other processes have been worked out for the purpose of taking 
 out the sulfur and restoring the double bond, in which case we 
 would expect a product similar to new rubber, and which would 
 vulcanize in the same manner. This is the ideal towards which 
 the researches have been directed, but it must be admitted that 
 as yet we have fallen far short of the ideal, and the reclaimed 
 rubber encountered in vulcanized compounds has been made by 
 one or the other of the two methods mentioned above, or some 
 variation of them. 
 
 Reclaimed rubber is added, under normal market conditions, 
 first of all because it is cheaper. Certain grades may be used 
 because they give desirable properties; for example, vulcanized 
 reclaimed rubber resists oil better than does new rubber, and the 
 use to which the article is to be put is worthy of notice in deciding 
 whether or not reclaimed rubber has been used on account of its 
 cost, or because in the case in question, it is actually better. 
 
 In the manufacture of pneumatic tires, there is always a con- 
 siderable amount of fabric trimmings, containing a large amount 
 of new, unvulcanized rubber. By the acid reclaiming process, the 
 fabric may be entirely removed, with a considerable portion of 
 the sulfur, without appreciably causing the rubber and sulfur to 
 combine. The product, known as "reclaimed or pure gum fric- 
 tion," is a valuable adjunct in rubber compounding. 
 
 Oil Substitutes. 
 
 In the preparation of certain articles, where the highest physi- 
 cal properties were not of primary importance, substitutes for 
 rubber have been used in order to lessen the cost of manufacture 
 (cf. footnote, page 26). One group of such substitutes is made 
 from oils of various kinds, and these substitutes are known com- 
 mercially as "oil substitutes." 
 
 When drying, or semi-drying oils, such as linseed, soya, corn, 
 
THE PREPARATION OF RUBBER COMPOUNDS 29 
 
 cottonseed, and similar oils, are treated with sulfur or sulfur 
 chloride, a solid plastic mass is obtained. These products have 
 been called vulcanized oils, because of the similarity of the 
 processes of preparation with those of rubber. The reaction with 
 sulfur requires heating, and the product varies in color from a 
 light to a very dark brown, or even black. The sulfur chloride 
 combines at ordinary temperatures, giving us the so-called "white 
 substitutes." 
 
 Mixed with these substitutes are various gums and oils, pro- 
 ducing an almost endless number of combinations. This need not 
 bother the analyst, however, for the treated oils are insoluble in 
 acetone and chloroform, whereas the untreated oils and gums are 
 usually soluble in one or the other of these solvents. They may 
 also be loaded with mineral pigments of various kinds. 
 
 Tests of Oil Substitutes. 
 
 An examination of the raw material should cover the un- 
 changed oil, loss on heating at 100C, free sulfur, and ash. Un- 
 changed oil acts in a totally different manner from the true 
 substitute, and the free sulfur is especially important, since it is 
 capable of combining with the rubber during vulcanization; 
 hence any free sulfur present must be taken into account when 
 figuring the amount of sulfur to be added as such to the rubber 
 compound. 
 
 Unchanged Oil. Reduce the sample to a fine state of division 
 by crumbling or cutting. Extract 2 gr. with acetone for eight 
 hours; dry the extract to constant weight at 90C, cool and weigh. 
 
 Free Sulfur. Treat the dried acetone extract with 50 to 75 cc. 
 of water, and 2 to 3 cc. of bromine, heat until colorless, or nearly 
 so, filter through a folded filter; heat the filtrate to boiling, add 
 10 cc. 10% barium chloride, and determine the precipitated 
 barium sulfate as usual. Calculate to sulfur, and deduct the 
 percentage of free sulfur from the total acetone extract. The 
 remainder is the unchanged oil. 
 
 Loss in Weight. Dry a 2 gr. sample in a neutral atmosphere at 
 90/100C until constant weight is secured. 
 
 Mineral Fillers. Ignite a 1 gr. sample, cool the residue and 
 weigh. Pure oil substitutes should have practically no ash; if 
 any pigments are added, the amount will be such as to leave no 
 
30 THE ANALYSIS OF RUBBER 
 
 doubt in the analyst's mind as to whether such additional was 
 accidental, or not. Oil substitutes are usually found in amounts 
 of from 1% to 5%, although we have seen some German made 
 rubber tubing that had nearly 50% of oil substitute. 
 
 Mineral Rubber. 
 
 The mineral rubbers are asphaltic or bituminous hydrocarbons 
 obtained either from natural or artificial sources. The natural 
 sources are from the minerals gilsonite and elaterite, while the 
 artificial mineral rubbers are made largely from the blown oils 
 from petroleum residues. 2 
 
 Mineral rubber possesses a melting point above that of the 
 usual vulcanization range, but its plasticity enables it to be 
 worked readily at much lower temperatures. In amounts up to 
 7 volumes, 3 it materially improves the tensile properties. It 
 serves to soften the uncured stock, makes it tackier reduces 
 blooming, and in a variety of ways proves itself to be an asset to 
 a rubber compound. It improves the waterproofing properties 
 of rubber. 
 
 Owing largely to the differences in the source of supply, and to 
 the various methods of preparation, the chemical and physical 
 properties vary widely. The acetone-soluble matter varies enor- 
 mously, running as low as 17%, and as high as 60%, the higher 
 percentages being the more common occurrence. Chloroform will 
 dissolve part of the residue, equal to about 10% of the whole. 
 They may contain as much as 10% of their weight in sulfur, all of 
 which is chemically combined. There is always a fair sized 
 amount which is soluble neither in acetone nor chloroform. 
 
 While the solvents do not give us exact data as to the quantita- 
 tive figures on mineral rubber, the color of the chloroform extract 
 is a very reliable index in determining the presence or absence of 
 this material. When present, this extract is deep brown to black 
 
 3 For the best and most recent work on Mineral Rubber, consult the article 
 by C. Olin North, "Mineral Rubber," read at the meeting of the Rubber Division 
 of the American Chemical Society at New York, September 6th to 10th, 1921. 
 Abstracts of this paper are to be found in the "India Rubber World." 65, 191-2 
 (1921), and "The Rubber Age," 10, 130-1 (1921). 
 
 s Since the specific gravity of the materials used in rubber compounding varies 
 widely, it affords a more exact method of comparing the effect of the different 
 substances if they are compared on the basis of volume rather than weight. 
 The volume is referred to the total volume of rubber, the latter being taken 
 as 100, 
 
THE PREPARATION OF RUBBER COMPOUNDS 31 
 
 in color, and is not likely to be confused with any other class of 
 material used in rubber manufacture. 
 
 During vulcanization, the percentage of soluble matter may 
 change somewhat; the acetone extract is usually somewhat lower 
 than when the material itself is subjected to extraction. The 
 chloroform extract shows little change. Various explanations 
 have been offered: (1) that the mineral rubber unites with the 
 rubber; (2) it combines with the sulfur to form insoluble prod- 
 ucts; (3) the dispersion of the mineral rubber on the crude rubber 
 produces an adsorption effect, and renders the former more diffi- 
 sult to dissolve out of the mix. Of these, the second seems to be 
 the most plausible, although admittedly the other two are 
 possibilities. 
 
 Mineral rubber has a specific gravity of about 1.00; the hard- 
 ness varies according to the melting point. The melting point 
 is anything that may be desired, but the most popular grade is 
 that melting in the neighborhood of 310F. 
 
 North 4 has determined that the best results with mineral rub- 
 ber are obtained when the proportion is 7 volumes of mineral 
 rubber to 100 of rubber. One is more likely to meet with less 
 rather than with more than this amount. 
 
 Tests for Mineral Rubber. 
 
 Acetone Soluble. Extract with acetone for four hours, a 1 gr. 
 sample of the mineral rubber; dry to constant weight, at 100C. 
 
 Chloroform Extract. Without drying the sample which has 
 been extracted with acetone, extract with chloroform for two 
 hours, or longer if at the end of that period the solvent is still 
 colored. Dry the extract to constant weight, at 100C. 
 
 Ash. Ignite 1 gr. in a porcelain crucible, cool and weigh. The 
 residue should be negligible. 
 
 Insoluble Matter. The difference between 100% and the sum 
 of the acetone and chloroform extracts, and the ash, shall be 
 called "insoluble matter." 
 
 Mineral Hydrocarbons. 
 
 The mineral hydrocarbons may be divided into two classes, 
 hard and soft. The former include ozokerite, ceresin, and par- 
 affin; the latter, petrolatum and heavy mineral oil. The hard 
 
 *Loc. cit. 
 
32 • THE ANALYSIS OF RUBBER 
 
 hydrocarbons are useful for their waterproofing effect, and are 
 to be found largely in materials intended for electrical purposes, 
 such as insulated wire and cable. The soft hydrocarbons are used 
 purely as softeners, to facilitate the handling of the stocks in the 
 factory, and whereas the hard hydrocarbons are without any 
 serious effect on the aging qualities, the soft hydrocarbons have a 
 decided deteriorating effect, and must be used in small quantities. 
 The explanation of this effect would appear to be that the mineral 
 oils are solvents for vulcanized rubber (as previously stated, 
 however, this is not a true solution, but rather a depolymerization 
 preceding solution). 
 
 Mineral hydrocarbons are rarely used to a greater extent 
 than 5%, and in the greatest number of cases the amount used 
 is between 1% and 2%. 
 
 Ozokerite. Ozokerite is a natural product, found in Austria, 
 Russia and southern Utah. It is dark brown to black in color, 
 with a specific gravity of about 0.90. The melting point should 
 exceed 65C (150F). Ceresin is ozokerite which has been purified 
 by treatment with fuming sulfuric acid; it is pale yellow in color, 
 with a resinous luster, non-crystalline in appearance, but in other 
 respects, similar to the parent substance. 
 
 Paraffin. Paraffin is a hard, white, crystalline substance, com- 
 posed of the higher boiling hydrocarbons from petroleum. Its 
 specific gravity is about 0.90, the melting point almost anything 
 that one desires, from soft paraffin which borders closely on 
 petrolatum, to the hard paraffins with melting points around 60C. 
 Ozokerite and ceresin are so much higher in price than paraffin, 
 that the temptation for adulteration is very great, and this is all 
 the more true because of the fact that paraffin, which is used 
 largely as the adulterant, is so near in chemical and physical 
 properties that rather large amounts can be added without fear of 
 detection. Ceresin in the pure state is much less crystalline than 
 paraffin, and less brittle, but it is doubtful if these advantages 
 warrant the extra cost of the pure article. 
 
 Paraffin and ceresin have the peculiar property of working 
 toward the surface of a rubber article, much in the same manner 
 as sulfur "blooms." It appears within a few days after vulcani- 
 zation, and if a slab of rubber containing paraffin be left un- 
 touched for say six months, it is possible to scrape a considerable 
 quantity of clean paraffin from the surface (possibly mixed with 
 
THE PREPARATION OF RUBBER COMPOUNDS 33 
 
 sulfur if the free sulfur is high). This fact is important in 
 analyzing such materials, for the ordinary handling, cleaning, 
 etc., in preparing a sample for analysis, will remove an appre- 
 ciable quantity, and hence, on this account, irrespective of the 
 errors of analysis, the determination of paraffin or ceresin is 
 likely to be low rather than high. 
 
 Oils and Waxes. 
 
 Rubber compounds may be made suitable for calendaring, tub- 
 ing, and other operations, either by excessive working on the 
 mixing mills, or by the use of elevated temperatures. Both 
 methods are objectionable in one sense or another, the excessive 
 working breaks down the rubber, producing a sticky, porous 
 material which is difficult to handle, to say nothing of its poorer 
 tensile properties. High temperatures are to be avoided in the 
 preliminary stages of manufacture, especially with organic accel- 
 erators, since some of the latter become very active at moderately 
 low temperatures, and a partial vulcanization will be effected 
 (what is technically known as "burnt" or "scorched" stock). 
 One method for avoiding these difficulties is to add a small 
 amount of oil (usually 1% to 3%), which softens the rubber 
 compound and brings about satisfactory working conditions. 
 We recognize two classes in these softeners, (a) in which the 
 oil or wax acts merely as a softener; (b) in which in addition 
 to its softening effect, it adds some distinct and desired property, 
 such as tackiness, etc. In class (a) we find palm oil, cottonseed 
 oil, petrolatum or vaseline, and heavy mineral oils; in class (b) , 
 Burgundy pitch, colophony or ordinary rosin, rosin oil, tar oils, 
 etc. The former may be expected in almost any stock, but the 
 latter are used chiefly in cement stocks, frictions, tapes, etc., 
 where adhesive properties have a particular value. 
 
 Palm Oil. Palm oil is obtained from the fruit of the palm tree, 
 Eloeis guineensis, and the west coast of Africa is practically the 
 only important commercial source of this oil. Specific gravity, 
 0.921-0.925; melting point 27-42C, solidification point 37-39C, 
 depending upon the age and origin of the oil. Iodine number, 
 53-57; the commercial oil contains water, sometimes as much as 
 7% ; other impurities up to 3%. It may be adulterated with bark 
 and dirt, and, before using, palm oil is melted, and the clean oil 
 
34 THE ANALYSIS OF RUBBER 
 
 skimmed from the surface. Palm oil is rarely adulterated with 
 other oils or fats, hence it is usually sufficient to determine water, 
 total impurities, and the solidifying point. The color varies from 
 orange yellow to a dark, dirty red. 
 
 Cottonseed Oil. Cottonseed oil is obtained from the seeds of 
 the cotton plant, Gossypium, of which the principal species are 
 G. Herbaceum in the United States, and G. Barbadense in Egypt. 
 Choice crude oil should be free from water and foots, possess a 
 sweet flavor and odor (i.e., should not be rancid), specific gravity 
 0.922-0.925; solidifying point 3-4C; iodine number 105-110. 
 
 Tests for Cottonseed Oil. 
 
 The best known test for cottonseed oil is Halphen's color test, 
 made as follows: 1-3 cc. of the oil is dissolved in an equal 
 volume of amyl alcohol, to this is added 1-3 cc. of carbon bisul- 
 fide holding in solution 1% of sulfur. The test tube is immersed 
 in boiling water, and the carbon bisulfide driven off. A deep red 
 color appears in about 30 minutes. The test depends upon the 
 presence of some chromogenetic substances which are destroyed 
 by high heating, so that rubber compounds containing cottonseed 
 oil may not show this test after vulcanization. 
 
 Petrolatum. Petrolatum, or vaseline, may be either light or 
 dark colored. Its specific gravity is between 0.85 and 0.90. At 
 ordinary temperatures, it is a soft paste, but at 40 to 50C it 
 melts to a clear fluorescent oil. It is not altered in composition 
 during vulcanization, and, unlike paraffin, it remains distributed 
 throughout the compound after vulcanization, and does not bloom 
 to the surface. 
 
 Heavy Mineral Oils. The heavy mineral oils are purely soften- 
 ers, but are more likely to be found as component parts of 
 reclaimed rubber and substitutes, than actually added to com- 
 pounds as such. They act in practically the same manner as 
 petrolatum. 
 
 Burgundy Pitch. Burgundy pitch is more important for its 
 adhesive properties than as a softener, although it acts in both 
 capacities. It is a dark, brittle substance, with a resinous luster, 
 and a specific gravity of about 1.10. It is soluble in acetone. 
 It is obtained from the Norway spruce, Picea Excelsa, by scarifi- 
 cation of the trees, and collecting the resin after it has hardened. 
 
THE PREPARATION OF RUBBER COMPOUNDS 35 
 
 The volatile oils which are present in the crude resin are removed 
 by boiling with water. It contains considerable bark and dirt, 
 and must be purified by melting and filtering through sieves. 
 It is frequently found in low grade frictions, insulating tape, 
 cements, etc. 
 
 Burgundy pitch is composed largely of abietic anhydride, and 
 gives a positive reaction with the Liebermann-Storch test. It is 
 so near ordinary rosin in composition that the latter is fre- 
 quently used as an adulterant, and it is one that is exceedingly 
 difficult to detect. 
 
 Rosin, or Colophony. Rosin is the residue remaining in the still 
 in the separation of oil of turpentine from crude turpentine. Its 
 principal constituent is abietic anhydride. Rosin is about 90% 
 saponifiable, the remaining 10% consisting of rosin oil. It melts 
 anywhere from 100 to 140C, specific gravity 1.08. Its color 
 varies from water white, pale amber, to black, but only the 
 lighter amber colors are used in rubber manufacture. It has very 
 little softening power, but is exceedingly tacky, so that it can be 
 used only in small amounts in cements, frictions, and varnishes. 
 
 Rosin Oil. By the destructive distillation of rosin, we obtain, 
 amongst other products, a reddish colored oil, commonly called 
 rosin oil. Its boiling point is around 360C, or over, specific 
 gravity 0.98-1.10; it usually contains 10% to 20% of rosin, which 
 is saponifiable, but the remaining 80% to 90% is an unsaponifi- 
 able hydrocarbon. It will be noticed that rosin always contains a 
 small amount of rosin oil, and vice versa, hence, both substances 
 give the same positive reaction in the Liebermann-Storch test. 5 
 
 Rosin oil adds very little to the tackiness of the rubber, and is 
 essentially a softener. It improves the waterproofing qualities 
 of rubber. 8 Rosin oil is not used very extensively, especially in 
 
 "The simplest way to make this test is to warm a few drops of the oil in 
 1-2 cc. of acetic anhydride, cool, and to a few drops on a porcelain test plate, 
 add a drop of sulfuric acid of sp. g. about 1.5. A reddish violet color indicates 
 rosin or rosin oil. It is believed that the unsaponifiable portion is really respon- 
 sible for the color, and when examining for rosin or rosin oil, the test may be 
 made much more delicate by making it upon the unsaponifiable portion. Bur- 
 gundy pitch. Venice turpentine and similar resins, give practically the same 
 color, so that the identification as rosin or rosin oil is not absolutely positive. 
 
 • Rubber compounds are so frequently used for waterproofing and in such 
 articles as rubber tubing, hot water bags, etc., that one is quite likely to over- 
 look the fact that rubber takes up a large amount of water when left in contact 
 with it for any length of time, and this holds true even after the rubber has 
 been vulcanized. Pure gum sheet, vulcanized, has been found to absorb as 
 much as 20% of its weight in water. C. O. Weber, in his book on India 
 
36 THE ANALYSIS OF RUBBER 
 
 high grade goods, since it is a solvent, or rather a depolymerizer 
 of rubber. The connection between the two substances, rosin oil 
 and rubber, can readily be seen in the fact that crude turpentine 
 is composed largely of the terpenes sylvestrene and australene, 
 the composition of which is C 10 H 16 ; which form tetrabromides, 
 ozonides, and polymerize easily. 
 
 Tar Oils. The tar oils are the residues from the destructive 
 distillation of wood or coal, the coal tars being the ones gener- 
 ally used. They are of varying composition, and act merely as 
 softeners. As a rule, they are soluble in acetone and alcohol, and 
 have a specific gravity of about 1.00. Their properties depend 
 largely upon the source of the crude material, and the degree of 
 rectification. 
 
 Glue. The glue used in rubber compounding is the ordinary 
 granulated bone glue. The moisture content varies between 7% 
 and 12%, and the specific gravity is about 1.25. Just as it comes, 
 it may be mixed directly with rubber on a fairly warm mill. It 
 is best to have the mixture refined while it is still hot in order 
 to thoroughly break up any particles of glue. Several other 
 methods are in vogue; 3 parts of glue are heated with 1 part of 
 water until a smooth mixture is obtained, 7 then cooled until it 
 sets to a firm jelly. This is mixed with rubber, and dried in a 
 vacuum dryer. In other preparations, the glue is melted and 
 mixed with oils or glycerin, and then allowed to cool; or it may 
 be dissolved in water, gas black or other fillers stirred in, the 
 solution concentrated, and the cooled mass mixed with rubber. 
 
 The effect of glue on rubber is to reduce the elongation, and 
 increase the permanent set. In many compounds, it has been 
 found to exert a stabilizing effect on the cure, flattening out the 
 
 Rubber and its Analysis, p. 14, says : "The water absorption of vulcanized 
 rubber is extremely small, certainly not large enough to appreciably affect the 
 insulation of a rubber cable after 5 years' continuous immersion." Weber did 
 not state what kind of a compound he had in mind when he made this state- 
 ment, but we have had experience with 40% fine Para compounds, containing 
 about 2% of paraffin, which became absolutely waterlogged after about two 
 or three years continuous immersion in water, and were utterly unfit for their 
 purpose. To secure the best waterproofing properties, we resort to the addition 
 of oils, waxes, and pitches. This is particularly true in electrical supplies. 
 
 7 At this stage, several possibilities are open. Some add formaldehyde in 
 sufficient quantities to produce an insoluble glue. Others have added glycerin, 
 about 5% of the dry glue, and concentrated the solution until the moisture con- 
 tent is from 15% to 20%. The purpose of this is to prevent the glue, on 
 cooling, becoming hard and brittle. This glue-glycerin-water combination mixes 
 readily with rubber, and in so doing, the moisture content is substantially 
 reduced. 
 
THE PREPARATION OF RUBBER COMPOUNDS 37 
 
 peak of the vulcanization curves, and reducing the danger of 
 either over or under cures. It has a special field in rubber tubing 
 for conducting gasoline, and other organic solvents, reducing 
 greatly the effect of such solvents on the rubber. Glue has a 
 slight accelerating effect on the vulcanization. 
 
 Other Organic Fillers. A large number of organic substances 
 are used in special articles, by reason of the real or fancied im- 
 provement in the quality or service, from such addition, or for 
 reasons of economy. Rubber soles may be stiffened with ground 
 cotton fabric; shellac, hard gums and resins are used in cements 
 and in waterproofing; ground cork or leather in some floor cover- 
 ings, etc., etc. 
 
 Vulcanizing Materials. 
 
 Sulfur. The sulfur used in rubber should be dry, and free from 
 acid, sand, or other impurities. Before using, it should be care- 
 fully sifted through a 50 mesh screen, excepting, of course, in low 
 grade compounds, where such refinements are of no value. The 
 purpose of the sifting is to remove dirt, splinters of wood, etc., 
 that may come from the container, and to remove agglomerations 
 or lumps of sulfur. 
 
 Tests for Sulfur. 
 
 Acidity. Ten gr. of the sample is placed in a flask, with 
 100 cc. of distilled water, heated on the water bath for 15-30 
 minutes, and any acidity titrated with N/10 sodium carbonate, 
 using methyl orange as the indicator. A blank is run on the 
 water used. Not over 2-3 drops should be required to make the 
 solution alkaline. 
 
 Moisture. Dry at 85C for one hour in a neutral gas, 1 gr. 
 of sample, cool and weigh. The loss should be negligible. 
 
 Ash. Ignite 1 gr. of sulfur in a porcelain crucible, performing 
 the burning in a hood with a strong draft. Cool the crucible and 
 weigh. The ash should be less than 1 mg. 
 
 Sulfur Chloride. The sulfur chloride used in rubber manufac- 
 ture is the monochloride, S 2 C1 2 . Since, however, chlorine acts on 
 the monochloride to give the dichloride, there is usually some of 
 the latter present in commercial sulfur monochloride. Pure 
 sulfur monochloride has a specific gravity of 1.709, boils at 138C, 
 fumes strongly in the air, is decomposed by water forming sulfur 
 
38 THE ANALYSIS OF RUBBER 
 
 dioxide, sulfur and hydrochloric acid. The sulfur liberated by 
 the reaction with water is readily dissolved by the sulfur chloride. 
 It is usually a red or a deep orange color. The dichloride, SC1 2 
 has a specific gravity of 1.62, boils at 64C, and at the boiling 
 point partially decomposes into S 2 C1 2 and Cl 2 . 
 
 The commercial sulfur monochloride usually has a gravity be- 
 tween 1.65 and 1.70, and a boiling point between 115C and 130C. 
 
 Sulfur monochloride should be stored in a cool, dry spot, in 
 clean earthenware jugs with tight fitting earthenware stoppers. 
 It should not be exposed to the air, on account of its affinity 
 for water. 8 
 
 Organic Accelerators. 
 
 The number of organic substances which accelerate the vul- 
 canization of rubber is so great that we have deemed it quite 
 unnecessary to attempt to deal with those which are only of 
 casual interest. Primarily, we are dealing with the analysis of 
 rubber goods, and are chiefly interested in the accelerators which 
 are now being used commercially, or which show possibilities of 
 becoming such. The most widely used organic accelerators today 
 are aniline, thiocarbanilide, and hexamethylenetetramine, and 
 the analyst should look first for these three before proceeding 
 further. 
 
 Most organic accelerators are used in small amounts. For 
 very fast curing purposes, such as tire repair stocks, the quan- 
 tity may be as high as 5% or 6% ; but for ordinary compounds 
 the amount is usually 1% or less of the amount of rubber pres- 
 ent. The amount used depends largely upon the time of cure 
 desired, and the nature of the accelerator. 
 
 Aniline. Aniline, or phenylamine (commonly called aniline 
 oil) , is colorless when freshly distilled, but on standing, acquires 
 a deep red color, and this is the condition in which it is found 
 commercially. It is an oily liquid, specific gravity 1.02, boiling 
 point 184.4C, melting point -6C. The melting point is a par- 
 ticularly useful test for purity. 
 
 • This reaction between sulfur monochloride and water will no doubt explain 
 a considerable amount of the trouble experienced with acid splices, and acid cured 
 goods in general, especially in the hot, sultry days in summer. The evaporation 
 of the solvent of a cement cools the surface below the dew point, resulting in a 
 deposit of a film of moisture. The latter reacts with the S 2 C1 2 , reducing the 
 amount of the active vulcanizing substance which, in extreme cases, may be 
 entirely destroyed before any vulcanization has taken place. 
 
THE PREPARATION OF RUBBER COMPOUNDS 39 
 
 Hexamethylenetetr amine. A white crystalline powder, com- 
 monly called hex, or hexa, melting point about 280C, but decom- 
 poses below its melting point. Specific gravity 1.25. It is quite 
 soluble in water, and slightly so in 95% alcohol. 
 
 Thiocarbanilide. Thiocarbanilide, diphenylthiourea, CS 
 (NHPh) 2 , commonly called thio, crystallizes in white plates, 
 M.P. 154C, specific gravity 1.32. It is made by heating carbon 
 bisulfide with aniline. The commercial product is usually a gray 
 powder, and may contain small amounts of sulfur. There are 
 at least a dozen trade names for this one accelerator, some of 
 the preparations being a mixture of thio with inert pigments. 
 
 Diphenylamine. Diphenylamine, or phenyl-aniline, NHPh 2 , 
 has a molecular weight of 169, specific gravity 1.16, melting point 
 54C, boiling point 302C. It is only slightly soluble in water. 
 
 Dimethylaniline. Dimethylaniline, PhNMe 2 , is a yellow liquid, 
 specific gravity 0.958, melting point 2.5C, boiling point 194C. It 
 is very slightly soluble in water. 
 
 Aldehyde Aniline. If well cooled formaldehyde is mixed with 
 aniline, anhydroformaldehyde-aniline (or trimethylenetrianiline) 
 is formed, melting point 140C. In alkaline solution, at ordinary 
 temperatures, formaldehyde and aniline give methylene-diphenyl- 
 diamine, CH 2 (NHPh) 2 , melting at 65C. This may also be pre- 
 pared by heating anhydroformaldehyde-aniline with alcoholic 
 aniline to 100C. 
 
 Commercial aldehyde-aniline is a mixture of several sub- 
 stances, the proportions varying with the differences in the con- 
 trol during the process of manufacture. 
 
 Ethylidene Aniline. Ethylidene aniline is made from acetalde- 
 hyde and aniline. It is a dark reddish liquid, very stiff at 
 ordinary temperatures, but it becomes quite fluid at the usual 
 working temperatures of the mixing mill (175F-200F). 
 
 P-nitrosodimethylaniline. P-nitrosodimethylaniline is obtained 
 in the form of large green, glistening leaflets, melting point 85C. 
 It stains paper or cotton a deep yellow. With caustic alkali, it 
 breaks down into nitrosophenol and dimethylamine, a reaction 
 of much interest in connection with the preparation of the dithio- 
 carbamates. 
 
 Other Aniline Derivatives. There are some other derivatives of 
 aniline which might be included here, but are not because they 
 
2) 
 
 40 THE ANALYSIS OF RUBBER 
 
 are of no importance commercially. We may mention p-pheny 
 lenediamine, p-aminodimethylaniline, etc. 
 
 Diphenylguanidine. Diphenylguanidine, NH:C:(NHPh) 
 melting point 147C. It is a mono-acid base; with carbon bisul- 
 fide, it forms thiocarbanilide and thiocyanic acid. One com- 
 mercial preparation consists of two thirds diphenylguanidine, and 
 one third magnesium oxide. 
 
 Triphenylguanidine. Two triphenylguanidines are known; 
 (a) PhN:C: (NHPh) 2 , is most easily prepared by heating thio- 
 carbanilide and aniline, and distilling off the excess of aniline. 
 Hydrogen sulfide splits off during the reaction. This is the tri- 
 phenylguanidine commonly used in rubber compounding. When 
 pure, it exists as white crystals, but the commercial product is 
 frequently colored yellow owing to the excess of aniline which 
 has not been distilled. It has a melting point of 143C. (6) The 
 second triphenylguanidine is derived from the HC1 salt of 
 diphenylamine and cyananilide, the formula being NH:C. 
 (PhNH).(Ph 2 N). It also has accelerating properties. 
 
 Diphenylcarboimide. Diphenylcarboimide, C 13 H 10 N 2 ; if tri- 
 phenylguanidine is heated under reduced pressure, aniline is given 
 off and diphenylcarboimide produced, PhN:C:NPh. The crude 
 substance is glassy, resinous, amorphous, with no definite melting 
 point, but softens gradually as it is heated. The pure substance 
 is said to have a melting point of 160C-170C. 
 
 Aldehyde Ammonia. When formaldehyde combines with am- 
 monia, instead of following the usual procedure, we get hexa- 
 methylenetetramine. Aldehyde ammonia is the product of the 
 combination of acetaldehyde and ammonia; Me.CHOH.NH 2 ; 
 melting point 70C-80C, boiling point 100C. It occurs as color- 
 less crystals, turning dark on exposure to the air; probably on 
 account of the reaction with the moisture in the air, since in 
 contact with water it forms hydroacetamide. 9 
 
 Furfnramide. Furfuramide, formed by the action of ammonia 
 on furfuraldehyde; a light brown crystalline substance, melting 
 point 117C. 
 
 Quinoidine. The product sold commercially under the name 
 quinoidine, is the residue remaining after the removal of the 
 alkaloids quinine, cinchonine, and cinchonidine, from the extract 
 of Peruvian bark. It is a dark brown to black resinous solid, 
 
 » Richter's Organic Chemistry, translation by E. F. Smith, II, p. 206. 
 
THE PREPARATION OF RUBBER COMPOUNDS 41 
 
 non-crystalline, which softens readily, and mixes well with 
 rubber. 
 
 Piperidine. Piperidine is a colorless liquid, with a peculiar 
 odor slightly resembling that of pepper; strongly basic, soluble 
 in alcohol and water; boiling point 106C. It is found in nature 
 in combination with piperic acid, as the alkaloid piperine, or 
 piperyl-piperidine, crystallizing in prisms, melting point 129C. 
 Piperine is chiefly of interest in combination with carbon disulfide, 
 when it forms one of the ultra-rapid accelerators (see following). 
 
 The So-Called "Ultra-rapid" Accelerators. The combination 
 of carbon bisulfide with secondary amines such as dimethylamine, 
 piperidine, piperine, pyrrolidine, etc., gives rise to the formation 
 of substances which are extremely powerful accelerators of vul- 
 canization; these are believed to be salts of dithiocarbamic acid, 
 and the accelerators of this class are usually called the thiocar- 
 bamates. They are so much more powerful than the organic 
 accelerators that some have attempted to distinguish them by 
 the name of "ultra-rapid accelerators." 10 
 
 The dithiocarbamates are mono-basic, and with zinc form salts 
 which form a second class of rapid accelerators. 
 
 A third class of rapid accelerators, the thiurams, is formed by 
 the oxidation of the dithiocarbamates ; the product is a derivative 
 of thiuramdisulfide, NH 2 C-S-S-S-S-CNH 2 ; for example, the 
 tetraethyl derivative would be (CSNEt 2 ) 2 S 2 , a white crystalline 
 substance, with a melting point of 70C. A few of these product? 
 may be mentioned as follows: 
 
 Dimethylamine and carbon bisulfide; C 5 H 14 N 2 S 2 , m.p. 103C. 
 
 Diethylamine and carbon bisulfide, C 9 H 22 N 2 S 2 ; m.p. 130C. 
 
 Thiuramdisulfide; NH 2 CS.S.S.SC.NH 2 . 
 
 Tetramethylthiuram disulfide, (CSNMe 2 ) 2 S 2 ; m.p. . 
 
 Tetraethylthiuram disulfide, (CSNEt 2 ) 2 S 2 ; m.p. 70C. 
 
 The above list includes practically all of the organic accelera- 
 tors which have reached any commercial significance, and per- 
 haps a few that have not as yet. There is still the derivatives of 
 quinoline, pyrrole, piperidine, and many others. In fact, it may 
 
 10 Some idea of their power to accelerate vulcanization may be gleaned from 
 the fact that a mixture of 50 parts each of rubber and zinc oxide, 3 parts of 
 sulfur, and only 0.1 part of the dimethyldithiocarbamate, will reach its maxi- 
 mum cure in three minutes. Some of the others in this class are even more 
 rapid in this, giving good cures in one minute, with slabs about one sixteenth 
 of an inch thick, hardly time enough for the heat to penetrate to the center 
 of the sheet. 
 
42 THE ANALYSIS OF RUBBER 
 
 not be going too far to say that any basic organic compound, 
 containing amino, or imino nitrogen, is a promising substance in 
 which to look for accelerating properties. 
 
 Inorganic Accelerators. 
 
 The inorganic accelerators are practically limited to com- 
 pounds of two elements, lead and magnesium. Calcium hydrox- 
 ide has accelerating power, but it can be used in such small 
 quantities, on account of its hardening effect on a compound, 
 that sufficient of it cannot be used to completely accelerate the 
 cure. Sodium hydroxide in small amounts acts as an accelerator, 
 while in amounts in the neighborhood of 5%, it actually retards 
 vulcanization. The lead compounds are litharge, red lead, basic 
 lead carbonate, sublimed white lead, sublimed blue lead, and 
 lead oleate. Magnesium oxide and carbonate are the only mag- 
 nesium compounds. 
 
 Litharge. 
 
 Litharge should be clean, dry, pale yellow in color, free from 
 copper; specific gravity 9.37. There should be only small 
 amounts of the dioxide. Litharge is used in quantities of from 
 5% to 20%. Of special interest is the manufacture of aprons 
 for the protection of workers with radio-active substances. These 
 contain about 90% of litharge, 9% of rubber, and 1% of sulfur, 
 by weight. 
 
 Tests for Litharge. 
 
 Moisture. Dry 2 gr. of the sample at 105C for 2 hours, cool 
 and weigh. 
 
 Lead Dioodde. 11 Treat 1 gr. of the sample in a beaker 
 with 15 cc. of nitric acid, sp.g 1.20. Stir the sample until all 
 trace of red color has disappeared. Add from a calibrated pipette 
 or burette exactly 10 cc. of dilute hydrogen peroxide (1 part 
 of 3% hydrogen peroxide to 3.5 parts of water). Add about 
 50 cc. of hot water, and stir until all of the lead dioxide has passed 
 into solution. In the case of some coarsely ground oxides, the 
 contents of the beaker may have to be heated gently to effect 
 complete solution. After the oxide has gone into solution com- 
 pletely, dilute with hot water to 250 cc, titrate with potassium 
 
 11 The Chemical Analysis of Lead and its Compounds, by John A. Scbaeffer 
 and Bernard S. White, pub. by Picher Lead Co.. Joplin, Mo. 
 
THE PREPARATION OF RUBBER COMPOUNDS 43 
 
 permanganate solution having an iron value of about .005. Run 
 a blank on the hydrogen peroxide. 
 
 If the permanganate has been standardized in terms of iron, 
 it can be calculated to lead dioxide, using the factor 2.134. From 
 this the total weight of the dioxide can be calculated. 
 
 Copper. Dissolve 20 gr. of litharge in dilute nitric acid, and 
 boil until solution is complete. Add 40 cc. dilute sulfuric acid, 
 boil gently for one hour, and allow to cool. Filter off the lead 
 sulphate and wash thoroughly. Nearly neutralize the acid with 
 ammonia, make acid with hydrochloric acid, warm the solution, 
 and pass in hydrogen sulfide. Filter the precipitate, without 
 washing, using some of the filtrate to transfer the last traces of 
 sulfide to the paper. Dissolve in nitric acid, and wash the paper 
 thoroughly with hot water. Add 3 cc. of cone, sulfuric acid, 
 evaporate until the fumes of sulfuric acid are evolved, cool, 
 dilute, and, after standing, filter again, washing with hot water 
 containing a little sulfuric acid. Precipitate the copper in the 
 filtrate as sulfide in an ammoniacal solution, filter, ignite and 
 weigh in a covered porcelain crucible. The residue will be a mix- 
 ture of CuO and Cu 2 S. Since the percentage of copper is the 
 same in both cases, calculate to copper using the factor 0.7988. 
 
 Fineness. Determine the residue on a 200 mesh screen, using 
 water to wash the pigment through, and breaking up any loose 
 lumps with a rubber policeman. 
 
 Red Lead. 
 
 Red lead is a mixture of the monoxide and dioxide, with a 
 specific gravity of 9.07. It should have a bright red color, be 
 clean and dry. The moisture, lead dioxide, copper and fineness 
 may be determined as under litharge. 
 
 White Lead. 
 
 White lead is the basic carbonate, containing about 80% 
 metallic lead, and 20% of carbon dioxide and combined water. 
 The specific gravity is 6.46. 
 
 Tests for White Lead. 
 
 Total Lead. 12 Weigh 1 gr. of the sample, moisten with water, 
 dissolve in acetic acid, and filter, ignite and weigh the impurities. 
 Add to the filtrate 25 cc sulfuric acid (1-1), evaporate until 
 
 " P. H. Walker, Bull. 109, Bureau of Chemistry, U. S. Dept. of Agriculture. 
 
44 THE ANALYSIS OF RUBBER 
 
 the acetic acid is driven off; cool and dilute to 200 cc. with water, 
 add 20 cc. ethyl alcohol, allow to stand for 2 hours, filter on 
 a Gooch crucible, wash with 1% sulfuric acid, ignite and weigh 
 as lead sulfate. Calculate to lead with the factor 0.6829 or to 
 the basic carbonate by 0.8526. 
 
 Carbonic Acid. A 1 gr. sample is placed in a flask containing 
 a side arm delivery tube connected with a train consisting of 
 two U-tubes containing sulfuric acid and potassium bichromate, 
 two U-tubes containing soda-lime, and the fifth U-tube contain- 
 ing the same solution as the second sulfuric-bichromate tube. 
 Add dilute nitric acid, and sweep out the liberated carbon dioxide 
 with a current of air which has been freed from carbon dioxide 
 by passing over soda-lime. Weigh the two soda-lime tubes, 
 and the fifth tube, containing sulfuric acid-bichromate; the in- 
 crease in weight is carbon dioxide. 
 
 Fineness. Treat as under litharge. 
 
 Sublimed White Lead. 
 
 Commercial sublimed white lead is a basic sulfate, containing, 
 on an average, of about 78.5% of lead sulfate, 16% of lead oxide, 
 and 5.5% of zinc oxide. It has a specific gravity of 6.20. It 
 should pass through a 200 mesh screen without appreciable 
 residue. • 
 
 Sublimed white lead is used for its accelerating properties, 
 which are almost entirely dependent upon the content of lead 
 oxide. A test mix would undoubtedly be the best method for 
 testing; the lead oxide may be calculated by determining the 
 total sulfur and total lead, and after calculating the sulfur to 
 lead sulfate the excess of lead may be calculated to lead oxide. 
 
 Sublimed Blue Lead. 
 
 Sublimed blue lead contains lead sulfate, sulfide, sulfite, oxide, 
 and zinc oxide, with occasional traces of carbon. The fineness 
 and accelerating properties are the only elements of interest; 
 the specific gravity will be about 6.50 to 7.0. 
 
 Lead Oleate. 
 
 Lead oleate is a yellowish soft waxy solid, used to replace 
 litharge because of the ease with which it may be distributed in 
 a rubber mixing. The specific gravity is 1.50. It is claimed that 
 
THE PREPARATION OF RUBBER COMPOUNDS 45 
 
 the lead oleate is much less harsh in its action than litharge, with 
 less danger of burning the stock. 
 
 Magnesium Oxide. 
 
 Magnesium oxide, MgO, is sometimes called calcined magnesia 
 from its method of preparation; it exerts a considerable influ- 
 ence on the vulcanization of rubber, although less than that of 
 litharge. It is prepared by precipitation as the carbonate, and 
 the latter ignited. It usually contains some calcium carbonate, 
 but the amount must be kept very low in order not to interfere 
 with its accelerating power. It has a specific gravity of from 3.20 
 to 3.45. 
 
 The calcium carbonate may be determined by solution of the 
 sample in hydrochloric acid, and the separation of the calcium 
 as oxalate from an ammoniacal solution, with ammonium 
 oxalate. The calcium may then be determined in any desired 
 way. 
 
 Because of its effect on the action of certain organic accel- 
 erators, magnesium oxide is sometimes used in amounts of 0.25% 
 to 1.0%, in which case the accelerating effect of the magnesium 
 oxide so used is small compared with that of the activated or- 
 ganic accelerator. As the principal, if not the only accelerator, 
 it will be found in amounts up to 10%. 
 
 Magnesium Carbonate. Magnesium carbonate is a light, white 
 powder, existing in a finer state of division than the oxide; its 
 specific gravity is around 2.22. It may also contain calcium 
 carbonate, which may be determined as under magnesium oxide. 
 
 The carbonate is not as powerful an accelerator as the oxide, 
 and hence will be found in somewhat larger amounts; it is sel- 
 dom used in less than amounts around 5%, and may go as high 
 as 20%. 
 
 In the absence of any appreciable amounts of calcium, deter- 
 mine the magnesia content of the dry pigment by igniting to a 
 dull red heat, to constant weight, taking care that the residue is 
 cooled in a desiccator, and weighed in a stoppered weighing 
 bottle, in order to prevent reabsorption of moisture. 
 
 Inorganic Fillers. 
 
 Aluminum Flake. Aluminum flake is essentially a mixture of 
 hydrated aluminium oxide and silicate. It is a white powder, 
 
46 THE ANALYSIS OF RUBBER 
 
 with a specific gravity of from 2.58 to 2.65; with 2.60 as a fair 
 average of the commercial lots. It contains very little moisture 
 which may be driven off by heating at 100C. Continued ignition 
 at a dull red heat shows an ignition loss of about 12% ; the resi- 
 due is the oxide and silicate. The ignited oxide is difficult to 
 get into solution in hydrochloric acid, even when fused for 
 a short time with sodium carbonate. This fact is important, 
 both in the examination of the pigment, and in the analyses of 
 ash. 
 
 On account of its low gravity and fineness, it is used to replace 
 some of the zinc oxide in a compound, although it does not give 
 as good tensile properties. 
 
 Ammonium Carbonate. 
 
 Commercial ammonium carbonate is a mixture of the car- 
 bonate and carbamate; it is used to supply the gas for making 
 sponge rubbers. 
 
 Asbestine. 
 
 Asbestine is the trade name for a fairly pure magnesium sili- 
 cate, specific gravity 2.60-2.80. It is used at times in place of 
 talc for dusting stocks, and replaces whiting in some mixes. It 
 is a cheap filling material. 
 
 Barytes. 
 
 Barium sulfate is used under various trade names, barytes, 
 blanc fixe, basofor, barium dust, etc. Wiegand has shown that 
 this pigment is a mere diluent; it is inert during vulcanization. 
 On account of the crystalline nature of this pigment, it is not 
 very well adapted for some lines of manufacture, but finds ex- 
 tensive use in mechanical goods. The specific gravity runs be- 
 tween 4.2 and 4.5. It should be free from grit and should leave 
 no residue on a 200 mesh screen. Some preparations of barytes 
 are claimed to have less than 1 % of residue on a 300 mesh screen. 
 The best means for telling the relative value of the various 
 brands of barytes is by means of vulcanization tests with experi- 
 mental batches. 
 
 Since barytes is used merely as a filler, it is seldom found in 
 amounts under 10%, and there may be as high as 30% in the 
 compound. 
 
THE PREPARATION OF RUBBER COMPOUNDS 47 
 
 Brown Pigments. 
 
 The principal brown pigments are the various mixtures of 
 iron and manganese oxides, the umbers. These are usually higher 
 in manganese than the siennas. They should be tested for grit, 
 and for change of color when heated. 
 
 Recent research has seemed to indicate that manganese is re- 
 sponsible for rapid deterioration of some rubber compounds; 
 should this be substantiated with further work, it would seem 
 to show that the manganese browns should be used with caution. 
 
 Calcium Sulfate. 
 
 Calcium sulfate is rarely used as such in rubber compounding, 
 but it exists as a part of many lots of commercial golden and 
 crimson sulfides of antimony. 
 
 Chinese Blue. 
 
 Blue is not a color which is used to any very great extent in 
 rubber manufacture. The chief blues are Chinese blue, ultra- 
 marine blue, and the blue organic dyes. 
 
 Chinese (or Prussian) blue, is precipitated from a mixture of 
 potassium ferrocyanide and ferric sulfate. It is an excellent blue 
 color, but has limited possibilities in rubber, owing to its turning 
 brown when mixed with alkalies, forming ferric oxide, and salts 
 of hydrocyanic acid. 
 
 Crimson Antimony. 
 
 Crimson antimony is largely an oxide or oxysulfide of anti- 
 mony, with a deep crimson, or red color; specific gravity varies 
 from 3.9 to 4.2. It is usually lower in free sulfur than golden 
 sulfide, and is used chiefly on account of its color. 
 
 Dyes. 
 
 The organic dyes are found chiefly in the sulfur chloride, or 
 acid, cured goods. Practically none of them are water soluble, 
 but most of them can be leached out with alcohol, acetone, or 
 benzene. The identification of these dyes is an exceedingly diffi- 
 
48 THE ANALYSIS OF RUBBER 
 
 cult proposition; they are, as a rule, merely coloring materials, 
 and have no other effect on the rubber, so that any dye which 
 will give the same color is no doubt of equal value, and the 
 positive identification of any one particular dye is not often a 
 matter of interest. 
 
 Fossil Flour. 
 
 Fossil flour (tripoli, diatomaceous earth) consists of the re- 
 mains of diatoms, and is nearly pure silica, with traces of alkali. 
 It may contain considerable moisture, and the loss in weight at 
 105C is an important indication of its availability for rubber 
 compounding. It is a very poor conductor of heat, and hence 
 is frequently used in steam valves, etc. The specific gravity is 
 about 2.00. 
 
 Gas Black. 
 
 Gas black is a very pure form of carbon, prepared by burning 
 natural gas with insufficient air for complete combustion. It is 
 the most finely divided pigment in use in rubber compounding; 
 it contains no oil or grease, and on ignition leaves no residue. 
 It has a specific gravity of 1.73, or less than one third of that 
 of zinc oxide, so that a pound of gas black has more than three 
 times the volume, and an even greater proportion of active sur- 
 face. It is hygroscopic to a considerable degree, taking up mois- 
 ture from the air to the extent of 2 or 3%. 
 
 Gas black should not be confused with lamp black, which is 
 made from the burning of oils, tars, or resins, also with insuf- 
 ficient air for complete combustion. The flame may impinge 
 on a revolving metallic cylinder, as in the case of gas black, or 
 the oil may be fired in a huge oven, and the smoke carried 
 through a series of chambers, thus making a partial separation 
 of the different grades of black. Those nearest the fire are, of 
 course, heavier, and contain a larger percentage of oil than the 
 black contained in those chambers furthest away from the fire. 
 These lamp blacks are further purified by heating, with the ex- 
 clusion of air, thus reducing the percentage of oil. Lamp black 
 is not as fine a pigment as gas black, and does not give the same 
 improvement in tensile properties that the latter does; in fact, 
 in this respect, it is rated below zinc oxide. It has the same 
 specific gravity as gas black. 
 
THE PREPARATION OF RUBBER COMPOUNDS 49 
 
 The only tests for gas black, or lamp black, are moisture, oils, 
 and ash. Moisture should be determined on a 1 gr. sample by 
 heating to 105C, cooling in a desiccator, and weighing in a stop- 
 pered weighing bottle. Oil is determined by extraction of a 5 gr. 
 sample with ethyl ether, and weighing the residue. Not less 
 than 5 gr. should be taken for the ash, and if the residue is an 
 appreciable amount, it shows an admixture of other blacks, or 
 dirt. 
 
 Owing to its low gravity, and fineness of particle size, gas 
 black seldom runs higher than 10%, although there have been 
 commercial articles manufactured containing 17-20%. 
 
 Golden Antimony. 
 
 Antimony sulfide, or golden sulfide, is a mixture of the tri- and 
 penta-sulfides of antimony, free sulfur, and it may contain little 
 or much calcium sulfate. The pigment varies from orange to a 
 reddish color, the red being due to the oxide or oxysulfide. The 
 composition varies within wide limits, as is shown by the varia- 
 tion in the specific gravity of from 2.5 to 2.9. It is not an accel- 
 erator of vulcanization; its real value consists in its ability to 
 give up the free sulfur to rubber during vulcanization and yet, 
 afterwards, to remain free from blooming. The free sulfur 
 should run about 17%, and the calcium sulfate should be low. 
 Caspari gives some figures showing that the free sulfur may vary 
 from 7 to 19% ; the calcium sulfate from 3 to 50% ; and the anti- 
 mony sulfides from 30 to 90%. 
 
 When used for coloring only, golden sulfide may be used only 
 to the extent of 1 or 2% ; when used as the source of sulfur for 
 vulcanization, 15 to 25% will be required, depending largely 
 upon the free sulfur and antimony sulfide content of the dry 
 pigment. 
 
 Tests for Golden Sulfide. 
 
 Calcium sulfate. Jacobson 13 recommends the following simple 
 test for calcium sulfate: Mix 1 gr. of the original sample 
 with 2 gr. of sublimed ammonium sulfate in a porcelain crucible. 
 Heat until the ammonium sulfate and antimony sulfide have been 
 driven off; cool and weigh. 
 
 "Chem. Ztg. 32, 984 (1908). 
 
50 THE ANALYSIS OF RUBBER 
 
 Free sulfur. Extract 1 gr. with acetone, or carbon bisulfide, in 
 the extractor described under "acetone extract." Distil off the 
 solvent, add 100 cc. of water and 3 to 5 cc. of bromine; proceed 
 with the determination as directed under the determination of 
 free sulfur in vulcanized articles. Or the solvent may be driven 
 off in a tared flask, the flask and contents dried to constant 
 weight at 90C, and the sulfur weighed directly. This method is 
 shorter, but as a rule, not as accurate. 14 
 
 Graphite. 
 
 Graphite, or plumbago, is a natural form of carbon, used to 
 some extent on account of its lubricating value in preventing 
 adhesion between rubber stocks and metal. It may be found in 
 some stocks where its acid and alkali resisting properties are 
 of peculiar value. 
 
 Greens. 
 
 Most of the green pigments used in rubber manufacture are 
 organic colors. Brunswick green, a mixture of Chinese blue and 
 chrome yellow (lead chromate), darkens when heated with sul- 
 fur. This green is sometimes marketed as "chrome green," but 
 the true chrome green is the oxide of chromium, Cr 2 3 , and is by 
 far the best mineral green for rubber work, since it is not readily 
 affected by heat, acids, or alkalies. 
 
 "Luff and Porritt, J. Soc. Chem. Ind. Ifi, 275-8T (1921), found by previously 
 heating antimony sulfide before extracting the free sulfur, the latter varied 
 considerably, as will be seen from the following table: 
 
 Sulfur Extracted from Antimony Sulfide. 
 Extraction for 5 Hours with Carbon Bisulfide. 
 
 
 Unheated 
 
 
 
 Heated 1 to 2 hours 
 
 
 
 1st 5 
 
 2nd 5 
 
 
 
 
 Sample 
 
 hours 
 
 hours 
 
 125 C 
 
 150 C 
 
 230 C 
 
 1 
 
 3.70 
 
 0.33 
 
 2.99 
 
 4.88 
 
 6.94 
 
 2 
 
 31.21 
 
 0.33 
 
 29.75 
 
 32.19 
 
 32.71 
 
 3 
 
 1.02 
 
 0.13 
 
 .95 
 
 1.01 
 
 .98 
 
 4 
 
 4.64 
 
 0.17 
 
 1.56 
 
 4.86 
 
 4.90 
 
 5 
 
 9.14 
 
 0.13 
 
 8.90 
 
 13.74 
 
 15.38 
 
 The presumption is that the sulfur extracted is available for vulcanization. 
 If during vulcanization a greater percentage of free sulfur than that indicated 
 at normal temperatures is available, this fact is of decided interest and value. 
 It is desirable that this subject be followed up — we should know more definitely 
 why at 125C the free sulfur drops off, and more particularly how long, after 
 heating, the additional free sulfur is capable of being extracted with carbon 
 bisulfide. 
 
THE PREPARATION OF RUBBER COMPOUNDS 51 
 
 Iron Oxides. 
 
 Red oxide of iron (Indian red, Venetian red) is one of our 
 most valuable pigments, not merely for its color, but for the 
 valuable tensile properties which it imparts to rubber, ranking, 
 in this respect, not very far behind zinc oxide. It is practically 
 pure Fe 2 3 , running over 98%, with small amounts of water. It 
 holds its color very well during vulcanization. The specific 
 gravity is between 5.0 and 5.20. 
 
 These iron oxides may be obtained in a great variety of shades, 
 depending largely on the method of preparation. The color 
 should always be matched against a standard, and it is best to 
 make a heat test at 150C, as recommended for golden sulfide. 
 
 Lime. 
 
 The lime which we use is the air slaked hydroxide, specific 
 gravity of 2.4. It is used largely because it will take up small 
 amounts of moisture which may be present in the compound, 
 and reduce the danger of ''blowing," or porosity. It has a decided 
 hardening effect on the rubber, and hence may not be used in 
 anything but small amounts. It also is believed to be responsible 
 for rapid deterioration. It has some accelerating effect on the 
 vulcanization, and due allowance must be made for this factor. 
 
 Lithopone. 
 
 Lithopone is a mixture of barium sulfate and zinc sulfide, con- 
 taining about 25 to 30% of the latter. It is not as fine a pigment 
 as the oxide, and does not produce as good tensile properties. It 
 is unaltered during vulcanization, and is often used as a substi- 
 tute for the more expensive zinc oxide. It must be low in water 
 soluble matter, lead, and chlorides. The specific gravity is 4.20. 
 
 Tests for Lithopone. 
 
 Moisture. Heat 1 gr. for 2 hours at 105C, cool and weigh. 
 
 Barium Sulfate. To 1 gr. of pigment, add 10 cc. cone, hydro- 
 chloric acid and 1 gr. of potassium chlorate in small portions. 
 Evaporate to half its volume, add 100 cc. of hot water, and a 
 
52 THE ANALYSIS OF RUBBER 
 
 few cc. of dilute sulfuric acid. Boil and filter, wash thoroughly, 
 ignite and weigh the barium sulfate. Any silica, and some of 
 the alumina, if present, would be included, but it is not worth 
 attempting to make a separation. 
 
 Total zinc. 15 Take 1 gr. of the pigment, and boil with the 
 following solution: Water 30 cc, ammonium chloride 4 grams, 
 cone, hydrochloric acid 6 cc. Dilute to 200 cc. with hot water; 
 add 2 cc. of a saturated solution of sodium thiosulfate, and titrate 
 with a standard solution of potassium ferrocyanide, using 5% 
 uranium nitrate as an outside indicator. Calculate the zinc to 
 zinc sulfide. 16 
 
 Fineness. Lithopone should leave practically no residue on a 
 200 mesh screen. 
 
 Sodium Bicarbonate. 
 
 Sodium bicarbonate is used in sponge rubber, since on heating 
 it breaks down into the carbonate, carbon dioxide, and water. 
 In the vulcanized article, it is found chiefly as the carbonate, 
 Na 2 C0 3 . 
 
 Talc. 
 
 Talc is used extensively as a lubricant, to prevent rubber sur- 
 faces from sticking together, and in molds, to prevent the rubber 
 stocks from sticking to the mold. It is rarely used as a filler, 
 but rubber has such a facility for absorbing talc that the analyst 
 will rarely fail to find 1% or 2% of talc in vulcanized compounds. 
 The specific gravity is about 2.7, and the color will vary from a 
 brilliant white to a dirty gray. 
 
 Talc usually has a considerable amount of grit, largely sand 
 and the iron minerals which are usually found associated with 
 talc (pyroxene, hornblende and biotite). 
 
 Ultramarine. 
 
 Ultramarine is probably a double silicate of sodium and 
 aluminium, with some sodium sulfide. The sulfide seems an 
 essential part; at least, if treated with acids, hydrogen sulfide is 
 given off and the blue color fades out. It is the best known blue 
 
 10 Low's Technical Methods of Ore Analysis, p. 284. 
 
 i« There Is a slight error here, owing to the fact that part of the zinc is 
 present as the oxide, but the error is usually negligible. 
 
THE PREPARATION OF RUBBER COMPOUNDS 53 
 
 pigment for hot vulcanization, but it is not safe to use it in goods 
 for acid curing, since sulfur chloride usually contains free acid, 
 and the latter would react with the ultramarine, and either par- 
 tially or wholly destroy the color. The specific gravity is 2.35. 
 
 Vermilion. 
 
 The true vermilion is the sulfide of mercury, a very heavy 
 pigment, specific gravity of about 8.00, but possessing a brilliant 
 red color. It is the most expensive pigment used in commercial 
 rubber goods, and since its color is its only good point, it is sel- 
 dom worth what it costs, and is not likely to be encountered by 
 the average analyst. Some so-called vermilions are merely red 
 lakes. In the dry pigment, they are easily recognized by the 
 difference in gravity. 
 
 Whiting. 
 
 On account of its low cost, whiting is extensively used. It is 
 essentially calcium carbonate, and should be entirely soluble in 
 dilute acids, and should contain no free alkali. It is somewhat 
 hygroscopic, specific gravity 2.67, and contains small amounts 
 of iron, alumina, and silica. It may be found in any amount up 
 to say 25 or 30%. 
 
 Tests for Whiting. 
 
 Moisture. Heat 2 gr. for 2 hours at 105C; cool and weigh. 
 
 Free Alkali. In an Erlenmeyer flask, shake 10 gr. of pigment 
 with 100 cc. of water, add a few drops of phenolphthalein ; the 
 color should not be deeper than a faint pink. 
 
 Water Soluble. Heat 10 gr. of pigment with 100 cc. of distilled 
 water, filter, evaporate to dryness in a weighed beaker or dish, 
 heat to 105C for 15 minutes, cool and weigh. 
 
 Fineness. Whiting should leave practically no residue on a 
 200 mesh screen. 
 
 Yellow Ochre. 
 
 The yellow ochres are practically all clays, containing large 
 amounts of hydrated iron oxide; the specific gravity will vary 
 enormously, probably more than any other pigment, from say 
 3.50 to 5.00. The higher gravity ochres are considered better 
 
54 THE ANALYSIS OF RUBBER 
 
 for the purpose; they hold their color better, have a stronger 
 color, and are less likely to change color during vulcanization. 
 The stronger colored ochres are to be preferred also, because less 
 is required to give a definite color in the finished article. 
 
 Zinc Oxide. 
 
 Zinc oxide is unquestionably the most widely used pigment 
 in rubber manufacture. Its extreme fineness makes it particu- 
 larly valuable where strength and wear-resisting qualities are 
 desired, it is unaffected in color during vulcanization, and hence 
 can be used in any color combination. It has a special field of 
 usefulness in that it also provides a rubber mix with an alkaline 
 reaction, which permits many of the organic accelerators to func- 
 tion. Thiocarbanilide, the dithiocarbamates, thiurams, etc., will 
 not accelerate vulcanization unless the mixture is basic, and zinc 
 oxide answers the purpose in a most acceptable manner. 
 
 With some accelerators, zinc oxide reacts during vulcaniza- 
 tion to form a new accelerator. The mechanism of such reac- 
 tions is still a matter under investigation, and while splendid 
 results have been accomplished by the workers in this field, we 
 can hardly feel that the last word has been said on the subject. 
 Probably the safest position to take is to say that practically all 
 of the organic accelerators are more active in the presence of a 
 basic oxide, such as magnesium, zinc and lead, and there are 
 some which will not react without some such basic substance. 
 In a few cases the marked difference between the reaction when 
 zinc oxide is present, compared with some other basic oxide, 
 suggests a possible reaction between the zinc oxide and the ac- 
 celerator. 
 
 Zinc oxide may be absent altogether, it may constitute only 
 a small percentage of the whole compound, or it may be as high 
 as 50%, as for example, in some of the white tire treads. 
 
 Tests for Zinc Oxide. 
 
 Zinc oxide should be tested for moisture, lead, chlorides, sul- 
 fates, sulfides, and water-soluble matter. The specific gravity 
 is 5.57, and the fineness such that there should be no residue on 
 a 200 mesh screen, and very little on a 300 mesh. Over 0.1% of 
 
THE PREPARATION OF RUBBER COMPOUNDS 55 
 
 lead renders it unfit for bright colored mixes, while much larger 
 amounts would so change the vulcanization as to prevent its use 
 altogether, unless, which seems unlikely, one could depend upon 
 getting a zinc oxide with absolutely constant lead content. 
 Chlorine is seldom found in amounts over 0.01, but cases have 
 been known in which the chlorine ran over 0.20%. Such an 
 amount will usually be reflected in an unusually high water 
 soluble extract. Metallic chlorides have a deleterious effect on 
 many rubber compounds, especially cements, and hence the 
 chlorine content must be kept low. 
 
 Moisture. Dry 2 gr. at 105C for 2 hours, cool, weigh, and cal- 
 culate the loss to percentage. 
 
 Insoluble Matter. In a 250 cc. beaker, treat 10 gr. with 50 cc. 
 of cone, hydrochloric acid; evaporate to dryness, take up the 
 residue with water and a few drops of hydrochloric acid, filter, 
 and wash thoroughly with hot water. Ignite the residue, cool 
 and weigh. 
 
 Water Soluble. Treat 10 gr. with 200 cc. of water, heat on a 
 hot plate for one hour, filter into a 250 cc. graduated flask, cool 
 to room temperature, and make up to the mark. Take a 50 cc. 
 portion, evaporate to dryness in a weighed beaker or dish, dry 
 for 2 hours at 105 C, cool and weigh. 
 
 Chlorides. From the water soluble extract take a 50 cc. por- 
 tion, make slightly acid with nitric acid, add 10 cc. N/10 silver 
 nitrate and a few drops of ferric chloride; titrate the excess of 
 silver nitrate with standard ammonium thiocyanate. 
 
 Sulfates. To another 50 cc. portion of the soluble matter, add 
 several drops of cone, hydrochloric acid, and heat; add 1 cc. of 
 10% barium chloride solution, allow to stand overnight; the 
 next day, if there is any precipitate, it can be determined as 
 usual. 
 
 Total Sulfur. Treat 10 gr. of pigment with 25 cc. of cone, 
 hydrochloric acid and 10 cc. bromine water. Evaporate to 
 dryness, take up with 50 cc. of hot water and a few drops of 
 hydrochloric acid, filter from any insoluble matter; heat nearly 
 to boiling, add 1 cc. of 10% barium chloride, and after standing 
 overnight determine any barium sulfate which may have pre- 
 cipitated in the usual manner. 
 
 Lead. The filtrate from the determination of insoluble matter 
 is nearly neutralized with sodium carbonate, and the lead pre- 
 
56 THE ANALYSIS OF RUBBER 
 
 cipitated with hydrogen sulfide. The qualitative test is usually- 
 sufficient; if desired, the lead sulfide may be determined by any 
 of the usual methods. 
 
 Fineness. Place 10 gr. on a 200 mesh screen sieve, and, with 
 a gentle current of water, wash the pigment through the screen. 
 Any loose aggregates may be broken up with a policeman. There 
 should be no residue. Repeat with a 300 mesh screen, and if any- 
 thing remains on the screen, it should be transferred to filter 
 paper, ignited, and weighed. 
 
Chapter IV. 
 The Theory of Vulcanization. 
 
 Having before us the proposition that we are at this time pri- 
 marily interested in the process of vulcanization as a change in 
 chemical composition, without necessarily dealing with the man- 
 ner in which such a change takes place, it seems as though a 
 detailed study of the various theories of vulcanization is quite 
 beyond the scope of the present work, and we will therefore go 
 into the subject only to the extent necessary to develop the facts 
 regarding the chemical changes during this process. 
 
 The term vulcanization has been used freely, and it will no 
 doubt clarify matters if we attempt to define it. 1 For our pur- 
 pose, we will assume that vulcanization will mean the addition 
 of any element or group of elements, of which we may use sulfur 
 and sulfur chloride as the principal examples, to crude rubber, 
 or a mixture of crude rubber with other substances, whereby 
 the crude rubber, or rubber mixing, is changed from a sticky, 
 plastic mass into a substance having a certain degree of tough- 
 ness, hardness, resiliency, and, in general, such properties as are 
 usually associated with what we know as vulcanized rubber. By 
 this definition, we purposely avoid including time or temperature 
 
 1 It is of peculiar interest that while this book was in press an article by 
 "Andrew H. King" appeared in Chem. & Met. Eng. 25, 1038-42 (1921), on "The 
 Aging of Rubber," in which he gave a definition of vulcanization which so 
 nearly parallels our own, as to make it well worth while quoting what he has 
 to say: 
 
 "By vulcanization, we mean the addition of a substance or substances to 
 rubber, which results in the production of a more elastic material, i.e., one with 
 less plasticity. When the change becomes of a sufficient magnitude that the 
 product becomes of commercial value, it is then known as 'cure.' It is well 
 known that substances other than sulfur or sulfur chloride — for example, 
 oxygen, chlorine, selenium, etc., produce a change in plasticity — in other words, 
 they vulcanize — but the products obtained in this way have not to date had any 
 commercial value, and therefore cannot be called 'cured.' In speaking of addi- 
 tional vulcanization, it is to be understood that we are not limiting ourselves 
 to sulfur or sulfur chloride." 
 
 Later on, in the same article, "King" says : "A surface aging which results 
 in hardening or checking of the surface, is probably due largely to additional 
 vulcanization by oxygen ; — internal aging may be sulfur and oxygen." 
 
 57 
 
58 THE ANALYSIS OF RUBBER 
 
 as a definite factor in the process; nor do we say that the 
 process can, or cannot take place in the presence or absence of 
 substances which may act as catalysts. The point which we wish 
 to make regarding catalysts is, that they change the reaction as 
 regards time, or temperature, or perhaps both, but they do not 
 change the principal reaction itself. Taking the reaction between 
 sulfur and crude rubber as an example, we finally come to the 
 point where the rubber is saturated with sulfur, and has the 
 formula (C 10 H 16 S 2 ) X . By the use of catalysts, we would get 
 exactly the same product, only in a shorter time, or at a lower 
 temperature. The use of these catalysts is therefore a matter of 
 commercial economy of time. It is true, when we use the longer 
 processes, or higher temperatures, we have side reactions, depoly- 
 merization, etc., but the main process is the same in each case. 
 
 C. 0. Weber found that when he heated rubber and sulfur to- 
 gether, he obtained a substance having as high as 32% of sulfur, 
 corresponding to (C 10 H 16 S 2 ) X ; with sulfur chloride, he obtained 
 (C 10 H 16 S 2 C1 2 ) X . He therefore decided that rubber was a poly- 
 prene, and that it combined with sulfur to form a series of poly- 
 prene sulfides, the final product being identified as above. He 
 was unable to isolate any of the intermediate products, and was 
 obliged to assume their existence. Ostwald, reviewing the work 
 of Weber, Hohn, Seeligmann, Axelrod, Hubener, Stern, Hinrich- 
 sen, and others, came to the conclusion that the chemical theory 
 did not satisfactorily explain the matter, and that the facts as 
 known were more in accordance with an adsorption process than 
 a chemical one. He based his conclusions on the following: 
 That there was always at least a small amount of free sulfur re- 
 maining after vulcanization (but which we now know is not so) ; 
 that the process was a reversible one and that the rate of adsorp- 
 tion depended upon the amount of working which the rubber 
 sustained. Special emphasis was laid on the temperature coeffi- 
 cient, 1.87, which seemed to agree more with the coefficient for 
 an adsorption process than a chemical one. Ostwald was perhaps 
 correct in saying that the evidence at the time was not sufficient 
 to sustain the contention that the process was a chemical one; 
 on the other hand, he himself included facts which as Loewen 2 
 has pointed out, are explainable only on the theory of a chemical 
 process. In the preparation of the bromine and nitrosite deriva- 
 
 *Z. Angew. Cbem. 25, 1553-60. 
 
THE THEORY OF VULCANIZATION 59 
 
 tives of rubber, it has been observed that the derivatives carry 
 all of the combined sulfur, which would seem to indicate a chemi- 
 cal bond between the rubber and the sulfur. Spence showed that 
 not only did the bromine derivatives carry all of the combined 
 sulfur, but that in a series of compounds, the bromine and sulfur 
 bore stoichiometric relations. Spence found evidence of an 
 adsorption effect between the free sulfur and the rubber, preced- 
 ing the actual chemical combination of the two. We shall see in 
 due course, when taking up the subject of the direct determination 
 of rubber, the importance of the conclusions which we reach 
 regarding the nature of the reaction between rubber and sulfur. 
 
 Having arrived at the conclusion that the reaction is a chemi- 
 cal one, we may pass on to the mechanism of the reaction. Crude 
 rubber will not combine with sulfur to any appreciable extent at 
 ordinary temperatures. With the exception of what we have 
 called the ultra-rapid accelerators (dithiocarbamates, etc.) there 
 is no appreciable reaction until a temperature of at least 100C 
 is obtained, and, for ordinary mixes, the rate at this temperature is 
 exceedingly slow. The ordinary commercial range may be said to 
 be between 125C and 150C. While exact data are lacking it has 
 been estimated that for each 6 to 8C increase in temperature, the 
 speed of the reaction is doubled, i.e., the time required for correct 
 vulcanization is reduced by one half. 3 Furthermore, the speed of 
 the reaction may be enormously altered by the addition of cat- 
 alysts. It will be noted that the reaction takes place best when 
 the mixture is weakly alkaline; acids, or strong alkalies, retard or 
 even prevent the combination of rubber and sulfur. About 0.1% 
 of caustic soda acts as an accelerating agent, 5% retards the 
 reaction almost completely. 
 
 These accelerators not merely affect the speed, but also lower 
 the temperature range at which appreciable vulcanization takes 
 
 •Probably every rubber chemist has some such formula on which he bases 
 changes in curing, and while such figures are only approximate, they do give 
 some idea of the order of magnitude of the change in the velocity of the reac- 
 tion. The point is of particular commercial importance, because it shows the 
 necessity for maintaining cures of rubber articles at exactly the prescribed time 
 and temperature. For example, in a cure of 60 minutes at 140C, an error of 
 1C would be equivalent to adding from 8 to 10 minutes to the regular cure. 
 Sufficient attention has not been paid to this important question, and the only 
 reason that more trouble has not resulted is that the maximum in most com- 
 pounds is not a point, but extends over a range of some minutes, and this 
 automatically provides a certain tolerance. With rapid curing compounds, how- 
 ever, the maximum is usually just a point in the curve and a variation in the 
 temperature results either in an under, or overcure, 
 
60 THE ANALYSIS OF RUBBER 
 
 place; some of them, as has been pointed out by Ostromuislenskii, 
 Bruni, Bedford and others, increase the speed at ordinary tem- 
 peratures to the point where it becomes noticeable. 
 
 One more noticeable action with these organic accelerators is 
 the difference in the change in the velocity of the reaction, at 
 ordinary temperatures, when the amount of the accelerator is 
 varied. For example, 0.05% of the dimethyldithiocarbamate may 
 be mixed with rubber, sulfur, and zinc oxide, at a temperature 
 around 100C, for some time, without any noticeable effect on the 
 rubber. With 0.25% of the same accelerator, in a few minutes a 
 decided change takes place, and the rubber becomes hard and 
 unworkable, and is clearly partially vulcanized. 
 
 Cold Vulcanization. 
 
 The acid cure process of cold vulcanization consists in sub- 
 mitting rubber to the action of sulfur monochloride, either in 
 vapor form or in solution. The reaction is similar to that of hot 
 vulcanization; the sulfur chloride adding at the double bond, 
 forming an addition compound, but in this case, both sulfur and 
 chlorine are added, and the resulting compound is different in 
 chemical composition, although greatly resembling the hot vul- 
 canization product in many of the tensile properties. One im- 
 portant fact stands out, that these properties are not as lasting 
 in the acid cured as they are in the hot vulcanized rubber. 
 
 The reaction between rubber and sulfur chloride is practically 
 instantaneous; consequently, an article to be manufactured by 
 this method must first be brought to its final form prior to vulcan- 
 ization. It has often been said that the reaction is a surface 
 one, but this does not exactly explain the true state of affairs. 
 In the case of a sheet of rubber exposed to the vapors of sulfur 
 chloride, the gas will be absorbed by the outer surface, but before 
 it can diffuse into the center of the sheet, chemical combination 
 between the two takes place. This will continue until the surface 
 has taken up all of the sulfur chloride with which it can combine. 
 In the meantime, especially if the sheet is very thick, the center 
 of the sheet is unchanged. A somewhat better distribution of the 
 sulfur chloride is effected by swelling the sheet in solvents like 
 benzene and then dipping the article in a solution of the sulfur 
 
THE THEORY OF VULCANIZATION 61 
 
 chloride in benzene. In this way, the penetration of the sulfur 
 chloride is facilitated, and better results obtained. 
 
 There is no excess of sulfur chloride remaining as long as the 
 rubber is at all unsaturated, and since there is no free sulfur, 
 acid cured articles do not show the sulfur blooming so common 
 with hot vulcanized articles. 
 
 Vulcanization With Mixed Gases. 
 
 A new method of cold vulcanization through the interaction 
 of two gases, has been proposed by Peachey. 4 It consists simply 
 in treating a rubber compound with sulfur dioxide, followed by 
 hydrogen sulfide. Sulfur is liberated in such an active form that 
 it can immediately combine with the rubber. In order to avoid 
 the possibility of having sulfuric acid remain in the rubber, it has 
 been found advantageous to use the sulfur dioxide first and 
 follow this by an excess of hydrogen sulfide, since the latter is 
 inert, and will, in time, be lost by diffusion. A control of the 
 extent of the vulcanization is obtained by adding exact quantities 
 of sulfur dioxide; since an excess of hydrogen sulfide is used, the 
 exact amount of sulfur to be added to the rubber can be cal- 
 culated. • 
 
 Since this process of vulcanization takes place at ordinary tem- 
 peratures, there is no doubt that, if practicable, it can be used with 
 many substances as fillers which it is not possible to use under 
 present conditions. This is especially true of some of the bright 
 organic colors. It is very noticeable, for example, that a much 
 wider range and more brilliant colors may be used with sulfur 
 chloride vulcanization than with the hot vulcanization. It is, 
 however, a question of time and temperature of heating; with the 
 ultra-rapid accelerators, it is quite possible that this advantage 
 will not be as marked as it is with the much slower accelerators. 
 
 1 British patent 136,716, Feb. 21, 1921 ; cf. also Caoutchouc and Guttapercha 
 18, 10744-5 (1921) ; Dubosc claims that in a discussion of the theory of vul- 
 canization, he stated that the reaction may be caused by the production of 
 colloidal sulfur. He showed that sulfur dioxide and hydrogen sulfide could be 
 generated by the ingredients of a rubber compound, and further stated that if 
 hydrogen sulfide and sulfur dioxide were simultaneously present, they would 
 combine to liberate sulfur in such a form as to enable it to immediately combine 
 with the rubber. In this instance, Dubosc was discussing the reaction in con- 
 nection with the theory of hot vulcanization, but the latter was merely used as 
 a source of the gases mentioned, and not necessarily the temperature at which 
 the gases would unite to give off sulfur as indicated. Whether or not this may 
 be called an anticipation of Peachey's patents remains to be decided. 
 
62 THE ANALYSIS OF RUBBER 
 
 Ostromuislenskii's Theories of Vulcanization. 
 
 Much has been said on the subject of the theories of vulcaniza- 
 tion advanced by Ostromuislenskii, but if we can maintain our 
 definition of vulcanization given in the beginning of this chapter, 
 we cannot see that there exists any fundamental difference be- 
 tween his theories, and the present-day practice. He has shown 
 that at ordinary temperatures, he can cause rubber and sulfur 
 to unite in the presence of piperidyl-piperidine-dithiocarbamate. 
 With a sufficient amount of accelerator, the same thing can be 
 done with dimethyldithiocarbamate, but if we reduce the quantity 
 of the accelerator to the neighborhood of 0.05%, then we will 
 find that the reaction will be so slow at ordinary temperatures as 
 to be commercially negligible. It now becomes merely a question 
 of concentration of accelerator in order to make the velocity of 
 the reaction at ordinary temperatures, which is so slow as to 
 approach zero, approach a finite quantity that will be visible to 
 the eye in a reasonably short time. 
 
 A much more distinctive process is the production of a vul- 
 canized rubber substance by the addition of trinitrobenzene ; 
 with benzoyl peroxide, with halides and halide esters. 5 These 
 products have some of the properties of rubber-sulfur vulcani- 
 zates. However, it must be concluded that we have here nothing 
 to invalidate our present conception of vulcanization, and that 
 what has been accomplished is to show that the change from the 
 sticky, plastic rubber, which was first thought to be a function of 
 sulfur, and was later extended to include sulfur chloride, is really 
 a property of a number of substances. Some of these may require 
 heating, and some do not; some require the presence of metallic 
 oxides, and still others do not. As far as can be seen, the chief 
 difference which has been noted up to this time, is the stability 
 of the various products vulcanized in the different ways, and it 
 may be that in order to arrive at a satisfactory definition of 
 what we mean by vulcanization, we shall not only have to state 
 that the vulcanized articles shall have certain definite properties, 
 
 ' There is an excellent analogy here between the various combinations of 
 rubber with elements, or groups of elements, and the similar reactions of the 
 unsaturated fatty acids, such as linoleic, linolenic, etc. With sulfur and sulfur 
 chloride, we have products quite similar to the addition product with oxygen, 
 having many properties in common, such as solubility, etc. 
 
THE THEORY OF VULCANIZATION 63 
 
 but that the rate of decomposition, or deterioration, shall be not 
 greater than a certain set figure. 
 
 To summarize the situation from the analyst's point of view: 
 vulcanization is the chemical combination of rubber with other 
 substances, without reference to time, temperature, catalysts 
 (except as these remain as constituent parts of the mixture), or to 
 any of the steps through which the products may have passed in 
 reaching the final form in which the rubber is found. For 
 example, there should be no chemical difference between rubber 
 and sulfur which has combined as such and which has combined 
 by reason of the treatment by Peachey's process. 
 
Chapter "V. 
 Sampling. 
 
 The sampling of rubber, and the materials to be used in the 
 manufacture of rubber compounds, as is the case with a great 
 many other commercial and natural materials, is usually done in 
 the most casual fashion, whereas the proper sampling, and the 
 care of the sample until the analysis has been completed, is funda- 
 mental. Unless the proper precautions are taken to make the 
 sample represent the material from which it was taken, and 
 maintain its condition and purity, not only is the accuracy of the 
 analysis affected, but the incorrect results may frequently lead to 
 ' false conclusions as to the manufacture or composition of the 
 article. Samples for analysis have been packed, without 
 adequate protection, in the same package with cans of oil; 
 ground rubber samples in unsealed paper envelopes with bits of 
 excelsior distributed throughout; inner tube samples which have 
 been light checked ; rubber articles with unmistakable evidence of 
 having been placed against steam radiators; these do not appeal 
 to the analyst as fertile fields for valuable results. Samples of 
 less than 1 gr. may be very flattering to the ingenuity and 
 ability of the receiver, but they can hardly be said to be repre- 
 sentative of lots of finished goods weighing hundreds, or even 
 thousands of pounds. 
 
 The process of sampling may be divided into three stages: (a) 
 the taking of the sample; (6) its removal to the laboratory; (c) 
 the preparation of the sample for analysis. The purpose of these 
 three steps is to have the actual material used in making the 
 various determinations of the same chemical composition as the 
 lot which it represents. If the sample is to represent a number of 
 pieces, the sample should be drawn to represent a fair average 
 composition of the lot. More often it is not advisable to take 
 more than one piece of a lot, or even a part of that. Under such 
 conditions, we cannot speak of average composition, but since the 
 supposition is that the entire lot is uniform, and that any one 
 piece (or part of it), will truly represent, not the average, but 
 the exact composition of all of the rest, in such cases we must 
 
 64 
 
SAMPLING 65 
 
 select our samples at random. Speaking generally, when we 
 sample raw materials we should draw more than one sample, 
 since these raw materials are apt to vary throughout the lot, 
 and also because raw materials are thoroughly mixed in the proc- 
 ess of manufacture, and it is the average composition which is 
 of chief interest. With finished materials, the averaging process 
 has already taken place, and it is a fair risk to assume that the 
 lot is uniform. 1 
 
 A. Taking the Sample. If the material is liquid, it should be 
 thoroughly stirred before drawing off the sample. Particular 
 attention should be taken to note whether there are two liquid 
 layers, or whether there is any suspended matter (such as water 
 in gasoline, or foots in vegetable oils). The liquid should be 
 bottled at once, and sealed with a stopper which is not attacked 
 by the liquid. The bottle should be scrupulously clean, both 
 inside and outside, and should be dry. 2 Greases, waxes, and 
 resins, are usually packed in small containers ; a few ounces may 
 be drawn from each container, or from a certain proportion of 
 them. These small samples are united to form a composite 
 sample, which is mixed and quartered until a final sample of 
 about 100 to 200 gr. is obtained. This should be placed in a 
 wide-mouthed bottle, or a can, and sealed. 
 
 Dry pigments are usually received in kegs or sacks. As in the 
 case of greases, a small portion is withdrawn from some propor- 
 tion of the containers; these are united, mixed and quartered, and 
 a final sample of 100 to 200 gr. bottled and sealed. 
 
 The sampling of crude rubber is discussed in connection with 
 the testing of crude rubber. 3 
 
 B. Transportation to the Laboratory. The distance between 
 the place where the sampling occurs, and that where it is to be 
 tested, may be a matter of only a few feet, or it may be hundreds 
 of miles. The principles involved are the same, irrespective of the 
 
 1 This is particularly true in rubber goods, so far as actual composition is 
 concerned ; but such a sample will not reveal any variation in the vulcanization, 
 since the latter process takes place in a relatively small number of units. We 
 have met cases of rubber belting, for instance, which is vulcanized a portion 
 at a time, where the manufacturer paid particular attention to the first and 
 last part of the belt because that was where the samples would be taken. No 
 amount of testing is proof against such chicanery. 
 
 - Samples of oil have been received, the container being an ink bottle in which 
 a few drops of ink were still to be seen at the bottom of the bottle ; and this 
 sample was to be tested for mineral acids! 
 
 a Cf. page 22. 
 
G6 THE ANALYSIS OF RUBBER 
 
 distance. The containers in which the raw materials are sent 
 should be tightly stoppered, so as to avoid the introduction of 
 dirt and other foreign matter, and also to prevent change in 
 composition through evaporation or leakage. 4 Manufactured 
 articles sent for analysis should be carefully wrapped. 
 
 The principal deteriorating agents of vulcanized rubber are 
 heat, light, and oils. It is quite essential to see that each package 
 is not only carefully wrapped, but that it will not come in contact 
 with oils, and on packages which are to be sent any distance 
 specific instructions should be written on the outside, that such 
 packages are to be kept in a clean, cool, dark and dry place. 
 These same precautions should be observed in the laboratory, 
 after the samples have been received. 
 
 If considerable stress has been laid on the care requisite for 
 delivering the sample to the laboratory, our justification is that 
 the analyst can test only what he receives; he cannot tell how 
 great a change, or even at times that any change at all, has taken 
 place. Questionable samples should be discarded at once ; failure 
 to do so will often lead to disagreeable controversies, which ac- 
 complish no good purpose, and tend to diminish that respect 
 which the analyses of the laboratory should inspire. 
 
 C. Preparation of the Samples for Analysis. 
 
 Raw Materials. Pigments, oils, waxes, etc., should be mixed 
 thoroughly before each portion is taken for analysis. 
 
 Unvulcanized Rubber Compounds. Sheet out rapidly on a cool 
 mill, roll between holland and place in a covered can. 
 
 Reclaimed Rubber. Treat as under unvulcanized rubber com- 
 pounds. 
 
 Cements. Cements should be stirred thoroughly before por- 
 tions are taken for analysis, and then immediately covered. A 
 fair sized quantity should be taken, the solvent removed, prefer- 
 ably by evaporation in thin layers at room temperature (if neces- 
 sary, to remove the last traces of the solvent, the rubber may be 
 heated for a short time between 80 and 90C, but it is better to 
 avoid heating of any kind), and the residue sheeted out and 
 rolled between holland, as in the case of other unvulcanized 
 (•(impounds. 
 
 ' A sample of gasoline was sent to the laboratory in a can without a cover. 
 It was delivered after working hours, and was not discovered until the next 
 morning. Considerable evaporation had taken place, and the residue was hardly 
 what it was expected to be. 
 
SAMPLING 67 
 
 Vulcanized Rubber Samples. Strip off all fabric, and see that the 
 rubber is homogeneous, i.e., that there are not two or more com- 
 pounds in the sample. 
 
 Grind about 50 gr. in a meat grinder, or coffee grinder, or 
 by passing between the tightly closed rolls of a laboratory mill. 
 Sift the ground material through a 20 mesh screen until about 
 25 gr. has been collected. It is not necessary to sift the entire 
 amount of 50 gr. 
 
 The type of grinder is immaterial, providing the following pre- 
 cautions are observed: The sample must be ground at room tem- 
 perature, without being appreciably heated up; no metal must 
 be introduced into the sample during the grinding; and prefer- 
 ence should be given to those grinders which tear the sample 
 rather than just cut it up, since the former gives the greater sur- 
 face for extraction. 
 
 Material containing fabric and rubber in such a manner as to 
 make it impossible to produce good separation, shall be cut with 
 scissors into as small pieces as is practicable. Rubber and fabric 
 cannot be ground together, since segregation will be certain to 
 occur on account of the difference in behavior on grinding, and 
 this holds true even if the entire sample is ground and sifted. 
 
 Hard rubber samples are prepared for analysis by rasping. 
 
 Insulated wire should be cleaned with a damp cloth, to remove 
 any adhering cotton or other adhering material, but care must be 
 exerted to see that waxy hydrocarbons are not removed from the 
 surface. If, however, a saturated braid sample must be used, 
 remove the braid, and sandpaper the insulation for a depth of at 
 least .005 in., and wipe with a damp cloth. This treatment will 
 probably give low results for waxy hydrocarbons, and hence 
 should be resorted to only when absolutely necessary, and a 
 statement regarding the treatment given the sample should al- 
 ways be included as a part of the report of the analysis. On the 
 other hand, it should always be indicated when analyses are made 
 on samples which have been braided, or which have been vul- 
 canized in contact with a saturated braid or tape, since there will 
 be a migration of the liquid hydrocarbons of the saturation from 
 the braid or tape, into the rubber insulation, and although the 
 waxy hydrocarbons may be a bit low, because of the sandpaper- 
 ing of the surface, the acetone extract and the liquid hydrocar- 
 bons will be high. 
 
Chapter VI. 
 Extractions. 
 
 Certain organic substances, mainly the oils and waxes, are 
 removed by extract with acetone, chloroform, or by saponification 
 with alcoholic potash. The results obtained by these three opera- 
 tions are largely qualitative, and from them may be obtained a 
 fair index as to the quality of the article as a whole. In addition, 
 there are some substances which may be determined quite accu- 
 rately in these extracts. 
 
 Extraction Apparatus. The extraction apparatus should com- 
 ply with the following requirements: It should be of the reflux 
 type, with the condenser placed immediately above the cup which 
 holds the sample; the sample must be suspended in the vapor of 
 the boiling solvent; the cup must be of the syphon type; the cup 
 must be far enough away from the sides of the extraction flask 
 that it will be maintained at the temperature of the boiling point 
 of the solvent; only glass or metal joints may be used — there 
 shall be no cork, rubber, or similar material in the extractor, with 
 which the solvent may come in contact, and from which extract- 
 able matter may be obtained. 
 
 The extraction flask may be of a size that will permit it to be 
 weighed directly, or it is permissible to transfer the extract to a 
 smaller flask for evaporation, drying, and weighing. The Cottle 
 (better known as the Underwriters), the Joint Rubber Insula- 
 tion Committee, and Bureau of Standards types are all satisfac- 
 tory, and may be relied upon to give equally accurate results, but 
 any extractor which fulfills the above requirements will do. 
 
 Acetone Extract. 
 
 The acetone used in this extraction must be redistilled, and 
 free from water or acid. It should be distilled over sodium or 
 potassium carbonate, and kept in clean dark-colored glass bottles. 
 
 Place 2.000 gr. of the sample in an acetone extracted paper 
 
 68 
 
EXTRACTIONS 69 
 
 thimble, or fold it in an extracted filter paper; insert in the 
 syphon cup, and extract continuously for eight hours. 1 The heat- 
 ing must be controlled so that the solvent syphons about 20 times 
 per hour. Remove the solvent, dry the flask and contents at 90C 
 to constant weight, 2 and calculate to percentage. This figure is 
 usually called the "acetone extract, uncorrected." Due record 
 should be made of the color and odor of the extract, and of any 
 other peculiarities which may be noticeable. With high free 
 sulfur, or waxy hydrocarbons, these substances will separate out 
 on the sides of the flask. 
 
 Reserve the residue for the chloroform extraction. 
 
 The acetone dissolves the unchanged or free sulfur, vegetable 
 fats or oils, rosin, mineral oils, paraffin, ceresin, ozokerite, a con- 
 siderable portion of bituminous substances such as the mineral 
 rubbers, tars, etc., and the so-called resins of the crude rubber. 
 In simple mixtures, the separate constituents may be determined 
 
 1 Eight hours should suffice for auy properly prepared sample extracted under 
 exact conditions. However, some uncured samples may fuse together into a solid 
 mass, and require a longer time for comparatively complete extraction. In such 
 cases, extract until the solution in the extraction cup is colorless, and continue 
 for four hours longer. Uncured samples should be sheeted thin and rolled 
 between hardened filter paper, to effect a thorough and more rapid extraction. 
 The expression "complete extraction" is a misnomer : the free sulfur actually 
 is extracted in the first four hours, but the soluble organic matter is extracted 
 with difficulty. Additional quantities of extract can be dissolved up to 48 hours, 
 or even more, but the amount so obtained is but a small proportion of the total, 
 and is more or less constant. Hence, if we interrupt the extraction at a definite 
 point, we secure results which serve the purpose of indicating the quality of 
 the rubber, and are comparable with other extracts made in a similar manner. 
 The same is true to a large extent with the chloroform and alcoholic potash 
 extractions, and we really deal with comparable, rather than with absolute 
 values. 
 
 The necessity for continuous extraction is explained on the same basis ; with 
 samples of approximately the same degree of fineness, the extraction is a matter 
 of time, rather than the number of times the syphon empties; hence, standing 
 overnight would permit the solvent to extract a considerable quantity of soluble 
 matter that would not otherwise be extracted. Many of the variations in check 
 results are really due to faulty manipulation, rather than to the type of extrac- 
 tor, or fineness of the sample. 
 
 2 There has been considerable discussion as to the adoption of a standard 
 time for drying. Some samples are dry in half an hour and it is a waste of 
 time to continue for hours longer. On the other hand, the Joint Rubber Insula- 
 tion Committee found some samples, notably those high in solid hydrocarbons, 
 which were not dry in two hours. Sometimes in the hot, humid months of 
 summer, water may condense on the outside of the condenser of the extractor, 
 and some of this may find its way into the extraction flask. If it does, it must 
 be removed, even if it does take more than half an hour ; it is not extract, and 
 must not be weighed as such. Of course, the longer periods for drying may 
 lose a little more of the free sulfur than the shorter periods ; especial care 
 must be taken to see that the temperature does not go over 90C, for even at 
 this temperature, there is some loss of free sulfur and at higher temperatures, 
 over an extended period, the loss may be very great. 
 
70 THE ANALYSIS OF RUBBER 
 
 with some accuracy, but in the more complex ones only a few 
 of the constituents may be determined with sufficient accuracy to 
 be of any value. The free sulfur may always be determined with 
 a high degree of accuracy; in the absence of tars and mineral 
 rubber, paraffin and ceresin are capable of being determined with 
 equal accuracy. Fatty oils will be associated with the rubber 
 resins, and if we assume that the latter are about 3.5 to 4% of 
 the rubber present, we may get a fair line on the quantity of 
 vegetable oils, but such a scheme is only approximate. 
 
 Rosin may be determined by the method of E. J. Parry. 3 
 The fatty acids are dissolved in 20 cc. of 95% alcohol, a drop of 
 phenolphthalein is added, and then strong caustic soda (one part 
 of alkali to two parts of water) until the reaction is just alkaline. 
 The solution is heated for a few minutes, allowed to cool, and 
 then transferred to a 100 cc. stoppered graduated cylinder. The 
 latter is filled to the mark with ether, 2 gr. of powdered silver 
 nitrate is added, and the mixture shaken vigorously for fifteen 
 minutes, in order to convert the acids into their silver salts. 
 When the insoluble salts have settled, 50 cc. of the clear solution 
 (containing the silver salts of rosin) is pipetted off into a second 
 100 cc. cylinder, and shaken with 20 cc. dilute hydrochloric acid 
 (1 acid to 2 water). The ethereal layer is drawn off, and the 
 aqueous layer is shaken twice with ether. The ether extracts are 
 united, washed with water, and the ether distilled off in a 
 weighed beaker. The residue, rosin, is dried at 110 to 115C, 
 cooled, and weighed. 
 
 This is an excellent means of separating fatty oils and rosin; 
 it is best performed by taking the water solution in the deter- 
 mination of unsaponifiable matter, making it acid with hydro- 
 chloric acid, and extracting the liberated fatty acids with ether. 
 The ether must be driven off, and the fatty acids dried, before 
 the method may be used. 
 
 The mineral oils can be partly separated from hard paraffin, 
 sufficiently so as to give some indication of the composition. So 
 far as our experience goes, no method has been given which will 
 determine the relative amounts of mineral rubber and paraffin 
 in a mixture of the two. The possibilities of such a method 
 being developed are very remote, in view of the wide variations in 
 
 8 Allen's Commercial Organic Analysis, 4th ed., Vol. V, p. 73. 
 
EXTRACTIONS 71 
 
 the composition of the mineral rubbers, and the fact that chemi- 
 cally they are so nearly like paraffin. 
 
 Chloroform Extract. 
 
 The chloroform extraction is performed in the same apparatus 
 used in making the acetone extraction. The chloroform should 
 be redistilled over alkali. 
 
 Extract for four hours, the residue from the acetone extraction 
 (it is not necessary to remove the acetone adhering to the 
 sample), using about 60 cc. of the solvent. If at the end of four 
 hours, the solvent in the syphon cup is still colored, continue to 
 extract until it is colorless. Filter the extract through fat free 
 filter paper into a small Erlenmeyer flask, distil off the solvent, 
 and dry the flask and contents to constant weight at 95C. 
 
 If the chloroform extraction cannot be started immediately 
 after the acetone extraction has been completed, the sample 
 should be protected against oxidation by keeping it in a vacuum 
 desiccator in a vacuum of at least 50 mm of mercury. Vulcanized 
 rubber which has been extracted with acetone oxidizes very 
 rapidly in the air and the resultant products are so soluble in 
 chloroform as to yield hopelessly false results, being as much as 
 five to ten times the true amount. 
 
 Reserve the residue from the chloroform extraction for treat- 
 ment with alcoholic potash. 
 
 The chloroform dissolves part of the rubber, particularly the 
 undercured, and the oxidized rubber. Its chief value is that it 
 dissolves part of the mineral rubber, the solution taking on an 
 intense brown or black color. It is an invaluable qualitative test 
 for mineral rubbers, the color being quite distinctive, and not 
 likely to be mistaken for anything else. 
 
 The chloroform extract in a well cured and unoxidized sample 
 of soft vulcanized rubber, will run from 1 to 3% of the rubber 
 present, with the average nearer the lower figure. It has been 
 suggested as means of determining whether or not the rubber has 
 been undercured, but the data available are largely limited to 
 insulation compounds, and are not entirely convincing. 
 
 Alcoholic Potash Extract. 
 
 Dry the residue from the chloroform extraction at 60C until 
 the odor of chloroform is no longer noticeable. Place the rubber 
 
72 THE ANALYSIS OF RUBBER 
 
 in a 200 cc. Erlenmeyer flask, and cover with 50 cc. normal alco- 
 holic potash. 4 Boil for four hours under reflux condenser. Filter 
 by decantation through a hardened filter paper, wash with two 
 portions of 25 cc. of hot alcohol, and then thoroughly with hot 
 water. Evaporate the filtrate to dryness, take up in warm water 
 and when the solution has been effected, cool to room tempera- 
 ture. Transfer to a separatory funnel, add 30 cc. N/5 hydro- 
 chloric acid and sufficient water to bring the total up to about 
 100 cc. Add 40 cc. of ethyl ether, shake thoroughly, allow to 
 stand until the two layers are completely separated, draw off the 
 water into a second separatory funnel, and continue to extract 
 with 20 cc. portions of ether until a colorless solution results, and 
 then twice more. Unite all the ether fractions in the first separa- 
 tory funnel, and wash with water until the water shows no 
 further acidity (test with silver nitrate solution). Filter the 
 ether through a plug of extracted cotton into a weighed beaker or 
 flask, evaporate to dryness, and dry to constant weight at 95C. 
 Another method for the determination of the alcoholic potash 
 extract is to dry the residue from the chloroform extract, cool, 
 and weigh. Place the rubber residue in a 200 cc. Erlenmeyer 
 flask, add 50 cc. N/1 alcoholic potash, and boil under a reflux 
 condenser for four hours. Filter off the rubber on a Gooch or 
 alundum crucible, wash with hot alcohol, and then hot water, un- 
 til the washings are free from alkali; dry in an inert atmosphere 
 to constant weight; the loss in weight is the oil substitute. 5 
 
 * We should not be led astray by those who wish to replace potassium 
 hydroxide with sodium hydroxide. When the former was difficult to obtain, one 
 did what could be done with the material which was available, but no question 
 of a slightly higher cost should interfere now with the use of a better and 
 more widely known reagent. On the other hand, with pure grain alcohol diffi- 
 cult to obtain under present conditions in the United States, the use of 
 methylated alcohol becomes almost obligatory. It is hard to see just what error 
 would be introduced by the presence of methyl alcohol ; it is difficult to con- 
 ceive of anything which might be present in a rubber compound which is soluble 
 in methyl alcohol, and insoluble in acetone, chloroform or ethyl alcohol. If 
 the analyst will see that his denatured alcohol has been denatured with methyl 
 alcohol, and will use this denatured alcohol only after redistillation over caustic 
 potash, the chances for error are very small indeed. Of course, in any event, 
 and regardless of the kind of alcohol used, a blank is always run, and due cor- 
 rection made for the results found. Again, no careful analyst will use an alco- 
 holic potash solution which has been standing a long time, and particularly if 
 it is badly discolored. 
 
 6 This method is not as accurate as the previously mentioned one, and is not 
 to be recommended. There is the greatest difficulty in washing out all of the 
 alkali, and the latter cannot be removed with acids on account of the proba- 
 bility of these acids attacking some of the pigments in the rubber. 
 
EXTRACTIONS 73 
 
 Ordinary crude rubber shows a small amount of material solu- 
 ble in alcoholic potash, usually around 1% of the amount of 
 rubber. This will be included in the results in either method. 
 In the first method, we weigh the fatty acids, although they were 
 present originally as the glycerides; i.e., we weigh only 95% of 
 the substitute. These two elements tend to neutralize each other, 
 and the result is a pretty accurate determination of the amount of 
 fatty substitute, not including, of course, any unchanged oil which 
 would have been extracted in acetone, or any pigments contained 
 in the substitute. 
 
 If vegetable oils are used, and there is sufficient sulfur present, 
 we may find that a part of the oil has been converted into an 
 insoluble form, and will appear at this point. There is no way 
 to distinguish substitute formed during vulcanization and that 
 added as such, excepting that the oil in the substitute is usually 
 very well changed into the acetone-insoluble form, whereas the 
 oil added as such will be changed to only a slight extent. 
 
 The method involving loss of weight is practically worthless, 
 because it is an almost hopeless task to thoroughly wash out the 
 alkali. In one case continuous washing for 8 hours did not suf- 
 fice, and acid cannot be used to neutralize the alkali, on account 
 of its effect on the acid-soluble fillers. 
 
 Analysis of Acetone Extract. 
 
 Unsaponifiable Matter. 
 
 Add to the acetone extract, 50 cc. of alcoholic potash, boil under 
 a reflux condenser for two hours, and evaporate to dryness. Add 
 10 cc. of water and 20 cc. of ether, heat until solution is complete ; 
 cool, and transfer to a separatory funnel, wash out with warm 
 water, and cool, then with two 20 cc. portions of ether; the separa- 
 tory funnel should contain 100 cc. of water, and not less than 40 
 cc. of ether. Shake vigorously, allow the two layers to separate, 
 and draw off the aqueous layer into a second separatory funnel. 
 Repeat the extraction until no further material can be extracted 
 (not less than four extractions should be made) . Unite the ether 
 portions of the extract, and wash with water until free from alkali 
 (the first two portions may be united with the original aqueous 
 solution, and the whole reserved for the determination of free 
 sulfur) . Filter the ethereal layer through extracted cotton, wash- 
 
74 THE ANALYSIS OF RUBBER 
 
 ing with ether and hot chloroform, using the latter to rinse the 
 original flask, and both separatory funnels. Evaporate to dry- 
 ness, dry to constant weight at 95 to 100C, cool and weigh. 
 
 The above method gives the liquid and solid hydrocarbons, and 
 the unsaponifiable resins. The difference between the total ex- 
 tract, and the sum of the free sulfur and unsaponifiable matter, 
 will consist of the saponifiable resins, and any fatty oils which 
 may have been extracted. The acetone soluble matter of the 
 mineral rubber will be found largely in the unsaponifiable por- 
 tion. Rosin, as its composition indicates, will be distributed be- 
 tween the two, about 90% being saponifiable. 
 
 Waxy Hydrocarbons. 
 
 The time-honored method for separating the solid paraffins has 
 been to dissolve the unsaponifiable portion in alcohol, and freeze 
 out the paraffin. However, some of the latter will always remain 
 in the alcohol, along with any liquid mineral oils, and the un- 
 saponifiable rubber resins. The Joint Rubber Insulation Com- 
 mittee devised a method for correcting for the alcohol soluble 
 paraffin, in the absence of mineral oils, or, if the latter were 
 present, to get the total soluble paraffins and the mineral oil 
 together. The alcohol insoluble paraffins are called "Waxy hydro- 
 carbons A" and the soluble paraffins are called "Waxy hydrocar- 
 bons B." If the latter are solid, the sum of the two is the total 
 paraffin in the sample. 
 
 If it is desired to know only the total mineral hydrocarbons, 
 then the method for Waxy hydrocarbons B is used directly. 
 
 Waxy Hydrocarbons A. 
 
 Add 50 cc. absolute alcohol to the unsaponifiable matter and 
 warm until solution is as complete as possible. Cool the solution 
 to — 4 or — 5C, and maintain at this temperature, or lower, by 
 packing the flask in a mixture of ice and salt. Filter out the 
 waxy hydrocarbons, using a funnel packed with ice and salt and 
 applying suction if necessary. Wash the flask and filter with 
 25 cc. of 95% alcohol which has been previously cooled to the 
 same temperature. Dissolve the residue on the filter paper in 
 hot chloroform into the original flask; evaporate the chloroform, 
 and dry the residue at 95 to 100C to constant weight. 
 
EXTRACTIONS 75 
 
 Waxy Hydrocarbons B. 
 
 Evaporate the alcohol from the determination of Waxy hydro- 
 carbons A, add 25 cc. of carbon tetrachloride, and transfer to a 
 separatory funnel. Shake with cone, sulfuric acid, drain off the 
 discolored acid, and repeat with fresh portions of the acid until 
 there is no longer any discoloration. Vigorous shaking is abso- 
 lutely necessary for the success of the method. After drawing 
 off all of the acid, wash the carbon tetrachloride solution with 
 repeated portions of water until all traces of acid are removed. 6 
 Transfer the carbon tetrachloride solution to a weighed flask, 
 evaporate off the solvent, and dry to constant weight at 95 to 
 100C. Note whether the residue is solid, liquid, or pasty. 
 
 • On account of the specific gravity of the carbon tetrachloride washing with 
 water is a very tedious proposition, because the carbon tetrachloride must be 
 drawn off with each washing, and returned to the flask. While the Joint Rubber 
 Insulation Committee did not recommend it, the carbon tetrachloride may be 
 diluted with ether until the mixed solvents have a gravity lower than that of 
 water ; the washing can then be continued as usual with separatory funnel 
 washings. Ether to the extent of about two and a half to three times the 
 volume of carbon tetrachloride will be necessary to have the ether-tetrachloride 
 mixture float on the water layer. 
 
Chapter VII. 
 The Determination of Rubber. 
 
 It is a peculiar fact concerning the analysis of rubber that the 
 determination of the principal constituent involved is seldom, if 
 ever, made by a direct determination. A tremendous amount of 
 research has been undertaken, methods and revisions of methods 
 have been suggested, but as yet no one method has succeeded in 
 securing the endorsement of the rubber analysts. 
 
 The methods for the determination of rubber may be classified 
 under three headings: (1) direct; (2) indirect; (3) difference. 
 In No. 1, the idea is to form a definite compound with rubber 
 and either weigh the compound directly or determine some part 
 of the compound and from these figures to calculate the total 
 rubber. The two principal methods in this group are the tetra- 
 bromide methods and the nitrosite methods. The indirect 
 methods (No. 2), proceed to separate the rubber, but the latter 
 is not determined as such, but is determined as the loss during 
 the solution. The difference methods comprise the third group, 
 and the principle involved is merely to determine every other 
 known constituent, subtract the total from 100%, and call the 
 remainder rubber. 
 
 The Tetrabromide Method. 
 
 The tetrabromide method was first advocated by Budde, for 
 use in determining the rubber in unvulcanized compounds, or 
 crude rubber. The bromination solution used was 6 gr. of bromide 
 and 1 gr. of iodine in 1000 cc. of carbon tetrachloride. The rubber 
 was swollen in carbon tetrachloride, and filtered. The clear solu- 
 tion was treated with 50 cc. of the bromine solution, allowed to 
 stand for 24 hours, diluted with an equal volume of alcohol, and 
 when the precipitate had settled it was filtered and washed with 
 carbon tetrachloride-alcohol (1-1), and finally, to remove the bro- 
 mine, with alcohol. The precipitate was weighed as C 10 H 16 Br 4 , 
 and calculated to rubber, using the factor 0.298. 
 
 76 
 
THE DETERMINATION OF RUBBER 77 
 
 The gravimetric method did not prove successful, and evoked 
 considerable criticism. Fendler, Harries, Hubener, Spence, and 
 others, presented various modifications, and, in the meanwhile, 
 Budde had published a volumetric method, which he claimed was 
 satisfactory for rubber vulcanized with sulfur chloride. In this 
 method, the rubber swollen in carbon tetrachloride was treated 
 with the brominating mixture, and after the tetrabromide had 
 been filtered free from bromine, it was treated with cone, nitric 
 acid and N/5 silver nitrate, and the bromine determined as in 
 Volhard's method. 
 
 The volumetric method did not prove any more acceptable than 
 the gravimetric. For some compounds, good results were ob- 
 tained, but in others, especially with vulcanized rubber, it was 
 found that the bromine did not replace the sulfur of vulcaniza- 
 tion. There was also found to be a loss of bromine during the 
 acid treatment, which Spence corrected by fusing the tetra- 
 bromide with alkali. Vulcanized samples gave low results, but 
 when it was noted that the sulfur combined with the double bonds 
 of rubber in stoichiometric proportions, it was seen that by adding 
 to the bromine found in the tetrabromide the bromine equivalent 
 of the sulfur of vulcanization (2Br = S) , the results were more 
 uniform, and more nearly correct. It was also noticed that 
 hydrobromic acid was formed during bromination, and efforts 
 were made to eliminate this factor by freezing the brominating 
 solution, but, on the whole, while cooling reduced the formation of 
 hydrobromic acid, it did not eliminate it. 
 
 Recently, further attempts have been made to make the method 
 practical. Lewis and McAdam 1 published a modification based 
 on Mcllhenny's 2 method for the determination of substitution, 
 and Fisher, Gray and Merling 3 have recommended some im- 
 provements in the Lewis and McAdam method. 
 
 When bromine adds to rubber, whether the latter be vulcanized 
 or not, there are a number of ways in which the reaction may 
 progress: 
 
 (1) HC : CH + 2 Br = HCBr.HCBr 
 
 (2) HCBr.HCBr = HC : CBr + HBr 
 
 (3) HC : CBr + 2 Br = HCBr.CBr 2 
 
 (4) HC : CH + HBr = HCBr.CH 2 " 
 
 (5) CH 2 .CH 2 + Br 2 = CHBr.CH 2 + HBr 
 
 1 J. Ind. Eng. Chem. 12, 675-6 (1920). 
 3 J. Am. Chem. Soc. U, 1084 (1899). 
 3 J. Ind. Eng. Chem. is, 1031-4 (1921). 
 
78 THE ANALYSIS OF RUBBER 
 
 The first reaction is purely additive; the second is a splitting 
 off of HBr, re-forming the double bond, which again combines 
 with 2 Br as shown in No. 3; the liberated hydrobromic acid may 
 unite at a new double bond, as in No. 4; and finally, No. 5 is a 
 case of straight substitution. The resulting product will contain 
 HCBr.HCBr; HCBr.CH 2 ; HCBr.CBr 2 ; HBr and Br. It is ap- 
 parent that every molecule of HBr remaining uncombined with 
 the rubber represents a loss of 2 Br from the excess over that 
 required for the double bonds. This has been one of the serious 
 errors, and it is one which varies greatly with variations in the 
 condition of time, temperature, and concentration of the solutions. 
 
 In determining the iodine number of "burnt" linseed oils, Smith 
 and Tuttle 4 found that concordant results could be obtained only 
 when a very exact procedure was followed, in which the weight 
 of the sample, volume and strength of the iodine solution, time 
 and temperature of the reaction, were specified within very nar- 
 row tolerances. The analogy in chemical reactions, between dry- 
 ing oils and rubber, is very striking, and we may expect to find 
 just as great difficulties with vulcanized rubber as with oxidized 
 linseed oil. Lewis and McAdam brominate for 2-4 hours, while 
 Fisher, Gray and Merling say 2.5 to 3.5 hours. The amount of 
 the sample is quite indefinite, not over 2.00 gr. for unvulcanized; 
 1.50 to 2.00 gr. for vulcanized. In the latter case, no special at- 
 tention seems to have been paid to whether the material contained 
 30 or 90% of rubber hydrocarbons; nor to whether they are deal- 
 ing with rubber having a vulcanization coefficient of 2.0 or 10.0. 
 There is a considerable amount of lack of definiteness in a reac- 
 tion which, in a similar determination (adding halogen to partly 
 oxidized oils), has been shown to be absolutely essential. It is 
 perhaps not to be wondered at that after so many years of trial, 
 there still remains such an element of doubt. 
 
 In spite of the fact that it requires an extra determination 
 (sulfur of vulcanization), the tetrabromide method offers an 
 excellent opportunity for a direct method, if some one will take 
 the time to ascertain the exact conditions which are necessary 
 for consistent results. 
 
 *J, In<J, and Eng. Cheni. 6, 994 (1914). 
 
THE DETERMINATION OF RUBBER 79 
 
 Nitrosite Method. 
 
 The nitrosite method was first described by Harries. 5 He 
 found that by allowing N 2 3 to react for a sufficient time with 
 a solution of rubber, he obtained a final product of constant 
 composition, C 10 H 15 N 3 O 7 , and by weighing this material, the rub- 
 ber hydrocarbon could be calculated. Rubber resins were re- 
 moved by extraction with acetone. Alexander in attempting to 
 repeat Harries work, claimed that during the formation of the 
 nitrosite, carbon dioxide was given off, and that the composition 
 of the final nitrosite of Harries was really C 9 H 12 N 2 O c . The loss 
 of carbon dioxide has not been confirmed by any other workers in 
 this field, so that if carbon dioxide was really lost, as Alexander 
 says, it was because of some condition not in accordance with 
 Harries method. However that may be, every one agreed upon 
 the difficulty in getting a constant composition for their end 
 product, and this method was more or less abandoned until Wes- 
 son 6 took it up from a new angle. He proved that Alexander 
 was wrong in his statement that carbon dioxide was lost, and 
 hence, although the final product varied in composition, it still 
 held all of its original carbon. If, therefore, the carbon in the 
 nitrosite was determined, a simple calculation could be made 
 directly to rubber which did not depend upon any definite com- 
 position of the nitrosite and was independent of the sulfur of 
 vulcanization. Wesson secured some good results on a few com- 
 pounds, but with some complicated commercial compounds, large 
 differences cropped out. Tuttle and Yurow, 7 while investigating 
 the possibilities of Wesson's method, found that his best results 
 were obtained by a fortunate balancing of errors, and that when 
 the causes for these errors were removed, accurate results could 
 be obtained directly in the presence of practically any known 
 organic or inorganic fillers. The only unfortunate circumstance 
 connected with this method is the fact that it requires a fairly 
 complicated combustion train, and this has prevented many 
 laboratories from testing it out. Perhaps further experimentation 
 along the lines of wet combustion would simplify it sufficiently to 
 permit its more general adoption. In the meantime, it stands as 
 
 °Ber. 81,, 2991-2 (1901) ; 35, 3256: 4429 (1902) ; 36, 1937 (1903). 
 • J. Ind. EDg. Chem. 5, 398 (1913) ; 6, 459-62 (1914) ; 9, 139-40 (1917). 
 7 India Rubber World, 57, 17-8 (1917) ; Bureau of Standards Technologic Paper 
 145 (1919), 
 
80 THE ANALYSIS OF RUBBER 
 
 our only accurate direct method for the determination of rubber, 
 irrespective of the condition of the rubber product, and the degree 
 of vulcanization. The method is as follows: 
 
 The preliminary extractions with acetone, chloroform, and 
 alcoholic potash will show whether mineral rubbers or oil sub- 
 stitutes are present. If the former, then acetone and chloroform 
 extractions are necessary ; and with oil substitutes, make acetone 
 and alcoholic potash extractions; and if both are present, make 
 all three extractions. When an alcoholic potash extraction is 
 made, wash the sample thoroughly with 5% hydrochloric acid, 
 hot water and alcohol. 
 
 Take 0.500 gr. of the finely ground sample (call this weight 
 W) , and extract with acetone for eight hours (make other extrac- 
 tions, if necessary, as stated above) . Dry the residue in hydrogen 
 (or other inert gas) for two hours at 100C. Place the sample in 
 50 to 75 cc. of chloroform and allow it to swell. Pass into this, 
 until the green color which is formed persists for 30 minutes, the 
 gases formed by heating arsenic trioxide and nitric acid of 
 specific gravity 1.30. To avoid contamination, it is important 
 that no rubber connections be used. Immerse the flask contain- 
 ing the rubber in cold water during the nitration. Allow the solu- 
 tion to stand overnight; the next day, filter off the nitrosite 
 through a Gooch crucible and wash with small quantities of 
 chloroform. Remove the acid gases and chloroform from the 
 flask by means of a gentle current of air. Evaporate the filtrate 
 to dryness. Dissolve the nitrosite remaining in the flask in the 
 Gooch crucible and in the residue from the filtrate in acetone, 
 and filter the solution through asbestos into a weight-burette. 
 The total volume should be about 100 cc. Allow this solution to 
 stand for a short time to permit any sediment which may form to 
 settle out in the bottom of the weight burette. 8 Weigh the 
 burette before and after filling, calling the difference N. Draw 
 off about 25 cc. into a small Erlenmeyer flask, reweigh the 
 burette, and call the difference O. Evaporate the portion drawn 
 off to a small volume, transfer to a porcelain boat (about 14 cm. 
 long and 1 cm. wide), which has been filled with alundum, and 
 wash the flask with acetone (it is best to make this transfer in 
 small portions, drying the boat and contents for a few minutes be- 
 
 8 The usual type of weight burette will not answer ; it is necessary to have 
 the solution drawn off at the side, about an inch from the bottom. See B. of 
 S. Tech. Paper 145 for a sketch of the weight burette. 
 
THE DETERMINATION OF RUBBER 81 
 
 tween each addition). After the final washing and drying, add 1 
 or 2 cc. of 1% solution of ammonia in distilled water, 9 and dry in 
 an inert gas for one hour at 90C. Repeat with a second portion of 
 ammonia and dry as before. By this means, all of the organic 
 solvent will be removed. 
 
 Place the boat in the furnace, and proceed with the combus- 
 tion. Pass the products of combustion through U-tubes or other 
 satisfactory absorption tubes, placed in the following order: a, 
 b, c, potassium bichromate-cone, sulfuric acid; d, 20-mesh 
 powdered zinc; e, f, soda-lime and calcium chloride; g, potassium 
 bichromate-cone, sulfuric acid; h, dilute palladium chloride solu- 
 tion, (very little palladium chloride is needed; use about a drop 
 of a 10% solution in 10 cc. of water) . Weigh e, f , and g before 
 and after each combustion; refill c and g frequently from the 
 same stock solution, so that the gases which enter e, and those 
 that leave g, will have the same moisture content. The pal- 
 ladium chloride serves to detect the presence of carbon monoxide 
 or other reducing gases; if there is any blackening, it shows in- 
 complete oxidation, and, in this event, the results must be dis- 
 carded and the determination repeated. 
 
 The carbon dioxide will equal the algebraic sum of the differ- 
 ences in tubes e, f, and g. Call this sum P. The factor for cal- 
 culating from carbon dioxide to rubber hydrocarbons, is .309. 
 The formula is therefore as follows: 
 
 n w — — = % of rubber hydrocarbons. 
 
 u x vv 
 
 Correct this figure for whatever extractions were made previ- 
 ous to nitration. 10 
 
 The Indirect Methods. 
 
 The indirect methods comprise those in which the rubber is 
 decomposed and rendered soluble in some solvent, and the ma- 
 terial which goes into solution is called rubber hydrocarbons. A 
 great many solvents have been suggested, heavy petroleum (B.P. 
 230-280) , anisole, phenetole, xylol, paraffin oil, camphor oil, tere- 
 
 • The nitrosite precipitate is insoluble in distilled water, and in acids ; it is 
 soluble in aqueous alkaline solutions. Ammonium hydroxide is used simply be- 
 cause it provides the necessary alkalinity, is volatile, and, even if it is not 
 completely driven off, introduces no error whatsoever. 
 
 10 In case the extractions contain other material than rubber, the correction 
 applied must be an arbitrary one. It will be recalled that the true resin con- 
 tent of high grade rubber is about 4% or less, and the normal chloroform and 
 alcoholic potash extracts of rubber are about 2% and 1% respectively. 
 
82 THE ANALYSIS OF RUBBER 
 
 bene, turpentine, salol, nitrobenzene, aniline, cumene and cymene. 
 The general procedure is the same for all of them; the acetone, 
 chloroform and alcoholic potash extractions remove the soluble 
 organic fillers; the solvent is then added and heated until the 
 rubber passes into solution. The fillers are separated by filtration 
 and weighed, and the loss is vulcanized rubber. Sometimes 
 special steps are taken to bring the fillers into a state in which 
 they can be easily filtered. All sorts of solvents are used to 
 wash the fillers. 
 
 The principal objection to these methods is that one can never 
 be sure that there are no organic or inorganic fillers passing into 
 solution, or that insoluble organic compounds are not formed 
 from the rubber and solvent. Correction after correction must be 
 applied, until finally, instead of a determination of rubber, we 
 have almost a complete system of analysis. 
 
 Difference Methods. 
 
 There is really no sharp line dividing the difference methods 
 from the indirect methods, but we have reserved the term "differ- 
 ence" methods for those which remove the rubber by ignition. 
 Such methods are widely used, largely because they are rapid, 
 and it is a fact that they are frequently quite accurate, espe- 
 cially in the absence of organic fillers. In most cases, the sample 
 is subjected to acetone, chloroform and alcoholic potash extrac- 
 tions, and then ignited. At times, only the acetone extraction is 
 made. A more complicated, and more exact procedure, is that 
 of the Joint Rubber Insulation Committee. 11 This method is 
 intended for 30% and 40% insulation compounds, but may be 
 used for any compound containing no organic fillers. Mineral 
 rubber, lampblack, glue, cellulose, etc., all give high results. The 
 method is as follows: 
 
 Add to the rubber residue from the alcoholic potash extrac- 
 tion, sufficient water to make the total volume 125 cc, and then 
 add 25 cc. cone, hydrochloric acid. Heat for one hour, decant 
 through a Buchener funnel, using hardened paper, and wash with 
 25 cc. of hot water; repeat this treat twice. The rubber should 
 be white, and free from black specks from undissolved fillers 
 (lead sulfide). Wash the rubber free from chlorides, transfer the 
 rubber to the filter paper, dry as much as possible by suction, 
 
 11 Am. Inst. Elec. Eng., April, X917. 
 
THE DETERMINATION OF RUBBER 83 
 
 wash with 50 cc. of 95% alcohol, and transfer the entire residue 
 to a weighing bottle. Dry to constant weight at 95 to 100C. Let 
 this weight be represented by C. 
 
 On a portion D of this residue C, determine the ash E (see 
 below) and the sulfur F in ash E. Determine the sulfur H in 
 another portion G of residue C. 
 
 To determine E, place about 0.50 of residue C into a weighed 
 porcelain crucible, heat gradually until the crucible has ceased 
 to smoke, then raise the temperature and heat until all organic 
 matter is destroyed. Cool and weigh, calling the residue E. If E 
 is small, the determination of sulfur in the ash may be omitted, 
 and F assumed to be zero. From the data thus obtained, the 
 percentage of rubber hydrocarbons in the original material is 
 calculated as follows: 
 
 — 1 1 — — ^ -p- ) = % rubber hydrocarbons. 
 
 The simplest method of all, is the determination of the ash. 
 Wrap a 1 gr. sample in filter paper, and extract with acetone for 
 four hours; ignite the residue in a porcelain crucible; cool and 
 weigh. Correct for the sulfur in the ash by adding a few drops 
 of nitric acid-bromine mixture, heat on the steambath, then add 
 5 gr. sodium carbonate, dry carefully until all moisture is 
 removed, fuse the residue, extract with hot water and filter; make 
 the filtrate just acid with hydrochloric acid, heat to boiling, add 
 barium chloride solution, and determine the barium sulfate as 
 usual. Deduct the sulfur thus found from the residue on ignition, 
 and the difference is called "ash, sulfur free." 
 
 In the ash method, rubber is the difference between 100% and 
 the sum of the total sulfur, ash (sulfur free), and any other 
 determinations of fillers which may have been made. 
 
 The only good word which can be said for the ash determina- 
 tion as a correct method for the determination of rubber, is that 
 it requires only a crucible, a Bunsen burner, and a balance. 12 
 
 12 It is quite within reason that those analysts who have for years been 
 adhering to the "ash method," will find fault with this statement. With a few 
 simple compounds, satisfactory results may be obtained, but beyond this nothing 
 is certain. Our experience with this method, in testing rubber materials for the 
 U. S. Government over a number of years, demonstrated only too clearly how 
 easy it was to be led astray by results obtained with the ash method. The 
 errors in this ash method are frequently very large — it is only the occasional 
 determination which comes with 2 or 3% of the truth. The plain truth of the 
 matter is that the ash method is used because it involves little labor, and re- 
 quires but a short time to complete, but if a reasonable degree of accuracy is not 
 necessary, one is tempted to ask "Why make the determination at all?" 
 
Chapter VIII. 
 
 Sulfur Determinations. 
 Total Sulfur. 
 
 Sulfur may occur in rubber compounds in any of the follow- 
 ing forms: 
 
 (a) Free sulfur. 
 
 (b) Sulfur combined with the rubber. 
 
 (c) Sulfur in organic compounds (mineral rubber, oil sub- 
 stitutes, accelerators). 
 
 (d) Sulfides (zinc, antimony, mercury, lead, cadmium). 
 
 (e) Sulfites and sulfates (calcium, barium, lead). 
 
 In addition to the above, there are other substances, such as 
 barium carbonate, lead oxide, etc., which, while not containing 
 sulfur, are important factors in deciding what method is satis- 
 factory for the determination of sulfur. We may say that all 
 rubber compounds will contain classes (a) and (6), but beyond 
 that, nothing definite may be assumed. It is obvious that if many 
 of these sulfur-bearing substances are present, the determination 
 of total sulfur becomes a difficult proposition, and, moreover, the 
 results obtained decrease in value. 
 
 In looking over the foregoing list, it is obvious that three steps 
 are essential: (1) oxidation of organic and inorganic substances; 
 (2) fusion of inorganic substances; (3) separation of metals 
 forming insoluble sulphates by filtration of the alkaline solution 
 of the fusion. 
 
 The purpose of the oxidation is obvious; the oxidizing agent 
 must be sufficiently effective to oxidize rather large amounts of 
 free sulfur (frequently 5%, and possibly 10% or 12%). Since it 
 is assumed that oxidation will be accompanied or followed by 
 fusion, it is not essential that the oxidizing treatment be carried 
 to the point of the complete oxidation of all of the organic sub- 
 stances. The period of fusion must suffice to convert insoluble 
 
 84 
 
SULFUR DETERMINATIONS 85 
 
 into soluble sulfates. These steps are fairly well agreed upon, 
 although various means are suggested for this part of the proce- 
 dure. The third step, nitration, is still a subject of controversy, 
 although why it should be so is difficult to understand. If the 
 solution is acidified before filtration, we may expect to form 
 calcium, lead, and barium sulfates. Calcium sulfate is highly 
 soluble, lead appreciably so, and barium sulfate scarcely at all. 
 Calcium may have been present originally as the carbonate 
 (whiting) , or as the sulphate (in antimony compounds) ; the lead 
 as oxide, sulfide, sulfite, or sulfate, and barium as the carbonate 
 or sulfate. The purpose of the filtration from an acid solution is 
 to thus eliminate the insoluble sulfates originally present in the 
 rubber mixture, but this will be accomplished only when barium 
 sulfate is the only sulfate originally present, and lead and calcium 
 are present in very small amounts, and barium carbonate is 
 absent. It is evident that only on rare occasions will the condi- 
 tions be such as to permit the nitration of the solution of the 
 melt from an acid solution, and with compounds of unknown 
 composition, it is impossible. 
 
 In the earliest attempts to determine the sulfur in rubber, 
 Henriques oxidized with cone, nitric acid. Later, Alexander 
 suggested sodium peroxide; Hinrichsen, a modification of Gas- 
 parini's electrolytic oxidation (afterwards improved by Spence) ; 
 Waters and Tuttle employed cone, nitric acid and bromine; 
 Pontio, manganese peroxide ; Frank and Markwald, fuming nitric 
 acid; and Kaye and Sharpe fused directly with zinc oxide and 
 potassium nitrate. Some advocate a solution, others direct fusion, 
 while practically all of those who had a preliminary wet oxida- 
 tion added a fusion later. Solution without fusion is obviously a 
 faulty procedure in the presence of lead and barium salts, and no 
 data have as yet been presented to show that the direct fusion 
 methods are accurate when the free sulfur is high. Of the 
 methods employing both solution and fusion, that of Waters and 
 Tuttle has given the most consistent results with all types of 
 compounds, especially when some minor changes from that 
 originally proposed are employed. The method recommended is 
 as follows: 
 
 Place 0.500 gr. of rubber in a porcelain crucible of about 50 cc. 
 capacity, add 20 cc. of cone, nitric acid saturated with bromine, 
 cover the crucible with a watch glass, and allow it to stand for 
 
86 THE ANALYSIS OF RUBBER 
 
 one hour. Heat the crucible gently for one hour, then remove 
 the watch glass, rinsing it with little water, and evaporate the 
 solution to dryness (with pure gum compounds before evaporat- 
 ing add 0.1 to 0.2 gr. of potassium nitrate). Add 5 gr. of 1-1 
 mixture of sodium carbonate and potassium nitrate, and 1 or 2 
 cc. of distilled water; digest for a few minutes, and then spread 
 the paste along the sides of the crucible, and dry on a steambath. 
 Fuse the mixture, being careful to avoid contamination of sulfur 
 from the flame. When the fusion is cold, place the crucible and 
 contents in a beaker with about 250 cc. of water and heat for 
 several hours; filter off the insoluble carbonates, washing with hot 
 water. The total volume of the filtrate should be between 300 
 and 40 cc. Add 7 to 8 cc. cone, hydrochloric acid, cover the 
 beaker, and heat on the steambath. Add 10 cc. 10% barium 
 chloride, and allow to stand overnight; filter off the precipitated 
 barium sulfate, ignite carefully over a Bunsen flame, cool and 
 weigh. Calculate to sulfur, using the factor 0.1373. 
 
 The principal change from the published method is the addi- 
 tion of the potassium nitrate before evaporating off the nitric 
 acid; it is necessary only in the absence of any basic fillers, and 
 serves the purpose of changing any free sulfuric acid into potas- 
 sium sulfate. With sulfuric acid, there is some danger of it being 
 reduced by the organic matter, and sulfur lost as S0 2 . 
 
 Probably the weakest feature of this method is the wear and 
 tear on the crucibles, if this indeed can be considered a weak 
 point. There are some makes of crucibles which will not last 
 through a determination ; on the other hand, some American and 
 Japanese crucibles last through five to ten fusions. At this rate, 
 the cost is negligible. It might be added further, in speaking of 
 crucibles, that the smaller and thinner crucibles last longer than 
 the thicker ones ; much of the crucible trouble has been due to the 
 use of extra large and thick crucibles, and to the use of inferior 
 makes. 
 
 Occasionally, one finds the statement that the solution of the 
 fusion should be evaporated to dryness with hydrochloric acid 
 to get rid of the last traces of nitric acid, and to render insoluble 
 the silica in the alkali silicates which have been formed. The 
 small amounts of nitrates remaining after the fusion will not 
 appreciably affect the accuracy of the determination and 
 
SULFUR DETERMINATIONS 87 
 
 dehydration of the silica is unnecessary, as Hiliebrand x has 
 shown. However, it is essential that the alkaline solution should 
 be reasonably cool when the acid is added, that undue excess of 
 acid be avoided, and that the solution be not allowed to concen- 
 trate to any large extent. We have frequently found consider- 
 able amounts of silica when the solution containing the precipi- 
 tated barium sulfate had been allowed to concentrate to 50 cc. 
 or less. 
 
 Allen and Johnston 2 have shown that the precipitate of barium 
 sulfate formed in the presence of alkali chlorides and hydrochloric 
 acid, is contaminated with chlorine and alkalies, and have 
 worked out a method for correcting these errors, and so arrive 
 at the true sulfur value. The precipitate obtained in the deter- 
 mination of total sulfur in rubber is subject to these same errors, 
 but if the barium chloride be added rapidly to the hot solution, 
 the solution never heated to boiling, and, further, if it is allowed to 
 stand for at least 18 hours before filtering, the contamina- 
 tion will be low, and the fortunate balancing of errors will give 
 results very close to the truth, so much so that it will not ordina- 
 rily pay to take the time for the corrections suggested by Allen 
 and Johnston. It should be noted, however, that any attempt to 
 improve the method of precipitation by eliminating only one of 
 the errors, will yield results which are not as accurate as if the 
 method was not changed. 
 
 If the free sulfur is low, the fusion method of the Joint Rubber 
 Insulation Committee 3 will be found acceptable: 
 
 Mix 0.500 gr. of rubber with 4 gr. sodium peroxide and 6 gr. 
 potassium carbonate in a dry 15 cc. iron crucible, and cover. 
 Insert the crucible in a hole in a heavy brass plate so that about 
 two thirds of the crucible projects through the hole. Heat cau- 
 tiously until the first part of the reaction has taken place, and 
 then increase the heat until the mixture fuses. Remove the 
 flame and cool; place the crucible and cover in a porcelain cas- 
 serole containing 200 cc. of water, add 5 to 10 cc. of bromine 
 water, and boil until the melt is dissolved. Allow the precipitate 
 
 1 Analysis of silicate and carbonate rocks, U. S. Geological Survey Bull. 422, 
 p. 198. 
 
 2 J. Am. Chem. Soc. 32, 588-617 (1910) ; see also Richards and Parker, Proc. 
 Am. Acad. SI, 67 (1896) ; Hulett and Duschak, Z. Anorg. Chem. 1,0, 196 (1904) ; 
 John Johnston and L. H. Adams, J. Am. Chem. Soc. S3, 829-45 (1911). 
 
 3 J. Ind. Eng. Chem. 6, 75-82 (1914). 
 
88 THE ANALYSIS OF RUBBER 
 
 to settle, decant the solution through a thick filter and wash with 
 hot water. 4 Make the filtrate faintly acid with hydrochloride 
 acid, heat to boiling, add 10 cc. of 10% barium chloride solution, 
 allow to stand overnight; filter the barium sulfate as usual. 
 
 This method was originally recommended for testing insulated 
 wire, in which the free sulfur was limited to 0.7%, and was found 
 quite satisfactory. Tuttle and Isaacs 5 found that with high free 
 sulfur, the results obtained were not accurate. It has been sug- 
 gested that by increasing the quantities of sodium peroxide and 
 potassium carbonate, even these high free sulfur samples could 
 be analyzed without any trouble, but data are lacking in support 
 of this contention. 
 
 It would be extremely desirable, from the time and labor-sav- 
 ing points of view, if the oxidation of the free sulfur and the 
 fusion could be accomplished in one treatment. Spence's 6 elec- 
 trolytic method eliminates the fusion, in the absence of lead and 
 barium salts. Evans and Merling 7 have devised a method, 
 using a Parr calorimeter: 0.200 gr. of rubber is packed in sodium 
 peroxide, with some sugar and potassium chlorate. The ignited 
 mass is extracted with water, filtered, acidified, and the sulfur 
 precipitated as usual. The authors claim to have secured some 
 excellent results so far, and the time required is very little, but 
 it seems desirable that others test this method to discover its 
 limitations, and faults, if it has any. 
 
 Free Sulfur. 
 
 The procedure to be adopted for determining the sulfur in the 
 acetone extract depends largely upon the nature of the material, 
 and whether it is desired to make further separation of the 
 constituents in the extract. If not, the simplest, and yet the most 
 accurate method we have, is as follows: 
 
 To the dried extract, add 100 cc. of water, and 3 to 5 cc. of 
 bromine. (If a very high free sulfur is indicated by the character 
 of the extract, the amount of bromine should be increased.) 
 Allow the flask to stand for half an hour to an hour, boil off 
 the bromine, and when the solution is practically colorless, filter 
 
 • The original method called for dehydration of silica, but, as previously noted, 
 this is unnecessary. 
 
 6 J. Ind. Eng. Chem. 7, 658 (1913). 
 •J. Ind. Eng. Chem. k, 413 (1912). 
 1 India Rubber World, 64, 658 (1921). 
 
SULFUR DETERMINATIONS 89 
 
 through a folded filter into a small beaker; cover the beaker, 
 heat to boiling, add 10 cc. of 10% barium chloride, and after 
 standing overnight, determine the barium sulfate as usual. 
 
 This method determines all of the sulfur in the extract; a great 
 many checks have been run by taking the insoluble residue, fus- 
 ing it with sodium carbonate and potassium nitrate, and deter- 
 mining the sulfur as is done in the Waters and Tuttle method, but 
 the sulfur has never exceeded 0.01 to 0.02% in this residue. The 
 oxidation is complete, rapid, requires no evaporation, furnishes 
 its own acidity by the reaction with the barium chloride ; in fact, 
 after ten years, there still remains to be found some objection to 
 its use. 
 
 If it is desired to make further examination of the acetone 
 extract, the method of the Joint Rubber Insulation Committee 8 
 is recommended: the method starts where the acetone extraction 
 has been treated with alcoholic potash, the alcohol removed, the 
 residue taken up in water, extracted with ether, and the ether 
 washed with water. 
 
 To the aqueous solution, add 2 gr. potassium nitrate; evaporate 
 to dryness in a nickel or silver dish, and heat to quiet fusion. 
 Transfer to a beaker, neutralize with hydrochloric acid, add 2 cc. 
 of acid in excess, filter and wash, making the filtrate up to 200 cc. 
 Heat to boiling, add a slight excess of barium chloride solution, 
 allow to stand overnight, and determine the barium sulfate as 
 usual. 
 
 Kelly 9 calls attention to the fact that what we have been deter- 
 mining as free sulfur is not the true free sulfur, but includes, in 
 addition to the sulfur which may be said to be available for 
 further vulcanization, such amounts of sulfur which may have 
 been combined with the organic resins extracted by acetone. 
 Obviously, this is so, and in the data presented by Kelly, which, 
 however, covers only one compound, there is about 0.40% of 
 sulfur combined with the organic matter in the extract. In such 
 cases, the free sulfur as determined in the past is quite misleading. 
 It is still a very great question as to whether the sulfur will 
 always be of the same order of magnitude as Kelly indicates. It 
 would have been very helpful if commercial samples had been 
 
 *Loc. cit. 
 
 •The determination of the true free sulfur, and the true sulfur of vulcaniza- 
 tion; J. Ind. Eng. Chem. 12, 875-8 (1920). 
 
90 THE ANALYSIS OF RUBBER 
 
 treated — in the many samples we have tested, we have frequently 
 found samples where the free sulfur, as determined by the bro- 
 mine method, was less than 0.10%. Upton 10 gives figures for 
 free sulfur on some samples of insulated wire, determined by two 
 methods, and several of these were below 0.20%, and in one case 
 only 0.07%. Without questioning the force of Kelly's argument, 
 it does seem as though we needed more data to show what varia- 
 tions there are between the free sulfur as at present determined, 
 and the amount he calls the true free sulfur. 
 
 Kelly's method is as follows: 
 
 The acetone extraction is performed as usual. The acetone is 
 driven off at not over 65C. To the residue, add 50 cc. of 75% 
 alcohol which has been saturated with sulfur. Weigh the flask 
 and contents to 0.5 gr. Heat for a few minutes, to get the organic 
 matter in solution, and then cool slowly. Allow to stand three 
 hours; reweigh, and add 75% alcohol, sulfur- free, to replace the 
 loss. Decant the solution, wash two or three times with 75% 
 alcohol saturated with sulfur, and then dry. The sulfur may 
 then be determined by any satisfactory method. 
 
 No word is said as to what is to be done with the alcoholic 
 solution of the resins; no scheme has been given for weighing 
 them, and at 65C the extract would not be sufficiently dried to 
 take that figure as the total acetone extract. If one knew just 
 how much alcohol was used, and the sulfur it contained, the 
 solution could be evaporated to dryness in a weighed flask, dried 
 to constant weight at 90C, and the organic extract determined 
 after making due allowance for the sulfur in the alcohol. As it 
 stands now, the new method requires a separate extraction for 
 the acetone extract. 
 
 Caspari X1 gives a similar method to the one used by Kelly. 
 He uses petroleum spirit, boiling point 60-100C, which dissolves 
 3.0 gr. of sulfur per litre, whereas the 75% alcohol dissolves only 
 0.08 gr. Kelly says nothing about the accuracy when mineral 
 rubber, tars, or paraffins are present, whereas Caspari speaks of 
 the difficulty in getting these in solution. However, as we are 
 interested only in separating the true free sulfur from the sulfur- 
 carrying organic substances, it is quite within reason that undis- 
 solved paraffin would create no impassable difficulties. 
 
 >"J. Ind. Eng. Chem. 10, 518 (1918). 
 
 11 India Rubber Laboratory practice, p. 110. 
 
SULFUR DETERMINATIONS 91 
 
 Sulfur of Vulcanization. 
 
 It is often desirable to know the amount of sulfur actually 
 combined with the rubber during the process of vulcanization, 
 both as regards determining the extent to which it has proceeded 
 and to attain a greater uniformity in manufacturing practice. 
 The simplest method for estimating uniformity, for comparative 
 results, is by means of stress-strain curves, but mechanical de- 
 fects operate to change values, so that comparisons are at best 
 difficult and uncertain. The noticeable effect on the vulcanization 
 by slight changes in sulfur content, demonstrate that the amount 
 of sulfur which actually unites with the rubber is the controlling 
 feature of the vulcanization. The value for the sulfur of vulcani- 
 zation is necessary for the calculation of the total rubber hydro- 
 carbons in some of the direct methods, and a further use is the 
 possible discovery of the presence of reclaimed rubber in a rubber 
 compound. 12 
 
 Several possibilities are available, depending upon the nature 
 of the rubber compound. The simplest case is that of pure rub- 
 ber and sulfur, and occurs but seldom in commercial articles, 
 although it is overworked as a formula for determining the value 
 or properties of crude rubber. In this case, if the total sulfur is 
 S, the free sulfur Sf, the percentage of rubber 100' — ■ S then the 
 
 sulfur coefficient, Sv will be: — 
 
 ~ S — Sf 
 
 100 — s 
 
 In samples containing no organic sulfur compounds, the fol- 
 lowing method, based upon the determination of sublimed white 
 lead by Schaeffer, 13 gives excellent results: 
 
 The sample is extracted with acetone for eight hours, and the 
 free sulfur determined in the extract by the bromine method. 
 The residue is placed in a porcelain boat, and transferred to a 
 
 ,2 This is not as simple a proposition as it was before the rapid accelerators 
 came into use. With inorganic accelerators, the proper cure for rubber was 
 approximately at a coefficient of 3.0 to 3.5. and hence higher coefficients were 
 fair indications of the presence of reclaimed rubber, especially in connection 
 with other qualitative tests. Today, the value of the coefficient of vulcanization 
 is almost nil, when, by the use of appropriate accelerators, good cures can be 
 obtained with sulfur coefficients below 2.0. Of course, if one can learn what 
 accelerator has been used, and determine the coefficient for the best cures with 
 that accelerator, such data might be quite valuable in determining the condition 
 of the rubber in the sample under observation, 
 
 13 J. Ind. Eng. Chem. J,, 837 (1912). 
 
92 THE ANALYSIS OF RUBBER 
 
 hard glass tube. Carbon dioxide is passed through the tube, 
 which is then heated, gradually at first, and then at a dull red 
 heat for a few minutes. The organic matter, together with the 
 rubber, is distilled out, but the mineral sulfides and sulfates are 
 unchanged. The sulfur in the fillers is determined by transferring 
 the residue to a porcelain crucible, and determining the sulfur 
 therein by the Waters and Tuttle method for total sulfur. The 
 calculations for this method require a separate determination of 
 rubber, R. Calling the sulfur in the residue Sr, then the sulfur 
 coefficient will be calculated as follows: 
 
 « _ S — (Sr + Sr) 
 R 
 This is the same formula as before, when R = 100 — S, and 
 Sr = 0. 
 
 The most difficult case is when, in addition to sulfides and sul- 
 fates, we have organic substances containing sulfur, such as 
 oil substitutes, mineral rubber, etc. There are several procedures 
 which may be followed, but the safest is probably to use Wes- 
 son's nitrosite method as revised by Tuttle and Yurow. 14 In his 
 original article, Wesson says: "If the statement of Alexander 15 
 proves to be true that the sulfur of vulcanization of the rubber 
 remains quantitatively in the nitrosite, this method could pos- 
 sibly admit of the simultaneous determination of the sulfur of 
 vulcanization. An aliquot portion of the clear acetone solution 
 of the nitrosite would be evaporated to dryness, and the sulfur 
 determined in the usual way." A few attempts were made to 
 determine the sulfur of vulcanization in this way, but not until 
 after the errors which were contained in Wesson's method had 
 been eliminated, was it possible to secure accurate determination 
 of the rubber, and until then, little effort was made to determine 
 the sulfur of vulcanization. When the final revision was in shape, 
 determinations of the sulfur of vulcanization were found to check 
 very well. 
 
 The sulfur coefficient figured by this method, is the result ob- 
 tained by dividing the combined sulfur by the percentage of 
 rubber hydrocarbons; such a calculation leaves no opening for 
 
 14 As a matter of fact, this method can be used for any compound, and is not 
 confined in its application to this single case where organic sulfur compounds 
 are present ; it is equally effective in rubber sulfur mixtures, and with mixtures 
 containing mineral sulfur bearing fillers. 
 
 15 Z. Apgew, Chem, 20, 1364 (1907) ; 2h 687 (1911) ; Ber. 1,0, 1077 (1907). 
 
SULFUR DETERMINATIONS 93 
 
 questions as to whether or not the sulfur was combined with 
 the rubber, or with something other than rubber. It is simple, 
 direct, and accurate. 
 
 When possible to make it, the direct determination of the 
 sulfur coefficient (or for that matter any determination) is pref- 
 erable to the difference methods, since all questions regarding 
 interfering substances are eliminated. Kelly points out that 
 not only is the figure usually determined as free sulfur really a 
 mixture of elemental sulfur and sulfur combined with the resins 
 and other soluble constituents of the rubber, but that part of 
 the sulfur insoluble in acetone is soluble in alcoholic potash. 
 There seems to be no doubt that our use of the term free sulfur 
 is not exactly correct; and that some of the residual sulfur 
 should be removed by alcoholic potash seems equally reasonable, 
 but, for ordinary length cures, the amount so removed is small 
 (Kelly shows 0.07% for 2V 2 hours) . 
 
 If we figure our coefficient on only the sulfur that is insoluble 
 in alcoholic potash, obviously we should also take into our cal- 
 culations the non-rubber constituents, and this would include 
 the acetone soluble matter, or resins. In our formula, we would 
 therefore have to correct R for the acetone extract A, and the 
 alcoholic potash extract P, and we would have to deduct the 
 sulfur in the alcoholic potash extract, Sp; hence, we would have 
 the rather involved equation: 
 
 S— (Sf + Sr + Sp) 
 bV ~~ R— (A + P) 
 
 As a matter of fact, the relative amounts of rubber and non- 
 rubber substances insoluble in acetone are such that even making 
 this additional correction changes the coefficient very slightly, 
 certainly within the limits of experimental error, as far as our 
 experience goes. Hence, although no doubt the published data 
 for coefficients of vulcanization are not absolute values, they are 
 probably relatively accurate, and are comparable. Hence any 
 deductions which may have been made from these data are no 
 doubt just as valid as though every correction had been made. 
 
 Sulfur in Fillers. 
 
 The sulfur in fillers is determined as given under the method 
 for the determination of rubber by the ash method (cf. page 83). 
 
Chapter IX. 
 
 Detection of Organic Accelerators. 
 
 There is very little published work on this subject; probably 
 a few laboratories have some special tests of their own, but as 
 yet no one has seriously taken up this field. The data given 
 below is largely from Twiss and Martin, 1 and Earle L. Reed. 2 
 
 Paranitrosodimethylaniline. Extract about 10 gr. of the 
 sample with acetone, and dry the extract; add 5 cc. dilute hydro- 
 chloric acid, shake thoroughly, and filter. A pink or carmine 
 color results if p-nitrosodimethylaniline is present. If the filtrate 
 is colorless, divide it into two portions, using one to test for 
 aniline, and the other for hexamethylenetetramine. 
 
 The above test is a better negative than a positive test — if no 
 color develops, the accelerator is not present, but there may be 
 other organic bases which will give a pink color on acidification 
 with dilute hydrochloric acid. 
 
 Twiss and Martin call attention to the color of the acetone 
 extract which, however, is too common a color to use as an indica- 
 tion of an organic accelerator. A more positive test is to treat 
 the dried acetone extract, or a dilute hydrochloric acid extract 
 of a finely ground sample, with hydrogen sulfide water and ferric 
 chloride solution, forming a blue, or greenish-blue, if paranitro- 
 sodimethylaniline is present. The reaction depends upon the re- 
 duction of part of the accelerator during vulcanization, to 
 p-aminodimethylaniline, which, when treated as stated, forms 
 methyleneblue. 
 
 Twiss gives the following alternative method: treat the hydro- 
 chloric acid solution of the dried acetone extract with a small 
 piece of metallic zinc. Filter off the solution, cool thoroughly, 
 and add a well cooled dilute aqueous solution of sodium nitrate. 
 Add a small amount of this mixture to a solution of beta-napthol, 
 with excess of aqueous sodium hydroxide. A deep blue results 
 in the presence of p-nitrosodimethylaniline. 
 
 'Rubber Age, 9, 379-80 (1921). 
 2 Unpublished data. 
 
 94 
 
DETECTION OF ORGANIC ACCELERATORS 95 
 
 It can also be tested for by means of the Liebermann reaction. 
 The dried acetone extract is boiled with a small amount of dilute 
 caustic, and filtered; the filtrate is evaporated to dryness, cone, 
 sulfuric acid and phenol added, the mixture diluted with water, 
 and made alkaline with caustic potash, when a deep blue colora- 
 tion will appear. 
 
 Aniline. Using the hydrochloric acid filtrate after testing for 
 paranitrosodimethylaniline, add a drop of freshly prepared and 
 filtered solution of bleaching powder. A violet color indicates 
 the presence of aniline. Thiocarbanilide will ordinarily give no 
 reaction to this test, unless present in very large amounts. It is 
 well, in order to make sure of its absence, to take another portion 
 of the dried extract, and heat, and look for the characteristic 
 odor of thiocarbanilide. 
 
 Thiocarbanilide. A portion of the dried acetone extra ct is 
 placed in a test tube, stoppered, and connected by a delivery 
 tube with a second test tube containing two or three cc. of distilled 
 water. The delivery tube must dip below the surface of the 
 water. The first test tube is now heated until bubbles escape 
 through the water in the second test tube, after which the heat- 
 ing is continued strongly for two or three minutes. Test the 
 water in the second test tube for aniline with the filtered bleach- 
 ing powder solution; a violet color will indicate thiocarbanilide 
 if the original aniline test was negative. 
 
 Thiocarbanilide has a very characteristic odor, which is 
 especially noticeable when heated. Heat the dried acetone ex- 
 tract, and compare the odor with that of some heated thio in a 
 second test tube. 
 
 Hexamethylenetetr amine. Using the second portion of the 
 hydrochloric filtrate from the test for p-nitrosodimethylaniline, 
 add 5 cc. of water, 1 cc. of phosphoric acid, a small amount of 
 phenylhydrazine hydrochloride, 2 drops of 10% ferric chloride 
 solution, and 2 drops of cone, hydrochloric acid. A cherry red 
 color is produced by the formaldehyde from the hexamethylene- 
 tetramine. 
 
 Extract a ground sample with water, and test the extract for 
 ammonia with Nessler's solution. A positive test indicates alde- 
 hyde ammonia or hexa — although some of the less commonly 
 used accelerators may yield small amounts of ammonia, and 
 hence respond to this test. 
 
96 THE ANALYSIS OF RUBBER 
 
 Diphenylamine. To the dried acetone extract from about 10 
 gr. of finely ground sample, add 2 cc. of cone, sulfuric acid, and 
 agitate gently. Add a small crystal of sodium nitrate — a blue 
 coloration results if diphenylamine is present. 
 
 This test can be made directly on light-colored compounds by 
 placing a few drops of cone, sulfuric acid on the rubber to be 
 tested, dipping a glass rod in dilute nitric acid, and touching it 
 to the edge of the sulfuric acid. 
 
 Quinoidine. Treat the dried acetone extract with dilute sul- 
 furic acid; quinoidine gives a blue fluorescence. Rochelle salts 
 precipitate the tartrates of quinine or cinchonidine, but not 
 quinidine. A saturated solution of potassium iodide added to an 
 acid solution gives quinidine hydroiodide. Quinine and quinidine 
 give the thalleioquin test, but cinchonine and cinchonidine do 
 not; to a solution of the acetone extract in dilute sulfuric acid, 
 add very weak bromine water, drop by drop, until a faint yellow 
 persists, but avoid an excess of bromine; add ammonia, drop by 
 drop, when a brilliant green color results. Making this solution 
 acid turns the color to red. 
 
 General Tests. Extract 10 gr. of finely ground sample with 
 dilute hydrochloric acid, cool thoroughly, and diazotize with cold 
 dilute aqueous sodium nitrite. (The simplest scheme is to put a 
 small piece of ice in the solution during the diazotizing; it can 
 be removed later.) After a few minutes, pour a little of this 
 mixture into a solution of beta-napthol in excess of caustic soda; 
 a red precipitate or coloration indicates the presence of a primary 
 aromatic amine, such as aniline toluidine, p-phenylenediamine, 
 etc., or of derivatives of such bases with aldehydes (formaniline, 
 methyleneaniline, benzylidene-aniline) , and with carbon bisulfide 
 (thiocarbanilide, or triphenylguanidine). 
 
Chapter X. 
 
 Mineral Analysis. 
 
 The first step in a fillers determination of a rubber compound 
 is to make a qualitative analysis of the metals which it contains. 
 In this work, the color of the sample will be of considerable 
 assistance in cutting out unnecessary steps, as will also a knowl- 
 edge of the use to which the article is to be put. Only in the 
 black compounds is there any necessity for making a fairly com- 
 prehensive examination. 
 
 Preparation of the Solution. The possible presence of lead, 
 barium and calcium in a mixture containing sulfur (as sulfuric 
 acid) makes the problem of making up a solution for qualitative 
 analysis quite an interesting one. While several choices are open, 
 the following procedure is recommended because of the fact that 
 it permits quantitative separations to be made on a number 
 of elements: 
 
 Place exactly 2.500 gr. of finely divided rubber in a porcelain 
 casserole (about 250 cc. capacity), cover with 25 cc. of fuming 
 nitric acid, and after standing in the cold for 15 to 30 minutes, 
 covered with a watch glass, heat on a steam bath or hot plate 
 until the rubber and all other organic matter is entirely de- 
 stroyed. Potassium chlorate and fresh acid should be added from 
 time to time. Evaporate the solution to dryness, add hydro- 
 chloric acid and a little water, and again evaporate to dryness 
 and heat to dehydrate silica. Take up the residue with 50 cc 
 boiling water and 2 or 3 cc. of cone, hydrochloric acid. Filter 
 into another porcelain casserole, and repeat the evaporation and 
 dehydration of silica. Take up with 50 cc. of hot water, and 2 
 or 3 cc. of cone, hydrochloric acid as before, and filter. Unite 
 the two portions of insoluble matter, and reserve for further 
 treatment. 
 
 Heat the filtrate from the above, and add, drop by drop, 10 
 cc. of barium chloride solution until no further precipitate is 
 formed, and then a few drops in excess. Allow to stand over- 
 
 97 
 
98 THE ANALYSIS OF RUBBER 
 
 night, filter off the barium sulfate (which may be discarded), 
 wash well and transfer the filtrate to a 250 cc. graduated flask. 
 
 The insoluble portions reserved above are fused with sodium 
 carbonate in a nickel crucible, cooled, and the melt taken up 
 with distilled water. If lead, barium, or calcium sulfates were 
 in the insoluble residue, they will now appear as insoluble car- 
 bonates, while the silica, if any, will be in solution. Filter off 
 the insoluble matter, wash free from alkali, and then dissolve 
 the carbonates off the filter with dilute hydrochloric acid and hot 
 water. Filter through the same filter paper, and unite the filtrate 
 with the solution already in the graduated flask. 
 
 The filtrate from the separation of the carbonates contains 
 the silica; it should be evaporated to dryness, and the silica 
 dehydrated and determined in the usual way. The filter paper 
 from the filtration of the lead and barium should be ignited, and 
 examined for silicates which may not have been attacked during 
 the fusion. 
 
 The solutions united in the graduated flask are now made up 
 to the 250 cc. mark at room temperature; 50 cc. of this solution 
 contains the fillers from 0.500 gr. of rubber. 
 
 By this procedure, we have eliminated the sulfuric acid, which 
 would prove so troublesome with lead, barium, and calcium, but 
 in so doing, have introduced barium into the solution. This is 
 of no importance, for barium is usually determined on a separate 
 sample by a short but excellent method. 
 
 Another element is introduced through the fusion in a nickel 
 crucible, but nickel is not likely to be found in rubber compounds 
 so that we need merely eliminate it in its turn, and proceed with 
 our analysis. On account of lead, fusion in platinum is impos- 
 sible, while fusion in iron would introduce serious complications. 
 
 The object in making up a standard solution, is that 50 cc. may 
 be taken for qualitative analysis, and further aliquot portions 
 may be drawn for such quantitative tests as may be desired. 
 In fact, with so few metals to be determined, 1 it is frequently 
 possible to combine qualitative and quantitative separations at 
 the same time. 
 
 If the silica is less than 0.5%, we may assume that it came 
 
 1 Lead, iron, aluminium, zinc, calcium, and magnesium are practically the only 
 metals to be determined. Antimony and barium are determined in special tests ; 
 manganese will be encountered where iron oxides are present, but is not neces- 
 sarily determined. 
 
MINERAL ANALYSIS 99 
 
 from the talc used in dusting, and that the silica pigments, such 
 as tripoli, talc, asbestine, aluminum flake, etc., have not been used 
 as fillers. 
 
 The procedure for making the qualitative and quantitative 
 separations may be taken from the standard text books, and 
 need not be repeated here. A few words of caution may not 
 come amiss. 
 
 In only two cases has vermilion been found amongst many 
 hundreds of samples tested; it is too costly, and since it is used 
 only for its color, there should be little difficulty in detecting this 
 substance from the color of the compound. 
 
 Green-colored samples should be tested for arsenic, not that it 
 is likely to be found, but merely to be on the safe side. Arsenic 
 colors should never be used in rubber compounding, but it is well 
 to see that no one is taking a chance. 
 
 Copper, even in traces, should be carefully looked for, because 
 even in small amounts its deteriorating influence on rubber com- 
 pounds is remarkable. 
 
 Note whether or not there is any appreciable quantity of mag- 
 nesium; a small amount may be expected from the talc used in 
 dusting stocks in the mill room, but it should be only a matter 
 of 0.10% or so. More than that requires a quantitative deter- 
 mination, owing to the practice of using small amounts of mag- 
 nesium oxide to activate organic accelerators. 
 
 If the nitric acid solution of the rubber shows insoluble 
 material, and yet no silica is present, it indicates insoluble sul- 
 fates of lead or barium, or both. 
 
 Black specks remaining after the fuming nitric acid treat- 
 ment of the rubber, indicates gas black or lamp black, for which 
 a separate determination is made. 
 
 It will be seen from the description of the mineral fillers used 
 in rubber manufacture, that the following metals may be found: 
 antimony, lead, iron, aluminium, chromium, zinc, calcium, 
 barium, magnesium, sodium, and ammonium salts. The com- 
 pounds formed with these metals, consist of oxides, sulfides, 
 sulfites, sulfates, carbonates, and silicates. 
 
 Oxides. The oxides are usually determined by difference ; after 
 the determination of the acid radicles, the excess of bases over 
 that required to combine with the acids is assumed to be present 
 as oxide. 
 
100 THE ANALYSIS OF RUBBER 
 
 Sulfides. Stevens 2 determines the sulfide sulfur as follows: 
 The apparatus consists of a Kipp generator for carbon dioxide, 
 a 250 cc. flask with an inlet tube reaching nearly to the bottom 
 of the flask, and a ground-in stopper carrying an outlet tube 
 (an all-glass wash bottle can readily be adapted for the pur- 
 pose), and connected to the outlet tube are two absorption 
 bottles containing lead acetate solution. Place in the flask 10 
 cc. of cone, hydrochloric acid and 20 to 30 cc. of ether, pass a 
 current of carbon dioxide through the apparatus until all air is 
 removed, then remove the stopper and add the sample (0.1 to 1.0 
 gr., depending upon the amount of sulfide expected; where noth- 
 ing is known regarding the sample, use 1.0 gr.). Again pass car- 
 bon dioxide through the apparatus for about 30 minutes, with an 
 occasional shaking of the flask. During this period, the hydro- 
 chloric acid attacks the sulfides, liberating hydrogen sulfide, 
 which is carried over to, and absorbed by the lead acetate solu- 
 tion. The purpose of the ether is to swell the rubber, and facili- 
 tate the penetration of the acid to all parts of the sample. 
 
 Heat gently to drive off the ether and the final traces of 
 hydrogen sulfide. Reserve the solution in the flask for the deter- 
 mination of sulfate sulfur. All of the sulfide sulfur is now com- 
 bined with the lead. 
 
 Stevens determines the sulfur from this point by adding acetic 
 acid to the lead acetate solution in order to decompose the car- 
 bonates formed, the lead sulfide is filtered off, and washed free 
 from lead salts, transferred to a stoppered flask, a standard 
 iodine solution added, and after the reaction is complete the 
 excess of iodine is titrated with sodium thiosulfate. However, 
 any other accurate method will answer the purpose; the lead 
 sulfide may be dissolved in nitric acid, taken to fuming with 
 sulfuric acid, and the lead sulfate determined gravimetrically. 
 
 If pure nitrogen is available for sweeping out the apparatus, 
 it will be found to be much simpler to use sodium hydroxide for 
 absorbing the hydrogen sulfide; the solution can be oxidized 
 with bromine, and after acidification, the sulfate can be precipi- 
 tated with barium chloride; altogether, much simpler, and 
 probably more accurate than the lead acetate method. 
 
 Sulfide sulfur, excepting antimony sulfides, may also be deter- 
 mined by the ignition method of Schaeffer, transferring the resi- 
 
 2 Analyst, 1,0, 275-81 (1915). 
 
MINERAL ANALYSIS 101 
 
 due to a flask similar to the one recommended by Stevens, and 
 proceeding as directed by him for driving over the hydrogen 
 sulfide. This procedure is best for lead sulfide; antimony and 
 mercury sulfides sublime unchanged. 
 
 Sulfites. Sulfites and sulfates are transposed by heating with 
 sodium carbonate. Schaeffer gives the following method for 
 determining the sulfite-sulfur in sublimed blue lead: 
 
 Boil 1.5 gr. of the sample with 3 gr. of sodium carbonate; 
 allow to stand, filter, and wash thoroughly. To the filtrate, add 
 3 cc. of bromine water, heat gently to oxidize the sodium sulfite 
 to sulfate, and precipitate the sulfate with barium chloride. The 
 barium sulfate formed will contain both the sulfur present as 
 sulfite, and sulfate; deduct the amount of sulfur present as 
 sulfate from the total, and the remainder is calculated to lead 
 sulfite. (See determination of sulfates in the presence of sulfites, 
 under sulfate-sulfur.) 
 
 Sulfates. Stevens determines the sulfate-sulfur in the residue 
 from the determination of sulfides, as follows: Extract the resi- 
 due with hydrochloric acid until no further material can be dis- 
 solved; unite the filtrates, and determine the sulfur as usual. 
 It will be noted that by this means Stevens dissolves out only 
 the lead sulfate and calcium sulfate; barium sulfate will be only 
 slightly attacked. This method is therefore not applicable for 
 the determination of lithopone, for example, or in any other case 
 where barium sulfate is present along with some sulfide. 
 
 We again find Schaeffer's ignition process of value in deter- 
 mining the sulfates. Boil the ignited residue with sodium car- 
 bonate as directed under sulfite-sulfur, and filter. The function 
 of the bromine water in the sulfite determination is to oxidize 
 the S0 2 to S0 3 ; if instead of adding bromine water we add 
 hydrochloric acid, and boil the solution, the sulfur dioxide will 
 be driven off, and we will have remaining only the sulfate-sulfur. 
 
 Carbonates. Carbonates can be determined in an apparatus 
 similar to Stevens' arrangement for sulfide-sulfur. Instead of a 
 Kipp generator, we use air which has first been passed through 
 a soda-lime tower, to remove traces of carbon dioxide. In this 
 case, the absorption train consists of two absorption bottles con- 
 taining cone, sulfuric acid and potassium bichromate (a and b) ; 
 two soda-lime tubes (c and d) ; and the fifth tube containing sul- 
 furic acid and bichromate (e). It is vital in this determination 
 
102 THE ANALYSIS OF RUBBER 
 
 that tubes b and e should be frequently refilled, and from the 
 same solution ; only with such precautions are we able to main- 
 tain the air at the same moisture content when it leaves e as 
 when it entered c. Tubes c, d, and e, are weighed before and 
 after the determination; the increase in weight is the carbon 
 dioxide. Cases are known where d actually lost weight, owing 
 to the fact that c absorbed all of the C0 2 , and the air withdrew 
 from d some of its moisture, which, however, was reabsorbed by e. 
 
 Any similar arrangement will do just as well, providing 3 the 
 gas used to wash the apparatus contains no carbon dioxide, or 
 organic matter which might be oxidized by the sulfuric acid- 
 bichromate mixture; the absorption tubes are adequate for the 
 purpose; and the balance of the moisture content of the gas is 
 preserved. 
 
 Silicates. These have already been separated by the method 
 of getting the metals of the fillers into solution. It is only neces- 
 sary here to repeat that all of the silica is not obtained by the 
 first dehydration and treatment with hydrochloric acid, no matter 
 how long the process be continued; the operation must be re- 
 peated or the error will show up in the determination of the other 
 constituents. 
 
 Special Determinations. 
 
 The qualitative and quantitative analyses made as prescribed 
 in the preceding paragraphs will suffice for the determination of 
 most of the metallic bases, or fillers, but some of these are better 
 determined by special tests; amongst the mineral fillers we find 
 in this list the antimony compounds, lead chromate, barium car- 
 bonate, etc., and amongst the organic, carbon black, blue, etc. 
 
 Antimony. The principal trouble with antimony is getting it 
 into solution without loss. There should be little difficulty once 
 this has been accomplished. Rothe 4 treats the sample with 
 10-20 cc. cone, nitric acid and 2 cc. sulfuric, and heats for 1 
 to 2 hours at a moderate heat; then increase the heat until 
 all nitric acid is driven off and the sulfuric acid fumes strongly. 
 More nitric acid is added, and taken to fuming, and this opera- 
 tion is repeated until the absence of darkening shows that the 
 
 •For a more complete discussion on this point, see Tuttle and Yurow. "The 
 Direct Determination of Rubber by the Nitrosite Method," U. S. Bureau of 
 Standards Tech. Paper, No. 145 (1919). 
 
 «Chem. Ztg. S3, 679 (1909). 
 
MINERAL ANALYSIS 103 
 
 organic matter is destroyed. Dilute to 100 cc. and boil to expel 
 all nitric fumes. Schmitz 5 takes from 2 to 4 gr. of finely cut 
 rubber (Frank and Marckwald think the quantity is too high, as 
 it no doubt is for most antimony compounds) , and treats it in a 
 Kjeldahl flask with 15 cc. cone, sulfuric acid per gram of rubber. 
 One drop of mercury and a small piece of paraffin (to prevent 
 foaming) are introduced. Heat until the solution starts to clear; 
 add 2-4 gr. of potassium sulfate, and heat until colorless. 
 Cool, dilute with water, add 1 to 2 gr. of potassium bisulfite, 
 with excess of tartaric acid ; heat until no sulfur dioxide remains, 
 add dilute hydrochloric acid, filter, and titrate the antimony. 
 Wagner 6 fuses in a porcelain crucible, 0.5 to 1.0 gr. of rubber 
 with 5 gr. of 1-4 sodium nitrate-potassium carbonate. The 
 rubber is mixed with part of the fusion mixture, placed in the 
 crucible, and covered with the remainder. The heat must be 
 applied gradually, and if any organic matter remains, more 
 sodium nitrate must be added, and the whole reheated. Wagner 
 claims good results, but the method looks risky; the danger of 
 loss of antimony by excessive heating is very great. When zinc 
 oxide or sulfide are present, Frank and Marckwald 7 separate 
 the rubber from the fillers with xylol ; otherwise, they oxidize the 
 organic matter with cone, nitric acid and potassium chlorate, 
 finally evaporating with hydrochloric acid. If organic matter is 
 still present, it must be eliminated. The antimony is precipi- 
 tated as sulfide, and weighed as such. Collier, Levin and 
 Scherrer 8 take advantage of the simultaneous determination of 
 the fillers by the cymene method to determine the antimony after 
 the rubber has been dissolved out. Their method is as follows: 
 
 Extract 0.500 gr. of the sample with acetone for 8 hours, 
 and with chloroform for 4 hours. Dry the residue in a vacuum 
 desiccator, transfer to a 300 cc. lipped assay flask, add 25 cc. of 
 cymene, and heat on an electric hot plate at 130-140C until the 
 rubber is dissolved. Cool the flask, dilute with 250 cc. of petro- 
 leum ether, and allow to stand overnight. Filter by decantation 
 through a tight Gooch pad of asbestos, previously washed with 
 alkali, cone, hydrochloric acid, and water, and dried. Wash by 
 decantation with petroleum ether until the filtrate is colorless. 
 
 'Gummi Ztg. 25, 1928-30 (1911). 
 
 •Chem. Ztg. SO, 638 (1906) ; J. Soc. Chem. Ind. 25, 583 (1906). 
 
 T Gummi Ztg. 23, 1046 (1909). 
 
 •Rubber Age, 8, 104-5 (1920). 
 
104 THE ANALYSIS OF RUBBER 
 
 Add 30 cc, of cone, hydrochloric acid to the assay flask, and 
 shake until all of the antimony sulfide has gone into solution; 
 filter slowly through the Gooch, using gentle suction. Wash 
 thoroughly, and dilute the filtrate to 250 cc. with hot distilled 
 water, pass in hydrogen sulfide until the antimony has been com- 
 pletely precipitated. 
 
 After the solution of the antimony has been effected, it may 
 be determined by any of the well known methods. Wagner, and 
 Frank and Marckwald weigh as sulfide, Schmitz recommends 
 titration, as do Collier, Levin and Scherrer. The methods recom- 
 mended by the last named are as follows: 
 
 Filter off the antimony sulfide, wash with hydrogen sulfide 
 water, and transfer the precipitate to the filter paper. Place 20 
 cc. of concentrated hydrochloric acid in the beaker, and set aside 
 temporarily. Transfer the antimony sulfide and the filter paper 
 to a Kjeldahl flask, add 12-15 cc. of concentrated sulfuric acid 
 and 5 gr. of potassium sulfate, place a funnel in the neck of the 
 flask, and heat until the solution is colorless. Wash the funnel, 
 and dilute the solution to about 100 cc. with water, add 1-2 gr. 
 of sodium sulfite, transfer the hydrochloric acid in the beaker 
 in which the antimony sulfide was precipitated to the Kjeldahl 
 flask, and boil until the sulfur dioxide is all driven out. Dilute 
 to 250-275 cc. with water, cool to 10-15C, and titrate with per- 
 manganate until a faint pink color is obtained. 
 
 Instead of filtering the antimony on filter paper, it may be 
 filtered on a Witt plate and asbestos. Transfer the plate, pad 
 and precipitate to an Erlenmeyer flask; remove any antimony 
 sulfide adhering to the beaker or funnel with hydrochloric acid. 
 Wash the beaker and funnel with hot distilled water, dilute the 
 solution to 250-275 cc, add 12 cc. of concentrated sulfuric acid, 
 boil the solution until no trace of hydrogen sulfide is obtained 
 with lead acetate paper, cool to 10-15C, and titrate with stand- 
 ard permanganate solution. 
 
 Barium Salts. Ignite a 1 gr. sample in a porcelain crucible, 
 cool, add 3 to 5 drops of nitric acid and 1 cc. of water, and stir 
 into a paste, add 5 gr. of 1-1 potassium nitrate-sodium carbonate, 
 dry on the hot plate or steam bath, fuse until the melt is soft 
 or pasty; allow it to cool, extract with hot water, and wash with 
 hot water containing a little sodium carbonate. Dissolve the in- 
 soluble carbonates in hydrochloric acid, and wash the filter paper 
 
MINERAL ANALYSIS 105 
 
 thoroughly. Nearly neutralize the nitrate with sodium carbon- 
 ate, and pass hydrogen sulfide through the solution until the lead 
 is entirely precipitated. Filter, heat the filtrate to boiling, and 
 add 10 cc. of 10% sulfuric acid; allow the precipitate to stand 
 overnight, and determine the barium sulfate as usual. 
 
 The only troublesome element is lead, and it may be com- 
 pletely eliminated. Check determinations of 0.10% of the 
 barium sulfate present may easily be obtained. 
 
 In some specifications, a maximum limit is placed on the total 
 sulfur, but barytes is a permissible filler, without having the 
 sulfur which it contains count as part of the total sulfur. In 
 such cases, the determination of barytes is obligatory; if made by 
 this method, the error in the total sulfur caused by the correc- 
 tion will not exceed 0.02%. 
 
 Barium Carbonate. 9 Place 1 gr. of the sample in a porcelain 
 boat, and ignite in an atmosphere of carbon dioxide as described 
 by Schaeffer. 10 After ignition, and when the ash is at room tem- 
 perature, remove the boat, grind the ash to a fine powder in an 
 agate mortar, transfer to a 250 cc. beaker, cover with 5-10 gr. of 
 ammonium carbonate, 15-20 cc. of strong ammonia, and 50 cc. of 
 distilled water. Ammonium carbonate transposes lead sulfate 
 into lead carbonate, but is practically without action on barium 
 sulfate. Boil the mixture for 15 to 30 minutes, filter, and wash 
 the precipitate thoroughly to remove all soluble sulfates. Wash 
 the residue on the filter paper back into the original beaker with 
 distilled water, add 10 cc. glacial acetic acid, and sufficient water 
 to make the volume up to 100 cc. Heat to boiling, and filter 
 through the same filter paper as before. Lead, barium calcium 
 and zinc carbonates pass into solution, whereas lead sulfide and 
 barium sulfate are not attacked. Pass hydrogen sulfide into the 
 filtrate, filter off the lead sulfide, heat the filtrate to boiling, and 
 
 9 The reason for working out a method for determining barium carbonate is 
 not without interest. In material made under specifications, some manufac- 
 turers evidently desired to use compounds which contained more than the pre- 
 scribed amount of sulfur. Realizing that the specifications exempted the sulfur 
 in the barytes from counting in the total sulfur, and knowing that the barium 
 sulfate was being estimated from the amount of barium found by analysis, they 
 felt that by adding barium carbonate, they would receive credit for sulfur 
 equal to the barium in the carbonate, and thus bring the total within the 
 specification limit. The trick was first discovered when, after correction for the 
 sulfur supposed to be present in combination with the barium, the total sulfur 
 was actually less than the free sulfur. 
 
 10 Cf. page 91. 
 
106 THE ANALYSIS OF RUBBER 
 
 precipitate the barium with 10 cc. of 10% sulfuric acid. Allow 
 to stand overnight, and determine the barium sulfate as usual. 
 
 If barium sulfate and no carbonate is present, a small amount 
 of precipitate will be found, showing a slight solubility of the 
 barium sulfate, or else reduction of the sulfate to sulfide. The 
 amount will usually be less than 1% of the amount of barium 
 sulfate present. In a mixture of the two, the carbonate will run 
 somewhat high, for the same reasons, but with proper attention 
 to details the results will be quite sufficient for every purpose. 
 
 Gas Black or Lamp Black. Chemical analysis alone will tell 
 nothing as to whether gas black or lamp black has been used. 
 Even the microscope is, as yet, of little value in distinguishing 
 between the two, and the only thing remaining for us to do is to 
 determine the total carbon, and assume from the physical prop- 
 erties of the article, whether or not the black is gas black or 
 lamp black. 
 
 The free carbon is determined as follows: X1 
 
 Extract 0.5 gr. of rubber for 8 hours with a mixture by volume 
 of 68% chloroform and 32% acetone. Transfer the sample to a 
 250 cc. beaker, and heat until it no longer smells of chloroform. 
 Add a few cc. of hot cone, nitric acid, and allow to stand in the 
 cold for about 10 minutes. Add 50 cc. more of hot cone, nitric 
 acid, taking care to wash down the sides of the beaker; heat on 
 the steam bath for at least an hour. While hot, decant the liquid 
 through a Gooch containing a thick pad of asbestos, taking care 
 to keep the insoluble material completely in the beaker. Wash 
 with hot nitric acid, and suck dry. Empty the filter flask. Wash 
 the insoluble residue with acetone, and then with a mixture of 
 equal parts of acetone and chloroform, until the filtrate is color- 
 less. The insoluble matter, which has been carefully retained in 
 the beaker, is digested on the steam bath for 30 minutes with 35 
 cc. of a 25% solution of sodium hydroxide. Dilute to 60 cc. with 
 hot water, filter the solution, and wash with a hot 15% solution 
 of sodium hydroxide. Test for the presence of lead by running 
 some warm ammonium acetate solution containing an excess of 
 the hydroxide through the pad into sodium chromate ; if a yellow 
 precipitate is obtained, the pad must be washed until the wash- 
 ings no longer give a precipitate with the sodium chromate 
 
 11 Smith and Epstein, U. S. Bureau of Standards Tech. Paper, No. 136 ; J. Ind. 
 Eng. Chem. 11, 33-6 (1919). 
 
MINERAL ANALYSIS 107 
 
 solution. Next wash the residue a few times with hot cone. 
 hydrochloric acid, and finally with warm 5% hydrochloric acid. 
 Remove the crucible from the funnel, taking care that the outside 
 is perfectly clean, and dry in an air bath at 150C to constant 
 weight. Burn off the carbon at a dull red heat, cool and reweigh; 
 the difference in weight is approximately 105% of the carbon 
 originally present in the form of lampblack or gas black. 
 
 Several points must be carefully watched during this pro- 
 cedure: the acetone and hot nitric acid must not be brought to- 
 gether, since they react with considerable violence. Again, care 
 must be used in the alkali washing to avoid carrying through 
 the filter some of the gas black; the pad must be unusually thick 
 and free from channels. This is one of the principal reasons for 
 keeping the fillers in the beaker until the last moment. 
 
 The published data of Smith and Epstein show that the loss in 
 weight on ignition is about 5% higher than the carbon a ctually 
 present; hence the factor 105. The 5% is probably organic 
 matter not removed by the preliminary steps of the method. 
 Mineral rubber has no effect on the determination. Calcium sul- 
 fate, if retained with the fillers, would be reduced during the 
 ignition of the carbon, and would give high results for the latter. 
 Quite apart from the point raised by Smith and Epstein that 
 calcium sulfate is rarely found in rubber compounds, usually only 
 when associated with antimony, the treatment with strong acids, 
 and boiling, would suffice to dissolve out a considerable quantity 
 of calcium sulfate, which is quite soluble in hot nitric or hydro- 
 chloric acid solutions. 
 
 Red Lead. The peroxide of lead contained in red lead is not a 
 particularly desirable constituent for rubber compounds, and 
 some specifications, notably those for 30 or 40% Para insulation, 
 forbid its use. The Joint Rubber Insulation Committee 1Z gives 
 the following test for red lead: Dissolve a 1 gr. sample, pre- 
 viously extracted with acetone, in xylol; when the rubber has 
 been completely dissolved, filter through a Gooch crucible, wash- 
 ing thoroughly with benzol, alcohol and acetone. Transfer the 
 Gooch pad to a distilling flask, add hydrochloric acid, and distil 
 over the chlorine into a potassium iodide-starch solution. If 
 more than 0.1 cc. of N/10 sodium thiosulfate is required to titrate 
 the iodine liberated, red lead may be assumed to be present. 
 
 "J. Ind. Eng. Chem. 6, 75-82 1914). 
 
108 THE ANALYSIS OF RUBBER 
 
 This method was suggested for insulation compounds, and, as 
 far as it has been tested, has given satisfactory results. The 
 method depends upon the liberation of chlorine by the action of 
 the peroxide on the hydrochloric acid. Some off-color litharge 
 samples have given positive tests under this method; which is 
 what we might expect, since these lots contain a greater amount 
 of peroxide than they should, and yet not enough to be classed 
 as red lead. They are really mixtures of red lead and litharge, 
 and should be so treated. 
 
 Chromates, such as chrome yellow, will give this reaction, but 
 they should cause no confusion, since the color of the sample will 
 usually tell whether chromates are present. It would be unusual 
 indeed to have both chromates and lead peroxide present in the 
 same sample. 
 
 Chromates. While chromium is not a frequent constituent of 
 rubber goods, it is a possibility, and should be determined. There 
 is considerable analogy between the analyses of the pigments in 
 printing inks, and those in rubber compounds, and the following 
 method, originally used in the analysis of printing inks, should 
 be equally available for rubber compounds. 
 
 Fuse 0.500 gr. of rubber with equal parts of sodium peroxide 
 and potassium carbonate, using a nickel crucible. The heating 
 must proceed cautiously until the organic matter is destroyed, 
 after which the melt can be heated strongly for 10 or 15 minutes. 
 Cool, extract with water, and filter. The chromium is in the fil- 
 trate as chromate. Pass carbon dioxide through the filtrate, and 
 heat on the steam bath, in order to precipitate any lead which 
 may be held up by the caustic alkali; filter if necessary. Cool, 
 acidify strongly with hydrochloric acid, add potassium iodide 
 and starch solution, and titrate with standard N/10 sodium 
 thiosulfate to a colorless solution. The solution may be stand- 
 ardized against potassium bichromate, and the chromium calcu- 
 lated to CrO s , in which condition it no doubt exists in the com- 
 pound. 
 
 This method has been found simple and accurate in the pres- 
 ence of lead, manganese, clay, and other fillers likely to be 
 present in printing ink*, and should be fully as satisfactory for 
 rubber goods. 
 
 Glue. Make a qualitative test as follows: Digest 1 gr. in 
 cresol, or xylol (any solvent for rubber which does not attack 
 
MINERAL ANALYSIS 109 
 
 glue will do just as well) until the rubber is decomposed. Dilute 
 with petroleum ether, and filter through filter paper. Wash the 
 residue with alcohol, and after allowing the alcohol to evaporate 
 wash the residue back into a beaker, cover with water, and boil. 
 Filter off the insoluble, and test the filtrate for glue with a solu- 
 tion of tannic acid. Traces of glue will give only a milky cloudi- 
 ness, but with large quantities a heavy precipitate is thrown 
 down. 13 
 
 The quantitative determination of glue is based on the deter- 
 mination of nitrogen by the Kjeldahl method. This procedure 
 assumes that the principal source of nitrogen, the organic accel- 
 erators, will be removed during the acetone extract. The U. S. 
 Bureau of Standards extracts with the mixed solvents, 68% by 
 volume of chloroform, and 32% of acetone. From this point on, 
 the procedures are alike: the dried sample is heated with sulfuric 
 acid, potassium or sodium sulfate and a small amount of copper 
 sulfate, the clear solution is diluted, made alkaline, and the 
 ammonia distilled into a standard solution of N/10 sulfuric acid. 
 
 Practically every one is agreed that the Kjeldahl method is 
 the most satisfactory means of approach in the quantitative de- 
 termination of glue, but differences arise as to the factor by which 
 to calculate from nitrogen to glue. The Bureau of Standards 
 uses 5.56; others prefer the factor of 6.25. Since glue is not a 
 pure chemical substance, we are bound to have differences of 
 opinion, but the weight of evidence seems to lean towards the 
 higher factor. The collagens have 17.9% of nitrogen, and even 
 assuming that in glue we have reasonably pure collagens, we must 
 take into consideration the water which is always present, and 
 which will average about 10%, hence, in the collagens, 5.56 
 would be the correct figure, and calculating that this is only 90% 
 of the whole, we get 6.18 as the corrected figure. 
 
 Another variable will be the amount of nitrogen in the in- 
 soluble matter in the rubber, which, as we have already discussed, 
 may run from 2 to 6% of the rubber, and may contain from 10 
 to 18% of nitrogen. 
 
 In view of the above facts, it is obvious that any factor will 
 
 ]3 A much shorter method of testing qualitatively for glue is as follows: 
 Heat 5 to 10 gr. finely divided sample of rubber with 25 cc. of water for 2 to 4 
 hours; decant, and test for glue with 2 or 3 ce. of a solution of tannic acid. 
 This test is not as safe as the one given above ; glue to the extent of 2 or 3% 
 may be easily overlooked, and hence the method is not recommended. 
 
110 THE ANALYSIS OF RUBBER 
 
 at best give only an approximation of the truth, but, even so, it 
 is believed that better results on the average will come from 
 the use of the factor 6.25, and which we recommend. 
 
 Ground Organic Wastes. In a mixture of rubber and wastes, 
 containing such materials as leather, cork, wool, silk, cotton or 
 other vegetable fibre, etc., the separate determination of these 
 wastes is usually of no consequence, and a direct determination 
 of the rubber by the nitrosite method will give practically all the 
 information that one really needs. Some of the solvents, such as 
 xylol, cymene, and possibly others, will determine the rubber 
 accurately enough in the presence of such materials. 
 
 There are occasions when we may be called upon to determine 
 cotton, as for example, in balloon fabrics, where it is difficult to 
 separate the rubber from the cotton, etc. For this purpose, the 
 method of Epstein and Moore 14 will suffice: 
 
 Treat a 0.500 gr. sample of the rubber with 25 cc. of freshly 
 distilled cresol (b.p.l98C) for 4 hours at 165C. Cool, add 
 200 cc. of petroleum ether very slowly, and with constant agita- 
 tion. Filter through a Gooch, and wash with petroleum ether, 
 then with hot benzene, and finally with acetone. Add hot 10% 
 hydrochloric acid, and transfer the contents of the flask to the 
 Gooch, and wash at least ten times with hot acid. Wash free of 
 chlorides, and then with acetone until the filtrate is colorless. 
 Wash with a mixture of equal parts of acetone and carbon 
 bisulfide. Wash with alcohol, and dry for V/% hours at 105C. 
 Transfer the asbestos pad and fillers to a weighing bottle, dry 
 for about 10 minutes further, cool and weigh. 
 
 Transfer the contents of the weighing bottle to a 50 cc. beaker 
 and pour over it 15 cc. of acetic anhydride and 1 to 2 cc. of cone, 
 sulfuric acid, and digest on the steam bath for one hour. Cool, 
 dilute with 25 cc. 90% acetic acid, and filter through a weighed 
 Gooch. Wash with hot 90%, acetic acid until the filtrate is color- 
 less, and then four times more. Wash about 5 times with acetone, 
 remove the crucible from the funnel, and dry to constant weight 
 at 150C. The cellulose has been dissolved out, and the usual 
 calculations are made. 
 
 Sponge Rubber. One of the interesting points in connection 
 with the analysis of sponge rubber is to determine the substance 
 
 14 U. S. Bureau of Standards Tech. Paper, 154; The Rubber Age, 6, 289 93 
 (1920). 
 
MINERAL ANALYSIS 111 
 
 used to produce porosity. Organic liquids, if used, will have been 
 dissipated by the time the sample reaches the analyst. If either 
 ammonium carbonate or sodium carbonate has been used suf- 
 ficient material will usually remain to give a qualitative test, 
 although a quantitative determination is out of the question. 
 
 Grind the sample into small particles, being particularly care- 
 ful to avoid heating. Digest 10 gr. of the sample in 25 cc. of 
 water for one hour, and filter. Divide into two portions; into 
 the first, add 10 cc. of 20% caustic soda, and note any odor of 
 ammonia which may escape. A positive test indicates ammo- 
 nium carbonate. A more delicate test may be made by adding 
 a little hydrochloric acid, and evaporating to dryness, and treat- 
 ing the dried residue with a small amount of strong alkali. Evap- 
 orate the second portion of the extract to dryness, take up with 
 25 cc. of water, and add a few drops of methyl orange. Titrate 
 with N/10 hydrochloric acid; any appreciable quantity of alka- 
 line carbonate, in the absence of ammonia, will be a fair indica- 
 tion that sodium bicarbonate was used. In case ammonium car- 
 bonate was used, the residue from the second filtrate should be 
 heated strongly to remove the ammonia, and thus determine 
 whether both substances were used. 
 
 Negative tests for both ammonium carbonate and sodium bi- 
 carbonate may be taken to indicate that organic liquids have 
 been employed. 
 
 Specific Gravity. 
 
 Rubber Compounds. For ordinary rough work, where great 
 accuracy is not necessary, and when pieces of from 2 to 5 gr. are 
 available, Young's gravitometer is a rapid and convenient instru- 
 ment. When the bearings are clean, and the instrument in good 
 working order, the results are usually with 0.02, plus or minus 
 and are frequently only half that. 
 
 For greater accuracy, the pyenometer is the best thing to use. 
 Weigh out about 5 gr. in small strips, place them in the pyenom- 
 eter bottle, and fill with distilled water to the mark, being 
 careful that no bubbles adhere to the rubber, and then weigh. 
 Knowing the weight of the bottle filled with water, the weight 
 of water displaced by the rubber is easily calculated, and from 
 this, the specific gravity of the rubber. Ordinarily the specific 
 gravity is expressed to two decimal places, but even without 
 
112 THE ANALYSIS OF RUBBER 
 
 bringing the pycnometer to constant temperature 2 the calculations 
 may be made to the third decimal. 
 
 It has been found convenient, both in using Young's gravi- 
 tometer and the pycnometer, to wet the rubber with a soap solu- 
 tion, brushing it on with a camel's hair brush, and then rinsing 
 the rubber with distilled water. It eliminates the risk of air 
 bubbles, and does not affect the accuracy of the determination. 
 
 Pigments and Fillers. Pigments in lumps may be handled as 
 in the case of rubber compounds; the pycnometer is probably 
 better for the purpose. 
 
 Oils are determined with the Westphal balance, or, for quicker 
 and less accurate work, a hydrometer will do. 
 
 For powders, or small particles, the pycnometer is required. 
 The liquid chosen must be such as to have no effect on the pig- 
 ment being tested. For many of them, water will answer, but 
 where this is impossible, any other liquid will do just as well, 
 providing it does not react with, or dissolve the pigment. With 
 liquids other than water, the coefficient of expansion may be 
 such as to make it imperative to hold to a standard temperature 
 of say 25C, the specific gravity being referred to water at that 
 temperature. 
 
 Weigh out 5 gr. of the pigment, transfer to a pycnometer, and 
 fill the latter about two-thirds full. Boil the liquid for 10 to 15 
 minutes, and then place under a vacuum bell jar. When the air 
 has been entirely removed from the sample, cool to room tem- 
 perature (or to a standard temperature of 25C), fill up to the 
 mark, and weigh. When a liquid other than water is used, deter- 
 mine its specific gravity as referred to water at 25C, and use 
 this to calculate the gravity of the pigment. 
 
 Reclaimed Rubber. One of the important values connected 
 with reclaimed rubber is its gravity, and yet it is frequently so 
 porous that ordinary methods fail to secure accurate results. 
 In thin sheets, and with boiling water, fair results may be ob- 
 tained. W T hen a small mixing mill has been available, the 
 following scheme has been found eminently satisfactory: 
 
 Mix 450 grams of reclaimed rubber and 50 grams of sulfur, 
 until thoroughly and evenly mixed. The total weight of the 
 batch after mixing should be within 1 gr. of 500. Vulcanize a 
 small strip from the mix, and from this strip determine the 
 specific gravity of the mixture, Calculate the specific gravity of 
 
MINERAL ANALYSIS 113 
 
 the reclaimed rubber, taking the specific gravity of sulfur as 2.0. 
 If the specific gravity of the mixture is a, and the specific gravity 
 of the reclaim x, the calculation is as follows: 
 
 100a — 20.00 
 
 x = 
 
 90 
 
 For example, if the gravity of a mixture is 1.370, the calcula- 
 tion would be: 
 
 __ 100 X 1-370 — 20.00 117.00 
 X ~" 90. " 90.00 
 
 x = 1.30 
 
 A chart can be drawn, so that given the specific gravity of a 
 mixture that of the reclaimed can be read off directly. A differ- 
 ent mixture of reclaimed rubber and sulfur may be employed, 
 making the necessary alterations in the formula, the only requi- 
 site being that there should be sufficient sulfur for vulcanization. 
 
Chapter XI. 
 Microsectioning and Micro-photography. 
 
 Microphotographs of rubber goods have been known for a 
 number of years, Weber showing some excellent photographs of 
 hard and soft rubber goods in his book on India Rubber. Re- 
 cently, there has been considerable attention paid to the use 
 of the microscope in mineral analysis of small amounts of 
 materials, and in the examination of commercial materials, mix- 
 tures, etc. It has been realized that the chemical analysis does 
 not give the last word, and that frequently the difference in 
 the properties of two materials may be a matter of their physical 
 state, rather than their average chemical composition. In the 
 rubber industry, many laboratories have been working along 
 the lines of preparing sections of rubber compounds thin enough 
 to be examined under transmitted light, instead of reflected light, 
 as had been so largely the practice. The problem very quickly 
 narrowed itself down to a question of mechanical manipulation, 
 for even the crude sections first prepared showed that the pro- 
 cedure was feasible, and that information could be obtained not 
 only regarding composition, but even the properties of rubber 
 compounds, if the proper sections could be prepared. 
 
 The microsectioning has largely been done with the Spencer 
 microtome, which seems adequate for the purpose. The main 
 difficulty has been to so stiffen the rubber compound that it 
 would have no motion when being cut. Freezing was resorted 
 to, the earliest attempts employing the expansion of carbon 
 dioxide directly on the stage of the microtome, or surrounding 
 the specimen to be cut with solid carbon dioxide. Further 
 stiffening of the rubber was obtained by imbedding it in such 
 materials as starch paste, water-glycerine solutions, paraffin, etc. 
 The best results are obtained with material which does not 
 become brittle at the low temperatures employed. Even carbon 
 dioxide cooling was found to be insufficient for the purpose, and 
 the use of liquid air was resorted to, with eminently satisfactory 
 
 114 
 
MICROSECTIONING AND MICROPHOTOGRAPHY 115 
 
 results. Sections thinner than lj« are now being prepared, a great 
 deal of work has been started, and we are beginning to see the 
 fruits of this work. 
 
 Liquid air is probably not available for many laboratories, 
 but in such cases the use of carbon dioxide alone will be found 
 to give results well worth the effort, even though better could 
 be obtained with the cooling effected by the liquid air. 
 
 Perhaps one of the most interesting points brought out by this 
 new phase of rubber testing came to light at the meeting of 
 the American Chemical Society at Rochester, in April, 1921. 
 Schippel x had previously shown by experiment that compounded 
 and vulcanized rubbers showed an increase in volume on stretch- 
 ing, and his explanation was that vacu were formed around the 
 mineral particles, caused by the rubber being pulled away from 
 the surface of the pigment. Green 2 exhibited some microphoto- 
 graphs of sections of rubber under strain, wherein the vacu 
 caused by the rubber leaving the surface of the pigment were 
 clearly visible. Still more important was the evident fact that 
 only the larger or coarser particles showed this phenomenon. 
 The mechanism of tearing, rapid wear, etc., when coarse pig- 
 ments are used, was quite apparent. Green's work reflects credit 
 on the soundness of Schippel's reasoning. 
 
 The work of Breyer and his coworkers Ruby, Depew, and 
 Green, and of I. C. Diner, should shortly put us in a position 
 where we can take a piece of rubber and at least qualitatively 
 tell what pigments are present. It is too much to expect any- 
 thing in the quantitative line, especially when one considers the 
 extremely small area covered by these microphotographs, and 
 the difficulty of securing even mixing of a plastic such as rubber 
 with dry fillers. We know that we have variations in compo- 
 sition from one part of a batch to another; and this variation 
 must be very much greater when the sample under observation 
 weighs less* than a milligram. It is quite within the range of 
 probability that we shall, by careful sectioning, be able to tell 
 whether we are dealing with carbon black or lamp black; and 
 particularly identify such substances as Tripoli, aluminum flake, 
 talc, asbestine, etc., in mixtures of two or more, under which 
 
 'J. Ind. Eng. Chem. IS, 33-7 (1920). 
 
 1 Henry Green. "Volume increase of compounded rubber under strain," Rub- 
 ber Division, American Chemical Society, Rochester, April, 1921. 
 
116 THE ANALYSIS OF RUBBER 
 
 conditions the identification by chemical or mechanical means is 
 practically impossible. 
 
 The general scheme for the examination of microsections 3 deals 
 with (a) reflected light; (b) transmitted light ; (c) polarized light. 
 With reflected light, we use not only vertical, but oblique rays, 
 so as to get some idea of the surface, as well as the color of the 
 section. In transmitted light we have a new color classification, 
 wherein some fillers which are opaque and colored in reflected 
 light may be translucent and show a different color by trans- 
 mitted light. In polarized light, we have the differences in opti- 
 cal behavior between crystalline and non-crystalline substances; 
 interference figures, extinction angles, etc., to further classify the 
 materials under observation. Considering the comparatively 
 limited number of substances one finds in rubber compounds, as 
 compared with the entire mineral field, the possibility of exact 
 identification is very great. 
 
 As far as the identification of fillers is concerned, the future 
 seems bright, and today practically all the work is being con- 
 ducted along these lines. We have still to consider the possibility 
 of identifying different rubbers, or the rubber plastics, such as 
 the mineral rubbers, substitutes, etc., reclaimed rubber, soften- 
 ing oils and waxes, etc. For some of these substances, notably 
 mineral rubber, paraffin, rosin, oil substitutes, we have excellent 
 chemical means of identification, and more or less accurate means 
 for their quantitative determination. The problem of the iden- 
 tification of reclaimed rubber, and the different grades of new 
 rubber, is still open for solution, and it may be that this new 
 means of research will prove of valuable assistance in investiga- 
 tions of this sort. 
 
 8 Some excellent text books for this type of work are found in "Minerals in 
 Rock Sections," by Luquer, D. Van Nostrand Co., and "Characters of Crystals," 
 by A. J. Moses, D. Van Nostrand Co. The preparation and identification of 
 minerals in rock sections, measurement of crystal faces, extinction angles, lines 
 of cleavage, etc., will be excellent, and withal comparatively simple preparation 
 for the study of microsections of rubber. Fred E. Wright (see bibliography) 
 has done some excellent work in the field of the identification of minerals in 
 rocks, through the aid of the petrographic microscope, and any one attempting 
 work in the field of the microscopic examination of rubber compounds will find 
 a careful study of Wright's work to be of great help. 
 
Chapter XII. 
 Calculation to Approximate Formulas. 
 
 The greater number of analyses are made for the purposes of 
 checking factory production, and for comparing finished goods 
 sold under chemical specifications. In such cases, a complete 
 analysis is seldom desired; for factory purposes, a few deter- 
 minations suffice, and for specification purposes the analysis is 
 carried just far enough to decide whether or not the specifications 
 have been complied with. In the latter case, it is usually suf- 
 ficient to report the percentage of the rubber present, the pres- 
 ence or absence of reclaimed rubber, the free, total, and barium 
 sulfate-sulfur, the presence and approximate amounts of oils, 
 waxes, mineral rubbers, substitutes, and any other organic fillers 
 likely to have a bearing on the analysis. 
 
 There are times when one is interested in learning everything 
 concerning an article, and then, in addition to the foregoing, we 
 need a complete analysis of the mineral fillers, both as to the 
 basic and acidic radicles. From these data, we build up an 
 approximate formula. The report of the analysis should cover 
 the following points: 
 Rubber hydrocarbons 
 Acetone extract, sulfur free 
 
 Color and appearance of extract 
 Saponifiable matter 
 Unsaponifiable matter 
 Mineral hydrocarbons 
 Vegetable hydrocarbons 
 Chloroform extract 
 
 Color and appearance of extract 
 Alcoholic potash extract 
 
 Color and appearance of extract 
 Total sulfur 
 Free sulfur 
 Sulfur of Vulcanization 
 
 117 
 
118 THE ANALYSIS OF RUBBER 
 
 Glue 
 Carbon 
 
 Other organic fillers 
 Mineral Fillers 
 Bases 
 Aluminium as A1 2 3 
 Antimony as Sb 2 S 8 
 Barium as BaO 
 Calcium as CaO 
 Iron as Fe 2 8 
 Lead as PbO 
 Magnesium as MgO 
 Zinc as ZnO 
 Any other bases 
 Acids 
 
 Carbonate as C0 2 
 Silica as Si0 2 
 Sulfide-sulfur as S 
 Sulfite-sulfur as S0 2 
 Sulfate-sulfur as S0 3 
 Organic Accelerators 
 Specific Gravity 
 
 With these data before us, we may proceed with the recon- 
 struction of the compound. 
 
 Rubber. The rubber is the sum of the rubber hydrocarbons 
 (sulfur free), and the acetone, chloroform and alcoholic potash 
 extracts, providing that no organic matter, other than that 
 originally present in the rubber, is shown by the analyses. 
 Ordinarily, with new rubber, the acetone extract will not exceed 
 4%, the chloroform extract in a properly cured article 2%, and 
 the alcoholic potash extract 1%, based upon the rubber. If any 
 appreciable quantity in excess of these amounts is found, it must 
 be explained. 
 
 Sulfur. The sulfur added as such is the sum of the free sulfur 
 and the sulfur of vulcanization, plus any sulfur which may have 
 combined with the fillers during vulcanization. This latter item 
 is often difficult, and sometimes impossible to determine, but a 
 knowledge of the general procedure in designing rubber com- 
 pounds will be a help. 
 
 Organic Fillers. The oils, fats, waxes, etc., are determined 
 
CALCULATION TO APPROXIMATE FORMULAS 119 
 
 from tests on the acetone, chloroform and alcoholic potash ex- 
 tracts. Mineral rubber at best can be only approximated. 
 Special fillers, such as glue, cellulose, carbon, etc., are set down 
 just as they are determined. 
 
 Inorganic Fillers. With a knowledge of what bases and acids 
 are present, we may start to build up the composition of the 
 mineral fillers. 
 
 Antimony Compounds. If only antimony, sulfur and calcium 
 sulfate are found, in addition to the rubber, we know that we 
 have a mixture of golden sulfide and rubber, and not only is 
 the calculation simple, but also we have the formula of the 
 golden sulfide used. 
 
 Barium. All barium should be calculated to sulfate, unless 
 by analysis barium carbonate is shown to be present. 
 
 Calcium. In the absence of antimony, calcium may be calcu- 
 lated to the carbonate, unless the quantity present is less than 
 1%. In such cases, especially in the absence of reclaimed rubber, 
 it may be assumed, with some assurance, that this small amount 
 was added as hydrated lime. In the presence of whiting, the 
 hydrated lime cannot be detected. 
 
 Aluminium. Aluminium is probably present as a silicate. The 
 microscope will be found to be an absolute necessity to determine 
 which silicate is present. In the absence of magnesium a white 
 compound will probably contain aluminum flake, or white clay. 
 Some clays contain titanium, and a positive qualitative test for 
 titanium would be sufficient indication that the substance is 
 clay. Titanium oxide, associated with barium sulfate, is used 
 as a paint pigment, but only in an experimental way in rubber. 
 
 Iron. Iron is usually present as the oxide, but frequently is 
 associated with clay. It is sufficient for the purpose to report 
 the oxide and clay separately; then, in rebuilding the compound, 
 any clay in excess of that found in the iron oxide used must be 
 added as such. 
 
 Lead. Without question, lead is one of the most difficult sub- 
 stances to work upon. If organic accelerators are present, it is 
 probable that lead oleate, sublimed white or blue lead is present. 
 Probably as safe a thing to do as any is to work up all of the 
 other fillers first, and then apply any sulfide, sulfite, or sulfate- 
 sulfur to the lead. 
 
 Magnesia. Magnesia may be present in one of three forms, the 
 
120 THE ANALYSIS OF RUBBER 
 
 oxide, carbonate, or silicate. With silica present, and no 
 aluminium, a magnesium silicate is probable. In the absence of 
 whiting, any carbon dioxide found is probably combined with 
 magnesium, although lead carbonate (white lead) may interfere. 
 The specific gravity of the compound as a whole is one means 
 for distinguishing between the oxide and carbonate. 
 
 Zinc. Zinc is usually present as the oxide, and the simul- 
 taneous presence of barytes is not evidence that lithopone is 
 present. In the absence of lead and antimony, any sulfide-sulfur 
 will undoubtedly be combined with zinc. It is best to calculate 
 all zinc as the oxide, and not to assume that lithopone is present 
 unless there is an excess of sulfide-sulfur over that required for 
 lead or antimony. 
 
 After the approximate amount of the probable ingredients of 
 the compound have been worked out as above, the sum should be 
 in the neighborhood of 100% — if anything, should exceed that. 
 The next step is to take this formula and calculate the specific 
 gravity, which should check within 0.02 the specific gravity of 
 the original compound. Any greater discrepancy than this re- 
 veals some error, which must be checked up. Obviously, if our 
 calculations are low, the high gravity substances are in error, 
 and vice versa. If the gravities agree closely, then the figures 
 may be rounded off to even percentages, to the nearest 0.25%, 
 and brought by adjustment exactly to 100%. 
 
 It must be very clear to every one that the interpretation of 
 analytical results is a matter requiring experience, ingenuity, and 
 a great deal of common sense. The intent of the above is cer- 
 tainly not to lay down exact rules, but merely to indicate the 
 general line of thought, permitting the analyst, with his first-hand 
 information as to the progress of the analysis, to make such 
 deductions as may seem wise. 
 
BIBLIOGRAPHY 
 
 Allen and Johnston 
 
 1. The exact determination of 
 
 sulfur in soluble sulfates. J. 
 Am. Chem. Soc. 3?, 588-617 
 (1910). 
 Paul Alexander 
 
 2. Determination of sulfur in rub- 
 
 ber. Gummi Ztg. 18, 729; 
 Z. Angew. Chem. 17, 1799 
 (1904). 
 
 3. Weber's method for the direct 
 
 determination of rubber. 
 Gummi Ztg. IS, 789-91 
 (1904); J. Soc. Chem. Ind. 
 28, 765. 
 
 4. Rubber nitrosite, and its use 
 
 for the analysis of crude 
 rubber and rubber products. 
 Ber. 38, 181-4 (1905). 
 
 5. Nitrosites of India rubber and 
 
 their application to the 
 Analysis of rubber. Z. 
 Angew. Chem. 24, 680. 
 
 6. The desulfurization of vulcan- 
 
 ized rubber. Chem. Ztg. 36, 
 1289, 1340, 1358. 
 Anon. 
 
 7. Determination of sulfur in 
 
 golden antimony sulfide. 
 Chem. Ztg. 28, 595 (1904). 
 
 8. Propositions for a uniform 
 
 execution of tests in the 
 evaluation of rubber. Caout- 
 chouc & Guttapercha 8, 
 5011-23 (1911). 
 
 9. Sulfur chloride substitute and 
 
 hot vulcanization. Gummi 
 Ztg. 26, 1594. 
 
 10. The quantitative determina- 
 
 tion of golden sulfide of 
 antimony. Gummi Ztg. 29, 
 137-9 (1914). 
 
 11. Specifications and methods of 
 
 analysis for mixtures con- 
 taining 30% of Hevea Rub- 
 ber. Caoutchouc & Gutta- 
 percha 12, 8587-92 (1915). 
 
 12. A study of factice and its 
 
 analysis. Mat. grasses 8, 
 4339-42 (1915). 
 
 13. The use of jar rings containing 
 
 lead oxide. Gummi Ztg. SO, 
 521 (1916). 
 
 14. Poisons used in the rubber in- 
 
 dustry. Engineering 102, 
 16-7 (1916). 
 Anon. 
 
 15. A brief review of the organic 
 
 accelerators. India Rubber 
 World 55, 190-2 (1917). 
 
 16. The Bayer patent vulcaniza- 
 
 tion accelerators. India 
 Rubber J. 52, 853-5 
 (1916). 
 
 17. Tests on life of rubber insu- 
 
 lating compounds. Elec. 
 World 70, 384 (1917). 
 
 18. Electrolytic methods for de- 
 
 termining lead, zinc and 
 antimony in rubber com- 
 pounds. India Rubber 
 World 57, 150 (1917). 
 
 19. Water Extract of raw Rubber. 
 
 J. Soc. Chem. Ind. 36, 1105 
 (1917). Rubber substitutes. 
 Caoutchouc & Guttapercha 
 16, 9709 (1919). 
 
 20. The Peachey vulcanization 
 
 process. India Rubber World 
 63, 409-10 (1921). 
 
 21. Magnesium carbonate as a 
 
 compounding ingredient in 
 rubber. India Rubber World 
 65, 112-3 (1921). 
 H. Apitzsch 
 
 22. Determination of sulfur in 
 
 organic compounds. Z. 
 angew. Chem. 26, 503. 
 L. Archbutt 
 
 23. Preparation of rubber for 
 
 analysis. Analyst 38, 550-4. 
 Austerweil 
 
 24. Passage of hydrogen through 
 
 rubber walls of balloons. 
 C. R. 154, 196. 
 S. Axelrod 
 
 25. A method for the direct deter- 
 
 mination of rubber in vul- 
 canized rubber mixings. 
 Gummi Ztg. 21, 1229 (1907). 
 
 121 
 
122 
 
 THE ANALYSIS OF RUBBER 
 
 26. Direct determination of rub- 
 
 ber in soft cured rubber 
 goods. Chem. Ztg. 33, 895 
 (1909). 
 G. Ban 
 
 27. Rubberized balloon fabrics. 
 
 The Rubber Ind., 259 (1914;. 
 
 28. Proofing airship fabric. Rub- 
 
 ber Age 7, 245-6 (1920). 
 S. Bary 
 
 29. Estimation of free sulfur in 
 
 vulcanized rubber. Rev. 
 gen. chim. 16, 142-5. 
 Charles Baskerville 
 
 30. Note on the preparation of 
 
 rubber samples for analysis. 
 J. Am. Chem. Soc. 28, 1511 
 (1906). 
 C. Beadle and H. P. Stevens 
 
 31. Examination of rubber tires. 
 
 Chem. News 96, 2488. 
 
 32. Analyses of vulcanized rub- 
 
 ber goods. Analyst 85, 11- 
 16. 
 
 33. Some analyses of Hevea latex. 
 
 Analyst 86, 6-9. 
 
 34. Influence of mineral ingredi- 
 
 ents on properties of rubber. 
 J. Soc. Chem. Ind. 80, 1421. 
 
 35. The nitrogenous constituent 
 
 of Para rubber and its bear- 
 ing on the nature of syn- 
 thetic rubber. J. Soc. Chem. 
 Ind. 31, 1099-1103 (1912). 
 
 36. Determination of the insoluble 
 
 in raw rubber. Analyst 37, 
 13-6. 
 
 37. Testing of crude rubber. 
 
 Caoutchouc & Guttapercha 
 9, 6296-300 (1912). 
 
 38. New extraction apparatus. 
 
 Gummi Ztg. 27, 2087. 
 R. Becker 
 
 39. Determination of mineral rub- 
 
 ber and similar products in 
 rubber goods. Gummi Ztg. 
 25, 598; Chem. Ztg. 85, 
 288. 
 
 40. Discussion of Hubener's tetra- 
 
 bromide method. Gummi 
 Ztg. 25, 531; 677 (1911); 26, 
 1503 (1912). 
 C. W. Bedford and Winfield Scott 
 
 41. Reactions of accelerators dur- 
 
 ing vulcanization. J. Ind. 
 Eng. Chem. 12, 31-3 (1920). 
 C. W. Bedford and L. B. Sebrell 
 
 42. Reactions of accelerators dur- 
 
 ing Vulcanization. III. 
 
 Carbo-sulfhydryl accelera- 
 tors and the action of zinc 
 oxide. J. Ind. Eng. Chem. 
 18, 1034-8 (1921). 
 
 43. Reactions of accelerators dur- 
 
 ing vulcanization. IV. 
 Mechanism of the action of 
 zinc compounds. J. Ind. 
 Eng. Chem. 14, 25-31 (1922). 
 Bedin 
 
 44. Analysis of manufactured soft 
 
 and hard rubbers. Ann. 
 chim. anal. 23, 57-9 (1918). 
 R. W. Belfit 
 
 45. A method for the direct deter- 
 
 mination of rubber in a 
 compound. J. Ind. Eng. 
 Chem. 8, 326-7 (1916). 
 Gustav Bernstein 
 
 46. Contribution to the study of 
 
 the cold vulcanization of 
 rubber. Z. Chem. Ind. Kol- 
 loide 11, 185. 
 John M. Bierer 
 
 47. Crimson antimony. India 
 
 Rubber World 63, 17-8 
 (1920). 
 Jules Bock 
 
 48. Determination of crude rub- 
 
 ber. Rev. Gen. Chim. 14, 
 209-21 (1911). 
 J. Boes 
 
 49. The investigation of rubber 
 
 goods. Apoth. Ztg. 22, 1105. 
 C. R. Boggs 
 
 50. Direct determination of rub- 
 
 ber. 8th Int. Cong. App. 
 Chem. 9, 45-58. 
 
 51. Vulcanization of rubber by 
 
 selenium. J. Ind. Eng. 
 Chem. 10, 117-8 (1918). 
 L. M. Bourne 
 
 52. Resin and sulfur in India rub- 
 
 ber. Chem. Eng. 6, 195. 
 Jean Boutaric 
 
 53. Analysis of rubberized fabrics. 
 
 Caoutchouc & Guttapercha 
 17, 10202-6. 
 Britland and Potts 
 
 54. Use of pyridine in rubber 
 
 analysis. J. Soc. Chem. Ind. 
 29, 1142. 
 
 55. Ceresin wax in rubber mixings. 
 
 India Rubber J. 43, 333. 
 Bruggeman 
 
 56. Rapid determination of fillers 
 
 in rubber compounds. 
 Gummi Ztg. 25, 1529; J. Soc. 
 Chem. Ind. SO, 908. 
 
BIBLIOGRAPHY 
 
 123 
 
 G. Bruni 
 
 57. Solubility of crystalline sub- 
 
 stances in rubber. Giorn. 
 chim. ind. appl., Feb., 1921. 
 G. Bruni and C. Pelizzola 
 
 58. The presence of manganese in 
 
 raw rubber and the origin of 
 tackiness. India Rubber J. 
 62, 101-2 (1921). 
 
 59. The tackiness of raw rubber 
 
 and the aging of vulcanized 
 rubber goods. India Rubber 
 J. 63, 415-6 (1922). 
 G. Bruni and E. Romani 
 
 60. Mechanism of action of cer- 
 
 tain accelerators of vulcan- 
 ization. India Rubber J. 
 62,63-6 (1921). 
 T. Budde 
 
 61. The determination of true 
 
 rubber in rubber goods. 
 Chem. Zentr. II, 173 (1905). 
 
 62. Determination of rubber in 
 
 cold cured rubber. Gummi 
 Ztg. 21, 1205-8 (1907). 
 
 63. Determination of rubber as 
 
 tetrabromide. Gummi Ztg. 
 
 22, 333 (1908). 
 
 64. The determination of vulcan- 
 
 ized rubber. Apoth. Ztg. 23, 
 318. 
 
 65. The valuation of cold vulcan- 
 
 ized rubber by the tetrabro- 
 mide method. Apoth. Ztg. 
 24, 529. 
 
 66. New method for determining 
 
 combined sulfur in vulcan- 
 ized rubber. Gummi Ztg. 
 
 23, 1143; 24, 4-6 (1909); 25, 
 269-70. 
 
 67. Estimation of rubber as tetra- 
 
 bromide. Z. Angew. Chem. 
 
 24, 954 (1911). 
 E. Bunschoten 
 
 68. Vulcanization with sulfur ac- 
 
 cording to Ostromuislenskii. 
 Chem. Weekblad 15, 257-68 
 (1918). 
 Burton of Standards (U. S.) 
 
 69. The testing of rubber goods. 
 
 Circular 38. Fourth edition, 
 1922. 
 E. M. Camerman 
 
 70. Analysis of manufactured rub- 
 
 ber. Eng. News 56, 551. 
 W. A. Caspari 
 
 71. Bromination of vulcanized rub- 
 
 ber. Caoutchouc & Gutta- 
 percha 8, 5289 (1911). 
 
 Caspari and Porritt 
 
 72. The theory of vulcanization. 
 
 Caoutchouc & Guttapercha 
 17, 10277-83 (1920). 
 D. Cheneau 
 
 73. General scheme for the analy- 
 
 sis of India rubber goods. 
 Z. Nahr. Genussm. 3, 312-8 
 (1900). 
 S. Collier and M. Levin 
 
 74. The direct determination of 
 
 the sulfur of vulcanization. 
 Rochester meeting of the 
 American Chemical Society, 
 April, 1921. 
 
 75. Analysis of rubber goods con- 
 
 taining antimony sulfide. 
 Rochester meeting of the 
 American Chemical Society, 
 April, 1921. 
 S. Collier, M. Levin and J. A. 
 Scherrer 
 
 76. Determination of antimony in 
 
 rubber goods. Rubber Age 
 8, 104 (1921). 
 David S. Collins 
 
 77. Infusorial Tripoli. Rubber 
 
 Age 5, 101 (1919). 
 D. F. Cranor 
 
 78. Effect of organic accelerators 
 
 on the vulcanization coeffi- 
 cient. India Rubber J. 58, 
 1199-1205. 
 Benton Dales and W. W. Evans 
 
 79. The use of the microscope and 
 
 photomicrographs in the 
 study of inorganic materials 
 used in rubber. New York 
 meeting of the American 
 Chemical Society, Septem- 
 ber, 1921. 
 Frederic Dannerth 
 
 80. Coal tar products used in the 
 
 rubber industry. Color Trade 
 J. 6, 113-6 (1920). 
 
 81. Solvents and thinners used in 
 
 the rubber industry. India 
 Rubber World 64, 487-9 
 (1921). 
 
 82. Oils, fats, waxes and resins 
 
 used in the rubber industry. 
 India Rubber World 64, 
 563-6 (1921). 
 
 83. Carbons and hydrocarbons 
 
 used in the rubber industry. 
 India Rubber World 64, 
 742-4 (1921). 
 
 84. Pitch hydrocarbons used in the 
 
 rubber industry. India Rub- 
 
124 
 
 THE ANALYSIS OF RUBBER 
 
 ber World 64, 821-4 
 (1921). 
 
 85. The action of volatile organic 
 
 solvents and vulcanizing 
 agents on organic compound- 
 ing materials and resinous 
 gums. New York meeting 
 of the American Chemical 
 Society, September, 1921. 
 
 86. Defects in industrial rubber 
 
 goods. Met. Chem. Eng. 18, 
 531-4 (1918). 
 
 87. Rubber and Jelutong. Met. 
 
 Chem. Eng. 18, 296-8 (1918). 
 
 88. The rubber industry as a user 
 
 of dyes and coal tar products. 
 Am. Dyestuff Rep. 9, 14-7 
 (1921). 
 Frederic Dannerth and R. M. Gage 
 
 89. A method for the valuation of 
 
 washed and dried rubber. 
 India Rubber World 56, 
 583-4 (1917). 
 E. R. Darling 
 
 90. Determination of combined 
 
 sulfur in sulfur chlorides. 
 Chem. Analyst 27, 21 
 (1918). 
 E. L. Davies 
 
 91. The determination of sulfur 
 
 in rubber. Chem. Analyst 
 15, 4 (1915). 
 C. David and L. J. Foucar 
 
 92. A rapid volumetric method for 
 
 the estimation of free sulfur. 
 J. Soc. Chem. Ind. 81, 
 100. 
 J'. Dekker 
 
 93. Estimation of the content of 
 
 unsaponifiable resins in vari- 
 ous kinds of rubber. Gummi 
 Ztg. 81, 824 (1917); Z. 
 Angew. Chem. 81, 11; 46. 
 
 94. The determination of sulfur in 
 
 vulcanized rubber. India 
 Rubber J. 69, 413-8 (1920). 
 
 95. The determination of substi- 
 
 tute in rubber. India Rub- 
 ber World 62, 654-5 
 (1920). 
 11'. A. Del Mar 
 
 96. Report of the Joint Rub- 
 
 ber Insulation Committee; 
 Specifications and analysis 
 of 30% Hevea Rubber. J. 
 Ind. . Eng. Chem. 6, 75-82 
 (1914). 
 
 97. Report of the Joint Rubber 
 
 Insulation Committee. II. 
 
 Proc. Am. Inst. Elec. Eng., 
 April, 1917. 
 Harlan A. Depew and I. R. Rub)/ 
 
 98. Some microsections cut from 
 
 vulcanized rubber articles. 
 J. Ind. Eng. Chem. 12, 1156- 
 9 (1920). 
 Ernest Deussen 
 
 99. Quantitative determination of 
 
 sulfur in rubber. Z. Angew. 
 Chem. 24, 494. 
 Dieterich 
 
 100. On the determination of rub- 
 
 ber. Chem. Ztg. 28, 974 
 (1904). 
 Irene C. Diner 
 
 101. Microscopy of rubber fillers. 
 
 New York meeting of the 
 American Chemical Societv, 
 September, 1921. 
 Rudolf Ditmar 
 
 102. New methods of analysis of 
 
 raw rubber. Gummi Ztg. 20, 
 364-6 (1906). 
 
 103. Relation between specific grav- 
 
 ity and the sulfur content of 
 vulcanized Para rubber. 
 Gummi Ztg. 20, 733 (1906). 
 
 104. Effect of heavy magnesia as a 
 
 filler upon India Rubber. 
 Gummi Ztsr. 20, 760 (1906). 
 
 105. Influence of light magnesia as 
 
 a filler upon India Rubber. 
 Gummi Ztg. 20, 816 
 (1906). 
 
 106. Vulcanization of India rubber 
 
 in the presence of litharge. 
 Gummi Ztg. 20, 1077-8 
 (1906). 
 
 107. Influence of pressure on rate 
 
 of vulcanization, strength, 
 and oxidation of different 
 kinds of rubber. Chem. Ztg. 
 81, 638-9 (1907). 
 
 108. The influence of zinc oxide on 
 
 the vulcanization of, and 
 oxidation of rubber. Gummi 
 Ztg. 21, 5 (1907). 
 
 109. The melting points of some 
 
 rubbers. Gummi Ztg. 21, 
 670 (1907). 
 
 110. The determination of sulfur 
 
 in vulcanized India rubber 
 and rubber substitutes by 
 Dennstedt's method. Gummi 
 Ztg. 21, 497 (1907). 
 
 111. The dyeing of rubber with or- 
 
 ganic dyes. Chem. Ztg. 37, 
 1162. 
 
BIBLIOGRAPHY 
 
 125 
 
 112. Vulcanization catalysts. Gum- 
 
 mi Ztg. 29, 424-6 (1915). 
 Rudolf Ditmar and O. Dinglinger 
 
 vulcanization and oxidation 
 of rubber. Gurnmi Ztg. 
 
 113. Effect of powdered glass on 
 
 2./, 234-5 (1907). 
 Rudolf Ditmar and Thieben 
 
 114. Changes occurring in the most 
 
 important inorganic fillers 
 during steam vulcanization. 
 Roll. Z. 11, 77-80. 
 E. D. Donaldson 
 
 115. Rapid electrolytic method for 
 
 total lead and zinc in rubber 
 compounds. Chem. Analvst 
 15, 11-12 (1915) ; India Rub- 
 ber J. 57, 1100 (1919). 
 Andre Dubosc 
 
 116. Action of ozone and oxozone 
 
 in the analysis of rubber. 
 Caoutchouc & Guttapercha 
 10, 7105 (1913). 
 
 117. Analysis of vulcanized rubber. 
 
 Caoutchouc & Guttapercha 
 13, 8782-3 (1916). 
 
 118. Method of determination and 
 
 identification of proteins in 
 rubber. Caoutchouc & Gut- 
 tapercha 13, 8810-1 (1916). 
 
 119. The analysis of rubber ma- 
 
 terials. Caoutchouc & 
 Guttapercha IS; 8939-44; 
 8980-3 (1916). 
 
 120. The role of analysis in the 
 
 manufacture of rubber 
 goods. Caoutchouc Sz Gut- 
 tapercha 13, 9055 (1916). 
 
 121. Action of amines in vulcaniza- 
 
 tion. Caoutchouc <fc Gutta- 
 percha, 13, 9064 (1916). 
 
 122. Changes in resin content of 
 
 rubber on vulcanization. 
 Caoutchouc & Guttapercha 
 
 13, 9094-5 (1916). 
 
 123. Physical and chemical analysis 
 
 of rubber thread. Caout- 
 chouc & Guttapercha 13, 
 9007-8 (1916). 
 
 124. Analysis of vulcanized rubber. 
 
 Caoutchouc & Guttapercha, 
 
 14, 9189-91 (1917); 9213-6 
 (1917); 9309-13 (1917). 
 
 125. Application of catalysis to 
 
 vulcanization. Rubber Age 
 3, 78-9 (1918). 
 
 126. A comparison of the effects 
 
 produced by organic accel- 
 erators in the vulcanization 
 
 of rubber. Caoutchouc & 
 Guttapercha 15, 9635-7 
 (1918). 
 
 127. Rubber substitutes, or vulcan- 
 
 ized oils. Chimie & Indus- 
 trie 1, 727-32 (1918). 
 
 128. Polymerization and oxidation 
 
 of crude rubber. Caoutchouc 
 & Guttapercha 15, 9478-82 
 (1918). 
 
 129. Action of lipase on white fac- 
 
 tice. Caoutchouc & Gutta- 
 percha, 16, 9722-7 (1919). 
 
 130. Vulcanization accelerators. 
 
 Caoutchouc & Guttapercha 
 16, 9853; 9856-65 (1919). 
 
 131. Analysis of vulcanized rubber. 
 
 Caoutchouc & Guttapercha 
 16, 9900-7 (1919). 
 
 132. Determination of sulfur exist- 
 
 ing as sulfides in vulcanized 
 rubber. Caoutchouc & Gut- 
 tapercha 16, 9952-3 (1919). 
 
 133. The use of furfural in rubber 
 
 analysis and in the rubber 
 industrv. Caoutchouc & 
 Guttapercha 16, 9957-9 
 (1919). 
 
 134. Estimation of rubber and tex- 
 
 tiles in impermeable fabrics. 
 Caoutchouc & Guttapercha 
 16, 9907-11 (1919). 
 
 135. Analysis of vulcanized rubber. 
 
 Caoutchouc & Guttapercha 
 16, 9901-6 (1919). 
 
 136. Analysis of crude rubber. 
 
 Caoutchouc & Guttapercha 
 
 16, 10051-3 (1919). 
 
 137. The use of carbon black in 
 
 tires. Caoutchouc & Gutta- 
 percha 17, 10274-5 (1920). 
 
 138. The discovery of accelerators. 
 
 Caoutchouc & Guttapercha 
 
 17, 10427 (1920). 
 
 139. Theory of the acceleration of 
 
 vulcanization. Caoutchouc 
 & Guttapercha 17, 10511-4 
 (1920). 
 
 140. Application of some amides 
 
 and amines of furfural to 
 vulcanization. Caoutchouc 
 & Guttapercha 17, 10495-505 
 (1920). 
 
 141. The new vulcanization and 
 
 accelerators. I. Caoutchouc 
 & Guttapercha 18, 11012-5 
 (1921). 
 
 142. Ibid. II. Caoutchouc & Gut- 
 
 tapercha IS, 11121-4 (1921). 
 
126 
 
 THE ANALYSIS OF RUBBER 
 
 143. Chemical analysis of rubber 
 
 articles. Ann. chim. anal, 
 appl. 8, 335-44 (1921). 
 
 144. The new vulcanization and ac- 
 
 celerators. Caoutchouc <fe 
 Guttapercha 19, 11171-6 
 (1922). 
 Andre Dubosc and Jean Wavelet 
 
 145. Mineral Rubber. Caoutchouc 
 
 & Guttapercha 16, 10037-40 
 (1919). 
 
 146. Action of lipases on oils vul- 
 
 canized with sulfur chloride 
 (white substitutes). Bull. 
 Soc. Ind. Rouen 47, 47-59 
 (1919). 
 W . A. Ducca 
 
 147. The determination of rubber 
 
 as a tetrabromide. J. Ind. 
 Eng. Chem. 4, 372-4 (1912). 
 Richard B. Earle 
 
 148. Report of Committee on or- 
 
 ganic accelerators, Rubber 
 Section, American Chemical 
 Society. J. Ind. Eng. Chem. 
 10, 865 (1918). 
 B. ./. Eaton 
 
 149. Vulcanization catalysts. Agr. 
 
 Bull. Federated Malay States 
 5, 38-43 (1916). 
 
 150. The natural accelerators ol 
 
 Para rubber. J. Soc. Chem. 
 Ind. 37, 51T (1918). 
 B. J. Eaton and F. W. Day 
 
 151. The distribution of nitrogen 
 
 in coagulum and serum of 
 Hevea latex on coagulation 
 with acetic acid. Agr. Bull. 
 Federated Malay States 4, 
 350-3 (1916). 
 
 152. A preliminary investigation on 
 
 the estimation of free and 
 combined sulfur in vulcan- 
 ized rubber, and on the rate 
 of combination of sulfur with 
 different types of plantation 
 rubber. J. Soc. Chem. Ind. 
 86, 16-20 (1917). 
 
 153. Investigation on sulfur in vul- 
 
 canized rubber. Agr. Bull. 
 Federated Malay States 6, 
 73-87 (1917). 
 Junius David Edwards 
 
 154. Methods of exposure and per- 
 
 meability tests of balloon 
 fabrics. Third Annual Re- 
 port. Nat. Advisory Comm. 
 for Aeronautics, 459-63 
 (1917), 
 
 H. A. Endres 
 
 155. The relative activity of cer- 
 
 tain accelerators In the vul- 
 canization of rubber. Caout- 
 chouc & Guttapercha 18, 
 11089-97 (1921). 
 S. W. Epstein and B. L. Gonyo 
 
 156. The extraction of rubber 
 
 goods. Rubber Age 6, 445- 
 7 (1920). 
 S. W. Epstein and. W. E. Lange 
 
 157. Detection and determination 
 
 of glue in rubber goods. 
 India Rubber World 61, 216- 
 7 (1920). 
 S. W. Epstein and R. L. Moore. 
 
 158. Determination of cellulose in 
 
 rubber goods. II. S. Bureau 
 of Standards Tech. Paper 
 154; Rubber Age 6, 289-93 
 (1920). 
 W. Esch 
 
 159. Determination of sulfur in 
 
 rubber. Chem. Ztg. 2S, 200 
 (1904). 
 
 160. Contributions to rubber in- 
 
 vestigations. Gummi Ztg. 
 22, 766. 
 
 161. Application of the bromide 
 
 derivative methods, for the 
 determination of vulcanized 
 rubber goods. Chem. Ztg. 
 35, 971-2 (1911). 
 W. W . Evans and Ruth Merling 
 
 162. A rapid bomb method for de- 
 
 termination of sulfur in rub- 
 ber compounds. Rochester 
 meeting of the American 
 Chemical Society, April, 
 1921. 
 G. Fendler 
 
 163. New method for the analysis 
 
 of rubber. Ber. Pharm. Ges. 
 14, 208-15 (1904). 
 
 164. Determination of rubber as 
 
 tetrabromide. Gummi Ztg. 
 24, 782 (1910). 
 
 165. Determination of rubber as 
 
 nitrosite. Gummi Ztg. 24, 
 1000 (1910). 
 G. Fendler and O. Kuhn 
 
 166. Studies on rubber and rubber 
 
 analysis. Gummi Ztg. 22, 
 132-4; 160-3; 215-9; 249-52 
 (1907). 
 
 167. The determination of rubber 
 
 as the tetrabromide. Gummi 
 Ztg. 22, 710 (1908), 
 
BIBLIOGRAPHY 
 
 127 
 
 Harry L. Fisher and Harold Gray 
 
 168. The tetrahydroxyphenol de- 
 
 rivative, and its tetramethyl 
 ether. New York meeting of 
 the American Chemical So- 
 ciety, Sept., 1921. 
 Harry L. Fisher, Harold Gray and 
 Ruth Merling. 
 
 169. A discussion of the tetrabro- 
 
 mide method for determin- 
 ing rubber hydrocarbons. J. 
 Ind. Eng. Chem. 13, 1031-4 
 (1921). 
 C. P. Fox 
 
 170. Technical determination of 
 
 rubber in Guayule. J. Ind. 
 Eng. Chem. 1, 735 (1908). 
 
 171. Effect of copper on crude rub- 
 
 ber. J. Ind. Eng. Chem. 9, 
 1092-3 (1917). 
 F. Frank 
 
 172. Determination of antimony in 
 
 red rubber goods. Gummi 
 Ztg. 25, 52, 2002 (1911). 
 Frank and Birken 
 
 173. Determination of mercuric sul- 
 
 fide and antimony trisulfide 
 in rubber goods. Chem. Ztg. 
 84, 49. 
 Frank and Jacobsohn 
 
 174. Determination of cinnabar 
 
 and antimony sulfide in red 
 vulcanized rubber. Gummi 
 Ztg. 23, 1046. 
 F. Frank and E. Marckwald 
 
 175. The analysis of rubber goods. 
 
 Chem. Ztg. 26, 385 (1902). 
 
 176. On the determination of total 
 
 sulfur. Gummi Ztg. 17, 71 
 (1903). 
 
 177. Method for the direct deter- 
 
 mination of the mineral mat- 
 ter in rubber goods. Gummi 
 Ztg. 22, 1344 (1908). 
 
 178. Control of the analysis of rub- 
 
 ber goods. Gummi Ztg. 23, 
 1522 (1909). 
 
 179. The chloral hydrate method 
 
 in rubber analysis. Gummi 
 Ztg. 23, 979 (1909). 
 
 180. Direct determination of nitrog- 
 
 enous substances and the 
 impurities in raw rubber. 
 Gummi Ztg. 26, 936 
 (1912). 
 
 181. Action of copper and other 
 
 metals on rubber. I. De- 
 terioration of cable insula- 
 tion due to insufficient 
 
 vulcanization. Gummi Ztg. 
 28, 1280-3 (1914). 
 
 182. Analytical methods for sulfur 
 
 chloride. Gummi Ztg. 28, 
 1580-1 (1914). 
 
 183. The "Dracorubin" test for sol- 
 
 vents used in the rubber in- 
 dustry. Gummi Ztg. 80, 
 524-5 (1916). 
 Frank, Marckwald and Leibschutz 
 
 184. On deresinated rubber. Gum- 
 
 mi Ztg. 21, 366 (1907). 
 H. C. T. Gardener 
 
 185. Rubber. Chemist Druggist 79, 
 
 3867. 
 Gasparini 
 
 186. Total sulfur by electrolytic 
 
 method. Chim. Ital. 87, II; 
 426-61 (1907). 
 R. Gaunt 
 
 187. The estimation of sulfur in 
 
 rubber. Analyst Jfi, 9-10 
 (1915). 
 W. C. Geer 
 
 188. Accelerated life of rubber com- 
 
 pounds. India Rubber 
 World 55, 127-30 (1916). 
 W. C. Geer and W. W. Evans 
 
 189. Ten years' experience with 
 
 aging tests. India Rubber 
 World 64, 887-92 (1921). 
 P. Goldberg 
 
 190. Determination of rubber in 
 
 India rubber goods. Chem. 
 Ztg. 37, 85. 
 A*. Gorier 
 
 191. Tackiness of crude rubber. 
 
 Gummi Ztg. 80, 351 (1916). 
 
 192. The viscosity index as a stand- 
 
 ard for the preliminary test- 
 ing of the quality of rubber. 
 Chem. Zentr. I. 393-4 
 (1916). 
 K. Gottlob 
 
 193. Influence of nitrous acid on 
 
 rubber. Z. angew. Chem. 
 20, 2213-21 (1907). 
 
 194. Nitrosites of rubber and their 
 
 use in analysis. Gummi Ztg. 
 26, 1561. 
 
 195. Vulcanization accelerators 
 
 Gummi Ztg. 80, 303-8; 326- 
 36 (1916). 
 Frank Gottsch 
 
 196. Specification of vulcanized rub- 
 
 ber by volume, and its de- 
 termination by a new solu- 
 tion method. J. Ind. Eng. 
 Chem. 7, 582-6 (1915), 
 
128 
 
 THE ANALYSIS OF RUBBER 
 
 J. Gram 
 
 197. Technical rubber analysis. 
 
 Chem. Ztg. 36, 249. 
 Henry Green 
 
 198. Volume increase of com- 
 
 pounded rubber under strain. 
 J. Ind. Eng. Chem. 13, 1029- 
 31 (1921). 
 
 199. Recent development in the art 
 
 of rubber microsectioning. 
 J. Ind. Eng. Chem. 13, 1130- 
 2 (1921). 
 ■I. M. Grove 
 
 200. The relative accelerating ac- 
 
 tion of different compounds 
 of lead in the vulcanizing 
 of rubber. India Rubber 
 World 64, 633-4 (1921). 
 Alice Hamilton 
 
 201. Industrial poisons used in the 
 
 rubber industry. U. S. Bu- 
 reau of Labor Statistics No. 
 179 (1915). 
 C. Harries 
 
 202. On the reaction of rubber 
 
 with nitric acid. Ber. 34, 
 2991-2 (1901). 
 
 203. The chemistrv of Para rubber. 
 
 Ber. 35, 3256; 4429 (1902). 
 
 204. The determination of rubber 
 
 (nitrosite). Ber. 36, 1937 
 (1903). 
 
 205. Remarks concerning Weber's 
 
 "dinitro-rubber." Ber. 38, 
 87-90 (1904). 
 
 206. Regarding the action of N-0 2 
 
 on rubber. Z. Angew. Chem. 
 20, 1969 (1907). 
 
 207. Identification of the eight 
 
 carbon-ring in normal rub- 
 ber. Ber. 46, 2590 (1912). 
 
 208. Detection of synthetic rub- 
 
 ber in technical analysis. 
 Gummi Ztg. 33, 222-3 
 (1920). 
 C. Harries and H. Rim-pel 
 
 209. Determination of rubber as 
 
 tetrabromide in crude rub- 
 ber. Gummi Ztg. 23, 1370 
 (1909). 
 E. Hatschek 
 
 210. The stress-elongation curve of 
 
 vulcanized India rubber. J. 
 Soc. Chem. Ind. Ifi, 251-3T 
 (1921). 
 L. G. D. Healy 
 
 211. An electric desiccator for the 
 
 analysis of India rubber and 
 other organic compounds. J. 
 
 Ind. Eng. Chem. 5, 489-90 
 (1913). 
 A. Helbronner and G. Bernstein 
 
 212. Rubber solutions vulcanized by 
 
 ultra-violet rays. Rubber 
 Ind., 156-63 (1914). 
 T r . Henri 
 
 213. Action of ultra-violet rays on 
 
 rubber. Caoutchouc & Gut- 
 tapercha 7, 4371 (1910). 
 R. Henriques 
 
 214. Contributions to the analysis 
 
 of rubber goods. Chem. Ztg. 
 16, 1595; 1623; 1644 (1892); 
 11, 634-8; 707-9 (1893); 18, 
 411-2; 442-4; 905 (1894); 19, 
 235; 382 (1895). 
 
 215. Valuation of crude and manu- 
 
 factured rubber. J. Soc. 
 Chem. Ind. 16, 566-7 
 (1897). 
 
 216. Analytical examination of rub- 
 
 ber goods. Z. angew. Chem. 
 34, 802 (1899). 
 
 217. Analysis of India rubber. 
 
 Gummi Ztg. 14, 149; 165; 
 183; 197 (1899). 
 G. H. Hillen 
 
 218. About rubber and guttapercha. 
 
 Arch. Pharm. 251, 94-121 
 (1913). 
 
 219. The determination of cellulose 
 
 in rubber. Gummi Ztg. 30, 
 670-1 (1916). 
 F. W. Hinrichsen 
 
 220. Communications upon the 
 
 Chemistry of rubber I; (a) 
 with W. Manasse: On the 
 tetrabromide of rubber; (b) 
 with W. Manasse: Deter- 
 mination of mineral fillers; 
 (c) with W. Manasse: The 
 determination of total sul- 
 fur; (d) with K. Meisen- 
 burg : Cold vulcanization ; 
 (e) with E. Stern: hot vul- 
 canization. Chem. Ztg. 3d, 
 735-56 (1909). 
 
 221. Analysis of vulcanized rubber 
 
 goods ; Determination of 
 fillers. Chem. Ztg. 33, 813, 
 979, 1061 (1909). 
 
 222. Control of rubber material for 
 
 insulated conductors. Chem. 
 Ztg. 34, 184 (1910). 
 
 223. On the chemistry of rub- 
 
 ber. II. The tetrabromide 
 method. Gummi Ztg. 26, 
 1408 (1912). 
 
BIBLIOGRAPHY 
 
 129 
 
 224. Present state of rubber analy- 
 
 sis. Gummi Markt. 6, 253-7. 
 
 225. The theory of the vulcaniza- 
 
 tion of rubber. Mitt. kgl. 
 Materialprufungsamt 33, 407- 
 15 (1915). 
 Hinrichsen and Kemp} 
 
 226. The action of iodine on rub- 
 
 ber. Ber. 46, 1287-91 (1913). 
 Hinrichsen and Kindscher 
 
 227. Identification of Para and 
 
 Cevlon rubber. Chem. Ztg. 
 34/230 (1910). 
 
 228. Direct determination of rubber 
 
 as tetrabromide. Chem. Ztg. 
 35, 329-30 (1911). 
 
 229. Hubener's method for the de- 
 
 termination of rubber as 
 bromide. Chem. Ztg. 36, 
 217; 236 (1912). 
 
 230. The desulfurization of vulcan- 
 
 ized rubber. Koll. Z. 10, 
 146-8 (1912); 11, 38-9 (1913). 
 
 231. The theory of the vulcaniza- 
 
 tion of rubber. Koll. Z. 11, 
 191 (1913). 
 
 232. Direct determination of rub- 
 
 ber as tetrabromide. Z. anorg. 
 Chem. 81, 70-82 (1913). 
 
 233. The action of sulfur chloride 
 
 and sulfur on rubber. Ber. 
 46, 1291 (1913). 
 Hinrichsen and Marcusson 
 
 234. Rubber resins. Z. angew. 
 
 Chem. 23, 49 (1910). 
 Hinrichsen, Quensell and Kindscher 
 
 235. Addition compounds of hydro- 
 
 gen ha 1 ides and halogens 
 with rubber. Ber. 46, 1283- 
 7 (1913). 
 If. II. Hodgson 
 
 236. Notes on a recent comparison 
 
 study of methods for deter- 
 mining sulfur in rubber. 
 India Rubber J. 47, 315 
 (1914). 
 I). Hold* 
 
 237. Analysis of India rubber 
 
 goods. Mitt. kgl. tech. 
 Versuchsamt. 1892; p. 366; 
 J. Soc. Chem. Ind. 12, 469 
 (1893). 
 Matthew Howie 
 
 238. Determination of nitrogen 
 
 content of rubber. J. Soc. 
 (hem. Ind. 37, 85T (1918). 
 G. Hubener 
 
 239. Analyses and analytical 
 
 methods for hard vulcanized 
 
 240. 
 241. 
 
 242. 
 243. 
 
 244. 
 
 215. 
 
 246. 
 247. 
 
 248. 
 
 249. 
 250. 
 
 rubber. Chem. Ztg. 33, 144; 
 155 (1909). 
 Method for determining rub- 
 ber in hot vulcanized soft 
 goods. Chem. Ztg. 
 662 (1909). 
 method for estimat- 
 sulfur of vulcaniza- 
 
 hard 
 Gummi 
 
 rubber 
 S3, 648, 
 A simple 
 ing the 
 tion in 
 rubber. 
 
 vulcanized 
 
 Ztg. 
 
 24, 
 
 213 (1910). 
 Concerning hard vulcanized 
 
 rubber. Gummi Ztg. 24; 627- 
 
 9; 740 (1910). 
 Determination of rubber in 
 
 crude rubber as tetrabro- 
 
 740 
 
 Ztg. 
 
 24, 
 
 mide. Gummi 
 (1910). 
 
 Direct determination of rubber 
 in vulcanized rubber goods. 
 Chem. Ztg. 35, 113-5 (1911). 
 
 Direct determination of rub- 
 ber in vulcanized rubber 
 goods. Chem. Ztg. 34, 1307- 
 8; 1315-6 (1910). 
 
 Rubber tetrabromide. Gummi 
 Ztg. 25, 634; 751-2 (1911). 
 
 The determination of total sul- 
 fur in rubber. C. A. 6, 2001) 
 (1912). 
 
 The bromide method for the 
 determination of rubber. 
 Gummi Ztg. 26, 1281 (1912). 
 Cf. also Gummi Ztg. 28, 320 
 (1914). 
 
 On the oxidation of rubber. 
 Gummi Ztg. 28, 237-9 (1913). 
 
 The determination of syn- 
 thetic rubber. Gummi Ztg. 
 88, 361-2 (1919). 
 
 ut in 
 
 Drying the acetone extract of 
 rubber. Ann. chim. anal. 
 B0, 212-3 (1915). 
 
 252. Determination of total sulfur 
 
 in the products of the rub- 
 ber industry. Ann. chim. 
 anal. 20, 214-6 (1915). 
 
 253. Observations on the subject of 
 
 the determination of total 
 chlorine in rubber substi- 
 tutes. Ann. chim. anal. 20, 
 241-2 (1915). 
 
 254. Rapid analysis of commercial 
 
 sulfides of antimony. Ann. 
 chim. anal. 21, 3-7 (1916). 
 
 255. Vulcanization accelerators. 
 
 of vulcanization. Mon. Sci. 
 7, 193-6 (1917). 
 
 A. H 
 251. 
 
130 
 
 THE ANALYSIS OF RUBBER 
 
 256. Vulcanization accelerators. 
 
 Caoutchouc & Guttapercha 
 15, 9596-8 (1918). « 
 H. Ingle 
 
 257. Some observations on rubber 
 
 resins. J. Soc. Chem. Ind. 
 SI, 272-3 (1912). 
 International Rubber Testing Com- 
 mittee 
 
 258. Suggestions for the standard 
 
 testing of rubber. Gummi 
 Ztg. 25, 1277 (1911). 
 Felix Jacobson 
 
 259. The determination of fillers. 
 
 Gummi Ztg. 27, 1906 (1913). 
 J. Jacoby 
 
 260. A new extraction apparatus. 
 
 Gummi Ztg. 27, 1870 (1913). 
 Joint Rubber Insulation Committee 
 
 261. See W. A. Del Mar 
 W. Jones 
 
 262. Analysis of rubber compounds. 
 
 Elec. World 59, 320. 
 H. W . Jones 
 
 263. The use of nitric acid as a sol- 
 
 vent for compounded and 
 vulcanized rubber. The 
 Rubber Ind. (1914), p. 189. 
 
 264. Methods of determining small 
 
 amounts of carbon dioxide 
 in rubber goods in the pres- 
 ence of sulfides. The Rub- 
 ber Ind. 1914, 191-2; J. Soc. 
 Chem. Ind. 84, 672 (1915). 
 
 265. A simple method for the de- 
 
 termination of mineral mat- 
 ter in rubber mixings. The 
 Rubber Ind. 1914, 199-200. 
 A. W. Jones, Jr. 
 
 266. Estimation of mineral matter 
 
 in vulcanized rubber. Chem. 
 Analyst 20, 21 (1917). 
 Kaye and Sharp 
 
 267. Rapid method for the estima- 
 
 tion of sulfur in vulcanized 
 rubber. India Rubber J. 44, 
 1189 (1912). 
 W. J. Kelly 
 
 268. The determination of the true 
 
 free sulfur, and the true co- 
 efficient of vulcanization in 
 vulcanized rubber. J. Ind. 
 Eng. Chem. 12, 875-8 
 (1920). 
 
 269. Determination of true free 
 
 sulfur, and the coefficient of 
 vulcanization in vulcanized 
 rubber. II. J. Ind. Eng. 
 Chem. 14, 196-7 (1922). 
 
 Andrew H. King 
 
 270. The chemical analysis of rub- 
 
 ber goods. Met. Chem. Eng. 
 14,581-4 (1916). 
 
 271. Rubber vulcanization accel- 
 
 erators. Met. Chem. Eng. 
 15, 231-4 (1916). 
 
 272. Rubber substitutes. Mot. 
 
 Chem. Eng. 18, 630-6 
 (1918). 
 F. Kirchhof 
 
 273. The direct determination of 
 
 rubber by titration with bro- 
 mine. Gummi Ztg. 27, 9 
 (1913). 
 
 274. On the oxidation of rubber. 
 
 Roll. Z. 13, 49-61 (1913). 
 
 275. The regeneration of rubber 
 
 from its tetrabromide. Koll. 
 Z. 15, 126-31 (1914). 
 
 276. The action of concentrated 
 
 sulfuric acid on natural and 
 artificial rubber. Koll. Z. 
 17,311-5 (1920). 
 Martin Klassert 
 
 277. Rubber resins. Z. angew. 
 
 Chem. 26, 471-2 (1913). 
 0. H. Klein, J. H. Link and F. 
 Gottsch 
 
 278. The determination of mineral 
 
 fillers in rubber by the ani- 
 line method. J. Ind. Eng. 
 Chem. 9, 140-1 (1917). 
 /. L. Kondakov 
 
 279. Syntheses of rubber. Caout- 
 
 chouc & Guttapercha 18, 
 10980-84; 11097-100 (1921); 
 19, 11169-71 (1922). 
 
 F. von Konek 
 
 280. Rapid estimation of sulfur in 
 
 organic compounds, coals, 
 
 mineral oils, etc. Z. angew. 
 
 Chem. 22, 516 (1903). 
 L. dr Konivgh 
 2S1. Estimation of mineral matter 
 
 in rubber goods. J. Am. 
 
 Chem. Soc. 19, 952-6 (1897). 
 Otto Korneck 
 
 282. Critical investigation of the 
 
 methods of analysis of crude 
 rubber. Gummi Ztg. 25, 
 4-9; 42-6; 77-88 (1910). 
 
 283. Determination of rubber as 
 
 tetrabromide. Gummi Ztg. 
 25, 424 (1911). 
 
 G. D. Kratz 
 
 284. Rate of vulcanization and 
 
 factory output. India Rub- 
 ber J. 51, 661-2 (1916). 
 
BIBLIOGRAPHY 
 
 131 
 
 G. D. Kralz and Arthur H. Flower 
 
 285. The vulcanization of rubber at 
 
 constant temperatures and 
 by a series of increasing tem- 
 peratures. J. Ind. Eng. 
 Chem. 11, 30-3 (1919). 
 
 286. Effect of certain accelerators 
 
 upon the properties of vul- 
 canized rubber. Chem. Met. 
 Eng. 20, 417-20 (1919). 
 
 287. The effect of certain accelera- 
 
 tors on the properties of vul- 
 canized rubber, II. J. Ind. 
 Eng. Chem. 12, 971-4 
 (1920). 
 G. D. Kratz, A. H. Flower, and Cole 
 Coolidge 
 
 288. The action of certain organic 
 
 accelerators in the vulcan- 
 izing of rubber. J. Ind. Eng. 
 Chem. 12, 317-24 (1920). 
 
 289. The action of certain organic 
 
 accelerators in the vulcan- 
 ization of rubber. J. Ind. 
 Eng. Chem. 13, 67-70 (1921) ; 
 Do. Ill; J. Ind. Eng. Chem. 
 13, 128-31 (1921). 
 H. Kuhl 
 
 290. Determination of lead and 
 
 zinc in rubber goods. Apoth. 
 Ztg. 51, 135. 
 A. A. Ladon 
 
 291. Testing rubber insulation. 
 
 Met. Chem. Eng. 14, 560 
 (1916). 
 Alan Leighton 
 
 292. The inflammability of gas 
 
 black. India Rubber World 
 60, 425-6 (1919). 
 M. Levin and S. Collier 
 
 293. An improved method for the 
 
 determination of sulfur in 
 rubber. Rubber Age 9, 47-8 
 (1921). 
 W. K. Lewis and W. H. McAdams 
 
 294. A direct method for the 
 
 determination of rubber hy- 
 drocarbons in crude and 
 vulcanized rubber. J. Ind. 
 Eng. Chem. 12, 675-6 (1920). 
 Lindley 
 
 295. Some types of mineral rubber. 
 
 India Rubber World 47, 17. 
 J. H. Link 
 
 296. Preparation of rubber samples. 
 
 Chem. Analyst 15, 7 (1915). 
 II. Loewen 
 
 297. Determination of fillers. 
 
 Gummi Ztg. 28, 7 (1914). 
 
 C. A. Lobry de Bruyn and F. II. 
 ran Lieant 
 
 298. Analysis of India rubber 
 
 articles. Chem. Ztg. IS, 309 
 (1894). 
 
 B. D. W. Luff and B. D. Porritt 
 
 299. The determination of avail- 
 
 able sulfur in golden sulfide 
 of antimony. J. Soc. Chem. 
 Ind. 40, 275-8T (1921). 
 W. C. McLewis 
 
 300. The properties of the colloidal 
 
 state and their applications 
 to industry. Caoutchouc & 
 Guttapercha 17, 10261-2 
 (1920). 
 Marckwald and Frank 
 
 301. Magnesia in rubber com- 
 
 pounds and vulcanization. 
 Gummi Ztg. 28, 222 (1914). 
 /. Marcusson and H. Descher 
 
 302. The determination of the sul- 
 
 fur and halogen content of 
 organic substances. Mitt. 
 kgl. Materialprufungsamt 30, 
 130. 
 Marcusson and Hinrichsen 
 
 303. The determination of fillers in 
 
 rubber mixtures. Chem. 
 Ztg. 84, 839 (1910). 
 Marquis and Heim 
 
 304. A method for determining 
 
 rubber in crude rubber. 
 Bull. Soc. Chim. 13, 862. 
 A. Maximov 
 
 305. The acceleration of vulcaniza- 
 
 tion by organic accelerators. 
 Caoutchouc & Guttapercha 
 18, 10944-7; 10986-8 (1921). 
 Morisson 
 
 306. Analysis of vulcanized rubber. 
 
 Caoutchouc & Guttapercha 
 17, 10233 (1920). 
 A. M. Munro 
 
 307. A new rapid method for the 
 
 determination of sulfur in 
 rubber compounds. India 
 Rubber World 62, 426 (1920). 
 R. Nicolardot 
 
 308. Examination of a sample of 
 
 synthetic rubber. Ann. 
 chim. anal. chim. Appl. 8, 
 176-9. 
 
 C. Olin North 
 
 309. The effect of compounding in- 
 
 gredients on the physical 
 properties of rubber. India 
 Rubber World 63, 98-102 
 (1920). 
 
132 
 
 THE ANALYSIS OF RUBBER 
 
 310. Mineral rubber. India Rub- 
 
 ber World 65, 191-2 (1921). 
 G. Noycr 
 
 311. Analysis of rubber overshoes. 
 
 Caoutchouc & Guttapercha 
 14, 9272-4 (1917). 
 
 312. Flowers of sulfur and sub- 
 
 limed sulfur. Caoutchouc & 
 Guttapercha, 15, 9661-3 
 (1918). 
 /. /. Ostromuislenskii 
 
 313. A new method of solid vulcan- 
 
 ization of rubber by means 
 of sulfur, trinitrobenzene or 
 benzoyl peroxide. J. Russ. 
 Phvs.'Chem. Soc. 47, 1885-92 
 (1915); C. A. 10, 1943. 
 
 314. Mechanism of the action of 
 
 amines and metallic oxides 
 on the process of the vul- 
 canization of rubber. J. 
 Russ. Phys. Chem. Soc. 47, 
 1892-8 (1915); C. A. 10, 
 1944. 
 
 315. A new method of vulcanizing 
 
 rubber by means of halo- 
 gen compounds, and the 
 mechanism of vulcanization. 
 J. Russ. Phys. Chem. Soc. 
 47, 1898-904 (1915); C. A. 
 10, 1945. 
 
 316. The preparation of vulcanized 
 
 rubber, colored with organic 
 dves. J. Russ. Phvs. Chem. 
 Soc. 47, 1993-5 (1915) ; C. A. 
 10, 1949. 
 
 317. A new method of vulcanizing 
 
 rubber with molecular oxy- 
 gen, ozone, or organic ozon- 
 ides. J. Russ. Phvs. Chem. 
 Soc. 47, 1904-9 (1915) ; C. A. 
 10, 1946. 
 
 318. Vulcanization or synthetic 
 
 rubbers. J. Russ. Phys. 
 Chem. Soc. 47, 1441-53 
 (1915); C. A. 10, 3176. 
 
 319. Mechanism of the process of 
 
 vulcanization of rubber. J. 
 Russ. Phvs. Chem. Soc. 47, 
 1453-61 (1915); C. A. 10, 
 3177. 
 
 320. Hot vulcanization of rubber 
 
 by nitro compounds without 
 sulfur. J. Russ Phvs. Chem. 
 Soc. 47, 1462-7 (1915) ; C. 
 A. 10, 3177. 
 
 321. Hot vulcanization of rubber 
 
 by peroxides or peracids, 
 without sulfur. J. Russ. 
 
 Phys. Chem. Soc. 47, 1467- 
 71 (1915); C. A. 10, 3177. 
 S. J. Peachey 
 
 322. The action of gaseous oxygen 
 
 on India rubber. J. Soc. 
 Chem. Ind. 31, 1103-5 
 (1912); 32, 179-80 (1913). 
 
 323. Vulcanization accelerators. 
 
 India Rubber J. 52, 693 
 (1916); J. Soc. Chem. Ind. 
 36, 321; 326; 424-9; 950 
 (1917). 
 S. J. Peachey and M. Leon 
 
 324. The oxidation of rubber. J. 
 
 Soc. Chem. Ind. 37, 55-60 
 (1918). 
 S. J. Peachey and A. Skipsey 
 
 325. A new process for the vulcan- 
 
 ization of rubber. J. Soc. 
 Chem. Ind. 40, 5-6T (1921). 
 M. J. B. Peregrin 
 
 326. Determination of fillers in rub- 
 
 ber goods. Ann. chim. anal. 
 22, 27-9 (1917). 
 0. Pfeiffer 
 
 327. Determination of sulfur by 
 
 Eschka's method. Chem. 
 Ztg. 28, 38 (1904). 
 H. Pohle 
 
 328. Contributions to the knowl- 
 
 edge of rubber, especially 
 regarding two dimethyl rub- 
 bers. Kolloidchem. Beihefte 
 13, 1-60 (1920). 
 M . Pontio 
 
 329. A new method for determin- 
 
 ing the free sulfur in vul- 
 canized rubber. Caoutchouc 
 & Guttapercha 5, 2194 
 (1908). 
 
 330. Determination of organic and 
 
 mineral impurities in crude 
 and vulcanized rubber. 
 Caoutchouc & Guttapercha 
 6, 2752 (1909). 
 
 331. Determination of total sulfur 
 
 in vulcanized rubber. Caout- 
 chouc & Guttapercha 6, 
 2751-2 (1909). 
 
 332. Commercial analysis of manu- 
 
 factured rubber. Bull. Soc. 
 Chim. 5, 423; 428-42 (1909). 
 
 333. A direct method for determin- 
 
 ing pure rubber in crude 
 rubber. Rev. gen. chim. 16, 
 400-1 (1913). 
 
 334. Direct estimation of pure rub- 
 
 ber in crude rubber, accord- 
 ing to Marquis and Heim. 
 
BIBLIOGRAPHY 
 
 133 
 
 Ann. chim. anal. 19, 60-62 
 (1914). 
 //. E. Potts 
 
 335. The oxidation of .sulfur by 
 
 nitric acid. Rubber Ind. 
 1911, 336-43. 
 
 B. D. Porritt 
 
 336. Isolation and examination of 
 
 textile in rubber-proofed 
 fabric. India Rubber J. 57, 
 559-63 (1919). 
 
 337. Action of light on rubber. 
 
 India Rubber J. 60, 1159-62 
 (1920). 
 Porritt and Anderson 
 
 338. The estimation of bitumens in 
 
 rubber mixings. The Rubber 
 Ind. 1914, 181-9. 
 Porritt and Wheatley 
 
 339. Experiments on the deter- 
 
 mination of mineral matter 
 in rubber mixings. The 
 Rubber Ind. 1914, 193-9. 
 D. H. Priestley and H. Skellon 
 
 340. Rubber substitutes and their 
 
 influence in vulcanization. 
 Caoutchouc & Guttapercha 
 19, 11181-3 (1922). 
 Robert S. Quinby 
 
 341. Benzene poisoning in rubber 
 
 manufacturing. Rubber Age 
 S, 451 (1921). 
 
 C. Reinhardt 
 
 342. Estimation of the ash of vul- 
 
 canized rubber. Stahl und 
 Eisen 1884, 649. 
 
 343. Analysis of vulcanized rubber. 
 
 Chem. News 52, 225 (1885). 
 
 D. Repony 
 
 344. An alcoholic potash extrac- 
 
 tion method. India Rubber 
 World 55, 277 (1917). 
 
 345. Analysis of antimony penta- 
 
 sulfide. India Rubber 
 World 60, 360-1 (1919). 
 
 346. Effect of light on unvulcan- 
 
 ized rubber compounds. 
 Rubber Age 5, 5 (1919). 
 
 E. Romani 
 
 347. Thiuram disulfides as vulcan- 
 
 izing agents for rubber. 
 Giorn. chim. ind. appl. 3, 
 197-9 (1921). 
 L. Rosenstein 
 
 348. The determination of sulfur in 
 
 rubber. India Rubber J. 
 67, 1100 (1919). 
 A. van Rossem 
 
 349. The water extract of crude 
 
 rubber. Coram. Netherlands 
 Govt. Inst. 1917. II. 66-8. 
 
 350. Indirect method to determine 
 
 the percentage of rubber. 
 J. Soc. ('horn. Ind. 36, 1105 
 (1917). 
 
 351. The degree of acidity of crude 
 
 rubber. J. Soc. Chem. Ind. 
 86, 1105 (1917). 
 
 352. The testing of rubber. Comm. 
 
 Netherlands Govt. Inst. 
 1917. V. 125-75. 
 
 353. Depolymerization of crude 
 
 rubber. Gummi Ztg. 31, 
 851 (1917). 
 A. van Rossem and P. Dekker 
 
 354. The so-called insoluble part of 
 
 crude rubber. J. Soc. Chem. 
 Ind. 36, 1105 (1917). 
 
 355. The chemical examination of 
 
 antimonv sulfides. India 
 Rubber J. 60, 905 (1920). 
 Rothe 
 
 356. The determination of cinnabar 
 
 and golden sulfide in rubber 
 goods. Chem. Ztg. 83, 679 
 (1909). 
 /. R. Ruby and Harlan A. Depew 
 
 357. The aging of certain rubber 
 
 compounds. J. Ind. Eng. 
 Chem. 12, 1155-6 (1920). 
 W. W. Sanders 
 
 358. Aniline oil poisoning. J. Ind. 
 
 Eng. Chem. 8, 854-5 (1916). 
 John A. Schaefjer 
 
 359. Determination of lead sulfate, 
 
 and the ultimate estimation 
 of sublimed white lead in 
 rubber. J. Ind. Eng. Chem. 
 4, 836 (1912). 
 Philip Schidrowitz 
 
 360. Recent advances in the analy- 
 
 sis and evaluation of rubber 
 and rubber goods. Analyst 
 40, 223-33 (1915). 
 Philip Schidrowitz and J. R. Bur- 
 nand 
 
 361. Vulcanization. B. Some ef- 
 
 fects of acceleration on the 
 rubber stress-strain curve. 
 J. Soc. Chem. Ind. 40, 268- 
 74T (1922). 
 Philip Schidrowitz and E. Kaye 
 
 362. On the composition of some 
 
 new crude rubbers. J. Soc. 
 Chem. Ind. 26, 129 (1907). 
 H. F. Schippel 
 
 363. Volume increase of com- 
 
 pounded rubber under strain. 
 
134 
 
 THE ANALYSIS OF RUBBER 
 
 J. Ind. Eng. Chem. 12, 33- 
 7 (1920). 
 W. Schmidt 
 
 364. The determination of nitrogen 
 
 in rubber. Gummi Ztg. 26, 
 1877 (1912). 
 
 365. The determination of anti- 
 
 mony in red rubber goods. 
 Gummi Ztg. 25, 1928 (1911). 
 
 366. The insoluble constituent of 
 
 rubber and its isolation. 
 Gummi Ztg. 27, 1085; 1131 
 (1913). 
 
 367. The direct volumetric estima- 
 
 tion of rubber by means of 
 bromine. Gummi Ztg. 27, 
 1342 (1913). 
 
 368. Analysis of antimony pentasul- 
 
 fide. Gummi Ztg. 28, 453-5 
 (1914). 
 
 369. The isolation of the in- 
 
 soluble constituent of rub- 
 ber. Gummi Ztg. 28, 1740 
 (1914). 
 
 370. Bromination and constitution 
 
 of rubber. Gummi Ztg. 34, 
 167-9; 193-5 (1919). 
 
 371. Rapid determination of golden 
 
 sulfide of antimony in rub- 
 ber goods and reclaimed 
 rubber. Gummi Ztg. 32, 2- 
 3 (1918). 
 Schneider 
 
 372. Valuation of India rubber. 
 
 Gummi Ztg. 17, 874 (1903). 
 Winfield Scott and C. W. Bedford 
 
 373. Reactions of accelerators dur- 
 
 ing vulcanization. II. A 
 theory of accelerators based 
 on the formation of polysul- 
 fides during vulcanization. 
 J. Ind. Eng. Chem. IS, 125-8 
 (1921). 
 Eugen Seidl 
 
 374. On the action of litharge dur- 
 
 ing the hot vulcanization of 
 rubber. 'Gummi Ztg. 24, 
 710-3; 748-51 (1910). 
 Setlik and Zojka 
 
 375. Analysis and testing of rubber 
 
 goods. 8th Int. Cong. App. 
 Chem. 9, 89-93. 
 J. W. Shields 
 
 376. Corrected stress-strain curves 
 
 for rubber. Rubber Age 10, 
 88-9 (1921). 
 D. A. Shirk 
 
 377. Antimony sulfuret (sulfide) ; 
 
 its manufacture and use. 
 
 Rubber Age 9, 87-90 
 (1921). 
 A. Short and F. H. Sharpe 
 
 378. Golden sulfide of antimony. 
 
 India Rubber J. 73, 679-80 
 (1922). 
 H. Skellon 
 
 379. The mass action of sulfur in 
 
 vulcanization. The Rubber 
 Ind. 1914, 172-80. 
 A. H. Smith and S. W. Epstein 
 
 380. The determination of free car- 
 
 bon in rubber goods. J. Ind. 
 Eng. Chem. 11, 33-6 (1919). 
 G. Harold Smith 
 
 381. The use of gas black in rubber 
 
 compounds. India Rubber 
 J. 63, 471-3 (1922). 
 D. Spence 
 
 382. The methods of analysis of 
 
 crude rubber. Gummi Ztsj;. 
 22, 188 (1908). 
 
 383. On the distribution of protein 
 
 in Para rubber. India Rub- 
 ber J. 34, 766. 
 
 384. The influence of protein on 
 
 the determination of rubber 
 as a tetrabromide. Gummi 
 Ztg. 24, 212 (1909). 
 
 385. The relationship of mechani- 
 
 cal to chemical properties of 
 vulcanized rubber. India 
 Rubber J. 52, 861 (1916). 
 
 386. Organic vulcanizing accelera- 
 
 tors. J. Soc. Chem. Ind. 36, 
 118-9 (1917). 
 
 387. Catalysts in vulcanization. J. 
 
 Ind Eng. Chem. 10, 115-7 
 (1918). 
 Spence. and Galletly 
 
 388. Action of chromyl chloride on 
 
 India rubber. J. Am. Chem. 
 Soc. 33, 190 (1911). 
 
 389. The determination of rubber 
 
 by Budde's tetrabromide 
 method; the influence of 
 protein. Gummi Markt. 5, 
 109. 
 
 390. On the estimation of rubber as 
 
 tetrabromide ; decomposition 
 of the bromide by igni- 
 tion with sodium carbonate. 
 Caoutchouc & Guttapercha 
 8, 5313 (1911); India Rub- 
 ber J. 43, 441 (1912). 
 Spence, Galletly and Scott 
 
 391. Communication on the chem- 
 
 istry of rubber, II; Physical 
 Chemical examination of 
 
BIBLIOGRAPHY 
 
 135 
 
 resin extraction. Koll. Z. 9, 
 274-80 (1911). 
 
 392. Determination of rubber as 
 
 tetrabromide ; decomposition 
 of the bromide by nitric 
 acid. Gummi Ztg. 25, 301-3 
 (1911). Cf. also Gummi 
 Markt. 5, 113. 
 D. Spence and G. D. Kratz 
 
 393. The insoluble constituent of 
 
 crude rubber; its isolation 
 and characterization. Koll. 
 Z. 14, 262 (1914). 
 /). Spence and C. A. Ward 
 
 394. The Chemistry of India rub- 
 
 ber. IV. The theory of 
 vulcanization. Z. Chem. 
 Ind. Kolloide 11, 274 
 (1913). 
 
 D. Spence and J. Young 
 
 395. Comparison of some methods 
 
 for the estimation of sulfur 
 in vulcanized rubber, with 
 especial reference to electro- 
 lytic oxidation. J. Ind. Eng. 
 Chem. 4, 413 (1912). 
 
 E. Stern 
 
 396. Technical reports of the Sec- 
 
 ond International Rubber 
 Conference in London. 
 Gummi Ztg. 25, 1570; 1604 
 (1911). 
 //. P. Stevens 
 
 397. The influence of various nitro- 
 
 gen and resinous substances 
 on the vulcanizing properties 
 of rubber. India Rubber J. 
 47, 403-5. 
 
 398. The estimation of sulfur in 
 
 rubber. Analyst 39, 74; 
 India Rubber J. 47, 785. 
 
 399. The estimation of sulfide and 
 
 sulfate-sulfur and the action 
 of solvents on vulcanized 
 rubber. Analyst Ifi, 275- 
 81. 
 
 400. The vulcanization of rubber 
 
 by agents other than sulfur. 
 J. Soc. Chem. Ind. 85, 107-9 
 (1917). 
 
 401. The coefficient of vulcaniza- 
 
 tion and the state of cure. 
 India Rubber J. 53, 220-3 
 (1917). 
 
 402. The function of litharge in the 
 
 vulcanization of rubber; the 
 influence of the resinous con- 
 stituents. J. Soc. Chem. Ind. 
 35, 874-7 (1916). 
 
 403. The natural accelerator of 
 
 Para rubber. J. Soc. Chem. 
 Ind. 86, 365-70 (1917). 
 
 404. Comparative methods for de- 
 
 termining the state of cure 
 of rubber. J. Soc. Chem. 
 Ind. 37, 280-4T (1918). 
 
 405. The nature of vulcanization. 
 
 J. Soc. Chem. Ind. 39, 192-6T 
 (1919). 
 
 406. The comparative effect of 
 
 organic and inorganic accel- 
 erators in vulcanizing rub- 
 ber. J. Soc. Chem. Ind. 37, 
 156-8T (1918). 
 
 407. The determination of sulfur in 
 
 vulcanized rubber. Analyst 
 43, 377-8 (1918). 
 
 408. The comparative effect of or- 
 
 ganic accelerators and mag- 
 nesia in vulcanizing rubber. 
 India Rubber J. 58, 527-31 
 (1919). 
 
 409. Notes on Accelerators. India 
 
 Rubber World 62, 719-20 
 (1920). 
 
 410. Methods of testing. Bull. 
 
 Rubber Growers Assocn. /, 
 34-8 (1919). 
 
 411. The analysis of rubber. Caout- 
 
 chouc & Guttapercha 17, 
 10455-8 (1920). 
 
 412. The aging of vulcanized rub- 
 
 ber. Bull. Rubber Growers 
 Assocn. 2, 270-6 (1920). 
 Thale 
 
 413. The analysis of ebonite. 
 
 Chem. Ztg. 80, 499 (1906). 
 M. Toch 
 
 414. The influence of pigments on 
 
 rubber. J. Ind. Eng. Chem. 
 9, 694-6 (1917). 
 J. Torrey 
 
 415. A new method for determin- 
 
 ing rubber. India Rubber J. 
 30, 417-8 (1905). 
 Tschirch and Schmitz 
 
 416. Method for determining ni- 
 
 trogenous substances in 
 crude rubber. Gummi Ztg. 
 26, 2079 (1912). 
 John B. Tuttle 
 
 417. The sampling of rubber goods. 
 
 J. Ind. Eng. Chem. 5, 618 
 (1913). 
 
 418. The determination of barium 
 
 carbonate and barium sul- 
 fate in rubber goods. U. S. 
 Bureau of Standards Tech. 
 
136 
 
 THE ANALYSIS OF RUBBER 
 
 Paper 64; J. Ind. Eng. Chem. 
 8, 324-6 (1916). 
 
 419. Chemical tests of balloon fab- 
 
 rics. Third Annual report,, 
 Nat. Advisory Coram, for 
 Aeronautics 1917, 463-6. 
 
 420. Action of heat and light on 
 
 vulcanized rubber. The 
 Rubber Age 8, 271-2 
 (1921). 
 
 421. Guides to analyses; The in- 
 
 terpretation of rubber analy- 
 sis. Chem. Bull. 7, 332-4 
 (1920). 
 
 422. The variability of crude rub- 
 
 ber. J. Ind. Eng. Chem. 13, 
 519-22 (1921). 
 John B. Tuttle and A. Isaacs 
 
 423. A study of some recent 
 
 methods for the determina- 
 tion of total sulfur in rub- 
 ber. U. S. Bureau of 
 Standards Tech. Paper 45; 
 J. Ind. Eng. Chem. 7, 658- 
 63 (1915). 
 John B. Tuttle and Louis Yurow 
 
 424. Preliminary note on the direct 
 
 determination of rubber. 
 India Rubber World 57, 17- 
 8 (1917). 
 
 425. The direct determination of 
 
 India rubber by the nitrosite 
 method. U. S. Bureau of 
 Standards Tech. Paper 145 
 (1919); C. A. 14, 1063. 
 Douglas F. Twiss 
 
 426. Hydrochloric acid-ether mix- 
 
 ture as a reagent for rubber 
 analysis. India Rubber J. 
 50, 199 (1915). 
 
 427. C. 0. Weber's chemical test 
 
 for sun cracking. India 
 Rubber J. 52, 325 (1916). 
 
 428. The chemistry of vulcaniza- 
 
 tion. J. Soc. Chem. Ind. 36, 
 782-9 (1917). 
 
 429. Vulcanization catalysts. J. 
 
 Soc. Chem. Ind. 36, 1072 
 (1917). 
 
 430. The discontinuity of vulcan- 
 
 ization in the presence of 
 organic accelerators. J. Soc. 
 Chem. Ind. 40, 242-8T 
 (1921). 
 
 431. Determination of available 
 
 sulfur in golden sulfide of 
 antimony. J. Soc. Chem. 
 Ind. 41, 20T (1922). 
 
 432. Rubber as a colloid. Caout- 
 
 chouc & Guttapercha 17, 
 10240-3 (1920). 
 1). F. Twiss and S. A, Brazier 
 
 433. Acceleration of vulcanization. 
 
 J. Soc. Chem. Ind. 39, 
 125-32T; 155-6T; 287T 
 (1920). 
 D. F. Twiss, S. A. Brazier, and F. 
 Thomas 
 
 434. The Mithiocarbamate accel- 
 
 erators of vulcanization. J. 
 Soc. Chem. Ind. 41, 81-8T 
 (1922). 
 D. F. Twiss and G, Martin 
 
 435. The detection of accelerators. 
 
 The Rubber Age 9, 379-80 
 (1921). 
 S. Uchida 
 
 436. Notes on some fatty oils (in- 
 
 cluding Para rubber tree 
 oil). J. Soc. Chem. Ind. 35, 
 1089-90 (1916). 
 
 A. J. Ultee 
 
 437. Stearins from Castilloa and 
 
 Ficus rubber. Chem. Zentr. 
 II, 1469 (1912). 
 
 438. Resin content of Ficus rubber. 
 
 Gummi Ztg. 26, 1427 (1912). 
 
 B. linger 
 
 439. Analysis of vulcanized rubber. 
 
 and the estimation of the 
 sulfur in the same. Z. anal, 
 chem. 24, 167-75. 
 H. S. Upton 
 
 440. Volumetric determination of 
 
 free sulfur in soft rubber 
 compounds. J. Ind. Eng. 
 Chem. 10, 518-20 (1918). 
 Vtz 
 
 441. Determination of rubber as 
 
 tetrabromide. Gummi Ztg. 
 26, 968 (1912). 
 
 442. Estimation of mineral fillers 
 
 in rubber goods. Kunst- 
 stoffe 2, 381; 409. 
 
 443. The investigation of golden 
 
 antimony. Gummi Ztg. 28, 
 126 (1914). 
 
 444. The determination of total sul- 
 
 fur in rubber. Gummi Ztg. 
 28, 631-2 (1914). 
 
 445. Recent methods for the esti- 
 
 mation of rubber. Z. angew. 
 Chem. 32, 235-6 (1919). 
 Vallery 
 
 446. Application of Koettstorfer's 
 
 Index to the determination 
 of resins and adulterations 
 in rubber. Mon. sci. 3, 82. 
 
BIBLIOGRAPHY 
 
 137 
 
 W. Vaubel 
 
 447. Contribution to the bromide 
 
 method for the determina- 
 tion of rubber. Gummi 
 Ztg. 26, 1879 (1912). 
 
 448. The properties of factice. 
 
 Gummi Ztg. 27, 1254 
 (1913). 
 W. Vaubel and E. Weinarth 
 
 449. Further comparative studies 
 
 on crude and vulcanized rub- 
 bers, by means of the bro- 
 mination method. Gummi 
 Ztg. 28, 92 (1914). 
 C. S. V enable and Tyler Fmoa 
 
 450. The solubility of gases in rub- 
 
 ber and rubber stock, and 
 effect of solubility on pene- 
 trability. J. Ind. Eng. Chem. 
 14, 139-42 (1922). 
 C. S. Venable and C. D. Greene 
 
 451. Solubilitv of sulfur in rubber. 
 
 J. Ind'. Eng. Chem. 14, 319- 
 20 (1922). 
 W. W. Vogt 
 
 452. Determination of the particle 
 
 size of pigments. India 
 Rubber World 66, 347-9 
 (1922). 
 0. de Vries 
 
 453. Rubber testing at the Central 
 
 Rubber Testing Station at 
 Buitenzorg, Java. India 
 Rubber J. 52, 717-9 (1916). 
 
 454. The coefficient of vulcaniza- 
 
 tion as a check on the state 
 of cure. India Rubber J. 
 53, 101-3 (1917). 
 
 455. The relation between the co- 
 
 efficient of vulcanization and 
 the mechanical properties of 
 rubber. J. Ind. Eng. Chem. 
 13, 1133-4 (1921). 
 0. de Vries and H. J. Hellendoorn 
 
 456. The aging of vulcanized rub- 
 
 ber. India Rubber J. 61, 87- 
 90 (1921). 
 
 457. Influence of factors in rubber 
 
 preparation on the aging 
 qualities of vulcanized rub- 
 ber. Arch. Rubbercult. 4, 
 458-72 (1920). 
 A. Wagner 
 
 458. A new modification of Hen- 
 
 riques method for the de- 
 termination of sulfur in 
 rubber. Gummi Ztg. 21, 
 552 (1907). 
 
 459. On the estimation of sulfur in 
 
 rubber. Gummi Ztg. 21, 
 657 (1907). 
 
 460. Estimation of antimony in 
 
 vulcanized rubber. Chem. 
 Ztg. 30, 638 (1906). 
 C. E. Waters and John B. Tut tie 
 
 461. Determination of total sulfur 
 
 in India rubber. J. Ind. Eng. 
 Chem. 3, 734-7 (1911). 
 Jean Wavelet 
 
 462. Observations on the work of 
 
 Stevens, and of Kratz and 
 Flower, on the action of 
 accelerators. Caoutchouc «fe 
 Guttapercha 17, 10143-5 
 (1920). 
 C. O. Weber 
 
 463. On the analysis of India rub- 
 
 ber articles. J. Soc. Chem. 
 Ind. 13, 476-85 (1894). 
 
 464. Concerning the vulcanization 
 
 of rubber. Chem. Ztg. 18, 
 837-8; 1695-6 (1894). 
 
 465. Upon the analysis of rubber 
 
 goods. Chem. Ztg. 18, 
 1003-5 ; 1040-1 ; 1064-9 
 (1894). 
 
 466. On the vulcanization of India 
 
 rubber. J. Soc. Chem. Ind. 
 13, 11-17 (1894); 14, 436-43 
 (1895). 
 
 467. The analysis of vulcanized 
 
 rubber. Z. angew. chem. 11, 
 313 (1898). 
 
 468. The nature of India rubber. 
 
 J. Soc. Chem. Ind. 19, 215- 
 21 (1900). 
 
 469. The determination of sulfur in 
 
 rubber. Gummi Ztg. 17, 
 179-80 (1902). 
 
 470. Analvsis of rubber goods. 
 
 Gummi Ztg. 17, 207-8 (1902). 
 Cf. also Ber. 36, 3103; 310S 
 (1903). 
 
 471. Methods for the determination 
 
 of rubber. Gummi Ztg. 18, 
 461 (1903). 
 
 472. A new method for rubber 
 
 analvsis. Gummi Ztg. 18, 
 339-41; 521-3 (1903). 
 
 473. The use of chloral hydrate in 
 
 rubber analvsis. India Rub- 
 ber J. 25, 375 (1903). 
 
 474. The action of light on India 
 
 rubber. India Rubber J. 25, 
 639 (1903). 
 
 475. The influence of litharge on 
 
 hot vulcanization. Gummi 
 Ztg. 17, 296 (1903). 
 
13S 
 
 THE ANALYSIS OF RUBBER 
 
 Lothar E. Weber 
 
 476. Preparation of rubber mixings 
 
 for analysis. Guinini Ztg. 
 21, 797 (1907). 
 
 477. The action of resins in the vul- 
 
 canizing of rubber. 8th Int. 
 Cong. Appl. Chem. 27, 
 71-2. 
 
 478. Increase of resins in the vul- 
 
 canization. India Rubber 
 World 55, 6 (1916). 
 
 479. The nature and uses of rub- 
 
 ber solvents. India Rubber 
 World 56, 317-8; 377-8 
 (1917). 
 
 480. The use of magnesia in rubber 
 
 compounds. India Rubber 
 World 67, 209 (1918). 
 
 481. An examination of German 
 
 synthetic rubber. India 
 Rubber World 61, 71-2 
 (1919). 
 Weber and Sweet 
 
 482. Antimony sulfide. Caoutchouc 
 
 & Guttapercha 15, 9468 
 (1918). 
 L. G. Wesson 
 
 483. Preliminary note on a new 
 
 method for the direct deter- 
 mination of rubber. J. Ind. 
 Eng. Chem. 5, 398 (1913). 
 
 484. The combustion method for 
 
 the direct determination of 
 rubber. J. Ind. Eng. Chem. 
 6, 459-62 (1914). 
 L. G. Wesson and E. S. Knorr 
 
 485. Wet combustion in the nitro- 
 
 site combustion method for 
 the direct determination of 
 rubber. India Rubber World 
 55, 69 (1916). J. Ind. Eng. 
 Chem. 9, 139-40 (1917). 
 R. Wheatly and B. D. Porritt 
 
 486. A new machine for the prep- 
 
 aration of vulcanized rubber 
 for analysis. J. Soc. Chem. 
 Ind. 34, 587-8 (1915). 
 G. Stafford Whitby 
 
 487. Moisture in crude rubber. J. 
 
 Soc. Chem. Ind. 37, 278-80T 
 (1918). 
 
 488. Variation in Hevea Brazilien- 
 
 sis. Ann. Bot. 31, 312-21 
 (1919). 
 
 489. The caoutchouc molecule. 
 
 India Rubber J. 62, 313 
 (1921). 
 
 G. Si afford Whitby and J. Doolid 
 
 490. Contributions to the knowl- 
 
 edge of the resins of Hevea 
 rubber. Rochester meeting 
 of the American Chemical 
 Society, April, 1921. 
 G. Stafford Whitby and 0. J. 
 Walker 
 
 491. Influence of certain organic 
 
 accelerators on the vulcan- 
 ization of rubber. J. Ind. 
 Eng. Chem. 13, 816-9 (1921). 
 
 492. The influence of piperidine- 
 
 piperidyl dithiocarbamate on 
 vulcanization. Rubber Age 
 9, 455-60 (1921). 
 G. Stafford Whitby and Arnold H. 
 Smith 
 
 493. Rapid and low temperature 
 
 vulcanization. New York 
 meeting of the American 
 Chemical Society, Septem- 
 ber, 1921. 
 Ernst Wleck 
 
 494. Contribution to the investiga- 
 
 tion of golden sulfide ; detec- 
 tion and determination of 
 calcium sulfate. Gummi Ztg. 
 29, 479-80 (1915). 
 Van Lear Woodward 
 
 495. Antimony sulfide; its place in 
 
 tube making. Rubber Age 
 1, 99-100 (1917). 
 Fred E. Wright 
 
 496. Oblique illumination in petro- 
 
 graphic microscope work. 
 Am. J. Sci. 35, 63-82. 
 
 497. The measurement of the re- 
 
 fractive index of a drop of 
 liquid. J. Wash. Acad. 4, 
 269-79; C. A. 8, 2516. 
 
 498. Determination of the relative 
 
 refringence of mineral grains 
 under the petrographic mi- 
 croscope. J. Wash. Acad. 4< 
 389-92; 5, 101-7. 
 
 499. Recent improvements in the 
 
 petrographic microscope. J. 
 Wash. Acad. 6, 465-7 (1916) ; 
 C. A. 10, 2419. 
 
 500. Petrographic microscope in 
 
 analysis. J. Am. Chem. Soc. 
 38, 1647-58 (1916). 
 
 501. Polarized light in the study of 
 
 ores and metals. Proc. Am. 
 Phil. Soc. 58, 401-47 (1919) ; 
 C. A. 14, 1645. 
 
APPENDIX A. 
 The Preparation of Materials for Rubber Manufacture. 
 
 While to many readers, the factory processes concerned with the prep- 
 aration of the materials which enter into the composition of a rubber 
 compound are every day affairs, for the benefit of those who are not 
 familiar with the technical processes used in rubber factories, it seems 
 worth while to give some brief description of the various steps. The crude 
 rubber is prepared by washing, if necessary, and by "milling" or "breaking 
 down"; the pigments may be screened or bolted. Certain pigments and 
 organic accelerators are incorporated with rubber to form master batches. 
 These, together with such pigments and crude rubbers as are used in the 
 form in which they are received, are fed to the compound room, where 
 the batches are weighed out according to the prescribed formulas. These 
 batches are mixed in the mill room, and then stored to cool and age. 
 The mixed rubber compound is prepared for manufacturing purposes by 
 (a) calendaring into sheets or strips; (b) tubing into special or irregular 
 shapes; (c) made into a rubber cement by dissolving the rubber com- 
 pound in benzene, gasoline, or a mixture of the two; (d) applied to fab- 
 ric, either on a calendar, or on a spreading machine. These four classes 
 may be called the "intermediates" of the rubber industry; from these, 
 practically all rubber articles are built. 
 
 Washing. The wild rubbers, and the inferior grades of plantation rubber, 
 contain large amounts of bark, dirt, and other foreign matter, and in this 
 condition are not suitable for rubber manufacture. These impurities are 
 removed by washing. The better grades of plantation rubber may be 
 washed in order to be certain that no grit or dirt remains in the rubber, 
 to cause trouble during the service of the articles of which they may form 
 a part. 
 
 The washing equipment consists of a vat for heating the rubber, and 
 three washing mills — a cracker, a refiner, and a finisher. Some factories 
 combine the refining and finishing on one mill. 
 
 Some of the wild rubbers, notably the fine Para sorts, come in large 
 lumps, or biscuits. Before they are ready for washing, they must be boiled 
 in water for several hours in order to soften them up sufficiently to be 
 worked easily. The large lumps are cut up before being heated. 
 
 The cracker consists of two rolls with heavy, coarse corrugations. The 
 rubber is passed through this mill two or three times, just enough to form 
 a thick, rough sheet. 
 
 The refiner is a two-roll mill, the front roll having finer corrugations 
 than the cracker. A stream of water plays on the rubber as it passes 
 between the rolls, and as fresh surfaces of the rubber are exposed, the 
 water washes out the impurities. 
 
 The finisher is a mill similar to the refiner, but having even smaller 
 corrugations on the front roll. Here the rubber is sheeted out thin, ready 
 for drying. 
 
 Drying. Two processes are used for drying: "air-drying" and "vacuum 
 drying." 
 
 In the air-drying, the lone sheets from the washing mills are hung over 
 poles in a drying room, the temperature of which is maintained at 90-110F. 
 
 139 
 
140 THE ANALYSIS OF RUBBER 
 
 Fresh air is kept circulating through the room, so as to facilitate drying. 
 In the plantation rubbers, the moisture is usually on the surface only, and 
 such rubbers can be dried in from two to four days. Many of the wild 
 rubbers contain a large amount of moisture, and take up a further quan- 
 tity during the boiling and washing. This moisture is distributed through- 
 out the rubber, and drying takes from three to five weeks. 
 
 In vacuum drying, the rubber is laid on pans, which are placed in a 
 steam-heated vacuum oven. The process requires only a few hours, but 
 it is not particularly well adapted for drying the wild rubbers, since the 
 rapid evaporation tends to form a hard surface film, which hinders further 
 evaporation, prolonging the drying to such a degree as to materially soften 
 and injure the rubber. Soft wild rubbers, such as Guayule, Pontaniak, 
 and the Africans, flow so readily at ordinary temperatures that they can- 
 not be dried on poles, and must be dried in pans, either in the drying 
 rooms or in the vacuum dryers. 
 
 Milling, or Breaking Down. Crude rubber varies considerably in quality 
 and in order to lessen the variation, lots of rubber are averaged by the 
 process of "milling" or "breaking down." For this purpose, we use the 
 regular mixing mill, consisting of two smooth rolls, the rear one having 
 a higher circumferential speed than the front one. The rolls are hollow, 
 so as to permit them to be heated, or cooled, in order that they may be 
 maintained at a temperature of about 175F. After the mills have been 
 running a short time, the friction of the rubber is more than sufficient to 
 maintain this temperature, and cold water is kept running through them 
 to prevent them getting so hot as to injure the rubber. 
 
 The rubber is fed into the mill, and as it softens it adheres to the front 
 roll. It is cut off, rolled up, and fed back into the mill until it is 
 homogeneous. By this time, it has attained a soft, wax-like consistency. 
 When sufficient uniformity has been attained, the rubber is cut off in the 
 form of sheets, or rolls. 
 
 This "milling" not only produces a uniform grade of rubber, but it 
 breaks up the hard, tough particles of crude rubber, and prepares it for the 
 following operations of mixing, calendaring, and tubing. 
 
 Master Batches. Some pigments, and many of the organic accelerators, 
 are first mixed with rubber, in what are known as "master batches," before 
 they are sent to the compounding room (where the batches for making 
 the various rubber compounds are weighed out). Pigments such as glue, 
 which requires a longer mixing than can safely be given the entire com- 
 pound, are first mixed with rubber. Glue, for example, is usually mixed 
 in the proportion of two or three parts of rubber to one of glue. Dusty 
 pigments, such as gas black, are treated in a similar way, frequently in a 
 building separated from the regular mixing room. This step is necessary 
 to prevent the black dust settling down on the other mills, and injuring 
 the colors of the other batches. Organic accelerators are used in small 
 amounts, sometimes 0.2% or less. It is highly important that these small 
 amounts be very evenly distributed throughout the entire compound. 
 After these accelerators have been thoroughly incorporated into a master 
 batch, it is very simple to distribute them throughout a compound. A 
 further advantage is found in the improved accuracy of weighing out the 
 small amounts needed, and elimination of losses of the accelerator during 
 the mixing. Small errors or losses in getting the accelerator into a com- 
 pound will produce more variation in the finished product than in any 
 other single ingredient. 
 
 Screening, or Boiling. Foreign matter, such as wood. grit, pieces of 
 paper or twine from the containers in which the pigments are shipped, are 
 objectionable in many of the better grades of rubber articles. Such 
 impurities are removed from the pigments by sifting the latter through 
 screens of from 40 to 90 mesh, the size depending largely upon the pig- 
 ment, and the article in which it is to be used. 
 
APPENDIX 141 
 
 Compounding. The compounding consists simply in weighing out the 
 proper amounts of pigments and rubbers. These are placed in large iron 
 boxes or pans, and the total weight checked up before leaving the com- 
 pound room. The usual tolerance in weighing is 0.5 lb. over or under, in 
 a 100 lb. batch, or a total variation of 1%. 
 
 Mixing. The mixing mills used for mixing rubber compounds vary from 
 36 to 84 inches in length, and 20 to 24 inches in diameter. The amount 
 of stock that can be mixed on a mill is largely a function of the width 
 and diameter of the front roll of the mill, and the specific gravity of the 
 stocks; these batches will range from 25 to 225 lbs. 
 
 The rubber is first thrown on the mill, together with any pigments put 
 up in master batches, 1 reclaimed rubber, and mineral rubber. After the 
 rubber softens, the pigments and oils are added. Any material which 
 drops between the rolls is caught in a pan, and returned to the rolls until 
 everything has been incorporated. The mixing is continued until the com- 
 pound is homogeneous, after which it is cut off from the rolls in slabs of 
 about one quarter to three-eighths of an inch in thickness, and weighing 
 15 to 25 lbs. each. These slabs are laid on racks to cool, after which they 
 are sent to storage bins to age for 24 to 48 hours. 
 
 Calendaring. A calendar consists of three smooth hollow steel rolls, 
 accurately ground, arranged vertically, the top and bottom rolls revolving 
 in the opposite direction to the middle roll. The stock, which has been 
 previously softened or "warmed-up" on the regular mixing mill, is fed 
 into the calendar between the top and middle rolls, passes around the 
 middle and under the bottom roll, after which it is wrapped up between 
 cotton liners. The function of these liners is to keep the fresh surfaces 
 of the rubber apart, and protect them from dust and dirt. The thickness 
 of the sheet is controlled by the distance between the top and middle 
 rolls; the width by adjustable knives placed against the rear of the middle 
 rolls. The middle and bottom rolls are further apart than the thickness 
 of the rubber sheet. The bottom roll is cooled, which serves to cool 
 the rubber, and toughen the sheet, before it is wound up in the liner. 
 
 The rubber sheet may be made of the desired thickness in one passing 
 through the calendar, or it may be built up from a number of thinner 
 sheets. 
 
 Calendaring Rubber to Fabric. When a rubber compound is to be 
 applied to a roll of fabric, the rubber is fed into the calendar as above, and 
 the fabric is fed from the opposite side, between the middle and bottom 
 rolls. The first coat is put on by having the middle roll rotate at a faster 
 speed than the bottom one. giving a grinding action which forces the 
 rubber into the meshes of the fabric. This is called "frictioning." If a 
 further layer of rubber is desired over the friction coat, the rubber and 
 fabric are fed in as before, but the middle and bottom rolls rotate at the 
 same speed, and the sheet of rubber is thus laid on the fabric, and pressed 
 
 1 We have here an added advantage in using organic accelerators in the form 
 of master batches. The accelerator is uniformly distributed throughout 
 the rubber before any of the sulfur is added, thus eliminating largely the 
 danger of partial vulcanization on the mixing mill (called burning, or scorch- 
 ing). It is a well established fact that high concentrations of many of the 
 organic accelerators greatly increase the reaction of vulcanization, especially 
 at the lower temperatures existing on the mixing mill. A compound containing 
 0.50% of dimethyldithiocarbnmato burned in less than one minute after the 
 sulfur has been added, whereas a similar stock, having only 0.05%, mixed with- 
 out difficulty. Andres (Caoutchouc & Guttapercha 18, 11089-97 [1921]), showed 
 that 2.5% of thiocarbanilide gave the best cure at 50 minutes, whereas 5% 
 gave a good cure in 3 minutes, and the best cure in 5 minutes. These higher 
 concentrations are easily obtained in a poorly mixed stock, and obviously in- 
 crease the probability of damage to the stock. 
 
142 THE ANALYSIS OF RUBBER 
 
 together just sufficiently for them to adhere firmly. This coating of rubber 
 is called the "skim" coat; tire fabric, for example, is usually frictioned on 
 both sides, and skim-coated on one side only. After the skim has been 
 applied the fabric is rolled up between a liner, to be cut up later as desired. 
 
 Tubing. The tubing machine is used for irregular strips of rubber, such 
 as tire treads, tire beads, rubber tubing, and the insulation on wire. The 
 essential parts of a tubing machine are a hopper for feeding in the stock, 
 the barrel, which can be heated by steam, and which contains the screw 
 for carrying the stock to the head, and the head, containing the die through 
 which the stock is forced. The head is often heated with a gas jet, to 
 prevent the tubed rubber cooling in the die and coming through rough or 
 cracked. The rubber is warmed up to the desired consistency, cut into 
 small strips, and fed into the hopper. The stock as it comes from the 
 tubing machine is cut in the desired lengths, placed between liners, and 
 set aside to cool. 
 
 Cement. In the manufacture of cement, the rubber is warmed up on 
 the mixing mill, and cut off in very thin sheets. A weighed amount of 
 this rubber is cut into small pieces, and thrown into a churn or mixer, 
 to which has already been added the measured amount of solvent. The 
 contents of the mixer are stirred until solution is complete, a matter of 
 from 4 to 12 hours, depending largely upon the nature of the 
 solvent, the grade of rubber, and the efficiency of the mixer. The first 
 mixing usually gives a heavier cement than desired, and this is thinned 
 down with more solvent until the right viscosity is obtained. The nature 
 of the service for which the cement is intended dictates the degree of 
 viscosity. 
 
 Spreading. In addition to the method of applying rubber to fabric by 
 calendaring, as described above, we may use the process known as "spread- 
 ing." A spreader consists of a rubber coated roll, against which rests a 
 heavy knife. Beyond the rubber roll are steam-heated coils or plates, 
 about 18 to 30 feet in length. The rubber compound is first made into 
 a very heavy cement (generally called "dough"). The fabric is passed 
 between the rubber-coated roll and the knife, the dough is applied to the 
 fabric just before the latter reaches the knife, and in passing between the 
 roll and the knife the latter scrapes off all but a thin coating of the cement. 
 As the fabric passes over the heated plates, the solvent in the cement 
 evaporates, and leaves a thin coating of rubber in very intimate contact 
 with the fabric. The space between the knife and roll controls the amount 
 of rubber left upon the fabric. The amount of rubber which may be added 
 at one passing depends upon the ability of the spreader to drive off the 
 solvent during the time when the fabric is passing over the heated rolls. 
 The factors are the temperature and length of the drying plates, and the 
 speed at which the machine is driven. A heavy coating of rubber is ob- 
 tained by passing the fabric through until the desired quantity of rubber 
 has been applied. 
 
APPENDIX B. 
 Physical Tests. 
 
 The chemist in the rubber factory is usually given the duty of making 
 whatever physical tests may be necessary to determine the properties of 
 the rubber compound or finished article. Similarly, the chemist in the 
 consumer's laboratory supervises and interprets the results of the physical 
 tests made upon samples taken from deliveries of manufactured goods. 
 It seems desirable, therefore, to point out what physical tests are usually 
 made, and their relation to the quality and life of the material. 
 
 The principal physical tests are (1) tensile strength, (2) ultimate elonga- 
 tion, (3) set at break, (4) friction. These tests are usually made on the 
 same testing machine. The tensile st length is the force required to break 
 a unit area of a rubber compound: the ultimate elongation is the extcn! 
 to which the rubber can be stretched before it will break; the set at 
 break is the increase in length of a measured length of rubber, taken at 
 some definite time after break; and the friction is the force required to 
 separate a rubber compound from a piece of fabric to which it has been 
 vulcanized. 
 
 Tensile Testing Machine. Three types of machines arc in more or less 
 common use in this country: (a) Scott; (b) Bureau of Standards; (c) 
 Schopper. 
 
 (a) The Scott is a machine of the dead weight type, the pull being 
 against a lever which moves outward as the tension is applied. There 
 are two clamps, into which are inserted the ends of the rubber test pieces. 
 The upper clamp is attached to the end of the weighted lever; the lower 
 clamp is attached to a rod driven at a uniform rate of speed (usually 20 
 inches per minute for tensile tests, and 2 inches per minute for friction 
 tests). The lever carries a set of pawls, which engage in the teeth of a 
 curved rack, preventing the lever from falling back when the tension is 
 released (as for example, when the test piece breaks). The tension is read 
 off from a dial, the indicator being actuated by the motion of the lever. 
 
 (6) The Bureau of Standards machine 1 differs from the Scott in that 
 the pull is against a spring balance, which directly records the pull. The 
 clamps are of the same type, and the operation of the machine is essen- 
 tially the same as that of the Scott. 
 
 (c) The Schopper machine is one of the dead weight type. The rack, 
 over w r hich the lever moves, is graduated, and the tension is read off oppo- 
 site the point where the lever stops. 
 
 There are a number of styles of clamps which may be used with these 
 machines, the principal ones being the eccentric grip, with its modification 
 consisting of a number of thin disks, mounted eccentrically, the zig-zag 
 grip, which is tightened by a screw, and the spool grips, for use with 
 ring-shaped test pieces. Any of these types may be used with any of the 
 machines mentioned, but the Scott and the Bureau of Standards machines 
 usually carry the eccentric grip clamps, for testing bar-shaped test pieces, 
 whereas the Schopper usually has only the spool grips. As long as the 
 Schopper is equipped only with grips for testing ring-shaped test pieces, 
 
 1 Cf. Bureau of Standards Circular 38, Fourth Edition, p. 53. 
 
 X43 
 
144 THE ANALYSIS OF RUBBER 
 
 it cannot be considered equivalent to the other machines. Between the 
 Scott and the Bureau of Standards machines there is little choice to be 
 made — providing they are both accurately calibrated, and the clamps sep- 
 arated at the same rate of speed, comparable results may be obtained. 
 The dead weight type is usually considered to be the more rugged, and 
 less likely to get out of order, than the spring balance. 
 
 When using the ring-shaped test pieces, any one of the three machines 
 may be used without affecting the results. In fact, the type of machine 
 is of importance, not so much for the accuracy of the determinations which 
 it will give, but from the point of view as to how it will stand up under 
 the service given to it, and the convenience of operation. 
 
 Shape of Test Pieces. The test piece commonly used in this country for 
 the determination of tensile strength, is the "bar-shaped" or "dumb-bell" 
 test piece. The constricted part is either ~hk or %" and 1 or 2" long. The 
 ends are enlarged to reduce to the minimum the danger of the test piece 
 tearing in the clamps. The enlarged ends are 1" wide for the V±" width, 
 and 1^4" wide for the %" width. A few use a V-k" width at the ends for 
 a Vi" width at the constricted part, particularly for testing compounds of 
 high rubber content (the so-called pure gum compounds). The style of 
 test piece is largely a matter of the operator's choice, influenced in pait 
 by the nature of the material to be tested. Some specifications define 
 exactly the shape, leaving nothing to the discretion of the operator. While 
 theoretically there should be no difference, as a matter of practice results 
 are comparable only when the same shaped test piece is used. 
 
 In the above, nothing has been said regarding the thickness of the test 
 piece. Except when slabs are prepared particularly for the purpose of 
 making tensile tests, this is not a matter which can be controlled easily, 
 but the thickness is quite likely to be an important factor, the thicker 
 pieces showing a greater tendency to tear, and hence giving lower results 
 than would be obtained from thinner ones. The most satisfactory prac- 
 tical range is from % to 3/16" (0.125 to 0.183"). 
 
 The ring-shaped test piece cannot be compared with the b;u-.<haped test 
 piece* Its only advantage lies in the fact that with it an autographic 
 chart may be made of the stress-strain curve. With the bar-shaped test 
 piece, to get the same data, it is necessary to use two operators, and from 
 their observations plot the stress-strain curve. 
 
 The' principal precautions to be taken in preparing test pieces of any 
 shape are that the edges be cut evenly and that the opposite sides of 
 the constricted part are parallel. With ring-shaped test pieces, the rings 
 must be very accurately centered, so as to obtain the same cross-section 
 at all points. If the top and bottom surfaces of the test pieces are not 
 smooth, they should be made so by buffing, so that accurate readings of 
 thickness may be made. 8 
 
 Tensile Strength. The tensile strength is usually expressed in pounds 
 per square inch, or kilograms per square centimeter. 4 The area is usually 
 referred to the cross-section at rest. However, before rubber can be 
 broken, it must be stretched from 300 to 900%, and since there is no 
 change in volume " during the stretching, the cross-section at break is very 
 much less than when the test piece is at rest. For this reason, the tensile 
 
 •This point is argued very convincingly in the Bureau of Standards Circular 
 38, Fourth Edition, p. 66, etc. 
 
 •A buffing machine, suitable for the purpose, is described in Bureau of Stand- 
 ards Circular No. 38, Fourth Edition, p. 48. 
 
 *To convert lbs. / sq. in. into kg. / sq. cm., multiply by 0.07031; to convert 
 kg. / sq. cm. to lbs. / sq. in., multiply by 14.222. 
 
 • Schippel's change in volume on stretching refers only to the vacua formed 
 around coarse particles of pigment. Such changes are negligible for the calcu- 
 lations under discussion. 
 
APPENDIX 145 
 
 strength has sometimes been referred to the cross-section at break, called 
 the "tensile product." This figure is obtained by multiplying the tensile 
 strength by the elongation at break. 
 
 The tensile strength is appreciably affected by a considerable number 
 of factors, some of which are within the control of the operator and some 
 are not. Of these, the most important are: rate of separation of the jaws, 
 temperature, size and shape of the test pieces, the direction of the cut 
 (i.e., whether along the length of a calendared sheet, or across), previous 
 stretching of the rubber, and the age of the rubber compound. These 
 factors have been discussed at some length by Whitby, 8 and the Bureau 
 of Standards, 7 and their conclusions may be briefly summarized as 
 follows: 
 
 Rate of Separation of the Jaws. The higher the speed, the higher will 
 be the results for tensile strength and ultimate elongation. The range in 
 speeds between 5" and 45" per minute may affect the results anywhere 
 from 5% to 20%. The speed 'generally employed is 20" per minute. 
 
 Temperature. The temperature at which tests are usually expected to 
 be made is 70F. Increasing the temperature lowers the tensile strength 
 and increases the elongation; lowering the temperature produces a reverse 
 effect. It is worthy of notice that for the range of temperature from 50F 
 to 90F. the tensile at break is much more constant than either the tensile 
 strength or ultimate elongation. 
 
 Size and Shape of the Test Pieces. There is a tendency for narrow test 
 pieces to develop higher values than wider ones; between Yi" and %", 
 differences as high as 20% have been noted. Unpublished data may con- 
 tain instances of even greater variation. 
 
 Direction of Cutting. The Bureau of Standards found that the tests 
 made on samples cut in the direction of calendaring show a higher ten- 
 sile and lower elongation than those cut in the transverse direction. 
 Some experimenters have not been able to duplicate these results, but 
 most of the data on the subject indicate that there is a decided difference 
 between the two directions in a calendared sheet. The ring-shaped test 
 pieces include rubber cut in all directions, and since the break occurs in 
 the direction of least resistance, it is obvious that the effects of calendaring 
 cannot be detected with such test pieces. 
 
 Previous Stretching of the Rubber Test Pieces. It is curious that while 
 Memmler and Schob, and the Bureau of Standards, agree that previous 
 stretching alters the results of tensile tests, the former obtained lower 
 results from test pieces subjected to previous stretching, whereas the latter 
 obtained higher figures. Memmler and Schob tested ring-shaped test 
 pieces by subjecting them to 50% of their normal breaking load for a 
 period of 30 minutes, and testing them after a rest period of 24 hours. 
 The Bureau of Standards employed bar-shaped test pieces, stretching to 
 200%, releasing, and then continuing with an increase of 100% until failure 
 ensued. With high grade material, Memmler and Schob found differences 
 of about 35% loss in tensile strength ; the Bureau of Standards found in- 
 creases of about 20%. Apparently what is needed to determine the exact 
 difference caused by previous stretching, is to combine the two sets of 
 ideas. The Bureau of Standards figures seem to show that short periods 
 of stretching increase the tensile strength. By increasing the time of 
 stressing the test pieces, we could find out whether or not there is a point 
 at which there is no further increase in tensile properties. Similar experi- 
 ments could be made with ring-shaped test pieces, for it is not at all im- 
 possible that the differences may be largely attributed to the differences 
 in the shape of the test pieces. 
 The greatest importance of these experiments lies in the fact that they 
 
 6 Whitby, Plantation rubber and the testing of rubber. 
 
 7 Bureau of Standards Circular 38, Fourth Edition. 
 
146 THE ANALYSIS OF RUBBER 
 
 emphasize the necessity for permitting rubber samples to age for at least 
 48 hours, in order to be certain that they have reached equilibrium. 
 
 Aging of the Rubber. Practically every one who has followed the test- 
 ing of rubber is agreed that a certain time is required after vulcanization 
 for the rubber to come to equilibrium. To be absolutely safe, many have 
 placed the period for aging at 3 days; others think that as little as 24 
 hours will suffice. In view of the results obtained in a study of the effects 
 of previous stretching, 24 hours seems hardly enough to be on the safe 
 side, and consequently a minimum of 48 hours is recommended. 11 
 
 Ultimate Elongation. The ultimate elongation has been defined as being 
 the extent to which a rubber compound may be extended before rupture 
 will occur. With bar test pieces, the elongation is determined by placing; 
 on the constricted portion of the test piece, parallel lines either 1" or 2" 
 apart, and then stretching until it breaks. The distance between the 
 marks at break (a) less the original distance (b) is. the elongation (c), 
 and is expressed in percentage. Some prefer to express the ultimate 
 elongation by dividing a by b, giving figures which are 100% higher than 
 the more commonly accepted figures. There is nothing gained by this 
 procedure, and it causes a great deal of confusion when making compari- 
 sons. In order to avoid this, many specifications are now being written 
 calling for an elongation in definite figures, such as from 1-5 inches, or 6 
 inches, or whatever length may be desired. 
 
 For correctly cured soft vulcanized rubber, the ultimate elongation is 
 affected most by the amount and grade of rubber present. Compounds 
 containing 90% of rubber will have an elongat'on of about 900% to 
 1000%, while compounds containing only 30% will have an elongation of 
 only 300% to 500%. Just as in the case of tensile strength, we find that the 
 ultimate elongation is a more or less arbitrary figure, the value of which 
 will depend to a considerable degree upon the manner of its determina- 
 tion. Practically all of the factors which influence the values for tensile 
 strength will be found to have an effect on those for elongation. 
 
 Stress-Strain Curves. If the tensile strength for each increment of elonga- 
 tion be determined, and plotted, the line drawn through these points 
 gives us what is known as the "stress-strain" curve. Generally the strains 
 are plotted as ordinates, and the stresses as abscissae. The Scott and 
 Schopper machines plot these curves autographically when ring-shaped test 
 pieces are used. With bar test pieces, one operator reads the strains, and 
 the other the stresses. By plotting the curves of a series of cures on one 
 sheet, the effect of time of vulcanization, or whatever other factor it is 
 desired to follow, may be easily observed. 
 
 The principal trouble with the stress-strain curve for rubber, is that in 
 making the curves of a series of cures, the first half or three-fourths of 
 the curves take practically the same course, and it is difficult, if not impos- 
 sible, to notice any appreciable difference until the last quarter of the 
 curve. The Goodyear laboratory has suggested a means for plotting the 
 results in such a fashion as to bring out differences in the early parts 
 of the curves. They plot stresses as ordinates, and the time of cure as 
 abscissae; for each time of cure, they plot the tensile strength for 100% 
 and each succeeding 100% elongation up to the break. Curves are drawn 
 through all points having the same elongation, and the final curve is drawn 
 
 • It is obvious that at times these precautions must yield to expediency, and 
 it is frequently more important in manufacturing work to get immediate results 
 which are approximately accurate, than to wait 48 or even 24 hours. We have 
 frequently taken slabs of rubber out of a mold, cooled them in running water for 
 15 minutes, and then proceeded with the tests. In all such cases, the proba- 
 bility of large errors being present was known and appreciated. Those samples 
 which showed any promise of being satisfactory were given the regular tests 
 48 hours later, and the latter figures only were used for record and comparison. 
 
APPENDIX 147 
 
 through the points of ultimate elongation. The latter is the usual "tensile- 
 strength-time-of-cure" curve used so much by investigators in this coun- 
 try. This system of plotting gives a much more satisfactory picture than 
 does the ordinary stress-strain curve. 
 
 Set at Break. After the test piece has been broken on the testing 
 machine it is laid aside for a period which ranges from 1 to 24 hours, 
 according to the methods adopted by the various laboratories, and the 
 increase in the distance between the marks is measured, and calculated 
 to percentage. The set at break for various cures of the same compound 
 passes through a maximum at the optimum cure, the shape of the curve 
 as plotted against time of cure very much resembles the tensile strength- 
 time-of-cure curve. Very little practical use is made of this determination. 
 
 A far more extensive use of the determination of set has been made 
 by determining the set on test pieces which have been stretched to less 
 than their ultimate elongation. The usual routine in such tests is to 
 stretch the test piece for ten minutes" and measure the increase in elonga- 
 tion ten minutes after releasing. With such tests, there is a drop in the 
 value of the set from an undercure to an overcure, the effect being most 
 noticeable in the former. 
 
 Friction. The adhesion between fabric and rubber is termed "friction." 
 There are two methods for its determination: (a) the amount of separa- 
 tion under definite load; (b) the load required to separate rubber and 
 fabric at a definite rate. 
 
 The first method is much employed in testing mechanical goods. In 
 testing belting, for example, a test piece is cut one inch wide and about 
 six inches long. Two plies of fabric are separated; the end of one ply is 
 mounted in a rigid position, while to the other ply is attached a weighted 
 clamp. A mark is made where the test is to start, and the weighted clamp 
 is then released. After a fixed time (generally 10 minutes), the amount 
 of separation is measured. This test merely gives a minimum value, and 
 does not measure the true adhesion. 
 
 In the second method, one ply is fastened to the upper clamp of a test- 
 ing machine, and the other ply to the lower one. The clamps are now 
 separated at a uniform rate, usually 2" per minute, and an auto- 
 graphic record made of the pull required to separate the two plies. This 
 method not only shows the maximum strength of the adhesion, but gives 
 the variation over the area tested, thus revealing any lack of uniformity. 
 
 Ordinarily, the adhesion between two rubber plies cannot be tested in 
 this manner, since the joint is usually stronger than either of the two 
 compounds. However, the second method is available for testing adhesions 
 such as are found in the acid cured splice of an inner tube, or between a 
 hard rubber and a soft rubber compound. 
 
 Heat Aging Tests. All of the tests described above are performed on 
 test pieces in the condition as they come from the vulcanizers. Such tests 
 give little, if any, idea of what the physical properties of the compound 
 will be at some future time. Artificial means for aging have been sug- 
 gested, but probably the most widely used is the one developed by W. C. 
 Geer. 10 The test pieces are maintained at a constant temperature of 70C, 
 and tests made at regular intervals until deterioration sets in. The most 
 serious fault with this method is that the heating takes place in air, so 
 that in addition to the deterioration caused by heating, we have the effect 
 of the oxygen of the air. In a great many cases, such as tire frictions, 
 breaker stocks, cushion stocks, etc., heat is undoubtedly the principal 
 agency in deterioration. In such cases, the heating should be done in an 
 
 9 Cf. Bureau of Standards Circular 38, Fourth Edition, pages 57-8, for a 
 description of a convenient form of apparatus for making these tests. 
 
 10 W. C. Geer, India Rubber World 55. 127-30 (1916) : W. C. Gecr and W. W. 
 Evans, India Rubber World 6h 887-92 (1921). 
 
148 THE ANALYSIS OF RUBBER 
 
 atmosphere of gas free from oxygen, or any gas which may have a tendency 
 to react with rubber. 
 
 The most which can be claimed for this test is that it is useful in com- 
 paring compounds of about the same type, or of cures of the same com- 
 pound. With oxygen excluded, it might be extended to include compounds 
 of different types, which are likely to be exposed to the same degree of 
 heat. Beyond this, we have not succeeded in developing anj'thing reliable 
 in the way of accelerated aging. 
 
APPENDIX d. 
 
 Table of Specific Gravities. 
 
 Minimum Maximum 
 
 Acetone 0.797 
 
 Aluminium silicate 2.61 3.02 
 
 Aluminum flake 2.56 2.65 
 
 Ammonium carbonate 1.50 1.60 
 
 Aniline 1.00 1.03 
 
 Antimony, red 2.87 
 
 golden 2.57 2.90 
 
 " crimson 3.11 4.20 
 
 " black 4.80 
 
 Arsenic yellow 2.75 
 
 Asbestine 2.60 2.82 
 
 Asphalt 0.99 
 
 liquid 0.99 
 
 Trinidad 1.20 
 
 Balata 1.05 
 
 Barytes (blanc fixe) 4.20 4.92 
 
 Beeswax 0.97 
 
 Benzene 0.745 
 
 Bitumen 1.07 1.16 
 
 Black substitute 1.10 
 
 Bone black 2.20 2.32 
 
 Brown substitute 1.07 1.32 
 
 Burgundy pitch 1.10 
 
 Camphene 0.865 
 
 Candelilla wax 0.99 
 
 Carbon bisulfide 1.26 1.29 
 
 Carbon black 1.68 1.89 
 
 Carbon tetrachloride 1.61 
 
 Carnauba wax 0.995 
 
 Castor oil 0.958 
 
 Ceresin 0.918 0.922 
 
 Chloroform 1.52 
 
 Chrome yellow, light 6.41 
 
 " " medium 5.73 5.84 
 
 " " deep 5.91 6.08 
 
 Chrome green 524 5.44 
 
 Clay, China 2.60 
 
 Blue Ridge 2.55 
 
 " Dixie 2.58 
 
 Coal tar 1.05 1.27 
 
 Cork 0.24 1.00 
 
 Cork dust 1.16 
 
 Corn oil 0.926 0.930 
 
 Cotton 1.45 1.55 
 
 Cottonseed oil 0.922 0.93 
 
 Dimethylaniline 0.958 
 
 149 
 
150 THE ANALYSIS OF RUBBER 
 
 Minimum Maximum 
 
 Dipheii3'lamine 1.16 
 
 Fossil flour 2.00 2.60 
 
 Fuller's earth 1.80 2.70 
 
 Gasoline, 72-75 Be 0.700 0.707 
 
 Glass, powdered 2.49 
 
 Glue 1.30 
 
 Glycerin 1.25 1.30 
 
 Graphite 1.95 2.40 
 
 Guttapercha 0.96 1.00 
 
 Hexamethylenetetramine 1.25 
 
 Indian red 4.80 525 
 
 Infusorial earth 1.66 1.95 
 
 Kaolin 2.75 
 
 Lampblack 1.53 1.75 
 
 Lead, chromate 5.65 6.12 
 
 oleate 1.50 
 
 " red 8.17 
 
 " sublimed blue 6.40 
 
 " " white 6.20 6.30 
 
 white 6.10 6.75 
 
 Leather fiber 1.40 
 
 Lime 2.21 2.40 
 
 Linseed oil 0.94 
 
 Litharge 8.90 9.52 
 
 Lithopone 3.60 4.25 
 
 Magnesia 2.16 3.65 
 
 Magnesium carbonate, light 1.74 2.22 
 
 heavy 3.00 3.07 
 
 oxide 3.20 
 
 Mica 2.80 3.20 
 
 Mineral rubber 1.00 1.06 
 
 Montan wax 1.04 
 
 Ochre 3.50 
 
 Ozocerite 0.90 0.95 
 
 Palm oil 0.94 
 
 Paraffin 0.869 0.91 
 
 oil 0.90 
 
 wax 0.91 
 
 Petrolatum 0.90 
 
 Pine tar 1.05 
 
 Pitch 1.23 1.28 
 
 Prussian blue 196 
 
 Red oxide 4.82 5.16 
 
 Rosin 1.05 1.08 
 
 Rosin oil 0.98 1.10 
 
 Rubber, Accra flake 1.02 
 
 " Amber crepe 0.92 
 
 Assam 0.967 
 
 Benguella 0.928 
 
 Borneo 0.916 
 
 " Cameroon 0.929 
 
 " Cameta 0.916 
 
 Caucho ball 0.915 
 
 Centrals 0.93 
 
 Congo 0.93 
 
 Guayule 0.975 
 
 extracted 0.995 
 
 " Madagascar 0.915 
 
APPENDIX 151 
 
 Minimum Maximum 
 
 Rubber, Manicoba 0.93 
 
 " Mozambique 0.939 
 
 Niger flake 0.93 
 
 " Para, coarse 0.95 
 
 " fine 0.94 
 
 Penang 0.918 
 
 Pontianak 0.99 
 
 " Roll brown crepe 0.95 
 
 Senegal 0.929 
 
 Sernamby 0.918 
 
 Sierra Leone 0.923 
 
 Singapore 0.937 
 
 Smoked sheets 0.91 0.95 
 
 " West Indies 0.935 
 
 Starch 1.50 
 
 Sulfur 1.96 2.07 
 
 Sulfur chloride 1.69 1.17 
 
 Talc 2.00 2.78 
 
 Thiocarbanilide 1.30 
 
 Tripoli 1.95 2.25 
 
 Ultramarine 2.30 2.40 
 
 Vaseline 0.84 0.945 
 
 Venetian red 1.96 2.07 
 
 Vermilion 7.89 8.10 
 
 Wax tailings 1.00 1.08 
 
 White substitute 1.04 1.14 
 
 Whiting 2.60 2.72 
 
 Wood pulp 1.43 1.46 
 
 Yellow ochre 3.50 5.00 
 
 Zinc, carbonate 4.42 4.45 
 
 " oxide 5.38 5.60 
 
 leaded 5.64 
 
 " sulfate 3.62 
 
 " sulfide 3.50 
 
SUBJECT INDEX. 
 
 Note. Figures in Roman refer to pages ; figures in Italic refer 
 to the article listed in the bibliography. 
 
 Accelerators, inorganic, 42, 106, 200, 
 4<)6, 408. 
 
 Accelerators, organic, 38, 94, 96, 15, 
 16, 41, 42, 43, 60, 78, 112, ISO, 
 133, 138, 140, 141, 142, m, W, 
 149, 150, 155, 195, 255, 256, 271, 
 286, 287, 288, 289, 305, 323, 347, 
 373, 386, 387, 403, 406, 4O8, 409, 
 429, 430, 433, 434, 485, 462, 491, 
 492, 493. 
 
 Accelerators, ultra-rapid, 41, 59. 
 
 Aging tests, 146, 147, 17, 59, 188, 
 189, 357, 412, 456, 457. 
 
 Aldehyde ammonia, 40. 
 
 Aldehyde aniline, 39. 
 
 Aluminium, 119. 
 
 Aluminum flake, 45. 
 
 Amber crepe, 20. 
 
 Ammonium carbonate, 46. 
 
 Analysis, Interpretation of, 4%1- 
 
 Aniline (aniline oil), 38, 95, 358. 
 
 Antimony, Crimson, 47, ^7. 
 
 Antimony, Determination of, 102, 
 119, 10, 18, 75, 76, 172, 173, 174, 
 254, 299, 345, 355, 356, 365, 368, 
 371, 481, 460. 
 
 Antimony, Golden, 49, 7, 377, 378, 
 443, 482, 494, 495. 
 
 Asbestine, 46. 
 
 Ash determinations, 83. 
 
 Balloon fabrics, 24, 27, 28, 184, 154, 
 
 836, 419. 
 Barium carbonate, 105, 418. 
 Barium, Determination of, 104 119 
 
 418. 
 Barytes (barium sulfate), 51. 
 Bitumens (see mineral rubber). 
 Blanc fixe (see barytes). 
 Bolting, 140. 
 
 Breaking-down rubber, 140. 
 Brown pigments, 47. 
 Burgundy pitch, 34. 
 
 Calcium carbonate (see whiting). 
 Calcium sulfate, 47, 49, 119, 494. 
 
 Calendaring, 141. 
 
 Carbonates, Determination of, 101, 
 
 264. 
 Carbon black (see gas black). 
 Castilloa, 17. 
 Caucho, 17. 
 Cellulose, Determination of, 158 
 
 219. 
 Cements, 66, 142. 
 Centrals, 17. 
 Ceresin, 32, 55. 
 Chinese blue, 47. 
 Chromates, 108. 
 Coagulation, 16. 
 Coefficient of vulcanization, 25 
 
 401, 454, 455. 
 Compounding, 141. 
 Compounds, Preparation of rubber, 
 
 26 
 Copper, 43, 99, 171, 181. 
 Cottonseed oil, 34. 
 Cure, Definition of, 14. 
 Cure tests, Formulas for, 25. 
 Curing tests, 24. 
 
 Dimethylaniline, 39. 
 Diphenylamine, 39, 96. 
 Diphenylcarboimide, 40. 
 Diphenylguanidine, 40. 
 Dithiocarbamates, 41. 
 Drying, 139. 
 Dyes, 47, 88, 111. 
 
 Elongation, 146. 
 Ethylidene aniline, 39. 
 Extraction apparatus, 68, 38, 260. 
 Extract, Acetone, 68, 73, 156, 251. 
 Extract, Alcoholic potash, 71, 844- 
 Extract, Chloroform, 69, 71. 
 
 Fillers, 37, 45, 112, 119, J, IS, 11 4, 
 
 221, 259, 278, 308, 309, 326. 
 Formulas, Calculation of, 117. 
 Fossil flour, 48. 
 Frictions, 147. 
 Furfuramide, 40. 
 
 153 
 
154 
 
 SUBJECT INDEX 
 
 Gas black, 48, 106, 137, 292, 380, 
 
 381. 
 Glue, 36, 108. 
 Graphite, 50. 
 Greens, 50. 
 Guayule, 18, 170. 
 
 Hevea Braziliensis, 16. 
 Hexamethylenetetramine, 39, 95. 
 
 Indian red, 51. 
 Iron oxides, 51, 119. 
 
 Landolphia, 17. 
 
 Lead, Determination of, 18, 115, 
 
 290, 359. 
 Lead oleate, 44. 
 Lead, Red, 43, 107. 
 Lead, Sublimed blue, 44. 
 Lead, Sublimed white, 44, 359. 
 Lead, White, 43. 
 Liebermann-Storch test, 35. 
 Light, Action of, 337, 31,6, 420, 47 J, . 
 Lime, 51. 
 
 Litharge, 42, 13, 106, 374, 402, 475. 
 Lithopone, 51, 120. 
 
 Magnesium carbonate, 45, 119, 21, 
 
 105, 301. 
 Magnesium oxide (heavy magne- 
 sia), 45, 119, 104, 301, 408, 480. 
 Manganese, 58, 59. 
 Master batches, 140, 141. 
 Microanalysis, 114, 7.9, 98, 101, 198, 
 
 199, 452, 496, 497, 498, 499, 500, 
 
 501. 
 Microphotographs, 114. 
 Mineral analysis, 97. 
 Mineral fillers, 29, 34, 56, 77, 177, 
 
 220, 266, 278, 281, 339, 342, 41',, 
 
 442. 
 Mineral hydrocarbons, 31. 
 Mineral oils, 34, 70. 
 Mineral rubber, 30, 31, 70, 71, 39, 
 
 83, 84, 145, 295, 310, 338. 
 Mixing, 141. 
 Moisture, 24, .$87. 
 
 Nitrogen in crude rubber, Determi- 
 nation of, 24, 364. 
 
 Oil substitutes, 28, 73, 9, 12, 19, 95, 
 127, 129, 146, 253, 340, 436, 448. 
 Oil substitutes, Tests for, 29. 
 Oils, Softening, 33, 82. 
 Oils, Tar, 36, 80. 
 Organic fillers, 37, 110, 118. 
 Oxides, Determination of, 99. 
 Ozokerite, 32. 
 
 Palm Oil, 33. 
 Pale crepe, 20. 
 Paraffin, 32, 70. 
 
 Paranitrosodimethylaniline, 40, 94. 
 Petrolatum, 34. 
 Piperidine, 41. 
 Piperine, 41. 
 Pontianak, 18, 87. 
 
 Resins, Rubber, 21, 24, 52, 93, 122, 
 234, 257, 277, 391, 402, 438, 446, 
 477, 478, 490. 
 
 Roll brown crepe, 20. 
 
 Rosin, 35, 70. 
 
 Rubber, Amber crepe, 20. 
 
 Rubber, Analysis of, 11, 19, 31, 32, 
 33, 44, 49, 54, 69, 70, 73, 96, 97, 
 102, 116, 117, 119, 120, 123, 124, 
 131, 1S5, 136, V,3, 160, 163, 166, 
 175, 178, 179, 197, 207, 208, 211, 
 214, 216, 217, 224, 237, 239, 262, 
 270, 282, 298, 306, 311, 330, 332, 
 343, 375, 382, 411, 413, 426, 439, 
 463, 465, 467, 470, 472, 473. 
 
 Rubber, Composition of crude, 22. 
 
 Rubber compound, 13, 139. 
 
 Rubber, Crude, 13, 21, 26, 37, 128, 
 184, 185, 218, 422. 
 
 Rubber, Definition of, 13. 
 
 Rubber, Determination of, 70, IIS, 
 48, 100, 190, 196, 240, 2'i>, 250, 
 304, 471. 
 
 Rubber, Difference methods for 
 determining, 82. 
 
 Rubber, Direct determination of 
 (see Nitrosite, Wesson's, and te- 
 trabromide methods), 3, 25, 26, 
 45 21,! h 245, 333, 334, 415, 445. 
 
 Rubber, Evaluation of, 8, 89, 215, 
 360, 372. 
 
 Rubber, Formula for, 12. 
 
 Rubber hydrocarbons, 13. 
 
 Rubber, Indirect method for deter- 
 mining, 81, 350. 
 
 Rubber, Insoluble matter in crude, 
 21, 15, 35, 36, 180, 238, 354, 366, 
 369, 383, 384, 393, 416. 
 
 Rubber, Nitrosite method for de- 
 termining, 79, 4, 5, 165, 193, 194, 
 202, 204, 205, 206. 
 
 Rubber, Pale crepe, 20. 
 
 Rubber, Para, 16, 17, 227. 
 
 Rubber, Plantation, 19. 
 
 Rubber for analysis, Preparation of, 
 23, 30, 296, 476, 486. 
 
 Rubber, Reclaimed, 27, 66, 112, 6. 
 
 Rubber, Roll brown crepe, 20. 
 
 Rubber, Smoked crepe, 20. 
 
 Rubber, Smoked sheets, 20. 
 
SUBJECT INDEX 
 
 155 
 
 Rubber substitutes, 28, 272. 
 
 Rubber, Synthetic, 279, 308, 328, 
 481. 
 
 Rubber, Tetrabromide method for 
 determining, 76, J,0, 50, 61, 62, 
 68, 6k, 65, 67, 71, W , 161, 164, 
 167 169 209, 220, 223, 22S, 229, 
 232, 235, 2^8, 246, 248, 273, 283, 
 294, 367, 370, 389, 390, 892, 441, 
 441, 449. 
 
 Rubber, Wesson's method for de- 
 termining, 79, 80, 92, 424, 4%5, 
 483, 484, 485. 
 
 Sampling, 23, 64, /,17. 
 
 Set, Permanent, 147. 
 
 Silicates, 102. 
 
 Sodium bicarbonate, 52. 
 
 Sodium hydroxide, 42. 
 
 Solvents for rubber, 81, 85, 188, 
 
 263, 479. 
 Specific gravity, HI, 103, 363. 
 Specific gravity table, 149. 
 Sponge rubber, 46, 110. 
 Spreading, 142. 
 Stress-strain curves, 146, 210, 861, 
 
 376. 
 Sulfates, 101, 1, 399. 
 Sulfides, 100, 7, 132, 264, 899. 
 Sulfites, 101. 
 
 Sulfur, 37, 50, 118, 233, 812, 335. 
 Sulfur, Combined (see sulfur of 
 
 vulcanization). 
 Sulfur, Free, 29, 88, 89, 29, 92, 152, 
 
 268, 269, 330, 44O. 
 Sulfur, Total, 84, 1, 2, 22, 52, 91, 
 
 94, 99, 110, 152, 159, 162, 176, 
 
 186, 187, 220, 236, 247, 252, 267, 
 
 280, 293, 307, 327, 331, 348, 895, 
 
 398, 407, 423, 439, 444, 458, 459, 
 
 461, 469, 
 Sulfur in fillers, 83, 93. 
 Sulfur of vulcanization, 25, 91, 66, 
 
 7//, 78, 153, 241. 
 Sulfur chloride, 37, 60 90, 182, 
 
 Talc, 52. 
 
 Tensile product, 145. 
 
 Tensile strength. 144. 
 
 Tensile tests, 143, 258, 291, 309, 
 
 352, 410, 458. 
 Tensile tests, Machine for making, 
 
 143. 
 Tensile tests. Test pieces for, 144. 
 Thiocarbanilide, 39, 95. 
 Thiurams, 41. 
 Triphenylguanidine, 40. 
 Tubing, 142. 
 
 Ultramarine, 52. 
 Unsaponifiable matter, 73. 
 
 Venice turpentine, 35. 
 Vermilion, 53, 178, 174, 356. 
 Vulcanization by selenium, 51. 
 Vulcanization by ultra-violet rays, 
 
 Pi 5 * 213 
 Vulcanization, Cold, 60, 61, 46, 220. 
 Vulcanization, Definition of, 1), 57. 
 Vulcanization, Hot, 107, 108, 121, 
 
 220, 284, 285, 374, 379, 397, 475. 
 Vulcanization, Ostromuislenskii's 
 
 theory of, 62, 68, 313, 314, 315, 
 
 316, 817, 318, 819, 320, 321, 400. 
 Vulcanization, Theory of, 57, 72, 
 
 125, 126, 139, 225, 230, 281, 233, 
 
 394, 404, 405, 428. _ 430, 4H, 466. 
 Vulcanization with mixed gases, 20, 
 
 825. 
 
 Washing, 22, 139. 
 Washing, Loss on, 22. 
 Waxes, 33, 82. 
 Waxy hydrocarbons, 74. 
 Whiting, 53, 119. 
 
 Yellow ochre, 53. 
 
 Zinc, Determination of, 18, 115, 
 
 290. 
 Zinc oxide, 54, 120.