650 



THE INDIA RUBBER WORLD 



(September 1, 1914. 



(Jinarily used for carbon determinations. The carbonic acid is 

 al)Sorl)ed by the soda lime in /' and is weighed in the usual 

 manner. 



In considering this nietliod it will be realized that after all 

 this only determines the carbon in the rubber and the com- 

 bustion method is known to be somewhat unreliable and trouble- 

 some. That this is so is shown by another paper, which has 

 also been published with the consent of the Bureau of Stand- 

 ards recently and read at the Cincimiati meeting of the Amer- 

 ican Chemical Society. 



In this paper, which is on the determination of carbon in 

 iron and steel, the author, J. F. Cain, points out some of the 

 difficulties in accurately determining carbon dioxide obtained 

 during combustions of steels by the use of weighed absorption 

 tubes. These are as follows : 



1. The elaborate precautions required tn prevent cliange in 

 weight of the tuiie due to loss or gain of moisture as the gas 

 is passed through. 



2. Difficulties in weighing caii.sed by electrical effects in 

 wi|)ing by buoyancy and by cban.ges in temperature. 



3. The necessity of maintaining constant conditions with 

 respect to the atmosphere within the tube at each weighing. 



4. Danger of getting chlorine or sulpliur in the absorption 

 tube from these substances in the sample. 



5. The difficulty of determining whether the increase in 

 weight of the tube is due solely to carbonic acid absorbed. 



All the above oijjcctions will apply to the determination of 

 carbon in rubber as well as in steel, as both are carried out in 

 practically the same apparatus and by the same process. 



Mr. Cain, therefore, recommends that the carbon be deter- 

 mined by absorbing the CO^. in barium carbonate, thus precipitat- 

 ing it and afterwards titrating the carbonate after filtering. The 

 barium carbonate is dissolved in an e.xcess of hydrochloric acid 

 and titrated back with standard barium hydrate. As phenol 

 phthalein reacts alkaline with barium carbonate, metliyl orange 

 is use<l as indicator. This method of determining carljon seems 

 far better than the one specified in the rubber analysis and 

 should be adopted. 



In using this method of dttermining rubljur, it nuist lie assumed 

 that no sub.stance present in the sample is insoluble in acetone 

 that might lie acted upon by nitrogen trioxide gas in the 

 chloroform solution. It is assumed that this method is to be 

 u.sed for all classes of ruliber goods and the substances which 

 may I'e added in the mixing are, of course, of very diverse char- 

 acter. In the first place in many classes of rublier articles it 

 would be impossible to separate the fibre from the sample liefore 

 analyzing. It would be necessary to be sure tliat none of the 

 fibres will be acted upon by the trioxide in any way that will 

 give an insoluble product in the chloroform and dissolve in 

 the acetone. It is certain that nitro-cellulose is such a product, 

 and it should be proved that there are no like products formed 

 from the trioxide of nitrogen. 



It is also agreed by the authorities that acetone does not 

 dissolve out completely all kinds of pitches or asphalts nor the 

 substitutes which have been vulcanized to a high degree with 

 sulphur when made. 



In fact, the number of substances containing carbon which 

 may be used in rubber compounding is almost unlimited and they 

 are not all soluble in acetone, so that it must be assumed tliat 

 the nitrogen trioxide has a highly selective affinity for the 

 caoutchouc molecule. 



Of course, if the method is to be limited to samples which 

 contain only known compouiuls in addition to the rubber, then 

 it is not of wide application. 



In I90.S Caspari showed' that gutta percha and balata formed 

 iiitrosites the same as caoutchouc, so that this method would 

 not distinguish between these substances. 



"Journal ."^ociely Chemic.ll Industry, 1905, p. 1274. 



The present method sometimes used of determining rubber 

 by difference is, of course, open to many objections, and it is 

 hoped that a method will be found that will not only determine 

 the rubber contents, but if possible determine the kind of rubber. 



\V. V'aubel and K. W'einruth* investigated the bromination 

 method of Huddle-Axelrod and Vaubel. They found it unre- 

 liable except on Para rubber. It failed to give concordant 

 results on vulcanized rubber or on crude. 



I'ontio' recommends boiling vulcanized rubber with toluene, 

 .xylene and cymene which dissolve the caoutchouc. h'ilter 

 while hot, then evaporate to a syrup and precipitate the pure 

 caoutchouc with acetone. 



The determination of carbon in the above precipitate would 

 not be open to objections of liability to contamination with other 

 carbcm compounds that the first method is. 



THE INFLUENCE OF NITROGEN COMPOUNDS ON 

 THE VULCANIZATION OF RUBBER. 



/^NE of the most important results of modern chemical and 

 ^^ scientific work is the recognition of the importance of small 

 (|uantities of compounds or elements when present in the main 

 commercial articles of chemical nature. Of course this recog- 

 nition is not altogether a recent phenomenon, for one of the 

 most important examples of this was the discovery of the con- 

 trolling influence of minute quantities in the manufacture of 

 Bessemer steel, and it was not till ores were obtained of low 

 phosphorus content that this industry became possible. In 

 Bessemer steel there must not be more than one-tenth per cent, 

 of phosphorus or the material is worthless. 



In the rubber industry it has not been until recently that this 

 subject has been .given much attention, particularly in relation 

 to the small quantity of nitrogen found in most commercial sam- 

 ples of rubber. 



Thus we fiiul that Weber, in his pioneer work on rubber chem- 

 istry, notes the insoluble constituent of rubber but concludes that 

 it was an oxidized hydrocarbon and a link between rubber and 

 cellulose, and he omitted to examine the insoluble matter for 

 nitrogen or to make any reference to nitrogen as being present 

 in rubber, except in a mere hint on page 5, where he quotes 

 Faraday's analysis of latex showing 1.90 per cent, of albumen. 

 The subject is dismissed with the idea that with good coagula- 

 tion work this "impurity" may be eliminated. 



In 1907 Dr. David S|ience ("Journal Society Chemical Indus- 

 try," p. 1287, of 1907) investigated the subject of the distrilnuion 

 of protein in rubber, and found that in the insoluble part there 

 was sometimes as much as 5.4 per cent, nitrogen. He consid- 

 ered it mainly protein; this would mean about 33 per cent, pro- 

 tein. 



Even Schidrowitz, in his l)ouk published as late as 1911, while 

 noting the work of Spence, does not disclose any knowledge of 

 any effect of this nitrogen constituent on rubber, though noting 

 that plantation varieties are deficient usually in nitro.gen. 



.\t the Eighth International Congress of Applied Chemistry, 

 held in Xew York in the fall of 1912, Clayton Beadle and II. P. 

 Stevens read a paper describing an investi,gation of Ilcvca latex 

 and giving determinations of the nitrogen in the latex and in the 

 dry crepe produced. They found that the latices contained from 

 1.31 per cent, to 1.56 per cent, protein; that the percentage of jiro- 

 tein of the total solids in the latex varied from 5.5 per cent, to 

 5.1 per cent., and that about one-half this protein was retained 

 in the washed and coagulated rubber in the form of dry crepe. 



In the proceedings of this same congress, Lothar E. Weber 

 gave results of experiments tending to show that the extraction 

 of the resins from plantation sheet prevented vulcanization of 

 the extracted sheet, but in an article by Stevens in the "Colloid 



'Chern. Zeit., 1914, i. 82 from Giimmi-Zeitung. 191.^. pp. 28-93, and Journal 

 Chemical Society, April, 1914, p. .^01. 



'Journal Chemical Society, .\pril, 1914. p. 301; .'\nn.il. Chim. Anal., 1914, 

 pp. 19-60. 



