172 



THE INDIA RUBBER WORLD 



December 1, 1920 



is advanced. Even the set uf curves reproduced by Schidrowitz' 

 would seein to show a similar difference. 



While therefore the upper, straight ends of the curves do not 

 run parallel for increasing times of cure, we have further to 

 consider whether the slope, determined by Schidrowitz's method, 

 that is, 0.4X (length at 1.04 kilograms— length at 0.60-kilugram ) 

 remains constant. Schidrowitz and coworkers find this to be 

 the case in their method of testing. We do not find constancy in 

 our testing ; the slope, determined by Schidrowitz's method, in- 

 creases with increasing times of cure. We have often controlled 

 this fact for the curves obtained in testing large samples, and 

 invariably found this increase in slope for increasing times of 

 cure. 



The figures in Table 1 may be cited as an example. The state 

 of cure is indicated m the first column by the position of the 

 stress-strain curve in our usual manner, i. c, by its length at 1.30 

 kilograms. The second column gives the difference in length 

 for 0.20-kilogram increase in load, or the slope of the upper 

 part of the stress-strain curve in per cent when drawn on the 

 scale of the Schopper machine. The third column contanis the 

 slope determined by Schidrowitz's method. 



Tabij; I 

 Length at 1.30 kg. Slope of Upper End Slope (SchidrowitzJ 



1070 



r.\BLE 11 



.'^Il1)c after Schidrowitz ... 34 



SioiM; of upper part of curve 37.5 



36 

 J9.9 



38 

 32.2 



4U 

 34.0 



42 



3;.o 



1050 



1030 



1010 



990 



970 



950 



33.6 

 32.2 

 30.8 

 30.1 

 29.9 

 29.7 

 29.5 



34.7 



35.1 



35.4 



35.7 



36.0 



36.35 



.16.7 



The changes in both properties arc clearly illustrated by this 

 table. It is of importance for the study of the e.xact nature of 

 stress-strain curves that the slope of the upper part ot the 

 curve becomes smaller (the curve flatter) on increasing the 

 time of cure, while the slope determined by Schidrowitz's 

 method shows a gradual increase. The two methods therefore 

 do not give results which run parallel for increasing states of 

 cure. 



It is not yet apparent what differences in method of curing or 

 testing cause this divergency in our case, while in Schidrowitz's 

 testing they remain constant. The mixture is nearly the same 

 (Schidrowitz's testing 8 sulphur on 100 rubber, in our own 7K 

 on 92'/2, or 8.1 on lOO). The temperature of vulcanization, room- 

 temperature during testing, and probably minor details in 

 method of curing and testing differ. 



While therefore the mathematical formulas for the stress- 

 strain curves, evolved by Schidrowitz and his coworkers, are 

 not strictly applicable to our testing results, there is a strong 

 indication that these authors are neverlheless on the right track 

 with their speculations on the deeper nature of the stress-strain 

 curves. They conclude that for each sample there is a correct 

 or optimum state of cure, represented by a conchoidal curve for 

 which o = fc, and giving an ideal balance of properties, the tough- 

 ness (tenacity) equalling the limit of extension. This optimum 

 state of cure is calculated to lie lower on the paper, the higher 

 the figure for "slope" or "type." We have shown' that the 

 maximum tensile strength is a property which follows this law ; 

 it is found lower on the paper, the higher the figure for slope. 

 This proves, in our view, that Schidrowitz and his coworkers, 

 in their mathematical speculations, have arrived at the truth, 

 though their formulas probably present the case in too simple 

 a form. In a former paper we have given some figures tor the 

 position of the maximum of tensile strength for samples with 

 different slopes. As the maximum of tensile strength in our 

 mixture is rather flat, it is difiicult to determine the exact posi- 

 tion of the curve, which gives a maximum tensile strength. 

 Combining all our published and unpublished results, we esti- 

 mate the length at 1.30 kilograms of the curves giving the max- 

 imum tensile strength at 992 and 957 for slopes of 36 and 40, 

 respectively. 



Schidrowitz and his coworkers give for the correct cure the 

 following extensions at a load of 1.04 kilograms per ^qua^e 

 millimeter: 884 and 850. This means a length of 984 and 950 

 times the original; using our figures from Table II, and assum- 

 ing, as is approximately true, that the curve has already reached 

 its straight part at a load of 1.04 kilograms, we calculate the 

 lcMj;ths at 1.30 kilograms for Schidrowitz's correct cure as 1023 

 and 995 for slopes of 36 and 40, respectively. These figures are 

 higher than ours, but the difference in length, distance between 

 the curves, is very similar, 38 against 35 units. 



Strict comparisons are impossible as the methods of curing and 

 testing differed ; it cannot be said whether Schidrowitz's correct 

 cure and the maximum of tensile strength coincide or not. A 

 relationship between the two is, however, very probable. 



SUMMARY 



The slope of the stress-strain curve, that is the increase in load 

 necessary to obtain a certain increase in length, or, in other 

 words, the resistance to stretching, is a property well worth 

 attention in ruhhcr testing. Determined after the formula of 

 Schidrowitz and Goldsborough it gives besides tensile strength 

 and rate of cure, an independent property, which is especially 

 typical in judging lower grades, and which, by its direct relation- 

 ship to permanent set seems to have a deeper meaning. 



The mathematical solution of the stress-strain curves as con- 

 choidal curves evolved by Schidrowitz, Goldsborough, and 

 Hatschek, does not strictly hold good for our method of testing; 

 the conclusions of the above authors as to a "correct" cure are, 

 therefore, not generally applicable. Still, the parallelism between 

 this supposed "correct cure" and the actual maximum of tensile 

 strength tends to show that a relationship exists, though the 

 mathematical formulation may be more complicated than that 

 supposed by Schidrowitz and his coworkers. 



PHENANTHRENE 



One of the newest coal tar derivatives for which commercial 

 applications are now being found is phenanthrene. This is an 

 aromatic hydrocarbon closely related to anthracene. In physical 

 appearance when pure, it is a white crystalline solid with a melt- 

 ing point of 100 degrees C. It is being marketed as .t distilled 

 product of approximately 80 per cent purity, which grade ap- 

 pears to be suitable for most commercial purposes, .■'i more re- 

 fined product could be furnished, however, to meet special re- 

 quirements. 



Phenanthrene is especially interesting to the rubber manu- 

 facturing industry as a wax substitute. It is used to replace 

 carnauba, ceresin, stearic acid, paraffine and other wa.xes in wire 

 insulation and other rubber work. 



DETERMINING THE SPECIFIC GRAVITY OF PIGMENTS 



For the determination of the specific gravity of a pigment a 

 SO-cc. pycnomcter with kerosene as wetting medium has been 

 found most satisfactory. In order to remove all occluded air 

 from the pigment, it is necessary to apply a vacuum of not more 

 than three millimeters for from one-half to two hours before 

 completely filling the pycnomcter and weighing at 15.6 degrees C. 

 The pigment must be thoroughly dry. 



•Journal of tlir Society of Chemical Industry. 1919. 347t. Fiuure 1. 



CHEMICAL PATENTS 



THE UNITED STATES 



SULPHUROl^S .^Nn Si;i,l'HURrC ACIDS AND THEIR ANHYDROUS AND 

 gaseous forms, generated by the oxidation of the rubber in 

 vulcanized rubber goods having a foundation fabric, arc neutral- 

 ized by subjecting the goods to a suitable heat and treating them 

 in a hermetically sealed chamber with undiluted ammonia gas 



