January, 1926 INDUSTRIAL AND ENGINEERING CHEMISTRY 73

content of alfalfa, the increases varying from 32.7 to 44.2 per cent. Since this increase in nitrogen occurred in crops which were much increased in yield, it is apparent that sulfur caused a very marked increase in nitrogen fixation.

Similar results for alfalfa and clover were obtained in other experiments. The data of one of the clover experiments (Table 111) show that, whereas the nitrogen content was not more than from 10 to 23 per cent greater than the checks, the use of sulfur actually caused over three times as much nitrogen to be fixed than was present in the checks.

Table 111-Yield a n d Nitrogen Content of First Cutting Clover on Sulfured a n d Unsulfured Ritzville Loams

SULFUR AND G y p s u m APPLICATIONS LBS. PER ACRE

159 500 200 1000 Check Sulfur Sulfur Cas04 Cas04

Yield, grams 2 0 . 4 9 8 . 3 82 .7 83.6 86.0 Nitrogen, per cent 2 0 . 7 2 . 5 4 2 .63 2 . 2 8 2 . 3 5 Yield increase, per cent 381.8 305.4 309.8 321.5 Nitrogen increase, per cent 2 2 . 7 2 7 . 0 1 0 . 1 13 .5

The sulfur content of the plant is often considerably in- creased, as shown in Table 11. There is less effect upon the intake of the other plant food elements. There is some evi- dence that the iron content, although not increased in amount, may be changed in its state of combination.

Sulfur and gypsum are selective in their actions in that they may affect legume crops, as shown above, but are not known to influence the nonlegumes. Moreover, the principal effect upon legumes appears to be that of increasing their nitrogen-fixing capacity.

An understanding of how and why sulfur and sulfates cause legumes to fix more nitrogen is far from complete. It is known, of course, that legumes are able to utilize atmos- pheric nitrogen by virtue of the bacteria that grow in nodules on their roots. It would thus seem that sulfur has an in- direct effect upon legumes through its direct action or effect upon the nitrogen-fixing organisms.

Rate of Combination of Sulfur with Rubber in Hard Rubber’

By W. E. Glancy, D. D. Wright, and K. H. Oon

HOOD RUBBER Co., WATERTOWN, MASS.

T THE Pittsburgh meeting of the AMERICAN CHEMICAL SOCIETY a paper2 was presented giving the results of A a n investigation of the influence of certain compound-

ing ingredients in hard rubber, more especially of their in- fluence upon the physical properties of hard rubber. At that time a suggestion was made that it would be desirable to correlate the changes in composition and the changes in physical properties which take place during the vulcanization of hard rubber. The information presented here is compiled with this end in view.

During the past ten years or more, investigation into the mechanism of vulcanization has been largely centered about the function of organic accelerators in hastening vulcaniza- tion. The characteristic curing curves, the most desirable temperatures of vulcanization, and the action of inorganic activators for various organic accelerators have been studied and theories evolved to explain the facts. It is not the in- tention to discuss here a theory of vulcanization, but to point out that in formulating any comprehensive theory the hard rubber field ought not to be neglected, especially since the one accepted compound of rubber and sulfur exists in this field.

Weber3 points out that the end product of vulcanization is polyprene disulfide, CloHl&. Other investigators have confirmed this statement. Hubner4 examined a sample of ebonite, which, however, showed less than 4 per cent combined sulfur, and reported that he had found only the monosulfide of rubber. Spence and Young5 have also shown that the rate of combination of sulfur with rubber is constant for a given temperature until 32 per cent of sulfur (estimated on the mix) is combined with the rubber. The writers’ previous work, on the changes in tensile strength as the vulcanization pro- ceeds, shows that the tensile strength increases slowly during the first part of the vulcanization, then very rapidly, and finally a t a much slower rate continues to a maximum. They

1 Presented before the Division of Rubber Chemistry a t the 69th Meeting of the American Chemical Society, Baltimore, Md., April 6 to 10, 1925.

3 THIS JOURNAL, 16, 359 (1925). 8 “The Chemistry of India Rubber,” p. D1. 4 Gumni-Zfg., $4, 627 (1910). * Kolloid-Z.. 13, 265 (1913).

have now determined approximately the amount of sulfur necessary to make a compound hard and the effect of several of the more common organic accelerators upon the coeffi- cient of vulcanization and the tensile strength.

Experimental

These mixings were made on a small laboratory mill and the usual precautions with regard to mastication, heat on the rolls, etc., were taken to insure uniformity of treatment. The mixed stocks, after aging for 24 hours or more, were vulcanized in a mold in a hydraulic press, the temperature of the press be- ing maintained a t 170” C. The test specimens were molded to form, so as to eliminate cutting, and the time of cure varied from 10 minutes in some cases to a maximum of 120 minutes.

Five mixings were made, as shown in Table I.

Number of mix First aualitv kiln-dried

smoked sheet Sulfur Diph;nylguanidine Ethylidene aniline Hexamethylenetetramine Tetramethyl thiuramdi-

sulfide

Table I 1 2

70 70 30 30

1 . 4

3

70 30

1.4

4

70 30

1 . 4

5

70 30

0 . 7

The test specimens were of the size and shape which it is customary to use for testing hard rubber-that is, 15.24 cm. (6 inches) long, with a restricted section in the center 1.27 cm. (0.5 inch) wide. It is that recommended by the Hard Rubber Division of the War Service Committee. The speci- mens mere broken on a horizontal Scott testing machine, the jaws of which were separated a t the rate of 0.5 cm. per minute, and the temperature was maintained a t 21’ C. during the testing. Previous to the testing the specimens were im- mersed in water at 21 O C. for one hour. The results reported are the averages of at least three tests, and experimental error has been reduced as far as possible by additional check tests when it seemed desirable.

The fragments from the tensile tests were ground to about 20 mesh and were used for the determination of coefficient of vulcanization. The method employed is that adopted by the Rubber Division of the AMERICAN CHEMICAL SOCIETY,

INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 18, No. 1

Figure 1

with two changes: (1) a 1-gram sample instead of a 2-gram sample; (2) the bromine was increased to 6 cc. instead of 3 cc. These changes seemed to be necessary to handle the large quantities of sulfur found in some cases, and to extract all of the uncombined sulfur. The combined sulfur has been es- timated by subtracting the free sulfur from the total sulfur. Several determinations of the total sulfur by the Pirelli method and by the method adopted by the Rubber Division of the AMERICAN CHEMICAL SOCIETY indicate that there is an

705MOKCD ZtiCCT

1 4 ~ T H Y i l D t N E . ANILINE

0

Figure 3

appreciable loss of sulfur during vulcanization when only rubber and sulfur are in the mix. This loss averaeed 0.4

Figure 2

combined sulfur estimated on the rubber, has been used in preference to the per cent combined sulfur estimated on the total mix.

Inasmuch as it was observed that there are changes in the specific gravity of hard rubber stocks which are cured for dif- ferent periods of time, the specific gravity has been determined for two of the stocks and the results recorded.

Results

The results are shown in Table I1 and also graphically in Figures 1 to 5. One very interesting fact shown in Figure 1 should be pointed out. Although sulfur has entered into combination with rubber until there is a coefficient of vulcani- zation of 28.14, there has been no corresponding increase in tensile strength and the specimen has remained flexible. As it seemed possible that dilution from the uncombined sulfur might retard the increase in tensile strength, several stocks were mixed which contained only smoked sheets and sulfur and which when fully vulcanized should have coeffi- cients of vulcanization of 17.6, 21.2, 23.5, and 28.0. The first two of the above-mentioned stocks were still soft when the vulcanization was continued for 6 hours a t 170' C. The stock which contained the largest amount of sulfur was no- ticeably hard in 60 minutes. The stock which contained 23.5 per cent sulfur, estimated on the rubber, was flexible when vulcanized for 3 hours. The evidence is, then, that the hard variety of vulcanized rubber exists only after a coefficient of vulcanization of approximately 23.5 has been reached. This amount of combined sulfur is the amount necessary that one sulfur atom be joined to one CIOH16 group. Apparently, no polyprene disulfide, c10H16s2, is formed until each ClOHl6 group has received one atom of sulfur.

Figure 2 shows the effect of 2 per cent diphenylguanidine on the base compound of 70 rubber-30 sulfur. This accel- erator, which has wide use in soft rubber goods, is evidently quite efficient in hard rubber. Various accelerators seem to give properties to soft rubber goods which cannot be obtained in unaccelerated stocks. Diphenylguanidine in hard rubber, however, seems to be purely a sulfur carrier, the rate of the

'

per cent. I n the stocks which contain accelerators &is loss is very small. The coefficient of vulcanization, or per cent

reaction'being greatly-speeded, but approximately the same maximum tensile strength is obtained as though no accel-

January, 1926 INDUSTRIAL AND ENGINEERING CHEMISTRY 75

Time of cure, min. 10 Stock

1 2 2 9 . 9 3 3 4 5

12

3 0 . 2 3

15

6 . 5 7 34 .87 3 3 . 9 8 2 2 . 2 2 27 .44

Table I1 30 35 4 0 4 5

Coe5cient of Vulcanization 2 7 . 5 7 3 2 . 8 9 3 4 . 5 0 3 7 . 4 2 4 0 . 3 7 4 0 . 9 3 3 7 . 9 0 3 9 . 9 s 3 5 . 7 5 3 7 . 2 4 3 8 . 1 8 3 9 . 9 3

Tensile Strength, Kg./Sq. Cm.

6 0

3 8 . 1 0 4 1 . 0 6 4 0 . 3 2 3 8 . 7 7 4 1 . 1 7

erator had been used. Zinc oxide is not necessary to acti- vate the accelerator.

Figure 3 shows the curing curve for ethylidene aniline. It reacts much like diphenylguanidine, but is seemingly a little less abrupt in its action.

Figure 4 shows the effect of 2 per cent hexamethylene- tetramine. Without an inorganic activator this material aids the combination of rubber and sulfur in the early stages of vulcanization, but is not nearly so efficient as diphenyl- guanidine or ethylidene aniline. In the later stages of vulcanization it seems actually to prevent the usual increase in tensile strength.

Figure 5 gives the curing curves for tetramethyl thiuram- disulfide. Only 1 per cent (based on the rubber) is used. Evidently this powerful accelerator, which is ordinarily used a t comparatively low temperatures, is very active in hard rubber at high temperatures. I ts use would probably be limited because of its cost.

Table I11

Stock I 5 30 35 40 45 60 75 90 120 .---- Time of cure, minutes----------

1 1 . 0 7 6 1 .148 1 .156 1 .156 1 . 1 6 2 1 . 1 6 2 1 .167 1 . 1 6 5 1 . 1 6 5 2 1 . 1 6 3 1 .169 1 . 1 7 4 1 . 1 7 1 1 .174 1 .173 1 .176

In considering the action of these accelerators, a state- ment can be made which would apply to all that have been included in this investigation-the organic materials are “regulators” of the rate of combination of sulfur with rubber.

The question of changes in specific gravity as vulcanization proceeds has been mentioned. The specific gravities of Stocks 1 and 2 are given in Table 111.

75

39 .17 4 1 . 3 4 4 0 . 6 0 39 .60 4 2 . 6 0

90

39 .93 4 1 . 7 1 4 1 . 2 1 4 0 . 0 7 4 2 . 2 7

120

40 .86 4 1 . 7 2 4 0 . 9 6 4 0 . 4 3 4 2 . 4 4

4 7 2 . 8 503 7 5 3 5 . 3 549 .6 5 7 7 . 3 584 1 519 2 577 .6 589 2 5 7 9 . 7 5 8 2 . 8 532 3 5 3 5 . 7 520 4 554 6 537 5 389 1 439 4 506 7 4 8 6 . 4 5 1 9 . 2 535 9 549 4 581 1 572 .0 563 1

It is possible to estimate the specific gravity of crude rub- ber as vulcanized in soft rubber goods by determining the spe- cific gravity of the vulcanized stock and making the necessary allowances for the other materials present. I n several in- stances the specific gravity of vulcanized rubber was found to be close t o 0.935. If, however, this same procedure is used for the determination of the apparent specific gravity of rub- ber in vulcanized hard rubber, this apparent specific gravity will be nearer 0.998 for a fully vulcanized hard rubber, as- suming the specific gravity of sulfur to be 2.0. It would appear, then, that in the combination of rubber with large quantities of sulfur there is a contraction in volume over that which might be expected if the rubber and sulfur exist in the same condition as in the low-sulfur vulcanized mixes.

Conclusions

1-Rapid changes take place in the physical properties of a rubber-sulfur mixture when the coefficient of vulcanization reaches approximately 23.5, or apparently when each CloHle group has received one atom of sulfur to form polyprene monosulfide.

2-An active hard rubber accelerator hastens the rate of combination of sulfur with rubber without appreciably in- creasing the strength of the final product.

3-Each accelerator has an influence peculiar to itself and may be considered a “regulator” of the rate of combination of sulfur with rubber.

4-In the formation of the hard rubber compound, the volume of the end product is found to be less than the vol- ume of the components.

70 SUOKED *CET

Figure 5