I N D U S T R I A L A N D ENGINEERING C H E X I X T R Y
262
canization of definite structural elements possibly along the lines referred to by Lunn, which serve to impede any readjustment of internal stresses during the early stages of vulcanization. It would be most stimulating to compare the displacement Criterion Z of a calendered sheet after vulcanization with the crawl or shrinkage of the same uncured sheet when exposed to intermediate temperatures, and also to rubber solvents.
VOl. 15, No. 3
Further study of ‘(grain” phenomena after vulcanization would include a systematic variation of both the amount and degree of dispersion of the pigment phases in the compound. It would also seem that the search for the ideal rubber pigment should include a study of the “grain” effects after cure in terms of not only size but crystal habit of the particles. The whole “grain” problem is replete with theoretical and practical significance.
Rubber Softeners’ By Paul M. Aultman and C. 0.North LEE TIRE & RUBBERCo., COWSHOHOCKBN, P A . , AND RUBBERSERVICE LABORATORIES, APRON,OHIO
T
HE
literature The following article describes the authors’ inoestigations with aggregate is broken into on the subject of regard to rubber softeners. smaller fragments. This is They haw studied the comparatioe action of various Softeners a reversible action as the rubber compoundon oulcanized rubber at the uulcanizing temperature of 140” c., rubber slowly recovers aping is comprised chiefly of articles on three raw mateand ham found thaf sulfur and accelerators aid in the depolymproximately its former properizafion of rubber, that the more adoanced the state of cure the erties. It is obvious that rials: (a) crude rubber, @) slower the depolymerization rate becomes, and that depolymerizing otherwise unexplained difpigments, and ( e ) accelersubstances usually accelerate the combination of sulfur. ferences in crude rubber atom. It covers quite thoroughly the variations of can be understood best plantation and wild rubby a conception of varirtbers, and lately it has considered the volume relations tions in the value of rz. However, many softeners when of the pigments as well as giving us better working theo- added in such small quantities as one-eighth of one per ries on accelerators. One other class of substances, how- cent, not only help to make the rubber more plastic, but ever, has not received the attention of the rubber chemists even prevent it from going back to its former state. in proportion to the importance it assumes in most practical These softeners aid in the process of breaking down, and mixes. We will designate this class as “rubber softeners,” hence are depolymerizing agents. This depolymerizing which includes substances such as mineral rubber, oils, tars, action of the softeners is undoubtedly carried on into the pitches, resins, and gums. The organic accelerators com- aging of the rubber, where it prevents the formation monly used may also be classified in this group, as they of the larger aggregates, thus tending toward preventing generally soften or render more plastic the uncured rubber. shortness. We assume that aging consists in the formation of larger aggregates through the agency of sulfur and through NEED FOR RUBBERSOFTENERS a natural tendency of the rubber aggregate to revert to larger Rubber softeners are added to mixes usually for two major Wgregates. reasons-first, to facilitate the processes of milling, calenderIn the curing of rubber we must assume that there are ing, and tubing; and, second, to cheapen the stock. The two major factors-one a breaking down of the rubber compounding of these into rubber gives us a wide variation aggregate, and the other a combination of sulfur and brokenin the characteristics of both cured and uncured rubber. down rubber aggregates. These processes can well be termed In the milled batch the rubber becomes more plastic and more depolymerization and polymerization. Our interest in softeneasily worked, even when very small amounts of Some sub- ers lies primarily in their effect on depolymerization. Their stances are incorporated. In the cured product the most effect on the combination of sulfur can be considered merely notable effects usually are the lowering of the stress values as a disturbing factor. of the stress-strain curve and the increasing of the elongation In order to find out this comparative depolymerizing effect, more than warranted by either any retardation of sulfur it was realized that the method of incorporating into a base combination or mere filling action. There is some effect stock in equal proportions and comparing the physical propon the properties of the rubber mix diametrically opposed erties with those of the base stock, when both were wlcanto that of finer pigments when added to rubber, as the pig- ized in the same cure, was the ultimate method of attack. ments make the stock less plastic and raise the values of the However, this method, even though used with some intereststress-strain curve, as well as cutting down the elongation. ing results in the case of pigments, has one serious faultThis effect varies greatly, depending on the softener used, namely, that it does not bring into consideration another probably because softeners have some action on the rubber variable, that of a change of the state of c u e . For instance, glue acts as a mild acce1erator;according to our combined aggregate itself. Rubber is generally considered a polymer of CloHle, the sulfur results, while clays often act as retarding agents. formula usually being written (CloH&, where n is a variable Comparison of any one effect can only be made when one number which must be large. It may well be called the variable is present. Inasmuch as the stress-strain curve coefficient of polymerization. The processes of milling is affected, not only by the state of the cure, but also by the and calendering, as well as heat alone, cause changes in the rapidity of the cure, it would be necessary for purposes of plasticity of rubber which can be explained only by assuming comparison to have the true combined sulfur the same a t other words, the rubber the cures tested. Any comparison tests which do not conthat the value n decreases-in d e r the coefficient of vulcanization cannot be accurate. 1 Presented before the Division of Rubber Chemistry at the 64th These caused us to another more direct Meeting of the American ChemicaI Society, Pittsburgh, P a , September 4 to 8. 1922. method for discovering the comparative actions of the various
INDUSTRIAL A N D ENGINEERING CHEMISTRY
March, 1923
substances on vulcanized rubber. In introducing this method we realize its crudity and that the results may be of little value in solving the problem, but we believe that the results are a t least of interest and that they may constitute a link in the chain of facts concerning rubber softeners. METHOD Briefly, this method consisted of heating vulcanized rubber in the softener until total disintegration ensued. An oil bath equipped with a constant-temperature controlling apparatus was used. I n this bath flasks containing 150 cc. of the softener were placed. The temperature was kept a t 140' C. When the temperature of the softener became 140" C., 1 g. of coarsely ground, vulcanized pure gum was stirred in, and the time required to secure disintegration noted. This procedure precluded the investigation of mineral rubbers or any tar-like bodies. The results also, owing to an indefinite end-point, are only approximations. Some of the materials tested were taken merely as a matter of interest. ACTIONOF VARIOUSSOFTENERS The time required for the various softeners to disintegrate the rubber was practically as follows:
............................ ................................. Paraffin oil.. . . . . . . . Xylol.. .............. Vaseline, . . . . . . . . . . . . . ...................... Naphthalene. Paraffin
................................. ................................... ............................... .............
Pine oil Rosin Oleic acid Coconut oil.. Stearic acid.. . . . . . . . . . . . . . . Lard oil.. . . . . . . . . . . . . . . . . . Cottonseed oil.. . . . . . . . . . . Corn oil., . . . . . . . . . . . . . . . . Soy-bean oil.. Linseed oil.. ............................ Castor oil ..............................
............
50 min. 1 hr. 35 min. 1 hr. 45 min. 2 hrs. 2 hrs. 20 min. 4 hrs:30 min. 5 hrs. 5 hrs. 10 hrs. 12 hrs. , 12 hrs. 12 hrs. 1 5 hrs. 16 hrs. 30 hrs. 34 hrs. (Not clear solution) 34 hrs. (Not clear solution) 38 hrs. (Not clear solution) 45 hrs. (No change) 45 hrs. (No change)
A number of other materials were tested but the results are not significant. The waxes and resins were very difficult to handle and did not give good results. It was observed that the time required for the saponifiable oils evidently had some relation to the iodine absorption number, as oils having a low number dissolved the rubber much faster than those of higher absorption number.
EFFECTOF BLOWIKG OILS Castor oil was an exception probably due to the hydroxyl groups of the ricinoleic acid. As several chemists claimed that blowing semi-drying oils introduced =O or -OH groups into the acid group, several samples of corn and cottonseed oils were blown a t different temperatures and their solvent effect compared with the untreated oil was note'd. In every case the blown oil took longer to disintegrate the rubber. EFFECTOF ADDIXGSULFUR The time required for the terpenes was not in accordance with practical results, as these are known to be strong depolymerizing substances. This observation led to the belief that other factors modified their action. As sulfur is always present as the vulcanizing agent in practical mixes and as sulfur also reacts rather readily with terpenes, 4 g. were added to pine oil and the resulting solution was tried with the untreated oil as a check. The time required for disintegration was approximately 2 hrs., instead of 12 for the untreated oil. This oil gave such a remarkable change in the presence of sulfur that representative samples of other oils were tried, using the same proportions. Representative results are as follows:
Paraffin,,
Aniline..
................
.................
Castor oil..
..............
263 No Sulfur
1 hr. 30 min.
2 hrs. 30 min. 14 hrs. 15 hrs. 5 hrs. 12 hrs. 35 hrs.
No solution
With 4 G. Sulfur 1 hr. 30 min. 2 hrs. 4 hrs. 30 min. 4 hrs. 30 min. 2 hrs. 2 hrs. 10 hrs.
It will be noticed that t.he addition of sulfur in most cases accelerated the disintegration. That the quantity of sulfur used might add in swelling the jell is easily explained, as it has been definitely proved that sulfur does dissolve in rubber. However, the fact that the disintegration itself was hastened leads us to believe that sulfur itself acts as a depolymerizing agent. The theory that a substance acting as a vulcanizing agent should also be a depolymerizer is not contrary to the evidence, as several softeners having a strong depolymerizing action accelerate the addition of sulfur to rubber. For instance, a base stock of rubber, zinc oxide, hexamethylenetetramine, and sulfur gave a combined sulfur of 0.93 per cent a t a cure of 40 min. a t 290" C. Three per cent vaseline added raised this to 1.10 per cent, and the addition of 3 per cent naphthalene, to 1.04 per cent. These are selected as examples because the vaseline and naphthalene should not react appreciably with the sulfur. The fact that the stock in which the rubber has been broken down for a long time on a hot mill will give a higher coefficient a t a given cure than the same normally broken down, is considered as contributory evidence. Depolymerization, then, aids sulfur addition, perhaps by increasing the number of aggregates available for sulfur combinations. ACTIONOF ACCELERATORS
As it follows that organic accelerators must also hasten depolymerization, the addition of these in the same amounts as the sulfur was tried. Thiocarbanilide, hexamethylenetetramine, and p-nitrosodimethylaniline were used. The time required to secure solution was generally greatly shortened. For instance, the time for stearic acid was cut from 14 hrs. to 6 hrs. and that of paraffin oil from 3 hrs. to 2 hrs. Here again we have an accelerating substance acting as a depolymerizer. It is noteworthy that accelerators usually soften the batch in which they are mixed. METHODOF SWELLING The end-point of these experiments was, as before stated, rather indefinite. Hence, it was thought necessary to check the observations. Weighed pieces of vulcanized test sheets about 1 in. square were immersed in the softeners, heated to various temperatures, and after definite periods of time were taken out and the excess clinging to them removed. These pieces were then reweighed. The weight of softener dissolved in the rubber was thus noted. Some representative results are listed below: RELATIVB VOLUME INCREASF-PBR CENT>
N o Sulfur Added Naphthalene., , .................... 273.0 Vaseline. . . . . . . . . .. . . . . . . . . . . . . . . . . . . 59.0 Pine oil and rosin. 44.0 Aniline. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 3 9 . 2 Cottonseed oil.. . . . . . . . . . . . . . . . . . . . . . 31.3 Castor oil.. . . . . . . . . . . . . . . . . . . . . . . . . . .... 1 These samples were heated to 100' C. for 3l/z hrs.
..
...................
Sulfur Added 282.0 81.2 69.0 52.5 47.0
....
It was definitely shown that sulfur aids in the swelling of the rubber. However, the difference between the degree of action of the various substances was not so great as in the disintegration method. Evidently, the rapidity of disintegration depended on two factors-the rate of swelling, and the rate of disintegration of the rubber aggregate. These have a distinct relation. In previous work on the di&tegration of rubber soles for the purpose of identifying pig: ments and fibers, it was noted that there was not nearly so
264
INDUSTRIAL A N D ENGINEERING CHEMISTRY
marked a difference between oils in their ability to swell the rubber as in their property of breaking down the rubber itself to form a clear solution. Hence, the rate of disintegration gives us more nearly the measure of relative depolymerization effects of the various softeners. By the swelling method it was also found that accelerators materially hasten the swelling of the rubber and that the addition of ?both accelerator and sulfur causes a further increase over either used alone. The retarding effect of four reinforcing pigments-zinc oxide, clay, glue, and carbon black-in a stock was also investigated, with the result that it was found that equal volumes had practically the same effect when the coefficient was approximately the same. Fine pigments apparently retard the swelling of the rubber.
EFFECTOF MIXTURES As oils are frequently used as mixtures, the effect of mixed oils was tried by the disintegrating method. I n general, if both oils had even slight depolymerizing action the time required was nearly an average of the time taken by each of the ingredients separately. For instance, vaseline and cottonseed oil required 4 hrs. 30 min. and 34 hrs., respectively, to dissolve the rubber, while a 50/50 mixture of the two required approximately 20 hrs. On the other hand, oils having no action on the rubber, such as linseed or castor, almost totally retarded the action of other oils with which they were mixed. A mixture of castor oil and paraffin was heated for 40 hrs. without causing more than a slight swelling of the rubber. Another peculiar result was that nothing was found which would even slightly retard a sulfur and terpene mixture. A mixture of 70 parts castor oil, 30 parts pine oil, and 4 g. sulfur required only 2 hrs. to secure a clear solution. In previous work on rubber soles it was observed that when ground, cured scrap had been incorporated in the stock, ;this scrap was extremely hard to dissolve. This was thought t o be due to the higher state of cure of this rubber as it had gone through two vulcanizations. To test this theory a number of tests were run by the solvating method. In every
Vol. 15, No. 3
case, except in a long overcure on a low sulfur stock after all sulfur had been combined, this theory proved to be true. However, in the case of the overcured stocks, an increase in the amount dissolved in the rubber was observed. This increase took place shortly after a weakening of the stock was observed, judging from the stress-strain curves. At this point the stock had evidently passed the equilibrium point where the rate of depolymerization was equal to the rate of polymerization, and was then suffering a depolymerization alone because practically all the sulfur had been combined. The state of aggregation was then becoming more nearly that of the lower cures. The essential difference between the aggregates of low cure and of this overcure was that more sulfur is combined with each aggregate combined in the overcure. An interesting conclusion can be made from this last experiment-the higher the state of cure the slower the rate. of depolymerization becomes. This means that the greatest depolymerizing action takes place in the raw stock or a t the lower states of cure. Whether or not this depolymerization is beneficial remains to be seen. From physical tests alone the evidence is strongly against undue depolymerization, as the more rapid the cure the higher the tensile which can be obtained. Thus, the thiocarbamates give us extraordinary tensiles, such as 3000 lbs. (with a 5-min. cure) per sq. in., in nearly pure gum stocks, while a suliurrubber mix requiring 3 hrs. to obtain an approximateIy optimum cure will give us only 1500 lbs. per sq. in. I n the first case less than 11/2 per cent of sulfur combines with the rubber, while in the Iatter more than 3 per cent combines. Howevei, as the aging is of supreme importance it remains to be proved whether or not the property is benefited. It remains to be proved that we must strive to keep depolymerization down to a minimum or plan to secure a certain amount in order to obtain good aging. The authors maintain that until depolymerization is thoroughly understood and measured and until the effect on aging is determined, very little progress can be made toward making rubber compounding a science instead of the meaningless mass of data which it is a t present. Tests are now being made with little or no consideration of the greatest of variables-depolymerization.
Improvement in Chemical Exports Considering the low index figure for prices in the chemical group, the export trade of the United States in chemicals and allied products for the eleven months, January to November, 1922, is most encouraging. For the first time since the postwar depression began there is an increase over the corresponding total of the preceding year. Inasmuch as the aggregate exports of all classes of American merchandise continued to fall below those of a year ago, this recovery in the foreign trade in chemicals, although small, has unusual significance. American chemicals and allied products exported in the eleven months of 1922 had a total value of $97,215,547, compared with $96,267,165 in the corresponding period cf 1921, a gain .of 1 per cent. (If naval stores, gums, and resins are included, an increase of 7 per cent is shown ) Deductions based on such a varying element as price are of necessity unreliable. Were it possible to procure the actual weight of all American chemicals shipped to foreign countries (some classes of which are recorded by value only), it is not improbable that the figures would show, not the falling off that values have a t times indicated, but a continued and healthy growth in this comparatively young industry. Although there still exists a loss of 4 per cent in the value of genera1 chemicals sent to foreign countries, gains occurred in medicinal and pharmaceutical preparations (13 per cent), fertilizers and fertilizer materials (4 per cent in quantity and 3 per cent in value), explosives (90 per cent in quantity and 46
per cent in value), and perfumery, cosmetics, and toilet preparations (33 per centuin value). Among the heavy chemicals, total foreign shipments of which fell in value from $49,625,532 in 1921.to $47,668,525 in 1922, the largest increases as t o quantity were in borax, 296 per cent, which rose from 3,658,059 lbs. (value $246,658) in 1921 t o 14,501,189 lbs. (value $755,612) in 1922; caustic soda, 238 per cent-from 40,460,561 lbs. ($1,627,334) to 136,900,479 lbs. ($4,934,944); chloride of lime, or bleaching powder, 149 per cent-from 14,892,143lbs. ($409,106) to 36,972,472 lbs. ($610,684);potassium chlorate, 75 per cent-from 297,002 Ibs. ($40,961) to 522,416 lbs. ($46,876); acetate of lime, 61 per cent-from 15,750,798 lbs. ($345,379) to 25,385,756 lbs. ($522,700); and copper sulfate, 54 per cent-from 3,153,278 lbs. ($198,260) t o 4,858,331 lbs. ($234,772). Some of the other American heavy-chemical sales abroad which have shown marked improvement during this period were calcium carbide, glycerol, sodium silicate, sodium bicarbonate, washing powder, and crude tar. On the other hand, benzene decreased only 4 per cent in quantity but 19 per cent in value-from 88,277,375 Ibs. ($2,888,126) in 1921 to 64,630,735 lbs. ($2,353!136) in 1922; formaldehyde, from $312,407 t o $195,961; sulfuric acid, from 12,155,349 Ibs. ($306,068) to 11,938,403 lbs. ($188,057); and soda ash, from 32,699,844 lbs. ($800,378) to 26,617,260 lbs. ($615,733). Salsoda increased in quantity but fell in value.
.
