I X D IJ S T R I A L A N D E N G I N E E R I N G C H E M I ST R Y
746
TABLE111.
VARIATION O F
FREEAND COXBINED FATIN FAT-LIQUORED CHROME HIDE POWDER LIQUORING FOR MIXTURESOF SULFONATED AND RAWCASTOR OIL
vARI.4TION I N
Sulfonated oil: 1 day 7 days 15 days 4 weeks 8 weeks 12 weeks
TABLEIv*.
25% 38.0 32.4 32.3 30.0 27.4 25.6
VARI.4TIOK O F
FREEF A T 75%
50% 32.9 27.8 27.3 25.6 24.1 20.0
25.8 21.1 20.1 19.2 15.3 10.8
V.4RIATIOh' IX C O M B I N E D
-Unoxidized7 25% 50% 75% % of d r y fat-liquored hide p o w d e r 3.45 7.39 11.80 4.90 9.73 14.81 5.26 10.32 16.08 5.43 10.72 16.65 5.49 10.2 17.0 5.96 13.7 20.8
FREEF A T 50% 75%
7
1 day 7 days 15 days 4 weeks 8 weeks 12 weeks
36.7 34.4 34.3
30.5
31.0 29.9
3.87 5.41 5.71 6.04 5.83 6.35
FAT
Total 50% 7.84 10.20 11.00 11.43 11.3 14.4
75%
.
12.40 15.47 16.60
17.35 19.5 22.3
Comw
--
34.0 31.8 28.8 27.1 26.6 26.6
29.6 27.8 25.5 23.1 24.5 22.8
-
-VARIATION Unovidized
IN
?5% 50% 75% % of d r y fat-laquored hide p o u d e r 2.21 4.16 6.68 2.14 4.80 7.88
2.31 2.48
2.26 2.63
4.20 4.83 4.60 4.95
extractive is usually very light or "oily" colored and seldom ever has the greenish cast of chromium compounds, whereas a chloroform extractive will in many cases show a greenish color indicative of the extraction of chromium compounds of one type or another. It is true that, if the chloroform extractives are analyzed for chromium, it will be found to be very small; but, if the polar groups should enter the chromium complex, the percentage of chromium would necessarily be small and the fatty acids large, especially if the complex is olated in character. In view of their work, the writers believe that petroleum ether extracts most of the oil that exists really free in the leather and that chloroform extracts not only free oil but oils actually in combination with any free
8.01 7.92 7.77 8.11
COWBINED F ~ T 25%
Total50%
2.93 3.69 4.62
5.40 6.34
4.80 4.55
8.14
5.09
8.01 8.20 8.27
OF
FREEFATTY ACIDS
25%
50%
%
%
%
7.5 11.7 15.7 20.2 29.4 35.0
13.6 17.9 19.5 24.5 31.5 39.1
24.6 23.3 25.2 30.2 38.9 46.0
COMBINED FATI N FAT-LIQUORED CHROME HIDE POWDER WITH LIQFORING FOR MIXTURESOF SULFON~TED AND R a w CODOIL 7
25%
25%
TIME 4 F T E R FAT
WITH
FREEAND
VaRI4TION I N
Sulfonated oil
7 -
Vol. 26, No. 7
75%
TIME AFTER F ~ T
C O W N OF FREEFATTY ACIDS 75% 8.12 9.22 11.32 11.4 11.70 11.49
75%
25%
50%
%
%
%
15.2
26.0 28.0 32.0 3i.1 38.4 41.5
8.6 21.2 21.7 24.5 25.9 30.6
22.2 28.6
31.5 32.1 37.2
chromium salts in the leather and with the chromium combined with the hide fiber. LITER.4TURE C I T E D
(1) Stather and Lauffmann, Collegium, 1932,391, 672, 940. (2) Stiasny and Riess, Ihid., 1925, 498; J . I n t e r n . SOC.Leather Trades Chem., 10, 150 (1926); J. Am. Leather Chem. Assoc., 21, 76 (1926). (3) Theis and Graham, Ibid., 28, 52 (1933). (4) Theis and Hunt, ISD.ENG.CHEW,23,50 (1931); 24,799 (1932). (5) Wilson and AMerrill, ".lnalysis of Leather," McGraw-Hill Book Co., N. Y., 1931. RECEIVED April 16, 1934. Presented before the Division of Leather and Gelatin Chemistry a t t h e 86th Meeting of the American Chemical Society, Chicago, Ill., September 10 to 15, 1933.
Relation of Volume Change to the Mechanism of Rubber Vulcanization IRAWILLIAMS,E. I. du Pont de Nemours & Company, Wilmington, Del.
A
M O S G the many partial e x p l a n a t i o n s of t h e mechanism of vulcanization, several require a polymerizing action at some step in t h e p r o c e s s ( I , 7 , 11, 14, 16, 17). The actual mechanism of the polymerization is u s u a l l y not and in c e r t a i n cases it is p r o b a b l e that the
Rubber-sulfur mixtures increase in density during in direct proportion to the amount of sulfur combined. No change in denSifY has been found Which would indicate either an increase or decrease in the number of double bonds during culcanization. T h e reaersion point of rubber during tulcanization, at least in certain cases, corresponds to the time of practical disappearance of accelerator.
term t~PolymeriZationli may be i n t e n d e d " t o cover some type of phenomena not involving primary valence forces. A concept of the mechanism of vulcanization involving polymerization is more or less logical follon-ing the discovery of the polymerization of butadienes to rubber-like bodies. The polymerization of isoprene which causes a loss of half of the double bonds produces a more or less plastic body, and a further polymerization to an increased molecular weight could conceivably cause the loss of plastic properties which results in rulcanization. This mechanism has been suggested for the vulcanization of plastic chloroprene polymer ( 3 ) , although little evidence exists t o support such a process.
The existing evidence points almost entirely to the absence of polymerization d u r i n g vulcanization. Spence a n d Scott (12 ) have p r e s e n t e d evidence based on iodine absorption to show that no decrease in unsaturation Occurs which be directly a c c o u n t e d for by of sulfur. Fisher and Gray ( 6 ) have s t u d i e d by the s a m e m e t h o d rubber vulcanized with nitro bodies or benzoyl peroxide. The authors conclude that any change in unsaturation was less than one per cent. The ability of rubber to combine with 32 per cent of sulfur without the evolution of more than small quantities of hydrogen sulfide is a further indication that little disappearance of unsaturation has taken place because of polymerization. It should be possible to determine the extent of polymerization by following the change in density. During the polymerization of isoprene the specific gravity changes from 0.68 to approximately 0.90, or by a value of 0.22 with a loss of
July, 1932
INDUSTRIAL
AND
E Y G I N E E R I N G CHEMISTRY
half of the double bonds ( 4 ) . If loss of the remaining unsaturation ihould cause a n increase in specific gravity of onlj 0.20, it should be possible b y measuring nithin t n o units of t h e third decimal place t o determine a loss of one per cent of the double bonds. Measurements have been recorded mliich indicate changes in densit? ( 2 , 5 , 9, I O ) , but data are not available to ceparate the changes due t o various exirting factors.
,9400
3MO
,9200
e100
pomoloea SuKur 1
2
3
4
E
I
6
The most efficient type of vulcanization is acconipanied liy coinbination of sulfur. The creation of the resulting organic sulfur compound will cause a change in density which will be measured along with any change in density caused by loss of unsaturation, and a suitable correction must be made. T o do this, advantage may be taken of the fact that, by choice of vulcanizing conditions, greatly different states of vulcanization, as measured by modulus or tensile qtrength, m a y be obtained with t h e same amount of combined sulfur. If rulcanization is accompanied by a change in the state of polymerization, then changes in density should be influenced by both the amount of combined sulfur and change in physical propertie., n-hile the absence of polymeric change would be indicated by a linear relationship between rombined sulfur and specific gravity, regardless of the p h of the rubber.
Rubber compounds cont>aiiiingzinc propionate even after long extraction still contained small amounts of zinc. The amount remaining, while enough t o have a measurable effect upon the specific gravity of the rubber, was not determined because of the difficulty of sufficiently accurate determination and because of t,he uncertainty in regard to the proportion existing as zinc sulfide or zinc propionate. T h e amount of zinc remaining appears to be great'er when the rubber is vulcanized a t a higher temperature, probably Iiecause of t,he production of zinc sulfide, and slou--curing compounds vulcanized a t high t,emperature appear to have slightly greater specific gravity than more rapid-curing compounds. The amount of zinc present is also probably influenced by the difference in solubilizing action of t h e different accelerators employed, n-liich assist in the extraction by the formation of a complex or salt with t'lie zinc. Rubber compounds containing zinc propionabe and accelerated with piperidinium pentamethylene dithiocarbamate and with a butyraldehyde butylamine reaction product were compared with a compound containing no accelerator. A base stock was first mixed which contained all but the a.ccelerator. This stock ?vas then dirided and accelerator added to two portions. The vulcanization time and temperature were adjusted t o suit each compound. The composition of each stock is show1 in Table I and the properties are shovvn in Table 11. TIBLE I. COMPOSITIOS OF COMPOVSDS COXT.UXISG ZIXC Extracted pale crepe rubber Zinc propionate Sulfur Phenyl-@-naphthylamine Piperidinium gentamethylene dithiocarbamate Butyraldehyde butylamine Vulcanization temp..
O
VELCASIZATION
TIME .lfinutes
25-.-i 100.0 4.0 6.0 0.5 1.0 0.0
25-B 100.0 4.0 6.0
0.5 0.0 0.0
39-B 100.0 4.0 6.0 0.5 0.0 1.0
115
145
140
C
1.0.4D
.It 500% At elongation break Kg. p e r s q . cm.
d'r9
0
COUBIKED
15
0
SULFCR
%
C O M P O U N D 25-21
EXPERIXIESTAL PROCEDCRE T h e following compounding practice was adopted : A quantity of pale crepe rubber was acetone-extracted and blended. The acetone extraction removed considerable resinous material which otherwise would have been converted at. least partially into an unextractable sulfur compound. Zinc propionate was used as the accelerator activator when required because of its solubility in rubber and ease of extract'ion. The rubber was vulcanized between sheets of aluminum foil which were not removed until just previous to making physical tests in order to protect the rubber from surface contamination and oxidation. Small samples of each cure were wrapped separately in cloth and acetone-ext,racted for 8 days. The position of the samples in the extractor v a s frequently changed t o insure uniform extraction. Specific gravity was determined indirectly by means of a hydrometer of such construction that the fourth decimal place could be estimated. The samples after acetone extract,ing were dried at a pressure of about 3 mm. of mercury i n a desiccator cont,aining calcium chloride. They were protected from air at all times as much as possible because of the rapid absorption of oxygen which is accompanied by an increase in specific gravity of a magnitude sufficient to cause serious difficulty. After drying, a strip approximately 1 mm. square and 4 em. long was cut from the rubber. This was dropped into an alcohol-water mixture which n-as held at a temperature within 0.1" C. and which contained a small amount of sodium butyl napht,halene sulfonate to promote wetting of the surface. The gravity of the liquid was adjusted approximately, and the vessel was connected to a vacuum line for 2 minutes. The final adjustment o f the gravity of the liquid was then made, after which the test' sample was cut into four pieces. I n case all four pieces when returned to the liquid did not have the same gravity, it was taken as an indication of bubbles or foreign material being contained in the rubber, and the test was repeated.
3 10 30 60 120
30 95 109 89 65
60 120 180 240
...
3 5 10 20 30 60
11 23 33 35 32
144 172 109
184 158
0.9197 0.9285 0.9332 0.9350 0.9355
3.17 3.34 3.13
1.29
0.9235 0.9295 0.9355 0.9428
1.13 2.05 2.74 3.71
0.9160 0.9212 0.9260 0.9320 0.9373 0.9484
0.80 1.38 2.30 3.22 4.00 5.40
2.50
C O M P O U N D 25-B
4 6 9
21 65 93 91 C O Y P O V X D 39-B
21
123 204 187 171 165 185
S o relationship is apparent between either combined sulfur or specific gravity and the state of vulcanization. While the compounds all include practically t h e same range of specific gravity and percentage of combined sulfur, more than a tenfold difference exists in the load-carrying capacity a t 500 per cent elongation. Such results emphasize t h a t the degree of vulcanization should be defined in physical rather than in chemical terms. The relationship between combined sulfur and specific gravity is shown in Figure 1. These curves indicate t h a t change in density is due t o nothing other than combination of sulfur. Curve 29--4, which includes greatly different physical states of vulcanization, shows no irregularity t h a t mould indicate density changes corresponding to t h e different physical states of the rubber. The slightly higher specific gravity of the unaccelerated compound is probably due t o a greater content of unextracted zinc.
INDUSTRIAL AND ENGINEERING CHEMISTRY
748
A similar set of experiments was conducted with a stock containing no zinc. In this case the accelerator employed was sparteine which is probably the most active known organic accelerator in the absence of metallic salts. The composition of the compounds is given in Table I11 and the physical properties are shown in Table IV. TABLE111. COMPOSITION OF COMPOUNDS CONTALYING KO
Vol. 26, No. 7
present method would not detect the polymerization of larger units than this. The present investigation has produced no direct evidence in regard to the nature of such dispersing or disaggregating actions as those proposed by Staudinger (IS), Toyabe (15), K6ln (8), and others, which precedes vulcanization. It should be noticed, however, that the most highly vulcanized compound reverts the most rapidly although vulcanized at
ZINC
37-A 100
Extracted pale crepe rubber Sulfur Sparteine Vulcaniaation temp.,
37-B 100
6
C.
--
I
6
1
0
135
145
TABLEIV. PROPERTIES OF COXFOUNDS CONTAIXING No ZINC LOAD o VULCANIZA-At SOOp/, At 15.0 TION Tma elongation break Unextracted Extracted Minutes
Kg. per
COMBIN~D SUI,FUR
sq. cm.
%
COMPOUND 37-A
5 10
20
30
29 37 91 178
61
94 157 204
0.946 0.947 0.948 0.951
0.9157 0.9201 0,9283 0.9371
0.32 1.00 2.18 3.59
0.9215 0.9301 0.9386 0.9465
1.14 2.48 3.69 4.76
COMPOUND 81-B
60
120 180 240
17 33 51 73
64 91 134 161
0.946 0.948 0.950 0.953
The difference in the state of vulcanization between the two compounds in the absence of zinc is less than in the previous examples, but exactly the same relationships are shown. In this case the change in specific gravity before extraction is given and shows that a small decrease in volume takes place during vulcanization. The relationship between specific gravity after extraction and the amount of combined sulfur is shown in Figure 2 and indicates that the density change is due only to combination of sulfur. I
I
L
I
a much lower temperature. This would indicate a change in the vulcanized material as distinct from any change in unvulcanized rubber which might precede vulcanization. In the case of compound 25-A, reversion coincides with the practical disappearance of accelerator. This was shown by acetone extraction which removed an extract of feeble accelerating properties. The absence of accelerator was also indicated by diffusing in a fresh portion of accelerator, after which the normal rate of combination of sulfur was resumed when vulcanization was continued.
CONCLUSIONS The present investigation shows that no direct relation exists between either combined sulfur or density and the degree of vulcanization. It is not probable that the various changes in physical properties of the rubber during vulcanization are due to changes in the polymeric state involving a change in primary valence forces. Neither is it apparent in what manner combination of sulfur has contributed directly to the change in physical properties of the rubber. It is probable that the change in physical state is due to a change in manner or degree of aggregation of the rubber molecules.
Rubber compounds when vulcanized for an extended time frequently begin to decline in modulus. This condition is commonly known as reversion and has been attributed to the dispersing or depolymerizing action of heat. Inspection of the data obtained with compounds 25-A and 39-B shows that the compounds continue to increase in density during cure, even during periods of reversion. This relationship which is shown in Figure 3 might be explained by some type of dispersion rather than by a depolymerization involving chemical valence which would in all probability cause a decrease in density. I t is possible that polymerization could take place beyond the limits which can be measured by the present method. The specific gravity determinations are probably correct, within 5 units of the fourth decimal place. If the writer's former assumption in regard to change in density with loss of double bonds is correct, this accuracy would correspond to one double bond in 400 or to 400 original isoprene units, which would represent a molecular weight of 27,200. The
LITERATURE CITED (1) Axelrod, Gummi-Ztg., 24, 352 (1909). (2) Blake, J. T . , IND.ENQ.CHEM.,22, 741 (1930). (3) Carothers, W.H.,e t al., J. Am. Chem. SOC.,53, 4203 (1931). (4) Conant and Tongberg, Ibid., 52, 1659 (1930). (5) Ditmar and Preuse, Rubber Age (N.Y.), 27, 595 (1930). (6) Fisher and Gray, IND.ENQ. CHEM., 20,294 (1928). (7) Kirchhof, F.,Kolloid-Z., 14,43 (1914);26,168 (1920). (8) K61n, Kautschuk,9, 119 (1933). (9) Kreps, India Rubber J . , 84,293 (1932). (10) McPherson, Bur. Standards, Sci.Paper 560,385-98 (1928). (11) Seidl, Cummi-Ztg., 34,797 (1920). (12) Spenoe and Scott, 2. Chem. Ind. Kolloide, 8 , 308 (1911). (13) Staudinger, H.,KoZZoid-Z., 54, 139 (1931). (14) Stevens and Stevens, J. SOC.Chem. Ind., 51, 44T (1932). (15) Toyabe, Rubber Chem. Tech., 3,384 (1930). (16) Whitby, G.S.,Inst. Rubber Ind. Trans., 6,61 (1930). . , 931 (1925). (17) Whitby and Simmons, IND.ENQ.C H ~ M 17, RFCEIYED -4pril 4,1934. Presented before the Division of Rubber Chemistry at the 87th Meeting of the American Chemical Society, St. Peter~biug,Fls., March 25 to 30, 1934. This paper is Contribution 29. Jackson Laboratory, E. I . du Pont de Nemours & Company.
