Water Vapor Permeability and Sorption of Hevea Latex It-ilrns WILLIAM W. BOWLER The Firestone Tire & Rubber Co., Akron 17, Ohio
I
T HAS been recognized
A
case, the number of grams considerable structural difference exists between Hevea for many years that of water vapor, &, passing latex films vulcanized after drying and those vulcanized Hevea latex films prepared through the area, A , of a in the latex state. Since these two types of films differ from vulcanized latex must film of thickness z in time t considerably in water vapor permeability and sorption, a differ greatly in structure is given by the relation careful study has been made of these phenomena. The from such films when vulweight of water permeating an uncompounded film was canized after drying. Of inversely proportional to the thickness of the film over a particular industrial imwide range of thicknesses and directly proportional to where A p is the vapor presportance is the fact that, A p , the vapor pressure difference across the film, except at sure difference across the films from vulcanized latex very high values of A p . However, films cast from vulcanfilm and P is the constant (prevulcanized films) are ized latex were much more permeable to water vapor, of proportionality called the very weak before complete sorbed considerably lass water, and had a greater negative permeability constant. P drying, yet, when dry, their temperature dependence of permeability than uncomhas the dimensions M * L / tensile strength compares pounded or dry vulcanized films. These experiments LZ.t,(pressure) and in this favorably with that of films indicate that diffusion takes place in prevulcanized films ' study is expressed as grams/ which are vulcanized in by a n activated diffusion through capillary holes and in cm./second/cm. of merthe dry state (dry vuldry vulcanized films by an activated diffusion through the cury. For the sorption of canized films). Hauser polymer itself or the adsorption layer. . water vapor by rubber, the (6), using a micromasorption coefficient, S, can nipulator, discovered that be defined by the expression during liquid vulcanization the Hevea latex particle changes from an easily deformable plastic mass to an elastic particle c=s.p (2) which in overvulcanization becomes crumbly. As a result of where C is the concentration of water in grams of water per CC. a later investigation, Hauser, le Beau, and Kao (7) advanced the theory that prevulcanized films depend for their strength on of rubber, and p is the aqueous vapor pressure in cm. of mercury. the formation of sulfur linkages between particles, and that these The constants, P and 8, can be related ( 4 ) , in the units menbonds cannot form until drying has proceeded far enough to tioned by the expression diminish Brownian motion and to allow the individual particles P =D*S (3) to approach each other more closely than is possible in the wet where D is the diffusion constant. However, D can only be state. calculated from P and S values when the processes represented After a thorough study of vulcanized latex, van Dalfsen (2) in Equations 1and 2 are followed. The permeation and sorption stated that in fresh latex, which has a thick adsorption layer on processes were first studied, therefore, to determine how closely the rubber particles, the high tensile strength of prevulcanized these two expressions are followed for latex films. films is a result of secondary van der Waals forces between the individual particles, These forces are large when the particles EXPERIMENTAL approach each other closely in a dried film but are necessarily The method used to study the permeability of latex films to much weaker in a wet film when the particles are protected by a water vapor consisted of sealing a film across an aluminum cup swollen adsorption layer. In old or purified latex, with a thinner containing water or a saturated salt solution, placing the cup in a adsorption layer, sulfur bonds may form from one particle to the desiccator containing aluminum oxide as desiccant, and deternext, according to van Dalfsen. In a film vulcanized in the mining the rate &twhich the cup lost weight when the desiccator dry state, on the other hand, the tensile strength depends to a was immersed as far as the rim in a constant temperature bath. great extent on the primary valence forces of sulfur bonds. The aluminum cups (Figure 1) were machined from block alumiThese sulfur bonds not only cross link the long, randomly coiled num, and were similar to the design suggested in the A.S.T.M. molecules within each particle, but join molecules from different procedure for testing permeability of plastic materials, A.S.T.M. particles as well. More recently, Humphreys and Wake (8) designation Df397-42T (1). The circular area through which stated that in prevulcanized films the cohesion between particles permeation took place was 7.92 square cm. Water or a saturated can only be due to the van der Waals or secondary valencies. salt solution provided a known vapor pressure, p , , on the lower Since van Dalfsen (a) found that prevulcanized and dry vulside of the film, and the desiccant a vapor pressure, p ~ of, zero canized films differ radically in water vapor permeability, an on the other side. The vapor pressure difference across the film, investigation of this phenomenon was uqdertaken. It was Ap, was then Ap = p l - p z . A few experiments were made hoped that further permeability studies would yield information with the desiccant in the cup and the aqueous solution in the not only on the mechanism of water vapor permeation through desiccator. Both directions of vapor flow yielded the same rate latex films, but also on the structure of the various types of for a given film in every test. The thickness of the film was films. measured by means of a dial gage graduated to 0.001 inch. There The permeation of water vapor through a rubber film can be was no deteetable change of thickness during an experiment. considered to consist of two processes, the sorption of water vapor at one side of the film, followed by diffusion of water vapor To prepare a cup for a run, water was first placed in it or, if through the film and evaporation a t the other side. In the ideal a salt solution was used, the solution was placed in a small glass
April 1952
INDUSTRIAL AND ENGINEERING CHEMISTRY
iw
I
,
ELASTOMERS-Latex container in the bottom of the cup. A circular film, cut slightly larger than the 1.25-inch diameter opening was placed over the mouth, and an aluminum ring, having an inside diameter equal t o that of the cup, was placed on the film and centered by means of a template fitting the raised outer rim of the cup. A heated beeswax-rosin mixture was poured around the edge of the ring t o seal off the edge of the film. The complete assembly, cup, film, and ring, was then placed in the desiccator. A fan was
Parts by Weight, Dry 100.0 1.75 0.5 1.0 0.25 3.0 1 .o
Latex rubber solids Sulfur Zinc diethyldithiocarbamate (ethyl zimate) Zinc salt of mercaptobenzothiazole (Zenite) Di-p naphthyl-p-phenylenediamine (Agerite white) Zinc oxide Potassium hydroxide
Prevulcanization was ordinarily carried out by allowing the compounded latex to stand a t room temperature for 3 weeks; the film was vulcanized in the dry state by heating at 100' C. for 1 hour. From ten determinations on films of slightly different thickness cast from one uncompounded latex, using water to supply a ~p value of 2.38 cm. of mercury, the average value of P was 3.10 X' 10-10 grams/cm./second/cm. of mercury, and the probable error of a single observation, 3~0.675( Z d e / n - 1)ll2,was 0.215 x 10-10. Consequently it was decided that three determinations on one film would yield results of sufficient accuracy for the present study; all measurements were made in triplicate. RESULTS
SORPTION B O T T L E
RING
FILM
WAX
-
CUP
The permeability of latex films was first studied as a function of the several variables in Equation 1. There can be no question of the direct proportionality between Q and t , since assembled cups lost weight a t a constant rate after the initial adjustment period. Likewise, when cups of different sizes were used, the permeability constant was found to be independent of the area. Schultz, Miers, Owens, and Maclay (9) found, in an investigation of thin pectinate films, that P varied with A and attributed this effect to the inability of the desiccant to take up water rapid]! enough when A was large. This difficulty was not encountered in the present study because of the relatively low rate of permestion through the much thicker latex films.
PERMEABILITY CUP Figure 1. Permeability C u p with Ring: Sorption Bottle inserted in the top of the desiccator through a mercury seal in order to keep the air in motion and retain a water vapor pressure of effectively zero at the upper surface of the film. Varying the speed of rotation of the fan did not alter the rate of permeation as long tm a gentle flow of air over the film was maintained. Weighings were recorded only after a constant equilibrium rate of vapor flow was established; this usually required about 3 days. Weighings were then made after successive 24-hour periods. On most films five weighings were sufficient to yield a reliable average. All measurements were made a t 25.0" zt 0.3' C. except in the experiments in which the effect of temperature was studied. It is recognized that with no circulation of air inside the cup, the vapor pressure, PI, a t the lower surface of the film might be less than that corresponding to the saturated aqueous pressure of the solution in the cup. However, the surface of the liquid was in all cases less than 0.5 em. from the film, and at the low rates of permeation encountered the error from this source was assumed to be negligible. Increasing the distance from the film to the surface of the solution in the cup to 1.0 cm. did not alter the permeation rate. Sorption measurements were made by the method of Evans and Critchfield ( 6 ) ,in which a dry film is suspended by a hook in a bottle (Figure 1) containing water or saturated salt solution supplying an atmosphere of known aqueous vapor pressure. Weighings were made by placing the bottle on a saddle across the pan of an analytical balance and suspending the wire holding the film from the end of the balance beam. When constant weight was attained the sorption coefficient was calculated from Equation 2. The films were prepared by pouring the latex, strained through cloth to remove coagulum and air bubbles, directly onto glass plates from a beaker and allowing the film to dry a t room temperature. Ammoniated, once-centrifuged natural latex, approximately 5 months old and of 60% dry rubber content, was used for the casting of most films. Basic compound as follows: 788
I
.
UNCOMPOUNDED
-
2 12
0
4
I6
20
- CM x 10' Figure 2. Effect of Film Thickness o n Permeability Constant, P; A p = 2.38 Cm. H g FILM
THICKNESS
The effect of film thickness, 5, on the permeability constant. P, is shown in Figure 2 for a prevulcanized film and also for a n uncompounded film. Each point is the average of several dctei minations. For the uncompounded film, P varies only slight1)Kith film thickness over a wide range of thicknesses, but P increases greatly with thickness with prevulcanized films. Di ) vulcanized films are entirely similar in their behavior here to uncoillpounded films, but the difference in behavior of the prevuleanizetl films is an indication that water vapor permeation in these film,proceeds by a different mechanism. I n all subsequent uorli, only those films were used whose thicknesses were between 0.04 and 0.06 cm. The effect of Ap on the permeability constant for a prevulc mieed and an uncompounded film is shown in Figure 3. Heie again, P is nearly independent of Ap for uncompounded films, but for prevulcanized films this is true only a t A p values brlow
INDUSTRIAL AND ENGINEERING CHEMISTRY
Vol. 44, No. 4
ELASTOMERS-Latex 1.5 em. of mercury. Above this value P increases very rapidly with A p . I n order to diork in a region in which the permeabilities of these two types of films are of significant difference, then, it was decided in all subsequent experiments to use a saturated solution of Ijotassium nitrate, giving an aqueous pressure of 2.22 cm. of mercury at 25.0' C. Since this is outside the region in which P is independent of A p , Equation 1 is not followed for pre35 vulcanized fibs, and P is not a true 30 proportionality constant. Hence 25 in what follows Q the value calcuX ?RE WJLCANIZED lated from E q u a :: zo 2 tion 1 as P will be designated as t I5 P 1 to show that i t is limited to the
t
'
; $
.
above specific con0. ditions. A pressure slightly below that corresponding to saturation was found to in0.5 1.0 I5 2.0 2.5 crease the reproVAPOR PRESSURE DIFFERENCE 'CMHG ducibility of the results on a single Figure 3. Effect of Vapor Pressure Difference on Permeability Confilm, but the A p stant, P; A p = 2.38 Cm. Hg value of 2.22 om. of m e r c u r y i s ~ i t h i n the region where PI for prevulcanized films is very sensitive to small changes in A p , and results on these films consequently varied from film to film. In another experiment with prevulcanized films P varied with the magnitude of pl and pz as well as with their difference, Ap. A A p value given by low values of p l and pz yielded a lower value of P than the identical A p value given by high values of pl and pz. The different behavior of prevulcanized and uncompounded films at varying A p values is additional evidence for the existence of different structures in these two types of films. The most reasonable explanation for this difference in behavior is that diffusion takes place in prevulcanized films through capillary openings, which, if present a t all in uncompounded or dry vulcanized films, are much sihaller because the relative softness of unvulcanized rubber permits the latex particles to flow together on drying. No significant effect on permeability could be observed when films were electrodeposited or coagulant dipped rather than cast, when the latex was diluted before casting, or when the ammonia content of the latex was varied. The effect of vapor pressure on the sorption of water vapor by uncompounded, dry vulcanized, and prevulcanized films is shown in Figure 4. Since both uncompounded and dry vulcanized film6 have larger S values than do prevulcaniaed films throughout the range studied, the difference must be attributed to a fundamental difference in structure of the prevulcaniaed films, rather than to the effect of compounding ingredients or to the altered nature of the polymer after vulcanization. The explanation for the lower sorption of prevulcaniaed films is not known. As shown in Figure 2, the permeability of prevulcanized films is greater than that for uncompounded films. Yet in the sorption experiments, prevulcanized films were found to sorb considerably less water per gram of rubber. Since the permeation process consists of sorption and diffusion, the actual diffusion rate of water through prevulcanized films must be considerably greater than that for uncompounded films and dry vulcanized films in order to offset the effect of the decreased sorption in prevulcanized
April 1952
Table I. Effect of Age of Latex, Presence of Compounding Ingredients, and Film Treatment on Permeability of Latex Films (Tern] erature
=
25.0° C . ; A p = 2.22 cm. Hg)
P?
Latex Used Uncompounded, 5 months old Uncompounded, 2 years old Uncompounded, e years old Basic compound Basic compound less 8 Basjc compound less ZnO Basic compound Basic compound plus 4.4 part8/100 KC1 Basic compound plus 0.5 part/100 casein Basic compound plus 1.0 partI100 potassium oleate Basic compound plus 3.3 parts/100 KOH Basic compound Basic compound Basic compound Basic compound Basic compound Basic compound
Treatment
G./Cm./ &c./Cm. Hg X 101
No cure No cure No cure No cure No cure Dry cure, looo C. Dry cure, looo C.
2 6 2.0 1.7 2.8 2 6 1.8 1.9
Dry cure, 100" C.
1.9
Dry cure, 100" C.
1.9
loo0 C. Dry cure, looo C.
2.4
Dry cure,
Wet cure, 1 day, 25O C. Wet cure 4 days 25O C Wet cure: 19 daGs, 25' C. Wet cure 45 days 25' C. Wet cure' 21 days ht 25O C. (film sbaked in water 1 week) Wet cure, 2 minutes, 78" C.
11 4.8
11 24 26
2.5 20
films. This is an indication that the difference in structure between -I5 0 - UNCOMPOUNDED the two types of DRY films is actually ; 00 -- VULCANIZED ?RE-VULCANIZLP x very great. Since the permeability of prevulcanized -10 film does not fol6 low Equation 1, and the sorption of water vapor by uncompounded and d r y - v u 1c a n i z e d films does not follow Equation 2, 05 10 I5 EO 25 i t is not possible t o calculate a true ' VAPOR PRESSURE - CM HG Figure 4. Effect of Vapor Pressure diffusion constant on Sorption Coefficient, S; Temperby meansof Equaature = 25" C. tion 3. The permeability, P1,of various films is shown in Table I. An uncompounded film cast from the 5-month old latex had a permeability under the conditions of the test of about 2.6 X 10-10 grams/cm./second/cm. of mercury, whereas this value was decreased about 25y0 with the 6-year old latex. However, since these were different latices, the difference cannot be definitely attributed to the age of the latex. Van Dalfsen ( 3 ) found that film permeability increased with the length of storage of the latex. With the 5-month old latex, compounding and heating, with or without zinc oxide, also reduced the permeability about 25%. Addition of potassium chloride and casein did not alter the permeability of dry cured films; addition of potassium oleate increased PI about 25% and addition of potassium hydroxide, about 500%. Curing latex in the wet state a t room temperature for this particular compound increased the permeability for about 21 days, after which no significant increase took place. A wet cure a t an elevated temperature had a similar effect, although heating for more than 2 minutes a t 78" C. could not be accomplished without the presence of added stabilizer because of coagulation of the mixture. The permeability of a prevulcaniaed film was reduced to about one tenth the original value by soaking in
. .
I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY
-
789
ELASTOMERS-Latedistilled water for 1 week; the permeability of uncompounded and dry vulcanized films was not altered by such soaking. The large increase in permeability of prevulcanized films with increasing precuring time can be attributed to the gradual stiffening of the rubber particles with vulcanization. Data illustrating the effect of stretching on the permeability of two uncompounded films are plotted in Figure 5 . These films were stretched a t various elongations for 1 hour and relaxed for 1 day before the permeability was measured. Under these conditions, crystallization would not be expected to play any important part in the process. The length of time during which the films were stretched was found to have no effect on the subsequent permeability. A film cured after drying suffered no change in permeability after stretching.
0 0
-
which do not pre-exist but are the result of thermal movements in the molecules of the materials composing the film. This type of diffusion undoubtedly takes place through both rubber polymer and the adsorbed layer. The lower sorption of water vapor by prevulcanized films may be a result of the continual desorption of water molecules into the large capillary spaces. Such desorption is impossible in uncompounded or dry vulcanized films, which do not have these large capillaries. In uncompounded films diffusion probably takes place by essentially the same mechanism as in dry vulcanized films. However, small capillaries, probably destroyed by heating in dry vulcanized films, must exist in uncompounded films, and some diffusion would then occur by the same mechanism as that in prevulcanized films. The decrease in permeability of uncompounded films on stretching would then result from the destruction of these residual capillary spaces, and the permeability of the stretched film Lends to fall to the value expected for a dry vulcanized film.
f FILM
I
FILM
2
-30
- 25 0
I
100
200
300
400
500
I
-20
600
I
X c
P E R C E N T ELONGATION
Figure 5. Effect of Stretching on Permeability of Uncompounded Films; A p = 2.22 cm. H g In Figure 6 are plotted the measured Pl values for uncompounded, dry vulcanized, and prevulcanized films a t four different temperatures from 15" to 45' C. The difference in the behavior of prevulcanized films is striking and indicates again that prevulcanized films are significantly different in structure from uncompounded and dry vulcanized films.
U U
5B
-I5
:
_p.'
-10
DISCUSSION
According to Doty, Aiken, and Mark ( 4 ) , three types of diffusion through polymer films can be recognized: (1) gaseous diffusion through capillary holes; (2) activated diffusion through the polymer itself; and (3) activated diffusion through capillary holm whose walls are lined with active centers of high adsorption power. IB seems reasonable to assume that diffusion takes place in prevulcanized films primarily by this third mechanism, since i t is likely that large capillaries, lined with hydrophyllic material, exist in such films as a result of the relative stiffness of the vulcanized rubber particles. This assumption would explain the great increase in the permeability constant with film thickness for prevulcanized films. It is only with pure gaseous diffusion, with no forces of attraction between diffusing molecules and the film, that P would be expected to be independent of film thickness. This assumption would also explain why, for prevulcanized films, P varies with the magnitude of pl and p2. At low values of p , and pz the capillaries will be lined mainly with hydrophyllic surface material, whereas a t high values of p l and p2 with adsorbed water molecules, and diffusion could conceivably be much greater through such water-lined capillaries. Likewise the reduction in permeability of prevulcanized films after soaking in water would follow as a result of the washing out of the hydrophyllic material containing the active centers for adsorption. On the other hand, diffusion in dry vulcanized films probably takes place b y an activated diffusion mechanism of a different type, the water molecules moving through temporary "holes"
790
I_.TEMPERATURE.
Figure 6.
Or.
Effect of Temperature on Permeability
The effect of temperature on permeability can be attributed, a t least in part, to a difference in the energy of activation for the two different diffusion processes. This temperature effect may also result from a positive temperature dependence of desorption of water molecules from active centers into the larger capillaries of prevulcanized films. The question of whether the strenglh of prevulcanized f i l m is due to secondary forces between the individual latex particles or to sulfur linkages between the particles is a difficult one. At first glance, one would expect that a prevulcanized film, if secondary forces between the particles are responsible for its strength, would resemble an uncompounded film in stress-strain properties. Actually a prevulcanized film is more nearly similar to a dry vulcanized film, in which there is present EL three-dimensional network of primary valence bonds. Humphreys and Wake (8) offer as evidence against the formation of interparticle sulfur bonds the collapse of a prevulcanized film into discrete particles when swollen in benzene. With films which are several months old, however, there is not such a great difference in behavior between those prevulcanized and those dry vulcanized, on swelling in benzene. Both types of films swell, become weaker, and recover a great part of their strength on drying. The prevulcanized film
INDUSTRIAL AND ENGINEERING CHEMISTRY
Vol. 44, No. 4
ELASTOMERS-Lateu becomes slightly weaker in the swollen state but crumbles on the application of large shearing stresses only to a slightly greater extent than the dry vulcanized film. When a vulcanized latex dries, sulfur and the other compounding ingredients are deposited a t the interfaces between particles. The extent to which sulfur bonds form from one particle to the next would seem to depend on the amount of interfering adsorption material present and on the subsequent treatment of the film. If little or none of such interfering material is present, it seems difficult to suppose that interparticle sulfur bonds would not form on subsequent heating of the film or even on standing a t room temperature for a sufficient length of time. If such sulfur bridges do exist between particles, then the problem of explaining the high tensile strength of prevulcanized films is much less difficult, since both prevulcanized and dry vulcanized latex films then possess an essentially similar primary valence network, despite the presence of larger capillary openings in the prevulcanized film. It would not be necessary for bonds t o form a t every interface between particles for a complete three-dimensional network to exist.
ACKNOWLEDGMENT
The author wishes to express his appreciation to L. A. Wohler and Ernst Schmidt, both of the Firestone Tire and Rubber Co., for their advice during the course of this work. LITERATURE CITED
(1) A.S.T.M. Standards, Part 111, Nonmetallic Materials, p. 1697, Figure 3 (c), 1944. (2) Dalfsen, J. W.van, Rubber Chem. and Technol., 16, 318 (1943). (3)Ibid., p. 388. (4) Doty, P. M.,Aiken, W. H., andaMark, H., IND.ENG.CHEM., ANAL.ED.. 16. 686 (1944). ( 5 ) Evans, W. D., and Critchfield, C. L., J . Research NutZ. Bur. Standards, 11, 151 (1933)(RP683). (6) Hauser, E.A.,India-Rubber J., 68, 725 (1924). (7) Hauser, E. A., le Beau, D. S., and Kao, J. Y . L., J . Phys. C h m . , 46, 1099 (1942). (8) Humphreys, N. C. H., and Wake, W. C., T ~ a n s Inst. . Rubber Ind., 25, 334 (1950). (9) Schultz, T.H., Miers, J. C., Owens, H. S., and Maclay, W. D., J . Phys. Colloid Chem., 53, 1320 (1949). RWEXVED for review Saptember 17, 1951.
ACCEPTED January 26, 1962.
GR-S LATICES IN F O A M RUBBER LEON TALALAY AND ANSELM TALALAY The Sponge Rubber Products Co., Shelton, Conn.
T
This procedure is basedon H E advent of low temT h e superiority of cold OR-Shigh solids latex over latex U. S. Patent 2,432,353 (IS). perature polymerized polymerized at 120' F. has been demonstrated in foam It was found to produce conGR-S latex has greatly benerubber in terms of greater tensile strength, better elongasistently foam rubber of fited the foam rubber techtion at break, higher modulus, improved rebound elasgood structure from a great nology. Prior t o the deticity, and enhanced low temperature properties. The variety of polymers. velopment of high solids cold The standard curing formagnitude of the effect observed is far greater than the mula contained uniformly 5 latex, foam rubber has been corresponding improvement reported in tire tread stocks partsof zinc oxide, l l / z parts made, of necessity, predom(5, 15). A number of polymer modifications in low temof antioxidant (Agerite inantly from Hevea. Such perature polymerized latex were investigated. I t was White), and l l / a parts each GR-S latex, as had been shown that by the proper choice of styrene content (apof zinc diethyldithiocarbac used, was blended in largely mate and zinc mercaptobenproximately 10% bound), OR-S foam rubber superior to zothiazole per 100 parts of for economic reasons, perHevea foam in subzero properties can be made. The elastomer. The sulfur was forming a function not modulus of foam rubber improve'd significantly with vraied from 1 t o 3 parts, as much different from a dilrising Mooney viscosity of the contained polymer, while indicated on the individual uent or extender. The supegraphs. the elongation at break was affected adversely. The deMinor adjustments in riority of cold GR-S latex gree of conversion was found to have little bearing on the soap stabilization, alkali over hot in all physical propproperties of the resultant foam rubber. content, and viscosity were erties, its better wet gel made, to compensate for strength, and the much revariations in the colloid duced odor level (9). . have serviceable foam rubber made entirely of GR-S. It is, therefore, COMPARISON OF HEVEA, 41' F. GR-5, AND 120' F. OR-S unfortunate that as yet the production of cold high solids GR-S Type V GR-S was chosen as a typical 120' F. latex. It was latex is so limited. compared to a 41 'F. latex of the same charge ratio (70-30 butadiA study was undertaken to compare the physical properties of ene-styrene) and Mooney viscosity (MS-4') of 53. The latter foam rubber made from Hevea latex with foam produced from was a pilot plant latex, made by the Copolymer Corp. as PF 500high solids GR-S latices, polymerized a t 41 ' and 120' F., respec489. It is referred t o as polymer 19 in Table I. Foam latex, made tively. In addition, a number of polymer modifications were infrom 62% centrifuged Hevea latex, was used as a standard of refvestigated in low temperature latex. erence. The physical properties of the foam rubber evaluated included Stresestrain measurements were made by the method detensile strength and elongation a t break, modulus of compression, scribed by Conant and Wohler ( I ) . The plotted results are averflexibility a t subzero temperatures, and rebound elasticity. ages of the best six out of ten data a t optimum cure. The tensile values obtained, in pounds per square inch, were PREPARATION OF SAMPLES divided by the density in pounds per cubic inch, and are so plotted Slab foam samples of 1 inch thickness were molded a t a density in Figure 1. This method compensates for minor density difof approximately 0.003 pound per cubic inch. ferences between samples, and is permissible since the tensileThe compounded latex was foamed by the catalytic decomposition of hydrogen peroxide and then rapidly frozen. The frozen density relationship is linear within the range encountered (0.003 structure was elled by permeation with a coagulating gas, carpound per.cubic inch f 10%). bon dioxide. %he gel waa vulcanized at 215" F., removed from In the left half of the graph the foam tensile density is plotted the mold, washed, and finally dried a t 180' F. in circulating hot against parts of sulfur per 100 parts of elastomer. There is a difair. , I
April 1952
I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY
791
