488
INDUSTRIAL AND ENGINEERING CHEMISTRY
With fillers steam-reduced asphalts were more favorable than oxidized asphalts; without fillers the reverse was true. 4. All coatings on red lead primer were superior to those on polished steel. 5. I n general, the improvement was proportional to the polybutene content, the polybutene film of higher molecular weight having better physical properties; but films containing over 10 per cent polybutene were either too tacky, for the low-molecular-weight polybutene, or too viscous for easy application in the case of the polybutene of higher molecular weight.
Vol. 35, No. 4
ACKNOWLEDGMENT
The advice, encouragement, and assistance of G. W. Oxley during the course of this work is greatly appreciated. LITERATURE CITED
(1) Staudinger, H., "Der Aufbau der hochmolekularen organisohen Verbindungen", 1932. (2) Thomas, R. M., Zimmer, J. C., Turner, L. B., Rosen, R., and Frolich, P. X., IND. EXG.CHEM.,32, 299-304 (1940). (3) Traxler, R. N., and Coombs, C. E., Proc. A m . SOC.Testing Materials, 37, 11, 549 (1937). PRESENTFAD before the Division of Paint, Varnish, and Plastics Chemistry a t the 104th Meeting of the AMERICAN CHEMICAL Socxnry, Buffalo. N. Y .
Brittle Temperature of
Rubber under Variable Stress A. R. KEMP, F. S. MALM, WINSPEAR
AND G. G.
Bell Telephone Laboratories, Murrab- Hill, N.J
HE projected use of synthetic elastomers on a large scale for outdoor service presents a pressing need for information dealing with their properties a t subzero temperatures. Under extreme cold weather conditions insulation, cable jackets, tires and tubes, and many other items must be capable of being (repeatedly) flexed without failure a t temperatures of -50" C. or lower. I n a previous paper (25) a method was presented for determining the critical temperature of fracture where the sample is bent rapidly through an angle of approximately 90". Although this test provides an excellent technique for the comparison of various rubber compositions, it was realized that the procedure should be revised to include the effect of varying the magnitude of stress as well as the rate of bending in order t o simulate more closely certain types of service. The present paper does this and also reviews some of the work already carried out in the field of low-temperature rubber testing. Early investigation (21) showed that crude rubber inoreased in tensile strength a t low temperatures. Le Blanc
T
~~
and Kroger (17) described this phenomenon as cold vulcanization. The observation that rubber which had been elongated and cooled in liquid air could be disrupted into fibrous fragments was reported by Hock (11). Mark and Valko (19) associated this phenomenon with brittleness and conducted experiments which showed that, under given test conditions, the "critical point" of raw rubber was -67" C. Ruhemann and Simon (89) observed that a sharp rise in the specific heat of smoked sheet rubber occurred in the range -65" to -75" C. Bekkedahl (1) found that a t -70" C. a sharp change of slope occurs in the temperature-volume curve with both supercooled amorphous and crystalline raw rubbers. This observation is important because it demonstrates that the state of brittleness develops at the same point independently of the existence of either physical state. Differences in the brittle points of crystalline and amorphous smoked sheet rubber, as determined by the fracture method described in the previous work (25),were found to be slight. Aside from the fracture method ( I S , 16,2626)other procedures used in investigations of rubber at low temperatures may be grouped according to the method of test into three general types involving deflection (15,18),penetration (go), and elastic
~
This paper supplies the need for a method to determine the temperature at which rubber and similar materials fracture under variable bending stress. Although the brittle temperature is sharply defined under high-speed bending through a sharp angle, it is lower as the speed of application or the magnitude of the stress is reduced. In some instances decreases of more than 28' C. in brittle temperature resulted from
reductions in bending stress such as might be encountered in service. Vulcanized pure gum natural rubber and plasticized polyvinyl chloride-acetate copolymer showed the largest changes, whereas the compounded and vulcanized natural and synthetic rubbers involved in this study exhibited a reduction in brittle temperature from 5' to 10' C. in going from the highest to the lowest stress employed.
April, 1943
INDUSTRIAL AND ENGINEERING CHEMISTRY
489
deformation (7, 8, 9, 14, 24, 96). The method of Koch (16) may be considered a representative test for measuring the resistance to deflection of an elastic material at low temperatures. Penetration measurements by a dead-weight indentation method as reported by Nagai (20)represent this type of low temperature test. The work of Sagajllo et al. (24) demonstrated the combined effects of elastic deformation and temperature lowering on rubber. This testing procedure is similar to the T-50 method (9) commonly used for determining the state of vulcanization in certain flat-curing carbon-black-rein-forced rubber compounds; the interpretation of the data is the important difference. Although the deflection, penetration, and elastic deformation tests detect appreciable changes in certain properties ofrubber at low temperatures, it would Figure 1. Apparatus for Determining Critical Temperature of Fracture be difficult to predict the performance on Bending of a given material tested by these methods with reference t o its being subjected to a sudden bending stress. Experience in manipulating the acetone-solid carbon dioxide The deflection method used by Koch (16) defines the conditioning bath has shown that certain precautions are necespoint of maximum stiffness in rubber and gives indications of sary to assure accurate temperature control. The procedure conincreased resistance to deformation at temperatures above the sists of placing 3.5 pounds (1.59 kg.) of crushed solid carbon dioxide in the insulated tank which is 14 X 2 X 8 inches (35.56 X brittle point, but woultl not permit the prediction of how a 5.08 X 20.32 cm.) and adding acetone in small quantities until a given material would respond t o severe or rapid mechanical saturated solution is obtained as indicated by a temperature of deformation in service. The data published by Nagai (20)on -78' C. Circulation is then induced by a motor stirrer, and the decreasing penetration values observed with lowering of the total acetone added is increased to 2 liters. Approximately 30 temperature are of little value aside from furnishing a rough minutes are required to reach a condition of e uilibrium where an immediate lowering of the temperature is notel upon the addition indication of increased hardness under these conditions. It of a small quantity of the pulverized refrigerant. The quantities will be shown later that hardness tests do not correlate with of a refrigerant and liquid mentioned roduce an equilibrium brittleness tests on bending. In determinin the temperature of approxunately -70' brittle fracture point, it is standard practice to establisa the approximate temperature of failure by preliminary tests a t 10" C. intervals and then continue testing a t 1' C. ascending intervals APPARATUS AND METHOD until the temperature of survival is noted. The heat transfer in In the improved apparatus shown in Figure 1the quadrant, A , the system resulting from inserting a test specimen and reimupon which the fixtures, B, for varying the bending stress are mersing the quadrant produces a rise of less than 1' C., and the mounted, is rotated by a worm drive, C, which utilizes intertemperature rise between -60' and -50' C. averages about changeable gears to vary the rate of deformation. This mechan0.25' C. per minute. Exactly 5 minutes are allowed for the conism is driven by a l/la-horseponer squirrel-cage motor, operatin ditioning of each test s ecimen exce t in the case of highly at 1140 r. p. m. and using a four-thread twelve-pitch worm whe8 plasticized materials, mc\ aa Korosear and Vinylite, which are of thirty teeth with 2.5-inch (6.35-cm.) pitch diameter and a tested after a 2-minute immersion to minimize the ossible effects sixty-tooth &inch (12.7-em.) pitch diameter wheel with a correof solvent action on their structure. The use ofboth acetone sponding worm gear; this arrangement gives quadrant speeds of and ethyl alcohol as cooling media have been found to produce 150 and 75 r. p. m. The fast s eed was selected on the basis of results which check closely with those from similar tests conduplication of results obtained for a large variety of rubber and ducted in an air atmosphere after longer conditioning periods. synthetic elastomer Compositions in tests usin the manually operated apparatus previously described (2%). %he ri ht-angle BRITTLE TEMPERATURE OF RUBBER COMPOSITIONS bend is obtained by settin the rigid breakin arm, D,t.25 inch (6.3 mm.) beyond the arc Sescribed by a bloc% 0.25 inch s uare; Table I contains brittle fracture point data for pure gum the block is mounted on the eriphery of trhequadrant a n a flush in which the test s ecimen, F, is and high-quality mechanical rubber compositions tested under with the back of the slot, inserted. When testin by the application of a &ending stress controlled conditions of variable bending stress and rate of controlled in intensity %y a selected arbor, a stri of 0.01-inch deformation. The results as stated were found by numerous (0.254-mm.) spring steel, G,is inserted in the slot gefore the test check tests on the same materials to be reproducible within specimen; the clearance of the breaking arm is adjusted so that in the test cycle the snmple is bent rapidly to conform with the 5t0.5" C. It was shown, however, in our previous work t h a t rofile of the arbor. The immersion tank is insulated H,by a raw rubber from different sources varies in brittle point over a Souble wrappin of aluminum foil, a 0.5-inch (1.27-cms layer of range of about 4" C. which results in a wider limit of reprofiber glass, a n f an asbestos board enclosure. At the brittle ducibility from batch t o batch. fracture point, determined by rotation at the fast speed with an arbor of 1-inch (2.54-cm.) radius, three or four breaks usually With the decreased magnitude of the bending stress and occur in a specimen 1.6 inches (3.81 cm.) in length, and a temrate of deformation, the pure gum wlcanizate does not fail at perature is found where a iven material will be intact after a -78' C., but the brittle fracture point of the carbon-blackbending cycle but will fail &en this temperature is lowered 1" C. reinforced rubber is lowered only 6.5" C. by the less severe When this point is determined, five specimens are tested successively at a temperature 1O C. above the point of failure; if all conditions of test, and a further decrease in rate of deare intact after being put through the test cycle, the brittle fracformation b y using a quadrant speed of 37.5 r. p. m. €ailed ture point is stated as the temperature between the observed t o lower further the temperature of fracture for the latter points of survival and failure. A calibrated Weston dial thermaterial. mometer was used for temperature measurements in this work.
8.
2
490
INDUSTRIAL AND ENGINEERING CHEMISTRY
TABLE I. EFFECTS OF VARIABLE BENDING STRESS AND RATEOF DEFORMATION ON BRITTLE FRAC~URE POINTS OF RUBBER VULCAh'IZSTES
Compound No. Smoked sheet Mercaptobenzothiazole Zinc dimethyl dithiocarba mate Sulfur Stearic acid Zinc oxide Channel black Semireinforcing black Phenyl-or-naphthylamine Sunlight aging inhibitor
1 100.00 0.50
... ...
2 100.00 0.75 0.10 3.00 2.00 10.00 30.00 30.00 2.00 2.00
Cure, min. (" C.)
SO(l26)
ZO(141.5)
Rate of Deformation (Quadrant Speed), R. P. M.
Radius of Bending Stress,
...
3.50 0.50 6.00
... ...
Brittle Fracture Point, C. Carbon black stock
Cm.
Pure gum
150
Right angle 2.54 3.81
-56.5 -6J.5
-58.5 -60.5 -63.5
75
Right angle 2.54 3.81
-58.5 -6J.5
-60.5 -62.5 -65.0
KO failures observed at -78O C. (-108.4'
F.).
Brittle fracture point data under variable stress for a series sf reclaimed rubber compositions (Table 11) showing the effects of adding semireinforcing black, clay and whiting, mineral oil, blown asphalt, and smoked sheet. The brittle point of the uncompounded reclaimed rubber \vas -490 c. These data show that the temperature of fracture varies with both the speed and magnitude of the bending stress. previously found ($5), the nature of the rubber used is important than minor changes in the undoubtedly lolver speeds or less severe bending stresses than those employed in these tests would result in still lower ternperatures a t which these compositions would fracture. BRITTLE TEMPERATURE OF SYNTHETIC ELASTOMERS
Vol. 35, No. 4
Fuoss (5) observed that plasticized polyvinyl chloride behaves similarly to rubber in that, a t a certain definite temperature, a sudden change in mechanical and electrical properties is noted. He refers to this point a t which brittleness occurs as an internal melting point. Since the unplasticized material is in a general class of hard plastics, including cellulose acetate and polystyrene which are also extensively used in their unplasticized form and as such are evaluated by their resistance to plastic flow, this terminology is correct. However, in the case of the plasticized materials which are useful in their flexible and elastic condition, the term "brittle point" is a more significant designation. Davies, Miller, and Busse (3), also studying the electrical properties of this material, stated that three variables-temperature, frequency of current, and ratio of plasticizer to polyvinyl chloride-can be made to produce roughly equivalent results. The chemical significance of plasticizer action on these materials is beyond the scope of this work; however, reference may be made to the work of Houwink (22) who suggests that the brittleness of a substance will depend upon the distance over which the interacting molecular forces work. Considerable insight on the mechanics of plasticizer action with particular reference to cellulose acetate is given in a recent article by Gloor and Gilbert (10). Table I11 lists brittle fracture data for plasticized polyvinyl chloride and P o b i n Y l chloride-acetate copolymers which may be considered aT'era,ge materials of their types. The samples tested were strips 1.75 X 0.5 X 0.075 inch (4.45 X 1.27 x 0.19 cm.) cut with scissors from sheets molded from the calendered mat'erials under a pressure of 500 pounds per square inch (89.25 kg. Per Sq. cm.1 for 5 minutes at 141.5' c. and cooled for 5 minutes under pressure. The immersion time in the acetone-solid carbon dioxide bath was exactly 2 minutes for each test, and the brittle fracture points stated were verified by the observation of five consecutive cases of survival a t temperatures 0.5' C. above the stated points. Some difficulty was encountered in testing these .materials. I n the various test cycles occasional cases of nonfailure were observed as much as 10' C. lower than the brittle fracture points stated: thus there is the possibility of a heterogeneous condition in plasticized mixtures of the types investigated. VULCAKIZABLE SYNTHETIC ELASTOMERS. Synthetic elastic polymers which more closely resemble rubber include butadiene polymers, acrylonitrile and styrene copolymers of butadiene, polychloroprene and its diolefin copolymers, polyiso-
I n view of the increasing amounts of synthetic elastomers now becoming available to the rubber industry and the immediate possibility of their extensive use, the need for information pertaining to their low-temperature properties becomes urgent. I n contrast to rubber, considerable improvement may be effected in some of these materials by the use of certain compounding ingredients. Wide variations in brittle points were observed during the previous work (25) in different TABLE11. BRITTLE FRACTURE POINTSOF ALL-RECLAIM AND types of the same general class of material, LOW-RUBBER-CONTENT COMPOSITIONS~ as well as in the same material from different Compound No. 3 4 5 6 7 8 sources. Whole tire reclaim 67.50 67.50 67.50 67.50 62.50 57.50 black 21.66 ... 21.65 21.65 21.65 SYNTHETICTHERMOPLASTIC ELASTOMERS. Semireinforcing .. . 3i:bo ... ... ... Hard clay Considerable information is available on the Ground Light-process limestone oil ... .., 32:40 ,., ... ... . . . 7.00 7.00 7.00 7.00 7.00 low-temperature properties of polyvinyl chloride Mineral rubber (m, p. 310' F. or 154' C.) ... ... ... 7.00 .. . ... which, although a hard plastic a t room temSulfur 1.00 1.00 1.00 1.00 1.00 1.00 Tetramethylthiurammonosulfide 0.15 0.15 0.15 0.15 0.15 0.15 perature, can be converted into a flexible and Di-o-tolylguanidine 0.10 0.10 0.10 0.10 0.10 0.10 elastic material by the incorporation of chemiSunlight aging inhihitor 2.00 2.00 2.00 2.00 2.00 2.00 Antioxidant 0.30 0.30 0.30 0.30 0.30 0.30 cal plasticizers in sufficiently large quantities. Smoked sheet ... ... ... .. , 5.00 10.00 Russell ($3) reported that the p form of Rate of polyvinyl chloride is brittle at 81' C., but Dei or mation (Quadrant Radius of that a mixture consisting of 60 parts of this Speed), Bending Stress, R. P. M . Cm. Brittle Fracture Point, C. material and 40 parts of dibutyl phthalate by I50 Rightangle -47.5 -46.5 -47.0 -43.5 -47.5 -50.0 weight (Koroseal) exhibited a brittle point of 2.54 -50.5 -51.5 -51.5 -46.0 -51.5 -51.5 -46" C. His method for brittle point deter3.81 -52.5 -53.5 -53.5 -49.5 -:3.5 -55.5 75 Rightangle -50 -49.5 -49.5 -44.5 -91.0 -;1.5 mination consisted of conditioning bars of the 2.54 -51.5 -55.5 -53.5 -46.0 -52.5 --53.5 materials for 5 minutes in an alcohol bath 3.81 -54.5 -57.5 -56.5 -50.0 -54.5 -56.5 at the temperatures stated, removing them, a Cured 20 minutes at 141.5O C. and immediately bending them double. 7
O
I N D U S T-FI I A L A N D E N G I N E E R I N G C H E M I S T R Y
April, 1943
TABLEIII. BRITTLE FRACTURE POINTSOF PLASTICIZED POLYVINYL CHLORIDE AND POLYVINYL CHLORIDE-ACETATE COPOLYMER Rate of, Deformation (Quadrant Speed) R. P . M. I
150 75
a
Pol vinyl chgride polymer 9
Right angle 2.54 3.81 Right angle 2.54 3.81
-25.5 -32.5 -46.5 -30.0 -38.8 -51.5
-50.0 -70.5 G
-56.0 -7,2.0
No failures observed at -78" C. ( - 108.4' F.1.
TABLEIV.
BRITTLE FRACTURE POINTSOF BUNAS COMPOS1TlONS" 11
Compound No. Bune S Smoked sheet Whole tire reclaim Carbon black (special) Blown asphalt
Rate of. Deformation
(%2$ft R. P. M.
12
13
...
50.00 50.00
5,j:,jo
50.00 25.00 50,00 50.00
25.00
25.00
25.00
100.00
gi.bo 2.00
Sulfur Stearic acid Zinc oxide Aocelerator A Aceelerator B
0
Brittle Fracture Point, O C. Polyvinyl chlorideacetate copolymer 10
Radius of Bending Stress, Cm.
1;:
1.50
Radius of Bending Cm.Stress'
'.O0 2.00
;:::1
2'oo 2.00 If:::
1.50
1.50
Brittle Fracture Point,
C.
150
Right angle 2.54 3.81
-51.5 -54.5 -59.5
-51.5 -52.5 -58.5
-49.5 -51.5 -53.5
75
Right angle 2.54 3.81
-54.5 -58.5 -63.5
-55.5 -56.5 -60.5
-51.5 -53.5 -57.0
Cured 20 minutes at 141.5'
c.
butylene and its diolefin copolymers, and various types of organic polysulfides. These materials, mixed in various proportions with vulcanizing agents and typical rubber compounding ingredients and vulcanized by heating, exhibit widely varying low-temperature properties (%) owing to their different chemical structures. Koch (16)applied his deflection test to an undisclosed butadiene-styrene copolymer composition and observed a freezing point of -66" C., which is in agreement with the brittle point range of -66" to -70" C. for this material observed in the previous work (25). Bekkedah1 and Scott (2) recently reported a second-order transition point of -23" C. for Hycar OR (modified butadiene-nitrile copolymer), using the technique previously described in theii work on rubber. These authors also stated that no firstorder transition was observed in the material studied. The brittle point of this material was found in our earlier work to be about -25" C. Garvey, Juve, and Sauser (6) reported low-temperature observations for a number of Hycar OR compositions. Table IV shows the brittle fracture points under variable stress of three butadiene-styrene copolymer compositions. These results are similar to those in Table I in respect to the effect of varying the nature of the stress. It was previously shown (25) that a good Buna S tire tread stock (without the blown asphalt) had a brittle fracture point of -66" to -70" C. under a right-angle bend a t high speed. These results show that the brittle temperature of synthetic rubber is more sensitive to compounding than natural rubber. For this reason greater care must be taken in compounding Buna S for low-temperature service. Yerzley and Fraser (27) recently investigated the effects of low temperatures on the Shore A hardness, elastic deforma-
'
491
tion, and mechanical deflection (torsion) of neoprene compositions: they proposed the term "freeze factor" to express the ratio of the observed change in hardness produced by low temperatures to the maximum possible increase in hardness for a given material. A new interpretation of this effect, designated as "F-50" (4, was recently proposed to define the temperature where half the maximum possible increase in Shore A hardness is observed. A comparison of data obtained by Shore A hardness measurements and brittle fracture point tests for the same series of Neoprene FR compositions is shown in Table V along with the Shore A hardness values for the various compositions a t their respective brittle fracture points. The fracture test indicates that compound 17, containing 30 parts of dibutyl sebacate, would be serviceable a t a temperature 5" C. lower than the best of the series as indicated by the hardness observations. Composition 17, subjected to bending tests a t a quadrant speed of 37.5 r. p. m. over a 3.8-cm. radius arbor, showed no failure a t -76" C.; compound 14 exhibited a brittle fracture point under these conditions at -67" C. The addition of dibutyl sebacate and diisobutyl adipate resulted in an appreciable lowering of the brittle fracture points of a typical neoprene GN composition (Table VI); the order of improvement was the same as in the series of tests with Neoprene FR compounds. Cooling curves made for the two plasticizers showed that both are in a solid state at temperatures above the fracture points of the compounds in which they were used, and that freeiing occurs well within the conditioning time allowed according to the testing procedure. Swelling tests on vulcanized base compound 19 in diisobutyl
TABLEV. COMPARISON OF LOW-TEMPERATURE PROPERTIES OF NEOPRENE FR COMPOSITIONSa AS DETERMINED B Y SHORE A HARDNESS AND BRITTLE FRACTURE TESTS Com ound $0.
14 15 16 17 18
Softener (30 Parts by Wt.) Diisobutyl adipate Dibutyl phthalate Coal-tar eoftener Dibutyl sebacate Light-process oil
F-50 Temp., Brittle Frtcture O C. Pointb, C.
Hardness at Brittle Fracture Point
-54
-58 5
80
-46
-61.5
92
- 43 - 40 -38
-50 5
93
-66.5
92
-53 0
95
a Base formula: Neoprene FR 100.0, stearic acid 1.0, Neozone A 2.0, soft black 100.0. sulfur 1.0, litharge 10.0. b Rete of deformation, 150 r. p. m.; right-angle bend.
TABLEVI.
BRITTLEFRACTURE POINTSOF NEOPRENE GN COMPOSITIONS
Compound No. Neoprene G N Magnesium oxide Zinc oxide Antioxidant Semireinforcingblack Petroleum softeners Stearic acid Accelerator Diisobutyl adigate Dibutyl sebaoate Rate of Deformation (Quadrant Speed), R. P. M.
19
20
21
100.0
100.0 7.0 2 0 1 0 35.0 2 0 0.25 0.25 15.0
100.0 7.0 2.0 1.0 35.0 2.0 0.25 0.25
7.0 2.0 1.0 35.0 2 0 0.25 0.25
...
15.0
Radius of Bending Stress, Cm.
Brittle Fracture Point,
150
Right angle 2.54 3.81
-41.5 -45.5 -46.5
-51.5 -56 -57
7s
Right angle 2.54 3.81
- 43
54 -56
-46.5 -60.5
-58
C.
-55.5 -58.5
- 60 - 57 - 59 - 62
Ternary Liquid
adipate and dibutyl sebacate showed appreciable increases in weight after 12-hour immersion a t 60” C. The neoprene immersed in dibutyl sebacate increased 80 per cent by weight as compared to a 70 per cent increase for the sample immersed in diiiobutyl adipate. On the basis of these observations and the brittle fracture test results, it appears that increased solvent action may offer a better guide to the selection of materials for lowering the brittle fracture point than the freezing point observations on the plasticizers.
A Method of Tie Line Interpolation
CONCLUSIONS
1. The temperature a t which natural and synthetic elastomers fracture on bending depends on the rate of application and the magnitude of the stress applied. The slower the rate of bending and the less the angle of bend, the lower will be the temperature of fracture. 2. A study of the stresses under varying types of service a t subzero temperatures must be made in order to select intelligently the laboratory test conditions which will best simulate performance in the field. 3. I n the case of synthetic elastomers having high fracture temperatures, the addition of certain types of plasticizers serve to correct this difficulty.
CHARLES E. DRYDEN’
RAPHICAL methods for the design of rectification
G
ACKNOW LEDGiMENT
The authors wish to acknowledge the assistance of W. H. Lockwood who performed a large part of the testing in connection with this paper. LITERATURE CITED
(1) Bekkedahl, N., J . Research Natl. Bur. Standards, 13, 411 (1934). (2) Bekkedahl, N., and Scott, R. B., Ibid., 29, 87 (1942). (3) Davies, J. M., Miller, R. F., and Busse. W. F., J. Am. Cheni. Soc.. 63, 361 (1941) (4) Fraser, D., du Pont Rubber Chemicals Div , Rept. 42-1 (1942). (5) Fuoss, R. M., J . Am. Chem. Soc., 63, 369 (1941). (6) Garvey, B. S., Juve, A. E., and Sauser, D. E., IND. ENG.CHEM., 33, 602 (1941). (7) Gee. G . , and Treloar, L. R. G., Trans. Inst. Rubber Ind., 16, 184 (1940). (8) Gehman, S. D., J . Applied Phys., 13, 402 (1942). (9) Gibbons, W. A., Gerke, R. H., and Tingey, H. C., IND.ENG CHEM.,AKAL.ED., 5, 279 (1933). (10) Gloor, W. E., and Gilbert, C. B., Ibid., 33, 597 (1941). (11) Hock, L., Z . Elektrochem., 31, 404-9 (1925). (12) Houwink, R., “Elasticity, Plasticity and Structure of Matter”, p. 71, Cambridge Univ. Press, 1937. (13) I. G. Farbenindustrie, Kunstofe, 28, 171 (1938). (14) Khvostovrtkaya, S., and Margaritov, V., J . Rubber Ind. (U. S . S. R.), 8, 231 (1933). (15) Koch, E. A., Kautschuk, 16, 151 (1940). (16) Kohman, G. T., and Peek, R. L., IND.ENG.CHIM., 20, 81 (1928). (17) Le Blanc, M., and Krager, M., Kolloid-Z., 37, 205 (1925). (181 McCortney, W. J., and Hendrick, J. V., IND.ENQ.CHEM.,33, 579 (1941). (19) Mark, H., and Valko, E., Re*. gen. caoutchouc, 7, 11 (1930). (20) Nagai, H., J. SOC.Rubber Ind. Japan, 8, 397 (1935); 9, 147 (1936). (21) Polanyi, M., and Schob, A., Mitt. Materialprfijungsamt, 42, 22 (1924). (22) Ruhemrtnn, M., and Simon. F., Z.physik. Chem., 138A, 1 (1928). (23) Russell, J. J., IND. ENG.CHIOM., 32, 509 (1940). (24) Srtgajllo, M., Bobinska, J.. and Saganowski, H., Proc. Rubber Tech. Conf., London, 1938,749. (25) Selker, M. L., Winspear, G. G., and Kemp. A. R., IND. ENG. CHEM.,34, 137 (1942). (26) Somerville, A . A., Proc. Rubber Tech. C m f . , London, 1938, 773. (27) Yerzley, F. L., and Fraser, D. F., IND.ENG.CHEM.,34, 332 (1942).
equipment have been presented by Ponchon (10) and Savarit (11) and extended t o liquid-liquid extraction by Maloney and Schubert (6); for practical use by these methods, equilibrium data for the distribution of the components between the phases must be on a solvent-free basis. All tie line interpolation methods so far reported (1, I, 4, 6) necessitate the use of the triangular diagram to calculate the distribution relation of the solute between the solvent and diluent layers on a solvent-free basis (also termed “concentration data”, 7) for each successive theoretical stage in an extraction system. A straight-line plot on rectangular coordinates of the solute distribution relations on a solvent-free basis may be obtained from three points on the ternary diagram of the system considered. Tie line data, on a solvent-free basis in the case of extraction, may be obtained from this plot and applied directly to the stepwise calculation of theoretical stages by the Maloney-Schubert method without further use of the ternary diagram. I n those cases where it is found applicable, the method thus has the advantage of limiting the use of the triangular diagram to the procurement of the above mentioned straight-line plot. This plot is readily applied to the computation of theoretical stages in solvent extraction by the Maloney-Schubert method; the triangular diagram is used together with any one of the tie line methods previously cited to accomplish the same purpose-i. e., to obtain solvent-free distribution relations for each successive stage in the design of a solvent extraction unit. THEORETICAL ASPECT
Varteressian and Fenske (14) found that for the system methylcyclohexane-aniline-n-heptane the distribution of one component (i. e., the solute) between two liquid phases, the solvent layer and the diluent layer, may be expressed by a hyperbolic equation of the type, Pr
= 1
+ ( P - 1)2
where z = weight fraction of solute in diluent laver on solventfree basis y = weight fraction of solute in solvent layer on solventfree basis p = a constant dependent on the system involved; it is a function of the osmotic pressure ratios of solvent and diluent Equation 1 may be rearranged in the form,
”-(
p--l )
2 f z1
?l
PRESENTED as part of the symposium on Compounding and Properties of Elynthetio Rubbers before the Division of Rubber Chemistry at the 104th Meeting of the AMERICAN CHQMICAL SOCIQTY. Buffslo, N. Y.
1
482
Present address, Sational Oil Products Company. Ilarrison, S J
