Document xdxLO3aVqNrZkZLY0kqK1GbDy

1M- T-/ c o 3 PLAINTIFFS Effect of Fillers in Polyethylene k EXHIBIT ^ uc-777 : : 5 By W. J. FRISSELL ftokeltte Company, A Division of Union Carbide and Carbon Corporation Reprinted from November, 1956 BAKELITE COMPANY A DIVISION OF UNION CARBIDE AND CARBON CORPORATION M / 30 East 42nd Street, New York 17, N. Y. V The term BAKELITE is a registered trade-mark of Union Carbide and Carbon Corporation. PLASTICS TECHNOLOGY Effect of Fillers in Polyethylene A study of the effects of 26 fillers on the physical properties of a medium molecular weight polyethylene resin. W. J. FRISSELL, Development Department Hakelite Co., Bound Brook, NJ. Previous work in this laboratory by Bostwick and Carey* has shown that inorganic fillers have a rein forcing effect in polyethylene, just as they do in other plastic resins. There are a number of attractive applica tions for filled polyethylene, such as hot water pipe, thermos botjle caps, electrical insulation, molded articles, kitchenware, and floor tile. Reduction of cost through the use of fillers should also be feasible in some applica tions. The object of this study was to define the effects of 5o*`*.Ick. R., ano Carey. R.H., -lad. Eng. Chem., 42. 040 (1950). a range of filler types on the properties of polyethylene. Primary emphasis was placed on physical properties, but effects on compound color and light stability were given only, cursory examination. Test Procedure The resin used for this evaluation was Bakelite poly ethylene resin DYNH, a medium molecular weight mate rial with a melt index of 1.5-2.4 grams/10 minutes. A list of the 26 fillers evaluated is given in Table I, along with their respective compositions and suppliers. All of the fillers are commercially available, except the November, 1956 Warren J. Frissell, Jr., was born in Dubuque, Iowa, in 1923. He attended the University of Dubuque and Iowa State College, receiving a G. S. in chemical engineering in 1948. From 1944*46, ha was employed by Tennessee Eastman Co. at Oak Ridge, Tenn., and joined Bakelite in 1948 as a chemical engineer in the development department. He is a member of Tau Bata Pi. Mr. Frissel! currently resides in Plainfield, N. J., with his wife and son. jSSQ&SS^MEiii. Table I. Fillers Tested in Polyethylene. No. Name 1. Aluminum Hydrate C-741 2. Asbestos fibers 3. Asbestos Shorts 7R 4. Atomite 5. Bentone 34 6. Cab-O-Sil 7. Dicalite Super-Aid 8. Fibrous Magnesium Silicate #380 9. Fibrous Talc 6N 10. Glajs Fibers: "A" II. "B" 12. "C" 13. "D" 14. Glass Film: *'A" 15. 16. "C" 17. D" 18. "E" 19. Hi-Sil 110 20. Iceberg pigment 21. Micronex W*6 22. Multiflex MM 23. Santocel C 24. Satintone 25. Silene EF 26. Sno-Brlte Composition Aluminum trihydrate, surface coated. Mixed fibers (South Africa). Short asbestos fibers. Wet-ground calcium carbonate. Basic clay. Very fine (0.020-micron) collodlal silica. Diatomaceous earth. Fibrous talc. l/j-In. chopped strand, Silane treated. (/2-In. chopped strand, heat cleaned. /a-lfw milled fibers, Silane treated. i/g-ln. chopped strand. Silane treated. Double flakes 8-microns thick, untreated. Same as film "A," but treated with potyisobutylene. Same as film '`A." but treated with compatible phenolic. Same as film "A," but treated with phenyl silane. Same as film "A," but treated with vinyl silane. Precipitated hydrous silica (0.025 micron). Calcined clay. Beaded channel black. Precipitated calcium carbonate. Hydrated silica. Calcined clay. Precipitated calcium silicate. Uncalcined clay. Supplier Aluminum Co. of America. Johns-Manville Corp. Thompson, Weinman & Co. National Lead Co. Godfrey L. Cabot, Inc. Dicalite Div., Great Lakes Carbon Corp. Whittaker, Clark & Daniels, Inc. international Talc Co., Inc. Owens-Corning Fiberglas Corp. Owens-Corning Fiberglas Corp. Owens-Corning Fiberglas Corp. Pittsburgh Plate Glass Co. Owens-Corning Fiberglas Corp. Owens-Corning Fiberglas Corp. Owens-Corning Fiberglas Corp. Owens-Corning Fiberglas Corp. Owens-Corning Fiberglas Corp. Columbia-Southern Chemical Corp. Burgess Pigment Co. Binney & Smith, Inc. Diamond Alkali Co. Monsanto Chemical Co. Minerals & Chemicals Corp. Pittsburgh Plate Glass Co. Thompson, Weinman & Co. asbestos fibers from South Africa and the glass film. The latter is a development-scale product of Owens-Coming Fiberglas Corp. Preparation of Samples Filler and resin were blended for five minutes in a Hobart kitchen-type mixer. A total batch weight of 400 grams was used. The blend was fluxed on an 8 by 16inch two-roll laboratory mill at 120 C. (steam outlet temperature), and rolled for 15 minutes. The resulting sheet was molded into appropriate plaques for the various tests. Test Methods Tensile properties at 23 and 80 C. were measured using a Baldwin-Tate-Emery Universal Testing Machine. Specimens were cut with die "A" of ASTM D412-5IT. A strain rate of 1.0 inch/ minute was used until the yield point was passed, then the rate was increased to 20 inches/minute until the specimen broke. The properties reported from this test are tensile strength, yield strength, ultimate elongation at yield point. Secant modulus at 23 C. was measured with the same test method, but at a strain rate of 0.1 inch/ minute. The secant modulus was taken as the slope of the line intersecting the stress-strain curve at the origin and 1.0% strain. Heat distortion temperature was determined in ac cordance with ASTM method D648-45T using a stress of 66 psi. Durometer "D" hardness was determined by the method of ASTM D676-49 using a 3.2-pound weight and taking the reading after 15 seconds contact. Brittle temperature (50% index) was run by modi fication of ASTM method D746-44T intended for use with materials having a wide transition range. XIA failure was determined in the Atlas XIA Weatherometer. Zero-strength time Is a type of creep test run above the softening point of the polymer. For this study, it was run at a temperature of 127 C. and a stress of 5.6 psi. The time to failure, in seconds, is recorded. The guillotine test was developed by Worthington Ball Co. to test the cutting resistance of golf ball covers. The apparatus is essentially a miniature guillotine. Depth of cut edgewise into a 0.125-inch thick sample is measured. Stress rupture was measured by loading a '/6-inch thick specimen in tension at a stress of 600 psi. and a temperature of 60 C. Time to failure, in days, and approximate per cent strain at failure were recorded. PLASTICS TECHNOLOGY Jbr Table 2. Tensile and Yield Properties at 23 C. of Filled Polyethylene at Various Loadings. Filler Filler con- No. centration, phr.: 21.. 3. 4. 65.. 78.. 9. 10. 11. 12. 13. 14. 15. 16. 17. 19. 20. 21. 22. 23. 24. 25. 26. None (control) Tensile Strength at 23* C.. Psi. 5 20 50 100 1680 -- 1480 1870 1790 ... 1540 1450 -- -- -- -- -- -- -- -- -- -- 1690 -- 1350 1890 -- 1710 1460 -- 1670 1460 I860 1520 1480 1560 1740 1650 1540 1730 1890 1580 1880 1490 1550 1600 1610 1750 1710 1630 1750 1690 1430 1600 1550 1560 1640 -- 1460 2530 1660 1450 1530 1900 1840 1790 1910 2680 1820 1980 1510 1770 1820 2060 1760 1980 1930 1690 1810 1310 1800 1730 1670 1650 -- 1380 2390 1760 1510 1440 -- 2170 1980 2110 3050 1790 2100 1510 -- 1540 2300 1600 2270 1150 1950 1390 1390 -- 1840 1730 2150 -- Yield Strength at 23" Ce, PiL 5 20 50 100 1340 1370 -- 1770 1340 1360 1340 1350 1280 1380 -- 1540 1360 1530 1350 1430 -- 1520 -- 1890 -- .1410 -- I860 -- 1260 -- 1450 -- 1430 -- 1520 -- 1730 -- 1540 1350 1540 -- 1550 1310 1500 1330 1330 -- 1450 1350 1460 1330 1430 -- 1450 1310 -- 1410 2530 1660 1450 1530 1900 1840 1790 1720 2680 1790 1970 1510 1770 1790 1970 1710 1900 1920 1640 1810 1310 1800 1650 1670 1650 -- 1380 2390 1760 1510 1440 -- 2170 1980 2050 3050 1790 2100 1510 -- 1540 2300 1600 2270 1150 1950 1390 1390 -- 1840 1730 2150 -- Ultimate Elongation at 23* C-. % 5 20 50 100 ...... ' 1 1 1 565 360 65 2 -- 22 15 5 355 140 25 6 590 630 110 30 590 160 36 0 -- 102 7-- 600 130 26 9 440 no 22 4 -- 98 49 19 -- 28 16 5 -- 65 15 3 -- 22 13 5 -- 270 16 3 -- 34 8-- -- 169 23 0 -- 43 27 0 -- 14 15 0 -- 41 19 0 545 185 15 0 -- 169 45 10 99 460 14 0 585 585 55 IS -- 87 12 -- 630 195 100 20 380 87 13 5 -- 197 16 10 560 -- -- -- Elongation at Yield Point at 23* C, % 5__ 20 50 100 76 60 20 12 -- 12 12 1 67 20 10 2 65 50 20 10 65 10 8 0 -- 35 6-- 57 40 17 5 60 22 15 3 -- 40 17 10 -- 12 10 5 -- 10 6 2 --77 1 -- 12 4 2 -- 6 3-- --850 -- 20 8 0 --2 2 0 -- 12 2 0 65 40 10 0 -- 50 24 9 40 50 7 0 70 60 20 15 -- 18 II -- 70 65 40 10 55 20 12 5 -- 53 7 8 72 -- -- -- Table 3. Tensile and Yield Properties at 80 C. of Filled Polyethylene at Various Loadings. Filler Filler con- No. eentration. phr.: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 2201.. 22. 23. 24. 25. 26. None (control) Tensile Strength at 80" C,, Psi. 5 20 19 100 590 520 415 330 -- 600 715 825 505 470 555 570 610 595 590 505 505 485 430 350 -- 630 660 -- 470 510 580 615 520 555 685 675 -- 600 650 750 -- 615 795 890 -- 580 680 570 -- 620 745 825 -- 530 450 360 -- 540 645 -- -- 525 605 620 -- 635 705 815 -- 640 605 630 -- 630 700 945 545 670 705 345 -- 730 760 850 490 635 715 595 610 610 540 445 -- 585 590 _ 600 575 635 650 605 520 540 570 -- 645 700 805 635 -- ___ _ Yield Strength at 80" C,, Psi. 5 20 50 100 545 485 415 330 -- 600 715 825 485 455 550 570 575 595 590 SOS 475 460 420 350 -- 630 660 -- 435 480 580 615 485 535 685 675 -- 600 650 750 -- 580 790 890 -- 550 680 570 -- 600 745 825 -- 520 450 360 -- 525 645 -- -- 525 605 620 -- 615 705 815 -- 630 595 630 -- 600 700 945 515 635 705 345 -- 670 710 850 455 545 680 595 575 580 515 445 -- 580 590 ___ 570 540 635 650 580 520 540 570 -- 605 675 805 545 -- -- Utimate Elongation at 80" C.. % 5 20 50 100 636 -- 201 398 440 -- 188 285 -- -- -- -- -- -- -- -- -- -- 303 -- 290 351 -- 528 153 -- 540 197 44 104 192 275 81 108 83 94 80 125 98 167 54 38 86 35 79 100 187 770 170 61 191 74 208 -- 57 14 51 63 no 18 31 33 54 10 14 15 52 26 23 21 31 21 25 96 203 80 10 61 25 107 -- 20 8 23 14 IS -- 12 II 18 4 10 17 36 8 9 7 II 0 32 13 8 -- 22 5 50 -- Elongation at Yield Point at 80" C.. / 5 20 50 100 75 57 52 16 -- 40 15 8 63 62 45 23 67 60 53 II 95 85 80 15 -- 65 12 -- 65 65 31 12 77 53 33 11 -- 62 30 15 --8 7 4 -- 70 9 7 -- 13 IS 8 -- no 13 5 -- 30 20 -- -- 33 17 8 -- 50 16 9 -- 20 15 7 -- 42 20 II 90 58 25 0 -- 60 50 28 70 87 62 13 77 60 37 8 -- 52 9-- 90 75 60 22 77 65 25 5 -- 80 60 45 81 -- -- -- November, 1956 Table 4. Secant Modulus, Heat Distortion, and ZST Values for Riled Polyethylene at Various Loadings, rmer No. 2. 3. 4. s. 6. 7. 8. 9. 10. II. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. centration, phr.: None (control) Secan Modulus at 23* C.t PsI. x 10-3 5 20 50 100 26 28 34 44 _ 48 83 108 31 40 58 81 29 33 38 56 30 __ 47 112 -- _ 39 42 76 64 __ 92 __ 39 58 91 -- 36 50 78 __ 64 108 166 __ 44 76 93 __ 55 77 125 44 69 89 __ 46 78 -- __ 58 92 123 5S 89 137 __ 76 80 108 __ 52 83 132 28 38 70 129 __ 33 43 72 31 40 67 94 29 33 43 66 __ 41 66 __ 25 33 48 71 30 39 60 132 __ 35 56 90 26 -- -- -- Heat Distortion Temp. at 66 Psi., C. 5 20 50 100 40 39 40 45 _ 53 72 69 44 46 56 62 38 38 43 48 46 46 54 63 __ 64 77 -- 44 44 57 69 42 46 59 67 59 60 72 __ 61 84 94 __ 68 85 82 __ 54 85 88 -- 56 70 88 -- 58 68 -- __ 59 71 80 -- 56 70 81 -- 65 66 78 -- S7 69 83 45 49 62 85 -- 53 60 67 47 50 54 68 39 42 46 56 -- 49 61 -- 40 43 49 62 44 50 59 86 -- 49 60 72 43 -- -- -- Zero Strength Time at 127* C., Sec. 5 20 50 100 --250 250 240 780 260 280 310 220 220 270 >860 280 270 270 260 210 240 360 -- 240 >600 -- -- 220 230 220 230 220 230 240 240 -- 200 190 200 -- 240 330 360 -- 210 220 230 -- 270 280 300 -- 220 210 230 -- 240 250 -- -- 250 230 250 -- 240 230 250 -- 250 250 250 -- 250 260 250 240 260 >860 >860 -- 200 200 190 210 240 >860 >860 260 260 260 250 -- 320 >860 -- 260 260 250 240 210 220 240 >860 -- 250 260 400 260 -- -- -- Table 5. Hardness, Cut Resistance, and Color of Filled Polyethylene at Various Loadings. Filler No. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. centration, phr.: None (control) Durometer "D" Hardness 5 20 so 49 49 51 -- 52 55 45 48 52 48 49 51 45 47 50 -- 52 59 44 54 47 51 - 48 51 -- 54 59 -- 49 54 -- 52 56 -- 49 53 -- 50 54 -- 48 51 -- 49 54 -- 52 54 -- 50 55 48 53 57 -- 49 52 47 50 55 4B 49 52 -- 50 57 48 50 54 46 49 55 -- 49 52 46 -- -- 100 54 60 55 55 53 -- 60 55 57 61 57 60 58 -- 59 59 59 58 67 56 65 56 -- 58 63 54 -- Guillotine Test Depth of Cut, Mm. 5 20 50 100 13 12 II II -- 11 9 Split 16 12 12 10 12 12 11 II 13 12 13 Split --H 8-- 12 13 10 9 16 14 12 -- -- 13 12 10 -- 1 1 10 9 -- 13 12 II -- 1 1 10 split 13 11 Split -- 12 II - -- -- 12 11 Split -- 12 11 Split -- 11 II Split -- 12 II Split 12 10 8 6 -- 13 11 10 15 12 10 6 14 12 12 Split -- II 9-- 14 12 II 10 13 12 10 8 -- 11 10 8 (4 -- -- -- Color Imparted by Filler While Dark gray Dark gray Grayish white Tan Very light tan Brown Gray Light gray Light gray Light gray Light gray Light gray Light gray Light gray Light gray Light gray Light gray Light gray While Black Cream Light tan Light gray Light gray Cream White PLASTICS TCCHNOLOGY Test Results Yield and Ultimate Tensile Four properties were measured with the tensile test: ultimate tensile strength, ultimate elongation, yield strength, and elongation at yield point. Since the yield point for most applications is the point of failure, the yield properties are considered more significant than the ultimate tensile properties. Tensile tests were run at both 23 and 80 C., with the results shown in Tables 2 and 3. Concentration of filler is shown in parts per hundred of resin. In general, the tensile properties at 23 C. were slightly affected by the addition of five parts of filler. Twenty parts of filler lowered the elongation, but had very little effect on tensile strength and yield strength. However, both 50 and 100 parts of filler reduced elonga tion sharply and, in most cases, raised the yidld and tensile strengths. Glass fibers "A" (Filler No. 10) raised yield strength to almost twice that of unfilled resin at a concentration of 100 parts. Other fillers which were particularly good in raising yield strength were asbestos fibers (No. 2), Dicalite Super-Aid (No. 7), glass film samples "C" and "D" (Nos. 16 and 18), and Sno-Brite (No. 26). With most of the fillers, increased concentration resulted in higher yield strength. However, in some cases, the yield strength was lower at 100 parts filler than at 50 parts. Two examples of the latter behavior occur with Hi-Sil 110 precipitated hydrous silica' (No. 19) and Micronex W-6 carbon black (No. 21). Apparently, in these cases, the surface area of the filler is so great that it cannot be wetted adequately by the resin. As a result, the compound is lacking in strength. It is interesting to note that at filler concentrations of 50 and 100 parts, the yield strength and ultimate tensile strength are usually the same; the ductibility normally associated with polyethylene is missing; and the compound breaks very soon after it yields. In general, the same behavior is found at 80 C. as at 23 C. Of course, the yield strength is much lower at 80 C. The elongation figures are of the same order, although at 100 parts of filler they are higher at 803 than at 23 C. At 80 C., the degree of increase in yield strength caused by the better fillers is not as great as at 23 C. The same fillers arc generally good, with the exception of Dicalite Super-Aid (No. 7). The phenomenon of a lower yield strength at 100 parts than at 50 parts is also found at 80 C., notably with Hi-Sii 110 (No. 19). Modulus of Elasticity In all cases, the secant modulus at 23 C. showed a rise with increasing filler concentration (see Table 4). At five parts phr., the fillers had little effect, but 20 parts raised the modulus as much as 300% in one case. The highest modulus, 166,000 psi., was found with 100 parts of glass fibers (No. 10). The result for unfilled resin was 26,000 psi. Other fillers which produced a high modulus were asbestos fibers (No. 2), Bentone 34 basic clay (No. 5). glass fibers "C" (No. 12), the glass films (Nos. 14-18), Hi-Sil 110 (No. 19), and Silenc EF precipitated cal cium silicate (No. 251. Atomite calcium carbonate (No. 4), and Aluminum Hydrate C-741 (No. 1) had the least effect on modulus. Heat Distortion Temperature Increases in heat distortion point ranged from 51 C. for glass fibers "A" (No. 10) to two degrees for Aluminum Hydrate C-741 (No. 1), as shown in Table 4. The most effective fillers here were the glass fibers (Nos. 10-13), the glass films (Nos. 14-18), Hi-Sil 110 (No. 19), and Silene EF (No. 25). It seems reasonable to expect a correlation between heat distortion tem perature and yield strength at 80 C. This, along with other possible correlations, is discussed in a later sec tion. , Zero-Strengfh Time (ZST) Results of this test (see Table 4) fall into two cate gories. Most of the filled compounds were not signif icantly different from the unfilled resin, but a few showed comparatively long zero-strength times. These were Aluminum Hydrate C-741 (No. 1), Asbestos Shorts 7R (No. 3), Cab-O-Sil silica pigment (No. 6), Hi-Sil 110 (No. 19), Micronex W-6 (No. 21), Santocel C hydrated silica (No. 23). and Silene EF (No. 25). The longer ZST values were found only at 50 or 100 parts of filler. Duromefer Hardness As shown in Table 5. the effect of fillers on Durometer "D" hardness was generally much less than effect on other properties. The highest hardness value found was 67 for a loading of 100 parts phr. of Hi-Sil 110 (No.. 19), as compared with 46 for unfilled resin. The differences among fillers on the hardness properties also were smaller than with other properties. The high hard ness value of 67 was somewhat below that of a rigid vinyl sheet which is around 80. Guillotine Test The resistance to cutting, as measured by this lest, was approximately doubled by the most effective fillers, as shown in Table 5. These were Hi-Sil 110 (No. 19) and Micronex W-6 (No. 21 ). Many of the compounds containing 100 parts of filler actually split, showing a definite brittle fracture. Brittle Temperature As shown in Table 6. the test for brittle temperature at 50% index was run on only five fillers so no over-all comparisons can be made. However, it is obvious that (he effect of any filler will be to raise the brittle tem perature drastically, especially at the higher loadings. Of the fillers tested. Iceberg pigment (No. 20) had the least effect. In connection with brittle temperature, it should be realized that the use of a filler changes polyethylene from a flexible material to one which is essentially rigid. The brittle temperatures of rigid plastics are usually above room temperature, and the brittle temperature test may be of questionable value. November, 1956 Table 6. Brittle Temperature, Stress Rupture, and Weatherometer Values for Riled Polyethylene at Various Loadings. Filler No. Concentration 3 50 5 50 6 20 SO 7 50 8 50 9 20 50 100 II 20 Brittle Temp. 50% Index. C. ____ -- -18 >25 -- -- -36 -14 >25 -34 12 13 19 20 21 24 Control 50 100 50 100 20 50 100 SO 20 50 100 50 100 -- 10 >25 -- -- -36 8 >25 -- -64 -25 3 -- -- -98 * These samples are not considered to have failed. Stress Rupture at 500 Psi. and 60 C. Days >274 16 -- -- 59 3 -- -- -- "" -- -- 70 105 -- -- -- 96 -- -- -- 0 2 2 Failure, % >15.9 20.0 -- -- 7.5 7.0 -- -- -- -- -- 3.5 1.0 -- -- -- 6.5 -- -- -- 5.0 39.1 XIA Weatherometer Failure Hours Type _ -- 500 240 -- _ 750 750 500 750* 750 500 ___ ___ 750* 750 750 ___ 500 500 500 ____ _ ____ Crating Crazing -- -- Crazing Crazing Crazing Slight darkening, White streaks White streaks White streaks Slight yellowing White streaks White streaks Crazing Crazing Crazing 750* Slight darkening Stress Rupture As shown in Table 6, straight resin shows the very short life of two days when tested for stress rupture at 600 psi. and 60 C. Three of the fillers, Satintone cal cined clay (No. 24), Micronex W-6 (No. 21), and fibrous magnesium silicate (No. 8) did not improve this performance. Among the other fillers tested. Asbestos Shorts 7R (No. 3) was the best with a life of over 274 days at the comparatively low strain of 15.9%. Glass fibers "C" (No. 12) and Hi-Sil 110 (No. 19), which lasted 105 and 96 days respectively, also were very good. It is apparent that properly chosen fillers can greatly improve the long-time strength of polyethylene resin. Light Stability A cursory study of the effects of five different fillers on the light stability of the resin was made using the X1A Weatherometer (see Table 6). After 750 hours exposure, straight resin showed only a slight discolora tion. All of the samples containing Cab-O-Sil (No. 6), Fibrous Talc 6N (No. 9), or Iceberg Pigment (No. 20) showed severe crazing at or before this exposure time. Fig. I. Correlation between yield strengths at 23* C. and 80 C. - Fig. 2. Correlation between yield strength at 80* C. and heat distortion at 66 psi. PLASTICS TECHNOLOGY The samples containing either of the glass fibers (Nos. 12 and 13) showed a curious white streaking effect in the exposed area which seemed like an erosion of the surface. The use of the fillers tested was definitely detrimental to the light stability of the resin. Color Imparted by Filler Most of the fillers imparted comparatively little color to the polyethylene, as shown in Table 5. However, the coloring effect of such materials as Micronex W-6 car bon black (No. 21) and the asbestos fillers (Nos. 2 and 3) would definitely limit their use. Of the fillers tested, the glass fibers (Nos. 10-13) and glass films (Nos. 1418) had the least opacifying effect. Summary of Properties To permit quick comparisons of fillers, a summary chart has been constructed and is shown in Table 7. A numerical rating system has been used as follows: (1) The filler had no effect as compared with un filled resin. (2) Slight effect. (3) Moderate effect. (4) Large effect. (5) Very large effect. In using a chart such as this, a specific application must be kept in mind. For example, if the modification of resin to make it suitable for boilable articles is being considered, an increase in modulus, heat distortion, and yield strength at 80 C. are required. The desired filler should have a large effect on these and related proper ties, combined with a small effect on other properties. In a second case, it may be desired to simply lower the cost by addition of a filler. Here the least effect on all properties is required. For the first application, the chart shows glass fibers "A" (No. 10) to be a good choice. For the second application. Aluminum Hydrate C-741 (No. 1) Atomite (No. 4), or Satintone (No. 24) would be indicated. Correlation Among Tests A number of graphs of related properties plotted against one another have been made to check correla tions. These graphs contain data on compounds with 20, 50 and 100 parts of filler, and are shown in Figures 1-4. Least square lines and correlation coefficients were calculated in each case. A correlation coefficient of 1.00 represents a perfect correlation. Figure I shows yield strength at 23 C. vs. yield strength at 80 C. There is some scatter, but the cor relation is fairly good throughout the graph. The cor relation coefficient is 0.78. in Figure 2, yield strength at Table 7. Numerical Ratings of Fillers by Effects on Polyethylene Properties at 50-Part Loading. No. Filler Name 1. Aluminum Hydrate C-741 2. Asbestos Fiber* 3. Asbestos Short* 7R 4. Atomite 5. Bentone 34 6. Ceb-O-Sil 7. Dicalita Super-Aid 8. Fibrous Magnesium Silicate ^380 9. Fibrous Tate 6N 10. Glass Fibers "A" It. "B" 12. "C" 13. "D" 14. Glass Film "A" 15. "B" 16. "C" 17. "D" 18. "E" 19. Hf-Sil 110 20. Iceberg Figment 21. Micronex W-6 22. Multifex MM 23. Sanfocel C 24. Santintone 25. Silane EF 26. Sno-Brife At 23C. At80C. - _c Tc5* 9 "O 9 % 21 o O. 9 w ? . o >- LU "oc 9 /) tJ 9 > 9 o . 2O" UJ > a ~oa 2<j C t/eO> <N n 0 "OE 3 Ja "5 O 9a Ju h- C ccn 9 a 2 9 ,Ep M - P 1S o T7 35 oX 1--s oa. sEe_o c SSL. to o C 9 * >< i 4 1 2 2 1 1 3 3 ---- 5 5 4 5 4 4 t 4 5---- 2 5 1 3 3 3 1 3 3 ---- 1 4 1 2 2 1 1 3 3 ---- 2 5 1 1 2 2 2 2 2 ---- 353 54555 554 3 4 1 4 3 3 1 4 4---- 3 5 3 4 3 3 1 3 3 ---- 24 3 4 3 3 1 3 3 4 2 5 5 5 5 5 5 2 5 4---- 353 54 5 1 4 3 5 2 3 5 5 5 4 5 1 4 4---- 2 5 1 5 4 4 1 3- 4 5 2 2 5 3 5 4 4 1 4 4---- 3 5 2 5 5 4 1 3 4 -- ____ 3 5 4 5 5 4 1 4 4 -- ___ 2 52 54 4 1 4 4 -- ____ 3 5 4 5 4 4 1 4 4 ___ 3 5 4 4 4 3 5 5 5 _ ____ 2453 2 3 1 3 4 3 3 3 542 3 2 544 14 1322 1 33 - 3 5 1 5 3 3 5 5 5__ 23222 2 1 4 4 ___ -- 25 143 3 144 _ 2542 3 3 1 3 4 -- ____ 9 9 * 9a. 5 -o 1- B is 5 *o o is> 3 i 4 45 22 3-- 2-- 33 31 2-- 2-- 2-- 24 2-- 2-- 2-- 2-- 2 2 ____ 24 1 51 2 ___ 2_ 21 2 2-- November, 1956 Effects of Fillers in Polyethylene 80 is plotted against heat distortion at 66 psi. This is the weakest of the group with a correlation coefficient of only 0.46. Figure 3 is a plot of Durometer "D" hardness against secant modulus at 23 C. A correlation coefficient of 0.82 shows this to be a fairly good correlation. In 'Figure 4, Durometer "D" hardness is plotted against depth of cut in the guillotine test. This is the best cor relation of the group with a coefficient of -0.88. As a result of these correlations, several tests could possibly be dropped in screening evaluations of fillers in polyethylene. The yield strength at 80 C. and the guillotine test give essentially the same information as yield strength at 23 C. and Durometer "D'' hardness, respectively, and thus are superfluous. Since Durometer "D" hardness and secant modulus correlate well, one of these tests also can be eliminated. These three tests may still be useful for the more comprehensive evalua tions, since exceptions to the correlations do exist. 45 50 55 6 0 65 70 0UR0METER *0* HARDNESS Fig. 3. Correlation between Durometer "D" hardness and secant modulus at 23 C. Summary and Conclusions A study has been made of the effects of 26 different fillers on the properties of a medium molecular weight polyethylene resin. The fillers varied widely in the de gree to which they changed such properties as tensile strength, yield strength, modulus of elasticity, brittle temperature, heat distortion temperature, and light stability. One of the glass fiber samples caused the largest in creases in yield strength and modulus. Asbestos shorts gave the greatest improvement in long-term strength. The least effect on all properties was found with cither aluminum hydrate or Atomite wet-ground calcium car bonate. The results indicate that the addition of fillers to polyethylene may be used to broaden the applications of this already versatile resin. Acknowledgements The author wishes to express his appreciation to his colleagues in the Development Laboratories for their aid in the collection of data and the preparation of this paper, and to The Worthington Ball Co.. Elyria. O.. for permission to discuss their Guillotine Test. --nu: i:m> /*~'X 1. * 45 SO 55 60 65 OUROMETER `O' HARDNESS 70 Fig. 4. Correlation betweel Durometer "D" hardness and depth of cut in guillotine test. PLASTICS TECHNOLOGY r : . T BAKELIT E COMPANY A DIVISION OF UNION CARBIDE AND CARBON CORPORATION DE3 9 30 EAST 42nd STREET, NEW YORK 17, N. Y. . j Atlanta 3, Ga. Boston 94, Mass. Chicago 1, 111. Cincinnati 6, Ohio Cleveland 14, Ohio Clifton, New Jersey Detroit 21, Michigan Hartford 3, Conn. Kansas City 5, Mo. Larchmont, N. Y. Los Angeles 58, Calif. New York 17, N. Y. Philadelphia 3, Pa. Pittsburgh 22, Pa. Rochester 4, N. Y. St. Louis 22, Mo. San Francisco 6, Calif. Washington 5, D. C. Direct Sales Offices 57 Forsyth Street 300 First Avenue, Needham Hgts. 230 No. Michigan Avenue 2330 Victory Parkway 1300 Lakeside Avenue, N. E. 1051 Bloomfield Avenue 10421 W. 7 Mile Road 410 Asylum Street 910 Baltimore Avenue 1877 Palmer Avenue 2770 Leonis Blvd. 30 East 42nd Street 117 So. 17th Street 537 Smithfield Street 130 Main Street, East 122 No. Kirkwood Road 22 Battery Street 777 14th Street, N.W. In Canada BAKELITE COMPANY Division of Union Carbide Canada Limited Bel levi I le-Toron to- M on trea I Outside Continental United States and Canada PLASTICS DEPARTMENT UNION CARBIDE INTERNATIONAL COMPANY A Division of Union Carbide and Carbon Corportion 30 East 42nd Street, New York 17, N. Y., U. S. A. March 1957 i.inoi Primed in U.S.A.