Document jz6vvNpoZp0pGmde0ZrjoQM5

REPRINTS FROM Rubber UJoricJ PLAINTIFF'S EXHIBIT August, 1971 pages 46 to 54 Non-black reinforcers and fillers for rubber by M. P. Wagner, Chemical Div. PPG Industries, !nc., Barberton, Ohio November, 1971 pages 45 to 50 Compounding with fibers for high performance elastomer compounds by George C. Derringer PPG Industries, Inc., Barberton, Ohio I Ii It j UNION CARBIDE CORPORATION RUBBER CHEMICALS 270PARK AVENUE, NEW YORK, N.Y. 10017 UCC 015000 6 Non-black reinforcers and fillers for rubber b Y M. P. Wagner, Chemical Div., PPG Industries, Inc., Barberton, Ohio Commercial development of the rub ber industry through the first two decades of this century depended heavily on non-black pigments, many still prevalent today (Fig. 1). Clay, whiting, zinc oxide, barytes, oxides of iron and lead, and many more such materials were used; materials readily available naturally or as by products of other industries were can didates as rubber fillers. The tire industry was born on non black pigments, arid, until World War I, zinc oxide and iron oxide were two of its major fillers. The success of the first carbon blacks in greatly im proving rubber properties introduced the age of reinforcement. With this came the quest for fillers having small particle size. Early efforts were directed to re ducing the particle size of zinc oxide. In the early 30's precipitated cal cium carbonates became available, first in relatively coarse particle size and later in die 30's in ultra-fine sizes below 0.1 micron. The first non-black filler having a particle size approaching that of car bon black was calcium silicate, first produced in 1939. Finally, in the early 40's, calcium silicate and cal cium carbonate fully comparable to carbon black in particle size were produced. Hydrated silica with a particle size equal to the finest carbon blacks ap peared in the early 50's. This was followed by anhydrous silica of still smaller particle size. Non-black fill CHRONOLOGY OF NON-BLACK PIGMENTS ers, through the hydrated and anhy drous silicas, have particle sizes as ZINC OXIDE, WHITING, CLAY. BARYTES, ETC. IN USE BETWEEN 1050-1900 small as any filler available, including carbon blacks. They provide unique 1900 ZINC OXIDE-PRIME FILLER IN TiRES ly high tensile strength, tear strength, flex resistance and stiffness. Abra sion resistance approaches, but does 1910 CARBON BLACKS START MOVE .25 TITANIUM DIOXIDE MIXED PIGMENTS I9E0 FINE PARTICLE ZINtfoXlDES TITANIUM DIOXIDE (ANATASE) 19l 30 .1 PRECSP. CALCIUM CARBONATES 1940 1950 I960 .07 .04 .03 .02 .015 CALCIUM SILICATES ULTRAFtNE CALCIUM CARBONATES EXTRAFINE CALCIUM SILICATES PRECIPITATED SILICAS ____ TITANIUM DIOXIDE (RUTILE)_ fumeo'silTcas SODIUM SILICOALUMINATES not yet equal that of the highest re inforcing fillers. Developments with in the last year point the way to over coming even this disparity between non-black fillers and carbon black. Together, these inexpensive rub- MATERIAL CONSUMPTION IN RUBBER INDUSTRY 1970 SILANE MODIFIED CLAYS FIBER REINFORCE MENT__ rig. 1. The first non-black filler having a particle size (center column) approaching that of carbon black was calcium carbonate. Today, hydrated and anhydrous silicas have particle sizes as small as any reinforcer. Table 1. Composite recipe (in pounds) for rubber industry in 1967. . Synthetic Rubber__ 67 f 100 Lbs. Rubber ) Carbon Blacks ...... . 44 44 lbs. Black Whitings.......... ............ (17) Est. Kaolin ..................... 15 \ Zinc Oxide _________ TiO;> + Barytes____ Talc ................. ........ Misc. Non-Black Filler 4.4 { ,, , ,. , > 48 lbs. Non-Black 2.3 ( 1,2 ) (8.3) Est. Fig. 2. Total consumption of non-black fillers for rub ber is expected to climb to 1,300,000 tons by 1975. UCC 015001 her fillers account for an estimated proximately classify non-black fillers produced from natural forms of cal '300,000 tons per year, almost as according to their reinforcement of cium carbonate are selected princi much as carbon blacks (Fig. 2). 'l aic consumption is decreasing, prob ably because other fillers are dis rubber. Solid particles larger than about 5 microns act to degrate prop erties--while most fillers are intend pally on the basis of origin and pur ity I Table 2). Common types vary in crystal structure and hardness. The placing it on a cost/performance ed to be smaller, inefficient separation softest, and hence the finest, is soft basis. The growth of titanium dioxide and barytes is slightly greater than during manufacture may leave a sig nificant fraction of coarse particles in limestone or chalk, commonly termed whiting. These are the most widely that of rubber, indicating that color the filler to lower rubber properties. used diluent fillers, and generally are is becoming a more important factor in the rubber industry. Those fillers with particles between 1 and 5 microns contribute very little the purest and most white, depend ing on the deposit. The most rapid growth in non- to the strength of rubber. Large vol The presence of very coarse par black pigments is occurring, how umes can be used without seriously ticles, chiefly quartz and mica, is the ever, in the miscellaneous category. impairing strength; hence, their clas next most important consideration Admittedly this is an estimate, but a sification as diluent fillers. Good (Fig. 4). Even the best grades of reasonable one. Included in this class grades of whiting and soft clays are whiting, if they contain a small are precipitated calcium carbonates, the dominant products here. amount of coai3e particles, will not calcium silicates, silicoaluminates, hy Below about 1 micron, most filler provide the best properries. Tear drated silicas and anhydrous silicas. particles begin to reinforce. Hard strength is severely reduced by even Figures for most of these products days and coarser precipitated cal a couple percent of residue on a 325- are unavailable, but 200,000 tons per cium carbonates are the principal rnesh screen. Tensile strength and vear represents a good estimate. semi-reinforcing fillers. Zinc oxide modulus are noticeably affected, also. That non-black fillers, collectively, and titanium dioxide also fall in the For higher strength and color po make up a sizeable portion ot the rub semi-reinforcing range, but are sel tential at moderate cost, the finer ber industry can be seen very readily dom used for this purpose. synthetic calcium carbonates are re from a composite rubber "master- Real enhancement of rubber quired (Fig. 5). A rather smooth batch ' of these materials (Table 1). strength is obtained with filler par increase in reinforcement is evident For every 100 pounds of rubber, 67 ticles below 0.1 microns. Ultrafine in tensile strength, tear strength, and of which are synthetic, there are 48 calcium carbonates, silicoaluminates, abrasion resistance as the coarser pounds of various non-black fillers calcium silicates, hydrated silicas and hard limestones are replaced by the vs. 44 pounds of carbon black. Of anhydrous silicas provide increasing finer soft whitings and, finally, by the 48 pounds of non-black pigments, reinforcement with decreasing par very fine precipitated calcium car two-thirds of these are whiting and ticle size. The last two of these are bonate. For optimum dispersion, the clay. The remaining one-third include below 0.025 microns in diameter, ultra fine calcium carbonate is fre 6 or 8 principal materials and an comparable to 1SAF and SAF car quently surface coated with oil or even larger number of very minor bon blacks in size. fatty acid. components. It is interesting to note that zinc oxide is utilized at levels typical for a vulcanization activator, emphasizing its minor role as a filler. 'ATiite sidewalls of tires represents its dominant use as a filler. Today's role of non-black fillers: general classification The decrease in particle size is, of course, accompanied by an increase in cost. The price of the common fillers range from S0.005-S0.08 per pound. Zinc oxide and titanium di oxide are somewhat higher, while the anhydrous silicas are considerably higher. Clays Rubber grade clays are chiefly kao lins, which occur predominantly in the southeastern United States. Two common grades are recognized by compounders (Fig. 6): soft, coarse sized clays for producing soft rubber compounds, and small particle-sized The classification of rubber fillers Calcium carbonates hard clay3 for harder rubber stocks. begins with particle size (Fig. 3). Calcium carbonates are available in Two processing treatments are used On this basis, it is possible to ap a range ot particle diameters. Those to iinish the clays. Classification by air, termed air-floated, is the most i LIGHT -10 MICROSCOPE -I ELECTRON ,, . MICROSCOPE "01 OEQRACMG filler GROUND MARBLE WHITING as DILUENT Fillers semi REINFORCING SOFT CLAYS HARO CLAYS L PCRtCEOCIjPITATED 10 /.5 5 RCmF$*C!N9 SJLICOAIUMIMATES CALCIUM SILICATES HYDRATED SILICAS S | a common. This is an efficient method Table 2. Calcium carbonates produced from natural forms are selected principally on the basis of origin and purity. 1. ORIGIN - Chalk (Whiting) Saccharoidal Marble 2. PURITY - Quartz, Mica Colcium Oxide (pH) Trace Metals (Cu or Pb) C-O.OI ANHYDROUS SILICAS I 3. MOISTURE Fig. 3. Classification of rubber fillers is by particle size, \vhich also approximates their reinforcing status. 4. COLOR UCC 015002 that remove* very large particle*, but depends on the deposit for purity, color, etc. Water-washed clays are generally lighter in color and more uniformly controlled for pH and particle size distribution. Very large particles are readily removed by screening the slurry of clay dispersed in water. Size distribution is controlled by isolating selected portions after settling in large tanks. More efficient separation of coarse and fine fractions 13 pos sible by this method. A convenient though not precise way to characterize kaolins is the "percent less than 2 microns." The extremes in hard and soft clays are obtained by water classification (Fig. 6), while the air-floated clays gen- CD ID CO r-. o emlly have a broader particle size distribution and are not as clearly different in rubber as are the water' washed varieties. Special treatments of clay are aim ed at improving dispersion, increas ing whiteness or reducing water ab sorption. Various organic surface active materials improve dispersion and, hence, rubber properties. Chem ical treatment during water washing removes colored impurities such as iron and copper and increases white ness. Calcining removes bound water between crystal layers and, conse quently, reduces water absorption. Silicas and silicates The highest reinforcing non-black pigments are the fine particle silicas and silicates. Rubber shoe soles sur passing leather in wear, comfort and flexibility contain one or more of the several pigments available in this class. Hydrated silica imparts exceptional tear resistance to most rubbers, and is used with carbon black to obtain maximum mileage in tires for mining and logging operations and for road building equipment. These and other Fig. 4. The presense of coarse particles in calcium carbonate reduces effectiveness in rubber reinforcing. CALCIUM CAR8CNATE FILLERS It* NR ' 230Or 300 oj i.a mEM DtA*tCTBy>wi>w ~ Fig. 5. The finer synthetic calcium carbonates are required where higher strength and color potential at a moderate level of reinforcer cost ore called for. Kg. 6, Rubber grade clays are chiefly kaolins. Two common grades used by compounders include: soft, coarse clays for soft compounds; and hard, small particle-size clays for compounding harder rubbers. UCC 015003 applications have become well estabiiih.eii over the past 20 years. Anhydrous silicas have the smallest oai tides available in non-black idlers. Thev also are very pure silica and have \erv low moisture. Consequent ly, they are chiefly used in silicone rubber for electrical applications. De manding requirements for very high temperature reinforcement, low mois ture sensitivity and high electrical resistivity justify die high cost of an hydrous silicas and silicone rubber in such applications. Tee properties of silicas and sili cates available in the U. S. differ, first of all, in chemical composition (Table 3.1, due to the raw materials and manufacturing method. The first four materials are made in an aque ous system by converting water-sol uble silicates into insoluble silica or silicates by use of acids .or salts. The products contain 65 to 88% silica and up tu 22% of other metal oxides. Lrider normal conditions they will also have 4-6% of adsorbed water. Not shown is the bound water, which, in each case, amounts to about 5%. Anhydrous silica is made by an entirely different process. Silicon tetrachloride is oxidized directly to silica by reaction with water and oxygen at elevated temperatures. The resultant product is essentially free of bound water, adsorbed water and metal oxides. The particle size varies according to the specific product, and is con trolled in the manufacturing process. The degree of reinforcement is again determined by particle size (Fig. 3). Current uses of non-black fillers The variety of rubber articles utiliz ing non-black fillers is tremendous, but their principal applications are based un specific attributes which can be grossly generalized as follows: whitings have as their chief utility low cost with a minimum of change in rubber properties; clays and pre cipitated calcium carbonates impart moderate properties at moderate cost; the hydrated silicas and silicates impart high tensile strength, tear strength and abrasion resistance. All, of course, are suitable for colors other than black, but titanium di oxides are required for white or pastel colors. Fillers quite often are used in com bination to gain the unique properties of each type, and sometimes there is Table 3. Typical compositional properties for silicas and silicates available in the United States. Calcium Silica**3 Sodium SilicaAlumina* Hydrated Silico Reinf.c Anhydrou* 5'ticQe SiOo, % ____ ____ 65 CaO, % ...... ____ 20 A1 nOj, % ... ___ . . NajO, %....... ___ Moisture .... . ..... . 5 pH ....---------- ...... 10 Particle Size, microns ........ .. 0 03 65 12 10 4 10 0.04 80 5 6 9.5 0.08 88 5 7 0.02 93 1 4 0.015 a Silen* F, PPG industries, loc. b Zeol#x 23, J. M, Hub*r Corp. c D, PPG Industries, Inc, d Hi-Sil 233, PPG lndustris, Inc. e Cob-O-Sil M-5, Cabot Corp. 60786C Table 4. Tear strength with Hi-Sil 233. Natural Rubber (SMR-H-5L) ............. 100 Zinc Oxide ............................................ . 5 Stearic Acid __________________ Sunolite 240 .......... ......... ...................._. 1.5 Wingstay 100 __________________ Thermoflex A______________ _________ 1.0 Plastogen __________________ 1.5 1.0 5.0 Hi-Sil 233 ______ _____ EPC Black (S-300) .... IISAF Black (N-285) .. Sulfur Santacure NS TMTD .............................. Mercaptosilane _____ 45 2.5 1.4 0.5 45 2.5 1.0 .5 1.0 45 2.7 1.0 15 30 2.5 1.0 .2 ML 4 - 212 ................... Mooney Scorch, Min./270 = F .......... 90% ODR Cure, Min./300F _____ 96 12 9 300% Modulus, psi .. 620 Tensile Strength, psi .. 3780 Elongation, % _______ 700 Hardness ........................ 60 Tear, Die C, ppi 470 Slit Tear, ppi RT ____ 217 170^ 171-296* Compression Set, % (B) 25 a Knotty Tear 77 11 7 1000 3860 660 62 620 187 193-340* 15 57 17 17 1390 4100 580 60 380 80 130 29 78 12 10 1330 3830 570 60 390 no 68-168' 19 a synergistic increase in properties when two or more fillers or compounding ingredients are used together. An example of this is the use of whiting in combination with hydrated silica (Fig. 7). Tear resistance as a function of silica and whiting levels was obtained from an UCC 015004 rlENQS 50 VOLUME RAYON COMPOSITE Fig. 7. Combinations of fillers such as whiting and hydrated silica are often employed to gain some com bination of the unique but complementary qualities. Table 5. Adhesion with the HRH modification of a typical carcass stock for passenger car tires. AD Natural Rubber _________________ 70 SBR _________ ___ _____ ___ ______ -- 30 SRF Black ... . ............... ... ............ --- 45 Hi-Sil 233s ..........-....... - ____________ Zinc Oxide _________________ -- Stearic Acid ____ Process Oil _______ ___ __________ Resorcinol __ _________ . 5 2 4 /A STS DPG _____ Sulfur _____ ___________________ ____ _ Hexamethylenetetramine ________ 0.85 0.35 2.4 Strip Adhesion, pounds per inch Nylon (Greige) ___ ______ ____ Rayon (Greige) ___ _____________ 1.5 3.0 70 30 30 15 5 2 4 2.5 0.85 0.35 2.4 1.6 90 + 55 Table 6. HRH adhesion to wire. Adhesion0, lbs. AD Brass-plated wire*1: Original (@ R. T.) .............- _______ 77 Agedb (& R. T.) ________ _______ 64 Agodb 250 :F,' _____ _______ 68 Bare steel wire: Original (@ R. T.) .... Agedb (@ R. T.) ........ Agedb (@ 250rF) ................... 41 35 15 196 164 150 70 60 23 - Wire Cord eomtructure: 5 x 7 x 3 x 1 (0.0059 ') & Aged 43 hr*, at 212F c ASTM D2229-T63, 1 -irveh put! through Fig. 8. The HRH system is also used to improve the performance of rubber reinforced with short fibers. Table 7. Effect of fillers on strip adhesion using resorcinol and hexamethylenetetramine (formulations similar to A and D). Strip Filler Adhesion, lbs.5 SRF -- Black SRF- Hi-Sil 233 SRF -- Silene D ... SRF -- Silene EF SRF-Clav 29 80 + 70 80 + 31 a To greige nylon, squore woven Table 8. Wire adhesion with modified HRH system. Natural Rubber (SMR-H-5L) ........__ 100 N-330 Black ___________ 45 Hi-Sil 233 ........ . 15 Stearic Acid 2 Antioxidant ________ 3 Naphthenic Oil ___ 8 Zinc Oxide . ____ 5 Resorcinol _______________ ____ _____ Morpholino-sulfenamide ................... 0.8 Sulfur ____ ______ ____ ___________ 3.0 Cyrez 966" ________ ____ ______ __ Cure, Min. 300~F ... ...... 30 300% Modulus, psi ____ 1320 Tensile Strength, psi ...... --.......... 2760 Elongation __________________ _ 490 100 45 15 2 3 8 5 2.5 0.8 3.0 3.0 30 1340 2570 420 Adhesion to Brass-Plated Wire, (3 x 1) + (6 x 1), Ib./ii Room Temp., Original ___ _______.. 69 Pulled at 250CF, Original _______ 82 Aged 72 hrs./250JF _____ 76 Wire covered, % . ............. ....... .30-40 128 150 156 90-100 * Hexamethoxyrnethyhnelamine, American Cyonomid Co. UCC 015005 experimental design. The resulting equations indicated that somewhat higher tear strength can be obtained b\ the partial substitution ol hy drated silica with whiting. An eco nomic saving is also realized. In another example of the com bined use of fillers (Table 41. the unique tear resistance from silicas and the abrasion resistance from I5AF black were combined in a tread jtock for heavy service tires (last column). An all-silica loading is superior to any carbon black, includ ing channel blacks (first column). In an effort to improve wear resist ance, a rriereaptosilane (Union Car bide's Silane A-189; was added with out sacrifice in tear resistance (sec ond column). New developments in non-block pigments: adhesion (Non-black pigments are not confined to ttvir traditional role of filler. One such exciting field of use is to pro mote adhesion of rubber to a variety of materials. Those fillers with small particle sizes are more effective. Zinc oxide has been noted for its ability to improve tack and adhesion of rubber. Hydrated silica improves the ad hesion of non-polar rubbers to metal and polyurethanes. In combination with certain resin formers, adhesion to untreated cord and metal is out standing. Jn fact, this combination of ingredients--called tire HRH system and including ffi-Sil 233 {precipi tated, hydrated silica), resorcinol and hexamethylenetetramine--is an effec tive method of obtaining adhesion to fibers and wire without need for cement or latex dips. As such, it is being used in belts, hose, proofed fabrics, and other fabric reinforced rubber goods. To some degree, the HUH system is also used in tires. The essential features of this tech nology are shown in Table 5. While adhesion of the carcass compound to RFL-dipped cord is excellent, its ad hesion to greige (or un-dipped) nylon or rayon is very low. However, the adhesion levels for both un-dipped fibers can be enormously improved by using the HRH system of Hi-Sil 233, resorcinol and hexamethylene tetramine. Actually, when properly compound ed, the HRH system provides ad hesion to greige cord surpassing that of RfL-dipped cord. Retention of adhesion at high temperatures and after aging is also excellent. Adhesion values for wire are also greatly improved with this system: by the simple addition of the HRH components to a conventional carcass formulation, adhesion values are ap proximately doubled (Table 6). The HRH system is particularly effective with brass-plated wire used in tires, belts and hose. Even at high tem peratures and after aging, adhesion is still high. While not as high, ad hesion to bare steel wire is still re spectable. When using the HRH system, however, it is essential that all ingredients be well dispersed and that premature reaction of hexa and resorcinol be avoided. Crystalline resorcinol can be added in the Banbury at a suf ficiently high mix temperature to ef fect dispersion. Alternatively, it can be added in the form of a rubber rnasterbatch or as a co-blend with stearic acid. These are readily dis pel sed in the Banbury or on the mill. Hexamethylenetetramine must be in a dispersible form and must be added at temperatures below 250CF. Several forms are available, including Hexa Flo Powder, a rubber masterbatch and various oil pastes. Not ah fillers function in the HRH system (Table 7). Only the fine par ticle silicas and silicates promote ad hesion. Apparently, the high energy surface and high surface area asso ciated with small particle size are required. Further, while hexamethylenetetra mine is an active component, objec tions to its use tan be overcome by using alternate materials. Two such materials are available: bexamethoxvmethyl melamine (Cyrez 966} and a similar material, Cohedur A. For ex ample, Cyrez 966 readily gives ad hesion values comparable to those previously shown for nylon and brass-plated wire; again, effective ness at high temperature and after aging is demonstrated (Table 8). Fiber reinforcement Another function of the HRH system is to improve the performance of rubber reinforced with short fibers, another class of non-black fillers (Fig. 8.) Chopped rayon, when add ed to rubber compounds, provides exceptional stiffness. However, poor adhesion in conventional compounds causes loss of stiffness at moderate extensions. The HRH modification overcomes this fault and provides ex ceptional strength and flexibility at reduced extensibility. In one series of tests, three short fibers at 12 volume percent loadings Table 9. Dynamic properties with fiber reinforcement. Natural Rubber _________ 100 Hi-Sil 233 _______ . ____ 12 Filler ____ ____ ___ 12 Vol. % Fiber, or 27 Vol. % FEF Bl ack Resorcinol ____ _ ____ ______ Zinc Oxide ........................ Hexa Flo Powder MBTS DPG __________ ____ _ Sulfur ....... ................................ 2.5 5 1.6 0.7 0.3 2.5 Rayon Fibor Giait Colidno HPP 100% Modulus, psi__ ___ Tensile .... ___ ____________ Elongation, % Yerzley Dynamic Modulus, psi __ Resilience, % .... ......... 1450 1840 150 1506 82 1650 1650 110 1444 88 2080 2510 120 2819 71 Goodrich Flexometer (212^ F, 22.5%, 24.5 lb.) Heat Build-Up, ~F .......... 43 25 Set, % ................. .......... 9.3 4.7 61 13.2 FEF Black 460 3560 430 1381 74 83 16.9 A07862 UCC 015006 were compared with FEF black at a 23 volume percent leading. The com pounds with fibers contained the in gredients of the HRH system. Chop ped rayon and fiber glass were typical of undipped tire cord material. Calidria HPP is an extremely fine asbes tos fiber with a diameter of only 0.02 microns. The effective stiffening of the rub ber by the fibers was much higher than that of the carbon black. Fiber glass and Calidria HPP had the highest efficiency, perhaps due to a much stiffer fiber. The high modulus contributed by the short fibers was obtained with considerably less loss of resilience than with conventional fillers. Some of the dynamic properties of these compounds are shown in Table 9. This shows that fiber reinforcetnent can provide higher resilience and lower heat build-up at a dynamic modulus equivalent to conventional fillers. It is also true that much higher dynamic modulus can be obtained with fiber reinforcement without sac rificing resilience or heat build-up. The combination of adhesion ob tained with the HRH system and the reinforcement produced by short fibers provides a new dimension in engineering with rubber. The in creased dynamic modulus/resilience ratio provides greater load support for resilient mounts, rubber bushings and many ether applications. Designed surface activity The chemical nature of most non black pigments suggests that strong bonding with an elastomer is im probable; and in some cases, noncompatability of the filler with the rubber has led to poor dispersion. However, a number of products have been surface modified to improve their performance in rubber. Zinc oxides, precipitated calcium carbo nates and clays have been coated with surface active materials to im prove dispersion and make the filleis more effective. Recently, clays have been modified with chemically re active silanes to substantially improve the properties they impart to rubber. It is, in fact, possible to tailor tbs reactivity of the filler surface to pro vide chemical bonding with the rub ber. Organofunctional silanes (Table 10) are effective materials for this purpose. Methoxysilyi (left side) re acts readily with fillers containing silanol groups--e.g., clays, silicas and silicates. The functional group on the right is chosen to react specifically with the rubber system. For most sulfur cured polymers the inercaptosilane is effective (the example at the bottom indicates the nature of the Table 10. ORGANO FUNCTIONAL SILANES (CHjO)j--Si t'- CHgCHgCHjSH I --CHgCHgCHgNHj ) -H=CH2 ] --1CH2)3--O-C--C=CH2 / 0 ch3 MERCAPTO AMINO VINYL METHACRYLYL V^j-fCHglj-O-CHg--CH--CH2 epoxy MERCAPTOSILANE COUPLING OF Hl-SIL 233 IN SBR-IS02 ----- --------------------------- - ' _' TYPtCAL FILLER-RUBBER BONDING-- STHES Fig. 9. Mercaptosilane significantly improves modulus and tensile strength of silica-filled SBRs. Two cure states show betfer crosslinking with coupling. Fig. 10. Even small quantities of mercaptosilane--less than 0.5 phr or 1 per cent on the silica--reduces heal build-up by 25F; increases tread wear by 40 per cent. UCC 015007 filler-rubber attachments). ` Thus, it is possible to introduce one or more rubber-functional groups on the surface of the filler. For poly urethane systems the amino- and epoxy- silanes are effective; for per oxide cured rubbers the vinyl and methocrylyl ore effective. The effect of mercaptosilane on a silica-filled SBR is shown in Figure 9. The SBR vulcanLzate containing 60 phr of Hi-Sil 233 and 0.7 phr of mercaptosilane shows a significantly greater stress at all elongations, lead ing to higher modulus and tensile strength, than an identical SBR vulcaaizate containing 60 phr of Hi-Sil 233 but no silane. Two cure states for each are shown to emphasize that an effective in crease in crosslinking occurs with the coupling reaction. This automatically gives higher modulus, hut coupling produces greater strength. This coupling technology has shown tremendous improvements in rubber properties containing hydrated silica or clay. In clay-filled EPDM, mercaptosilane increases modulus and tensile strength significantly. At the same time, elongation and heat build up are reduced. Similar changes are seen to occur in clay-filled polychloroprene. Here, the reduction in elongation is quite large, indicating a substantial in crease in effective cure state with mercaptosilane. This alone accounts for a sizeable increase in modulus. Even so, the retention of tensile strength at the high state of cure indi cates significant coupling activity. In polyurethane rubber, the aminosiiane is an effective coupling agent. The very large increase in tensile strength indicates coupling between the clay and the rubber chains. To obtain maximum abrasion re sistance, a combination of small par ticle size and high filler-rubber inter action seems essential. Extensive evaluations have been carried out in silica-filled rubbers. A few of these will be reported. In SBR, mercaptosilane improves the reinforcement with Hi-Sil 233 (Table 11). A typical Hi-Sil-fitled SBR was- compared with and without mercaptosilane and also with the tread black compound. Mercaptosilane increases modulus and tensile to values comparable to those of the black compound. At the same time, heat build-up is reduced to practical Table 11. Effect of mercaptosilane coupling on a Hi -Sil-fiDed SBR compound. S8R 1502 _________________________ Hi-Sil 233 ......... ....... -............. ......... N-285 Black ...................... ..............--- Process Oil ............................................. Zinc Oxide ___________ _______ _______ Stearic Acid ....................................... -- PBNA ______ Flexamine ........... --............. ........... MBTS ..................... ..................... . .... DOTG _______ _____ __________________ Sulfur ......................... ...................... - Carbowax 4000 ------ ------- --------------- Mercaptosilane ... --....................... 100 60 10 4 1 1 1.5 1.5 2.75 1.5 100 60 10 4 1 1 1.5 1.5 2.75 1.5 100 60 10 4 2 1 1 0.8 0.3 1.85 Ml 4 - 212 ... ............... ................ ........ 100 77 76 300% Modulus, psi . _____ ______ 720 Tensile ------------ --------------------------------- 2690 Elongation, % ____________ _________ 580 Hardness . ... ...... ....... ....... .................. 71 1980 3770 460 67 2210 3510 460 74 Goodrich Flexometer AT, F -------- 85 (BO) 49 73 Compression Set, % (B) ..................... 25 12 20 Pico Abrasion Index, % .................... 81 131 170 Road Wear Index _________________ 79 114 no AQ7864 Fig. 11. The brood spectrum of particle size and structure of non-black fillers already provides a wide variety of properties; but, with the coming of tailored surface activity of these fillers, the possibilities for improved modified rubbers could become enormously greater in the future. UCC 015008 [eve'j. The most significant change was in tread wear. Mercaptosilane improved the Road Wear Index of the Ili-Sil tread from the 80% level to the 110% level or equal to the black tread. The effect of mercaptosilane on heat build-up and tread wear is re markable (Fig. 10). Even very small quantities, less than 0.5 phr or 1 c/o on the silica, reduces heat build-up by 2o'-F and increases tread wear by 40%. Maximum benefit to tread wear occurs at about 2% on the silica. This demonstrates the flexi bility of this technique in that prop erties can be adjusted to meet requir ..nents. Similar property improvement is obtained with mercaptosilane in a modern passenger tread. Using a blend of oil extended SBR and poly butadiene, tread wear equivalent to the black control was realized. Lab oratory properties were unable to pre dict this increase in abrasion resist ance. This emphasizes the care that must be exercised in comparing black and non-black pigments on the basis of laboratory properties alone. In a nitrile rubber (Table 12), mercaptosilane is similarly effective. Only 1.2 phr of the silane enhances modulus nearly threefold, while de creasing compression set to accept- able levels. A decrease in viscosity, a benefit frequently obtained with the silane, is observed in this data. Further, the mercaptosilane has no effect on aging in this sulfurless cure. Abrasion resistance is significantly increased, evidence that reinforce ment is improved through coupling. Many other studies have been conducted in natural rubber, EPDM, chloroprene rubber and epichlorohydrin rubber. All respond to the coupling activity of mercaptosilane. Much improved modulus, tensile and dynamic properties are obtained in all polymers. The cost of the improvement ob tained with silane coupling may be uneconomical for many applications. However, the special requirements of some rubber products can only be achieved in this manner. These in clude nitrile, EPDM and epichlorohydrin compounds which need high resilience, low set, gas permeability, and in some cases color that can only be obtained through a silica-silane combination. Future of non-black pigments What of the future of non-black pig ments? There will, of course, be con tinued growth in the existing prod ucts. There will also be new varieties of some. For example, there is a de mand for a non-black pigment with better resilience and modulus and with more normal viscosity and cure behavior. Such a product will be forthcoming in a few months. Look for new innovations in the application of fibrous fillers. With increased bonding, the full utilization Table 12. Effect of coupling agents; NBR - Sulfurless Cure. Medium High Nitrile Rubber .. Zinc Oxide . ______ Srsaric Acid_________ ________ Lubricant _____ . ....... ............ DO? ___________ _____ ________ Antioxidant _____ ___ _____ Hi-Sil 233 TMTD Sulfur...................................... . A-189 100.0 5.0 2.0 1.0 10.0 1.0 70.0 3.0 0.2 A*b A C> V 1.2 ML-4 212 ___________________ .. 106 82 Stress-Sfrain at 10'/300 Original Aged1 Original Aged* 300% Modulus, psi .... .. 820 Tensile, psi ____ ________.. 3060 Elongation, % ........... .. ._ 570 Hardness ______________ 78 Compression Set, % (B) ......... 35 N3S Abrasion Index, % ____ 86 3080 560 84 2010 3640 440 78 22 234 3570 370 84 11 70 Hr*. 212'F AUTHOR Melvin P. Wagner of PPG Industries predicts an exceptionally bright future for non-black fillers. of fiber character can be obtained. Tailoring of surface activity of non black fillers will become common place. The possibilities are enormous. Some examples have been shown in this presentation, but it is possible to speculate even further. Heretofore, filler particle size and structure have been varied at will, but surface activity has either been low or high (Fig. 11). Beneficial properties are obtained at both ends of the activity spectrum. For exam ple, high tear strength is apparently associated with low activity, while high abrasion resistance requires high activity. We have seen that tt is pos sible to obtain both in a single pig ment by adjusting surface activity. Placing of rubber pigments in this cube (Fig. 11) shows that a wide variety of properties is already avail able. Examples of low and high sur face, structure and activity are present in a number of materials. None pos sess the potential for all variations. Several which have wide flexibility in particle size and structure are amenable to changes in surface ac tivity. The silicas, for example, al ready exist in a range of particle sizes and structures, and their activity can readily be altered by using silanes. Other possibilities also exist. The future of non-black fillers is exceptionally bright. Perhaps a more universal filler will eventually exist which will have the flexibility of variable size, structure and activity by the judicious choice of special compounding additives. At present, this pigment may well be a non-black one. Q UCC 015009 Rubber LUorid NOVEMBER, 1971 Compounding with fibers for high performance elastomer compounds Studies indicate that while elastomeric composites loaded with short inorganic 'fiberettes' con be easily dispersed directly in a Banbury or aon mill, it proves more advantageous with organic fibers--particularly with high loadings--to first prepare a stiff fiber-rubber masterbatch, and then incorporate this stiff masterbatch into the main compound. A0786G Jv C. Derringer l'l'(r lmiusiries, Inc. Barberton, Ohio Discontinuous fibers, because of their large length-to-diameter ratios, are potentially very effective fillers for modulus reinforcement of elastomers, [his is illustrated in the equation by Gnth1: G -- Gt, if + 0.67 fc + l.o2 Pc-) where: c = volume concentration of filler t -- length-to-diameter ratio of filler particle G --" modulus of filled com pound C., -- modulus of unfilled rubber At a gireti fiber concentration, modulus increases with the square of the length-to-diameter ratio. Thus, ?:nre the length-to-diameter ratios of t v iibers employed in this study are m the vicinity nf 10-50, the modulus of the resulting composites can he expected to be 2-3 orders of magni tude greater than obtainable with re i;ifi)rcir:g blacks. This equation is ui K valid, however, when good ndh.--ion exists between fibers and elast'.i:it-.-i jc- matrix. Lnfortunately. with- : Z. G-'>\ R. Simt xti 0 Gold, Kalljxl Z?;i . 193-3, 74. 3'CS, Thesis, Vienna, 1337. out special odditis es, adhesion be tween most fibers and elastomers is typically very pool. HRH system A li ieomponeut system consisting ot hexamethylenetetramine, resorcinol and fine particle hydrated silica (i.e., HRH i has been found- to bring about extremely good adhesion be tween most types of elastomers and most common wowii fabrics. Good adhesion was also found to result when the HRH system was added to many discontinuous fiber-elastomer composites3. This is illustrated in Figure 1 which shows stress strain curie; for an SRF filled NR/SBR compound containing chopped, un coated fiber glass with and without HRH. The formulations are given in Table I. Without HRH the composite ex hibited high modulus up to approxi mately 30(7 elongation. Beyond this strain (stress) level, however, the fiber-matrix adhesion apparently fail ed. resulting in poor, high strain modulus. The composite with HRH, on the other hand, exhibited a nearly linear stress strain curve and modulus - 0.7,csv:t TtcHri'bChe Btrieve. "Ailtesion of R`JbCofr.pjunCs cn Teililes arv5 Melals*' (19G/J. ; Psrrtrver. G. C.. Advdrcfrs ir. Coemi^l'y Series, M3, si 1971 PHASE MICROSCOPE hod to ba used to get details of microtomed com posites such as the above fiber glass in rubber composite (magnification, 400X). Microscope provided sharper contrasts between matrices and fi bers with similar refractive indices. UCC 015010 Table I. Fiber glass--NR/SBR composite with and with out HRH. Parts by V/eight (without HRH) (with HRH) Natural Rubber .. ________ S3R 1712 ................. ....... SRF .............. . ...................... Hydrated Silica3 .................... Zinc Oxide ________ _____ .. Stearic Acid ___ _________ Napthenic Oil .___ .. .......... Fiber Gloss1' ................... .. ResorciroH .............. 70 30 30 5 2 4 150 70 30 30 20 5 2 4 150 2.5 Hexamethylenetetramine11* Benzat'niazyl Disulfide* . Diphenylguanidine* ______ Sulfur* _____ _______________ 0.85 0.35 2.4 1.6 0.85 0.35 2.4 2 Hi-SH 233, PPG Industries, Inc. c Technical grade- J Hexc Flo Powder, Heyden Chemlcol Div,, Tenneco, Inc. * Added on mill. ktooo Fig. 2. Variotions of Young's Modulus with fiber glass loading in a polychloroprene compound containing HRH. y* << 5 2*00 500 2200 400 VT / / / T,, * ;V T0 2022.6 c i!M5 6 isa.r / E0 ` 410.225 f * 0.J7 ^---- 9 0J | 1 1 i ;-------------------- "0 5 IO IS 20 2S 30 35 40 VOLUME LOAOING OF rsEft GLASS Fig. 3. Tensile and elongation at break; fiber glass loading in a polychloroprene compound containing HRH. 'X CO 200 STRAIN, S'. ELONGATION Fig. 1. HRH effect on fiber glass -NB/SBR composite. O Table II. Standard polychloroprene formulation containing HRH. Polychloroprene* ........... 100 N-phenyl-alpha-napthylamine 2 Magnesium Oxide ............ 4 Hydrated Silica _____________ 25 Naphthenic Oil ............... 15 Resorcinol ........ 1.7 Stearic Acid __________________ 2.5 Diorthofolylguanidine* ___ .... 0.5 Sulfur* ............ 1.0 Hexamethylenetetramine* _. 1.6 * Napr*ne W, E. I. dvfont. ' Added on mill. with fiber loading is iiFStrated in Figures 2 and 3 for fiber glass in polychloroprene formulation is given in Table II. For modulus, tensile and elougation the following empirical equations have been found adequate for most of the fiber and elastomer types investigated3, Young's Modulus G = Go --1 + exp (aVb) Tensile T = To + cV - dVV2 Elongation E = E co + (Eo-E ) exp (--fV) where: levels higher than can be obtained with conventional reinforcing pig ments. Variation with fiber loading Variation of stress strain properties G " Young's modulus of com posite, psi Gq -- Young's modulus of matrix compound, psi T = Tensile strength of composite, psi To = Tensile strength of matrix compound, psi E = Elongation at break of com posite, % E co -- Limiting elongation of highly ioaded compound Eg = Elongation at break of mat rix compound, % a, b, c, d, f and g are empirical parameters V = volume loading of fiber, cm3 per 100 g rubber The modulus curve (f ig. 2) could also be represented by a quadratic function to conform to Guth's model, but the empirical exponential model has been found more flexible and capable of representing all of the fiber composites studied to date. In Figure 3 it is seen that small fiber loadings have a very large effect on both tensile and elongation. For tensile strength, the initial sharp loss is gradually recovered as the fiber loading is increased. The matrix ten sile strength is regained at approxi mately four times the loading giving minimum tensile, by virtue of the pro posed equation. Elongation at break UCC 015011 Table III. Fibers vs. FEF in natural rubber. Formulations (parts by weight) Naturgi Rubber ____________ Hydrated Silica ____ ________ Rayon Fiber3 ____ ____ _________ Fiber Glass Ultrafine Asbestos1" rFr .............................. ......... .................. XR124RC* ...._ ....... ....__..... Zinc Oxide* _____________ ____ Hexamethylenetetramine* Benzctniazyl Disulfide ... Diphenylguanidine* ___________ Sulfur* ___ ____ ____ ____ _____ TOO 12 9 4.2 5 1.6 0.7 0.3 2.5 18 15 30 15 30 35 50 65 ODR RFteometer Properties at 3003F Minimum Torque, in.-lb............ . Maximum-Torque, in.-lb. .. 95% Cure Time, minutes 3.8 67.4 13.5 2.6 7B.4 14.5 4.0 64.4 13.5 5.5 74.3 13.0 5.0 79.2 15.0 5.9 103.0 16.5 6.4 62.0 13.0 9.5 67.0 14.0 11.3 67.9 16.5 Stress Strain Properties Cure at 300 ; F, minutes _____ 30 Young's Modulus, psi _____ .. 760 20% Modulus, psi _________ 150 100% Modulus, psi ......... 795 Tensile, psi ......... 1680 Ultimate Elongation, % ___ ... 350 Hardness, Shore A ............ . 63 Compression Modulus at Zero Strain, psi ........................ . 732 30 1330 255 1290 1710 195 68 30 840 150 885 1920 310 57 596 30 1960 360 1680 1800 120 67 30 1400 300 1140 2340 295 67 1138 30 3376 615 1890 2550 175 82 15 295 70 280 3850 500 53 596 15 389 90 430 3120 420 63 759 15 486 100 565 2840 320 67 894 Yerzley Oscillograph Resilience, % __________ _______ Dynamic Modulus, psi................ Static Modulus ........................ 88 1107 880 82 1506 1 180 91 995 760 88 1444 1140 81 1709 1010 72 2819 1670 74 1381 850 Goodrich Flexometer (212F - 163 psi - 17.5%) Static Compression, % ......... ,, Dynamic Compression, % ____ Dynnmlc Drift, % Permonent Set, % _ Heat Build-Up, F^ .. . .. 19,5 7.9 3.5 6.5 27 15.6 3.5 8.7 9.3 43 Goodyear Resilience, % . 85 20.5 8.7 3.6 6.2 23 84 16.7 - 5.1 2.8 4.7 25 15.1 3.5 14.2 9.1 39 78 9.0 13.2 61 20.3 8.1 17.2 16.0 71 17.2 5.4 35.9 33.0 120 15.9 5.2 33.1 29.8 111 76 70 66 3 60 cuf's per inch. Microfibers Inc. b Cofedrea HPP, Union Carbide. c 60/40 Peiorcinol/Stearic Acid Masterbotch, Horv/ick Standard Chemicol Co. ' Added on rrilf. A07tib. UCC 015012 Fig. A. Effect of applied fore reiolive to fiber glass orientation on extension, compressive force. 28 VOLUMES FEE 9 Fig. 6. Effect of fibers and FEF on heat buildup. 28 Volumes of FEF 37.5 drops continuously but most rapidly at the lower loadings. For most of the various types of composites studied, elongation leveled out at 20-50%. Because the rapid loss of elongation with increased fiber loading is found only when good fiber-matrix adhesion js obtained, ultimate elongation is a good index of fiber-matrix adhesion, especially at higher fiber loadings. NR-HRH composites at low fiber loadings The remainder of this exposition deals with natural rubber composites con taining relatively low volume load ings of fiber glass, rayon, and ultrafine asbestos (Table III). The hexa methylenetetramine was added on the mill to prevent premature reaction with the resorcinol. In all extension tests the fiber was oriented parallel to the direction of applied force. In all compression tests the fiber was oriented perpendicular to the direc tion of applied force. Orientation All rubber compounds demonstrate what is known as a grain effect, or anisotropy. That is, physical proper ties are dependent upon the direction UCC 015013 i,t fatn-tilied -trees relative tn the *g~ rain &1 r-rilOU. Hie ~o-called grain is a result of the milling operation, which tends to '.jUbL molecular orientation in (lie tnliitt.e direction. When the reiufurtiu": material has a letigtli-to-diauieter rat;.i ereater than 1 vve might expect i; t... also become oriented during i.'-ilh.-.g. Tliis is actually what hap pen* lvitii fiber reinforcement. Firure 4 illustrates stress strain curies for a 6-volunv; i 15 pbr; loudit:a <A fiber glass with HKH iti the standard natural rubber formulation shown in Table III. Curves are shown for 11.j:'r extension ) A 7>1.) and cornpie-sit.-r. ' ,\>1) with tlie force ap plied hot!; parallel and perpendicular to tile tiher orieutatioti direction. Lambda. A . is the extension ratio if A is > 1 or the compression ratio if A is < 1. In any case, A -~ deform ed dimension original dimension. It is quite interesting that in ex tension. greatest stress results from a parallel application of force, whereas in compression it results irons a per pendicular force application. The I'ionuression sample was a cylinder I inch in length by "b inch in diam eter. L pon compression the deforma tion was perpendicular to the direc tion ..I applied force rather than par allel. which is the case for extension. In short, greatest reinforcement is thus obtained when the fiber orien tation direction is parallel to the di rection of deformation of the test piece. The same general behavior is obtained with high structured blacks su'd! as FEF: however, the magnitude of the effect is much smaller. For tin- fiber glass composite, no extra milling was given to the com pound so that it was far from com pletely orientated. If complete orien tation could, in fact, be accomplished the two curves in Figure 4 would be considerably more separated. Higher fiber loadings also increase the degree of anisotropy. The prop erty. of anisotiupy, previously barely obtainable via conventional com pounding. can lie very useful in the desirn of rubber articles. Modulus and resilience Bv tar the most unique, property of lIKlt-fiber composites is their ex tremely high extension modulus. Mod ulus values can be obtained which arc rnnipb-teh out of reach with any type ol carbon Mark presently avail able. For example, the 2.0'.c modulus values in Table III which range as high as 6lfi psi cannot be matched by an equivalent loading of any exist ing carbon black. At high fiber load ings 12U-7U volumes>, A dung's mod ulus values in the 2U-O09--10,01)0 psi range have been obtained. This is two orders of magnitude higher than can be obtained with anv carbon black. Perhaps as important as the high modulus is the relatively high resil ience which is simultaneously ob tained. This applies to both static and dynamic modulus, for example, in Tabic HI. 50 phr l 2d vol.l of f El gave a Youngs modulus oi 309 psi, Yerzeley static compression modulus of 050 psi, Yerzeley dynamic com pression modulus of L3<j1 psi and a Yerzeley resilience of i4'/< . The corlesponding values for the 30 phr *12 volume I fiber glass composite were I960 psi. 1140 psi, 1444 psi. and Eli'4, respectively. Thus at a considerably lower loading, the fiber glass resulted in significantly higher modulus levels, especially in exten sion. in addition to considerably higher resilience. The effect of fiber loading on Goodyear resilience is compared to that of fSAF-HM black in Figure o. The formulation is the same as that uf Table HI with the exception that 25 phr of hydrated silica was em ployed instead of 12. Thus the fiber curves begin at a volume loading of 10. It is the slopes of these lines that we are interested in and the fiber glass slope is 0.55 percent per volume vs. 1 percent per volume for the black. In other words, one part bv volume of I5AF-HM reduced the iesilience about twice as much as an equal amount of fiber glass. Heat buildup and permanent set Along with higher resilience. HRH fiber composites exhibited consider ably lower heat buildup and perma nent set than reinforcing carbon blacks. A comparison of the 9 phr rayon composite with the 50 phr FEF loading illustrates this difference quite well. The two compounds ex hibited roughly equal hardness and static compression modulus while the composite had slightly lower dynamic compression modulus. The heat build up and permanent set of the com posite. however, were 27 F and 6.5/c- respectively, compared to UCC 015014 120' F and \YV,r Mr the FEF. figure 6. which shows heat build up as j function of time, gives a very good comparison of the behavior of the composites vs. FEE black. Where as the comp.c-ites reached equilibrium heat buildup at 20 minute;, the FEF v ukanizato exhibited a continual in crease until the sample blew out. Figure 7 is a photograph of (lie test specimens alter 60 minutes of testing. 1 he difference in permanent set is quite striking. Whereas the com posites exhibited only very slight di mensional changes, the hljck com pounds were extremely deformed, with llie high loading showing a blow out. Creep To study creep behavior, the 6 vol ume loading composites and 28 vol ume FEF comparison were subjected to a load of 390 psi at 212;l and elongation measured as a function of time. The linear extension ratio vs. log time I hours t plot is given in Figure 8. As can be seen, the com posites exhibited considerably lower creep than the FEF in spite of the considerably lower loadings. The val ues of fractional increase in extension ratio per decade of time are given in Tabic IV. Stress softening Early during the course of work with FIRFf fiber composites, it was ob served that the composites exhibited a rather high degree of stress soften ing. To determine the importance of this, the 6 volume composites and all three loadings of I FF were stretched repeatedly to I0U|t elongation. The results are shovvn in Figure 9. Each curve was fitted to the following equation: G = C cc 4 S exp [--a (A-l)b} where: G = iOO'/l modulus at anv given stretch G -- limiting modulus, i.e. JOO' e modulus after an infinite number of stretches S -- amount oi softening A -- number of stretches a, b -- fitted constants The equation parameters are sum marized in Table V. A plot of 100Ft modulus on the first stretch > i.e., ~t Si vs. Moo is shown in Figure 10. The slope of the resulting least square line was 0.199. In other words, all vulvanizates (fiber or FEE filled) ultimately lost YOG, of their A0787C Table V. Parameters for softening equation. G -- G x r S exp f - a(N-lb)] Loading M Sa 6 Volumes Ravon ____ .. 380.575 6 Volumes Fiber Glass............... 351.75 6 Volumes Ultrafine Asbestos 606.82 19 Volumes FEF _________________ 176.10 28 Volumes FEF ......... 268.75 36 Volumes FEF ---------------- --------- 324.11 409.365 498.253 593.21 90.90 134.29 235.92 0.682 0.653 0.906 0.448 0.333 0.980 b 0.404 0.403 0.447 0.319 0.653 0.489 number of stretches, n Fig. 9. A comparison of the effect of fibers and FEF on stress softening. Fig. 10. Initial and limiting modulus; composites end FEF filled vulcanizates. original modulus so that comparison of modulus on the first sfrelch is valid. Processing Of al! the fiber types evaluated, fiber glass is hv far the easiest to disperse. It can either be added in the Banbury or on the mill during the final stage of mixing. The latter results in longer fiber lengths because fiber breakage is minimized. Ullrafine asbestos is not as easy to disperse but disperses well when added in the Banbury early in the mix cycle. Organic fibers such as rayon, ny lon. and orlun are quite difficult io disperse, especially at high loadings, f or best results, a stiff masterbatch of liber and rubber, usually a; equal parts bv weight, is first prepared in the Danbury. This masterbatch is then added in the Banbury to the final compound. For the HUH system to be effec tive. resorcinol should be added in the Banbury and the hexamethylene tetramine in the second stage of mix at temperatures lower than 150'F. This prevents premature reactions of resorcinol and hexamethylenetetra mine. It is further recommended that the hydrated silica constitute at least 1/3 of the total pigment loading. Fiber and elastomer types To date, rayon, orlon, fiber glass, asbestos, and nylon have been found to be effective in HRH composites with natural rubLer, NBR, 5BR, BR or polycldoroprene matrices. The HRli system is not effective with polyester fiber in any elastomeric matrix. Initial work with the HRH system indicated that it was ineffective in highly saturated elastomers such as butyl and EFDM, Recent work, how ever, has shown that when sufficient zinc oxide and stearic acid li.e., ap proximately 5 phr of each I are in cluded in the formulation, good ad hesion to nylon can be obtained. It is, therefore, expected that with ade quate stearic acid and zinc oxide, high performance EPDM and butyl composites will be possible with the HRH system. Applications The most promising applications for HRH composites are those in which high modulus is required in combi nation with high resilience. Applica tion examples: gaskets, motor mounts and all types of rubber springs. The high degree of anisotropy will be useful in such applications as tubing, where swell can he minimized without decreasing elasticity. Furthermore, since the HRH sys tem promotes adhesion to metal and many other substrates, the HRH com posites have a dual advantage of high performance and good adhesion. It is expected that these new com posite materials will find their way into manv diverse applications since they offer properties previously un obtainable in robber. f~| F-3,S r2 - 1M UCC 015015