THERMOSET ELASTOMERS

209

THERMOSET ELASTOMERS J. E. MARK Department of Chemistry and the Polymer Research Center University of Cincinnati, Cincinnati, Ohio 45221-0172

Introduction Basic Concepts Some Historical High Points Some Rubberlike Materials Preparation of Networks Some Typical Applications Experimental Details Mechanical Properties Swelling Optical and Spectroscopic Properties Scattering Stress-Strain Behavior Control of Network Structure Networks at Very High Deformations Non-Gaussian Effects Ultimate Properties Multimodal Chain-Length Distributions Other Types of Deformation Biaxial Extension Shear Torsion Swelling Filler-Reinforced Elastomers and Elastomer-Modified Ceramics Current Problems and Future Trends

Introduction Basic Concepts. Elastomers are defined by their very large deformability with essentially complete recoverability. In order for a material to exhibit this type of elasticity, three molecular requirements must be met: (i) the material must consist of polymeric chains, (ii) the chains must have a high degree of flexibility and mobility, and (ill) the chains must be joined into a network structure (1-3).

210

J.E. Mark

The first requirement arises from the fact that the molecules in a rubber or elastomeric material must be able to alter their arrangements and extensions in space dramatically in response to an imposed stress, and only a long-chain molecule has the required very large number of spatial arrangements of very different extensions. The second characteristic required for rubberlike elasticity specifies that the different spatial arrangements be accessible, i. e., changes in these arrangements should not be hindered by constraints as might result from inherent rigidity of the chains, extensive chain crystallization, or the very high viscosity characteristic of the glassy state (1.2.4 .58). The last characteristic cited is required in order to obtain the elastomeric recoverability. It is obtained by joining together or "cross linking" pairs of segments, approximately one out of a hundred, thereby preventing stretched polymer chains from irreversibly sliding by one another. The network structure thus obtained is illustrated in Figure 1, in which the cross links are generally chemical bonds (as would occur in

Figure 1. Sketch of a typical elastomeric network, with an interchain entanglement depicted In the lower right-hand corner. sulfur-vulcanized natural rubber). These elastomers are frequently included in the category of "thermosets", which are polymers having a network structure which is generated or "set" by thermally-induced chemical cross-linking reactions. The term has now frequently taken on the more specific meaning of networks that are very heavily cross linked and below their glass transition temperatures. Such materials, exemplified by the phenol-formaldehyde and the epoxy resins, are very hard materials with none of the high extensibility associated with typical elastomers. The cross links in an elastomeric network can also be temporary or physical aggregates, for example the small crystallites in a partially crystalline polymer or the glassy domains In a multi-phase triblock copolymer (3.6). The latter materials are considered separately in the chapter on "Thermoplastic Elastomers". Additional information on the cross linking of chains is given below.

Thermoset Elastomers

211

Some Historical High Points. The earliest elasticity experiments involved stress-straintemperature relationships, or network "thermoelasticlty". They were first carried out many years ago, by J. Gough, back in 1805 n.2.4.5.9.10). The discovery of vulcanization or curing of rubber into network structures by C. Goodyear and N. Hayward in 1839 was important in this regard since it permitted the preparation of samples which could be investigated in this regard with much greater reliabiity. Such more quantitative experiments were carried out by J. P. Joule, in 1859. This was, in fact, only a few years after the entropy was introduced as a concept in thermodynamics in general! Another important experimental fact relevant to the development of these molecular ideas was the fact that mechanical deformations of rubberlike materials generally occurred essentially at constant volume, as long as crystallization was not induced (1). (In this sense, the deformation of an elastomer and a gas are very different). A molecular interpretation of the fact that rubberlike elasticity is primarily entropic in origin had to await H. Staudinger's much more recent demonstration, In the 1920's, that polymers were covalently-bonded molecules, and not some type of association complex best studied by the colloid chemists (1). In 1932, W. Kuhn used this observed constancy in volume to point out that the changes in entropy must therefore involve changes in orientation or configuration of the network chains (4.6). Later in the 1930's, W. Kuhn, E. Guth, and H. Mark first began to develop quantitative theories based on this idea that the network chains undergo configurational changes, by skeletal bond rotations. In response to an Imposed stress (1.2). More rigorous theories began with the development of the "Phantom Network" theory by H. M. James and E. Guth in 1941, and the "Affine Model" theory by F. T. Wall, and by P. J. Flory and J. Rehner, Jr. in 1942 and 1943. Modern theories generally begin with the phantom model and extend it, for example by taking into account interchain interactions

Some Rubberlike Materials. Since high flexibility and mobility are required for rubberlike elasticity, elastomers generally do not contain stiffening groups such as ring structures and bulky side chains (2.4). These characteristics are evidenced by the low glass transition temperatures Tg exhibited by these materials. Such polymers also tend to have low melting points, if any, but some do undergo crystallization upon sufficiently large deformations. Examples of typical elastomers include natural rubber and butyl rubber (which do undergo strain-induced crystallization), and poly(dimethylsiloxane), poly(ethyl acrylate), styrene-butadiene copolymer, and ethylene-propylene copolymer (which generally don't). The most widely used elastomers are natural rubber, synthetic polyisoprene and butadiene rubbers, styrene-butadiene copolymers, ethylene-propylene rubber (specifically EPDM), butyl and halobutyl elastomers, polyurethanes, polysiloxanes, polychloroprenes, nitrile rubber, polyacrylic rubbers, fluorocarbon elastomers, and thermoplastic elastomers ( H ) . The examples which have unsaturation present in the repeat units (such as the diene elastomers) have the advantage of easy cross linkability, but the disadvantage of increased vulnerability to attack by reactants such as oxygen and ozone.

212

J.E.Mark

Some polymers are not elastomeric under normal conditions but can be made so by raising the temperature or adding a diluent ("plasticizer"). Polyethylene is in this category because of its high degree of crystallinity. Polystyrene, poly(vinyl chloride), and the biopolymer elastin are also of this type, but because of their relatively high glass transition temperatures require elevated temperatures or addition of diluent to make them elastomeric (4.5). A final class of polymers is inherently nonelastomeric. Examples are polymeric sulfur, because of its chains are too unstable, poly(p - phenylene) because its chains are too rigid, and thermosetting resins because their chains are too short (4). Preparation of Networks. One of the simplest ways to introduce the cross links required for rubberlike elasticity is to carry out a copolymerizatlon in which one of the comonomers has a functionality (^ of three or higher (4.12). This method, however, has been used primarily to prepare materials so heavily cross linked that they are in the category of hard thermosets rather than elastomeric networks (12) • The more common techniques include vulcanization (addition of sulfur atoms to unsaturated sites), peroxide thermolysis (covalent bonding through free-radical generation), end linking of functionally-terminated chains (isocyantes to hydroxyl-terminated polyethers, organosilicates to hydroxyl-terminated polysiloxanes, and silanes to vinyl-terminated polysiloxanes). For commercial materials, the compounding recipe generally contains numerous ingredients in addition to the polymer and cross-linking agent (for example, sulfur, a peroxide, or a isocyanate) (14). Examples are activators (to increase cross-linking efficiency), retarders (to prevent premature cross linking or "scorch"), accelerators, peptizing agents, antioxidants and antiozonants, softeners, plasticizing aids, extenders, reinforcing fillers (typically carbon black or silica), and processing aids. Specific applications can require still additional additives, for example blowing agents in the case of elastomeric foams, thermally conducting particles \n the case of heated rollers, fiber meshes in the case of high pressure tubing, etc. A sufficiently stable network structure can also be obtained by physical aggregation of some of the chain segments onto filler particles, by formation of microcrystallites, by condensation of ionic side chains onto metal ions, by chelation of ligand side chains to metal ions, and by microphase separation of glassy or crystalline end blocks in a triblock copolymer (4.5). The main advantage of these materials is the fact that the cross links are generally only temporary, which means that such materials frequently exhibit reprocessability. This temporary nature of the cross linking can, of course, also be a disadvantage since the materials are rubberlike only so long as the aggregates are not broken up by high temperatures, presence of diluents or plasticizers, etc. Some Tvpical Applications. Typical non-biological applications are tires, gaskets, conveyor belts, drive belts, rubber bands, stretch clothing, hoses, balloons and other inflatable devices, membranes, insulators, and encapsulants. Biological applications include parts of living organisms (skin, arteries, veins, heart and lung tissue, etc.), and various biomedical devices (contact lens, prostheses, catheters, drug-deliver systems, etc.). It is interesting to note that most of these applications require only small

Thermoset Elastomers

213

deformations; relatively few take advantage of the very high extensibility that is characteristic of most elastomeric materials! Frequently, specific applications require a particular type of elastomer (15). For example, a hose should have as large a mismatch of solubility parameters with the fluid it will be transporting. Thus, a polar elastomers such as polychloroprene would be best for hoses used with hydrocarbon fluids such as gasoline, jet fuel, greases, oils, lubricants, etc. Some Experimental Details Mechanical Properties. The great majority of studies of mechanical properties of elastomers have been carried out in elongation, because of the simplicity of this type of deformation (3.4). Results are typically expressed in terms of the nominal stress f* s f/A* which, in the simplest molecular theories, is given by f = (vkTA/) (a - a-2)

(1)

where vA/ Is the density of network chains, i. e., their number per unit volume V, k is the Boltzmann constant, T is the absolute temperature, and a Is the elongation or relative length of the stretched elastomer. Also frequently employed is the modulus, defined by [f*] = fV^/3/ (a - a-2) = VkTA/

(2)

where V2 is the volume fraction of polymer in the (possibly swollen) elastomer. There are a smaller number of studies using types of deformation other than elongation, for example, biaxial extension or compression, shear, and torsion. Some typical studies of this type are mentioned below. Swelling. This non-mechanical property is also much used to characterize elastomeric materials (1.2.4.12V It is an unusual deformation in that volume changes are of central importance, rather than being negligible. It is a three-dimensional dilation in which the network absorbs solvent, reaching an equilibrium degree of swelling at which the free energy decrease due to the mixing of the solvent with the network chains is balanced by the free energy increase accompanying the stretching of the chains. In this type of experiment, the network is typically placed into an excess of solvent, which it imbibes until the dilational stretching of the chains prevents further absorption. This equilibrium extent of swelling can be interpreted to yield the degree of cross linking of the network, provided the polymer-solvent interaction parameter %i is known. Conversely, if the degree of cross linking is known from an independent experiment, then the interaction parameter can be determined. The equilibrium degree of swelling and its dependence on various parameters and conditions provide, of course, important tests of theory. Optical and Spectroscopic Properties. An example of a relevant optical property is the birefringence of deformed polymer network {^Z)^ This strain-induced birefringence can

214

J.E. Mark

be used to characterize segmental orientation, both Gaussian and non-Gaussian elasticity, crystallization and other types of chain ordering, and short-range correlations (2.4). Other optical and spectroscopic techniques are also important, particularly with regard to segmental orientation. Some examples are fluorescence polarization, deuterium NMR, and polarized Infrared spectroscopy (4.12.16). Scattering. The technique of this type of greatest utility in the study of elastomers is small-angle neutron scattering, for example, from deuterated chains in a non-deuterated host (4.12.17). One application has been the determination of the degree of randomness of the chain configurations in the undeformed state, an issue of great importance with regard to the basic postulates of elasticity theory. Of even greater importance is determination of the manner in which the dimensions of the chains follow the macroscopic dimensions of the sample, i. e., the degree of "affineness" of the deformation. This relationship between the microscopic and macroscopic levels in an elastomer is one of the central problems in rubberlike elasticity. Some small-angle X-ray scattering techniques have also been applied to elastomers. Examples are the characterization offillersprecipitated into elastomers, and the corresponding incorporation of elastomers into ceramic matrices, in both cases to improve mechanical properties (4.18-21). Typical Stress-Strain Behavior A typical stress-strain isotherm obtained on a strip of cross-linked elastomer such as natural rubber is shown schematically in Figure 2 (1-3). The units for the force are gen-

Figure 2. Stress-elongation curve for an elastomer showing an upturn in modulus at high elongations. erally Newtons, and the curves obtained are usually checked for reversibility. In this type of representation, the area under the curve is frequently of considerable interest since It is proportional to the work of deformation w = J fdL. Its value up to the rupture point is thus a measure of the toughness of the material.

Thermoset Elastomers

215

The upturn in modulus at high elongation is of particular interest since it corresponds to an increase in toughness. It is generally due to strain-induced crystallization, resulting from increase in melting point of the network chains. This is, in turn, due to the decreased entropy of the stretched chains and the fact that the melting point is inversely proportional to the entropy of melting. In some cases, however, the upturns can be due to the limited extensibility of the chains. These instances are easy to identify, since these upturns will not be diminished by decreasing the amount of crystallilzatlon by increase in temperature or by addition of a diluent. It is in this sense that the stretching "induces" the crystallization of some of the network chains (5). The initial part of the stress-strain isotherm shown in Figure 2 is of the expected form in that f approaches linearity with a as a becomes sufficiently large to make the subtractive a-2 term in Equation (1) negligibly small. The large increase in f* at high deformation in the particular case of natural rubber is due largely, if not entirely, to strain-induced crystallization. Additional deviations from theory are found in the region of moderate deformation upon examination of the usual plots of modulus against reciprocal elongation (2.22). Although Equation (2) predicts the modulus to be independent of elongation, it generally decreases significantly upon increase in a (22). The intercepts and slopes of such linear plots are generally called the Mooney-Rivlin constants 2Ci and 2C2, respectively, in the semi-empirical relationship [f] = 2Ci + 2C2a-''. As described above, the more refined molecular theories of rubberlike elasticity (6.23-26) explain this decrease by the gradual increase in the non-affineness of the deformation as the elongation increases toward the phantom limit. Control of Network Structure Until recently, there was relatively little reliable quantitative information on the relationship of stress to structure, primarily because of the uncontrolled manner in which elastomeric networks were generally prepared (1-4). Segments close together in space were linked irrespective of their locations along the chain trajectories, thus resulting in a highly random network structure in which the number and locations of the cross links were essentially unknown. Such a structure is shown above in Figure 1. New synthetic techniques are now available, however, for the preparation of "model" polymer networks of known structure (3.4.6.27-44). An example is the reaction shown in Figure 3, in which hydroxyl-terminated chains of poly(dimethylsiloxane) (PDMS) are end-linked using tetraethyl orthosilicate. Characterizing the uncross-linked chains with respect to molecular weight Mn and molecular weight distribution, and then running the specified reaction to completion gives elastomers In which the network chains have these characteristics, in particular a molecular weight Mc between cross links equal to Mn, and cross links having the functionality of the end-linking agent. The end-linking reactions described above can also be used to make networks having unusual chain-length distributions (45-48). Those having a bimodal distribution are of particular interest with regard to their ultimate properties, as will be described below.

216

J.E.Mark

\ HO ^ \ / ^ OH

+ (C2H50)4SI

^

/

/^'v

"^ 4 C2H5OH

In this endlinking reaction, HO ^ \ / ^ OH represents a hydroxyl-terminated poly(dimethylsiloxane) chain. The average molecular weight and its distribution for the precursor chains becomes the average molecular weight and its distribution for the network chains.

Figure 3. A typical end-linking scheme for preparing an elastomeric network of known structure. Networks at Very High Deformations Non-Gaussian Effects. As already described in Figure 2 (1-3). some (unfilled) networks show a large and rather abrupt increase in modulus at high elongations. This increase (49.50) is very important since it corresponds to a significant toughening of the elastomer. Its molecular origin, however, has been the source of considerable controversy (2.4.49.51-57). It had been widely attributed to the "limited extensibility" of the network chains (55), i. e., to an inadequacy in the Gaussian distribution function. The issue has now been resolved (6.55.58-6QV however, by the use of endlinked, non-crystallizable model PDMS networks. These networks have high extensibilities, presumably because of their very low incidence of dangling-chain network irregularities. They have particularly high extensibilities when they are prepared from a mixture of very short chains (molecular weights around a few hundred g moh'') with relatively long chains (around 18,000 g mol-''), as further discussed below. Apparently the very short chains are important because of their limited extensibilities, and the relatively long chains because of their ability to retard the rupture process. Comparisons of stress-strain measurements on such blmodal PDMS networks with those in crystallizable polymer networks such as natural rubber and c/s-1,4polybutadiene were carried out, particularly as a function of temperature and presence of a plasticizing diluent (55.61). The results showed that the anomalous upturn in modulus observed for crystallizable polymers such as natural rubber is largely if not entirely due to strain-induced crystallization. Ultimate Properties. The ultimate properties of interest are the tensile strength, maximum extensibility, and toughess (energy to rupture), and all are affected by straininduced crystallization (SSJ. The higher the temperature, the lower the extent of crystallization and, correspondingly, the lower the ultimate properties. The effects of increase in swelling parallel those for increase in temperature, since diluent also suppresses network crystallization. For non-crystallizable networks, however, neither change is very important, as is illustrated by the results reported for PDMS networks (62).

Thermoset Elastomers

217

In the case of such non-crystallizable, unfilled elastomers, the mechanism for network rupture has been elucidated to a great extent by studies of model networks similar to those already described. For example, values of the modulus of bimodal networks formed by end-linking mixtures of very short and relatively long chains as illustrated in Figure 4, were used to test the "weakest-link" theory (6) in which rupture

Figure 4. Sketch of a network having a bimodal distribution of network chain lengths. The very short and relatively long chains are arbitrarily shown by the thick and thin lines, respectively. was thought to be initiated by the shortest chains (because of their very limited extensibility). It was observed that increasing the number of very short chains did not significantly decrease the ultimate properties. The reason (55) is the very non-affine nature of the deformation at such high elongations. The network simply reapportions the increasing strain among the polymer chains until no further reapportioning is possible. It is generally only at this point that chain scission begins, leading to rupture of the elastomer. The weakest-link theory implicitly assumes an affine deformation, which leads to the prediction that the elongation at which the modulus increases should be independent of the number of short chains in the network. This assumption is contradicted by relevant experimental results, which show very different behavior (55): the smaller the number of short chains, the easier the reapportioning and the higher the elongation required to bring about the upturn in modulus. Multimodal Chain-Lenath Distributions As already mentioned, there turns out to be an exciting bonus if one forms a multimodal distribution of network chain lengths by end linking a very large number of short chains into a long-chain network. The ultimate properties are then actually improved! Bimodal networks prepared by these end-linking techniques have very good ultimate properties, and there is currently much interest in preparing and characterizing such networks

218

J.E. Mark

(4.35.37.44.63-71). and developing theoretical interpretations for their properties (7277). The types of improvements obtained are shown schematically in Figure 5. The

Unimodal, all short chains

~7i t^

BImodal

1

Unimodal, all long chains

1

]L^—^

a Figure 5. Typical plots of nominal stress against elongation for unimodal and bimodal networks obtained by end linking relatively long chains and very short chains. The area under each curve represents the rupture energy (a measure of the "toughness" of the elastomer). results are represented in such a way that the area under a stress-strain isotherm corresponds to the energy required to rupture the network. If the network Is all short chains it is brittle, which means that the maximum extensibility is very small. If the network is all long chains, the ultimate strength is very low. In neither case Is the material a tough elastomer because the areas under the curves are relatively small. As can readily be seen from the figure, the bimodal networks are much improved elastomers in that they can have a high ultimate strength without the usual decrease in maximum extensibility. A series of experiments were carried out In an attempt to determine if this reinforcing effect in bimodal PDMS networks could possibly be due to some intermolecular effect such as strain-induced crystallization. In the first such experiment, temperature was found to have little effect on the isotherms (5.47V This strongly argues against the presence of any crystallization or other type of intermolecular ordering. So also do the results of stress-temperature and birefringence-temperature measurements (47). In a final experiment, the short chains were pre-reacted in a two-step preparative technique so as possibly to segregate them in the network structure (45.61) as might occur in a network cross linked by an incompletely soluble peroxide. This had very little effect on elastomeric properties, again arguing against any type of intermolecular organization as the origin for the reinforcing effects. Apparently, the observed increases in modulus are due to the limited chain extensibility of the short chains, with the long chains serving to retard the rupture process. This can be thought of in terms of what executives like to call a "delegation of responsibilities".

Thermoset Elastomers

219

There is an another advantage to such bimodality when the network can undergo strain-induced crystallization, the occurrence of which can provide an additional toughening effect (78). Decrease in temperature was found to increase the extent to which the values of the ultimate strength of at least some bimodal networks exceed those of the corresponding unimodal ones. This suggests that bimodality facilitates strain-induced crystallization. In practical terms, the above results demonstrate that short chains of limited extensibility may be bonded into a long-chain network to improve its toughness. It is also possible to achieve the converse effect. Thus, bonding a small number of relatively long elastomeric chains into a relatively hard short-chain PDMS thermoset greatly improves its impact resistance (79). Since dangling chains represent imperfections in a network structure, one would expect their presence to have a detrimental effect on the ultimate properties (f/A*)r and ttr, of an elastomer. This expectation is confirmed by an extensive series of results obtained on PDMS networks which had been tetrafunctionally cross linked using a variety of techniques (gO). The largest values of the ultimate strength (f/A*)r were obtained for the networks prepared by selectively joining functional groups occurring either as chain ends or as side groups along the chains. This is to be expected, because of the relatively low incidence of dangling ends in such networks. Also as expected, the lowest values of the ultimate properties generally occurred for networks cured by radiation (UV light, high-energy electrons, and y radiation) (8Q). The peroxidecured networks were generally intermediate to these two extremes, with the ultimate properties presumably depending on whether or not the free-radicals generated by the peroxide are sufficiently reactive to cause some chain scission. Similar results were obtained for the maximum extensibility ar (§Q). These results were supported by more definitive results obtained by investigation of a series of model networks prepared by end-linking vinyl-terminated PDMS chains (§0). Other Tvpes of Deformation Biaxial Extension. There are numerous other deformations of interest, including compression, biaxial extension, shear, and torsion (12). Some of these deformations are considerably more difficult to study experimentally than simple elongation and, unfortunately, have therefore not been as extensively investigated. Measurements in biaxial extension are of particular importance since they are important in packaging applications. This deformation can be imposed by the direct stretching of a sample sheet in two perpendicular directions within its plane, by two independently-variable amounts. In the equi-biaxial case, the deformation is equivalent to compression. Such experimental results (81) have been successfully interpreted in terms of molecular theories 14.6). Biaxial extension studies can also be carried out by the inflation of sheets of the elastomer (2.). Upturns in the modulus (§2) were seen to occur at high biaxial extensions, as expected.

220

J.E. Mark

Shear. Experimental results on natural rubber networks In shear (83) are not well accounted for by the simple molecular theory of rubberlike elasticity. The constrainedjunction theory, however, was found to give excellent agreement with experiment (6). The upturns in modulus in shear (M) were found to be very similar to those obtained in elongation. Torsion. Very little work has been done on elastomers in torsion. There are, however, some results on stress-strain behavior and network thermoelasticity (2.85). More results are presumably forthcoming, particularly on the unusual bimodal networks and on networks containing some of the unusual in-situ generated fillers described below. Swelling. Most studies of networks in swelling equilibrium give values for the cross-link density or related quantities that are in satisfactory agreement with those obtained from mechanical property measurements (1.2). A more Interesting area involving some swollen networks or "gels" is their abrupt collapse (decrease in volume) upon relatively minor changes in temperature, pH, solvent composition, etc. (4.6.86.87). Although the collapse is quite slow In large, monolithic pieces of gel, it is rapid enough in fibers andfilmsto make the phenomenon interesting with regard to the construction of switches and related devices. Gels are also formed, of course, when elastomers are used to absorb liquids, for example in diapers and in attempts to control oil spills over bodies of water. Filler-Reinforced Elastomers and Elastomer-Modified Ceramics One class of multi-phase elastomers are those capable of undergoing strain-induced crystallization, as was mentioned above. In this case, the second phase is made up of the crystallites thus generated, which provide considerable reinforcement. Such reinforcement is only temporary, however, in that it may disappear upon removal of the strain, addition of a plasticizer, or increase in temperature. For this reason, many elastomers (particularly those which cannot undergo strain-induced crystallization) are generally compounded with a permanent reinforcing filler (4.6.8.88-98). The two most Important examples are the addition of carbon black to natural rubber and to some synthetic elastomers (90.92.99). and the addition of silica to siloxane rubbers (£1). In fact, the reinforcement of natural rubber and related materials is one of the most important processes in elastomer technology. It leads to increases in modulus at a given strain, and improvements of various technologically-important properties such as tear and abrasion resistance, resilience, extensibility, and tensile strength (90.92.94.97.100103). There are also disadvantages, however, including increases in hysteresis (and thus of heat build-up), and compression set (permanent deformation). There Is an incredible amount of relevant experimental data available, with most of these data relating to reinforcement of natural rubber by carbon black (92.94.100). Recently, however, other polymers such as PDMS, and other fillers, such as precipitated silica, metallic particles, and even glassy polymers, have become of interest (18.104-139). The most important unsolved problem in this area is the nature of the bonding between the filler particles and the polymer chains (103). The network chains may

Thermoset Elastomers

221

adsorb strongly onto the particle surfaces, which would increase the effective degree of cross linking. This effect will be especially strong If particles contain some reactive surface groups which may cross link (or end link) the polymer chains. Chemisorption, with permanent chemical bonding between filler particles and polymer chains, can be dominant, particularly if the filler is precipitated into the elastomer in-situ during curing (18.107.108.126). Another type of adsorption which can occur at a filler surface is physisorption, arising from long-range van der Waals forces between the surface and the polymer. Contrary to chemisorption, this physical adsorption does not severely restrict the movement of polymer chains relative to the filler surface when high stresses are applied. The available experimental data suggest that both chemisorption and physisorption contribute to reinforcement phenomena, and that the optimal degree of chemical bonding is quite low (of the order 0.2 bonding sites per nm^ of filler surface) (97). Excessive covalent bonding, leading to immobilization of the polymer at the filler surface, is highly undesirable. A filler particle may thus be considered a cross link of very high functionality, but transient in that it can participate in molecular rearrangements under strain. There are probably numerous other ways in which a filler changes the mechanical properties of an elastomer, some of admittedly minor consequence (8.103). For example, another factor involves changes in the distribution of end-to-end vectors of the chains due to the volume taken up by the filler (102.103.140). This effect is obviously closely related to the adsorption of polymer chains onto filler surfaces, but the surface also effectively segregates the molecules in its vicinity and reduces entanglements. Another important aspect of filler reinforcement arises from the fact that the particles influence not only an elastomer's static properties (such as the distribution of its end-to-end vectors), but also its dynamic properties (such as network chain mobility). More specifically, the presence of fillers reduces the segmental mobility of the adsorbed polymer chains to the extent that layers of elastomer close to the filler particles are frequently referred to as "bound rubber" (141-144). As is obvious from the above comments, the mechanism of the reinforcement is only poorly understood. Some elucidation might be obtained by precipitating reinforcing fillers into network structures rather than blending badly agglomerated fillers into the polymers prior to their cross linking. This has, in fact, been done for a variety of fillers, for example silica by hydrolysis of organosilicates, titania from titanates, alumina from alumlnates, etc. (4.6.106.108.138.139). A typical, and important, reaction is the acid- or base-catalyzed hydrolysis of tetraethylorthosilicate: Si(OC2H5)4 + 2H2O

> Si02 + 4C2H5OH

(3)

Reactions of this type are much used by the ceramists in the new sol-gel chemical route to high-performance ceramics (145-152). In the ceramics area, the advantages are the possibility of using low temperatures, the purity of the products, the control of ultrastructure (at the nanometer level), and the relative ease of forming ceramic alloys. In the elastomer reinforcement area, the advantages include the avoidance of the difficult, time-consuming, and energy-intensive process of blending agglomerated filler into high molecular weight and high-viscosity polymers, and the ease of obtaining extremely good dispersions.

222

J.E. Mark

In the simplest approach to obtaining elastomer reinforcement, some of the organosilicate material is absorbed into the cross-linked network, and the swollen sample placed into water containing the catalyst, typically a volatile base such as ammonia or ethylamine. Hydrolysis to form the desired silica-like particles proceeds rapidly at room temperature to yield the order of 50 wt % filler in less than an hour Impressive levels of reinforcement can be obtained by this in-situ technique (6.19.20). The modulus [f] generally increases substantially, and some stress-strain isotherms show the upturns at high elongation that are the signature of good reinforcement. As generally occurs in filled elastomers, there can be considerable irreversibility in the isotherms, which is thought to be due to irrecoverable sliding of the chains over the surfaces of thefillerparticles. If the hydrolyses in organosilicate-polymer systems are carried out with increased amounts of the silicate, bicontinuous phases can be obtained (with the silica and polymer phases interpenetrating one another) (18). At still-higher concentrations of the silicate, the silica generated becomes the continuous phase, with the polymer dispersed in it (6.153-167). The result is a polymer-modified ceramic, variously called an "ORMOCER" (153-155V "CERAMER" (156-158V or "POLYCERAM" (162-164V It is obviously of considerable importance to determine how the polymeric phase, often elastomerlc, improves the mechanical properties of the ceramic in which it is dispersed. Current Problems and Future Trends There is a real need for more high-performance elastomers, which are materials that remain elastomeric to very low temperatures and are relatively stable at very high temperatures. Some phosphazene polymers, [-PRR'N-] (168-170V are in this category. These polymers have rather low glass transition temperatures in spite of the fact that the skeletal bonds of the chains are thought to have some double-bond character. There are thus a number of interesting problems related to the elastomeric behavior of these unusual semi-inorganic polymers. There is also increasing interest in the study of elastomers that also exhibit mesomorphic behavior (g). A particularly challenging problem is the development of a more quantitative molecular understanding (171-174) of the effects offillerparticles, in particular carbon black in natural rubber and silica in siloxane polymers (5.90.92.175.176). Such fillers provide tremendous reinforcement in elastomers in general, and how they do this is still only poorly comprehended. A related but even more complex problem involves much the same components, namely one that is organic and one that is inorganic. When one or both components are generated in-situ, however, there is an almost unlimited variety of structures and morphologies that can be generated (S). How physical properties such as elastomeric behavior depend on these variables is obviously a challenging but very important problem. An example of an important future trend is the study of single polymer chains, particularly with regard to their stress-strain isotherms (177-187). Although such studies are obviously not relevant to the many unresolved issues that involve the interactions of chains within an elastomeric network, they are certainly of interest in their own right.

Thermoset Elastomers

223

Literature Cited (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24)

Flory, P. J. "Principles of Polymer Chemistry"; Cornell University Press: Ithaca NY, 1953. Treloar, L. R. G." The Physics of Rubber Elasticity"; 3rd ed.; Clarendon Press: Oxford, 1975. Mark, J. E. J. Chem. Educ. 1981, 58, 898. Mark, J. E.; Erman, B. "Rubberlike Elasticity. A Molecular Primer"; WileyInterscience: New York, 1988. Mark, J. E. In "Physical Properties of Polymers"; 2nd ed.; J. E. Mark, A. Eisenberg, W. W. Graessley, L. Mandelkern, E. T. Samulski, J. L. Koenig and G. D. Wignall, Ed.; American Chemical Society: Washington, DC, 1993; p 3. Erman, B.; Mark, J. E. "Structures and Properties of Rubberlike Networks"; Oxford University Press: New York, 1997. Mark, J. E.; Erman, B. In "Polymer Networks"; R. F. T. Stepto, Ed.; Blackle Academic, Chapman & Hall: Glasgow, 1998. Mark, J. E.; Erman, B. In "Performance of Plastics"; W. Brostow, Ed.; Hanser: Cincinnati, 1999. Mason, P. "Cauchu. The Weeping Wood"; Australian Broadcasting Commission: Sydney, 1979. Morawetz, H. "Polymers: The Origins and Growth of a Science"; WileyInterscience: New York, 1985. "Rubber Technology"; Third ed.; Morton, M., Ed.; Van Nostrand Reinhold: New York, 1987. Erman, B.; Mark, J. E. Ann. Rev. Phys. Chem. 1989, 40, 351. "Characterization of Highly Cross-Linked Polymers"; Labana, S. S.; Dickie, R. A., Ed.; American Chemical Society: Washington, DC, 1984. "Elastomers and Rubber Compounding Materials"; Franta, I., Ed.; Elsevier: Amsterdam, 1989. "Engineering with Rubber. How to Design Rubber Components"; Gent, A. N., Ed.; Hanser Publishers: New York, 1992. Noda, I.; Dowrey, A. E.; Marcott, C. In "Fourier Transform Infrared Characterization of Polymers"; H. Ishida, Ed.; Plenum Press: New York, 1987. Wignall, G. D. In "Physical Properties of Polymers Handbook"; J. E. Mark, Ed.; American Institute of Physics Press: Woodbury, NY, 1996; p 299. Schaefer, D. W.; Mark, J. E.; McCarthy. D. W.; Jian, L.; Sun, C.-C; Farago, B. In "Polymer-Based Molecular Composites"; D. W. Schaefer and J. E. Mark, Ed.; Materials Research Society: Pittsburgh, 1990; Vol. 171; p 57. McCarthy, D. W.; Mark, J. E.; Schaefer, D. W. J. Polym. Sci.. Polvm. Phys. Ed. 1998,36, 1167. McCarthy, D. W.; Mark, J. E.; Clarson, S. J.; Schaefer, D. W. J. Polvm. Sci.. Polvm. Phys. Ed. 1998.36.1191 Breiner, J. M.; Mark, J. E. Polymer 1998,39. 5483. Mark, J. E. Rubber Chem. Technol 1975, 48, 495. Ronca, G.; Allegra, G. J. Chem. Phvs. 1975, £3, 4990. Florv. P. J. Proc. R. Soc. London. A 1976. 351. 351.

224

J.E. Mark

(25) (26) (27)

Flory, P. J. Polymer 1979,2Q, 1317. Flory, P. J.; Erman, B. Macromolecules 1982, 15, 800. Gottlieb, M.; Macosko, C. W.; Benjamin, G. S.; Meyers, K. O.; Merrill, E. W. Macromolecules 1981, 14-1039. Mark, J. E. Adv. Polvm. Sci. 1982, M , 1 • "Elastomers and Rubber Elasticity"; Mark, J. E.; Lai, J., Ed.; American Chemical Society: Washington, 1982; Vol. 193. Queslel, J. P.; Mark, J. E. Adv. Polym. Sci. 1984, 65, 135. Mark. J. E. Ace. Chem. Res. 1985. 18. 202. Mark, J. E. Polym. J. 1985, 12, 265. Mark, J. E. Brit. Polym. J. 1985, 17, 144. Miller, D. R.; Macosko, C. W. J. Polvm. Sci.. Polvm. Phvs. Ed. 1987, 25, 2441. Lanyo, L. C ; Keiley, F. N. Rubber Chem. Techno!. 1987, gQ, 78. Mark, J. E. In "Frontiers of Macromolecular Science"; T. Saegusa, T. Higashimura and A. Abe, Ed.; Blackwell Scientific Publishers: Oxford, 1989; p 289. Smith, T. L.; Haidar, B.; Hedrick, J. L. Rubber Chem. Technol. 1990, g3, 256. Mark. J. E. J. Inorg. Organomet. Polvm. 1991. 1. 431. Mark. J. E. Anaew. Makromol. Chemie 1992. 202/203. 1. Mark, J. E.; Eisenberg, A.; Graessiey, W. W.; Mandelkern, L.; Samulski, E. T.; Koenig, J. L.; Wignall, G. D. "Physical Properties of Polymers"; 2nd ed.; American Chemical Society: Washington, DC, 1993. Sharaf, M. A.; Mark, J. E.; Hosani, Z. Y. A. Eur. Polym. J. 1993, 22, 809. Sharaf, M. A.; Mark, J. E. Makromol. Chemie 1994, 25,13. Mark, J. E. J. Inoro. Organomet. Polvm. 1994, 4, 31. Mark, J. E. Ace. Chem. Res. 1994,22, 271. Mark, J. E.; Andrady, A. L. Rubber Chem. Technol. 1981,54, 366. Llorente, M. A.; Andrady, A. L.; Mark, J. E. Coll. Polym. Sci. 1981, 2 ^ , 1056. Zhang, Z.-M.; Mark, J. E. J. Polvm. Sci.. Polym. Phvs. Ed. 1982, 20, 473. Mark, J. E. In "Elastomers and Rubber Elasticity"; J. E. Mark and J. Lai, Ed ; American Chemical Society: Washington, DC, 1982. Mullins, L. J. ADD!. Polvm. Sci. 1959, 2, 257. Mark, J. E.; Kato, M.; Ko, J. H. J. Polvm. Sci.. Part C 1976, 54, 217. Smith, K. J., Jr.; Greene, A.; Ciferri, A. Kolloid-Z. Z. Polym. 1964, 1M, 49. Morris. M. C. J. ADDI. Polvm. Sci. 1964. 8. 545. Treloar, L. R. G. Ren. Proa. Phvs. 1973. 35, 755. Chan, B. L.; Elliott, D. J.; Holley, M.; Smith, J. F. J. Polym. Sci.. Part C 1974, 48, 61. Andrady, A. L.; Llorente, M. A.; Mark, J. E. J. Chem. Phys. 1980,22, 2282. Doherty, W. O. S.; Lee, K. L.; Treloar, L. R. G. Br. Polvm. J. 1980, 15, 19. Furukawa, J.; Onouchi, Y.; Inagaki, S.; Okamoto, H. Polym. Bulletin 1981, £, 381. Su, T.-K.; Mark, J. E. Macromolecules 1977, m, 120. Chiu, D. S.; Su, T.-K.; Mari<, J. E. Macromolecules 1977, Ifl, 1110. Mari<, J. E. Polvm. Ena. Sci. 1979, IS, 409.

(28) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40) (41) (42) (43) (44) (45) (46) (47) (48) (49) (50) (51) (52) (53) (54) (55) (56) (57) (58) (59) (60)

Thermoset Elastomers (61)

(62) (63) (64) (65) (66) (67) (68) (69) (70) (71) (72) (73) (74) (75) (76) (77) (78) (79) (80) (81) (82) (83) (84) (85) (86) (87) (88) (89) (90) (91) (92) (93) (94) (95) (96) (97)

225

Mark, J. E.; Eisenberg, A.; Graessley, W. W.; Mandelkem, L.; Koenig, J. L. "Physical Properties of Polymers"; 1 st ed.; American Chemical Society: Washington, DC, 1984. Chlu, D. S.; Mark, J. E. Coll. Polvm. Sci. 1977, 225. 644. Mark, J. E. Makromol. Chemie. Suppl. 1979, 2, 87. Sllva, L. K.; Mark, J. E.; Boerio, F. J. Makromol. Chemie 1991,122, 499. Hanyu, A.; Stein, R. S. Macromol. Symp. 1991, 45, 189. Roland, C. M.; Buckley, G. S. Rubber Chem. Technol. 1991, 64, 74. Oikawa, H. Polvmer 1992, 33, 1116. Hamurcu, E. E.; Baysal, B. M. Polymer 1993, 34, 5163. Subramanian, P. R.; Galiatsatos, V. Macromol. Svmo. 1993, 76, 233. Sharaf, M. A.; Mark, J. E.; Al-Ghazal, A. A.-R. J. Apol. Polvm. Sci. Svmo. 1994, 55, 139. Besbes, S.; Bokobza, L.; Monnerie, L.; Bahar, I.; Erman, B. Macromolecules 1995,28,231. Erman, B.; Mark, J. E. J. Chem. Phys. 1988, 29, 3314. Termonia, Y. Macromolecules 1990, 23,1481. KloczkowskI, A.; Mark, J. E.; Erman, B. Macromolecules 1991, 24, 3266. Sakrak, G.; Bahar, I.; Erman, B. Macromol. Theory Simul. 1994, 3, 151. Bahar, I.; Erman, B.; Bokobza, L; Monnerie, L. Macromolecules 1995, 28, 225. Erman, B.; Mark, J. E. Macromolecules 1998, 31, 3099. Sun, C.-C; Mark, J. E. J. Polvm. Sci.. Polym. Phys. Ed. 1987, 25, 2073. Tang, M.-Y.; Letton, A.; Mark, J. E. Coll. Polym. Sci. 1984, 262, 990. Andrady, A. L.; Llorente, M. A.; Sharaf, M. A.; Rahalkar, R. R.; Mark, J. E.; Sullivan, J. L.; Yu, C. U.; Falender, J. R. J. Appl. Polym. Sci. 1981, 26, 1829. Obata, Y.; Kawabata, S.; Kawai, H. J. Polym. Sci.. Part A-2 1970, 8, 903. Xu, P.; Mark, J. E. J. Polym. Sci.. Polym. Phvs. Ed. 1991, 29, 355. Rivlin, R. S.; Saunders, D. W. Philos. Trans. R. Soc. London. A 1951, 243, 251. Wang, S.; Mark, J. E. J. Polvm. Sci.. Polym. Phvs. Ed. 1992, 30, 801. Wen. J.; Mark, J. E. Polym. J. 1994, 26,151. Tanaka, T. Phvs. Rev. Lett. 1978, 4Q, 820. Tanaka, T. Sci. Am. 1981, 244, 124. Oberth, A. E. Rubber Chem. Technol. 1967, 40, 1337. Boonstra, B. B. In "Rubber Technology"; M. Morton, Ed.; Van Nostrand Reinhold: New York, 1973; p 51. Boonstra, B. B. Polvmer 1979, 20, 691. Warrick, E. L.; Pierce, O. R.; Polmanteer, K. E.; Saam, J. C. Rubber Chem. Technol. 1979. 52.437 RIgbi, Z. Adv. Polym. Sci. 1980, 36, 21. Queslel, J. P.; Mark, J. E. In "Encyclopedia of Polymer Science and Engineering, Second Edition"; R. A. Meyers, Ed.; Wiley-lnterscience: New York, 1986; p 365. Donnet, J.-B.; Vidal, A. Adv. Polym. Sci 1986, ZS, 103. Ahmed, S.; Jones, F. R. J. Mater. Sci. 1990, 25, 4933. Enikolopyan, N. S.; Fridman, M. L.; Stalnova, I. O.; Popov, V. L. Adv. Polvm. Sci. 1990,96,1. Edwards, D. C. J. Mater. Sci. 1990, 25, 4175.

226

J.E. Mark

(98)

Medalia. A. I.; Kraus, G. In "Science and Technology of Rubber"; 2nd ed.; J. E. Mark, B. Erman and F. R. Eirich, Ed.; Academic: New York, 1994; p 387. Karasek, L.; Sumita, M. J. Mats. Sci. 1996, 31, 281. Kraus, G. Adv. Polym. Sci. 1971, 8, 155. "Reinforcement of Elastomers"; Kraus, G., Ed.; Interscience: New York, 1965. Kloczkowski, A.; Sharaf, M. A.; Mark. J. E. Comput. Polvm. Sci. 1993, 2. 39. Kloczkowski, A.; Sharaf, M. A.; Mark, J. E. Chem. Eno. Sci. 1994, 4S, 2889. Matijevic, E.; Scheiner, P. J. Coll. Interfacial Sci. 1978, 62, 509. Mark. J. E. Kautschuk + Gummi Kunstoffe 1989. 42. 191. Mark. J. E. CHEMTECH 1989. 19. 230. "Polymer-Based Molecular Composites"; Schaefer, D. W.; Mark, J. E., Ed.; Materials Research Society: Pittsburgh, 1990; Vol. 171. Mark, J. E.; Schaefer, D. W. In "Polymer-Based Molecular Composites"; D. W. Schaefer and J. E. Mark, Ed.; Materials Research Society: Pittsburgh, 1990; Vol. 171; p 51. Yasrebi, M.; Kim, G. H.; Gunnison, K. E.; Milius, D. L.; Sarikaya, M.; Aksay, I. A. In "Better Ceramics Through Chemistry IV"; B. J. J. Zelinski, C. J. Brinker, D. E. Clark and D. R. Ulrich, Ed.; Materials Research Society: Pittsburgh, 1990; Vol. 180; p 625. Chung, Y. J.; Ting, S.-J.; Mackenzie, J. D. In "Better Ceramics Through Chemistry IV"; B. J. J. Zelinski, C. J. Brinker, D. E. Clark and D. R. Ulrich, Ed.; Materials Research Society: Pittsburgh, 1990; Vol. 180; p 981. Saegusa, T.; Chujo, Y. J. Macromol. Sci. - Chem. 1990, A2Z, 1603. Mauritz, K. A.; Jones. C. K. J. Appl. Polym. Sci. 1990, 4Q, 1401. Mauritz, K. A.; Scheetz, R. W.; Pope, R. K.; Stefanithis, I. D.; Wilkes, G. L.; Huang, H.-H. Preprints. Div. Polvm. Chem.. Inc.. Am. Chem. Soc. 1991, 32(3), 528. Bianconi, P. A.; Lin, J.; Strzelecki, A. R. Nature 1991, 249, 315. Okada, A.; Fukumori, K.; Usuki, A.; Kojima, Y.; Sato, N.; Kurauchi, T.; Kamigaito, O. Preprints. Div. Polym. Chem.. Inc.. Am. Chem. Soc. 1991, 32(3), 540. Yano, K.; Usuki, A.; Okada, A.; Kurauchi, T.; Kamigaito, O. Preprints. Div. Polym. Chem.. Inc.. Am. Chem. Soc. 1991.32(1). 65. Calvert, P. In "U.S.-Japan Workshop on Smart/Intelligent Materials and Systems"; I. Ahmad, A. Crowson, C. A. Rogers and M. Aizawa, Ed.; Technomic Pub. Co.: Lancaster, 1991; p 162. Calvert, P. In "Ultrastructure Processing of Advanced Materials"; D. R. Uhlmann and D. R. Ulrich, Ed.; Wiley: New York, 1992; p 149. Mackenzie, J. D.; Chung, Y. J.; Hu, Y. J. Non.-Crvst. Solids 1992, 147&148. 271. Hu. Y.; Mackenzie, J. D. J. Mat. Sci. 1992, 27, 4415. Brennan, A. B.; Rodrigues, D. E.; Wang, B.; Wilkes, G. L. In "Chemical Processing of Advanced Materials"; L. L. Hench and J. K. West, Ed.; Wiley: New York, 1992; p 807. Huang, H.; Glaser, R. H.; Brennan, A. B.; Rodigues, D.; Wilkes, G. L. In "Ultrastructure Processing of Advanced Materials"; D. R. Uhlmann and D. R. Ulrich, Ed.; Wiley & Sons: New York, 1992; p 425.

(99) (100) (101) (102) (103) (104) (105) (106) (107) (108) (109)

(110) (111) (112) (113) (114) (115) (116) (117)

(118) (119) (120) (121) (122)

Thermoset Elastomers

227

(123) Schmidt, H. In "Ultrastructure Processing of Advanced Materials"; D. R. Uhlmann and D. R. Ulrich, Ed.; Wiley: New York, 1992; p 409. (124) Schmidt, H. K. In "Submicron Multiphase Materials"; R. H. Baney, L. R. Gllliom, S.-l. Hirano and H. K. Schmidt, Ed.; Materials Research Society: Pittsburgh, PA, 1992; Vol. 274; p 121. (125) Ellsworth, M. W.; Novak, B. M. In "Submicron Multiphase Materials"; R. H. Baney, L. R. Gilliom, S.-l. Hirano and H. K. Schmidt, Ed.; Materials Research Society: Pittsburgh, 1992; Vol. 274; p 67. (126) Mark, J. E.; Wang, S.; Xu, P.; Wen, J. In "Submicron Multiphase Materials"; R. H. Baney, L. R. Gilliom, S.-l. Hirano and H. K. Schmidt, Ed.; Materials Research Society: Pittsburgh, PA, 1992; Vol. 274; p 77. (127) Sun, L; Akionis, J. J.; Salovey, R. Polym. Eno. Sci. 1993, 33,1308. (128) Matijevic. E. Chem. Mater. 1993. 5. 412. (129) Novak. B. M. Adv. Mats. 1993. 5. 422. (130) Mark, J. E.; Calvert, P. D. J. Mats. Sci.. Part C 1994,1, 159. (131) Mark, J. E. In "Frontiers of Polymers and Advanced Materials"; P. N. Prasad, Ed.; Plenum: New York, 1994; p 403. (132) "Hybrid Organic-Inorganic Composites"; Mark, J. E.; Lee, C. Y.-C; Bianconi, P. A., Ed.; American Chemical Society: Washington, 1995; Vol. 585. (133) Mark, J. E. In "Diversity into the Next Century"; R. J. Martinez, H. Arris, J. A. Emerson and G. Pike, Ed.; SAMPE: Covina, CA, 1995; Vol. 27. (134) Schmidt. H. K. Macromol. Svmp. 1996. 101. 333. (135) Calvert, P. In "Biomimetic Materials Chemistry"; S. Mann, Ed.; VCH Publishers: NewYork, 1996; p 315. (136) Giannelis, E. P. In "Biomimetic Materials Chemistry"; S. Mann, Ed.; VCH Publishers: NewYork, 1996; p 337. (137) Wen, J.; Wilkes, G. L. In "Polymeric Materials Encyclopedia: Synthesis, Properties, and Applications"; J. C. Salamone, Ed.; CRC Press: Boca Raton, 1996. (138) Mark. J. E. Hetero. Chem. Rev. 1996. 3. 307. (139) Mark. J. E. Polvm. Ena. Sci. 1996. 36. 2905. (140) Yuan, Q. W.; Kloczkowski, A.; Mark, J. E.; Sharaf, M. A. J. Polvm. Sci.. Polvm. Phvs.Ed. 1996.34.1674. (141) Litvlnov, V. M.; Spless, H. W. Macromol. Chem. 1992, 1^3, 1181. (142) Meissner, B. J. Appl. Polvm. Sci. 1993, 5Q, 285. (143) Karasek, L.; Meissner, B. J. Appl. Polym. Sci. 1994, 52, 1925. (144) Leblanc. J. J. Appl. Polym. Sci. 1997. 66. 2257. (145) "Ultrastructure Processing of Ceramics, Glasses, and Composites"; Hench, L. L.; Ulrich, D. R., Ed.; Wiley: New York, 1984. (146) "Ultrastructure Processing of Advanced Ceramics"; Mackenzie, J. D.; Ulrich, D. R., Ed.; Wiley: NewYork, 1988. (147) Ulrich. D. R. J. Non-Cryst. Solids 1988. 100. 174. (148) Ulrich, D. R. CHEMTECH 1988, tg, 242. (149) Mackenzie, J. D. J. Non-Crvst. Solids 1988, IQO, 162. (150) Ulrich. D. R. J. Non-Cryst. Solids 1990. 121. 465.

228

J.E. Mark

(151) Brinker, C. J.; Scherer, G. W. "Sol-Gel Science"; Academic Press: New York, 1990. (152) "Better Ceramics Through Chemistry VIII: Hybrid Materials"; Brinker, C. J.; Giannelis, E. P.; Laine, R. M.; Sanchez, C , Ed.; Materials Research Society: Warrendale, PA, 1998; Vol. 519. (153) Schmidt, H. In "Inorganic and Organometallic Polymers"; M. ZeldIn, K. J. Wynne and H. R. Allcock, Ed.; American Chemical Society: Washington, DC, 1988; p 333. (154) Schmidt, H.; Wolter, H. J. Non-Cryst. Solids 1990, 121, 428. (155) Nass, R.; Arpac, E.; Glaubitt, W.; Schmidt, H. J. Non-Cryst. Solids 1990, 121, 370. (156) Wang, B.; Wilkes, G. L J. Polvm. Sci.. Polym. Chem. Ed. 1991, 29, 905. (157) Wilkes, G. L.; Huang, H.-H.; Glaser, R. H. In "Silicon-Based Polymer Science"; J. M. Zeigler and F. W. G. Fearon, Ed.; American Chemical Society: Washington, DC, 1990; Vol. 224; p 207. (158) Brennan, A. B.; Wang, B.; Rodrigues, D. E.; Wilkes, G. L J. Inorg. Organomet. Polym. 1991. 1.167. (159) Sobon, C. A.; Bowen, H. K.; Broad, A.; Calvert. P. D. J. Mat. Sci. Lett. 1987, 6, 901. (160) Calvert, P.; Mann, S. J. Mat. Sci. 1988, 22, 3801. (161) Azoz, A.; Calvert, P. D.; Kadim, M.; McCaffery, A. J.; Seddon, K. R. Nature 1990, 344. 49. (162) Doyle, W. F.; Uhlmann, D. R. In "Ultrastructure Processing of Advanced Ceramics"; J. D. Mackenzie and D. R. Ulrich, Ed.; Wiley-lnterscience: New York, 1988; p 795. (163) Doyle, W. F.; Fabes, B. D.; Root, J. C; Simmons, K. D.; Chiang, Y. M.; Uhlmann, D. R. In "Ultrastructure Processing of Advanced Ceramics"; J. D. Mackenzie and D. R. Ulrich, Ed.; Wiley-lnterscience: New York, 1988; p 953. (164) Boulton, J. M.; Fox, H. H.; Neilson, G. F.; Uhlmann, D. R. In "Better Ceramics Through Chemistry IV"; B. J. J. Zelinski, C. J. Brinker, D. E. Clark and D. R. Ulrich, Ed.; Materials Research Society: Pittsburgh, 1990; Vol. 180; p 773. (165) Mark, J. E.; Sun, C.-C. Polym. Bulletin 1987, IS, 259. (166) Ning, Y. P.; Zhao, M. X.; Mark, J. E. In "Frontiers of Polymer Research"; P. N. Prasad and J. K. Nigam, Ed.; Plenum: New York, 1991; p 479. (167) Zhao, M. X.; Ning, Y. P.; Mark, J. E. In "Advanced Composite Materials"; M. D. Sacks, Ed.; American Ceramics Society: Westen/ille, OH, 1993; p 891. (168) Mark, J. E.; Yu, C. U. J. Polym. Sci.. Polym. Phys. Ed. 1977,15, 371. (169) Andrady, A. L.; Mark, J. E. Eur. Polym. J. 1981,17, 323. (170) Mark, J. E.; Allcock, H. R.; West, R. "Inorganic Polymers"; Prentice Hall: Englewood Cliffs, NJ, 1992. (171) Heinrich, G.; Vilgis, T. A. Macromolecules 1993, 26, 1109. (172) Witten, T. A.; Rubinstein, M.; Colby, R. H. J. Phys. II France 1993. 3, 367. (173) Kluppel. M.; Heinrich, G. Rubber Chem. Technol. 1995, fig, 623. (174) Kluppel, M.; Schuster, R. H.; Heinrich, G. Rubber Chem. Technol. 1997, 70, 243. (175) Polmanteer, K. E.; Lentz, C. W. Rubber Chem. Technol. 1975,4§, 795. (176) Kraus, G. Rubber Chem. Technol. 1978,51, 297.

Thermoset Elastomers (177) (178) (179) (180) (181) (182) (183) (184) (185) (186) (187)

229

Chu,S. Science 1991, 253, 861. Perkins, T. T.; Smith, D. E.; Chu, S. Science 1994, 264, 819. Perkins, T. T.; Quake, S. R.; Smith, D. E.; Chu, S. Science 1994, 264, 822. Perkins, T. T.; Smith, D. E.; Larson, R. G.; Chu, S. Science 1995, 2g8, 83. Smith, S. B.; Cui, Y.; Bustamante, C. Science 1996, 2Z1, 795. Kellermayer, M. S. Z.; Smith, S. B.; Granzier, H. L.; Bustamante, C. Science 1997, 276,1112. Rief, M.; Oesterhelt, F.; Heymann, B.; Gaub, H. E., Science 1997, 275,1295. Oberhauser, A. F.; Marszaiek, P. E.; Erickson, H. P.; Fernandez, J. M. Nature 1998,393,181. Li. H.; Rief, M.; Oesterhelt, F.; Gaub, H. E., Adv. Mater. 1998, 3, 316. Marszaiek, P. E.; Oberhauser, A. F.; Pang, Y-P.; Fernandez, J. M. Nature 1998, 396,661. Ortiz, C; Hadziioannou, G. Macromolecules 1999, 32, 780.