Characteristics of antiplasticized thermosets: effects of network architecture and additive chemistry on mechanical fortification.
Thermosets are commonly used as matrix materials for filled composites. In such applications, filler material is introduced to impart enhanced mechanical properties. It has been well established, however, that the compressive strength of such composites is dominated by the mechanical properties of the matrix material. For fiber-reinforced composites, failure occurs through a microbuckling mechanism dictated by the shear modulus of the matrix (1, 2). In particulate-filled composites, compressive failure is shear-dominated, suggesting that the yield strength of the matrix is of critical importance (3). In both cases, the interfacial adhesion between matrix and filler is also important. In order to enhance the compressive or flexural performance of filled composites, either or both of these limitations must be addressed. The current work focuses on increasing the yield strength and modulus of the matrix through the use of molecular additives.
At low concentrations, some plasticizers have shown to increase, rather than decrease, the stiffness and strength of polymers. Such additives are termed antiplasticizers (4, 5). Well known in thermoplastic polymers such as polycarbonate (6-9), polystyrene (10), and poly(vinyl chloride) (11-13), antiplasticization has also been demonstrated in thermoset epoxy resins with a variety of additives (14-16). The antiplasticization effect, so called because the mechanical response is opposite that of a plasticized material while still exhibiting a depressed glass transition temperature, has been related to reduced mobility on the molecular scale. Using dynamic mechanical methods, this reduction in mobility has been attributed to an amplitude decrease in the sub-[T.sub.g] [beta]-relaxation peak (11, 17, 18). A decrease in sub-[T.sub.g] relaxations correlates with increased stiffness of the backbone and is manifested by enhanced mechanical properties. This decrease is also usually accompanied by an increase in the density of the material, which is thought to be the actual cause of the reduction in molecular mobility (19-22). The present study is a systematic investigation into the molecular parameters that affect antiplasticization.
Recently, we reported a new class of antiplasticizer for epoxy thermosets that demonstrate enhanced strength and modulus of networks while significantly reducing the viscosity of the resin prior to cure (23). These materials were also shown to exhibit improved flame retardance through an increase in the char yield and a decrease in the rate of thermal degradation, which was attributed to the phosphorus present in the additives. Organophosphorus additives are commonly used to impart flame retardance to polymers and have been examined as replacements for hazardous chlorinated additives (24, 25).
In the current work, a family of organophosphorus additives is examined in order to study the effects of fortifier chemistry and network architecture on antiplasticization. The family of trialkyl phosphates was chosen in order to investigate the effects of molecular weight, size, and flexibility of the additive on antiplasticization by changing the length of the alkyl chain. Likewise, by changing the nature of the curing agents used in hardening the resin, the role of [M.sub.c] and network free volume can be probed. Both model and commercial curing agents were investigated in this study.
The antiplasticized networks investigated in this work are based on an epoxy (Epon 825 from Shell) crosslinked with a stoichiometric amount of curing agent. A variety of curing agents were used to probe the effects of network architecture and crosslink chemistry on antiplasticization (Table 1). Two diamine curing agents (EDA, PDA) were used to formulate model networks of well-characterized, small [M.sub.c]. This is due to the low molecular weight of these curing agents. The commercially cured systems (D230, D400), by contrast, result in loosely crosslinked networks. In addition to the four amine curing agents, an anhydride curing agent, hexahydrophthalic anhydride (HHPA), was investigated.
In addition to altering the polymer matrix through the use of various curing agents, the role of the antiplasticizer was investigated by changing the structure of the additives themselves (Table 2) and are of the general structure of trialkyl phosphates, P(O)[(OR).sub.3]. In addition to the alkyl phosphates, triphenyl phosphate (TPhP), a triaryl phosphate, was studied. All the additives retain similar chemical properties while differing in molecular weight, density, and flexibility, allowing for careful comparison of these parameters on antiplasticization. All the phosphates were purchased from Aldrich and used as received.
The fortified polymers were made by blending a given antiplasticizer into the epoxy resin (Epon 825) at 50[degrees]C and mixing with a stoichiometric amount of hardener. The epoxy prepolymer is a purified diglycidyl ether of bisphenol A (DGEBA) of known molecular weight. After degassing under vacuum at 50[degrees]C for 15 min, the mixture was transferred into molds for curing. Plaques were made by pouring into polytetrafluoroethylene molds between glass plates treated with a silating release agent. Surfasil (Pierce Chemical). Compression samples were cured in 8-mm-diameter glass test tubes. The amine samples were cured at 75[degrees]C for 3 hr, followed by an additional 3 hr at 125[degrees]C. The anhydridecured epoxy also included 1 phr (one part per hundred) accelerator, 2,4,6-tris-(dimethylaminomethyl) phenol, and was reacted under a nitrogen blanket to reduce oxidation for 3 hr at 75[degrees]C, followed by 2 hr at 200[degrees]C. Complete cure was verified using differential scanning calorimetry (DSC), where the glass transition temperature did not change following successive heating.
Mechanical measurements of modulus and strength were conducted both in tension and compression. Samples were prepared according to ASTM D638 and ASTM D695 for tension and compression, respectively. All samples were tested on an Instron 1123 machine. Tension and compression samples were loaded at constant crosshead speed of 2 mm/min. Extensometers were used. where appropriate, for measurements of strain. Densities were measured using a buoyancy method at room temperature in degassed, deionized water (ASTM D792). The thermal properties of the cured resins were investigated using DSC and a scanning rate of 10[degrees]C/min.
Results are reported as a function of the molecular percent of additive based on the initial molar amounts of epoxy resin, fortifier, and curing agents used to make the thermoset. Such a comparison allows for additives of different molecular weights to be compared more easily and without ambiguity. It should be noted, however, that the additives never exceeded 30 wt% and that optimal conditions, although varying by additive, were achieved at 10 to 20 wt% additive.
Additive Effects on Antiplasticization
In a previous study (23), we reported that a small molecular additive could be used as a processing aid for epoxy formulations and was capable of enhancing mechanical properties of the resin after cure. The additive, dimethyl methyl phosphonate (DMMP), was shown to increase the modulus and yield strength of D230-cured Epon 825 by as much as 12% when loaded in tension. DMMP was also found to significantly reduce the viscosity of the resin prior to cure, allowing for greater processability and a reduction in defects. The intrinsic viscosity of a 20-wt% DMMP-Epon 825 mixture was reduced by an order of magnitude both at room temperature and at 50[degrees]C. Likewise, the incorporation of the phosphorus moiety into the polymer served to reduce the flammability of the material by decreasing the heat release rate in a manner similar to that previously published for phosphorus flame retardants (24, 25).
In order to further understand the molecular mechanisms for antiplasticization, a family of phosphates was used. These exhibit similar decreases in viscosity and flammability to the DMMP antiplasticizer. However, investigating a number of similar chemical structures of various molecular weights, solubility parameters, and densities allows for a systematic look into the molecular parameters affecting the ability of antiplasticizers to increase mechanical properties.
Diluents, plasticizers and antiplasticizers alike, reduce the glass transition ([T.sub.g]) of a polymer as a function of concentration. The addition of trialkyl phosphate additives into an epoxy thermoset depresses the glass transition linearly with additive concentration (Fig. 1). There is, however, no evidence of phase separation, as only one transition is observed over all concentrations. In the case of the larger additives, the decrease in [T.sub.g] follows with the size of the additive, with the largest additive, TBP, having the most severe effect on the thermal properties of the polymer. A similar phenomenon has been reported in the plasticization of PVC with phthalate diluents (26-28). Early work in the area of polymer plasticization demonstrates the effect of alkyl chain branching in phthalic diesters on plasticization of polymers, where the degree of branching is inversely proportional to the efficiency of the additive, indicating the importance of additive volume (29). The larger the plasticizer, the greater the decrease in [T.sub.g]. A similar effect is evident for the case of TPrP and TBP. By increasing the van der Waals volume, the decrease in [T.sub.g] is more pronounced. The smallest of the fortifiers investigated, TMP, behaves decidedly outside the trend of the remaining diluents, increasing the glass transition temperature at 5 mol% and not significantly affecting the [T.sub.g] of the mixture until higher concentrations.
[FIGURE 1 OMITTED]
An important characteristic of antiplasticized polymers is the apparent increase in density of the polymer upon inclusion of the fortifier (Fig. 2). This effect has been exhibited in most antiplasticized polymer systems (16), including the amine-cured epoxy antiplasticized with DMMP. The increase in density, or decrease in free volume, of the mixture is manifested through a reduction in secondary relaxations, which generally accompany antiplasticization. Mixing of small molecules with glassy polymers can lead to a departure from specific volume additivity, and a number of theoretical models exist to predict this behavior based on the equilibrium liquid-state specific volumes of polymer and diluent. One such model, proposed by Paul and Ruiz-Trevino (22), uses volume additivity of the two components, polymer and additive, in their equilibrium liquid states and incorporates a thermal expansion term that results in the departure from simple additivity (Eq 1).
[FIGURE 2 OMITTED]
[V.sub.mg] (T) = [[omega].sub.a] [V.sub.al] (T) + [[omega].sub.p][V.sub.pl] (T) + ([[d[V.sub.ml]]/[dT]] - [[d[V.sub.mg]]/[dt]]) ([T.sub.gm] - T) (1)
The specific volume of the mixture in its glassy state, [V.sub.mg] (T), at a test temperature (T) below [T.sub.g] in Eq 1 is expressed as a weight fraction rule of mixtures of the liquid specific volume of the additive, [V.sub.al](T), and the amorphous liquid state of the polymer, [V.sub.pl](T). The predicted specific volume of the matrix, [V.sub.mg](T), can then be translated into a density term through Eq 2.
[FIGURE 3 OMITTED]
[[rho].sub.mg] (T) = [1/[[V.sub.mg](T)]] * [[rho].sub.[H.sub.2]O](T) (2)
All the components for this model can be either experimentally obtained or estimated from literature (30). The experimental densities measured for all the additives, with the exception of TMP, follow the predictions of these models closely (Fig. 3). Densification through the addition of the trialkyl phosphates other than TMP results in an increase in density predicted by the RuizTrevino model. The trimethyl phosphate fortifier, already shown to demonstrate unique glass transition behavior, does not fit the model to any degree of success and even exceeds the maximum density predicted by the model. Covalent bonding between the polymer backbone and TMP, leading to increased crosslink density and reduced density, however, can be discounted. After the modified thermoset is subjected to 24 hr under vacuum above the [T.sub.g] of the polymer and the boiling point of the additive, significant weight loss is observed. The loss of additive can then be verified via elemental analysis through a reduction in phosphorus content following such a treatment. This is indication, therefore, that the TMP is increasing the density of the polymer by filling the free volume and is not simply a result of a [T.sub.g] mismatch between the additive and polymer as indicated by the available models.
The strength of the polymer-additive interaction can be explored using the concept of the Hildebrand solubility parameter, [[delta].sub.a]. In addition to varying the molecular volume and density of the additives by changing the nature of the R-group in the trialkyl phosphates, P(O)(OR)[.sub.3], the solubility parameter is also varied. By increasing the alkyl group, [[delta].sub.a] of the molecule is reduced, decreasing the strength of the intermolecular interaction. The solubility parameter for amine-cured epoxy resins is reported (31) between 18.5 and 20.5 MP[a.sup.1/2], depending on the hydrogen-bonding strength of the solvent. This evaluation predicts that TMP would have the greatest interaction polymer. Both TPhP and TPrP have the same [[delta].sub.a] and exhibit very similar thermal properties.
The mechanical properties of epoxies antiplasticized with phosphate diluents show marked increases in the strength and stiffness of such networks. Both the tensile modulus and yield strength of the 825-D230 system were increased upon incorporation of the phosphates (Fig. 4). These improvements follow the general trends already observed for the diluents' effects on thermal and physical properties, wherein increased additive size leads to a decrease in the efficiency of that additive to antiplasticize the matrix. The smallest of the additives, TMP, increased the strength and stiffness a more significant amount over the other additives (34% increase in tensile modulus, 29% increase in tensile yield strength). The larger additives show decreased efficiency with increasing molecular weight such that the TBP, the largest of the alkyl phosphates, resulted in a 6% increase in tensile modulus and 5% increase in tensile yield stress at optimal concentrations. The concentration range over which the additives resulted in improved mechanical properties was also decreased with increasing additive molecular weight. Because antiplasticizers lead to improved strength and modulus over a well-defined concentration range, one comparison for antiplasticizer efficiency is the concentration at which the mechanical properties are returned to those of the unmodified polymer. This concentration, also known as the threshold concentration (27), is a dividing line between polymer plasticization and antiplasticization. Beyond the threshold concentration, the additive behaves as a plasticizer, decreasing mechanical performance. Here the threshold concentration is defined as that required for the material to return to the tensile modulus of the unmodified resin. The threshold concentration is 36 mol% for TMP, and decreases to 18, 15, 13 and 18 mol% for TEP, TPrP, TBP and TPhP, respectively. Although these numbers differ slightly for compressive modulus and yield strength, the general trend remains the same with a significant difference between TMP and the other, larger, phosphates.
[FIGURE 4 OMITTED]
The strength of the polymer-additive interaction, however, correlates strongly with the effective concentration range for each additive (Fig. 5). A solubility parameter of 20 MP[a.sup.1/2] is used for the epoxy network in a strongly hydrogen-bonding solvent. The results in Fig. 5 are plotted against the solubility parameter difference between that of the epoxy, [[delta].sub.e], and additive, [[delta].sub.a]. Given that the antiplasticized materials behave slightly differently when observing the increase in modulus and the increase in [[sigma].sub.y], the concentration ranges obtained from each plot are presented above. Ideally, the slope of each curve would be identical. That the slopes are not identical is related to the thermal properties of the antiplasticized resins. The yield behavior of epoxy thermosets is known to be highly dependent on the thermal properties. or [T.sub.g], of the resin (32, 33). While [[sigma].sub.y] changes with [T.sub.g] at constant backbone stiffness, the modulus does not (34). Because the [T.sub.g] of the resins antiplasticized with low [[sigma].sub.a] additives (TPrP and TBP) is decreased with increasing additive concentration, the [[sigma].sub.y] that is measured is likewise slightly depressed (low [T.sub.g] usually translates to a low [[sigma].sub.y]). As a result, the correlations in Fig. 5 based on the yield strength measurements include the effect of depressed [T.sub.g] and threshold concentrations are slightly reduced. The modulus-based correlation is unaffected by changing [T.sub.g] of the resin.
[FIGURE 5 OMITTED]
The difference in mechanical reinforcement for chemically related additives demonstrates the role of the additive in dictating the mechanical properties of antiplasticized epoxies. The chemical structure of the additive controls the final thermal and physical properties of the modified resin. Large increases in density and small decreases in [T.sub.g] then translate to greater modulus and strength. The strength of the polymer-additive interaction is also evident in the threshold concentration of each additive. It is likely that the effectiveness of TMP in increasing the mechanical properties stems from the large increase in density, a small decrease in [T.sub.g] and almost ideal match in solubility parameter upon inclusion of this additive into the resin.
Effect of Nework Architecture
The most obvious effect of changing the curing agent in an epoxy thermoset is the molecular weight between crosslinks ([M.sub.c]) of the system. The [M.sub.c] can be increased directly by increasing the molecular weight of the curing agent. This increases the spacing between tetrafunctional crosslinks and loosens the network. One can also increase [M.sub.c] by employing a combination of tetrafunctional curing agents, typically a diamine, with difunctional reagents such as aniline or other monoamines (32). The monoamine reagents loosen the network by reacting two diepoxies without forming a chemical crosslink and serving as chain extenders. The [M.sub.c] of an epoxy resin influences the thermal and mechanical properties of the network, and is one of the major factors in controlling epoxy resin properties (34).
In the present study, the crosslink density is altered by using a number of tetrafunctional diamine curing agents to form both commercial and model networks of varying [M.sub.c]. The goal is to investigate whether the free volume of the network, which is proportional to [M.sub.c], affects the efficiency of a given antiplasticizer in a manner similar to the effect of different additives on a given thermoset system. Here, TMP is the lone antiplasticizer investigated, and only the molecular weight of the curing agent is altered. The solubility parameter of epoxy resins cured with chemically similar curing agents is not expected to change significantly (35). Because the antiplasticizer is also unchanged, the strength of the polymer-additive interaction is also unchanged.
By using a variety of different curing agents, the nature of the polymer-additive interaction can be controlled. In the case of the amine-cured systems, such interactions have been shown to be associated with the 2-hydroxylpropylether [OC[H.sub.2]CH(OH)C[H.sub.2]] unit of the polymer backbone. Various NMR and creep studies have shown that the mobility of this unit is suppressed upon introduction of an antiplasticizer (18, 36). In the case of phosphate antiplasticization, hydrogen bonding between the hydroxyl moiety and the phosphate is believed to be responsible for the strong diluent-polymer interactions. It is interesting that antiplasticization occurs in resins that do not express an ability to form strong hydrogen bonds with the antiplasticizer.
As seen in Fig. 6, the mechanical properties of the antiplasticized networks do in fact differ significantly with the [M.sub.c] of the resin. As the [M.sub.c] of the network increases, the efficiency of the antiplasticizer is decreased. The D400-cured network, which has the largest [M.sub.c], is simply plasticized by TMP with no apparent mechanical enhancements at any concentration. The most tightly crosslinked materials, however, produce the greatest increases in both strength and stiffness. The EDA and PDA systems both exhibit increases in modulus and strength values of approximately 50% at TMP concentrations of 25-30 mol%. Note, however, that the EDA-cured and PDA-cured networks fail in a brittle fashion under tension, and the failure stress for these materials, rather than the yield stress, is reported. Nevertheless, the increases in modulus and strength as a function of [M.sub.c] clearly show that for diamine-cured epoxy networks antiplasticized with TMP, a smaller [M.sub.c] is better.
The significant departure from expected behavior for the D400 network prompted the need for an examination of a thermoset with [M.sub.c] between that of the D230-cured system and the D400. In order to achieve such a [M.sub.c], a mixture of equal molar amounts of D230 and D400 was used to cure the epoxy. The resulting network, D230-D400, has a [M.sub.c] of 550 g/mol. When TMP was used to antiplasticize this network, the mechanical response, as seen in Fig. 6, is an average of the D230 and D400 responses, indicating a competition between antiplasticization of the network and plasticization of the same. The role of the [M.sub.c] of the resin in dictating reinforcement potential lies in the molecular mechanism for antiplasticization. The additive enhances modulus and strength by eliminating or reducing local molecular motions along the polymer backbone. By increasing the [M.sub.c] of the resin, this localized effect is diluted as more degrees of freedom are introduced into the network.
[FIGURE 6 OMITTED]
The curing agent used has significant effects on the glass transition temperature of the thermoset (33). It is known that, while keeping the backbone stiffness constant, increasing the [M.sub.c] of network results in a linear reduction in the [T.sub.g]. The D230-D400 network, therefore, has a [T.sub.g] halfway between that of the D230 network and the D400. The effect of the antiplasticizer on [T.sub.g] is superimposed on the network contribution and results in the curve offsets seen in Fig. 7. The trend in [T.sub.g] of the antiplasticized networks follows the mechanical response closely, with the high-[M.sub.c] networks (D400) experiencing the largest drops in [T.sub.g]. Three of the networks appear to behave similarly: EDA, PDA, and D230. Increasing the [M.sub.c] of the network beyond that of the D230 results in decreased [T.sub.g] with increased TMP concentration, and a consequent reduction in mechanical performance.
[FIGURE 7 OMITTED]
The density increases are also matrix-dependent (Fig. 8). The slopes of the EDA, PDA, and D230 curves are very similar, indicating a comparable interaction between the additive and the matrix. As the [M.sub.c] of the network increases, the slope of the density vs. additive concentration curve also decreases. This drop-off is evident in the D230-D400 network and continues through to the D400 network, which, as has already been shown, is simply plasticized by TMP. This is then apparent in the reduction in mechanical properties of these materials.
[FIGURE 8 OMITTED]
Anhydride-cured epoxy does not contain hydrogen bond-donating moieties. Rather, the reaction between the epoxide and the anhydride leads to the creation of ester linkages, and no creation of hydroxyl groups as per the amine-curing reaction. Using this resin system, TMP is shown to exhibit antiplasticization behavior, although to a diminished degree. The HHPA resins were tested in compression to eliminate brittle failure of the polymer and observe the effect of the additive on the yield strength of the material. The fact that the anhydride resin is antiplasticized by TMP (Fig. 9) demonstrates that hydrogen bonding between additive and polymer is not a necessary condition for antiplasticization of epoxy networks. Increases of 17% and 5% of the compressive modulus and compressive yield stress were observed. Because the strength of the interaction is diminished, the threshold concentration is also reduced. The anhydride-cured resin exhibits a threshold concentration of only 10 mol%, significantly lower than that of the amine-cured resin. It can therefore be shown that polar-polar interactions between the additive and polymer are sufficient to ensure antiplasticization. However, because of the weaker interaction between the additive and polymer, the level of antiplasticization is reduced.
The mechanism of antiplasticization of epoxy thermosets has been shown to be dependent on both the architecture of the network and the molecular weight and size of the additive used. By systematically altering the size of the additive, the efficiency of the additive is probed. In the case of diamine-cured epoxy, the smallest additive investigated, TMP, was found to be the most efficient. Enhancements of 34% and 29% were observed in the modulus and yield strength of the 825-D230 system with TMP. This efficiency correlates with the high increase in density upon inclusion of the additive, as well as the minimal decrease in glass transition temperature of the resulting polymer. The strength of the interaction, captured by the solubility parameters of the two components, has been shown to correlate with the useful concentration range of each additive. As the size of the additive increases, the antiplasticization efficiency decreases significantly, both in concentration range and absolute mechanical enhancement. A triaryl phosphate, TPhP, was shown to antiplasticize the epoxy matrix in a similar manner to the trialkyl phosphates. With the exception of TMP, all the phosphates could be fitted to models predicting the decrease in specific volume upon incorporation of diluents into a polymer.
[FIGURE 9 OMITTED]
A number of curing agents, which result in different [M.sub.c], were also investigated. Changing the [M.sub.c] of the network demonstrates the effect of free volume and network mobility on antiplasticization. TMP acts most efficiently as an antiplasticizer in a tightly crosslinked network, where, at optimal concentrations, the tensile modulus increases 49% and the tensile strength 60%. In such tightly crosslinked networks, a reduction in sub-[T.sub.g] backbone relaxations is more effective at increasing static mechanical properties by mitigating the effects of long-chain flexibility. In tightly crosslinked thermosets, therefore, a low [M.sub.c] together with a small additive capable of interacting strongly with the polymer backbone can lead to significant improvements in the mechanical strength and stiffness of the mixture. The fortification is diminished for high-[M.sub.c] networks where long-chain motion dilutes the effect of reduced mobility of the hydroxylpropylether linkage.
The role of hydrogen bonding as a necessary condition for mechanical fortification was also examined. Using a curing agent that does not generate hydrogen bonding-donating moieties (HHPA), it was found that the formation of strong hydrogen bonds between the additive and the polymer is not a requirement for antiplasticization.
Table 1. Curing Agent Structures. Curing Agent Structure Mol. Wt.(g/mol) [M.sub.c] (g/mol) Ethylene diamine EDA 60 400 Propylene diamine PDA 74 410 Jeffamine D230 (x = 2.6) D230 241 500 Jeffamine D400 (x = 5.6) D400 453 600 Hexahydrophthalic anhydride HHPA 154 450 Table 2. Phosphate Additives. Phosphate [P(O)[(OR).sub.3]] R = Mol. Wt. (g/mol) Trimethyl phosphate TMP C[H.sub.3] 140 Triethyl phosphate TEP C[H.sub.2]C[H.sub.3] 182 Tripropyl phosphate TPrP C[H.sub.2]C[H.sub.2] C[H.sub.3] 224 Tributyl phosphate TBP C[H.sub.2]C[H.sub.2] C[H.sub.2]C[H.sub.3] 266 Triphenyl phosphate* TPhP [C.sub.6][H.sub.5] 326 Density [V.sub.w] Phosphate [P(O)[(OR).sub.3]] (g/[cm.sup.3] ([cm.sup.3]/mol) Trimethyl phosphate 1.197 63 Triethyl phosphate 1.072 94 Tripropyl phosphate 1.012 124 Tributyl phosphate 0.979 155 Triphenyl phosphate* 1.266 159.5 [[delta].sub.a] Phosphate [P(O)[(OR).sub.3]] (MPa)[.sub.1/2] Trimethyl phosphate 19.6 Triethyl phosphate 17.9 Tripropyl phosphate 17.6 Tributyl phosphate 17.3 Triphenyl phosphate* 17.6 *TPhP is a solid at room temperature ([T.sub.m] = 48[degrees]C).
1. H. T. Hahn and J. G. Williams, ASTM Spec. Tech. Publ., 115 (1986).
2. B. Budiansky and N. A. Fleck. J. Mech. Phys. Solids, 41, 183 (1993).
3. S. Ahmed and F. R. Jones, J. Mater. Sci., 25, 4933 (1990).
4. W. J. Jackson and J. R. Caldwell. J. Appl. Polym. Sci., 11, 227 (1967).
5. W. J. Jackson and J. R. Caldwell. J. Appl. Polym. Sci., 11, 221 (1967).
6. L. Makaruk. H. Polanska, and T. Mizerski, J. Appl. Polym. Sci., 23, 1935 (1979).
7. T. M. Don. Polym. Eng. Sci., 36, 2601 (1996).
8. M. K. Gupta, B. Ripmeester, D. J. Carlsson, and D. M. Wiles, J. Polym. Sci., Polym. Lett., 21. 211 (1983).
9. M. Shuster. M. Narkis, and A. Siegmann. Polym. Eng. Sci., 34, 1613 (1994).
10. S. L. Anderson. Macromolecules, 28, 2944 (1995).
11. L. Mascia, Polymer, 19, 325 (1978).
12. S. L. Brous and W. L. Semon, Ind. Eng. Chem., 667 (1935).
13. J. K. S. Darby and J. R. Darby. The Technology of Plasticizers. John Wiley & Sons: New York (1982).
14. G. Haldankar, E. Shockey, and A. Garton. Polym. Mater. Sci. Eng. Proc. ACS Div. Polym. Mater. Sci. Eng., 62, 120 (1990).
15. J. Daly, A. Britten, and A. Garton. J. Appl. Polym. Sci., 29, 1403 (1984).
16. V. G. Khozin, A. G. Farrakhov, and V. A. Voskresenskii, Poly. Sci. U.S.S.R., 21. 1757 (1979).
17. J. F. Shi, P. T. Inglefield, A. A. Jones, and M. D. Meadows, Macromolecules, 29, 605 (1996).
18. L. Heux, J. L. Halary, F. Laupretre, and L. Monnerie. Polymer, 38, 1767 (1997).
19. Y. Maeda and D. R. Paul, J. Polym. Sci., Part B: Polym. Phys., 25, 957 (1987).
20. Y. Maeda and D. R. Paul, J. Polym. Sci., Part B: Polym. Phys., 25, 981 (1987).
21. J. S. Vrentas, J. L. Duda, and H.-C. Ling. Macromolecules, 21, 1470 (1988).
22. F. A. Ruiz-Trevino and D. R. Paul, J. Polym. Sci. Part B: Polym. Phys., 36, 1037 (1998).
23. A. S. Zerda and A. J. Lesser. J. Appl. Polym. Sci., 84, 302 (2001).
24. C. S. Wang and C. H. Lin. J. Appl. Polym. Sci., 75, 429 (2000).
25. Y. L. Liu, G. H. Hsiue, and Y. S. Chiu, J. Polym. Sci. Part A: Polym. Chem., 35, 565 (1997).
26. F. Wurstlin and H. Klein, Makromolek. Chem., 16, 1 (1955).
27. P. B. Stickney and L. E. Cheyney. J. Polym. Sci., 3, 231 (1948).
28. E. A. Dimarzio and J. H. Gibbs. J. Polym. Sci. Part A: Polym. Chem., 1, 1417 (1963).
29. P. D. Ritchie, ed., Plasticizers, Stabilizers and Fillers, Iliffe Books. Ltd., London (1972).
30. D. W. van Krevelen, Properties of Polymers, Elsevier Science Publ., New York (1990).
31. J. Brandrup and E. H. Immergut, eds., Polymer Handbook. 3rd Ed., John Wiley & Sons, New York (1989).
32. A. J. Lesser and R. S. Kody, J. Polym. Sci. B: Polym. Phys., 35, 1611 (1997).
33. A. J. Lesser and E. Crawford. J. Appl. Polym. Sci., 66, 387 (1997).
34. E. Crawford and A. J. Lesser. J. Polym. Sci. Part B: Polym. Phys., 36, 1371 (1998).
35. V. Bellenger. E. Morel, and J. Verdu, J. Appl. Polym. Sci., 37, 2563 (1989).
36. M. E. Merritt, J. M. Goetz, D. Whitney, C. P. P. Chang, L. Heux, J. L. Halary, and J. Schaefer, Macromolecules, 31, 1214 (1998).
ADAM S. ZERDA and ALAN J. LESSER*
Polymer Science and Engineering
University of Massachusetts
Amherst, MA 01003
*To whom correspondence should be addressed.
|Printer friendly Cite/link Email Feedback|
|Author:||Zerda, Adam S.; Lesser, Alan J.|
|Publication:||Polymer Engineering and Science|
|Date:||Nov 1, 2004|
|Previous Article:||Mechanical, thermal, and barrier properties of NBR/organosilicate nanocomposites.|
|Next Article:||Preparation and properties of biodegradable poly(propylene carbonate)/starch composites.|