Curing kinetics and reaction-induced homogeneity in networks of poly(4-vinyl phenol) and diglycidylether epoxide cured with amine.
Thermosetting epoxies possess various desirable properties, such as high tensile strength and modulus, excellent chemical and solvent resistance, dimensional and thermal stability, good creep resistance, and excellent fatigue characteristics. These characteristics make them ideal candidate matrices for various important applications, including adhesives, electronic encapsulants, and matrix resins for high-performance fiber-reinforced composites. However, epoxy resins are normally brittle because they have high cross-linking densities. Some recent studies have shown that polymeric thermoplastics, such as poly(ether sulfone)s (PESs) [1, 2], poly(ether imide)s (PEIs) [3, 4], polycarbonate (PC) [5, 6], and poly(2,6-dimethyl phenylene oxide) (PPO)  have enhanced fracture toughness without sacrificing the glass transition temperature ([T.sub.g]) strength, stiffness, or other desirable characteristics of the thermosetting resin systems.
Complete network homogeneity in cured polymer/epoxy systems has been rarely reported. Only a few systematic works [4, 6, 8-11] have been performed on the compatibility and phase behavior of blends that comprise thermosetting resins and elastomer or thermoplastics. Curing with particular selected hardeners (4,4'-diaminodiphenylsulfone (DDS)) enables cross-linked epoxy systems that consist of polyether oxide (PEO) and diglycidyl ether of bisphenol-A (DGEBA) to form a miscible network . The hydrogen bonding interactions between the ether group of PEO and the hydroxyl group of the epoxy were observed before and after curing. These interactions have been considered to be one of the main driving forces of the miscibility in these thermoplastic-thermosetting polymer blends [12, 13]. However, the homogeneity of the cured epoxy/PC is caused by the extensive chemical links between the epoxy and PC molecules [6, 14, 15]. An earlier work  noted that the extensive chemical reactions between the polymer and the epoxy generate a rare homogeneous network formed by the DDS-curing of mixtures of epoxy resins with PC. The various structures of the hardeners can also produce various chain segments in the cured epoxy networks, which in turn affect the ultimate phase morphology of the cured epoxy/polymer solids . For example, Guo et al.  reported immiscibility in a cured PEO/epoxy system. However, the DGEBA epoxy in their work was cured with an aliphatic amine (tetraethylenepentamine (TEPA)) rather than DDS.
The curing agent, catalysts, or thermoplastics added to the epoxy blends importantly affect the mechanism of the epoxy curing [6, 16]. Different hardeners or catalysts can be used to enable individual epoxy resins to exhibit various kinetic mechanisms (such as the n-th order and autocatalytic kinetic mechanisms, among others). However, Lee et al.  reported that the curing reaction of the DGEBA/4,4'-methylene (MDA)/phenyl glycidyl (PGE)-acetamide (AcAm) system is autocatalytic, independently of the catalytic reaction, because of the preexistence of the hydroxyl group in PGE-AcAm as a catalyst. Polymeric additives, such as bisphenol-A polycarbonate (PC), polyetherimide (PEI), and poly(hydroxyl ether of bisphenol-A) (phenoxy) have been added to various epoxy resin systems so that their effects on the curing kinetics and network structures could be studied [4, 6, 14, 17]. The curing of the DDS-cured epoxy blends with the thermoplastic additive, PC, does not proceed from the original autocatalytic mechanism for curing epoxy with DDS, because of the chemical reactions that occur between the hydroxyl group (in epoxy) and the carbonate group in PC . An n-th order reaction mechanism effectively describes the curing behavior of the epoxy/PC blends up to vitrification. It is known that introducing PEI does not change the curing reaction mechanism of epoxies, but systematically shifts the phase-separated morphology of the cured polymer/epoxy networks . However, although the hydroxyl groups in the phenoxy can have a catalytic effect on the opening of the epoxide ring between the epoxy and the amine, its addition does not change the curing reaction mechanism in DDS-cured tetraglycidyl-4,4'-diaminodiphenylmethane (TGDDM) blends with phenoxy .
Moreover, one of our recent works  showed that some chemical reactions may proceed between an epoxide and certain hydroxyl-containing polymers, such as poly(4-vinyl phenol) (PVPh), when heated at high temperatures. Adding the thermoplastic additive to epoxy resins probably alters not only the network morphology but also thermal, physical, and mechanical characteristics. In this work we aim to clarify the reactive characteristics and accurately predict the curing behavior of DDS-cured epoxy/PVPh blends.
Materials and Sample Preparation
The PVPh was obtained from a specialty polymer supplier (Polysciences, Inc.), with a viscosity-average molecular weight [bar.M]v = 2.2 X [10.sup.4] g/mol and [T.sub.g] = 153[degrees]C. Diglycidylether of bisphenol-A (DGEBA) was obtained from Fluka Co., with an epoxide equivalent weight of 178 g (or a degree of polymerization, n = 0.04). The epoxy or epoxy/PVPh mixtures were cross-linked (cured) with DDS, supplied by Ciba-Geigy as HT-976. The amino hydrogen in the hardener had an equivalent weight of 62 g. The chemical structures of DGEBA and PVPh are as follows:
Hot-melt blending was performed to prepare the mixture samples. To minimize the thermal history, the PVPh powder was weighed and mixed in a predetermined quantity of the epoxy resin at 150[degrees]C by hot-melt blending, until the liquid epoxy/polymer mixtures appeared homogeneous. The homogeneity thus achieved was visually checked and double-checked with an optical microscope. The homogeneous epoxy/PVPh mixtures were immediately cooled to a temperature of 120[degrees]C. The DDS, already in powder form, was then introduced into the liquid mixture of epoxy/polymer, and the mixtures were stirred by hand until they appeared homogeneous. Then, one part of all of the mixtures was quenched with liquid nitrogen to analyze the curing kinetics using DSC. Another part was continuously heated to the predetermined curing temperature for a particular period as quickly as possible. The well-blended DGEBA/PVPh/DDS molten mixtures were spread onto potassium bromide (KBr) pellets (for examination by Fourier transform infrared spectroscopy (FT-IR)) or onto glass slides (to be examined under an optical microscope). The portions of PVPh added to the mixtures varied between 0 and 50 phr (phr represents the number of parts of PVPh per hundred parts of the DGEBA epoxy resin). It was more difficult to melt-blend the DGEBA/PVPh mixtures with higher PVPh contents because the viscosity of the whole mixtures increased greatly, making it difficult to stir the mixtures.
The glass transition temperatures ([T.sub.g]) and other thermal transitions of the blend samples were measured with a differential scanning calorimeter (Perkin-Elmer DSC-7) equipped with a mechanical intracooler (down to -60[degrees]C) and a computer for data acquisition/analysis. Additional subambient DSC runs (temperatures down to -80[degrees]C) were cooled with a liquid nitrogen tank and helium gas purging. Unless otherwise specified, all [T.sub.g] measurements were made at a scan rate of 20[degrees]C/min on the second scanning, and taken as the onset temperature of the transition of the heat flow curves.
FT-IR (Nicolet Magna-560) was used to examine molecular interactions between the constituents. Spectra were obtained at 4 [cm.sup.-1] resolution, and averages were obtained from at least 64 scans in the standard wave-number range of 400-4000 [cm.sup.-1].
We investigated the morphology of the cured DGEBA/PVPh/DDS blends using a polarized-light optical microscope (Nikon Optiphot-2 POL) with Ufx-DX automatic exposure to confirm the phase structure of the polymer mixtures before and after curing. Our objective was to examine the optical homogeneity of the epoxy/PVPh mixtures before and after curing, to provide additional evidence of phase structures. Heating and temperature control were provided by a microscope heating stage (Linkam THMS-600 with a TP-92 temperature programmer).
A conventional model of autocatalytic cure reaction was assumed for most of the amine-cured epoxy systems. Such a model is generally expressed as follows [20, 21]:
r = [[d[alpha]]/[dt]] = ([k.sub.1] + [k.sub.2][[alpha].sup.m])(1 - [alpha])[.sup.n] (1)
where [alpha] is the extent of conversion; r = d[alpha]/dt is the rate of the reaction, [k.sub.1] and [k.sub.2] are the apparent rate constants, and m and n are the kinetic exponents of the reactions. In most amine-cured epoxy systems or blends with thermoplastic polymers other than polycarbonate (PC), the cure reactions are autocatalytic [4, 18]. Exceptions to the autocatalytic reaction mechanism are normally found in anhydride-cured epoxy systems. For example, the curing kinetics of anhydride-cure epoxy systems have been shown to be n-th order . The isothermal curing rate of an n-th order reaction is given by r = d[alpha]/dt = [k.sub.1](1 - [alpha])[.sup.n]. N-th order kinetics can be considered to be a special case of autocatalytic mechanisms, in which the autocatalytic kinetic constant, [k.sub.2], approaches zero.
Horie et al.  assumed that the two main epoxy-amine reactions (those involving primary and secondary amines) are autocatalytic and have the same reactivity. They then derived the kinetic relationship d[alpha]/dt = ([k.sub.1] + [k.sub.2][alpha])(1 - [alpha])(B - [alpha]), where B = 1 (stoichiometry ratio). The model of Horie et al.  is a special case of Eq. 1 with m = 1 and n = 2. However, the third type of reaction in epoxy curing (the etherification reaction) and the observed deviation of the model at a high level of conversion have been attributed to gelation [22-27].
Equation 1 reveals that the constant [k.sub.1] can be calculated from the initial reaction rate near [alpha] = 0. The kinetic constants [k.sub.1] and [k.sub.2] are assumed to be in Arrhenius form:
[k.sub.i] = [A.sub.i]exp(-[[E.sub.ai]/[RT]]) (2)
where [A.sub.i] is the preexponential constant, [E.sub.ai] is the activation energy, R is the gas constant, and T is the absolute temperature. A first estimate of the reaction order n, Eq. 1, is obtained by modifying the reaction order n as follows:
ln([d[alpha]]/[dt]) = ln([k.sub.1] + [k.sub.2][[alpha].sup.m]) + n ln(1 - [alpha]). (3)
Except for the initial region, a plot of ln(d[alpha]/dt) vs. ln(1 - [alpha]) is expected to be linear, with slope n. Equation 1 can then be further rearranged to give
ln[[[d[alpha]/dt]/[(1 - [alpha])[.sup.n]]] - [k.sub.1]] = ln [k.sub.2] + m ln [alpha]. (4)
The first term in Eq. 4 can be determined from the previously estimated [k.sub.1] and n values. A plot of the left-hand term in Eq. 4 vs. ln([alpha]) is expected to yield a straight line whose slope and intercept can be used to estimate the reaction order, m, and the autocatalytic kinetic constant, [k.sub.2], respectively. The described procedure can be applied to obtain the preliminary kinetic parameters from the first trial. However, an iterative procedure is required to yield more values. Equation 4 can also be rearranged into the following form:
[FIGURE 1 OMITTED]
ln(d[alpha]/dt) - ln([k.sub.1] + [k.sub.2][[alpha].sup.m]) = n ln(1 - [alpha]) (5)
where [k.sub.2], m, and n are estimated according to the stated procedures; the left-hand terms in Eq. 5 can be plotted against ln(1 - [alpha]), and a new value of the reaction order n is checked against the one obtained earlier. The same iterative procedure can be repeated until the values of m and n converge to within 1% deviation.
RESULTS AND DISCUSSION
Glass Transition and Morphology of DDS-Cured Epoxy Blends With PVPh
Figure 1 shows thermograms of the 30 phr of PVPh in the cured DGEBA/PVPh/DDS samples with various contents of DDS. The single [T.sub.g] is related to the composition over the range of 10-70 phr DDS. The DGEBA/PVPh/DDS blend after curing exhibited a homogeneous morphology with no discernible phase separation domains.
As shown in Fig. 2, we plotted the [T.sub.g] compositions of the DDS-cured DGEBA network system without and with 30 phr of PVPh, to determine the effect of the PVPh contents on the DDS-cured epoxy. Before PVPh was added to the DDS-cured DGEBA network system, the [T.sub.g]s of the cured epoxy initially increased to and then declined from the maximum, suggesting that an appropriate amount of curing agent, DDS, could increase the cross-linking density of the epoxy and the [T.sub.g]s. However, the excess DDS reduced the extent of curing of the epoxy and the [T.sub.g]s of the whole system. After PVPh was added to the pre-cured DGEBA/DDS, the single [T.sub.g] of the cured DGEBA/PVPh/DDS system was found to depend on the composition. A plot of [T.sub.g]s vs. compositions was assumed to be parabolic with a maximum. Theoretically, the [T.sub.g] values of the thermoplastic/thermosetting blends should be between those of its individual components. Interestingly, the [T.sub.g] values of the DGEBA epoxy/PVPh cured with DDS were substantially increased, and were even higher than those of the cured neat DGEBA/DDS--probably because of chemical reactions between DGEBA epoxy and PVPh at high temperatures, and the regular cross-linking reactions between DGEBA and DDS. However, as the contents of DDS in cured DGEBA/PVPh systems increased to over 40 phr, the [T.sub.g]s of the cured DGEBA/PVPh/DDS systems were between individual [T.sub.g] values of the cured epoxy and PVPh. A recent report  showed that chemical reactions occur at higher temperatures in the DGEBA/PVPh blends in the absence of the curing agent DDS.
[FIGURE 2 OMITTED]
Figure 3A shows DSC thermograms of the 30 phr DDS-cured DGEBA/PVPh samples with 0, 5, 10, 20, 30, and 50 phr of PVPh. A single slightly broadened, composition-dependent [T.sub.g] over the range of PVPh contents from 0 to 50 phr was observed. The result reveals that PVPh is homogeneously distributed within the DDS-cured DGEBA network to achieve a one-phase solid state, which remains independent of the PVPh contents in the epoxy blends. Figure 3B shows [T.sub.g] as a function of PVPh content in the 30-phr DDS-cured DGEBA/PVPh network system. The figure shows that the [T.sub.g]'s initially increase with the PVPh content in the blends, and then fall after reaching a maximum, for the PVPh contents between 10 and 20 phr. Restated, the relationship between [T.sub.g] and PVPh content for the cured DGEBA epoxy/DDS appears to be quadratic with a maximum. Interestingly, the single [T.sub.g] is higher than that of either the cured DGEBA/DDS or that of neat PVPh, suggesting that DGEBA can react with DDS. In addition, DGEBA may also undergo other chemical reactions with PVPh during curing. Additionally, adding PVPh to epoxy/DDS increases the curing density of the mixtures and thus increases the [T.sub.g] value of DGEBA/PVPh/DDS following curing. Adding an excess of PVPh to epoxy/DDS may prevent the [T.sub.g] values of all of the mixtures from continuously increasing; however, the [T.sub.g] of the DGEBA/PVPh/DDS mixtures remained higher than that of both the DDS-cured epoxy and neat PVPh, indicating that more unreacted PVPh with DGEBA would reduce [T.sub.g] for DDS-cured epoxy blends with PVPh.
[FIGURE 3 OMITTED]
We further identified the effect of PVPh on the cured epoxy by increasing the content of DDS. Figure 4A shows the DSC themograms for the 50 phr DDS-cured DGEBA/PVPh samples with 0, 10, 20, 30, and 50 phr of PVPh. Similarly, a single [T.sub.g] is a function of the content in the range of 0-50 phr PVPh, indicating that the miscibility of the cured DGEBA/PVPh blends remained at high DDS content (50 phr), according to the thermal results of DSC, which were not affected by the DDS content in the cured DGEBA/PVPh/DDS mixtures. In summary, Fig. 4B shows that [T.sub.g] is a function of the PVPh content in the 50 phr DDS-cured DGEBA network system, and that the [T.sub.g] value of the cured DGEBA/PVPh/DDS (DDS = 50 phr) decreases as the PVPh content increases, gradually leveling off at a constant value. This finding demonstrates that introducing higher DDS content into the cured epoxy/PVPh mixtures increased the rate of cross-linking of epoxy, making PVPh less reactive than the DGEBA monomers. Even when more PVPh was added to the DDS-cured epoxy blends, the [T.sub.g] of all of the mixtures after curing decreased only slowly. Restated, PVPh competed with DDS to react with DGEBA monomers in the curing process.
[FIGURE 4 OMITTED]
Figure 5 shows the corresponding hydroxyl stretching region from 3720 to 3100 [cm.sup.-1] for DDS-cured epoxy blends with PVPh. The bottom spectrum of the cured DGEBA/DDS includes a broad band centered at 3410 [cm.sup.-1], which is assigned to a wide distribution of hydrogen-bonded hydroxyl groups. Additionally, an absorbance shoulder at a higher frequency was attributed to free or non-hydrogenbonded hydroxyls. The relative intensity of the non-hydrogen-bonded hydroxyls slowly declined as the PVPh content in the DGEBA/PVPh/DDS mixtures increased. However, the absorbance peaks of the hydroxyl groups (spectrum IV) shifted downward in frequency (3390 [cm.sup.-1]) as 50 phr of PVPh was added to the cured mixtures. The hydrogen bonding interactions became stronger as PVPh, which consisted of phenyl-hydroxyl groups in DDS-curing DGEBA blends, was added. Therefore, the factors that affected the miscibility of the DDS-cured DGEBA/PVPh mixtures were not only the chemical reactions between DGEBA epoxy and PVPh, but also the physical interactions among the hydroxyl groups. These results reveal that linear polymer PVPh completely interpenetrates with the DDS-cured epoxy network. We used an optical microscope for a preliminary examination of the cured epoxy/PVPh networks, and found that the networks were transparent with no discernible phase domains/boundaries at 800X (not shown for brevity). The cured networks of all epoxy/polymer blend compositions, as revealed by OM, were domain-free and transparent.
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
We investigated the curing of epoxy/DDS with the thermoplastic PVPh polymer using dynamic DSC heating scans. Figure 6 shows the DSC thermogram of the uncured DGEBA/PVPh/DDS samples with 0, 10, 30, and 50 phr PVPh contents at a scanning rate of 10[degrees]C/min. The single composition-dependent [T.sub.g] of the uncured DGEBA/PVPh/DDS blends increased with the content of PVPh in the mixtures. Additionally, exothermic peaks revealed the heat of the chemical reactions among DGEBA epoxy, DDS, and PVPh. The exothermic peak temperatures of the curing reactions, as obtained from the traces from the bottom to the top of Fig. 6, were approximately 210[degrees]C, 195[degrees]C, 175[degrees]C, and 170[degrees]C, respectively. The addition of PVPh to the DDS/epoxy mixtures substantially affected the chemical reactions in the mixtures. As more PVPh was added, the mixtures became more reactive. However, the total exothermic enthalpy, resulting from the dynamic DSC heating scans, apparently decreased as the content of PVPh increased, because PVPh competed with the curing agent (DDS) to undergo chemical reactions with DGEBA at high temperatures. The DGEBA/PVPh/DDS mixtures exhibited considerably less exothermic enthalpy than the curing reactions between DGEBA and DDS. These results reveal that both the DDS content and the PVPh content in the DGEBA/PVPh/DDS mixtures increase the reactivity and rate of ring opening for DGEBA epoxy.
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
Analyzing Curing Kinetics
The isothermal curing reaction of DSC was conducted at three temperatures: 177[degrees]C, 187[degrees]C, and 197[degrees]C, respectively. The reaction, which normally took about 2 hr, was considered complete when the isothermal DSC thermogram leveled off to the baseline. We used the total area under the exothermal curve, determined by extrapolating the baseline at the end of the reaction, to calculate the isothermal heat of curing, [DELTA][H.sub.I] (J/g). After the isothermal curing reaction in the DSC was complete, the sample was cooled to 40[degrees]C. The samples after isothermal curing were scanned at 10[degrees]C/min from 40[degrees]C to 300[degrees]C to determine the residual heat of reaction, [DELTA][H.sub.R] (J/g). The DSC scanning traces did not yield [DELTA][H.sub.R] for the DDS-cured DGEBA/PVPh, perhaps because the investigated epoxy (DGEBA with an epoxide equivalent weight of 178 g) had such a short chain that the curing reactions were more easily completed at the selected temperatures for a certain time, yielding the reduced curing density. Therefore, the total heat of curing, [DELTA][H.sub.T], was taken as [DELTA][H.sub.I], while [DELTA][H.sub.R] was identified as zero. The conversion, [alpha](t) at time t, was defined as the ratio of the partial heat of curing to the total heat of the isothermal curve ([alpha](t) = [DELTA][H.sub.I](t)/[DELTA][H.sub.T]). Moreover, the derivative of the curve of the extent of conversion with respect to time is normally regarded as determining the isothermal curing rate (r = d[alpha](t)/dt; Eq. 1). Notably, the calculation of the reaction heats of the blends was based on the net weight of the epoxy in the blends, excluding DDS and PVPh. Table 1 lists the heats of the curing reactions for DDS-cured epoxy blends with PVPh. The exothermal enthalpy ([DELTA][H.sub.T]) of the DDS-cured DGEBA/PVPh slightly increased with the isothermal temperatures. The result shows that the higher curing temperature associated with reactions of cured epoxy of higher degree could generate a greater exothermal enthalpy. Furthermore, [DELTA][H.sub.T] decreased as the PVPh content in the DGEBA/PVPh/DDS mixtures increased. However, the exothermal enthalpy of the reactions between PVPh and DGEBA was lower than that of the curing of epoxy with DDS, so the exothermal heat of the isothermal treatment of the DGEBA/PVPh/DDS mixtures became lower as the PVPh content in the mixtures increased. Accordingly, the curing enthalpy of DDS-cured DGEBA/PVPh was substantially lower than that of cured epoxy/DDS without PVPh. This result agrees closely with the results of the dynamic DSC heating scans.
Figure 7 shows the extent of reaction ([alpha]) as a function of time for the cured DGEBA/PVPh/DDS mixtures. The isothermal reaction curves of the DDS-cured epoxy system with PVPh exhibited a typical sigmoid shape, indicating that this system could be interpreted as one of autocatalyzed curing. Similar results were obtained for samples isothermally heated at the two higher temperatures (187[degrees]C and 197[degrees]C); therefore, for brevity, the DSC results are not presented here. Figure 8 plots the conversion rate (d[alpha]/dt) as a function of time. For the blends of PVPh and epoxy cured with DDS at various temperatures (177[degrees]C, 187[degrees]C, and 197[degrees]C), the reaction rate initially increased with time to a maximum, before it gradually fell to zero. The kinetics are autocatalytic. Restated, the addition of PVPh to DDS-cured epoxy blends did not affect the mechanism of the curing reactions. Additionally, the maximum reaction rate apparently increased with the PVPh content in DGEBA/PVPh/DDS mixtures. However, the time required by the mixture to reach the maximum reaction rate decreased as the PVPh content in the mixtures increased. These results were attributable to the chemical reactions between DGEBA and PVPh, which increased the extent of the curing reactions. These results differ from those observed for the miscible epoxy/PEO system with physical interactions , in which the concentration of the epoxy or hardener components was reduced by dilution caused by polymeric additives. Nevertheless, when the PVPh content in DDS-cured epoxy/PVPh exceeded 30 phr, the maximum curing rate for the DGEBA/PVPh/DDS mixtures was slightly reduced, but remained higher than that of neat epoxy cured with DDS. However, the time required by the mixture to reach the maximum reaction rate was shorter when 50 phr of PVPh were added to the mixtures, indicating that the high [T.sub.g] of PVPh increased [T.sub.g] values of the uncured DGEBA/PVPh/DDS mixtures (Fig. 6). Therefore, molecules moved more slowly, reducing the overall curing rate, when the PVPh content in the mixtures exceeded 30 phr. An examination of Fig. 8A-C reveals that for mixtures with the same contents at a high isothermal curing temperature, epoxy blends exhibited a high conversion rate over a given time. Based on the above results, we tested the experimental data using the autocatalytic model.
[FIGURE 9 OMITTED]
Table 2 lists the rate constants obtained with the use of iterative and graphic procedures. The reaction orders, m and n, varied slightly with the isothermal temperature, and increased with the content of PVPh in the DGEBA/PVPh/DDS mixtures, indicating that epoxy had a higher reaction order when PVPh was added to DDS-cured epoxy systems. Additionally, more epoxy in DGEBA/PVPh/DDS mixtures was exhausted during curing, because, like DDS, the epoxide groups of DGEBA underwent other chemical reactions with the phenyl-hydroxyl groups of PVPh. Figure 9 plots ln[k.sub.1] and ln[k.sub.2] vs. 1/T, yielding the activation energies for the epoxy and its blends. The values of [DELTA][E.sub.1] and [DELTA][E.sub.2] obtained in this work for the neat amine-cured DGEBA epoxy were 79.4 and 40.8 kJ/mol, respectively, which are the same as those found in the literature [21, 22]. The two kinetic constants of the DGEBA/PVPh/DDS mixtures, [k.sub.1] and [k.sub.2], are substantially larger than those of the neat amine-cured DGEBA epoxy, indicating that the DGEBA/PVPh/DDS mixtures cured faster than the DDS-cured epoxy. However, when the PVPh content in the blends was 50 phr, the increase in [k.sub.1] and [k.sub.2] became smaller, or the values actually dropped a little, but they remained larger than those of the neat epoxy system. Adding PVPh to the DDS-cured epoxy accelerated the curing reactions by causing chemical reactions between PVPh and the epoxy. However, when the PVPh contents in the mixtures exceeded 30 phr, the excess of PVPh did not continue to promote the reactions, but suppressed them. Adding PVPh to the DDS-cured epoxy promoted the complete curing reactions of cured DGEBA/PVPh/DDS blends, up to a particular PVPh content. These blends thus differed markedly from the DDS-cured epoxy blends with PEI  and phenoxy , respectively. The PEI component has no catalytic effect, and only the later-stage reaction exhibits an increased rate . However, in the phenoxy-modified epoxy system, the -OH group in the phenoxy catalyzes the opening of the epoxide ring between the epoxy and the amine in an early stage of curing .
[FIGURE 10 OMITTED]
Figure 10 shows that the empirical conversion curves fit the experimental data quite closely only in the early stage of curing for the DGEBA/PVPh mixtures with three different contents cured with 50 phr DDS at three different temperatures. The autocatalytic model specifies the kinetics accurately, except in the later stage of the curing reactions. The diffusion of species may be a controlling factor in the vitrified state, by preventing the reactions from proceeding. Following vitrification, curing almost stops, and thus the extent of curing is limited [4, 18, 28]. In this work, the [T.sub.g]s of the DDS-cured epoxy/PVPh blends did not exceed the curing temperature even after complete curing for 2 hr (Table 3).
Figure 11 plots the reaction rate versus the extent of conversion for the epoxy resin and its blends with three PVPh contents (10, 30, and 50 phr) at 177[degrees]C. The curing reaction of the DGEBA/PVPh/DDS mixtures at the lower levels and at the same conversion levels proceeded at a higher rate than that of neat epoxy resin. However, the addition of more than 30 phr PVPh did not further increase the curing rate of the epoxy blends. Therefore, when conversion was more extensive, the increase in the curing rate was reduced when 50 phr PVPh was added to the DGEBA/PVPh/DDS mixtures. The curing rate vs. extent of conversion of the epoxy blends at the other two cure temperatures (187[degrees]C and 197[degrees]C) exhibited similar trends, but are not shown here. A higher PVPh content may have been responsible for the higher [T.sub.g] value of the uncured epoxy blends, retarding the reactive molecules but not accelerating the curing of the DDS-cured epoxy blends with PVPh. Higher PVPh contents tended to suppress curing reactions. Moreover, as the extent of conversion during curing increased, the curing rate slowly declined, and the gelled epoxy network retarded the curing reaction.
[FIGURE 11 OMITTED]
The DDS-cured mixtures of polymeric PVPh (containing a pendant phenol) and DGEBA epoxide exhibited a homogeneous network (no phase-separation domains). The homogeneity of the DDS-cured PVPh networks was attributable not only to the physical hydrogen bonding interactions between the epoxy and PVPh, but also to the extensive chemical reactions between the epoxy and polymeric PVPh that occurred during the curing reactions between DDS and epoxy.
In this study, the effects of PVPh on the curing kinetics of DGEBA/PVPh mixtures were investigated. The DDS-curing of the mixed polymer/epoxy resin was autocatalytic in a manner similar to the curing of neat epoxy (which was not blended with PVPh). However, at a fixed DGEBA/DDS ratio, the curing rates of mixtures with higher PVPh contents were found to be substantially higher. Since more PVPh was added to the epoxy mixtures, the higher [T.sub.g] values of the vitrifying network may have retarded the cure reactions. However, although this effect changed the curing rates in the final stages, it did not change the general curing mechanism of the epoxy/PVPh/DDS mixtures. The addition of PVPh to the DDS/epoxy mixtures accelerated curing, perhaps for reasons related to the pendant phenol -OH group in PVPh. The kinetic parameters of the epoxy blends were obtained and the proposed kinetic model was found to describe accurately the DDS-curing of the DGEBA/PVPh blends up to the vitrification point.
TABLE 1. The isothermal cure heats of DDS (50 phr)-cured DGEBA with various PVPh contents at three temperatures. Isothermal cure heat (J/g) Cure temperatures PVPh (phr) 177[degrees]C 187[degrees]C 197[degrees]C 0 577.9 610.7 661.1 (neat epoxy) 10 431.5 460.0 556.1 30 423.6 442.4 507.0 50 272.7 279.6 300.0 TABLE 2. Autocatalytic modeling kinetic constants for DDS (50 phr)- curing of DGEBA/PVPh mixtures. [k.sub.1] X [k.sub.2] X PVPh [T.sub.Curing] [10.sup.3] [10.sup.3] (phr) ([degrees]C) m n ([min.sup.-1]) ([min.sup.-1]) 0 177 0.42 1.37 18.0 135.5 187 0.34 1.36 29.9 147.7 197 0.37 1.65 44.4 216.0 10 177 0.30 1.43 71.6 170.9 187 0.35 1.47 77.4 338.9 197 0.36 1.47 78.4 344.6 30 177 0.39 1.69 148.5 247.2 187 0.49 1.95 165.5 521.7 197 0.66 2.28 269.3 587.5 50 177 0.42 1.61 115.0 397.9 187 0.47 1.98 155.4 502.5 197 0.57 2.04 225.5 593.0 PVPh [E.sub.a1] [E.sub.a2] (phr) (kJ [mol.sup.-1]) (kJ [mol.sup.-1]) ln [A.sub.1] ln [A.sub.2] 0 79.4 40.8 17.2 8.9 10 8.1 62.1 -0.4 14.9 30 52.1 76.5 12.0 19.1 50 59.2 35.1 13.6 8.5 TABLE 3. The glass transition temperatures of DDS (50 phr)-cured DGEBA with various PVPh contents at three temperatures for 2h. [T.sub.g] of the cured epoxy mixtures ([degrees]C) Cure temperatures PVPh (phr) 177[degrees]C 187[degrees]C 197[degrees]C 0 182.4 184.4 185.4 (neat epoxy) 10 173.9 175.4 176.5 30 170.3 169.3 171.9 50 171.8 171.6 171.9
Contract grant sponsor: National Science Council of the Republic of China; contract grant number: NSC 92-2216-E-168-001.
1. C.B. Bucknall and I.K. Partridge, Polymer, 24, 639 (1983).
2. C. B. Bucknall, C. M. Gomez, and I. Quintard, Polymer, 35, 353 (1994).
3. D.J. Hourston and J.M. Lane, Polymer, 33, 1379 (1992).
4. C.C. Su and E.M. Woo, Polymer, 36, 2883 (1995).
5. T. Ohsako, K. Nagura, and I. Nozue, Polymer, 34, 5080 (1993).
6. C.C. Su, J.F. Kuo, and E.M. Woo, J. Polym Sci., Part B: Polym. Phys., 33, 2235 (1995).
7. R.A. Pearson, and A.F. Yee, Polymer, 34, 3658 (1993).
8. T.J. Horng and E.M. Woo, Polymer, 39, 4115 (1998).
9. Q. Guo, Polymer, 36, 4753 (1995).
10. S. Ritzenthaler, E. Girard-Reydet, and J.P. Pascault, Polymer, 41, 6375 (2000).
11. Q. Guo, X. Peng, and Z. Wang, Polym. Bull., 21, 593 (1989).
12. X. Luo, S. Zheng, N. Zhang, and D. Ma, Polymer, 35, 2619 (1994).
13. S. Talibuddin, L. Wu, and J. Runt, Macromolecules, 29, 7527 (1996).
14. C.C. Su and E.M. Woo, Macromolecules, 28, 6779 (1995).
15. M. Romg and H. Zeng, Polymer, 38, 269 (1997).
16. J.Y. Lee, H.K. Choi, M.J. Shim, and S.W. Kim, Thermochim. Acta, 343, 111 (2000).
17. E.M. Woo, Polym. J., 29, 523 (1997).
18. H.K. Hseih and E.M. Woo, Polymer, 39, 2175 (1998).
19. Y.P. Huang, E.M. Woo, P.L. Wu, and C.C. Su, J. Macromol. Sci., Phys., B43, 363 (2004).
20. K.C. Cole, J.J. Hechler, and D. Noel, Macromolecules, 24, 3098 (1991).
21. L. Chiao, Macromolecules, 23, 1286 (1990).
22. K. Horie, H. Hiura, M. Sawada, I. Mida, and H. Kambe, J. Polym. Sci. (Part A1), 8, 1357 (1970).
23. S. Sourour and M.R. Kamal, Thermochim. Acta, 14, 41 (1976).
24. J. Mijovic, A. Fishbain, and J. Wijana, Macromolecules, 25, 979 (1992).
25. J. Mijovic and J. Wijana, Polymer, 35, 2683 (1994).
26. L. Buckley and D. Roylance, Polym, Eng. Sci., 22, 166 (1982).
27. P. Johncock, L. Porecha, and G.F. Tudgey, J. Polym. Sci., 23, 291 (1985).
28. B.J. Enns and J.K. Gillham, ACS Adv. Chem. Ser., 203, 27 (1983).
Chean Cheng Su
Department of Chemical and Materials Engineering, National University of Kaohsiung, No. 700, Kaohsiung University Rd., Nan-Tzu District, Kaohsiung 811, Taiwan
E.M. Woo, Yin Ping Huang
Department of Chemical Engineering, National Cheng Kung University, Tainan 701, Taiwan
*Correspondence to: C.C. Su; e-mail: firstname.lastname@example.org
|Printer friendly Cite/link Email Feedback|
|Author:||Su, Chean Cheng; Woo, E.M.; Huang, Yin Ping|
|Publication:||Polymer Engineering and Science|
|Date:||Jan 1, 2005|
|Previous Article:||Study of the miscibility and mechanical properties of NBR/HNBR blends.|
|Next Article:||In-line optical detection in the transient state of extrusion polymer blending and reactive processing.|