Poly(vinylphenol-co-methyl methacrylate) with Poly(ethylene oxide)
Abstract
Experimental results indicate that PEO was miscible with PVPh-r-PMMA as shown by the existence of single composition-dependent glass transition temperature over the entire composition. However, the PVPh-b-PMMA blend with PEO shows a closed loop immiscible region in the ternary polymer blend system. Furthermore, FTIR reveals that at least three competing equilibrium are present in these blends;
self-association of PVPh-co-PMMA copolymer (hydroxyl-hydroxyl), interassociation of hydroxyl-carbonyl, and hydroxyl-ether interassociation between PVPh and PEO. Based on the Painter-Coleman Association Model (PCAM), a value for KC = 300 is obtained in PVPh-b-PMMA/PEO blend system at room temperature.
Although the relative ratio of interassociation equilibrium constant of PEO to PMMA is larger in PVPh-b-PMMA/PEO blend system, the PVPh-r-PMMA/PEO blend system has greater ∆ν and greater homogeneity at the molecular scale than the PVPh-b-PMMA/PEO blend system because of the ∆K effect.
5-1 Introduction
Polymer blends are one of the most important topics in polymer science during the last decade. For macromolecules, the Gibbs free energy of mixing is mainly controlled by its enthalpic contribution because the entropic term is usually favorable but very small. Therefore, polymer pairs with complementary chemical structures favoring specific interactions (and, thus a negative enthalpy of mixing) usually lead to miscible systems.1-4 Ideally, one polymer should possess donor sites and the other possesses acceptor sites on their respective chains. The most commonly observed interactions are of the acid-base type, i.e., hydrogen bonding,5 dipole-dipole, and charge-transfer interactions. On the contrary, when only dispersive forces can be expected between components, a positive (unfavorable) contribution of the enthalpic term is expected and as a consequence these systems are usually immiscible.
In order to obtain miscible polymer blends with desirable properties, it is very important to understand factors affecting the miscibility of polymer mixtures.
The miscibility of homopolymer/copolymer blends has been successfully described by the binary interaction model.6-8 According to this model, the mutual repulsive force between dissimilar segments in the copolymer can lead to negative heat of mixing necessary to attain miscibility. Paul et al.,9-11 Karasz et al.,12,13 and Jo et al.14 have further extended the above binary interaction model to several types of blends containing copolymers and applied the model to interpret the effect of the copolymer composition in the miscibility of blends. Nevertheless, the effect of the copolymer microstructures is not always explained in terms of neighboring repulsions. In the case of poly(vinyl acetate-co-vinyl alcohol) (ACA)15 or poly(vinylphenol-co-methyl methacrylate) (PVPh-co-PMMA),16 the specific interaction between carbonyl groups
and hydroxyl groups competes with hydroxyl self-association. Therefore, the monomer sequential distribution in these copolymers must play an important role in determining resulted hydrogen bond distribution. The possibility of hydrogen bond formation between hydroxyl groups and neighboring carbonyl groups in the copolymer attributes to a short scale competition of specific interactions. In addition, the sequential distribution of the copolymer in a homopolymer/copolymer blend will also affect the charge distribution and the probability of contact between interaction sites, and consequently affect the miscibility of the blend.17
In the ternary polymer blend, however, when all three binary pairs (B-A, B-C, and A-C) are individually miscible, a completely homogeneous or a closed immiscibility loop phase diagram has been observed.18 The phase separation is caused by the difference in the interaction energy of the binary system, the so-called
“∆χ” effect and “∆K” effect in ternary polymer blends such as phenoxy/PMMA/PEO,19 poly(vinylphenol) (PVPh)/poly(vinyl acetate) (PVAc)/PEO,20 and poly(styrene-co-acrylic acid)/PMMA/PEO.21 According to previous literature,5,22-24 binary pairs of PVPh/PMMA and PVPh/PEO are totally miscible over the entire compositions in the amorphous phase due to the formation of the interassociation hydrogen bonding between the hydroxyl group of PVPh and the carbonyl group of PMMA and the ether group of PEO, respectively. PEO is a highly crystalline polymer that is miscible with several weakly interacting polymers, such as PMMA,25-28 PVAc,29 and poly(vinylpyrrolindone).30 Miscible blends of PEO with PMMA have been well investigated in the literature, and the results indicate that the blend components are miscible in the melt and in the amorphous phase. PEO can act as a Lewis base since the oxygen atom bears a partial negative charge, while the carbonyl carbon atom of PMMA possesses a partially positive charge. In this study, we intend to the effect of different sequential distribution of a copolymer on
the miscibility in homopolymer/copolymer blend by FTIR and DSC analyses. In addition, the ∆K effect was found to play a key role resulting in different phase behaviors.
5-2 Experimental Section
5-2.1 Materials.
The poly(vinylphenol-co-methyl methacrylate) (PVPh-co-PMMA) copolymers studied in this work were prepared by different synthetic routes as described elsewhere.16 The block copolymer PVPh-b-PMMA and random copolymer PVPh-r-PMMA were synthesized by anionic and free radical copolymerization of 4-tert-butoxystyrene and methyl methacrylate, respectively. The tert-butoxy protective group was selectively removed through hydrolysis reaction. Table 1 shows the sequential distributions and molecular weights of several PVPh-co-PMMA copolymers employed in this study. Poly(ethylene oxide) (PEO) with Mn = 20 000 was obtained from Aldrich Co.
5-2.2 Blend Preparation
Blends of various (PVPh-co-PMMA)/PEO compositions were prepared by solution casting. Tetrahydrofuran (THF) solution containing 5 wt% polymer mixtures was stirred for 6-8 h and then cast onto a Teflon dish and was left to evaporate slowly at room temperature for 1 day. The blend film was then dried at 50 oC for 2 days.
5-2.3 Measurements.
All infrared spectra were recorded at 25 oC at a resolution of 1cm-1 on a Nicolet AVATAR 320 FTIR spectrometer. Each sample was dissolved in THF and then cast directly onto KBr pellets. All films were vacuum–dried and were thin enough to be within the absorbance range where the Beer-Lambert law is obeyed. Since these samples containing hydroxyl groups that are water sensitive, a pure nitrogen flow was used to purge the IR optical box in order to maintain sample films dryness.
Thermal analysis was performed on a DSC instrument from Du-Pont (DSC-9000) at
a scan rate of 20 oC/min over a temperature ranging from 20 to 250 oC. The sample was quenched to -120 oC from the melt state for the first scan and then rescanned between -120 oC and 250 oC at 20 oC/min. The glass transition temperature was obtained at the inflection point of the jump heat capacity.
5-3 Results and Discussion
5-3.1 Thermal Analyses
The PVPh is totally miscible with PEO and PMMA in the amorphous phase due to the formation of the interassociation hydrogen bonding between the hydroxyl group of the PVPh and the carbonyl group of the PMMA or the ether group of the PEO. In addition, PEO and PMMA are also fully miscible in the amorphous phase.
In general, differential scanning calorimetry (DSC) is one of the convenient methods to determine the miscibility in polymer blends. Meanwhile, a single compositionally dependent glass transition is an indication of full miscibility at a dimensional scale between 20 and 40 nm. Figure 1 shows the conventional second run DSC thermograms of PVPh-r-PMMA/PEO blends of various compositions; it reveals that each blend composition has only single glass transition temperature. A single value of Tg strongly suggests that the PVPh-r-PMMA/PEO blend system is fully miscible in the homogeneous amorphous phase. Table 2 summarizes the thermal properties of PVPh-r-PMMA, PEO, and their blends. At higher PEO contents, PEO crystallizes from the molten mixture of PEO and PVPh-r-PMMA copolymer. The melting temperature of the PEO component decreases with increasing the PVPh-r-PMMA copolymer content in the blend. The melting temperature also decreases with increasing the PVPh content in the PVPh-r-PMMA copolymer at the same blend ratio. This phenomenon suggests that the PVPh-r-PMMA copolymer in the blend hinders the crystallization of PEO, which is a typical phenomenon of a miscible blend in which the glass transition temperature of an amorphous polymer is higher than that of a crystalline component. Furthermore, the dependence of Tg on the composition of the miscible PVPh-r-PMMA/PEO blends is shown in Figure 2A.
Clearly, the value of Tg of each PVPh-r-PMMA/PEO blend shift to lower
temperatures as the PEO content increases in the blend. However, the upturn of the values of Tg at higher PEO content is due to the crystallization of PEO during quenching. This phenomenon suggests that not only the crystallization of PEO in the blends can change the amorphous phase but also the crystal of PEO is able to act as a physical cross-linking point that may hinder the molecular mobility of amorphous phase.31,32
Figure 3 displays the second run DSC thermograms of PVPh-b-PMMA/PEO blends. Again, the binary PVPh/PMMA, PVPh/PEO, and PMMA/PEO blends are all fully miscible in the amorphous phase. However, the DSC thermograms of 30-b-70/PEO = 8/2, 7/3, and 40-b-60/PEO = 8/2 show two Tg’s, implying that they are immiscible in the amorphous phase. The thermal properties of PVPh-b-PMMA/PEO blends were summarized in Table 3 and the dependence of Tg
on the composition of the miscible PVPh-b-PMMA/PEO blends is shown in Figure 2B. It is observed that the thermal behavior of PVPh-b-PMMA/PEO blends have the similar tendency to the PVPh-r-PMMA/PEO blends.
5-3.2 FT-IR analyses
Infrared spectroscopy has been used to detect the existence of specific interactions in polymer blends. This tool can be used to study the mechanism of interpolymer miscibility through the bond formation both qualitatively and quantitatively. Several regions within the infrared spectra of PVPh-co-PMMA/PEO blends are influenced by the hydrogen-bonding interaction.
Figure 4A shows infrared spectra in the 2700-3800 cm-1 range for the pure PVPh. This broad band can be considered to be composed of narrow contributions corresponding to hydroxyl groups surrounded by different environments: hydroxyl
groups hydrogen bonded with other hydroxyl groups in the same or vicinal chains (forming dimmers, trimers, etc.), and non-hydrogen bonded hydroxyl groups.33 In addition, the spectra of the miscible 92-r-8/PEO blends show significant changes in this region, suggesting a redistribution in the arrangement of the hydroxyl group association. These data indicate that there are many different types of hydroxyl groups present in PVPh-r-PMMA copolymers and PVPh-r-PMMA/PEO blends. In the meantime, the spectrum of pure PVPh shown in Figure 4A is characterized by a very broad band centered at 3350 cm-1, indicating that these hydroxyl groups are hydrogen bonded to other hydroxyl groups as dimmers and chain-like multimers. A second narrower band, observed at 3525 cm-1 as a shoulder on the high frequency side of the broad hydrogen bonded band, is assigned to free hydroxyl groups.5 Taking into account the effect of composition, the carbonyl groups of methyl methacrylate units compete with self-associated hydroxyl groups for hydrogen bonding and cause the shift of the hydroxyl band toward higher wavenumbers at lower vinylphenol content. In this situation, majority of only one type of hydroxyl group from the hydrogen–carbonyl inter-association is expected, and thus the hydroxyl stretching band is relatively narrower. On the contrary, the free, dimmer, or multimer hydrogen bonded hydroxyl groups will exist at higher vinylphenol contents, resulting in broader absorptions. Therefore, the spectrum of 30-r-70 copolymer shown in Figure 4B is reasonable to assign the band at 3440 cm-1 to the hydroxyl groups interacting with carbonyl groups because the small number of the hydroxyl groups tend to interact completely with carbonyl groups. When comparing the spectra corresponding to the same system as a function of composition, a progressive shift of this band toward lower wavenumber is observed for increasing content of ether oxygen of PEO (Figure 4). This behavior suggests that a significant part of the hydroxyl groups involved in the association processes previously
described for PVPh-r-PMMA copolymer are now hydrogen bonded to ether oxygen groups in PEO.
In the case of PVPh-b-PMMA/PEO blends (Figure 5), infrared spectra show a clear shift of the hydroxyl band toward lower wavenumbers, relative to that of the pure PVPh-b-PMMA copolymer. As the PEO content in the blend increases, the band gradually shift to lower frequencies, providing additional evidence for the existence of hydrogen bonding interaction between the PEO ether group and the hydroxyl group of PVPh. The frequency difference between the free hydroxyl absorbance and the hydrogen bonded hydroxyl (∆ν) is a measure of the average strength of the intermolecular interactions.34,35 For instance, we use the position of the free hydroxyl stretching vibration at 3525 cm-1 as a reference, then the median frequency difference for hydroxyl-hydroxyl self-association in PVPh is about 175 cm-1 while that of the interassociation between hydroxyl groups of PVPh and carbonyl groups of PMMA is 85 cm-1. Figure 6 displays the frequency difference (∆ν) in FTIR of all PVPh-r-PMMA/PEO and PVPh-b-PMMA/PEO blends at room temperature vs. PEO content. The ∆ν of PVPh-r-PMMA/PEO and PVPh-b-PMMA/PEO blend system exhibits the same increasing trend as the PEO content is increased. In the 30-r-70/PEO = 4/6 blend, there is no evidence for the presence of free hydroxyl and the concentration of hydroxyl-hydroxyl interactions appears insignificant. Therefore, it is reasonable to assign the band at 3150 cm-1 to the hydroxyl groups of PVPh hydrogen bonded to ether oxygens of the PEO in the PVPh-r-PMMA/PEO system. The frequency difference between the free hydroxyl band and the band attributed to hydroxyl groups hydrogen bonded to ether oxygens is about 375 cm-1. This is somewhat greater than that observed for PVPh-b-PMMA/PEO system (c.a. 350 cm-1) and presumably reflects a moderate
interaction; a reasonable conclusion considering the enhanced affinity for hydrogen bonding of the hydroxyl groups in PVPh-r-PMMA compared to that in PVPh-b-PMMA. However, these values of ∆ν in all copolymer/PEO blends also imply that the interassociation between PVPh and PEO is considerably stronger than either the self-association of hydroxyl groups in PVPh or the interactions between hydroxyl groups in PVPh and carbonyl groups in PMMA.
The CH2 wagging region of the pure PEO and its blends with PVPh-co-PMMA copolymer is now examined. Figure 7 and Figure 8 show infrared spectra in the 1320-1380 cm-1 region of the pure PEO, and various PVPh-co-PMMA/PEO blends at room temperature. The pure PEO has two bands at 1360 and 1343 cm-1 that represent the crystalline phase of the PEO.36 These bands decrease as the PVPh-co-PMMA content is increased. These crystalline bands disappear on PVPh-r-PMMA/PEO = 4/6 and PVPh-b-PMMA/PEO = 5/5 blends and are replaced by a broad band roughly centered at 1350 cm-1 corresponding to the amorphous phase. That means the PEO crystallization is being retarded or even inhibited by adding the amorphous PVPh-co-PMMA copolymer. As a result, we can confirm that the hydroxyl group of PVPh is more favorable to form the interassociation with ether group of PEO than with carbonyl group of PMMA.
However, the difference in crystallization behaviors was investigated by DSC and FTIR on 30-r-70/PEO = 4/6 and 30-b-70/PEO = 5-5 blends. In general, the polymer crystallinity measured by FTIR is from direct sample measurement, and no thermal history is involved in preparing the sample. On the contrary, the polymer crystallinity detected by DSC depends on the thermal history because recrystallization may occur during cooling or heating scan.
We now turn our attention to Figure 6 again. The frequency difference (∆ν) increases with increasing PEO content and approaches a maximum for the blend
containing 60 wt% PEO except the 92-r-8/PEO and 75-b-25/PEO blends, in which they show ∆ν decrease with further increase of the PEO content. The observed slight decrease in ∆ν at higher PEO content (80 wt%) can be explained by the fact that these blend systems possess PEO crystalline phase. Consequently, the hydroxyl-ether interassociation tends to decrease between the PVPh and PEO segments due to reduced chain mobility in the PEO crystalline phase.37 In another aspect, since more hydroxyl groups are present at the interface between PVPh-co-PMMA copolymer and PEO crystalline phase in the 92-r-8/PEO = 2/8 and 75-b-25/PEO = 2/8 blends, the hydroxyl stretching bands do not shift back to higher wavenumber.
The carbonyl stretching band for PMMA appears at 1730 cm-1. Figure 9 and Figure 10 show the infrared spectra of the carbonyl stretching measured at room temperature ranging from 1670 to 1770 cm-1 of PVPh-co-PMMA and PVPh-co-PMMA/PEO blends. The carbonyl stretching of the pure PVPh-co-PMMA is split into two bands, absorption by free and hydrogen bonded carbonyl groups at 1730 and 1705 cm-1, respectively. Obviously, the relative contribution of these two types of carbonyl groups must be dependent on copolymer composition. As has been observed in infrared spectra, we can expect a higher fraction of hydrogen bonded carbonyl groups for copolymers rich in vinylphenol units where carbonyl groups are more surrounded by the donor medium. As can be seen in Figure 9 and Figure 10, the presence of ether groups in these blend leads to a competition with carbonyl groups for hydrogen bonding with hydroxyl groups. This competition is evident in the evolution of the carbonyl band in the blend. Thus for a particular PVPh-co-PMMA copolymer, the hydrogen bonded carbonyl groups (progressive decrease of the shoulder at relatively lower wavenumber) decrease with the increase
association prevails over the hydroxyl-carbonyl association. This result agrees with the evolution previously reported for the hydroxyl stretching region.
These bands can be readily decomposed into two Gaussian peaks, the free carbonyl (1730 cm-1) and the hydrogen bonded carbonyl (1705 cm-1) absorptions.
Using the known respective absorptivity coefficients, fractions of these two types carbonyl groups can be calculated from the relative intensities of these two bands.
To obtain the fraction of the hydrogen bonded carbonyl group, a known absorptivity ratio for hydrogen bonded and free carbonyl is required. We employed a value of αHB/αF = 1.5 which was previously calculated by Moskala et al.38 Table 4 and Table 5 summarize fractions of hydrogen-bonded carbonyl calculated through curve fitting of the data from both copolymers and their blends. Table 4 and Table 5 show that the fraction of hydrogen bonded carbonyl decreases with increasing the relative ratio of PEO to PMMA. This result implies that the interassociation equilibrium constant of hydroxyl-ether is greater than the interassociation equilibrium constant of hydroxyl-carbonyl and the self-association equilibrium constant of hydroxyl-hydroxyl at room temperature. In our previous study,37,39 we have used three competing functional groups to predict the fraction of hydrogen bonded carbonyl group. According to the Painter-Coleman Association Model (PCAM), we designate B, A, and C as PVPh, PMMA, and PEO, respectively. K2, KB, KA, and KC
are their respective association equilibrium constants.
(1) 2 B2
These four equilibrium constants can be expressed as follows in terms of volume VC/VB are the ratios of segmental molar volumes.
The self-association constant of PVPh (hydroxyl-hydroxyl), the interassociation constant between PMMA and PVPh (carbonyl-hydroxyl), and the interassociation constant between PEO and PVPh (ether-hydroxyl) in PVPh-r-PMMA/PEO blend have been reported in the literature.16,40,41 The interassociation constant of PEO in PVPh-b-PMMA/PEO blend is determined indirectly from a least-squares fitting procedure. If these equilibrium constants (K2, KB, KA), segment molar volume, and the fraction of hydrogen bonded carbonyl group are known, the KC value can be calculated from eqs 5-9 by using a
The self-association constant of PVPh (hydroxyl-hydroxyl), the interassociation constant between PMMA and PVPh (carbonyl-hydroxyl), and the interassociation constant between PEO and PVPh (ether-hydroxyl) in PVPh-r-PMMA/PEO blend have been reported in the literature.16,40,41 The interassociation constant of PEO in PVPh-b-PMMA/PEO blend is determined indirectly from a least-squares fitting procedure. If these equilibrium constants (K2, KB, KA), segment molar volume, and the fraction of hydrogen bonded carbonyl group are known, the KC value can be calculated from eqs 5-9 by using a