2.3.1 PMMA-co-PVP Copolymer Characterization. A series of copolymers was prepared using various VP and MMA monomer concentrations. Table 2-1 lists the MMA contents (mol%) of the copolymers, determined through 1H NMR spectroscopy and EA. Because of the compositions of the copolymers determined through 1H NMR spectroscopy were affected by the interaction between water and PMMA-co-PVP, their VP contents (mol%) are slightly overestimated. EA provided better accuracy, although the H atom contents determined this way are inaccurate because of compositions of the copolymers were affected by the presence of water. Accordingly, we applied only the EA-determined N and C contents to calculate the VP content using Eq. (2-1). In the following discussion, the sample codes for these copolymers are based on the MMA contents obtained through EA.
We calculated reactivity ratios (r1 for MMA; r2 for VP) using the methodology of Kelen and Tudos.35–37 Table 2-1 summarizes the monomer feed ratios and the resultant compositions
of the copolymers. To minimize errors resulted from changes in the feed ratios, the polymerization was terminated at monomer conversions of less than 10%. The values of r1
and r2 are the ratios of the homo-propagation and cross-propagation rate constants for each monomer (i.e., k11/k12 and k22/k21, respectively). Figure 2-1 displays the Kelen–Tudos plot for the PMMA-co-PVP copolymers. The values of r1 and r2 were 0.94 and 0.97, respectively. In a previous study,38 we defined a copolymerization to be “ideal” when the product r1×r2 was unity. When r1 and r2 both equal 1, the two monomers possess equal reactivity toward both propagating species; the behavior of the resulting copolymer is referred to as random or Bernoullian. Thus, the copolymers synthesized through free radical polymerization in this study were essentially random. (i.e., close to an ideal copolymer: r1 × r2 = 0.91).
2.3.2. Analyses of OP-POSS/Homopolymer Binary Blends. Figure 2-2 presents DSC thermograms of the PMMA/OP-POSS and PVP/OP-POSS blends. The star-shaped OP-POSS was similar to a linear oligomer of poly(vinyl phenol), i.e., it has a degree of polymerization equal to 8. The glass transition temperature of OP-POSS (25 °C) was lower than that of a typical high-molecular-weight (Mn = 10,000 g/mol) PVPh (150 °C) because of molecular weights and structural differences. Single values of Tg existed in both blends, implying that all of these binary and ternary blends are miscible. Several equations have been suggested to predict the variation of the glass transition temperature of a random copolymer or miscible blend as a function of its composition. In this study, we employed the Kwei equation to predict the variation of the glass transition temperature:
2 transition temperature of the corresponding blend components, and k and q are fitting constants. Furthermore, the value of q, a parameter corresponding to the strength of hydrogen
bonds in the blend, correlated to the balance between the breaking of the self-association and the forming of the inter-association hydrogen bonds. Figure 2-3 displays the dependence of Tg
on compositions of the PVP/OP-POSS and PMMA/OP-POSS blends. We obtained the values of k and q based on non-linear least-squares best fits. In the PVP/OP-POSS blends, q had a value of +100, revealing the presence of a strong intermolecular interaction between PVP and OP-POSS. On the other hand, a negative value of q (–40) was obtained for the PMMA/OP-POSS blends, indicating that the intermolecular hydrogen bonding was weaker than the intramolecular hydrogen bonding.
Figure 2-4 displays partial IR spectra (2700–3700 cm–1) of the PMMA/OP-POSS and PVP/OP-POSS blends. The pure OP-POSS exhibits two bands in the OH stretching region in the IR spectrum; one corresponding to the hydrogen-bonded OH groups (a broad band centered at 3350 cm–1) and the other to “free” OH groups (a shoulder centered at 3525 cm–1).
When the PMMA (PVP) was mixed with OP-POSS and the C=O oxygen atoms of PMMA (PVP) interacted with the OH groups of OP-POSS, thus the broad band shifted to higher (lower) frequency at 3450 (3190) cm–1. This behavior reflected the competition between the hydroxyl–hydroxyl and hydroxyl–carbonyl interactions. Additionally, the hydroxyl–carbonyl interactions predominated over the hydroxyl–hydroxyl interactions in the PMMA (PVP)-rich blends; thus, we assigned the band at 3450 (3190) cm–1 to be the OH groups interacting with the C=O units. Moskala et al. used the frequency difference (Δ ν ) between the hydrogen-bonded and free OH absorptions to estimate the average strength of the intermolecular interaction.39 Accordingly, we used the free OH stretching at 3525 cm–1 as a reference, the hydroxyl–carbonyl inter-association was weaker than the hydroxyl–hydroxyl self-association interaction in the PMMA/OP-POSS blends, but stronger in the PVP/OP-POSS blends. This finding is consistent with the negative and positive values of q for the PMMA/OP-POSS and PVP/OP-POSS blends, based on Kwei equation.
Figure 2-5 displays the C=O stretching regions in the IR spectra of PVP/OP-POSS and PMMA/OP-POSS blends. In Figure 2-5(a), pure PVP exhibits a broad band centered at 1680 cm–1, corresponding to the “free” C=O groups. Painter et al.40 reported that the pyrrolidone group strongly self-associates through transitional dipole coupling. Therefore, the signal for
“free” C=O groups at 1680 cm–1 is not that of “truly free” C=O groups, which would be centered at 1708 cm–1. The signal for C=O stretching was split into two bands at 1680 and 1650 cm–1 corresponding to “free” and the hydrogen-bonded C=O groups, respectively; these signals fitted the Gaussian function well. As the concentration of OP-POSS increased, the probability of PVP/OP-POSS interactions increased, resulting in an increased intensity of the hydrogen-bonded C=O band at the expense of the “free” C=O band. We calculated the fraction of the hydrogen-bonded C=O groups (fb) using Eq. (2-4):41
Af
where Ab and Af denote the peak areas corresponding to the hydrogen-bonded and “free” C=O groups, respectively. In this case, we employed a ratio for the two absorptivities (a2/a1) of 1.3 based on a previous calculation.40 Table 2-2 summarizes the results of curve fitting for the PVP/OP-POSS blends. As expected, the fraction of hydrogen-bonded C=O groups increased upon increasing the OP-POSS content.
Figure 2-5(b) reveals that the PMMA blend system has a sharp (compared with that of the pure PVP) IR band at 1730 cm–1 and a shoulder at 1710 cm–1, representing the free and the hydrogen-bonded C=O groups, that also fitted the Gaussian function well. We applied the method described above to analyze the PMMA/OP-POSS blends, but in this case, we used a value of a2/a1 of 1.5.42 Table 2-2 lists the results of curve fitting of the PMMA/OP-POSS system. Again, the fractions of hydrogen-bonded C=O groups increased upon the increase of the OP-POSS content.
In previous studies,27,28 we confirmed that certain interactions occur between the POSS moieties and OH group. In this study, the situation was more complicated than those in previous studies because the OH groups were attached to the POSS cage. To further understand the interaction phenomena, we employed the Painter–Coleman association model (PCAM) to analyze these systems:41
where A, B, and C are descriptors representing the siloxane groups, the phenol groups of the POSS cages, and PMMA, respectively; KA, KB, and KC are their respective association equilibrium constants; K2 is the equilibrium constant of forming dimers between phenol groups. These equilibrium constants can be expressed in terms of volume fractions:
(2-5)
(VC/VB) are the ratios of the segmental molar volumes.28 Furthermore, we adopted the self-association equilibrium constants of PVPh45 (K2 = 21 and KB = 66.8), for phenol groups in this study to describe the formation of dimers and multimers, respectively. The
inter-association equilibrium constant (KC) of the PVPh/PMMA blend has been reported previously to be 37.4.44 Using the value of KC together with the phenol group self-association equilibrium constants (K2 and KB), we obtained a theoretical curve for the fraction of hydrogen-bonded C=O groups at 25 °C as a function of the weight fraction of OP-POSS content (Figure 2-6). When the value of KA was equal to 2, the experimental data agreed fairly well with the predictions of the PCAM. A slight deviation was observed at weight fractions less than 0.4 because the most accurate range for determining the fraction of hydrogen-bonded C=O groups was from 0.4 to 0.7, where the bands for both the free and hydrogen bonded C=O bands were well separated and had significant absorbances.45 The ratio of the inter-association equilibrium constant (KA) corresponding to the interaction between OH and siloxane groups of POSS cages to the inter-association equilibrium constant (KC), was 0.05, implying that inter-association between the phenol and siloxane groups of the POSS cages was insignificant and, thus, could be ignored. The structure of the OP-POSS and the hydroxyl–hydroxyl interaction formed through the phenol groups are the reasons for this behavior because both the arms of OP-POSS which were steric barriers and the presence of hydroxyl–hydroxyl interaction blocked the OH groups from interacting with the siloxane groups.
Using the value of the KB and K2 above and ignoring inter-association between the phenol and siloxane groups of OP-POSS, we employed the PCAM again to determine the “real”
value of KA for the PMMA/OP-POSS blend. The approximate equations were simplified as follows:46,47
⎟⎟⎠
where ΦA and ΦB denote the volume fractions of the non-self associated species A (PMMA) and the self-associating species B (OP-POSS), respectively: ΦA1 and ΦB1 are the corresponding volume fractions of the isolated PMMA and OP-POSS, respectively; r is the ratio of molar volumes (VA/VB). Using the least-squares method, we obtained theoretical curves for the fraction of hydrogen-bonded C=O groups at 25 °C as a function of the weight fraction of the OP-POSS content (Figure 2-7). When KA was equal to 29.0, the experimental data agreed fairly well with the predictions of the PCAM. Table 2-3 lists all the parameters required for the PCAM to estimate the thermodynamic properties of these blends.
Furthermore, the inter-association equilibrium constant (KA = 29.0) of the PMMA/OP-POSS blend system was smaller than that of the self-association equilibrium constant (66.8) of the OP-POSS oligomer; this finding implies that the tendency toward self-association of two OH groups dominates over the hydroxyl–carbonyl interactions in the PMMA/OP-POSS blends.
Most importantly, the value of KA for PMMA/OP-POSS blend system was smaller than those for poly(vinyl phenol) PVPh/PMMA(KA = 37.4)44 and ethyl phenol (EPh)/PMMA (KA = 101)48 blends, implying that the OH groups in the PMMA/OP-POSS blend system have less of a chance to interact with C=O groups than they do in the other two blends. In a previous study,44 we found that the value of the inter-association equilibrium constant is affected by the spacing between the hydrogen-bonding functional groups. In this case, the spacing between the OH groups attached to the POSS cage (a star-shaped macromolecule) can be smaller than those of the other two blend systems, making them less accessible for inter-association, resulting in a decrease in the ratio of the inter-association and self-association equilibrium
constants.
We could not, however, use this approach to determine the KA for the PVP/OP-POSS blend because the pyrrolidone groups exhibited strong self-association through transitional dipole coupling and the signal for “free” carbonyl group at 1680 cm–1 was not that of “truly free”
C=O groups, which would have been centered at 1708 cm–1. Furthermore, for the value of KA for the PVP/OP-POSS blends to be calculated accurately, it would have to be less than 6000±
2000 based on the results above.40 In previous studies,49–51 we determined that the value of q for the PMMA/PVPh and PVP/PVPh blend systems where 0 and +140, respectively—much greater than those of –40 and +100 for the PMMA/OP-POSS and PVP/OP-POSS blend systems, respectively—indicating that the intermolecular hydrogen bonding in the PMMA/OP-POSS and PVP/OP-POSS blends was weaker than that in the PMMA/PVPh and PVP/PVPh blends. The values of KA are in good agreement with these results obtained from curve fitting of the Kwei equation.
2.3.3. Analyses of Binary Blend OP-POSS/Copolymers. Figure 2-8 displays the C=O regions for the MMA and VP units in the IR spectra of the MMA61/OP-POSS blend. Table 2-4 summarizes the results of curve fitting. The addition of a small content of OP-POSS (20%) in the blend resulted in a new band at 1658 cm–1 assigned to the “hydrogen-bonded” C=O groups of the PVP. The signal for the “hydrogen-bonded” C=O groups of PMMA appeared when the OP-POSS content was increased to 40 wt%. The OH groups of OP-POSS prefer to interact with PVP rather than with PMMA, as evidenced by the difference between the value of KA for the PMMA/OP-POSS and PVP/OP-POSS blend systems. When the OP-POSS content was greater than 40%, the OH groups interacted simultaneously with both the PMMA and PVP units.
As indicated in Table 2-1, the presence of VP units in each copolymer system resulted in an increase in the value of Tg with all of the DSC traces revealing a single glass transition
temperature, implying that these copolymers are miscible. Figure 2-9 presents DSC thermograms of the MMA61/OP-POSS blend system. Essentially all of the blends displayed the same trend; the presence of OP-POSS results in a lowering of the glass transition temperature of the copolymer. We used the Kwei equation to predict the variation of the glass transition temperature of the miscible blend as a function of its composition. Figure 2-10 illustrates the dependence of Tg on the composition of the MMA81/OP-POSS, MMA61/OP-POSS, and MMA53/OP-POSS miscible blends. Again, based on the non-linear least-squares best fit, the values of k and q were obtained. For the MMA53/OP-POSS blend system, we obtained the largest value of q (+75) implying that the MMA53/OP-POSS blend system featured stronger intermolecular interactions between OP-POSS and itself than did the other two copolymer/OP-POSS blends. As indicated in Figure 2-10, the addition of a lower content of VP (20 wt%) in the PMMA chain resulted in a change in nature of the interaction.
The intermolecular hydrogen bonds in the PMMA/OP-POSS blends were weaker than the intramolecular hydrogen bonds. After copolymerization at a 20 wt% VP content, the intermolecular hydrogen bonding became stronger than the intramolecular hydrogen bonding because the OH groups interacted more preferably with the VP segments. Therefore, the value of q increased gradually upon increasing copolymerized-VP content, reflecting the fact that the intermolecular hydrogen bonding of the copolymer/OP-POSS blend system became stronger accordingly.
2.3.4. Analyses of Ionic Conductivity. In a previous study,34 we found that a polymer electrolyte composed of LiClO4/MMA61 had a higher ionic conductivity at room temperature than did the LiClO4/PMMA, LiClO4/PVP, LiClO4/MMA81, and LiClO4/MMA53 blend systems at the same LiClO4 content. As mentioned above, the interactions between polymer chains are affected by the presence of OP-POSS. In addition, because the mobility of charge carriers is directly related to the mobility of the polymer matrix, we were interested in
comparing the ionic conductivity of ternary and binary blends. Figure 2-11 displays the plots of the ionic conductivity with respect to the MMA content in copolymers at room temperature for LiClO4/PMMA-co-PVP and LiClO4/OP-POSS/PMMA-co-PVP blends containing a fixed LiClO4 content of 20 wt%. The polymer electrolyte LiClO4/OP-POSS/MMA61 exhibited a higher ionic conductivity at room temperature than did the binary blend of LiClO4/PMMA-co-PVP because the OP-POSS might lead to an increase in the chain mobility of the polymer electrolyte. To clarify the complicated nature of the interactions in the LiClO4/OP-POSS/polymer ternary blend, we are presently performing additional studies that we will discuss in the near future.
2.4. Conclusions
We have employed DSC and FTIR spectroscopy techniques to investigate in detail the miscibility behavior and mechanisms of interaction for polymer blends of OP-POSS and PMMA-co-PVP. For the OP-POSS/PMMA blends, the value of KA of 29 was smaller than those for the PVPh/PMMA (KA = 37.4) and EPh/PMMA (KA = 101) blends, implying that the spacing between phenol groups attached to the POSS nanoparticle was smaller than those of the other two blend systems, resulting in a decrease in the ratio of the inter-association and self-association equilibrium constants. Moreover, intermolecular hydrogen bonding became stronger than intramolecular hydrogen bonding after copolymerization with VP content because the OH groups preferred to interact with the VP segments. Furthermore, the presence of OP-POSS in a LiClO4/OP-POSS/MMA61 ternary blend played an important role in enhancing the ionic conductivity of the polymer electrolyte.
References
1. Kuo, S. W.; Lin, H. C.; Huang, W. J.; Huang, C. F.; Chang, F. C. J. Polym. Sci., Polym.
Phys. 2006, 44, 673.
2. Lee, Y. J.; Huang, J. M.; Kuo, S. W.; Chang, F. C. Polymer 2005, 46, 10056.
3. Zhang, H.; Kulkarni, S.; Wunder, S. L. J. Phys. Chem. B 2007, 111, 3583.
4. Kopesky, E. T.; Haddad, T. S.; Cohen, R. E.; McKinley, G. H. Macromolecules 2004, 37, 8992.
5. Li, G. Z.; Wang, L.; Toghiani, H.; Daulton, T. L.; Koyama, K.; Pittman, C. Y.
Macromolecules 2001, 34, 8686.
6. Matejka, L.; Strachota, A.; Plestil, J.; Whelan, P.; Steinhart, M.; Slouf, M. Macromolecules 2004, 37, 9449
7. Lin, H. C.; Wang, C. F.; Kuo, S. W.; Tung, P. H.; Huang, C. F.; Lin, C. H.; Chang, F. C. J.
Phys. Chem. B 2007, 111, 3404.
8. Leu, C. M.; Chang, Y. T.; Wei, K. H. Chem. Mater. 2003, 15, 2261.
9. Leu, C. M.; Chang, Y. T.; Wei, K. H. Chem. Mater. 2003, 15, 3721.
10. Phillips, S. H.; Haddad, T. S.; Tomcrzak, S. J. Curr. Opin. Solid State Mater. Sci. 2004, 8, 21.
11. Li, G. Z.; Wang, L. C.; Ni, H. L. J. Inorg. Organomet. Polym. 2001, 11, 123.
12. Chen, W. Y.; Wang, Y. Z.; Kuo, S. W.; Huang, C. F.; Tung, P. H.; Chang, F. C. Polymer 2004, 45, 6897.
13. Tamaki, R.; Choi, J.; Laine, R. M. Chem. Mater. 2003, 15, 793.
14. Lee, Y. J.; Huang, J. M.; Kuo, S. W.; Lu, J. S.; Chang, F. C. Polymer 2005, 46, 173.
15. Ye, Y. S.; Chen, W. Y.; Wang, Y. Z. J. Polym. Sci., Part A: Polym. Chem. 2006, 44, 5391.
16. Liu, Y. L.; Chang, G. P.; Hsu, K. Y.; Chang, F. C. J. Polym. Sci., Part A: Polym. Chem.
2006, 44, 3825.
17. Do Carmo, D. R.; Paim, L. L.; Dias, N. L. Appl. Surf. Sci. 2007, 253, 3683.
18. Voronkov, M. G.; Lavrent’ev, V. I. Top. Curr. Chem. 1982, 102, 199.
19. Baney, R. H.; Itoh, M.; Sakakibara, A.; Suzuki, T. Chem. Rev. 1995, 95, 1409.
20. Provatas, J. G. Trends Polym. Sci. 1997, 5, 327.
21. Loy, D. A.; Shea, K. J. Chem. Rev. 1995, 95, 1431.
22. Lichtenhan, J. D.; In Polymeric Materials Encyclopedia, Salamone, J. C. Ed.; CRC Press:
Boca Raton, FL, 1996, 7768.
23. Ni, Y.; Zheng, S. Chem. Mater. 2004, 16, 5141.
24. Kuo, S. W.; Chang, F. C. Macromolecules 2001, 34, 4089.
25. Kuo, S. W.; Lin, C. L.; Chang, F. C. Polymer 2002, 3943.
26. Eastwood, E.; Viswanathan, S.; O’Brien, C. P.; Kumar, D.; Dadmun, M. D. Polymer 2005, 3957.
27. Xu, H. Y.; Kuo, S. W.; Chang, F. C. Macromolecules 2002, 35, 8788.
28. Huang, C. F.; Kuo, S. W.; Huang, W. J.; Chang, F. C. Macromolecules 2006, 39, 300.
29. Maitra, P.; Wunder, S. L. Electrochem. Solid-State Lett. 2004, 7(4), A88.
30. Lin, H. C.; Kuo, S. W.; Huang, C. F.; Chang, F. C.; Macromol. Rapid Commun. 2006, 27, 537.
31. Xu, Y.; Painter, P. C.; Coleman, M. M. Polymer 1993, 34, 3010.
32. Luo, L.; Ranger, M.; Lessard, D. G..; Le Garrec, D.; Gori, S.; Leroux, J. C.; Rimmer, S.;
Smith, D. Macromolecules 2004, 37, 4008.
33. Watanabe, M.; Ohashi, S.; Sanui, K.; Ogata, N.; Kobayashi, T.; Ohtaki, Z.
Macromolecules 1985, 18, 1945.
34. Chiu, C. Y.; Yen, Y. J.; Chang, F. C. Polymer 2006, 48, 1329.
35. Kennedy, J. P.; Kelen, T.; Tüdös, F. J. Polym. Sci., Polym. Chem. Ed. 1975, 13, 2277.
36. Kelen, T.; Tüdös, F. Macromol. Sci. Chem. 1975, A9, 1.
37. Kuo, S. W.; Chang, F. C. Polymer 2001, 42, 9843.
38. Odian, G.. Principles of Polymerization, 4th ed.; John Wiley & Sons: New York, 2004.
39. Moskala, E. J.; Varnell, D. F.; Coleman, M. M. Polymer 1985, 26, 228.
40. Hu, Y.; Motzer, H. R.; Etxeberria, A. M.; Fernandez-Berridi, M. J.;Iruin, J. J.; Painter, P.
C.; Coleman, M. M. Macromol. Chem. Phys. 2000, 201, 705.
41. Coleman, M. M.; Graf, J. F.; Painter, P. C. Specific Interactions and the Miscibility of Polymer Blends; Technomic Publishing: Lancaster, PA, 1991.
42. Moskala, E. J.; Howe, S. E.; Painter, P. C.; Coleman, M. M.; Macromolecules 1984, 17, 1671.
43. Kuo, S. W.; Chan, S. C.; Chang, F. C. Polymer 2002, 43, 3653.
44. Lin, C. L.; Chen, W. C.; Liao, C. S.; Su, Y. C.; Huang, C. F.; Kuo, S. W.; Chang, F. C.
Macromolecules 2005, 38, 6435.
45. Coleman, M. M.; Pehlert, G. J.; Painter, P. C. Macromolecules 1996, 29, 6820.
46. Lee, Y. J.; Kuo, S. W.; Huang, W. J.; Lee, H. Y.; Chang, F. C. J. Polym. Sci. 2004, 42, 1127.
47. Kuo, S. W.; Chang, F. C. Macromol. Chem. Phys. 2002, 203, 868.
48. Hu, Y.; Painter, P. C.; Coleman, M. M.; Butera, R. J. Macromolecules 1998, 31(10), 3394.
49. Lee, H. F.; Kuo, S. W.; Chang, F. C. Macromolecules 2006, 39, 5458
50. Chen, W. C.; Kuo, S. W.; Jeng, U. S.; Chang, F. C Macromolecules 2008, 41, 1401.
51. Kuo, S. W.; Chang, F. C. Macromolecules 2001, 34, 4089.