A study on the ternary blends of polyphenylenesulfide, polysulfone, and liquid crystalline polyesteramide.
Recently the "in-situ composite forming technique" as been proposed to improve the mechanical properties of general purpose engineering plastics (1). In this technique, the thermotropic liquid crystalline polymer (LCP) reinforcing agent comes out as a low viscous melt during the compounding process, and crystallizes upon cooling to form needle-like reinforcing structures in the final fabrication step. This in-situ crystallization technique is advantageous over the conventional composites containing an inorganic reinforcement since melt processability and mechanical properties can be simultaneously improved. Most of the in-situ composites have been made up of two polymers; isotropic matrix polymer and anisotropic reinforcing polymer. The isotropic polymer may be either non-crystallizable (2-12) or crystallizable (13-20), and the anisotropic polymer are usually an LCP.
In general, the LCP blends have been reported to exhibit tensile strengths lower than the predicted values by rule of mixtures over a wide range of blend compositions (21-23). Particularly in the blends of two immiscible crystalline polymers such as LCP/poly(phenylene sulfide) (PPS) blend, unsatisfactory mechanical properties are frequently observed (24-27). Seppalla et al. (28) have reported that injection molded blends of Vectra A and PPS have a tensile modulus slightly exceeding the predicted value by rule of mixtures at LCP contents higher than 20wt%, but have a tensile strength lower than the predicted value. Similar tendencies of mechanical properties have also been reported by Harada et al. (29) for the blends of Vectra A950 and nylon 6 or 12.
A polysulfone (PSF) has been introduced to the PPS/LCP blend systems in the hope that PSF may improve the interface properties through some degree of chemical interactions with PPS because both polymers contain aromatic rings and sulfur (30). The effect of PSF on the physical properties of PPS/LCP blends has been discussed.
A crystalline poly(p-phenylene sulfide) (PPS), Ryton GR-02 of Phillips Petroleum Co., was used as matrix polymer, and the reinforcing component was a wholly aromatic thermotropic liquid crystalline copolyesteramide, Vectra B950 (LCP) of Hoechst Celanese Co., which is made up of 2,6-hydroxynaphthoic acid (60%), terephthalic acid (20%), and aminophenol (20%). As a third component incorporated to the PPS/Vectra B950 blends, an amorphous polysulfone (PSF), Udel P-1700 of Amoco Co., was tested.
Melt Blending and Preparation of Test Specimens
The materials were dried at 100 [degrees] C in a forced convection oven for at least 24 h before use. The loading level of Vectra in the blends was 10, 25, 50, and 75 wt% and that of PSF was 15 and 30 wt%. The blends were prepared by melt extruding the formulated components through a Brabender twin-screw compounder, quenching in a cooling water bath, followed by drawing and pelletizing. The extrusion temperature was controlled at 290 [degrees] C for PPS, at 300 [degrees] C for PPS/LCP blend compositions.
The specimens for tensile and rheological experiments were prepared by injection molding the blends. The stock temperature was 350 [degrees] C, the mold temperature, 100 [degrees] C, the injection pressure, 7000 psi, and the cycle time, 1 min. The blend fibers were obtained by spinning through a capillary rheometer (Instron model 3211). The capillary radius was 1.245 mm and the length to diameter ratio, 41. The spin draw ratio was determined by dividing the capillary diameter by the diameter of the spun fiber.
Measurement of Physical Properties
The thermal properties were measured with a DuPont Thermal Analyzer 2000 equipped with a 910 DSC apparatus. The heating and cooling rates were 10 [degrees] C/ min, respectively. The isothermal crystallization data were obtained by heating the sample up to 330 [degrees] C in the nitrogen atmosphere, holding for 3 min at 330 [degrees] C, and then quenching at the rate of 160 [degrees] C/min to a predetermined crystallization temperature, 245 [degrees] C, at which the exothermic crystallization peak was recorded. The total crystallization time was determined as the time required from the onset of the peak till a steady baseline was obtained.
The rheological measurements were carried out by Rheometrics Dynamic Spectrometer RDS 7700 in the nitrogen atmosphere at 290, 300, and 310 [degrees] C. In the RDS measurements, a parallel plate geometry, plate diameter 25mm, and gap size 1.2 mm, was adopted. The tests were performed in a frequency sweep mode from 0.1 to 500 rad/s at the shear strain level 10%.
The mechanical properties of the blend fibers were measured using an Instron Model 4201 tensile tester. The gauge length and crosshead speed were 30 mm and 5 mm/min, respectively. The mechanical properties of the injection molded blend specimen were measured with an Instron testing machine operated at the crosshead speed 5 cm/min at room temperature, according to the ASTM D638 (5 cm gauge length). The average value of six measurements was taken. The tensile fractured surface of the blend fibers was observed using a scanning electron microscope, Hitachi S-510. The optical texture was observed by a Leitz Ortholux II microscope attached with heating stage and programmable PID temperature controller (model 350).
RESULTS AND DISCUSSION
The thermal transition behavior of polymer blends can give important information on compatibility as well as crystallization behavior. The DSC analysis results of PPS/LCP blends are summarized in Table 1.
[TABULAR DATA FOR TABLE 1 OMITTED]
The DSC curve of PPS clearly shows glass transition temperature ([T.sub.g]) at 90 [degrees] C, cold crystallization temperature ([T.sub.cc]) at 119 [degrees] C, melt crystallization temperature ([T.sub.mc]) at 206 [degrees] C, and melting temperature ([T.sub.m]) at 279 [degrees] C (20, 23-25). LCP exhibits a crystalline to nematic transition at 282 [degrees] C (31-35). It should be noted that the thermal transitions of PPS such as [T.sub.g], [T.sub.m], and [T.sub.cc] appear to be independent of the LCP content. This suggests that PPS and LCP used in this study are incompatible in all compositions examined. However, increasing the LCP content raises the melt crystallization temperature ([T.sub.mc]) but reduces the crystallization half-time at 245 [degrees] C, suggesting that LCP finely dispersed within the PPS matrix plays the role of a nucleating agent in the crystallization of PPS (31, 35).
The viscosity curves of PPS/LCP blends at 290 and 300 [degrees] C are presented in Figs. 1 and 2, respectively.
PPS exhibits a Newtonian flow behavior over the entire frequency range observed at both temperatures, which is usually found in flexible chain polymers as shown in Fig. 1 (36). On the other hand, LCP does not give a Newtonian flow region at 290 [degrees] C, suggesting yield behavior. Further the viscosity of PPS/LCP blends are generally increased with increasing LCP content at 290 [degrees] C. At 300 [degrees] C, however, the PPS/LCP blend shows very peculiar rheological behavior as shown in Fig. 2. First, both PPS and LCP produce a Newtonian flow region at low frequencies. Second, the blend viscosity shows a maximum value at the blend ratio 50/50. The mechanism of the viscosity increase up to 50 wt% LCP content is believed to be quite complicated. Several suggestions have been made to explain the intricate theological behavior including structural change of PPS/LCP blend resulting from PPS curing in air (37, 38). Further suggestions include formation of distorted crystalline structures in the anisotropic melt (39, 40) and filler effect of the small fraction of crystallites in the polymer melt as a result of phase separation in the immiscible PPS/LCP blend (36, 39). Unlike the fibrillar structure suggested by Mantia et al. (19), this system develops a spherical LCP structure in the PPS matrix. In addition, the occurrence of crosslinking and chain extension reaction on heating PPS in air can be included (38). In our judgement, hence, the viscosity increase in this case may be due to combined factors of crosslinking of PPS and formation of spherical LCP structures (as will be morphologically shown later).
In the viscosity curves the disappearance of the lower Newtonian flow region is indicative of a non-zero yield stress frequently observed in heterogeneous systems (36, 41). It is empirically established that the yield behavior is clearly shown in the plot of square root of stress against square root of shear rate, the so-called Casson plot (2, 3, 41). Similar plotting of square root of loss modulus ([G[double prime].sup.1/2]) against square root of frequency ([[Omega].sup.1/2]) also gives a measure of system inhomogeneity. It should be noted in the Casson-type plot shown in Fig. 3 that LCP and PPS/LCP blends with high LCP content give a nonzero loss modulus, analogously to a non-zero yield stress in the Casson plot, which indicates that the system is not homogeneous.
PPS possesses dual characteristics of thermoplastic and thermoset. Consequently, the rheological properties of a PPS melt is expected to be changed with time during melt processing. The time dependence of the rheological properties of PPS, 50/50 PPS/LCP blend, and LCP at 300 [degrees] C is compared in Fig. 4.
In these three cases, the dynamic viscosity is substantially increased with time, most notable in the 50/50 PPS/LCP blend. The main reason for the viscosity increase of PPS/LCP blends may be the well recognized rheological responses of two respective components. With PPS, the increase of viscosity during melting processing is attributable to chain extension through oxidative crosslinking because of air exposure during the melt blending process in this study. Oxygen uptake followed by a loss of S[O.sub.2] is also considered as a factor to increase the viscosity as reported by others (37, 38). In the case of LCP, the rearrangement of crystal structure is known to cause a viscosity increase (39, 40, 42, 43).
Fig. 5 shows the effect of PSF on the rheological properties of PPS/LCP blends containing 10 phr LCP at 290 [degrees] C.
For better and finer dispersion of one polymer in another polymer in melt blending, it is prerequisite that the viscosity ratio of dispersed to matrix phases should be smaller than unity (12, 22). Comparing Fig. 5 with Fig. 1 reveals that at a fixed 10 wt% LCP content, addition of PSF notably increases the viscosity of the blend systems. This implies that PSF increases the viscosity of matrix (PPS) phase, and hence this may increase the droplet deformation of dispersed LCP domains because it is expected to facilitate the load transfer from the matrix to the dispersed phase (13, 45, 46). In addition, it is self-evident that mixing temperature and mixing rate (viz., shear rate during mixing) also have a very significant effect on the resultant dispersed morphology because they determine the viscosity ratio.
The tensile modulus and tensile strength of as-spun PPS/LCP and PPS/LCP/PSF fibers are plotted against the LCP content in Fig. 6.
Up to 10 wt% LCP content, the tensile modulus and tensile strength are increased with increasing LCP content. However, the binary blend fiber gives tensile modulus and tensile strength falling below those predicted by the additivity rule of mixing if the LCP content exceeds 20 wt%. Of two tensile properties, the tensile strength exhibits a more serious deviation from the additivity rule. In fact, the tensile strength has a minimum value in the vicinity of 50/50 blending ratio. The poor tensile properties of PPS/LCP blends are attributable to cleavage, cavitation, interfacial debonding, poor dispersion and destruction of heterogeneities resulting from the phase separation as reported by others (47-50). However, one can see a synergistic effect on tensile strength if the third component PSF is incorporated into the binary blend as shown in Fig. 6b (denoted as an open rectangle). In fact, incorporation of 30 wt% of PSF into the 90/10 PPS/LCP blend increases the tensile strength ca. 20%. As predicted, the effect of draw ratio on the tensile strength of the PPS/LCP fiber is more pronounced in the LCP rich blend fiber. The SEM photomicrographs of the tensile fractured surface of PPS/LCP and PPS/LCP/PSF fibers are shown in Figs. 7 and 8, respectively.
These SEM photomicrographs clearly show that the spherical or ellipsoidal domains observed in the binary PPS/LCP blend system as seen in Fig. 7 are deformed into rod-like or thread-like fibrils by addition of PSF as shown in Fig. 8.
The tensile modulus and tensile strength of the injection molded PPS/LCP and PPS/LCP/PSF specimens are plotted against the LCP content in Fig. 9.
The tensile properties of the injection molded article also exhibit similar trends to those of the as-spun PPS/LCP fibers all over the compositions examined. That is, incorporation of PSF into the PPS/LCP blends increases the tensile strength by ca. 20% particularly at low LCP content. The tensile fractured morphology of the injection molded PPS/LCP and PPS/LCP/PSF specimens are shown in Figs. 10 and 11, respectively.
The injection molded blends also exhibit similar tendency in mechanical properties to the blend fibers. That is, the microstructure also reveals that the higher elongation of the dispersed LCP domains upon incorporating PSF is responsible for the improved tensile strength.
In addition, a loss of S[O.sub.2] takes place when PPS is processed at the elevated temperatures as evidenced by other researchers (37, 44). Ramanathan et al. (27) have reported that the extruded blends of Vectra and PPS exhibit a foamed cellular structure due to the reaction between two components during extrusion. In fact, the outgases seem to pass through the LCP domains as illustrated in the polarized optical micrographs of 50/50 PPS/LCP blend at 350 [degrees] C in Fig. 12.
The 50/50 PPS/LCP blend forms a co-continuous network at 350 [degrees] C in the static state whereas applying shear to the blend system leads to an aggregated structure as shown in Fig. 12b. When this specimen under shear is cooled to 190 [degrees] C, below [T.sub.mc], instabilities at the interface of PPS and LCP and in the LCP domains are observed as shown in Fig. 12c. Further, in the PPS/Vectra blends, the acidic gases from PPS such as S[O.sub.2] and [H.sub.2]S evolved at the elevated temperature may deteriorate the physical properties of LCP phase through promotion of the hydrolysis of ester groups in LCP (51).
In the binary in-situ composites, poor dispersion and poor interface adhesion frequently lead to deteriorated mechanical properties, more seriously if the matrix is a crystalline polymer such as PPS. This problem may be solved to some degree by incorporating a carefully selected amorphous polymeric additive to the blend system. In the case of PPS/Vectra B950 blends, introducing an amorphous PSF has turned out to be effective in enhancing the elongation of the dispersed LCP domains, which produces a positive effect in tensile strength particularly at low LCP content. In these incompatible LCP blends, compatibilizing between two phases as well as higher elongation of the LCP domains is essential for improving mechanical performance of the resultant in-situ composites. Although incorporation of PSF into PPS/LCP blends increases the deformation of LCP domains, it is not so successful in improving the interface adhesion. Future studies on the compatibilized LCP blend systems will be carried out to manufacture in-situ composites with better mechanical performance.
1. G. Kiss, Polym. Eng. Sci., 27, 410 (1987).
2. S. M. Hong, B. C. Kim, K. U. Kim, and I. J. Chung, Polym. J., 23, 1347 (1991).
3. S. M. Hong, B. C. Kim, S. S. Hwang, and K. U. Kim, Polym. Eng. Sci., 33, 630 (1993).
4. S. M. Hong and B. C. Kim, Polym. Eng. Sci., 34, 1605 (1994).
5. A. Golovoy, M. Kozlowski, and M. Narkis, Polym. Eng. Sci., 32, 854 (1992).
6. D. Beery, S. Kenic, A. Siegmann, and M. Narkis, Polym. Eng. Sci., 32, 14 (1992).
7. T. M. Malik, P. J. Carreau, and N. Chapleau, Polym. Eng. Sci., 29, 600 (1989).
8. B. Y. Shin and I. J. Chung, Polym. J., 21, 851 (1989).
9. A. I. Isayev and M. Modic, Polym. Compos., 8, 158 (1987).
10. K. G. Blizard and D. G. Baird, Polym. Eng. Sci., 27, 653 (1987).
11. S. H. Jung and S. C. Kim, Polym. J., 20, 73 (1988).
12. C. Carfagna, E. Amendola, L. Nicolais, D. Acierno, O. Francescangeli, B. Yang, and F. Rustichelli, J. Appl. Polym. Sci., 43, 839 (1991).
13. M. R. Nobile, E. Amendola, L. Nicolais, D. Acierno, and C. Carfagna, Polym. Eng. Sci., 29, 224 (1989).
14. A. Kohli, N. Chung, and R. A. Weiss, Polym. Eng. Sci., 29, 573 (1989).
15. C. U. Ko and G. L. Wilkes, J. Appl. Polym. Sci., 37, 3063 (1989).
16. W. G. Perkins, A. M. Marcelli, and H. W. Frerking, Jr, J. Appl. Polym. Sci., 43, 329 (1991).
17. A. K. Mithal and A. Tayebi, Polym. Eng. Sci., 31, 1533 (1991).
18. B. Y. Shin and I. J. Chung, Polym. Eng. Sci., 30, 13 (1990).
19. F. P. L. Mantia, A. Valenza, M. Paci, and P. L. Magagnini, Rheol. Acta, 28, 417 (1989).
20. M. Amano and K. Nakagawa, Polymer, 28, 263 (1987).
21. W. C. Lee and A. T. Dibenedetto, Polymer, 34, 684 (1993).
22. S. Lee, S. M. Hong, Y. Seo, T. S. Park, S. S. Hwang, and K. U. Kim, Polymer, 35, 519 (1994).
23. M. H. B. Skovby, J. Kops, and R. A. Weiss, Polym. Eng. Sci., 31, 954 (1991).
24. Y. Oyanagi, Plast. Mold. Technol., 4, 89 (1987).
25. D. G. Baird, T. Sun, D. S, Done, and R. Ramanathan, Am. Chem. Soc., Div. Polym. Chem., Polym. Prepr., 30, 546, (1989).
26. P. R. Subramanian and A. I. Isayev, Polymer, 32, 1961 (1991).
27. R. Ramanathan, K. G. Blizard, and D. Baird, SPE ANTEC. Tech. Papers, 34, 1123 (1988).
28. J. Seppala, M. Heino, and C. Kapanen, J. Appl. Polym. Sci., 44, 1051 (1992).
29. T. Harada, K. Tomari, S. Tonogai, S. Nagai, and K. Yamaoka, "The Polym. Proc. Soc., Intern. 5th Annual Meeting," Japan, 219, April (1989).
30. U.S. Pat. 4,021,596 (1977).
31. S. M. Hong, B. C. Kim, K. U. Kim, and I. J. Chung, Polym. J., 24, 727 (1992).
32. G. P. Desio and L. Rebenfeld, J. Appl. Polym. Sci., 39, 825 (1990).
33. V. M. Nadkarni and J. P. Jog, J. Appl. Polym. Sci., 32, 5817 (1986).
34. V. M. Nadkami, V. L. Shingankuli, and J. P. Jog, Intern. Polym. Proc., 2, 55 (1987).
35. L. Minkova, M. Paci, M. Pracella, and P. Magagnini, Polym. Eng. Sci., 32, 57 (1992).
36. L. E. Nielsen, Polymer Rheology, Marcel Dekker, Inc. (1977).
37. C. C. M. Ma, H. C. Hsia, E. L. Liu, and J. T. Hu, Polym. Compos., 8, 256 (1987).
38. C. C. M. Ma and L. T. Hsiue, J. Appl. Polym. Sci., 39, 1399 (1990).
39. C. R. Gochanour and M. Weinberg, J. Rheol., 30, 101 (1986).
40. D. S. Kalika, D. W. Giles, and M. M. Denn, J. Rheol., 34, (1990).
41. R. M. Ottenbrite, L. R. Utracki, and S. Inoue, Current Topics in Polymer Science, Vol II, pp. 149-65, Hanser Publishers, Munich (1987).
42. S. S. Kim and C. D. Han, Macromolecules, 26, 3176 (1993).
43. Y. G. Lin and H. H. Winter, Macromolecules, 24, 2877 (1991).
44. N. S. J. Christopher, J. L. Cotter, G. J. Knighy, and W. W. Wright, J. Appl. Polym. Sci., 12, 863 (1968).
45. H. B. Chin and C. D. Han, J. Rheol., 23, 557 (1979).
46. C. D. Han, Multiphase Flow in Polymer Processing, Chapt. 5, Academic Press, New York (1981).
47. M. F. Cheung, A. Golovoy, H. K. Plummer, and H. V. Oene, Polymer, 31, 2299 (1990).
48. M. F. Cheung, A. Golovoy, and H. V. Oene, Polymer, 31, 2307 (1990).
49. M. F. Cheung and H. K. Plummer, Jr., Polym. Bull., 26, 349 (1991).
50. S. Akhtar and J. L. White, Polym. Eng. Sci., 31, 84 (1991).
51. A. Golovoy, M. F. Cheung, and M. Zinbo, Polym. Commun., 30, 322 (1989).
|Printer friendly Cite/link Email Feedback|
|Author:||Kim, Byoung Chul; Hong, Soon Man; Hwang, Seung Sang; Kim, Kwang Ung|
|Publication:||Polymer Engineering and Science|
|Date:||Feb 1, 1996|
|Previous Article:||Epoxidation reaction of trimethylolpropane with epichlorohydrin: kinetic study of chlorohydrin formation.|
|Next Article:||A field equation for the distribution of striation thickness from impingement mixers.|