Compatibilizers for thermotropic liquid crystalline polymer/polyethylene blends prepared by reactive mixing.
Polymer blends containing a thermotropic liquid crystalline polymer (TLCP) are of interest because of the ability to form a fibrillar, reinforcing TLCP phase during processing. However, the interfacial adhesion between TLCPs and conventional thermoplastics is generally weak, and as a result, the expected reinforcement of the TLCP microfibrillar phase is rarely achieved (1-6). Improvements in the mechanical properties of these blends can be achieved by adding compatibilizers (7-11). Compatibilization may be a result of physical interactions between the compatibilizer and the two polymers (10) or it may involve the formation of covalent bonds between what is called a reactive compatibilizer and one or both of the other components (11).
Blends of a thermotropic liquid crystalline polyester with polyolefins such as polyethylene or polypropylene are of interest with regard to barrier properties, and recent work has shown that these blends can be compatiblized with a partially neutralized poly(ethylene-co-r-acrylic acid), EAA, ionomer (11, 12). Zhang et al. (11) reported that the sodium salt of EAA reacted with the TLCP at elevated temperature by an acidolysis reaction, see Scheme 1, to form a graft-copolymer of the TLCP and the ionomer, and the EAA-TLCP graft copolymers were effective compatibilzers for blends of the TLCP with low-density polyethylene, high-density polyethylene and polypropylene (12). This paper provides a more detailed analysis of the acidolysis reaction and the properties of compatibilized TLCP/polyethylene blends.
The TLCP used was Vectra[TM] A950 (Ticona/Celanese), which is a wholly aromatic copolyester of 73% hydroxybenzoate and 27% hydroxynaphthanoate; see Scheme 2. The ionomer was Iotek[TM] 8000 (Exxon Chemical Co.), which is a 45% neutralized [Na.sup.+]-salt of an EAA containing 15 wt% acrylic acid; see Scheme 2. Hereafter, these two materials are referred to as TLCP and EAA-Na, respectively. The thermal and flow properties of these and the low density polyethylene (LDPE and high density polyethylene used are given in Table 1.
Graft-copolymer compatibilizers with various compositions were prepared by reactive mixing of the TLCP and EAA-Na in a Wayne Machine and Die Co., 2.54 cm single-screw extruder equipped with a Maddock mixing screw (L/D = 25). The processing temperature and screw speed were varied from 280 to 330[degrees]C and 20 to 80 rpm, respectively, to control the extent of graft-copolymer formation. The feed composition of the TLCP/EAA-Na mixture was varied from 10 to 50 wt% TLCP. The notation used for the compatibilizers was Comxy-tuv-rs where xy is the composition expressed as wt% TLCP, tuv was the nominal extrusion temperature ([degrees]C) and rs was the screw speed (rpm) used: the compatibilizers prepared are summarized in Table 2.
Blends of TLCP and polyethylene (either LDPE or HDPE) with and without a pre-formed TLCP-(EAA-Na) graft copolymer compatibilizer were prepared with the single-screw extruder. Unless otherwise indicated, the ratio of TLCP to polyethylene was 1:3 (w/w). Pellets of LDPE, EAA-Na and the compatibilizer were dried at 70[degrees]C and the TLCP was dried at 150[degrees]C in a vacuum oven for at least 12 h prior to extrusion. The dried pellets were then tumbled together in predetermined compositions in a polyethylene bag, and the dry blends were extruded at 290[degrees]C using a screw speed of 40 rpm. The in situ formation of the compatibilizer was also evaluated by mixing the TLCP, LDPE and EAA-Na at 310[degrees]C in the single-screw extruder using various screw speeds ranging from 20 to 80 rpm. The pelletized extrudates were dried at 70[degrees]C and then molded into type V tensile specimens (ASTM D638) with an Arburg Allrounder injection molding machine using the following conditions: 150 rpm screw speed, 65 MPa in jection pressure, 60 MPa holding pressure and a temperature setup in the four zones from the throat to the die of 250, 270, 290 and 290[degrees]C. The mold temperature was maintained at ca. 25[degrees]C using chilled water.
Tensile testing of the injection molded specimens was carried out with an Instron Universal Testing Machine, Model 1011, using a crosshead speed of 5 mm/min. The arithmetic mean and standard deviation of the tensile properties were calculated using a minimum of five samples. The notched impact strength was measured with an Izod impact tester at room temperature according to ASTM Standard D256. The dimensions of the samples were 64 mm in length, 10.5 mm in width and 3.2 mm in thickness, and the average impact strength was calculated using at least five specimens. A rolling saw was used to make a 1 mm notch with a 45[degrees] angle.
The morphology of the blends was observed with an Hitachi environmental scanning electron microscopy (SEM). Injection molded samples were cryogenically fractured and the fracture surfaces were coated with gold and platinum. The SEM micrographs reported in this paper show the center of the fracture surface, which was entirely within the core region in samples that exhibited a skin-core morphology.
The complex viscosity of the TLCP/LDPE blends with and without compatibilizer was measured at 290[degrees]C as a function of frequency with an Advanced Rheometric Expansion System (ARES[TM]) rheometer (Rheometric Scientific) using 25-mm parallel plates. The strain was kept at 0.1, which was within the linear region of response of the materials. All measurements were conducted under a nitrogen atmosphere in order to prevent oxidative degradation. The viscosity of the pure component polymers was also measured at 290[degrees]C using the as-received pellets.
The extent of graft-copolymer formation was characterized by Fourier transform infrared (FTIR) spectroscopy using a Nicolet Magna-IR 560 FTIR spectrometer. A total of 32 scans were averaged with a resolution of 2 [cm.sup.-1]. Blends of 7 wt% TLCP and 93% EAA-Na were prepared in a Brabender Plasti-Corder at 290[degrees]C for various mixing times up to 60 min, and were then pressed into thin films by compression molding in a Wabashi automatic press at 200[degrees]C, which was below the melting point of the TLCP in order to suppress the further reaction.
RESULT AND DISCUSSION
Zhang et al. (11) showed that the reaction of a 50/50 mixture of TLCP and EAA-Na resulted in the formation of ester linkages with aromatic and aliphatic substituents; the evidence of that reaction was a weak shoulder at ca. 1761 [cm.sup.-1] in the infrared spectrum. In this study, we attempted to better characterize the acidolysis reaction by using compositions with excess ionomer, specifically a ratio of TLCP to EAA-Na of 7/93 (w/w). The FTIR spectra of (7/93) TLCP/EAA-Na blends mixed in a Brabender intensive mixer at 290[degrees]C for various time up to 60 min are shown in Fig. 1. The neat EAA-Na was also mixed in the intensive mixer at 290[degrees]C for 60 min to determine whether any of the changes in the spectra in Fig. 1 were due to chemical decomposition of the ionomer; no changes were evident in the FTIR spectra of the ionomer.
The acidolysis reaction between the TLCP and EAA-Na produced an ester linkage with aromatic and aliphatic substituents which absorbed at 1761 [cm.sup.-1] (11). The aliphatic-aromatic ester acidolysis product at 1761 [cm.sup.-1] progressed from a weak shoulder for short mixing times to a distinct peak at the longer mixing times. In addition to the 1761 [cm.sup.-1] absorbance, the spectra for the TLCP/EAA-Na blends in Fig. 1 also show four other new absorbances not present in either of the two neat polymer components, at 1633, 1493, 1198 and 1116 [cm.sup.-1], which increased in intensity with increasing reaction time.
The strong absorbances at ca. 1695 and 1576 [cm.sup.-1] in Fig. 1 were due to the C=O stretching vibration for COOH and the asymmetric [COO.sup.-] stretching vibration, respectively, in the EAA-Na (13). The acidolysis reaction should also produce carboxylic acid with aromatic substituents, Ar-COOH, and if the [COO.sup.-] [Na.sup.+] salt groups from the EAA-Na participate in the acidolysis reaction, aromatic carboxylate salts, Ar-[COO.sup.-] [Na.sup.+], may also be formed. For carboxylic acid and carboxylates, an aromatic substituent should shift the infrared absorbance to lower wave number compared to those groups with aliphatic substituents, and accordingly, it is reasonable to attribute the peaks at 1633 and 1493 [cm.sup.-1] to the C=O stretching vibration in Ar-COOH and Ar-[COO.sup.-], respectively. The origin of the other two new absorbances, at 1198 and 1116 [cm.sup.-1], is not as clear, though they, too, may be a consequence of the replacement of an aliphatic substituent on a carboxylic acid or carboxyl ate anion with an aromatic substituent.
The tensile properties for 1:3 TLCP/LDPE blend containing a graft-copolymer compatibilizer at a concentration of 10 parts per hundred parts polymer (php) are given in Table 3 for different graft-copolymer compatibilizers formed from various TLCP/EAA-Na ratios. The composition of the compatibilizer had little effect on the properties of the compatibilized blends. For the remainder of the work reported in this paper, the compatibilizer composition used was 50/50 TLCP/ionomer.
The effect of adding the graft-copolymer compatibilzer Com-50-310-40 on the morphology of fracture surfaces of injection molded bars of (1:3) TLCP/HDPE blends is shown in Fig. 2. Without the compatibilizer, Fig. 2a, gross phase-separation of the TLCP occurred, with domain sizes of the order of ~ 15 [mu]m. The large domain size is typical of high interfacial tension between the two phases, as would be expected in an immiscible, incompatible blend. The TLCP domain size was significantly reduced by the addition of about 5 php compatibilizer, Fig. 2b, and the phase-separated TLCP domains are difficult to resolve in the micrographs. The TLCP domains in the injection molded samples for all TLCP/HDPE blends actually exhibited a fibrillar structure, making it difficult to compare the domain sizes between the uncompatibilized and compatibilized blends. Comparisons of the domain volume were made by annealing the samples at 320[degrees]C for 10 mm to relax the deformed TLCP domains into spherical domains. While this protocol may have promoted further acidolysis reaction, it was assumed that the volume of the spherical domains that were achieved by the annealing represent the volume of the fibrils in the injection molded samples. Further acidolysis reaction, however, between the TLCP and the EAA-Na may have improved the interfacial adhesion in the blends and affected how fracture occurred, e.g., cohesive fracture through the TLCP domains rather than adhesive fracture at the interface of the two phases. As the amount of compatibilizer increased to 10 and 15 php, a very fine microstructure was observed on the fracture surface, see Figs. 2c and 2d, which may be remnants of fracture through the TLCP domains as a consequence of very good adhesion between the TLCP and polyethylene phases.
The effect of a TLCP/ionomer graft-copolymer compatibilizer on the morphology of injection molded (1:3) TLCP/LDPE blends is shown in Fig. 3. Two different compatibilizers, com50-280-40 and com50-330-40, were evaluated, the major difference being that the extrusion reaction was run at 280[degrees]C for the first compatibilizer and 330[degrees]C for the latter one. The size of the TLCP domains decreased significantly with the addition of the compatibilizer, and the reduction of the TLCP domain volume was greater for the compatibilizer prepared at the higher temperature (330[degrees]C vs. 280[degrees]C), which is a consequence of more graft copolymer formation by the acidolysis reaction at the higher temperature (12).
The use of a graft-copolymer compatibilizer improved the notched impact strength of the TLCP/HDPE blends. This is shown in Fig. 4 for the addition of compatibilizer com50-310-40 to (1:3) TLCP/HDPE blends; the addition of 15 php compatibilizer improved the impact strength by ca. 60% compared to that without compatibilizer. Thus, it appears that the graft-copolymer compatibilizer not only exhibits interfacial activity, but it also improves the interfacial adhesion between the TLCP and HDPE phases.
The processing conditions for preparing the graft-copolymer compatibilizer, especially the extrusion temperature, affected its efficacy as a compatibilizer. This is shown in Fig. 5, in which the impact strength of compatibilized (1:3) TLCP/LDPE, each containing 10 php compatibilizer is plotted against the temperature at which the acidolysis reaction was carried out in the extruder. In all cases, the extrusion screw speed was held fixed at 40 rpm. The impact strength of the blends increased as the extrusion temperature was increased from 280[degrees]C to ca. 300[degrees]C, after which further increases in temperature had no noticeable affect on the impact strength. The improvement in impact strength when the extrusion temperature was increased is consistent with the conclusion that increasing the reaction temperature for preparing the compatibilizer promotes more rapid reaction kinetics and more graft-copolymer formation, and hence, better interfacial activity and interfacial adhesion. When the compatibilizer was prepared at 280[degrees]C, the impact strength of the compatibilized blend was no better than that of the uncompatibilized blend; see Fig. 5. That result is consistent with earlier observations that temperatures of 290[degrees]C or greater were needed to promote the acidolysis reaction (11).
The insensitivity of the impact strength as temperature was increased above 300[degrees]C may indicate either that the graft copolymer formation proceeded to completion in the extruder residence time at 300[degrees]C 50 that increasing the kinetics further without changing the residence time had no effect on the graft copolymer formation or that sufficient graft copolymer formation occurred at 300[degrees]C so that further production of graft copolymer formation contributed little to the compatibilization. An alternative explanation of the trend in Fig. 5 may be that the effect of graft copolymer formation is offset by the increasing difficulty of mixing the components as the TLCP viscosity decreased with temperature.
The effect of the addition of a copolymer on the viscosity of an immiscible blend can be complex (14-16). The viscosity of the blend depends on the melt morphology and the interfacial properties. The addition of a compatibilizer can reduce the dispersed phase size and alter the interfacial adhesion between phases. The effect of adding 10 php of compatibilizers com50-280-40, com5-310-40 and com50-330-40 on the complex melt viscosity of a 1:3 TLCP/LDPE blend at 290[degrees]C is shown in Fig 6. The complex viscosity of the uncompatibilized blend was nearly the same as for LDPE, except at the lowest frequencies where the viscosity of the blend was higher. The enhancement of the viscosity at the lower frequencies also corresponded to an upturn or yielding phenomenon observed with the neat ionomer.
The data in Fig. 6 for the pure components shows that the TLCP viscosity at 290[degrees]C is much greater than that of the polyethylene. When the viscosity ratio of the dispersed to continuous phase is much greater than one, as it was for the uncompatibilized TLCP/LDPE blends, it is very difficult to deform or break up the dispersed phase during mixing. As a consequence, for the uncompatibilized blends, one would expect poorer mixing and a larger and less deformed dispersed TLCP phase. For the compatibilized blends, however, the compatibilizer is expected to be located at the interface between the LDPE and TLCP and entangle with both polymers. That reduces or eliminates the importance of the difference between the viscosities of the continuous and dispersed components, and should facilitate dispersion and breakup of the dispersed phase.
The complex viscosity of the blend increased significantly with the addition of the compatibilizers, which is probably due to a combination of the higher viscosity of the compatibilizer itself and the increased interfacial adhesion between the polyethylene and TLCP phases as a consequence of entanglements between the graft-copolymer compatibilizer and the two homopolymers. At low frequency, the compatibilized blends exhibited a yield stress, which is consistent with improved interfacial adhesion between the phases that improves stress transfer and essentially acts as a virtual crosslink between the two phases. The effect of the different compatibilizers on the melt viscosity at high frequency was similar. At lower frequencies, where the yielding behavior was observed, the viscosity increased with increasing temperature for the formation of the compatibilizer, i.e., from 280[degrees]C to 330[degrees]C.
The tensile stress-strain data in Figs. 7 and 8 contrast the properties of blends of TLCP, LDPE and ionomer for different mixing procedures. The overall composition of the blends were identical for all of the curves in Figs. 7 and 8. However, for the blends in Fig. 7, EAA-Na ionomer was dry blended with TLCP and LDPE (5/30/75 by wt) and the compatibilizer was formed in situ during extrusion of the ternary mixture. In contrast, for Fig. 8. a 50/50 TLCP/EAA-Na graft-copolymer compatibilizer was prepared separately and then added to a (1:3) TLCP/LDPE blend. The use of the preformed compatibilizer enhanced the toughness (as measured by the area under the stress-strain curve) over what was achieved by the in situ formation of the compatibilizer. Part of this enhancement may be due to the fact that the two extrusions used, i.e., one to prepare the compatibilizer and one to mix the compatibilizer with the blend, provided additional residence time for the acidolysis reaction, so that more graft copolymer could be for med by the latter method.
The in situ approach produced a higher tensile modulus, which was due to the formation of an oriented TLCP microfibrilar phase, which appeared to be favored by poorer interfacial adhesion (12). That observation is contrary to the finding of O'Donnell and Baird (17) that for TLCP/polypropylene blends, a lower interfacial tension improved the fibrillation of the TLCP phase, which increased the modulus of the blend. However, other factors such as the viscosity ratio and the stress affect the ability to deform the dispersed phase, and these may account for the different observations for the two systems.
Figures 7 and 8 also indicate that the toughness of the blend can be improved by using a slower screw speed either during the separate reactive mixing step used to form the graft copolymer or during in situ formation of the compatibilizer. A slower screw speed results in longer residence time in the extruder, which in either process promotes more formation of the graft copolymer and, therefore, improved compatibilization of the TLCP/LDPE blend.
TLCP-ionomer graft copolymer compatibilizers prepared by reactive extrusion effectively reduced the interfacial tension between a TLCP and polyethylene (LDPE or HDPE) phases leading to improvements in elongation and toughness of TLCP/polyethylene blends. The efficiency of the graft copolymers as compatibilizers for TLCP and polyethylene improved when the extrusion reaction was run at higher temperatures. The TLCP/EAA-Na ratio used to prepare the compatibilizer had little effect on the properties of the compatibilized blends. Although the compatibilizer can be prepared in situ by mixing the EAA-Na directly to a mixture of TLCP and polyethylene and processing the ternary mixture above 290[degrees]C, better toughness and elongation of the final blends were attained by pre-reacting the TLCP and EAA-NA and then mixing the resulting graft-copolymer product with the blend of interest.
[FIGURE 1 OMITTED]
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
Table 1 Properties of Polymers Used. Sample Tg Designation Manufacturer Grade ([degrees]C)1 TLCP Ticona Vectra[TM] A950 100 Ionomer Exxon Chemical Co. Iotek[TM] 8000 LDPE Dow Chemical Co. 530A HDPE Union Carbide DMDC 6146 Sample Tm Melt Flow Rate Designation ([degrees]C) (1) (g/10 min.) TLCP 273 99 (2) Ionomer 82 1.9 (3) LDPE 111 2.78 (3) HDPE 126 0.16 (3) (1)DSC (heating rate = 20[degrees]C/min) (2)290[degrees]C; 2.06 kg (3)200[degrees]C; 2.06 kg Table 2 Compatibilizer Notation. Extrusion Screw Composition (wt%) Temperature Speed Compatibilizer ([degrees]C) (rpm) TLCP EAA-NA Com50-280-40 280 40 50 50 Com50-290-40 290 40 50 50 Com50-300-40 300 40 50 50 Com50-310-40 310 40 50 50 Com50-320-40 320 40 50 50 Com50-330-40 330 40 50 50 Com50-310-20 310 20 50 50 Com50-310-60 310 60 50 50 Com50-310-80 310 80 50 50 Com10-310-40 310 40 10 90 Com20-310-40 310 40 20 80 Com25-310-40 310 40 25 75 Com35-310-40 310 40 35 65 Table 3 Tensile Strength and Elongationat TLCP/HDPE/Compatibilizer (25/75/10) Blends. Tensile Tensile Elongation Modulus Strength at Break Compatibilizer (GPa) (MPa) (%) Com10-310-40 1.33 [+ or -] 0.04 46.5 [+ or -] 0.3 7.9 [+ or -] 0.5 Com20-310-40 1.34 [+ or -] 0.05 47.8 [+ or -] 0.5 7.6 [+ or -] 0.4 Com25-310-40 1.28 [+ or -] 0.05 47.8 [+ or -] 0.5 9.0 [+ or -] 0.5 Com35-310-40 1.27 [+ or -] 0.02 47.3 [+ or -] 0.3 8.4 [+ or -] 0.2 Com50-310-40 1.24 [+ or -] 0.02 45.9 [+ or -] 1.1 6.6 [+ or -] 0.5
This work was supported by grants from the Polymer Division of the National Science Foundation (DMR 9712194) and from Connecticut Innovations, Inc. (99CT007).
(1.) D. Dutta, H. Fruitwala, A. Kohil. and R. A. Weiss, Polym. Eng. Sci., 30, 1005 (1990).
(2.) G. T. Pawlikowski, D. Dutta, and R. A. Weiss, Annu. Rev. Mater. Sci., 21, 159 (1991).
(3.) F. P. LaMantia and A. Valenza, Makromol. Chem., Makromol. Symp., 56, 151 (1992).
(4.) A. A. Handlos and D. G. Baird, J. Macromol. Chem. Phys.--Rev. Macromol. Chem. Phys., C35(2), 183 (1995).
(5.) D. Aceimo and F. P. La Mantia, Processing and Properties of Liquid Crystalline Polymers and LCP Based Blends, ChemTec Publishing, Toronto, 1993.
(6.) Liquid Crystalline Polymers, National Academy Press. Washington, D.C., 1990.
(7.) W. C. Lee and A. F. Dibenedetto, Polym. Eng. Sci., 32, 400 (1992).
(8.) S. M. Hong, S. S. Hwang. Y. Seo, I. J. Chung, and K. U. Kim, Polym. Eng. Sci., 37, 646 (1997).
(9.) J. Y. Lee, S. M. Hong, S. S. Hwang, Y. Seo, and K. U. Kim, Int. Polym. Process., 12, 19 (1997).
(10.) R. A. Weiss, Y. Ghebremeskel, and L. Charbonneau, Polymer, 41, 3471 (2000).
(11.) H. Zhang, R. A. Weiss, J. E. Kuder, and D. Cangiano, Polymer, 41, 3069 (2000).
(12.) Y. Son and R. A. Weiss, Polym. Eng. Sci., 41, 329 (2001).
(13.) H. C. Kim, K. H. Nam, and W. H. Jo, Polymer, 34, 4043 (1993).
(14.) B. Brahimi, A. Ait-Kadi, A. Ajji, R. Jerome, and R. Fayt. J. Rheol., 35, 1068 (1991).
(15.) M. Bousmina, P. Bataille, S. Sapieha, and H. P. Schreiber, J. Rheol., 39, 499 (1995).
(16.) S. Kutsumizu, H. Hara, H. Tachino, K. Shimabayashi, and S. Yano, Macromolecules, 32, 6340 (1999).
(17.) H. J. O'Donnell and D. G. Baird, Polymer, 36, 3113 (1995).
|Printer friendly Cite/link Email Feedback|
|Author:||Son, Younggon; Weiss, R.A.|
|Publication:||Polymer Engineering and Science|
|Date:||Jun 1, 2002|
|Previous Article:||Morphology development during blending of immiscible polymers in screw extruders.|
|Next Article:||Mechanical properties of thermoplastic polyurethanes containing aliphatic polycarbonate soft segments with different chemical structures.|