Self-reinforced polypropylene/LCP extruded strands and their moldings.
Self-reinforced composites involving LCPs have many advantages over conventional glass fiber reinforced composites, such as lower energy consumption in melt blending, less machine abrasion, etc. Such thermoplastic/LOP composites have received substantial interest, and many studies have been done in the past decade. A great deal of effort regarding thermoplastic/LCP blends has also been made in the past decade (1-5). Among early researchers, Isayev has acquired extensive patents on blends of LOP with thermoplastics (6-10), another LCP (11), an elastomer (12) and thermoplastic/LCP laminates (13-16) and making web (17) and injection moldings of pre-generated thermoplastic/LCP and LCP/LCP strands (18).
There are two essential steps with regard to the processing of self-reinforced, or in situ, composites: 1) the dispersion of the LCP phase in the thermoplastic matrix as fine droplets, and 2) the extension of these droplets into microfibrils with a high aspect ratio in a strong flow field. The enhancement of thermoplastic /LCP blends arises from the formation of an oriented microfibrillar LCP phase in the solidified blend under the strong flow. The size, shape, and distribution of the LCP domains in the matrix, which determine the ultimate performance of the composites, depend upon the LCP content, processing conditions, viscosity ratio of the components. interfacial adhesion between the components, and the rheological characteristics of the matrix (1-5, 19). Many experimental studies have been done on self-reinforced composites, in particular on the conditions for LCP fibrillation, since this is the crucial step in their successful fabrication. The flow field has been discovered to be of principal importance for successful production of thermoplastic/LCP blends among other factors. Simple extensional flows, such as melt drawing, are most efficient to fabricate LOP microfibrils, and the effectiveness increases with draw ratio (1-5, 20-23). LCP content has an effect on the morphology of the LCP phase, and subsequently the mechanical properties of the blend. For instance, in some cases, it is found that a certain minimum LCP content is required for fibril formation (21, 24-28). The effects of viscosity ratio on the morphology of immiscible thermoplastic/LCP blends have been reported by various researchers (29-31). Studies on blends of thermoplastics with LCPs indicate that in order to achieve good fibrillation, the viscosity of the dispersed LOP phase should be lower than that of the matrix (31-33).
Compatibilization is a process by which an incompatible blend is rendered less incompatible, so that the resulting material is useful for engineering purposes. This compatibility does not imply thermodynamical compatibility, in which the polymers exist in a single molecularly blended homogeneous phase (34). Compatibilization can be either physical or chemical in nature, involving partial miscibility, entanglements of polymer chains, physical or chemical interactions. Chemical similarity may lead to partial miscibility, and the presence of functional groups to a reaction, and both of these may increase compatibility. One route to a compatibilized blend is thus to functionalize either blend component or both. A different type of compatibilization involves the addition of a compatibilization agent, or compatibilizer, to act as a bridge between the binary blend components at their interfaces. The action of such compatibilizers, usually block or graft copolymers, is based on the reactivity or miscibility of their segments with at least one of the blend components. They act, in fact, as interfacial agents, since they tend to enrich and act at the interfaces as emulsifiers enhancing the interaction between the blend components through covalent or other types of bonding. The compatibilizers chosen for the current experiments belong to this group. Besides block and graft copolymers, a variety of lower molecular weight reactive chemicals promote copolymer formation or crosslinking reactions and thereby improve compatibility (35, 36).
The usual effects of compatibilization are the reduction in interfacial energy, a stabilized morphology with perhaps a finer dispersion, and an increase in the interfacial adhesion (37). In the case of LCP blends, it should be emphasized that the goal is not miscible blends, but rather, two-phase blends that combine the good properties of the LCP and thermoplastic components. The main requirement for a thermoplastic/LCP blend is the formation of fine oriented LCP fibrils firmly attached in the thermoplastic matrix. Thus, improving the compatibility in most cases means increasing the interfacial adhesion; however, the reduction of interfacial energy and fine dispersion lead to the deterioration of LCP fibrillation. It is of paramount importance to balance these two competing factors to optimize the final performance of the blends.
In the present work, polypropylene (PP) of various molecular weights were extruded with a LCP through a circular die and the melts were drawn to produce strands with unidirectional orientation of dispersed LCP phase. These strands were subsequently pelletized and injection molded at temperature below the melting point of LCP to preserve the generated LCP fibrils. The mechanical properties and their relation to the morphology were investigated as functions of draw ratio, LCP content, and molecular weights of PPs. Also, six different polymers were selected for studying the compatibilization of the PP/LCP blends.
MATERIALS AND EXPERIMENTAL
Six grades of isotactic PPs were used for studying the dependence upon molecular weight of the properties of PP / LCP strands and their moldings. These grades included Profax 6823, 6723 and 6523 and Valtec H444, HH441 and PH910S (Himont USA Inc., Table 1). PP6823, PP6723 and PP6523 are simply different melt flow grades of homopolymer PP with a general purpose stabilization package. HH444 and HH441 are stabilized reactor spheres coated with a primary antioxidant and an acid neutralizing agent. PH910S is also in the form of reactor spheres, but with minimal stabilization consisting of less than 100 ppm of a primary antioxidant. The LCP used was Vectra A950 (Hoechst-Celanese), a random copolyester of 73 mol% of p-hydroxybenzoic acid (HBA) and 27 mol% of 2,6-hydroxy naphthoic acid (HNA). Several polymers, suggested by the manufacturers to be used as compatibilizers, were selected. The information concerning the compatibilizers is listed in Table 2.
The materials were dried at 100[degrees]C in a vacuum oven overnight. Blends of various compositions were all prepared using a one-inch single-screw extruder (Killion Inc.) followed by a six-element static mixer (Koch Industries, Inc.). The screw speed was operated at 70 RPM, and the temperature was maintained at 300[degrees]C for the whole system except at the feeding zone (260[degrees]C). A 90[degrees] angle-fitting die with 10-mm diameter was attached to the static mixer for strand drawing processing. The extrudate from the die was drawn by a take-up device, and then subsequently quenched in a water bath. Different take-up speeds in the range of about 7.5 to 60 cm/s were utilized in order to study the effect of drawing. The draw ratio was defined as the ratio of the take-up speed to the exit speed at the die. It was found that at temperature above the melting point of LCP, the viscosities of PPs Valtec HH444, HH441 and PH910S became so low because of their low molecular weights and high melt flow indices t hat it was not possible to obtain usable samples. Thus, no results from specimens containing these three PPs were included in the results section.
The PP/LCP strands were chopped into pellets using a pelletizer and dried at 100[degrees]C in a vacuum oven overnight to remove residual moisture. Injection molding was performed using a Boy 15S machine. The temperature was chosen at 200[degrees]C to minimize the deterioration of LCP fibrillar structures. Dumbbell-shaped specimens of the size 6.35 cm x 0.30 cm X 0.15 cm (mini-tensile bar, MTB) were prepared for tensile testing. The tensile properties were measured using an Instron mechanical tester (Model 4204) at room temperature and at a crosshead speed of 5 mm/min. All tests were carried out with a 1-KN load cell without an extensiometer. For each sample, at least five specimens were tested. To determine the LCP morphology, samples were first fractured in the direction perpendicular to the flow cryogenically in liquid nitrogen, coated with gold-palladium alloy and then examined by means of scanning electron microscopy (SEM, Hitachi S-2150).
RESULTS AND DISCUSSION
Mechanical Properties and Morphology of PP/LCP Strands
Figure 1 shows the variation of modulus and tensile strength of PP6723/LCP strands at several compositions as a function of draw ratio. Tensile strength here refers to the yield stress for the samples where yielding took place and the stress at break for the samples where breaks occurred without yielding. Typically, the 50/50 PP/LCP blends did not yield and simply broke as brittle materials at elongations in the range of 3.0% to 3.5%. The 75/25 blends first yielded at elongations in the range of 3.8% to 4.3%, then broke later at a higher elongation. The 87.5/12.5 blends first yielded at elongations in the range of 5.6% to 7.6%, and then necking occurred, followed by breakage. Drawing had very little effect on the pure PP strands, but when LCP was added into the blends, the tensile properties were improved by drawing. The more LCP added, the better was the reinforcement by drawing. For example, at 87.5/12.5, the modulus and tensile strength of the PP6823/LCP strand increased from 1.0 GPa and 34.7 MPa to 1.5 GP a and 37.9 MPa between the draw ratios of 6.7 and 22.2. However, between the draw ratios of 5.7 and 19.4, the modulus and tensile strength of the 50/50 PP6823/LCP strand were improved from 3.0 GPa and 70.1 MPa to 4.9 GPa and 106.9 MPa. The mechanical performance of the blends was closely related to the LCP morphology. The reinforcement of composites was generated from the alignment of the fibrils with high length-to-diameter ratio, l/d. As will be seen in the SEM micrographs, the size of dispersed LCP droplets in the extruded samples was enlarged by the LCP concentration, and the larger the droplets, the easier it was to achieve the elongation and alignment of the droplets by the stretching of the extrudates. That is why drawing improved the tensile properties of thermoplastic/LCP composites more effectively when the LCP concentration was sufficiently high. Drawing had little influence on 87.5/12.5 PP6523/LCP strands, but began to enhance the tensile properties when LOP increased to 25 wt%. A similar phenomen on was observed in the film extrusion process (38). PP6523 had the lowest molecular weight matrix with the lowest viscosity among Profax PPs used in the experiments. The viscosity ratio, [[eta].sub.LCP]/[[eta].sub.matrix], was largest when PP6523 was used as the matrix. The viscosity ratio is one of the most important factors determining the droplet dispersion and deformation in shear and elongational flows. As shown below, the higher the matrix viscosity, the smaller the viscosity ratio, the smaller the dispersed particle size, and the greater the deformation of the particle. This conclusion was also drawn in the film casting process of PP/LCP blends (38).
Figure 2 illustrates the effects of LCP concentration on the tensile properties of the PP/LCP strands at a draw ratio of 24. Note that the data plotted in Fig. 2 is obtained by interpolating the experimental results shown in Fig. 1. Similar data (not shown) were plotted at a draw ratio of 12. The tensile strength and modulus of PP6823/LCP and PP6723/LCP strands at a draw ratio of 12 showed continuous increases with an increase in LCP concentration. The tensile strength of PP6523/LCP strands did not show any improvement at 12.5 wt% of LCP, but started to rise as the LCP content continued to increase. The modulus of PP/LCP strands containing 12.5 wt% of LCP showed only little change, and then was greatly improved as the amount of LCP increased. Notice that the tensile properties, especially the tensile strength, of PP/LCP 50/50 strands were much greater than those of the corresponding prepregs (drawn films or sheets) at the same draw ratios (38). For instance, PP6523/LCP 50/50 strand had a modulus of 3.7 GPa an d a tensile strength of 77.6 MPa at a draw ratio of 12, but the prepreg of the same composition had a modulus of only 3.2 GPa and a tensile strength of 34.1 MPa at the same draw ratio. We offer some experimental observations to explain this phenomenon. It is well known that based on the principle of minimum viscous dissipation, it is expected that the less-viscous fluid will migrate toward the wall of the die, with the more-viscous fluid pushed to the center of the die, in order to minimize the energy of viscous dissipation (39,40). In the current experiments, the LCP, being the less-viscous fluid, was expected to move toward the die wall. Comparing the geometry of the coat-hanger die used for film casting and the circular die used for fiber spinning, the coat-hanger die had a much larger surface coming in contact with the melt during processing. Observing the samples of films and fibers of 50/50 composition, we found that the fibers had a very smooth surface, while the film surface was usually rough with lon g pieces of LCP dispersed in it. We suspected that in the case of fiber spinning, because of the smaller contact surface area in the die, the LCP was able to encapsulate the more-viscous polypropylene. But in the case of films, the LCP was incapable of wrapping the polypropylene in the center completely because of the much larger surface area in the coat-hanger die. This was supported by the observation mentioned earlier. We have also observed (38) that it was difficult to prepare films at 50 wt% of LCP because of the tremendous viscosity differences and the poor interfacial adhesion between the LCP and polypropylene phases. When the film was subjected to stretching, we usually found defects at the interfaces between the LCP and PP phases on the film surfaces. On the other hand, this phenomenon noticed in film casting was not seen in fiber spinning, and the strand surface was much less defective. The tensile properties vs. LCP content at the draw ratio of 24 also showed similar patterns, as shown in Fig. 2.
Figure 3 illustrates the modulus of PP/LCP strands as a function of LCP volume fraction at draw ratio of 24. Similar plots (not shown) were obtained at draw ratio of 12. Clearly, at draw ratios of 12 and 24, the moduli of PP6823/LCP and PP6723/LCP strands are, respectively, below and above those predicted by the rule of mixture. The change of 50/50 PP6723/LCP modulus was the most significant. This figure suggests that drawing not only increased the aspect ratio of the LCP droplet, but also increased the modulus and tensile strength of the LCP phase.
The relations between the matrix molecular weight and the tensile properties are plotted in Fig. 4. Both the modulus and tensile strength of pure PP strands were independent of the molecular weight. Tensile strength of PP/LCP blends increased with matrix molecular weight at draw ratios of 12 (not shown) and 24. The modulus also increased with matrix molecular weight. The decline of PP6823/LCP 50/50 modulus at draw ratio of 18 observed in the case of films (38) was also seen here in the fibers at a draw ratio of 12.
Although the skin-core morphology was seen in all strand samples, their effect on mechanical properties was not investigated in the present study. The presence of the skin-core structure in the sample is seen in Fig. 5 as an example. During the extrusion, the shear stress is highest at the wall, so therefore the rate of viscous dissipation per unit length of tube would be smaller when the less-viscous component of the two wetted the tube wall, surrounding the more-viscous component staying in the core. Since the LCP was the less-viscous fluid in this case, it would migrate toward the wall during the extrusion. LCPs were often formed from highly aromatic stiff monomers, so they were readily elongated. Thus, we observed a more fibrillated morphology in the skin section than in the core. This distinction was more evident at higher LCP concentration. The skin-core structure in thermoplastic/LCP blends is very well documented in the literature (1-6). Notice that all the SEM micrographs presented in the following d iscussion were taken from the skin area.
Figure 6 shows the morphology of 87.2/ 12.5 PP/LCP strands. Typically, without drawing, the LCP was dispersed in the matrix as spherical droplets. As drawing became stronger, the spherical droplets deformed into elongated fibrils aligned along the drawing direction. This morphological transformation was responsible for the reinforcement seen in the films (38) and observed here in the strands at high draw ratios. On the other hand, there were few changes in the LCP phase suspended in the PP6523 matrix because of the low viscosity of PP6523, which explained why the modulus and tensile strength of PP6523/LCP strands did not vary with the draw ratios at this low LCP concentration.
The morphology of 75/25 PP/LCP strands is shown in Fig. 7. The particle size was larger because of the increase in LCP composition. At 25 wt% of LCP, microfibril formation was seen in all three PP matrices, which corresponded to the tensile property enhancement of the strands with drawing. Strands containing a higher molecular weight matrix had better properties because the high molecular weight matrix generated finer dispersions. At 50 wt% of LCP (Fig. 8), the LCP became highly oriented, by drawing with long fibrils with high aspect ratios. The strands were significantly reinforced by this morphology and showed high tensile properties.
Mechanical Properties and Morphology of Injection-Molded PP/LCP Composites
The injection-molded specimens were made from pellets prepared by chopping the PP/LCP strands. In order to preserve the LCP microfibrils created in the fiber drawing process, the processing temperature chosen in injection molding was much lower than the melting temperature of the LCP phase. The modulus and tensile properties of injection-molded samples of PP6723/LCP blends are shown in Fig. 9. The modulus and tensile strength of original strands are also plotted for comparison purposes. The tremendous improvement of tensile properties with drawing in the case of strands was not observed here in the injection-molded samples. Instead, only very limited increases, if any, were seen in the injection-molded samples. At 12.5 wt% of LCP, the modulus of PP/LCP samples was about the same magnitude as that of the strands at low draw ratios, but unlike the modulus of the strands, the modulus of injection-molded samples increased only very slightly with drawing. At 25 wt% of LCP, the modulus and tensile strength of PP/LC P samples showed a great deal of decline compared to the corresponding strands. As the LCP concentration increased to 50 wt%, the decline of modulus and tensile strength from strands to injection-molded specimens was even more severe. For example, at 25 wt% of LCP, the modulus and tensile strength of PP6723/LCP composites fell off by 50.3% and 13.8%, respectively, at a draw ratio of 24.8. However, at 50 wt% of LCP, the deterioration of modulus and tensile strength increased to 72.1% and 68.4%, respectively, at a draw ratio of 24.3. Sabol et al. (41) reported similar conclusions. Their studies of the use of pelletized pre-generated strands to form composites through injection molding and sheet extrusion also led to a less than ideal morphology for mechanical property enhancement. They suggested that these processes caused the LOP fibrils formed in the strand drawing process to agglomerate and deform, leading to poor mechanical reinforcement of the matrix.
Figure 10 illustrates the effects of LCP content on the tensile properties at a draw ratio of 24. The modulus increased with LCP concentration, as in the case of films (38) and strands. However, the modulus increased only modestly when LCP increased from 25 to 50 wt%, in contrast to great improvements observed in films (38) and strands. The tensile strength behaved in quite a different way from what we have seen so far. At a draw ratio of 12, the tensile strength of PP6823/LCP and PP6723/LCP samples increased with LCP content, reached a maximum at around 20 wt% of LCP, and then deteriorated drastically at 50 wt% of LCP. The tensile strength of PP6523/LCP first decreased, and then climbed back up, only to drop again. Similar patterns were seen at a draw ratio of 12 (not shown). For the purpose of comparison, we also prepared injection-molded samples made from mixtures of PP and LCP pellets at several compositions. The results are plotted in Fig. 11. The modulus and tensile strength at each composition were alm ost identical regardless of the polypropylene molecular weight. This suggests that the high molecular weight polypropylenes were unable to produce a better LCP dispersion and fibrillation than low molecular weight polypropylenes in the flow field of the injection-molding process. But on the other hand, the injection-molding machine (Boy 15S) used in the experiment was equipped with a screw without a mixing section, and therefore, it was possible that the LCP was not fully dispersed and fibrillated.
Comparing Fig. 10 and Fig. 11, we find out that at 12.5 wt% of LCP, the injection-molded samples made from pelletized strands had lower tensile properties than those from mixtures of PP and LCP pellets. At 50 wt% of LCP, samples made from mixtures of PP and LCP pellets had moduli of about 3 GPA and tensile strengths of about 70 MPa, while those samples made from pelletized strands had moduli of about 2.3 GPA and tensile strengths of about 42 MPa at best. However, Isayev (18) and Sabol et al. (41) have reported that products made using certain systems of pre-generated fibers show much better mechanical properties than those made by the direct processing of blends. Figure 12 illustrates the tensile properties vs. the LCP volume fraction. Unlike previous cases, the modulus did not move above the line predicted by the rule of mixtures when the draw ratio increased. Figure 13 shows the same plot for samples injection-molded from PP/LCP pellet mixtures. Comparing Fig. 12 and Fig. 13 with Fig. 3, we find out that in jection molding cannot create samples with modulus and tensile strengths higher than the values predicted by the rule of mixtures, but in previous cases, we have observed that by drawing the strands and films (38), the modulus and tensile strength can rise above the rule of mixtures. This suggests that elongational flows are more effective in enhancing the modulus and tensile strength of the composites in the machine direction than shear flows.
Figure 14 shows the modulus and tensile strength of injection-molded samples made from pelletized strands vs. the matrix molecular weight. While the matrix molecular weight had little influence on the modulus, it increased the tensile strength. Notice the dramatic decline of the tensile strength of 50/50 PP/LCP blends. In some cases, the tensile strength of 50/50 blends even dipped below that of pure PPs.
Figure 15 shows the morphology of injection-molded samples made from a 75/25 PP/LCP pellet mixture. The skin-core morphology was also observed in these specimens. The SEM micrographs were taken from the skin region. The morphology observed here was very different from what we saw in the strands. Owing to the absence of a mixing section in the injection molding screw, the LCP phase was neither well dispersed nor aligned in one direction. At each composition, the morphology was indistinguishable regardless of the matrix molecular weights. We pointed out earlier that the tensile properties of these samples were almost identical and independent of their matrix molecular weights. This observation agreed with the morphology observation. The high molecular weight polypropylene, in this case, was unable to generate fine dispersion and aligned fibrillation.
Figure 16 shows an example of the morphology of injection-molded samples made from pelletized strands. The prime distinction between strands and injection-molded specimens was that the fibril alignment seen in the strands no longer existed after the injection molding. The LCP domains seen in this case became deformed and disoriented. The skin-core morphology in this case was not observed. In the previous section we pointed out that the tensile property decline from strands to injection-molded samples was greater at higher LCP concentration. The explanation in terms of morphology was the following: we mentioned earlier that in the strand drawing, higher LCP concentration reinforced the composites better because at higher concentration, the size of LCP particles became larger, and larger droplets were deformed more easily, and the elongated fibrils had higher aspect ratios. These long, oriented fibrils gave the strands their high modulus and tensile strength. But when the strands were pelletized and then inject ion-molded, these long fibrils were difficult to align in the mainly shear flow fields of injection molding. Longer fibrils were also more easily fractured during further processing. To demonstrate this point, we prepared strand and injection-molded samples for SEM by immersing the samples in boiling xylene at [sim]140[degrees]C until the polypropylene was completely dissolved, so we could observe the size of the LCP phases. The results are shown in Fig. 17. The difference of LCP phases in the strands and in the injection-molded samples was clear: the long fibrils that existed in the strands became much shorter in the injection molded samples. As a matter of fact, we have observed that in the strands containing 50 wt% of LCP, there were fibrils with length much longer than the pellets, which meant that these long fibrils could very likely be fractured even in the pelletization process. All these factors contributed to the decline of modulus and tensile strength from fibers to injection molded samples, and the se factors were intensified at higher LCP concentration.
To further investigate the fibril fractures in the pelletization/injection molding process, we prepared composites through compression molding of randomly aligned fibers. The PP/LCP strands were randomly placed in the mold of the compression molding press, heated up to 178[degrees]C and then consolidated under a force of 35.6 kN for 5 minutes. The results of the measurements of mechanical properties of the compression-molded composites are shown in Table 3. Notice that the tensile strength and modulus of the 50/50 composite were a lot higher than those of injection-molded samples.
We believe that in this case, compression molding was a better alternative because it was able to preserve the optimal properties of the reinforcing phase in the drawing process and then maintained the high aspect ratio of the LCP fibrils during composite formation. Injection molding, on the other hand, might have destroyed the fibrils especially when the fibrils were long. Sabol et al. (41) actually suggested that the compression molding of randomly oriented strands could serve as a tool to predict the limits of properties possible in composites produced from pre-generated strands.
Mechanical Properties and Morphology of Compatibilized PP/LCP Strands and Injection Moldings
Figure 18 represents the tensile strength and modulus of 75/25 PP/LcP strands with 5 wt% of compatibilizers as a function of draw ratios. Improvements in modulus and tensile strength were observed in the blends containing maleic anhydride modified polypropylenes (Polybond 3200, Polybond 3001 and UNITE MP660) and acrylic acid modified polypropylene (Polybond 1001) as compatibilizers. However, the improvement was only modest, and it diminished as the draw ratio increased. For instance, at a draw ratio of 5, a maximum gain of 11.2% in tensile strength was obtained using Polybond 1001, but this gain decreased to 5.6% when the draw ratio increased to 20. The highest gain of 13.5% in modulus was observed in blend with Polybond 3200. At a draw ratio of 20, the modulus gain was down to 5.4%. The addition of compatibilizer reduced the interfacial tension between two incompatible phases and led to a finer dispersion.
But at the same time, smaller droplets were more difficult to deform. We suspected that at low draw ratios, the droplets were not under a lot of force to deform; the improvement in tensile properties was mainly due to the increased interfacial adhesion from the presence of compatibilizers. At high draw ratios, the improvement of blend properties was less because the LCP fibrillation diminished as a result of the reduction in Interfacial energy. The effects of compatibilizer on the modulus were close, but amongst these compatibilizers, Polybond 3200 and Polybond 1001 gave the blends the highest tensile strength. The addition of these compatibilizers resulted in strand diameter fluctuations to take place at lower draw ratio, probably due to the decrease In the blend viscosity.
In contrast, the maleic anhydride modified EPDMs reduced the modulus and tensile strength. The reduction was more drastic in the case of Royaltuf 485, possibly due to its lesser polypropylene composition. The highest draw ratios for these samples were significantly lower than the others; adding modified EPDMs increased the elasticity of the blends considerably, which made the spinning process more difficult.
Figure 19 illustrates the morphology of compatibilized PP/LOP strands. The skin-core morphology was still present. Notice that the SEM micrographs were taken from the skin region. The basic theme was still the same: The LCP phase was dispersed as spherical droplets in the PP matrix when there was no drawing, and then started to elongate into fibrils as drawing stepped up. Comparing the particle size in the extruded samples, it seems that compatibilized blends had smaller particle size, especially in the blends containing Polybond 3001 and UNITE MP660.
Figure 20 shows the tensile strength and modulus of injection-molded samples made from pelletized strands containing compatibilizers. The figure is very similar to its strand counterpart; the modulus and tensile strength of injection-molded samples declined drastically from that of the strands, however, they were still improved by the maleic anhyrdride and acrylic acid modified polypropylenes, but weakened by the maleic anhydride modified EPDMs, when compared with those samples containing no compatibilizers.
O'Donnell and Baird (42) studied the compatibilizing effects of a maleic anhydride grafted polypropylene for blends of polypropylene and several different LCPs, including Vectra A950, by injection molding. They reported that the addition of such compatibilizer led to significant mechanical property improvements in PP/LCP blends. The tensile strength in all the cases that were studied, increased with increasing amount of compatibilizer. However, a possible plateau in the tensile strength of Vectra A blends occurred, and a decrease in the modulus occurred at the highest maleic anhydride polyprophylene concentration (50 wt%). This indicated that unlimited increases in mechanical property with increasing compatibilizer content did occur. On the other hand, significant increases in the strength could occur, and these increases were dependent upon the particular LCP. It did appear that the properties of the LCP with a structure of copoly(ester amide), which could undergo strong hydrogen bonding, were strongly affec ted by the concentration of maleic anhydride polypropylene. The tensile strength of this blend increased without limit in the range of compatibilizer investigated, and the modulus showed large increase as well. They also reported that the addition of compatibilizer in the PP/LCP blends increased the dispersion of the LCP phase, reduced interfacial tension between the phases and enhanced adhesion. The addition of maleic anhydride polypropylene to these PP/LCP blends led to a finer dispersion of LCP within the matrix but with a more fibrillar structure being formed, which led to the reinforcement of the matrix. But as shown in the PP/Vectra A blends, there was a point where mechanical properties diminished, and at this point the morphology appeared to exhibit LCP structure with lower aspect ratios.
Heino and Seppala (43) reported similar results. The tensile strength and stiffness were enhanced by the presence of maleic anhydride grafted polypropylene as compatibilizer. However, the impact strength was not improved. They also conducted experiments of adding EPDM in PP/LCP blend in order to improve the impact behavior. The results were not successful; EPDM rubber did not toughen the PP/LCP blend because the EPDM particles did not locate at the interfaces where they might have formed a bridge between the blend components. The tensile properties were also weakened by the addition of EPDM. They further concluded that a successful compatibilization of incompatible blends depended on the chemical suitability of the compatibilzer for the system (chemical or physical action), the amount and distribution of the modifier, and the viscosity ratios of the blend components. An effective blending was usually required. The morphology of ternary systems was more complex and perhaps more difficult to control. The accomm odation of small amount of compatibilizer just at the interfaces might be especially difficult to achieve.
Figure 21 shows the morphology of compatibilized injection-molded samples made from the pelletized strands. The SEM micrographs were taken from the skin area. The loss of fibril orientation was evident. There is no distinct skin-core morphology observed in this case.
Self-reinforced composites of LCP in matrices of various molecular weight polypropylenes were extruded into strands with different draw ratios and compositions. In general, the modulus and tensile strength increased with draw ratio and the LCP content. The morphological transformation of LCP phase in the strands due to drawing was clearly observed. This reinforcement was intimately connected to their morphological transformation from spherical droplets to oriented, and elongated microfibrils. The blends containing high molecular weight matrices generally had higher properties because their higher viscosity was more effective in generating fine dispersion and fibrillation.
Injection molding was performed using pelletized strands. Though at 12.5 wt% of LCP content, the modulus and tensile strength of injection-molded specimens were close or even slightly higher than that of the original strands, serious deteriorations in tensile properties were observed in the injection-molded specimens containing 50 wt% of LCP. The effect of melt drawing on the original strands diminished considerably after injection molding. The SEM micrographs indicated the LCP fibrils were deformed and disoriented, even fractured in high LCP content, in the injection-molded specimens, and these deformation, disorientation and fracture were responsible for the significant decline in tensile properties of injection-molded specimens. The higher the LCP concentration, the more severe the decline in properties. At 50 wt% of LCP, the properties of injection-molded samples made from pelletized fibers were considerably lower than samples made from mixtures of PP/LCP pellets, but composites made by compression moldin g from randomly aligned strands demonstrated the highest propel-ties. This suggests that compression molding is a better option than injection molding, in particular when dealing with high LCP content blends; the long LCP Microfibrils generated in the melting drawing process were better preserved without breakage. which provided stronger enhancement in the composites. The aligned LCP fibrils generated in the strands were unable to be transferred successfully via injection molding.
Six different polymers were chosen as compatibilizers for PP/LCP strands. Among these polymers, maleic anhydride and acrylic anhydride modified polypropylenes provided some modest improvements in the modulus and tensile strength of the strands, but reduced the highest draw ratio, perhaps due to the reason that the low viscosity of compatibilizing agents promoted the diameter fluctuation during the strand drawing operation. Maleic anhydride modified EPDMs, on the other hand, not only undermined the strands, but also seriously increased the difficulties in the strand drawing by increasing the elasticity of the blends. Microfibril formation was observed in the strands, as well as finer dispersion of LCP.
Injection molding was performed using pelletized compatibilized PP/LCP strands. The injection-molded specimens were considerably weaker compared to the corresponding strands due to the loss of microfibril orientation. But the modulus and tensile strength of injection-molded specimens were still reinforced by the presence of maleic anhydride and acrylic anhychide modified polypropylenes, and weakened by the maleic anhydride modified EPDMs.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
[FIGURE 11 OMITTED]
[FIGURE 12 OMITTED]
[FIGURE 13 OMITTED]
[FIGURE 14 OMITTED]
[FIGURE 18 OMITTED]
[FIGURE 20 OMITTED]
Table 1. The Characteristics of Polypropylenes. * Lot Polydispersity MFR (dg/min) Type Number Index 230[degrees]C/2.16 kg 6823 LK15077 3.9 0.51 6723 LK15026 3.9 0.92 6523 BE37228 4.3 4.1 HH444 LK61306 4.4 73 HH441 LK06817 4.6 440 PH910S LK06323 4.7 748 Type [M.sub.w] from correlation 6823 670,000 6723 557,000 6523 351,000 HH444 144,000 HH441 83,000 PH910S 70,000 * Information provided by Himont. Table 2. The Characteristics of Compatibilizers. Polymer Composition Supplier Polybond 1001 * Acrylic acid modified (6 wt%) Uniroyal Chemicals polypropylene Polybond 3001 * Maleic anhydride modified Uniroyal Chemicals polypropylene (MAH index 0.3) *** Polybond 3200 * Maleic anhydride modified Uniroyal Chemicals polypropylene (MAH index 2.5) *** Royaltuf 465 * Maleic anhydride modified Uniroyal Chemicals ethylene/propylene non-conjugated diene elastomer (Total maleic anhydride/Acid 0.5%, EPDM E/P Ratio 55/45) Royaltuf 485 * Maleic anhydride modified Uniroyal Chemicals ethylene/propylene non-conjugated diene elastomer (Total maleic anhydride/Acid 0.5%, EPDM E/P Ration 75/25) Unite MP660 ** Maleic anhydride grafted (0.1 wt%) polypropylene Aristech Chemicals * Information provided by Uniroyal Chemicals. ** Information provided by Aristech Chemicals. *** The MAH Index is determined via an IR method and is not equivalent ot weight% maleic anhydride. Data provided by Uniroyal Chemicals. Table 3. Tensile Properties of PP6723/LCP Composites Made From Randomly Aligned Fibers. Composition Draw Ratio Modulus (GPa) Tensile Strength (MPa) 87.5/12.5 35.9 1.5 35.3 75/25 30.1 2.2 41.5 50/50 24.3 3.6 69.8
(1.) A. I. Isayev and T. Limtasiri, in The international Encyclopedia of Composites, vol. 3, p. 55, S. M. Lee, ed., VCH Publishers, New York (1990).
(2.) D. Dutta, H. Fruitwala, A. Kohli, and R. A. Weiss, Polym. Eng. Sci., 30, 1005 (1990).
(3.) W. Brostow, Polymer, 31, 979 (1990).
(4.) A. A. Handlos and D. G. Baird, Macromol Reviews, Chem. Phys., C35, 183 (1995).
(5.) A. I. Isayev, in Liquid Crystalline Polymer Systems: Technological Advances, p. 1, A. I. Isayev, T. Kyu and S. Z. D. Cheng, eds., ACS Publishers, Washington D.C., (1996).
(6.) A. I. Isayev and M. J. Modic, U.S. Patent 4,728,698 (1988).
(7.) A. I. Isayev and S. Swaminathan, U.S. Patent 4,835,047 (1989).
(8.) A. I. Isayev, U.S. Patent 5,006,402 (1991).
(9.) A. I. Isayev, U.S. Patent 5,006,403 (1991).
(10.) A. I. lsayev. U.S. Patent 5.283,114 (1994).
(11.) A. I. Isayev and P. R. Subramanian, U.S. Patent 5,070,157 (1991).
(12.) A. I. Isayev, U.S. Patent 5,021,475 (1991).
(13.) A. I. Isayev, Eur. Patent Appl. WO 91/01879 (1991).
(14.) A. I. Isayev, U.S. Patent 5,238,638 (1993).
(15.) A. I. Isayev, U.S. Patent 5,275,877 (1994).
(16.) A. I. Isayev and P. R. Subramanian. U.S. Patent, 5,032,433 (1991).
(17.) A. I. Isayev, U.S. Patent 5,268,225 (1993).
(18.) A. I. Isayev, U.S. Patent 5,260,380 (1993).
(19.) D. G. Baird, T. Sun, D. S. Done, and G. L. Wilkes, J. Thermoplast Compos. Mat, 3(1), 81(1990).
(20.) Q. Lin, J. Jho, and A. F. Yee, Polym. Eng. Sct., 33,789 (1993).
(21.) G. Crevecoeur and G. Groeninckx, Polym. Eng. Sci., 30, 532 (1990).
(22.) D. Dutta, R. A. Weiss, and K. Kristal, Polym. Compos., 13, 394 (1992).
(23.) M. T. Heino and J. V. Seppala, J. Appl. Polym, Sci., 44. 2185 (1992).
(24.) C. U. Ko and G. L. Wilkes, J. Polym, Sci., Phys. Ed., 37, 3063 (1989).
(25.) A. I. Isayev and P. R Subramanian, Polym. Eng. Sct., 32, 85 (1992).
(26.) A. Mehta and A. I. Isayev, Polym, Eng. Sci., 31, 971(1991).
(27.) G. Crevecoeur and G. Groeninckx, J. Appl Polym. Sci., 49, 839 (1993).
(28.) G. Crevecoeur and G. Groeninckx, Polym, Compos., 13, 244 (1992).
(29.) A. I. Isayev and M. J. Modic, Polym. Compos., 8, 158 (1987).
(30.) M. T. Heino, P. T. Hietaoja, T. P. Vainio, and J. V. Seppala, J. Appl. Polym. Sci., 51, 259 (1994).
(31.) D. Beery, S. Kenig, and A. Siegmann, Polym. Eng. Sci., 31, 451 (1991).
(32.) K. G. Blizard and D. G. Baird, Polym. News, 12, 44 (1986).
(33.) D. Done, A. Sukhadia, A. Dutta, and D. G. Baird, SPE ANTEC, 37, 988 (1991).
(34.) A. Y. Coran and R. Patel, Rubber Chem. Technol., 56, 1045 (1983).
(35.) M. Xanthos, Polym. Eng. Sci., 28, 1392 (1988).
(36.) M. Xanthos and S. S. Dagli, Polym. Eng. Sci., 31, 929 (1991).
(37.) R. L. Markham, Compalloy Europe '90. 1 (1990).
(38.) Y. C. Liang and A. I. Isayev, Polym. Compos., 23, No. 5 (2002).
(39.) C. D. Han, Multiphase Flow in Polymer Processing, Academic Press, New York (1981).
(40.) Z. Tadmor and C. G. Gogos, Principles of Polymer Processing, Wiley, New York (1979).
(41.) E. A. Sabol, A. A. Handlos, and D. G. Baird, Polym. Composites, 16, 330 (1995).
(42.) H. J. O'Donnell and D. G. Baird, Polymer, 36, 3113 (1995).
(43.) M. T. Heino and J. V. Seppala, J. Appl. Polym. Sci., 48, 1677 (1993).
A.I. ISAYEN (*)
(*) Corresponding author.
|Printer friendly Cite/link Email Feedback|
|Author:||Liang, Y.C.; Isayen, A.I.|
|Publication:||Polymer Engineering and Science|
|Date:||May 1, 2002|
|Previous Article:||Tensile modulus of polymer nanocomposites.|
|Next Article:||Studies of spinodal decomposition in a ternary polymer-solvent-nonsolvent system.|