Exfoliation and dispersion enhancement in polypropylene nanocomposites by in-situ melt phase ultrasonication.
The class of materials known as nanocomposites has enjoyed increased interest since the initial development of a nylon-based material by Usuki et al. at Toyota in 1993 (1). In general, polymer nanocomposites combine an organic polymer with an inorganic layered silicate or smectite clay (in the work of Usuki et al., the thermoplastic resin nylon 6 and a montmorillonite clay were used). Layered silicates are made up of several hundred thin platelet layers stacked into orderly packets or particles. Each of the individual, disk-shaped platelets is of very large aspect ratio (L/D on the order of 100-1000). In the Toyota work, when the clay was dispersed homogeneously and exfoliated as individual platelets throughout the polymer matrix, dramatic increases in strength, modulus, and heat distortion temperature were observed at very low filler loadings (<10% by weight), a result of the large surface area contact between polymer and filler (1). The nanocomposites were produced by intercalation of caprolactam monomers into the layered silicate galleries followed by in-situ polymerization of the monomers. In order to enhance monomer intercalation and compatibility between phases, the layered silicate was first rendered organophilic by cation exchange with ammonium ions of 12-aminolauric acid. While melt compounding of nylons with organically modified clays (also known as nanoclays) has also been attempted, the mechanical properties and degree of clay dispersion and exfoliation are slightly short of those of the in-situ polymerized type (2). Efforts to generate similar nanocomposites using other types of thermoplastics and thermosets have enjoyed varying degrees of success (3-21).
Because of their desirable combination of properties, nanocomposites have many applications in the automotive industry. Plastic-based materials are continually finding new uses in a variety of automotive components. In general, plastics have certain advantages over metals, including dent/ding resistance, lighter weight, increased design flexibility, and corrosion resistance. Furthermore, studies have shown that nanocomposite materials could provide considerable increases in fuel economy. According to Vaia and Giannelis (22) (and the references therein), reductions in vehicle weight due to the widespread usage of nanocomposites could result in fuel savings of 1.5 billion liters of gasoline over a single year of vehicle production. Automotive structural applications have traditionally been made from continuous glass mat composites and highly filled plastic materials such as sheet molding compound (SMC) where the polymeric component can be as little as 15% by weight. Both SMC and glass mat thermoplastic (GMT) materials, however, are still relatively high in density. Trim and semi-structural components, on the other hand, are commonly fabricated from injection moldable thermoplastics and thermosets. These lighter-weight composites, such as short fiber and mineral filled thermoplastics, could be substituted for metals or SMC and GMT composites in the same applications if their mechanical properties could meet the more stringent requirements. Virtually all bumper fascias and air-intake manifolds have transitioned from metallic to plastic construction. As new high-performance plastics and composites are developed, the transition will also encompass both more structural components, as well as Class A (a level of surface smoothness required of exterior surfaces) body panels and high-heat underhood applications.
Injection moldable thermoplastics have traditionally been reinforced by the addition of particulate and fiber fillers in order to improve mechanical properties such as stiffness, dimensional stability, and temperature resistance. Typical fillers include chopped glass fiber and talc, which are added at loadings of 20%-40% to obtain significant mechanical reinforcement. At these loading levels, however, low-temperature impact performance and material toughness are sacrificed. This is an area where nanocomposites have an advantage. Because of the small size yet large aspect ratio of the nanoclay fillers, an extremely low loading is required for the same level of reinforcement, resulting in a robust material with good impact strength.
Because of the polar nature of layered silicate clays, attempts to generate nanocomposites in nonpolar polyolefin matrices have been only marginally successful. Many groups have attempted melt compounding of polypropylene and polyethylene nanocomposites by adding maleic anhydride-grafted polypropylene oligomers (PP-MA) to aid in compatibilization and dispersion (6, 10, 13, 14, 23, 24). While this strategy is effective in improving nanoclay exfoliation, it requires almost 25% PP-MA for significant exfoliation, which is costly and has the deleterious effect of softening the matrix. To circumvent this issue, a few groups have also attempted intercalation of olefin monomers and in-situ polymerization to generate polyolefin-silicate nanocomposites (9, 25-27). In 1996, Tudor et al. attempted in-situ polypropylene polymerization with a Ziegler-Natta catalyst, which produced oligomers, but did not succeed in producing an intercalated or exfoliated structure because of catalyst instability (25). In 1999, Giannelis and co-workers were able to generate an exfoliated polyethylene by in-situ polymerization with a new class of catalyst (9). More recently, two groups have been successful in the synthesis of polypropylene nanocomposites by in-situ polymerization. In 2001, Ma et al. (26) generated several PP nanocomposites at various nanoclay loadings with a Ziegler-Natta catalyst. Following this in 2002, Sun and Garces (27), from the Dow Chemical Company, developed a novel method for in-situ polymerization of PP nanocomposites with the use of metallocene-type catalysts. In both types of polymerization, the specific nanoclay loadings were obtained through controlling the polymerization time. Although much of the nanocomposite samples had regions of well-exfoliated nanoclay platelets, both groups also observed some regions in which the polymerization conditions were unable to break up all the nanoclay aggregates. Dramatic increases in Young's modulus were found by both groups for the in-situ polymerized PP nanocomposites. As is the case for nylon 6 nanocomposites, the in-situ polymerized PP nanocomposites show superior properties to those generated by conventional melt compounding. This is most likely due to the increased exfoliation present in the materials generated by the former method.
In the current work, we have focused on the development of polypropylene-silicate nanocomposites by inexpensive and direct methods. Of particular interest is the development of processing methods that are independent of the specific polymer and filler types. For example, previous studies (28, 29) indicate that the use of supercritical fluids is effective in the pre-exfoliation of layered nanoclays for use in melt compounding into a polymer matrix. By utilizing this method, the addition of compatibilizers was unnecessary for the exfoliation of platelets throughout the matrix. Furthermore, the study indicated that supercritical fluids could also be effective for in-situ melt processing/compounding of polymer-clay nanocomposites, independent of the material system. In the current study, the effect of melt compounding with in-situ ultrasonication has been studied for various polypropylene-montmorillonite clay nanocomposite systems. At elevated temperature, the additional ultrasonic energy enhances the dispersion and exfoliation of the silicate layers throughout the polymer matrix. Typically, ultrasonication has been used in liquid processing applications in chemical and biological systems for the emulsification, dispersion, acceleration of reactions, or disruption of cellular structures. Aside from ultrasonic joining, the use of ultrasound has been limited in polymeric systems. For processing of polymer nanocomposites, ultrasonication has been used to disperse nanoclays in low-viscosity monomers prior to in-situ polymerization, as well as for the direct dispersion of clay and or fiber fillers prior to melt mixing (30-35). At present, only two groups have studied the application of ultrasonic energy to a high molecular weight polymeric melt for the generation of nanocomposite materials (36-39). Ryu et al. found that ultrasonication was effective for the copolymerization of polystyrene and polypropylene by generating free radicals of styrene and PP macroradicals. Furthermore, they noted that the process was also useful for the dispersion of clay throughout the polymer matrix and improved the mechanical properties of the resulting nanocomposite. Here we also show that ultrasonication can be an effective means of enhancing clay dispersion and exfoliation in polypropylene homopolymer nanocomposites without the added benefit of mechanical agitation or shear. Furthermore, the effect of polypropylene molecular weight has been studied in a homologous series, as well as the effects of compatibilizer and supercritical fluid pretreatment of the clay. The optimization of ultrasonic processing conditions, continuous processing with shear, and polymer degradation are beyond the scope of the current study and are the subjects of current research in our laboratory.
The layered silicates used in this study were montmorillonite clays that were cation exchanged with various alkyl ammonium salts. Cloisite 20A (referred to as 20A) was obtained from Southern Clay Products, Inc. and 1.30E was obtained from Nanocor, Inc. The alkyl ammonium cation exchanged montmorillonite clays are referred to as nanoclays. The Southern Clay Products' nanoclay is cation exchanged in the presence of excess ammonium salt, while the Nanocor nanoclays are rinsed of excess salts and purified. Specifications for the nanoclays can be found in Table 1.
Three commercially available polypropylenes (PP) of various melt flow indices and molecular weights were also studied. The PPs (denoted as Ph020, PP6523, and PP6823) were obtained from Basell Polyolefins and have molecular weight ranges and melt flow rates as shown in Table 2. Also shown in the table are properties of a maleic anhydride-grafted polypropylene, Polybond 3200 from Basell Polyolefins (denoted as PP-MA), which was used as a compatibilizer in some of the nanocomposite formulations. The anhydride content of the PP-MA is also shown in the table in terms of KOH equivalents (40, 41).
Ultrasonication. Ultrasonication was performed using a Sonics and Materials, Inc., 60 Watt ultrasonicator with a 1/2" diameter horn, operating at a frequency of 20 kHz. The required energy input was determined by preliminary ultrasonication experiments on a nanoclay-mineral oil model system at ambient temperature, which was subsequently examined by shear rheology and X-ray diffraction (XRD) measurements. Although the viscosity range of the mineral oil suspensions was significantly lower than that of the polypropylene melt, we chose this system as a model because of the similar interactions between the polar clay surfaces and the nonpolar matrix. Polymer nanocomposite samples were ultrasonicated with the same apparatus. Precompounded pellets (processed by conventional means) of PP/nanoclay mixtures (nominal 1 g samples) were placed in a Teflon cup and inserted into a heating block. The heating was maintained at 180[degrees]C-200[degrees]C during ultrasonication of the sample to ensure that the sample remained in the melt phase. A nitrogen blanket was also used during the heating and ultrasonication process, limiting the extent of oxidation of the polymer.
Rheology. Steady shear rheology on mineral oil samples was performed on a TA Instruments CS[L.sup.2] 500 Carri-Med rheometer. Parallel plate geometry was used with 20-mm aluminum plates set at a gap of 0.5 mm. Plates were temperature controlled at 20[degrees]C. Because of the extreme thixotropy of the ultrasonicated mineral oil/clay suspensions, care was taken to load the suspensions in an identical manner to ensure similar shear history for each sample. Viscosity measurements were taken from steady shear experiments at shear rates between 10 and 500 [s.sup.-1].
Processing. Conventional processing of the polymer nanocomposites was performed by a number of methods, including single-screw and twin-screw extrusion, and batch mixing. Single-screw and twin-screw extrusion were carried out at 200[degrees]C on a Haake Rheocord 90 unit. Post-extrusion processing consisted of solidifying the extrudate in a water bath and cutting in a pelletizer. Batch mixing was performed on the Haake system with a Model 600 mixer operating at 200[degrees]C and 70 rpm for 1-2 minutes. Samples containing supercritical fluid-treated nanoclay were processed by batch mixing. All others were processed by single or twin screw extrusion, as noted. Composite mixtures and nanocomposite samples had nanoclay loadings of 4%-5% by weight; those with compatibilizer were generated at a loading of 5% PP-MA by weight, unless otherwise noted.
Sample plaques and sheets were prepared by compression molding at 200[degrees]C and 25,000 psi for plaques and 5000 psi for sheets. Samples were quenched in a water bath for several minutes. Plaques were molded to nominal 3 mm thickness between steel plates with an aluminum mold, while thin films were prepared between Teflon sheets. Injection molded test bars were prepared on either a Boy 80M (80 ton) injection molding machine or a Mini-jector Wasp model #55 mini injection molding machine at 205[degrees]C-210[degrees]C and 5.2 MPa injection pressure. Clamping in the Mini-jector is accomplished by the injection pressure force exerted on the V-mold, and is not controlled independently. Mold temperature was controlled to 24[degrees]C on the Boy 80M system, and uncontrolled at ambient conditions for the Mini-jector.
XRD. X-ray diffraction (XRD) measurements were taken on powdered nanoclays, nanoclay-mineral oil gels, neat PPs, and nanocomposite samples to evaluate the nanoclay basal spacing and polypropylene crystallization. The degrees of nanoclay dispersion, polymer intercalation, and exfoliation in the PP matrix subsequent to compounding were evaluated. Neat polymer and conventional nanocomposite samples were examined as compression molded thin sheets or injection molded bars. Ultrasonicated nanocomposite samples were examined as compression molded thin sheets only. Powdered nanoclay samples were mounted on zero background quartz holders with Vaseline. Nanoclay-mineral oil gels were also mounted on zero background quartz holders. XRD measurements were performed on a Scintag X2 theta-theta powder diffractometer using Cu Ka radiation and a Si(Li) energy dispersive detector. The diffractometer was set up in Bragg-Brentano geometry. Samples were run with a beam divergence of 0.17[degrees].
TEM. Degrees of nanoclay dispersion, intercalation, and exfoliation in nanocomposite samples were also evaluated by means of transmission electron microscopy (TEM). TEM was performed on a JEOL 2000 microscope operating at 200 kV. Cryo-microtomed sections were cut at -120[degrees]C to approximately 50 nm in thickness. Sections were stained for approximately 30 minutes using Ru[O.sub.4] to stabilize samples from the electron beam.
The amount of ultrasonic energy required for dispersion of clay in the polymer melt was first estimated by gelation experiments with suspensions of 4 wt% 20A nanoclay in mineral oil. As can be seen in Fig. 1, XRD measurements indicate that an increase in interlayer spacing occurs as the input ultrasonic energy is increased for the model mineral oil samples. The samples are labeled by total input energy/power (rate of energy input)/additives. In one of the mineral oil-nanoclay mixtures, dodecyltrimethyl ammonium bromide (DTMA Br) was present as an additive. For the unsonicated sample, labeled 0 kJ, a distinct peak is seen in the XRD spectrum, indicating a base nanoclay layer spacing of 24 [Angstrom]. As the ultrasonic energy is increased, the peak broadens and continues to shift towards lower 2[theta], implying increased interlayer spacing. At the same time, an increase in the ultrasonic energy leads to gelation, or the formation of a Bingham fluid with shear-dependent yield stress, of the mineral oil-clay suspension. Note that in the sample that received 20 kJ of energy at 100% with added DTMA Br alkyl ammonium salt (labeled 20 kJ/100%/excess salt), the presence of salt is indicated by the very distinct crystal peaks.
Steady shear rheology of the same nanoclay-mineral oil mixtures was also studied. Rheometric measurements show an increase in the viscosity as well as enhanced shear thinning for samples with larger input energies and excess amounts of DTMA Br alkyl ammonium salt (see Fig. 2). Steady shear viscosities at 300 [s.sup.-1] were determined from the measurements in Fig. 2, and compared as a function of total ultrasonic energy. The results (see Fig. 3) show an increase in viscosity as the total energy input is increased. Flocculation of the discotic clay platelets is responsible for the gel-like structure that develops in clay suspensions (42), so the formation of a gel is a good indication of clay disorder (30). There are several theories, however, on the flocculation structure and interactions of the clay platelets that lead to a strong gel structure, including edge-to-face, edge-to-edge, and house-of-cards, among others ((42-44) and the references therein). Furthermore, a dramatic increase in viscosity was observed for the sample containing excess salt. In aqueous clay suspensions, salt concentration is known to play an important role in the resulting gel structure (42). Here, in an organic suspension, we observe that the presence of excess salt further strengthens the gel.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
I.30E/PP6523 Base Case: Effect of In-situ Ultrasonication
Based on the mineral oil experiments, and because of the extremely high melt viscosity of PP, a high input ultrasonic energy of 40 kJ was chosen for preliminary experiments with the nanocomposites (nominal sample size of 1 gram). The XRD pattern of the I.30E nanoclay (examined in powdered form) can be seen in the lower part of Fig. 4. The well-defined peak at 2[theta] ~ 3.8[degrees] indicates a (001) basal spacing of the nanoclay galleries of 23.1 [Angstrom], with a second order peak present at ~8.3[degrees]. Figure 4 also shows the XRD pattern of neat PP6523 (from an injection molded sample), which has crystalline peaks at approximately 14.1[degrees], 16.8[degrees], and 18.5[degrees] (corresponding to 6.3 [Angstrom], 5.3 [Angstrom], and 4.8 [Angstrom], respectively) (45). Conventional extrusion of the nanoclay and PP leads to two incompatible phases with large particles of nanoclay, which consist of many platelet layers. In this case, there is very little intercalation and negligible exfoliation of the nanoclay platelets. The pattern shown at the top of Fig. 4 is that of the PP6523/I.30E composite processed by conventional twin-screw extrusion in which the crystalline PP peaks agree well with those found in the neat samples, and a slight shift to larger interlayer spacing (25.6 [Angstrom]) is found for the nanoclay. The relative intensities of the crystalline PP peaks differ slightly, but may be attributed to uncontrolled variables during the molding process (e.g., orientation of the polymer chains). Upon ultrasonication, the peak indicating interlayer spacing of the nanoclay shifts to lower 2[theta], giving a spacing of 31.7 [Angstrom], as shown in Fig. 5. A second-order peak of the nanoclay spacing appears at 5.3[degrees]. The XRD spectrum of the unsonicated sample is again shown with the ultrasonicated sample for direct comparison.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
As XRD measurements give only an indication of the structure or order of the material, TEM analysis was also used to supply complementary information about the degree of dispersion, intercalation, and foremost the exfoliation of the nanocomposites. Figure 6a shows a typical micrograph of a PP6523 /I.30E sample at 68 kX magnification prior to ultrasonication. Large particles consisting of many platelet layers can be clearly seen. The dispersion is poor and there is little intercalation or exfoliation. This micrograph is a very localized picture, but is representative of the overall material. Figure 6b shows a micrograph under similar magnification of a typical sample of the same nanocomposite after ultrasonic processing. On average, the overall particle size is much smaller, and there is notable intercalation, as inferred by the spacing between platelet layers. One also observes several tactoids (small stacks of three or four platelet layers (44, 46)) as well as many single platelets in the PP6523 matrix. Again, this local image is illustrative of the overall material.
Effect of Compatibilizer
The effect of adding maleic anhydride-grafted polypropylene (PP-MA) as a compatibilizer between the nanoclay and PP phases, and the effect of ultrasonication on the compatibilized materials, were examined. Samples with 0%, 5%, and 10% loadings of PP-MA were compounded with Ph020 and 5% 20A in a single screw extruder. The mechanical properties were evaluated as a function of compatibilizer loading and the results are shown in Table 3. It can also be clearly noted in the table that the addition of 5 wt% of PP-MA to neat PP softens the overall material significantly (by approximately 18%). This softening effect was also noted by Alexandre and Dubois (and the references therein) (47). On the other hand, the addition of PP-MA improves the mechanical properties by enhancing intercalation and exfoliation of the nanoclay platelets in the PP matrix (10, 13, 14, 23). As implied by Table 3, there is an optimum amount of PP-MA that can be added in order to address these competing effects. This is in contrast to the results of Hasegawa et al., who show that the stiffness continues to increase for increasing PP-MA loading up to 21.6% (10). Usuki and co-workers (10, 13, 14) also show that the acid content of PP-MA plays a significant role in the compatibilization, so our results may be explained by the differences in PP-MA grade used. Ultrasonication of the nanocomposites in the presence of PP-MA was also examined. Figure 7 shows XRD spectra of PP6523/I.30E with 5% PP-MA compatibilizer prior and subsequent to melt ultrasonication. The comparison of the peak locations shows that the change in layer spacing in the presence of PP-MA is enhanced from 27.5[Angstrom] to 35.5[Angstrom] upon ultrasonication. Furthermore, there is a significant reduction in peak intensity, implying that there is an increased disordering of the nanoclay layers. Recall that individual exfoliated platelets show no signal in XRD. Referring back to Fig. 5, we see that an uncompatibilized sample of PP6523/I.30E (prior to ultrasonication) shows not only a tighter initial interlayer spacing, but also a smaller change in the spacing (after ultrasonication) than for the compatibilized PP-MA sample in Fig. 7. These data are presented in Table 4.
[FIGURE 6A OMITTED]
[FIGURE 6B OMITTED]
TEM analysis of the samples containing 5% PP-MA compatibilizer provides complementary data to the XRD measurements, as seen in Figs. 8a and b. Figure 8a shows a TEM micrograph of PP6523/5% PP-MA/5% I.30E following conventional processing in a twin-screw extruder. The presence of PP-MA enhances the dispersion of nanoclay tactoids throughout the polymer matrix (compared to a conventionally prepared sample without PP-MA in Fig. 6a). The micrograph in Fig. 8b shows a region in the ultrasonicated sample. The nanoclay loading appears to be higher in the ultrasonicated sample even though both samples are of identical composition. This is due to the exfoliation of the nanoclay layers in the ultrasonicated sample versus more tightly layered tactoids in the conventionally processed sample. Note that the nanoclay tactoids are much less defined in the sample after ultrasonic processing (Fig. 8b). Many exfoliated. Single platelets are also present throughout the matrix. These images of the nanocomposites are in agreement with the XRD analysis shown in Fig. 7.
[FIGURE 7 OMITTED]
Effect of PP Molecular Weight/Viscosity
Three PPs of varying molecular weight and viscosity were conventionally compounded with 20A nanoclay (in the presence of 5% PP-MA compatibilizer), in order to examine molecular weight effects. Table 5 shows the flexural moduli for the three neat PP grades as well as their corresponding nanocomposites. Surprisingly, the mechanical properties do not change monotonically with the molecular weight or viscosity. The nanocomposite generated from the mid-range molecular weight/viscosity PP, PP6523, shows the greatest percentage increase in the flexural properties. Owing to this nonmonotonality in the conventionally processed nanocomposites, the dependence of PP molecular weight and viscosity on the effectiveness of in-situ sonication was examined.
[FIGURE 8A OMITTED]
[FIGURE 8B OMITTED]
Precompounded samples of I.30E/PP (no PP-MA compatibilizer) for all three molecular weights were ultrasonicated under the same conditions with an input energy of 40 kJ in the system described previously. For the low molecular weight PP, Ph020, there is a dramatic increase in the interlayer spacing due to ultrasonication. The shift in the peak location indicates that the interlayer spacing of the nanoclay in Ph020 increases from 23.6 [Angstrom] to 36 [Angstrom] after ultrasonication. For the high molecular weight PP, PP6823, the effect of in-situ ultrasonication is more modest, shifting the interlayer spacing from 35 [Angstrom] to 37 [Angstrom]. While the absolute spacing indicates significant intercalation, the change due to ultrasonic processing is small for PP6823. Comparing to the effect in the mid-molecular weight sample, PP6523 (see Fig. 5), one observes that the nanoclay interlayer spacing increases from 23 [Angstrom] to 31.7 [Angstrom] after ultrasonication. The comparison of the three molecular weight samples shows that although the conventionally compounded samples prior to ultrasonication display differing degrees of intercalation, the change after ultrasonication is more dramatic in the low molecular weight/low viscosity sample, Ph020. One also observes that the shape of the peaks in the Ph020 sample is much more diffuse and broadened than for the higher molecular weight samples. These results for ultrasonication are in contrast to the molecular weight scaling observed in the measurement of mechanical properties of the conventionally processed materials. Table 4 shows a summary of the results from the molecular weight studies.
Prior to ultrasonication, large tactoids with tightly packed layers are present. After ultrasonication, however, TEM evaluation reveals that the tactoids are broken up to a much smaller size and the nanoclay layers are significantly more disordered. The resulting interlayer spacings estimated from TEM evaluation agree well with the values found by XRD measurements. The same trends were observed with the PP6823/I.30E samples. A comparison of all the data for the three molecular weight ranges reveals that the ultrasonication is most effective in the dispersion, intercalation, and exfoliation of the low molecular weight PP, Ph020.
Effect of Nanoclay Pre-Treatment
Finally, the effect of pretreatment (28, 29) of the nanoclays with supercritical carbon dioxide (denoted as SCF) was examined in PP6523 prior to and after ultrasonication. The nanoclay interlayer spacing was found to increase as a result of melt ultrasonication of a PP6523/PP-MA/I.30E sample in which the I.30E nanoclay was treated with SCF using the method described previously by two of the authors (28, 29). In the asmolded sample, the interlayer spacing is measured to be 29 [Angstrom], and shifts to 36.5 [Angstrom] as a result of the ultrasonication. Similar results are seen for the pretreated nanoclay (SCF I.30E) even without the addition of PP-MA compatibilizer. XRD data are presented in Table 4.
The values in Table 4 together with TEM analysis show that the ultrasonic technique is more effective with the use of SCF pretreated nanoclay over the base case of untreated PP6523/I.30E. The SCF pretreated nanoclay yields enhanced exfoliation over the untreated nanoclay prior to ultrasonication. The pretreatment step does not, however, provide an advantage over the untreated nanoclay when combined with ultrasonication.
CONCLUSIONS AND DISCUSSION
This study has shown that in-situ ultrasonication of the polymer melt phase is an effective method to enhance the dispersion, intercalation, and exfoliation of nanoclays in thermoplastic-based nanocomposites. Three polypropylene homopolymers of increasing molecular weight and viscosity were studied, as were the effects of compatibilizer addition and nanoclay pretreatment with SCF.
All samples showed some increase in nanoclay interlayer spacing as a result of in-situ ultrasonic processing. The effect of molecular weight in the conventionally processed samples was significant and non-monotonic. Furthermore, the interlayer spacings that were determined by XRD analysis do not correlate with the mechanical properties measured for the same samples. Ultrasonic processing technique, however, was found to be more effective in dispersing and exfoliating nanoclays in lower molecular weight polypropylenes. This may be due to increased diffusivity of the lower molecular weight species at the same melt temperature. Additionally, the effectiveness decreased with increasing molecular weight and viscosity of the polymer. Mechanical property measurements, however, have not yet been measured for the ultrasonically processed samples because of the limited sample size produced during batch processing. Further optimization of ultrasonic processing conditions and scale-up are currently under way in our laboratory. It is likely that the dispersion and exfoliation of the nanoclay will be further enhanced upon optimization of the processing conditions, e.g., higher ultrasonic energy input, higher temperature, or simultaneous shear mixing. Once optimal conditions have been found, scale-up studies are also planned.
It must be pointed out, however, that there are important limitations to the methods of characterizing exfoliation. Neither XRD nor TEM alone can give a full picture of the nanoscale dispersion, intercalation, and exfoliation that exist in the nanocomposite samples. If the sample is not uniform, many different regions within a single sample must be analyzed using TEM to understand the global structure. XRD, on the other hand, can give a global average of the degree of intercalation, but is not able to provide quantitative data about the degree exfoliation or disorder. In other words, XRD can detect if structure is present, but exfoliation can be inferred only by the absence of information (i.e., basal peaks). One must consider TEM results in complement with the XRD analysis (48).
Compatibilizer and SCF pretreated nanoclay were found to be equally receptive to ultrasonic treatment for improving exfoliation over the base case. This implies that the ultrasonic processing technique could be a less expensive method for achieving a well-exfoliated sample with exceptional mechanical properties. The use of expensive compatibilizers and the organic modification and/or SCF pre-treatments of clays may not be necessary for some polymer-clay systems. Furthermore, although only polypropylenes were studied in this work, the ultrasonic processing technique should be applicable to any polymer-layered silicate nanocomposite system.
These results indicate that in-situ melt ultrasonic processing may be an effective means of producing a well-dispersed and well-exfoliated nanocomposite material. The technique has been demonstrated in a small-scale batch process without the utilizing supplementary mechanical mixing. Scale-up of the process in conjunction with a continuous process such as extrusion or injection molding that provides additional shear mixing should improve results.
Table 1. Specifications of Alkyl Ammonium Exchanged Nanoclays. Layered Silicate Organic Modifier Cloisite 20A (20A) dimethyl dihydrogenated tallow ammonium ion *(C[H.sub.3])[.sub.2](HT)[.sub.2][N.sup.+] 1.30E stearyl amine [C.sub.18][H.sub.37]N[H.sub.2] Layered Silicate Cation Conc. Basal Spacing Cloisite 20A (20A) 95 meq/100 g 24.5 [Angstrom] 1.30E 119 meq/100 g 23 [Angstrom] *Where HT is hydrogenated tallow, which consists of ~65% C18, ~30% C16, ~5% C14. Table 2. Specifications of Polypropylenes. Melt Flow Rate* Molecular KOH Polymer (g/10 min) Weight Range Equivalents Ph020 37 low -- PP6523 4 medium -- PP6823 0.5 high -- PP-MA 110 [double dagger] very low 0.43 wt% * Melt flow conditions 230[degrees]C/2.16 g. [double dagger] Melt flow conditions 190[degrees]C/2.16 g. Table 3. Effect of Compatibilizer, PP-MA, on Mechanical Properties of Nanocomposites. Nanoclay PP-MA Flex Modulus/ Sample (wt%) (wt%) Std. Dev (MPa) Ph020 0 0 1340/22 Ph020/5% PP-MA 0 5 1109/88 Ph020/5% 20A 5 0 1273/18 Ph020/5% PP-MA/5% 20A 5 5 1912/90 Ph020/10% PP-MA/5% 20A 5 10 1621/92 Table 4. Summary of Nanoclay Interlayer Spacing Results. Spacing Prior to PP Matrix Nanoclay Ultrasonication ([Angstrom]) Ph020 (low mol. wt) I.30E 23.6 PP6523 (med. mol. wt) I.30E 25.6 PP6823 (high mol. wt) I.30E 35 PP6523/PP-MA I.30E 27.5 PP6523/PP-MA SCF I.30E 29 Spacing After % Change in PP Matrix Ultrasonication Spacing ([Angstrom]) Ph020 (low mol. wt) 36 53% (12.4 [Angstrom]) PP6523 (med. mol. wt) 31.7 24% (6.1 [Angstrom]) PP6823 (high mol. wt) 37 6% (2 [Angstrom]) PP6523/PP-MA 35.5 29% (8 [Angstrom]) PP6523/PP-MA 36.5 26% (7.5 [Angstrom]) Table 5. Flexural Modulus Percent Increase From Neat PP to Nanocomposite as a Function of Molecular Weight. Molecular Flexural Modulus (MPa) Sample Weight Neat 5% 20A % Increase Ph020 low 1340 1912 42.7 PP6523 medium 1277 1968 54.1 PP6823 high 1466 1478 0.82
The authors are grateful to Ms. Gail Cunningham for the cryo-microtoming sections and preparing samples for TEM analysis.
1. A. Usuki, Y. Kojima, M. Kawasumi, A. Okada, Y. Fukushima, T. Kurauchi, and O. Kamigaito, J. Mater. Res., 8, 1179 (1993).
2. J. W. Cho and D. R. Paul, Polymer, 42, 1083 (2001).
3. P. B. Messersmith and E. P. Giannelis, Chem. Mater., 6, 1719 (1994).
4. R. A. Vaia, K. D. Jandt, E. J. Kramer, and E. P. Giannelis, Macromolecules, 28, 8080 (1995).
5. Z. Wang and T. J. Pinnavaia, Chem. Mater., 10, 3769 (1998).
6. A. Usuki, M. Kato, A. Okada, and T. Kurauchi, J. Appl. Polym. Sci., 63, 137 (1997).
7. R. A. Vaia, H. Ishii, and E. P. Giannelis, Chem. Mater., 5, 1694 (1993).
8. M. W. Weimer, H. Chen, E. P. Giannelis, and D. Y. Sogah, J. Am. Chem. Soc., 121, 1615 (1999).
9. J. S. Bergman, H. Chen, E. P. Giannelis, M. G. Thomas, and G. W. Coates, Chem. Comm, 21, 2179 (1999).
10. N. Hasegawa, M. Kawasumi, M. Kato, A. Usuki, and A. Okada, J. Appl. Polym. Sci., 67, 87 (1998).
11. Y. Hu, L. Song, J. Xu, L. Yang, Z. Chen, and W. Fan, Colloid Polym. Sci. 279, 819 (2001).
12. G. Jimenez, N. Ogata, H. Kawai, and T. Ogihara, J. Appl. Polym. Sci., 64, 2211 (1997).
13. M. Kato, A. Usuki, and A. Okada, J. Appl. Polym. Sci., 66, 1781 (1997).
14. M. Kawasumi, N. Hasegawa, M. Kato, A. Usuki, and A. Okada, Macromolecules, 30, 6333 (1997).
15. M. Kawasumi, N. Hasegawa, A. Usuki, and A. Okada, Mater. Sci. Engng., 6, 135 (1998).
16. Y. Kurokawa, H. Yasuda, and A. Oya, J. Mater. Sci. Lett., 15, 1481 (1996).
17. J. Lee and E. P. Giannelis, ACS Polym. Preprints, 38, 688 (1997).
18. S. Lee, J. S. Park, and H. Lee, Proceedings of the ACS, PMSE, 83, 417 (2000).
19. E. Manias, A. Touny, L. Wu, B. Lu, K. Strawhecker, J. W. Gilman, and T. C. Chung, Proceedings of the ACS, PMSE, 82, 282 (2000).
20. E. Manias, A. Touny, L. Wu, K. Strawhecker, B. Lu, and T. C. Chung, Chem. Mater., 13, 3516 (2001).
21. G. Qian, J. W. Cho, and T. Lan, Proceedings of the Polyolefins Conference (2001).
22. R. A. Vaia and E. P. Giannelis, MRS Bulletin, 26, 394 (2001).
23. G. Galgali, C. Ramesh, and A. Lele, Macromolecules, 34, 852 (2001).
24. D. Marchant and K. Jayaraman, Industrial & Engineering Chemistry Research, 41, 6402 (2002).
25. J. Tudor, L. Willington, D. O'Hare, and B. Royan, Chem. Comm., 17, 2031 (1996).
26. J. Ma, Z. Qi, and Y. Hu, J. Appl. Polym. Sci., 82, 3611 (2001).
27. T. Sun and J. M. Garces, Adv. Mater., 14, 128 (2002).
28. E. C. Lee and D. F. Mielewski, Ford Technical Report SRR-2000-0262 (2000).
29. C. W. Manke, E. Gulari, D. F. Mielewski, and E. C. Lee, U.S. Patent 6,469,073 B1 (2002).
30. Y. Zhong and S.-Q. Wang. J. Rheol., 47, 483 (2003).
31. Y. K. Kim, A. F. Lewis, P. K. Patra, S. B. Warner, and P. Calvert, National Textile Center Research Briefs-Materials Competency, June (2002).
32. J. G. Ryu, H. S. Kim, and J. W. Lee, Proceedings, SPE ANTEC Conference, 2 (2002).
33. S. T. Lim, H. J. Choi, and M. S. Jhon, J. Ind. Eng. Chem., 9, 51 (2003).
34. G. D. Barber, R. F. Storey, and R. B. Moore, ACS Polymer Preprints, 148, 768 (1999).
35. N. Artzi, Y. Nir, M. Narkis, and A. Siegmann, J. Polymer Science, Part B: Polymer Physics, 40, 1741 (2002).
36. E. C. Lee and D. F. Mielewski, Ford Technical Report, SRR-2001-0179 (2001).
37. E. C. Lee and D. F. Mielewski, U.S. Patent Application No. 60/347,536 (2002).
38. J. G. Ryu, P. S. Lee, H. Kim, and J. W. Lee, Korea-Australia Rheology Journal, 13, 61 (2001).
39. J. G. Ryu, P. S. Lee, H. S. Kim, and J. W. Lee, Proceedings, SPE ANTEC Conference, 2135 (2001).
40. M. J. Solomon, A. S. Almusallam, K. F. Seefeldt, A. Somwangthanaroj, and P. Varadan, Macromolecules, 34, 1864 (2001).
41. M. Sclavons, V. Carlier, B. D. Roover, P. Franquinet, J. Devaux, and R. Legras, J. Appl. Polym. Sci., 62, 1205 (1996).
42. P. F. Luckham and S. Rossi, Advances in Colloid and Interface Science, 82, 43 (1999).
43. F. Pignon, J.-M. Piau, and A. Magnin, Phys. Rev. Lett., 76, 4857 (1996).
44. F. Pignon, A. Magnin, J.-M. Piau, B. Cabane, P. Lindner, and O. Diat, Phys. Rev. E, 56, 3281 (1997).
45. G. Natta and P. Corradini, Del Nuovo Cimento, Vol. 15, Series 10 (1960)
46. R. Krishnamoorti and K. Yurekli, Current Opinion in Colloid & Interface Science, 6, 464 (2001).
47. M. Alexandre and P. Dubois, Mater. Sci. Engng. R: Reports, 28, 1 (2000).
48. A. B. Morgan, J. W. Gilman, and C. L. Jackson, Proceedings of the ACS, PMSE, 82, 270 (2000).
E. C. LEE (1*), D. F. MIELEWSKI (1), and R. J. BAIRD (2[dagger])
(1) Materials Science Department
(2) Physics Department
Ford Research Laboratory
Ford Motor Company
2101 Village Road, MD3182-SRL, Dearborn, MI 48124
*To whom correspondence should be addressed. E-mail: email@example.com
([dagger])Present address: Department of Chemistry, Wayne State University, Detroit, MI 48202-3489.
|Printer friendly Cite/link Email Feedback|
|Author:||Lee, E.C.; Mielewski, D.F.; Baird, R.J.|
|Publication:||Polymer Engineering and Science|
|Date:||Sep 1, 2004|
|Previous Article:||A low-cost, low-fiber-breakage, injection molding process for long sisal fiber reinforced polypropylene.|
|Next Article:||Pre-necking and post-necking relaxation and creep behavior of polycarbonate: a phenomenological study.|