Structure and properties of microfibrillar-reinforced composites based on thermoplastic PET/LDPE blends after manufacturing by means of pultrusion.
Pultrusion (Pult) processing is one sector of the composites industry that is steadily growing. Compared with the common composite production processes, it offers the advantage of continuous production of profiles with a constant cross-section. Until now, the process has been quite exclusively developed for thermosetting matrices; however, thermoplastic Pult shows a growth of interest, and considerable efforts have been made over the last decade. This can be explained by certain advantages associated with thermoplastics, e.g., good mechanical performance, high temperature resistance, good chemical resistance, good recyclability, and their ability to be postshaped (1-5).
Nevertheless, some difficulties still remain to achieve a good mingling of the matrix with the reinforcing fibers (6). The process parameters having an important effect on the properties of the pultruded composites were found to be the pulling speed, the preheating conditions, the heating temperature in the die, and the cooling rate (7-10).
The main problem in using thermoplastic matrices lies in the full impregnation of the fiber reinforcement because of the high matrix melt viscosities. To overcome this problem, the distance of the polymer melt flow should be reduced. A great number of semifinished products have been developed, where the matrices and the reinforcing fibers (mainly glass or carbon) are closely mingled, for instance, in preimpregnated tapes, hybrid yarns, or powder-impregnated bundles.
If polymeric fibers are used as reinforcement, the impregnation problem can be easily solved by means of Pult of orientated polymer blends, consisting of polymers with different melting temperature. In the subsequent processing step, when melting of the component with the lower melting point occurs, it is only necessary to ensure that the oriented fibrillar structure of the polymer with the higher melting temperature is preserved.
It should be mentioned here, however, that in the literature no information exists about the processing of polymer-polymer composites, i.e., microfibrillar-reinforced composite (MFC)-structured blends by Pult. The MFC is a new type of polymer composite, satisfying to a great extent the peculiarities of traditional polymer blends and composites, and having reinforcing elements with sizes between those of the micro- and nanoreinforced composites (11-18). Unlike the classical composites (e.g., glass- or carbon-fiber reinforced ones), this new group of polymer composites is reinforced with polymer fibrils or more frequently with bundles of them.
As a matter of fact, the structure of the fibrillized blends is very similar to that of the commingled fibers (parallel-aligned fibrils and bundles of them from blend components), but with much small diameters of the reinforcing polymer fibrils (from 300 nm to 2 [micro]m). Besides, the reinforcement with polymeric fibrils is an alternative to common fiber reinforcements, which has important environmental consequences. In addition, MFC offers other potential advantages: polymer matrix reinforcement by conventional (flexible) thermoplastics, no mineral additives, reduction in weight, increased processability, a better control of matrix crystallization, improved mechanical integrity, and ability for recycling.
The challenge of this study is to adjust the Pult parameters (preheating, die temperature, pulling speed, etc.) of a polymer/polymer material group in such a way that no rupture of the MFC cables (or bristles) occurs before a final consolidation in the Pult die system. Such rupture could be possible due to the noncontinuous and polymeric nature of the reinforcing phase (as compared e.g., to the continuous, nonmelting glass, or carbon fibers in commingled yarns). Expected advantages are to end-up with better recyclable, lighter, less-expensive products, without any remarkable losses in the relevant material properties of the pultruded profile.
This study concentrates on the Pult of poly(ethyleneter-ephthalate) (PET)/low-density polyethylene (LDPE)-orientated bristles for producing MFC-structured rectangular profiles. The starting material for Pult (drawn PET/LDPE bristles) was manufactured according to the MFC concept on an industrially relevant production line (19-21). Pultruded profiles have been manufactured using a self-designed Pult facility. The influence of various processing parameters, such as temperature and pulling speed, on the final properties of the profile was determined. For comparison, additional experiments were also performed on compression (CM) and injection-molded (IM) samples from the same drawn blends.
The Pult experiments have been conducted using orientated bristles of a 50/50 wt% PET/LDPE blend. The PET was received from EASTMAN[R] PLASTICS, "EASTAR Copolyester EN058," Natural, and the LDPE came from Exxon Mobil Lot: 7047 LD152MB BAGNr 3841. The orientated cables were manufactured and characterized (as described in Refs. 19-21). After production, the drawn material was wound up by a winding device.
The glass transition ([T.sub.g]) of PET fraction in the drawn blend is 112[degrees]C (about 40[degrees]C higher than [T.sub.g] of the undrawn one). During scanning in DSC, PET fraction crystallizes and then melts at 256[degrees]C. The melting temperature of the LDPE is 115[degrees]C.
Processing of the Drawn Blend
Pultruded rectangular profiles have been manufactured using the facility available at the IVW GmbH (TU Kai-serslautern, Germany). Figure 1 shows a scheme of the Pult line. From right to left, the drawn PET/LDPE bristles are stored in a creel stand and are then guided into the preheating chamber in which the nominal air temperature was set well above the melting temperature of the LDPE. This does, however, not mean that all the LDPE in the cables get fully molten, since the moving time through this zone is relatively short.
[FIGURE 1 OMITTED]
Directly after preheating, the cables are pulled through the heating die, where the matrix (LDPE) is supposed to fully melt and weld together, so as to completely fill the gaps between the incoming bristles. A final consolidation is supposed to take place in the cooling die. The profile is pulled through the die with the help of a pulling device placed at the end of the Pult line. The length of the preheating zone amounted to 600 mm, and its temperature could be varied. Hot convection air was used alternatively, since the possibility of a contact preheating device could be used, but it was not applied because an overheating could lead to an insufficient strength of the cables.
The die system consists of a heating die and a cooling die, separated by a narrow gap. The heated die cavity is 80 mm in length, and it is tapered at the entrance to facilitate the supply of the incoming bristles and to buildup some internal consolidation pressure profile. The final section geometry is 4 X 10 [mm.sup.2]. The cooling die, placed right after the heating die, usually ensures a good consolidation, using circulating water as a cooling aid. The die consists of a lower and an upper part that are screwed together so that it is possible to vary the height of the cavity. During all the experiments presented in this work, the cooling die was maintained at room temperature. The dies are machined from tool steel, and the inside of their cavities are chrome-platinated to reduce the friction in the die. Seventy-five cables from the material PET/LDPE were required to fill the die cavity. All the cables were preliminary dried in an oven for 10 hr at 80[degrees]C.
The temperatures of the preheating systems as well as heating die were steadily controlled by the use of thermocouples located in the preheating chamber and the heating die. The processing parameters examined in this study were the preheating and heating die temperature as well as the line speed. The Pult parameters used for the preparation of profiles from the materials are given in Table 1.
TABLE 1. Process parameters for pultrusion of PET/LDPE orientated bristles. Sample Preheating Hot die Pulling Yarns Cooling die number temperature temperature speed number temperature ([degrees]C) ([degrees]C) (cm/min) ([degrees]C) Pult 1 170 250 6 75 25 Pult 2 170 240 6 75 25 Pult 3 170 230 6 75 25 Pult 4 170 240 12 75 25 Pult.5 180 250 12 75 25 Pult 6 180 220 18 75 25 Pult 7 180 230 18 75 25 Pult 8 180 240 18 75 25 Pult 9 180 250 18 75 25
For the preparation of CM samples, pultruded bars that exhibited a poor quality, such as bad consolidation, were used. They were cut into 10-cm long pieces and placed in a mold. The bars were dried in an oven for 8 hr at 80[degrees]C before CM. A poly(tetrafluoroethylene) film was used to prevent the samples to stick to the mold. A conventional spray release agent was not used because it could get in-between the individual bars and prevents them to melt or weld together. The mold was put in a hot press (PW 10 Paul Weber), having a temperature of 240[degrees]C, and a pressure of 5 MPa was applied over a period of 5 min to ensure the complete melting of the LDPE. After that, the mold was cooled to room temperature under a constant pressure of 5 MPa. The plates obtained were about 4 mm thick. The latter were then cut into samples similar to the geometry of the bars obtained after Pult: length 10 cm and width 1 cm. Thus, the prepared samples were then tested with regard to their mechanical properties.
For the preparation of the 1M samples, the drawn cables of the PET/LDPE blend were cut into pellets (with a length of 3-4 mm), dried at 80[degrees]C for 8 hr, and subsequently processed by IM, using a Kloeckner Ferromatik, FM 20 machine. The temperature zones from the hopper to the die were set at 165, 175, 185, and 210[degrees]C. Testing samples similar to the Pult ones (length 10 cm, width 1 cm and a thickness of 4 mm) were obtained.
METHODS OF ANALYSIS
Scanning Electron Microscopy
Scanning electron microscopic (SEM) observations were performed on a JEOL JSM 5400 and JSM 5510 microscope, with an acceleration voltage of 10 kV. Samples from each processing methods (IM, CM, and Pult) were immersed in liquid nitrogen and fractured, or split to study their morphology on the respective fracture surfaces. All specimens were coated with a thin gold layer prior to SEM analysis.
Wide-Angle X-ray Scattering
Wide-angle X-ray Scattering (WAXS) patterns of the blends and IM samples were obtained using a Siemens D 500 X-ray machine. Diffraction patterns were registered by means of a photographic film. In this way, information about orientation and crystallization phenomena of the drawn blends, final pultruded, and IM specimens were obtained.
The flexural properties of the pultruded, CM, and IM samples were determined by performing the three-point bending experiments at room temperature. The tests were carried out on a Zwick 1474 Universal Testing Machine. The span between the supports was 64 mm, and the crosshead speed amounted to 5 mm/min. Five specimens from each Pult conditions were tested.
Impact tests with unnotched samples (according to DIN EN ISO 179:2000) were performed at room temperature using a Charpy Impact Testing Machine (CEAST company, Italy). The clamp distance was 71 mm, and the impact speed amounted to 3.7 m/s. Five specimens from each Pult condition were tested.
RESULTS AND DISCUSSION
Investigation of the Processing Parameters
The influence of the preheating temperature was investigated by conducting experiments at increasing pulling speed. Former works have shown the importance of a sufficient matrix preheating on the final impregnation quality (3), (9). Here, the air temperature in the preheating zone was chosen to be 170-180[degrees]C, which corresponds to temperatures of about 50 and 60[degrees]C above the melting point of the LDPE, and about 80 and 70[degrees]C below the melting point of the PET. A sufficient high preheating temperature offers the advantage that the matrix viscosity is low enough, which leads to a decrease of the pulling force in the heating die, and in this way, it also reduces the risk of a rupture of the complete cables in the heating die.
It is important to mention again about the problem encountered when using drawn bristles containing discontinuous PET fibrils. As the LDPE matrix is softening, the drawn cables gradually lose their strength, which can result in the risk mentioned earlier. Therefore, particular attention should be paid to avoid the complete melting of the LDPE in the preheating chamber. The best condition seems to be, when the preheating temperature can be adjusted in such a way, that the surface of the cables just reaches the melting temperature of the LDPE, but the cables themselves still preserve their integrity. A load cell placed in line with the die system permitted the control of the pulling force during the experiments. Although the load cell acted only as a qualitative measure, the following statement can be made.
As long as the load level could be kept on a relatively low, but constant level, the process could be kept stable. But if the load becomes too low, rupture of the cable because of overheating in the preheating zone took place. If the load increased too much because of high-friction in the die system, cables could rupture as a result of mechanical overloading.
Further experiments were systematically performed to determine the influence of pulling speed and heating die temperature on the profile properties. The heating die temperatures were chosen to be 220, 230, 240, or 250[degrees]C (Table 1), which corresponds to temperatures of about 35, 25, 15, or only 5[degrees]C below the melting point of the PET and about 100-130[degrees]C above the melting of the LDPE. The pulling speeds were set at 6, 12, or 18 cm/min, respectively.
It should be noted that the die temperature and the pulling speed have a combined influence on the consolidation quality of the separated bristles. First, the temperature influences the LDPE viscosity and, consequently, the pressure inside the die. An increase in the temperature leads to a decrease of LDPE viscosity, which in turn tends to reduce the pressure. In general, a low viscosity combined with a sufficient high pressure should act favorably on the consolidation of the individual bristles. Second, the pulling speed determines the residence time of the polymer in the die system and, consequently, influences the temperature the polymer can reach, which in turn influences the matrix viscosity and pressure. On the other hand, the high pulling speed can lead to an insufficient consolidation time. However, it should be pointed out that at a low pulling speed the blend partners are exposed to a relatively high temperature over a long time, which can lead to thermal degradation.
Furthermore, also the pulling force encountered in the Pult process should be considered. The different mechanisms contributing to the pulling force are known to be the friction resistance, which refers the friction of the materials against the die wall as well as the viscous resistance, which is the force generated by the shear flow in the thin region between the pultruded mass and the die wall. In the pulling force model (22), it is demonstrated that the viscous and compaction components increase, and, consequently, the pulling force, by increasing pulling speed. On the other hand, the pulling force increases if the die temperature is low, because of a high LDPE matrix viscosity. As already mentioned, a serious increase in the pulling force can cause breaking and pulling out of the PET fibrils and thus be fatal to the process. Finally, all the mentioned reasons explain the difficulty to clearly separate the influence of both pulling speed and die temperature.
Additionally, one should pay attention to the cooling process. In the present case, the cooling temperature was kept constant (25[degrees]C). At low pulling speeds, the profile spent enough time in the cooling die and can solidify over its entire cross-section. However, at a higher pulling speed, only the outer layer of the profile was solidified after leaving the die, while its center cooled down slowly. This problem was not investigated here in detail, but this could also partly explain the difference observed in surface quality. It was found that, at low pulling speed, the pultruded profiles exhibited a relatively smooth surface, like a matrix-rich layer, whereas an increase of the pulling speed finally leads to a profile with a roughened surface.
Figure 2 shows pictures of the row material (the orientated PET/LDPE blend bristles) and the pultruded profiles. It should be mentioned here that the IM and CM specimens possess a smoother surface when compared with the pultruded ones. This is due to the different processing windows of the three processing methods.
[FIGURE 2 OMITTED]
Structural and Morphological Characterization
In contrast to all previous experiments on MFC manufacturing, the isotropic matrix was obtained through annealing, CM, or IM of the drawn blends (11-21); in this present case, this stage was carried out in a different way. Here, the drawn blends were processed via Pult, keeping the processing window close to but below the melting of the higher melting component (PET in the present case). In doing so, one expects to convert the lower melting component (LDPE) into an isotropic matrix reinforced by the PET fibrils. To verify this structure, one has to characterize samples from various manufacturing stages. As demonstrated for many polymer blends used for MFC (11-21), the most appropriate techniques for this purpose are SEM and WAXS. Although the former offers a good idea on the fibrillar morphology, the latter provides information about the presence or absence of orientation of the composite components.
Figures 3-5 show SEM micrographs of the main manufacturing and processing stages of the MFC samples. As expected, in accordance with the previous findings (11-21), after drawing, the blend components are transformed into a highly oriented state (Fig. 3a and b). The SEM observations of the peeled fracture surfaces show very well-orientated PET fibrils with a high aspect ratio. Better information about the length and thickness of the reinforcing PET fibrils one be drawn from PET/LDPE samples, of which the LDPE was selectively extracted (Fig. 3c). An estimation of the thickest and the thinnest PET fibrils leads to values of 1.5 [micro]m and 350 nm.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
The morphological peculiarity of the IM samples is shown in Fig. 4. Subsequent IM of the cut bristles at a temperature, being considerably higher than the melting temperature, [T.sub.m], of LDPE, but still below [T.sub.m] of PET, leads to a melting and converting of the LDPE fraction into an isotropic semicrystalline matrix.
In contrast, the PET is supposed to preserve its fibrillar structure. On the microphotograph with the lowest magnification (Fig. 4a), one can see a well-expressed skin-core structure. Within the skin zone, the orientations of the PET fibrils follows the flow direction (from right to left) and form a layer-like structure (Fig. 4b). A partly transverse orientation of the fibrils can be observed, on the other hand, in the core region (Fig. 4c). At the same time, the shape and thickness of the fibrils are quite different than those of the as-drawn blend (see Fig. 3). This difference is related to the already-discussed structural changes of the PET fibrils (relaxation and shrinkage) during processing. In this case, additional to the effect of temperature, the high injection and postpressure contributed to the changing of the shape of the reinforcing material.
In a previous investigation (20), the so-called phenomenon of transcrystallization was observed in the IM sample by means of transmission electron microscopy (TEM). The TEM micrograph demonstrates the basic difference between the crystalline mass in the bulk of the matrix and the crystalline layers around the fibrils. In the bulk material, i.e., far away from the PET fibrils, the LDPE lamellae with a thickness of about 6-7 nm are dispersed quasi-homogeneously, showing no preferred orientation direction. This situation contrasts drastically to the organization of the crystalline lamellae closest to the PET fibrils. Here, the LDPE-crystalline lamellae with the same thickness are placed parallel to each other and perpendicular to the surface of the PET fibrils, in this way, composing uniformly organized lamellae with around 150-200 nm wide layers. The formation of these transcrystalline layers of LDPE is related to the nucleation effect of the PET fibrils. The nucleation ability of PET components in the blends can also contribute to a change in crystallization rate of the matrix material and, in this way, influence the LDPE structure. In addition, it can be assumed that an important role of transcrystalline layers on the adhesion between the reinforcing elements and the matrix exists.
The SEM observation of the fracture surfaces is quite different in the case of Pult and CM samples. Because of the preservation of the high orientation of the PET fibrils in the Pult and CM samples, it was not possible to prepare fractured samples (with the fracture plane transverse to the fibrils' orientation), even after quenching in liquid nitrogen. On the other hand, breaking the pultruded samples with the fracture plane parallel to the fibril orientation, i.e., splitting the pultruded bars in their length axis, gave an evidence that both samples possess in their well-consolidated regions (which was the complete cross-section of the CM samples and at least the skin structure of most of the Pult samples) uniaxially oriented PET fibrils, which are embedded in an isotropized LDPE matrix (see Fig. 5). As a final result, a unidirectional composite-like structure is observed. This morphology is due to the quasi-static nature of the CM-and Pult-processing methods, in contrast to the IM mode, in which processing occurs in a very dynamic fashion. Only in the core region of some of the Pult samples, not well-consolidated parts could be found, but these details were not photographed.
The results of the SEM studies are in quite good agreement with the WAXS data presented in Fig. 6. It is worth mentioning that all the WAXS patterns were taken at room temperature, first from drawn bristles (Fig. 6a), then after IM (Fig. 6b) and Pul (Fig. 6c). This was necessary to better visualize the crystalline structure and the degree of orientation during the different processing modes. In fact, the original samples that were crystallized during drawing and processing of MFC did not develop their typical degree of crystallinity, particularly not in the case of PET component. In some cases, only an additional crystallization step allowed to derive adequate conclusions regarding the type of crystallization (e.g., oriented vs. isotropic). It also seems to be important to remember that the two components of the present blend (LDPE and PET) represent two significantly different behaviors with respect to the crystallization rate.
[FIGURE 6 OMITTED]
Figure 6a shows the WAXS pattern of the as-drawn PET/LDPE (50/50 by weight) sample without applying any additional thermotreatment. Only the reflections of LDPE can be observed, indicating a very high degree of orientation, with chains strongly aligned along the drawing direction or fiber axis (FA). This suggestion is supported by the well-known fact that LDPE crystallizes much faster than PET.
Figure 6b reproduces the WAXS pattern from the IM sample. Isotropic Debye rings for LDPE can be recognized, demonstrating the isotropic crystallization of LDPE. At the same time, isointensity arcs indicate a more or less isotropic state of the PET fibrils. The observation that after IM PET crystallizes in an isotropic state is also supported by the SEM observation (see Fig. 4), where quite randomly dispersed fibrils can be seen.
After Pult, a completely different blend structure is created (Fig. 6c). The PET fraction in the blend practically preserves its orientation in the FA direction, as evidenced from PET hkl spots being situated on the equator. At the same time, one can see a strong increase of the crystallization and perfection of the PET crystals. The LDPE component in the blend shows a tendency to isotropization after Pult. However, some orientation of LDPE crystallites can be observed because the Debye rings of LDPE are not completely of an isointensity type, as it was the case after IM (Fig. 6a). A possible explanation for this is probably because of the aforementioned (20) phenomenon of trans-crystallization of LDPE on the PET microfibril surfaces. Another case for this observation is due to a lack of melting of the LDPE fraction in some bristles situated in the middle of the profiles. This was assumed from optical observation of the fractured samples.
From the SEM observations (Figs. 3-5) and the WAXS analyses (see Fig. 6), three important conclusions can be drawn: (i) the morphologically well-defined microfibrils of PET after drawing of the blends (see Fig. 5) are in a glassy state; (ii) during IM, the glassy microfibrils lose their uniaxial orientation (in the drawing direction) and are displaced in the isotropic LDPE matrix more or less in various directions (Figs. 4 and 6b); and (iii) after Pult, PET fibrils crystallize and preserve their high orientation in the Pult direction (Fig. 6c). This last observation has very important consequences for the mechanical behavior of the polymer-polymer composites.
The flexural properties of the pultruded profiles were investigated using three-point bending tests. The results are shown in Fig. 7. For comparison, the same figure presents the flexural properties of IM neat LDPE- and PET/LDPE-drawn blends as well as CM samples.
[FIGURE 7 OMITTED]
It can be seen that the flexural modulus and strength of the pultruded samples are 30-35 and about 10-12 times higher than neat LDPE, respectively. The highest modulus and strength exhibited the profiles, which were pultruded in a temperature range of 230-240[degrees]C and at a speed of 6-12 cm/min (Pults 3 and 4). At the same time, these values are comparable to those of the CM sample. Most of the profiles that were processed with speeds of 12-18 cm/min exhibited about 10-15% lower values compared with sample Pults 2-4. The lowest flexural properties possessed samples that were processed at 220[degrees]C (Pult 6), followed by the pultrudates prepared at 250[degrees]C (Pults 1, 5, and 9). This could be related to a poor consolidation of the individual bristles at 220[degrees]C as well as to a partial melting of the PET fibrils mainly on the profile surfaces at 250[degrees]C. Actually, this temperature is in the melting range of PET.
Comparing the flexural properties of PET/LDPE-drawn blends processed by different methods, one can see that the pultruded profiles (see Fig. 7) exhibit much high values of the mechanical properties than the IM samples, i.e., 3.5-4.5 times higher strength and 2.5-3.3 times higher modulus. This is due to a much higher aspect ratio, as well as due to the unidirectional orientation of the PET fibrils in the case of pultruded samples. All these experimental results demonstrate the strong reinforcing effect of the PET fibrils on the LDPE matrix. It should be mentioned here that this effect is much more expressed in the case of Pult and CM processing than in the case of IM.
The results of the Charpy impact tests, performed with samples manufactured in the different ways, are presented in Fig. 8. One can observe an opposite trend, i.e., when the impact energy is compared to the flexural properties; a higher flexural strength corresponds to a lower impact energy. A possible explanation for this behavior was found by optical observation of the fractured samples. As mentioned earlier, it was found that, especially in the middle of the profiles processed with a speed of 18 cm/min (Pults 6-9), several bristles were not melted and consequently not well impregnated, acting in this way as "energy absorbers." These samples exhibited ~25% higher impact energy than the samples with higher flexural properties (Pults 2-6 and CM sample). These experimental results demonstrate the effect of pulling rate and die temperature on the impregnation quality of the samples. In other words, the samples possessing a poor impregnation quality (Pults 6-9) exhibited the highest impact energy (see Figs. 7 and 8). This is related to a greater amount of splitting of these samples during impact perpendicular to the length direction of the pultruded bars. The observation corresponds, in some sense, to the structural design of materials for bulletproofed vests or composite armors, in which an intended poor interfacial quality between the fibers and matrix causes a much greater area of damage absorption. One can also conclude from this observation that the Charpy impact energy can indirectly deliver information about the impregnation quality.
[FIGURE 8 OMITTED]
In this study, the pultruded profiles, using PET/LDPE-orientated bristles, were successfully produced. In the best of our knowledge, this study presents the first successful attempt for pultruding profiles from a drawn polymer blend. Under certain conditions of pulling speed and temperature, the profiles manufactured exhibit mechanical properties that are similar to those of MFC-structured composites, manufactured by CM molding. Both are, however, much superior in their properties than corresponding IM samples. This is due to the different morphology of the samples manufactured with different processing methods: a much higher aspect ratio and a unidirectional orientation of the reinforcing PET fibrils in the case of pultruded and CM samples, compared with the quasi-random orientation of the PET fibrils in the LDPE matrix in the case of the IM samples.
Further work is needed, however, to better control the Pult processing parameters and to utilize the advantages of MFCs in continuous profiles in a very reliable way and to combine high stiffness and strength with good impact properties at the same time.
The authors are grateful for the hospitality at the Institut fur Verbundwerkstoffe GmbH (IVW), University of Kaiserslautern, Germany.
(1.) C. Gensewich, "Pultrusion von Konstruklionswerkstoffen aus nachwachsenden Rohstoffen," in 2nd International Wood and Natural Fiber Composites Symposium. Kassel, Germany (1999).
(2.) M. Karus and M. Kaup, J. Ind. Hemp., 7. 119 (2002).
(3.) G. Bechtold. "Pultrusion von geflochtenen und axial ver-starkten Thermoplast-Halbzeugen und deren zerstorungsfreie Porengehaltsbestimmung," Dissertation, Institut fur Verbund-werkstoffe GmbH, Universitat Kaiserslautern (2000).
(4.) J. Blaurock and J.W. Michaeli, "Pultrusion von endlosfaser-verstaerkten Prorilen mit thermoplastischer Matrix, Proceedings 27," in 27th Internationale AVK-Tagung, Baden-Baden. Germany, B20-1-B20-8 (1996).
(5.) W.J. Tomlinson and J.R. Holland, Reinforc. Plast., 37, 46 (1993).
(6.) S. Wiedmer and M. Manolesos, J. Thermoplast. Comp. Mater., 19, 97 (2006).
(7.) V. Kerbiriou, VDI-Verlag, Impragnieren und Pultrusion von ther-moplastischen Verbundprofilen, Dusseldorf, Germany (1997).
(8.) J.G. Vaughan. T.W. Dillard. and E. Seal, J. Thermoplast. Comp. Mater., 3. 131 (1990).
(9.) B.T. Astrom, P.H. Larsson. P.J. Hepola, and R.B. Pipes, Composites, 25, 814 (1994).
(10.) C. Poli, Manufacturing Handbook and Buyers' Guide: Plastic Technology, Bill Communications, New York, 601 (1996).
(11.) M. Evstatiev and S. Fakirov. Polymer, 33, 877 (1992).
(12.) M. Evstatiev, S. Petrovich, and S. Fakirov, Macromolecules, 26, 5219(1993).
(13.) S. Fakirov. M. Evstatiev, and J.M. Schultz. Polymer, 34, 4669 (1993).
(14.) S. Fakirov and M. Evstatiev. Adv. Mater., 6. 395 (1994).
(15.) M. Evstatiev, S. Fakirov, and K. Friedrich, Appl. Comp. Mater., 2, 93 (1995).
(16.) M. Evstatiev. J.M. Schultz. S. Fakirov. and K. Friedrich, J. Appl. Polym. Sci., 67, 723 (1998).
(17.) M. Evstatiev, S. Fakirov, and K. Friedrich, Structure Development During Polymer Processing, A.M. Cunha and S. Fakirov, Eds.. Kluwer, Dordrecht, 311 (2000).
(18.) S. Fakirov, Handbook of Thermoplastic Polyesters. Wiley-VCH, Weinheim, 2, 1093 (2002).
(19.) M. Evstatiev, S. Fakirov, B. Krasteva, K. Friedrich, J. Covas, and A. Cunha, Polym. Eng. Sci., 42. 826 (2002).
(20.) K. Friedrich. H. Kamo, M. Evstatiev. and S. Fakirov, J. Macromol. Sci. Phys., 43, 776 (2004).
(21.) K. Friedrich, S. Fakirov, and Z. Zhang, Eds., Polymer Composites: from Nano- to Macro Scale, Springer, New York, 149 (2005).
(22.) T.B. Astrom and R.B. Pipes. Polym. Comp., 14. 173 (1993).
M. Evstatiev, (1) I. Angelov, (1) K. Friedrich (2), (3)
(1) Laboratory on Structure and Properties of Polymers, Faculty of Chemistry, "St. Kliment Ohridsky" University of Sofia, 1164 Sofia, Bulgaria
(2) Institute for Composite Materials (IVW GmbH), Technical University of Kaiserslautern, D-67663 Kaiserslautern, Germany
(3) CEREM, College of Engineering, King Saud University, Riyadh, Saudi Arabia
Correspondence to: Michael Evstatiev; e-mail: firstname.lastname@example.org
Contract grant sponsor: Deutsche Forschungsgemeinschaft (DFG).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||polyethylene terephthalate/low-density polyethylene|
|Author:||Evstatiev, M.; Angelov, I.; Friedrich, K.|
|Publication:||Polymer Engineering and Science|
|Date:||Feb 1, 2010|
|Previous Article:||Crystallization behavior of PIPD/nylon 6 blends.|
|Next Article:||Effect of processing parameters, antistiction coatings, and polymer type when injection molding microfeatures.|