Barium sulfate-filled blends of polypropylene and poly(styrene-co-acrylonitrile) - dynamic mechanical melt state properties and interlayer effects.
Filled immiscible thermoplastic blends have a complex phase structure, which depends on the processing parameters, chemical character, volume shares, melt viscosity, and surface tension of the components (1, 2). The rheological properties can be used to analyze filled immiscible blends, but the behavior is complex and the properties of a given system can vary substantially depending on the morphology. Because of this, great care must be taken in the analysis of these systems. The most important aspects of the rheology of filled polymers, covering the degree of filling, the influence of particle size and size distribution, and the sensitivity of the dynamic shear amplitude, are very well described in review articles (3-5), and the behavior of multiphase systems is described in more detail by Han (6) and others (7, 8). Basic rules for how to control the morphology of immiscible blends are described by Paul and Barlow (9) and dual phase continuity has been reviewed by Lyngaae-Jorgensen (10). Filled polymer blends have been investigated in studies regarding electrical conductive composites containing carbon black (11, 12). The key issue in this area is to control the phase and filler structure to reach double percolation, i.e. polymer phase continuity and filler-filler continuity. This can be reached at low filler content when the filler is occluded in one of the phases or at the phase boundaries. The mechanical behavior and viscoelastic properties related to the phase morphology of PC/SAN polymer blends has been investigated by Quintens and co-workers (13, 14). This work is of interest here, because of the use of dynamic mechanical measurements as an analytical tool to describe the change of phase morphology during thermal annealing above the glass temperature of both polymers.
In our previous work (15), the microstructure and mechanical properties of filled immiscible blends were investigated by scanning electron microscopy (SEM) and solid state dynamic mechanical spectroscopy, and the measurements were compared with theoretical calculations using a core shell model structure (16, 17). The melt mixed blends consisted of a polypropylene matrix and a poly(styrene-co-acrylonitrile) dispersed phase with a barium sulfate filler. The PP-g-MAH affects the chemical character of the polypropylene matrix and thereby the potential for interactions with the filler and thus offers means for control, in which phase the filler is occluded. This is described by [TABULAR DATA FOR TABLE 1 OMITTED] the SEM micrographs in Fig. 1 a-b, showing filled blends with an average particle size of 1 [[micro]meter], where the residues of the PP/SAN/BaS[O.sub.4] blends (60/20/20 vol %) are visible after selective extraction of the SAN phase by THF. In Fig. 1a, the barium sulfate filler is visible in the cavities after the dissolved SAN phase. However, if enough of the PP matrix is replaced by PPg-MAH, as in Fig. 1b, the mineral filler will be occluded in the PP phase. The microstructure and the dynamic mechanical properties in the solid state could be correlated. The intensity of the SAN glass transition was studied as a function of filler particle size and amount and type of maleic anhydride-grafted polypropylene in the polypropylene matrix. A theoretical model, called the interlayer model, was used as a tool to qualitatively analyze the properties of the investigated blends.
The aim of this work was to study the dynamic mechanical properties of filled blends with different microstructures in the melt state, which consists of a polypropylene matrix and a poly(styrene-co-acrylonitrile) dispersed phase with barium sulfate filler. The properties were studied as a function of filler particle size and amount and type of maleic arthydride-grafted polypropylene in the polypropylene matrix. The results of the measurements were compared with the interlayer model to distinguish the effect of a PP-gMAH interlayer between the filler particles and the polypropylene matrix, the effect of the occlusion of the filler in either the SAN or PP phase, and the effect of a filler network formation.
Table 1 shows the properties of the polymers and of the BaS[O.sub.4] fillers used in this study. The molecular masses of the polymers were determined by size exclusion chromatography, calibrated with broad polyethylene standards for PP and PP-g-MAH, and with narrow polystyrene standards for the SAN copolymer. All the BaS[O.sub.4] grades were of precipitated qualities. Melt mixing in a Brabender AEV 330, with a batch volume of 50 ml, was preceded by drying overnight under vacuum at 60 [degrees] C for all materials except the polypropylene. The polymers were first preblended for 5 minutes, with an additional 10 minutes' melt mixing with the BaS[O.sub.4] filler added at a melt temperature of 190 [degrees] C and 40 rpm. The filled polymer composites were then pressed at 190 [degrees] C into 25 mm circular samples for dynamic mechanical characterization. The samples were analyzed with a Rheometrics Dynamic Analyzer RDAII with a 25 mm parallel plate fixture in the melt state at 170 [degrees] C. Dynamic strain sweeps from 0.1 to 100% at 1 rad/s and dynamic frequency sweeps from 500 to 0.1 rad/s at low strains, typically 1% for filled blends with PP-g-MAH and 0.5% for filled blends without PP-g-MAH, were used in the measurements. The pure materials were characterized at higher strains. Scanning electron microscope analysis, with a Zeiss DSM 940A apparatus, was performed on samples fractured at liquid nitrogen temperature followed by 2 hours' selective extraction of SAN in THF at room temperature. The densities used to estimate the volume fraction were 910 kg/[m.sup.3] for the PP and the PP-gMAH, 1080 kg/[m.sup.3] for the SAN copolymer and 4400 kg/[m.sup.3] for the BaS[O.sub.4]. The volume share polypropylene plus PP-g-MAH was kept constant at 60% in the filled blends, and the concentration of PP-g-MAH is given as volume percentage of the polypropylene matrix. The SAN and filler volume shares were kept equal and constant at 20% each.
The interlayer model (16, 17) is used to simulate the dynamic mechanical properties of the filled blends. The model is derived from the van der Poel model (18) corrected by Schwarzl (19) and Smith (20). The representative volume element in the interlayer model has the geometry of an elastic sphere covered by a shell of interlayer, which in turn is surrounded by a shell of matrix material. Finally, there is a homogeneous phase with properties equivalent to those of the heterogeneous dispersion surrounding the spheres. The model assumes perfect dispersion of spherical particles in the matrix without taking into account any interactions, particle size or particle size distribution. The method accounts only for the pure mechanical-geometrical effect of the presence of the particle and layers. The most important condition is that the representative volume element reacts to external stresses as a cube of homogeneous composite material and that displacements and radial and tangential stresses are continuous at the boundaries of the phases. With these limiting prerequisites, a set of linear equations is obtained as a result of the continuity conditions in a rotationally symmetrical shear stress field. The determinants of the coefficients of the linear equation system should be zero to obtain a solution that is not trivial with which to calculate the shear modulus. This leads to a quadratic equation with the following form:
[Mathematical Expression Omitted] (1)
[G.sub.c] and [G.sub.m] are the shear moduli for the composite and matrix, respectively, and [absolute value of X], [absolute value of Y], [absolute value of Z] and [absolute value of T] are tenth order determinants of the 10 x 10 matrixes, X, Y, Z and T, respectively. The determinants depend on the shear moduli, Poisson ratios and the volume fractions of the constituents. The extension of the elastic solution for Equation 1 to a viscoelastic solution is made on the basis of the correspondence principle (16, 17, 21). The dynamic mechanical properties of the polymers used in the model as the constitutive materials are shown in Fig. 2 a-b. The values of the PP and SAN polymers are the measured ones, and the properties of the low molecular weight MA4 graft copolymer are smoothed data from the measurements, with an average phase angle of 74 [degrees].
Calculations of the mechanical properties of the filled blends without PP-g-MAH were always made with a polypropylene matrix, SAN interlayer and BaS[O.sub.4] filler particle. The properties of the mineral-filled PP phase with a PP-g-MAH interlayer were calculated with MA4 in the model blend. The properties of the filled polypropylene matrix with different volume shares of filler and MA4 were then able to be used in the calculations of the properties of the system (PP/MA4/BaS[O.sub.4])/SAN/BaS[O.sub.4], which represents matrix/interlayer/filler. If all the filler was assumed to be located in the polypropylene phase, the calculations were performed for a system with a (PP/MA4/BaS[O.sub.4])matrix filled with spherical SAN occlusions without interlayer. Furthermore, Poisson ratios for SAN, PP and MA4 were assumed to be 0.5 in the melt state. The Poisson ratio for the BaS[O.sub.4] is 0.3 (22), and the dynamic shear modulus, [G.sub.d]([Omega]), for BaS[O.sub.4] is calculated to have a value (22) of 1.15 x [10.sup.10][Pa]. The filler is considered in the simulations to be pure elastic, i.e. the dynamic shear loss modulus, G[double prime][([Omega]).sub.filler] = 0.
The melt state dynamic mechanical spectra of the polypropylene, SAN and MA4 are depicted in Fig. 2 ab. It is clear from the Figure that the SAN copolymer has a higher dynamic mechanical modulus than PP in the melt state. The MA4 low molecular weight maleic anhydride grafted polypropylene has a low dynamic modulus and an almost Newtonian behavior at 170 [degrees] C.
The dynamic mechanical properties of the filled blends was seen to be very sensitive towards the amplitude of the oscillatory shear in the measurements. The known reason for this sensitivity is a structure formed by the filler particles. The dependence of the strain amplitude for the properties is shown in Fig. 3, depicting the quotient of the dynamic modulus as a function of the strain amplitude at 1 rad/s in the numerator and of the dynamic modulus at 0.1% strain amplitude at 1 rad/s in the denominator, [G.sub.d]([Gamma])/[G.sub.d,[Gamma] = 0.1%], plotted as a function of strain amplitude. The average filler size and concentration of MA4 are varied in Fig. 3 in the PP/SAN/BaS[O.sub.4] blends. We concluded from our previous study (15) that the filler particles are occluded solely in the SAN phase in PP/SAN/BaS[O.sub.4] blends without MA4. With 2.5% of the PP matrix substituted by the graft copolymer and with fillers of 1 [[micro]meter] average particle diameter, the filler win be occluded in both phases. Finally with more than 5% MA4, the filled blend reaches a saturation level with all the filler particles occluded in the PP phase. From the strain sweep in Fig. 3, it can be seen that the nonlinearity starts at low strain values. The sensitivity is weakest for the smallest particles with an average particle diameter of 0.1 [[micro]meter]. The amplitude dependence becomes less pronounced when PP-g-MAH is added. The reason for this is twofold. First, as we deduce from our previous study, the filler concentration decreases when the filler is present in both phases or only in the PP phase and not in the SAN phase, which would be highly filled and very strain-sensitive. Secondly, the PP-g-MAH has an affinity to the BaS[O.sub.4] filler and probably acts on the filler as a dispersive agent in the PP phase, with less filler-filler contact changing the filler particle structure and thereby lowering the strain dependence.
Figures 4 and 5 shows the frequency dependency at 170 [degrees] C of the dynamic mechanical properties for the different PP/SAN/BaS[O.sub.4] (60/20/20 vol%) blends. The concentration of PP-g-MAH is varied for three different average particle sizes (0.1, 1 and 3.5 [[micro]meter]) in Fig. 4 a-c, and the type of PP-g-MAH is then varied for blends with 1 [[micro]meter] filler size in Fig. 5 a-b. In the blends without PP-g-MAH, the filler is occluded in the SAN phase, meaning that 40 vol% of the blend is a highly filled polymer (50/50 vol%) with a strong modulus-enhancing effect in the melt. Another important reason for the high modulus values without PP-g-MAH is probably the deviation from the ideal dispersed spherical phase structure.
The frequency dependency of [G.sub.d]([Omega]) and [Delta]([Omega]) for the filled blends with fillers of an average particle size of 0.1 [[micro]meter] and different concentrations of MA4 can be seen in Fig. 4a. The measurements were carried out with different strain amplitudes, but at the expected low amplitudes for filled polymers of 0.5-1% (4, 5). The lower amplitude of 0.5% was used for blends without PP-g-MAH and a strain amplitude of 1% was used in blends with PP-g-MAH because of the above mentioned phenomena. The dynamic modulus is at its highest value for all frequencies shown when no MA4 is present. With 2.5% and 5% MA4, the modulus is lowered, but with 7.5 and 10% MA4 the [G.sub.d]([Omega]) values increase as a function of PP-g-MAH concentration, especially at low frequencies. This behavior is also described by the phase angle, which decreases with frequency and increases with MA4 content up to 5%. At 7.5% and 10% MA4, the phase angle, [Delta]([Omega]), is lowered as compared with 5% MA4. At low frequencies, the phase angle even decreases as a function of decreased angular frequency for the blends with 7.5% and 10% MA4.
The average filler size is increased to 1 [[micro]meter] in Fig. 4b. This lowers the dynamic modulus and increases the phase angle at the investigated frequencies, as compared with the corresponding blends with an average filler size of 0.1 [[micro]meter]. The tendencies of the mechanical properties when increasing MA4 concentration are similar to the tendencies in Fig. 4a with 0.1 [[micro]meter] fillers. In Fig. 4c, the average barium sulfate filler size is increased to 3.5 [[micro]meter], and the [G.sub.d]([Omega]) and [Delta]([Omega]) are further lowered and increased, respectively.
Different PP-g-MAH types are used in Fig. 5a and 5b, namely MA1 and MA7. In an earlier work (15), they were shown to have quite a different capability to occlude the filler in the polypropylene phase as a function of concentration. The MA7 must be added in higher concentrations, compared with MA4, and MA1 is weak in its ability to control the phase in which the filler is occluded. When MA1 and MA7 are used, there is an experimental indication of the influence of the low molecular PP-g-MAH on the dynamic mechanical properties, especially the MA1, which is not very effective in controlling the phase in which the filler is occluded. This can be observed in Fig. 5a, where the dynamic modulus decreases systematically as a function of the concentration of MA1. When MA1 is present in the blends, the phase angle, [Delta]([Omega]), increases for all concentrations from 2.5% up to 10% MA1. MA7 is used in Fig. 5b, and the pattern is closer to the behavior of the blends with MA4, but the [G.sub.d]([Omega]) does not increase as a function of MA7 concentration at higher concentrations, as do blends with MA4. Neither does [Delta]([Omega]) decrease as a function of decreasing frequency at low frequencies and high MA7 concentration. The lowest modulus, in the blends with a filler size of 1 [[micro]meter], is reached with MA4 and MA7 at 5% and 10% PP-g-MAH, respectively. This is probably because of the change in microstructure and dilution of the filler in the PP matrix phase of low viscosity compared with SAN, and because of the PP-g-MAH interlayer around the filler particles occluded in the PP phase. The MA1 does not have the ability to occlude an the filler particles in the PP phase and cannot cause these effects, and thus there is a lower modulus decrease at the highest MA1 concentration compared with MA4 and MA7.
The larger drop in the [G.sub.d]([Omega]) curves as a function of PP-g-MAH concentration in Figs. 4 and 5 can be coordinated with morphological changes in the filled blends. The morphology as a function of PP-g-MAH concentration was investigated earlier by SEM (15). When MA4 is added, the PP/SAN/BaS[O.sub.4] blend goes from a 60/40 PP/(SAN/BaS[O.sub.4]) system with a cocontinuous type of microstructure to a 80/20 (PP/MA4/BaS[O.sub.4])/SAN system with a matrix and an elongated dispersed SAN phase. This phenomenon can be seen in Fig. 4a when 5% MA4 is used and in Figs. 4b and 4c when 2.5% MA4 is used. The absence of one dominant jump in the [G.sub.d]([Omega]) curves in Fig. 5a is a result of the fact that the filler is never occluded solely in the PP phase as a function of MA1 concentration. The viscosity-decreasing effect of the polypropylene phase resulting from the fact that the PP-g-MAH does not act as an interlayer is therefore much more visible with the MA1 graft copolymer than with MA4 and MA7. Another effect shown in Fig. 4 is the increase of the [G.sub.d]([Omega]) curves at the highest MA4 concentrations or the decrease of the phase angle at low angular frequencies and high MA4 concentrations. This may be explained by the formation of an elastic component in the matrix, caused by the well-dispersed BaS[O.sub.4] forming a structure in the polypropylene matrix. On the other hand, the introduction of a low molecular mass interlayer around the filler particles should not cause a higher modulus, but on the contrary should lower the dynamic modulus. The effects of the filler surface area can also be seen in Fig. 4, from which it can be concluded that the drop of modulus, [G.sub.d]([Omega]), as a function of MA4 concentration in the blends is lowest for the smallest sized filler of 0.1 [[micro]meter]. Accordingly, the phase angle decreases with increased filler surface area in Fig. 4 a-c. At higher frequencies, the phase angle in Fig. 4 a-c first increases with the concentration of PPg-MAH. However, for the smallest filler in Fig. 4a, the phase angle decreases slightly with increased MA4 concentration at the highest MA4 concentrations. This could also be a result of a more efficient filler structure formation at higher MA4 concentrations compared with the fillers of larger size and smaller surface area. The increase in modulus in Fig. 4a when increasing the concentration of MA4 from 5% could be logical if the PP-g-MAH has a compatibilizing effect on the BaS[O.sub.4] filler in the PP phase and causes a break-up of agglomerated structures resulting in particles with a smaller effective diameter. However, once again, the same counter-argument can be raised; a low molecular mass interlayer should lower the modulus and increase the phase angle. The results so far show that the dynamic mechanical properties in the melt vary as a complicated response to morphology, dilution of filler, polymer filler compatibility, interlayer formation and a filler structure formation.
Some of the simulations with the interlayer model made are shown in Fig. 6, where the calculated dynamic modulus, [G.sub.d]([Omega]), is plotted as a function of angular frequency at 170 [degrees] C. The model curve with the highest modulus is reached in the simulation with an the filler occluded in the SAN phase and no MA4 added to the blend. In the case in which all the filler is occluded in the polypropylene phase, the modulus curve drops a relatively small step compared with the curve with the highest modulus, but, when an interlayer of MA4 is introduced in the model, the [G.sub.d]([Omega]) curve drops more significantly. The drop of [G.sub.d]([Omega]) then continues with the layer thickness in a systematic way. The theoretical interlayer effects are analyzed by the quotas of the calculated [G.sub.d]([Omega]) curves in Fig. 7; the four uppermost curves show the quota between the [G.sub.d]([Omega]) for a blend without MA4 and an the filler occluded in the SAN phase in the numerator, and, in the denominator, the modulus of the blends with a MA4 interlayer between the polypropylene and the filler. It can be seen that the theoretical influence of the PP-gMAH interlayer in the calculations is more pronounced at lower frequencies, especially with a thicker interlayer. The theoretical frequency dependence of the phase in which the filler is occluded in a filled blend without PP-g-MAH, illustrated by the lowest curve, is relatively weak in Fig. 7, but it must be remembered that the model does not take the degree of filling or particle-particle interaction or interlocking into consideration. The theoretical calculations in the melt state support the analysis of the melt state dynamic mechanical measurements and previous SEM study that a PP-g-MAH interlayer forms around the filler particles and filler agglomerates, and this results in great morphological and mechanical effects. The use of functionalized polymers has shown to be a viable route to morphological control.
Effects of an interlayer formation and of the filler size can be seen in the results in Fig. 4, depicting the dynamic mechanical measurements with different filler sizes and the same PP-g-MAH, MA4. The effect of an interlayer formation is the large drop in modulus as a function of an increased concentration of MA4, described earlier. The influence of the filler size is illustrated [TABULAR DATA FOR TABLE 2 OMITTED] by the higher absolute modulus of the corresponding blends containing the smaller filler. The theoretical modulus is, as expected, always lower than the measured ones for the equivalent filled blends. This is an indication of filler-filler contact, interlocking and other deviations from the ideal conditions in the theoretical model, assuming perfect dispersion of spherical particles with spherical interlayer and matrix. Another important question is whether we can regard the PP-g-MAH as an interlayer, with the properties of the pure material, around every particle, regardless of the particle size or the concentration of PP-g-MAH. Theoretically, the random coil end-to-end size, r, can be calculated by using the formula (23):
[<[r.sup.2]>.sub.0] = [C.sub.[infinity]] [multiplied by] n [multiplied by] [l.sup.2]
where the first subscript signifies the theta conditions, which can be regarded in the melt state, and n is the number of carbon-to-carbon bonds of length, l = 1.54[Angstrom]. [C.sub.[infinity]] is a constant which depends on the nature of the polymer when n, [approaches][infinity] and is set to 5.7, which is the value for isotactic PP in accordance with Flory (23). With a SEC number average molecular mass for the MA4 of 8.4 kg/mole, the end-to-end size is calculated to 7.4 nm. This is to be compared with the theoretical interlayer thickness in Table 2, if all particles are surrounded by a PP-g-MAH interlayer regardless of the random coil size, which must indicate the minimum layer thickness. A strong interaction between the BaS[O.sub.4] filler and the PP-g-MAH is probably not the case here, but could otherwise change the conformation of the graft copolymer on the filler surface. The theoretical interlayer thickness in Table 2 shows that the thickest interlayer in the case with the smallest filler (0.1 [[micro]meter]) is of lower value than the minimum layer thickness indicated by the end-to-end size. When the largest filler of 3.5 [[micro]meter] is used, the interlayer is probably thick enough to show bulk properties, and the situation with the 1-[[micro]meter] filler is an intermediate case. An interlayer of only a few nanometers does not have the properties of the pure material, especially when the interlayer consists of modified PP and must be compatible to some degree with the matrix PP.
We can therefore speculate that agglomerated filler structures consisting of BaS[O.sub.4] filler particles with an average size of 0.1 [[micro]meter] are occluded in the PP matrix because of the MA4 graft copolymer. A further increase of MA4 results in deagglomeration of the filler.
The filler particles are probably surrounded by an interlayer consisting of PP-g-MAH molecules attached to the filler surface, and the graft copolymer is also affected by the matrix, resulting in an interlayer with a higher modulus than that of the pure MA4. This phenomenon could give rise to a filler structure in the PP matrix able to increase the dynamic mechanical modulus. The process described may also be true in the case with the barium sulfate filler, with an average particle size of I [[micro]meter], but with much less emphasis on the deagglomeration part. The speculation is supported in the literature, where there is a description of the formation of filler networks by spherical particles in an unpolar thermoplastic matrix (24). Factors such as filler size and level of filler-matrix interaction capabilities, which affect filler network formation in carbon black-filled thermoplastic polymers, are described by Gandhi and Salovey (25). The phenomenon of polymer bound on solid filler is discussed by Lipatov (26), and adsorbed polyethylene on kaolin was investigated by Maurer et al. (27). The bound polymer results in a layer with properties different from the bulk properties of the pure polymer.
The dynamic mechanical properties of the PP/SAN/BaS[O.sub.4] blends in the melt are systematically influenced by the BaS[O.sub.4] average particle size and the amount of MA4 added to the blend. Decreasing the average filler particle size decreases the drop in the dynamic shear modulus as a function of PP-g-MAH concentration in the blends and increases the absolute modulus. Increased concentration of the low viscous MA4 can even enhance the modulus of the blends, especially at lower frequencies; from the study it is reasonable to believe that this is a result of a filler network formation. The use of different low molecular weight PP-g-MAH qualities sheds some light on the effect of a low molecular component in the PP phase with the filler occluded in both phases. MA4 shows a complicated response in its dynamic mechanical properties because of its ability to control the polymer phase in which the filler is occluded and a probable ability to deagglomerate the barium sulfate filler of the smallest average size. On the other hand, MA1 shows a simpler response, with a monotonously decreasing modulus as a function of concentration at all frequencies studied.
Comparisons of measurements and theoretical calculations indicate that the phase in which the filler is occluded is of lesser importance. The more important factor for the dynamic mechanical properties in the melt is the introduction of an interlayer of low viscosity between the filler and the polypropylene and the filler size.
We gratefully acknowledge the National Swedish Board for Industrial and Technical Development (NUTEK) for financial support of this work. We are also grateful for the assistance of A. Martensson with the SEM apparatus, L-I. Kulin and M. Bjorklund for SEC analysis, and M. Schmidt for valuable advice regarding the dynamic mechanical analysis.
1. J. Kolarik, F. Lednicky, J. Jancar, and B. Pukanszky, Polymer Comm., 31, 201 (1990).
2. B. Pukanszky, J. Kolarik, and F. Lednicky, Compos. Polym., 3, 271 (1990).
3. A. B. Metzner, J. Rheol., 29, 739, (1985).
4. M. R. Kamal and A. Mutel, J. Polym. Eng., 5, 293 (1985).
5. A. Y. Maikin, "Rheology of Filled Polymers," in Advances in Polymer Science, 96, 69 (1990).
6. C. D. Han, Multiphase Flow in Polymer Processing, Academic Press, New York (1981).
7. L. A. Utracki, Polymer Alloys and Blends, Thermodynamics and Rheology, Hanser Publishers, Munich (1989).
8. C. W. Macosko, P. Guegan, A. Khandpur, A. Nakayama, P. Marekal, and T. Inoue, Macromolecules, 29, 5590 (1996).
9. D. R. Paul, and J. W. Barlow, J. Macromol. Sci. Rev. Macromol. Chem. C, 18, 109 (1980.)
10. J. Lyngaae-Jorgensen, and L. A. Utracki, Makromol. Chem. Macromol. Symp., 48/49, 189, (1991).
11. K. Levon, A. Margolina, and A. Z. Patashinsky, Macromolecules 26, 4061 (1993).
12. F. Gubbels, S. Blacher, S. Vanlathem, R. Jerome, R. Deltour, F. Brouers, and Ph. Teyssie, Macromolecules, 29, 1559 (1995.)
13. D. Quintens, G. Groeninckx, M. Guest, and L. Aerts, Polym. Eng. Sci. 30, 1474 (1990.)
14. D. Quintens, G. Groeninckx, M. Guest, and L. Aerts, Polym. Eng. Sci. 31, 1207 (1991.)
15. C. O. Hammer, and F. H. J. Maurer, J. of Adhesion, 64, 61 (1997).
16. F. H. J. Maurer, in "Polymer Composites", p. 399, B. de Sedlacek, ed., W, de Gruyter & Co., Berlin, (1986).
17. F. H. J. Maurer, in Controlled Interphases in Composite Materials, pp. 491-504, H. Ishida, ed., Elsevier, New York, pp. 491-504, (1990).
18. C. van der Poel, Rheol. Acta, 1, 198 (1958).
19. F. R. Schwarzl and J. van der Eikhoff, in Calculation of Elastic Properties of Composites by Means of the van der Poel Theory, Central Lab. TNO Delft, The Netherlands, Report no. CL 71/55 (1971).
20. J. C. Smith, J. Res. Nat. Bur. of Standards, 78A, 355 (1974).
21. R. M. Christensen, Mechanics of Composite Materials, Wiley, New York (1979).
22. H. S. Katz and J. V. Milewski, Handbook of Fillers and Reinforcements for Plastics, p. 53, Van Nostrand Reinhold Co., New York, (1978).
23. P. J. Flory, Statistical Mechanics of Chain Molecules, p. 11, Hanser, New York, (1989).
24. L. Sun, J. J. Aklonis, and R. Salovey, Polym. Eng. Sci., 33, 1308 (1993).
25. K. Gandhi, and R. Salovey, Polym. Eng. Sci., 28, 877 (1988).
26. Y. S. Lipatov, Colloid Chemistry of Polymers, A.D. Jenkins, ed., Elsevier, Amsterdam, (1988).
27. F. H. J. Maurer, R. Kosfeld, and Th. Uhlenbroich, Col. Pol Sci. 263, 624, (1985).
|Printer friendly Cite/link Email Feedback|
|Author:||Hammer, Claes O.; Maurer, Frans H.J.|
|Publication:||Polymer Engineering and Science|
|Date:||Aug 1, 1998|
|Previous Article:||Free volume, mobility, and structural relaxations in poly(ethylene oxide)/poly(methyl methacrylate) blends.|
|Next Article:||Poly(styrene-g-ethylene oxide) copolymers as interfacial agents in immiscible polymer blends.|