Comparative X-ray scattering, microscopical, and mechanical studies on rectangular plates injection molded from different types of isotactic polypropylene.
It is well known that the interior of plastic articles injection molded from semicrystalline polymers such as polypropylene (PP) is far from being homogeneous and isotropic, but shows a multilayered structure, parallel to the surface of the moldings (1, 2). Details of this layered structure (texture) depend on the molecular properties of the polymer and, to a high degree, on the processing conditions. Moreover, the structure usually varies with increasing distance from the gate. Since the anisotropies of this layered structure are important for the end-use properties of the molded articles, numerous studies have been performed to obtain information on the internal structure of moldings and to correlate the observed structural details with the processing conditions on one hand and with mechanical and further properties of the moldings on the other hand, for example, Refs. 3-11.
Since the sandwich-like structure in a molding may consist of several structurally different layers, e.g., skin and core, etc., being themselves quite heterogeneous (see Ref. 2), a thorough structural characterization of the cross section of the molding necessitates an adequate spatial resolution of the applied method of investigation. Among the methods able to yield information about the internal structure and texture of a molding, optics using visual light (polarization microscopy, birefringence measurements), electron microscopy, and X-ray methods (small-angle and wide-angle X-ray scattering) have been used. X-ray methods were applied in different approaches for studying the layered structure, e.g., by taking film diffractograms at selected positions in the cross section by means of a micro camera (2), or by sectioning the molding parallel to its surface in three (9) or numerous (7) slices prior to difractometric investigation. Recently, spatially resolved X-ray scattering, performed by scanning the cross section of plastic parts with a fine X-ray beam (collimated by means of a Kratky small-angle camera) and registering the intensity of scattering as a function of position in the cross section, was shown to be a potent tool for the structural characterization at high spatial resolution (12-19).
In continuation of our previous work, we present in this paper a thorough structural study on six rectangular plates, which were injection molded from two different grades of isotactic PP at pressures from 293 to 1560 bar and investigated by applying spatially resolved wide-angle X-ray scattering, transmission electron microscopy, polarization optical microscopy, birefringence measurements, and mechanical tests to obtain comprehensive information about the layered structures in the plates, about the influence of polymer grade and processing conditions, and about the relations between structure and end-use properties. The samples investigated in this study represent a subset of a large number of specimens molded in context with the national working party S33 on the "Influence of molecular structure and processing parameters on the properties of molded plastic parts," initiated by H. Janeschitz-Kriegl (see Ref. 20). Some preliminary results of our investigation have been published elsewhere (18, 19, 21, 22).
MATERIALS AND METHODS
Two commercial isotactic PP grades, Daplen KS10 [MFI(230/2.16) = 8.0 g/10 min, [M.sub.W] = 322,000 g/mol, [M.sub.W]/[M.sub.N] = 6.8] and Daplen PT55 [MFI(230/2.16) = 18.6 g/10 min, [M.sub.W] = 205,000 g/mol, [M.sub.W]/[M.sub.N] = 3.5], produced by PCD Polymere/Linz, were selected for the present investigations (20).
Preparation of Moldings
Rectangular plates with dimensions of 230 x 70 x 2 mm, with film gate, were injection molded without packing, using an injection-molding machine model Engel ES 250 with microprocessor CC-90. Injection-molding conditions were as follows: melt temperature 200 [degrees] C, mold temperature 20 [degrees] C, flow front velocity 100 mm/s. The cavity was mechanically sealed when a desired pressure [p.sub.i,max] (Table 1) was reached.
Preparation of Samples
Sections of the PP specimens were taken perpendicular to the surface of the plates by different methods of microtomy. The thickness of the sections extracted from the plates in different positions and orientations depended on the method of investigation that followed the preparation step.
Ultrathin sections (thickness about 0.1 [[micro]meter]) of Ru[O.sub.4] stained samples for transmission electron microscopical (TEM) investigations were produced using a Reichert Ultracut E ultramicrotome with cryo unit, using diamond knives with a cutting edge of 45 degrees.
For the light microscopic observation, we prepared semithin sections (5-12 [[micro]meter] thick) for high resolution, using a Reichert Supercut 2050 microtome with diamond knives. In addition thin platelets of 30-100 [[micro]meter] thickness were prepared for a lower resolution general view. For this purpose we used a microtome with a rotating internal-hole saw blade (Sagemikrotom 1600, Leitz/Wetzlar). This microtome renders platelike samples over a specimen length of up to 35 mm without mechanical deformation (deformation is unavoidable if sectioning is performed with a cutting knife).
For X-ray investigations, blocks of about 15 x 3 x 2 mm edge-length were taken in the same way with the internal-hole saw microtome, following the thin section removal.
Microtomed sections taken in flow direction (FD) or perpendicular to it were examined by means of a polarizing light microscope (Diapan, Reichert/Wien), equipped with a [Lambda]/2 platelet. Subsequently micrographs were taken to fix the thickness of the different layers and to register the morphology characteristic of the observed sector. The magnification of the micrographs generally ranged from 20- to 600-fold, the thickness of visible layers varied from a few microns to hundreds of microns.
For TEM investigations, for improved contrast a special staining procedure (23) was applied to the trimmed specimen blocks before sectioning. By exposing the trimmed specimens only to the vapor phase above a freshly prepared aqueous staining mixture of Ru[Cl.sub.3] and hypochlorite at room temperature, the amorphous regions of the PP samples were stained, without destroying or swelling the samples (various other methods of staining failed when applied to the PP specimens of average commercial quality). This procedure rendered the fine structure of exactly defined areas, already investigated by preceding light microscopy, in much higher resolution. The TEM micrographs had a final magnification of 50,000-fold to resolve the fine structure composed of alternating crystalline lamellae and amorphous interlayers having an average thickness of about 10 nm. Serial TEM micrographs enabled us to follow the development of periodic structures from the surface of the molding down to a depth of more than 250 [[micro]meter] in one ultrathin section.
Table 1. Variation of the Maximum Pressure in the Cavity.
PP grade [p.sub.i,max] (bar) plate
KS10 293 1K 702 2K 1559 3K
PT55 465 1P 923 2P 1540 3P
Birefringence measurements were performed on 40-[[micro]meter] thick microtomed sections cut perpendicular to the plate surface in FD. These sections were mounted between microscope slides using xylene-free Canada balsam so that any refractive-index mismatch was tolerated. A polarizing light microscope (Leitz/Gottingen) equipped with a tilting compensator was used to determine the optical birefringence. The measurements lead to birefringence profiles of the oriented structure in the cross section of the plates.
Wide-angle X-ray Scattering (WAXS)
Block-shaped samples were cut from each plate [ILLUSTRATION FOR FIGURE 1 OMITTED] at different distances from the gate (20 mm, 95 mm (PT55) or 100 mm (KS10), 210 mm) in different orientations (parallel and normal to FD).
These samples were mounted on appropriate holders and investigated by means of a conventional Kratky small-angle camera (Anton Paar/Graz) adapted to serve as a wide-angle diffractometer, using a very fine and short X-ray beam (15, 17-19). The X-ray source was a tube with Cu target (Philips PW 2253/11), operated at 2.5 kW; a Ni filter was used to suppress the [K.sub.[Beta]] line. The sample was aligned in such a way that the line-shaped beam (cross section of the beam at the sample about 60 [[micro]meter] x 2 mm) transmitted the sample exactly parallel to the original surface of the plate. The distance of the illuminated zone from the surface was varied by displacing the sample in steps of 20 [[micro]meter] along the direction normal to its surface (direction ND), thereby scanning the entire thickness of the cross section of the sample. Measurements were performed with the primary beam directed either normal to FD or in and against FD. The scattered radiation was registered as a function of distance from surface in the range of scattering angles 2[Theta] from about 12 to 23 degrees, using a linear detector (position sensitive detector Braun OED 50M; 500 channels, counting time 10 min) parallel to ND. The geometry of the setup enabled the equatorial measurement of the most intense and informative reflections of crystalline PP, at a high spatial resolution in the sample along direction ND.
The registration device was coupled to a personal computer that controlled all operations of the measurement, including the positioning of the sample and the intermediate storage of data. The measured intensities were first corrected for absorption effects and then inspected in terms of "intensity maps" showing scattering curves (= intensity vs. scattering angle) plotted vs. the distance from surface as parameter.
For quantitative evaluation of the data, background curves due to noncrystalline (amorphous) material were determined, by fitting a fourth-order polynomial to 5 appropriately chosen points of each scattering curve, and subsequently subtracted from the measured curves. The residual "crystalline" scattering curves were then approximated by a variable number of pseudo-Voigt functions [i.e., linear combinations of Gaussians and Lorentzians (24)] corresponding to the different X-ray reflections of PP. Eventually the fitted reflections were analyzed in terms of various parameters (18, 19).
Apparent interplanar spacings [d.sub.hkl] were determined from the angular position of reflections by applying Bragg's equation. Apparent crystallite dimensions [L.sub.hkl] were determined from the half-width of reflections by applying Scherrer's equation. The following parameters were derived by comparing the intensities of different reflections (for definitions of parameters see Table 2):
a) The orientation index [A.sub.110] [after Trotignon and Verdu (7)], and a similarly defined orientation index [A.sub.130]; these indices characterize the degree of orientation of [Alpha]-PP crystallites. The determination of both indices is based on the vanishing (or decrease of intensity) of the reflections (111) and (131 + 041) in the equatorial measurement when the crystallites orientate in the direction of flow (hence A = 1 for highly oriented [Alpha]-crystallites, otherwise A [less than] 1).
b) An index C, which may serve as a measure of the epitaxial double orientation of [Alpha]-PP crystallites (that means the crystallographic axis c and also the crystallographic axis [a.sup.*] may form a fiber axis in the direction FD, see Ref. 29). This index is based on the fact that in the equatorial measurement both [a.sup.*]- and c-oriented crystallites contribute to the (040) reflection, but only c-oriented crystallites contribute to the (110) and (130) reflections (hence C = 1 for [a.sup.*]-oriented crystallites; for c-oriented crystallites or for isotropic material C [less than] 1).
c) The index B, quantifying the relative amount of [Beta]-PP according to Turner-Jones et al. (26).
d) An index G, quantifying the relative amount of [Gamma]-PP.
For selected samples a further parameter, [T.sub.40], was calculated from the intensities of the (040) reflection of [Alpha]-PP, obtained from measurements with three different directions of the primary beam [in FD, against FD, normal to FD (15, 17, 18)]. This parameter is derived from the standard deviation [[Sigma].sub.u] (in degrees) of the density distribution of (040) poles on the equator with respect to the fiber axis (the underlying theory has been outlined in Ref. 17). It is suitable for characterizing the degree of orientation of [Alpha]-PP crystallites especially in those regions where the indices [A.sub.110] and [A.sub.130] already are at their upper limit (e.g., in highly oriented surface layers).
On 40-[[micro]meter] thick microtomed sections, removed parallel to the plate surface, tensile tests were carried out to get profiles of mechanical properties in the cross section of the plates. The sections were taken in FD at different distances from the gate. A special tensile test machine developed at the Leoben institute was used to perform the tensile tests. A strain rate of 4.2% elongation/s was applied. From the registered stress-strain diagrams, the elongation at break was determined. This mechanical property is especially suitable to reveal differences in the degree of molecular orientation.
WAXS Intensity Maps
Wide-angle X-ray scattering intensity maps of the plates from PP KS10 and PT55, molded with low and with high pressure, respectively, are presented in Figs. 2 and 3. The maps reflect the layered structure of the cross section of the moldings and clearly reveal the influence of processing conditions and of the molecular properties of the polymer on the layered architecture and its variation in dependence on the distance from gate. General features of the maps are the occurrence of very intense reflections in the surface layers of the samples from positions A and B [20 mm and 95 (100) mm from gate], especially pronounced in the case of KS10 plates, and less intense reflections in the core of these samples. At the third position (C, 210 mm from gate), the intensity of reflections from the surface layers is much less enhanced in comparison with the core region (KS10 plates, [ILLUSTRATION FOR FIGURE 2 OMITTED]) or even lower than in the core (PT55 plates, [ILLUSTRATION FOR FIGURE 3 OMITTED]).
In general, all intensity maps which were obtained from measurements with the primary beam normal to FD are quite symmetrical around the middle of the cross section, whereas intensity maps derived from measurements with the primary beam either in or against FD are rather unsymmetrical [ILLUSTRATION FOR FIGURE 4 OMITTED]. In previous papers (15, 17) such asymmetries in the intensity distribution were shown to be related to the orientation of crystallites, indicating that the direction of preferred orientation (fiber axis) is not parallel but slightly inclined toward the surface of the plates (30). Using the model of fiber orientation, the variation of the intensities of the different reflections with the distance from surface, as observed in the intensity maps in Figs. 2 and 3 (primary beam normal to FD), is generally ascribed to variations in the degree of orientation rather than to variations in the crystallinity (though an influence of the latter effect cannot be excluded).
It is noteworthy that in the range of the surface layers even the background curves show a marked dependence on the direction of the primary beam [ILLUSTRATION FOR FIGURE 5 OMITTED]. This behavior apparently reflects an orientation of the noncrystalline material. The effects are most pronounced with samples from positions A and B of KS10 plates.
A closer inspection of the intensity maps in Figs. 2 through 4 unveils the occurrence of reflections due to different modifications of PP (see the indices of reflections given in [ILLUSTRATION FOR FIGURE 4 OMITTED]). The predominant modification is [Alpha]-PP, its equatorial reflections (110), (040), and (130) corresponding to [d.sub.hkl] of about 6.26, 5.24, and 4.78 [angstrom], respectively, are almost ubiquitous except for the surface layers of PT55 plates at position C; the reflections (111) and (131 + 041) of [Alpha]-PP show up in the core region of all samples, but nearly disappear in the surface layers of the KS10 plates at positions A and B. The second important modification, [Beta]-PP, may be followed by means of its strong equatorial reflection (300) corresponding to [d.sub.hkl] of about 5.51 [angstrom]. This modification is found in all plates, always concentrated in certain regions of the cross section (in the inner part of surface layers and the adjacent transition zone); it is also clearly present at position C in KS10 plates, whereas its occurrence at position C in PT55 plates is only sporadic (see sample 1P in [ILLUSTRATION FOR FIGURE 3 OMITTED]). A third crystallite type of PP, the [Gamma]-modification, may be identified by its (130) reflection ([d.sub.hkl] = 4.40 [angstrom]). This reflection is observed mainly in the surface layers of KS10 plates at positions A and B; its intensity is found to increase considerably with the injection pressure.
Apart from the clearly crystalline [Alpha]-, [Beta]- and [Gamma]-modifications of PP and the noncrystalline (amorphous) form, the presence of which is reflected by the background curves showing a broad maximum at about 18 degrees [ILLUSTRATION FOR FIGURE 5 OMITTED], the occurrence of another modification of PP is indicated by the scattering behavior in the outermost region of the surface layers, particularly expressed in the PT55 plates at large distance from gate (at position C). The scattering curves obtained from this region just beneath the surface of the PT55 plates only show two broad peaks, at about 15 degrees and 21.5 degrees. This type of scattering is usually ascribed to the mesomorphic modification of PP, which is sometimes also referred to as a "smectic" modification. Additional hints for the occurrence of a mesomorphic modification may be found in the behavior of the background curves near the surface. Both for KS10 and PT55 plates, these curves clearly reflect a shift of the maximum of the background from about 18 degrees in the core and inner surface layers to about 15-15.5 degrees in the outermost region of the surface layers [ILLUSTRATION FOR FIGURE 5 OMITTED].
The "mesomorphic scattering behavior" of the outermost layer in the PT55 plates is obviously the more pronounced the higher the injection pressure. With increasing distance from surface a continuous structural transition appears to occur, in that the broad maximum at 15 degrees is gradually transformed and split up into the much narrower (110), (040), and (130) reflections of [Alpha]-PP. This finding suggests that the mesomorphic scattering behavior is due to the presence of very small and poorly ordered crystallites of the [Alpha]-type. TEM micrographs taken from PT55 plates, from the surface down to a depth of more than 200 [[micro]meter], unveil the presence of only locally ordered structures [ILLUSTRATION FOR FIGURE 6 OMITTED]. The magnitude of the white ribs and the range of ordering grow with increasing distance from surface. This behavior, indicating a growing amount of ordered lamellar (crystalline) structures, is in full accord with the X-ray results.
Orientation Parameters From WAXS
Profiles of the various orientation and concentration parameters (see Table 2) derived from the experimental data of all six plates at positions A and B are presented in Figs. 7 and 8. The indices [A.sub.110], B, and C are shown for all samples. The orientation index [A.sub.130] is only shown for the PT55 samples (with the KS10 samples the profile of [A.sub.130] is very much alike that of [A.sub.110], apart from the always lower level in the core region). In addition, profiles of the parameters G and [T.sub.040] are also shown for selected samples.
Already a rough inspection of the profiles of orientation parameters reveals the pronounced differences in the layered structures in the PT55 and KS10 plates, respectively.
In none of the samples from the PT55 plates [ILLUSTRATION FOR FIGURE 8 OMITTED] do the orientation indices [A.sub.110] and [A.sub.130] reach the upper limit of 1 anywhere; typical values of [A.sub.110] in the maxima in the surface layers are between 0.8 and 0.9, the maximum values of [A.sub.130] are even lower. On the other hand, values of [A.sub.110] (and [A.sub.130]) very close to or even identical with 1 can be observed in the surface layers of all samples from positions A and B of the KS10 plates [ILLUSTRATION FOR FIGURE 7 OMITTED]. This different behavior indicates that the degree of orientation in the surface layers is much higher in the KS10 samples than in the PT55 samples.
Further differences between the PT55 and KS10 plates become obvious when the shape of the [A.sub.110] profiles is compared. With the PT55 samples, the [A.sub.110] profile is markedly dependent on both the injection pressure and the distance from gate, which leads to quite different shapes of the [A.sub.110] profile in the region between the surface and a depth of about 0.5 mm [ILLUSTRATION FOR FIGURE 8 OMITTED]. With the KS10 plates, however, the [A.sub.110] profile is very similar for all samples from positions A and B [ILLUSTRATION FOR FIGURE 7 OMITTED]. Similar statements can be made about the [A.sub.130] profiles. While the [A.sub.130] profiles of the KS10 samples from positions A and B (profiles not shown) are all very much alike, the [A.sub.130] profiles of the PT55 samples show variations with pressure and distance from gate like: the [A.sub.110] profiles, but do not always follow the [A.sub.110] profiles in all details.
An example for different behavior of [A.sub.110] and [A.sub.130] profiles is the range at a depth of about 0.3 to 0.4 mm in the samples from position B of PT55 plates 1P and 2P [ILLUSTRATION FOR FIGURE 8 OMITTED]. Here the [A.sub.110] profiles show pronounced minima while in the [A.sub.130] profiles the minima are less expressed (sample 1P) or even missing (sample 2P). These minima in the [A.sub.110] profiles of the PT55 samples clearly coincide with maxima of the index B, in the case of sample 2P also with maxima of the index C. This behavior reflects a decrease of [Alpha]-crystallinity in favor of [Beta]-crystallinity in that region, paralleled by a decrease of c-orientation in favor of [a.sup.*]-orientation of [Alpha]-crystallites. The obviously different response of [A.sub.110] and [A.sub.130] to changes in the fractions of [a.sup.*]- and c-oriented [Alpha]-PP crystallites can be understood by taking into account that the (110) reflection of [a.sup.*]-oriented crystallites appears close to the meridian (in the scattering experiment on the plates equivalent to FD) while the (130) reflection appears just between the equator (the direction of detection in the scattering experiment) and the meridian. Considering a not too high degree of orientation (that means a rather broad orientation distribution), poorly [a.sup.*]-oriented crystallites will contribute to the intensity [I.sub.130] measured on the equator more readily than to the intensity [I.sub.110].
It may be helpful in this context to analyze the profiles of the orientation index C of the PT55 samples [ILLUSTRATION FOR FIGURE 8 OMITTED] in more detail. At position A of plate 1P, the profile shows a pronounced maximum in the center of the surface layers; the maximum coincides with minima of the indices [A.sub.110] and [A.sub.130] thus indicating a high fraction of [a.sup.*]-oriented [Alpha]-crystallites (the profile of [A.sub.130] is not shown to avoid overcrowding of the plot). At position B of the same plate, the maximum of index is shifted more toward the inner boundary of the surface layers. A similar shift can also be observed with the samples from plate 2P, but with these samples an additional maximum of the index C appears in the region adjacent to the surface layer. While at position A this second maximum is not much expressed, it is very pronounced at position B (see the interpretation given above). At both positions, the final decay of the index C toward the low level in the core region occurs within a relatively narrow range. In plate 3P, at both positions, the profile of index C also exhibits two maxima in the surface layers and the adjacent region; at position A the peaks are, however, much broader than in plate 2P and on the left side the values of index C decay quite slowly toward the core while on the right side (the gate-side) the index C decays much faster. Together with the accompanying changes in the shape of the profiles for [A.sub.110] and [A.sub.130], the observed variations in the profile of index C provide insight into the structural complexity of the surface layers and the adjacent regions in the PT55 plates.
In comparison with the behavior of the index C as observed with the PT55 plates, the shape of the profiles of index C for the KS10 plates unveils much less structural detail [ILLUSTRATION FOR FIGURE 7 OMITTED]. The shape of the profiles in the range of surface layers is quite plain, exhibiting only an increase of the index C with increasing depth; the effect is most pronounced at the highest injection pressure. The maximum of index C lies within or close to the boundary of the surface layers toward the core.
Since the [A.sub.110] profiles of the KS10 samples do not reveal more details of the structural organization within the surface layers, we calculated the orientation parameter [T.sub.040] (see Table 2) for the KS10 samples as an alternative source of information. The profiles of [T.sub.040] turn out to be similar for all KS10 samples presented in Fig. 7. They show a pronounced maximum situated about in or near the center of the surface layers; the height of this maximum does not depend much on the injection pressure or the distance from gate (except for sample 2K). From the theory, the position of the maximum of [T.sub.040] is expected to indicate the range where the degree of fiber orientation of [Alpha]-crystallites is highest. For drawing conclusions one should, however, take into account that the position of the maxima might be slightly erroneous (by [greater than or equal to]20 [[micro]meter]) as a consequence of the calculation of [T.sub.040] from three independent measurements, each requiring a separate alignment of the sample in the camera.
Analogous determinations of the parameter [T.sub.040] were also performed with selected PT55 samples. The profiles of [T.sub.040] presented in Fig. 8 exhibit the behavior as expected on the basis of the other orientation indices. At position B, one finds at low injection pressure (sample 1P) a single maximum, while at high pressure (sample 3P) a maximum and a broad shoulder can be observed. The positions of maxima and shoulder are about the same as found for the profiles of the index [A.sub.110]. A similar behavior is observed at position A. On the whole, the range with significant orientation is markedly broader at high injection pressure (sample 3P) than at low pressure (sample 1P). The height of the maxima in the [T.sub.040] profiles of the PT55 samples is much lower than the maxima in the corresponding profiles for the KS10 samples. This finding once more demonstrates the generally lower degree of orientation in the surface layers of the PT55 plates.
Light Optical and Mechanical Studies
The information extracted from the analysis of the various orientation parameters of the PT55 plates is corroborated and supplemented by polarization optical micrographs [ILLUSTRATION FOR FIGURE 9 OMITTED] from the same plates and positions as investigated by the X-ray technique. The micrographs clearly distinguish between different zones in the surface layer region of the samples from positions A. The micrographs for position B (not shown) are similar to those for position A. In position C, we find a differentiated behavior, most clearly to be seen when samples from plates 1P and 3P are compared. The sample from plate 1P exhibits the onset of spherulitic morphology in a depth of about 100 [[micro]meter] and sporadic occurrence of [Beta]-PP spherulites in deeper regions (about 300 [[micro]meter] below surface). The sample from plate 3P only shows a continuously growing granularity, beginning at a depth of about 200 [[micro]meter].
The results of the X-ray analysis of the KS10 and PT55 plates in terms of profiles of the orientation parameters [A.sub.110] and especially [T.sub.040] may also be compared with the profiles of birefringence and elongation at break (21, 22), which were determined from plates molded under the same or at least similar conditions as the plates investigated in the X-ray studies (compare [ILLUSTRATION FOR FIGURE 10 WITH FIGURES 7 AND 8 OMITTED]). This comparison reveals a high degree of accordance. In general, the maximum of [T.sub.040] is found to be situated between the maximum of birefringence and the minimum of elongation at break.
Concentration Parameters From WAXS
A comparison of the profiles of the Turner-Jones index B in Figs. 7 and 8 reveals differences in the distribution of [Beta]-PP in the cross section of KS10 and PT55 plates. Differences are found, e.g., in the general shape of the profiles (usually several peaks in the KS10 plates and single peaks in the PT55 plates) and in the height of the peaks (showing different dependence on injection pressure and distance from gate in the KS10 and the PT55 plates, respectively). By integration over the profiles of the index B, one can compare the different samples with respect to the total [Beta]-PP content in the cross section. In this way the highest values of total [Beta]-PP content were found both in the KS10 and the PT55 plates molded at medium. injection pressure (plates 2K and 2P), however, in plate 2P at position A and in plate 2K at position [Beta].
Another obvious difference between PT55 and KS10 plates concerns the occurrence of [Gamma]-PP. Measurable concentrations of this modification only appear in the surface layers of KS10 samples. While at low injection pressure (sample 1K) the values obtained for the concentration index G are all below 0.1, values of G up to about 0.6 are observed in plate 3K molded with the highest injection pressure. It may be noted that although [Beta]-PP and [Gamma]-PP can apparently coexist in the same region of the cross section (compare the profiles of indices B and G in [ILLUSTRATION FOR FIGURE 7 OMITTED]), the maxima of their distributions do not coincide. In the case of PT55 plates, the presence of small amounts of [Gamma]-PP in the surface layers is just indicated in the intensity maps of samples from position A [ILLUSTRATION FOR FIGURE 3 OMITTED]; the [Gamma](130) reflections are, however, too weak for a quantitative evaluation. The values of index G estimated for sample 3P certainly do not exceed a level of 0.1.
Apparent Crystallite Dimensions and Interplanar Spacings
That the structural organization in the surface layers of the KS10 plates is not as uniform as it seems to be from the [A.sub.110] profiles becomes evident when also apparent crystallite dimensions [L.sub.hkl] and interplanar spacings [d.sub.hkl] are included in the comparison of samples. For shortness, we only present results obtained for the samples from position B, annotating that the results obtained for samples from position A are not essentially different.
Figure 11 shows in the upper half a comparison of the profiles of dimensions [L.sub.110], [L.sub.040], and [L.sub.130] of [Alpha]-crystallites for all KS10 plates as well as some values obtained for [L.sub.130] of [Gamma]-PP and [L.sub.300] of [Beta]-PP. According to the profiles shown in the figure, the crystallite size varies with the distance from surface and in dependence on the injection pressure. In general, the values of [L.sub.040] are higher than those of [L.sub.110] and [L.sub.130], and all these parameters are in the surface layers larger than in the core region. With increasing injection pressure, the shape of the profiles changes; the most pronounced differences are observed between plates 2K and 3K. Apart from the changes in the surface layers, a decrease of crystallite dimensions in the core region with increasing injection pressure is also suggested by Fig. 11. The apparent crystallite dimension [L.sub.[Gamma]-130] obtained for the [Gamma]-crystallites in the surface layers of plate 3K is of similar size as the dimension [L.sub.130] of [Alpha]-crystallites in the core. It may be noted that the occurrence of enhanced concentrations of [Gamma]-crystallites in the surface layers of plate 3K is paralleled by a significant decrease of size [L.sub.110] of [Alpha]-crystallites in that region; [L.sub.130] and [L.sub.040] exhibit a maximum in the same region. In plates 1K and 2K, on the other hand, the zones of enhanced [Gamma]-PP concentrations overlap only partially with the regions where the [L.sub.130] values exceed the [L.sub.110] values.
The values obtained for the dimension [L.sub.300] of [Beta]-crystallites (data are only shown for the range of the main maximum of the index B) are generally larger than the values of [L.sub.040] of [Alpha]-crystallites.
Drastic variations with the distance from surface have also been found in the profiles of the apparent interplanar spacings [d.sub.hkl] of the KS10 plates, most pronounced with [d.sub.110] and less expressed with [d.sub.040]. particularly at medium pressure (plate 2K). In the lower half of Fig. 11, we present the profiles for [d.sub.110] and [d.sub.040]. One characteristic feature of these profiles is the very deep minimum of [d.sub.110] values in the surface layer region of samples 1K and 3K (corresponding to a decrease of the cell parameter a by up to about 1% as compared to the core region) and the less deep double minimum found for sample 2K. Another noteworthy feature is the shift of the position of the minimum form a depth of about 0.43 mm in sample 1K to a depth of about 0.18 mm in sample 3K, whereby the double minimum observed with sample 2K obviously represents an intermediate stage. The profiles of the distances [d.sub.hkl] are found to depend on the direction of the primary beam, however, to a different extent for the different parameters. This finding is not unexpected. It is in accord with the assumption that the deformation of the lattice cells in the surface layers of the plates is different in the different directions. A detailed analysis of the various [d.sub.hkl] profiles, in dependence on the direction, is in progress and will be presented elsewhere.
To complete our results we present in Fig. 12 profiles of the parameters [L.sub.hkl] and [d.sub.hkl] for the samples from position B of PT55 plates. The comparison with Fig. 11 shows that the variation of the crystallite dimensions with the distance from surface is less pronounced in the PT55 plates. Except for the plates molded with low pressure, the values of [L.sub.hkl] in the core region of the PT55 plates are similar to the values obtained for the KS10 plates in the same region, whereas in the surface layers of the PT55 plates lower [L.sub.hkl] values are observed than in the surface layers of the KS10 plates. It may be noted that the maxima of the [L.sub.040] profiles of the PT55 plates occur at somewhat larger depths than the maxima of the [L.sub.110] and [L.sub.130] profiles; the two latter profiles exhibit a shoulder on the core-side, most expressed at the highest pressure.
The variation of the parameters [d.sub.110] and [d.sub.040] with the distance from surface and with the injection pressure is also much less expressed in the PT55 plates. A shift of the minimum in the surface layers toward the surface with increasing pressure can also be observed with the PT55 plates.
DISCUSSION AND CONCLUSIONS
According to the results presented in the foregoing, the gross architecture of the layered structure in the plates of KS10 and of PT55 as well consists of a core of monotonously low degree of orientation, surrounded on both sides by a surface layer of distinctly higher degree of orientation. There are, however, a number of differences in the layered structures of KS10 and PT55 plates. The most significant differences, found from spatially resolved WAXS, are as follows:
1) Thickness and degree of orientation of surface layers
The surface layers of the KS10 plates are much thicker and exhibit a considerably higher degree of orientation than the surface layers of the PT55 plates. At large flow lengths, the degrees of orientation and of order (crystallinity) in the surface layers of the PT55 plates strongly decrease, whereas in the KS10 plates both order and orientation can be found to persist even at large flow lengths, however, at a lower level than near the gate or in the middle of the plates. While the thickness of the surface layers of the KS10 plates does not change considerably with the pressure, a significant increase of thickness of these layers with increasing pressure is clearly found for the plates molded from PT55; thus, at the highest pressures used in this study, the thickness of the surface layers of the PT55 plates approaches the values observed with the KS10 plates. According to the profiles of orientation parameters, the internal structures of the surface layers in KS10 plates and in PT55 plates are different, partly due to differences in the axial orientation of [Alpha]-PP crystallites.
Some similar statements concerning the thickness of surface layers and the orientation in the plates can also be made from the results of measurements of birefringence and/or elongation at break and by light microscopy.
It is plausible to explain the differences in the orientation behavior of the two polymer grades by the differences in their molar mass distribution. Due to the lower molar mass and narrower mass distribution, the relaxation times of PT55 are shorter than those of KS10. Since oriented states can relax in PT55 more readily than in KS10, the surface layers in PT55 plates are Thinner and less oriented than in KS10. That the thickness of the surface layer in PT55 plates increases considerably with increasing pressure can be explained by an additional orientation in the compression phase before the mold is mechanically sealed. Possibly this orientation by compression (22) occurs through pressure-induced formation of nuclei (see below) in the still oriented, unrelaxed melt.
2) Crystallite dimensions
The differences in crystallite size of [Alpha]-PP observed between the surface layers and the core are in the KS10 plates significantly larger than in the PT55 plates. The crystallite dimensions in the surface layers of the KS10 plates are larger than in the PT55 plates, and these differences become more pronounced with increasing pressure, whereas in the core regions the differences in crystallite size between KS10 and PT55 plates decrease and eventually disappear with increasing pressure. A decrease of crystallite size in the core with increasing pressure can be observed in the KS10 and the PT55 plates as well.
The pressure dependence of crystallite dimensions in the core may be explained by pressure-induced nucleation (22). It appears plausible that an enhanced concentration of nuclei in the core region may lead to smaller crystallites. The occurrence of a finer spherulitic texture in the core region at high pressures has been reported (22). The formation of larger crystallites in the surface layers might be due to the better orientation of the polymer chains by shear. Then the differences in the crystallite dimensions in the surface layers of KS10 and PT55 plates may again result from the different relaxation times of the two polymer grades. While the differences in the shape of [L.sub.110] and [L.sub.040] profiles of the PT55 plates [ILLUSTRATION FOR FIGURE 12 OMITTED] can be explained by variations of axial orientation (c, [a.sup.*]) of the [Alpha]-crystallites, the interpretation of the differences found in the KS10 plates [ILLUSTRATION FOR FIGURE 11 OMITTED] should also take into account the observed formation of [Gamma]-crystallites (see below). Neglecting an involvement of [Gamma]-crystallites in the shape of [L.sub.110], [L.sub.130], and [L.sub.040] profiles of KS10 plates, the behavior of these profiles in the inner region of the surface layers of plate 3K suggests that in this range the dimension [L.sub.040] is larger and [L.sub.110] is smaller than in the outer region of the surface layer. This finding could mean that the [Alpha]-crystallites are anisometric to a different extent. If one takes into account also the observed variation in the distribution of c- and [a.sup.*]-oriented [Alpha]-crystallites, one might, on the other hand, conclude that the size [L.sub.040] of [a.sup.*]-oriented crystallites exceeds that of c-oriented crystallites.
3) Occurrence of [Gamma]-PP
High pressures cause the enhanced formation of [Gamma]-PP crystallites in the surface layers of the KS10 plates, whereas in the PT55 plates a significant increase of [Gamma]-PP concentration at high pressure does not occur. That high pressure favors the crystallization of PP in the [Gamma]-lattice is well known (28, 31-33). The pronounced formation of [Gamma]-crystallites in the KS10 plate molded with the highest pressure is obviously accompanied by a decrease of the size of [Alpha]-crystallites along the crystallographic axis a. Since the unit cells of [Alpha]- and [Gamma]-PP are related, the aforementioned finding may be interpreted to suggest a pressure-induced transformation or deformation of [Alpha]-PP into [Gamma]-PP. The shorter polymer chains and shorter relaxation times of PT55 might explain why the PT55 plates do not exhibit a formation of [Gamma]-PP comparable to that in the KS10 plates.
For the following considerations, it must be borne in mind that the (040) reflections of [Alpha]-crystallites and [Gamma]-crystallites coincide; thus, the values of [L.sub.040] measured in the range of enhanced concentrations of [Gamma]-crystallites will also contain contributions from this crystallite type. In view of the relatively low values found for the dimension [L.sub.[Gamma]-130] of [Gamma]-crystallites [ILLUSTRATION FOR FIGURE 11 OMITTED] it is not very likely that the observed increase of [L.sub.040] values at the inner side of the surface layers reflects properties of the [Gamma]-crystallites. The interpretations given above (see Crystallite dimensions) become more likely.
4) Interplanar spacings
The variation of interplanar spacings with the distance from surface is more pronounced in the KS10 plates than in the PT55 plates. The observed variation of interplanar spacings is itself a noteworthy phenomenon, indicating a shear-induced deformation of the unit cell of [Alpha]-crystallites in the surface layer. It cannot be overlooked that in the KS10 plates 1K and 2K the minimum in the [d.sub.110] profiles at a depth of about 0.43 mm coincides with the maximum of [Gamma]-PP content, whereas the minimum of [d.sub.110] values at a depth of about 0.18 mm in the plates 2K and 3K lies at the margin of the region with enhanced concentrations of [Gamma]-PP. From these findings, we may conclude that the shear-induced stress in the surface layers of the KS10 plates, indicated by the minima of [d.sub.110] values, was effectively reduced at the inner side of the surface layers by the enhanced formation of [Gamma]-crystallites at higher pressures, resulting in a outward-shift of the minimum of [d.sub.110] values. In a sense, the deformation of the cell of [Alpha]-PP, reflected by the significant decrease of [d.sub.110] values at low pressure, may be seen as a precursor of the crystallization in the [Gamma]-PP lattice at high pressure.
The differences in the molecular properties, e.g., in the relaxation times, of the two polymer grades may again be held responsible for the, in comparison with KS10, less pronounced variation of [d.sub.hkl] values (less pronounced deformation of [Alpha]-PP cell) in the PT55 plates. This would also explain why in the PT55 plates even at the highest pressure no significantly enhanced [Gamma]-crystallization can be observed.
5) Distribution of [Beta]-PP
In the PT55 plates, regions with enhanced concentrations of [Beta]-PP lie much nearer to the surface than in the KS10 plates. Moreover, the amount of [Beta]-PP decreases faster with increasing flow length than in the KS10 plates. The crystallization of PP in the [Beta]-lattice is favored by shear and by temperature (26, 34-37). Since in the PT55 plates the surface layers are thinner, shear zones are closer to the surface than in the KS10 plates. On the other hand, the shorter relaxation times of PT55 explain the faster decay of [Beta]-PP content with the flow length.
The authors are very much indebted to the Austrian "Fonds zur Forderung der wissenschaftlichen Forschung" for sponsoring this work, first in the course of a national research program on injection-molded articles (projects S3304 and S3305) and later as project P7446. Thanks are due to PCD (formerly Chemie Linz GesmbH) for supplying the test material and to Maschinenfabrik Engel KG for assistance in preparation of the plates. The excellent assistance by Mrs. M. Brunegger, Mrs. Ch. Knabl, and Ing. E. Wrentschur in the preparation and/or investigation of samples and by Dr. P.M. Abuja in the development of computer programs is gratefully acknowledged.
Table 2. Parameters Derived from the Wide-angle Reflections.
[Mathematical Expression Omitted]
[Mathematical Expression Omitted]
C = [I.sub.040]/[I.sub.110] + [I.sub.040] + [I.sub.130] (3)
[T.sub.040] = 4.25/[[Sigma].sub.u] (4)
B = [I.sub.[Beta]-300]/[I.sub.[Beta]-300] + [I.sub.110] + [I.sub.040] + [I.sub.130] (5)
G = [I.sub.[Gamma]-130]/[I.sub.[Gamma]-130] + [I.sub.130] (6)
The Miller indices (110), (040), (130), (111), and (131 + 041) belong to the mono-clinic cell of [Alpha]-PP (25); [Beta](300) refers to the hexagonal cell B of [Beta]-PP (26), [Gamma](130) to the triclinic cell of [Gamma]-PP (27, see also Ref. 28). Since the (040) reflection of [Gamma]-PP superimposes that of [Alpha]-PP, Eq 3 and 5 were modified in the case of high concentrations of [Gamma]-PP by including the intensity [I.sub.[Gamma]]-130 in the denominator. The standard deviation [[Sigma].sub.u] of (040) pole density distribution is obtained from [I.sub.040] measured with three different directions of the primary beam (17).
1. M. R. Kantz, H.D. Newman, and F. H. Stigale, J. Appl. Polym. Sci., 16, 1249 (1972).
2. Z. Mencik and D. R. Fitchmun, J. Polym. Sci. Polym. Phys. Ed., 11, 973 (1973).
3. G. Menges, G. Wubken, and B. Horn, Colloid Polym. Sci., 254, 267 (1976).
4. B. Heise, L. Klostermann, and W. Woebcken, Colloid Polym. Sci, 260, 487 (1982).
5. S. S. Katti and J. M. Schultz, Polym. Eng. Sci., 22, 1001 (1982).
6. J. M. Schultz, Polym. Eng. Sci., 24, 770 (1984).
7. J. P. Trotignon and J. Verdu, J. Appl. Polym. Sci., 34, 1 (1987).
8. J. P. Trotignon and J. Verdu, J. Appl. Polym. Sci., 34, 19 (1987).
9. M. Fujiyama, T. Wakino, and Y. Kawasaki, J. Appl. Polym. Sci., 35, 29 (1988).
10. J. Koppelmann, E. Fleischmann, and G. Leitner, Rheol. Acta, 26, 548 (1987).
11. E. Fleischmann and J. Koppelmann, Kunststoffe, 77, 405 (1987).
12. P. Zipper, E. Wrentschur, A. Janosi, and W. Geymayer, Prague Meetings on Macromolecules, 10th Discussion Conference: Small-Angle Scattering and Related Methods, Prague, 1987. Programme, p. 37 (1987).
13. H. Janeschitz-Kriegl, P. Zipper, E. Wrentschur, J. Koppelmann, E. Fleischmann, G. Leitner, H. Muschik, M. Radax, and W. Geymayer, in: "Integration of Fundamental Polymer Science and Technology 2", P. J. Lemstra and L. A. Kleintjens, Eds., p. 477, Elsevier, London (1988).
14. E. Fleischmann, P. Zipper, A. Janosi, W. Geymayer, J. Koppelmann, and J. Schurz, Polym. Eng. Sci., 29, 835 (1989).
15. P. Zipper, A. Janosi, and E. Wrentschur, Osterr. Kunststoff-Z., 21, 54 (1990).
16. P. Zipper, A. Janosi, and E. Wrentschur, Sixth Annual Meet. Polym. Proc. Soc., Nice, 1990. Abstracts, 06-08 (1990).
17. P. Zipper, A. Janosi, E. Wrentschur, P.M. Abuja, J. Appl. Cryst., 24, 702 (1991).
18. P. Zipper, A. Janosi, E. Wrentschur, C. Knabl, and P.M. Abuja, Osterr. Kunststoff-Z., 24, 162 (1993).
19. P. Zipper, A. Janosi, E. Wrentschur, J. Phys. IV, 3, 33 (1993).
20. H. Janeschitz-Kriegl, ed., "Influence of Molecular Structure and Processing Parameters on the Properties of Moulded Plastic Parts," Final Report of the National Working Party S 33, Univ. Linz (1991).
21. E. Fleischmann and J. Koppelmann, Kunststoffe, 78, 453 (1988).
22. E. Fleischmann, Thesis, Univ. Leoben (1990).
23. W. Geymayer and E. Ingolic, Inst. Phys. Conf. Ser. No. 93, Vol. 2, p. 453 (1988).
24. R. A. Young and D. B. Wiles, J. Appl. Cryst., 16, 430 (1982).
25. G. Natta and P. Corradini, Nuovo Cimento, 15, 40 (1960).
26. A. Turner-Jones, J. M. Aizlewood, and D. R. Becket, Makromol. Chem., 75, 134 (1964).
27. D. R. Morrow and B. A. Newman, J. Appl. Phys., 39, 4944 (1968).
28. S. Bruckner, S. V. Meille, V. Petraccone, and B. Pirozzi, Progr. Polym. Sci., 16, 361 (1991).
29. F. L. Binsbergen and B. G. M. De Lange, Polymer, 9, 23 (1968).
30. P. Zipper, A. Janosi, and E Wrentschur, Osterr. Kunststoff-Z., 24, 106 (1993).
31. J. L. Kardos, A. W. Christiansen, and E. Baer, J. Polym. Sci. Part A-2: Polym. Phys., 4, 777 (1966).
32. K. D. Pae, D. R. Morrow, and J. A. Sauer, Nature, 211, 514 (1966).
33. J. He and P. Zoller, J. Polym. Sci. Part B: Polym. Phys., 32, 1049 (1994).
34. J. M. Crissmann, J. Polym. Sci. Part A-2: Polym. Phys., 7, 389 (1969).
35. H. Dragaun, H. Hubeny, and H. Muschik, J. Polym. Sci. Polym. Phys. Ed., 15, 1779 (1977).
36. J. Varga, F. Schulek-Toth, and A. Ille, Colloid Polym. Sci., 269, 655 (1991).
37. B. Fillon, A. Thierry, J. C. Wittmann, and B. Lotz, J. Polym. Sci. Part B: Polym. Phys., 31, 1407 (1993).
|Printer friendly Cite/link Email Feedback|
|Author:||Zipper, Peter; Janosi, Andras; Geymayer, Wolfgang; Ingolic, Elisabeth; Fleischmann, Ernst|
|Publication:||Polymer Engineering and Science|
|Date:||Feb 1, 1996|
|Previous Article:||Degradation of polyolefin pipes in hot water applications: simulation of the degradation process.|
|Next Article:||Structure and properties of biaxially stretched poly(ethylene terephthalate) sheets.|