Effects of SBS on Phase Morphology of iPP/aPS Blends.

I. SMIT [1]

G. RADONJIC [2]

The supermolecular structure of binary isotactic polypropylene/poly(styrene-b-butadiene-b-styrene) (iPP/SBS) and isotactic polypropylene/atactic polystyrene (iPP/aPS) compression molded blends and that of ternary iPP/aPS/SBS blends were studied by optical microscopy, scanning and transmission electron microscopy, wide-angle X-ray diffraction and differential scanning calorimetry. Nucleation, crystal growth, solidification and blend phase morphology are affected by the addition of amorphous components (SES and aPS). As a compatibilizer in immiscible iPP/aPS blends, SBS formed interfacial layer between dispersed honeycomb-like aPS/SBS particles and the PP matrix, thus influencing the crystallization process in iPP. The amount of SBS and aPS, and compatibilizing efficiency of SBS, determine the size of dispersed aPS, SBS, and aPS/SBS particles and, consequently, the final blend phase morphologies: well-developed spherulitic morphology, cross-hatched structure with blocks of sandwich lamellae and co-continuous mo rphology. The analysis of the relationship between the size of spherulites and dispersed particles gave the criterion relation, which showed that, in the case of a well-developed spherulitization, the spherulites should be about fourteen times larger than the incorporated dispersed particles; i.e. to be large enough to engulf dispersed inclusions without considerable disturbing of the spherulitic structure.

INTRODUCTION

Blending of immiscible polymers usually leads to multiphase systems characterized with different morphologies. The type of morphology depends on: composition [1, 2], viscosity ratio of the components [3-5], interfacial tension [6, 7], and processing conditions [4, 8, 9]. The introduction of polymeric interfacial agents, i.e., compatibilizers, modifies the interactions between different polymers and, additionally affects morphology.

Among immiscible thermoplastic polymer blends, ductile, semicrystalline isotactic polypropylene (iPP) can form interesting combinations with brittle, amorphous atactic polystyrene (aPS). Numerous studies have demonstrated the influence of different factors on the properties and phase morphology of iPP/aPS blends [10-21]. Owing to their immiscibility, PP/aPS blends showed a deterioration of some mechanical properties in comparison to the pure components [10-12]. Compatibility of these immiscible polymers can be improved by the addition of a suitable compatibilizer; i.e. block- or graft copolymers. Effective compatibilizer, located at the interface between two polymers, improves the interfacial adhesion and increases the stability against coalescence by reducing the interfacial tension [22, 23]. Poly(styrene-b-butadiene-b-styrene) (SBS) is an effective compatibilizer for iPP/aPS blends, and, in addition, it can be added separately to the pure iPP as an impact modifier. A limited number of studies are reported concerning the effect of SBS on nucleation and spherulitization in the binary iPP/SBS blends [24-27].

Recently, various investigations of ternary iPP/aPS/SBS blends have been described [28-42]. Santana and Muller [28] examined the effects of SBS triblock copolymer to the iPP/aPS blends with aPS as a matrix phase. They found no improvements in the tensile and the impact properties of such blends. A lower amount of SBS (2 wt%) did not change the size of dispersed PP particles, but reduced the homogeneous nucleation of iPP. Horak et al. [29-32] used di-, tri, and pentablock types of styrene/butadiene copolymers as compatibilizers for blends of high impact polystyrene (PS-HI) and iPP. Hlavata and Horak [29] showed that the addition of diblock (SB) and triblock (SBS) copolymers slightly reduced the degree of crystaflinity and preserved a-phase of IPP in iPP/ PS-HI blends. Rosenberg et al. [30] showed that the viscoelastic properties depended primarily on the blend composition. Horak et al. [31] also reported the influence of the PS-terminated multiblocks (in SBS and SBSBS) on the interfacial layer around disperse d PP particles. They found that some mechanical properties were improved compared to the blends in which diblock (SB) copolymer was added. In a recent paper, Horak et al. [32] tested the effect of three types of styrene-butadiene copolymers (SB, SBS, and SBSBS) on the gas permeability and the mechanical properties of iPP/aPS blends. Obieglo and Romer [33] demonstrated that the addition of a few percent of SBS could improve notched impact strength of recycled iPP/aPS blends. Recently, Fortelny et al. [5, 34] found that the admixture of SBS to iPP/aPS blends reduced the average size of the dispersed PS particles, but did not affect uniformity of the phase structure. D'Orazio et al. [35] showed how the addition of unsaturated polypropylene copolymer grafted with styrene (uPP-g-aPS) to iPP/aPS blends (prepared by solvent-casting method) significantly changed their phase morphology. In order to obtain more general and comparable results, Radnjic et al. [21, 36-42] have undertaken systematic investigations on the c ompatibilizing and structural effects of the different styrenic/rubber block copolymers in iPP/aPS blends. They compared the compatibilizing effects of poly(styrene-b-ethylene-co-propylene) (SEP) [21, 36-39], poly(styrene-b-butadiene-b-styrene) (SBS) [21, 37-42], and poly(styrene-b-ethylene-co-butylene-b-styrene) (SEBS) [21, 38] in iPP/aPS blends.

In this paper we present the results of a study on the influence of SBS on the phase structure of iPP/aPS blends. The structural effects of SES and aPS, phase morphology of ternary iPP/aPS/SBS blends were compared to the binary iPP/aPS and iPP/SBS blends. This research work is the continuation of the previous studies, which described the compatibilization effect of SBS in iPP/aPS blends [40, 42].

EXPERIMENTAL

Materials

The polymers used in this study were Novolen 1100L polypropylene (BASF, Germany), polystyrene GP-678E (DOKI, Croatia), and poly(styrene-b-butadiene-b-styrene) Kraton D-1102 CS (Shell Chem. Co.) with polystyrene/poly-butadiene weight ratio 29/71. Polymer characteristics are shown in Table 1.

Blend Preparation

aPS pellets were dried overnight at 70[degrees]C before use and premixed with iPP and SBS pellets before being fed into the kneading chamber. Blends of the different compositions were prepared by melt blending in an oil heated Brabender kneading chamber at 200[degrees]C for 6 min with a rotor speed of 50 rpm. After finishing the blending process, they were transferred rapidly between two aluminum sheets placed in the preheated hydraulic press at 220[degrees]C. Samples of blends used for structural investigations were prepared by compression molding. A load of 100 bar was used, and after 10 min, 1-mm-thick and 4-mm-thick plates were removed and cooled to room temperature in air. The weight ratios of iPP and aPS in binary iPP/aPS blends were 97.5/2.5, 95/5, 90/10, 70/30, and 50/50. The iPP/SBS blends were prepared with weight ratios 97.5/ 2.5, 95/5, 90/10, and 70/30. Compatibilizer was used in iPP/aPS/SBS blends with 90/10, 70/30, and 50/50 iPP/aPS weight ratios with the concentrations of 2.5. 5, and 10 weight percent (wt%).

Optical Microscopy

Thin cross microtomed sections of the 1-mm-thick plates were examined with a Leitz Orthoplan microscope at crossed and parallel polars. Compression force in polarizing and optical micrographs took the lateral direction. The particle diameters, [d.sub.p]' of the aPS or aPS/SBS phase or channels width were measured on several optical micrographs of each sample with micrometer scale and quantified as a number average particle diameter [d.sub.p] by Eq 1:

[d.sub.p] = [sigma][N.sub.i] * [d.sub.t]/[sigma][N.sub.i] (1)

where [N.sub.i] is the number of the particles with the diameter [d.sub.i].

The average diameter of the spherulites was measured on several polarization micrographs of each sample and was quantified as a number average spherulite diameter [d.sub.sph] by Eq 2:

[d.sub.sph] = [sigma][N.sub.i] * [d.sub.i]/[sigma][N.sub.i] (2)

where [N.sub.i] is the number of the measured spherulites with the average diameter [d.sub.i].

Scanning Electron Microscopy (SEM)

A Jeol JSM-840A scanning electron microscope was used for studying the morphology. Samples were fractured in liquid nitrogen and covered with gold before being examined with the microscope at an acceleration voltage of 10 kV. All SEM micrographs are secondary electron images.

The particle size of the dispersed aPS phase was determined from several SEM micrographs with the micrometer scale. Five hundred to fifteen hundred dispersed particles were counted for the determination for each studied blend. The particle diameter was determined as shown previously [39]. The diameter of the dispersed phase [d.sub.p] was quantified through Eq 1.

Transmission Electron Microscopy (TEM)

Ultrathin sections (approximately 70 nm thick) were cut at room temperature from 4-mm-thick plates with Reichert-Jung Ultracut E microtome equipment with a diamond knife. Before microtoniing, samples were exposed first to [OsO.sub.4] vapor for 3 days. After that, overnight exposure to [RuO.sub.4] was performed because of the additional contrasting and hardening of the samples. Microtomed ultrathin sections then were placed on Cu grids and micrographs were taken at an acceleration voltage of 80 kV with a Philips 3000 microscope.

Wide-Angle X-Ray Diffraction (WAXD)

The wide-angle X-ray diffractograms of the rotated samples (1-mm-thick plates) were taken by a Philips diffractometer with monochromatized CuK[alpha] radiation in the diffraction range 2[theta] = 4-50[degrees]. Because of the broadened diffraction contribution from the aPS and SBS amorphous phases, degree of crystallinity, [w.sub.c,x], was evaluated by the Hermans-Weidinger method [43], using an angular range of 2[theta] = 6-34[degrees]. Theoretical values of degree of crystallinity ([w.sub.c,calc]) were calculated by the additivity rule [44]:

[w.sub.c,calc] = [w.sub.c,x,1] * [f.sub.1] + [w.sub.c,x,2] * (1 - [f.sub.1]) (3)

where [w.sub.c,x,1] and [w.sub.c,x,2] are crystallinities of polymers 1 and 2 in the blend and [f.sub.1] is the weight fraction of polymer 1.

Orientation parameters [A.sub.110] and C, used as a measure for orientations of corresponding (110) and (040) planes, were calculated by formulas proposed by Trotignon and Verdu [45] and Zipper et al. [46] and defined by the following equations:

[A.sub.110] = [I.sub.110]/[I.sub.110] + [I.sub.111] + [I.sub.131+041] (4)

C = [I.sub.040]/[I.sub.110] + [I.sub.040] + [I.sub.130] (5)

where I represents the intensities of the corresponding reflections.

From half-maximum width of 110 and 040 reflections the crystallite sizes [L.sub.110], and [L.sub.040], by using the Scherrer formula [47] were calculated after the correction for instrumental broadening with a 111 germanium diffraction profile.

Differencial Scanning Calorimetry (DSC)

Thermal analysis was performed with a Perkin Elmer DSC-7 calorimeter. The samples (9-12 mg) were cut from 1-mm-thick compression molded plates and encapsulated in aluminum pans. The samples were heated in a nitrogen atmosphere from room temperature to 200[degrees]C under a controlled heating rate of 20[degrees]C/min. The melting temperatures ([T.sub.m]) of iPP were obtained from the maximum of the melting peaks and enthalpies of melting ([delta]h) from the melting peak area, respectively. The crystallinity ([w.sub.x,h]) of iPP and of the blends was calculated by Eq 6:

[w.sub.c,h] = [delta]h/[delta][[h.sup.0].sub.PP] X 100 (6)

where [delta]h is the enthalpy of fusion per gram of the sample and [delta][[h.sup.0].sub.PP] the enthalpy of fusion per gram of 100% crystalline iPP. For [delta][[h.sup.0].sub.PP] we used the value of 148 J/g [48]. Theoretical values of [w.sub.c,calc] were calculated by the additivity rule [44] by replacing [w.sub.c,x] with [w.sub.x,h] (Eq 3).

RESULTS

Optical Microscopy Observations

Polarizing micrographs of the compression molded pure iPP are shown in Fig. 1. It is evident that skincore morphology in pure iPP is absent. Micrographs of plain iPP reveal spherulitic structure with well-developed positive radial (type I) [alpha]-spherulites [49, 50] ([d.sub.sph] = 45 [micro]m as shown in Table 2) over the entire cross section of the sample.

Polarization and optical micrographs of the binary iPP/aPS blends exhibit also the uniform morphology without skin-core effect (Fig. 1). The addition of aPS to iPP dramatically changes the morphology of the iPP matrix. The addition of small amount of aPS (2.5 and 5.0 wt%) to PP causes the formation of well-developed [alpha]-spherulites with [d.sub.sph] = 57 and 56 [micro]m, respectively (Table 2).

Polarization and optical micrographs of the binary iPP/SBS blends, shown in Fig. 2, exhibit also the morphology without skin-core effect. Well-developed spherulites can be seen up to 10 wt% of SBS with [d.sub.sph] = 53-60 [micro]m (Table 2). The large spherulites' morphology seems to disappear (at least it is not clearly recognizable) and the average dispersed particles' diameter increases to [d.sub.p] = 10 [micro]m (Table 2) by adding 30 wt% of SBS.

In comparison to the 95/5 iPP/aPS blend, significantly greater aPS particles in 90/10 iPP/aPS belnd ([d.sub.p] = 11 [micro]m) disturb the growth of the regular spherulites so that a well-developed spherulitic structure cannot be recognized (Fig. 3). The morphology of this blend, and the size of the dispersed particles, are similar to those of 70/30 iPP/SBS blend. When SBS compatibilizer is added to 90/10 iPP/aPS blend, typical morphology with well-developed spherulities with [d.sub.sph] = 85-99 [micro]m (Table 2) appears again as in iPP/aPS blends with the lower aPS contents (Fig. 3). It can be explained by the smaller size of the dispersed particles in ternary iPP/aPS/SBS blends than is found in the binary 90/10 iPP/aPS blend. In ternary iPP/aPS/SBS blends with a 90/10 iPP/aPS weight ratio, spherulites become more diffuse with coarsened inclusions and appear like ring spherulites. The most diffuse spherulitic structure in the blend with 10 wt% SBS can be explained by the considerable content of the amorphous (aPS plus SBS) phase.

In the 70/30 iPP/aPS blend, dispersed spherical aPS particles become so large that they hinder the regular growth of the spherulites, which are transformed into morphology with the cross-hatched bundles of the sandwich lamellae (Fig. 4) as reported for iPP/SEBS blends [51]. The addition of SBS (Fig. 4) considerably reduces the dispersed particle size in ternary iPP/aPS/SBS blends with 70/30 iPP/aPS weight ratio in comparison to the binary 70/30 iPP/aPS blend. However, this particle size reductions are still not enough to cause possible regular spherulitization. Only 70/30 iPP/aPS sample with 2.5 wt% of SBS added shows hardly noticeable diffuse large spherulites with [d.sub.sph] = 92 [micro]m (Table 2). It seems that this blend is a limit of the possibility for arising of the recognizable large spherulites structure.

Figure 5 shows the incompletion of the phase co-continuity of binary 50/50 iPP/aPS blend. Dispersed aPS particles coalesced into three-dimensionally co-continuous network with the average width of the co-continuous channel-like domains of 40 [micro]m (Table 2). The spherical dispersed aPS particles still exist in iPP phase beside the co-continuous channels (see also SEM micrographs in Ref. 40). The incompletion of the phase co-continuous morphology in this blend is in accordance to the observations that in 50/50 iPP/aPS blends aPS can appear as the dispersed phase [5, 12, 15]. Addition of SBS block copolymer causes significantly narrower co-continuous channel-like domains of both phases (16-25 [micro]m) and, as a consequence, co-continuous phase morphology is almost completely developed in ternary iPP/aPS/SBS blends with 50/50 iPP/aPS weight ratio. Co-continuous channel-like domains tend to orient perpendicular to the compression molding direction which takes lateral direction in Fig. 5. Further, small iPP s pherulites arise in the channel-like domains of iPP phase.

Electron Microscopy Obsevations

Figure 6 shows the morphology of the cryogenically fractured samples with iPP/aPS weight ratio of 90/10 obtained by scanning electron microscopy (SEM). SEM micrographs of iPP/aPS blends with the weight ratios of 70/30 and 50/50 (Fig. 6a) and as presented in our previous papers [36, 40, 42] show aPS dispersed phase in JPP matrix indicating their immiscibility. The addition of 10 wt% of SBS copolymer substantially decreases the size of the dispersed particles (Fig. 6b) confirming the compatibilizing activity of SBS in iPP/ aPS blends. Moreover, dispersed particles become more uniform in size compared to noncompatibilized blend (Fig. 6a).

Figures 7(a-b) show the morphology of iPP/aPS blends with weight ratio 70/30 compatibilized with 10 wt% of SBS obtained by transmission electron microscopy (TEM). Unsaturated blocks of SBS, that is, polybutadiene (PB) blocks, were selectively contrasted by [OsO.sub.4] and [RuO.sub.4] vapor, and appear black in the TEM micrographs. PS blocks of SBS and pure aPS appear brighter. Dispersed particles in iPP matrix (Fig. 7a) are, actually, the complex aPS/SBS aggregates with well-developed internal honeycomb-like structure. SBS encapsulates and joins together small aPS particles thus forming a complex dispersed aPS/SES aggregates and surrounds each particle at the interface with iPP matrix. In the previous papers [38, 40, 42] the honeycomb-like morphology of ternary iPP/aPS/SBS blends in more detail was described. Higher magnification of this blend phase morphology shows the internal structure of the dispersed aPS/SBS aggregates (Fig. 7b). Figure 7b also revealed more or less a perpendicular orientation of the bl ocks of iPP stacked sandwich lamellae to the surface of the dispersed aPS/SES aggregates. The transcrystal lamellar orientation, probably caused by interaction with interfacial SBS layers, causes interconnection of two dispersed aPS/SES particles by sandwich lamellae, which continuously pass from particle to particle. Beside the interconnection of the aPS/SBS particles with transcrystal lamellae, interrupted transcrystal lamellae as well as lamellae passing by these particles also might exist. With the increase of the dispersed particles size, i.e. with the increase of aPS content, the number of transcrystal lamellae, interrupted at particle surfaces, increases. The orientation effect of SBS layers and the increase of dispersed particles size transform proposed structure of the cross-hatched lamellae into thick blocks of stacked sandwich lamellae. This supermolecular structure of iPP matrix will have to be further investigated by the other methods.

Wide-Angle X-ray Diffraction

Diffractograms of the pure iPP and all blends taken by wide-angle X-ray diffraction (WAXD) exhibit only stable monoclinic, [alpha]-form of iPP crystalline phase as reported previously [39-40]. Although small amounts of added aPS (up to 5 wt%) and SBS (at least up to 10 wt%) promote the spherulitization in iPP matrix, their interactions with iPP matrix seem to be different While the addition of aPS does not change the intensity relationships of ce-phase reflections (see also ref. 40 and 42), even 2.5 w% of SBS considerably intensifies 110 peak and depresses 040 peak of a-iPP in all blends. As a result, relatively low value of orientation parameter [A.sub.110] for iPP/aPS blends (defined by Eq 4) increases with the addition of SES to the plain iPP or to iPP/aPS blends (Fig. 8). It is interesting that [A.sub.110] values decrease in the binary IPP/SBS and increase in the ternary iPP/aPS/SBS blends with increasing SBS content except for blends with 2.5 w% of SBS. It seems that the interactive effect between SBS a nd iPP becomes stronger either with the decrease of the individual SBS particles size in iPP (nucleating effect) or with SBS layer thickening to certain thickness of the interfacial layers in ternary iPP/aPS/SBS blends. The decrease of [A.sub.110] values for blends with 2.5 wt% SBS (curve 8b) indicates a greater orientation efficiency of the tiny SES particles in iPP (97.7/2.5 iPP/SBS) than that of thin SBS layers which envelope aPS/SBS particles in ternary iPP/aPS/SBS blends.

The orientation parameter C (defined by Eq 5) exhibits a completely opposite trend as shown in Fig. 9. All blends with SBS show considerably lower C values in comparison to the pure iPP and to the binary iPP/aPS blends. The addition of aPS to iPP exhibit a slight decrease of C values in comparison to the pure iPP (curve 9a). Binary iPP/aPS/SBS and ternary iPP/aPS/SBS blends exhibit a very low C values and very close curves (curves 9c-d). These C values decrease and above 70/30 iPP/aPS weight ratio increase.

Figures 10 and 11 show the iPP/aPS weight ratio dependence of the [L.sub.110] and [L.sub.040] crystallite sizes, respectively. The growth of the crystallites is more influenced by the addition of aPS than SBS as already reported in ref. 40. There are several factors that may affect crystallites growth. The addition of aPS to the plain iPP (curves 10a and 11a) causes two opposite effects. According to ref. 16, 24, 25, 28 and 52, low amounts of aPS or SBS can act as a nucleating agent, thus increasing a heterogenous nuclei density. In fact, low amounts of aPS (up to 10 wt%) decreases crystallite sizes, especially [L.sub.040] crystallite size. Low amount of SES (up to 5 w%) in the binary iPP/SBS blends shows analogous crystallite size decrease (see curves 1 5a and 1 6a in ref. 40). The solidification effect, which enhances crystal growth, seems to appear (more expressed for [L.sub.110]) at the higher amount of aPS and SBS (above 10 wt%). Obviously, nucleation and solidification are manifested in different inte nsity of [110] and [040] directions of crystal growth. Generally, the ternary iPP/aPS/SBS blends show a uniform increase of [L.sub.110] as well as [L.sub.040] crystallite sizes with the addition of aPS (curves 10b-d and 11b-d). The presence of SBS additionally influences the crystal thickening.

While the experimental crystallinity values, [w.sub.c,x'] (determined by WAXD) of the binary iPP/aPS blends decrease according to the additivity rule, the crystallinity of the ternary iPP/aPS/SBS blends exceeds the values calculated by the additivity rule [w.sub.c,calc] (calculated by the Eq 3) from 3% to 5% (Table 3). Although such small exceeding of [w.sub.c,x] values could be considered as a value near the resolution limit of the WAXD method, the systematic increase of [w.sub.c,x] with the addition of SBS indicates that SBS affects the crystallization process in iPP. SBS interlayers or individual dispersed SBS particles in iPP matrix affect iPP crystallization in ternary blends most probably through interaction of PB blocks of SBS with iPP chains. Even 2.5 w% of SBS can promote the nucleation and spherulitization. The latter was confirmed by optical microscopy (Figs. 2-4).

Thermal Properties

Differential scanning calorimetry (DSC) measurements were performed only by first heating in order to have the same thermal history as samples used for WAXO and OM determinations. Otherwise, the comparison of [w.sub.c] values obtained by different experimental methods is questionable. All thermograms exhibit only one smooth profile of endotherm, which corresponds to the melting of monoclinic [alpha]-form of iPP crystalline phase [21]. The enthalpies of melting ([delta]h) were calculated from the integration of the melting peak areas in the thermograms.

For the comparison with WAXD measurements, [w.sub.c] values were determined from the enthalpy values and by the use of Eq 6. Theoretical values of [W.sub.c,h] were determined by Eq 3. For [delta][[h.sup.0].sub.pp] the value of 148 J/g was used [48]. Values for [w.sub.c], determined by WAXD and DSC methods, are summarized in Table 3. The experimental and the calculated [w.sub.c,h] values, determined by DSC analysis, are higher compared to the [w.sub.c,x] values determined by WAXD analysis. Although the different [w.sub.c] values are caused by the characteristics of methods used, the similar trend of [w.sub.c] decreasing is obtained. It can be seen that [w.sub.c,exp] and [w.sup.c,calc] decrease with the increasing amount of the amorphous components in the blends irrespective of the experimental methods used.

DISCUSSION

Size of Dispersed Particles

The size of the dispersed aPS, SBS and aPS/SBS particles increases with the increase of aPS or SBS contents. In the blends with 50/50 iPP/aPS weight ratio it causes the appearence of the co-continuous morphology. The increase of the diameter of the dispersed aPS particles in iPP/aPS blends is higher than the increase of the dispersed SBS particles in iPP/SBS blends with added (aPS or SBS) components.

Both optical and electron microscopy observation show a considerable compatibilizing effect of SBS block copolymer for reducing the dispersed honeycomb-like aPS/SBS particles diameter. The analysis of the SEM micrographs of the binary 70/30 iPP/aPS blend gives a very broad particle size distribution (Fig. 12a). Especially big dispersed particles (up to 50 [micro]m in diameter) in this blend take random distribution. Even with the addition of 2.5 wt% of SBS to the 70/30 iPP/aPS blend the particle size distrubution becomes significantly narrower (Figs. 12b). Maximal dispersed particle size is reduced approximately five-times for all SBS contents. A more uniform dispersed particles distribution is obtained in ternary iPP/aPS/SBS blends as a consequence of the compatibilization activity of SBS.

Figure 13 represents the dependence of the number average diameter ([d.sub.p]) of the dispersed particles on the SBS concentration for blends with 90/10 and 70/30 iPP/aPS weight ratios. values were determined from the SEM micrographs shown in the previous papers [38, 40]. The reduction of [d.sub.p] values of the dispersed particles in 90/10 iPP/aPS blends is less pronounced by the SBS addition than in 70/30 iPP/aPS blends. The addition of 5 wt% of SES to 90/10 iPP/aPS reduces [d.sub.p] about two-times (from 1.2 [micro]m to 0.6 [micro]m), whereas in 70/30 iPP/aPS blend for more than five-times. In 70/30 iPP/aPS blends major effect of [d.sub.p] reduction is obtained by the addition of the first 2.5 wt% of SBS copolymer. Plateau of the curves in Fig. 13 beyond 5 wt% of SBS content indicates no significant reduction of [d.sub.p] values. Some authors propose that the concentration of the compatibilizer beyond no reduction effects on dispersed particles size corresponds to the interfacial saturation by the block co polymer molecules [23, 53, 54].

In distinction from the scanning electron microscopy (SEM), optical microscopy (OM), covers narrower dispersed particles size distribution (Table 2), but the particle sizes exhibit a similar trend of change. Optical microscopy confirms that the influence of SBS on the reduction of aPS particles size is the most expressed by adding up to 5 wt% of SBS.

Spherulite Morphology

Uniform morphologies with well-developed spherulites, like those in plain iPP, arise in iPP/aPS blends at least up to 5 wt% of the added aPS. in iPP/SBS blends, at least, up to 10 wt% of the added SBS, and in all ternary iPP/aPS/SBS blends with 90/10 iPP/aPS weight ratio. The analysis of the spherulite size shows their relatively narrow size distribution for all the samples (Table 2). It shows also the considerable increase of the number average spherulite diameter ([d.sub.sph]) in the binary iPP/aPS and iPP/SBS blends in comparison to [d.sub.sph] value in pure iPP. In iPP/SBS blends with SBS amount up to 25 wt%, iPP matrix crystallizes exclusively through heterogeneous nucleation [24]. In such a case, the spherulite size primarily depends on the heterogeneous nuclei density. Small aPS, SBS and aPS/SBS particles can act as a nucleation agent, and consequently, the decrease of the spherulites size with the addition of aPS and SBS it should be expected, as was reported by different authors [16, 24-26, 52]. Opp osite observed effect of the spherulite size increase at the chosen compression molding conditions in our experiments observed, even in blends with low amounts of aPS and SBS, can be caused by different factors which compensate aPS and SBS hetereogenous nucleation ability in crystallization process:

* according to Han, et al. [10-12] the addition of low amounts of aPS and SBS can decrease the melt viscosity increasing thus chains mobility at the same temperature (plastification effect);

* aPS and SBS particles oclude heterogenous nuclei and decrease heterogenous nuclei density;

* the interactions at the iPP-aPS and iPP-SBS interfaces affect spherulite growth;

* compatibilizing effect of SBS (SBS makes migration of iPP chains from the melt easier during the solidification process in the blend).

The reduction of the dispersed particles size in the binary 90/10 iPP/aPS blend by the introduction of SBS leads to the appearance of the spherulites which are considerably larger than that in a pure iPP as well as in binary iPP/aPS and iPP/SBS blends. The development of significantly bigger spherulites in ternary iPP/aPS/SBS blends with 90/10 iPP/aPS weight ratio can be explained by superimposing the following factors to the mentioned effects:

* SBS compatibilizer diminishes coalescence and consequently reduces dispersed particles diameter (see Figs. 12 and 13, and Table 2). Smaller dispersed particles enable regular growth of radial lamellae and well-developed spherulites to a larger extent. aPS and SBS affect the cooling rate of blends melt and the iPP chains mobility influencing prolonged crystallization process in iPP (solidification effect);

* SES layers can isolate the aPS/SBS particles with the heterogeneities, thus decreasing the heterogenous nuclei density.

Higher Young's modulus (see Fig. 20 in ref. 40) for ternary iPP/aPS/SBS blends with 90/10 iPP/aPS than for other ones with 70/30 and 50/50 iPP/aPS weight ratios can be explained by a well-developed spherulitic structure.

Relationships Between Spherulites and Dispersed Particle Size

Phase morphology of the ternary iPP/aPS/SBS blends, as well as the binary iPP/aPS and iPP/SBS blends, depends, in the first place, on dispersed aPS, SBS, and aPS/SBS particles size. It seems that well-developed spherulites can arise in the crystallization process of the iPP matrix until they can engulf dispersed aPS particles incorporated in iPP matrix. When aPS particle size reaches some critical limit value above which hypothetical iPP spherulites cannot engulf aPS particles, they hinder the regular growth of the long radial lamellae as well as the development of regular, well-developed spherulites.

Very diffuse spherulites (as seen in polarization micrograph of the 70/30 iPP/aPS blend with 2.5 wt% of SBS in Fig. 4) can be considered as a limit case of arising the well-developed spherulitic structure. In all these blends [d.sub.p] values do not exceed 5 [micro]m (see Table 2). If the dispersed particles size is considered as a limiting factor for the arising of well-developed spherulites, the criterion for their arising can be estimated by the comparison of the spherulites' and dispersed particles' diameters in 70/30 iPP/aPS blend with 2.5 wt% SBS only. The problem of the determination of the right criterion value is a very distinctive distribution width of the spherulites compared to the dispersed particles' distributions = ([d.sub.sph,max]/[d.sub.sph,min] = 1.6; [d.sub.p,max]/[d.sub.p,min] = 33.3 from SEM micrographs; [d.sub.p,max]/[d.sub.p,min] = 3 from OM micrographs). It means, by comparing the value of a number average spherulites' diameter ([d.sub.sph] = 92 [micro]m; see Table 2) wi th the value of the number average particles' size diameter estimated from SEM micrographs ([d.sub.p] = 2 [micro]m; see Fig. 16) one can obtain the value [d.sub.sph]/[d.sub.p] = 92/2 = 46. In order to achieve the most reliable criterion value for the possibility of well-developed spherulites to appear, the number average particles' diameter inside distribution width of the spherulites [d.sub.sph,max]/[d.sub.sph,min] = 1.6 should be considered. If [d'.sub.p] is defined as the number average particles' diameter inside distribution width of the spherulites [d.sub.sph,max]/[d.sub.sph,min] = 1.6 (for the case of 70/30 iPP/aPS blend with 2.5 wt% SBS observed by OM) the distribution range [d.sub.p,max]/[d'.sub.p,min] = 1.6 can be calculated from following data: [d.sub.p,max] = 10 [micro]m and [d'.sub.p,min] = 6.25 [micro]m. [d'.sub.p], defined in such manner, has, therefore, the value of 6.5 [micro]m. Thus, the relationship between the number average spherulites' diameter ([d.sub.sph] = 92 [micro]m) and the number av erage particles' diameter ([d'.sub.p] = 6.5 [micro]m gives the value [d.sub.sph]/[d'.sub.p] = 92/6.5 14. The following criterion relation is then:

[d.sub.sph]/[d'.sub.p] [greater than or equal to] 14 (7)

This means that the spherulites should be about fourteen-times bigger than the incorporated dispersed aPS, SBS or aPS/SBS particles to be able to engulf these inclusions without significantly disturbing the large spherulitic structure. To develop the large spherulites the effective compatibilizer should be able to diminish coalescence in the blends in such a way so that the resulting dispersed particles' size is, at least, fourteen-times smaller than the hypothetical spherulites' size. The comparison of the maximal values of the spherulites and particles diameters ([d.sub.sph,max]/[d.sub.p,max]) in Table 2 provides the ratios equal or higher than fourteen for all blends with well-developed spherulites. In 70/30 iPP/aPS blend with 2.5 wt% SBS the relationship value of the upper limits is a bit smaller ([d.sub.sph,max]/[d.sub.p,max] = 115/9 = 12.8). It seems, from this value (12.8). that the regular spherulitization depends on the geometric factors according to the relation:

[d.sub.sph,max]/[d.sub.p,max] [greater than] 4[pi] or [d.sub.sph,max] [greater than] 4[pi] [d.sub.p,max] (8)

Therefore, for regular spherulitization, the diameter of maximal spherulite should be greater than four circumferences of included maximal dispersed particle.

Taking into account the hindering geometric factor and the maximal particles diameters [d.sub.p,max],it is clear why 90/10 and 70/30 iPP/aPS blends, as well as 70/30 iPP/SBS blend, cannot give well-developed spherulitic structures. If the maximal particle diameters [d.sub.p,max] are 20 [micro]m for 90/10 iPP/aPS and 50 [micro]m for 70/30 iPP/aPS, the maximal diameters of the spherulites for engulfing such dispersed particles should be [d.sub.sph,max] = 20 X 14 = 280 [micro]m 90/10 iPP/aPS blend and [d.sub.sph,max] =14 X 50 = 700 [micro]m in 70/30 iPP/aPS blend, respectively. Also, 70/30 iPP/SBS blend with = 18 [micro]m requires big spherulites with maximal diameter of 252 [micro]m.

Crystallinity, Crystallite Size and Orientation

The increase of [A.sub.110] values and decrease of C values with the addition of SBS to iPP and iPP/aPS indicate the increase of the degree of c-axial orientation and a simultaneous decrease of the [a.sup.*]-axial orientation [46]. According to Lovinger [55], if [a.sup.*] corresponds to the axial direction of the lamellar growth, the number of [a.sup.*]-axis- oriented lamellae in planes parallel to the sample surface decreases. The increase of c-axial orientation indicates the increase of the number of c-axis-oriented lamellae in the planes parallel to the sample surface [56]. Such orientation implies the possibility of preferential transcrystalline growth of sandwich lamellae. These findings are in good accordance to the TEM micrograph (Fig. 7b) which revealed the trend of perpendicular orientation of PP sandwich lamellae to the SBS layers on the surface of the dispersed aPS/SBS particles.

All blends with large spherulites, i.e., binary iPP/aPS and iPP/SBS, as well as ternary iPP/aPS/SBS blends with 90/10 iPP/aPS weight ratio, exhibit a relativelly small and similar [L.sub.110] crystallite size values. Small amounts of aPS and SBS may act as a crystallite nucleation agent as well as the promoter of the spherulite growth (see Table 2). The many very tiny dispersed aPS and SBS particles seem to act as the centers for crystallite nucleation, but do not act as the centers for spherulite nucleation. Nucleation effect is

more expressed at the [L.sub.040] crystallite size.

The increase of the [L.sub.110] and [L.sub.040] crystallite sizes with further addition of aPS (especially for ternary iPP/aPS/SBS blends with 70/30 and 50/50 iPP/aPS weight ratios) seems to be influenced by the solidification rather than by the phase morphology of the blend systems: well-developed spherulites, blocks of sandwich lamellae or co-continuous morphology.

The addition of the SBS to iPP/aPS blends influences the crystal thickening additionally. In spite of the reported ability of transferring heterogeneities to the PP matrix in the phase of nucleation [28], SES layers can isolate aPS/SBS particles with included heterogeneities to some degree. During the crystal growth. interactions of SBS with iPP segments at iPP-SBS interfaces can affect crystal thickening additionally to the effect of solidification. It is interesting that [L.sub.110] crystallite size uniformly increases, while [L.sub.040] crystallite size decreases with the increase of the SBS content, particularly in the case of 2.5 wt%. The uniform crystal thickening in [110] direction, affected by SBS interlayer, suits well with [A.sub.110] values behavior. [L.sub.040] crystallite size decreases with the higher SBS contents in binary PP/SBS blends most probably from the same reason as orientation and crystal growth are favored in [110] direction in comparison to [040] direction.

Higher crystallinity values of the ternary PP/PS/ SBS blends (from 3% to 5%) indicate obvious effect of SBS interlayers or individual dispersed SBS particles on crystallization process in iPP matrix, probably through the interaction of PB blocks of SBS with PP chains. It is in accordance to the crystallite orientation and size behavior, too. Even 2.5 wt% of SES can promote the nucleation and spherulitization. The latter was confirmed by optical microscopy (Figs. 2-4).

Blend Phase Morphology

Size of the dispersed aPS, SBS, and aPS/SBS particles, affected by aPS and SBS amounts and by the compatibilizing efficiency of SBS, dominantly influences the final phase morphology of PP matrix in all blends. Compression molded iPP/aPS, iPP/SBS, and iPP/aPS/SBS blends in investigated composition ranges exhibit three different global phase morphologies as illustrated in Fig. 14:

* [d.sub.sph]/[d'.sub.p] [greater than] 14

If the dispersed particles are fourteen-times smaller than the spherulites, well-developed spherulites arise (Fig. 14a). Well-developed spherulitic structure appears by adding aPS at least up to 5 wt% and SBS at least up to 10 wt% to the plain iPP, as well as SBS to 90/10 iPP/aPS blend. Small additions of aPS and SBS to plain PP and iPP/aPS blend enlarge spherulites contrary to the reported decrease of the spherulites size as a consequence of the nucleation effect (16, 24-26, 52).

* [d.sub.sph]/[d'.sub.p] [less than] 14

If the spherulites cannot grow fourteen-times bigger than the dispersed particles, they are not able to engulf that big particles. Consequently, a well-developed spherulitic structure is disturbed. Our results are in good agreement with the conclusion of Bartczak et al. [16] that bigger aPS inclusions disturb radial orientation of lamellae within the spherulites. It is supposed that the morphology of well-developed spherulites is transformed into morphology with cross-hatched bundles of sandwich lamellae above critical dispersed particle size as already reported for 70/30 PP/SEBS blend [51]. A cross-hatched lamellar structure transforms by the addition of SBS to 70/30 iPP/aPS blend to the structure with dominantly transcrystalline oriented blocks of sandwich lamellae (Fig. 14b).

* co-continuous morphology

In the binary 50/50 iPP/aPS blend the diameter of the dispersed aPS particles become large enough so that particles coalesce into a three-dimensionally co-continuous network. The addition of SBS to this blend causes narrowing of the channel-like domains of co-continuity. That leads to the completion of the co-continuous phase morphology (Fig. 14c). In the iPP co-continuous channel-like domains, only small spherulites of PP may arise.

CONCLUSIONS

From the results obtained by different experimental methods, the following conclusions referring to the phase structure of the binary iPP/aPS and iPP/SBS as well as ternary iPP/aPS/SBS compression molded blends can be summarized:

1. All studied blends show a phase-separated morphology due to the immiscibility of the blend components.

2. All samples studied exhibit a uniform morphological texture without a skin-core effect. Ternary iPP/aPS/SBS blends with iPP/aPS weight ratio of 50/50 exhibit co-continuous morphology.

3. SBS block copolymer acts as a compatibilizer in iPP/aPS immiscible blends and forms an interfacial layer between iPP matrix and dispersed aPS/SBS particles. It strongly interacts with iPP and aPS and significantly changes their phase morphologies.

4. Dispersed particles are complex aPS/SBS "honeycomb-like" aggregates consisting of small PS particles enveloped and joined together with SBS. Thin SBS layers envelop such aggregates of aPS/SBS particles and form interfacial layers between aPS/SBS particles and iPP matrix.

5. Even 2.5 wt% of SBS diminishes the coalescence of the dispersed aPS/SBS particles and reduces their average particle size.

6. SBS and aPS affect the crystallization process of iPP matrix and change the final supermolecular structure of iPP:

* aPS in lower amounts acts as a nucleating agent and decreases the crystallite size more than SBS.

* Higher amounts of aPS and SBS make a solidification effect and promote crystallite growth.

* Strong interactions of SBS interfacial layers with the iPP matrix cause the orientation growth of transcrystal sandwich lamellae.

7. The diameter of the dispersed aPS, SBS, and aPS/SBS particles are determined by aPS and SBS amounts and by the compatibilizing efficiency of SBS. They dominantly influence the formation of three different blend phase morphologies:

* [d.sub.sph]/[d'.sub.p] [greater than] 14

Blends exhibit well-developed spherulitic structure if the dispersed particles are fourteen-times smaller than the spherulites.

* [d.sub.sph]/[d'.sub.p] [less than] 14

Disturbed spherulitic structure or cross-hatched lamellar structure with dominantly transcrystalline oriented block of sandwich lamellae arise if the spherulites cannot grow fourteen-times bigger than the dispersed particles.

* co-continuous morphology

In the case of the binary 50/50 iPP/aPS blend the diameter of the dispersed aPS particles became large enough so that particles coalesce into a three-dimensionally co-continuous network. The addition of SBS ends the completion of co-continuous phase morphology.

ACKNOWLEDGMENTS

This work was supported by Ministry of Sciences and Technology of the Republic Slovenia and Ministry of Sciences and Technology of the Republic of Croatia. The authors also thank Dr. E. Ingolic from Technical University Graz, Austria, Dr. I. Anzel, Dr. V. Babic-Invancic, and Mr. J. Pohleven for their help in the experimental work.

(*.) Correspondence author: Ivan Smit, Ruder Boskovic Institute. Bijenicka 54, P.O. Box 1016, 10001 Zagreb, Croatia.

(1.) Ruder Boskovic Institute Bijenicka 54, P.O. Box 180 10002 Zagreb, Croatia

(2.) University of Maribor, EPF Maribor, Institute for Technology Razlagova 14, 2000 Maribor, Slovenia

REFERENCES

(1.) T. Kunori and P. H. Geil, J. Macromol. Sci. Phys., 18, 93 (1980).

(2.) B. D. Favis and J. P. Chalifoux, Polymer, 29, 1761 (1988).

(3.) B. D. Favis and J. P. Chalifoux, Polymer, 27, 1591 (1987).

(4.) I. Fortelny, D. Michalkova, and J. Mikesova, J. Appl. Polym. Sci., 59, 155 (1996).

(5.) E. Navratilova and I. Fortelny, Polym. Networks Blends, 6, 127 (1996).

(6.) S. Wu, Polym. Eng. Sci., 27, 335 (1987).

(7.) C. C. Chen and J. L. White, Polym. Eng. Sci., 33, 923 (1993).

(8.) C. J. Nelson, G. N. Avgeropoulos, F. C. Weissert, and G. G. Bohm, Angew. Makrornol. Chem., 60/61, 49 (1977).

(9.) C. E. Scott and C. W. Macosko, Int. Polym. Proc., 10, 36 (1995).

(10.) C. D. Han and T. C. Yu, J. Appl. Polym. Sci., 15, 1163 (1971).

(11.) C. D. Han and T. C. Yu, J. Appl. Polym. Sci., 19, 2831 (1975).

(12.) C. D. Han, C. A. Villamizar, and Y. W. Kim, J. Appl. Polym. Sci., 21, 353 (1977).

(13.) N. P. Krasnikova, E. V. Kotova, G. V. Vinogradov, and Z. Pelzbauer, J. Appl. Polym. Sci., 22, 2081 (1978).

(14.) R. Greco, G. Ragosta, E. Martuscelli, and C. Silvestre, in Polymer Blends: Processing, Morphology and Properties, p. 281, E. Martuscelli, R. Palumbo, and M. Kryszewski, eds., Plenum Press, New York (1980).

(15.) E. Martuscelli, C. Silvestre, R. Greco, and G. Ragosta, in Polymer Blends: Processing, Morphology and Properties, p. 295, E. Martuscelli, R. Palumbo, and M. Kryszewski, eds., Plenum Press, New York (1980).

(16.) Z. Bartczak, A. Galeski, and N. P. Krasnikova, Polymer, 28, 1627 (1987).

(17.) W. Wenig, H. W. Fiedel, and A. School, Colloid Polym. Sci., 268, 528 (1990).

(18.) M. Mucha, Colloid Polym. Sci., 264, 859 (1986).

(19.) M. Fujiyama, J. Appl. Polym. Sci., 63, 1015 (1997).

(20.) S. Rabiej, Eur. Polym. J., 29, 625 (1993).

(21.) G. Radonjic, PhD thesis, University of Maribor (June 1998).

(22.) J. Noolandi, Polym. Eng. Sci., 24, 70 (1984).

(23.) R. Fayt, R. Jerome, and Ph. Teyssie, Makromol. Chem., 187, 837 (1986).

(24.) A. Ghijsels, N. Groesbeek, and C. W. Yip, Polymer, 23, 1913 (1982).

(25.) J. Karger-Kocsis, A. Kallo, A. Szafner, G. Bodor, and Zs. Senyei, Polymer. 20, 37 (1979).

(26.) M. Saroop and G. N. Mathur, J. Appl. Polym. Sci., 71, 151 (1999).

(27.) Th. Laus, Angew. Makromol. Chem, 60/61, 87 (1977).

(28.) O. O. Santana and A. J. Muller, Polym. Bull, 32, 471 (1994).

(29.) D. Hlavata and Z. Horak, Eur. Polym. J., 30, 597 (1994).

(30.) R. Rosenberg, Z. Horak, and I. Fortelny, Polym. Networks Blends, 5, 185 (1995).

(31.) Z. Horak, V. Fort, D. Hlavata, F. Lednicky, and F. Vecerka, Polymer, 37, 65 (1996).

(32.) Z. Horak, J. Kolarik, M. Sipek, V. Hynek, and F. Vecerka, J. Appl. Polym. Sci., 69, 2615 (1998).

(33.) G. Obieglo and K. Romer, Kunststoffe, 83, 926 (1993).

(34.) I. Fortelny and D. Michalkova, Polym. Networks Blends, 7, 125 (1997).

(35.) L. D'Orazio, R. Guarino, C. Mancarella, E. Martuscelli, and G. Cecchin, J. Appl. Polym. Sci., 65, 1539 (1997).

(36.) G. Radonjic and V. Musil, Angew. Makromol. Chem., 251, 141 (1997).

(37.) G. Radonjic, V. Musil, and M. Makarovic, Acta Chim. Sloven., 22, 29 (1997).

(38.) G. Radonjic, J. Appl Polym. Set, 72, 291 (1999).

(39.) G. Radonjic and V. Musil, Kovine, zlitine, tehnologije, 30, 75 (1996).

(40.) G. Radonjic, V. Musil, and I. Smit, J. Appl Polym. Sci, 69, 2625 (1998).

(41.) G. Radonjic and V. Musil, Kovine, zlitine, tehnologije, 31, 97 (1997).

(42.) G. Radonjic, V. Musil, and I. Smit, Kovine, zlitine, tehnologije, 32, 81 (1998).

(43.) P. H. Hermans and S. A. Weidinger, Makromol Chem., 50, 98 (1961).

(44.) U. W. Gedde, Polymer Physics, Chapman and Hall, London (1995).

(45.) J. P. Trotignon, J. Verdu, J. Appl Polym. Sci, 34, 1(1987).

(46.) P. Zipper, A. Janosi, and E. Wrentschur, J. Physique W, Suppl J. Phys. I, 3, 33 (1993].

(47.) L. E. Alexander, X-Ray Diffraction Methods in Polymer Selenee, John Wiley, New York (1969).

(48.) B. Monasse and J. M. Haudin, in Polypropylene: Structure Blends and Composites, Vol. 1: Structure and Morphplogy (J. Karger-Kocsis, Ed.), Chapman and Hall, London (1995), Ch. 1, p. 3.

(49.) F. J. Padden and H. D. Keith, J. Appl. Phys., 30, 1479 (1959).

(50.) J. Varga. J. Mater. Sci, 27, 2557 (1992).

(51.) S. Setz, F. Snicker. J. Kessler, T. Duschek, and R. Millhaupt, J. Appl. Polym. Sci, 59, 1117(1999).

(52.) Z. Bartczak, E. Martuscelli, and A. Galeski, in Polypropylene: Structure Blends and Composites, Vol.2: Copolymers and Blends (J. Karger-Kocsis, Ed.), Chapman and Hall, London (1995), Ch. 2.

(53.) B. D. Favis, Polymer, 35, 1552 (1994).

(54.) T. Tang and B. Huang, Polymer, 35,281 (1994).

(55.) A. J. Lovinger, J. Polym. Sci, Polym. Phys. Ed, 21, 97 (1983).

(56.) M. Fujiyama, T. Wakino, and Y. Kawasaki, J. Appl. Polym. Sci, 35, 29 (1988).