Processing and Characterization of Microcellular Foamed High-Density Polyethylene/Isotactic Polypropylene Blends SAEED DOROUDIANI Department of Chemical Engineering and Applied Chemistry University of Toronto Toronto, Ontario, Canada M5S 3E5 CHUL B. PARK* Department of Mechanical and Industrial Engineering University of Toronto Toronto, Ontario, Canada M5S 3G8 and MARK T. KORTSCHOT Department of Chemical Engineering and Applied Chemistry University of Toronto Toronto, Ontario, Canada M5S 3E5 In this paper, a study on the batch processing and characterization of microcellular foamed high-density polyethylene/isotactic polypropylene (HDPE/iPP) blends is reported. A microcellular plastic is a foamed polymer with a cell density greater than 109 cells/cm3 and fully grown cells smaller than 10 mm. Recent studies have shown that the morphology and crystallinity of semicrystalline polymers have a great influence on the solubility and diffusivity of the blowing agent and on the cellular structure of the resulting foam in microcellular batch processing. In this research, blends of HDPE and iPP were used to produce materials with variety of crystalline and phase morphologies to enhance the subsequent microcellular foaming. It was possible to produce much finer and more uniform foams with the blends than with neat HDPE and iPP. Moreover, the mechanical properties and in particular the impact strength of the blends were significantly improved by foaming. INT RODUCT ION M icrocellular plastics are foamed polymers with a cell-population density greater than 109 cells/cm3 and fully grown cells smaller than 10 mm. A microcellular foamed structure can be developed by first saturating a polymer sample with a volatile blowing agent followed by rapidly decreasing the solubility of the blowing agent in the polymer (1–8). Our recent studies have shown that the morphology and crystallinity of semicrystalline polymers have a great influence on the *To whom correspondence should be addressed. POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 solubility and diffusivity of the blowing agent in the polymer and on the cellular structure of the resulting foam in microcellular batch processing (9). In this work, the microcellular batch processing and properties of HDPE and iPP blends were investigated. First, standard samples were produced from HDPE and iPP and their blends with compositions of 90/10, 50/50, and 10/90 (wt/wt). Then, the solubility and diffusivity of CO2 in these samples were investigated along with the morphologies and the crystallinities of the blends. The samples were foamed in a microcellular process, and the structures and properties of the resulting foams were compared. The results of this 1205 Saeed Doroudiani, Chul B. Park, and Mark T. Kortschot work show that the presence of another phase in polyolefins has a great influence on the foaming of HDPE and iPP in microcellular batch processing. Although a foam structure was difficult to achieve in neat HDPE and iPP, a microcellular foam structure was easily obtained with the blends. Also, the mechanical properties that were degraded by blending were improved significantly by foaming. The experimental results reveal that in some cases the addition of a low percentage of another polymer greatly affects the morphologies, the mechanical properties (10), the surface properties (11), and the foamability. Polyolefins comprise a large group of polymers with various macromolecular structures and different degrees of branching. The density of polyolefins is the lowest among the polymers, and their properties are strongly affected by the density. Polyolefins, and in particular PE and PP, are major components of postconsumer plastics waste. Separation of plastics waste into individual parts and sorting is costly, and usually technically difficult. Therefore, developing processes for consumption of plastics wastes without sorting would be valuable. Polymers, depending on their structures, may form crystalline or amorphous morphologies during cooling from the melt. The crystalline structure of the polymer strongly depends on the crystallization conditions. HDPE and iPP are semi-crystalline polymers and form lamellar structures and spherulites when cooled from the melt. Blending is commonly used to enhance certain properties of the polymers. Two polymers may be miscible and/or compatible depending on their thermodynamic properties. A miscible polymer blend is a homogeneous mixture of two polymers at the molecular level with a negative value of the free energy of mixing (DGM > DHM # 0) (12, 13). In an immiscible polymer blend, the free energy of mixing is positive. Miscible blends are characterized by a single glass transition temperature and a single homogeneous phase, while immiscible blends display two glass transition temperatures and have two or more phases. It is known that HDPE and iPP are immiscible and incompatible, in spite of the similarity of their chemical structures (13, 14). Blending affects the crystalline structures of both polymers. In the case of HDPE/iPP blends, solidification from the melt usually involves the crystallization of the iPP phase followed by that of the HDPE phase if the rate of cooling is not too fast (15, 16). When HDPE/iPP blends are cooled rapidly from the melt, simultaneous crystallization of the HDPE and the iPP will take place, and the phase boundary as well as the phase morphology will be different from those obtained through slow cooling. The effect of the second polymeric phase on the morphology of the first phase must be similar to the known effect of low molecular weight additives on the crystalline structure (13). Hence, the addition of iPP to HDPE might cause a decrease in the size of the spherulites. 1206 Blending also affects the surface (or interface) properties. The surface phenomena in polymer blends, even in simple systems, are very complex, and not well understood. In general, blends of both compatible and incompatible polymers show pronounced surface activity, which is more pronounced in incompatible blends than in compatible blends (11). The surface properties of the polymers have a great effect on the mechanism and kinetics of microcellular foaming (17). The activation energy for bubble nucleation is much lower for the heterogeneous interfaces developed in the blends. Therefore, it is expected that the foamability will be greatly enhanced by blending. It is the purpose of this study to make use of blending to enhance the microcellular foamability of polyolefin materials. PROBL EM ST AT EMENT In previous work (9), it was shown that the batch foam processing HDPE and iPP was strongly dependent on their crystal morphology and the degree of crystallinity. In the batch foam processing of HDPE and iPP, microcellular structures were difficult to achieve, unless the polyolefin materials were quenched during cooling from the melt so that they had a relatively low crystallinity. It was believed that the foamability of polyolefin materials was enhanced by increasing the solubility of gas in the polymer and by decreasing the stiffness of the polymer matrix when the crystallinity was lowered. In this study, the effects of the blending of HDPE and iPP on the foamability and the properties of the resulting foams were investigated. Blends of HDPE/iPP were prepared by injection molding, and the samples were characterized in terms of the crystalline parameters, the sorption parameters, and the mechanical properties. Then the microcellular foamability of these blends was studied as the composition was varied. Finally, the mechanical properties of the resulting foams were examined. Particular emphasis was given to the impact strength of the foamed blends. EXPERIMENT AL Materials Injection-molding-grade HDPE from DuPont (SCLAIR 2909, melt index 13.5 dg/min at 190°C, density 960 kg/m3) and isotactic-PP homopolymer from Himont (Profax 6331, melt index 10 dg/min at 230°C, density 902 kg/m3) were used as received. Commercial grade carbon dioxide (Matheson Gas Products) was used as a blowing agent without any further purification. Standard samples of HDPE, iPP and their blends with compositions of 90/10, 50/50, and 10/90 (denoted as PP10, PP50, and PP90, respectively) were produced by an injection molding machine (Engel ES28) directly from neat and manually-mixed pellets, using a standard test specimen mold (the dimensions conformed to type I specimens of ASTM Method D638 for tensile and ASTM Method D256 for impact tests). POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 Processing and Characterization Fig. 1. Normalized sorption curves of HDPE, iPP and their blends. The barrel temperatures of the injection molding machine were 163°C–193°C–193°C (325°F–380°F–380°F) in the hopper-to-nozzle direction for HDPE and PP10 materials, and 177°C–218°C–218°C (350°F–425°F –425°F) for PP50, PP90 and iPP materials. Sorption Experiments The solubility and diffusivity of the blowing agent in the specimens were measured in sorption experiments. For the purposes of this test, the specimens were pressurized in a pressure vessel with carbon dioxide. At various times, the mass uptake was recorded. From the uptake sorption curve, the diffusivity was derived (18) as follows: Mt yM` 5 4 (Dyp)0.5 (t 0.5yL) (1) t 0.5/L A plot of Mt/M` as a function of (Fig. 1) yields essentially a straight line (for the initial time period of the test) with a slope of 4(D/p)0.5, which is readily solved for D. Here Mt is the total amount of gas that has diffused into the polymer strip at time t, and M` is the total amount of gas that has diffused at infinite time. The solubility is calculated to be M` divided by the weight of the sample. Microcellular Foaming Figure 2 shows an experimental setup for microcellular foam processing (the drawing is not scaled). The pressure vessel design is detailed in reference (19). First, specimens with a thickness of 3.2 mm (0.1250) POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 Fig. 2. Experimental setup for microcellular batch foam processing. were saturated in a pressure vessel with carbon dioxide at room temperature (23–25°C) and 5.51 MPa (800 psi) for 110 hours. The required saturation time was determined from the sorption curves. Then, the pressure was released and the samples were immersed in a glycerin bath for 20 seconds at 133°C (for HDPE and PP10 specimens), for 20 seconds at 140°C (for PP50 specimen), and for 10 seconds at 160°C (for PP90 and iPP specimens). It was not possible to foam all the blends at a single temperature: when the temperature was too high, the foam formed but then collapsed, while a low temperature resulted in little expansion because of the high viscosity of the polymers. Hence, the foaming temperature for each blend was determined from the foaming behaviors at various temperatures. 1207 Saeed Doroudiani, Chul B. Park, and Mark T. Kortschot Morphology and Structure Studies Samples were prepared for morphology studies by microtoming the injection-molded specimens at room temperature with a Sorvall MT-6000 equipped with a glass knife, and their morphologies were studied using an Olympus BH-2 polarized optical microscope equipped with a 35 mm camera. The foamed samples were characterized using a Hitachi S-520 scanning electron microscope (SEM) at an acceleration voltage of 20 KV. The samples were immersed in liquid nitrogen for 30 minutes, then fractured and mounted on stubs. The fracture surfaces were sputter coated with gold prior to the microscopy work. The fracture surfaces resulting from impact testing were also characterized directly using the SEM. Differential Scanning Calorimetry The crystallinites of HDPE, iPP and their blends were investigated using a differential scanning calorimeter (DSC, Model DuPont 2000) at a scanning rate of 10 K/min under a nitrogen environment. The crystallinity (x) of a single-phased material (i.e., HDPE and iPP) was calculated by measuring the specific heat required for melting (DHm) through integration of the appropriate peak and dividing this value by the heat of fusion for the pure crystalline phase (DH 0m) (20–22). On the other hand, the calculation of crystallinity of a blend was complicated. The crystallinities of HDPE (xHDPE ) and iPP (xPP) in the blends were separately calculated by dividing the specific enthalpies required for melting of each phase (i.e., DHm,HDPE and DHm,PP ) by the heats of fusion for each crystalline 0 0 phase (i.e., DH m, HDPE and DH m,PP ). 0 ) 3 100% xHDPE 5 (DHm,HDPEyDH m,HDPE (2) 0 ) 3 100% xPP 5 (DHm,PPyDH m,PP (3) DHm,HDPE and DHm,PP in Eqs 2 and 3 were determined from the DSC thermograms of the blends using Eqs 4 and 5: DHm,HDPE 5 Mechanical Properties T esting Tensile tests were performed on a Sintech Model 20 testing machine equipped with a computer, according to the ASTM D-638 standard method. Tensile properties and statistical data were calculated using the Testwork program (Version 2.10, Sintech Inc.). For each sample, five (for foamed samples) or ten (for unfoamed samples) specimens were tested. The tensile properties of the foamed samples were calculated based on the expanded cross-sectional area (denoted as “Foamed”) and based on the cross-sectional area of the unfoamed material (denoted as “Foamed*”). The impact strength of the samples was determined using a Tinius Olsen 92T impact tester. The tests were run at room temperature according to ASTM D-256 (Izod) with notched samples. All the samples were allowed to desorb the gas for at least two weeks before the property testing to remove the effect of the residual gas (23, 24). RESUL T S AND DISCUSSION In the microcellular processing of semicrystalline polymers, the crystalline structure and the stiffness of the polymer matrix play a critical role (4, 5, 9). Therefore, in order to understand the effect of blending on the foam processing of HDPE/iPP blends, the crystalline morphology, crystallinity and the mechanical behavior of the unfoamed blends were studied first. Since foaming affects the physical and mechanical properties of the materials, the results of tensile and impact testing of the foamed samples are also discussed, and the mechanical properties are compared with those of unfoamed specimens. Morphology Studies Dhm,HDPE (1 2 x) ? w DHm,PP 5 the integrated areas of each peak in the thermogram). x and w are the weight fraction of iPP in the blend and the weight of blend sample used in DSC, respectively. The weight of the blend samples ranged from 3.5 to 5.0 mg in this study. (4) Dhm,PP x?w (5) where DHm,HDPE and DHm,PP are, respectively, the heats required for melting the HDPE and iPP phases, (i.e., The crystalline morphologies of the microtomed sections of the injection-molded HDPE, iPP, and their blend samples were investigated using an optical microscope with polarized light. The photomicrographs in Fig. 3 show that the presence of materials in one phase had a definite and pronounced effect on the crystalline structure of material in the other phase. In Table 1. Melting and Crystallization Parameters of Blend Samples Calculated from DSC Data. Sample Tm,HDPE (°C) DHm,HDPE (J/g) xHDPE (%) Tm,PP (°C) DHm,PP (J/g) xPP (%) HDPE PP10 PP50 PP90 iPP 134.8 134.4 131.5 130.2 — 175.7 176.9 186.6 162.9 — 60.0 60.4 63.7 55.6 — — 164.2 164.1 166.1 165.7 — 45.2 64.9 77.6 90.8 — 21.6 31.0 37.1 43.4 xHDPE and xPP are the crystalline fractions of HDPE and iPP, respectively, calculated from the measured enthalpies of fusion for HDPE (DHm,HDPE) and ipp (DHm,PP). Tm,HDPE and Tm,PP are the temperatures at melting peaks of HDPE and iPP, respectively. 1208 POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 Processing and Characterization (a) (b) (c) (d) (e) Fig. 3. Morphology of (a) HDPE, (b) PP10, (c) PP50, (d) PP90, and (e) iPP. Scale bars 20 mm. POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 1209 Saeed Doroudiani, Chul B. Park, and Mark T. Kortschot k 5 [kam,HDPE(1 2 xHDPE) 1 kcr,HDPE xHDPE](1 2 x) 1 [kam,PP (1 2 xPP) 1 kcr,PP xPP]x (7) where kam,HDPE, kcr,HDPE, kam,PP, and kcr,PP are the solubilities of CO2 in the amorphous and crystalline regions of HDPE and iPP, respectively; and xHDPE and xPP are the crystalline fractions of HDPE and PP, respectively. Since CO2 dissolves only in the amorphous phase (4, 5, 9), the second term in each square bracket can be ignored. Therefore, the solubility of CO2 in the blend can be written as a function of the blend composition and the crystalline fraction of each component: Fig. 4. Solubility of CO2 in the HDPE/iPP blends as a function of blend composition. this work, the size of the spherulites for HDPE and iPP samples were 30–40 and 50–60 mm, respectively. It was observed that the regular spherulite patterns of HDPE and iPP (Figs. 3a and e) tended to become irregular with blending (Figs. 3b, c, and d). These observations are consistent with the results of Lovinger and Williams (22), and Teh (25). The crystalline fraction of each component in the blends was measured in the DSC experiments, and the results are summarized in Table 1. Various values have been reported for the heats of fusion of HDPE and iPP depending on the mode of crystallization (26–33). Since it was not clear which crystallization mode was dominant for the HDPE and iPP crystallites in our blend samples, the most representative values of 293 J/g (26–31) and 209 J/g (28–33) were chosen as the heat of fusion for HDPE and iPP, respectively, in calculating the crystalline fractions. The crystalline fraction of iPP decreased as the HDPE component increased, whereas the crystalline fraction of HDPE did not change much with the weight fraction of iPP in the blends. It seemed that blending tended to reduce the crystallinity of the iPP. These results are in agreement with the observations made by Martuscelli with iPP/LDPE blends (34), and Bartczak et al. with HDPE/iPP blends (35). The presence of PE melt was found to retard the crystallization of iPP. k 5 kam,HDPE(1 2 xHDPE)(1 2 x) 1 kam,PP(1 2 xPP)x (8) The solubilities of CO2 in the amorphous phases of HDPE and iPP (i.e., kam,HDPE and kam,PP) were calculated from the measured solubilities of CO2 in the HDPE and iPP and the corresponding crystallinities of the pure HDPE and iPP samples (measured from the DSC experiments): k 5 44.0 (1 2 xHDPE) (1 2 x) 1 79.5 (1 2 xPP)x (9) Based on the measured crystalline fraction in each component of the blends, the expected solubilities of CO2 were derived using Eq 9. The solubilities calculated using Eq 9 (based on the DSC results, see Table 1) and the measured solubilities of CO2 in the blends are compared in Fig. 4. Note that the crystallinity of each phase in the blend was less than that found in the pure polymers (Table 1). As a result, the total fractions of amorphous region for the blends were greater than the values predicted by the rule of mixtures, and therefore, the solubilities predicted by Eq 9 were greater than those calculated by the rule of mixtures (Fig. 4). The measured solubilities were slightly lower than the predicted ones. Few models have been proposed to describe the diffusion of small molecules in polymer blends. In the simplest model, the polymer is assumed to consist of Solubility and Diffusivity The effects of blending on the solubility and diffusivity of the blowing agent in and through the samples were investigated using a sorption test (Figs. 4 and 5). At equilibrium, the rule of mixtures should describe the solubility of CO2 in the blends. The solubility of CO2 in the blend is k 5 kHDPE(1 2 x) 1 kppx (6) where kHDPE and kPP are the solubilities of CO2 in the HDPE and iPP phases, respectively. Since each component actually consists of both amorphous and crystalline regions, the solubility may be rewritten as 1210 Fig. 5. Diffusivity of CO2 through the HDPE/iPP blends as a function of blend composition. POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 Processing and Characterization (a) (b) (c) (d) (e) Fig. 6. Scanning electron micrographs of foamed (a) HDPE, (b) PP10, (c) PP50, (d) PP90, and (e) iPP. POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 1211 Saeed Doroudiani, Chul B. Park, and Mark T. Kortschot laminae of each component arranged in either series or parallel configurations (36). In our case, the diffusion of CO2 in the HDPE/iPP blend is more complicated because the matrix has four different phases: the amorphous and crystalline HDPE phases, and the amorphous and crystalline iPP phases. Despite the complex phases in the blend matrix, the measured diffusivity of CO2 in the blend almost followed the rule of mixtures due to the minor change in the crystalline fraction in each component of the blends (Fig. 5). Microcellular Foaming In microcellular foam processing, after a saturated solution of the polymer/blowing agent is prepared, a large number of cells should be nucleated. The resulting microcellular foam structure is determined mainly by the number of nuclei. The nucleation of microcells in semicrystalline polymers strongly depends on the crystallinity and morphology (9). This section describes the effects of blending on the microcellular foamability of semicrystalline polymers. The microcellular foamability of the materials was greatly enhanced by blending. Since both the HDPE and iPP were injection molding grades with relatively low molecular weights, the crystallinities were high. Because of the resulting increase in the stiffness and decrease in the gas solubility (9), a microcellular foamed structure was difficult to achieve with these pure materials. Cellular structures were developed only locally near the edges of the samples. However, when these materials were blended, the foamability was significantly improved. A uniformly distributed microcellular foamed structure was obtained with the blend samples of PP10 (Fig. 6b). A microcellular structure was also developed in PP90. However, the foam structure was not very uniform (Fig. 6c). The volume expansion of foamed PP50 was much larger than those of foamed PP10 and PP90, and a large degree of non-uniformity was observed in the resulting foam samples of PP50. Because this sample was foamed at 140°C, it is probable that the bulk of expansion was confined to the HDPE phase. The volume expansion ratios of foamed PP10, PP50, and PP90 (denoted as FPP10, FPP50, and FPP90, respectively) were 11%, 71%, and 14% respectively. The SEM pictures of microcellular foamed HDPE, PP10, PP50, PP90, and iPP are shown in Fig. 6. The particulate pattern on the fracture surface of foamed HDPE (Fig. 6a) seems to be due to the ductile nature of the HDPE materials (37). The improved foamability of blends can be attributed to many factors such as the interface developed between the two immiscible materials, the crystalline morphology of blends, the reduced crystallinity, etc. First of all, the foamability of HDPE/iPP blend samples must have been strongly affected by the interface in the blends. HDPE and iPP are immiscible and form stable interfaces. Because it is well known that the poorly bonded interfacial regions have a lower activation energy for bubble nucleation (17), it is believed that the interface developed in immiscible blends of 1212 HDPE and iPP must have provided favorable heterogeneous nucleating sites for bubble formation. Therefore from the viewpoint of the surface characteristics, blending should be advantageous to microcellular foaming. The foam structure of HDPE/iPP blends must also have been affected by the crystalline morphology of blends. Our previous study demonstrated that the morphology of a semicrystalline polymer affects the foam structure (9). The study of the blend samples showed that by blending, an irregular texture of crystallites was obtained. Since the blending of two semicrystalline materials influences the morphology in each phase of the blend, the resulting foam structure must be affected by the changed morphology of HDPE and iPP in the blends. The degree of crystallinity could also have affected the foamability of the HDPE/iPP blends. Since the crystalline fraction of iPP in the blends decreased as the weight fraction HDPE increased, the solubility of CO 2 in the blend increased, as discussed above. Although there was only a slight increase in the solubility by blending, its effect may not be negligible because the amount of thermodynamic instability produced by the rapid pressure decrease and temperature rise during microcellular foaming is very sensitive to the concentration of gas dissolved in the polymer (6). Furthermore, the reduced crystallinity must have reduced the stiffness of the blend matrix. Since the stiffness of the polymer matrix affects the foamed structure significantly during microcellular foaming (9, 38), the reduced crystallinity must have been favorable to the foaming. The unique foaming behaviors of the HDPE/iPP blends must also have been affected by the gas loss during foaming. Matuana et al. (8) and Behravesh et al. (39) showed that the void fraction of a microcellular foamed sample is strongly affected by the amount of gas lost through the foam skin at various foaming temperatures. Behravesh et al. (39) showed that for a 5 wt% concentration of CO2, the maximum achievable volume expansion ratio would be 29.5 times if there is no gas loss through the foam skin. For the typical solubilities of 2.1, 3.3, and 4.4 wt% in the HDPE/iPP blends, the maximum achievable volume expansion ratios would be 12.4, 19.5, and 26.0 times. Since the obtained volume expansion ratios of the HDPE/iPP blend foams were much lower than the maximum achievable ones, it is speculated that a significant amount of gas was lost during the foaming in the batch microcellular process. Because of the difficulty involved in measuring the amount of gas lost in the microcellular processing, it was not easy to identify the effect of the desorption on the resultant foam morphology for each blend. A further study will be carried out to quantify this effect. Mechanical Properties One of the interesting features of HDPE/iPP blends is their tensile properties. The tensile properties of the POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 Processing and Characterization (a) (b) (c) (d) Fig. 7. Tensile properties of unfoamed and foamed samples as a function of composition: (a) stress at yield, (b) ultimate tensile strength, (c) elastic modulus, and (d) elongation at peak. The properties of foamed samples are presented based on the expanded cross-sectional area (Foamed) and the cross-sectional area of the unfoamed material (Foamed*). blend samples are shown as a function of composition in Fig. 7. Each data point represents the mean of at least five measurements, with related standard deviations shown as error bars. The changes of the stress at yield for the unfoamed blend samples were close to what is predicted by the rule of mixtures (Fig. 7a); the yield stress increased monotonically as the percentage of iPP increased from 0 to100%. But the nominal tensile strength (Fig. 7b) and the elastic modulus (Fig. 7c) of blends showed a positive synergism, whereas the elongation at break was deteriorated slightly by blending (Fig. 7d). These results support the observations made by Noel and Carley (40) and by Teh et al. (10). Deanin and Sansone (41) and Lovinger and Williams (22) also found similar results with 25/75 composition. From their detailed studies of the tensile properties and the morphology of HDPE/iPP blends, Lovinger and Williams concluded that morphological effects may be the main reasons for the tensile behavior of HDPE/iPP blends (22); these include spherulite sizes, intercrystalline links between lamellae, and some interactions between the two incompatible phases and their mutual boundaries. The tensile properties of the foamed samples (based POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 on the total foamed cross-section) were in general lower than those of the unfoamed samples. In particular, the tensile properties of the foamed samples at a 50/50 composition deteriorated significantly because of the large volume expansion ratio obtained from the foamed samples. The normalized yield stress and elastic modulus of the foamed samples (denoted as Foamed*) at all compositions were almost the same as those of the unfoamed samples; the one exception was the ultimate tensile strength. Although the introduction of a microcellular foam structure into these blends lowered the stress at yield and the elastic modulus (based on the same volume), the ultimate tensile strength seemed to be improved by foaming. For example, Fig. 7b shows that the ultimate tensile strength of the foamed blend at a 10/90 composition was twice as high as that of the unfoamed blend. This reinforcing effect caused by the microcellular structure may be due to the orientation of macromolecular chains in the cell walls during foaming. The elongation at peak of the HDPE/iPP blends did not seem to be much affected by microcellular foaming. In contrast to what was observed for the tensile properties, the effect of blending and microcellular 1213 Saeed Doroudiani, Chul B. Park, and Mark T. Kortschot Fig. 8. Notched Izod impact strength of HDPE, iPP, foamed and unfoamed samples of blends. foaming on the impact strength of blends was quite pronounced. First, it should be emphasized that blending produced a significant loss in the impact strength. For example, the impact strength of the PP10 sample (containing 10% iPP) was less than the half of that of the neat HDPE (Fig. 8). The lowered impact strength seemed to be due to the weak interface between the neat HDPE and iPP phases in the blends. It is interesting to note that the trend of impact strength as a function of composition was similar to that of the elongation at peak (Fig. 7d) (42). When a microcellular foam structure was developed in the blends, the impact strength was improved significantly. The notched Izod impact strengths of foamed PP10, PP50 and PP90 were three, four and two times as high as those of the corresponding unfoamed ones, respectively (Fig. 8). One possible explanation is that the microcells blunt the notch and retard its propagation. A mechanism has been suggested for describing this behavior of foamed samples in analogy with the toughening of brittle plastics by incorporating rubber particles (1, 43). The toughening mechanism depends on the failure of the matrix under impact. When the craze initiation stress of the matrix is lower than the yield stress, the failure mechanism is due to crazing, the dispersed rubber particles act as craze initiators, and toughening is achieved. Another toughening mechanism is yielding, in which the craze initiation stress is higher than the yield stress. Sometimes crazing and yielding may occur at the same time (44, 45). CONCL USIONS The microcellular processing and the mechanical properties of HDPE, iPP, and their blends have been studied. It was observed that blending decreased the crystallinity of iPP in the blends. The presence of another phase also had a great influence on the crys1214 talline morphology in all blends. The results of this work also show that the presence of another phase influences the foaming of HDPE and iPP in microcellular batch processing. All blends were foamed with a fine cellular structure, whereas little or no foaming took place in the neat polymers. The blend with a 50/50 composition was foamed with a very high volume expansion ratio and had a nonuniform structure, which was reflected in the large standard deviations in the test results. Based on the experimental results, it is suggested that a minor phase of another polymer should be added to the major phase materials to facilitate the microcellular foam processing of polyolefin materials. The tensile properties and impact strength of blends and their foams were investigated. Overall, the tensile properties of blends nearly followed the rule of mixtures. Some tensile properties, such as the ultimate tensile strength and the elastic modulus, showed a synergistic effect with blending while the elongation at the peak was deteriorated by blending. When a microcellular foam structure was developed in the blends, the specific tensile properties (based on the same weight) were almost the same as those of the unfoamed blends except for the ultimate tensile strength. The ultimate tensile strength of PP10 was improved significantly by microcellular foaming. However, the tensile properties based on the same volume were lower than those of the unfoamed blends. On the other hand, the impact strength was deteriorated dramatically by blending. Although the impact strength was deteriorated by blending, it was improved significantly by introducing a microcellular foam structure into the materials. NOMENCL AT URE D 5 Diffusivity, cm2/s. FPP10 5 Foamed HDPE/iPP blend with a composition of 90/10. FPP50 5 Foamed HDPE/iPP blend with a composition of 50/50. FPP90 5 Foamed HDPE/iPP blend with a composition of 10/90. k 5 Solubility of CO2 in the blend, g/g. kam,HDPE 5 Solubility of CO 2 in the amorphous phase of HDPE, g/g. kam,PP 5 Solubility of CO 2 in the amorphous phase of iPP, g/g. kcr,HDPE 5 Solubility of CO2 in the crystalline phase of HDPE, g/g. kcr,PP 5 Solubility of CO2 in the crystalline phase of iPP, g/g. kHDPE 5 Solubility of CO2 in HDPE, g/g. kPP 5 Solubility of CO2 in iPP, g/g. L 5 Thickness, cm. Mt 5 Amount of gas in the polymer at time t, g. M` 5 Amount of gas in the polymer at equilibrium saturation, g. PP10 5 HDPE/iPP blend with a composition of 90/10. POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 Processing and Characterization PP50 5 HDPE/iPP blend with a composition of 50/50. PP90 5 HDPE/iPP blend with a composition of 10/90. t 5 Time, s. Tm,HDPE 5 Temperature at melting peak of HDPE, °C. Tm,PP 5 Temperature at melting peak of iPP, °C. x 5 Weight fraction of iPP. w 5 Weight of sample in DSC, g. DGM 5 Free energy of mixing, J. DHM 5 Enthalpy of mixing, J. DHm 5 Specific enthalpy required for melting a polymer, J/g. 0 5 Specific enthalpy of fusion of a pure DH m, crystalline phase (or heat of fusion), J/g. DHm,HDPE 5 Specific enthalpy required for melting HDPE, J/g. DHm,PP 5 Specific enthalpy required for melting iPP, J/g. 0 DH m,HDPE 5 Heat of fusion of HDPE, J/g. 0 DH m,PP 5 Heat of fusion of iPP, J/g. Dhm,HDPE 5 Specific heat required for melting the HDPE phase in a blend, J/g. Dhm,PP 5 Specific heat required for melting the iPP phase in a blend, J/g. x 5 Crystallinity. xHDPE 5 Crystalline fraction of HDPE. xPP 5 Crystalline fraction of iPP. REFERENCES 1. J. E. Martini, F. A. Waldman, and N. P. Suh, SPE ANTEC Tech. Papers, 28, 674 (1982). 2. C. B. Park, D. F. Baldwin, and N. P. Suh, Polym. Eng. Sci., 3 5 , 432 (1995). 3. C. B. Park and N. P. Suh, Polym. Eng. Sci., 3 6 , 34 (1996). 4. D. B. Baldwin, C. B. Park, and N. P. Suh, Polym. Eng. Sci., 36, 1437 (1996). 5. D. B. Baldwin, C. B. Park, and N. P. Suh, Polym. Eng. Sci., 36, 1446 (1996). 6. S. Cha, PhD thesis, Massachusetts Institute of Technology (1994). 7. L. Matuana-Malanda, C. B. Park, and J. J. Balatinecz, Cellular Plast., 32, 449 (1996). 8. L. M. Matuana, C. B. Park, and J. J. Balatinecz, Polym. Eng. Sci., 37, 1137 (1997). 9. S. Doroudiani, C. B. Park, and M. T. Kortschot, Polym. Eng. Sci., 36, 2645 (1996). 10. J. W. Teh, H. P. Blom, and A. Rudin, Polymer, 35, 1680 (1994). 11. S. Wu, Polymer Interface and Adhesion, Ch. 3, Marcel Dekker, New York (1982). 12. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, New York (1953). 13. L. A. Utracki, Polymer Alloys and Blends-Thermodynamics and Rheology, Hanser, New York (1989). 14. K. Solk (Ed.), Polymer Compatibility and Incompatibility, Principles and Practices, Harwood, New York (1982). 15. J. W. Teh, A. Rudin, and J. C. Keung, Adv. Polym. Tech., 13, 1 (1994). 16. B. Ke, J. Polym. Sci., 61, 47 (1962). POLYMER ENGINEERING AND SCIENCE, JULY 1998, Vol. 38, No. 7 17. J. S. Colton and N. P. Suh, Polym. Eng. Sci., 27, 493 (1987). 18. R. W. Vieth, Diffusion In and Through Polymers: Principles and Applications, Hanser Publishers, Munich (1990). 19. S. Doroudiani, Project Report, Department of Chemical Engineering and Applied Chemistry, University of Toronto (1995). 20. V. A. Bershtein and V. M. Egorov, Differential Scanning Calorimetry of Polymers: Physics, Chemistry, Analysis, Technology, Ellis Horwood Ltd., Chichester, England (1994). 21. J. F. Rabek, Experimental Methods in Polymer Chemistry, Wiley Interscience Publication, New York (1980). 22. A. J. Lovinger and M. L. Williams, J. Appl. Polym. Sci., 25, 1703 (1980). 23. D. I. Collias and D. G. Baird, Polym. Eng. Sci., 35, 1167 (1995). 24. D. I. Collias and D. G. Baird, Polym. Eng. Sci., 35, 1178 (1995). 25. J. W. Teh, J. Appl. Polym. Sci., 28, 605 (1983). 26. Ser van der Ven, Polypropylene and Other Polyolefins: Polymerization and Characterization, Elsevier Science Publishers, The Netherlands (1990). 27. B. Wunderlich and G. Czornyj, Macromolecules, 10, 906 (1977). 28. B. Wunderlich, Macromolecular Physics, Vol. 1, Crystal Structure, Morphology, Defects, Academic Press, New York (1973). 29. B. Wunderlich, Macromolecular Physics, Vol. 3, Crystal Melting, Academic Press, New York (1980). 30. Advanced Thermal Analysis Laboratory Databank, University of Tennessee, Department of Chemistry, URL: http://funnelweb.utcc.utk.edu/;athas/databank/alken.html (1996). 31. P. Martuscelli, C. Silvestre, and G. Abate, Polymer, 23, 299 (1982). 32. E. J. Clark and J. D. Hoffmann, Macromolecules, 1 7 , 878 (1984). 33. H. S. Bu, Z. D. Cheng, and B. Wunderlich, Makromol. Chem., Rapid Commun., 9, 75 (1988). 34. E. Martuscelli, Polym. Eng. Sci., 24, 563 (1984). 35. Z. Bartczak, A. Galeski, and M. Parcella, Polymer, 27, 537 (1986). 36. J. Sax and J. M. Ottino, Polym. Eng. Sci., 2 3 , 165 (1983). 37. D. Vesely, Polym. Eng. Sci., 36, 1586 (1996). 38. D. B. Baldwin, N. P. Suh, and M. Shimbo, in Cellular Polymers, V. Kumar and S. G. Advani, eds., MD-Vol. 38, p. 109, ASME, New York (1992). 39. A. H. Behravesh, C. B. Park, and R. D. Venter, in Cellular and Microcellular Materials, V. Kumar and K. A. Sealer, eds., MD-Vol. 76, p. 47, ASME, New York (1996). 40. O. F. Noel and J. F. Carley, Polym. Eng. Sci., 15, 117 (1975). 41. R. D. Deanin and M. F. Sanson, Polym. Prepr., 19, 211 (1978). 42. L. E. Nielson and R. F. Landel, Mechanical Properties of Polymers and Composites, Marcel Dekker, New York (1994). 43. D. I. Collias, D. G. Baird, and R. J. M. Borggreve, Polymer, 35, 3978 (1994). 44. I. Walker and A. A. Collyer, in Rubber Toughened Engineering Plastics, A. A. Collyer, ed., Chapman & Hall, London (1994). 45. S. Wu, Polym. Eng. Sci., 30, 753 (1990). Received August 1996 Revised July 1997 1215