Properties of blends of linear and branched polypropylenes in film blowing.

Fang, Yunli ; Sadeghi, Farhad ; Fleuret, Guillaume 等

INTRODUCTION

As one of the most important resins, polypropylene (PP) has many advantageous properties, such as high melting point, high tensile strength, stiffness, low density, and excellent chemical resistance, compared to polyethylene (PE). These advantages make PP widely used in thermoforming, foaming, blow moulding, film sheeting, and blown film industries. However, the low melt strength, which comes from its highly linear chains and relatively narrower molecular weight distribution (MWD), limits its operating window in film blowing that is dominated by elongational flow. The major issue for PP in blown film is bubble instability. The analysis of bubble stability in film blowing has been extensively performed for PE, mainly because of its superior melt strength, as previously shown (Kanai and White, 1984; Han and Park, 1975). Some experiments on the bubble stability of PP have been carried out, as shown by Ghaneh-Fard et al. (1996). One question rises if there is a way to promote bubble stability for PP by adding a non-linear PP.

Adding a low-density polyethylene (LDPE) to a linear low-density polyethylene (LLDPE) is widely applied in the blown film industry to improve the processability (bubble stability). LDPE has good processability due to the presence of long-chain branches (LCBs), but relatively poor mechanical properties compared to a linear PE. On the other hand, LLDPE is popular for its better mechanical properties while showing poor processability due to its linear structure and relatively narrower M WD. Blending of the two resins is a compromise between these two issues, processability and mechanical property enhancement.

PP resins with LCBs are in market mostly for foam applications and compared to L-PP show a significant strain-hardening behaviour and higher melt strength, as indicated by Lagendijk et al. (2001). A stronger network formed as a result of larger entanglement density in the melt state for LCB-PP that reduces the probability of rupture and coalescence of formed bubbles during bubble growth, as shown previously (Gotsis and Zeevenhoven, 2004). In the light of PE applications (blending of LDPE with LLDPE), the main objective of this work was to investigate the improvement of bubble stability and the properties of blown films made of blends of L-PP with LCB-PP.

EXPERIMENTAL

Materials

Two commercial PP resins, one with a linear structure (L-PP) and an other with long-chain branches (LCB-PP) were used in this study. Both resins were supplied by Basell. The L-PP (PF1280) has a melt flow index of 1.2 g/10 min whereas LCB-PP (Profax PF-814) has an index of 2.49 g/10 min under the same conditions (230[degrees]C, 2.16 kg). We will refer to these resins as L-PP and LCB-PP. The main characteristics of the resins used are listed in Table 1.

Blending Procedure

Blend samples were prepared using a Leistritz AG 34 mm co-rotating twin-screw extruder (TSE). The screw is 960 mm long and 34 mm in diameter and its configuration can be found in Bourry and Favis (1998). The temperature profile was controlled at 190/ 210/230/230/230/230/230[degrees]C from feed to exit. The screw speed was held at 50 rpm and the extruded materials were pelletized after passing through water at room temperature. The unblended resins were also subjected to the same procedure to give them the similar thermo-mechanical history as the blends. No stabilizer was added. Samples used for rheological tests were prepared using a 50 ml, Haake Rheocord 90 Chamber. The screw speed was set to 50 rpm and 210[degrees]C under a nitrogen environment. Also, 0.5 wt % of Irganox B225 antioxidant was added to avoid degradation during blending in the internal mixer and during rheological measurements. To compare the effect of the process on rheological properties, two types of samples were prepared, one using the Haake with antioxidant and the other using the TSE without antioxidant.

Rheological Characterization

Dynamic rheological measurements were carried out using a Rheometric SR5000 stress controlled rheometer. A time sweep under 0.1 rad/s was performed over 1 h to check the thermal stability for all the samples. The linear viscoelastic properties were determined for frequencies ranging from 0.01 to 500 rad/s, using strain and stress values determined to lie within the linear regime. The extensional behaviour of the resins and blends was determined using a Rheometric ARES rheometer, equipped with the new SER universal testing platform from Xpansion Instruments. The model used in our experiment was SER-HV-A01, which is a dual windup extensional rheometer and has been specifically designed for the ARES rheometer platform. It is capable of generating elongational rates up to 20 [s.sup.-1] under controlled temperature (up to 250[degrees]C).

Film Preparation

A lab-scale blown film unit was used to study the bubble stability. It is equipped with an in-line camera system and consists of a 45 mm Killion single-screw extruder with a helical annular die (outer diameter of 63.5 mm and a gap of 1.5 mm), dual-lip air ring and take-up device. The screw length to diameter ratio (LID) is 24 and the compression ratio is around 3. The extrusion temperature profile from the hopper to the die was set as 160/190/200/200/ 200/200[degrees]C for all the resins used. The mass flow rate was maintained at 2.0 [+ or - ] 0.1 kg/h during the bubble stability tests. Cooling was performed by using a dual-lip air ring, which is well known for its better cooling efficiency compared to that of a single-lip air ring. No internal cooling device was used. The room temperature was not regulated. The optical system used for bubble stability study was developed in our laboratory. Detailed information can be found in our previous publication (Kim et al., 2004).

Film Characterization

Differential scanning calorimetry (DSC) tests were carried out on a TA Instruments Q1000, with the heating rate of 20[degrees]C/min. Tensile tests were performed using an Instron 5500R machine. The procedure used was based on the D638-02a ASTM standard. For Fourier transformed infrared (FTIR) measurements, the spectra were recorded on a Nicolet Magna 860 FTIR instrument from Thermo Electron Corp., Waltham, MA (DTGS detector, resolution 4 [cm.sup.-1], accumulation of 128 scans). The beam was polarized by a Spectra-Tech zinc selenide wire grid polarizer from Thermo Electron Corp.

Tapping mode atomic force microscopy (TM-AFM) measurements were performed with a Digital Instruments NanoScope (R) III (from Veeco), using silicon cantilevers with a force constants of approximately 40 N/m and resonance frequency of approximately 320 kHz. The samples were imaged after cleaning by acetone and nitrogen in a relative humidity of about 35%. A field emission scanning electron microscope (FESEM-Hitachi S4700) was also employed for the observation of films. Although no particular preparation of the film surface was required, it was found preferable to rinse the surface of the films with acetone prior to the tests.

Haze (defined as the percentage of light that is scattered from a film sample for an incident beam of more than 2.5[degrees]) was measured in accordance with the procedures specified in the ASTM D 100397. The measurements were carried out on a Haze Guard Plus[TM] instrument (Model 4725) made by the BYK-Gardner Inc.

RESULTS AND DISCUSSION

Rheological Properties

Both samples obtained using the internal mixer and the TSE were tested and the rheological data showed differences of less than 3% (data not reported). The frequency sweep tests were performed at 230[degrees]C for the L-PP, LCB-PP, and their blends and the results are presented in Figure 1. Time sweep tests carried out for the samples at 230[degrees]C showed a thermal degradation less than 5% during the time required for running a frequency sweep test. As observed in Figure 1, the terminal zone and the zero-shear plateau were not reached at the lowest frequency for all the samples. The L-PP exhibits much larger complex viscosity values than the LCB-PP and the transition from the low frequency to the high frequency zone is somewhat broader for LCB-PP.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

To account for molecular weight effect, the data of Figure 1 were re-organized using the elastic modulus normalized by the zero-shear viscosity, determined using the Carreau-Yasuda model, which could be found in Carreau et al. (1997), and the results are shown in Figure 2. It is clear that the values of the normalized elastic modulus of the LCB-PP and blends are larger than that of the L-PP at low frequencies. The effect is inversed at high frequencies.

The semi-log additivity rule has been used to assess the miscibility of PE blends for LLDPE and LDPE blends, as suggested by Utracki and Schlund (1987). The results are reported here in Figure 3 for the complex viscosity of the PP blends versus blend composition at different frequencies (including the zero-shear viscosity). The complex viscosities of the blends 20:80 and 40:60 (LCB-PP/L-PP) at different frequencies including the zero-shear viscosity show a positive deviation from the semi-log additivity rule, suggesting some immiscibily of the PP components. As expected, the deviations decrease with increasing frequency. For lower temperature of 190[degrees]C, the agreement of the semi-log additivity rule was slightly better, indicative of improved miscibility with decreasing temperature (data not shown).

[FIGURE 3 OMITTED]

Stange et al. (2005) found that L-PP and LCB-PP blends could follow the semi-log additivity rule blends up to 50% of LCB-PP. This is not true in our case. Usually, the structure of the branched component such as the length and content of branches is an important factor on the miscibility, as previously shown (Hameed and Hussein, 2004).

The relaxation time spectra were calculated from the dynamic modulus data (G', G", [omega])) using the NLREG (non-linear regularization) computer software (Honerkamp and Weese, 1993) and the results are reported in Figure 4. The spectra show only one peak that shifts to longer times with the addition of LCB-PP in the range of 1-10 s while its shape gets broader. LCB-PP entangles with linear chains resulting in more entanglements that probably delay the relaxation of the L-PP chains. The intensity of the peak and the area under the peaks are also decreasing as a result of the reduction of the average molecular weight. The presence of single peaks of the spectra suggests miscibility of the blend components in contrast to the indication of the semi-log additivity rule. However, the deviations of the semi-log additivity rule are small as shown in Figure 3 and we cannot confirm immiscibility/miscibility from the rheological data.

Extensional Rheology

The transient elongational viscosities for L-PP, 20:80 LCB-PP/L-PP, 40:60 LCB-PP/L-PP, and LCB-PP at different elongational rates measured at 180[degrees]C are presented in Figure 5. Due to difficulties we met and the limitations of the device, the data could not be obtained at 230[degrees]C and only the results at strain rate larger than 1 [s.sup.-1] are presented. The L-PP does not show any strain-hardening behaviour, as expected. However, with the addition of 20% LCB-PP, pronounced strain-hardening behaviour is observed for all strain rates. The strain hardening becomes more pronounced with increasing LCB-PP content. It has been shown by Stange et al. (2005) that such systems even with less than 10 wt% of LCB-PP showed quite a pronounced strain-hardening behaviour. LCB-PP chains with tree-like architecture locally interlock with linear chains resulting in early strain hardening and enhanced melt strength of the blends. We also repeated the elongational tests for another research project at 200[degrees]C and a similar behaviour for the blends was observed.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

Before going into the bubble stability analysis, it is important to note that for a process such as film blowing, melt strength by itself is not a sufficient factor to claim good processability. In fact, the other factor called drawability is equally important. Melt strength is defined as the force required to pull a filament from an extruded melt onto a rotating wheel under given conditions, while drawability is the maximum speed that can be applied in fibre spinning by wrapping an extruded molten filament onto a rotating roll under constant acceleration, as described by Dupire and Michel (2004). Usually for a practical case a compromise between high melt strength and drawability is needed. To produce a stable film at a reasonable production rate, a molten film needs to sustain a large enough strain rate at the die.

Online Bubble Stability

As mentioned previously, bubble stability is a major issue in the film blowing of PP. The various types of bubble instabilities for PP can be classified as (1) draw resonance (DR), characterized by a periodic oscillation of the bubble diameter, (2) helicoidal instability, characterized by a helicoidal motion of bubble around its axial direction, and (3) FLH (frost line height) instability, characterized by variations in the vertical position of the FLH, as previously shown (Kim et al., 2004). It is usual to report instabilities as functions of the blow-up ratio (BUR) and take-up ratio (TUR). BUR is defined as the ratio of the bubble to the die external diameter and TUR is the ratio of the film velocity at the take up device to the polymer velocity in the die. The film thickness (gauge) has a linear relation with BUR and TUR for a given mass flow rate, which can be defined as follows:

Film gauge = C/TUR x BUR (1)

where

C = [H.sub.o] [[rho].sub.0][[rho].sub.1]

and [H.sub.o] is the die gap, [rho] is the density of the polymer. 0 stands for the position of the die exit and 1 is for the position where the film reaches the solid zone (above the frost line height).

[FIGURE 6 OMITTED]

[FIGURE 7 OMITTED]

One may easily convert either BUR or TUR to film gauge. However, the density at different temperatures could not be readily evaluated. That is why most publications are based on TUR versus BUR instead of film gauge. If we neglect the density changes with temperature, the equation will be simplified to:

Film gauge = [H.sub.0]/ TUR x BUR (2)

In this work, for the conditions used, the film gauge varied from approximately from 13 to 350 [Lm. Due to the un-uniformity of the film, which could be up to [+ or -] 30 % in cases, only BUR and TUR were used to make the instability plots.

The criteria for differentiating bubble instability behaviours for various PE resins were extensively studied and can be found elsewhere (Kim et al., 2004). Three parameters were used to evaluate the bubble instability: (1) eccentricity, d, which describes how far [alpha] bubble moves away from the die centre, (2) rotation angle, a (reference angle of the bubble right side with respect to an arbitrary y-plane), (3) variations in the radius of the left or right sides of the bubble ([R.sub.Left] and [R.sub.Right]). Relations of these parameters and geometrical variables in reference to Figure 6 are as follows:

Rotation angle ([alpha]) = arctan ([Y.sub.c]/[X.sub.c]) (3)

Eccentricity (d) = [square root of [X.sup.2.sub.c] + [Y.sup.2.sub.c] (4)

where [X.sub.c] and [Y.sub.c] are the Cartesian coordinates that define the bubble centre.

During bubble stability tests, the acquisition rate of 0.26 s/data was used for all measurements. Figure 7 presents the data collected on-line for the L-PP sample at a mass flow rate of 2.0 kg/h, extrusion temperature of 200[degrees]C, and BUR of 1.2. It is observed that at a TUR equal to 18 (Figure 7a) the bubble showed DR as observed for various PE resins (Kim et al., 2004) with important variations of the diameter. We arbitrarily set the variations of [+ or -] 2.5 % of diameter as the limit of a stable bubble. Increasing TUR to 23 without changing the other parameters (BUR, FLH), the radius variations got smaller; however, it is still beyond the stable bubble limit. When TUR was further increased to 29 and 34, the radius variations for these two conditions were almost the same and at the limit of a stable bubble. The rotation angle ([alpha]) and eccentricity (d), as shown in Figure 7b, are also within their respective fluctuation limit of a stable bubble ([alpha] < 60, d < 10 mm). Also, the amplitude of the rotation angle and eccentricity decrease with increasing TUR. It should be mentioned that no helicoidal instability was found for all experiments on the L-PP, LCP-PP, and their blends, even though FLH fluctuations were observed for most unstable bubbles.

[FIGURE 8 OMITTED]

To evaluate the improvement of the stable region with the addition of LCB-PP, stability maps are plotted in Figure 8 for the LPP (Figure 8a), 20:80 (Figure 8b), 40:60 (Figure 8c), and LCB-PP (Figure 8d), respectively. It was very difficult to obtain a thin film (going to large TUR and BUR) for the LCB-PP. This is more likely due to the poor drawability of this resin of low molecular weight, as shown by Dupire and Michel (2004). We observe that the area for the stable region increases with increasing content of the LCB-PP (Figures 8a to 8d).

Blending an LCB-PP with an L-PP represents a compromise between achieving melt strength and drawability that results into a stable bubble at a high production rate (large TUR). As it is seen in Figure 8, blending enhances the bubble stability range enabling us to achieve larger BUR and TUR values. To assess the stable region for all the PP systems, all the curves for the transition from stable to unstable bubbles have been drawn on the same graph of Figure 9. As expected, the stable region for the LCB-PP is much larger than that of the L-PP due to the presence of LCBs. However, we experienced leaking problems for the LCB-PP when the film was quite thin. The leaking, as explained above, could be due to the poor drawability of this resin, which limits the processing to low draw ratio, as indicated by Dupire and Michel (2004).

[FIGURE 9 OMITTED]

Thermal, Mechanical, and Optical Properties The DSC heating curves, which illustrate the melting behaviour of the L-PP, LCB-PP, and the 40 % blend are shown in Figure 10. The melting point of the L-PP is the highest (165[degrees]C) and with the addition of LCB-PP it decreases down to 152[degrees]C for the unblended LCB-PP. This suggests that the lamellae thickness of the L-PP diminishes with the addition of LCB-PP. However, it is observed that the crystallinity degree (characterized by the area under the curve) does not change significantly by blending with LCB-PP. Even though the LCB-PP has long chain branches, the back bone chains in comparison to the L-PP are shorter and the inclusion of such short chains leads to the formation of thinner lamellae. However, the main role of the long LCB-PP chains is to improve the melt processing prior to crystallization.

[FIGURE 10 OMITTED]

The orientation of the amorphous and crystalline phases (c-axis with respect to the machine direction, MD) of the samples has been measured by FTIR spectroscopy and the results are presented in Figure 11 (see Sadeghi et al., 2007 for details). The orientation function was calculated as (Ward et al., 2000):

[f.sub.i,MD] = (D - 1/D + 2) (5)

where D is the ratio of the absorbance in the machine (parallel) to that in the transverse direction (TD). For the orientation function of the crystal phase (c-axis with respect to MD), [f.sub.c], the band 998 [cm.sup.-1] is considered so D will be [([A.sub.II]/[A.sub.[perpendicular]]).sub.998], where A is the absorbance. To measure the total orientation that includes both the crystalline and the amorphous phase orientations the band at 972 [cm.sup.-1] was selected and [f.sub.av] (average orientation) was calculated based on ([A.sub.II]/[A.sub.[perpendicular]]).sub.972]. The orientation of the amorphous phase, [f.sub.a], can be determined from these two values by:

[f.sub.av] = [X.sub.c] [f.sub.c] + (1 - [X.sub.c] [f.sub.a] (6)

where [X.sub.c] is the degree of crystallinity and was calculated based on the DSC results. In uniaxial stretching the orientation function changes from 0 (for a non-oriented sample) to 1 (fully oriented sample). As it is observed in Figure 11, the orientation of the crystalline phase increases overall with adding LCB-PP (with a maximum around 20% LCB-PP) while the orientation of the amorphous phase slightly drops with blending. The increase in the orientation of the crystalline phase is likely attributed to the formation of a particular crystalline structure for the blend. The larger entanglement density for the blend system increases the number and the rate of formation of initial elongated fibrils in the nucleation stage of stress induced crystallization. This results in the formation of a row nucleated lamellar structure that is more oriented. The LCB-PP that contains shorter chains probably forms less tie chains, which are the major part of the amorphous phase, and that could explain the slight drop in the amorphous orientation with blending. It is worthwhile to mention that FTIR (if the results are analyzed properly) in transmission mode is a powerful technique for measuring the orientation of thin films with an error of less than 1 % for the orientation functions.

[FIGURE 11 OMITTED]

[FIGURE 12 OMITTED]

The surface morphology of the blown films was examined and two micrographs are shown in Figure 12. A typical spherultic structure for the L-PP films is shown in Figure 12a from SEM while for the 10% blend a lamellar structure is evident as shown in Figure 12b from AFM. Unfortunately, the SEM micrographs for the blends and the AFM micrographs for the L-PP films were not clear enough to be of any use.

Tensile tests were carried out on the films to examine the response of the crystalline structure to stretching. The main impact of adding LCB-PP to L-PP is a reduction of the mechanical strength. As depicted in Figure 13, yielding occurs almost for the all blends and becomes more significant as the content of LCB-PP increases. It is also observed that the elongation at break increases with the LCP-PP content. Incorporating a branched PP, which has a lower molecular weight than the L-PP (see Table 1), likely induces a smaller density of the tie chains for the crystalline structure. A direct relationship between the mechanical properties and the fraction of the tie chains has been assumed by Nitta and Takayanagi (2000).

[FIGURE 13 OMITTED]

The most important negative impact of adding LCB-PP to L-PP is a deterioration of the mechanical properties in the TD as shown in Figure 14. The film samples show small strength range as observed in MD (Figure 13) but bear a much smaller strain. Yielding is observed for L-PP while upon adding LCB-PP, the samples failed at a small strain of around 0.05 without showing any yielding, in contrast to the opposite behaviour observed for the tensile properties in MD (see Figure 13).

The haze value for the L-PP, LCB-PP, and their blends as a function of blend composition is presented in Figure 15. Haze decreases dramatically with the presence of LCB-PP and shows a minimum value for a blend containing 10% of the branched PP. We also observe that the haze value for all the blend compositions is smaller than that of either L-PP or LCB-PP. The high transparency of the blend films is attributed to the thinner crystallites and the row nucleated lamellar structure as shown in Figure 12b.

[FIGURE 14 OMITTED]

[FIGURE 15 OMITTED]

CONCLUSIONS

In this work, we have investigated the effects of adding a long-chain branched polypropylene (LCB-PP) into a linear-polypropylene (L-PP). Adding an LCB-PP significantly improved the melt strength resulting in a strain-hardening behaviour. The rheology characterization in small amplitude oscillatory flow showed a deviation from the log additivity rule for the blends, but the relaxation spectra revealed a single peak for the blends that shifts to longer times with increasing LCB-PP content. From bubble stability tests, it was found that area in the stability map increased with LCB-PP content. However, the addition of LCB-PP reduced the mechanical strength of the blown films with a more pronounced impact on the mechanical strength in TD. Upon blending the yielding behaviour in TD disappeared. The addition of long branches created a row nucleated lamellar structure for the blend films, which favoured orientation of the crystals blocks along MD. However, the amorphous orientation decreased with blending. The lamellar structure formed for the blend films showed thinner lamellae that could explain the reduction in the mechanical strength of the blown film samples. This could also explain the drastic decrease in the haze value for the films made with the blends. This could be of interest for applications where film transparency is an important issue.

ACKNOWLEDGEMENTS

Financial support from NSERC (Natural Science and Engineering Research Council of Canada) and from FQRNT (Fonds Quebecois de Recherche en Nature et Technologies) is gratefully acknowledged.

NOMENCLATURE

A absorbance
AFM atomic force microscopy
BUR blow-up ratio (film diameter divided by die
 diameter)
D ratio of the absorbance in the machine
 (parallel) to that in the transverse direction
d eccentricity (mm)
[f.sub.a] orientation of amorphous phase
[f.sub.av] average orientation
[f.sub.c] orientation function of the crystal phase
[f.sub.i] orientation function (defined by Equation (3))
FLH frost line height (mm)
G' storage modulus (Pa)
H([tau]) relaxation spectrum (Pa)
[DELTA]H heat flow (W/g)
L-PP linear polypropylene
LCB-PP long-chain branch polypropylene
[M.sub.w] molecular weight (g. mol-1)
[M.sub.w]/[M.sub.n] molecular weight distribution
MD machine direction
MFI melt flow index (g)
PP polypropylene
[R.sub.Left] radius of bubble on left side (mm)
[R.sub.Right] radius of bubble on right side (mm)
SEM scanning electron microscopy
T temperature ([degrees]C)
TD transverse direction
TUR take-up ratio (roll velocity divided by the
 extrudate velocity at die)
t time (s)
[X.sub.c] crystallinity degree

Greek Symbols

[alpha] rotation angle of bubble ([degrees])
[[eta].sub.o] zero shear viscosity (Pa x s)
[[eta].sup.*] complex viscosity (Pa x s)
[[eta].sub.E] uniaxial elongational viscosity (Pa x s)
[tau] relaxation time (s)
[omega] frequency (rad/s)

Manuscript received March 22, 2007; revised manuscript received June 15, 2007; accepted for publication June 22, 2007.

REFERENCES

Bourry, D. and B. D. Favis, "Cocontinuity and Phase Inversion in HDPE/PS Blends: Influence of Interfacial Modification and Elasticity," J. Polym. Sci.: Part B Polym. Phys. 36,1889-1899 (1998).

Carreau, P. J., D. De Kee and R. P. Chhabra, "Rheology of Polymeric Systems: Principles and Applications," New York, Hanser, Munich, Germany (1997).

Dupire, M. and J. Michel, "Polypropylene with High Melt Strength and Drawability," US Patent 6723795 (2004).

Ghaneh-Fard, A., P. J. Carreau and P. G. Lafleur, "Study of Instabilities in Film Blowing," AIChE J. 42, 1388-1396 (1996).

Gotsis, A. D. and B. L. F. Zeevenhoven, "The Effect of Long Chain Branching on the Processability of Polypropylene in Thermoforming," Polym. Eng. Sci. 44, 973-982 (2004).

Hameed, T. and I. A. Hussein, "Effect of Short Chain Branching of LDPE on Its Miscibility with Linear HDPE," Macromol. Mater. Eng. 289, 198-203 (2004).

Han, C. D. and J. Y. Park, "Studies on Blown Film Extrusion. III. Bubble Instability," J. Appl. Polym. Sci. 19, 3291-3297 (1975).

Honerkamp, J. and J. Weese, "A Non Linear Regularization Method for the Calculation of Relaxation Spectra," Rheol. Acta 32, 65-73 (1993).

Kanai, T. and J. L. White, "Kinematics, Dynamics and Stability of the Tubular Film Extrusion of Various Polyethylenes," Polym. Eng. Sci. 24,1185-1201 (1984).

Kim, S., Y. Fang, P. G. Lafleur and P. J. Carreau, "Dynamics and Criteria for Bubble Instabilities in a Single Layer Film Blowing Extrusion," Polym. Eng. Sci. 44, 283-302 (2004).

Lagendijk, R. P., A. H. Hogt, A. Buijtenhuijs and A. D. Gotsis, "Peroxydicarbonate Modification of Polypropylene and Extensional Flow Properties," Polymer 42, 10035-10043 (2001).

Nitta, K. H. and M. Takayanagi, "Tensile Yield of Isotactic Polypropylene in Terms of a Lamellar-Cluster Model," J. Polym. Sci.: Part B Polym. Phys. 38,1037-1044 (2000).

Sadeghi, F., A. Ajji and P. J. Carreau, "Analysis of Microporous Membranes Obtained from Polypropylene Films by Stretching," J. Membr. Sci. 292, 62-71 (2007).

Stange, J., C. Uhl and H. Munstedt, "Rheological Behavior of Blends from a linear and a Long-Chain Branched Polypropylene," J. Rheol. 49, 1059-1079 (2005).

Utracki, L. A. and B. Schlund, "Linear Low Density Polyethylenes and Their Blends: Part 4 Shear Flow of LLDPE Blends with LLDPE and LDPE," Polym. Eng. Sci. 27, 1512-1522 (1987).

Ward, l. M., P. D. Coates and M. M. Dumoulin, Eds., "Solid Phase Processing of Polymers," Hanser, Munich, Germany (2000).

Yunli Fang, Farhad Sadeghi, Guillaume Fleuret and Pierre J. Carreau *

Center for Applied Research on Polymers and Composites (CREPEC), Chemical Engineering Department, Ecole Polytechnique, P. O. Box 6079, Stn Centre-Ville, Montreal, QC, Canada H3C 3A7

* Author to whom correspondence may be addressed. E-mail address: pcarreau@a polymtl.ca

DOI 10.1002/cjce.20011

Table 1. Polymers used and their main characteristics

Sample LCB-PP

Manufacturer Basell (PF 814)
MFI (g/10 min) 2.49 (230[degrees]C/2.16 kg)
Density (kg/[m.sup.3]) 900
[M.sub.w] (g/mol) 290 000
[M.sub.w]/[M.sub.n] 4.9

Sample L-PP

Manufacturer Basell (PF 1280)
MFI (g/10 min) 1.2 (230[degrees]C/2.16 kg)
Density (kg/[m.sup.3]) 902
[M.sub.w] (g/mol) 425 500
[M.sub.w]/[M.sub.n] 3.4