# Novel polypropylenes for foaming on conventional equipment.

Polypropylene (PP) has a high melting point (-160[deg.]C), excellent chemical resistance, and a relatively high tensile modulus, properties that have made it one of the most widely used thermoplastics. However, PP's low elongational viscosity and poor melt strength have prevented its widespread use in extruded, low-density foams. The processing window for the semicrystalline PP, unlike many amorphous polymers such as polystyrene, is relatively narrow-only -4[deg.]C. The lower bound of the foaming window is set by the melting point of the polymer, and the upper limit is simply the temperature at which the cell walls begin to collapse because the viscositY is too low. Significant improvements in PP technology have resulted in increased extensional viscosity, thus making possible the production of low-density foam on conventional equipment.

This article demonstrates the importance of extensional or elongational viscosity in the foam process, compares new and conventional PPs in extensional flow, and shows how these rheological differences allow production of low-density foam on tandem extrusion equipment. Bubble Dynamics The bubble dynamics of foam extrusion are quite complex (reviewed in C.D. Han, Multiphase Flow in Polymer Processing, Academic Press, 1981), but a simple analysis of the growth of a spherical viscous bubble demonstrates the salient features of the foaming process. The normal stress difference that acts over the wall of a spherical bubble is the hoop stress, given by: [mathematical expression omitted] where, [delta]P is the pressure drop across the expanding bubble (assumed constant in this analysis), R is the radius of the bubble or cell, and h is the thickness of the bubble. Introducing the constitutive equation for a Newtonian fluid, one obtains a simple differential equation for bubble growth. Assuming h [is not greater than] R: [mathematical expression omitted] where [eta] is the viscosity of the fluid. In a real system this pressure drop is generally the partial pressure of the blowing agent in the polymer minus ambient pressure. Introducing the continuity equation and solving Equation 2 for R as a function of time, one gets the result: [mathematical expresson omitted] where R. is the initial volume after nucleation, t is time, and V is the volume of material in the cell wall. This equation, though simple, predicts the features of an actual foaming system, namely the importance of polymer viscosity and blowing agent partial pressure. Figure I shows that at a given viscosity, the rate of bubble growth increases rapidly with time. This phenomenon explains why large bubbles grow faster than small ones and why it is critical for foaming to occur uniformly at the exit of the die. The value of the viscosity has a pronounced effect on bubble growth and is a parameter controlling cell collapse.

A similar analysis F.N. Cogswell, Polymer Melt Rheology, John Wiley & Sons, 1981) shows that the stress level in the polymer increases with the cube of the radius. Therefore, if a polymer's viscosity decreases rapidly with increasing stress or if the viscosity function decreases as strain increases, unstable cell growth can occur or nonuniform cell size will dominate. If the elongational viscosity increases with stress or time, cell growth can be stabilized. The actual dynamics of a real foaming system are confounded by bubble interactions, nonisothermal and nonisobaric foaming resulting from blowing agent evaporation, and the nonNewtonian behavior of polymeric fluids. Therefore, the extruded foam process cannot readily be rigorously analyzed. Rheological Analysis A 3.5-melt-flow conventional homopolymer (Pro-Fax 6523) and a 7.0-melt-flow high-melt-strength, foamable homopolymer were compared. Zero-shear viscosities were 1.0 X 10[sup.4] and 7.3 x 110[sup.3] Pa-sec, respectively. The extensional viscosity growth functions were measured on a Rheometrics RER 9000 extensional rheometer at 180[deg.]C and at a constant strain rate. The instrument pulls a cylindrical rod of molten polymer uniaxially. Force and strain are measured as a function of time, from which the viscosity growth function is calculated by the equation: [mathematical expression omittee] where [mathematical expression omitted] is the extensional viscosity growth function, F(t) is the measured force, i; is the rate of strain, A[sub.o] is the initial cross-sectional area, and t is time. Measurements of the elongational viscosity by converging flow and melt tension by a Goettfert Rheotens apparatus supplemented the RER 9000 results. Converging flow analysis measures the apparent extensional viscosity from the entrance pressure loss into a small L/D die. Streamlines at the die entrance converge, forcing the volume elements to stretch. Cogswell describes how to estimate the extensional viscosity from the entrance pressure loss, which can be taken as the Bagley end corrections for simplicity. Following Cogswell: [mathematical expression omitted] where, [eta[sub.e] is the apparent extensional viscosity, n is the power-law index, P[sub.o]. is the Bagley ends correction, [eta] is the apparent shear viscosity for fully developed flow at the apparent shear rate, [gamma]. The values obtained by this technique are not true elongational viscosities, but relative comparisons, much like a melt index.

A second indexer type test was conducted at 200[deg.]C on the melt tension apparatus, a device similar to a small fiber line. Polymer melt exits a capillary rheometer and is stretched by a pair of gears. The frequency of rotation increases linearly with time, and force is measured by a transducer. This experiment is nonhomogeneous and nonisothermal, but it is widely used in industry to gage the melt strength and processability of polymers. Foam Line and ProcedureBecause standard equipment for PS and LDPE foam production is not available in laboratory scale, experiments were conducted on a single screw, 38-mm-diameter, vented, 24:1-L/D extruder that was modified to simulate a tandem system as closely as possible. The compression ratio in the first section is 2.7:1. The polymer is fully melted in the first stage. The melt passes over a 1.5-flight blister section with a 1.6-mm clearance and into the second stage. CFC-1 14 blowing agent is metered by a positive displacement pump into the melt stream through the extruder vent located at the feed section of the second stage. (Environmentally friendly blowing agents are suitable for foamable PP, but because of facility limitations, CFC-114 was used for experimentation.) The melt is recompressed and metered into a 25-cm-long, electrically heated, static mixing section where the mixing and cooling of the polymer/ blowing agent solution is accomplished. The die has a 28.6-mm wide opening, with the gap adjustable from 0 to 3.2 mm.

Sample preparation consisted of coating the pellets with 0.2% mineral oil and adding 0.75% Cantal 350 talc as the cell nucleator by tumble blending. The line was preheated and started up at the conditions listed in Table 1.

Because the melt viscosity is reduced by the CFC- 114 blowing agent, the temperatures must be reduced after start-up to prevent internal expansion. To obtain the optimum melt viscosity and die pressure, the temperatures were systematically lowered in the barrel, mixer, and die zones after blowing agent was introduced. The die gap at this point was approximately 0.5 mm. As the viscosity increased, and die pressure reached a minimum of 17 bars, expansion began to occur outside the die. As the melt temperature was further reduced, the die gap was opened. At too low a temperature, solid polymer was found in the extrudate. At optimum conditions, the die gap was adjusted to the maximum opening where only external expansion is achieved. In these experiments, the foam expanded freely from the die. Rheological Measurements The shear viscosities of the two polymers Fig. 2) are in accord with the melt flow rates, the conventional polymer having the higher shear viscosity However, the extensional viscosity growth functions are markedly different for the two polymers (Fig. 3). The viscosity function increases slowly with time and then drops sharply for the conventional PP. The sharp drop corresponds to ductile failure of the specimen. The high-melt-strength PP shows a pronounced strain-hardening behavior. In general, these PPs do not fail in a ductile manner, but if strained rapidly, they will undergo cohesive failure. Plots of the viscosity functions as functions of Hencky strain Fig. 4) also demonstrate the marked difference between the two polymers. Data in Figs. 3 and 4 were taken at 180[deg.]C.

The high-melt-strength PP has an elongational viscosity Fig. 5), as measured by converging flow, about an order of magnitude greater than that of the conventional resin, and even greater than that of the 0.75-MFR PP (Pro-Fax 6823). The melt tension results reflect similar behavior. The maximum melt tension of the foamable PP is about 13 cN, compared with 2.1 cN for the conventional PR These experiments clearly show that the elongational flow behavior of the high-melt-strength resins is different in character and higher in value than conventional resins of similar melt flow. Foam Processability Resin processability was based on maximum expansion, density reduction, and maximum die gap opening. Cell size, cell uniformity, and surface appearance were characterized by ASTM methods or slight variations. Physical properties were not measured because the foam sample geometries were unsuitable for testing.

Injection of CFC-114 into the conventional PP melt stream resulted in an unsteady-state condition, as the blowing agent was observed exiting through the die without mixing with the polymer. Reducing the melt temperature to 171'C increased the viscosity sufficiently to prevent internal expansion with the die opening at 100% (3.2 mm). The reduced pressure at the die exit allowed the melt to expand. However, the low extensional viscosity caused the rupture of the cell walls and a total collapse of the Structure. Further lowering of the temperature to increase the melt strength by raising the viscosity only

resulted in the inability to produce a homogeneous polymer/blowing agent solution.

In comparison, the high elongational viscosity of the high-melt-strength PP enabled it to resist cell wall rupture. This resulted in retention of the blowing agent and significant expansion and density reduction with little or no collapse or shrinkage. The uniform, closed-cell structure formed when the polymer solidified is shown in Fig. 6, where micrographs of the conventional PP, high-melt-strength PP, and commercial polystyrene foams are compared. The average cell diameter for the high-melt-strength PP was 0.9 mm, and a density reduction of [is greater than] 90% was achieved (Table 2). Similar processing characteristics for the foamable PPs were observed on commercial-scale equipment (114-mm-152-mm tandem extruder). Density reductions as high as 95% and densities as low as 15 kg/ m[sup.3] have been reported.
``` TABLE 1. Extruded Foam Pro
- cessing Temperatures, [deg.]C
- Temperature Start-up state
Extruder zone
1 177 177
2 205 201
3 205 185
4 205 176
Adaptor zone 218 175
Melt 210 170
```
COPYRIGHT 1991 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.