Library

Notes: Summary An experimental study has been carried out to better understand the flow behavior of viscoelastic polymeric melts in a converging channel which consists of two nonparallel planes. For the study, measurements were taken of both stresses and velocities, the former by means of the flow birefringence technique and the latter by means of streak photography. The material used was polystyrene. A comparison was made of the stress-birefringent patterns obtained in the present study with those obtained in an earlier study byHan andDrexler, who used a slit die having a tapered entrance. It has been found that both stresses and velocities in the converging flow field of polymeric melts do not exhibit any noticeable evidence of secondary motion, contrary to the theoretical prediction made by earlier investigators. The apparently unnoticeable secondary motion in the present study may be attributable to the extremely slow motion of the polymeric melts investigated. Also carried out was a theoretical analysis of converging flow. This was essentially the same as that carried out earlier byHan andDrexler, using a modified second-order fluid model which assumes that all three material functions depend on the second invariant of the rate of deformation. A comparison was made of the experimentally determined stress distributions with the theoretically predicted ones.

Notes: Summary Measurements were taken of the wall normal stress of viscoelastic polymeric melts (high density polyethylene and polypropylene) flowing through a converging channel having a half-angle of 15 degrees. It has been found that as the melt approaches the die exit, the wall normal stress rapidly increases, which is believed due to the rapid acceleration of the melt in the converging flow field. Using a modified secondorder fluid, theoretical expressions have been derived for the wall shear stress, normal stress difference at the channel wall, and the wall normal stress gradient in the flow through a converging channel and through a conical duct. It has been shown that, contrary to the situation in a viscometric flow field, in the converging flow field of viscoelastic fluids measurement of wall normal stress alone does not permit us to determine the pressure gradient of viscoelastic fluids.

Notes: Multilayer flat film coextrusion was studied, both experimentally and theoretically. For the experimental study, a sheet-forming die with a feedblock was designed, and plastic films of three and five layers were coextruded. The die was provided with three pressure transducers in the axial direction in order to determine the pressure gradient in the die, allowing the determination of the reduction in pressure drop when different combinations of two polymer melts were coextruded. Polymers used for coextrusion were: (1) low density polyethylene and ethylene-vinyl acetate; (2) low density-polyethylene and high density polyethylene; (3) low density polyethylene and polystyrene. For the theoretical study, the z-component of the equations of motion for steady fully-developed flow were solved using a power law non-Newtonian model, Comparisons were made between the experimental and the theoretically predicted volumetric flow rates. Predictions of the velocity distributions, shear rate profiles, and shear stress distributions were made as functions of the processing conditions and the rheological properties of the individual polymers concerned.

Notes: An experimental study was carried out to investigate the phenomenon of interfacial instability in multilayer flat-film coextrusion. For the study, a sheet-forming die with a feed block was used to coextrude three-and five-layer flat films. Polymers coextruded were: (a) low-density polyethylene with polystyrene, and (b) high-density polyethylene with polystyrene. It was observed that, for a given polymer system, there is a critical value of wall shear stress at which an irregular (i.e., unstable) interface between the layers sets in, giving rise to a pattern similar to that usually found in a wood panel. Once the instability sets in, the severity of interfacial instability is found to depend on both the total volumetric flow rate (hence wall shear stress) of the combined streams and the ratio of the individual layer thicknesses. An attempt is made to correlate the critical conditions for the onset of interfacial instability in terms of the layer thickness ratio, and the viscosity and elasticity ratios of the two polymers being coextruded.

Notes: An experimental study was carried out to gain a better understanding of the dynamic behavior of gas bubbles during the structural foam injection molding operation. For the study, a rectangular mold cavity with glass windows on both sides was constructed, which permitted us to record on a movie film the dynamic behavior of gas bubbles in the mold cavity as a molten polymer containing inert gas was injected into it. The mold was designed so that either isothermal or nonisothermal injection molding could be carried out. Materials used were polystyrene, high-density polyethylene, and polycarbonate. As chemical blowing agents, sodium bicarbonate (which generates carbon dioxide), a proprietary hydrazide and 5-phenyl tetrazole, both generating nitrogen, were used. Injection pressure, injection melt temperature, and mold temperature were varied to investigate the kinetics of bubble growth (and collapse) during the foam injection molding operation. It was found that the processing variables (e.g., the mold temperature, the injection pressure, the concentration of blowing agent) have a profound influence on the nucleation and growth rates of gas bubbles during mold filling. Some specific observations made from the present study are as follows: an increase in melt temperature, blowing agent concentration, and mold temperature brings about an increase in bubble growth but more non-uniform cell size and its distribution, whereas an increase in injection pressure (and hence injection speed) brings about a decrease in bubble growth but more uniform cell size and its distribution. Whereas almost all the theoretical studies published in the literature deal with the growth (or collapse) of a stationary single spherical gas bubble under isothermal conditions, in structural foam injection molding the shape of the bubble is not spherical because the fluid is in motion during mold filling. Moreover, a temperature gradient exists in the mold cavity and the cooling subsequent to mold filling influences bubble growth significantly. It is suggested that theoretical study be carried out on bubble growth in an imposed shear field under nonisothermal conditions.

Notes: The effect of deformation history on the elongational behavior and spinnability of polypropylene melt was investigated by carrying out isothermal melt-spinning experiments. For the study, spinnerettes of different die geometries were used to investigate the effect, if any, of the entrance angle, the capillary length-to-diameter (L/D) ratio, and the reservoir-to-capillary diameter (DR/D) ratio on the elongational behavior of molten threadlines. An experimental study was also carried out to investigate the phenomenon of draw resonance in the extrusion of polypropylene melts through spinnerettes of different die geometries. Draw resonance is the phenomenon which gives rise to pulsations in the threadline diameter when the stretch ratio is increased above a certain critical value. The results of our study show that the critical stretch ratio at which the onset of draw resonance starts to occur decreases as the L/D ratio is decreased, as the entrance angle is increased, as the DR/D ratio is increased, as the melt temperature is decreased, and as the shear rate in the die is increased. Of particular interest is the observation that, at 180°C, the severity of fiber nonuniformity increases as the stretch ratio is increased, whereas at 200°C and 220°C, the severity of fiber nonuniformity first increases and then decreases as the stretch ratio is increased considerably above the critical value. A rheological interpretation of the observed onset of draw resonance is presented with the aid of the independently determined rheological data.

Notes: An experimental study of sandwich foam coextrusion was carried out, using a sheet-forming die with a feedblock. Polymers used for the experiment were low-density polyethylene (LDPE) for the outer layers and ethylene-vinyl acetate (EVA), with the chemical blowing agent azodicarbonamide, for the foamed core component. The present study has shown that the cell size and its distribution in the foamed core and the mechanical properties of the sandwiched foam product can be controlled by a judicious choice of the thickness ratio of the core to skin components, the meltextrusion temperature, and the concentration of chemical blowing agent.

Notes: The effects of molecular structure and cooling conditions on the severity of draw resonance was investigated by carrying out carefully controlled melt spinning experiments. For the study, two types of polymeric materials were used: one which exhibits viscoelastic behavior (high-density polyethylene, polypropylene, and polystyrene), and the other which exhibits almost Newtonian behavior [nylon-6 and poly(ethylene terephthalate)]. In order to investigate the effect of cooling on the severity of draw resonance, different methods of cooling the molten threadline were employed. In one set of experiments, isothermal chambers of various lengths (3, 6, and 12 in.) were attached to the spinnerette face, so that the molten threadline, upon exiting from the spinnerette, began to cool in the ambient air only after it had passed through the isothermal chamber. This method of cooling is called “delayed cooling,” providing both an isothermal region (inside the isothermal chamber) where only stretching occurs, and a nonisothermal region (outside the isothermal chamber) where both stretching and cooling occur simultaneously. In other experiments, the temperature profile of the molten threadline was controlled by adjusting the temperature of the heated chamber. This method of cooling provides a gradual drop of the threadline temperature, compared to the more sudden drop when spinning into a cold environment provided at the spinnerette exit. The severity of draw resonance was recorded on movie film, and the thread tension was measured with a low-force load cell transducer and recorded on a chart recorder. The temperature of the threadline along the spin direction was measured using a fiber optical probe attached to a Vanzetti Infrared Thermal Monitoring System (Model TM-1). It was found that the severity of draw resonance depended on the molecular structure and the way the molten threadline was cooled. Of particular interest is the observation that, for the viscoelastic materials investigated, cooling destabilized the molten threadline outside the isothermal chamber. This gave rise to more severe resonant behavior, at and above the critical draw-down ratio, in contradiction to the theoretical prediction by Fisher and Denn. It was observed, also, that the elasticity of the materials tended to destabilize the molten threadline (i.e., it increased the severity of draw resonance), again in contradiction to the theoretical prediction of Fisher and Denn. It is believed that morphological changes of polymers may play an important role in the occurrence of draw resonance when a melt threadline is stretched under cooling. Our study indicated that a good understanding of draw resonance of viscoelastic fluids requires more careful study than the classical hydrodynamic stability analysis reported by Fisher and Denn. They based their analysis on several convenient and yet unjustified assumptions, and solely on phenomenological considerations. We suggest that future theoretical analysis of draw resonance be carried out by considering a fluid model with a nonlinear memory function in order to properly account for the deformation history of the fluid, and the relaxation and cooling processes in the die swell region and the region below it.

Notes: An experimental study was carried out to investigate the flow behavior of gas-charged molten polymers in foam extrusion. For the study, a rectangular slit die with glass windows was constructed to permit visual observations, from the direction perpendicular to flow, of the dynamic behavior of gas bubbles when a gas-charged molten polymer flows between two parallel planes. Pictures were taken of gas bubbles in the flow channel with the aid of a camera attached to a microscope, and these were later used to determine the position at which gas bubbles start to grow. Using three melt pressure transducers mounted on the short side of the rectangular slot, pressure distributions were measured along the longitudinal centerline of the die. The polymeric materials used were high-density polyethylene and polystyrene, and the chemical blowing agents used were a proprietary hydrazide which generates nitrogen, and sodium bicarbonate which generates carbon dioxide. It was observed that the gas-charged molten polymer shows a curved pressure profile as the melt approaches the die exit, whereas the polymer without a blowing agent shows a linear pressure profile. The visual observations of the bubble growth in the flow channel, together with the pressure measurements, permitted us to determine the bubble inflation pressure, often referred to as the critical pressure for bubble inflation. It was found that the critical pressure decreases with increasing melt extrusion temperature, and increases with increasing blowing agent concentration. It was also found that the bulk viscosity of gas-charged molten polymers decreases with increasing blowing agent concentration and with increasing melt temperature. A general remark is made concerning the precaution one should take when an Instron rheometer is used for determining the bulk viscosity of gas-charged molten polymers.