Volume 7, Issue 1

The Farrel Continuous Mixer has been an important compounding machine for a quarter century. There have, until this time, been no efforts to simulate flow in this machine. We apply a hydrodynamic lubrication theory-based analysis to this problem. We present a model of flow in a Farrel Continuous Mixer which predicts pressure and flux profiles for a fully filled machine. Calculations are reported for both Newtonian and non-Newtonian models.

A summary of recent technological developments related to large diameter screw extruders is presented. Special attention is given to cascade screw extruders.

Reciprocating extruders are used in injection molding. The number of reciprocating extruders used in the plastics processing industry is larger than conventional extruders. Therefore it is surprising that only few studies have been made of the conveying and melting process in reciprocating extruders.

The present work is a study of the thermo-viscous effects in the extrusion of a “power law” liquid with temperature dependent viscosity by including the effects of solidification. This is achieved by using a novel inverse isotherm finite element method. The method is based on an inversely defined (but not inverted) formulation where the isotherms are characteristic transverse coordinate lines. The method allows the simultaneous evaluation of all the field variables as well as the apriori unknown geometric parameters. It is shown that under certain conditions the extrudate can be significantly affected by the presence of the solidification front. Results are presented for die-swell, pressure, and location of solidification as a function of the material properties.

A basic study of the kinematics and dynamics of the double bubble tubular film process as applied to polyethylene terephthalate (PET), poly p phenylene sulfide (PPS), polycarbonate (PC) and low density polyethylene (LDPE) is presented. The stability of the four resins was investigated for both a single bubble tubular film and a second bubble of a double bubble process. For the single bubble the LDPE is very stable and the PET, PPS and PC have more limited regions of stability with periodic axisymmetric instabilities near blowup ratios of unity and helical instabilities at higher blowup ratios. It is found that in the first bubble the PC vitrifies to a glass, PET and PPS crystallize to 4 to 9% levels and LDPE to nearly 50%. In the second bubble it is found possible to operate in a stable manner with a heated air ring and stable bubble for PET and PPS but not PC and LDPE. In the second bubble PET and PPS show substantial levels of increase of crystallinity. This is not the case for PC and LDPE. It appears that a strain hardening process such as crystallization is necessary for stability of the second bubble. Stress fields in the second bubble are computed using membrane theory.

Melt strength data are presented on different types of binary polyethylene blends as linear low density polyethylene/low density polyethylene (LLDPE/LDPE), linear low density polyethylene/high density polyethylene (LLDPE/HDPE) and LDPE/LDPE. Melt strength was assessed from uniaxial tensile experiments with a Rheotens apparatus. In addition, the blends were characterised in terms of a few other rheological properties such as the zero-shear viscosity and the extrudate swell in capillary extrusion. For the LLDPE/LDPE blend pairs synergistic effects were observed for the melt strength and some other properties. No melt strength synergism was observed for the LLDPE/HDPE and the LDPE/ LDPE blends. The results are discussed in terms of the Bersted model on the effects of long chain branching on the low-shear melt viscosity of polyethylenes.

On-line uniaxial deformation behavior and deformation induced crystallization characteristics of initially amorphous syndiotactic polystyrene (s-PS) sheets were studied using an Iwamoto biaxial stretcher operated in the uniaxial mode. The stress strain curves obtained under various conditions indicate that this polymer can be stretched uniformly at 110°C which is between its glass transition temperature and cold crystallization temperature. The principal refractive indices were determined as a function of deformation conditions. Because of the negative intrinsic birefringence in the s-PS polymer chains the refractive indices in stretching direction decrease and those in the transverse direction increase with deformation.

The present paper describes a three-dimensional finite element method for calculating thermoviscoelastic stresses generated during cooling of injection molded parts. The initial temperature field corresponds to the end of the filling stage. The fully time-dependent algorithm is based on the calculation of the elastic response at every time step. The time-temperature equivalence is assumed for the non-isothermal behavior. At the end of the calculation, one obtains the residual stresses and the shape of molded parts at room temperature. The method is illustrated by means of four examples.

A numerical approach is taken to the analysis and simulation of structural reaction injection molding (SRIM). Formulation is based on anisotropic Darcy flow through a porous medium. Governing equations are developed for thin gap mold cavities with no gapwise flow. The reaction of the resin and its effect on the local viscosity are accounted for by using constitutive models. A numerical grid generation method is employed to handle irregularly shaped flow domains during the filling stage. A new finite difference mesh is generated at each time step to cover the flow domain as described by the mold boundaries and the moving resin front. The technique is shown to be capable of handling a wide range of irregularly shaped planar geometries in monitoring the instantaneous state of the cure along with the temperature and the flow fields in a detailed form. The results indicate a highly pronounced disturbance of the temperature field due to the exothermic nature of the chemical reaction. It is also shown that this temperature field and the state of cure significantly influence the flow field through fluid viscosity. The method reduces to the simulation of resin transfer molding (RTM) processes by dropping the reaction terms. The comparisons with limited experimental data for isothermal flows indicate high accuracy in both predicting the resin front locations and the fill pressures.

Flexural test specimens were injection-molded from random and block polypropylene copolymers with various ethylene contents at cylinder temperatures of 200 to 320°C. Properties such as flexural modulus (FM), flexural strength (FS), heat distortion temperature (HDT), Izod impact strength (IIS), and mold shrinkage (MS), and degrees of molecular orientation such as thickness of skin layer (ST) and crystalline orientation functions and their distributions in the flow and thickness directions were measured. The relationships between these properties and the degrees of molecular orientation were investigated. These properties decreased with increasing cylinder temperature. FM, FS and HDT were decreased by copolymerization with ethylene, in particular, by random copolymerization. IIS was increased by copolymerization with ethylene. MS was decreased by random copolymerization and slightly increased by block copolymerization. ST and crystalline c-axis orientation function f c decreased with increasing cylinder temperature and ethylene content and on increasing distance from the gate. f c is high at the surface region and decreases toward the interior. The higher the cylinder temperature and ethylene content, the lower is f c at the surface region. There are cases at high cylinder temperatures where a shoulder or a peak of f c is observed at a middle point in the thickness direction, which is assumed to be caused by secondary flow during cooling and pressure holding process. Rectilinear relationships between the properties and ST or f c were obtained for each sample. This means that the properties cannot be unified only with degree of molecular orientation and the character of sample still remains. The character of the sample is assumed to be originated from higher-order structures beside molecular orientation, for example, crystallinity, spherulite size, and thickness of lamellae, etc.