Band 7, Heft 4

Waste from polymer materials – no matter whether filled or unfilled, extremely impure or mixed – can be divided more or less into their molecular components by chemical means.

The mixing of plastic into filled and unfilled polymer blends has been an important issue in the polymer industry. Processing difficulties with these polymers have been encountered in the mixing quality as well as in the thermal degradation due to viscous heating. Mixing often occurs as an element of the processing step, e.g., inside single and twin screw extruders used in the fabrication of final parts or sheets, and inside internal mixers such as the Banbury type mixer. Quantifying the mixing inside an extruder or an internal mixer and predicting the thermal degradation due to viscous heating is an extremely difficult task. A better understanding of the mixing process and control of viscous heating will lead to optimal parts and may eventually allow us to increase the relative amount of fillers within the material. This paper presents a boundary element simulation of the flow of filled and unfilled polymer blends inside extruders and internal mixers. First, the general equations that govern such flows are shown. The boundary integral equation and the fundamental solutions are then formulated followed by the numerical implementation and logistics of the simulation. After the theoretical background is presented, a numerical example is given. The paper then shows how the simulation was used to analyze the flow inside a single screw extruder and internal batch mixers. Such a simulation can be used to eliminate some of the tedious trial-and-error tasks that are typically performed in the early stages of material synthesis and can also be used by manufacturers of mixing equipment when optimizing the geometries of cavities and mixing heads to achieve optimum mixing with reduced viscous heating. This research will significantly increase our knowledge of the behavior of filled and unfilled polymer blends and expand our understanding of the complex phenomena that take place during their mixing process.

High Reynolds' number flows between narrow gaps are very common in the reaction injection mold filling process (RIM) where low viscosity polymers are injected into thin mold cavities in relatively short time intervals. The resulting short cycle times have given RIM advantages over other manufacturing techniques in the polymer industry. In addition to the lower temperatures encountered during the RIM process, the low material viscosities enables mold filling using smaller injection pressures and mold clamping forces than in the conventional thermoplastic injection mold filling process. Considering the low viscosity of the injected fluid, short injection times and considerably large parts, such as automotive body panels, it is clear that in RIM processes inertia effects are not negligible. In fact Reynolds' numbers in such processes can reach 10 and above. This paper presents an experimental study performed on the high Reynolds' number flows between narrow gaps. First, the general equations that govern such flows are derived and written in dimensionless form. The magnitudes and influence of inertia forces, surface tension and viscous forces are analyzed in terms of Reynolds' number, capillary number and Weber number. The theoretical background is followed by a short explanation of the experimental set-up. Several experiments are shown in different mold configurations and at various mold filling speeds. The results showed that inertia greatly influences the mold filling pattern and capillary effects are a major cause of air bubble entrapment.

A new screw pump/extruder is analyzed with regard to the effect of the pump components on the pump output. The pump is constructed such that the barrel, screw core, and screw helix can all be rotated independently. Contrary to conventional wisdom, it is demonstrated that the major physical component of the pump which contributes to the output is the screw helix. A new theory is developed for pumping characteristics of this new pump and the theoretical analysis is compared with laboratory data. The theory compares quite well with the data.

Surface roughness and thickness uniformity are two of the most important parameters to control in the manufacture of polymeric films. In this paper, we present the results of a detailed experimental study on the effect of biaxial stretching variables on these parameters in polyethylene terephthalate and poly p-phenylene sulfide films. Decreasing processing temperature and increasing stretch ratio were found to improve surface smoothness and thickness uniformities of PET films. However, in the PPS films, biaxial stretching resulted in rougher surfaces. The main mechanism to improve these variables especially the thickness uniformity was found to be the strain hardening. In PET stretching this mechanism was present and in PPS it was found to be absent in the processing range investigated. This resulted in the improvement in surface roughness and thickness uniformities in the PET films with the increase of stretch ratio and with the decrease of stretching temperature. In PPS, increase of stretch ratios had an opposite effect on the surface roughness. This may be due to branching in the PPS.

In cast film extrusion, the polymer is extruded through a slit die, stretched in air and then cooled on a chill roll. Isothermal and non-isothermal models have been developed to predict the polymer melt behavior between the die exit and the chill roll. The mechanical resolution is based on the assumption of a uniform thickness in the film cross section. It takes gravity and inertia forces into account and is performed in two steps: the equilibrium equation for a small film element is first written in a curvilinear coordinate system and the stress tensor is then calculated in a local Cartesian system. The temperature field in the film thickness is determined using local heat transfer coefficients, whose calculation is based on a precise description of the radiation and convection phenomena. This local description requires the measurement of air temperature close to and far from the film. The coupling between mechanical and thermal equations is achieved by introducing the temperature dependence of the viscosity. An iterative scheme is used to adjust the stretching force, the length of the film and the contact point on the roll. The theoretical predictions (film profile, velocity, width, temperature) are compared with experimental measurements performed on a pilot line. The agreement is always good, except for the width, whose computed linear variation is in contradiction with the experimental observations.

The complex crystalline orientation character of injection molded high density polyethylene was examined. The principal axes and level of orientation along these axes were evaluated using wide angle x-ray diffraction pole figure analyses for various injection molding conditions. The principal axes of each layer were determined by means of ellipsoidal representations from wide angle x-ray diffraction pole figure data. It has been found that the principal axes of crystalline orientation do not exactly coincide with the flow direction. The crystalline orientation state for each layer was evaluated in terms of biaxial orientation factors with appropriate principal axes of orientation. The general behavior of crystalline orientation indicates b-axis orientation normal to the wall and perpendicular to the flow and width directions and c, a-axes biaxial orientation in the plane of the flow and width directions. A mechanism leading to this distribution of crystalline axes is discussed.

A polypropylene was filled with 0 to 30 wt.-% of flaky fillers such as glass flake, mica, or talc, which had similar particle sizes and different aspect ratios and crystallization temperatures. Flexural test specimens were injection-molded from the filled polypropylenes at cylinder temperatures of 200 to 320°C. Properties such as flexural modulus, flexural strength, heat distortion temperature, Izod impact strength, and mold shrinkage and higher-order structures such as crystallinity, β-crystal content, thickness of skin layer, a'-axis-oriented component fraction, and crystalline orientation fraction were measured and structure-property relationships were studied. A simulation of molecular orientation was carried out from a viewpoint of melt orientation at the gate and its relaxation in the mold cavity.