Band 15, Heft 1

Starting from our previous results concerning the melt flow during the dispersive mixing zones of Maddock and Egan type, one establishes relations for the pressure variation, appropriate for incremental calculation. Because generally the melt flow along the mixing zone is non-isothermal, one establishes relations for the temperature variation along the mixing zone, namely: temperature variation caused by external heating, by internal friction, by external friction and due to melt compressibility. By using the established relations one develops a network procedure for pressure and temperature variation calculation along the mixing zone. The computer program based on the network procedure was used to calculate a usual mixing zone for a 63.5 mm extruder-screw. The program gives the possibility to represent the pressure and temperature variation between the inlet and the outlet of the mixing zone. This is useful when we are interested in the optimum design of such mixing zones.

The rhomboidal mixing section is becoming very popular among processors to provide distributive mixing. Currently, several different designs are used but the details of the flow behavior and mixing efficiency is not well understood. This information is needed to be able to design and find the most efficient rhomboid geometry. In this investigation nine different geometries with various pitches (helix of rhomboids) were analyzed using a 3-dimensional boundary element method (BEM). The geometries were compared according to mixing efficiency, pressure and energy consumption. The results were compared to experiments performed with a conventional single screw extruder that was fitted with three different rhomboidal mixing sections. The investigation led to the conclusion that the most effective distributive mixing sections were those with neutral rhomboids (pine-apple mixer). However, the neutral rhomboidal mixing section consumes the most pressure in the extruder. It was also concluded that rhomboidal mixing sections deform the material by shear, making them poor dispersive mixing sections.

The ∊-caprolactone bulk polymerisation, initiated by tetra-propoxy titanium, was investigated in a co-rotating twin screw extruder. On the basis of pressure measurements, the process dynamic modelling was achieved. The pressure drop at the exit die of the extruder was correlated to the rheological behaviour of the melt polycaprolactone owing to equations which characterise the flow inside the die. The residence time distribution modelling was determined with an in-line Ultra-Violet fluorescence method. Thus, the combination of both modellings leads to the process dynamic modelling. This work constitutes the basis for an approach of the process regulation by the way of an internal model control. Simulation has given good results. However, the large response time to the control, relative to the process dynamic, constitutes effectiveness limiting factor for experimental control of the process.

An objective function is proposed for use in optimisation of profile die design, and defined in terms of mass flow rates to topological partitions of the profile cross section. In a series of uPVC extrusion trials its dependence on land channel height is investigated for two profiles to assess its suitability for gradient based optimisation of flow distribution. Preliminary finite element flow simulations are carried out, and comparisons of computed and experimental objective function values show that the mathematical modelling should be extended to include extensional effects and draw down at the die exit. Examination of the experimental profiles indicates the limitations of die optimisation in determining the detailed geometry of the finished profile cross sections.

Numerical simulation of turbulent air flow tangentially impinging on a blown film bubble has been carried out. The cooling air is assumed to originate from a dual orifice adjustable air-ring. The stream line pattern and heat flux are determined through a finite volume numerical technique using a version of the so called k-∊ modeling. It is shown that as the air flows through the air ring passage relatively large under-pressure is created (Venturi effect), which applies a suction to the bubble. Depending on the flow rates and geometrical configurations the air jet may attach itself (Coanda effect) to either the air ring surface and/or to the bubble. The Coanda effect has a significant influence on local heat flux. The numerical results suggest that the Venturi and Coanda effects influence significantly process stability, production rate and final film properties.

The non isothermal crystallization behaviour of two model blends of immiscible polymers (an isotactic polypropylene and a polyamide-6) was examined by a quenching procedure within a range of cooling rates from 0.1 to above 1000°C/s in order to emulate the solidification conditions arising during polymer processing. The final structure of the blends was analyzed by density and WAXD and the transition from stable to metastable phases in the blends and the pure homopolymers was compared examining their dependence on cooling rate. This shows that crystallization in the blends is always faster since the transitions move to larger cooling rates. Observation of morphological details suggests that this can be due to enhanced nucleation at the interface. The density variation in the case of the polyamide-6 rich blend is larger than expected from an additive volume contribution which may be ascribed to development of microvoids at the interface. In the case of the isotactic polypropylene rich blend the additive volume contribution is consistent with experimental observations.

High-strength, high-modulus products with a characteristic phase morphology were prepared via the injection molding of blends of isotactic polypropylene (PP) and polyamide-12 (PA-12) under special processing conditions (low melt temperature, high holding pressure, and a convergent sprue design). These conditions generated a situation in which the dispersed phase can be distributed in the self-reinforced matrix as oriented fibrils. Optical (polarized light optical and phase contrast) microscopic and SEM investigations revealed a uniform distribution of the dispersed phase within the matrix. These products can be regarded as in situ thermoplastic composites. The strength level of the self-reinforced blends is approximately twice as high as that of the products processed under conventional conditions. The PA-12 phase dispersed as ultrathin fibrils enhances the flexural impact strength of the blends, decreases their notch sensitivity, and alters its fracture mechanism.

The compressive squeeze flow is encountered in various processes like compression molding and in rheometry. It involves a complex deformation field driven by both shear and normal stresses. Here the flow and deformation field generated during the compressive squeeze flow of generalized Newtonian fluids, subject to wall slip, was analyzed using both numerical simulation based on the 2-D Finite Element Method and the 1-D analytical model, which is based on the lubrication assumption. The agreement between the analytical and numerical solutions deteriorates with increasing Navier's wall slip coefficient and the increasing yield stress value of the viscoplastic fluid. The lubrication assumption is valid under various squeeze flow conditions but numerical analysis in conjunction with experimental studies are necessary to determine the ranges of material parameters and operating conditions for which the lubrication assumption is valid.

Injection molding of the isotactic polypropylenes polymerized using metallocene and Ziegler-Natta catalysts and having the same molecular weights was performed. Rheological properties and shear-induced crystallization kinetics were measured and their comparative analysis is made. These data are fitted to the equations necessary for simulation of the injection molding process. Numerous injection molding runs were carried out and pressure measurements during molding were performed. The gapwise distribution of crystallinity, spherulite size and thickness of flow-induced crystallized layer in moldings were measured. The effect of various processing conditions on these values was determined. The measured values were compared with the results obtained from the simulation of injection molding process using unified theory of crystallization proposed earlier. Comparative analysis is made between flow conditions and crystallization behavior of these polymers and their effect on the pressure development and structural characteristics of the moldings.

The pressure-volume-temperature (PVT) melt behavior of 12 polyethylene resins was evaluated at pressures up to 200 MPa, using both isothermal and isobaric measurements in a GNOMIX high pressure dilatometer. The resins included high density polyethylene (HDPE), low density polyethylene (LDPE) and linear low density polyethylene (LLDPE). They were produced using a variety of catalysts, including Ziegler-Natta (ZN) and metallocene catalysts. The PVT data were used to evaluate two empirical equations of state (the Tate and Inverse Volume equations) in predicting PVT behavior of the melt, the isothermal compressibility, and the thermal expansion coefficient. The dependence of the melting and crystallization temperatures on pressure was also evaluated, and compared to existing equations. It was not possible to identify structural effects on the PVT melt behavior, as well as the isothermal compressibility and thermal expansion coefficient. However, some slight dependence on density was observed for the parameters of the equations of state. The isobaric experiments revealed that the pressure effect on the melting and crystallization temperatures was similar for all resins. While the melting and crystallization temperatures varied widely for the various resins, the pressure coefficients of the melting and crystallization temperatures were approximately equal (0.25°C/MPa.) The main differences in the behavior of the resins were in the transitional region during melting and crystallization.

It is well known that the thermal behaviour of moulds used for thermoplastics processing has an important effect on both the part quality and the process productivity. The present work reports a study on the evaluation of the relative effect of different geometric and operational parameters on the thermal behaviour of injection moulds. It uses a mathematical model developed to simulate the thermal transient exchanged during the entire moulding cycle. The model is based on a 2D Fourier equation, uses an adaptive grid and describes the thermal contact resistance at mould/polymer interface. The cooling time and the interface temperature distribution are calculated from the computed temperature field in the domain under study. Experimental data, obtained with an instrumented tool during typical production conditions, is used to validate the developed model. The evaluation analysis is based on the Taguchi methodology used to rank the relative effect of the parameters on the selected variables (cooling time, temperature distribution and number of cycles to reach the thermal steady state).

Polymer solidification under processing conditions is a complex phenomenon in which the kinetics of flow, high thermal gradients and high pressures determine the product morphology. The study of polymer structure formed under pressure has been mainly made using conventional techniques such as dilatometry and differential scanning calorimetry under isothermal conditions or non isothermal conditions but at cooling rates several orders of magnitude lower than those experienced in industrial processes. A new equipment has been recently developed and improved to study the crystallization of polypropylene when subjected to pressure and cooled rapidly. An experimental apparatus essentially constituted of a special injection mould has been employed. Polymer samples can be cooled at a known cooling rate and under a known pressure. Micro Hardness (MH), Wide angle x-ray diffraction (WAXD), Polarised Optical Microscopy (POM) and density measurements are then used to characterize the sample morphology. The results of rapid cooling experiments under pressure on an iPP sample display a lower density and a lower density dependence on cooling rate for increasing pressure. Micro hardness confirms the trend. A deconvolution technique of WAXD patterns is used to evaluate the final phase content of samples and to assess a crystallization kinetics behaviour. Phase distribution results indicate that the decrease of alpha phase with pressure is balanced by an increase of the mesomorphic phase leaving unaffected the amorphous phase content. This peculiar behaviour can be easily related to a negative influence of pressure on the kinetics of the crystallization of alpha phase.