Volume 12, Issue 4

This article presents melt temperature measurements with an infrared thermometer during extrusion and investigates the influence by the surface emissivity and the rotating screw. This influence can be used for on line residence time measurements in the extruder line, to measure the number of revolutions of the screw, and to measure both the surface temperature and the bulk's temperature. The results show that colouring the PE with a high emisffisivity tracer allows us to measure the melt's surface temperature and minimise the influence of the rotating screw. Its influence is weaker in complete filled section as in partial filled sections of the extruder and the mean bulk's temperature is measured when the emissivity is lower than one. Furthermore it is shown that it is possible to measure the residence time distribution on line with an infrared thermometer, by taking advantage of its sensibility on a changing surface emissivity. Measurements with classical methods shows the same residence time distribution as obtained with the infrared spectroscopy. The advantage of this method is its wide range of applications, such like as the use in pressure zones, for a wide range of temperature and to measure on line the local residence time distribution all along the screw.

Modular self-wiping co-rotating twin screw extruders have been widely used in polymer blending and compounding over the past four decades since they were first commercialized by Werner and Pfleiderer [1]. Despite its importance in the polymer processing industry, studies of the initial mixing process in a modular co-rotating twin screw extruder of binary polymer blends have received little attention until the last three years [2 to 9]. Among the various factors controlling the mixing of thermoplastic blends, the viscosity characteristics are a major determinant the degree of deformation of a disperse phase domain depends greatly upon the viscosity ratio. When one component of the mixing system has a very low viscosity, and there is an extremely large viscosity ratio, it is well known to be very difficult to mix [10 to 12].

Of the different mixing modules contained within the closely intermeshing, co-rotating twin-screw extruder (CICo-TSE), the trilobal elements and mixing discs are of major importance to the mixing efficiency. These two types of mixing element impose contrasting mixing behaviour on the viscous melt and are thus used selectively to perform differing mixing tasks. Problems encountered when solving the flow problem within these two mixing element zones arise from both the complex 3D geometry and the time dependent flow boundaries as each respective element type rotates about its fixed axes. Following on from the 2D results reported, the computational fluid dynamics package, Polyflow, was employed to investigate the mixing mechanisms exhibited by both a set of three trilobe element pairs and by a pair of staggered mixing discs, as utilised within the Betol BTS40 CICo-TSE. The flow fields were simulated using the ideal rheological properties of polypropylene and characterised in terms of velocity vectors, shear stresses generated and a mixing parameter λ, which quantifies the elongational and rotational flow components. Each flow field exhibited different characteristics for the configurations investigated and it has been found that the average flow field parameter values for the trilobe elements change periodically, whereas minimal changes are observed as the mixing discs rotate.

A composite model simulation of solids conveying, melting and melt flow in a modular co-rotating twin screw extruder has been developed. This is based on combining existing melt conveying models with new melting and solids conveying models for the various elements of a modular twin screw extruder. Computations are made for axial fill factor, pressure, temperature, melting profiles and also the power consumption, torque, specific energy consumption and average residence times. The results are compared with experiment.

In this paper a model has been developed for the approximate evaluation of the temperature rise in the extrusion of filled plastic materials. The model developed in this paper considers the influence of thermal and rheological properties of extrudate, operating conditions of extrusion and geometry of screw channel on the temperature of extrudate. The optimal parameters of screw channel in terms of productivity and temperature build-up are discussed. The model was tested by experimental studies in which filled plastic material was produced using ground rubber tyre particles dispersed in LDPE. The agreement between the model prediction and the experimental data is discussed.

Internal mixers used in rubber processing and other industries are always partially filled. This results in the establishment of multiple random free surfaces in the flow field generated inside these mixers. Therefore successful mathematical modelling of internal mixing process depends on the development of efficient techniques for the reliable simulation of complex free surface flows. Various interface tracking and boundary capturing methods have been used in the past to model this kind of flow regimes. In particular, in recent years the well-known volume of fluid (VOF) method has been frequently used to model a variety of processes involving free surface and moving boundary flows. Both Eulerian and Lagrangian frameworks can be adopted in the VOF scheme to simulate free surface regimes. Under realistic flow conditions, however, the straightforward application of the technique in both frameworks may yield inaccurate results unless elaborate solution strategies are used to avoid errors. In many cases the use of such elaborate schemes requires excessive computational costs and effort or the solution scheme becomes complex and inflexible. In this paper we describe a relatively simple free surface tracking method based on the application of the VOF method in an Eulerian framework. In this scheme the flow field inside a partially filled internal mixer is treated as a two-phase system consisting of incompressible and compressible phases. The sections filled with the fluid which is being mixed are always regarded as an incompressible phase. The parts which are filled with air (or voids in some applications) form the second phase in the present two-phase flow analysis. The latter phase is treated as a compressible or an incompressible fluid (or pseudo fluid in the case of voids) depending on the value of the pressure calculated at each time step for the sections which contain it. We show that free surface flow of highly viscous fluids in partially filled internal mixers can be very successfully simulated by this method.

Isotactic polypropylene (PP) was reinforced with Vectra B950 (VB), a thermotropic liquid crystalline polymer (TLCP), in two novel fiber spinning processes which both employ two extruders to allow separate thermal histories to be applied to PP and VB prior to their combination. In one process, a single VB melt stream was surrounded by a sheath of PP, in a manner comparable to a wire coating process, just prior to melt spinning to generate fibers possessing a sheath-core structure. The other process involved the distribution of continuous VB melt streams within a separately plasticated PP melt to form a composite melt which was drawn to create matrix-fibril fibers consisting of a PP matrix reinforced with VB fibrils. Both processes allow polymers with non-overlapping processing temperatures to be combined and melt spun to form composite fibers. The effect of fiber draw ratio on the tensile mechanical properties was assessed for composite fibers reinforced with 50 wt.-% (39 vol.-%) VB. Numerical simulations of the non-isothermal fiber drawing processes and an investigation of fiber morphology were undertaken in order to explain the mechanical properties. The composite fibers were compression molded at a temperature 80°C below the melting temperature of VB to form unidirectionally reinforced plaques. For both fiber types, the plaques exhibited excellent retention of fiber properties, illustrating the potential for using the fibers to form reinforced composite parts.

In order to study the process of rectangular fiber melt spinning, experimental and theoretical investigations are conducted. Several different values of such process variables as die dimension, flow rate, take-up speed, quench air speed and temperature, and die temperature are chosen for manufacturing poly(ethylene terephthalate) fibers. Both 1- and 2-dimensional formulations are employed under the Newtonian fluid assumption. Two separate mathematical schemes are then combined via the Picard iteration to provide a convenient tool estimating the process with all variables taken into account and with as little computational effort as possible. From this experiment, it is found that the extrudate exhibits negative swelling in the fiber width direction, which results from the negative second normal stress direction in shear flow inside the die. Fiber cross-section takes elliptical shape at the point of maximum thickness swell and then gets flat as it approaches the take-up point. As to numerical simulation, all the quantities such as fiber dimension, velocity, temperature and tension except variables related to cross-sectional area are relatively well described by this simple theory. Neglect of material elasticity and over-simplification of temperature variation along fiber cross-section are main causes for the deviation of calculated values from experimental data.

The properties of polymer material can be shaped by crystallinity distribution during the film extrusion process. This paper addresses the implementation of design sensitivity analysis to determine the process inputs which can generate the desired crystallinity distribution. To achieve such a goal, the performance index is defined in terms of the difference between the desired and the actual crystallinity distributions. The sensitivity of the performance index to process inputs is determined in order to update these process inputs in every iteration of optimization. The design sensitivity analysis presented here allows a systematic movement through the search space to reach a sub-optimal set of process inputs. The simulation results show that the proposed methodology can determine the process inputs in order to generate the crystallinity distribution close to the desired one.

Multilayer feedforward neural networks with error backpropagation algorithm have been applied to analyze the drawing process of PET films. Six kinds of process conditions, four kinds of material structures and six kinds of material properties were used as inputs and outputs chosen to analyze the relationships between process conditions, material structures and properties. It was shown that the trained neural networks represent a small error between the target and the corresponding estimated values. It has been shown quantitatively that material studies play an intermediate role in the relationship between process conditions, material structures and properties. It was also shown that model-free simulator predictions by neural networks are more reliable than the corresponding conventional non-linear regression analysis in the complex polymer processing.

Cavity pressure plays an important role in determining the quality of injection molded articles. The non-linear and time-varying nature of the process causes difficulties in implementing efficient control systems, which usually employ linear, time-invariant design strategies. Conversely, advanced control methods, such as self-tuning control, do not provide physical information about the process. A physically-based adaptive controller has been developed for cavity pressure control during the filling phase. Different control algorithms, employing different tuning criteria, were implemented on a microcomputer to control the cavity pressure. Experimental results indicate that this control yields accurate results for the injection molding of polyethylene. The results also show that this method, using the IMC algorithm, is superior to the classical optimal controller tuning approach.

The problem of shrinkage in injection moulding has been studied by means of a new technique based on strain gages placed on mould surface before injection. The local shrinkage development from the moment it starts inside the mould to soon after ejection can be followed by means of this method. With reference to a simple rectangular cavity, the effects of different holding pressures, of geometrical constraints placed inside the mould, of position in the cavity and of total cavity length on local shrinkage of injected PS samples are analysed. Shrinkage inside the mould is, indeed, registered: it starts later if higher holding pressures are applied; furthermore, any factor which produces an enhancement of shrinkage evolution before complete solidification gives rise to a corresponding increase of final shrinkage.

Injection molded parts often show several types of surface defect. It has been hypothesized that wall slip is associated with some of these defects. Wall slip of molten plastics has been observed above a certain critical shear stress in rheological measurements. The objectives of the proposed research were to analyze the occurrence of flow marks during the injection molding of linear polyethylene and evaluate its possible relation to wall slip. Various variables were studied in terms of their influence on flow mark formation: mold thickness, mold temperature, melt temperature, gate, injection speed and Teflon TM and silicone oil coatings. It was found that injection speed is the controlling factor for the generation of flow marks during injection molding of the linear polyethylenes studied. Since one of the resin studied had shown no tendency to slip in capillary flow experiments, and since a Teflon TM coating on the mold walls did not affect the occurrence of flow marks, we conclude that there is no relationship between wall slip and the generation of flow marks. Microscopic observation of molded surfaces suggests instead that flow marks result from the filamentaton and stretching of semi-solidified material in the neighborhood of the three-phase contact line.