Prediction of mechanical properties of polyethylene mouldings based on laminate theory and thermomechanical indices

The properties of moulded plastic products are dependent on the processing technology used in their manufacture and in particular on the structural mo rphology resulting from the thermomechanical environment imposed on the mel t. This paper presents a unified approach to describe the behaviour of the products based on knowledge of the thermomechanical conditions imposed duri ng processing. A linear medium density polyethylene has been processed usin g rotational moulding, compression moulding, and injection moulding in orde r to achieve different thermomechanical conditions (i.e. shear races and co oling rates). The processing conditions used were typical of those in commo n use in the respective industries. The moulding parts were mechanically te sted to determine the tensile, flexural, and impact properties. These measu rements were performed both on samples corresponding to the entire thicknes s of the moulding and on slices taken from across the section of the mouldi ngs. On the basis of these measurements, two models were developed. One is based on laminate theory, in which, from a knowledge of the mechanical prop erties of the individual layers through the wall thickness, it is possible to predict the tensile and flexural properties of the full thickness mouldi ng. The other is an empirical model that predicts the tensile modulus of a plastic part as a function of two thermomechanical indices. It is shown tha t the type of dependence of the mechanical performance on the thermomechani cal conditions imposed during processing is similar for the three moulding techniques used. A good agreement is achieved between the experimental data and those predicted by the thermomechanical model. It is also shown that v ia the combined use of the thermomechanical indices concept and the laminat e analysis, good predictions of the mechanical behaviour of plastic mouldin gs with complex microstructures can be achieved. It is proposed that this a pproach could provide a very valuable addition to existing melt how simulat ion packages. This would enable not only processing conditions to be optimi sed but the properties of the end product could be predicted.