Process analysis using multivariate regression models exemplified by WPC processing with a single-screw extruder

In the polymer industry, single-screw extruders are important extruding pure plastics; however, they are rarely used for the production of wood-plastic composites (WPC). The objective of this study is to enhance the understanding of WPC extrusion using a smooth bore single-screw extruder by determining the impact on product quality through process parameters and raw material properties using multivariate regression models. A general factorial design was used to examine the influence of variation of raw material moisture content and screw speed on processibility. To evaluate the performance of the single-screw extrusion in the production of WPC profiles, a reference was produced by use of the conical, co-rotating twin-screw extruder. The feasibility of PVC-based WPC production on single-screw extruder was demonstrated. However, only the profiles produced from absolutely dry raw material and with a low screw speed were of superior quality, i.e. showing higher density and mechanical strength properties.