Analysis of process parameters related to the single-screw extrusion of recycled polypropylene blends by using design of experiments

Abstract

The process dynamics of single-screw extrusion on mixtures of polypropylene (PP) and recycled PP were studied using a statistical, design of experiments (DoE) approach. For a conventional screw design, the barrel temperature, screw speed and two vastly different melt viscosity polypropylene mixtures were selected as the independent factors, whilst melt pressure, mass output, screw torque and temperature rise at the die due to shear heating were the dependent responses. A central composite design (CCD) in the framework of response surface methodology (RSM) was constructed, and an analysis of variance (ANOVA) was carried out to determine the significance of the response surface models. The resulting statistical and response surface predictions have demonstrated that the low viscosity component concentration in the blend is a dominating factor on melt pressure and screw torque, apart from the expected effect of screw speed on output. Viscous heating is affected only by screw speed and recycled polypropylene concentration. Furthermore, the predictions have identified a wider process operating window with increased low-viscosity component concentration. The data confirm that statistical tools make quantitative predictions for the effects of experimental process variables, in accordance with the expected qualitative trends towards process optimisation, providing scope towards its application in scaled-up industrial processes.