Evaluation of particle and fibre degradation during processing of wood plastic composites (WPC) using dynamic image analysis

Abstract

This work was performed within the scope of the DFG Research Training Group 1703 “Resource Efficiency in Interorganizational Networks – Planning Methods to Utilize Renewable Resources”. The aim of this work was to investigate the influence of different processing parameters on the filler morphology of Wood Plastic Composites (WPC). The results have partially been published or are submitted for publication elsewhere (four publications in total) and are reproduced within individual chapters of this work. WPC are a group of materials that combine the properties of a wood filler material with those of a polymer matrix. A review of recent research was conducted to identify the potential of WPC to contribute to efficient resource utilisation. The results show that a variety of waste and by-products from wood and agricultural industry, e.g. offcuts, sawdust, residues from board manufacturing, pulping sludge, can be used for the production of WPC. Also recycled polymers and biopolymers can serve as raw materials. In defining the final WPC properties, the morphology of the filler material, i.e. fibres or particles, plays an important role. During processing the material is subject to high temperature and shear leading to degradation of the wood filler. The use of dynamic particle analysis for the characterisation of WPC filler material was introduced to verify its suitability for the analysis of filler degradation during processing. The polymer was dissolved and extracted from the compound and particle morphologies before and after processing were compared. The length-based size distribution proved to be most suitable for the analysis of processing effects since particles on both ends of the distribution are well represented. The effects of process parameters like filler content, feeding method, pre-heating of filler material, polymer viscosity, rotor/screw speed, feed rate and screw design on filler degradation were studied. Therefore, Norway spruce (Picea abies) wood particles were compounded with polypropylene (PP) either in an internal mixer or in a twin-screw extruder (TSE). To study the influence of polymer viscosity different grades of PP and high density polyethylene (HDPE) having different melt flow rates (MFR) were used. After compounding, overall particle size was reduced by > 97 %. For PP composites, particle degradation increased with increasing filler content in both internal mixing and extrusion, and for higher number of kneading elements in the extruder screw. For HDPE composites, the effect of filler content was only marginal. Feeding wood particles and polymer as a dry-blend resulted in smaller particles compared to feeding the wood into the polymer melt. Also pre-heating the particles to prevent melt freezing caused stronger degradation. Particle degradation was reduced when polymer matrices with high MFR were used. The effect of screw speed and feed rate varied with filler content and screw design. Since the compounding conditions on the laboratory scale are not comparable to industrial scale processes, industrial scale compounding conditions were mimicked on a laboratory scale TSE to study the effect of realistic processing. The degradation of radiata pine (Pinus radiata) wood fibres was compared to glass fibres since they are a standard filler in composite industry. Via dead-stop experiments and sampling along the screws the influence of screw design, screw speed, and feed rate was analysed. Process conditions related to low specific mechanical energy (SME) input and gentle screw design led to more gradual fibre length reduction along the screw. This effect was more distinct for initially longer glass fibres than for initially shorter wood fibres. Final fibre lengths at the end of the screws showed no dependence on process settings. A difference in final length due to screw design was more distinct for glass fibres than for wood fibres and was also represented in the composite properties. A more severe screw design resulted in lower composite strength for glass fibre compounds but not for wood fibre compounds.