On effective thermal properties of wood particles reinforced HDPE composites

Plant fiber biocomposites have the potential of being thermal insulating materials with strong mechanical resistance. However, they are vulnerable to defects due to the incompatibility of the materials. In that prospect, this study focuses on the determination of thermal properties of wood particles reinforced HDPE composites (WPC) using experimental, theoretical, and computational modeling approaches. From the laser flash analysis in the experimental setup, the results show that the addition of 30 and 60 \(\text{wt\%}\) of wood particles leads to a reduction of 20 and 44% in thermal conductivity, respectively, as compared to the neat HDPE. This demonstrates the thermal insulation capabilities of the wood particles. The use of several theoretical models to determine the WPC’s effective thermal conductivity shows good consistency with the experimental results, with a maximum discrepancy error of 4%. However, a noticeable divergence is observed as the wood content increases. Further study on the effective properties is carried out using the pixel-based computational homogenization analysis. This computational method carries out finite element calculations on representative 2D images of the WPC. The computational results are also consistent with the experimental results, but similar divergence like the theoretical models is still observed. The divergence is suspected to originate from the presence of porosity in the experimental specimens. The percentage of porosity is estimated using theoretical models and is integrated into the computational homogenization technique by introducing another phase in a different color code. This leads to the convergence of computational and experimental results. The pixel-based homogenization method has shown great flexibility and is able to integrate defects like porosity to yield precise results of the effective thermal conductivity of WPC which can then be used in structural engineering analysis.