Effective elastic properties of biocomposites using 3D computational homogenization and X-ray microcomputed tomography

Abstract

A 3D computational homogenization method based on X-ray microcomputed tomography (μCT) was proposed and implemented to investigate how the fiber weight fraction, orthotropy and orientation distribution affect the effective elastic properties of regenerated cellulose fiber-polylactic acid (PLA) biocomposites. Three-dimensional microstructures reconstructed by means of the X-ray μCT were used as the representative volume elements (RVEs) and incorporated into the finite element solver within the computational homogenization framework. The present method used Euclidean bipartite matching technique so as to eliminate the generation of artificial periodic boundaries and use the in-situ solution domains. In addition, a reconstruction algorithm enabled finding the volume and surface descriptions for each individual fiber in a semi-automatic manner, aiming at reducing the time and labor required for fiber labeling. A case study was presented, through which the method was compared and validated with the experimental investigations. The present study is thus believed to give a precise picture of microstructural heterogeneities for biocomposites of complex fiber networks and to provide an insight into the influences of the individual fibers and their networks on the effective elastic properties.