Large Scale FEM of the effective elastic properties of particle reinforced composites

Over the years several methods have been proposed for the determination of the effective elastic properties of particle reinforced composites. The material microstructures used in the present analysis is a real microstructure and a numerically generated microstructure. X-ray microtomography is used to determine the material microstructure and with this method the interior microstructure is determined in a non-destructive way. Using the commercially available equipment, SkyScan 1072, the maximum resolution is approximately 5 microns. The data obtained from the tomographic examination is used to generate three-dimensional finite element models of the microstructure. The models contain a large number of elements, up to 1 million, and are solved iteratively using an element-by-element formalism. Models containing 100 particles have been statistically generated and the material properties of each particle is assigned using a Gaussian distribution of the properties. Various distributions have been used to determine how the variation of particle properties influences the effective elastic properties. From this examination it was found that the larger the variation of the particle properties is the softer the composite material gets.Over the years several methods have been proposed for the determination of the effective elastic properties of particle reinforced composites. The material microstructures used in the present analysis is a real microstructure and a numerically generated microstructure. X-ray microtomography is used to determine the material microstructure and with this method the interior microstructure is determined in a non-destructive way. Using the commercially available equipment, SkyScan 1072, the maximum resolution is approximately 5 microns. The data obtained from the tomographic examination is used to generate three-dimensional finite element models of the microstructure. The models contain a large number of elements, up to 1 million, and are solved iteratively using an element-by-element formalism. Models containing 100 particles have been statistically generated and the material properties of each particle is assigned using a Gaussian distribution of the properties. Various distributions have been used to determine how the variation of particle properties influences the effective elastic properties. From this examination it was found that the larger the variation of the particle properties is the softer the composite material gets