Application of Statistically Similar Representative Volume Elements of Metal Matrix Composites for the Efficient Simulation of Crack Propagation on the Microscale

Abstract

AbstractThe efficiency of various drilling and cutting processes is strongly determined by the wear resistance of the applied tools. For example in mechanized tunnel drilling, metal matrix composites, used as reinforcements on the chisels, are primarily exposed to surface spalling. This wear mechanism is governed by subcritical crack propagation through the material's microstructure, which consists of brittle carbide inclusions surrounded by a ductile matrix. The microstructure morphology strongly influences the crack propagation and thus, the resistance against wear. In order to improve the material's microstructure regarding wear, numerical simulations on the microscale are an important tool to gain knowledge about the influence of the morphology on the crack propagation. The investigated microstructure is given as voxel data obtained from a µCT scan. Because of the high complexity of the microstructure, simulations of crack propagation through it are computationally costly, in particular under cyclic loads. Hence, simplified artificial microstructures are constructed which resemble the morphological as well as mechanical properties of the full microstructure. Here, these less complex microstructures are constructed according to the method in [1,2] for the generation of so‐called Statistically Similar Representative Volume Elements (SSRVE). For the efficient simulation of crack propagation through heterogeneous microstructures based on voxel data, the framework in [10] is applied on the SSRVEs. In numerical simulations, the capability of the SSRVEs to represent the crack behavior of the full microstructure is investigated by comparing with the results are compared to simulations on a cutout of the full microstructure.