

Multi-scale Modelling of Void Growth and Coalescence within Heterogeneous Void Clusters
Abstract
The ductility of advanced high strength automotive alloys is limited by their complex microstructures that contain heterogeneous distributions of reinforcing particles. To describe the ductile fracture mechanism in these alloys, Butcher [1] proposed a multi-scale material model that accepts measured particle distributions to capture void nucleation and evolution from the individual particle-scale to the macro-scale. While this model is promising, it is limited by the use of nucleation and evolution rules that assume the voids and particles to be isolated, thereby neglecting the critical interactions within the void and particle clusters that ultimately govern fracture. In the present work, we propose an attractive solution to this issue by implementing a non-local averaging scheme into the model of Butcher [1] to incorporate a microstructure-based lengthscale into the evolution submodels. For each void, the nearest neighbour information is used to compute the average morphology in the region defined by the lengthscale. This information is used to evaluate the evolution models for isolated voids to approximate the interactions within the clusters.