Ductile fracture prediction of HPDC aluminum alloy based on a shear-modified GTN damage model

Abstract

In this paper, we investigate how the shear-modified Gurson-Tvergaard-Needleman (GTN) model can be used to reveal the effect of manufacturing-process-induced porosity on the scatter of ductile fracture properties of a high-pressure die-casting aluminum alloy. The GTN model exhibits great advantages in predicting the variation of failure strain/displacement by altering the initial void volume fraction (f0), compared with uncoupled ductile damage models. For a specific metallic material, one unique set of model parameters is usually determined from experiments, which leads to the fact that only a deterministic failure strain/displacement can be obtained under a fixed porosity. To overcome this shortcoming of the shear-modified GTN model, a novel parameter calibration scheme is proposed and validated in the present study. Following the physical background of the GTN model, the twelve material parameters of the shear-modified GTN model have been determined based on a combination of macroscopic mechanical tests covering a wide range of stress states and micromechanics-based unit cell simulations. Different from the existing calibration strategies, the simulation results of unit cells embedded with a spherical void have revealed a mutual dependence between the f0 and critical void volume fraction (fc). By conducting interrupted shear tests, the evolution of secondary nucleated void volume fraction was identified according to the statistical results of the fractured brittle silicon particles. The remaining parameters have been further iteratively determined. In the end, the effectiveness of the proposed parameter determination approach is confirmed based on the accurate prediction of stochastic ductile failure properties (both the global failure displacements and the local failure patterns) in different fracture tests. Given the proposed parameters identification strategy, the GTN material models have been enriched to predict the ductile failure behavior of metallic materials, which possess relatively high porosities and more pronounced scattering in failure properties.