Inverse engineering based analytical and numerical prediction on ductile fracture for AZ60M magnesium alloy under wide stress triaxiality

Abstract

The lightweight automotive and construction industries increasingly rely on magnesium (Mg) alloy structural components as they have the potential to enhance service performance and enable cost-effective manufacturing. However, the extrusion of Mg alloys presents significant challenges, primarily due to their limited formability and the strong crystallographic textures that develop during extrusion. This study investigates the influence of the fracture behavior of AZ60M Mg alloy bars under different stress states, employing several uncoupled damage criteria to enable reliable ductile failure predictions. Tensile and compression tests were conducted using AZ60M specimens of varying geometries and dimensions, revealing distinct deformation scenarios, stress–strain relationships, and fracture forms. Digital Image Correlation technology was employed to capture detailed deformation behavior under diverse stress triaxialities across the extrusion loading direction. Initially, finite element models were developed for each specimen and then optimized using an experiment-based inverse engineering approach, through which the fracture strains were derived by correlating simulation results with experimental data. Additionally, plastic deformation behavior was accurately simulated using the Swift-Voce hardening law. Finally, the Rice-Tracey, DF2014, and DF2016 fracture criteria were calibrated using the experimental data, with the DF2016 model showing superior predictive performance. These findings provide valuable insights for improving the formability and ductile failure prediction of Mg alloys, which is critical for their broader application in lightweight structural components.