A criterion for ductile failure in age-hardenable aluminum alloys

Abstract

In the present work, a new constitutive model is proposed to predict failure in age-hardenable aluminum alloys. In this regard, the existing nano-void theory of ductile failure is extended to precipitation hardened aluminum alloys by incorporating the effects of dynamic microstructure evolution due to the formation of deformation and precipitation induced cellular dislocation substructure. As per the nano-void theory, nano-voids form at pre-existing nano-particles and grow by accumulation of point defects that are generated dynamically during deformation, ultimately leading to ductile failure by void coalescence. The newly proposed constitutive model accounts for the effects of precipitation induced dislocation substructure on point defect generation by incorporating a new evolution law for the effective obstacle strength associated with substructure evolution. The formation of deformation - precipitation induced dislocation substructure is confirmed using transmission electron microscopy (TEM). The proposed model is calibrated and successfully validated against experimental data by predicting failure strains for tensile loading along different in-plane directions for three different AA6xxx series age-hardened aluminum alloys with very different starting microstructures and processing histories. The error between predicted failure strains and corresponding experimental values is less than 7%. In addition, the versatility of the proposed constitutive model is demonstrated by coupling the failure criterion with stress-strain data generated through crystal plasticity simulations, to predict failure strain for arbitrary loading – stress triaxiality conditions. Such failure strain versus stress-triaxiality data is invaluable as it is routinely used as an input for modeling fracture within finite element (FE) based simulations. Furthermore, the approach developed herein, is suitable for use with computer aided engineering (CAE) tools to overcome current limitations in failure prediction for aluminum components.