Two-dimensional mesoscale simulations of quasielastic reloading and unloading in shock compressed aluminum

Two-dimensional (2D) mesoscale simulations of planar shock compression, followed by either reloading or unloading, are presented to examine and understand the quasielastic response observed experimentally in shocked polycrystalline aluminum. The simulations included a realistic representation of the grain ensembles in polycrystalline samples to identify heterogeneous deformation features deemed important to model the continuum measurements. The simulations were carried out using a 2D Lagrangian finite element code (ISP-TROTP) that incorporated elastic-plastic deformation in grain interiors and utilized a contact/cohesive methodology to analyze the response of finite strength grain boundaries. Local heterogeneous response due to mesoscale features was quantified by calculating appropriate material variables along in situ Lagrangian tracer lines and comparing the temporal variation of their mean values with results from 2D continuum simulations. A series of initial calculations ruled out effects due to finite element size and width of the representative volume element used in our simulations. Simulations using a variety of heterogeneities were performed to identify the heterogeneities that were most important for simulating the experimentally observed quasielastic response. These were inclusions, hardened grain boundaries, and microporosity. Mesoscale simulations incorporating these effects demonstrate that the shock-deformed state in polycrystalline aluminum is strongly heterogeneous with considerable variations in lateral stresses. The simulated velocity profiles for a representative reloading and unloading experimental configuration were found to agree well with experimental data, and suggest that hardened grain boundaries are the most likely source of mesoscale heterogeneities in shocked 6061-T6 aluminum. The calculated shear strength and shear stresses were also found to be in good agreement with the reported experimental values. (c) 2006 American Institute of Physics.