Modelling of Dynamic Behaviour of Orthotropic Metals Including Damage and Failure

A physically based material model for metals, with elastic plastic and damage/failure orthotropy is proposed in this paper. The model is defined within the frameworks of irreversible thermodynamics and configurational continuum mechanics and integrated in the isoclinic configuration. The use of the multiplicative decomposition of deformation gradient makes the model applicable to arbitrary plastic and damage deformations. To account for the physical mechanisms of failure, the concept of thermally activated damage initially proposed by Klepaczko (Klepaczko, 1990) was adopted as the basis for the new damage evolution model. This makes the proposed damage/failure model compatible with the Mechanical Threshold Strength (MTS) model (Follansbee and Kocks, 1988; Chen and Gray, 1996; Goto et al., 2000; Gray et al., 1999; Chen et al., 1998) which was used to control evolution of flow stress during plastic deformation. In addition the constitutive model is coupled with a shock equation of state which allows for modelling of shock wave propagation in the material. The new model was implemented in DYNA3D and our in-house non-linear transient SPH code, MCM (Meshless Continuum Mechanics). Parameters for the new constitutive model for AA7010 (a polycrystalline aluminium alloy, whose orthotropy is a consequence of grain morphology), were derived on the basis of the tensile tests and Taylor anvil tests. The tensile tests were performed for the range of temperatures between 343.15K and 413.15K, and strain rates between and . The new model was validated in two stages. The first stage comprised a series of single element tests design to separately validate elasticity, plasticity and damage related parts of the model. The second stage comprised a series of numerical simulations of Taylor anvil and plate impact tests for AA7010 and comparison of the numerical results with the experimental data. The numerical results illustrate the ability of the new model to predict experimentally observed behaviour.