A micromechanical constitutive model based on crystal plasticity for thermo-mechanical cyclic deformation of NiTi shape memory alloys

Based on crystal plasticity, a new micromechanical constitutive model is constructed to describe the cyclic deformation of polycrystalline NiTi shape memory alloy presented under different thermo-mechanical cyclic loading conditions. At the scale of single crystal, the phase transformation and transformation-induced plasticity of the NiTi shape memory alloys are considered to be related with the 24 martensite variants and the friction systems at austenite-martensite interfaces, respectively. Three kinds of internal variables are included in the single crystal model, i.e., the reversible martensite volume fraction, the residual martensite volume fraction, and the friction slip at austenite-martensite interfaces. The Helmholtz free energy for the representative volume element of NiTi single crystal is constructed and the thermodynamics driving forces for internal variables are obtained by corresponding dissipation inequalities, respectively. An explicit scale-transition rule is adopted to extend the proposed single crystal model to the polycrystalline version. Also, the initial crystallographic texture is addressed in order to reflect the anisotropic phase transformation behavior of the NiTi shape memory alloys presented in the tension and compression cases. The proposed model is firstly verified by comparing the simulations with the corresponding uniaxial cyclic deformation experiments of polycrystalline NiTi shape memory alloys, and then is discussed by describing the multiaxial cyclic deformation of the polycrystalline NiTi shape memory alloy under the strain-controlled and stress-controlled cyclic loading conditions with different multiaxial loading paths and predicting the recovery of residual martensite phase during the sequential heating. Finally, some details about the cyclic deformation of polycrystalline NiTi shape memory alloy in intragranular scale are also addressed with the help of the proposed model. (C) 2013 Elsevier Ltd. All rights reserved.