Abstract
We recently demonstrated that pseudoelastic NiTi tubes compressed axially buckle first into an axisymmetric wrinkling mode but subsequently collapse by progressive development of buckle lobes with three circumferential waves. This behavior is governed by the material nonlinearity associated with the austenite to martensite solid-state transformation. On unloading, the material transforms back to austenite, deformation is recovered, and the buckling lobes are erased (Jiang et al., 2016). The problem was simulated numerically using a constitutive model framework for the pseudoelastic behavior of NiTi based on a J2-type nonlinear kinematic hardening model with the back stress represented through a potential. The first generation of the model did not account for the significant tension/compression asymmetry exhibited by the material so the buckling response did not reproduce well the experiments underestimating the dissipated energy. In the present work the constitutive model is extended to include the tension/compression asymmetry. It is introduced through a weighted mix of two potentials that are calibrated to the tensile and compressive stress-strain responses of the material. In addition, the present model allows for plastic deformations at the completion of martensite transformation. Incorporation of the tension/compression asymmetry significantly improves the calculated response of the buckling and recovery of NiTi tubes both qualitatively and quantitatively. The addition of plastic deformations to the model further improves the predictions. The initial axisymmetric wrinkling and the switch to the buckling mode with three circumferential waves are now reproduced correctly, and so are the evolution of localized collapse and the recovery of deformation on unloading. Furthermore, the stress levels during collapse and recovery follow the experimental levels quite closely thus reproducing the energy dissipated.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have