Abstract

Most engineering approaches to model the mechanical response of metals rely on correlations between phenomenological formulations and macroscopic experiments, which area costly and time-consuming. These approaches are reliable for predictions within testing conditions of the experiments used to calibrate the correlation. However, model fidelity can rapidly decreases when the applied loading conditions depart from those use to calibrate the correlations. To mitigate modelling uncertainty, the author presented a crystal plasticity framework [1] in which virtually all parameters can be quantified independently at micro- and mesoscales. The key value added by this approach relies on the low dependence on loading conditions of small scale damage mechanisms. Although the parameterization of small scale mechanisms is, in principle, independent from macroscopic experiments, the uncertainty of low scale mechanisms is still large and propagates to macroscopic predictions. This presentation will discuss the role of physics-based parameters on modelling the cyclic response of FCC metallic materials at low to medium temperatures. We will demonstrate approaches to parameterize atomic scale unit processes and mesoscale mechanisms independently. We will explore modelling Cu and Ni single crystals under cyclic loading at various temperatures and we will discuss the modelling of hydrogen-charged Ni, for which the dearth of experimental data demands confident physics-based models.

[1]    Castelluccio and McDowell, Int. J. Plast., 2017, 28, 1.