Atomistic modelling of the orientation dependence of shock-induced phase transitions in tin

Tin is an excellent model material as it exhibits a variety of mechanisms that are relevant to high dynamic loading due to its low-symmetry crystal structure. However, shock-induced phase transformations are known to affect tin's material properties, making their precise characterisation particularly challenging. In particular, the β-γ phase transformation has been shown to occur down to low shock pressures, a phenomenon that is made harder to predict by tin's highly anisotropic crystal structure. Indeed, while shock-induced phase transitions have been experimentally explored, their dependence on crystal orientation remains poorly understood. To this end, molecular dynamics simulations of shock loading in single-crystal tin across multiple orientations are performed here to identify precisely the mechanical conditions under which these transformations occur. These simulations utilise a recent machine learning-based interatomic potential, trained on density functional theory data across a wide range of pressures, to resolve the mechanisms driving orientation-dependent phase transitions. In this work, we show the preliminary results of these efforts. In particular, we explore the local conditions governing these polymorphic phase transitions. Eventually, this approach will complement current state-of-the-art experimental investigations and allow for further upscaling at the continuum scales for the establishment of continuum models for real components.