Several pressure-induced phase transformations have been predicted and observed in tin; its phase diagram reflects its special position in group IV of the periodic table of elements, where the lighter elements (C, Si, Ge) tend to form covalent bonds. The most stable phase at 0K corresponds to a diamond phase similar to those found in lighter elements. However, pressure and temperature transitions are observed, associated with a change in the nature of the interatomic bonds. The thermodynamic and mechanical properties of the different phases of tin, as well as the structural transitions, are fairly well known today, but are still difficult to reproduce using electronic structure calculations.

For classical Molecular Dynamics (MD) simulations, a number of semi-empirical potentials have been proposed in the literature, which can reproduce certain parts of the phase diagram or certain properties, but which are limited in their ability to predict certain properties, in particular the elastic constants. Recently, Machine Learning Interatomic Potentials (MLIPs) have been developed to improve the description of the properties of the different crystalline phases. However, these potentials, trained on crystalline phases at temperature and liquid configurations, do not take into account the specific distortions of the lattices encountered during deformation of the material (dislocation formation, maculation).