Machine-learning interatomic potential for radiation damage and defects in tungsten

Tungsten will be used as a plasma-facing material in fusion power reactors, where the absorption of high-energy neutrons leads to permanent damage in the crystal structure. A comprehensive understanding of the atom-level damage in tungsten has been limited by the slowness of quantum simulations and the insufficient accuracy of classical simulations. This study bridges the gap between the two by developing a machine-learning interatomic potential that allows the simulation of tungsten in extreme environments with quantum accuracy.