SC Seminar Bastiaan Braams (MD)

Force fields for fusion materials and plasma material interaction

I will present a perspective on the status and needs of computational tools for plasma-material interaction and primary radiation damage, using as motivation work done in a recent International Atomic Energy Agency (IAEA) Coordinated Research Project on Plasma-Wall Interaction with Irradiated Tungsten and Tungsten Alloys in Fusion Devices. In my perspective the interaction potential for molecular dynamics simulations is (almost) everything and present interaction potentials used in modelling leave much to be desired. There are very interesting ongoing developments in machine learning for interaction potentials and I will review some approaches: the Gaussian Approximation Potential (GAP) approach led by Gábor Csányi of Cambridge University, the Deep Potential Molecular Dynamics (DeepMD) approach led by Weinan E en Roberto Car of Princeton University with contributions from IAPCM (Beijing), and the Solid Harmonic Wavelet Scattering approach led by Stéphane Mallat of ENS in Paris. New approaches such as these will be important to provide computational support for studies of primary radiation damage and they will be essential in connection with experiments that rely on surrogate (charged particle) irradiation or on irradiation by neutrons from fission, spallation or a stripping reaction.