Empirical Force Field Development

While the results from first-principles calculations can directly impact materials design, these calculations also have an indirect, essential role as a foundational step in a multi-scale simulation strategy. Such multi-scale simulation strategies seek to address phenomena that require either a large number of atoms – for example, complex materials with small levels of dopants or defects – or relatively long time scales (nano to microseconds). Given the relatively large computational expense of first-principles methods, simulating such large systems (> 10,000 atoms) or probing long-time dynamics is not viable at the first principles level. A common practice is to use first-principles data on smaller systems (a ‘training set’) to develop an empirical force field (EFF). Within the CoMET program, EFF development runs through all materials research themes.