Ab initio for Millions - The Power for Machine-learned Potentials
Since its initial development for hydrocarbons in 2001 [1], we have found that this concept is transferable to applications to elements all across the periodic table, including all first row elements, metals, ceramics and ionic materials [2]. For all these elements and associated materials we have demonstrated that ReaxFF can accurately reproduce quantum mechanics-based structures, reaction energies and reaction barriers, enabling the method to predict reaction kinetics in complicated, multi-material environments at a relatively modest computational expense. At this moment, over 1000 publications including ReaxFF development of applications have appeared in open literature and the ReaxFF code – as implemented in LAMMPS, ADF, or in standalone-format – has been distributed around the world.
