Moment tensor Potentials as a Promising Tool to Study Diffusion Processes

I. I. Novoselov; A. V. Yanilkin; A. V. Shapeev; E. V. Podryabinkin

Moment tensor Potentials as a Promising Tool to Study Diffusion Processes

Computational Physics 2018-12-11 v2

Authors: I. I. Novoselov , A. V. Yanilkin , A. V. Shapeev , E. V. Podryabinkin

Abstract

A recently proposed class of machine-learning interatomic potentials --- Moment tensor potentials (MTPs) --- is investigated in this work. MTPs are able to actively select configurations and parametrize the potential on-the-fly. It is shown that MTPs accurately reproduce energies, forces and stresses calculated ab initio. As a more comprehensive test, MTPs are employed to calculate vacancy diffusion rates in Al, Mo and Si. We demonstrate that the results are in a good agreement with ab initio data for the materials considered.

Keywords

machine learning for materials science molecular dynamics simulations

Cite

@article{arxiv.1812.02946,
  title  = {Moment tensor Potentials as a Promising Tool to Study Diffusion Processes},
  author = {I. I. Novoselov and A. V. Yanilkin and A. V. Shapeev and E. V. Podryabinkin},
  journal= {arXiv preprint arXiv:1812.02946},
  year   = {2018}
}

Comments

28 pages, 9 figures, 1 table, 42 references

Related papers

View all related →