Moment tensor potentials as a promising tool to study diffusion processes

[Display omitted] •Moment Tensor Potentials (MTPs) are a robust model of interatomic interactions.•Active learning improves reliability of MTPs.•MTPs trained at a given temperature can also be used at a lower one.•MTPs actively trained on-the-fly accurately describe diffusion in Al, Mo, Si. A recent...

Full description

Saved in:
Bibliographic Details
Published in:Computational materials science 2019-06, Vol.164, p.46-56
Main Authors: Novoselov, I.I., Yanilkin, A.V., Shapeev, A.V., Podryabinkin, E.V.
Format: Article
Language:English
Subjects:
Active learning
Aluminum
Diffusion
Molybdenum
Moment tensor potentials
Silicon
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:[Display omitted] •Moment Tensor Potentials (MTPs) are a robust model of interatomic interactions.•Active learning improves reliability of MTPs.•MTPs trained at a given temperature can also be used at a lower one.•MTPs actively trained on-the-fly accurately describe diffusion in Al, Mo, Si. 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.
ISSN:0927-0256
1879-0801
DOI:10.1016/j.commatsci.2019.03.049