Constructing machine learning potential for metal nanoparticles of varying sizes via basin-hoping Monte Carlo and active learning

Nanoparticles, distinguished by their unique chemical and physical properties, have emerged as focal points within the realm of materials science. Traditional theoretical approaches for atomic simulations mainly include empirical force field and ab initio simulations, with the former offering effici...

Full description

Saved in:
Bibliographic Details
Main Authors: Gong Fu-Qiang, Xiong Ke, Cheng Jun
Format: Article
Language:English
Published: Science Press 2024-03-01
Series:National Science Open
Subjects:
Online Access:https://www.sciengine.com/doi/10.1360/nso/20230088
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850236037080547328
author Gong Fu-Qiang
Xiong Ke
Cheng Jun
author_facet Gong Fu-Qiang
Xiong Ke
Cheng Jun
author_sort Gong Fu-Qiang
collection DOAJ
description Nanoparticles, distinguished by their unique chemical and physical properties, have emerged as focal points within the realm of materials science. Traditional theoretical approaches for atomic simulations mainly include empirical force field and ab initio simulations, with the former offering efficiency but limited reliability, and the latter providing accuracy but restricted to systems of relatively small sizes. Herein, we propose a systematic strategy and automated workflow designed for collecting a diverse types of atomic local environments within a training dataset. This includes small nanoclusters, nanoparticles, as well as surface and bulk systems with periodic boundary conditions. The objective is to construct a machine learning potential tailored for pure metal nanoparticle simulations of varying sizes. Through rigorous validation, we have shown that our trained machine learning potential is capable of effectively driving molecular dynamics simulations of nanoparticles across a wide temperature range, especially within the nanoscale regime. Remarkably, this is achieved while preserving the accuracy typically associated with ab initio methods.
format Article
id doaj-art-dba259b7efc24ed8b417413e6fbf6009
institution OA Journals
issn 2097-1168
language English
publishDate 2024-03-01
publisher Science Press
record_format Article
series National Science Open
spelling doaj-art-dba259b7efc24ed8b417413e6fbf60092025-08-20T02:02:04ZengScience PressNational Science Open2097-11682024-03-01310.1360/nso/20230088eb33e642Constructing machine learning potential for metal nanoparticles of varying sizes via basin-hoping Monte Carlo and active learningGong Fu-Qiang0Xiong Ke1Cheng Jun2["State Key Laboratory of Physical Chemistry of Solid Surface, iChEM, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China"]["State Key Laboratory of Physical Chemistry of Solid Surface, iChEM, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China"]["State Key Laboratory of Physical Chemistry of Solid Surface, iChEM, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China","Laboratory of AI for Electrochemistry (AI4EC), Tan Kah Kee Innovation Laboratory (IKKEM), Xiamen 361005, China","Institute of Artificial Intelligence, Xiamen University, Xiamen 361005, China"]Nanoparticles, distinguished by their unique chemical and physical properties, have emerged as focal points within the realm of materials science. Traditional theoretical approaches for atomic simulations mainly include empirical force field and ab initio simulations, with the former offering efficiency but limited reliability, and the latter providing accuracy but restricted to systems of relatively small sizes. Herein, we propose a systematic strategy and automated workflow designed for collecting a diverse types of atomic local environments within a training dataset. This includes small nanoclusters, nanoparticles, as well as surface and bulk systems with periodic boundary conditions. The objective is to construct a machine learning potential tailored for pure metal nanoparticle simulations of varying sizes. Through rigorous validation, we have shown that our trained machine learning potential is capable of effectively driving molecular dynamics simulations of nanoparticles across a wide temperature range, especially within the nanoscale regime. Remarkably, this is achieved while preserving the accuracy typically associated with ab initio methods.https://www.sciengine.com/doi/10.1360/nso/20230088condensed matter physicsnanoparticlesmachine learning potentialworkflow
spellingShingle Gong Fu-Qiang
Xiong Ke
Cheng Jun
Constructing machine learning potential for metal nanoparticles of varying sizes via basin-hoping Monte Carlo and active learning
National Science Open
condensed matter physics
nanoparticles
machine learning potential
workflow
title Constructing machine learning potential for metal nanoparticles of varying sizes via basin-hoping Monte Carlo and active learning
title_full Constructing machine learning potential for metal nanoparticles of varying sizes via basin-hoping Monte Carlo and active learning
title_fullStr Constructing machine learning potential for metal nanoparticles of varying sizes via basin-hoping Monte Carlo and active learning
title_full_unstemmed Constructing machine learning potential for metal nanoparticles of varying sizes via basin-hoping Monte Carlo and active learning
title_short Constructing machine learning potential for metal nanoparticles of varying sizes via basin-hoping Monte Carlo and active learning
title_sort constructing machine learning potential for metal nanoparticles of varying sizes via basin hoping monte carlo and active learning
topic condensed matter physics
nanoparticles
machine learning potential
workflow
url https://www.sciengine.com/doi/10.1360/nso/20230088
work_keys_str_mv AT gongfuqiang constructingmachinelearningpotentialformetalnanoparticlesofvaryingsizesviabasinhopingmontecarloandactivelearning
AT xiongke constructingmachinelearningpotentialformetalnanoparticlesofvaryingsizesviabasinhopingmontecarloandactivelearning
AT chengjun constructingmachinelearningpotentialformetalnanoparticlesofvaryingsizesviabasinhopingmontecarloandactivelearning