Balasaheb J. Nagare, Sajeev Chacko, Dilip. G. Kanhere
{"title":"Machine-Learned Potential Energy Surfaces for Free Sodium Clusters with Density Functional Accuracy: Applications to Melting","authors":"Balasaheb J. Nagare, Sajeev Chacko, Dilip. G. Kanhere","doi":"arxiv-2309.08937","DOIUrl":null,"url":null,"abstract":"Gaussian Process Regression-based Gaussian Approximation Potential has been\nused to develop machine-learned interatomic potentials having\ndensity-functional accuracy for free sodium clusters. The training data was\ngenerated from a large sample of over 100,000 data points computed for clusters\nin the size range of N = 40 - 200, using the density-functional method as\nimplemented in the VASP package. Two models have been developed, model M1 using\ndata for N=55 only, and model M2 using additional data from larger clusters.\nThe models are intended for computing thermodynamic properties using molecular\ndynamics. Hence, particular attention has been paid to improve the fitting of\nthe forces. Interestingly, it turns out that the best fit can be obtained by\ncarefully selecting a smaller number of data points viz. 1,900 and 1,300\nconfigurations, respectively, for the two models M1 and M2. Although it was\npossible to obtain a good fit using the data of Na55 only, additional data\npoints from larger clusters were needed to get better accuracies in energies\nand forces for larger sizes. Surprisingly, the model M1 could be significantly\nimproved by adding about 50 data points per cluster from the larger sizes. Both\nmodels have been deployed to compute the heat capacities of Na55 and Na147 and\nto obtain about 40 isomers for larger clusters of sizes N = 147, 200, 201, and\n252. There is an excellent agreement between the computed and experimentally\nmeasured melting temperatures. The geometries of these isomers when further\noptimized by DFT, the mean absolute error in the energies between DFT results\nand those of our models is about 7 meV/atom or less. The errors in the\ninteratomic bond lengths are estimated to be below 2% in almost all the cases.","PeriodicalId":501259,"journal":{"name":"arXiv - PHYS - Atomic and Molecular Clusters","volume":"19 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Atomic and Molecular Clusters","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2309.08937","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Gaussian Process Regression-based Gaussian Approximation Potential has been
used to develop machine-learned interatomic potentials having
density-functional accuracy for free sodium clusters. The training data was
generated from a large sample of over 100,000 data points computed for clusters
in the size range of N = 40 - 200, using the density-functional method as
implemented in the VASP package. Two models have been developed, model M1 using
data for N=55 only, and model M2 using additional data from larger clusters.
The models are intended for computing thermodynamic properties using molecular
dynamics. Hence, particular attention has been paid to improve the fitting of
the forces. Interestingly, it turns out that the best fit can be obtained by
carefully selecting a smaller number of data points viz. 1,900 and 1,300
configurations, respectively, for the two models M1 and M2. Although it was
possible to obtain a good fit using the data of Na55 only, additional data
points from larger clusters were needed to get better accuracies in energies
and forces for larger sizes. Surprisingly, the model M1 could be significantly
improved by adding about 50 data points per cluster from the larger sizes. Both
models have been deployed to compute the heat capacities of Na55 and Na147 and
to obtain about 40 isomers for larger clusters of sizes N = 147, 200, 201, and
252. There is an excellent agreement between the computed and experimentally
measured melting temperatures. The geometries of these isomers when further
optimized by DFT, the mean absolute error in the energies between DFT results
and those of our models is about 7 meV/atom or less. The errors in the
interatomic bond lengths are estimated to be below 2% in almost all the cases.