Savitribai Phule Pune University |
Title | Machine learned potential energy surfaces for free sodium clusters with density functional accuracy: Applications to melting |
Author/s |
Balasaheb J. Nagare, Sajeev Chacko
Department of Physics, University of Mumbai, Kalina Campus, Santacruz (E), 400 098, Mumbai, India Dilip G. Kanhere Department of Scientific Computing, Modeling and Simulation, Savitribai Phule Pune University, Pune 411 007, India |
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, where N denotes the number of atoms in the cluster, using the density functional method as implemented in the VASP code. 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. 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. Thus, it turns out that the best ﬁt can be obtained by carefully selecting a small number of data points viz. 1,900 and 1,300 conﬁgurations for the two models M1 and M2, respectively. These two models have been deployed to compute the heat capacities of Na55. The heat capacities of Na147 and about 40 isomers for larger clusters of sizes N = 147, 200, 201, and 252, have been obtained using the ﬁnal model M2. There is an excellent agreement between the computed and experimentally measured melting temperatures. The geometries of these isomers have been further optimized by density functional theory. The mean absolute error with respect to DFT energies is found to be about 7 meV/atom or less. The errors in the lowest interatomic bond lengths are below 2 % in almost all the cases. |
Keywords | Machine learning, interatomic potentials, potential energy surfaces, density functional theory, free sodium clusters, melting |
Download | arXiv:2309.08937 |
Citing This Document | Balasaheb J. Nagare, Sajeev Chacko, and Dilip G. Kanhere , Machine learned potential energy surfaces for free sodium clusters with density functional accuracy: Applications to melting . Technical Report CMS-TR-20240401 of the Centre for Modeling and Simulation, Savitribai Phule Pune University, Pune 411007, India (2024); available at http://scms.unipune.ac.in/reports/. |
Notes, Published Reference, Etc. | Accepted for publication in Physica Scripta |
Contact | bjnagare AT physics.mu.ac.in (Balasaheb J. Nagare), sajeev.chacko AT physics.mu.ac.in (Sajeev Chacko), dgkanhere AT gmail.com (Dilip. G. Kanhere) |
Supplementary Material |