Recommended by

Indexed by

Solvation free energies from machine learning molecular dynamics

Nicephore Bonnet1*, Nicola Marzari1*

1 Theory and Simulation of Materials (THEOS), École Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland

* Corresponding authors emails: nicephore.bonnet@epfl.ch, nicola.marzari@epfl.ch
DOI10.24435/materialscloud:a0-jh [version v1]

Publication date: May 27, 2024

How to cite this record

Nicephore Bonnet, Nicola Marzari, Solvation free energies from machine learning molecular dynamics, Materials Cloud Archive 2024.80 (2024), https://doi.org/10.24435/materialscloud:a0-jh


In this paper, we propose an extension to the approach of [Xi, C; et al. J. Chem. Theory Comput. 2022, 18, 6878] to calculate ion solvation free energies from first-principles (FP) molecular dynamics (MD) simulations of a hybrid solvation model. The approach is first re-expressed within the quasi-chemical theory of solvation. Then, to allow for longer simulation times than the original first-principles molecular dynamics approach and thus improve the convergence of statistical averages at a fraction of the original computational cost, a machine-learned (ML) energy function is trained on FP energies and forces and used in the MD simulations. The ML workflow and MD simulation times (≈200 ps) are adjusted to converge the predicted solvation energies within a chemical accuracy of 0.04 eV. The extension is successfully benchmarked on the same set of alkaline and alkaline-earth ions. The record includes all molecular-dynamics trajectories, energies and forces used to obtain the solvation energies of alkaline and alkaline-earth ions in water, as reported in Table 2 of referenced paper.

Materials Cloud sections using this data

No Explore or Discover sections associated with this archive record.


File name Size Description
2.5 GiB For each system (water cluster, Li, Na, ...), the LAMMPS MD trajectory, energies and forces are provided. See detailed description in README.txt.
384 Bytes README file.


Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.


solvation machine learning molecular dynamics

Version history:

2024.80 (version v1) [This version] May 27, 2024 DOI10.24435/materialscloud:a0-jh