Swagata Roy¹, Johannes Dürholt², Thomas Asche², Federico Zipoli³, Rafael Gómez-Bombarelli^1*

1 Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA

2 Evonik Operations GmbH, Marl, Germany

3 IBM Research, Zurich, Switzerland

* Corresponding authors emails: rafagb@mit.edu

DOI10.24435/materialscloud:61-x8 [version v1]

Publication date: Jul 06, 2023

How to cite this record

Swagata Roy, Johannes Dürholt, Thomas Asche, Federico Zipoli, Rafael Gómez-Bombarelli, Learning a reactive potential for silica-water through uncertainty attribution, Materials Cloud Archive 2023.106 (2023), doi: 10.24435/materialscloud:61-x8.

Description

The reactivity of silicates in aqueous solution is relevant to various chemistries ranging from silicate minerals in geology, to the C-S-H phase in cement, nanoporous zeolite catalysts, or highly porous precipitated silica. While simulations of chemical reactions can provide insight at the molecular level, balancing accuracy and scale in reactive simulations in the condensed phase is a challenge. Here, we demonstrate how a machine-learning reactive interatomic potential can accurately capture silicate-water reactivity. The model was trained on a new dataset comprising 400,000 energies and forces of molecular clusters at the 𝜔-B97XD\def2-TVZP level. To ensure the robustness of the model, we introduce a new and general active learning strategy based on the attribution of the model uncertainty, that automatically isolates uncertain regions of bulk simulations to be calculated as small-sized clusters. Our trained potential is found to reproduce static and dynamic properties of liquid water and solid crystalline silicates, despite having been trained exclusively on cluster data. Furthermore, we utilize enhanced sampling simulations to recover the self-ionization reactivity of water accurately, and the acidity of silicate oligomers, and lastly study the silicate dimerization reaction in a water solution at neutral conditions and find that the reaction occurs through a flanking mechanism.

Materials Cloud sections using this data

Files

File name	Size	Description
silica_Painn_model.zip MD5md5:54896b829ec630425f13e43dfa57049a	451.9 MiB	Neural network interatomic potential (PaiNN arXiv:2102.03150 ) for silica-water. Predicts energy and forces with units kcal/mol and Kcal/mol-Å respectively. Code for training and running MD simulations are found at https://github.com/learningmatter-mit/NeuralForceField
README.md MD5md5:0170f0f718b1cb66f52236a07bb44991	1.6 KiB	README

License

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.

External references

Keywords

free energy neural network potential chemical reactions silicates Active learning