Machine learning potentials for complex aqueous systems made simple

TL;DR

This paper introduces a user-friendly machine learning framework for developing accurate potential energy models tailored to specific thermodynamic states of complex aqueous systems, enabling efficient large-scale simulations.

Contribution

The authors propose a simple, active learning-based method to quickly develop state-specific machine learning potentials for complex aqueous systems, reducing human effort and increasing applicability.

Findings

Validated models for diverse aqueous systems including ions, surfaces, and confined water.

Achieved high accuracy in reproducing ab initio structural and dynamical properties.

Demonstrated the approach on water on TiO2 surface, revealing detailed water behavior.

Abstract

Simulation techniques based on accurate and efficient representations of potential energy surfaces are urgently needed for the understanding of complex aqueous systems such as solid-liquid interfaces. Here, we present a machine learning framework that enables the efficient development and validation of models for complex aqueous systems. Instead of trying to deliver a globally-optimal machine learning potential, we propose to develop models applicable to specific thermodynamic state points in a simple and user-friendly process. After an initial ab initio simulation, a machine learning potential is constructed with minimum human effort through a data-driven active learning protocol. Such models can afterwards be applied in exhaustive simulations to provide reliable answers for the scientific question at hand. We showcase this methodology on a diverse set of aqueous systems with increasing degrees of complexity. The systems chosen here comprise bulk water with different ions in solution, water on a titanium dioxide surface, as well as water confined in nanotubes and between molybdenum disulfide sheets. Highlighting the accuracy of our approach with respect to the underlying ab initio reference, the resulting models are evaluated in detail with an automated validation protocol that includes structural and dynamical properties and the precision of the force prediction of the models. Finally, we demonstrate the capabilities of our approach for the description of water on the rutile titanium dioxide (110) surface to analyze the structure and mobility of water on this surface. Such machine learning models provide a straightforward and uncomplicated but accurate extension of simulation time and length scales for complex systems.