Machine Learning Assisted Simulations of Electrochemical Interfaces: Recent Progress and Challenges

Abstract

The electrochemical interface, where the adsorption of reactants and electrocatalytic reactions take place, has long been a focus of attention. Some of the important processes on it tend to possess relatively slow kinetic characteristics, which are usually beyond the scope of ab initio molecular dynamics. The newly emerging technique, machine learning methods, provides an alternative approach to achieve thousands of atoms and nanosecond time scale while ensuring precision and efficiency. In this Perspective, we summarize in detail the recent progress and achievements made by the introduction of machine learning to simulate electrochemical interfaces, and focus on the limitations of current machine learning models, such as accurate descriptions of long-range electrostatic interactions and the kinetics of the electrochemical reactions occurring at the interface. Finally, we further point out the future directions for machine learning to expand in the field of electrochemical interfaces.

Electric Double Layer Effects in Electrocatalysis: Insights from Ab Initio Simulation and Hierarchical Continuum Modeling

“…Such machine-learning potential-based simulation methods are especially important to enable the investigation of EDL effects in more complex reactions, such as the electrocatalytic coupling system and the electrocatalytic transformation of biomass/organic molecules. Recently, several advances have been made. − Third, the ab initio simulation results of dynamic interfaces are mainly verified by the comparison between computational vibration spectroscopy and experimental vibration spectroscopy at present. However, the combination between them is largely qualitative, and the computational spectral peaks tend to have larger widths. , This may be due to the differential spectrum processing of experimental spectroscopy, the rationality of the simulated interface model and reaction condition settings, or the effectiveness of current vibrational spectroscopy calculation methods.…”

Corrosion of Electrochemical Energy Materials: Stability Analyses Beyond Pourbaix Diagrams

“…Another approach to overcoming the sampling limitations of DFT is to combine it with machine learning (ML) approaches. Such ML-assisted first-principles simulations may help elucidate the atomistic details of dynamically evolving electrochemical interfaces under reaction conditions. − For example, in a recent work combining microscopy, spectroscopy and accelerated ML molecular dynamics a detailed atomistic picture of dynamic surface restructuring of Pd deposited on Ag was revealed . It was demonstrated in simulations and confirmed by surface science experiments that in contrast to the thermodynamically favored states, where Pd is dispersed in the Ag bulk, nontrivial metastable structures should prevail at mild temperatures.…”

“…Unfortunately, the application of DFT-ML methods to understand the mechanisms of electrochemical corrosion in aqueous environments is still quite rare. − Nevertheless, we believe that a combination of DFT and ML tools can be extremely helpful in tackling complex corrosion problems. Recent advances in deep ML including the approaches based on Boltzmann generators can significantly help with sampling of equilibrium states.…”

Photocatalytic activity of dual defect modified graphitic carbon nitride is robust to tautomerism: machine learning assisted ab initio quantum dynamics

“…It is developing rapidly and has already been applied to a wide range of systems and processes. [33][34][35][36][37][38][39] ML methods are paving the way to uncover complex reaction paths and to correlate and predict material structure and properties, providing a balance between accuracy and efficiency. Generation of long MD trajectories with ab initio quality results is now feasible with the aid of ML force fields (MLFFs).…”