Machine Learning: A New Paradigm in Computational Electrocatalysis

Abstract

Designing and screening novel electrocatalysts, understanding electrocatalytic mechanisms at an atomic level, and uncovering scientific insights lie at the center of the development of electrocatalysis. Despite certain success in experiments and computations, it is still difficult to achieve the above objectives due to the complexity of electrocatalytic systems and the vastness of the chemical space for candidate electrocatalysts. With the advantage of machine learning (ML) and increasing interest in electrocatalysis for energy conversion and storage, data-driven scientific research motivated by artificial intelligence (AI) has provided new opportunities to discover promising electrocatalysts, investigate dynamic reaction processes, and extract knowledge from huge data. In this Perspective, we summarize the recent applications of ML in electrocatalysis, including the screening of electrocatalysts and simulation of electrocatalytic processes. Furthermore, interpretable machine learning methods for electrocatalysis are discussed to accelerate knowledge generation. Finally, the blueprint of machine learning is envisaged for future development of electrocatalysis.

Tuning OER Electrocatalysts toward LOM Pathway through the Lens of Multi-Descriptor Feature Selection by Artificial Intelligence-Based Approach

“…Although it is quite challenging to rapidly screen excellent electrocatalysts from the huge materials space using time-consuming computational approaches, ML can provide an effective technology to evaluate the performance of electrocatalysts. As a result, it can be used to predict efficient electrocatalysts that have not been found or proposed. , For example, researchers used ML models to predict the adsorption free energies of intermediates in electrocatalysis and used for the selection of electrocatalysts . Therefore, as soon as the optimum descriptor or feature subset is generated, ML models can be trained using wide variety of algorithm.…”

“…The sluggish kinetics of anodic OER is the primary factor in determining the overall effectiveness of electrocatalytic water splitting process. , Accordingly, boosting hydrogen energy production is undoubtedly requires an effective and robust OER electrocatalyst. In several attempts, many transition metal oxides because of their intriguing catalytic properties, abundance, and low-cost have received a lot of attention as OER electrocatalysts. − However, as the data of OER electrocatalysts in chemical space is virtually unlimited, ML modeling approach is appealingly suitable for quick screening of prospective electrocatalysts from materials database. ,,− …”

“…With this ML model (Figure ), the group was able to screen other promising spinel oxides, and [Mn][Al 0.25 Mn 0.75 ] 2 O 4 was identified due to its confirmed superior performance. This implies that ML can guide rational design of the OER electrocatalysts using descriptors. , …”

“…As a result, it can be used to predict efficient electrocatalysts that have not been found or proposed. 142,143 For example, researchers used ML models to predict the adsorption free energies of intermediates in electrocatalysis and used for the selection of electrocatalysts. 143 Therefore, as soon as the optimum descriptor or feature subset is generated, ML models can be trained using wide variety of algorithm.…”

“…This implies that ML can guide rational design of the OER electrocatalysts using descriptors. 142,143 Considering the fact that feature selection deals with process of identification and selection of a subset of input features that are most relevant to the target decision or variable, the technique evaluates information from each feature in the context of target decision or variable and selects the feature. This maximizes gain of information, which in turn minimizes the input features into subset�a representation that is close to the original input.…”