Modeling the potential energy surface by force fields for heterogeneous catalysis: classification, applications, and challenges

Abstract

The concept of the potential energy surface (PES) in computational simulations is essential for studying material properties and heterogeneous catalytic processes. However, constructing the PES using quantum mechanical methods is computationally expensive and typically limited to small systems. Force field methods, which rely on quantum mechanical data, use simple functional relationships to establish a mapping between system energy and atomic positions or charges. Force field methods are more efficient for handling large-scale systems, such as catalyst structures, adsorption and diffusion of reaction molecules, and heterogeneous catalytic processes. To further promote in-depth research in this field, this review introduces the classification, development, and characteristics of various force field methods including classical force fields, reactive force fields, and machine learning force fields. It summarizes the forms, fitting methods, and distinct periods of these force field methods. Additionally, these force field approaches are compared in terms of their applicability, accuracy, efficiency, and fitting methods. Finally, the optimization and challenges of force field methods in constructing the PES are discussed. It is expected that this review will assist researchers in selecting and applying different force field methods more effectively to promote in-depth understanding of catalytic reaction mechanisms and the efficient design of catalysts.