Global Neural Network Potential for Material Simulation and Catalysis

While the underlying potential energy surface (PES) determines the structure and other properties of material, it has been frustrated to predict new materials from theory even with the advent of supercomputing facilities. The accuracy of PES and the efficiency of PES sampling are two major bottlenecks, not least because of the great complexity of material PES. This lecture introduces our recent progress in SSW-NN method and its application in catalysis. We designed a “Global-to-Global” approach for material discovery by combining the novel global optimization method with neural network (NN) techniques. We describe in detail the current implementation of the SSW-NN method with particular focuses on the size of the global data set and the simultaneous energy/force/stress NN training procedure. All these methods have been implemented in LASP software (www.lasphub.com). A number of important reactions and functional materials, in particular those related to catalysis e.g. ZnCrO oxides and Titania supported Au particles, are utilized as the examples to demonstrate the automated global data set generation, the improved NN training procedure and the application in reaction discovery and catalysis. As a general tool for reaction/material simulation, the SSW-NN method provides an efficient and predictive platform for large-scale computational material screening.

1. Pei-Lin Kang, Cheng Shang, Zhi-Pan Liu , Large-Scale Atomic Simulation via Machine Learning Potentials Constructed by Global Potential Energy Surface Exploration, Acc. Chem. Res. 2020, 53, 10, 2119