EXTENDED ABSTRACT: Two-dimensional (2D) materials have been developed into various catalysts with high performance, but employing them for developing highly stable and active nonprecious hydrogen evolution reaction (HER) catalysts still encounters many challenges [1]. To this end, the artificial intelligence (AI) screening of HER catalysts is accelerated by using different appropriate methods (simple element information, genetic programming, improved crystal graph convolutional neural network) for various layered 2D materials. The values of the Gibbs free energy of hydrogen adsorption (ΔGH*) are accurately and rapidly predicted via these AI models. Based on these ffndings, more than ten layered 2D materials are discovered to be active catalysts with thermodynamical and dynamical stability as ΔGH* approaches to zero. Overall, a multistep workffow is developed through ML models combined with DFT calculations for efffciently screening the potential HER catalysts from 2D materials with the different situations, which is believed to have signiffcant contribution to catalyst design and fabrication.
Keywords:Hydrogen evolution reaction; Density functional theory; Artiffcial intelligence algorithm
REFERENCES:
[1] Turner, J. A, Science, 305, (2004) 972−974
Xiang Sun, Ph.D., Senior Engineer. He obtained his Ph.D. in Chemical Engineering from Zhejiang University of Technology in 2020 and continued his postdoctoral work at the same university. He has been working for Zhejiang Institute of Chemical Industry since 2022 until now. Currently, he is mainly engaged in the prediction of long-term cycling performance of lithium batteries with fluorine-containing additives, molecular fundamental properties simulation of ffuorine molecules, and related reaction mechanism research. He has published more than 10 papers in journals such as Adv. Funct. Mater., Appl. Catal. B Environ., ACS Appl. Mater. Interfaces, Chem. Eng. J., and J. Phys. Chem. C. He has also been granted one invention patent and three software copyrights.