Deep Learning-Based Energy Mapping of Chlorine Effects in an Epoxidation Reaction Catalyzed by a Silver-Copper Oxide Nanocatalyst

Deep learning is poised to revolutionize the field of heterogeneous catalysis. In this study, we harness its potential to predict energy values across a catalyst surface, a task traditionally relegated to computationally intensive density functional theory (DFT). We propose a novel deep learning approach to construct an exhaustive energy map, pinpointing the optimal locations for adsorbed chlorine in the ethylene epoxidation reaction. Leveraging the power of trained neural networks, we achieved a staggering reduction in computational time, cutting down the duration of energy calculations by over 50 million times compared with traditional methods. This groundbreaking integration of artificial intelligence not only accelerates this process but also effectively surpasses the limitations of conventional methods. By highlighting the transformative potential of deep learning in catalysis, this research paves the way for future studies and stands to revolutionize efficiency in the chemical industry, fostering an urgent need to delve deeper into the implications and applications of this technology.