Investigating High-Performance Non-Precious Transition Metal Oxide Catalysts for Nitrogen Reduction Reaction: A Multifaceted DFT–kMC–LSTM Approach

The need for non-precious-metal catalysts for the nitrogen reduction reaction (NRR) is growing due to the high cost of precious-metal catalysts. Transition metal oxides (TMOs) are a promising option, but there is limited experimental and computational evidence for their use. The present work is a comprehensive investigation of multiple TMOs utilizing a multifaceted approach. Specifically, we integrated density functional theory (DFT) to analyze thermodynamic and electronic properties, kinetic Monte Carlo (kMC) to simulate the entire reaction pathway for candidate materials, and long–short-term memory (LSTM) methods to incorporate the long-term surface degradation within the kinetic model. Using this approach, we predicted that V2O3 would exhibit superior NRR activity compared to a noble Ru catalyst, with a turnover frequency 1000 times higher, while also demonstrating stable performance for over 10,000 h. It is noteworthy that one of the key factors contributing to the high performance of TMOs is their unique ability to upshift the O 2p-band due to the tensile strain generated by N2-TM site binding. This allows for swift H transfer to adjacent N-containing species on TM sites by weakening H binding strength, leading to an accelerated NRR process. Overall, this work provides a holistic investigation of TMOs through a comprehensive DFT–kMC–LSTM approach and demonstrates their unique ability to leverage complementary H transfer between TM and O sites rather than suppressing the competing hydrogen evolution reaction (HER) processes.