A large part of the research in the group involves synergistic integration of experiments with computations. While DFT remains the main workhorse for elucidating reaction mechanisms and predicting transition metal (TM) complex properties for catalyst design calculations, widespread exploration with different combinations of ligands and transition metals remains challenging due to computational costs and inaccuracies that limit its predictive performance for localized d-electron complexes. Research using predictive AI models in organic chemistry structure-activity relationships has seen exponential growth over the last decade. However, the use of AI in the inorganic chemistry space remains challenging due to the multifaceted reactivity exhibited by TM. Any implementation of predictive and generative artificial intelligence strategies for inorganic and organometallic chemistry will first require the development of agents (machine learning models/neural networks) that can accurately predict specific property trends across different TM complexes. 

The Haber-Bosch (HB) process is one of the most important industrial chemical reactions, affording ammonia from atmospheric nitrogen. The “fixed” nitrogen is a source of almost all artificial fertilizers, supporting half of the world’s food production. HB, however, is also one of the most energy-intensive reactions that use approximately 2% of the world’s fossil fuel as a source of hydrogen and energy. In recent years, there has been renewed interest in developing electrochemically driven nitrogen reduction to ammonia (ENR) using sustainable sources of power with the ultimate goal of using greener sources of protons (H+) and electrons. 

The electrochemical up-conversion of carbon dioxide to liquid fuel provides an attractive strategy for storing renewable energy from sunlight. Therefore, carbon dioxide reduction reactions to access alcohols and long-chain hydrocarbons are extremely important and present one of the greatest challenges of modern-day catalysis. One of the key transformations in the mechanism of such a reduction is the conversion of carbon dioxide to carbon monoxide, and many electrocatalysts have already been reported to carry out this conversion with high Faradaic efficiency. However, to access liquid fuels, further reduction of CO is necessary to obtain products like methanol, ethanol, or hydrocarbons. Thus, reductively coupling CO to access hydrocarbons or hydrogenating CO to methanol has been of long-standing interest to inorganic chemists.