(706b) High-Throughput Quantum Chemical Screening of Homogeneous Catalysts
After decades of advances in simulation of electronic structure and chemical reactivity, the ability to computationally design novel catalyst structures is close on the horizon. The ideal simulation tools would start from a class of desirable reactions and predict the most efficient, selective, and stable catalyst for the transformation. This lofty goal, however, is likely beyond reach due to the astronomic number of conceivable catalyst structures that would need to be evaluated. Instead, catalyst optimization—where catalyst features are varied started from existing structures—is aimed at the more modest goal of increasing the effectiveness of existing catalytic processes.
Research in the Zimmerman group has approached catalyst optimization using an evolutionary strategy to search for ligand variations that lead to higher activity in catalysis. Taking as an example the reduction of carbon dioxide by dihydrogen, cobalt diphosphine catalysts are known to efficiently effect this transformation in the presence of a base. Existing Co diphosphines, however, require very strong bases to turn over the catalytic cycle. Our computational screening has identified a number of phosphine ligands which increase the acidity of the Co complex and allow weaker bases to be effective. These catalysts are being synthesized in collaboration with Melanie Sanford at the University of Michigan to demonstrate the effectiveness of the screening procedure. Our results for this optimization will be presented alongside detailed descriptions of the computational techniques that are required for successful catalyst design.