Computational, Data Science, and Experimental Evaluation and Design of Homogeneous Transfer Hydroformylation Catalysts

The computational design of transition-metal catalysts remains a grand challenge in chemistry. Here, we used density functional theory (DFT) calculations, data science, and experiments to tackle the analysis and design of new catalysts for transfer hydroformylation because only a single known bisphosphine ligand combined with Rh has been reported to provide high reactivity. We first used experiments to screen commercially available and common bisphosphines, and this data was then combined with DFT calculations and data science to screen >600 bisphosphine ligands from our ReaLigands library and develop interpretable chemical space mapping. Following another round of DFT calculations, we identified several promising bisphosphine ligands that were subsequently synthesized and validated experimentally.