A Data Science-Guided Approach for the Development of Nickel-Catalyzed Homo-Diels–Alder Reactions

The Ni-catalyzed homo-Diels-Alder (hDA) reaction represents a convergent but under-investigated approach to preparing bridged bicyclic ring systems. Using the <i>kraken</i> monophosphine descriptor library, Ni-catalyzed hDA reactions of acyclic and cyclic electron deficient olefins were investigated, and key ligand effects required for reactivity were identified using classification models. This analysis guided the discovery of the monophosphine (<i>S</i>)-AntPhos as a chiral ligand for the enantioselective hDA of acyclic dienophiles. However, these conditions were not compatible with cyclic substrates. Further mechanistic and computational studies revealed a putative role of Ni(I) species and mechanistic divergence between cyclic and acyclic enone dienophiles. Using reaction space design and Bayesian optimization, conditions were developed that expanded the scope to cyclic dienophiles. The resultant cycloadducts were transformed into bicycloheptane structures via cyclopropane cleavage reactions, demonstrating the ability to rapidly access structurally complex scaffolds using this method.