Predicting the Stability of Base-mediated C─H Carboxylation Adducts Using Data Science Tools.

Base-mediated C-H carboxylation is a versatile pathway for utilizing carbon dioxide (CO2) as a C1 building block in organic synthesis. However, CO2 constitutes a notorious thermodynamic sink, which restricts this approach to activated or intrinsically reactive nucleophiles. To qualitatively assess the stability of CO2 adducts, we present a computational approach that integrates quantum chemistry with statistical modeling to build a predictive workflow. The target property is the CO2 affinity, specifically the negative Gibbs free reaction energy. This predictive workflow has been applied to 60 novel carbon-centered nucleophiles, suggesting reactions that yield stable carboxylation adducts. The results have been validated through experimental methods for five carbanions, which include three stable and two unstable adducts in DMSO according to our predictions. In addition, we examined two further carbanions that were suggested to form stable CO2 adducts in DMSO, to further assess the experimental protocol and broaden its scope to structurally distinct motifs.