Determining desired sorbent properties for proton-coupled electron transfer-controlled CO2 capture using an adaptive sampling-refined classifier

Abstract

Electrochemical CO2 capture technologies have been found to consume less energy than the industry standard of thermal separations, but their real-world applicability requires that they also operate at comparable rates. Optimizing for both low energy demands and high capture rates is complicated by trade-offs between the two objectives and the many manipulable solution chemistry variables, including species type and concentration. Here, we computationally identified the solution chemistries that are most likely to outperform thermal separations in both energy demand and capture rate for electrochemical capture driven by proton-coupled electron transfer reactions by using an adaptive sampling contour estimation method. This approach provided high confidence inferences with few simulation runs by selecting the most informative conditions to test. We found that moderately basic pKa values of the reduced form of the redox-active compound were the most important variables for low energy and high rate CO2 capture.