Insights into CO₂/N₂ Selectivity in Porous Carbons from Deep Learning

Porous carbons are an important class of porous material for carbon capture. The textural properties of porous carbons greatly influence their CO 2 adsorption capacities. But it is still unclear what features are most conductive to achieving high CO 2 /N 2 selectivity. Here, we trained deep neural networks from the experimental data of CO 2 and N 2 uptakes in porous carbons based on textural features of micropore volume, mesopore volume, and BET surface area. We then used the model to screen porous carbons and to predict CO 2 and N 2 uptakes, as well as CO 2 /N 2 selectivity. We found that the highest CO 2 /N 2 selectivity can be achieved not at the regions of highest CO 2 uptake but at the regions of lowest N 2 uptake where mesopores disrupt N 2 adsorption. This insight will help guide experiments to synthesize better porous carbons for post-combustion CO 2 capture.