Elucidating nitrogen doping effects in carbon-based CO2 adsorbents through data-driven analysis for rational adsorbent design

The influence of nitrogen doping on the CO 2 adsorption performance of carbon-based materials remains debated, with unclear conclusions regarding whether surface chemistry or textural properties primarily govern adsorption behavior. In this study, we systematically examined the effects of nitrogen content and surface area on CO 2 adsorption capacity using a data-driven statistical approach. A comprehensive dataset of 1215 experimental data points compiled from 100 publications was analyzed through correlation analysis, multiple linear regression, and mediation analysis. The results reveal that nitrogen content and surface area are interrelated and may exert competing effects on adsorption. Multiple linear regression confirms that surface area exerts a substantially greater influence on CO 2 adsorption than nitrogen content, as indicated by higher standardized regression coefficients at both 273 K and 298 K. Mediation analysis further demonstrates that the influence of nitrogen on CO 2 adsorption occurs predominantly indirectly through its modification of surface area, with mediated proportions of 70.1 % and 58.2 %, respectively. A comparative evaluation of nitrogen-doped and undoped carbons supports these findings, showing consistent positive correlations between surface area and CO 2 adsorption capacity regardless of nitrogen incorporation. These results provide quantitative evidence that textural development and surface accessibility play a more decisive role than nitrogen functionalities in CO 2 capture. The insights obtained here establish a data-driven foundation for the rational design and future machine-learning optimization of advanced carbon-based CO 2 adsorbents.