Machine learning-assisted QnSPR study of structurally diverse nanoporous carbon materials and their capacitive behavior in dilute ionic liquid electrolyte
Maike Käärik
In the development of efficient energy storage solutions, it is important to understand the relationship between electric double-layer (EDL) capacitance and the structural properties of electrode materials. Using machine learning, 67 nanoporous carbon materials, characterized with determined original experimental materials textural descriptors and their measured electric double-layer capacitance in a 1.9 M EMIm-TFSI solution in acetonitrile (ACN) were analyzed. A linear regression model was developed based on the materials specific surface area, pore volumes, and electrode density, enabling the prediction of both cathodic and anodic capacitance. The model achieved an R² of 0.93 for cathodic capacitance and 0.94 for anodic capacitance. The results showed that very small pores are most suitable for the adsorption of electrolyte ions—0.4–0.5 nm for the EMIm⁺ cation and less than 0.4 nm for the TFSI⁻ anion. This confirms that tailoring the pore size of nanoporous carbon to the dimensions of the ions is important for improving its energy storage capacity.
