Classical density functional theory as a fast and accurate method for adsorption property prediction of porous materials

Abstract

Physical adsorption is crucial in many industrial processes, prompting researchers to develop new materials for energy-efficient processes. Porous adsorbents are particularly promising due to their design flexibility, and computational screening has accelerated the search for optimal materials. Recently, classical density functional theory (cDFT) has emerged as a faster screening alternative to state-of-the-art computational methods. However, its predictions have not been extensively validated, especially for materials involving strong Coulombic interactions. This article validates cDFT by calculating adsorption properties for over 500 Metal-Organic Frameworks with three adsorbates $({\text{CH}}_4$, ${\text{N}}_2$, ${\text{CO}}_2)$ and comparing them to results from Grand Canonical Monte Carlo (GCMC) simulations. For ${\text{CO}}_2$, accounting for Coulombic interactions is crucial for accurate predictions. Our findings show that cDFT closely replicates GCMC results while reducing computation time to a median of six minutes per material, making it a strong candidate for estimating adsorption properties in porous materials.