Inverse Design of Metal-Organic Frameworks for Adsorption Processes: Learning Numerically Optimizable Surrogate Models of Isotherm Parameters.

Abstract

Metal-Organic Frameworks (MOFs) have gained considerable attention due to their exceptional adsorption properties, offering promising solutions for various applications, including gas separation and storage. In this study, we present an alternative approach for the inverse design of MOFs at the process level, utilizing optimizable surrogate models of isotherm parameters. Our methodology integrates adsorption isotherm equations and process models, facilitating the numerical optimization of MOF descriptors. We develop accurate and computationally efficient surrogate models, which serve as a bridge between MOF descriptors and their adsorption isotherm parameters. The optimizable nature of the surrogate models enables their seamless integration with process models, such as pressure swing adsorption (PSA) and vacuum swing adsorption (VSA) cycles. Moreover, our methodology offers a framework that can be extended to other applications and materials, advancing the field of material design and optimization.