Accurate Predictions of the Adsorption Space and Efficient Sorbent Discovery in Metal-Organic Frameworks
Loading...
Author(s)
Yu, Xiaohan
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
Adsorption-based separations using metal–organic frameworks (MOFs) are promising candidates for replacing common energy-intensive separation processes. The so-called adsorption space formed by the combination of billions of possible molecules and thousands of reported MOFs is vast. It is very challenging to comprehensively evaluate the performance of MOFs for chemical separation through experiments. Molecular simulations and machine learning (ML) have been widely applied to make predictions for adsorption-based separations. Previous ML approaches to these issues were typically limited to smaller molecules and often had poor accuracy in the dilute limit. The present thesis addresses this limitation by first developing accurate ML models predicting Henry’s constants and heats of adsorption. We then developed accurate ML models predicting adsorption isotherms of diverse molecules in large libraries of MOFs. By combining molecular simulation data, ML predictions with Ideal Adsorbed Solution Theory, we tested the ability of these approaches to make predictions of adsorption selectivity and loading for challenging near-azeotropic mixtures. We then focused on exploring MOFs for direct air capture (DAC). We presented Open DAC(ODAC) 2023 database with over 38 million quantum chemistry calculations on thousands of MOFs containing CO2 and/or H2O. We introduced a tool to automatically generate missing-linker defects in MOFs and applied the tool to include more than three thousand defective MOFs to the database. Over two hundreds of promising MOFs were identified and the influence of defects was studied. Machine learning models were developed based on this database to accelerate the development of MOFs for DAC.
Sponsor
Date
2025-04-07
Extent
Resource Type
Text
Resource Subtype
Dissertation