Unifying Chemistry and Machine Learning for the Study of Noncovalent Interactions

Gas separations are in great demand for carbon emission reduction, natural gas purification, oxygen isolation, and much more. Many of these separations rely on cost-prohibitive methods such as cryogenic distillation or strong-binding solvents. As a result, novel materials are being developed to subvert the energetic expense of gas separation processes. These studies focus on improving the performance of alternative materials, including (but not limited to) metal-organic frameworks, covalent organic frameworks, dense polymeric membranes, porous polymers, and ionic liquids.

In this work, the atomistic effects of functional units are explored for gas separations processes using electronic structure theory and machine learning. In particular, the crucial role of noncovalent interactions for these processes is investigated. Specifically, the atomistic effects are studied for the separation of CO2 and N2 using vinyl-addition polynorbornenes to interpret how structural modification can contribute to macroscopic permeability properties. An in-depth examination elucidated the role of functionalizing boranes to increase N2 interaction for the selective separation from O2. Here, a small-scale screening was performed to append electron-withdrawing moieties with boranes to develop a stronger Lewis acid to interact with N2. As a next step, electronic structure theory and machine learning were employed to suggest promising functional units for CO2/N2 separations. Novel techniques are developed to perform a high-throughput screening of a large molecular database with great accuracy. Lastly, the efficiency and applicability of accurate calculations for noncovalent interactions was improved by reducing computational expense of coupled cluster calculations using machine learning. This work introduces novel methods to study noncovalent interactions crucial for gas separations through the incorporation of accurate electronic structure theory methods and advanced machine learning techniques.