The majority of research literature on chemical separations focuses on a relatively small collection of chemicals. In many ways this is appropriate because of the dramatic economic and environmental impacts of these species. The space of possible chemicals, however is vast; Carl Sagan's famous "billions and billions" is many orders of magnitude too small for chemical space. This talk will describe efforts using data from molecular simulations and machine learning tools to efficiently predict the adsorption properties of arbitrary molecules in large libraries of crystalline nanoporous materials. In addition to using machine learning to make extremely rapid predictions, we will illustrate how a hierarchy of modeling methods needs to be adopted to move from these predictions to specific guidance for experimental materials development.

Towards accurate prediction of adsorption equilibrium for arbitrary molecules in large libraries of nanoporous materials

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.