Metal–organic frameworks (MOFs) are a class of crystalline materials composed of metal nodes connected by organic linkers. Due to their high degree of synthetic tunability, MOFs have been considered for a wide range of applications, including many that rely on a change in oxidation state. While most MOFs are generally considered to be redox-inactive, a growing number of MOF structures have been synthesized that can support redox processes, oftentimes via the presence of open-shell transition metal cations and/or redox non-innocent linkers. These so-called redox-active MOFs have been investigated for challenging catalytic oxidation reactions, the selective adsorption of reducible gas species, and next-generation electronic devices. In this dissertation, a computational screening approach based on density functional theory calculations is used to gain insight into the reactive properties of redox-active MOFs for three main application areas. The first portion of this dissertation is focused on the catalytic oxidation of strong C–H bonds, such as those of light alkanes, via the formation of high-valent metal-oxo and metal-oxyl species at the inorganic nodes of MOFs. Switching focus to adsorption processes, the second portion of this dissertation is centered around the design of MOFs with redox-active metal centers that can selectively bind O2 over N2 via charge transfer interactions. In the final portion of this dissertation, a high-throughput periodic DFT workflow is used to create the first large-scale quantum-mechanical property database for MOFs, which is then used to train machine learning models that can guide the discovery of MOFs with targeted band gaps. Throughout this work, several methodological studies are also carried out to better understand the qualitative and quantitative shortcomings of different density functional approximations with the goal of making more actionable predictions in future computational screening efforts. Collectively, this dissertation demonstrates the ability to use high-throughput quantum-mechanical simulations to discover new structure–property relationships, identify promising MOFs for challenging oxidation reactions, and more efficiently explore the vast expanse of MOF chemical space.

Creator

• Rosen, Andrew Scott

DOI

• https://doi.org/10.21985/n2-j52z-ey73

Subject

• Chemical engineering

Language

• en

Alternate Identifier

• etdadmin_upload_838355
• http://dissertations.umi.com/northwestern:15675

Keyword

• Computational screening
• Density functional theory
• Machine learning
• Metal-organic framework

Date created

• 2021-01-01

Resource type

• Dissertation

Rights statement

• In Copyright