Computational Studies of Electrocatalytic Sites

QC 20260929

Abstract

Electrocatalysis is important for energy conversion and chemical synthesis, like water splitting and \ce{CO2} reduction. Porous framework materials are very promising electrocatalysts because they can hold a high density of reactive sites. These sites can be spatially separated, structurally well defined, and chemically tunable. However, once they are incorporated into a framework, their performance can be affected by the local microenvironment, such as neighboring sites, solvent, and ions. These microscopic effects are difficult to resolve experimentally because many measurements reflect the average response of many sites rather than the behavior of each individual site.

This thesis develops and applies multiscale computational methods to connect the chemistry of individual reactive sites to the framework environment. Firstly, an automated modeling platform was developed to efficiently construct dynamic simulation-ready framework models. Dynamic simulations showed that molecular accessibility cannot always be understood from a single static structure, because the accessible pore environment can change over time. Secondly, quantum-chemical calculations were used to describe local reaction mechanisms. Furthermore, empirical valence bond theory combined with free-energy perturbation methods and molecular dynamic simulations was employed to investigate reactions in complex framework environments. This extends the local reaction description to systems containing multiple reactive sites, solvent, ions, and the surrounding framework environment. The studies show that framework effects depend on both reactions and systems. In some cases, the reaction behavior of an isolated catalytic site remains largely preserved after it is incorporated into the framework. In other cases, the assembled material creates different local environments that cause nominally equivalent sites to show different reaction energetics. Molecular accessibility can also vary independently.

Overall, reactive sites in framework materials cannot generally be treated as repeated copies of isolated molecular catalysts. Their behavior reflects both their local chemistry and the microenvironment created by the assembled material. At the same time, the well-defined and tunable structure of framework materials makes them useful for connecting molecular-scale chemistry with material-scale behavior. Together with experimental characterization, computational modeling can help connect microscopic catalytic behavior with material-level performance and support the design of more effective framework electrocatalysts.

Link to DiVA