Date of Award
Program or Major
Doctor of Philosophy
With the growth of the population, society’s energy demands are mostly reliant on petroleum products that come from the refining of crude oil. Most of these refining reactions have been developed through averaging spectroscopic techniques, but scientists do not know exactly what is happening in these processes at the nano and atomic levels. This information is crucial when designing an efficient refining process that produces petroleum products that emit fewer harmful gases when combusting. Scanning probe microscopy techniques have become a powerful tool to look into the chemical structures found in petroleum products, to understand catalytic reactions in refining processes, and to find new non-combustible uses for these products. In this dissertation, I show how scanning probe microscopy (SPM) techniques, especially non-contact atomic force microscopy (NC-AFM) can provide an atomic-level understanding of the chemical structures and active catalytic sites that play a role in these refining processes. First, I studied hydrodesulfurization reactions that use molybdenum disulfide as a main catalyst to explore the effect of layer thickness, strain, and underlying substrates on its electronic and catalytic properties. Here, I present the first NC-AFM experiments investigating the active catalytic sites of molybdenum disulfide on industrially relevant substrates. Through these experiments, I found how NC-AFM techniques on insulators need to be improved to achieve high-resolution images that are comparable to those collected on metal substrates. Second, I created Auto-HR-AFM, a machine-learning script that collects optimal high-resolution NC-AFM images. Auto-HR-AFM is a modular and open-source script that provides an initial framework for a fully automated SPM. Expanding on this framework will widen the use of scanning probe microscopy techniques to non-experts and the automation will increase the time the system is kept running to collect large optimal datasets. Ultimately, these studies will broaden the use of high-resolution SPM techniques and help create more efficient catalysts and refining processes to produce cleaner and more efficient petroleum products.
Arias, Steven, "Scanning Probe Microscopy Studies of Petroleum Chemistry: Substrate-Dependent Catalytic Properties of MoS2 and Automating Scanning Probe Microscopy with Machine Learning" (2023). Doctoral Dissertations. 2732.