First-Principle Investigations of C-H bond Activation on Nickel Oxide: Electrochemical Activation of Methane and Oxidative Dehydrogenation of Ethane

Computational design of heterogeneous catalysts requires a detailed understanding of reaction mechanisms and the identification of active sites. Over the last decades, density functional theory (DFT) has evolved as the standard tool for investigating catalytic reactions, because it can make ab initio predictions of surface properties, allowing for the estimation of thermodynamic and kinetic parameters of elementary reaction steps. In this dissertation, DFT calculations were performed to elucidate the mechanism of C-H bond activation thermodynamically stable hydrocarbons such as methane and ethane, the primary components of abundant natural gas. First, methane conversion to methanol assisted by pre-adsorbed CO3 and O on both Ni(111) and NiO(100) surfaces were studied. The direct formation of methanol, either activated by CO3 or O on Ni(111) has high activation barriers and thus, is not likely to proceed. Conversely, a direct pathway to methanol was discovered on the NiO(100) surface, which involves an oxygen adatom and mimics the radical rebound mechanisms previously reported for various single site catalysts. Despite the transient formation of a methyl radical, this pathway is energetically preferred over other mechanisms involving the highly activated C-O bond formation step between adsorbed CH3 and OH intermediates. Relevant to the electro-oxidation of methane, we then investigated the sensitivity of the methane-to-methanol reaction on NiO(100) to the presence of (oscillating) electric fields. Our DFT results show that positive and negative electric fields favor different elementary steps in the reaction cycle, indicating that the overall reaction can be accelerated with a dynamically applied, oscillating potential. The oxygen adatom site on NiO(100) also plays a critical role as active site for the first C-H bond activation step in the oxidative dehydrogenation (ODH) of ethane. This conclusion is supported CO2 adsorption experiments, FTIR spectroscopy, and reactivity measurements contributed by our experimental collaborators. These experiments showed that CO2 can be used as probe molecule to titrate two distinct active oxygen sites, and DFT calculations suggest that nearby Ni vacancies are required. These sites were identified by matching the calculated CO2 vibrational fingerprints and the experimental FTIR spectra of CO2 adsorption. Combining the experimental results of the ODH reaction and the calculated reaction enthalpy, we were able to propose a reaction cycle involving two active oxygen sites near Ni vacancies with different activity: the more active non-stoichiometric oxygen adatoms are readily passivated by CO2, whereas the less reactive lattice oxygen persists in the presence of CO2 at reaction temperature. These outcomes provide fundamental insight into the role of non-stoichiometric oxygen in metal oxides and help guide the future design of catalysts with tailored surface oxygen population for ODH reactions. Overall, this dissertation demonstrates the power of ab initio simulations in assisting the interpretation of experimental observations. This concept can be extended to other non-catalytic systems, such as providing an explanation for the reported odd-even effect in the experimentally measured wettability of organic self-assembled monolayers (SAMs). Here, our ab initio molecular dynamics (AIMD) simulations provided insight into dipole moments at the SAMs-liquid interface that can be tailored by different tail groups and number of CH2 units in the carbon chain. Other than aiding the interpretation of experiments, we have also discovered novel mechanisms and used DFT to predict how oscillating electric fields may be leveraged to accelerate methane electro-oxidation to methanol.