Investigating metal oxides for C1 chemistry : a computational and experimental study
Date of Issue2019-04-29
School of Chemical and Biomedical Engineering
With huge global natural gas reserves, coupled with depletion of fossil fuels, researchers are exploring potential technologies to convert methane into high value-added chemicals. Transition metal oxides (TMOs) are an important class of catalytic materials widely used in the direct and indirect catalytic methane conversion methods. Fundamental understanding of the reaction chemistry is crucial to the realization of commercial catalytic methane conversion processes. Computational tools like Density Functional Theory (DFT), and experimental surface characterization techniques like Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) and X-ray Photoelectron Spectroscopy (XPS), are often used independently by computational and experimental researchers, to provide insights into metal oxide catalyzed reaction mechanisms, pathways, and energetics. Identifying realistic reaction intermediates and their corresponding IR and XPS spectral peaks presents a challenge for researchers to study complex catalytic reactions on TMO surfaces. DFT calculations are widely employed to assist in experimental analysis and provide mechanistic insights into reaction intermediates, and pathways for TMO catalyzed reactions. Localised nature of electrons leads to the strongly correlated nature of TMOs, and the standard approximations in the DFT exchange-correlation functional fail to describe these electron localizations accurately. DFT+U method is a widely used extension of DFT, where the Hubbard U term is an onsite potential that penalizes electron delocalization, successfully describing such systems. This U-value is usually chosen based on its accuracy in reproducing bulk properties, however, using the bulk properties-based U-values in a locally changing surface reaction environment may not describe the surface reaction energetics correctly. CuO is a widely-used TMO with applications in heterogeneous and environmental catalysis. In this thesis, two novel DFT+U approaches are developed with CuO as a model TMO and present methods for U-value determination that accurately capture the surface chemistry on TMOs. Both these approaches provide consistent U-values of 4-5 eV that accurately predicts the surface catalytic properties, as opposed to the widely used bulk property, optimized U-value of 7 eV that fails to correctly predict both surface reaction energetics and XPS shifts for CuO catalytic systems. In the first method, DFT+U calculations are performed to investigate the dissociative chemisorption of H2 on CuO, and the appropriate U-value is determined from the comparison of DFT+U calculated adsorption enthalpies for different U-values with the experimental adsorption enthalpy. In the second method, comparison of experimental XPS shifts with DFT+U derived XPS shifts (for a range of surface moieties on the catalyst) for different U-values leads to the determination of the U-value. The second method not only benchmarks the U-value but also establishes unknown surface adsorbates, their configurations and predict their experimental XPS shifts synergistically, thus addressing the bottlenecks associated with the application of integrated computational and experimental methods for studying TMO catalyzed reactions. After establishing a robust DFT+U method that accurately captures surface chemistry on TMOs, the sequential activation of C-H bonds of methane on CuO is investigated using DFT+U and experimental FTIR study. The primary process and operational challenges for methane catalytic conversion over TMO surfaces are associated with the activation of the highly stable and weakly polarized C-H bonds of methane, which leads to the employment of high operating temperatures, consequently favoring undesired reactions, leading to a loss in activity, selectivity, and yield of desired products. In literature, the presence of moisture and surface hydroxyls on transition metal systems is known to reduce the barriers of various catalytic methane conversion reactions, but their role in TMO catalyzed reactions has not been investigated. The current thesis reveals that the presence of surface hydroxide species on CuO reduces the activation barriers associated with methane activation and dissociation. Experimental FTIR study was performed to investigate the reaction under dry and moisturized reaction conditions and revealed the presence of similar intermediates (methoxy, formic, CO2, and surface hydroxyl species) under both conditions. Presence of surface hydroxide species even under dry conditions is predicted by DFT and confirmed by FTIR experiments, thus indicating that the C-H bond of methane may not be activated by CuO lattice oxygen directly but gets oxidized with the formation of moisture and surface hydroxide species, henceforth reducing the C-H activation barriers. In summary, this thesis reveals: (i) A novel surface reactivity based semi-empirical DFT+U approach that accurately describes surface catalytic reactions over TMOs and establishes the failure of widely applied bulk property optimized U-value to study these surface catalytic reactions, (ii) The Synergistic application of DFT and experimental XPS shifts to determine the reaction intermediates and surface species for TMO catalyzed reactions without any explicit knowledge of the U-value for respective TMO. This is the first study that integrates DFT and XPS methods to study TMO catalyzed reactions and addresses the bottleneck constraining the application of combined experimental and computational approaches to such systems. (iii) The promoting role of moisture and surface hydroxides in the activation and dissociation of methane from computational and experimental studies. This study provides evidence that even under dry conditions, lattice oxygen may not directly activate the C-H bonds of methane but indirectly via the formation of surface hydroxides formed on the surface. It also establishes that moisturized surface approach can lead to an increased lifetime of the catalysts due to water being the oxygen donor for the initial part of the methane activation.