Fundamental Insights into Catalyst Design: Energetics of Metal Atom Adsorption and Nanoparticle Adhesion on Oxide Surfaces
Hemmingson, Stephanie Leah
MetadataShow full item record
Heterogeneous catalysts consisting of transition metal nanoparticles dispersed on high surface area oxide supports are ubiquitous in industrial-scale chemistry and alternative energy technology. Despite their importance, our fundamental understanding of the physical and chemical properties that make these systems catalytically effective is still incomplete. This dissertation details the use of surface-sensitive ultrahigh vacuum (UHV) techniques to study model catalysts consisting of late transition metals that are vapor-deposited onto single-crystal oxide films in order to understand how their fundamental physical and chemical properties affect their catalytic properties, such as activity, selectivity, and resistance to deactivation. Specifically, the binding energies of metal atoms and nanoparticles – and the adhesion energy of metal nanoparticles and films – are measured as a function of the size and type of nanoparticle, and the surface or site that they are adsorbed on. The key technique utilized in this study is single-crystal adsorption calorimetry (SCAC), which uses a highly sensitive detector to directly measure the heats of adsorption of metal vapor adsorbing onto oxide thin films. These calorimetric data are combined with surface-sensitive spectroscopic techniques to characterize the oxide surface, and to model the size of the nanoparticles a function of total metal coverage. This approach allows the heat data to be correlated with the surface structure, composition, particle size, or electronic character of the system. Several improvements to the instrument are discussed that allow for the study of metals with very low vapor pressures (high heats of vaporization), specifically Au, which is the focus of this work. This new instrument can also be operated at temperatures as low as 100 K, which is used to study metal atom adsorption under conditions where adatom diffusion is slower, which increases the particle density (reducing the size of particles formed) and increasing the likelihood that nucleation occurs at less favorable sites. Au adsorption is discussed here on three difference surfaces with increasing complexity: Pt(111), MgO(100), and CeO2-x(111). These results are compared to Cu adsorption on Pt(111) and CeO2-x(111) (also presented in this work), and to Cu adsorption on MgO(100) as well as Ag adsorption on MgO(100) and CeO2-x(111) (from previous work done by this research group). The first ever experimental measurements of gold adsorption onto an oxide surface as a function of particle size are presented, and the adsorption energy of both a single Au atom and a single Cu atom was measured on CeO1.95(111). This represents the first experimental measurements of any metal atom adsorption onto any oxide surface. The effect of defects on Au and Cu adsorption, such as step edges and electron-rich oxygen vacancies is discussed in detail. These defects are either introduced in film preparation (in the case of oxygen vacancies on CeO2-x(111)), or are inherently present in thin film preparations (morphological defects). The metals studied here, like most late transition metals, adsorb more strongly on morphological defects such as steps, while only Au and Ag bind more strongly to oxygen vacancies on CeO2-x(111) (Cu does not). Additionally, Au was found to nucleate significantly more strongly than Ag and Cu on morphological defects of MgO(100), and to form 2D islands on MgO(100) within the first 0.4 monolayers (ML) coverage. This work presents the first ever measurement of the heat of adsorption (and thus the chemical potential) of metal atoms in metal nanoparticles as a function of their 2D island diameter. A similar interpretation is used to present the chemical potential of Cu atoms in Cu nanoparticles as a function of their 3D particle dimeter on CeO2-x(111). The work contained in this dissertation has added critically important thermodynamic and structural data to the library of research in the catalysis and surface science communities, and has addressed some of the most relevant materials that could not be studied with previous generations of SCAC instruments. These model catalyst systems are of substantial fundamental and practice interest due to their immediate use in industrial catalysis, and combining these newest results with existing data (both from our group and from the literature) has allowed us to propose a new trend for metal-oxide adhesion. These investigations, and specifically this trend, will aid in the global effort to expedite the future design and testing process of new catalytic materials through improved understanding of their fundamental properties.
- Chemistry