A Theoretical Study on the Surfaces of Zinc Oxide

Zinc oxide is an important wide bandgap n-type semiconductor with uses ranging from electronics to catalysis. The chemical and physical properties related to its surfaces are of fundamental interest and also key to the material’s design. In this Thesis, computational methods have been used to model the surfaces of ZnO. We report a detailed theoretical study on the four main low-index wurtzite ZnO surfaces. For nonpolar surfaces, we focus on the stability, atomic structure and electronic properties of both clean and defective surfaces. Our calculations explain why steps are common on the (10-10) surface, as seen in experiment. We calculate the ionisation potential which is in good agreement with experiment. The electronic band edges of the nonpolar surfaces are seen to behave differently, with a local rise of the VBM and CBM for (10-10) and (11-20), respectively. For ZnO polar surfaces, our results can explain why experimental findings reported have been varied and even contradictory at times. The calculated surface energies indicate on average a slightly higher stability of the (000-1) surface compared to the (0001) surface. Structurally, triangular and hexagonal patterns are seen among the stable structures but a high level of disorder is predicted. We also report new interatomic potentials (IP) for the Cu/ZnO system. Our IP can work as a fast and reliable method to filter low energy Cu/ZnO structures. Global optimisation calculations show a preference for planar Cu clusters over the (10-10) surface, with a strong interaction between the Cu and Zn species. Finally, we study the surface atomic configurations for the MoO3/Fe2O3 catalytic system. The lowest energy structure was used in the fitting of EXAFS parameters. Overall, our Thesis shows the great utility of theoretical calculations in the explanation of experimental findings in surface science