| The primary objective of this dissertation was to explore the feasibility of using scanning tunneling microscopy and spectroscopy (STM/S) to observe and quantify spatially-localized variations in the electric potential on transition metal oxide surfaces. This work was carried out on three separate systems. First, ZnO, which is known to have an orientationally-dependent work function. Second, segregation in TiO{dollar}sb2{dollar} provided a system in which to explore the effect of an electric potential gradient on the tunneling spectra. Thirdly, TiO{dollar}sb2{dollar} dosed with the same element which segregated. The procedure and method used to measure the local variations in the electric potential on these systems was primarily accomplished with STM/S. Both polar ZnO surfaces show a continuous variation in the local electronic structure across a step of a mesa consistent with the work function varying. Specifically the carrier density varied a factor of six between the terrace and the step on the Zn-terminated surface. Dosing this surface with oxygen resulted in an increase of one electron per 1000 ions, whereas dosing the O-terminated surface resulted in an increase of three. Upon segregation of Cu, one micron diameter CuO hillocks were uniformly distributed on the surface, which was determined to be insulting. The Cu-segregation to the rutile surface and the insulating property were predicted using an ionic segregation model. After reduction, the variations in the tunneling spectra on and off the hillocks were recorded, and modeled as due to primarily two factors: first, tip-induced band bending occurs off but not on the hillocks, and second, the work function is lower on the hillocks. The same amount of Cu was deposited onto a TiO{dollar}sb2{dollar}(001) surface, which also formed hillocks, and exhibited a variation in the electronic structure. The hillock clusters for the Cu-deposited surface were 100 times smaller than the Cu-segregated. Also the hillocks on the Cu-deposited surface tended to cluster. |
