Doctor of Philosophy (PhD)


Physics and Astronomy

Document Type



Metal-nonmetal transition, or more specifically, metal-insulator transition (MIT) has been one of the most intriguing topics in condensed matter physics. Two theories describing fundamental driving mechanisms of MIT has been well-established over time: Mott-Hubbard theory and Anderson localization theory. The former mainly deals with contribution of electron interactions/correlations to the MIT, and the latter focuses on the role of disorder. However, it is an open topic how a system behaves when both effects exist in a system. This study mostly takes interest in a type of MIT induced by dimensionality-crossover in the transition metal oxides (TMOs) systems. TMO system is a perfect playground for studying the underlying mechanisms behind MIT due to wide presence of strong electron interactions in the d-band, and the existence of oxygen vacancies as an unavoidable form of disorder. The focus of this study is mainly on the role played by disorder. A metallic TMO system SrVO3 (SVO) was chosen to perform the investigation due to its simple structure and lack of magnetic ordering. Well-ordered SVO thin films have been fabricated in a layer-by-layer fashion on crystalline SrTiO3 (001) substrates. Surface structural characterization and morphology images suggest that the SVO films are of high quality with correct symmetry and atomically flat surfaces. The structural and chemical composition characterization indicates the existence of a significant amount of oxygen vacancies in the first three layers of the SVO films, coinciding with the critical thickness for the MIT, which has been confirmed by spectroscopic analysis which reveals zero density of states at the Fermi level for films with thickness below 3 unit cell (u.c.). Transport measurements reveal weakly localized lnT behavior for metallic SVO films close to the critical thickness, agreeing with the picture of a 2D disordered correlated system. Negative magnetoresistance observed in the weakly localized films is consistent with the prediction that disorder dominates over correlation effects. Moreover, by deliberately introducing more disorder into metallic SVO films, MIT can also be induced. Through our research, we conclude that the disorder effect is the major driving mechanism for MIT.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Zhang, Jiandi