Ab initio study of electronic and optical properties of transparent conducting oxides

A variety of industrial applications such as solar cells, touch screen displays and LED require materials combining the two antagonistic properties of transparency to the visible light and high conductivity. In transparent conducting oxides (TCOs), the transparency is achieved thanks to the wide band gap (> 3 eV) of the host oxide, while the high conductivity results from the presence of charge carriers due to intrinsic defects and doping. n-type TCOs (mostly based on In2O3, SnO2 and ZnO) are already present in many devices. In contrast, current p-type TCOs are not largely commercialized as they exhibit much lower carrier mobilities. New applications (e.g for the transistors industry) would be possible if better performing p-type TCOs could be found. TCOs have been and are still extensively studied with both experimental and theoretical approaches. In this work we apply ab initio methods to investigate the electronic structure (band gap, charge carrier effective masses) and compute optical spectra of selected TCO hosts, in particular SnO2, as well as new candidates p-type TCOs. Calculations are based on different approaches: (i) density functional theory, which is accurate for the description of ground state properties, but fails to describe excited states (since it underestimates the band gap) (ii) many-body perturbation theory in the GW approximation, which is a computationally expensive approach but allows for an accurate description of the excited states. First, we probe the effects of approximating the frequency dependence of the dielectric function in GW and find that the use of different plasmon pole models can lead to significant discrepancies in the band gap. Secondly, we consider the effects of a recently proposed exchange-correlation functional (the Tran-Blaha functional) on the electronic states of tin oxides and find that it leads to improved band gaps but it strongly underestimates the band widths. We then investigate trends in the structural, electronic and optical properties of rutile SnO2 under different types of compressive stress. In the last part of the thesis work we focus on the calculation of the GW band gap of candidates TCOs, selected through a high-throughput computational screening based on the Materials Project database, and contribute to the identification of some new potential p-type TCOs.(FSA 3) -- UCL, 201