In this thesis defects and impurities in CdTe have been studied with ab initio methods. CdTe is a II-VI semiconductor with many important applications such as γ- and X-ray detectors, solar cells and medical imaging. Even though CdTe has been studied for more than 70 years, some of its properties connected with defects and impurities, are still shrouded in mystery. Todays experimental techniques are highly developed and can provide rather detailed data, but require elaborate theoretical analysis. Here ab initio modelling comes into play and in particular density functional theory (DFT). When reviewing different theoretical studies of defects and impurities in CdTe, one finds a vast number of discrepancies between experiment and theory. Mismatches appear even between different theoretical studies. Although many problems, such as, for example, the semiconductor band gap underestimation or the spurious interaction between charged defects, are avoided by employing corrections or implementing new functionals, some of them still remain. Employing the hybrid functional HSE06, the following topics were studied in this thesis: - Te antisites: Experimental data predict the defect state to appear in the middle of the band gap, thus "pinning" the Fermi level. In contrast, our calculations show that Te antisite alone cannot be the reason for the Fermi level pinning, since it does not form a defect level in the middle of the band gap. Instead we propose that charge compensation between Te antisites in a (+2) state and Cd vacancies in a (-2) state explains the Fermi level pinning. - Cd vacancy: Electron paramagnetic resonance experiments clearly show the existence of a hole polaron for the (-1) charged vacancy. But DFT studies report a completely delocalised hole. In our studies, for the first time, this state was found in its proper geometrical configuration with a hole localisation stabilised by a Jahn-Teller distortion, thereby removing the discrepancy between experiment and theory. - Cd chalcogenides: Additionally, with particular focus on the hole localisation problem, the series of isovalent compounds (CdTe, CdSe and CdS) was studied to understand the mechanism of hole polaron formation. We explain the trend of the hole localisation in terms of Coulomb interaction, explicitly showing that the effect of electron correlation is negligible. - Cl-doped CdTe: The formation of a Cl - Cd vacancy complex explains the selfcompensation and selfpurification mechanism. We find Cl to annihilate the hole polaron. - Te antisite under deformation: In an attempt to tailor the energy position of the Te antisite defect level in the CdTe band gap, we studied CdTe under different deformations. It is shown that by a carefully chosen deformation the defect levels can be pushed closer to the valence and/or conduction band and hence the CdTe detector performance may be improved.

In this dissertation, I present the results of my research on a prototypical interface of the metal oxide ZnO and the organic semiconductor C₆₀. I establish that the physics at such oxide / organic interfaces is complex and very different from the extensively investigated case of organic semiconductor / metal interfaces. The studies presented in this dissertation were designed to address and improve the understanding of the fundamental physics at such hybrid organic / inorganic interfaces. Using photoemission spectroscopies, I show that metal oxide defect states play an important role in determining the interfacial electronic properties, such as energy level alignment and charge carrier dynamics. In particular, I show that for hybrid interfaces, electronic phenomena are sensitive to the surface electronic structure of the inorganic semiconductor. I also demonstrate applications of photoemission spectroscopies which are unique in that they allow for a direct comparison of ultrafast charge carrier dynamics at the interface and the electronic structure of defect levels. The research presented here focuses on a achieving a significant understanding of the realistic and device relevant C₆₀ / ZnO hybrid interface. I show how the complex surface structure of ZnO can be modified by simple experimental protocols, with direct and dramatic consequences on the interfacial energy level alignment, carrier dynamics and carrier collection and injection efficiencies. As a result of this careful study of the electronic structure and dynamics at the C₆₀ / ZnO interface, a greater understanding of the role of gap states in interface hybridization and charge carrier localization is obtained. This dissertation constitutes a first step in achieving a fundamental understanding of hybrid interfacial electronic properties.

A study of defects in semiconductors is vital as defects tend to influence device operation by modifying their electrical and optoelectronic properties. This influence can at times be desirable in the case of fast switching devices and sometimes undesirable as they may reduce the efficiency of optoelectronic devices. ZnO is a wide bandgap material with a potential for fabricating UV light emitting diodes, lasers and white lighting devices only after the realization of reproducible p-type material. The realization of p-type material is greatly affected by doping asymmetry. The self-compensation behaviour by its native defects has hindered the success in obtaining the p-type material. Hence there is need to understand the electronic properties, formation and annealing-out of these defects for controlled material doping. Space charge spectroscopic techniques are powerful tools for studying the electronic properties of electrically active defects in semiconductors since they can reveal information about the defect “signatures”. In this study, novel Schottky contacts with low leakage currents of the order of 10-11 A at 2.0 V, barrier heights of 0.60 – 0.80 eV and low series resistance, fabricated on hydrogen peroxide treated melt-grown single crystal ZnO samples, were demonstrated. Investigations on the dependence of the Schottky contact parameters on fabrication techniques and different metals were performed. Resistive evaporation proved to produce contacts with lower series resistance, higher barrier heights and low reverse currents compared to the electron-beam deposition technique. Deep level transient spectroscopy (DLTS) and Laplace-DLTS have been employed to study the electronic properties of electrically active deep level defects in ZnO. Results revealed the presence of three prominent deep level defects (E1, E2 and E3) in the as-received ZnO samples. Electron-beam deposited contacts indicated the presence of the E1, E2 and E3 and the introduction of new deep level defects. These induced deep levels have been attributed to stray electrons and ionized particles, present in the deposition system during contact fabrication. Exposure of ZnO to high temperatures induces deep level defects. Annealing samples in the 300°C – 600°C temperature range in Ar + O2 induces the E4 deep level with a very high capture cross-section. This deep level transforms at every annealing temperature. Its instability at room temperature has been demonstrated by a change in the peak temperature position with time. This deep level was broad, indicating that it consists of two or more closely spaced energy levels. Laplace-DLTS was successfully employed to resolve the closely spaced energy levels. Annealing samples at 700°C in Ar and O2 anneals-out E4 and induces the Ex deep level defect with an activation enthalpy of approximately 160 – 180 meV. Vacuum annealing performed in the 400°C – 700°C temperature range did not induce any deep level defects. Since the radiation hardness of ZnO is crucial in space applications, 1.6 MeV proton irradiation was performed. DLTS revealed the introduction of the E4 deep level with an activation enthalpy of approximately 530 meV, which proved to be stable at room temperature and atmospheric pressure since its properties didn’t change over a period of 12 months. / Thesis (PhD)--University of Pretoria, 2013. / Physics / unrestricted