The global demand for electrical power is increasing, with a greater emphasis on green energy generation with minimal environmental impact. Thermoelectric materials offer a solution to capture waste heat in several key areas, contributing to the increased energy efficiency and reduction in greenhouse gas emissions. Strontium titanate (STO) is considered a promising oxide material and a candidate for use in thermoelectric (TE) devices. However, more work is needed to improve its thermoelectric performance. Literature studies have shown that the doping of STO can improve the figure of merit, via a decrease in the thermal conductivity. The aim of this thesis is to use density functional theory methods to investigate the structural, energetic and vibrational properties of doped strontium titanate (STO). In this work, particular interest is paid towards the generation of high quality phase diagram (stability) using different levels of approximation for the inclusion of temperature effects for solid phases, and to methods that enable the approximation of thermal conductivity values which are conventionally computationally prohibitive. Chapter 1 outlines the current electrical energy generation landscape and introduces thermoelectric generators and their possible applications. This is followed by a brief introduction in to conventional and oxide based thermoelectric materials. We then highlight the underlying theory, techniques and methods used in this work (Chapter 2). The generation of phase diagrams has been investigated using different levels of theory required to accurately reproduce phase stability (e.g. quasi harmonic approximation) without the inclusion of experimental ad hoc corrections or inclusion of minimal experimental data (Chapter 3). The results showed that a variety of different levels of theory can be implemented through my development of the Surfinpy V2 code. However, the inclusion of quasi harmonic approximation for temperature effects of solid phases only resulted in small shifts in phase stability compared to the harmonic approximation. We then outline a methodology for the reproduction of experimentally observed properties of cubic STO and investigate the light p-element doping of bulk STO (Chapter 4). We achieved this through the characterisation of defect structures, mapping of energetics of complex defects through phase diagrams and approximation of the thermal conductivity values through calculation of the vibrational properties. Results showed that the doping of STO with light p-elements is not thermodynamically favourable but feasible. Additionally, the constant relaxation time approximation was successfully used to approximate changes in the thermal conductivity for defective structures. The adsorption of graphene onto the (001) surface of SrTiO3 and la-doped SrTiO3 was investigated (Chapter 5). Results showed thermodynamically favourable adsorption energies for graphene onto STO and La-doped STO with a large predicted decrease in the thermal conductivity of the composite material formed. Additionally, the interaction of defects with graphene coated STO and La-doped STO has been investigated through the use of phase diagrams. Results showed that under standard conditions the most stable surface configuration is on the phase boundary between the pristine surface and the graphene coated surfaces, and increasing oxy3gen partial pressure increases the stability of phases that have oxygen adsorbed onto the surface. Finally the conclusions and future work are presented in Chapter 6.