A Computational study of mixed metal oxides

In this thesis, we investigated using computational methods two types of mixed metal oxides: tetragonal tungsten bronzes (TTB) of interest for thermoelectric applications and Ni, Mn, Co (NMC) oxides with a layered structure that are used as cathodes in rechargeable Li-ion batteries. An atomic level understanding of the defect chemistry, electrochemical behavior and crystal structures of the materials is provided. In the first part, we investigated the feasibility of inducing disorder by doping in materials with the TTB structure. Disorder has been successfully induced by doping different metal sites, and a number of synthesizable compositions with high level of cation disorder are found. We found that the ionic size determines the positions of cations in TTB structure. The most stable octahedra to accommodate dopants is determined by both ionic size and charge. In the second part of the thesis, we studied the layered cathodes LiXO2 (X=Ni, Co, Mn, Zn and Zr) for Li-ion batteries. This study provides a systematic understanding of the effect of compositional variations and defect chemistry in the layered structure. Compositional variations lead to different electrochemical behavior, and the mixing of different TMs may result in intrinsic redox reactions (e.g., Ni/Mn and Co/Mn), which are shown to influence the electrochemical properties of the layered cathodes. In LixNiO2, structural transitions and activation of lattice oxygen are observed and investigated. Our results indicate that the local environment plays the dominant role in determining the electrochemical properties of the layered cathodes, and control over the local environment is crucial in tailoring the properties of the layered cathodes.