Search

One of the most promising families of catalyst support materials from the point of view of durability in the PEFC are metal oxides. SnO2 can only corrode, in a strong acid, at potentials ! 1.4V, progressing through surface hydroxide formation. The bulk stability is backed up by simulated potential cycling, with minimal ESCA loss at 1.6V. The conductivity of pure SnO2, which forms the rutile crystal structure, is very small compared to C but can increase by orders of magnitude with formation of certain intrinsic defects which create electron donor levels in the band gap (which, at 3.6 eV, classes the undoped oxide as a wide band gap semiconductor). The ab-initio study evaluates the potential of extrinsic dopants to increase conductivity through adding n-type charge carriers to the system, investigating the influence of Ta on the thermodynamic stability and electronic properties through DFT calculations. Particular focus has been given to Ta, as TaO2 also forms with the rutile structure. This allows investigation of the full range of Ta doping concentration from an isolated defect to alloy by including it as a substitutional defect. A cluster expansion parameterised in terms of binary occupation variable that equal Sn or Ta is used to calculate the orderings (distributions of Ta:Sn atoms over the lattice at fixed concentration). It is found that alloys are stabilized thermodynamically by lattice distortions, with the creation of donor levels near the conduction band minimum (CBM) that with the relaxation of the lattice turn into states lying deep in the band gap that are calculated to give inadequate donation of electrons to the conduction band. This is indicative of collaborating Jahn-Teller active (JT) distortion modes of the disordered crystallite oxide which is shown through analysis of the spin density. The results suggest that alloying of metals in oxide systems is not a feasible approach to increasing conductivity with thermodynamic stability due to JT effects. The second part of this thesis explores the potential of hydrogen defects, which have been shown to form at interstitial and host oxygen sites under laboratory conditions. The same methodology is followed to establish the efficacy of hydrogen as an electron donor and its relation to other defects in SnO2. Much theoretical work has been devoted to the study of hydrogen in n-type semiconductors and experimental work supports that hydrogen can be expected to act as a conventional n-type donor in SnO2. It is found that interstitial H not only makes SnO2 metallic as recorded by the electronic structure, but further that the defect can bring about a large downward shift in the effective band gap. This also raises the possibility of activating Ta-induced deep donor levels.