A thesis submitted to the Faculty of Science, in fulfilment of the requirements for the degree of Doctor of Philosophy. University of the Witwatersrand, Johannesburg May 23, 2016 / Amongst the major technological challenges of the twenty rst century is the harvesting of renewable energy sources. We studied the solar cell performance of the ternary compounds AgAlX2 (X = S, Se and Te) and AgInS2 as promising materials for meeting this challenge. Structural, electronic and optical properties of the compounds were investigated by means of the density functional theory and many body perturbation theory. Using cohesive energy and enthalpy, we found that among six potential phases of AgAlX2 and AgInS2, the chalcopyrite and the orthorhombic structures were very competitive as zero pressure phases. We predicted a low pressure-induced phase transition from the chalcopyrite phase to a rhombohedral phase. For the chalcopyrite phase, we found that the tetragonal distortion and anion displacement were the cause of the crystal eld splitting. The bandgaps from the general gradient approximation PBEsol were underestimated when compared to experiment and accurate bandgaps were obtained from the hybrid functioanl HSE06, the meta-general gradient approximation MBJ and GW approximation. Optical absorption from the Bethe-Selpeter equation indicated the presence of bound exciton in AgAlX2. We estimated the solar cell performance of the compounds using the Shockley and Queisser model and the spectroscopy limited maximum e ciency approach. We found that apart from AgAlS2, the estimated theoretical e ciency of the other compounds was greater that 13 %.

Hybrid lead halide perovskites, AMX 3 compounds in which A = CH 3 NH 3 (MA), CH(NH 2 ) 2(FA), Cs; M = Pb,Sn; X = I, Br, Cl, display remarkable performance in solution-processed optoelectronic devices, including > 22% efficient thin film photovoltaic cells. These compounds represent the first class of materials discovered to exhibit properties associated with high performance compound semiconductors, while being formed at or near room temperature using simple solution chemistry techniques. This thesis is focused on the synthesis, structural characterisation and phase behaviour of MAPbI 3 , FAPbI 3 , A-site solid solutions and novel organic metal halide framework materials. The complete atomic structure and phase behaviour of methylammonium lead iodide is elucidated for the first time, including hydrogen positions, using high flux, constant wave-length neutron powder diffraction. At 100 K an orthorhombic phase, space group Pnma, is observed, with the methylammonium cations ordered as the C–N bond direction alternates in adjacent inorganic cages. Above 165 K a first order phase transition to tetragonal, I4/mcm, occurs with the unlocking of cation rotation, which is disordered primarily in the ab plane. Above 327 K a cubic phase, space group Pm3m, is formed, with the cations isotropically disordered on the timescale of the crystallographic experiment. The high temperature phase of formamidinium lead iodide, α-FAPbI 3 is shown for the first time to be cubic, (Pm3m), at room temperature using time-of-flight, high resolution neutron powder diffraction. Polymorphism and the low temperature phase behaviour of FAPbI 3 have been further investigated using reactor and spallation neutron sources with high resolution in temperature. A tetragonal phase, P4/mbm, is confirmed in the temperature range 140-285 K.The composition, structural and optical parameters of ’A’ site solid solutions (MA/FA)PbI 3 have been investigated by single crystal X-ray diffraction, UV-vis spectroscopy and 1 H solution NMR. A composition-dependent transition in the crystal class from tetragonal to cubic(or pseudo-cubic) at room temperature is identified and correlated to trends in the optical absorption. Novel hybrid materials with inorganic frameworks of varying dimensionality from 0D to 2D, including imidazolium lead iodide and piperazinium lead iodide, have been synthesised using various templating organic cations and their atomic structures solved by single crystal X-ray diffraction.

Two main subjects are addressed in this study. The ability of a bright red material with metallic behaviour to be used as a visible light photocatalyst for hydrogen evolution and the feasibility of visible light photocatalytic water splitting using Z-schemes constituted from different kinds of photocatalysts and materials used as mediators. Strontium niobate (Sr₁₋ₓNbO₃) is an A-site deficient perovskite with intense red colour. It is an unusual material that displays both metallic type conduction and- as we present- photocatalytic activity. Specifically, photocatalytic visible light hydrogen production with oxalic acid as a sacrificial reagent is achieved from this material even without the need for a co-catalyst or other alteration. This photocatalytic activity is screened with time and related to different parameters that might influence it, like crystal structure, surface area and surface chemistry. The crystal structure of strontium niobate is A site stoichiometry dependant and the materials acquires a cubic symmetry for Sr≤ 0.92 and orthorhombic for 0.92≤ Sr≤ 0.97. The change of crystal structure from cubic to orthorhombic symmetry seems to have a negative effect on the photocatalytic activity, as the NbO₆ octahedra become distorted and unfavourable for d-orbital overlapping. The highest photocatalytic activity is exhibited at the turning point of one structure to the other. Increase in the photocatalytic activity is also exhibited by enlarging the surface area through ball milling, nevertheless, a clear trend for surface area effect on activity is not obtained among samples with different Sr content. Additionally, an enrichment of Sr on the surface of strontium niobate is observed by XPS, which apart from the fact that seems to be a governing factor improving stability it is also considered a key point for the exhibited photocatalytic activity altogether. Full water splitting under visible light from Z-schemes is studied by fabricating three general categories of systems. These three different categories depend on the mediator used to fabricate the Z-schemes and are: redox couple Z-schemes (with Fe⁺³/Fe⁺²), solid mediator Z-schemes (with GO) and no mediator Z-schemes. The materials used as photocatalysts for the fabrication of the Z-schemes are: Sr₀.₉₂NbO₃ for hydrogen production and both WO₃ and BiVO₄ independently for oxygen production. The photocatalytic activity for water splitting is evaluated in production of hydrogen and oxygen with time and the ratio of their production rates is frequently checked to see whether the ideal hydrogen to oxygen 2:1 is achieved. The general idea acquired from the results of all the three types of systems is that, water splitting with Z-schemes is a complicated process and in most cases governed by many subreactions. More specifically, in all cases of redox couple Z-schemes we got hydrogen to oxygen ratio imbalances and with the most prominent one being the lack of hydrogen production. Thankful is the fact that a certain type of system, the one consisting of WO₃ as oxygen photocatalyst and Fe⁺² as initial mediator species gives results very close to the ideal one and with a high degree of reproducibility indicating this way the probable formation of a Z-scheme that has overcome more of the imbalances. In between the two other categories, solid mediator and no mediator Z-schemes, subreactions seem to be the governing factor hence imbalances are always present. A case study in the no mediator Z-schemes on an attempt to investigate sources of imbalances, reveals that a big source of imbalance is most probably from the trapping of protons from WO₃.