The development of new types of solar cells is driven by the need for clean and sustainable energy. In this respect dye sensitized solar cells
(DSC) are considered as a promising route for departing from the traditional solid state cells. The physical insight provided by computational modeling may help develop improved DSCs. To this end it is important to obtain an accurate description of the electronic structure, including the fundamental gaps and level alignment at the dye-TiO2 interface. This requires a treatment beyond ground-state density functional theory (DFT). In this talk I will present a many-body perturbation theory study, within the G0W0 approximation, of TiO2 clusters, dye molecules, and dye-sensitized TiO2 clusters. I will show how the combination of DFT-based basin hopping with G0W0 calculations enables identifying the isomers of (TiO2)2-10 clusters that are observed in photoemission experiments. I will explain the mechanism of selection for clusters with a high electron affinity, rather than the most stable isomers. In addition, I will discuss some of the issues pertaining to
G0W0 calculations, namely: (i) dependence on the mean field starting point and (ii) the validity of the assumption that the DFT wave-function is a good approximation to the quasi-particle wave-function. I will show how these issues are manifested for dye molecules and for dye-sensitized
TiO2 clusters.
