Proceedings of 6th International Conference on Hybrid and Organic Photovoltaics (HOPV14)
Publication date: 1st March 2014
The dye-sensitized solar cell (DSSC) is a low cost and technically promising alternative to the conventional silicon-based photovoltaic device [1]. Lower efficiency levels are a major drawback but an increase of only a few percent would afford a highly competitive solar-cell technology. At the heart of the cell is a dye molecule which is responsible for light capture and electron injection. The experimental search for novel dye molecules is expensive, time consuming and often chemically biased. On the other hand, high-throughput computational screening is cheap, fast and can search a large portion of chemical space. However, there are a number of fundamental problems associated with the accurate determination of dye properties on the large-scale. First of all, the molecules are often highly flexible and possess a complex potential energy surface with multiple minima. A so called NP-hard problem, finding low-energy energy structures requires navigating this complex space. We address this problem by considering what information can be obtained from crystallographic data and what shortcuts can be made in the search for low-energy structures. In addition, there is no consensus on the optimal method for determining the electronic structure of molecules, with particular difficulty for modelling the excited state. We consider the capabilities of DFT and TDDFT by comparison with experimental and higher-level computational results. More specifically, the alignment of ground and excited state energy levels and the calculation of the overall absorption profile. The general framework for the batch processing of molecules that combines quantum chemistry and cheminformatics tools will also be discussed.
[1] O'Regan, N.; Grätzel, M. A low-cost, high-efficiency solar cell based on dye-sensitized colloidal TiO2 films. Nature 1991, 353, 737-740.