Modeling Excited States and Alignment of Energy Levels in Dye-Sensitized Solar Cells: Successes, Failures, and Challenges

Theoretical and computational modeling is a powerful tool to investigate and characterize the structural, electronic, and optical properties of the main components of dye-sensitized solar cells (DSCs). In this article we focus on the description of the ground and excited state properties of both standalone and TiO2-adsorbed metallorganic and fully organic dyes, relevant to modeling the dye->semiconductor electron injection process, which is the primary charge generation step in DSCs. By reviewing previous data from our laboratory, integrated with new calculations, we wish to critically address the potential and limitations of current DFT and TDDFT computational methods to model DSCs. While ruthenium dyes are accurately described by standard DFT approaches, for highly conjugated organic dyes, characterized by strong charge transfer excited states, specifically tailored exchange-correlation functionals are needed. For ruthenium dye/semiconductor interfaces, a strategy is presented, which accurately describes the electronic and optical properties and the alignment of ground and excited state levels at the same time, allowing us to discuss the coupling and the energetics of the excited states underlying the ultrafast electron injection. For donor-acceptor organic dyes, a simple picture based on the dye lowest unoccupied molecular orbital (LUMO) broadening accounts for the different interfacial electronic coupling exhibited by dyes with different anchoring groups. The explored DFT/TDDFT methods, however, are not capable to deliver at the same time a balanced description of the dye@TiO2 excited states and of the alignment of the dye excited states with the semiconductor manifold of unoccupied states. This represents a challenge which should be addressed by next generation DFT or post-DFT methods.