Michael McGehee, Shawn Scully, and Melissa Summers. Materials Science and Engineering, Stanford University, Building 550, Stanford, CA 94305-2205
Almost all organic photovoltaic cells are based on either planar or bulk heterojunctions of two semiconductors. After light is absorbed, excitons must get to the interface between the two semiconductors to dissociate by electron transfer. In some cases, such as in dye-sensitized cells or polymer-fullerene bulk heterojunctions with very high fullerene concentrations, excitons are formed right at the interface and exciton transport is therefore not a limiting factor on the performance of the cells. In many other cases, such as in polymer-nanowire or polymer-titania cells, excitons need to travel at least five nanometers, if not more. For this reason exciton diffusion is a very important process to understand and optimize. We have carefully measured the exciton diffusion length in several polymers and found that the values are less than reported in the literature. Common sources of error in diffusion length measurements are neglecting interdiffusion between the donor and acceptor, interference effects and resonance energy transfer. Since the diffusion length in most polymers is 6 nm or less, we have explored ways to enhance exciton transport. One is to use resonance energy transfer from a donor to an acceptor with a slightly smaller energy gap. We will show an example where the effective diffusion length and consequently the efficiency of the photovoltaic cell is enhanced by a factor of three. We will also show that resonance energy transfer occurs in many previously studied donor-acceptor blends, including polymer-fullerene blends with low fullerene concentrations. Finally we will discuss how high the effective diffusion length could be if the donor-acceptor pair were optimized for resonance energy transfer.