Modelling Current-Voltage Characteristics in Organic Photovoltaics
MetadataShow full item record
Thin film organic photovoltaics have the potential to meet the world’s growing energy needs cheaply and cleanly. However, organic photovoltaics have not yet been able to match the efficiencies of conventional silicon solar devices. The shape of the current-voltage (IV) curve provides information about where these efficiency shortfalls are occurring in a device. For instance, different regions of the curve can indicate whether the device is leaking current, whether shunting is present, and how far the device deviates from the ideal. However, extracting all of this information requires a model to fit the IV curve. Researchers have found that IV curves for these organic devices can be modelled by a power law. The higher the power law, the steeper the slope of the curve at the turn on voltage and the more the device behaves like an ideal diode. However, the origin of this power law value has been attributed to many factors, including injection, transport, or screening of charge. In this study, we propose a model for the IV curve of organic photovoltaics consisting of a bilayer heterojunction between an electron donor and acceptor. The model is based on the theory of space-charge-limited currents with energy traps laid out by Lampert and Mark and is extended to include the recombination rate of charges at the interface between the donor and electron acceptor. The theory predicts that the recombination rate should be limited by the difference in energy between the highest occupied molecular orbital (HOMO) of the donor and the lowest unoccupied molecular orbital (LUMO) of the acceptor. Devices fabricated using the same donor (SubPC) and a variety of acceptors with different LUMO levels yield a power law directly proportional to the HOMO-LUMO energy gap. The model is capable of fitting the full extent of the IV curve exactly, giving diagnostic information about leakage, shunting, transport, and recombination in existing devices. We anticipate that the model can also serve a predictive role based solely on intrinsic and measurable material properties.