Energy Transport in Organic Photovoltaics

Great progress has been made since the turn of the millennium in understanding the basic physical processes which lead to charge generation in organic photovoltaic (OPV), leading
to an increase in device efficiency from the low single digits to above 12%. However, much work remains. The goal of this dissertation is to improve the understanding of the two main energy transport processes in OPV, exciton (bound electron-hole excited states optically excited in OPV) transport to the heterojunction and free charge transport out of the device.

We begin by examining the method of SR-PLQ, originally developed to measure the exciton diffusion length LD, the average distance an exciton travels in a material before
recombining, in optically thick layers. This is an important parameter in device design, as diffusion length (LD) determines the maximum thickness of planar layers. By including an
optical model, we show it is possible to extend this technique to measure LD in optically thin films.

We further extend SR-PLQ to examine the behavior of excitons at interfaces. Until recently, interfaces in devices were assumed to either be perfectly blocking (reflecting all incident excitons) or quenching (removing all incident excitons, whether through recombination or dissociation). However, recent work has shown that the common buffer material molybdenum oxide (MoOx) is partially quenching, and new fullerene:large band-gap material mixed buffers are partially blocking. We have therefore developed techniques to model and measure the blocking efficiency of an interface, the fraction of incident excitons that it blocks. Knowing the behavior of these non-ideal interfaces is vital for understanding and modeling devices.

We have also examined the behavior of interfaces between materials with identical or nearly-identical energy levels but different LD and exciton lifetime  . We derive general
methods for calculating and measuring the resulting blocking characteristic of these interfaces, and show how this can be sued to measure very small energy offsets between
materials. These principles are applied to the fullerenes C60 and C70, with C60 observed to have an energy gap 18  5 meV wider than C70.

The second energy transport mechanism in OPV, charge transport, is examined in the context of a mixed fullerene:large energy-gap material buffer. The exciton-blocking characteristics of the buffer were examined using the techniques above, and the exploration of the charge conduction mechanisms of the buffer allows for a full physical picture of how it improves OPV performance. We examine the charge transport properties of the buffer using transient photocurrent and resistivity measurements, correlated with morphological studies using conductive-tip atomic force microscopy (cAFM) and X-ray diffraction (XRD). These
show that even a small amount of fullerene in the buffer dramatically improves charge transport, enabling non-dispersive transport and improved conductivity. Interestingly, the buffer does not show results consistent with percolation theory, remaining conductive at mixing fractions where theory predicts a phase transition to insulating behavior.