UC Berkeley

Harnessing the energy in photoexcitations requires efficient and controlled charge separation whether to drive a photochemical reaction or to create electricity for a photovoltaic device. Semiconducting nanocrystals, or quantum dots, have shown particular promise as light absorbers for these applications. They possess large extinction coefficients, size-tunable band gaps, and an accessible and functionalizable surface. The accessibility of the surface not only enables efficient charge extraction, but it also allows for undesirable charge trapping on ill-defined surface states. This dissertation will explore the mechanism of charge transfer from photoexcited quantum dots in order to better understand both charge extraction and trapping. It is specifically focused on hole transfer by using hole-accepting moieties since this process tends to limit the efficiency of quantum dot based devices and is understudied compared to electron transfer.

A model system for studying hole transfer is presented that uses cadmium selenide/cadmium sulfide core/shell nanocrystals or quantum dots with high photoluminescence quantum yields, which therefore selects dots with relatively few traps. The surface of these quantum dots are functionalized with ferrocene-derived hole-accepting molecules and the effect of this functionalization on the photoluminescence is used to determine the rate constant for hole transfer. The rates are found to exponentially decay with increasing CdS shell thickness or increasing linker length between the ferrocene and the quantum dot surface. This is well modeled by a tunneling process.

The relationship between driving force and rate in this system was investigated by using six distinct ferrocene acceptors with a range in driving force of 800 meV. The resulting relationship between rate and energetic driving force for hole transfer is not well modeled by the standard two-state Marcus model. Alternative mechanisms for charge transfer are posited, including an Auger-assisted mechanism that provides a successful fit to the results. The observed relationship can be used to design QD-molecular systems that maximize interfacial charge transfer rates while minimizing energetic losses associated with the driving force.

The temperature dependence of the hole transfer rate in this model system is also explored. The observed Arrhenius slopes for these rates do not depend on driving force, which suggests that surface state mediated hole transfer may dominate direct hole transfer. A model for trap-mediated transfer via shallow and reversible hole traps is posited and matches well with the data.

Finally, the results presented within are used to inform the creation of a generalizable model for hole transfer through which previously published work on hole transfer to both engineered and native hole traps can be analyzed.