Energy and information transfer via coherent exciton wave packets

Abstract

Electronic excitons are bound electron-hole states that are generated when light interacts with matter. Such excitations typically entangle with phonons and rapidly decohere; the resulting electronic state dynamics become diffusive as a result. However, if the exciton-phonon coupling can be reduced, it may be possible to construct excitonic wave packets that offer a means of efficiently transmitting information and energy. This thesis is a combined theory/computation investigation to design condensed matter systems which support the requisite coherent transport. Under the idealizing assumption that exciton-phonon entanglement could be completely suppressed, the majority of this thesis focuses on the creation and manipulation of exciton wave packets in quasi-one-dimensional systems. While each site could be a silicon quantum dot, the actual implementation focused on organic molecular assemblies for the sake of computational simplicity, ease of experimental implementation, potential for coherent transport, and promise because of reduced structural uncertainty. A laser design was derived to create exciton wave packets with tunable shape and speed. Quantum interference was then exploited to manipulate these packets to block, pass, and even dissociate excitons based on their energies. These developments allow exciton packets to be considered within the arena of quantum information science. The concept of controllable excitonic wave packets was subsequently extended to consider molecular designs that allow photons with orbital angular momentum to be absorbed to create excitons with a quasi-angular momentum of their own. It was shown that a well-defined measure of topological charge is conserved in such light-matter interactions. Significantly, it was also discovered that such molecules allow photon angular momenta to be combined and later emitted. This amounts to a new way of up/down converting photonic angular momentum without relying on nonlinear optical materials. The associated excitations were dubbed \emph{twisted excitons}. Twisted exciton packets can be manipulated as they travel down molecular chains, and this has applications in quantum information science as well. In each setting considered, exciton dynamics were initially studied using a simple tight-binding formalism. This misses the actual many-body interactions and multiple energy levels associated real systems. To remedy this, I adapted an existing time-domain Density Functional Theory code and applied it to study the dynamics of exciton wave packets on quasi-one-dimensional systems. This required the use of high-performance computing and the construction of a number of key auxiliary codes. Establishing the requisite methodology constituted a substantial part of the entire thesis. Surprisingly, this effort uncovered a computational issue associated with Rabi oscillations that had been incorrectly characterized in the literature. My research elucidated the actual problem and a solution was found. This new methodology was an integral part of the overall computational analysis. The thesis then takes up the a detailed consideration of the prospect for creating systems that support a strong measure of transport coherence. While physical implementations include molecular assemblies, solid-state superlattices, and even optical lattices, I decided to focus on assemblies of nanometer-sized silicon quantum dots. First principles computational analysis was used to quantify reorganization within individual dots and excitonic coupling between dots. Quantum dot functionalizations were identified that make it plausible to maintain a measure of excitonic coherence even at room temperatures. Attention was then turned to the use of covalently bonded bridge material to join quantum dots in a way that facilitates efficient exciton transfer. Both carbon and silicon structures were considered by considering the way in which subunits might be best brought together. This resulted in a set of design criteria which were then evaluated using first-principles, excited state analyses. It was found that efficient exciton transfer is indeed possible. When coupled to the previous quantum dot functionalizations, the notion that quantum dot materials could support partially coherent exciton wave packets was determined to be quite reasonable.