The field of theoretical chemistry has provided undeniably useful insights about molecular systems that otherwise, through experiment, would not be obtainable. We are constantly developing new and improved methods to fill in the gaps about how various factors including the electronic structure can affect the chemistry seen experimentally. The goal of most quantum chemistry methods is to develop a method that is widely applicable, has low computational costs, but with as much accuracy as possible. Some of the most challenging systems in our field include those that are considered strongly correlated. Strong correlation is usually referring to the need for a large number of configurations to properly model the chemistry. These systems can not be solved exactly, thus various approximations must be made. A set of methods that take advantage of truncating only the unimportant configurations to solve these challenging systems are selected configuration interaction methods. Even though these selected CI methods can often provide accurate results, their general application is limited by memory bottlenecks. In 2020, our group developed the Tensor Product Selected Configuration Interaction (TPSCI) method to overcome these memory bottlenecks. We take advantage of the local character of these strongly correlated systems by doing a change of basis into tensor products, then do a selected CI algorithm in that basis.
In this dissertation, we discuss how we have extended TPSCI to compute excited states. We first test on a set of polycyclic aromatic hydrocarbons that were previously studied with TPSCI. We find very high accuracy and dimension reduction as compared to state of the art selected CI approaches. We then validate TPSCI's ability to study the electronic structure involved in the singlet fission process in tetracene tetramer with extending analysis using a Bloch effective Hamiltonian. This effective Hamiltonian allows for intuitive analysis of the singlet fission process. We also show how accurate and interpretable TPSCI can be on an open-shell biradical transition bimetallic complex, in addition to, hexabenzocoronene that is not straightforward clustering due to the conjugated benzene rings.
To alleviate the previous system size limitations, we recently implemented a Restricted Active Space Configuration Interaction as a local solver for clusters. We present novel results of using this new solver on a tetracene dimer. We comment on specific coupling strengths and show the electronic dynamics of our TPSCI effective Hamiltonian which support a CT-mediated mechanism for the tetracene dimer singlet fission. / Doctor of Philosophy / The field of theoretical chemistry has used some of the fundamental principles in quantum mechanics to study the electronic structure of molecular systems for many years. The power of computational resources has increased over the years, facilitating the increased complexity and accuracy of quantum chemistry methods. This dissertation lies in the realm of pushing past previous molecular system computational limits with rewarding accuracy and increased interpretability.
We achieve these goals by taking advantage of the localized nature in most of our chemistry vocabulary by using tensor product methods. Tensor product methods are able to separate a large problem into smaller units to overcome previous system size limitations while maintaining the desired accuracy. The main method focused on in this dissertation is a tensor product method called Tensor Product Selected Configuration Interaction (TPSCI) established by our research group in 2020.
This dissertation covers the required background information including basic terminology and previously developed methods then presents very recent and novel research using TPSCI. We first focus on extending TPSCI to excited states since excited states are extremely important for many photochemical processes, spectral analysis, and chemical sensing. We then test TPSCI on a spectrum of systems that range from very local character (separated molecular units) to bimetallics to very delocalized (carbon-based conjugation) chemistry. We find TPSCI is able to handle this variety of systems with very high accuracy and allows for very in-depth qualitative analysis. Finally, we present novel results incorporating an additional approximation in the local solver to further extend TPSCI's applicability. To test this new local solver, we focus on a process called singlet fission which is promising to help overcome solar cell efficiency limits. We are able to match previously reported results for the tetracene dimer which supports the use of TPSCI to study larger singlet fission systems in future work. With the work presented in this dissertation, we have aimed to highlight the potential utility of TPSCI, motivating further developments and research in this direction.
| Identifer | oai:union.ndltd.org:VTETD/oai:vtechworks.lib.vt.edu:10919/121340 |
| Date | 14 October 2024 |
| Creators | Braunscheidel, Nicole Mary |
| Contributors | Chemistry, Mayhall, Nicholas, Economou, Sophia Eleftherios, Crawford, Daniel, Valeyev, Eduard Faritovich |
| Publisher | Virginia Tech |
| Source Sets | Virginia Tech Theses and Dissertation |
| Language | English |
| Detected Language | English |
| Type | Dissertation |
| Format | ETD, application/pdf |
| Rights | In Copyright, http://rightsstatements.org/vocab/InC/1.0/ |
Page generated in 0.0023 seconds