Advancing ab initio electronic structure theory and computational methods for excited state molecular dynamics
Abstract/Contents
- Abstract
- Photochemistry studies chemical reactions caused by absorption of light. Developing theoretical and computational tools for photochemistry will not only help better understand photochemical processes such as photosynthesis and vision, but can also provide guidelines about how molecular photodevices can be better designed. Therefore, the goal of my graduate research is to develop a set of computational tools for studying photochemical processes. Physical systems have a hierarchical structure, i.e. basic particles like nuclei and electrons interact leading to the formation of molecules, and molecules interact and change conformations giving rise to chemical reactions. Naturally, the corresponding theoretical methods should also follow this hierarchy. At the bottom level, we need molecular integrals to describe different types of interactions between basic particles. I introduced the automated code engine (ACE) that generates optimized codes for computing integrals on the graphical processing units, and developed several variants of tensor hyper-contraction (THC) approximations. ACE reduces the computational prefactor of integral evaluations whereas THC reduces the formal scaling. On top of the integrals, we then need electronic structure methods to describe the energies and forces for a molecule at any given nuclear configuration; including electron correlation is the key to having an accurate description. Here, I first developed single reference THC-MP2 to capture the dynamic correlations, and then developed multi-reference THC-CASPT2 method to incorporate static correlations simultaneously. These methods were later generalized to THC-MSPT2 to enable descriptions for excited states and conical intersections, both are critical for photochemistry. Finally, given the electronic structure methods, we then need methods to explore the potential energy surfaces. In particular, critical point search methods locate the important configurations (e.g. Franck-Condon point, conical intersections), while molecular dynamics methods generate trajectories describing how the molecules move and interact with each other. By interfacing the electronic structure methods that I developed with the geomeTRIC geometry optimizer and G-AIMS non-adiabatic dynamics framework, a complete toolbox for understanding photochemistry is provided.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2018; ©2018 |
| Publication date | 2018; 2018 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Song, Chenchen |
|---|---|
| Degree supervisor | Martinez, Todd J. (Todd Joseph), 1968- |
| Thesis advisor | Martinez, Todd J. (Todd Joseph), 1968- |
| Thesis advisor | Markland, Thomas E |
| Thesis advisor | Solomon, Edward I |
| Degree committee member | Markland, Thomas E |
| Degree committee member | Solomon, Edward I |
| Associated with | Stanford University, Department of Chemistry. |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Chenchen Song. |
|---|---|
| Note | Submitted to the Department of Chemistry. |
| Thesis | Thesis Ph.D. Stanford University 2018. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2018 by Chenchen Song
- License
- This work is licensed under a Creative Commons Attribution Non Commercial No Derivatives 3.0 Unported license (CC BY-NC-ND).
Also listed in
Loading usage metrics...