"Developing methods for the accurate and efficient simulation of photochemistry"
The ability to address matter with light to affect desired changes presents opportunities across chemistry, biology, and materials science. Creating new light-driven molecules and materials, or improving existing ones, requires design principles informed by atomistic insight. This insight is difficult to obtain due to the complicated physics that arise upon the absorption of light, which can depend critically upon mechanistic details in the ultrafast (sub-picosecond) regime and on small structural details, necessitating very high temporal and spatial resolution. Experimentally, these processes can be interrogated on ultrafast timescales by time-resolved pump-probe techniques, but their interpretation can be difficult. Theoretical approaches to simulating nonadiabatic dynamics have achieved significant postdictive success but can struggle to perform in a predictive context, particularly due to the sensitivity of dynamical observables to the treatment of the underlying electronic structure. A prerequisite for rational design is routine prediction, and my research has therefore focused on developing methods to describe excited electronic states efficiently and accurately for nonadiabatic dynamics and applying these methods to simulate photodynamics of small molecules. I will highlight the ultrafast dynamics of cyclobutanone as an example demonstrating method development in electronic structure theory and application to simulate time-resolved experimental observables, in this case ultrafast electron diffraction (UED). This work helps establish the footing needed to move to larger timescales and longer length-scales while maintaining sufficient accuracy to make quantitative predictions.