Ultrafast X-Ray Absorption Spectroscopy as a Probe of Conical Intersection-Mediated Dynamics: Theoretical Tools and Studies

Ultrafast time-resolved X-ray absporption spectroscopy (TRXAS) has emerged as a powerful tool for probing excited-state non-adiabatic molecular dynamics. Of particular interest is the sensitivity of TRXAS to the changes in valence electron density that occur as an excited-state wavepacket approaches a region of conical intersection between electronic states. In particular, the mixing of state characters in these high-coupling regions can be expected to modulate the amount of localisation of the valence electron density around the individual atomic centres, resulting in shifts in the core-orbital energies. Resultantly, shifts in the peaks in the TRXAS are expected to occur, which may be directly related to the non-adiabatic nuclear dynamics.

As the emergence of new light sources make excited-state TRXAS studies a reality, there is a need for accompanying theoretical simulations to enable a complete characterisation and understanding of the recorded spectra. In particular, rigorous, general, and tractable methods for the simulation of excited-state X-ray absorption spectra (XAS) are needed. Up until now, however, there has been a dearth of methods that can treat initial states of arbitrary character, while maintaining a low cost and black box usage. To address this issue, we have developed a new theoretical framework employing the core-valence separated combined DFT/MRCI method (CVS-DFT/MRCI). This is a method that is, perhaps, uniquely suited to the simulation of XAS for large molecules and initial states of multi-reference character, which are encountered in the vicinity of conical intersections.

In this talk, I will present recent studies using CVS-DFT/MRCI for the exploration of the sensitivity of TRXAS to conical-intersection-mediated dynamics in a range of molecules, ranging from the disentangling of competing relaxation pathways in allene to the Jahn-Teller effect-mediated dynamics in the graphene nanoflake coronene.