Modern photochemistry covers a vast application space including photocatalysis, optically switchable materials, photopharmacology, optogenetics, optoelectronics and more. Despite the enormous chemical diversity of the engineered materials and samples, all these processes have in common that photons initially trigger nuclear and electronic wavepackets at defined molecular units. Understanding this initial photochemistry and its engineerability at the most fundamental level is thus a paramount goal in the design of more efficient, robust, and sustainable devices. The fate of virtually all photoinduced molecular processes is determined by conical intersections (CIs). These are regions of energetic degeneracy between electronic states in polyatomic molecules that open ultrafast, non-radiative relaxation channels that are much faster than, for example, fluorescence. Nuclei and electrons move on comparable timescales, leading to a breakdown of the Born-Oppenheimer approximation and a complex interplay of coupled wavepacket motions. CIs can therefore be regarded as critical junctions in the chemical compound space where the photochemical decision-making leading to different photoproducts takes place. Our prime objectives are to simulate (i) the quantum molecular dynamics of these events[1], (ii) their spectroscopic observables using novel X-ray pulses from free-electron lasers[2-4], and (iii) how quantum control can be used to steer their outcomes and enhance the detectability of the intricate, often elusive spectroscopic signatures[5-6]. This “marriage of three” key methods generates a variety of datasets about photochemical processes, their observables, and their controllability. I will provide insight into these datasets, in the hope that presentation to this community can improve their standardization in generation, prominence in other communities, and their usability by data science methods.
Daniel Keefer