Norell, Jesper: Fingerprints of light-induced molecular transients - from quantum chemical models of ultrafast x-ray spectroscopy. , Stockholm University, 2020
http://su.diva-portal.org/smash/record.jsf?pid=diva2%3A1412962&dswid=-4792
https://www.diva-portal.org/smash/get/diva2:1412962/FULLTEXT01.pdf
Open Access Version
Abstract:
Abstract Absorption of sunlight generates renewable electricity and powers the growth of plants, but also causes severe damage both to synthetic materials and biological tissue. The wildly varying outcomes of these light-induced processes are ultimately determined by much slighter differences in their underlying reaction pathways, induced by the transient properties of shortlived and miniscule molecules; a powerful approach to their detection and characterization is offered by ultrafast x-ray spectroscopy, with identification of spectral fingerprints and further guidance from quantum chemical models. This thesis contains the computational half of three experimentally joint projects that push the limits for detection of electronic, spin and structural dynamics of small molecular systems in solution. A wide selection of theoretical frameworks are combined to model various aspects of the measurements: from multi-configurational descriptions of non-adiabatic couplings in the photo-dynamics and multi-electron transitions in the x-ray spectroscopy, to affordable simulations of extensive aqueous solutions by density functional theory and classical mechanics. Applied to experimental data, the presented quantum chemical results allowed in particular to: simultaneously identify molecular forms and electronic states of aqueous 2-thiopyridone, to determine a detailed pathway for its excited-state proton-transfer; characterize the charge-transfer state of aqueous ferricyanide, to extend well-known concepts from steadystate spectroscopy into the ultrafast domain; establish the newly implemented framework of multi-configurational Dyson orbitals, as a powerful tool for simulation of photoelectron spectroscopy. A number of computational predictions are additionally presented for hitherto-unexplored experimental regions, which may help to guide and optimize future measurements.