SAMOP 2021 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
MO: Fachverband Molekülphysik
MO 6: Ultrafast
MO 6.3: Talk
Thursday, September 23, 2021, 14:30–14:45, H2
Unraveling collective coordinates influencing time-resolved x-ray absorption spectra of ionized urea and its dimer through machine learning — •Yashoj Shakya1,2, Ludger Inhester1, Caroline Arnold1,2,3, Ralph Welsch1,3, and Robin Santra1,2,3 — 1Center for Free-Electron Laser Science, DESY, Hamburg, Germany — 2Department of Physics, Universität Hamburg, Hamburg, Germany — 3Hamburg Centre for Ultrafast Imaging, Hamburg, Germany
Probing the early dynamics of chemical systems following ionization is essential for our understanding of radiation damage. Time-resolved x-ray absorption spectroscopy (TRXAS) on a femtosecond timescale can provide crucial insights into the ultrafast processes occuring upon ionization due to its element-specificity. However, even with theoretical simulations, getting a clear interpretation of the spectra can be challenging due to the high dimensionality of the data.
In this theoretical study, we investigate the response of urea and its dimer to ionizing radiation and how it can be probed via TRXAS. We show how statistical analysis techniques, in particular partial least square regression, can unravel specific structural dynamics in a molecule that induce changes in its TRXAS. By applying this technique, collective coordinates that most influence TRXAS are obtained from simulated ab initio nonadiabatic trajectories of valence-ionized urea and its dimer. For urea, this leads to the possibility of tracing specific molecular vibrations in its TRXAS. For its dimer, where ionization triggers a proton transfer, we show how the spectra can reveal the progress of the transfer.