Parallel Session: Theory, Contributed Talk (15min)

CC6

Energy-Corrected Sudden approach to the non-Markovian relaxation matrix for two linear colliders

A finer picture of collisional effects on spectral band shapes is urgently demanded by a number of atmospheric and combustion explorations [1]. For that, simulations of band profiles in large spectral intervals should be done at various thermodynamic conditions, in particular for elevated gas densities where the line-mixing effects are strongly pronounced. The problem is solved when the fundamental, frequency-dependent relaxation (super)matrix Γ(ω) is known. Presently, the only affordable receipt to calculate Γ(ω) without limitations of the perturbation theory is due to the Energy- and Frequency Corrected Sudden Approximation (EFCSA) model developed initially for the structureless bath [2] and extended recently to linear perturbers [3], assuming collision durations to be finite, yet much shorter than the period of collider's rotation.

In a previous study [4], we suggested an approach based on spectral moments to model the translational interaction spectral functions (ISF). This approach requires, however, refined potential energy surfaces available only for a limited number of molecular systems. To make non-Markovian calculations feasible for an arbitrary molecular pair, we develop here a semi-empirical approach to ISF simulation based on the Energy-Corrected Sudden model. While there exist papers presenting ECS-modeling of the Markovian relaxation matrix for a linear perturber [5] or the non-Markovian ECS matrix for a structureless perturber [6], to our knowledge, there is no published work describing non-Markovian relaxation matrices that accounts for the anisotropy of the perturbing molecule.

After having applied the basic hypotheses of the ECS approximation and having factorized the ISFs into frequency- and anisotropy-dependent parts, we consider the particular case of the isotropic Q-branch, which allows identification of the frequency-dependent factor with the adiabaticity factor and enables relating the anisotropy-dependent factor to the bimolecular transition rates. We propose various analytical multi-parameter models for bimolecular basic transition rates that are used, together with a Lorentzian-type adiabaticity factor, for ISF modeling and are shown, on the example of high-pressure anisotropic Raman spectra of nitrogen, to be suitable for theoretical prediction of large-band spectra of dense gases of linear molecules.

AS and AK acknowledge the financial support from the RFBR (project number 19-33-90244).

[1] Jean-Michel Hartmann, Christian Boulet, Daniel Robert, *Collisional Effects on Molecular Spectra,* Elsevier, Amsterdam, **2008**.

[2] Alexandre P. Kouzov, *Physical Review A*, **1999**, 60, 2931.

[3] Alexander P. Kouzov, Jeanna V. Buldyreva, Andrei V. Sokolov, *Journal of Chemical Physics, ***2018**, 149, 044305.

[4] Andrei Sokolov, Alexander Kouzov, Jeanna Buldyreva, *Journal of Raman Spectroscopy*, **2020**, 51, 2053.

[5] Sheldon Green, *Journal of Chemical Physics, 1993, *98, 257.

[6] Jeanna V. Buldyreva, Lionel Bonamy, *Physical Review A*, **1999**, 60, 370.