Structure and dynamics of the semirigid HHe2+ and the fluxional HHe3+ cations
The HHen+ cationic complexes consist of a strongly-bound linear HHe2+ chromophore surrounded by weakly-bound solvating He atoms. The structures and the vibrational dynamics of the HHen+ (n=3-6) complexes have been previously investigated using action-spectroscopic experiments in a cryogenic ion-trap machine and high-level electronic-structure computations [1-3]. In these previous studies, the anharmonic frequencies for the vibrations of the chromophore have been computed using perturbation theory, resulting in a reasonable agreement with experiment. Understanding the large-amplitude motions of the weakly-bound solvating He atoms requires variational nuclear-motion computations. Based on neural-network potential energy surfaces (NN-PES) [4-5] developed as part of this study, we determined rovibrational energies of HHe2+ and HHe3+ using the GENIUSH code [6-8]. The training set of the NN-PES consist of CCSD(T*)-F12b/AVQZ computations for configurations sampled within an active learning framework from path integral molecular dynamics simulations [5]. For HHe2+, the computed rovibrational transitions corresponding to the asymmetric stretch show outstanding agreement with experiment, demonstrating the excellent quality of the NN-PES. In the case of HHe3+, the third, solvating He orbits around the HHe2+ unit forming a torus. We studied the shape of the fluxional HHe3+ cation and its vibrations involving the solvating He atom, both with and without the consideration of the permutational symmetry of the He atoms. We identified the bending and stretching modes of the solvating He atom, as well as states corresponding to a linear secondary minimum structure. Visualization of the vibrations and the temperature-dependent dynamical structure of the cation is helped considerably by computing and plotting the nuclear density.
[1] Attila G. Császár, Tamás Szidarovszky, Oskar Asvany, Stephan Schlemmer, Molecular Physics, 2019, 117, 9-12, 1559-1583.
[2] Oskar Asvany, Stephan Schlemmer, Tamás Szidarovszky, Attila G. Császár, Journal of Physical Chemistry Letters, 2019, 10, 5325-5330.
[3] Matthias Töpfer, Anders Jensen, Keigo Nagamori, Hiroshi Kohguchi, Tamás Szidarovszky, Attila G. Császár, Stephan Schlemmer, Oskar Asvany, Physical Chemistry Chemical Physics, 2020, 22, 22885-22888.
[4] Fabien Brieuc, Christoph Schran, Harald Forbert, Dominik Marx, RubNNet4MD: Ruhr-Universität Bochum Neural Networks for Molecular Dynamics Software Package Version~1, https://www.theochem.rub.de/go/rubnnet4md.html
[5] Fabien Brieuc, Christoph Schran, Felix Uhl, Harald Forbert, Dominik Marx, Journal of Chemical Physics, 2020, 152, 210901.
[6] Edit Mátyus, Gábor Czakó, Attila G. Császár, Journal of Chemical Physics, 2009, 130, 134112.
[7] Csaba Fábri, Edit Mátyus, Attila G. Császár, Journal of Chemical Physics, 2011, 134, 074105.
[8] Csaba Fábri, Martin Quack, Attila G. Császár, Journal of Chemical Physics, 2017, 147, 134101.