Berlin 2024 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

O: Fachverband Oberflächenphysik

O 2: Focus Session: Frontiers of Electronic-Structure Theory – Advances in Time-Dependent and Nonequilibrium Ab Initio Methods I

O 2.4: Talk

Monday, March 18, 2024, 11:15–11:30, HE 101

Anharmonic fingerprints from THz modes of naphtalene crystals enabled by machine-learning — •Paolo Lazzaroni, Shubham Sharma, and Mariana Rossi — Max Planck Institute for the Structure and Dynamics of Matter, Hamburg, Germany

Organic molecular crystals exhibit strong lattice anharmonicity, especially in the collective motions that are governed by intermolecular interactions and lie in the low-frequency THz range [1]. Inspired by recent observations that the polarization-orientation (PO) Raman spectra can give exquisite insight into the anharmonic couplings between modes [2], we devise a first-principles framework that aims at reproducing, explaining and give quantitative insight into the type and strengths of mode coupling. This framework is based on machine-learned potentials and polarizability tensors trained on ab-initio molecular dynamics trajectories [3]. We obtain results, even for large system supercells, through the time-correlation formalism for PO Raman signals, retaining the full anharmonic nature of the potential. In order to do this, a procedure has been developed that allows us to isolate the Γ-point Raman signal from our real-space molecular dynamics simulations. [1] M. Asher et al., Adv. Mater. 32, 1908028 (2020) [2] N. Benshalom et al., J. Phys. Chem. C 127, 36 (2023) [3] N. Raimbault et al., New J. Phys. 21 105001 (2019)

Keywords: machine-learning; molecular dynamics; Raman spectroscopy; anharmonicity