BSE@GW-based protocol for spin-vibronic quantum dynamics using the linear vibronic coupling model. Formulation and...

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Spectroscopy

Background:

• Accurate simulation of photoinduced dynamics is crucial for understanding molecular behavior after light absorption.
• Existing methods often face challenges with complex systems, particularly transition metal complexes.
• Nonadiabatic effects and multidimensional wave packet propagation are key to describing excited-state dynamics.

Purpose of the Study:

• To present a novel computational protocol for generating potential energy surfaces.
• To perform photoinduced nonadiabatic multidimensional wave packet propagation.
• To enable accurate modeling of excited-state dynamics in complex molecular systems.

Main Methods:

• Parameterization of a linear vibronic coupling (LVC) Hamiltonian using the Green's function - Bethe-Salpeter equation (BSE@GW) approach.
• Multi-layer multi-configurational time-dependent Hartree (ML-MCTDH) wave packet propagation.
• Spectral clustering algorithm for automated ML tree generation based on time-dependent Hartree (TDH) simulations.

Main Results:

• BSE@GW offers a more robust description of transitions in absorption spectra compared to TD-DFT for the tested transition metal complex.
• The linear approximation in LVC parameterization is validated over a wide range of normal mode elongations.
• Spectral clustering allows for the generation of diverse ML trees, impacting the numerical efficiency of ML-MCTDH propagation.

Conclusions:

• The developed protocol provides a robust framework for simulating photoinduced nonadiabatic dynamics.
• The protocol demonstrates applicability to challenging systems like the transition metal complex [Fe(cpmp)]^2+.
• The flexibility in ML tree generation offers tunable numerical efficiency for wave packet propagation.