Performance Evaluation of the Symmetrical Quasi-Classical Dynamics Method based on Meyer-Miller Mapping Hamiltonian in the Treatment of Site-Exciton Models

TL;DR

This study benchmarks the symmetrical quasi-classical dynamics (MM-SQC) method against quantum dynamics (ML-MCTDH) for site-exciton models, showing it performs well with proper bath mode treatment, especially at low bath frequencies.

Contribution

The paper provides a comprehensive evaluation of MM-SQC's accuracy across various site-exciton models, highlighting the importance of proper bath mode treatment for reliable results.

Findings

MM-SQC performs well at low bath frequencies, matching ML-MCTDH.

High-frequency bath modes require adiabatic renormalization for accurate MM-SQC results.

Proper treatment of bath modes is crucial for the reliability of MM-SQC in nonadiabatic dynamics.

Abstract

The symmetrical quasi-classical dynamics method based on the Meyer-Miller mapping Hamiltonian (MM-SQC) shows the great potential in the treatment of the nonadiabatic dynamics of complex systems. We performed the comprehensive benchmark calculations to evaluate the performance of the MM-SQC method in various site-exciton models with respect to the accurate results of quantum dynamics method multilayer multiconfigurational time-dependent Hartree (ML-MCTDH). The parameters of the site-exciton models are chosen to represent a few of prototypes used in the description of photoinduced excitonic dynamics processes in photoharvesting systems and organic solar cells, which include the rather board situations with the fast or slow bath and different system-bath couplings. When the characteristic frequency of the bath is low, the MM-SQC method performs extremely well, and it gives almost the identical results to those of ML-MCTDH. When the fast bath is considered, the deviations exist between the MM-SQC and ML-MCTDH results if the high-frequency bath modes are improperly treated by the classical manner. When the so-called adiabatic renormalization was employed to construct the reduced Hamiltonian by freezing high-frequency modes, the MM-SQC dynamics can give the results comparable to the ML-MCTDH ones. Thus, the MM-SQC method itself provide reasonable results in all test site-exciton models, while the proper treatments of the bath modes must be employed. The possible dependence of the MM-SQC dynamics on the different initial sampling methods for the nuclear degrees of freedom is also discussed.