Numerical tests of coherence-corrected surface hopping methods using a donor-bridge-acceptor model system

被引:16
|
作者
Sifain, Andrew E. [1 ,2 ,3 ]
Wang, Linjun [4 ,5 ]
Tretiak, Sergei [2 ,3 ,6 ]
Prezhdo, Oleg V. [1 ,7 ]
机构
[1] Univ Southern Calif, Dept Phys & Astron, Los Angeles, CA 90089 USA
[2] Los Alamos Natl Lab, Ctr Nonlinear Studies, Theoret Div, Los Angeles, NM 87545 USA
[3] Los Alamos Natl Lab, Ctr Integrated Nanotechnol, Los Angeles, NM 87545 USA
[4] Zhejiang Univ, Ctr Chem Novel & High Performance Mat, Hangzhou 310027, Zhejiang, Peoples R China
[5] Zhejiang Univ, Dept Chem, Hangzhou 310027, Zhejiang, Peoples R China
[6] Skolkovo Inst Sci & Technol, Moscow 143026, Russia
[7] Univ Southern Calif, Dept Chem, Los Angeles, CA 90089 USA
基金
美国国家科学基金会; 中国国家自然科学基金;
关键词
NONADIABATIC MOLECULAR-DYNAMICS; ELECTRON-TRANSFER; QUANTUM DECOHERENCE; CHARGE-TRANSFER; PROTON-TRANSFER; RECOMBINATION; SEPARATION; TRANSPORT; RATES;
D O I
10.1063/1.5092999
中图分类号
O64 [物理化学(理论化学)、化学物理学];
学科分类号
070304 ; 081704 ;
摘要
Surface hopping (SH) is a popular mixed quantum-classical method for modeling nonadiabatic excited state processes in molecules and condensed phase materials. The method is simple, efficient, and easy to implement, but the use of classical and independent nuclear trajectories introduces an overcoherence in the electronic density matrix which, if ignored, often leads to spurious results, such as overestimated reaction rates. Several methods have been proposed to incorporate decoherence into SH simulations, but a lack of insightful benchmarks makes their relative accuracy unknown. Herein, we run numerical simulations of common coherence-corrected SH methods including Truhlar's decay-of-mixing (DOM) and Subotnik's augmented SH using a Donor-bridge-Acceptor (DbA) model system. Numerical simulations are carried out in the superexchange regime, where charge transfer proceeds from a donor to an acceptor as a result of donor-bridge and bridge-acceptor couplings. The computed donor-to-acceptor reaction rates are compared to the reference Marcus theory results. For the DbA model under consideration, augmented SH recovers Marcus theory with quantitative accuracy, whereas DOM is only qualitatively accurate depending on whether predefined parameters in the decoherence rate are chosen wisely. We propose a general method for parameterizing the decoherence rate in the DOM method, which improves the method's reaction rates and presumably increases its transferability. Overall, the decoherence method of choice must be chosen with great care and this work provides insight using an exactly solvable model. Published under license by AIP Publishing.
引用
收藏
页数:8
相关论文
empty
未找到相关数据