量子退相干
物理
量子力学
连贯性(哲学赌博策略)
激发态
表面跳跃
量子耗散
耗散系统
电子
时间演化
量子
原子物理学
量子电动力学
作者
Kim F. Wong,Peter J. Rossky
摘要
An electronic state and nuclear configuration dependent mechanism for electronic coherence loss is integrated into the mean field with surface hopping (MF/SH) algorithm for nonadiabatic (NA) mixed quantum–classical molecular dynamics (MQC-MD). The characteristic decoherence time scale between a pair of states is evaluated from differences in the instantaneous Hellmann–Feynman forces on the two surfaces at each phase space point along the quantum–classical trajectory. Within this instantaneous decoherence mean field with surface hopping (id-MF/SH) formalism, both the primary evolution that is responsible for transition probabilities and the auxiliary equations governing the nuclear dynamics are described by the same dissipative MQC Liouville–von Neumann equation. Decoherence, therefore, impacts both the transition probabilities and the realization of the quantum–classical trajectory. The method is implemented for the solvated electron in water and methanol and applied to trajectories describing photoexcitation of equilibrium ground-state species. Analysis reveals that, in general, both the decoherence time scale and the NA coupling increase with decreasing eigenstate energy gap. The cooperative combination of both strong coupling and large coherence as the energy levels approach each other gives rise to localized regions of comparatively higher transition probabilities. Excited-state survival probability analysis extracts a decay time of 1540 fs for the solvated electron in water and 2617 fs for the methanol case. The approximate agreement with the decay of time 2102 fs for the aqueous system, based on simulations employing a fixed 6 fs decoherence parameter, suggests that an average prescription of coherence loss may be adequate for the first-excited-state solvated electron system. The self-contained form of the id-MF/SH formalism, however, makes the new method a general approach to NA MQC-MD for condensed phase systems.
科研通智能强力驱动
Strongly Powered by AbleSci AI