Mean-field dynamics with stochastic decoherence (MF-SD): a new algorithm for nonadiabatic mixed quantum/classical...

Area of Science:

• Quantum Chemistry
• Computational Chemistry
• Molecular Dynamics

Background:

• Nonadiabatic mixed quantum/classical (MQC) simulations are crucial for modeling quantum systems.
• Existing MQC algorithms often treat quantum decoherence and nonadiabatic transitions separately.
• This separation can lead to complexities and reduced computational efficiency.

Purpose of the Study:

• To develop a unified theoretical framework for decoherence in MQC simulations.
• To introduce a novel MQC algorithm, mean-field dynamics with stochastic decoherence (MF-SD).
• To enhance the accuracy and computational efficiency of nonadiabatic molecular dynamics simulations.

Main Methods:

• Derivation of a nuclear-induced decoherence rate from the system's density matrix using the frozen Gaussian approximation.
• Development of the MF-SD algorithm incorporating this decoherence rate.
• Stochastic determination of decoherence events at each time step, leading to wave-function collapse and velocity adjustment.

Main Results:

• MF-SD successfully unifies surface hops and decoherence into a single process.
• The algorithm avoids artificial constructs like reference states or trajectory swarms, improving efficiency.
• MF-SD demonstrates quantitative accuracy comparable to existing MQC algorithms, with superior performance in some cases.

Conclusions:

• MF-SD offers a computationally efficient and accurate approach to nonadiabatic MQC simulations.
• The unified treatment of decoherence simplifies the modeling of quantum-classical interactions.
• This new method provides a valuable tool for studying complex molecular systems.