三量子比特Dicke模型中的两体和三体纠缠动力学

The dynamics of the bipartite and tripartite entanglement in the three-qubit Dicke model

本文利用绝热近似方法和精确对角化方法研究三量子比特Dicke模型中的纠缠动力学.处于两种典型的纠缠态GHZ态和W态上的量子比特在时间演化过程中与辐射光场发生强耦合作用,在各种子系统间产生纠缠,通过分析这些纠缠的演化特性发现初始GHZ态的三体纠缠鲁棒性比W态强,这与旋波近似结论一致.与旋波近似下结果不同的是,两种态中任意一对量子比特间的纠缠都随时间演化到几乎为零,而三体纠缠随时间周期演化,且纠缠程度相对较强,说明系统中的强耦合作用通过抑制量子比特中的对纠缠来支持三体纠缠.

We study the entanglement dynamics of the three-qubit Dicke model by means of the adiabatic approximation and the exact diagonalization in the parameter regime where the qubit transition frequencies are far off-resonance with the radiation field and the interaction strengths reach the ultrastrong-coupling regime.The single-mode field is prepared in the coherent state and two typical states GHZ and W are chosen as the initial three-qubit states.In the process of evolution,the interaction between the quantized field and three-qubit system leads to the generation of entanglement between the field and qubits,as well as between different parties in the three-qubit system,i.e.the pairwise entanglement of two qubits and the tripartite entanglement,which are of ongoing interest in quantum information process.The generalized concurrence and negativity are adopted to quantify different kinds of entanglement.The qubit-field entanglement never reaches the maximum and no sudden death occurs in the the tripartite entanglement for GHZ state,but it is exactly the opposite for W state.This reflects that the tripartite entanglement of the GHZ state is more robust than W sate,which is the same as in the rotating wave approximation.The results beyond the rotating wave approximation show that the pairwise entanglement gradually decreases and vanishes in the evolution of both initial states,with the tripartite entanglement periodically reaching relatively high level.This means that the interaction in system supports the tripartite entanglement at the cost of pairwise entanglement.The conclusions provide theoretical reference for the robustness of entanglement state and quantum information processing using Dicke model.