Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Multipartite entanglement generation with high-order non-Hermitian exceptional points from dressing-controlled atomic nonlinearity

Open Access Open Access

Abstract

Multipartite entanglement has emerged as a valuable quantum resource for constructing large-scale quantum networks. However, the presence of non-Hermitian features induced by natural microscopic quantum systems significantly modifies the overall response of nonlinear parametric processes, thereby enabling direct manipulation of multipartite entanglement properties. In this study, we demonstrate the generation of multimode entanglement through atomic four-wave mixing (FWM) and analyze the properties of exceptional points (EP) under dressing control in non-Hermitian systems. By leveraging dressing-controlled atomic nonlinearity, we achieve versatile EPs and higher-order EPs by carefully tuning the atomic multi-parameter in the cascading FWM system. Additionally, we investigate the entanglement properties of various permutations of the output signal modes using the positive partial transpose (PPT) criterion. Notably, under non-Hermitian control, the application of single-, double-, and N-dressing splits leads to coherent multichannel control and further extends the scale of quantum entanglement. The outcomes of our research offer a novel approach to actively control non-Hermitian quantum phenomena without relying on artificial photonic structures. Furthermore, this paves the way for the realization of complex quantum information tasks by exploiting the non-Hermitian characteristics of the light-matter interaction.

© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

1. Introduction

Hermiticity is a fundamental and essential property observed in a wide range of physical systems, arising from the conservation of energy, and is characterized by the presence of real-valued eigenvalues. However, in realistic scenarios involving energy exchanges with the surrounding environment, the description of such open systems necessitates the use of a non-Hermitian Hamiltonian. Extensive investigation in this field has unveiled the remarkable ability of non-Hermitian processes to significantly transform system behaviors and give rise to counterintuitive phenomena when compared to their Hermitian counterparts [13]. In particular, the occurrence of degeneracy points, where the eigenvalues of the complex-valued non-Hermitian Hamiltonian coincide, provides a novel perspective for the study of physical properties in non-Hermitian systems. These special points, known as exceptional points (EPs), represent branch points where two eigenvalues and their corresponding eigenvectors simultaneously merge into a a value [46]. Furthermore, the intriguing phenomenon of higher-order EPs arises when more than two eigenvalues converge simultaneously [79].

Recently, exceptional points (EPs) have attracted considerable attention and have been the subject of extensive investigation within the field of optical systems. This research encompasses a diverse range of systems, including photonic crystals [10], optical microcavities [1113], optical waveguides [14], and atomic systems [15,16]. Of particular significance are systems that exhibit parity-time (PT) symmetry, achieved through precise manipulation of gain or dissipation mechanisms [1719]. Concurrently, there has been a growing interest in exploring EPs in other physical systems, such as plasmonics [20], acoustics [21,22], and optomechanics [23,24], among others. Moreover, variations in parameters within non-Hermitian systems can prompt broken-symmetry phase transitions, resulting in profound alterations in the sensitivity surrounding EPs. This exceptional characteristic of EPs finds widespread application in enhancing sensitivity measurements [9,13,25], pump or loss-induced lasers [26,27], laser mode selectivity [28,29], and mode switching [23]. However, previous studies on EPs have predominantly focused on fixed regimes, specifically the weak or strong coupling regime, within coupled artificial optical or photonic structures. As a consequence, the exploration of EPs in the context of light-matter coupling remains largely unexplored.

The realization of large-scale quantum networks relies heavily on multipartite entanglement [3033]. Such networks have significant applications in the field of quantum information processing, spanning from quantum computing to quantum secure communications [3436]. Despite the emergence of biphoton processes with nonlinear crystals as one of the most developed technologies for generating quantum entanglement, there remains a pressing need to discover scalable methods for constructing large-scale quantum states [30,37]. In recent years, the parametric amplifier four-wave mixing (PA-FWM) technique utilizing atoms has demonstrated promise as an effective approach for generating multimode entanglement [3840]. Atomic media offer several inherent physical advantages such as strong nonlinear coefficients [41], narrow bandwidth [42,43], and long coherence time [44]. These properties make atomic systems highly suitable for advanced quantum technologies, including multi-photon generation [45,46], sources of multipartite quantum entanglement [47], and all-optical quantum information protocols [4850].

The atomic four-wave mixing (FWM) process relies on third-order nonlinear effects to generate a pair of entangled beams [51]. To effectively modulate the nonlinear susceptibility, atomic coherence-assisted non-Hermitian processes are employed. These processes take advantage of the dressing effect, which arises from the constructive interference among different transition paths. Through this mechanism, the nonlinear susceptibility can be optimized to control the gains of the FWM [52], Rabi oscillations [53,54], and shape the correlation waveform of biphotons [55]. Furthermore, the dressing effects enable energy level splitting, facilitating the creation of multimode and multi-coherent channels in the FWM processes [52,56]. Consequently, this allows for the generation of high-dimensional time-energy entangled photon states [57] and enhances the degree of squeezing in the state [5860].

However, coherently-prepared, multilevel atoms represent naturally occurring microscopic dissipative systems that are inherently endowed with EPs. In contrast to less controllable systems, the presence of atomic coherence within these dissipative atomic structures enables precise adjustment of the light-matter coupling between atomic states and external optical fields surrounding the EPs. This, in turn, facilitates active control over the quantum correlations generated within such non-Hermitian parametric processes.

This study explores the analysis of the dressing-controlled energy-level cascade four-wave mixing (ELC-FWM) process, a natural non-Hermitian nonlinear system, to demonstrate the presence of versatile exceptional points (EPs) and higher-order EPs, as well as multipartite entanglement within a single Rb cell. The coherent preparation of multilevel atoms offers distinct advantages in achieving flexible non-Hermitian control. We investigate the evolution of EPs in these non-Hermitian systems using three different doubly dressing structures, along with a single dressing structure. By adjusting the atomic multi-parameter, these systems can exhibit a wide range of EPs and higher-order EPs. Moreover, through the regulation of atomic coherence, the multipartite entanglement among the output modes can be modulated via multiple coherent channels originating from the degeneracy of the real and imaginary parts of the eigenvalues. Additionally, by utilizing generalized N-dressing fields in the interaction with the ELC-FWM process, non-Hermitian control can expand the quantum information capacity by generating a large-scale of multipartite entangled states with multiple modes. This research lays the foundation for studying the effects of quantum entanglement around EPs in non-Hermitian systems of light-matter coupling.

2. ELC-FWM process model

Using Four-wave mixing with the third-order nonlinear effect in rubidium atomic medium has been proven to be a promising candidate for generating two quantum correlated beams of light. Here we consider a multimode FWM system consisting of two pump fields E1 and E2 with one probe field ES2, as shown in Fig. 1(a), which depicts the cascade energy level of three modes scheme using a double-Λ FWM process in hot rubidium vapor. Here, we use two strong coherent pump beams cross in an 85Rb atomic vapor at a small angle, then another seed beam (weak coherent beam) is injected into medium vapor which has frequency detuning Δ1 to the D1 line of the 85Rb (5S1/2, F = 2→5P1/2). The strong pump E1 (frequency ω1, wave vector k1, and Rabi frequency Ω1) drives the atom transition from energy |1〉 to |3〉. When it decays to energy |2〉, the process will emit a photon with the mode ES1. Then the photon of E1 mode is excited from energy |2〉 to |3〉, from where it decays to the ground state, emitting a photon with the mode ES2 (ω2, k2, Ω2). The second strong beam E3 (ω3, k3, Ω3) is also excited from the ground state to energy |3〉, from where it decays to the energy |2〉, emitting a photon of the mode ES3.

 figure: Fig. 1.

Fig. 1. The scheme of the three modes is based on the energy-level-cascade four-wave mixing process. (a) The ELC-FWM processes in a single rubidium vapor. (b) Energy level diagram of the whole ELC-FWM system with two subsystems of FWM1 and FWM2. (c) The spatial structures of the two pump beams E1 and E2 with three output modes ES1, ES2 and ES3. (d) The diagram of the phase-matching conditions for the parametric amplified FWM1 and FWM2.

Download Full Size | PDF

The whole system consists of dual-pump fields process by a cascade of energy levels, and two FWM processes will occur simultaneously (see Fig. 1(b)). Figure 1(c) that shows the spatial distribution amongst the correlated FWM output modes. In contrast to the single FWM process which annihilates two pump E1 photons to generate a single photon ES1 and an idler photon ES2, the new photon of ES3 will be generated by the second cascaded FWM interaction between pump E1 and E2. The two FWM processes are considered to be a whole process, and it also satisfies the phase-matching conditions ks1+ ks2 = 2k1 and ks2+ ks3= k1 +k3 (see Fig. 1(d)) with a cascade of energy levels in a single Rb vapor.

In the entire process, the interaction Hamiltonian of the entire system can be expressed as ${H_1} = i\hbar {\kappa _1}{\hat{a}^\dagger}_1{\hat{a}^\dagger}_2 + h.c$ and ${H_2} = i\hbar {\kappa _2}{\hat{a}^\dagger}_2{\hat{a}^\dagger}_3 + h.c$, where ${\hat{a}^\dagger}_i$(i = 1,2,3) are the boson creation operators of the three generated modes and the commutation relations can be expressed as $[{{{\hat{a}}_i},\hat{a}_j^\dagger} ]= {\delta _{ij}}$. Therefore, two pump parameters ${\kappa _1} = |{ - i{\mu_0}{\varpi_1}\chi_1^{(3)}E_1^2/2{k_s}} |$ and ${\kappa _2} = |{ - i{\mu_0}{\varpi_2}\chi_2^{(3)}{E_1}{E_3}/2{k_s}} |$ depend on the central frequency of Stokes and anti-Stokes signals ${\varpi _i}$(i = 1,2) as well as nonlinear susceptibility tensor $\chi _i^{(3)}$. Here, ${{\mathbf E}_i}$ is the pump optical field amplitude, and µ0 is the diploe matrix element of an atomic transition.

In the Heisenberg picture, the associated boson creation (annihilation) operators obey the time-evolution motion equation within the dipole approximation. The dynamic equations have the form

$$\textrm{PA - FWM1}\;\frac{{d{{\hat{a}}_1}}}{{dt}} = {\kappa _1}\hat{a}_2^\dagger \begin{array}{c} , \end{array}\frac{{d{{\hat{a}}_2}}}{{dt}} = {\kappa _1}\hat{a}_1^\dagger ,$$
$$\textrm{PA - FWM2}\;\frac{{d{{\hat{a}}_2}}}{{dt}} = {\kappa _2}\hat{a}_3^\dagger \begin{array}{c} , \end{array}\frac{{d{{\hat{a}}_3}}}{{dt}} = {\kappa _2}\hat{a}_2^\dagger .$$

After the algebraic operation, the output of cascade FWM can be written as follows:

$${\hat{a}_{1out}} = {G_1}{\hat{a}_{1in}} + {g_1}\hat{a}_{2in}^\dagger, $$
$${\hat{a}_{2out}} = {g_1}{G_2}\hat{a}_{1in}^\dagger + {G_1}{G_2}{\hat{a}_{2in}} + {g_2}\hat{a}_{3in}^\dagger, $$
$${\hat{a}_{3out}} = {g_1}{g_2}{\hat{a}_{1in}} + {G_1}{g_2}\hat{a}_{2in}^\dagger + {G_2}{\hat{a}_{3in}}, $$
where G1 = cosh(κ1 t) and G2 = cosh(κ2 t) and they also satisfy Gi2gi2 = 1 (i = 1, 2), which stand for the amplitude gain of the FWM processes respectively, depending on the ELC-FWM process strength of interaction.

3. Non-Hermitian exceptional points under multi-dressing

In this section, let us consider the effect of the non-Hermitian system and the EPs under the control of the single- and doubly-dressing fields. Due to the third-order density matrix element $\rho _i^{(3)}$ being proportional to the nonlinear polarization χi(3), the study of its resonance modes can simplify the study evolution of exceptional points in the non-Hermitian system under the multi-dressing coherent control.

With the dipole and rotating wave approximation, the Hamiltonian of atomic interaction in the cascade energy level PA-FWM1 process can be expressed

$$\begin{aligned} {{\mathbf H}_i} &={-} \hbar (({\Delta _1} - {\Delta _s})|2 \rangle \left\langle 2 \right|+ {\Delta _1}|3 \rangle \left\langle 3 \right|)\\ & + \hbar (({\Delta _1} - {\Delta _s} + {{\Delta ^{\prime}}_1})|4 \rangle \left\langle 4 \right|)\\ & + \hbar ({\Omega _1}|3 \rangle \left\langle 1 \right|+ {\Omega _{s1}}|3 \rangle \left\langle 2 \right|)\\ & + \hbar ({\Omega _{s2}}|4 \rangle \left\langle 1 \right|+ {\Omega _1}|4 \rangle \left\langle 2 \right|+ H.c.), \end{aligned}$$
where H.c represents the Hermitian conjugate item, Ωi=µijEi/ℏ is the Rabi frequency which indicates the strength of the interaction of the field Ei with the atomic ensemble; µij is atomic dipole moment matrix element, and Δiiji is the frequency detuning between transition levels and the coherent fields. Under this situation, the density matrix operators can be described by the master equation in the atomic ensemble
$$\dot{\boldsymbol{\mathrm{\rho}}} ={-} \frac{i}{\hbar }[{{\mathbf H}_i},\boldsymbol{\mathrm{\rho}}] - \Gamma \boldsymbol{\mathrm{\rho}}, $$
where $\boldsymbol{\mathrm{\rho}}$ is the density operator and $\Gamma \boldsymbol{\mathrm{\rho}}$ represents the decoherence process of the atomic system. Based on Eq. (3) and Eq. (4), three output modes ES1, ES2 and ES3 optical fields can be derived via perturbation chains as follows
$$\textrm{PA - FWM2}\,\rho _{11}^{(0 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{31}^{(1 )}\buildrel {{\omega _{S1}}} \over \longrightarrow \rho _{21}^{(2 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{31({S2} )}^{(3 )}, $$
$$\textrm{PA - FWM2}\,\rho _{22}^{(0 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{32}^{(1 )}\buildrel {{\omega _2}} \over \longrightarrow \rho _{12}^{(2 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{32({S1} )}^{(3 )}, $$
$$\textrm{PA - FWM2}\;\rho _{22}^{(0 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{32}^{(1 )}\buildrel {{\omega _2}} \over \longrightarrow \rho _{12}^{(2 )}\buildrel {{\omega _3}} \over \longrightarrow \rho _{32({S3} )}^{(3 )}, $$
$$\textrm{PA - FWM2}\;\rho _{11}^{(0 )}\buildrel {{\omega _3}} \over \longrightarrow \rho _{31}^{(1 )}\buildrel {{\omega _{S3}}} \over \longrightarrow \rho _{21}^{(2 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{31({S4} )}^{(3 )}, $$
where the superscript of the element of the density matrix ${\rho ^{(j)}}$(j = 0,1,2,3) represents the perturbation order, and ${\omega _i}$ is the optical field frequency. Each of the chains describes the transition path of particles between different energy levels. For example, first the PA-FWM1:$\rho _{11}^{(0)}\buildrel {{\omega _1}} \over \longrightarrow \rho _{31}^{(1)}$ denotes the atom at the ground energy level |1〉 absorbing a photon and transitions from the ground state $\rho _{11}^{(0)}$ to the excited state $\rho _{31}^{(1)}$. According to the above perturbation chains, the third order of density matrix of ES1, ES2 and ES3 can be written as
$$\textrm{PA - FWM1}\;\rho _{31(S2)}^{(3 )} ={-} i\Omega _1^2{\Omega _{S1}}\textrm{/}{d_{31}}{d_{21}}{d^{\prime}_{31}}, $$
$$\textrm{PA - FWM1}\;\rho _{32(S1)}^{(3 )} ={-} i\Omega _1^2{\Omega _\textrm{2}}\textrm{/}{d_{32}}{d_{12}}{d^{\prime}_{32}}, $$
$$\textrm{PA - FWM2}\;\rho _{31(S2)}^{(3)} ={-} i{\Omega _1}{\Omega _3}{\Omega _{S3}}/{d^{\prime\prime}_{31}}{d^{\prime}_{21}}{d^{\prime}_{312}}, $$
$$\textrm{PA - FWM2}\;\rho _{32(S3)}^{(3 )} ={-} i{\Omega _1}{\Omega _2}{\Omega _3}\textrm{/}{d_{32}}{d^{\prime}_{12}}{d^{\prime\prime}_{32}}, $$
where $d_{ij}^{} = {\Gamma _{ij}} + i({\Delta _i} - {\Delta _k})$ represents the different density matrix elements items in a whole FWM process. The ${\Gamma _{ij}} = ({\Gamma _i} + {\Gamma _j})/2$ is the decoherence rate between state |i〉 and |j〉. According to the order of particle transitions between the different energy levels, the parameters ${d_{31}}\textrm{ = }{\Gamma _{31}} + i{\Delta _1}$, ${d_{21}}\textrm{ = }{\Gamma _{21}} + i({{\Delta_1} - {\Delta_{S1}}} )$, ${d^{\prime}_{31}}\textrm{ = }{\Gamma _{31}} + i({{\Delta_1} - {\Delta_{S1}} + {{\Delta^{\prime}}_1}} )$, and other items can also be obtained using the perturbation chains. For the two-photon detuning in the parametric ELC-FWM processes, the δi is used to represent the quantum deviations around the central frequency ${\varpi _{s1}}$, which can be expressed as δi1s1, satisfying ${\delta _i}$${\varpi _{s1}}$. Due to the uncertainty in the transitions of the particles, the deviation δi is related to the quantum property of the biphoton from the FWM processes.

3.1. Single dressing with non-Hermitian EP

With the strong coherent field E1 acting on the energy level |1〉, the energy level will be split into two dressed states. The density matrix elements of different orders can be modified by the dressing field. For convenience, without losing generality, here we consider the modified second-order density matrix element dressed by the pump field E1, as an example, to discuss atomic coherence effect within the coherent channel of PA-FWM1.

As shown in Fig. 2(a), with the dressing field E1, the dressed states|Ω1 + 〉 and |Ω1-〉 by splitting the ground energy |1〉 can be obtained, and the two eigenvalues define the two positions of the quasi-energy levels. Under the dressing effect of the E1 field, the density matrix element in Eq. (6a) can be expressed which can be written as

$$\rho _{31(DS2)}^{(3)} = \frac{{ - i\Omega _1^2{\Omega _{S1}}}}{{{d_{31}}({d_{21}} + \Omega _1^2/{d_{23}}){{d^{\prime}}_{31}}}},$$
where the ${d_{21}} = {\Gamma _{21}} + i{\delta _1}$, ${d_{23}} = {\Gamma _{23}} + i{\delta _1} - i{\Delta _1}$, and ${d^{\prime}_{31}} = {\Gamma _{31}} + i{\delta _1} + {\Delta ^{\prime}_1}$. The PA-FWM1 and PA-FWM2 processes in the entire system satisfy the energy conservation:${\delta _1} + {\delta _2} = {\delta ^{\prime}_2} + {\delta _3}$, which is also the typical association of these outputs of photon frequency.

 figure: Fig. 2.

Fig. 2. Energy level diagram and EPs in the ELC-FWM processes under the single dressing field. (a) two dressed states splitting diagram with single dressing effect. The blue bidirectional dotted arrow indicates the dressing effect caused by the E1 field. (b) and (e) The eigenvalue surfaces of real and imaginary parts in the multi-coherent channels with the parameters Ω131 and frequency detuning Δ131. (c) and (f) The eigenvalues of the real and imaginary parts. The exceptional points occur at Ω131 = 0.5 with Δ1 = 0. (d) and (g) The level repulsion of the eigenvalues in the real and imaginary parts with Δ1 = 0.15Γ31.

Download Full Size | PDF

As the dressed states can be considered as the two-level system coupled with an electric field, the complex eigenvalues of the dressing system with the density matrix can be obtained

$$\delta _ \pm ^{} = \frac{{i({\Gamma _{21}} + {\Gamma _{23}}) + {\Delta _1}}}{2} \pm \sqrt {\frac{{{{({\Delta _1} + i({\Gamma _{21}} - {\Gamma _{23}}))}^2}}}{4} + \Omega _1^2} .$$

Figure 2(b), and 2(e) illustrate the eigenvalues surfaces of the real parts and imaginary parts around the exceptional point singularity in the parameter space. When the detuning Δ1 = 0, the two complex eigenvalues show repulsion at the EP with Ω1=|Γ2123|/2 in Fig. 2(c) and 2(f). Also, with no-zero valued detuning, the corresponding eigenvalues in the real part and imaginary part will occur level repulsion in Fig. 2(d) and 2(g).

As such, the ELC-FWM systems exhibit all-optical controllable non-Hermitian characteristics with the dressing effect. As seen in Fig. 2(c) and 2(f), when we set Ω131 < 0.5, the eigenvalues δ± of the real parts coalesce but the imaginary parts occur degenerate. In this case, the coupled system takes the same frequency mode with different dissipative rates, which, as a result, can induce a single absorption peak from the interference between the two coherent channels. When Ω131 > 0.5, the real parts repel each other, but the imaginary part merges at the EP. With the increase of the strength of the dressing field, the interference between the two coherent channels become gradually disappears and it corresponds to different resonance frequencies and degenerate dissipation near the EP. Furthermore, such non-Hermiticity is involved within the parametric FWM processes, so it is directly related to bi-photons generation. The corresponding quantum entanglement property will be analyzed in Sec. 4. Note that, for the dressing effects and non-Hermitian control, in practice, it does not have to add extra dressing fields. One can directly modulate the coupling strength via changing the detuning and the power of the pump fields.

3.2. Multi-dressing with non-Hermitian higher-order EP

When the two strong pump fields exhibit the dressing effect on the energy level |1 > and |3>, the energy level split can be further induced [61], which can support higher-order EPs by tuning the system parameters. Due to the influence of the different split energy levels, the doubly dressing effect shows the three different structures in the ELC-FWM process, i.e., parallel-cascade scheme, sequential-cascade scheme, and nested-cascade scheme. These three structures will generate different effects for atomic coherence and parametric gain, as shown in Fig. 3.

 figure: Fig. 3.

Fig. 3. The doubly dressed states energy level diagram with two strong coherent fields E1 and E2. (a) The doubly dressed FWM of the parallel-cascade scheme with effecting on different density matrix element (b) The doubly dressing FWM of sequential-cascade scheme same as (a), but dressed same density matrix element. (c) The doubly dressed FWM of the nested-cascade scheme. (d) The entanglement scheme from multi-coherent channels of the parallel-cascade scheme (left), the other two schemes (right). (a)-(c) The blue and green bidirectional dotted arrow indicates the dressing effect by the fields E1 and E2.

Download Full Size | PDF

According to the dressing effect acting on the different energy levels, it corresponds to the different density matrix elements in the ECL-FWM system. In the parallel-cascade scheme, the energy level |1〉 and |3〉 are dressed by two strong coherent fields independently in Fig. 3(a), while the sequential-cascade and nested-cascade scheme structural fields modify the level |3〉 at the same time, but with a different order, as seen in Fig. 3(b) and 3(c). So different dressing effect can be expressed by the corresponding density matrix element. Moreover, the multi-coherent channels via the doubly dressing control can further increase the scale of the produced states. As shown in Fig. 3(d), the parallel dressing regime and the other two dressing structures correspond to the four- and three-mode tripartite quantum states, respectively. The transitions of these particles between different splitting energy levels exhibit frequency modes expansion under the multi-dressing effect.

Firstly, we analyze the parallel-cascade doubly dressing (${\cal P}{\cal C}{\cal D}{\cal D}$) scheme which is the two strong coherent fields that modify the different density matrix elements and different energy levels. As shown in Fig. 3(a), the energy |1〉 is split into the two dressed states |Ω1 + 〉and |Ω1-〉 by the coherent field E1. Similarly, the energy |3〉 is split into the two dressed states |Ω3 + 〉 and |Ω3-〉 by the field E3, respectively. Thus, the corresponding output modes are regulated by the different dressing fields. Based on Eq. (7a), the third-order density matrix element of the anti-Stocks optical field ES2 with doubly dressing effect can be expressed by the perturbation chain $\rho _{11}^{(0 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{31}^{(1 )}\buildrel {{\omega _{S1}}} \over \longrightarrow \rho _{2D\Omega 1 \pm }^{(2 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{3D\Omega 3 \pm ({S2} )}^{(3 )}$, where the subscript of $\rho _{2D\Omega 1 \pm }^{(2 )}$ and $\rho _{3D\Omega 3 \pm ({S2} )}^{(3 )}$ indicates the two fields E1 and E3 dress the energy |1〉 and |3〉, respectively.

The third-order density matrix element of single dressing in Eq. (7) are modified by the two dressing fields which can be rewritten

$$\rho _{31(S2)}^{(3 )} = \frac{{ - i\Omega _1^2{\Omega _{S1}}}}{{({{\Gamma _{31}} + i{\Delta _1}} )\left( {{\Gamma _{21}} + i{\delta_1}\textrm{ + }\frac{{\Omega _1^2}}{{{d_{23}}}}} \right)\left( {{\Gamma _{31}} + i{\delta_1} + i{{\Delta^{\prime}}_1} + \frac{{\Omega _3^2}}{{{d_{11}}}}} \right)}}, $$
where ${d_{23}} = {\Gamma _{23}} + i{\delta _1} - i{\Delta _1}$ and ${d_{11}} = {\Gamma _{11}} + i{\delta _1} + i{\Delta ^{\prime}_1} - i{\Delta _3}$ stand for the dressing items due to the splitting energy level by the two dressing fields E1 and E3. The second and third items of the third-order density matrix element $\Omega _1^2/{d_{23}}$ and $\Omega _3^2/{d_{11}}$ represent the modification items. According to Eq. (9), the splitting of the energy level will induce the generation of multiple modes in the ELC-FWM process. Considering the resonant response condition, i.e., maximizing the density matrix element, the complex eigenvalues can be obtained with the dressing field E1
$$\delta _{1 \pm }^{} = \frac{{i({\Gamma _{21}} + {\Gamma _{23}}) + {\Delta _1} \pm \sqrt {{{({\Delta _1} + i({\Gamma _{21}} - {\Gamma _{23}}))}^2} + 4\Omega _1^2} }}{2}. $$

Similarly, the complex eigenvalues with the dressing field E3 also can be obtained

$$\delta _{2 \pm }^{} = \frac{{i({\Gamma _{11}} + {\Gamma _{31}}) + {\Delta _3} - 2{{\Delta ^{\prime}}_1} \pm \sqrt {{{({\Delta _3} + i({\Gamma _{11}} - {\Gamma _{31}}))}^2} + 4\Omega _3^2} }}{2}.$$
With changing the Rabi frequencies Ω1 and Ω3, respectively, the eigenvalue surfaces are mutually relative cross together in Fig. 4(a), 4(c). In Fig. 4(b) and 4(d), we plot the real parts and imaginary parts of the eigenvalues under the dressing fields E1 or E3 as a function of the Rabi frequencies Ω1 or Ω3. If the detuning is not considered, two independent second-order EP(1) and EP(2) can be observed in determining one of the Rabi frequencies. When we set Ω1 = 0.5Γ31, in the Ω331 > 0.24 case, two eigenvalues show repulsion at EP to Ω3=|Γ1131|/2. When Ω331 < 0.24, the real part of the eigenvalue become degenerate, and the imaginary part split from the EP. When Ω3 = 0.5Γ31, in the Ω131 > 0.5 case, the eigenvalues become bifurcation near the EP to Ω1=|Γ2123|/2. The coherent fields E1 and E3 couple the ground state |1 > and the excited state |3>, which drive the particle transition |1〉→|3〉→|1〉. As the two dressing fields work in parallel, i.e., acting on different orders of the density matrix, two independent EPs can be observed in such ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme.

 figure: Fig. 4.

Fig. 4. Exceptional points in the ELC-FWM processes under the doubly dressing ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme. (a) and (c) The eigenvalue surfaces of real and imaginary parts in the multi-coherent channels with the parameters Ω131 and Ω331; (b) and (d) the real parts and imaginary parts depending on the two parameters Ω131 and Ω331 with Ω1 = 0.5Γ31 and Ω3 = 0.5Γ31.

Download Full Size | PDF

Secondly, the sequential-cascade doubly dressing (${\cal S}{\cal S}{\cal D}{\cal D}$) scheme has a different dressing level structure compared to the ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme. Here the energy level |3〉 is split by two strong coherent fields E1 and E3, and the corresponding second-order density matrix element is modified by the two dressing fields simultaneously. For the ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme, the third-order density matrix element can be obtained by the perturbation chain of the ELC-FWM process $\rho _{11}^{(0 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{31}^{(1 )}\buildrel {{\omega _{S1}}} \over \longrightarrow \rho _{2D\Omega 1 \pm \Omega 3 \pm }^{(2 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{3({S2} )}^{(3 )}$, where the subscript $\rho _{2D\Omega 1 \pm \Omega 3 \pm }^{(2 )}$ indicates the dressing fields E1 and E3 dress the energy |3 > simultaneously. From the perturbation chain, the fields E1 and E3 drive the particle transition |2〉→|3〉→|2〉 and |1〉→|3〉→|1〉, respectively. According to Eq. (7), the third-order density matrix element can be rewritten as,

$$\rho _{31(S2)}^{(3 )} = \frac{{ - i\Omega _1^2{\Omega _{S1}}}}{{({{\Gamma _{31}} + i{\Delta _1}} )\left( {{\Gamma _{21}} + i{\delta_1}\textrm{ + }\frac{{\Omega _1^2}}{{{d_{31}}}} + \frac{{\Omega _3^2}}{{{{d^{\prime}}_{23}}}}} \right)({{\Gamma _{31}} + i{\delta_1} + i{{\Delta^{\prime}}_1}} )}}, $$
where the dressing items represent ${d_{31}} = {\Gamma _{31}} + i{\delta _1} + i{\Delta ^{\prime}_1}$ and ${d^{\prime}_{23}} = {\Gamma _{23}} + i{\delta _1} - i{\Delta _3}$. The second item of the third-order density matrix element is modified to the $\Omega _1^2/{d_{31}}$ and $\Omega _3^2/{d^{\prime}_{23}}$ together. Thus, the atomic coherence effect is jointly influenced by the two dressing fields E1 and E3, which suggests generating high-order EPs

Via maximizing the third-order density matrix element, the three complex eigenvalues δ1, δ2, and δ3 can be obtained by analytically solving a cubic equation (more details seen in Supplement 1). Considering the higher-order EPs conditions, the three eigenvalues degenerate with,

$$\begin{aligned} {\Omega _1} &={\pm} \frac{{{{(i(2{\Gamma _{31}} - {\Gamma _{23}} - {\Gamma _{21}}) - 2{{\Delta ^{\prime}}_1} - {\Delta _3})}^{3/2}}}}{{3\sqrt 3 \sqrt {i({\Gamma _{23}} - {\Gamma _{31}}) + ({{\Delta ^{\prime}}_1} + {\Delta _3})} }},\\ &{\Omega _3} ={\pm} \frac{{{{(i({\Gamma _{21}} - 2{\Gamma _{23}} + {\Gamma _{31}}) - {{\Delta ^{\prime}}_1} - 2{\Delta _3})}^{3/2}}}}{{3\sqrt 3 \sqrt {i({\Gamma _{23}} - {\Gamma _{31}}) + ({{\Delta ^{\prime}}_1} + {\Delta _3})} }}. \end{aligned}$$

Here, the higher-order EPs can be achieved when Eq. (13) takes positive real values (for dressing effect, Ω1 and Ω3 are taken as the strength of the Rabi frequency), and the corresponding parameters satisfy the constraint ${\Delta ^{\prime}_1} = {\Delta _3} = 0$ and ${\Gamma _{23}} \ne {\Gamma _{31}}$, which indicates the three eigenvalues can coalesce one value.

When changing the Rabi frequency of the dressing fields, the eigenvalue surfaces of real and imaginary parts are mutually relative cross together as shown in Fig. 5(a) and 5(e). In Fig. 5(b)-5(h), without considering the fields detuning, the real parts and imaginary parts of the three eigenvalues are plotted as a function with Ω1 = 0.1Γ31, 0.22Γ31, 0.35Γ31 in the ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme. As seen in Fig. 5, the three eigenvalues δ1, δ2 and δ3 are represented by the red solid line, the blue and black dashed line, respectively. For Ω1 = 0.1Γ31, the two eigenvalues δ1 and δ2 coalesce one value at the Ω3 = 0.3Γ31, and the two eigenvalues δ1 and δ3 coalesce one value at the Ω3 = 0.46Γ31. In this case, the non-Hermitian system corresponds to two EPs. With increasing the Ω1 = 0.22Γ31, three eigenvalues δ1, δ2 and δ3 coalesce one value at the Ω3 = 0.55Γ31, corresponding to the higher-order EP in Fig. 5(c) and Fig. 5(g). Here, we find the two EPs can transform into the higher-order EP in the ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme. If we continue to increase the Rabi frequency of the dressing field E1 to Ω1 = 0.35Γ31, it can be observed that the original degenerate EP has a level repulsion, and the eigenvalues of the ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme are no longer degenerate, seen in Fig. 5(d) and Fig. 5(h). It is worth noting that, different from those in the ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme, here the two dressing fields co-modify the same item in the density matrix element in the ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme. This makes the system appear higher-order non-Hermitian EPs. Therefore, higher-order EPs can be understood as a result of the coupling between multi-field and the atoms, while ordinary EPs are from the coupling between a single field and the atoms, similar to single dressing and ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme.

 figure: Fig. 5.

Fig. 5. Exceptional points in the ELC-FWM processes under the doubly dressing ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme. (a) and (e) The eigenvalue surfaces of the real and imaginary parts with the two dressing fields parameters Ω1/Γ31 and Ω3/Γ31. (b) and (f) The real and imaginary part of the eigenvalues with Ω1 = 0.1Γ31. (c) and (g) Ω1 = 0.22Γ31. (d) and (h) Ω1 = 0.35Γ31.

Download Full Size | PDF

Thirdly, a different coupling between multi-field and the atoms, i.e., the nested-cascade doubly dressing (${\cal N}{\cal C}{\cal D}{\cal D}$) scheme, is considered. However, here the two dressing fields have hierarchy structural characteristics, i.e., the strong coherent field E1 is the primary dressing field and E3 is the secondary dressing field. If the coherent field E1 is equal to zero, there is no dressing effect regardless of the presence of coherent field E3. The third-order density matrix element can be obtained by the perturbation chain $\rho _{11}^{(0 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{31}^{(1 )}\buildrel {{\omega _{S1}}} \over \longrightarrow \rho _{2D\Omega 1 \pm{/}\Omega 3 \pm }^{(2 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{3({S2} )}^{(3 )}$, where the subscript $\rho _{2D\Omega 1 \pm{/}\Omega 3 \pm }^{(2 )}$ indicates the dressing field E1 and E3 acting on the same energy |3 > . So, also the atomic coherence effect is influenced in a hierarchy order by the two dressing fields E1 and E3. The third-order density matrix element can be rewritten as

$$\rho _{31(S2)}^{(3 )} = \frac{{ - i\Omega _1^2{\Omega _{S1}}}}{{({{\Gamma _{31}} + i{\Delta _1}} )\left( {{\Gamma _{21}} + i{\delta_1}\textrm{ + }\Omega _1^2/({{d^{\prime}}_{31}} + \frac{{\Omega _3^2}}{{{d_{21}}}})} \right)({{\Gamma _{31}} + i{\delta_1} + i{{\Delta^{\prime}}_1}} )}}, $$
where the corresponding items are the ${d^{\prime}_{31}} = {\Gamma _{31}} + i{\delta _1} + i\Delta ^{\prime}$, ${d_{21}} = {\Gamma _{21}} + i{\delta _1} - i{\Delta _3} + i\Delta ^{\prime}$. In Eq. (14), the dressing item of the third-order density matrix element $\Omega _3^2/{d_{21}}$ is nested in the item $\Omega _1^2/{d^{\prime}_{31}}$.

The three complex eigenvalues δ1, δ2 and δ3 can also be obtained by solving the cubic equation. (more details seen in Supplement 1). The higher-order EP conditions, with the three degenerate eigenvalues, are obtained as

$$\begin{aligned} {\Omega _1} &={\pm} \frac{{i{{(i{\Gamma _{21}} - i\Gamma {}_{31} + 2{{\Delta ^{\prime}}_1} - {\Delta _3})}^{3/2}}}}{{3\sqrt 3 \sqrt {{{\Delta ^{\prime}}_1} - {\Delta _3}} }},\\ &{\Omega _3} ={\pm} \frac{{i{{({\Gamma _{21}} - \Gamma {}_{31} + i{{\Delta ^{\prime}}_1} - 2i{\Delta _3})}^{3/2}}}}{{3\sqrt 3 \sqrt {i{\Delta _3} - i{{\Delta ^{\prime}}_1}} }}. \end{aligned}$$
When Eq. (15) takes positive real values, the higher-order EPs can be achieved, and these values can be satisfied as
$$\begin{aligned} \frac{{{\Gamma _{21}} - {\Gamma _{31}}}}{{2{{\Delta ^{\prime}}_1} - {\Delta _3}}} &= \left\{ {\begin{array}{c} {\tan (\frac{\pi }{3} + \frac{{2\pi }}{3}n)}\\{\tan (\frac{{2\pi }}{3}m)} \end{array}} \right.(n,m \in Z),\\ \frac{{{{\Delta ^{\prime}}_1} - 2{\Delta _3}}}{{{\Gamma _{21}} - {\Gamma _{31}}}} &= \left\{ {\begin{array}{c} {\tan (\frac{\pi }{6} + \frac{2}{3}k\pi )}\\{\tan (\frac{\pi }{2} + \frac{2}{3}l\pi )} \end{array}} \right.(k,l \in Z). \end{aligned}$$

Based on Eq. (16), the relevant parameters become ${\Delta ^{\prime}_1} ={-} {\Delta _3}({\Delta _3} \ne 0)$ and ${\Gamma _{21}} = {\Gamma _{31}} - 3\sqrt 3 {\Delta _3}$.

In Fig. 6(a) and 6(e), we plot the eigenvalue surfaces of real and imaginary parts in the multi-coherent channels with the two dressing fields parameters Ω131 and Ω331. Considering the detuning of dressing fields, the real and imaginary parts of eigenvalues can be plotted with the different parameters Ω1 = 0.4Γ31, 0.5Γ31 and 0.6Γ31. The curves of the red solid line, blue and black dashed line represent the three complex eigenvalues of δ1, δ2 and δ3 in the ${\cal N}{\cal C}{\cal D}{\cal D}$ scheme. For Ω1 = 0.4Γ31, the three eigenvalues occur level repulsion in the real and imaginary parts, seen in Fig. 6(b) and Fig. 6(f). At the critical condition Ω1 = 0.5Γ31, the three complex eigenvalues coalesce one value at the Ω3 = 0.5Γ31, corresponding to the higher-order EP in Fig. 6(c) and Fig. 6(g). With increasing the Ω1 = 0.6Γ31, the three eigenvalues occur level repulsion again in Fig. 6(d) and Fig. 6(h). It is interesting to see, in contrast to the ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme, when continuously altering the parameters, only the higher-order EP can exist, but the merger of the two eigenvalues, corresponding to ordinary EPs, does not. This exactly indicates the high-order dressing property in the ${\cal N}{\cal C}{\cal D}{\cal D}$ scheme. As here the two dressing fields E1 and E3 co-modify the atoms in a hierarchy order, and it cannot be reduced the ordinary EPs case, which is quite different from all the previous coupling regime. And, thus, we can think this corresponds to the genuine higher-order EP.

 figure: Fig. 6.

Fig. 6. Exceptional points in the ELC-FWM processes under the doubly dressing ${\cal N}{\cal C}{\cal D}{\cal D}$ scheme. (a) and (e) The eigenvalue surfaces of real and imaginary parts with the two dressing fields parameters Ω131 and Ω331. (b) and (f) The real and imaginary part of the eigenvalues with Ω1 = 0.4Γ31. (c) and (g) Ω1 = 0.5Γ31, (d) and (h) Ω1 = 0.6Γ31

Download Full Size | PDF

4. Coherent dressing control of entanglement

The presence of multiple coexisting FWM processes within the ELC process can be attributed to an increase in coherent channels via the dressing effect. This dressing effect leads to the emergence of multiple resonance peaks and the subsequent appearance of multiple frequency modes in the original output mode ES1, ES2, and ES3. Consequently, it is possible to modify the nonlinear susceptibility of the ELC-FWM process to exert control over the multipartite entanglement within the non-Hermitian system. As shown in Fig. 7, the normalized real parts of the third-order density matrix are plotted to reflect the influence of the dressing effect on the field ES2. Figure 7(a) illustrates the third-order density matrix influenced by the single dressing effect and depicts the presence of two resonance peaks as the deviation δ is varied from -40 MHz to 40 MHz. The resonance positions of the two coherent channels are (I) ${\delta _{RE( - )}} = ({\Delta _1} - \sqrt {\Delta _1^2 + 4{\Gamma _{21}}{\Gamma _{23}} + 4\Omega _1^2} )/2$ and (II)${\delta _{RE( + )}} = ({\Delta _1} + \sqrt {\Delta _1^2 + 4{\Gamma _{21}}{\Gamma _{23}} + 4\Omega _1^2} )/2$.

 figure: Fig. 7.

Fig. 7. The normalized the real parts of the third-order density matrix intensity under the single and doubly dressing effect. (a) The condition of the existing single-dressing effect with two resonance peaks (I) and (II). (b) The condition of the doubly-dressing effect of the ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme with four resonance peaks (I), (II), (III), and (IV). (c) The condition of the doubly-dressing effect of the ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme and (d) ${\cal N}{\cal C}{\cal D}{\cal D}$ scheme with three resonance peaks (I), (II) and (III).

Download Full Size | PDF

Figure 7(b) represents four resonance peaks (I), (II), (III) and (IV) of the third-order density matrix under the doubly-dressing effect in the ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme. Additionally, Fig. 7(c) and 7(d) exhibit the three resonance peaks (I), (II) and (III) under the doubly-dressing effect in the ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme and the ${\cal N}{\cal C}{\cal D}{\cal D}$ scheme, respectively. It is worth noting that analogous outcomes to those observed in the real part can also be identified when examining the imaginary part of the eigenvalues across multiple coherent channels. Attributed to the multi-dressing effect, which causes additional splitting of energy levels, it generates multiple coexisting FWM channels within the ELC-FWM scheme. Consequently, the system exhibits multiple frequency modes for each output signal. These additional frequency modes not only construct multimode entanglement but also enhance the system's capacity for quantum information transmission [62].

We use the PPT criterion to describe the entanglement properties of the doubly-dressing effect and theoretically investigate the entanglement of the tripartite in the ELC-FWM system. Here, we consider the multipartite entanglement of two partitions in the type of ${\hat{a}_i}|{\hat{a}_j}$ and ${\hat{a}_i}|({\hat{a}_j},{\hat{a}_k})$(i, j, k = 1, 2, 3). We set gain G2 = 1.3, and vary the deviation δ to trace out the entanglement of the partition of three output modes.

In Fig. 8, the PPT values below 0 indicate the existence of entanglement among the output modes of the ELC-FWM process. Each of the rectangle pink regions in Fig. 8 corresponds to the coherent channels in Fig. 7, where the parametric gain is maximized. According to Eq. (2), the dressing effect can directly modulate the parametric gain of the ELC-FWM process. Consequently, the entanglement characteristics of the FWM system can be actively controlled in the preparation of entangled sources. Figure 8(a) and 8(b) illustrate the modulation of the bipartite and tripartite entanglement under the single dressing effect with corresponding coherent channels. Similarly, Figs. 8(c)-8(h) demonstrates the variation trend of the multipartite entanglement under the doubly-dressing effect. The absence of entanglement between the type of ${\hat{a}_1}|{\hat{a}_3}$ is not given in the three doubly-dressing schemes.

 figure: Fig. 8.

Fig. 8. Entanglement criteria in the single and doubly-dressing effect with three output modes. (a) and (b) The PPT value of the single dressing. (c) and (d) The ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme under the doubly-dressing effect. (e) and (f) The ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme. (g) and (h) The ${\cal N}{\cal C}{\cal D}{\cal D}$ scheme. The gray region shows the separability state, and below it is the entanglement state. The rectangle pink region (I), (II), (III), and (IV) indicates the change in the PPT criterion for each coherent channel.

Download Full Size | PDF

5. Generalized N-dressing effect

We illustrate the N-dressing fields interact with the cascade energy level of the atomic system in Fig. 9(a). Each of the dressing fields Edn couples the ground state |1〉 and excited state |3〉. Under the approximation condition, the N-dressing PA-FWM process can be considered as a superposition of the multiple dressing effects. As a result, the generalized Hamiltonian describing the atomic interaction in the N-dressing cascade energy level system can be expressed

$$\begin{aligned} {{\mathbf H}_{iN}} &={-} \hbar (\sum\limits_{n = 1}^N {{\Delta _n}} |3 \rangle \left\langle 3 \right|+ \sum\limits_{n = 1}^N {({{\Delta _n} - {\Delta _{s1}}} )} |2 \rangle \left\langle 2 \right|)\\ & + \hbar (\sum\limits_{n = 1}^N {({{\Delta _n} - {\Delta _{s1}} + {{\Delta^{\prime}}_1}} )} |4 \rangle \left\langle 4 \right|)\\ & + \hbar (\sum\limits_{n = 1}^N {{\Omega _n}} |3 \rangle \left\langle 1 \right|+ {\Omega _{s1}}|3 \rangle \left\langle 2 \right|))\\ & + \hbar ({\Omega _{s2}}|4 \rangle \left\langle 1 \right|+ {\Omega _1}|4 \rangle \left\langle 2 \right|+ H.c.), \end{aligned}$$
where Δn (n = 1, …N) is the detuning of the nth dressing field, and Ωn denotes the corresponding Rabi frequency of the dressing fields. According to the master equation in the atomic ensemble, the perturbation chain of the multi-dressing effect in the PA-FWM process can be obtained
$$\rho _{11}^{(0 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{31}^{(1 )}\buildrel {{\omega _{S1}}} \over \longrightarrow \rho _{2D\sum\limits_{n = 1}^N {\Omega n \pm } }^{(2 )}\buildrel {{\omega _1}} \over \longrightarrow \rho _{3({S2} )}^{(3 )}. $$

 figure: Fig. 9.

Fig. 9. The energy level diagram and theoretically calculated normalized the real parts of the third-order density matrix with the multi-dressing fields in the ELC-FWM processes. (a) the energy level diagram of interaction with the multilevel atomic system under the N-dressing fields. (b)-(e) the multiple coherent channels with a different number of multi-dressing fields, i.e., the number of dressing fields is N = 3, N = 4, N = 5, and N = 6.

Download Full Size | PDF

The subscripts $2D\sum\limits_{n = 1}^N {\Omega n \pm }$ denote the second-order density matrix element $\rho _{21}^{(2)}$ dressed by the N-dressing fields. Multiple coherent fields acting on the energy level |3〉 simultaneously constitute the cascade multi-dressing structure. Subsequently, the third-order density matrix element can be expressed

$$\rho _{31(NS2)}^{(3 )} = \frac{{ - i\Omega _1^2{\Omega _{S1}}}}{{({\Gamma _{31}} + i{\Delta _1})\left( {{\Gamma _{21}} + i{\delta_1}\textrm{ + }\frac{{\Omega _1^2}}{{{d_{31}}}} + \frac{{\Omega _3^2}}{{{{d^{\prime}}_{23}}}} + \sum\limits_{n = 4}^N {\frac{{\Omega _n^2}}{{{\Gamma _{31}} + i{\delta_1} + i{\Delta _i}}}} } \right)({\Gamma _{31}} + i{\delta _1} + i{{\Delta ^{\prime}}_1})}},$$
where the summation term $\sum\limits_{n = 4}^N {\frac{{\Omega _n^2}}{{{\Gamma _{31}} + i{\delta _1} + i{\Delta _i}}}}$ in the denominator represents the modification of the multi-dressing fields on the density matrix elements.

Under the interaction of the multi-dressing effect, the atomic coherence induces the multi-channel nonlinear optical processes in the multilevel system. As a result, the theoretically calculated normalized real parts of the third-order density matrix show the rich multiple resonance peaks structure in Fig. 9(b)-9(e). Here, we consider the cases of the multi-dressing fields N = 3, N = 4, N = 5, and N = 6. Furthermore, the coherent channels can be extended by simply adding the number of control fields without altering the overall quantum system structure.

In Fig. 10, the triple partitions of the entanglement criterion of the ECL-FWM process can be further optimized through the utilization of multiple dressing fields (ie., N = 3, 4, 5, 6). The dotted line indicates the position corresponding to the resonance peaks as in Fig. 9. Therefore, the multipartite entanglement of the ELC-FWM process can be effectively modulated by increasing the number of dressing fields. The atomic coherence control using multi-dressing fields can further extend the entanglement modes, which provides an effective and scalable method to improve the quantum information capacity in non-Hermitian system.

 figure: Fig. 10.

Fig. 10. Entanglement criterion in the multi-dressing coherent control with three output modes (a) the positivity under transposition (PPT) criterion under the number of N = 3 dressing fields. (b)-(d) N = 4,5,6 dressing fields. The dotted lines indicate the position of the multichannel regulation.

Download Full Size | PDF

6. Conclusion

In summary, our study demonstrates the capability of the ELC-FWM process to achieve versatile EPs and higher-order EPs, as well as to establish scalable and all-optically controlled multimode entanglement, based on the multi-dressing effects of atomic coherence. Remarkably, these achievements are obtained without the need for modifying the experimental optics architecture or introducing artificial photonic structures in the non-Hermitian systems. Furthermore, we employ the PPT criterion to characterize the multipartite entanglement under single- and doubly-dressing as well the generalized N-dressing fields. The results reveal that the properties of multipartite entanglement can be actively, flexibly, and continuously controlled around the EPs. This research opens up new avenues for exploring non-Hermitian physics through the integration of light-matter coupling, nonlinear optics, and entanglement generation, thereby paving the way for non-Hermiticity-enhanced quantum information processing.

Funding

National Natural Science Foundation of China (11904279, 12074303, 12074306, 12174302, 61975159, 62022066); Natural Science Foundation of Shandong Province (ZR2022LLZ004); Shaanxi Key Science and Technology Innovation Team Project (2021TD-56).

Disclosures

The authors declare no competing interests.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Supplemental document

See Supplement 1 for supporting content.

References

1. C. M. Bender and S. Boettcher, “Real spectra in non-Hermitian Hamiltonians having PT symmetry,” Phys. Rev. Lett. 80(24), 5243–5246 (1998). [CrossRef]  

2. C. M. Bender, S. Boettcher, and P. N. Meisinger, “PT-symmetric quantum mechanics,” J. Math. Phys. 40(5), 2201–2229 (1999). [CrossRef]  

3. R. El-Ganainy, K. Makris, D. Christodoulides, and Z. H. Musslimani, “Theory of coupled optical PT-symmetric structures,” Opt. Lett. 32(17), 2632–2634 (2007). [CrossRef]  

4. M.-A. Miri and A. Alù, “Exceptional points in optics and photonics,” Science 363(6422), eaar7709 (2019). [CrossRef]  

5. ŞK Özdemir, S. Rotter, F. Nori, and L. Yang, “Parity–time symmetry and exceptional points in photonics,” Nat. Mater. 18(8), 783–798 (2019). [CrossRef]  

6. E. J. Bergholtz, J. C. Budich, and F. K. Kunst, “Exceptional topology of non-Hermitian systems,” Rev. Mod. Phys. 93(1), 015005 (2021). [CrossRef]  

7. I. Mandal and E. J. Bergholtz, “Symmetry and higher-order exceptional points,” Phys. Rev. Lett. 127(18), 186601 (2021). [CrossRef]  

8. Z. Lin, A. Pick, M. Lončar, and A. W. Rodriguez, “Enhanced spontaneous emission at third-order Dirac exceptional points in inverse-designed photonic crystals,” Phys. Rev. Lett. 117(10), 107402 (2016). [CrossRef]  

9. H. Hodaei, A. U. Hassan, S. Wittek, H. Garcia-Gracia, R. El-Ganainy, D. N. Christodoulides, and M. Khajavikhan, “Enhanced sensitivity at higher-order exceptional points,” Nature 548(7666), 187–191 (2017). [CrossRef]  

10. B. Zhen, C. W. Hsu, Y. Igarashi, L. Lu, I. Kaminer, A. Pick, S.-L. Chua, J. D. Joannopoulos, and M. Soljačić, “Spawning rings of exceptional points out of Dirac cones,” Nature 525(7569), 354–358 (2015). [CrossRef]  

11. S.-B. Lee, J. Yang, S. Moon, S.-Y. Lee, J.-B. Shim, S. W. Kim, J.-H. Lee, and K. An, “Observation of an exceptional point in a chaotic optical microcavity,” Phys. Rev. Lett. 103(13), 134101 (2009). [CrossRef]  

12. J. Zhu, ŞK Özdemir, L. He, and L. Yang, “Controlled manipulation of mode splitting in an optical microcavity by two Rayleigh scatterers,” Opt. Express 18(23), 23535–23543 (2010). [CrossRef]  

13. W. Chen, Ş Kaya Özdemir, G. Zhao, J. Wiersig, and L. Yang, “Exceptional points enhance sensing in an optical microcavity,” Nature 548(7666), 192–196 (2017). [CrossRef]  

14. S. Klaiman, U. Günther, and N. Moiseyev, “Visualization of branch points in PT-symmetric waveguides,” Phys. Rev. Lett. 101(8), 080402 (2008). [CrossRef]  

15. P. Peng, W. Cao, C. Shen, W. Qu, J. Wen, L. Jiang, and Y. Xiao, “Anti-parity–time symmetry with flying atoms,” Nat. Phys. 12(12), 1139–1145 (2016). [CrossRef]  

16. Z. Zhang, Y. Zhang, J. Sheng, L. Yang, M.-A. Miri, D. N. Christodoulides, B. He, Y. Zhang, and M. Xiao, “Observation of parity-time symmetry in optically induced atomic lattices,” Phys. Rev. Lett. 117(12), 123601 (2016). [CrossRef]  

17. C. E. Rüter, K. G. Makris, R. El-Ganainy, D. N. Christodoulides, M. Segev, and D. Kip, “Observation of parity–time symmetry in optics,” Nat. Phys. 6(3), 192–195 (2010). [CrossRef]  

18. A. Regensburger, C. Bersch, M.-A. Miri, G. Onishchukov, D. N. Christodoulides, and U. Peschel, “Parity–time synthetic photonic lattices,” Nature 488(7410), 167–171 (2012). [CrossRef]  

19. B. Peng, ŞK Özdemir, F. Lei, F. Monifi, M. Gianfreda, G. L. Long, S. Fan, F. Nori, C. M. Bender, and L. Yang, “Parity–time-symmetric whispering-gallery microcavities,” Nat. Phys. 10(5), 394–398 (2014). [CrossRef]  

20. H. Benisty, A. Degiron, A. Lupu, A. De Lustrac, S. Chénais, S. Forget, M. Besbes, G. Barbillon, A. Bruyant, and S. Blaize, “Implementation of PT symmetric devices using plasmonics: principle and applications,” Opt. Express 19(19), 18004–18019 (2011). [CrossRef]  

21. R. Fleury, D. Sounas, and A. Alù, “An invisible acoustic sensor based on parity-time symmetry,” Nat. Commun. 6(1), 5905 (2015). [CrossRef]  

22. K. Ding, G. Ma, M. Xiao, Z. Zhang, and C. T. Chan, “Emergence, coalescence, and topological properties of multiple exceptional points and their experimental realization,” Phys. Rev. X 6(2), 021007 (2016). [CrossRef]  

23. H. Xu, D. Mason, L. Jiang, and J. Harris, “Topological energy transfer in an optomechanical system with exceptional points,” Nature 537(7618), 80–83 (2016). [CrossRef]  

24. J. Zhang, B. Peng, ŞK Özdemir, K. Pichler, D. O. Krimer, G. Zhao, F. Nori, Y.-X. Liu, S. Rotter, and L. Yang, “A phonon laser operating at an exceptional point,” Nat. Photonics 12(8), 479–484 (2018). [CrossRef]  

25. J. Wiersig, “Sensors operating at exceptional points: General theory,” Phys. Rev. A 93(3), 033809 (2016). [CrossRef]  

26. M. Brandstetter, M. Liertzer, C. Deutsch, P. Klang, J. Schöberl, H. E. Türeci, G. Strasser, K. Unterrainer, and S. Rotter, “Reversing the pump dependence of a laser at an exceptional point,” Nat. Commun. 5(1), 4034 (2014). [CrossRef]  

27. B. Peng, Ş Özdemir, S. Rotter, H. Yilmaz, M. Liertzer, F. Monifi, C. Bender, F. Nori, and L. Yang, “Loss-induced suppression and revival of lasing,” Science 346(6207), 328–332 (2014). [CrossRef]  

28. H. Hodaei, M.-A. Miri, M. Heinrich, D. N. Christodoulides, and M. Khajavikhan, “Parity-time–symmetric microring lasers,” Science 346(6212), 975–978 (2014). [CrossRef]  

29. L. Feng, Z. J. Wong, R.-M. Ma, Y. Wang, and X. Zhang, “Single-mode laser by parity-time symmetry breaking,” Science 346(6212), 972–975 (2014). [CrossRef]  

30. C. Fabre and N. Treps, “Modes and states in quantum optics,” Rev. Mod. Phys. 92(3), 035005 (2020). [CrossRef]  

31. P. van Loock and S. L. Braunstein, “Multipartite entanglement for continuous variables: a quantum teleportation network,” Phys. Rev. Lett. 84(15), 3482–3485 (2000). [CrossRef]  

32. S. Armstrong, M. Wang, R. Y. Teh, Q. Gong, Q. He, J. Janousek, H.-A. Bachor, M. D. Reid, and P. K. Lam, “Multipartite Einstein–Podolsky–Rosen steering and genuine tripartite entanglement with optical networks,” Nat. Phys. 11(2), 167–172 (2015). [CrossRef]  

33. M. Navascues, E. Wolfe, D. Rosset, and A. Pozas-Kerstjens, “Genuine network multipartite entanglement,” Phys. Rev. Lett. 125(24), 240505 (2020). [CrossRef]  

34. M. V. Larsen, X. Guo, C. R. Breum, J. S. Neergaard-Nielsen, and U. L. Andersen, “Deterministic multi-mode gates on a scalable photonic quantum computing platform,” Nat. Phys. 17(9), 1018–1023 (2021). [CrossRef]  

35. H.-S. Zhong, H. Wang, Y.-H. Deng, M.-C. Chen, L.-C. Peng, Y.-H. Luo, J. Qin, D. Wu, X. Ding, and Y. Hu, “Quantum computational advantage using photons,” Science 370(6523), 1460–1463 (2020). [CrossRef]  

36. S. Wengerowsky, S. K. Joshi, F. Steinlechner, H. Hübel, and R. Ursin, “An entanglement-based wavelength-multiplexed quantum communication network,” Nature 564(7735), 225–228 (2018). [CrossRef]  

37. Y. Cai, J. Roslund, G. Ferrini, F. Arzani, X. Xu, C. Fabre, and N. Treps, “Multimode entanglement in reconfigurable graph states using optical frequency combs,” Nat. Commun. 8(1), 15645 (2017). [CrossRef]  

38. Y. Liu, Y. Cai, B. Luo, J. Yan, M. Niu, F. Li, and Y. Zhang, “Collective multipartite Einstein-Podolsky-Rosen steering via cascaded four-wave mixing of rubidium atoms,” Phys. Rev. A 104(3), 033704 (2021). [CrossRef]  

39. X. Pan, S. Yu, Y. Zhou, K. Zhang, K. Zhang, S. Lv, S. Li, W. Wang, and J. Jing, “Orbital-angular-momentum multiplexed continuous-variable entanglement from four-wave mixing in hot atomic vapor,” Phys. Rev. Lett. 123(7), 070506 (2019). [CrossRef]  

40. X. Wang, S. Yu, S. Liu, K. Zhang, Y. Lou, W. Wang, and J. Jing, “Deterministic generation of large-scale hyperentanglement in three degrees of freedom,” Adv. Photonics Nexus 1(01), 016002 (2022). [CrossRef]  

41. Y. Zhang, A. W. Brown, and M. Xiao, “Opening four-wave mixing and six-wave mixing channels via dual electromagnetically induced transparency windows,” Phys. Rev. Lett. 99(12), 123603 (2007). [CrossRef]  

42. Y. Mei, Y. Zhou, S. Zhang, J. Li, K. Liao, H. Yan, S.-L. Zhu, and S. Du, “Einstein-podolsky-rosen energy-time entanglement of narrow-band biphotons,” Phys. Rev. Lett. 124(1), 010509 (2020). [CrossRef]  

43. C. Liu, J. Chen, S. Zhang, S. Zhou, Y.-H. Kim, M. M. Loy, G. K. Wong, and S. Du, “Two-photon interferences with degenerate and nondegenerate paired photons,” Phys. Rev. A 85(2), 021803 (2012). [CrossRef]  

44. R. M. Camacho, P. K. Vudyasetu, and J. C. Howell, “Four-wave-mixing stopped light in hot atomic rubidium vapour,” Nat. Photonics 3(2), 103–106 (2009). [CrossRef]  

45. D.-S. Ding, W. Zhang, S. Shi, Z.-Y. Zhou, Y. Li, B.-S. Shi, and G.-C. Guo, “Hybrid-cascaded generation of tripartite telecom photons using an atomic ensemble and a nonlinear waveguide,” Optica 2(7), 642–645 (2015). [CrossRef]  

46. M.-X. Dong, W. Zhang, Z.-B. Hou, Y.-C. Yu, S. Shi, D.-S. Ding, and B.-S. Shi, “Experimental realization of narrowband four-photon Greenberger–Horne–Zeilinger state in a single cold atomic ensemble,” Opt. Lett. 42(22), 4691–4694 (2017). [CrossRef]  

47. K. Zhang, W. Wang, S. Liu, X. Pan, J. Du, Y. Lou, S. Yu, S. Lv, N. Treps, and C. Fabre, “Reconfigurable hexapartite entanglement by spatially multiplexed four-wave mixing processes,” Phys. Rev. Lett. 124(9), 090501 (2020). [CrossRef]  

48. Y. Chen, Q. Zhu, X. Wang, Y. Lou, S. Liu, and J. Jing, “Deterministic all-optical quantum state sharing,” Adv. Photonics 5(2), 026006 (2023). [CrossRef]  

49. S. Liu, Y. Lou, and J. Jing, “Orbital angular momentum multiplexed deterministic all-optical quantum teleportation,” Nat. Commun. 11(1), 3875 (2020). [CrossRef]  

50. S. Liu, Y. Lou, Y. Chen, and J. Jing, “All-optical entanglement swapping,” Phys. Rev. Lett. 128(6), 060503 (2022). [CrossRef]  

51. V. Boyer, A. M. Marino, R. C. Pooser, and P. D. Lett, “Entangled images from four-wave mixing,” Science 321(5888), 544–547 (2008). [CrossRef]  

52. Y. Liu, J. Wei, M. Niu, Y. Lin, Z. Chen, J. Yan, B. Luo, F. Li, Y. Cai, and Y. Zhang, “Atomic-coherence-assisted multipartite entanglement generation with dressing-energy-level-cascaded four-wave mixing,” Phys. Rev. A 106(4), 043709 (2022). [CrossRef]  

53. J. Wen, S. Du, and M. H. Rubin, “Biphoton generation in a two-level atomic ensemble,” Phys. Rev. A 75(3), 033809 (2007). [CrossRef]  

54. K. Li, S. Li, C. Li, Y. Cai, F. Li, and Y. Zhang, “Dressing-Shaped Rabi Oscillation of Photon Pairs in Hot Atomic Ensemble,” Adv. Quantum Technol. 4(2), 2000098 (2021). [CrossRef]  

55. L. Zhao, Y. Su, and S. Du, “Narrowband biphoton generation in the group delay regime,” Phys. Rev. A 93(3), 033815 (2016). [CrossRef]  

56. B. Luo, J. Yan, Y. Jiang, S. Li, Y. Li, S. Zhang, F. Li, Y. Cai, and Y. Zhang, “Experimental realization of multimode nonlinear parametric amplification from cascading four-wave mixing of dressed atoms,” Opt. Express 31(4), 6982–6995 (2023). [CrossRef]  

57. H. Yan, S. Zhang, J. Chen, M. M. Loy, G. K. Wong, and S. Du, “Generation of narrow-band hyperentangled nondegenerate paired photons,” Phys. Rev. Lett. 106(3), 033601 (2011). [CrossRef]  

58. D. Zhang, C. Li, Z. Zhang, Y. Zhang, Y. Zhang, and M. Xiao, “Enhanced intensity-difference squeezing via energy-level modulations in hot atomic media,” Phys. Rev. A 96(4), 043847 (2017). [CrossRef]  

59. C. Li, B. Gu, D. Zhang, W. Chen, Z. Shen, G. Lan, and Y. Zhang, “Modulation gain and squeezing by dressed state in hot atomic system,” Laser Phys. Lett. 16(5), 055401 (2019). [CrossRef]  

60. Y. Liu, K. Li, W. Li, S. Li, Y. Cai, and Y. Zhang, “Dressing-Controlled Quantum Steering in Energy-Level Cascaded Parametric Amplified Four-Wave Mixing Processes,” Adv. Quantum Technol. 3(8), 2000029 (2020). [CrossRef]  

61. P. Li, H. Zheng, Y. Zhang, J. Sun, C. Li, G. Huang, Z. Zhang, Y. Li, and Y. Zhang, “Controlling the transition of bright and dark states via scanning dressing field,” Opt. Mater. 35(5), 1062–1070 (2013). [CrossRef]  

62. O. Kovalenko, Y.-S. Ra, Y. Cai, V. C. Usenko, C. Fabre, N. Treps, and R. Filip, “Frequency-multiplexed entanglement for continuous-variable quantum key distribution,” Photonics Res. 9(12), 2351–2359 (2021). [CrossRef]  

Supplementary Material (1)

NameDescription
Supplement 1       Supplementary materials for multipartite entanglement and density matrix equations

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (10)

Fig. 1.
Fig. 1. The scheme of the three modes is based on the energy-level-cascade four-wave mixing process. (a) The ELC-FWM processes in a single rubidium vapor. (b) Energy level diagram of the whole ELC-FWM system with two subsystems of FWM1 and FWM2. (c) The spatial structures of the two pump beams E1 and E2 with three output modes ES1, ES2 and ES3. (d) The diagram of the phase-matching conditions for the parametric amplified FWM1 and FWM2.
Fig. 2.
Fig. 2. Energy level diagram and EPs in the ELC-FWM processes under the single dressing field. (a) two dressed states splitting diagram with single dressing effect. The blue bidirectional dotted arrow indicates the dressing effect caused by the E1 field. (b) and (e) The eigenvalue surfaces of real and imaginary parts in the multi-coherent channels with the parameters Ω131 and frequency detuning Δ131. (c) and (f) The eigenvalues of the real and imaginary parts. The exceptional points occur at Ω131 = 0.5 with Δ1 = 0. (d) and (g) The level repulsion of the eigenvalues in the real and imaginary parts with Δ1 = 0.15Γ31.
Fig. 3.
Fig. 3. The doubly dressed states energy level diagram with two strong coherent fields E1 and E2. (a) The doubly dressed FWM of the parallel-cascade scheme with effecting on different density matrix element (b) The doubly dressing FWM of sequential-cascade scheme same as (a), but dressed same density matrix element. (c) The doubly dressed FWM of the nested-cascade scheme. (d) The entanglement scheme from multi-coherent channels of the parallel-cascade scheme (left), the other two schemes (right). (a)-(c) The blue and green bidirectional dotted arrow indicates the dressing effect by the fields E1 and E2.
Fig. 4.
Fig. 4. Exceptional points in the ELC-FWM processes under the doubly dressing ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme. (a) and (c) The eigenvalue surfaces of real and imaginary parts in the multi-coherent channels with the parameters Ω131 and Ω331; (b) and (d) the real parts and imaginary parts depending on the two parameters Ω131 and Ω331 with Ω1 = 0.5Γ31 and Ω3 = 0.5Γ31.
Fig. 5.
Fig. 5. Exceptional points in the ELC-FWM processes under the doubly dressing ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme. (a) and (e) The eigenvalue surfaces of the real and imaginary parts with the two dressing fields parameters Ω1/Γ31 and Ω3/Γ31. (b) and (f) The real and imaginary part of the eigenvalues with Ω1 = 0.1Γ31. (c) and (g) Ω1 = 0.22Γ31. (d) and (h) Ω1 = 0.35Γ31.
Fig. 6.
Fig. 6. Exceptional points in the ELC-FWM processes under the doubly dressing ${\cal N}{\cal C}{\cal D}{\cal D}$ scheme. (a) and (e) The eigenvalue surfaces of real and imaginary parts with the two dressing fields parameters Ω131 and Ω331. (b) and (f) The real and imaginary part of the eigenvalues with Ω1 = 0.4Γ31. (c) and (g) Ω1 = 0.5Γ31, (d) and (h) Ω1 = 0.6Γ31
Fig. 7.
Fig. 7. The normalized the real parts of the third-order density matrix intensity under the single and doubly dressing effect. (a) The condition of the existing single-dressing effect with two resonance peaks (I) and (II). (b) The condition of the doubly-dressing effect of the ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme with four resonance peaks (I), (II), (III), and (IV). (c) The condition of the doubly-dressing effect of the ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme and (d) ${\cal N}{\cal C}{\cal D}{\cal D}$ scheme with three resonance peaks (I), (II) and (III).
Fig. 8.
Fig. 8. Entanglement criteria in the single and doubly-dressing effect with three output modes. (a) and (b) The PPT value of the single dressing. (c) and (d) The ${\cal P}{\cal C}{\cal D}{\cal D}$ scheme under the doubly-dressing effect. (e) and (f) The ${\cal S}{\cal S}{\cal D}{\cal D}$ scheme. (g) and (h) The ${\cal N}{\cal C}{\cal D}{\cal D}$ scheme. The gray region shows the separability state, and below it is the entanglement state. The rectangle pink region (I), (II), (III), and (IV) indicates the change in the PPT criterion for each coherent channel.
Fig. 9.
Fig. 9. The energy level diagram and theoretically calculated normalized the real parts of the third-order density matrix with the multi-dressing fields in the ELC-FWM processes. (a) the energy level diagram of interaction with the multilevel atomic system under the N-dressing fields. (b)-(e) the multiple coherent channels with a different number of multi-dressing fields, i.e., the number of dressing fields is N = 3, N = 4, N = 5, and N = 6.
Fig. 10.
Fig. 10. Entanglement criterion in the multi-dressing coherent control with three output modes (a) the positivity under transposition (PPT) criterion under the number of N = 3 dressing fields. (b)-(d) N = 4,5,6 dressing fields. The dotted lines indicate the position of the multichannel regulation.

Equations (28)

Equations on this page are rendered with MathJax. Learn more.

PA - FWM1 d a ^ 1 d t = κ 1 a ^ 2 , d a ^ 2 d t = κ 1 a ^ 1 ,
PA - FWM2 d a ^ 2 d t = κ 2 a ^ 3 , d a ^ 3 d t = κ 2 a ^ 2 .
a ^ 1 o u t = G 1 a ^ 1 i n + g 1 a ^ 2 i n ,
a ^ 2 o u t = g 1 G 2 a ^ 1 i n + G 1 G 2 a ^ 2 i n + g 2 a ^ 3 i n ,
a ^ 3 o u t = g 1 g 2 a ^ 1 i n + G 1 g 2 a ^ 2 i n + G 2 a ^ 3 i n ,
H i = ( ( Δ 1 Δ s ) | 2 2 | + Δ 1 | 3 3 | ) + ( ( Δ 1 Δ s + Δ 1 ) | 4 4 | ) + ( Ω 1 | 3 1 | + Ω s 1 | 3 2 | ) + ( Ω s 2 | 4 1 | + Ω 1 | 4 2 | + H . c . ) ,
ρ ˙ = i [ H i , ρ ] Γ ρ ,
PA - FWM2 ρ 11 ( 0 ) ω 1 ρ 31 ( 1 ) ω S 1 ρ 21 ( 2 ) ω 1 ρ 31 ( S 2 ) ( 3 ) ,
PA - FWM2 ρ 22 ( 0 ) ω 1 ρ 32 ( 1 ) ω 2 ρ 12 ( 2 ) ω 1 ρ 32 ( S 1 ) ( 3 ) ,
PA - FWM2 ρ 22 ( 0 ) ω 1 ρ 32 ( 1 ) ω 2 ρ 12 ( 2 ) ω 3 ρ 32 ( S 3 ) ( 3 ) ,
PA - FWM2 ρ 11 ( 0 ) ω 3 ρ 31 ( 1 ) ω S 3 ρ 21 ( 2 ) ω 1 ρ 31 ( S 4 ) ( 3 ) ,
PA - FWM1 ρ 31 ( S 2 ) ( 3 ) = i Ω 1 2 Ω S 1 / d 31 d 21 d 31 ,
PA - FWM1 ρ 32 ( S 1 ) ( 3 ) = i Ω 1 2 Ω 2 / d 32 d 12 d 32 ,
PA - FWM2 ρ 31 ( S 2 ) ( 3 ) = i Ω 1 Ω 3 Ω S 3 / d 31 d 21 d 312 ,
PA - FWM2 ρ 32 ( S 3 ) ( 3 ) = i Ω 1 Ω 2 Ω 3 / d 32 d 12 d 32 ,
ρ 31 ( D S 2 ) ( 3 ) = i Ω 1 2 Ω S 1 d 31 ( d 21 + Ω 1 2 / d 23 ) d 31 ,
δ ± = i ( Γ 21 + Γ 23 ) + Δ 1 2 ± ( Δ 1 + i ( Γ 21 Γ 23 ) ) 2 4 + Ω 1 2 .
ρ 31 ( S 2 ) ( 3 ) = i Ω 1 2 Ω S 1 ( Γ 31 + i Δ 1 ) ( Γ 21 + i δ 1  +  Ω 1 2 d 23 ) ( Γ 31 + i δ 1 + i Δ 1 + Ω 3 2 d 11 ) ,
δ 1 ± = i ( Γ 21 + Γ 23 ) + Δ 1 ± ( Δ 1 + i ( Γ 21 Γ 23 ) ) 2 + 4 Ω 1 2 2 .
δ 2 ± = i ( Γ 11 + Γ 31 ) + Δ 3 2 Δ 1 ± ( Δ 3 + i ( Γ 11 Γ 31 ) ) 2 + 4 Ω 3 2 2 .
ρ 31 ( S 2 ) ( 3 ) = i Ω 1 2 Ω S 1 ( Γ 31 + i Δ 1 ) ( Γ 21 + i δ 1  +  Ω 1 2 d 31 + Ω 3 2 d 23 ) ( Γ 31 + i δ 1 + i Δ 1 ) ,
Ω 1 = ± ( i ( 2 Γ 31 Γ 23 Γ 21 ) 2 Δ 1 Δ 3 ) 3 / 2 3 3 i ( Γ 23 Γ 31 ) + ( Δ 1 + Δ 3 ) , Ω 3 = ± ( i ( Γ 21 2 Γ 23 + Γ 31 ) Δ 1 2 Δ 3 ) 3 / 2 3 3 i ( Γ 23 Γ 31 ) + ( Δ 1 + Δ 3 ) .
ρ 31 ( S 2 ) ( 3 ) = i Ω 1 2 Ω S 1 ( Γ 31 + i Δ 1 ) ( Γ 21 + i δ 1  +  Ω 1 2 / ( d 31 + Ω 3 2 d 21 ) ) ( Γ 31 + i δ 1 + i Δ 1 ) ,
Ω 1 = ± i ( i Γ 21 i Γ 31 + 2 Δ 1 Δ 3 ) 3 / 2 3 3 Δ 1 Δ 3 , Ω 3 = ± i ( Γ 21 Γ 31 + i Δ 1 2 i Δ 3 ) 3 / 2 3 3 i Δ 3 i Δ 1 .
Γ 21 Γ 31 2 Δ 1 Δ 3 = { tan ( π 3 + 2 π 3 n ) tan ( 2 π 3 m ) ( n , m Z ) , Δ 1 2 Δ 3 Γ 21 Γ 31 = { tan ( π 6 + 2 3 k π ) tan ( π 2 + 2 3 l π ) ( k , l Z ) .
H i N = ( n = 1 N Δ n | 3 3 | + n = 1 N ( Δ n Δ s 1 ) | 2 2 | ) + ( n = 1 N ( Δ n Δ s 1 + Δ 1 ) | 4 4 | ) + ( n = 1 N Ω n | 3 1 | + Ω s 1 | 3 2 | ) ) + ( Ω s 2 | 4 1 | + Ω 1 | 4 2 | + H . c . ) ,
ρ 11 ( 0 ) ω 1 ρ 31 ( 1 ) ω S 1 ρ 2 D n = 1 N Ω n ± ( 2 ) ω 1 ρ 3 ( S 2 ) ( 3 ) .
ρ 31 ( N S 2 ) ( 3 ) = i Ω 1 2 Ω S 1 ( Γ 31 + i Δ 1 ) ( Γ 21 + i δ 1  +  Ω 1 2 d 31 + Ω 3 2 d 23 + n = 4 N Ω n 2 Γ 31 + i δ 1 + i Δ i ) ( Γ 31 + i δ 1 + i Δ 1 ) ,
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.