betaFeedback

WorksPhysics and Astronomy

Excitation-Selective Control of Protected Subradiant Antiresonances

Abstract

Subradiant cavity antiresonances exhibit narrow spectral features, yet controlling their nonlinear response without perturbing the protected resonance remains challenging. Here we demonstrate excitation-selective control of a subradiant antiresonance in a driven two-emitter cavity-QED system. An auxiliary collective channel leaves the antisymmetric single-excitation state unaffected while shifting the doubly excited state, introducing tunable excitation anharmonicity without altering the linear antiresonance. This separation enables control of two-photon excitation pathways and photon correlations. Independently, a spatial coupling phase tunes optical access to the subradiant state without changing its intrinsic decay rate. We find that the interferometric phase information per incident photon is optimized when cavity-induced readout approximately matches intrinsic subradiant decay, while steady-state atomic entanglement peaks at an intermediate coupling phase. We further extend phase-selective optical control to collective subradiant modes in correlated emitter arrays. These results establish collective symmetry and spatial mode matching as distinct controls of nonlinear photon correlations, optical phase information, and atomic entanglement.

Introduction.—Interference between collective radiative pathways gives rise to superradiance and subradiance with decay rates markedly different from those of isolated emitters [1, 2, 3, 4]. Subradiant states support long-lived optical coherence, but their weak radiative coupling often limits optical excitation and readout [5, 6, 7, 8, 9, 10, 11, 12, 13]. Ordered emitter arrays offer spatial control over collective optical modes, enabling cooperative scattering, selective radiance, and long-lived collective excitations [14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25]. Their collective response is sensitive to geometry, atomic motion, and disorder [26, 27, 28, 29, 30]. Cavity and nanophotonic platforms provide additional control over collective excitations and their radiative properties [31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41]. In particular, cavity antiresonances provide a sensitive probe of the emitter response [42], while coupling to subradiant collective modes can yield narrow antiresonances with enhanced cooperativity and steep phase dispersion [43]. These developments raise a question beyond linear spectroscopy: can higher-excitation dynamics be controlled without perturbing a subradiant antiresonance?

This question is nontrivial because subradiant protection, optical accessibility, and nonlinear response involve distinct aspects of collective dynamics. A long-lived excitation may couple weakly to an optical probe, whereas stronger cavity coupling can deepen the antiresonance and suppress transmission. Thus, steep phase dispersion alone does not guarantee greater phase information per incident photon, an important consideration for collective optical metrology [5, 44, 45, 46, 48, 49] [47]. Beyond linear response, photon correlations depend on higher-excitation manifolds and their interference pathways, which are not determined by the single-excitation spectrum [12, 50, 51, 52]. Strong photon bunching can arise when single-photon excitation is suppressed relative to higher-order processes, as recently explored in cavity-mediated subradiance [40]. Collective dissipation, spatially selective coupling, and cavity-mediated interactions provide routes to controlling dark states and collective excitation dynamics [25, 53, 54, 55, 56, 57, 58, 59, 60]. Related reservoir-engineering approaches have also enabled the preparation and stabilization of entangled states [61, 62, 63, 64, 65]. The central challenge is to modify higher-excitation pathways without disturbing the collective state responsible for the linear antiresonance.

Here we demonstrate excitation-selective control in a driven two-emitter cavity-QED system. The emitters couple to a primary cavity with a tunable relative phase, while an auxiliary cavity mediates spin-exchange interactions (SEI) and collective dissipation. The antisymmetric single-excitation state is dark to the auxiliary channel, whereas the symmetric single-excitation and doubly excited states remain coupled. The SEI shifts the latter states while leaving the antisymmetric state unaffected, introducing a tunable anharmonicity into the excitation ladder. When the primary cavity selectively addresses the antisymmetric state, the linear antiresonance remains unchanged, while the modified two-excitation spectrum allows photon correlations to be tuned. Independently, the primary-cavity coupling phase controls optical access to the subradiant state without altering its intrinsic decay rate. We further investigate the resulting phase-sensitive optical response and steady-state atomic entanglement, and extend phase-selective optical control to collective subradiant modes in correlated emitter arrays.

Model and excitation-selective protection.—We consider NN identical two-level emitters coupled to a driven primary cavity and an auxiliary cavity Figure 1(a). We first establish the excitation-selective mechanism for two emitters and then extend the linear optical response to correlated emitter arrays. The primary cavity mode a^\hat{a} couples to the emitters with position-dependent strengths gj=gcos⁡[(j−1)ϕ]g_j=g\cos[(j-1)\phi], where ϕ\phi is the relative coupling phase, while the auxiliary mode b^\hat{b} couples uniformly with strength gBg_B. In the dispersive regime, adiabatic elimination of the auxiliary cavity induces an effective infinite-range SEI and collective dissipation. Both are governed by the symmetric collective lowering operator J^B−=∑j=1Nσ^j−\hat{J}_{B}^{-}=\sum_{j=1}^{N}\hat{\sigma}_{j}^{-}, where σ^j−=∣g⟩j⟨e∣j\hat{\sigma}_{j}^{-}=\lvert g\rangle_j\langle e\rvert_j. The coherent interaction VJ^B+J^B−V\hat{J}_{B}^{+}\hat{J}_{B}^{-} includes both spin exchange between emitters and collective energy shifts, with V=−gB2Δb/(Δb2+κb2)V=-g_B^2\Delta_b/(\Delta_b^2+\kappa_b^2) and ΓB=gB2κb/(Δb2+κb2)\Gamma_B=g_B^2\kappa_b/(\Delta_b^2+\kappa_b^2). Here Δb\Delta_b is the auxiliary-cavity detuning from the atomic transition, κb\kappa_b is the auxiliary-cavity field decay rate, and ΓB\Gamma_B characterizes the induced collective dissipation [31, 66]. Then the resulting effective Hamiltonian reads (ℏ=1\hbar=1)

Cavity-QED model and excitation-selective protection

Figure 1. Cavity-QED model and excitation-selective protection. (a) An array of NN two-level emitters coupled to a driven primary cavity and a far-detuned auxiliary cavity, with spatially correlated atomic decay. (b) Collective-state level structure for N=2N=2 at ϕ=π\phi=\pi. The primary cavity (blue arrows) couples the ∣G⟩↔∣A⟩↔∣E⟩\lvert G\rangle\leftrightarrow\lvert A\rangle\leftrightarrow\lvert E\rangle ladder, while the auxiliary channel (orange dashed arrows) couples the symmetric sector, leaving ∣A⟩\lvert A\rangle protected but modifying the two-excitation response.

H^eff=Δaa^†a^+δ∑j=1Nσ^j+σ^j−+VJ^B+J^B−+a^†∑j=1Ngjσ^j−+H.c.+iη(a^†−a^),(1)\hat{H}_{\mathrm{eff}}=\Delta_a\hat{a}^{\dagger}\hat{a}+\delta\sum_{j=1}^{N}\hat{\sigma}_{j}^{+}\hat{\sigma}_{j}^{-}+V\hat{J}_{B}^{+}\hat{J}_{B}^{-} +\hat{a}^{\dagger}\sum_{j=1}^{N}g_j\hat{\sigma}_{j}^{-}+\mathrm{H.c.}+i\eta(\hat{a}^{\dagger}-\hat{a}), \tag*{(1)}

where Δa\Delta_a and δ\delta are the primary-cavity and atomic detunings, respectively.

Including complete dissipations of system, the dynamics of density matrix is described by the master equation

ρ^˙=−i[H^eff,ρ^]+2κD[a^]ρ^+2ΓBD[J^B−]ρ^+∑i,j=1Nγij(2σ^i−ρ^σ^j+−{σ^j+σ^i−,ρ^}),(2)\dot{\hat{\rho}}=-i[\hat{H}_{\mathrm{eff}},\hat{\rho}]+2\kappa\mathcal{D}[\hat{a}]\hat{\rho}+2\Gamma_B\mathcal{D}[\hat{J}_{B}^{-}]\hat{\rho} +\sum_{i,j=1}^{N}\gamma_{ij}\left(2\hat{\sigma}_{i}^{-}\hat{\rho}\hat{\sigma}_{j}^{+}-\{\hat{\sigma}_{j}^{+}\hat{\sigma}_{i}^{-},\hat{\rho}\}\right), \tag*{(2)}

where D[O^]ρ^=O^ρ^O^†−{O^†O^,ρ^}/2\mathcal{D}[\hat{O}]\hat{\rho}=\hat{O}\hat{\rho}\hat{O}^{\dagger}-\{\hat{O}^{\dagger}\hat{O},\hat{\rho}\}/2 denotes the Lindblad type of dissipation and κ\kappa is the primary-cavity field decay rate. To account for spatially correlated atomic decay,

as in photonic-crystal waveguides (PCWs) [34, 67, 68], we adopt the Markovian decay matrix γij=γpδij+γ1Dcos⁡[ϕg(i−j)]\gamma_{ij}=\gamma_p\delta_{ij}+\gamma_{1\mathrm{D}}\cos[\phi_g(i-j)] for equally spaced emitters coupled to a bidirectional guided reservoir [20, 69]. Here γp\gamma_p and γ1D\gamma_{1\mathrm{D}} denote the independent and guided-mode decay rates, respectively, and ϕg\phi_g is the propagation phase between adjacent emitters, giving a single-emitter coherence decay rate γ=γp+γ1D\gamma=\gamma_p+\gamma_{1\mathrm{D}}. We neglect coherent waveguide-mediated exchange to isolate correlated dissipation, with ϕ\phi and ϕg\phi_g specifying the primary-cavity coupling profile and reservoir correlations, respectively.

We first consider two emitters (N=2N=2), for which the excitation-selective mechanism is illustrated in Figure 1(b). The collective atomic states are ∣G⟩=∣gg⟩\lvert G\rangle=\lvert gg\rangle, ∣E⟩=∣ee⟩\lvert E\rangle=\lvert ee\rangle, and ∣S,A⟩=(∣eg⟩±∣ge⟩)/2\lvert S,A\rangle=(\lvert eg\rangle\pm\lvert ge\rangle)/\sqrt{2}. Introducing S^−=(σ^1−+σ^2−)/2\hat{S}^{-}=(\hat{\sigma}_{1}^{-}+\hat{\sigma}_{2}^{-})/\sqrt{2} and A^−=(σ^1−−σ^2−)/2\hat{A}^{-}=(\hat{\sigma}_{1}^{-}-\hat{\sigma}_{2}^{-})/\sqrt{2}, the primary-cavity interaction takes the form a^†(gSS^−+gAA^−)+H.c.\hat{a}^{\dagger}(g_S\hat{S}^{-}+g_A\hat{A}^{-})+\mathrm{H.c.}, where gS,A=g(1±cos⁡ϕ)/2g_{S,A}=g(1\pm\cos\phi)/\sqrt{2}. The correlated reservoir gives γc=γ1Dcos⁡ϕg\gamma_c=\gamma_{1\mathrm{D}}\cos\phi_g, with collective coherence decay rates γA=γ−γc\gamma_A=\gamma-\gamma_c, γS=γ+γc+2ΓB\gamma_S=\gamma+\gamma_c+2\Gamma_B, and γE=2(γ+ΓB)\gamma_E=2(\gamma+\Gamma_B) for the states ∣A⟩\lvert A\rangle, ∣S⟩\lvert S\rangle, and ∣E⟩\lvert E\rangle, respectively. At ϕ=π\phi=\pi, gS=0g_S=0 and gA=2gg_A=\sqrt{2}g, so the primary cavity couples only to the antisymmetric state within the single-excitation manifold. Since this state is decoupled from the auxiliary channel, its resonance frequency and intrinsic decay rate are unaffected by VV and ΓB\Gamma_B, leaving the corresponding linear cavity ant iresonance unchanged [7, 42, 43]. For ϕ≠π\phi\ne\pi, the symmetric state also contributes to the linear response.

This protection follows directly from the collective selection rules J^B−∣A⟩=0\hat{J}_{B}^{-}\lvert A\rangle=0, J^B−∣S⟩=2∣G⟩\hat{J}_{B}^{-}\lvert S\rangle=\sqrt{2}\lvert G\rangle, and J^B−∣E⟩=2∣S⟩\hat{J}_{B}^{-}\lvert E\rangle=\sqrt{2}\lvert S\rangle. The auxiliary-induced non-Hermitian term H^nh,B=(V−iΓB)J^B+J^B−\hat{H}_{\mathrm{nh},B}=(V-i\Gamma_B)\hat{J}_{B}^{+}\hat{J}_{B}^{-} therefore leaves ∣A⟩\lvert A\rangle unaffected but shifts ∣S⟩\lvert S\rangle and ∣E⟩\lvert E\rangle by 2(V−iΓB)2(V-i\Gamma_B). At ϕ=π\phi=\pi, the primary cavity couples the ∣G⟩↔∣A⟩↔∣E⟩\lvert G\rangle\leftrightarrow\lvert A\rangle\leftrightarrow\lvert E\rangle ladder, whereas the auxiliary channel connects ∣E⟩\lvert E\rangle to ∣S⟩\lvert S\rangle Figure 1(b). The auxiliary interaction thus modifies the complex two-excitation spectrum and two-photon pathways without altering the protected linear antiresonance. This separation is enforced by collective symmetry rather than a fine-tuned cancellation between coherent and dissipative contributions. Unless otherwise specified, we use N=2N=2, g=γ=κg=\gamma=\kappa, V=κV=\kappa, ΓB=0.05κ\Gamma_B=0.05\kappa, γc=0.95γ\gamma_c=0.95\gamma, η=0.1κ\eta=0.1\kappa, and Δa=0\Delta_a=0.

**Symmetry-protected antiresonance and metrology.—**In the weak-drive limit, cavity transmission amplitude is t(δ,ϕ)=κ/[κ+iΔa+gS2/(γS+iδ+2iV)+gA2/(γA+iδ)]t(\delta,\phi)=\kappa/[\kappa+i\Delta_a+g_S^2/(\gamma_S+i\delta+2iV)+g_A^2/(\gamma_A+i\delta)]. Figure 2(a) shows how ϕ\phi controls the relative contributions of the symmetric and antisymmetric collective states to the cavity transmission. At ϕ=0\phi=0, only the symmetric state couples to cavity, producing an antiresonance near δ=−2V\delta=-2V. As ϕ\phi increases, the coupling to the antisymmetric state grows, giving rise to a second antiresonance near δ=0‘xref‘[Fig.2‘(c)].At\delta=0` {xref}`[Fig. 2`(c)]. At\phi=\pi$, the symmetric state decouples from cavity, and the transmission reduces to tπ(δ)=κ/[κ+iΔa+2g2/(γA+iδ)]t_\pi(\delta) = \kappa/[\kappa+ i\Delta_a + 2g^2/(\gamma_A + i\delta)]. Its independence from VV and ΓB\Gamma_B establishes a symmetry-protected antiresonance whose intrinsic resonance frequency and coherence decay rate remain unaffected by the auxiliary channel. The coupling phase thus tunes the cavity response without altering the intrinsic properties of the subradiant state [42, 43].

Figure 2. Phase-controlled subradiant antiresonance and optical response. (a) Cavity transmission T=∣t∣2T = |t|^2 versus detuning δ\delta and phase ϕ\phi. (b) Cavity-induced decay Γcav\Gamma_{\mathrm{cav}}, versus VV and ϕ\phi. The dashed line marks Γcav=γA\Gamma_{\mathrm{cav}} = \gamma_A. (c) Transmission spectra for different ϕ\phi. (d) Relative transmission phase Δθ=θ(δ)−θ(0)\Delta\theta= \theta(\delta) - \theta(0) at ϕ=π\phi= \pi for different γc\gamma_c.

To quantify the distinction between intrinsic protection and tunable cavity coupling, we consider the self-energy of the antisymmetric state. Near δ=0\delta= 0, the symmetric contribution provides a slowly varying background K(ϕ,V)=κ+iΔa+gS2/(γS+2iV)K(\phi, V) = \kappa+ i\Delta_a + g_S^2/(\gamma_S + 2iV), provided the two collective resonances are sufficiently separated. The corresponding local self-energy is

ΣA=gA2K≡Γcav−iΔcav,(3)\Sigma_A = \frac{g_A^2}{K} \equiv\Gamma_{\mathrm{cav}} - i\Delta_{\mathrm{cav}}, \tag*{(3)}

where Γcav=Re⁡ΣA\Gamma_{\mathrm{cav}} = \operatorname{Re}\Sigma_A and Δcav=−Im⁡ΣA\Delta_{\mathrm{cav}} = -\operatorname{Im}\Sigma_A describe the cavity-induced coherence decay and frequency shift, respectively. Figure 2(b) shows the ratio Γcav/γA\Gamma_{\mathrm{cav}}/\gamma_A, which characterizes the cavity-induced decay relative to intrinsic subradiant decay. For ϕ≪1\phi\ll1, gA≈gϕ2/(22)g_A \approx g\phi^2/(2\sqrt{2}), yielding Γcav≈(g2ϕ4/8)Re⁡(1/K0)\Gamma_{\mathrm{cav}} \approx(g^2\phi^4/8)\operatorname{Re}(1/K_0), with K0=K(0,V)K_0 = K(0,V). Thus, cavity-induced decay vanishes as ϕ4\phi^4 near ϕ=0\phi= 0, while γA\gamma_A remains unchanged. This provides independent control over cavity-induced response while preserving the intrinsic subradiant decay rate.

The protected subradiant state also exhibits a pronounced transmission-phase response. Figure 2(d) shows that increasing γc\gamma_c sharpens the phase dispersion near the antisymmetric antiresonance. At ϕ=π\phi= \pi and Δa=0\Delta_a = 0, the transmission phase θ=arg⁡t\theta= \arg t has a central slope ∂δθ∣δ=0=2g2/[γA(κγA+2g2)]\left.\partial_\delta\theta\right|_{\delta=0} = 2g^2/[\gamma_A(\kappa\gamma_A + 2g^2)]. For 2g2≫κγA2g^2 \gg\kappa\gamma_A, the slope approaches 1/γA1/\gamma_A, directly relating the phase response to the intrinsic decay rate of antisymmetric state. However, the steep phase dispersion occurs near suppressed transmission, so a large phase slope does not necessarily yield optimal measurement sensitivity. This motivates a Fisher-information analysis that accounts for both phase dispersion and transmitted photon flux.

The steep phase dispersion of the protected antiresonance offers enhanced phase sensitivity, but the suppressed transmission limits the available signal. To quantify this trade-off, we consider a perturbation λ\lambda that shifts the atomic detuning, δ(λ)=δ0+χλ\delta(\lambda) = \delta_0 + \chi\lambda, with χλ=∂δ/∂λ\chi_\lambda= \partial\delta/\partial\lambda. For a coherent probe and ideal homodyne detection of the phase quadrature, the classical Fisher information per incident photon is Iλ,θ(in)=4T(∂θ)2\mathcal{I}^{(\mathrm{in})}_{\lambda,\theta} = 4T(\partial_\theta)^2 [70, 71]. This expression accounts for the coherent phase response but excludes information from amplitude variations and output fluctuations. We introduce the dimensionless phase Fisher information

I~θ=κ2χλ2Iλ,θ(in)=4T∣∂θ∂(δ/κ)∣2.(4)\widetilde{\mathcal{I}}_\theta= \frac{\kappa^2}{\chi_\lambda^2}\mathcal{I}^{(\mathrm{in})}_{\lambda,\theta} = 4T\left|\frac{\partial\theta}{\partial(\delta/\kappa)}\right|^2. \tag*{(4)}

Unlike the phase slope alone, this quantity incorporates the loss of transmitted photons near the antiresonance. Figure 3(a) shows that the Fisher information is maximized at an intermediate coupling phase. Increasing γc\gamma_c enhances this maximum while shifting the optimal phase toward weaker coupling to the antisymmetric state. The optimum thus reflects a balance between the intrinsic subradiant decay and cavity-induced optical readout.

Figure 3. Phase Fisher information and optimal readout. I~θ\widetilde{\mathcal{I}}_\theta versus ϕ\phi for different (a) γc\gamma_c with V/κ=1V/\kappa= 1 and (b) VV with γc/γ=0.95\gamma_c/\gamma= 0.95. (c) ϕopt\phi_{\mathrm{opt}} from full calculation (solid), readout-decay matching Γcav=γA\Gamma_{\mathrm{cav}} = \gamma_A (dashed), and asymptotic scaling ϕopt∝γA1/4\phi_{\mathrm{opt}} \propto\gamma_A^{1/4} (dotted). (d) I~θ\widetilde{\mathcal{I}}_\theta versus γA\gamma_A for different VV, with dashed line showing asymptotic inverse-square scaling.

To identify the origin of this optimum, we examine the transmission near the antisymmetric resonance. Treating the symmetric contribution as the slowly varying background KK defined above yields

t(δ)≈κKγA+iδγA+Γcav+i(δ−Δcav).(5)t(\delta) \approx\frac{\kappa}{K}\frac{\gamma_A+i\delta}{\gamma_A+\Gamma_{\mathrm{cav}}+i(\delta-\Delta_{\mathrm{cav}})}. \tag*{(5)}

For a negligible or compensated cavity-induced frequency shift, the on-resonance transmission and phase slope are T0≈Tbg[γA/(γA+Γcav)]2T_0 \approx T_{\mathrm{bg}}[\gamma_A/(\gamma_A+\Gamma_{\mathrm{cav}})]^2 and ∂δθ∣0≈Γcav/[γA(γA+Γcav)]\left.\partial_\delta\theta\right|_0 \approx\Gamma_{\mathrm{cav}}/[\gamma_A(\gamma_A+\Gamma_{\mathrm{cav}})], respectively, where Tbg=∣κ/K∣2T_{\mathrm{bg}}=|\kappa/K|^2. The resulting phase Fisher information is

Iλ,θ(in)χλ2≈4TbgΓcav2(γA+Γcav)4.(6)\frac{\mathcal{I}^{(\mathrm{in})}_{\lambda,\theta}}{\chi_\lambda^2}\approx4T_{\mathrm{bg}}\frac{\Gamma_{\mathrm{cav}}^2}{(\gamma_A+\Gamma_{\mathrm{cav}})^4}. \tag*{(6)}

For a fixed optical background, maximizing this expression with respect to Γcav\Gamma_{\mathrm{cav}} gives Γcav=γA\Gamma_{\mathrm{cav}}=\gamma_A. The optimal phase information therefore occurs when the cavity-induced coherence decay matches the intrinsic subradiant decay, rather than when the optical coupling or phase slope is maximized. This readout-decay matching condition provides a local criterion for converting protected subradiant coherence into measurable phase information.

The auxiliary interaction further controls the readout-decay balance. Figure 3(b) shows that varying VV shifts the information maximum through the symmetric-state contribution to KK, without affecting the protected antisymmetric state. In the strongly subradiant regime, ϕopt→0\phi_{\mathrm{opt}}\to0, where Γcav≈(gA2ϕ4/8)Re⁡(1/K0)\Gamma_{\mathrm{cav}}\approx(g_A^2\phi^4/8)\operatorname{Re}(1/K_0). The matching condition Γcav=γA\Gamma_{\mathrm{cav}}=\gamma_A then yields optimal phase

ϕopt≈[8γAgA2Re⁡(1/K0)]1/4∝γA1/4.(7)\phi_{\mathrm{opt}}\approx\left[\frac{8\gamma_A}{g_A^2\operatorname{Re}(1/K_0)}\right]^{1/4}\propto\gamma_A^{1/4}. \tag*{(7)}

This fourth-root scaling follows from gA∝ϕ2g_A\propto\phi^2 near ϕ=0\phi=0. Figure 3(c) compares this prediction with the full transmission calculation, showing agreement within the validity of the local approximation.

At the matching point, the transmission remains finite, T0≈Tbg/4T_0\approx T_{\mathrm{bg}}/4, while ∣∂δθ∣0∣≈1/(2γA)|\left.\partial_\delta\theta\right|_0|\approx1/(2\gamma_A), giving Iλ,θ,max⁡(in)≈χλ2Tbg/(4γA2)\mathcal{I}^{(\mathrm{in})}_{\lambda,\theta,\max}\approx\chi_\lambda^2T_{\mathrm{bg}}/(4\gamma_A^2). For a finite, nonzero background transmission and a valid local approximation as γA→0\gamma_A\to0, the optimized dimensionless phase information scales as I~θ,max⁡∝(γA/γ)−2\widetilde{\mathcal{I}}_{\theta,\max}\propto(\gamma_A/\gamma)^{-2} for γA/γ≪1\gamma_A/\gamma\ll1, with other frequency scales fixed. Figure 3(d) illustrates this scaling, with VV modifying the prefactor through the optical background. Thus, phase-controlled readout converts long-lived subradiant coherence into enhanced phase information while maintaining finite transmission.

Photon correlations and atomic entanglement.— We next examine how symmetry protection shapes photon correlations and steady-state atomic entanglement beyond the linear response. From the steady-state master equation, we calculate the intracavity population na=⟨a^†a^⟩n_a=\langle\hat{a}^{\dagger}\hat{a}\rangle and second-order correlation g(2)(0)=g^{(2)}(0)=

⟨a^†2a^2⟩ss/na2\langle\hat{a}^{\dagger2}\hat{a}^2\rangle_{\mathrm{ss}}/n_a^2. Figure 4(a) and Figure 4(b) show pronounced photon bunching at the antisymmetric antiresonance, accompanied by a strongly suppressed cavity population. The large g(2)(0)g^{(2)}(0) reflects the relative survival of two-photon excitation pathways but does not necessarily imply an enhanced photon-pair flux.

Figure 4. Photon correlations and atomic entanglement. (a) na/η2n_a/\eta^2 and (b) log⁡g(2)(0)\log g^{(2)}(0) versus δ\delta and ϕ\phi. (c) Concurrence CssC_{\mathrm{ss}} versus ϕ\phi for different γc/γ\gamma_c/\gamma at δ=0\delta=0. (d) g(2)(0)g^{(2)}(0) (solid) and TT (dashed) versus γc/γ\gamma_c/\gamma at (δ,ϕ)=(0,π)(\delta,\phi)=(0,\pi). (e) CssC_{\mathrm{ss}} (solid) and PAP_A (dashed) versus Γcav\Gamma_{\mathrm{cav}} at γc/γ=0.99\gamma_c/\gamma=0.99. (f) ϕentopt\phi_{\mathrm{ent}}^{\mathrm{opt}} from the master equation (solid), readout-decay matching (dashed), and fourth-root scaling (dotted).

This contrast originates from the distinct action of the auxiliary collective channel in different excitation manifolds. At ϕ=π\phi=\pi, the single-excitation pathway ∣0,G⟩↔∣1,G⟩↔∣0,A⟩|0,G\rangle\leftrightarrow|1,G\rangle\leftrightarrow|0,A\rangle involves the protected state ∣A⟩|A\rangle, satisfying JˉB∣A⟩=0\bar{J}_B|A\rangle=0. The auxiliary channel therefore leaves the linear antiresonance unchanged. In contrast, the two-excitation pathway ∣2,G⟩↔∣1,A⟩↔∣0,E⟩|2,G\rangle\leftrightarrow|1,A\rangle\leftrightarrow|0,E\rangle involves ∣E⟩=∣ee⟩|E\rangle=|ee\rangle, for which JˉB∣E⟩=2∣S⟩\bar{J}_B|E\rangle=\sqrt{2}|S\rangle. Its diagonal complex energy in the no-jump effective Hamiltonian is ϵE=2δ+2V−2i(γ+ΓB)\epsilon_E=2\delta+2V-2i(\gamma+\Gamma_B). The auxiliary channel thus modifies two-photon excitation pathways without shifting the protected linear resonance, allowing strong bunching near suppressed transmission, in contrast to conventional photon blockade [50]. Figure 4(d) further shows that increasing correlated decay enhances g(2)(0)g^{(2)}(0) while reducing transmission.

Beyond photon correlations, the protected antisymmetric state also supports steady-state atomic entanglement. We quantify the entanglement of reduced atomic state ρ^at=Tr⁡cavρ^ss\hat{\rho}_{\mathrm{at}} = \operatorname{Tr}_{\mathrm{cav}} \hat{\rho}_{\mathrm{ss}} using Wootters concurrence CssC_{\mathrm{ss}} [72]. Figure 4(c) shows that increasing correlated decay enhances CssC_{\mathrm{ss}}, which peaks at an intermediate coupling phase rather than at ϕ=π\phi= \pi. As shown in Figure 4(e), the concurrence closely follows the antisymmetric-state population PA=⟨A∣ρ^at∣A⟩P_A = \langle A|\hat{\rho}_{\mathrm{at}}|A\rangle, identifying the antisymmetric Bell state as the dominant entangled component. Indeed, for states restricted to the {∣G⟩,∣A⟩}\{|G\rangle, |A\rangle\} subspace, C=PAC = P_A, even in the presence of coherence between these states.

The finite-phase entanglement maximum can be understood from the weak-excitation population of the antisymmetric state, PA≈η2gA2/[K(γA+iδ)+gA2]P_A \approx\eta^2 g_A^2/[K(\gamma_A + i\delta) + g_A^2]. Using gA2/K=Γcav−iΔcavg_A^2/K = \Gamma_{\mathrm{cav}} - i\Delta_{\mathrm{cav}}, we obtain PA≈(η2/Re⁡K)Γcav/[(γA+Γcav)2+(δ−Δcav)2]P_A \approx(\eta^2/\operatorname{Re} K)\Gamma_{\mathrm{cav}}/[(\gamma_A + \Gamma_{\mathrm{cav}})^2 + (\delta- \Delta_{\mathrm{cav}})^2] for Re⁡K>0\operatorname{Re} K > 0. For negligible effective detuning and a slowly varying background, Css≈PA∝Γcav/(γA+Γcav)2C_{\mathrm{ss}} \approx P_A \propto\Gamma_{\mathrm{cav}}/(\gamma_A + \Gamma_{\mathrm{cav}})^2, yielding an optimum at Γcav=γA\Gamma_{\mathrm{cav}} = \gamma_A. This approximate matching condition accounts for the finite-phase maximum in Figure 4(e).

Figure 4(f) compares the optimal entanglement phase from the full master equation with the weak-excitation prediction. In the strongly subradiant regime, ϕentopt→0\phi_{\mathrm{ent}}^{\mathrm{opt}} \to0, where gA2≈g2ϕ4/8g_A^2 \approx g^2\phi^4/8 and K≈K0K \approx K_0. At δ=0\delta= 0, maximizing the antisymmetric-state population yields gA,opt2≈γA∣K0∣g_{A,\mathrm{opt}}^2 \approx\gamma_A|K_0| and hence ϕentopt≈(8γA∣K0∣/g2)1/4∝γA1/4\phi_{\mathrm{ent}}^{\mathrm{opt}} \approx(8\gamma_A|K_0|/g^2)^{1/4} \propto\gamma_A^{1/4}. This reproduces the fourth-root scaling found for optimal phase sensing, although the two optimal phases generally differ because they obey distinct optimization conditions. Saturation and variations in the optical background can lead to deviations from the weak-excitation prediction. These results show that symmetry protection preserves the linear antiresonance while enabling tunable photon correlations, atomic entanglement, and phase-sensitive optical readout.

Collective subradiant response in emitter arrays

Extending the linear-response analysis to correlated emitter arrays, we find that spatially structured cavity coupling provides tunable optical access to protected collective modes. The two-emitter results show that subradiance-enhanced phase sensing requires both long-lived collective coherence and finite optical coupling. We now examine how this interplay extends to emitter arrays, where multiple collective modes can contribute to the cavity response. For NN equally spaced emitters, the main-cavity coupling acquires the spatial profile gj=gcos⁡[(j−1)ϕ]g_j = g\cos[(j-1)\phi], while the auxiliary channel remains uniform, J^B=∑j=1Nσ^j−\hat{J}_B = \sum_{j=1}^{N}\hat{\sigma}_j^-. The independent phases ϕg\phi_g and ϕ\phi then control the collective decay structure and the optical accessibility of its eigenmodes, respectively.

In the weak-excitation regime, the atomic coherences s=(⟨σ^1−⟩,…,⟨σ^N−⟩)T\mathbf{s} = (\langle\hat{\sigma}_1^-\rangle,\ldots,\langle\hat{\sigma}_N^-\rangle)^T satisfy M(δ)s=−igα\mathbf{M}(\delta)\mathbf{s} = -ig\boldsymbol{\alpha}, where α=⟨a^⟩\boldsymbol{\alpha} = \langle\hat{a}\rangle. Eliminating the atomic coherences yields

M(δ)=Γ+(ΓB+iV)uuT+iδI,\mathbf{M}(\delta) = \mathbf{\Gamma} + (\mathbf{\Gamma}_B + iV)\mathbf{u}\mathbf{u}^T + i\delta I,
t(δ,ϕ)=κκ+iΔa+gTM−1(δ)g(8)t(\delta,\phi) = \frac{\kappa}{\kappa+ i\Delta_a + \mathbf{g}^T\mathbf{M}^{-1}(\delta)\mathbf{g}} \tag*{(8)}

where u=(1,…,1)T\mathbf{u} = (1,\ldots,1)^T and g=(g1,…,gN)T\mathbf{g} = (g_1,\ldots,g_N)^T. The correlated decay matrix can be written as Γ=γpI+γ1D(ccT+ssT)\mathbf{\Gamma} = \gamma_p I + \gamma_{1D}(\mathbf{c}\mathbf{c}^T + \mathbf{s}\mathbf{s}^T), with cj=cos⁡[(j−1)ϕg]c_j = \cos[(j-1)\phi_g] and sj=sin⁡[(j−1)ϕg]s_j = \sin[(j-1)\phi_g]. Since the guided-mode contribution has rank at most two, at least N−2N-2 collective modes retain only the residual decay γp\gamma_p. Imposing orthogonality to the uniform auxiliary channel further defines the protected subspace D=ker⁡[c,s,u]T\mathcal{D} = \ker[\mathbf{c},\mathbf{s},\mathbf{u}]^T, whose dimension is N−rank⁡[c,s,u]N-\operatorname{rank}[\mathbf{c},\mathbf{s},\mathbf{u}], generically N−3N-3. For any v∈D\mathbf{v} \in\mathcal{D}, M(δ)v=(γp+iδ)v\mathbf{M}(\delta)\mathbf{v} = (\gamma_p + i\delta)\mathbf{v}, showing that these modes are unaffected by both the auxiliary-induced coherent interaction and collective decay.

The existence of protected collective modes does not, however, guarantee their optical accessibility. Introducing the orthogonal projector PD\mathbf{P}_{\mathcal{D}} onto D\mathcal{D}, the protected contribution to the cavity self-energy is

χD(δ,ϕ)=gTPDgγp+iδ(9)\chi_{\mathcal{D}}(\delta,\phi) = \frac{\mathbf{g}^T\mathbf{P}_{\mathcal{D}}\mathbf{g}}{\gamma_p + i\delta} \tag*{(9)}

The optical weight WD(ϕ)=gTPDg=∥PDg∥2W_{\mathcal{D}}(\phi) = \mathbf{g}^T\mathbf{P}_{\mathcal{D}}\mathbf{g} = \lVert\mathbf{P}_{\mathcal{D}}\mathbf{g}\rVert^2 quantifies the coupling of the cavity field to the protected subspace. In particular, WD=0W_{\mathcal{D}} = 0 when g\mathbf{g} lies in the span of

c\mathbf{c}, sg\mathbf{s}_g, and u\mathbf{u}, rendering these long-lived modes optically dark. Varying ϕ\phi therefore controls their optical accessibility without changing their intrinsic linewidths, while the remaining collective modes contribute an additional optical background. This extends the separation between subradiant protection and optical readout from a single antisymmetric state to a collective subspace.

Figure 5(a) and Figure 5(b) show the resonant transmission T=∣t∣2T=|t|^2 and phase susceptibility Sθ=∣∂θ/∂(δ/κ)∣δ=0S_\theta=\left|\partial\theta/\partial(\delta/\kappa)\right|_{\delta=0} as functions of ϕ\phi and NN, where θ=arg⁡t\theta=\arg t. Their phase dependence reflects the redistribution of optical coupling among collective modes with different decay rates. Defining Dc=κ+iΔa+gTM−1gD_c=\kappa+i\Delta_a+\mathbf{g}^{T}\mathbf{M}^{-1}\mathbf{g} and using ∂δM−1=−iM−2\partial_\delta\mathbf{M}^{-1}=-i\mathbf{M}^{-2}, we obtain

Figure 5. Phase-sensitive optical response of correlated emitter arrays. (a) Transmission TT and (b) phase susceptibility SθS_\theta versus coupling phase ϕ\phi and emitter number NN. White dashed curves indicate the optimal phase ϕopt(N)\phi_{\mathrm{opt}}(N) maximizing I~θ=4TSθ2\widetilde{\mathcal{I}}_\theta= 4TS_\theta^2. (c) Phase Fisher information I~θ\widetilde{\mathcal{I}}_\theta versus ϕ\phi for N=4N = 4, 88, and 1616. (d) Optimized phase Fisher information I~θ,max⁡\widetilde{\mathcal{I}}_{\theta,\max} versus NN for the indicated reservoir phases ϕg\phi_g. Panels (a)–(c) use ϕg=0.4π\phi_g = 0.4\pi. Other parameters are γ1D=0.95γ\gamma_{1D} = 0.95\gamma and γp=0.05γ\gamma_p = 0.05\gamma.

Sθ=κ∣Re⁡[gTM−2gDc]∣δ=0.(10)S_\theta=\kappa\left|\operatorname{Re}\left[\frac{\mathbf{g}^{T}\mathbf{M}^{-2}\mathbf{g}}{D_c}\right]\right|_{\delta=0}. \tag*{(10)}

As in the two-emitter system, strong phase dispersion can be accompanied by suppressed transmission. We therefore characterize the phase sensitivity by the phase Fisher information per incident photon, I~θ=4TSθ2\widetilde{I}_\theta=4TS_\theta^2, which accounts for the transmission penalty. The dashed curves in Figure 5(a) and Figure 5(b) mark the phases that maximize I~θ\widetilde{I}_\theta within the selected interval.

Figure 5(c) shows that I~θ(ϕ)\widetilde{I}_\theta(\phi) develops multiple local maxima as NN increases, reflecting the phase-dependent coupling to different collective response channels. Unlike the two-emitter system, the array generally involves several collective modes and a nontrivial optical background. Consequently, the single-mode matching condition Γcav=γA\Gamma_{\mathrm{cav}}=\gamma_A does not universally determine the optimal phase. Figure 5(d) further shows that substantial optimized phase information persists for arrays of up to N=60N=60, with a pronounced dependence on the reservoir phase ϕg\phi_g. The optimization is restricted to 0<ϕ/π≤0.60<\phi/\pi\leq0.6; the resulting size dependence does not imply universal collective enhancement. These results extend the separation between subradiant protection and optical accessibility from a single antisymmetric excitation to a collective subspace. While reservoir correlations establish long-lived collective modes, spatially selective cavity coupling determines their contribution to the measured phase response.

The protection stems from the symmetry of the auxiliary collective coupling rather than a cancellation of coherent and dissipative effects. Frequency disorder and nonuniform auxiliary couplings can mix bright and dark states, while additional radiative losses broaden the subradiant resonance. Coherent photon-mediated exchange can further reshape the collective spectrum [20, 69]. Recent studies of collective photon correlations and multiexcitation subradiance highlight the importance of nonlinear excitation pathways [73, 74, 75, 76]. Extending excitation-selective protection to higher-excitation manifolds in larger arrays remains an important direction for controlling multiphoton correlations.

Acknowledgments

Acknowledgments.—This work was supported by the National Natural Science Foundation of China (Grant No. 12674338 and Grant No. 12374365), Quantum Science and Technology-National Technology Major Project (Grant No. 2025ZD0300400) and Guangdong Provincial Quantum Science Strategic Initiative (Grant Nos. GDZX2505001 and GDZX2604004).

Conclusion

We have demonstrated that collective symmetry enables independent control of nonlinear excitation pathways and optical readout while preserving a subradiant cavity antiresonance. The auxiliary collective channel decouples from the antisymmetric single-excitation state but remains active in the two-excitation manifold, allowing photon correlations to be tuned without shifting or broadening the protected linear resonance. Independently, the cavity coupling phase controls optical access to the subradiant state without altering its intrinsic decay rate. This separation enables pronounced photon bunching and tunable atomic entanglement, while the competition between optical readout and subradiant decay yields an optimal phase Fisher information per incident photon. This phase-selective readout mechanism further extends to protected collective modes in correlated emitter arrays.

References

  1. [1]R. H. Dicke, Phys. Rev. 93, 99 (1954).
  2. [2]R. H. Lehmberg, Phys. Rev. A 2, 883 (1970).
  3. [3]M. Gross and S. Haroche, Phys. Rep. 93, 301 (1982).
  4. [4]Z. Ficek and R. Tanaś, Phys. Rep. 372, 369 (2002).
  5. [5]L. Ostermann, H. Ritsch, and C. Genes, Phys. Rev. Lett. 111, 123601 (2013).
  6. [6]L. Ostermann, D. Plankensteiner, H. Ritsch, and C. Genes, Phys. Rev. A 90, 053823 (2014).
  7. [7]S. Filipp, A. F. van Loo, M. Baur, L. Steffen, and A. Wallraff, Phys. Rev. A 84, 061805 (2011).
  8. [8]W. Guerin, M. O. Araujo, and R. Kaiser, Phys. Rev. Lett. 116, 083601 (2016).
  9. [9]G. Facchinetti, S. D. Jenkins, and J. Ruostekoski, Phys. Rev. Lett. 117, 243601 (2016).
  10. [10]D. F. Kornovan, N. V. Corzo, J. Laurat, and A. S. Sheremet, Phys. Rev. A 100, 063832 (2019).
  11. [11]G. Ferioli, A. Glicenstein, L. Henriet, I. Ferrier-Barbut, and A. Browaeys, Phys. Rev. X 11, 021031 (2021).
  12. [12]Y.-X. Zhang and K. Mølmer, Phys. Rev. Lett. 122, 203605 (2019).
  13. [13]M. Eltohfa and F. Robicheaux, Phys. Rev. A 112, 023112 (2025).
  14. [14]S. D. Jenkins and J. Ruostekoski, Phys. Rev. A 86, 031602 (2012).
  15. [15]S. D. Jenkins and J. Ruostekoski, Phys. Rev. Lett. 111, 147401 (2013).
  16. [16]R. J. Bettles, S. A. Gardiner, and C. S. Adams, Phys. Rev. A 92, 063822 (2015).
  17. [17]R. J. Bettles, S. A. Gardiner, and C. S. Adams, Phys. Rev. Lett. 116, 103602 (2016).
  18. [18]E. Shahmoon, D. S. Wild, M. D. Lukin, and S. F. Yelin, Phys. Rev. Lett. 118, 113601 (2017).
  19. [19]J. Perczel, J. Borregaard, D. E. Chang, H. Pichler, S. F. Yelin, P. Zoller, and M. D. Lukin, Phys. Rev. A 96, 063801 (2017).
  20. [20]A. Asenjo-Garcia, J. D. Hood, D. E. Chang, and H. J. Kimble, Phys. Rev. A 95, 033818 (2017).
  21. [21]J. Rui, D. Wei, A. Rubio-Abadal, S. Hollerith, J. Zeiher, D. M. Stamper-Kurn, C. Gross, and I. Bloch, Nature 583, 369 (2020).
  22. [22]K. E. Ballantine and J. Ruostekoski, Phys. Rev. Research 2, 023086 (2020).
  23. [23]K. E. Ballantine and J. Ruostekoski, PRX Quantum 2, 040362 (2021).
  24. [24]K. Skrakaew, P. Weckesser, S. Hollerith, D. Wei, D. Adler, I. Bloch, and J. Zeiher, Nat. Phys. 19, 714 (2023).
  25. [25]O. Rubies-Bigorda, V. Walther, T. L. Patti, and S. F. Yelin, Phys. Rev. Research 4, 013110 (2022).
  26. [26]S. J. Masson, I. Ferrier-Barbut, L. A. Orozco, A. Browaeys, and A. Asenjo-Garcia, Phys. Rev. Lett. 125, 263601 (2020).
  27. [27]O. Rubies-Bigorda and S. F. Yelin, Phys. Rev. A 106, 053717 (2022).
  28. [28]N. O. Gjonbalaj, S. Ostermann, and S. F. Yelin, Phys. Rev. A 109, 013720 (2024).
  29. [29]A. Yadav and D. D. Yavuz, Phys. Rev. A 110, 023709 (2024).
  30. [30]B. Olmos and I. Lesanovsky, Phys. Rev. Lett. 134, 243602 (2025).
  31. [31]H. Ritsch, P. Domokos, F. Brennecke, and T. Esslinger, Rev. Mod. Phys. 85, 553 (2013).
  32. [32]R. Reimann, W. Alt, T. Kampschulte, T. Macha, L. Ratschbacher, N. Thau, S. Yoon, and D. Meschede, Phys. Rev. Lett. 114, 023601 (2015).
  33. [33]B. Casabone, K. Friebe, B. Brandstatter, K. Schüppert, R. Blatt, and T. E. Northup, Phys. Rev. Lett. 114, 023602 (2015).
  34. [34]J. S. Douglas, H. Habibian, C.-L. Hung, A. V. Gorshkov, H. J. Kimble, and D. E. Chang, Nature Photonics 9, 326 (2015).
  35. [35]H. Pichler, T. Ramos, A. J. Daley, and P. Zoller, Phys. Rev. A 91, 042116 (2015).
  36. [36]Y. Solomons and E. Shakhnovich, Phys. Rev. A 107, 033709 (2023).
  37. [37]Z. Yan, J. Ho, Y.-H. Lu, S. J. Masson, A. Asenjo-Garcia, and D. M. Stamper-Kurn, Phys. Rev. Lett. 131, 253603 (2023).
  38. [38]C. Hotter, L. Ostermann, and H. Ritsch, Phys. Rev. Research 5, 013056 (2023).
  39. [39]X. Zhou, D. A. Suresh, F. Robicheaux, and C.-L. Hung, Phys. Rev. Lett. 135, 113601 (2025).
  40. [40]K.-Y. Kim, J. H. Lee, W. B. Jeon, D. H. Park, S. I. Park, J. D. Song, C. Lee, and J.-H. Kim, Nat. Commun. 16, 6346 (2025).
  41. [41]S. J. Masson and A. Asenjo-Garcia, Phys. Rev. Research 2, 043213 (2020).
  42. [42]C. Sames, H. Chibani, C. Hansen, P. A. Altin, T. Wilk, and G. Rempe, Phys. Rev. Lett. 112, 043601 (2014).
  43. [43]D. Plankensteiner, C. Sommer, H. Ritsch, and C. Genes, Physical Review Letters 119, 093601 (2017).
  44. [44]D. Meiser, J. Ye, D. R. Carlson, and M. J. Holland, Phys. Rev. Lett. 102, 163601 (2009).
  45. [45]J. G. Bohn, Z. Chen, J. M. Weiner, D. Meiser, M. J. Holland, and J. K. Thompson, Nature 484, 78 (2012).
  46. [46]D. Barberena, R. J. Lewis-Swan, J. K. Thompson, and
  47. [48]M. A. Norcia, J. R. K. Cline, J. A. Muniz, J. M. Robinson, R. B. Hutson, A. Goban, G. E. Marti, J. Ye, and J. K. Thompson, Phys. Rev. X 8, 021036 (2018).
  48. [49]H. Liu, S. B. Jager, X. Yu, S. Touzard, A. Shankar, M. J. Holland, and T. L. Nicholson, Phys. Rev. Lett. 125, 253602 (2020).
  49. [50]K. M. Birnbaum, A. Boca, R. Miller, A. D. Boozer, T. E. Northup, and H. J. Kimble, Nature 436, 87 (2005).
  50. [51]H. Flayac and V. Savona, Phys. Rev. A 96, 053810 (2017).
  51. [52]H. J. Snijders, J. A. Frey, J. Norman, H. Flayac, V. Savona, A. C. Gossard, J. E. Bowers, M. P. van Exter, D. Bouwmeester, and W. Loffler, Phys. Rev. Lett. 121, 043601 (2018).
  52. [53]M. Gegg, A. Carmele, A. Knorr, and M. Richter, New J. Phys. 20, 013006 (2018).
  53. [54]A. Shankar, J. T. Reilly, S. B. Jager, and M. J. Holland, Phys. Rev. Lett. 127, 073603 (2021).
  54. [55]A. Piñeiro Orioli, J. K. Thompson, and A. M. Rey, Phys. Rev. X 12, 011054 (2022).
  55. [56]L. Qiao and J. Gong, Phys. Rev. Lett. 129, 093602 (2022).
  56. [57]M. Zanner, T. Orell, C. M. Schneider, R. Albert, S. Oleschko, M. L. Juan, M. Silveri, and G. Kirchmair, Nat. Phys. 18, 538 (2022).
  57. [58]T. L. Patti, D. S. Wild, E. Shahmoon, M. D. Lukin, and S. F. Yelin, Phys. Rev. Lett. 126, 223602 (2021).
  58. [59]B. Sundar, D. Barberena, A. M. Rey, and A. Piñeiro Orioli, Phys. Rev. Lett. 132, 033601 (2024).
  59. [60]B. Sundar, D. Barberena, A. M. Rey, and A. Piñeiro Orioli, Phys. Rev. A 109, 013713 (2024).
  60. [61]H. Krauter, C. A. Muschik, K. Jensen, W. Wasilewski, J. M. Petersen, J. I. Cirac, and E. S. Polzik, Phys. Rev. Lett. 107, 080503 (2011).
  61. [62]S. Shankar, M. Hatridge, Z. Leghas, K. M. Sliva, A. Narla, U. Vool, S. M. Girvin, L. Frunzio, M. Mirrahimi, and M. H. Devoret, Nature 504, 419 (2013).
  62. [63]A. Vivas-Viña, D. Martín-Cano, and C. Sánchez Muñoz, Phys. Rev. Lett. 133, 173601 (2024).
  63. [64]A. Vivas-Viña, D. Martín-Cano, and C. Sánchez Muñoz, Phys. Rev. Research 6, 043051 (2024).
  64. [65]N. A. Ron, M. Carmi, and R. Bekenstein, Phys. Rev. Research 6, L042051 (2024).
  65. [66]F. Reiter and A. S. Sørensen, Phys. Rev. A 85, 032111 (2012).
  66. [67]A. Goban, C.-L. Hung, J. D. Hood, S.-P. Yu, J. A. Muniz, O. Painter, and H. J. Kimble, Phys. Rev. Lett. 115, 063601 (2015).
  67. [68]J. D. Hood, A. Goban, A. Asenjo-Garcia, M. Lu, S.-P. Yu, D. E. Chang, and H. J. Kimble, Proc. Natl. Acad. Sci. U.S.A. 113, 10507 (2016).
  68. [69]A. S. Sheremet, M. I. Petrov, I. V. Iorsh, A. V. Poshakinskiy, and A. N. Poddubny, Rev. Mod. Phys. 95, 015002 (2023).
  69. [70]V. Giovannetti, S. Lloyd, and L. Maccone, Nature Photonics 5, 222 (2011).
  70. [71]R. Demkowicz-Dobrzanski, M. Jarzyna, and J. Kołodyński, in Progress in Optics, Vol. 60 (Elsevier, 2015) pp. 345–435. W. K. Wootters, Physical Review Letters 80, 2245 (1998).
  71. [73]K. Singh, A. Cidrim, A. Kovalenko, T. M. Pham, O. Číp, L. Slodička, and R. Bachelard, Phys. Rev. Lett. 134, 203602 (2025).
  72. [74]L. A. Williamson, M. O. Borgh, and J. Ruostekoski, Phys. Rev. Lett. 125, 073602 (2020).
  73. [75]O. Rubies-Bigorda, S. Ostermann, and S. F. Yelin, Phys. Rev. A 107, L051701 (2023).
  74. [76]M. Tečer, M. Di Liberto, P. Silvi, S. Montangero, F. Romanato, and G. Calajò, Phys. Rev. Lett. 132, 163602 (2024).

Paper details

vlnc:W_ab5pq9f4
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 3

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents