Superbunching from coherently driven atoms in a waveguide

G\"oran Johansson; Maryam Khanahmadi; Therese Karmstrand; Zeidan Zeidan

REVIEW 2 major objections 5 minor 65 references

Superbunching from coherently driven atoms in a waveguide

T0 review · 2 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Weak drive makes N atoms transmit only in N+1-photon bunches.

desk verdict A clean, clearly derived result on (N+1)-photon superbunching in waveguide QED, worth refereeing, but the main quantitative claim is tied to an idealized symmetric geometry and the bunching statistics include one fitted comparison. read the letter →

arxiv 2506.05147 v1 pith:Y7I6B2MB submitted 2025-06-05 quant-ph

Zeidan Zeidan , Therese Karmstrand , Maryam Khanahmadi , Göran Johansson This is my paper

classification quant-ph

keywords waveguidequantumelectrodynamicssuperbunchingDickesubspacemulti-photonscatteringheraldedstatepreparationphotonbunchingcoherentdriveinput-outputtheory

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper studies N two-level atoms placed one resonant wavelength apart in a one-dimensional waveguide and driven by a weak coherent field. It claims that in this regime transmission through the atoms is no longer a single-photon event but an (N+1)-photon process: the atoms act as a mirror for up to N photons, and only when N+1 photons arrive together can light get through. The transmitted light is predominantly incoherent and strongly superbunched, with a second-order coherence that diverges as the drive weakens. The counterintuitive core is that the first transmitted photon of a bunch heralds the atomic ensemble in its fully excited state, because transmission is unlocked only by collectively exciting all N atoms. This turns an otherwise passive mirror into a heralded multi-photon source.

What carries the argument

The central object is the driven Dicke model restricted to the symmetric (N+1)-dimensional subspace, with collective dipole operators $S_\pm = \sum_i S_{i\pm}$ and input-output relations $a^R_{\mathrm{out}} = \alpha + \sqrt{\gamma} S_-$, $a^L_{\mathrm{out}} = \sqrt{\gamma} S_-$. The steady state $\rho = D^{-1} \sum_{m,n} (S_-/g^*)^m (S_+/g)^n$ (Eq. 2), inherited from the Puri\textendash Lawande solution, supplies all correlation functions. The no-click master equation\textemdash the Lindblad equation with click terms removed\textemdash is the mechanism that reveals the hidden path: its conditioned steady state is the fully excited state, showing that a transmitted photon can only be emitted after every atom has been excited. The timescales $T_{\mathrm{in}}$ (Eq. 7) and the hypoexponential relaxation $T_{\mathrm{out}}$ close the argument by showing that the N+1 input photons must arrive within a short window and that the excited ensemble decays in a cascaded superradiant chain.

What would settle it

Drive the waveguide with a weak coherent field and measure the second-order coherence of the transmitted light for N = 2 or 3 atoms: the paper predicts $g^{(2)}_{\mathrm{trans}}(0) \propto |g|^{-2(N+1)}$ in the weak-drive limit, i.e., a divergence as the drive amplitude goes to zero. Alternatively, prepare N atoms, apply a weak drive, and detect the first transmitted photon; the predicted conditional atomic state is the N-truncated geometric distribution with the fully excited state most likely, so a measurement that finds the ensemble predominantly in lower excitation states would refute the central claim.

Watch

Extended reading notes

Core claim

For atoms separated by exactly the drive wavelength, the weak-drive transmission rate is $\bar{n}_{\mathrm{trans}} \approx \gamma |g|^{2(N+1)} \frac{N+1}{(N!)^2}$ (Eq. 3), proportional to the rate of (N+1)-photon components in the Poissonian drive, and this transmission is overwhelmingly incoherent. The $n$-th order zero-delay correlation $G^{(n)}_{\mathrm{trans}}(0)$ for $n \leq N+1$ scales as $|g|^{2(N+1)}$, so $g^{(2)}_{\mathrm{trans}}(0) \approx |g|^{-2(N+1)}$ diverges as the drive tends to zero: transmission, when it happens, happens in superbunched bunches. Conditioning on a transmitted click yields a maximally mixed Dicke state (Eq. 5); conditioning on a first transmitted click after a no-click evolution yields an $N$-truncated geometric distribution (Eq. 6), with the fully excited state as the most likely origin; conditioning on no clicks in either output yields exactly the fully excited state $|N\rangle_{\mathrm{ex}}$. The paper interprets this as proof that transmission proceeds through collective excitation of all N atoms, and verifies the analytical bunching statistics against Monte Carlo trajectory simulations.

Load-bearing premise

The results assume the atoms sit exactly one drive wavelength apart and couple symmetrically to the waveguide, so that all dynamics stays in the symmetric Dicke subspace; any position-dependent phase between the collective dipole operators would change or destroy the predicted (N+1)-photon scaling.

Editorial extensions

If this is right

The first transmitted photon of a bunch can be used as a herald that the atomic ensemble is fully excited, enabling preparation of the entangled binomial state between left- and right-propagating photons.
For up to eight atoms the output state filtered to the optimal mode retains at least 90% of the photons, so the scheme is immediately useful for quantum-enhanced metrology.
Adding a mirror at the right distance channels all N emitted photons in one direction, giving a route to beating the diffraction limit in quantum lithography.
The superbunching divergence, $g^{(2)}_{\mathrm{trans}}(0) \propto |g|^{-2(N+1)}$, is a sharp experimental fingerprint that grows with atom number, so few-atom waveguide experiments can test the scaling directly.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

I would expect the same conditioning logic to carry over to time-delayed (non-Markovian) waveguides only if the symmetric subspace remains closed; a numerical check on a delay-line model would show whether the first-click geometric distribution survives.
The (N+1)-photon threshold suggests a passive nonlinearity: the ensemble is a near-perfect mirror for Fock states with up to N photons and transparent only to N+1-photon components, which could act as a photon-number filter.
A natural extension is to detuned or anharmonic emitters such as transmons; the paper lists this as future work, and the predicted divergence in $g^{(2)}_{\mathrm{trans}}(0)$ gives a quantitative target for measuring how dephasing erodes the bunching.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

I'd send this to referees. The paper makes a clean, genuinely new claim: for N atoms spaced a wavelength apart, weak resonant transmission scales as gamma|g|^{2(N+1)}(N+1)/(N!)^2 and is dominated by (N+1)-photon incoherent events. The derivation from the driven-Dicke steady state is clear, and the no-click master equation argument that a transmitted click heralds the fully excited state is elegant. The supplemental material has the details; I checked the steps for Eq. (3) and the correlation functions and they are consistent. The main contribution is not the input-output formalism, which is standard, but the identification of this particular (N+1)-photon superbunching mechanism and the all-atom-excitation condition. That goes beyond the single-atom bunching results the authors cite. The Monte Carlo bunching statistics for N=1,2,3 match the analysis, so the paper does not rest on fitting alone. Where I have reservations: first, the central quantitative prediction assumes exact wavelength spacing and symmetric coupling. This is stated in the model section, and the SM notes that half-wavelength spacing changes the signs of the dipole operators, but there is no quantitative robustness analysis: no statement about how much position disorder or coupling asymmetry is tolerated before the (N+1)-photon scaling breaks. This is a scope limitation rather than an internal inconsistency, but for an experimental proposal it matters. The stress-test note makes this point correctly. Second, the bunching probability distribution Eq. (8) is compared to simulation using timescales T_in and T_out obtained by fitting that same equation to the simulation data. The authors are upfront about this, but it means Fig. 2(c) is not a parameter-free prediction. The time-integral picture is suggestive, not a validated quantitative prediction. The core result Eq. (3) is independent of this fitting, so the circularity does not bite the main claim. Overall, the paper is internally consistent and honest about the Markov-approximation caveat. I would bring it to a reading group, and I would cite it if I worked in wQED. A serious referee should see this; it deserves a fair review.

Referee Report

2 major / 5 minor

Summary. This paper studies N two-level atoms in a one-dimensional waveguide, separated by one drive wavelength and symmetrically coupled, driven by a weak coherent field. Working in the symmetric Dicke subspace and using input-output relations, the authors derive the exact steady state and from it the reflected and transmitted photon rates, showing that the weak-drive transmission rate scales as |g|^{2(N+1)} and is predominantly incoherent. They compute zero-time correlation functions, finding superbunching, and derive conditional atomic states after transmission clicks using the no-click master equation, concluding that the first click in a bunch heralds the fully excited state. They introduce timescales T_in and T_out, simulate Monte Carlo trajectories for bunch-size statistics, and present fidelities for generating entangled binomial states. The central derivations in Eqs. (2)-(7) are analytic, and the supplemental material provides detailed step-by-step derivations.

Significance. The ideal-model results are significant and mostly self-contained. Equation (3) is a parameter-free asymptotic prediction, Eq. (4) gives closed-form correlation functions, and the no-click conditioning argument in Eqs. (5)-(6) and SM S5 provides a clear mechanistic picture of (N+1)-photon transmission. If the predicted scaling holds, the setup offers a simple heralded multiphoton source. The Monte Carlo simulations support the ideal-model predictions at the level of conditional state populations, although the bunch-size distribution comparison is weakened by the use of fitted timescales. The robustness of the results to geometric and coupling disorder is not addressed.

major comments (2)

[Photon detection statistics; Eq. (8) and Fig. 2] The analytical bunch-size probabilities p(k) in Eq. (8) are compared with the Monte Carlo data in Fig. 2(c) using timescales \hat{T}_in and \hat{T}_out that are obtained by fitting Eq. (8) to the same simulated data. This makes the agreement in Fig. 2(c) partly circular: the two fitted parameters absorb the difference between the idealized Poisson-arrival model and the actual detection protocol, so the dash-dotted curves are not an independent verification of the bunching mechanism. The fitted values also differ substantially from the theoretically defined timescales (e.g., \hat{T}_in=0.37 versus T_in=1.10 for N=3), so the relationship between Eq. (8) and Eqs. (7) and T_out,p is unexplained. Please either derive the effective timescales from the model, or present the comparison with the independently computed T_in and T_out and quantify the discrepancy.
[Model; SM S2] The central scaling result Eq. (3), the correlation functions Eq. (4), and the no-click conditional states all rely on the assumption that the atoms are exactly one drive wavelength apart and symmetrically coupled, so that the collective operators S_\pm have no position-dependent phases. SM S2 notes that half-wavelength spacing changes the collective operators to an alternating-sign sum, but the paper gives no quantitative indication of how deviations from the ideal spacing or coupling nonuniformity affect the (N+1)-photon transmission rate. Since the applications section and the experimental outlook refer to waveguide implementations, an estimate of the tolerance (e.g., adding small random phases \phi_i and computing the leading correction to \bar{n}_{trans}) is needed to establish the robustness of the claim beyond the perfectly ordered geometry.

minor comments (5)

[Model, Eq. (2)] The displayed steady-state solution is typeset in a way that makes the normalization ambiguous; the expression should read \rho = (a_R^{\dagger} a_R)^{-1} / \mathrm{Tr}[(a_R^{\dagger} a_R)^{-1}] = D^{-1} \sum ... , and D should be defined in the main text rather than only in the Supplemental Material.
[Abstract and Conditional atomic state section] The statement that 'transmission is only possible through a process where all N atoms are excited' is stronger than Eq. (5), which gives a maximally mixed state with average N/2 excitations after any transmitted click; the fully excited statement holds for the first click after a no-click interval, as derived in Eq. (6). Please qualify the abstract wording accordingly.
[Model section and Fig. 1 caption] The symbol \gamma is used both as the rate to each propagation direction and as the total single-atom decay rate (SM S1 defines 2\gamma as the single-atom decay rate); please make the convention explicit in the main text.
[Photon detection statistics and Fig. 2] The fitted timescales \hat{T}_in and \hat{T}_out are reported in the captions, but the reason they are systematically shorter than T_in and T_out (e.g., by factors of about 3 and 1.4 for N=3) is not discussed; a sentence explaining this would help the reader interpret the dashed lines.
[SM S6C, Eq. (S80)] The bound on p(N+1) uses a Chernoff bound with \lambda = |\alpha|^2 T_in < N+2; since the regime of interest is weak driving, \lambda is indeed small, but the text should state this condition explicitly when invoking the bound.

Circularity Check

1 steps flagged · score 4.0 of 10

Fitted Eq. (8) validation is partially circular; central Eqs. (3)-(6) are independently derived.

fitted input called prediction [Photon detection statistics section, Eq. (8) and Fig. 2(c) caption]
"The random variables Y and Z have the parameters |α|2 ˆTin and |α|2 ˆTout, respectively, where the timescales ˆTin and ˆTout are obtained by fitting Eq. (8) to the simulation data."

The 'analytical prediction' for bunching probabilities is evaluated with timescales fitted to the same Monte Carlo data against which it is plotted, so the excellent agreement in Fig. 2(c) is partly forced by construction rather than an independent test. The paper itself notes the fitted timescales are shorter than the Tin and Tout used to define bunches, so the comparison validates the functional form of Eq. (8) only after two free parameters are adjusted to the target data. The central transmission rate Eq. (3) and conditional states Eqs. (5)-(6) are derived without such fitting, so the circularity is localized and partial.

full rationale

The main quantitative claims are derived analytically from the driven-Dicke steady state (Eqs. 2-4) and from no-click quantum trajectory equations (Eqs. 5-6). These derivations are self-contained: Eq. (2) is the standard Puri-Lawande steady state, Eqs. (3)-(4) follow by algebra, and Eqs. (5)-(6) come from the Bayesian update and the no-click conditional master equation with no fitted parameters. The wavelength-spacing and symmetric-coupling assumption is a model premise, not a circularity; the paper's brief note about half-wavelength spacing (SM S2) is a robustness limitation rather than a circular step. Citations to [31] point to the authors' own Supplemental Material, but the derivations are provided in that same submission, so no load-bearing self-citation chain is present. The one genuine circular step I can exhibit is the photon-detection statistics comparison: Eq. (8) is plotted as an 'analytical prediction' but its timescale parameters are fitted to the very simulation data it is compared against. This makes that particular validation a two-parameter postdiction. Because the headline scaling and the fully-excited-state mechanism do not rely on that fit, I score 4 rather than 6: the circularity is real but localized to one validation curve.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central analytic results (Eqs. 3-6) are derived from the known driven-Dicke steady state and input-output relations, requiring no free parameters. The bunching probability model Eq. (8) introduces two timescales that are fitted to simulation data. Five background assumptions are listed, including the symmetric-subspace and Markov approximations.

free parameters (2)

\hat{Tin} (fitted input timescale) = 0.37 (N=3), 0.83 (N=2), 2.58 (N=1)
Used in Eq. (8) to model the initial photon arrival distribution; obtained by fitting Eq. (8) to the simulation data.
\hat{Tout} (fitted output timescale) = 0.93 (N=3), 1.05 (N=2), 1.80 (N=1)
Used in Eq. (8) to model additional photon arrivals during emission; obtained by fitting Eq. (8) to the simulation data.

assumptions (5)

standard math The steady state of the driven Dicke master equation is given by Eq. (2) (Puri-Lawande).
Used to compute all atomic correlations; borrowed from prior literature (Refs. [34-39]).
domain assumption Input-output relations aR_out = aL_in + sqrt(gamma) S- and aL_out = aR_in + sqrt(gamma) S- hold for the waveguide.
Standard input-output theory for waveguide QED, cited to Refs. [30,31].
domain assumption Atoms spaced by one wavelength and symmetrically coupled, so dynamics is confined to the symmetric Dicke subspace.
Explicit model assumption in the Model section; if spacing or couplings differ, the collective operators change (SM S2).
domain assumption Markov approximation: time delays between atoms are negligible.
Stated in the Model section; the authors acknowledge in Discussion that this warrants future investigation.
domain assumption Weak-drive approximation |g| << 1 (and implicitly |g| << 1/sqrt(N) for the D approximation).
Used to derive Eqs. (3)-(6); the stronger condition for large N is not stated explicitly.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Superbunching from coherently driven atoms in a waveguide." pith.science (2026). https://pith.science/paper/Y7I6B2MB

@misc{pith2026250605147,
  author       = {Pith},
  title        = {Pith review of: Superbunching from coherently driven atoms in a waveguide},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Y7I6B2MB}},
  note         = {Machine review of arXiv:2506.05147}
}

abstract

We investigate the scattered field from $N$ identical two-level atoms resonantly driven by a weak coherent field in a one-dimensional waveguide. For atoms separated by the drive wavelength, increasing the number of atoms progressively suppresses transmission while enhancing photon bunching. Transmission becomes a superbunched $(N+1)$-photon scattering process that is predominantly incoherent. Remarkably, we find that this transmission is only possible through a process where all $N$ atoms are excited, enabling heralded multi-photon state generation with applications in long-distance entanglement and quantum metrology.

Figures

Figures reproduced from arXiv: 2506.05147 by the authors.

**Figure 2.** FIG. 2. Bunching probability versus effective drive amplitude [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** FIG. 3. Fidelity of the output field with the entangled bino [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: FIG. 4. Bunching probability versus effective drive amplitude Ω [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

65 extracted references · 59 canonical work pages

[1]

as ˙Ä = 2D(aR out)Ä, to immediately obtain the steady-state solution [ 34–39]: Ä = (aR† outaR out)− 1 Tr ( aR† outaR out ) − 1 = 1 D N∑ m,n =0 ( S− g∗ ) m ( S+ g ) n , (2) where the normalized drive amplitude g = ³/√µ = iΩ/µ and the normalization constant arXiv:2506.05147v1 [quant-ph] 5 Jun 2025 2 λ λaL in aL out aR out γ γ γ γ FIG. 1. N identical atoms, ...

work page Pith review arXiv 2025
[2]

(5) In the strong driving regime, |g| k 1, the conditional state in Eq

on detecting a transmitted pho- ton, we obtain the maximally mixed state Äc = aR outÄaR† out ⟨ aR† outaR out ⟩ = IN +1 N + 1. (5) In the strong driving regime, |g| k 1, the conditional state in Eq. ( 5), to leading order, coincides with the steady state in Eq. ( 2), as the drive saturates the atomic system and photon detections have a negligible eﬀect on ...

work page
[3]

reveals that transmission requires driving the atomic system with suﬃcient photons to excite all N atoms. This behavior can be understood through two perspectives: (1) N atoms can coherently reﬂect Fock states of up to N photons, while states with N +1 or more photons enable incoherent transmission. (2) Although the average drive amplitude is in the linea...

work page
[4]

R. H. Dicke, Physical Review 93, 99–110 (1954)

work page 1954
[5]

This raises the question of how to understand this state in the weak driving regime, |g| j 1, where the steady state is close to the ground state

is independent of the normalized drive amplitude g. This raises the question of how to understand this state in the weak driving regime, |g| j 1, where the steady state is close to the ground state. In this regime, while the prob- ability of detecting a transmitted photon is small, pro- portional to |g|2(N +1), when such a photon is detected, Eq. (

work page
[6]

This can be understood by highlighting the fact that we are conditioning on detecting any trans- mitted photon

tells us that the atomic system has an average of N/2 excitations. This can be understood by highlighting the fact that we are conditioning on detecting any trans- mitted photon. Transmission only occurs in bunches, as seen from Eqs. (

work page
[7]

This lack of information about which excitation level the transmitted photon was emitted from leads to maximal uncertainty in the atomic state

and ( 4), following the collective exci- tation of all atoms, and each photon in the bunch could have been emitted from any excitation level of the sys- tem. This lack of information about which excitation level the transmitted photon was emitted from leads to maximal uncertainty in the atomic state. To determine the atomic state after a speciﬁc trans- mi...

work page
[8]

without transmitted clicks by removing the term aR outÄaR† out. When we condition the steady state of this no-click evolution on an inevitable 3 transmitted detection, we obtain the atomic state after the ﬁrst transmitted click in a photon bunch. Using |kðex to denote the state with k excited atoms, this ﬁrst-click state is given by [ 31]: Äﬁrst c = N∑ k=...

work page

Show all 65 references

[9]

without both reﬂected and trans- mitted clicks by removing the jump terms aL outÄaL† out and aR outÄaR† out. Indeed, conditioning the steady state of this no-click evolution on an inevitable transmitted detection yields the fully excited atomic state |N ðex, conﬁrming that tra...
[10]

37 and ˆTout = 0

with the ﬁtted timescales ˆTin = 0. 37 and ˆTout = 0. 93. decay times between adjacent states follow exponential distributions with rates Γ k. The total relaxation time follows a hypoexponential distribution [ 43–45], which is the sum of independent exponential random variable...
[11]

Here we present results for N = 3 atoms, while results for N = 1 and 2 atoms can be found in the End Matter. For our analysis, a photon bunch is deﬁned by the number of clicks in the reﬂected and transmitted ﬁelds within a bunching time Tout af- ter, and including, the ﬁrst tr...
[12]

For example, an ( N + 2)-photon bunch arises from ei- therN + 2 initial photons with no additional arrivals, or N +1 initial photons plus one additional arrival

captures all possible combinations of initial and additional photons that result in a k-photon bunch. For example, an ( N + 2)-photon bunch arises from ei- therN + 2 initial photons with no additional arrivals, or N +1 initial photons plus one additional arrival. The ran- dom ...
[13]

Applications—Naturally, our simple setup enables multi-photon state generation and entanglement distri- bution, heralded by the detection of the ﬁrst transmitted photon

can be approximated by p(N + 1) = e− Ω 2T /γ , showing an exponential convergence to (N + 1)-photon bunches [ 31]. Applications—Naturally, our simple setup enables multi-photon state generation and entanglement distri- bution, heralded by the detection of the ﬁrst transmitted ...

2021
[14]

N. V. Corzo, J. Raskop, A. Chandra, A. S. Sheremet, B. Gouraud, and J. Laurat, Nature 566, 359–362 (2019)

2019
[15]

A. S. Sheremet, M. I. Petrov, I. V. Iorsh, A. V. Poshakin- skiy, and A. N. Poddubny, Reviews of Modern Physics 95, 015002 (2023)

2023
[16]

A. S. Prasad, J. Hinney, S. Mahmoodian, K. Ham- merer, S. Rind, P. Schneeweiss, A. S. Sørensen, J. Volz, and A. Rauschenbeutel, Nature Photonics 14, 719–722 (2020)

2020
[17]

Gross and S

M. Gross and S. Haroche, Physics Reports 93, 301–396 (1982)

1982
[18]

Goban, C.-L

A. Goban, C.-L. Hung, J. Hood, S.-P. Yu, J. Muniz, O. Painter, and H. Kimble, Physical Review Letters 115, 063601 (2015)

2015
[19]

Paulisch, M

V. Paulisch, M. Perarnau-Llobet, A. Gonz´ alez-Tudela, and J. I. Cirac, Physical Review A 99, 043807 (2019)

2019
[20]

Perarnau-Llobet, A

M. Perarnau-Llobet, A. Gonz´ alez-Tudela, and J. I. Cirac, Quantum Science and Technology 5, 025003 (2020)

2020
[21]

Khanahmadi and K

M. Khanahmadi and K. Mølmer, Physical Review A 107, 013705 (2023)

2023
[22]

Blauensteiner, I

B. Blauensteiner, I. Herbauts, S. Bettelli, A. Poppe, and H. H¨ ubel,Phys. Rev. A 79, 063846 (2009)

2009
[23]

Maﬀei, D

M. Maﬀei, D. Pomarico, P. Facchi, G. Magniﬁco, S. Pas- cazio, and F. V. Pepe, Phys. Rev. Res. 6, L032017 (2024)

2024
[24]

A. N. Boto, P. Kok, D. S. Abrams, S. L. Braunstein, C. P. Williams, and J. P. Dowling, Physical Review Letters 85, 2733–2736 (2000)

2000
[25]

Bj¨ ork, L

G. Bj¨ ork, L. L. S´ anchez-Soto, and J. S¨ oderholm,Physical Review Letters 86, 4516–4519 (2001)

2001
[26]

D. E. Chang, A. S. Sørensen, E. A. Demler, and M. D. Lukin, Nature Physics 3, 807–812 (2007)

2007
[27]

Vetsch, D

E. Vetsch, D. Reitz, G. Sagu´ e, R. Schmidt, S. T. Dawkins, and A. Rauschenbeutel, Phys. Rev. Lett. 104, 203603 (2010)

2010
[28]

Pennetta, M

R. Pennetta, M. Blaha, A. Johnson, D. Lechner, P. Schneeweiss, J. Volz, and A. Rauschenbeutel, Phys- ical Review Letters 128, 073601 (2022)

2022
[29]

Lodahl, A

P. Lodahl, A. Floris van Driel, I. S. Nikolaev, A. Irman, K. Overgaag, D. Vanmaekelbergh, and W. L. Vos, Nature 430, 654–657 (2004)

2004
[30]

Astaﬁev, A

O. Astaﬁev, A. M. Zagoskin, A. A. Abdumalikov, Y. A. Pashkin, T. Yamamoto, K. Inomata, Y. Nakamura, and J. S. Tsai, Science 327, 840–843 (2010)

2010
[31]

A. A. Abdumalikov, O. Astaﬁev, A. M. Zagoskin, Y. A. Pashkin, Y. Nakamura, and J. S. Tsai, Physical Review Letters 104, 193601 (2010)

2010
[32]

I.-C. Hoi, C. M. Wilson, G. Johansson, T. Palomaki, B. Peropadre, and P. Delsing, Physical Review Letters 107, 073601 (2011)

2011
[33]

I.-C. Hoi, T. Palomaki, J. Lindkvist, G. Johansson, P. Delsing, and C. M. Wilson, Physical Review Letters 108, 263601 (2012)

2012
[34]

I.-C. Hoi, C. M. Wilson, G. Johansson, J. Lindkvist, B. Peropadre, T. Palomaki, and P. Delsing, New Jour- nal of Physics 15, 025011 (2013)

2013
[35]

A. F. van Loo, A. Fedorov, K. Lalumi` ere, B. C. Sanders, A. Blais, and A. Wallraﬀ, Science 342, 1494–1496 (2013)

2013
[36]

Ficek and S

Z. Ficek and S. Swain, Quantum Interference and Coher- ence (Springer-Verlag, 2005)

2005
[37]

J. T. Shen and S. Fan, Optics Letters 30, 2001 (2005)

2005
[38]

Shen and S

J.-T. Shen and S. Fan, Phys. Rev. Lett. 95, 213001 (2005)

2005
[39]

Zheng, D

H. Zheng, D. J. Gauthier, and H. U. Baranger, Physical Review A 82, 063816 (2010)

2010
[40]

Peropadre, J

B. Peropadre, J. Lindkvist, I.-C. Hoi, C. M. Wilson, J. J. Garcia-Ripoll, P. Delsing, and G. Johansson, New Jour- nal of Physics 15, 035009 (2013)

2013
[41]

X. H. H. Zhang and H. U. Baranger, Physical Review A 97, 023813 (2018)

2018
[42]

Lalumi` ere, B

K. Lalumi` ere, B. C. Sanders, A. F. van Loo, A. Fedorov, A. Wallraﬀ, and A. Blais, Physical Review A 88, 043806 (2013)

2013
[43]

See Supplemental Material at URL-inserted-by-publishe r for lengthier derivations, which includes Refs. [32-33]
[44]

Riordan, Combinatorial Identities , 99th ed., Wiley se- ries in probability and mathematical statistics (John Wi- ley & Sons, Nashville, TN, 1968)

J. Riordan, Combinatorial Identities , 99th ed., Wiley se- ries in probability and mathematical statistics (John Wi- ley & Sons, Nashville, TN, 1968)

1968
[45]

Mitzenmacher and E

M. Mitzenmacher and E. Upfal, Probability and Comput- ing: Randomized Algorithms and Probabilistic Analysis (Cambridge University Press, 2005)

2005
[46]

R. R. Puri and S. V. Lawande, Physics Letters A 72, 200–202 (1979)

1979
[47]

Puri and S

R. Puri and S. Lawande, Physica A: Statistical Mechanics and its Applications 101, 599–612 (1980)

1980
[48]

P. D. Drummond, Phys. Rev. A 22, 1179 (1980)

1980
[49]

S. V. Lawande, R. R. Puri, and S. S. Hassan, Journal of Physics B: Atomic and Molecular Physics 14, 4171–4189 (1981)

1981
[50]

Hassan, R

S. Hassan, R. Bullough, R. Puri, and S. Lawande, Phys- ica A: Statistical Mechanics and its Applications 103, 213–225 (1980)

1980
[51]

S. S. Hassan, G. P. Hildred, R. R. Puri, and R. K. Bullough, Journal of Physics B: Atomic and Molecular Physics 15, 2635–2655 (1982)

1982
[52]

Mølmer, Y

K. Mølmer, Y. Castin, and J. Dalibard, Journal of the Optical Society of America B 10, 524 (1993) . 6

1993
[53]

Breuer and F

H.-P. Breuer and F. Petruccione, The Theory of Open Quantum Systems (Oxford University Press, 2007)

2007
[54]

H. M. Wiseman and G. J. Milburn, Quantum Measure- ment and Control (Cambridge University Press, 2009)

2009
[55]

Amari and R

S. Amari and R. Misra, IEEE Transactions on Reliability 46, 519–522 (1997)

1997
[56]

Bolch, S

G. Bolch, S. Greiner, H. de Meer, and K. S. Trivedi, Queueing Networks and Markov Chains: Modeling and Performance Evaluation With Computer Science Appli- cations (Wiley, 2006)

2006
[57]

Legros and O

B. Legros and O. Jouini, Applied Mathematical Mod- elling 39, 4971–4977 (2015)

2015
[58]

Lambert, E

N. Lambert, E. Gigu` ere, P. Menczel, B. Li, P. Hopf, G. Su´ arez, M. Gali, J. Lishman, R. Gadhvi, R. Agarwal, A. Galicia, N. Shammah, P. Nation, J. R. Johansson, S. Ahmed, S. Cross, A. Pitchford, and F. Nori, Qutip 5: The quantum toolbox in python (2024)

2024
[59]

A. H. Kiilerich and K. Mølmer, Phys. Rev. Lett. 123, 123604 (2019)

2019
[60]

Khanahmadi, M

M. Khanahmadi, M. M. Lund, K. Mølmer, and G. Jo- hansson, Physical Review Research 5, 043071 (2023)

2023
[61]

S. R. Sathyamoorthy, A. Bengtsson, S. Bens, M. Simoen, P. Delsing, and G. Johansson, Phys. Rev. A 93, 063823 (2016). End Matter 10.0 0.2 0.4 0.6 (a) |0⟩ |1⟩ |2⟩ 0.5 0.33 0 1 (b) |0⟩ |1⟩ |2⟩ 0.5 0.5 1.0 1.5 Ω γ √ N 0.0 0.2 0.4 0.6 0.8 1.0 (c) |0⟩ |1⟩ |2⟩ 13 6 4 7 5 Bunching pro...

2016
[62]

( S6a), and the leading-order term in Eq

Strong driving regime In the strong driving regime, we have that D ∼ N + 1 from Eq. ( S6a), and the leading-order term in Eq. ( S17) is given by setting r = n in Eq. ( S5c), and since Hn = ( N +n+1 2n+1 ) (n!)2, we obtain G(n) ref (0) = γ n ( N + n + 1 2n + 1 ) (n!)2 N + 1 + O...
[63]

( S6a), and the leading-order term is given by setting r = N in Eq

Weak driving regime For weak external ﬁeld driving, we have that D ∼ (N !)2|g|−2N from Eq. ( S6a), and the leading-order term is given by setting r = N in Eq. ( S5c), and since HN = (N !)2, we obtain G(n) ref (0) = γ n|g|2n + O ( |g|2(N +1)) , |g| j 1, (S20) and the coherence ...
[64]

( S6a), and the leading-order term is given by setting k = j = r = 0 in Eq

Strong driving regime Once again, for strong external ﬁeld driving, we have that D ∼ N + 1 from Eq. ( S6a), and the leading-order term is given by setting k = j = r = 0 in Eq. ( S24), and since H0 = (N +1)! N ! = N + 1, we obtain G(n) trans(0) = γ n|g|2n + O ( |g|2(n−1)) , |g|...
[65]

Weak driving regime The leading-order term for weak driving is obtained for the largest r in Eq. ( S24). At ﬁrst glance, it seems that r = N would give us the leading-order term but then the outer summations sum to zero. To proceed, we instead notice that N∑ r=max(k,j ) |g|−2r...

1979

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

as ˙Ä = 2D(aR out)Ä, to immediately obtain the steady-state solution [ 34–39]: Ä = (aR† outaR out)− 1 Tr ( aR† outaR out ) − 1 = 1 D N∑ m,n =0 ( S− g∗ ) m ( S+ g ) n , (2) where the normalized drive amplitude g = ³/√µ = iΩ/µ and the normalization constant arXiv:2506.05147v1 [quant-ph] 5 Jun 2025 2 λ λaL in aL out aR out γ γ γ γ FIG. 1. N identical atoms, ...

work page Pith review arXiv 2025

[2] [2]

(5) In the strong driving regime, |g| k 1, the conditional state in Eq

on detecting a transmitted pho- ton, we obtain the maximally mixed state Äc = aR outÄaR† out ⟨ aR† outaR out ⟩ = IN +1 N + 1. (5) In the strong driving regime, |g| k 1, the conditional state in Eq. ( 5), to leading order, coincides with the steady state in Eq. ( 2), as the drive saturates the atomic system and photon detections have a negligible eﬀect on ...

work page

[3] [3]

reveals that transmission requires driving the atomic system with suﬃcient photons to excite all N atoms. This behavior can be understood through two perspectives: (1) N atoms can coherently reﬂect Fock states of up to N photons, while states with N +1 or more photons enable incoherent transmission. (2) Although the average drive amplitude is in the linea...

work page

[4] [4]

R. H. Dicke, Physical Review 93, 99–110 (1954)

work page 1954

[5] [5]

This raises the question of how to understand this state in the weak driving regime, |g| j 1, where the steady state is close to the ground state

is independent of the normalized drive amplitude g. This raises the question of how to understand this state in the weak driving regime, |g| j 1, where the steady state is close to the ground state. In this regime, while the prob- ability of detecting a transmitted photon is small, pro- portional to |g|2(N +1), when such a photon is detected, Eq. (

work page

[6] [6]

This can be understood by highlighting the fact that we are conditioning on detecting any trans- mitted photon

tells us that the atomic system has an average of N/2 excitations. This can be understood by highlighting the fact that we are conditioning on detecting any trans- mitted photon. Transmission only occurs in bunches, as seen from Eqs. (

work page

[7] [7]

This lack of information about which excitation level the transmitted photon was emitted from leads to maximal uncertainty in the atomic state

and ( 4), following the collective exci- tation of all atoms, and each photon in the bunch could have been emitted from any excitation level of the sys- tem. This lack of information about which excitation level the transmitted photon was emitted from leads to maximal uncertainty in the atomic state. To determine the atomic state after a speciﬁc trans- mi...

work page

[8] [8]

without transmitted clicks by removing the term aR outÄaR† out. When we condition the steady state of this no-click evolution on an inevitable 3 transmitted detection, we obtain the atomic state after the ﬁrst transmitted click in a photon bunch. Using |kðex to denote the state with k excited atoms, this ﬁrst-click state is given by [ 31]: Äﬁrst c = N∑ k=...

work page