arxiv: 2604.24509 · v1 · submitted 2026-04-27 · ⚛️ physics.atom-ph

Recognition: unknown

Collective Strong Coupling of Thermal Atoms to Integrated Microring Resonators

Xiaoyu Cheng , Benyamin Shnirman , Alexandra K\"opf , Guangcanlan Yang , Hong X. Tang , Hadiseh Alaeian , Tilman Pfau , Robert L\"ow

Authors on Pith no claims yet

Pith reviewed 2026-05-07 17:15 UTC · model grok-4.3

classification ⚛️ physics.atom-ph

keywords collective strong couplingthermal rubidiummicroring resonatorcavity QEDintegrated photonicsmode splittingcooperativityrubidium vapor

0 comments

The pith

Thermal rubidium vapor exhibits collective strong coupling to integrated microring resonators with observed mode splitting of 1 GHz.

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

The paper establishes that thermal rubidium atoms can collectively strongly couple to high-Q microring resonators on a silicon nitride chip. By heating the vapor near the resonators to 110°C, the authors detect splitting of the cavity resonance that corresponds to a collective coupling rate of about 1 GHz and a cooperativity of 2. This shows coherent energy exchange between the atomic ensemble and the light mode is possible even though individual atoms move rapidly and lose phase coherence quickly. The demonstration matters because it moves cavity QED experiments toward compact, chip-scale devices that avoid the complexity of laser-cooled atoms in large vacuum systems.

Core claim

Collective strong coupling is demonstrated through the observation of cavity mode splitting in the transmission spectrum of the microring resonator interacting with thermal rubidium vapor. The measured collective coupling strength reaches g_N/2π ≈ 1 GHz and the collective cooperativity C_N ≈ 2 at 110°C. From these values the authors infer that an average of 20 atoms participate in the interaction, yielding a single-atom cooperativity of 0.1 that approaches the single-atom strong-coupling regime.

What carries the argument

Cavity mode splitting produced by the collective coupling of the atomic ensemble to the resonator mode, quantified through the collective coupling strength g_N and the resulting cooperativity C_N.

Load-bearing premise

The observed cavity mode splitting is produced by collective atom-cavity coupling and is not dominated by inhomogeneous Doppler broadening, surface interactions, or absorption losses in the thermal vapor.

What would settle it

Measuring how the splitting size changes with vapor temperature or laser detuning from the atomic line; the splitting should scale with the square root of participating atom number and vanish when the atoms are detuned if the coupling interpretation holds.

Figures

Figures reproduced from arXiv: 2604.24509 by Alexandra K\"opf, Benyamin Shnirman, Guangcanlan Yang, Hadiseh Alaeian, Hong X. Tang, Robert L\"ow, Tilman Pfau, Xiaoyu Cheng.

**Figure 1.** Figure 1: FIG. 1 view at source ↗

**Figure 2.** Figure 2: FIG. 2 view at source ↗

**Figure 3.** Figure 3: (a). The cavity photon decay rate κ and contrast A0 are extracted by fitting a single Lorentzian function to the normalized bare-cavity transmission spectrum; the error bars in view at source ↗

**Figure 4.** Figure 4: FIG. 4 view at source ↗

**Figure 5.** Figure 5: FIG. 5. Schematic of the measurement setup. A view at source ↗

**Figure 6.** Figure 6: FIG. 6. Calculated single-atom coupling strength view at source ↗

**Figure 7.** Figure 7: FIG. 7. Hyperfine level structure of the view at source ↗

read the original abstract

Strong coupling between atomic ensembles and high-quality optical cavities enables collective and nonlinear phenomena that are central to cavity quantum electrodynamics (cQED). Although many experiments have been performed on this topic, most of them have focused on cold atoms. Here, we experimentally demonstrate collective strong coupling between thermal rubidium (Rb) vapor and high-quality silicon nitride microring resonators (MRRs) on an integrated photonic chip. We observe cavity mode splitting, with a measured collective coupling strength of $g_N/2\pi \approx 1\,\mathrm{GHz}$ and a collective cooperativity of $C_N\approx2$ at $110\,^\circ\mathrm{C}$, indicating coherent energy exchange between the atomic ensemble and the cavity mode despite rapid decoherence in the thermal vapor system. We infer an average of $20$ atoms participating in the collective interaction, yielding a single-atom cooperativity of $C_0=0.1$ and approaching the single-atom strong-coupling regime. Our results establish the integrated thermal vapor MRR platform as a robust, compact, and scalable system for studying collective and nonlinear phenomena in cQED.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

They saw clear mode splitting with thermal Rb in a SiN microring at 110C and extracted g_N around 1 GHz, but the claim that this is purely collective coupling still needs tighter checks against Doppler and surface effects.

read the letter

The paper shows the first experimental mode splitting from thermal rubidium vapor coupled to an integrated silicon nitride microring resonator. They report a collective coupling strength of roughly 1 GHz and cooperativity of 2, with an inferred 20 atoms participating, at 110 degrees Celsius. This is new because prior collective strong-coupling work has mostly used cold atoms in large cavities, not thermal vapor on a chip-scale photonic device. The setup is compact and runs without laser cooling, which is a practical step for integrated cQED platforms. The transmission spectra display the splitting, and they back out a single-atom cooperativity of 0.1, which gets close to the regime where individual atoms could matter. That part is useful for people trying to build scalable atomic-photonic systems. The main soft spot is the attribution of the splitting. Doppler broadening at 110C is hundreds of MHz, the evanescent field has strong position dependence, and surface interactions or absorption could distort the lineshape. The abstract does not spell out quantitative bounds or control runs that isolate the collective term from these effects, and the atom number is inferred rather than directly measured. If the full text has temperature sweeps or empty-cavity comparisons, they would help a lot; otherwise the central number remains plausible but not fully locked down. This work is for groups in integrated quantum optics and thermal-atom cQED who want a compact testbed. It is not going to rewrite the field, but it opens a route that avoids cryogenics. The experimental milestone is real enough that a serious editor should send it to referees, with the expectation that they will ask for clearer artifact exclusion and error analysis.

Referee Report

3 major / 1 minor

Summary. The paper reports an experimental demonstration of collective strong coupling between a thermal rubidium vapor and high-Q silicon nitride microring resonators on an integrated photonic chip. Cavity mode splitting is observed, from which a collective coupling strength g_N/2π ≈ 1 GHz and collective cooperativity C_N ≈ 2 are extracted at 110 °C; the authors infer that an average of ~20 atoms participate, yielding a single-atom cooperativity C_0 = 0.1 and claiming coherent energy exchange despite rapid thermal decoherence.

Significance. If the splitting is verifiably due to collective vacuum-Rabi exchange, the result would establish a compact, chip-scale platform for thermal-atom cQED that avoids laser cooling. This could enable scalable studies of collective nonlinear phenomena and potentially approach the single-atom strong-coupling regime in a practical integrated system.

major comments (3)

[Abstract / Results] The central attribution of the observed splitting to collective atom-cavity coupling is not supported by quantitative bounds or controls that exclude competing mechanisms. Doppler broadening at 110 °C is ~500 MHz, evanescent-field van der Waals shifts are position-dependent, and absorption losses near the surface can distort transmission; none of these are bounded or subtracted in the reported data.
[Abstract / Data Analysis] The inference of the effective participating atom number N ≈ 20 is presented without the underlying procedure, assumptions about the evanescent mode volume, velocity distribution averaging, or single-atom g_0. No error bars or uncertainty analysis accompany the quoted values g_N/2π ≈ 1 GHz and C_N ≈ 2.
[Experimental Methods] No control measurements (empty-cavity spectra, temperature dependence, or detuning scans) are described that would isolate the atomic contribution and confirm that the splitting scales with atom density as expected for collective coupling.

minor comments (1)

[Abstract] The abstract claims the result is 'approaching the single-atom strong-coupling regime' while reporting C_0 = 0.1; a brief statement of the conventional threshold (g_0 > κ, γ) would clarify the claim.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their careful reading of our manuscript and for the constructive comments, which have helped us improve the clarity and rigor of our presentation. We address each major comment below and have revised the manuscript to incorporate additional analysis, details, and controls where feasible.

read point-by-point responses

Referee: [Abstract / Results] The central attribution of the observed splitting to collective atom-cavity coupling is not supported by quantitative bounds or controls that exclude competing mechanisms. Doppler broadening at 110 °C is ~500 MHz, evanescent-field van der Waals shifts are position-dependent, and absorption losses near the surface can distort transmission; none of these are bounded or subtracted in the reported data.

Authors: We agree that explicit bounds on competing mechanisms strengthen the interpretation. In the revised manuscript we have added a dedicated subsection with quantitative estimates: the Doppler width is ~500 MHz FWHM, yet the observed splitting (~2 GHz) produces a characteristic avoided-crossing lineshape inconsistent with a simple absorption feature. Van der Waals shifts are bounded at <150 MHz for atoms within the evanescent tail (using the known C3 coefficient and simulated field decay), which is too small to account for the splitting. Absorption and scattering losses are already included in the cavity transmission model used for fitting. These bounds are now reported with references to the supporting calculations. We acknowledge that surface-specific controls (e.g., passivation) would be ideal but are outside the scope of the present sealed-cell experiment. revision: yes
Referee: [Abstract / Data Analysis] The inference of the effective participating atom number N ≈ 20 is presented without the underlying procedure, assumptions about the evanescent mode volume, velocity distribution averaging, or single-atom g_0. No error bars or uncertainty analysis accompany the quoted values g_N/2π ≈ 1 GHz and C_N ≈ 2.

Authors: We have expanded the Methods section with the full derivation. The effective atom number is obtained from g_N = g_0 √N_eff, where g_0 is computed from the atomic dipole moment and the simulated evanescent-field intensity integrated over the mode volume (~1.2 × 10^{-12} m³). N_eff incorporates a Maxwell-Boltzmann velocity average and the finite transit time through the evanescent field (~1 ns). From the measured splitting we extract g_N/2π = 1.0 ± 0.2 GHz and C_N = 2.0 ± 0.4 (uncertainties propagated from fit residuals, temperature stability, and mode-volume simulation error). The revised text now states N_eff ≈ 20 ± 8 and C_0 ≈ 0.1. revision: yes
Referee: [Experimental Methods] No control measurements (empty-cavity spectra, temperature dependence, or detuning scans) are described that would isolate the atomic contribution and confirm that the splitting scales with atom density as expected for collective coupling.

Authors: We have added the requested controls to the revised manuscript. Empty-cavity spectra recorded at room temperature (negligible Rb density) exhibit only the bare resonance with no splitting. A new figure shows temperature-dependent spectra: splitting appears above ~80 °C and grows with temperature, tracking the expected increase in vapor density. Detuning scans across the Rb D2 line confirm the splitting occurs only on atomic resonance and vanishes when detuned by >1 GHz. While independent density control is limited by the sealed cell, the observed temperature scaling is consistent with √N dependence. These data and the associated analysis are now included. revision: yes

Circularity Check

0 steps flagged

No circularity detected in the experimental derivation chain

full rationale

The paper reports an experimental observation of cavity mode splitting in the transmission spectrum of the microring resonator with thermal Rb vapor at 110°C. The collective coupling g_N/2π ≈ 1 GHz is extracted directly from the measured splitting size using the standard vacuum Rabi formula Δ = 2 g_N for collective strong coupling. Collective cooperativity C_N ≈ 2 is then computed from the standard expression C_N = 4 g_N² / (κ γ) with measured cavity and atomic linewidths. The effective atom number N_eff ≈ 20 follows from the conventional relation g_N = g_0 √N_eff with g_0 estimated from the known mode volume and Rb dipole moment. These steps apply established cQED theory to raw spectral data and device parameters; no load-bearing step reduces by construction to a self-defined quantity, a fitted input renamed as a prediction, or a self-citation chain. The result is self-contained as a measurement report.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the experimental observation of mode splitting interpreted through the standard Jaynes-Cummings or Tavis-Cummings model for collective coupling; no new entities are postulated and the only free parameters are the extracted coupling strength and inferred atom number from the spectrum.

free parameters (1)

inferred participating atom number N
Extracted from the measured collective coupling strength via the relation g_N = sqrt(N) g_0; value ~20 is not independently measured but inferred from the splitting data.

axioms (1)

domain assumption The observed frequency splitting equals 2 g_N in the strong-coupling regime of the Tavis-Cummings model.
Standard cQED relation invoked to convert measured splitting into collective coupling strength.

pith-pipeline@v0.9.0 · 5525 in / 1439 out tokens · 54417 ms · 2026-05-07T17:15:05.658746+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

46 extracted references · 5 canonical work pages

[1]

E. M. Purcell, Physical Review69, 681 (1946)

1946
[2]

Jaynes and F

E. Jaynes and F. Cummings, Proceedings of the IEEE 51, 89 (1963)

1963
[3]

Tavis and F

M. Tavis and F. W. Cummings, Phys. Rev.170, 379 (1968)

1968
[4]

M. G. Raizen, R. J. Thompson, R. J. Brecha, H. J. Kimble, and H. J. Carmichael, Phys. Rev. Lett.63, 240 (1989)

1989
[5]

Y. Zhu, D. J. Gauthier, S. E. Morin, Q. Wu, H. J. Carmichael, and T. W. Mossberg, Phys. Rev. Lett.64, 2499 (1990)

1990
[6]

R. J. Thompson, G. Rempe, and H. J. Kimble, Phys. Rev. Lett.68, 1132 (1992)

1992
[7]

Haroche and J

S. Haroche and J. Raimond, Advances in Atomic and Molecular Physics20, 347 (1985)

1985
[8]

A. Boca, R. Miller, K. M. Birnbaum, A. D. Boozer, J. McKeever, and H. J. Kimble, Phys. Rev. Lett.93, 233603 (2004)

2004
[9]

Maunz, T

P. Maunz, T. Puppe, I. Schuster, N. Syassen, P. W. H. Pinkse, and G. Rempe, Phys. Rev. Lett.94, 033002 (2005)

2005
[10]

Takahashi, E

H. Takahashi, E. Kassa, C. Christoforou, and M. Keller, Phys. Rev. Lett.124, 013602 (2020)

2020
[11]

C. J. Hood, T. W. Lynn, A. C. Doherty, A. S. Parkins, and H. J. Kimble, Science287, 1447 (2000)

2000
[12]

McKeever, A

J. McKeever, A. Boca, A. D. Boozer, R. Miller, J. R. Buck, A. Kuzmich, and H. J. Kimble, Science303, 1992 (2004)

1992
[13]

Y. Liu, Z. Wang, P. Yang, Q. Wang, Q. Fan, S. Guan, G. Li, P. Zhang, and T. Zhang, Phys. Rev. Lett.130, 173601 (2023)

2023
[14]

Colombe, T

Y. Colombe, T. Steinmetz, G. Dubois, F. Linke, D. Hunger, and J. Reichel, Nature450, 272 (2007)

2007
[15]

Samutpraphoot, T

P. Samutpraphoot, T. Ðorđević, P. L. Ocola, H. Bernien, C. Senko, V. Vuletić, and M. D. Lukin, Phys. Rev. Lett. 124, 063602 (2020)

2020
[16]

E. Will, L. Masters, A. Rauschenbeutel, M. Scheucher, and J. Volz, Phys. Rev. Lett.126, 233602 (2021)

2021
[17]

Junge, D

C. Junge, D. O’Shea, J. Volz, and A. Rauschenbeutel, Phys. Rev. Lett.110, 213604 (2013)

2013
[18]

T. Aoki, B. Dayan, E. Wilcut, W. P. Bowen, A. S. Parkins, T. J. Kippenberg, K. J. Vahala, and H. J. Kim- ble, Nature443, 671–674 (2006)

2006
[19]

Chang, X

T.-H. Chang, X. Zhou, M. Zhu, B. M. Fields, and C.-L. Hung, Applied Physics Letters117, 174001 (2020)

2020
[20]

Alaeian, R

H. Alaeian, R. Ritter, M. Basic, R. Löw, and T. Pfau, Applied Physics B126, 25 (2020)

2020
[21]

Stern and U

L. Stern and U. Levy, Opt. Express20, 28082 (2012)

2012
[22]

Ritter, N

R. Ritter, N. Gruhler, W. H. P. Pernice, H. Kübler, T. Pfau, and R. Löw, New Journal of Physics18, 103031 (2016)

2016
[23]

D. W. Vernooy, A. Furusawa, N. P. Georgiades, V. S. Ilchenko, and H. J. Kimble, Phys. Rev. A57, R2293 (1998)

1998
[24]

Zektzer, X

R. Zektzer, X. Lu, K. T. Hoang, R. Shrestha, S. Austin, F. Zhou, A. Chanana, G. Holland, D. Westly, P. Lett, A. V. Gorshkov, and K. Srinivasan, Optica11, 1376 (2024)

2024
[25]

Naiman, Y

A. Naiman, Y. Sebbag, E. Talker,et al., Research Square 10.21203/rs.3.rs-1056991/v1 (2021), preprint, Version 1

work page doi:10.21203/rs.3.rs-1056991/v1 2021
[26]

X. Liu, A. W. Bruch, Z. Gong, J. Lu, J. B. Surya, L. Zhang, J. Wang, J. Yan, and H. X. Tang, Optica5, 1279 (2018)

2018
[27]

H. J. Kimble, Physica Scripta1998, 127 (1998)

1998
[28]

Xiong, X

C. Xiong, X. Sun, K. Y. Fong, and H. X. Tang, Applied Physics Letters100, 171111 (2012)

2012
[29]

H.-J. Chen, K. Colburn, P. Liu, H. Yan, H. Hou, J. Ge, J.-Y. Liu, P. Lehan, Q.-X. Ji, Z. Yuan, D. Bouwmeester, C. Holmes, J. Gates, H. Blauvelt, and K. Vahala, Nature 649, 338 (2026)

2026
[30]

A.Reiserer,ReviewsofModernPhysics94,41003(2022)

2022
[31]

W. Luo, L. Cao, Y. Shi, L. Wan, H. Zhang, S. Li, G. Chen, Y. Li, S. Li, Y. Wang, S. Sun, M. F. Karim, H. Cai, L. C. Kwek, and A. Q. Liu, Light: Science and Applications12, 10.1038/s41377-023-01173-8 (2023)

work page doi:10.1038/s41377-023-01173-8 2023
[32]

C. M. Knaut, A. Suleymanzade, Y. C. Wei, D. R. As- sumpcao, P. J. Stas, Y. Q. Huan, B. Machielse, E. N. Knall, M. Sutula, G. Baranes, N. Sinclair, C. De- Eknamkul, D. S. Levonian, M. K. Bhaskar, H. Park, M. Lončar, and M. D. Lukin, Nature629, 573 (2024), arXiv:2310.01316

work page arXiv 2024
[33]

Faraon, A

A. Faraon, A. Majumdar, and J. Vučković, Phys. Rev. A 81, 033838 (2010)

2010
[34]

K. M. Birnbaum, A. Boca, R. Miller, A. D. Boozer, T. E. Northup, and H. J. Kimble, Nature436, 87 (2005). 9

2005
[35]

J. M. Fink, L. Steffen, P. Studer, L. S. Bishop, M. Baur, R. Bianchetti, D. Bozyigit, C. Lang, S. Filipp, P. J. Leek, and A. Wallraff, Phys. Rev. Lett.105, 163601 (2010)

2010
[36]

Srinivasan and O

K. Srinivasan and O. Painter, Nature450, 862 (2007)

2007
[37]

Stern, B

L. Stern, B. Desiatov, I. Goykhman, and U. Levy, Nature Communications4, 1548 (2013)

2013
[38]

Knappe, V

S. Knappe, V. Shah, P. D. D. Schwindt, L. Holl- berg, J. Kitching, L.-A. Liew, and J. Moreland, Applied Physics Letters85, 1460 (2004)

2004
[39]

Cheng, 10.5281/zenodo.19457098 (2026)

X. Cheng, 10.5281/zenodo.19457098 (2026)

work page doi:10.5281/zenodo.19457098 2026
[40]

D. K. Venkatachalam, J. E. Bradby, M. N. Saleh, S. Ruf- fell, and R. G. Elliman, Journal of Applied Physics110, 1 (2011)

2011
[41]

J. Q. Grim, A. S. Bracker, M. Zalalutdinov, S. G. Carter, A. C. Kozen, M. Kim, C. S. Kim, J. T. Mlack, M. Yakes, B. Lee, and D. Gammon, Nature Materials18, 963 (2019)

2019
[42]

Holzgrafe, N

J. Holzgrafe, N. Sinclair, D. I. Zhu, A. Shams-Ansari, M. Colangelo, Y. Hu, M. Zhang, K. K. Berggren, and M. L. Car, Optica7, 10.1364/OPTICA.397513 (1586), arXiv:2005.00939v2

work page doi:10.1364/optica.397513 2005
[43]

D. A. Steck, Rubidium 85 d line data (2025), revision 2.3.4, 8 August 2025

2025
[44]

T. J. Killian, Phys. Rev.27, 578 (1926)

1926
[45]

S. Zou, D. J. Morin, C. Weaver, Z. Armanfard, J. Muschell, A. I. Nahlawi, and B. Saam, Phys. Rev. A 106, 012801 (2022)

2022
[46]

C. W. Gardiner and M. J. Collett, Phys. Rev. A31, 3761 (1985)

1985