Pith. sign in

REVIEW 3 major objections 7 minor 38 references

Compact Fiber-Coupled Narrowband Two-Mode Squeezed Light Source

T0 review · 3 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A rubidium four-wave-mixing source delivers 4.4 dB of two-mode squeezing after single-mode fibers at 795 nm.

desk verdict A credible, useful engineering demonstration of a fiber-coupled narrowband squeezed-light source, but the headline 4.4 dB needs error bars and a tighter shot-noise calibration audit. read the letter →

arxiv 2507.03755 v1 pith:NQ3ADQX5 submitted 2025-07-04 quant-ph physics.atom-ph

Umang Jain , Jae Choi , Christopher Hull , Alberto M. Marino This is my paper · ORCID
classification quant-phphysics.atom-ph
keywords two-modesqueezedlightfour-wavemixingrubidiumvaporfiber-coupledquantumsourceintensity-differencesqueezingnarrowbandshot-noiselimitatomicmemory
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 reports a compact source of narrowband two-mode squeezed light at 795 nm, generated by four-wave mixing in a hot $^{85}$Rb vapor cell. The twin beams are coupled into separate single-mode fibers, and the measured intensity-difference noise sits 4.4 dB below the shot-noise level at 1 MHz after the fibers, with a pump power of only 135 mW. The result matters because narrowband, fiber-delivered squeezed light at the rubidium D1 wavelength is what atomic quantum memories and atomic sensors require, and a low size, weight, and power design would make such quantum resources usable outside the laboratory. The paper also documents the cost of fiber delivery: bypassing the fibers gives 7.2 dB of squeezing, and the roughly 2 dB penalty is attributed mainly to uncorrelated spatial modes entering the fibers rather than to simple optical loss.

What carries the argument

The central object is the intensity-difference squeezed twin-beam pair produced by four-wave mixing in a double-$\Lambda$ configuration in $^{85}$Rb, where two pump photons are absorbed and correlated photon pairs are emitted at the probe and conjugate frequencies. The carrying mechanism is spatial mode matching: the pump and seed are sized so the bright part of each twin beam matches a single coherence area of the process, ensuring that mostly correlated spatial modes enter the single-mode fibers, while uncorrelated vacuum modes that couple in add thermal noise. A balanced-detection calibration, with the coherent probe split into two equal beams and the pump blocked, sets the shot-noise reference used for the reported squeezing values.

What would settle it

Replace the shot-noise calibration with an independent reference, for example a coherent beam attenuated to the same total photocurrent and measured on a separate balanced detector, then compare the twin-beam intensity-difference noise at 1 MHz; the 4.4 dB claim fails unless the twin-beam noise remains about 4.4 dB below that independently calibrated floor.

Watch

Extended reading notes

Core claim

The paper's central claim is that a double-$\Lambda$ four-wave-mixing process in isotopically pure $^{85}$Rb vapor can be packaged as a modular, all-fiber-interface source of two-mode squeezed light that preserves quantum correlations in the delivered beams. With the pump and seed beams delivered through polarization-maintaining fibers and overlapping at 0.4° in a 12 mm cell heated to about 99°C, the generated probe and conjugate beams carry roughly 200 µW each. After coupling into dedicated single-mode fibers at greater than 90% efficiency, the intensity-difference noise is 4.4 dB below the shot-noise level at 1 MHz, whereas bypassing the fibers yields 7.2 dB. The paper argues that the gap is dominated by coupling of uncorrelated spatial modes of the spatially multimode twin beams into the fibers, and it models that excess noise as thermal noise entering a beamsplitter through the unused port.

Load-bearing premise

The reported 4.4 dB rests on treating a coherent probe beam, with the pump blocked and split into two equal beams, as an accurate shot-noise reference for the twin beams at the same total power; if balanced detection leaves technical noise uncancelled or that calibration is biased, the squeezing figure would be overestimated.

Editorial extensions

If this is right

  • A rubidium-based quantum memory or sensor can receive 795 nm squeezed light through a standard single-mode fiber pigtail, removing the free-space alignment usually needed at an atomic interface.
  • The 135 mW pump power, about one quarter of the power used in earlier atomic-vapor four-wave-mixing setups, makes the thermal and electrical budget compatible with portable platforms.
  • The 4.4 dB measured after the fibers, not the 7.2 dB before them, is the resource available to a downstream application, so system designers should budget for at least 2 dB of delivery penalty.
  • With fiber coupling efficiency already above 90%, the clearest route to higher delivered squeezing is reducing reflections from uncoated fiber ends and improving spatial mode matching into the fibers.

Reading between the lines

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

  • Beyond the paper's claims, the same fiber-mode-matching strategy should transfer to other spatially multimode quantum light sources, where uncorrelated modes similarly degrade fiber-delivered squeezing.
  • The paper's own loss budget suggests a concrete next experiment: antireflection-coating the output fiber ends should recover the reflection portion of the roughly 2 dB gap, and any residual gap would test whether thermal uncorrelated modes account for the rest.
  • A direct check of the thermal-mode interpretation would be to measure the noise of each fiber-coupled beam separately; uncorrelated thermal modes should appear as excess single-beam noise above the coherent-state level.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 7 minor

Summary. The manuscript reports a compact fiber-coupled source of two-mode squeezed light at 795 nm based on four-wave mixing in hot 85Rb vapor. Pump and probe beams are delivered through polarization-maintaining fibers and intersect in a heated vapor cell; the generated probe and conjugate beams are coupled into separate single-mode PM fibers. The authors measure intensity-difference squeezing of -7.2 dB before the output fibers and -4.4 dB after the fibers at 1 MHz, with >90% coupling efficiency and 135 mW pump power. They attribute the 1.9 dB gap between the after-fiber value and the loss-only expectation (-6.3 dB) to coupling of uncorrelated spatial modes, quantifying this with a heuristic beam-splitter model. The central claim is the compact, low-SWaP, fiber-delivered >4 dB squeezed source.

Significance. If the measured -4.4 dB after-fiber squeezing is correct, the result would be a useful step towards deployable narrowband squeezed-light sources for atomic memories, quantum networking, and sensing. The paper's strengths are its compact 12-inch by 12-inch footprint, low pump power (135 mW), full fiber-coupling of both input and output beams, and an honest comparison between measured and loss-limited squeezing. However, the quantitative claim depends on a shot-noise calibration whose validity is not demonstrated, and no statistical uncertainties are provided, so the result must be treated as preliminary until those are supplied.

major comments (3)
  1. [§II, paragraph 4 and §III, Fig. 4] The reported -4.4 dB after-fiber squeezing is a ratio to a shot-noise level measured with the pump blocked and the probe split 50/50 on a balanced detector. The manuscript gives no evidence that this reference is shot-noise limited at 1 MHz: it does not report the electronic noise floor relative to the signal, the detector common-mode rejection ratio, a linear-power scaling check, or the residual technical noise with the pump blocked. Without these checks, a calibration floor elevated by technical noise would make the measured squeezing appear larger and could explain part of the 1.9 dB discrepancy with the -6.3 dB loss-only expectation. This verification is load-bearing for the central claim.
  2. [§III and Fig. 4] All squeezing values are reported as single traces with no error bars, no repeated measurements, and no confidence intervals. Since the central deliverable is 'over 4 dB of squeezing after fibers,' the authors should report statistics over repeated alignments and measurements at the analysis frequency (e.g., mean ± standard deviation or a confidence interval) to establish that 4.4 dB is a robust value and not a single alignment optimum.
  3. [§III, paragraph 3 and Eq. (1)] The model S' = ηS0 + (1−η)[εNth + (1−ε)Nv] is used to attribute the 1.9 dB discrepancy to the coupling of uncorrelated spatial modes, but its inputs are not measured: ε is fixed at 0.5 and Nth is described only as 'consistent with that of the initial level of squeezing.' The authors should either measure the spatial-mode-matching penalty directly (e.g., by varying the fiber input aperture or comparing coupling of bright and dark regions) or clearly label the calculation as an illustrative bound. As written, the statement that the discrepancy is 'mainly due to' mode matching is not established.
minor comments (7)
  1. [§III, Eq. (1)] The quantities in Eq. (1) are not defined as being in linear or dB units; please state that all terms are in linear units (or provide the conversion), and define Nth precisely.
  2. [§III and Fig. 4] The green trace is computed assuming 10% loss, but the text reports coupling efficiencies 'greater than 90%' for both beams; please specify the exact coupling efficiencies used for the expected curve and acknowledge that 10% is a lower bound.
  3. [§II, paragraph 5] The phrase 'f-to-f optical system' should be defined (e.g., a telescope or a pair of lenses separated by the sum of their focal lengths) for readers unfamiliar with the term.
  4. [§III] Please give the spectrum analyzer settings (resolution bandwidth, video bandwidth, averaging, and measurement time) used for the traces in Fig. 4.
  5. [§III] The pump power range for maximum squeezing (135–250 mW) is stated but no supporting data are shown; a simple squeezing-versus-pump-power curve would strengthen the optimization claim.
  6. [§II] The paper states that fiber coupling efficiencies are greater than 90% but does not describe the measurement method; add one sentence on how these values were obtained.
  7. [§IV] The conclusion uses the phrase 'over 4 dB' while the measured value is 4.4 dB with no uncertainty; please phrase accordingly.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central squeezing claims are experimental measurements, and the loss model is explicitly illustrative rather than used to derive the reported result.

full rationale

This paper is an experimental demonstration, so the main quantitative claims are measurements rather than derivations from fitted parameters. The reported -4.4 dB after-fiber squeezing is a measured ratio of the twin-beam intensity-difference noise to a shot-noise calibration performed with the pump blocked and a balanced detector. That calibration is an experimental assumption about what constitutes the shot-noise floor; it is not derived from the squeezing result, and no fitted parameter is renamed as a prediction. The loss model S' = eta*S0 + (1-eta)[epsilon*Nth + (1-epsilon)*Nv] is presented with illustrative numbers ('for a coupling efficiency of 90%, initial squeezing of -7.2 dB, normalized excess thermal noise consistent with that of the initial level of squeezing, and equal fractions of coupled thermal and vacuum noise') and the paper explicitly says the model 'overestimates' the contribution of uncorrelated modes. This model does not feed back into the central measured claim. Self-citations to prior FWM work (e.g., Ref. [31]) and to a mode-coupling model (Ref. [35]) provide context and independent support; they are not invoked as a uniqueness theorem or as an ansatz that forces the result. The skeptical concern about the shot-noise calibration being biased by technical noise is a measurement-validity issue, not a circularity issue, because the calibration is not constructed from the quantity it is used to verify. Overall, the derivation chain is self-contained with respect to the experimental data presented, and no circularity is present.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The central claim rests on standard quantum-optics assumptions about FWM-generated twin beams and on the calibration of the shot-noise reference; no new theoretical entities are introduced.

free parameters (2)
  • Fraction of thermal noise epsilon = 0.5
    Chosen by hand to illustrate the effect of uncorrelated spatial modes in the loss model (Section III). Not fitted; the authors note the model overestimates the contribution.
  • Normalized excess thermal noise Nth = consistent with S0 = -7.2 dB (normalized)
    Assumed to match the initial squeezing level in the loss-model estimate; this is a stated assumption, not a measured fit.
assumptions (3)
  • domain assumption The FWM process in the double-Lambda configuration of 85Rb generates twin beams whose intensity-difference noise is squeezed below the shot-noise limit.
    This is the physical basis of the source, established in prior work [30,31] and used without derivation here. It is not proved in this paper. Location: Section II, paragraph 1.
  • domain assumption The shot-noise limit can be calibrated using a coherent probe beam of the same total power, split equally, with the pump blocked.
    The SNL calibration assumes coherent-state noise at the same power equals the quantum noise reference for the twin beams, and that balanced detection cancels technical noise. Location: Section II, calibration paragraph.
  • domain assumption Fiber-coupling inefficiency can be modeled as a beamsplitter mixing in vacuum and thermal noise.
    The loss model treats imperfect coupling as a beamsplitter with vacuum and thermal noise inputs, following [35]; this is an assumption used to interpret the after-fiber squeezing. Location: Section III, model paragraph.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Compact Fiber-Coupled Narrowband Two-Mode Squeezed Light Source." pith.science (2026). https://pith.science/paper/NQ3ADQX5

@misc{pith2026250703755,
  author       = {Pith},
  title        = {Pith review of: Compact Fiber-Coupled Narrowband Two-Mode Squeezed Light Source},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NQ3ADQX5}},
  note         = {Machine review of arXiv:2507.03755}
}
abstract

Quantum correlated states of light, such as squeezed states, are a fundamental resource for the development of quantum technologies, as they are needed for applications in quantum metrology, quantum computation, and quantum communications. It is thus critical to develop compact, efficient, and robust sources to generate such states. Here we report on a compact, narrowband, fiber-coupled source of two-mode squeezed states of light at 795 nm based on four wave mixing (FWM) in a $^{85}$Rb atomic vapor. The source is designed in a small modular form factor, with two input fiber-coupled beams, the seed and pump beams required for the FWM, and two output fibers, one for each of the modes of the squeezed state. The system is optimized for low pump power (135 mW) to achieve a maximum intensity-difference squeezing of 4.4 dB after the output fibers at an analysis frequency of 1 MHz. The narrowband nature of the source makes it ideal for atomic-based quantum sensing and quantum networking configurations that rely on atomic quantum memories. Such a source paves the way for a versatile and portable platform for applications in quantum information science.

Figures

Figures reproduced from arXiv: 2507.03755 by the authors.

Figure 1
Figure 1. (b) shows the schematic of the experimental setup. Implementation of the FWM requires orthogonally polarized probe and pump beams overlapping at a slight angle at the center of a Rb vapor cell. To implement such a configura￾tion, two independent single mode, polarization maintaining optical fibers are used to deliver the orthogonally polarized probe and pump beams. After the fibers, pump and seed probe beams are mad… view at source ↗
Figure 3
Figure 3. FIG. 3. Picture of system implemented on a 12 in by 12 in bread [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. FIG. 4. Intensity-difference noise measurements normalized to the [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 34 canonical work pages

  1. [1]

    U. L. Andersen, T. Gehring, C. Marquardt, and G. Leuchs, Phys. Scripta 91, 053001 (2016)

    work page 2016

  2. [2]

    B. J. Lawrie, P. D. Lett, A. M. Marino, and R. C. Pooser, ACS Photonics 6, 1307–1318 (2019)

    work page 2019

  3. [3]

    E. S. Polzik, J. Carri, and H. J. Kimble, Phys. Rev. Lett. 68, 3020–3023 (1992)

    work page 1992

  4. [4]

    M. R. Grace, C. N. Gagatsos, Q. Zhuang, and S. Guha, Phys. Rev. Appl. 14, 034065 (2020)

    work page 2020

  5. [5]

    C. Xu, L. Zhang, S. Huang, et al., Photonics Res.7, A14 (2019)

    work page 2019

  6. [6]

    U. L. Andersen, J. S. Neergaard-Nielsen, P. van Loock, and A. Furusawa, Nat. Phys. 11, 713–719 (2015)

    work page 2015

  7. [7]

    Pirandola, B

    S. Pirandola, B. R. Bardhan, T. Gehring, et al., Nat. Photonics 12, 724–733 (2018)

    work page 2018

  8. [8]

    Weedbrook, S

    C. Weedbrook, S. Pirandola, R. García-Patrón, et al., Rev. Mod. Phys. 84, 621 (2012)

    work page 2012

Show all 38 references
  1. [9]

    Abadie, B

    J. Abadie, B. P. Abbott, R. Abbott et al., Nat. Phys. 7, 962–965 (2011)

    2011

  2. [10]

    J. Aasi, J. Abadie, B. P. Abbottet al., Nat. Photonics 7, 613–619 (2013)

    2013

  3. [11]

    N. C. Menicucci, P. van Loock, M. Gu, et al., Phys. Rev. Lett. 97, 110501 (2006)

    2006

  4. [12]

    Lloyd and S

    S. Lloyd and S. L. Braunstein, Phys. Rev. Lett.82, 1784 (1999)

    1999

  5. [13]

    S. L. Braunstein and H. J. Kimble, Phys. Rev. Lett. 80, 869 (1998)

    1998

  6. [14]

    Grosshans and P

    F. Grosshans and P. Grangier, Phys. Rev. Lett. 88, 057902 (2002)

    2002

  7. [15]

    V . C. Usenko, A. Acín, R. Alléaume, et al., arXiv:2501.12801 (2025)

    2025 arXiv

  8. [16]

    Arnbak, C

    J. Arnbak, C. S. Jacobsen, R. B. Andrade, et al., Opt. Express 27, 37877 (2019)

    2019

  9. [17]

    Kashiwazaki, N

    T. Kashiwazaki, N. Takanashi, T. Yamashima, et al., APL Pho- tonics 5, 036104 (2020)

    2020

  10. [18]

    Takanashi, T

    N. Takanashi, T. Kashiwazaki, T. Kazama,et al., IEEE J. Quan- tum Electron. 56, 1 (2020)

    2020

  11. [19]

    H. Liu, M. L. Iu, N. Hamdash, and A. S. Helmy, arXiv:2406.19991 (2024)

    2024

  12. [20]

    Appel, E

    J. Appel, E. Figueroa, D. Korystov, et al., Phys. Rev. Lett. 100, 093602 (2008)

    2008

  13. [21]

    Jensen, W

    K. Jensen, W. Wasilewski, H. Krauter, et al., Nat. Phys. 7, 13 (2010)

    2010

  14. [22]

    Kitching, S

    J. Kitching, S. Knappe, and E. A. Donley, IEEE Sensors J. 11, 1749 (2011)

    2011

  15. [23]

    Ye and P

    J. Ye and P. Zoller, Phys. Rev. Lett.132, 190001 (2024). 5

    2024

  16. [24]

    Colombo, E

    S. Colombo, E. Pedrozo-Peñafiel, and V . Vuleti ´c, Appl. Phys. Lett. 121, 210502 (2022)

    2022

  17. [25]

    Schulte, C

    M. Schulte, C. Lisdat, P. O. Schmidt, et al., Nat. Commun. 11, 5955 (2020)

    2020

  18. [26]

    Z. Qin, J. Jing, J. Zhou, et al., Opt. Lett. 37, 3141 (2012)

    2012

  19. [27]

    Z. Qin, L. Cao, H. Wang, et al., Phys. Rev. Lett. 113, 023602 (2014)

    2014

  20. [28]

    R. Ma, W. Liu, Z. Qin, X. Jia, and J. Gao, Phys. Rev. A 96, 043843 (2017)

    2017

  21. [29]

    G. Sim, H. Kim, and H. S. Moon, Sci. Rep. 15, 7727 (2025)

    2025

  22. [30]

    C. F. McCormick, A. M. Marino, V . Boyer, and P. D. Lett, Phys. Rev. A 78, 043816 (2008)

    2008

  23. [31]

    Dowran, A

    M. Dowran, A. Kumar, B. J. Lawrie, et al. , Optica 5, 628 (2018)

    2018

  24. [32]

    Boyer, A

    V . Boyer, A. M. Marino, R. C. Pooser, and P. D. Lett, Science 321, 544 (2008)

    2008

  25. [33]

    Corzo, A

    N. Corzo, A. M. Marino, K. M. Jones, and P. D. Lett, Opt. Express 19, 21358 (2011)

    2011

  26. [34]

    M. W. Holtfrerich and A. M. Marino, Phys. Rev. A 93, 063821 (2016)

    2016

  27. [35]

    Gupta, R

    P. Gupta, R. W. Speirs, K. M. Jones, and P. D. Lett, Opt. Express 28, 652 (2020)

    2020

  28. [36]

    Kumar, G

    A. Kumar, G. Nirala, and A. M. Marino, Quantum Sci. Technol. 6, 045016 (2021)

    2021

  29. [37]

    Kumar, H

    A. Kumar, H. Nunley, and A. M. Marino, Phys. Rev. A 98, 043853 (2018)

    2018

  30. [38]

    Nirala, S

    G. Nirala, S. T. Pradyumna, A. Kumar, and A. M. Marino, Sci. Adv. 9, eadf9161 (2023)

    2023

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.