Pith. sign in

REVIEW 3 major objections 6 minor 23 references

Cavity-enhanced continuous-wave microscopy using unstabilized cavities

T0 review · 3 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read A self-imaging 4f cavity enhances continuous-wave microscopy contrast and signal-to-noise even when the cavity length drifts, and yields a dark-field mode based on optical path length.

desk verdict First CW self-imaging cavity-microscope demo with a real new dark-field mode, but the 'unstabilized' robustness claim is only proven for piston noise and the baseline comparison needs strengthening. read the letter →

arxiv 2412.16909 v1 pith:3JY3MNSR submitted 2024-12-22 physics.optics physics.atom-ph

Oliver Lueghamer , Stefan Nimmrichter , Clara Conrad-Billroth , Thomas Juffmann , Maximilian Prüfer This is my paper · ORCID
classification physics.opticsphysics.atom-ph
keywords cavity-enhancedmicroscopyself-imagingcavitycontinuous-waveunstabilizeddark-fieldopticalpathlengthdispersiveimagingsignal-to-noiseenhancement
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 aims to show that the gains of cavity-enhanced microscopy—more signal per probe photon and better signal-to-noise at fixed sample damage—survive even when the cavity cannot be actively stabilized. It uses a self-imaging 4f cavity, a resonator whose lenses make every ray retrace its path after one round trip, so all transverse image modes transmit together. The experiments show enhanced contrast on holes in a 10 nm silicon nitride membrane and on human cheek cells, and the theory predicts that averaging over one free spectral range keeps half the resonant contrast gain. If correct, the result makes continuous-wave cavity-enhanced microscopy practical without locking, including dispersive imaging of ultracold atoms, and adds a dark-field mode that selects scattered light by optical path length.

What carries the argument

The central object is the self-imaging 4f cavity: two mirrors with a pair of intracavity lenses arranged so that any ray retraces its path after one full round trip, making all transverse modes degenerate and allowing an image to circulate while the field builds up by roughly $1/(1-\sqrt{R_1R_2})$. The argument for unstabilized operation rests on a common-mode fluctuation assumption—every transverse mode and the undiffracted reference see the same cavity-length jitter—modeled as a uniform average over one free spectral range at fixed $4kf$. A Fabry-Perot toy model with a phase-locked reference passing the same unstable cavity shows that half the resonant phase sensitivity survives the average, and the reflected field carries the missing half; the multimode imaging theory generalizes this to spatially varying phase shifts.

What would settle it

Drive one cavity mirror with a tilt that grows beyond the paraxial common-mode limit while integrating images over one free spectral range; if contrast and SNR drop below the predicted factor-two retention relative to resonance, the common-mode premise is falsified.

Watch

Extended reading notes

Core claim

The paper's central claim is that a degenerate self-imaging cavity operating in continuous wave enhances the weak phase contrast of an optically thin sample, and that the enhancement is not lost when the cavity length is unstable. In the linear-response model, the transmitted image carries a phase term multiplied by the cavity build-up factor $1/(1-\sqrt{R_1R_2})$, giving optimum contrast $C_{\max}\approx \frac{2\sqrt{R_2}}{T_1+T_2}|(\chi-\chi_0)\sin 4kf|$ for a weak phase object. If the cavity length is averaged uniformly over one free spectral range, the contrast becomes $C_{\rm avg}\approx \frac{\sqrt{R_2}}{T_1+T_2}|(\chi-\chi_0)\sin 4kf|$, a factor two lower than the resonant value, provided all transverse modes and the reference light undergo the same length fluctuations. The paper reports experimental confirmation on hole structures in a 10 nm Si$_3$N$_4$ membrane and on epithelial cells, including a dark-field regime in which the cavity length selects scattered light by optical path length. It also argues that SNR and SNR at fixed damage are enhanced in the same way, and that monitoring the reflected output recovers the information lost in transmission.

Load-bearing premise

The result rests on all transverse modes and the reference unscattered light experiencing the same cavity-length fluctuations, so their relative phase is preserved while the cavity length is uniformly sampled over one free spectral range.

Editorial extensions

If this is right

  • Cavity-enhanced microscopy can be operated without active frequency locking, because a time average over one free spectral range keeps roughly half the resonant contrast gain.
  • Thick, optically thin phase samples can be dark-field imaged with forward-scattered light, with the cavity length selecting optical path length rather than scattering angle.
  • Dispersive imaging of ultracold atoms should gain signal-to-noise at fixed probe-induced damage, since coherent forward scattering is amplified while incoherent scattering is not.
  • The factor-two information lost in transmission through an unstabilized cavity is recoverable by monitoring the reflected field, so the unstabilized scheme can in principle match resonant sensitivity.

Reading between the lines

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

  • Beyond the paper, the resonance-shift dark-field contrast could be inverted into a quantitative optical-thickness map by recording the cavity-length position of peak brightness per pixel; the paper demonstrates the shift but stops short of extracting thickness maps.
  • Beyond the paper, the common-mode averaging argument should transfer to other fully degenerate cavities, such as confocal resonators, as long as probe and reference share the same fluctuating optical path.
  • Beyond the paper, a stress test at higher finesse or with aberrating samples would locate where the paraxial common-mode assumption breaks, a boundary the paper does not fix.
  • Beyond the paper, combining the transmitted and reflected ports of the unstabilized cavity in one experiment could restore the full resonant sensitivity rather than half, since the toy model shows the missing factor is in reflection.
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 / 6 minor

Summary. The manuscript reports continuous-wave cavity-enhanced microscopy using a self-imaging 4f cavity. The authors demonstrate image transmission through the degenerate cavity, contrast enhancement for holes in a 10 nm Si3N4 membrane, and a detuning-based dark-field modality for cheek cells. The theoretical part includes a multimode linear-optics derivation (Supplement) of the bright-field contrast on and off resonance, and a single-mode Fabry-Perot toy model (Appendix) for an unstabilized cavity. The central claim is that signal, SNR, and SNR per damage are enhanced even when the cavity length is not stabilized, provided the reference and signal fields share the same cavity fluctuations.

Significance. If the unstabilized-cavity claim is robust, this would considerably extend the applicability of cavity-enhanced microscopy to settings where active stabilization is impractical, notably dispersive imaging of ultracold atoms. The paper provides a detailed and self-contained linear-optics model, explicit analytic contrast formulas, and an instructive toy model. The experimental results on a fabricated test sample and on biological cells are novel, and the path-length dark-field modality is an interesting new contrast mechanism. However, the robustness claim is currently established only for common-mode (piston) fluctuations, and the experimental contrast comparison lacks a phase-contrast baseline.

major comments (3)
  1. [Supplement, Eqs. (38)-(39); Appendix] The unstabilized-cavity theory models only a global piston displacement δz1 of M1 and averages uniformly over kL at fixed 4kf. The central claim that enhancement persists when the cavity cannot be stabilized assumes that all transverse modes and the reference light experience identical path-length fluctuations. The justification in the Appendix, that common noise on the cavity path length holds for small path length changes on the order of λ and in the paraxial limit, does not cover differential fluctuations: a mirror tilt δz1(r⊥)=δz0+θx produces a position-dependent round-trip phase, so after FSR averaging the intensity difference between sample and background at transverse separation Δx can cancel when 2kθΔx ≳ π (θ ≳ λ/(4Δx) ≈ 20 μrad for Δx = 10 μm). Such tilts are plausible in an unstabilized cavity, and the experimental unstabilized demonstration uses a controlled piezo scan, which is a piston displacement and does not test this failure mode. Therefore the claim of robust enhancement for unstabilized cavities is not established for realistic differential noise.
  2. [§Demonstration of cavity-enhanced microscopy and Fig. 2] The measured cavity-enhanced contrast (≈10%) is compared only to the single-pass bright-field contrast (1.5%), while the paper itself notes that defocus phase contrast can be as large as 24%. For a phase object, the appropriate single-pass baseline is phase-contrast or defocused imaging; the reported 10% is below the 24% upper bound, and no direct single-pass phase-contrast measurement is presented. Consequently, the experimental claim of contrast enhancement over the best single-pass phase-contrast modality is not supported by the data as reported. The authors should either measure the single-pass phase-contrast contrast in the same setup, or explicitly restrict the comparison to bright-field imaging.
  3. [Abstract; Supplement, Eq. (39)] The abstract and conclusion claim enhanced SNR and SNR per damage for unstabilized cavities, but the multimode imaging theory in the Supplement derives only the contrast (Eqs. (40)-(41)). The SNR/SNRD enhancement for the imaging case is asserted by analogy to the single-mode toy model in the Appendix, without a shot-noise calculation for the multimode detection. Since the relationship between contrast gain and SNR depends on the detected background level, which varies with cavity detuning, the authors should provide the explicit SNRD derivation for the imaging scenario, or state clearly that only the contrast enhancement is proven in the multimode case.
minor comments (6)
  1. [Fig. 3 caption] The phrase 'the incidence light is resonant' should read 'the incident light is resonant'; the caption would also benefit from specifying what is integrated over time (e.g., the camera exposure).
  2. [Supplement, resolution estimation] The reported σ≈3 μm is used as the resolution, but the convention relating σ to a standard resolution metric (e.g., FWHM or Rayleigh criterion) is not stated.
  3. [§Demonstration of cavity-enhanced microscopy] The defocused single-pass image in Fig. 2b is not characterized by a defocus distance, which makes the phase-contrast comparison difficult to reproduce.
  4. [Eq. (4)] The definition of the effective number of round trips N would be clearer if the intensities were defined as measured at the same plane; the current expression mixes intracavity and output intensities without specifying the beam area.
  5. [General] The paper alternates between 'defocused' and 'defocussed'; please standardize the spelling.
  6. [Title and abstract] The use of 'unstabilized cavities' is somewhat overbroad because the experimental unstabilized demonstration is a controlled piezo scan over one FSR; consider a more precise phrase such as 'cavities without active stabilization' to avoid implying free-running random drift.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the FSR-averaged cavity enhancement is derived explicitly from the paper's own linear-optics model, and the self-cited SNRD scaling is independently re-derived in the Appendix.

full rationale

All load-bearing theoretical results are derived in-manuscript from textbook Fabry-Perot input-output relations and Fourier optics. The resonant contrast Cmax (Supplement Eq. 40) and the FSR-averaged contrast Cavg (Supplement Eq. 41) follow from explicit uniform averaging over the cavity length kL at fixed 4kf, with only mirror reflectivities R1, R2 and sample phase shifts chi, chi0 as inputs; no target contrast or SNR is inserted as a fitted quantity. The main-text Eq. (3) is not a separate claim but the same averaged result quoted for the experiment. The SNRD improvement, attributed in passing to the self-cited reference [11], is independently re-derived in the Appendix toy model (Eqs. 9-13 and surrounding text), so the self-citation is not load-bearing. The paper explicitly states the central validity condition: 'This enhanced performance of an unstabilized cavity relies on all modes within the cavity undergoing the same fluctuations,' and the Appendix adds that common noise is 'justified for small path length changes on the order of lambda and in the paraxial limit.' This is an openly stated physical limitation rather than a circular definition. Experimental contrast and SNR are compared against separately acquired single-pass images, and the unknown phase 4kf is an unmeasured parameter, not a fitted value used to force the claimed enhancement. No derivation step reduces by construction to its own input.

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

The central derivation is self-contained linear optics with standard assumptions. The main costs are the unmeasured geometric phase 4kf, the common-mode noise assumption for unstabilized operation, and the weak-sample, paraxial, and shot-noise assumptions. No new physical entities are introduced.

free parameters (2)
  • Intracavity geometric phase 4kf = unknown fixed value
    The contrast formulas Cmax and Cavg in the Supplement depend on |sin 4kf|. The authors state that 4kf takes an unknown fixed value because the lens focal length is not known at wavelength level; no measurement or fit of this phase is provided, so the predicted contrast magnitude is not quantitatively benchmarked.
  • Gaussian blur standard deviation sigma = approximately 3 micrometers
    In the Supplement, sigma is obtained by fitting a box-convolved Gaussian to a single-pixel line profile of a 10 micrometer hole to estimate resolution. This fit characterizes the imaging system but is not part of the central contrast claim.
assumptions (5)
  • standard math Paraxial scalar wave optics and thin-lens Fourier transform property for ideal lenses.
    Used in the Supplement to propagate fields through the 4f cavity; standard Fourier optics.
  • domain assumption Weak, thin, lossless sample: sample coefficients expanded to first order in chi and all second-order reflection terms neglected.
    Central for the linear-response contrast formulas in the Supplement; fails for thick or strongly scattering samples.
  • domain assumption Symmetric illumination Ein(r_perp)=Ein(-r_perp) and sample features coarse compared to the wavelength.
    Needed to derive the closed-form output field in the Supplement; not tested experimentally in the paper.
  • domain assumption All transverse modes and the reference beam experience identical cavity-length fluctuations (common-mode noise), with uniform averaging over one free spectral range as the model of an unstabilized cavity.
    This is the key premise for the unstabilized-cavity enhancement claim; the paper provides only a paraxial justification and no experimental verification of common-mode correlations or uniform averaging.
  • domain assumption Shot-noise-limited balanced homodyne detection and a lossless sample in the toy model.
    Assumed in the Appendix to convert signal averages into SNR and SNRD claims; real detectors and samples will degrade these.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Cavity-enhanced continuous-wave microscopy using unstabilized cavities." pith.science (2026). https://pith.science/paper/3JY3MNSR

@misc{pith2026241216909,
  author       = {Pith},
  title        = {Pith review of: Cavity-enhanced continuous-wave microscopy using unstabilized cavities},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3JY3MNSR}},
  note         = {Machine review of arXiv:2412.16909}
}
read the original abstract

Microscopy gives access to spatially resolved dynamics in different systems, from biological cells to cold atoms. A big challenge is maximizing the information per used probe particle to limit the damage to the probed system. We present a cavity-enhanced continuous-wave microscopy approach that provides enhanced signal-to-noise ratios at fixed damage. Employing a self-imaging 4f cavity, we show contrast enhancement for controlled test samples as well as biological samples. For thick samples, the imaging cavity leads to a new form of dark-field microscopy, where the separation of scattered and unscattered light is based on optical path length. We theoretically show that enhanced signal, signal-to-noise, and signal-to-noise per damage are also retrieved when the cavity cannot be stabilized. Our results provide an approach to cavity-enhanced microscopy with unstabilized cavities and might be used to enhance the performance of dispersive imaging of ultracold atoms.

Figures

Figures reproduced from arXiv: 2412.16909 by the authors.

Figure 1
Figure 1. Cavity setup and characterization. a) Ray tracing picture showing the self-imaging properties of the 4f cavity. The sample is illuminated with a collimated beam (red); the diffracted light (green) is imaged onto the sample after one full roundtrip. b) Cavity length scan monitored by the output voltage (normalized) of a photodiode placed at the output of the cavity. The separation between the two resonance peaks yiel… view at source ↗
Figure 2
Figure 2. Cavity-enhanced microscopy of holes in a Si3N4 membrane. a) Scanning electron microscopy image of holes cut into a Si3N4 membrane. The red and blue circles indicate holes with a diameter of 10 µm, and the yellow circle marks a hole with a diameter of 6 µm. The inset shows a close-up of the hole we use as our main target. b) Defocused single-pass image of the membrane; the inset shows the region of the image with the… view at source ↗
Figure 3
Figure 3. Cavity dark-field imaging. a) Bright Field image of cheek cell in single pass. b) Calculated SNR as a function of mirror displacement represented by the driving voltage of the piezo. The blue curve shows the cavity output detected by the photodiode. c) Yellow: Integrated images over time, while the incidence light is resonant and the re￾fracted light is out of resonance. d) Dark Blue: Integrated images over time whi… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Set-up for toy model calculation. a) The out￾put of a Fabry-P´erot cavity setup is homodyned with a local oscillator. b) Additionally, the local oscillator is transmit￾ted through the same unstable cavity picking up the phase shift of the detuned cavity. For both cases…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

23 extracted references · 22 canonical work pages

  1. [1]

    Hawkes, P. W. & Spence, J. C. Science of microscopy, vol. 1 (Springer, 2007)

    work page 2007

  2. [2]

    Morris, J. D. & Payne, C. K. Microscopy and cell biol- ogy: new methods and new questions. Annual review of physical chemistry 70, 199–218 (2019)

    work page 2019

  3. [3]

    Ketterle, W., Durfee, D. S. & Stamper-Kurn, D. M. Mak- ing, probing and understanding Bose-Einstein conden- sates. arXiv:cond-mat/9904034 (1999)

    arXiv 1999

  4. [4]

    B., Frankort, T

    Juffmann, T., Klopfer, B. B., Frankort, T. L., Haslinger, P. & Kasevich, M. A. Multi-pass microscopy. Nature Communications 7, 12858 (2016)

    work page 2016

  5. [5]

    Arnaud, J. A. Degenerate Optical Cavities. Applied Op- tics 8, 189 (1969)

    work page 1969

  6. [6]

    Klopfer, B. B. B., Juffmann, T. & Kasevich, M. M. A. Iterative creation and sensing of twisted light. Optics Letters 41, 5744 (2016)

    work page 2016

  7. [7]

    Phase-shift amplification for precision measure- ments without nonclassical states

    Luis, A. Phase-shift amplification for precision measure- ments without nonclassical states. Physical Review A65, 025802 (2002)

    work page 2002

  8. [8]

    & Maccone, L

    Giovannetti, V., Lloyd, S. & Maccone, L. Quantum metrology. Physical Review Letters96, 10401 (2006)

    work page 2006

Show all 23 references
  1. [9]

    L., Berry, D

    Higgins, B. L., Berry, D. W., Bartlett, S. D., Wiseman, H. M. & Pryde, G. J. Entanglement-free Heisenberg- limited phase estimation. Nature 450, 393–396 (2007)

    2007

  2. [10]

    A., Israel, Y., Bowman, A

    Koppell, S. A., Israel, Y., Bowman, A. J., Klopfer, B. B. & Kasevich, M. A. Transmission electron microscopy at the quantum limit. Applied Physics Letters120, 190502 (2022)

    2022

  3. [11]

    B., Kasevich, M

    Nimmrichter, S., Chen, C.-F., Klopfer, B. B., Kasevich, M. A. & Juffmann, T. Full-field cavity enhanced mi- croscopy techniques. Journal of Physics: Photonics 1, 015007 (2018)

    2018

  4. [12]

    & Fabre, C

    Gigan, S., Lopez, L., Treps, N., Ma ˆ ıtre, A. & Fabre, C. Image transmission through a stable paraxial cavity. Physical Review A72, 023804 (2005)

    2005

  5. [13]

    L., Klopfer, B

    Israel, Y., Reynolds, J. L., Klopfer, B. B. & Kasevich, M. A. Continuous wave multi-pass imaging flow cytom- etry. Optica 10, 491 (2023)

    2023

  6. [14]

    Cao, H., Chriki, R., Bittner, S., Friesem, A. A. & David- son, N. Complex lasers with controllable coherence. Na- ture Reviews Physics1, 156–168 (2019)

    2019

  7. [15]

    Slobodkin, Y. et al. Massively degenerate coherent per- fect absorber for arbitrary wavefronts. Science 377, 995– 998 (2022)

    2022

  8. [16]

    Kolobov, M. I. Quantum Imaging (Springer New York, 2007)

    2007

  9. [17]

    & Juffmann, T

    Bouchet, D., Dong, J., Maestre, D. & Juffmann, T. Fun- damental bounds on the precision of classical phase mi- croscopes. Phys. Rev. Appl.15, 024047 (2021)

    2021

  10. [18]

    Zuo, C. et al. Transport of intensity equation: a tutorial. Optics and Lasers in Engineering135, 106187 (2020)

    2020

  11. [19]

    Mazzinghi, C. et al. Cavity-enhanced polarization rota- tion measurements for low-disturbance probing of atoms. Optics Express 29, 40854 (2021)

    2021

  12. [20]

    Andrews, M. R. et al. Direct, Nondestructive Observa- tion of a Bose Condensate. Science 273, 84–87 (1996)

    1996

  13. [21]

    & Spielman, I

    Altunta¸ s, E. & Spielman, I. B. Quantum back-action lim- its in dispersively measured Bose-Einstein condensates. Communications Physics 6, 66 (2023)

    2023

  14. [22]

    M., Marsh, B

    Kroeze, R. M., Marsh, B. P., Lin, K.-Y., Keeling, J. & Lev, B. L. High cooperativity using a confocal-cavity– qed microscope. PRX Quantum 4, 020326 (2023)

    2023

  15. [23]

    & Weber, H

    Hodgson, N. & Weber, H. Optical Resonators: Funda- mentals, Advanced Concepts, Applications (Springer Se- ries in Optical Sciences)(Springer London, 2005). 6 APPENDIX Cavity parameters.— The cavity length is defined by the 4 f requirement to be ∼ 30 cm. Accounting for the opti...

    2005

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.