Searches for ultralight vector and axion dark matter with KAGRA

Atsushi Nishizawa; Hinata Takidera; Hiroki Fujimoto; Ippei Obata; Jun'ya Kume; Kentaro Komori; Koji Nagano; Soichiro Morisaki; Takumi Fujimori; Tomohiro Fujita

REVIEW 2 major objections 3 minor 32 references

Searches for ultralight vector and axion dark matter with KAGRA

T0 review · 2 major / 3 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read This paper reports that KAGRA, a laser interferometric gravitational wave detector in Japan, has used its first joint observing run with GEO600 to set upper limits on the B-L gauge coupling at the 10^-19 level for ultralight vector dark…

desk verdict A clean review/status report whose one genuinely new claim—that KAGRA is hardware-ready for an axion search—needs commissioning evidence before it can carry the paper's message. read the letter →

arxiv 2501.08930 v2 pith:JLTFCI2A submitted 2025-01-15 hep-ph astro-ph.IMphysics.optics

Yuta Michimura , Takumi Fujimori , Hiroki Fujimoto , Tomohiro Fujita , Kentaro Komori , Jun'ya Kume , Yusuke Manita , Soichiro Morisaki

show 5 more authors

Koji Nagano Atsushi Nishizawa Ippei Obata Yuka Oshima Hinata Takidera

This is my paper · ORCID

classification hep-phastro-ph.IMphysics.optics PACS 95.35.+d04.80.Nn14.80.Va

keywords ultralightdarkmattervectoraxionKAGRAgravitationalwavedetectorB-Lgaugebosonaxion-photoncouplingpolarizationrotation

topics Dark Matter

open problems Dark Matter

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

This paper reports on two concrete steps toward using the KAGRA gravitational wave detector as a dark matter instrument. An analysis of data from the 2020 O3GK run with GEO600 set upper limits on the B-L gauge coupling of ultralight vector dark matter at the $10^{-19}$ level in the mass range $10^{-13}\,\mathrm{eV} \lesssim m_A \lesssim 10^{-12}\,\mathrm{eV}$. The team also installed polarization optics at the arm cavity transmission ports in 2021, making KAGRA ready to collect its first axion dark matter data during the O4c observing run in 2025. The projected sensitivities in both channels could surpass existing equivalence-principle and astrophysical bounds if the detector reaches its designed noise performance.

What carries the argument

The vector search relies on KAGRA being the only gravitational wave detector with test masses (sapphire) and auxiliary mirrors (fused silica) of different materials, so an oscillating force from a B-L gauge boson field produces a differential signal in the MICH, PRCL, and SRCL channels instead of cancelling as it does in the common DARM channel. The axion search uses the axion-photon coupling, which induces a tiny oscillating circular birefringence; the ADAM-GD scheme detects the resulting polarization rotation, and reading out the beam transmitted through the arm cavity rather than reflected makes the rotation add coherently over the odd number of cavity transits.

What would settle it

Measure the noise spectral density in the MICH, PRCL, and SRCL channels and at the transmission ports during the O4c run; if the integrated noise exceeds the design curves used to generate Fig. 1 by more than an order of magnitude, the claimed readiness and projected sensitivities for both searches are contradicted.

Watch

Extended reading notes

Core claim

Using the auxiliary length channels (MICH, PRCL, SRCL) that sense differential acceleration between KAGRA's sapphire test masses and fused silica auxiliary mirrors, the O3GK analysis produced the first KAGRA upper limit on B-L vector dark matter: a $10^{-19}$ level bound on the coupling $\epsilon_{B-L}$ over $10^{-13}\,\mathrm{eV} \lesssim m_A \lesssim 10^{-12}\,\mathrm{eV}$. For axions, the paper shows that measuring oscillating polarization rotation of the beams transmitted through the 3-km arm cavities avoids the cancellation that affects reflected-beam readouts at the anti-symmetric port, and reports that the required polarization optics are now installed and ready for the 2025 run.

Load-bearing premise

The projected dark matter sensitivities assume KAGRA will reach its designed noise levels in the auxiliary length channels and at the arm-cavity transmission ports; if those channels remain several orders of magnitude above design during the 2025 run, the projected reach will not be achieved.

Editorial extensions

If this is right

If KAGRA reaches its designed sensitivity, the vector dark matter limits will surpass the best equivalence-principle tests in the mass range near $10^{-13}$ to $10^{-12}\,\mathrm{eV}$.
The upcoming O4c run in 2025 is expected to deliver the first axion dark matter constraints from a gravitational wave detector, complementing tabletop experiments such as DANCE.
The demonstration that auxiliary length channels are scientifically useful will motivate dedicated noise reduction in these channels, turning them into standard dark matter observables.
If no signal is found, the new bounds will tighten constraints on ultralight vector and axion dark matter parameter space in mass windows not fully covered by existing experiments.

Reading between the lines

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

A null axion result in the O4c run could still be significant if the achieved limits improve on astrophysical bounds from SN1987A or M87 in the low-mass region, though the paper does not quantify this.
The mixed-material strategy could be adopted by future detectors or retrofits, potentially extending the technique to other observatories and mass ranges.
The polarization-rotation hardware may also be sensitive to axion-like particles over a broader mass range if cavity parameters are optimized, which the current configuration does not fully explore.

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.

Desk Editor's Note

read the letter

Quick take: this is a review/status report, not a new measurement. The only genuinely new claim is that KAGRA installed polarization optics at the arm-cavity transmission ports in 2021 and plans to take axion data in O4c in 2025. Everything else—the O3GK vector limits, the ADAM-GD sensitivity projections, the axion polarization equations—is already in the cited papers (refs. 6, 7, 13, 16, 31). The authors are transparent about this; the abstract and intro frame it as a review.

What the paper does well: it gives a clean, accurate summary of the vector force search using auxiliary length channels (MICH, PRCL, SRCL), explains why KAGRA's mixed-material mirrors (sapphire vs. fused silica) give a unique advantage over LIGO/Virgo, and lays out the ADAM-GD axion scheme clearly. The sensitivity curves in Fig. 1 match the published versions. The citation pattern looks appropriate; no sign of overclaiming or hiding prior work.

The soft spot is exactly the one the stress test flags: the readiness statement in Sec. 3. "We installed polarization optics... in 2021 and are now ready" carries no supporting evidence—no measured extinction ratio, no transmitted power, no noise spectrum, no reference to a KAGRA technical note. For a review paper that's a smallish flaw; for a paper whose central message is that KAGRA is hardware-ready for the first axion search, it's load-bearing. If the optics are in place and working, the paper is a useful status update. If not, it's just a proposal dressed as a report. This is easily fixed by adding a reference to commissioning documentation or a brief description of the measured performance.

The other caveat is the designed-sensitivity assumption. The text says "once the detector reaches its designed sensitivity"—that's a projection, not a result. The paper doesn't hide this, but a reader skimming Fig. 1 could take the projected reach as actual sensitivity.

Bottom line: it's a solid review with one unsupported empirical claim. It deserves peer review in the sense that a referee can quickly verify the references and ask for the commissioning evidence. I'd bring it to a reading group if you're interested in GW-detector dark matter searches, but I wouldn't cite it for anything beyond the status update. For the record, the stress-test concern holds up; that's the one thing I'd push on.

Referee Report

2 major / 3 minor

Summary. This paper reviews the status of ultralight vector and axion dark matter searches using the KAGRA gravitational wave detector. It summarizes the published O3GK upper limit on the B-L gauge coupling from KAGRA's 2020 data, and it reports the installation of polarization optics at the arm cavity transmission ports in 2021, stating that KAGRA is ready for the first axion dark matter search in the planned O4c run of 2025. The theoretical background, sensitivity projections, and limits from other experiments are taken from earlier work by the authors and from published LIGO-Virgo-KAGRA results.

Significance. If the hardware-readiness claim is correct, the paper documents KAGRA as the first gravitational-wave detector equipped to search for axion dark matter via polarization rotation, complementing its already-published vector dark matter limit. The paper is a concise and well-referenced status report rather than a technical derivation; its novelty rests almost entirely on the Section 3 installation claim. The ability to verify that claim will determine whether the paper carries new information beyond a review of existing results.

major comments (2)

[Section 3] The sole new empirical assertion in the manuscript is the sentence "we installed polarization optics at the transmission ports of the two arm cavities in 2021 and are now ready to collect the first axion data during the upcoming O4c observing run in 2025." No supporting evidence is provided: there is no measured transmitted power at the end test mass ports, no polarization extinction ratio, no noise spectrum, and no citation to a KAGRA technical document or commissioning note. Because this statement underlies the paper's central claim that KAGRA is hardware-ready for a first axion search, it is load-bearing. The authors should either cite a verifiable technical reference or present quantitative commissioning data (e.g., achieved extinction ratio or transmitted power) to support the readiness claim.
[Section 2] The text states "Once the detector reaches its designed sensitivity, it will surpass equivalence principle test limits" and Fig. 1 shows projected sensitivities. The paper does not explicitly state which auxiliary length channels (MICH, PRCL, SRCL) are assumed to reach design noise in the projections, nor does it quantify the margin by which they would surpass limits. Since the projected reach depends directly on the noise in those channels, a sentence clarifying that the curves assume the design sensitivity in each channel and referencing the noise model would make the basis of the projection transparent.

minor comments (3)

[Figure 1 caption] In the left panel caption, "fifth-fore searches" should read "fifth-force searches."
[Section 2, around Eq. (1)] The coupling is introduced as ϵ_B-L, but later in the same paragraph the quantity g_B-L/M appears; the relation between ϵ_B-L and g_B-L should be defined explicitly to avoid notational confusion.
[References] Reference 27 is formatted inconsistently: "C. P. Salemi C P et al." should be "C. P. Salemi et al."

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the vector limit and axion projections rest on published, externally benchmarked analyses, and the new readiness claim is an unsupported but non-circular hardware assertion.

full rationale

The paper does not present any fitted parameter disguised as a prediction or any result that reduces by construction to its inputs. The O3GK vector dark matter limit is reported as a published joint-analysis result (ref. 16) with a dedicated pipeline (ref. 31) and is compared against external limits from Eot-Wash, MICROSCOPE, and LIGO-Virgo O3. The axion sensitivity projections in Fig. 1 are taken from the authors' earlier proposals (refs. 6 and 7), but those proposals are benchmarked against independent experiments (CAST, SHAFT, ABRA, DANCE, and LIGO projections), and the underlying equations (2)-(3) follow from the standard axion-photon interaction; no input is renamed as an output. The only new empirical assertion is the statement in Sec. 3 that polarization optics were installed at the transmission ports in 2021 and that KAGRA is ready for the O4c axion run. This assertion is not backed by measured transmission, extinction, or noise data, so it is a missing-support gap, but it is not circular: it is not derived from any parameter fitted to the same claim, and it does not rely on an unverified theorem. The paper's reliance on the authors' prior papers is transparent self-citation in a review context, and none of those citations is load-bearing in a way that forces the present conclusions. No circular step can be identified with the required quotation and reduction evidence.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

No numbers are fitted in this paper. The sensitivity projections rely on standard assumptions (local dark matter density, virial velocity, KAGRA design sensitivity), all stated in the text, and on the authors' earlier derivations in the cited literature.

free parameters (1)

Local dark matter density rho_DM = 0.4 GeV/cm^3 (assumed)
Used in Eq. (1) to set the field amplitude A0 and therefore the force amplitude; projected limits scale with the square root of this assumed density. The paper states it as an assumption, not a measurement.

assumptions (5)

domain assumption The dark matter in the searched mass range is composed entirely of the candidate ultralight vector or axion field.
The projected limits in Fig. 1 assume 100% abundance of the searched species; a lower local abundance would weaken the projected constraints. Stated in Section 2 as 'We assumed that the field energy density equals the local dark matter density.'
domain assumption Local dark matter velocity v ~ 10^-3.
Used for the coherence time tau = 2*pi/(m_A v^2) in Section 2; a different velocity dispersion changes the signal linewidth and search pipeline.
ad hoc to paper KAGRA will reach designed sensitivity in the relevant auxiliary length channels and at the transmission ports.
The text says 'Once the detector reaches its designed sensitivity, it will surpass equivalence principle test limits,' and the axion curves in Fig. 1 are projections, not measured noise floors.
domain assumption The vector force formula and the ADAM-GD axion detection scheme from Refs. 5, 6, 7, 13, and 14 are correct.
Equations (1)-(3) are quoted from the authors' prior work; the review provides no independent derivation.
domain assumption Sapphire and fused silica have sufficiently different B-L charge-to-mass ratios to produce a measurable differential force.
The text in Section 2 states g_B-L/M = 0.510/m_n for sapphire and 0.501/m_n for fused silica; this difference is the basis for KAGRA's supposed advantage, but no calculation is shown in this paper.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Searches for ultralight vector and axion dark matter with KAGRA." pith.science (2026). https://pith.science/paper/JLTFCI2A

@misc{pith2026250108930,
  author       = {Pith},
  title        = {Pith review of: Searches for ultralight vector and axion dark matter with KAGRA},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JLTFCI2A}},
  note         = {Machine review of arXiv:2501.08930}
}

read the original abstract

We have proposed using laser interferometric gravitational wave detectors to search for ultralight vector and axion dark matter. Vector dark matter can be probed through oscillating forces on suspended mirrors, while axion dark matter can be detected via oscillating polarization rotation of laser beams. This paper reviews these searches with the KAGRA detector in Japan, including the first vector dark matter search with KAGRA's 2020 data and installation of polarization optics for axion dark matter search during the upcoming 2025 observing run.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages

[1]

LIGO-Virgo Collaboration, Phys. Rev. Lett. 116, 061102 (2016)

work page 2016
[2]

S. M. Vermeulen et al. , Nature 600, 424 (2021)

work page 2021
[3]

Morisaki, T

S. Morisaki, T. Suyama, Phys. Rev. D 100, 123512 (2019)

work page 2019
[4]

Fukusumi, S, Morisaki, T, Suyama, Phys

K. Fukusumi, S, Morisaki, T, Suyama, Phys. Rev. D 108, 095054 (2023)

work page 2023
[5]

Obata, T

I. Obata, T. Fujita, Y. Michimura, Phys. Rev. Lett. 121, 161301 (2018)

work page 2018
[6]

Nagano, T

K. Nagano, T. Fujita, Y. Michimura, I. Obata, Phys. Rev. Lett. 123, 111301 (2019)

work page 2019
[7]

Nagano et al

K. Nagano et al. ., Phys. Rev. D 104, 062008 (2021)

work page 2021
[8]

Oshima et al

Y. Oshima et al. , J. Phys. Conf. Ser. 2156, 012042 (2021)

work page 2021

Show all 32 references

[9]

Fujimoto et al

H. Fujimoto et al. , J. Phys. Conf. Ser. 2156, 012182 (2021)

2021
[10]

Oshima et al

Y. Oshima et al. , Phys. Rev. D 108, 072005 (2023)

2023
[11]

Heinze et al

J. Heinze et al. , Phys. Rev. Lett. 131, 191002 (2024)

2024
[12]

Pandey, E

S. Pandey, E. D. Hall, M. Evans, Phys. Rev. Lett. 133, 111003 (2024)

2024
[13]

Michimura et al

Y. Michimura et al. , Phys. Rev. D 102, 102001 (2020)

2020
[14]

Morisaki et al

S. Morisaki et al. , Phys. Rev. D 103, L051702 (2021)

2021
[15]

LIGO-Virgo-KAGRA Collaboration, Phys. Rev. D 105, 063030 (2022)

2022
[16]

LIGO-Virgo-KAGRA Collaboration, Phys. Rev. D 110, 042001 (2024)

2024
[17]

A. L. Miller and L. Mendes, Phys. Rev. D 107, 063015 (2023)

2023
[18]

Frerick et al

J. Frerick et al. , Phys. Lett. B 848, 138328 (2024)

2024
[19]

Manita, K

Y. Manita, K. Aoki, T. Fujita, S. Mukohyama, Phys. Rev. D 107, 104007 (2023)

2023
[20]

Manita et al

Y. Manita et al. , Phys. Rev. D 109, 095012 (2024)

2024
[21]

Michimura et al

Y. Michimura et al. , J. Phys. Conf. Ser. 2156, 012071 (2021)

2021
[22]

Schlamminger et al

S. Schlamminger et al. . Phys. Rev. Lett. 100, 041101 (2008)

2008
[23]

A. T. Wagner et al. , Class. Quantum Grav. 29, 184002 (2012)

2012
[24]

Berg´ eet al

J. Berg´ eet al. , Phys. Rev. Lett. 120, 141101 (2018)

2018
[25]

CAST Collaboration, Phys. Rev. Lett. 133, 221005 (2024)

2024
[26]

V. A. Gramolin et al. , Nat. Phys. 17, 79 (2021)

2021
[27]

C. P. Salemi C P et al. , Phys. Rev. Lett. 127, 081801 (2021)

2021
[28]

Payez et al

A. Payez et al. , J. Cosmol. Astropart. Phys. 02, 006 (2015)

2015
[29]

M. C. D. Marsh et al. , J. Cosmol. Astropart. Phys. 12, 036 (2017)

2017
[30]

Sisk-Reyn´ eset al

J. Sisk-Reyn´ eset al. , Mon. Notices Royal Astron. Soc. 510, 1264 (2022)

2022
[31]

Nakatsuka et al

H. Nakatsuka et al. , Phys. Rev. D 108, 092010 (2023)

2023
[32]

Obata, J

I. Obata, J. Cosmol. Astropart. Phys. 09, 062 (2022)

2022

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

LIGO-Virgo Collaboration, Phys. Rev. Lett. 116, 061102 (2016)

work page 2016

[2] [2]

S. M. Vermeulen et al. , Nature 600, 424 (2021)

work page 2021

[3] [3]

Morisaki, T

S. Morisaki, T. Suyama, Phys. Rev. D 100, 123512 (2019)

work page 2019

[4] [4]

Fukusumi, S, Morisaki, T, Suyama, Phys

K. Fukusumi, S, Morisaki, T, Suyama, Phys. Rev. D 108, 095054 (2023)

work page 2023

[5] [5]

Obata, T

I. Obata, T. Fujita, Y. Michimura, Phys. Rev. Lett. 121, 161301 (2018)

work page 2018

[6] [6]

Nagano, T

K. Nagano, T. Fujita, Y. Michimura, I. Obata, Phys. Rev. Lett. 123, 111301 (2019)

work page 2019

[7] [7]

Nagano et al

K. Nagano et al. ., Phys. Rev. D 104, 062008 (2021)

work page 2021

[8] [8]

Oshima et al

Y. Oshima et al. , J. Phys. Conf. Ser. 2156, 012042 (2021)

work page 2021