arxiv: 2604.20734 · v1 · submitted 2026-04-22 · ✦ hep-ph · hep-ex

Recognition: unknown

Search for Axion Like Particles produced via the Primakoff process at COMPASS

Authors on Pith no claims yet

Pith reviewed 2026-05-10 00:00 UTC · model grok-4.3

classification ✦ hep-ph hep-ex

keywords Axion-like particlesPrimakoff processALP-photon couplingCOMPASS experimentCompton scatteringMerged photon clustersDark matter mediators

0 comments

The pith

Reinterpreting COMPASS Compton data yields exclusion limits on ALP-photon couplings between 0.2 and 600 MeV.

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

This paper reanalyzes 2009 COMPASS data from 190 GeV pion and muon beams striking a nickel target to hunt for axion-like particles via the Primakoff process. ALPs in the MeV mass range are produced with large boosts so their decay photons merge into single clusters in the electromagnetic calorimeter, exactly mimicking the single-photon signature of standard Primakoff Compton scattering. The authors quantify how much of the observed Compton-like sample could actually come from these ALPs and use the absence of excess to set limits. The resulting bounds exclude couplings larger than about 0.1 GeV inverse at 95 percent . This approach reuses existing data to constrain ALP parameter space that connects beam-dump results to collider searches.

Core claim

The Primakoff production of ALPs followed by their two-photon decay produces merged clusters that contaminate the single-photon Compton scattering sample; by estimating this contamination in the 2009 COMPASS dataset the analysis excludes ALP-photon couplings g_a gamma gamma greater than or equal to 0.1 GeV inverse for masses from 0.2 to 600 MeV at 95 percent .

What carries the argument

The merged photon cluster from a boosted ALP decay that is reconstructed as a single electromagnetic shower, indistinguishable from the Primakoff Compton single-photon signal.

If this is right

The limits provide independent constraints on the ALP-photon coupling that bridge beam-dump experiments and high-energy colliders.
The reinterpretation framework can be applied to other existing datasets with similar calorimeter resolutions.
This work supplies a baseline for future searches that resolve the two photons separately in different kinematic regimes such as Primakoff neutral-pion production.
The method demonstrates how null results in one channel can be turned into new bounds on light dark-sector particles.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Other fixed-target experiments with high-energy beams and electromagnetic calorimeters could apply the same merged-cluster analysis to their archived data.
Tighter limits in this mass window would further restrict models in which ALPs mediate interactions between the Standard Model and a dark sector.
Detectors with finer granularity could test the prediction by separating the diphoton signal and directly observing the ALP decay.

Load-bearing premise

Any ALP-induced merged clusters can be accurately quantified as contamination in the existing Compton sample without large unaccounted backgrounds or systematic errors in the production rate and detector response.

What would settle it

An independent calculation or measurement that shows the expected number of merged clusters from ALPs in the COMPASS setup differs significantly from the observed rate of Compton-like events after subtracting known backgrounds would invalidate the derived limits.

Figures

Figures reproduced from arXiv: 2604.20734 by Mehran Dehpour.

**Figure 2.** Figure 2: FIG. 2: Primakoff ALP production in [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3: Comparison of the measured ratio [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗

**Figure 4.** Figure 4: FIG. 4: The 95% C.L. exclusion limits on the ALP-photon coupling as a function of ALP [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗

read the original abstract

Axion-Like Particles (ALPs) are well-motivated candidates for dark matter and potential mediators to the dark sector. We present a search for ALPs coupled to photons, based on a reinterpretation of COMPASS data. Using the 2009 dataset consisting of $190~\text{GeV}$ $\pi^-$ and $\mu^-$ beams impinging on a fixed nickel target, we investigate the Primakoff production of ALPs. Due to the high beam energy, ALPs in the MeV mass range are produced with a significant Lorentz boost, leading to strongly collimated decay photons. Consequently, these photons are not spatially resolved by the electromagnetic calorimeter and are instead reconstructed as a single merged cluster. This signature mimics the single-photon signal of Standard Model Primakoff Compton scattering, which was the primary focus of the original COMPASS analysis. By quantifying this potential ALP contamination in the Compton scattering sample, we derive exclusion limits on the ALP-photon coupling $g_{a\gamma\gamma}$ in the mass range $0.2 \lesssim m_a \lesssim 600~\text{MeV}$. Our results exclude couplings $g_{a\gamma\gamma} \gtrsim 10^{-1}~\text{GeV}^{-1}$ at 95% C.L., providing independent constraints on the parameter space that bridges beam dump experiments and high energy colliders. While current collider-based limits remain more stringent, this work establishes a novel reinterpretation framework and provides a baseline for future studies of resolved diphoton states in complementary kinematic regimes, such as Primakoff $\pi^0$ production.

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

Reinterpreting COMPASS data for ALP merged-photon contamination gives weak new limits but needs better documentation of the analysis chain.

read the letter

This paper reuses the 2009 COMPASS run with 190 GeV pion and muon beams on nickel to look for axion-like particles produced via Primakoff scattering. The key move is to treat the case where boosted ALP decays produce two photons that merge into one electromagnetic calorimeter cluster as contamination in the original Compton scattering sample, then set limits from that excess not appearing in the data. It produces exclusion curves for g_aγγ above roughly 0.1 GeV^{-1} in the 0.2–600 MeV mass window. That is the main new element: a concrete application of an existing dataset to a merged-photon signature that had not been exploited here before. The approach is sensible on paper and connects beam-dump and collider regimes in a straightforward way. It also sketches how future work could look at resolved diphoton states in similar kinematics. The execution, however, stays thin on the numbers that matter. There is no visible breakdown of the Primakoff differential cross section with nuclear form factors, the Lorentz-boosted decay length distribution, the probability that two showers reconstruct as one cluster, or the efficiency and background subtraction taken from the original Compton analysis. The stress-test concern about 30–50 % uncertainties in merged-cluster modeling is therefore still live; even modest errors would move the already modest limit by a comparable amount. The result is therefore more of a proof-of-concept than a definitive constraint. This is the sort of note that belongs in the ALP and light dark-sector community as an example of data repurposing. A serious editor should send it to referees, but only with the expectation that the authors add the missing Monte Carlo and systematic sections before acceptance. I would not cite it until those details are filled in.

Referee Report

2 major / 2 minor

Summary. The manuscript reinterprets the 2009 COMPASS dataset (190 GeV π⁻ and μ⁻ beams on nickel) to search for ALPs produced via the Primakoff process. Boosted ALPs in the 0.2–600 MeV range decay to two photons that merge into a single ECAL cluster, contaminating the single-photon sample originally analyzed for Primakoff Compton scattering. By estimating this contamination, the authors derive 95% CL exclusion limits excluding g_{aγγ} ≳ 10^{-1} GeV^{-1}.

Significance. If the modeling holds, the result supplies an independent constraint in the intermediate-mass ALP window between beam-dump and collider experiments and demonstrates a reusable reinterpretation technique for existing Primakoff data. The limits themselves are comparatively weak, but the framework is novel and could be strengthened with resolved diphoton analyses in future runs.

major comments (2)

[Analysis and signal modeling] The central limit extraction requires accurate inputs for (i) the differential Primakoff cross section including nuclear form factors, (ii) the Lorentz-boosted decay length and opening-angle distribution, and (iii) the merged-cluster reconstruction efficiency in the ECAL. The manuscript provides no quantitative description, validation plots, or systematic uncertainty budget for these components; any 30–50% mismatch in efficiency directly rescales the quoted g_{aγγ} bound (see skeptic note on unaccounted systematics).
[Results and limit setting] The abstract and results state that the ALP contribution is quantified as contamination in the existing Compton sample, yet no details are given on how the original background subtraction, selection efficiency, or data-driven contamination estimate are adapted for the ALP hypothesis. This information is load-bearing for the 95% CL claim.

minor comments (2)

The title omits the hyphen in 'Axion-Like Particles'.
Notation for the coupling is written inconsistently as g_{aγ γ} in the abstract; standardize to g_{aγγ}.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thorough review of our manuscript on the reinterpretation of COMPASS data for ALP searches via the Primakoff process. The comments highlight important areas where additional details are required to strengthen the presentation of our analysis. We have revised the manuscript to incorporate quantitative descriptions of the signal modeling and limit-setting procedures as requested. Below, we address each major comment point by point.

read point-by-point responses

Referee: [Analysis and signal modeling] The central limit extraction requires accurate inputs for (i) the differential Primakoff cross section including nuclear form factors, (ii) the Lorentz-boosted decay length and opening-angle distribution, and (iii) the merged-cluster reconstruction efficiency in the ECAL. The manuscript provides no quantitative description, validation plots, or systematic uncertainty budget for these components; any 30–50% mismatch in efficiency directly rescales the quoted g_{aγγ} bound (see skeptic note on unaccounted systematics).

Authors: We agree that the original submission lacked sufficient quantitative details on these critical components of the analysis. In the revised manuscript, we have added a new subsection detailing the differential Primakoff cross section, including the incorporation of nuclear form factors using the standard dipole parameterization for the nickel target. For the ALP decay, we now provide explicit formulas and plots for the Lorentz-boosted decay length and the photon opening-angle distribution, generated from Monte Carlo simulations that account for the beam energy and ALP mass range. Regarding the merged-cluster reconstruction efficiency, we describe the ECAL cluster merging criteria based on the calorimeter granularity and shower shape, with an estimated efficiency of 85-95% depending on mass, validated against full detector simulation. We have also included a systematic uncertainty budget, with the dominant contribution from efficiency variations estimated at 25%, which is propagated to the final limits. This addresses the concern about potential rescaling of the bounds. revision: yes
Referee: [Results and limit setting] The abstract and results state that the ALP contribution is quantified as contamination in the existing Compton sample, yet no details are given on how the original background subtraction, selection efficiency, or data-driven contamination estimate are adapted for the ALP hypothesis. This information is load-bearing for the 95% CL claim.

Authors: We acknowledge the need for more explicit information on adapting the original analysis framework. In the revision, we have expanded the results section to explain that the ALP signal is modeled as an additive contamination to the single-photon Compton sample. The original data-driven background subtraction is retained, as the ALP contribution is treated as a signal-like excess in the same kinematic region. We detail how the selection efficiency for merged clusters is derived from the original Compton efficiency by applying the merging condition, and provide a data-driven estimate of the contamination fraction using the expected ALP yield folded with acceptance. A profile likelihood ratio method is now described for setting the 95% CL limits, incorporating the systematic uncertainties from the modeling. This ensures the limit extraction is fully transparent and reproducible. revision: yes

Circularity Check

0 steps flagged

No circularity: limits derived from standard expected-signal calculation against fixed data sample

full rationale

The derivation computes expected ALP event yield N_ALP(g_aγγ) via Primakoff cross section, Lorentz-boosted decay kinematics, and merged-cluster efficiency, then sets 95% CL exclusion where this yield would exceed the observed Compton sample counts within uncertainties. No parameter is fitted to the target result and then renamed as a prediction; the original COMPASS Compton analysis is an external dataset whose event counts are taken as fixed input. No self-citation chain, ansatz smuggling, or self-definitional loop appears in the chain from production rate to limit. The result is therefore a conventional reinterpretation limit, not forced by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The analysis rests on standard particle-physics modeling of Primakoff production and calorimeter response with no new free parameters or postulated entities introduced by the authors.

axioms (2)

domain assumption The Primakoff production cross-section for ALPs follows standard QED calculations.
Invoked when describing ALP production from the beam photon interacting with the nickel nucleus.
domain assumption Boosted ALP decays produce two photons that are spatially unresolved and appear as a single merged cluster in the electromagnetic calorimeter.
Central premise that allows the ALP signal to mimic the Compton scattering signature.

pith-pipeline@v0.9.0 · 5588 in / 1424 out tokens · 43576 ms · 2026-05-10T00:00:15.154464+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

31 extracted references · 14 canonical work pages · 1 internal anchor

[1]

Abelet al., Measurement of the Permanent Electric Dipole Moment of the Neutron, Phys

C. Abelet al., Measurement of the Permanent Electric Dipole Moment of the Neutron, Phys. Rev. Lett.124, 081803 (2020), arXiv:2001.11966 [hep-ex]

work page arXiv 2020
[2]

R. D. Peccei and H. R. Quinn, CP Conservation in the Presence of Instantons, Phys. Rev. Lett.38, 1440 (1977)

1977
[3]

Weinberg, A New Light Boson?, Phys

S. Weinberg, A New Light Boson?, Phys. Rev. Lett.40, 223 (1978)

1978
[4]

Wilczek, Problem of StrongPandTInvariance in the Presence of Instantons, Phys

F. Wilczek, Problem of StrongPandTInvariance in the Presence of Instantons, Phys. Rev. Lett.40, 279 (1978)

1978
[5]

Axions In String Theory

P. Svrcek and E. Witten, Axions In String Theory, JHEP06, 051, arXiv:hep-th/0605206. 15

work page Pith review arXiv
[6]

String Axiverse

A. Arvanitaki, S. Dimopoulos, S. Dubovsky, N. Kaloper, and J. March-Russell, String Axi- verse, Phys. Rev. D81, 123530 (2010), arXiv:0905.4720 [hep-th]

work page internal anchor Pith review arXiv 2010
[7]

A. E. Nelson and N. Seiberg,Rsymmetry breaking versus supersymmetry breaking, Nucl. Phys. B416, 46 (1994), arXiv:hep-ph/9309299

work page arXiv 1994
[8]

Preskill, M

J. Preskill, M. B. Wise, and F. Wilczek, Cosmology of the Invisible Axion, Phys. Lett. B120, 127 (1983)

1983
[9]

L. F. Abbott and P. Sikivie, A Cosmological Bound on the Invisible Axion, Phys. Lett. B120, 133 (1983)

1983
[10]

Navaset al.(Particle Data Group), Review of Particle Physics, Phys

S. Navaset al.(Particle Data Group), Review of Particle Physics, Phys. Rev. D110, 030001 (2024)

2024
[11]

Probing MeV to 90 GeV axion-like particles with LEP and LHC,

J. Jaeckel and M. Spannowsky, Probing MeV to 90 GeV axion-like particles with LEP and LHC, Phys. Lett. B753, 482 (2016), arXiv:1509.00476 [hep-ph]

work page arXiv 2016
[12]

Searching for Axionlike Particles with Ultraperipheral Heavy-Ion Collisions,

S. Knapen, T. Lin, H. K. Lou, and T. Melia, Searching for Axionlike Particles with Ultrape- ripheral Heavy-Ion Collisions, Phys. Rev. Lett.118, 171801 (2017), arXiv:1607.06083 [hep-ph]

work page arXiv 2017
[13]

Abbonet al.(COMPASS), The COMPASS experiment at CERN, Nucl

P. Abbonet al.(COMPASS), The COMPASS experiment at CERN, Nucl. Instrum. Meth. A 577, 455 (2007), arXiv:hep-ex/0703049

work page arXiv 2007
[14]

Adolphet al.(COMPASS), Measurement of the charged-pion polarizability, Phys

C. Adolphet al.(COMPASS), Measurement of the charged-pion polarizability, Phys. Rev. Lett.114, 062002 (2015), [The values fort target = 0.285X 0,l target = 4.2 mm, andl ECAL2 = 33.98 m are extracted from the Sept. 2012 internal release note. The effective spectrometer thicknesst spec ≈0.026X 0 was obtained via private communication.], arXiv:1405.6377 [hep-ex]

work page arXiv 2015
[15]

Aloni, C

D. Aloni, C. Fanelli, Y. Soreq, and M. Williams, Photoproduction of Axion-like Particles, Phys. Rev. Lett.123, 071801 (2019), arXiv:1903.03586 [hep-ph]

work page arXiv 2019
[16]

J. R. Pybuset al., Search for axion-like particles through nuclear Primakoff production using the GlueX detector, Phys. Lett. B855, 138790 (2024), arXiv:2308.06339 [hep-ex]

work page arXiv 2024
[17]

Bethe and W

H. Bethe and W. Heitler, On the Stopping of fast particles and on the creation of positive electrons, Proc. Roy. Soc. Lond. A146, 83 (1934)

1934
[18]

C. F. von Weizsacker, Radiation emitted in collisions of very fast electrons, Z. Phys.88, 612 (1934)

1934
[19]

E. J. Williams, Nature of the high-energy particles of penetrating radiation and status of ionization and radiation formulae, Phys. Rev.45, 729 (1934). 16

1934
[20]

K. J. Kim and Y.-S. Tsai, Improved Weizsacker-Williams method and its application to lepton andWboson pair production, Phys. Rev. D8, 3109 (1973)

1973
[21]

Y. M. Andreev and E. V. Bugaev, Muon bremsstrahlung on heavy atoms, Phys. Rev. D55, 1233 (1997)

1997
[22]

A. S. Galperin and G. Mitselmakher, Calculation of Ultrarelativistic Pion Bremsstrahlung Cross-section on Nuclei With the Allowance for Pion Polarization, JINR Report (1980)

1980
[23]

Gasser, M

J. Gasser, M. A. Ivanov, and M. E. Sainio, Revisitingγγ→π +π− at low energies, Nucl. Phys. B745, 84 (2006), arXiv:hep-ph/0602234

work page arXiv 2006
[24]

Georgi, D

H. Georgi, D. B. Kaplan, and L. Randall, Manifesting the Invisible Axion at Low-energies, Phys. Lett. B169, 73 (1986)

1986
[25]

C. P. Oliveira, D. Hadjimichef, and M. V. T. Machado, Production of the axion-like particles on electron-nucleus and ultraperipheral heavy ion collisions, J. Phys. G48, 085005 (2021)

2021
[26]

V. M. Budnev, I. F. Ginzburg, G. V. Meledin, and V. G. Serbo, The Two photon particle production mechanism. Physical problems. Applications. Equivalent photon approximation, Phys. Rept.15, 181 (1975)

1975
[27]

R. R. Dusaev, D. V. Kirpichnikov, and M. M. Kirsanov, Photoproduction of axionlike particles in the NA64 experiment, Phys. Rev. D102, 055018 (2020), arXiv:2004.04469 [hep-ph]

work page arXiv 2020
[28]

O’Hare, cajohare/axionlimits,https://cajohare.github.io/AxionLimits/

C. O’Hare, cajohare/axionlimits,https://cajohare.github.io/AxionLimits/
[29]

Friedrich (COMPASS), The pion polarisability and more measurements on chiral dynamics at COMPASS, PoSCD15, 015 (2016)

J. Friedrich (COMPASS), The pion polarisability and more measurements on chiral dynamics at COMPASS, PoSCD15, 015 (2016)

2016
[30]

Letter of Intent: A New QCD facility at the M2 beam line of the CERN SPS (COMPASS++/AMBER)

B. Adamset al.(AMBER), A New QCD facility at the M2 beam line of the CERN SPS: COMPASS++/AMBER, SPSC-I-250 (2018), arXiv:1808.00848 [hep-ex]

work page Pith review arXiv 2018
[31]

Ecker (COMPASS), Testing predictions of the chiral anomaly in Primakoff reactions at COMPASS, Nuovo Cim

D. Ecker (COMPASS), Testing predictions of the chiral anomaly in Primakoff reactions at COMPASS, Nuovo Cim. C47, 217 (2024), arXiv:2310.09138 [hep-ph]. 17

work page arXiv 2024