Pith. sign in

REVIEW 4 major objections 6 minor 12 references

Cygnus X-3's PeV photons trace to protons accelerated in its jet

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-03 14:51 UTC pith:PJ72CRW7

load-bearing objection A short, honest addendum that ties a hadronic-jet model to the new LHAASO Cygnus X-3 data, but whose agreement with the sub-PeV drop rests on an unvalidated ~2000-fold wind-photon enhancement, so the quantitative match is provisional. the 4 major comments →

arxiv 2512.18786 v3 pith:PJ72CRW7 submitted 2025-12-21 astro-ph.HE

Cygnus X-3 as a PeVatron and the LHAASO 2025 data

M. Kachelriess , E. Lammert This is my paper
classification astro-ph.HE
keywords Cygnus X-3PeVatroncosmic-ray accelerationmicroquasar jetphoto-hadronic interactionsgamma-ray emissionorbital modulationline-driven stellar wind
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Using the recently measured photon flux from Cygnus X-3 that extends to 4 PeV and varies with the binary orbit, this paper argues that the source is a cosmic-ray accelerator in its jet, not just a leptonic emitter. The authors show that protons accelerated beyond tens of PeV produce the gamma rays by colliding with the companion star's UV photons and with wind gas; the UV channel naturally peaks near a PeV and is orbitally modulated because the viewing angle between the jet and the stellar light changes during the orbit. To explain the sharp drop of the flux just below 1 PeV, they propose that the line-driven wind of the Wolf-Rayet star traps and multiplies UV photons, increasing pair-production absorption. If this interpretation holds, Cygnus X-3 becomes the first binary system identified as a PeVatron.

Core claim

The central claim is that the PeV photon emission observed from Cygnus X-3 is generated by hadronic interactions of ultra-relativistic protons accelerated in the jet. The paper demonstrates that photo-hadronic interactions on stellar UV photons produce a spectrum that rises toward a PeV and exhibits orbital modulation with the same phase as the GeV emission, consistent with observations. The fast drop below the PeV peak is then attributed to pair-production absorption on an enhanced UV photon density in the line-driven stellar wind, where multiple line scatterings amplify the photon density by a factor of order 2000 within a few stellar radii. With this adjustment, the predicted average spec

What carries the argument

The paper's key objects are (1) the photo-hadronic reaction rate on an anisotropic UV photon background, approximated by integrating the cross section with a fixed cosine µ0 of the angle between the proton and photon momenta—this angle changes over the orbit and produces the modulation—and (2) the enhanced pair-production absorption in the wind, modeled as a random walk of photons with a density enhancement factor N^{1/2} ~ v∞/Δv ~ 2000 in the region where the wind is still accelerating. The first element converts the spectrum into a probe of jet physics; the second explains the abrupt turnover near 1 PeV.

Load-bearing premise

The linchpin is that the line-driven wind of the Wolf-Rayet star raises the UV photon density near the star by a factor of roughly 2000, making pair-production absorption strong enough to turn over the spectrum just below 1 PeV—the paper notes that without this enhancement, the predicted flux rises too slowly to match the data.

What would settle it

Measure the spectrum of Cygnus X-3 between 0.1 and 1 PeV in two orbital phase bins (inferior and superior conjunction): the absorption hypothesis predicts a significant phase-dependent difference in the turnover energy, while free-streaming models predict a monotonic power law; observing no phase dependence would falsify the wind-absorption claim.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • If correct, Cygnus X-3 is a PeVatron, establishing that microquasar jets accelerate cosmic rays to at least tens of PeV.
  • The neutrino flux from the same photo-hadronic interactions is predicted to share the orbital modulation, so a time-dependent search in neutrino telescopes would improve sensitivity.
  • The sharp sub-PeV turnover becomes a diagnostic of the wind's UV photon density and can be used to measure the structure of the line-driven wind.
  • In the authors' parameter regime, leptonic emission cannot reach PeV energies, so future PeV observations would strengthen the hadronic interpretation.

Where Pith is reading between the lines

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

  • If this mechanism is generic, other high-mass X-ray binaries with Wolf-Rayet companions might also be PeVatrons, and could contribute to the galactic cosmic-ray flux; searching for orbitally modulated PeV emission from similar systems would test this.
  • The wind-absorption explanation makes a phase-dependent prediction: the depth of the sub-PeV drop should vary with orbital phase, because the line of sight crosses different amounts of the dense wind; a phase-resolved spectral analysis could confirm or refute it.
  • The random-walk enhancement factor is a rough estimate; a detailed radiative-transfer model of the line-driven wind could replace it with a first-principles prediction, turning the absorption hypothesis into a falsifiable quantitative model.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper is a short addendum to the authors' earlier model of Cygnus X-3 as a PeV cosmic-ray accelerator. It confronts the KL25 hadronic-jet model with the LHAASO 2025 measurement of orbitally modulated photons up to 4 PeV. The model has two photon components: hadronic interactions on stellar-wind gas, which follow the injected proton spectrum, and photo-hadronic interactions on stellar UV photons, which produce a component rising up to about 1 PeV. To reproduce the sharp sub-PeV rise in the data, the authors add gamma-gamma absorption on an enhanced UV photon density inside about 3 stellar radii; the enhancement is estimated as ~2000 from a random-walk argument for line-driven winds. They also argue that X-ray target photons are negligible and that a leptonic origin cannot reach PeV energies, and they predict an orbitally modulated neutrino flux. The agreement with LHAASO is presented by visual comparison in Fig. 1.

Significance. Should the interpretation survive scrutiny, it would be an important step: it identifies a microquasar as a PeV hadronic accelerator and connects the LHAASO spectrum to wind physics and a testable neutrino orbital modulation. The paper has several strengths: it is based on a published Monte Carlo framework, separates the gas and UV target components, uses the phase information of the LHAASO data qualitatively, and makes falsifiable predictions for neutrino time templates and for a declining modulation below 100 TeV. Its main weakness is that the quantitative match to the sub-PeV drop rests entirely on an absorption enhancement factor that is calibrated to the data and derived with a simplified, possibly misapplied, line-scattering argument. As presented, the result is a proof-of-concept rather than a validated model; a revision that turns the enhancement into a derived or scanned parameter with a real fit statistic is necessary.

major comments (4)
  1. [Numerical results; Fig. 1] The central agreement with LHAASO is obtained by tuning the absorption. The text states that the UV-only spectrum is 'too slow to reproduce well the measurements' and that matching the fast rise requires absorption 'increased by a factor of order 100-1000 compared to the standard treatment'; the adopted value, ~2000 from v_inf/Delta_v, is then used to display a curve that 'agrees better with the data.' Because this factor is not measured independently and no sensitivity scan is shown, the procedure is circular: the model is adjusted to the same spectral feature that it is then used to explain. Please provide a scan in the enhancement factor (including 0, 100, 1000, 2000) and a quantitative fit statistic, or derive the factor from a concrete wind model.
  2. [Numerical results; random-walk argument] The random-walk estimate N ~ (v_inf/Delta_v)^2 and the density enhancement sqrt(N) ~ 2000 describe photons trapped in spectral lines of a line-driven wind. It is not established that the continuum differential density n_gamma(epsilon) that enters Eq. (1) and the gamma-gamma opacity is enhanced by the same factor at the photon energies relevant for p-gamma and gamma-gamma interactions. The paper applies this single factor to the full UV target field. Please specify the frequency-dependent photon density after line transport, and identify which target energies are actually enhanced; otherwise the claim tau_gamma_gamma ~ 1 at 1 PeV is not supported.
  3. [Numerical results; Fig. 1 and Fig. 3] The comparison to LHAASO data is qualitative. No statistical test, residuals, or quantified uncertainty is presented, and the predicted orbital modulation in Fig. 3 is not compared with the measured modulation amplitude or phase. For a paper whose main conclusion is agreement with data, a simple chi-squared or likelihood over the binned spectrum, plus a comparison of the phase curve, should be added. This is particularly important because the data set has only a handful of energy bins.
  4. [Reaction rates; neglect of X-ray targets] The assumption that X-ray photons are a negligible target is based on the absence of orbital modulation and on the IXPE picture. However, X-rays from the inner envelope could still contribute a non-modulated opacity in the acceleration region, and the previous KL25 model included them. Please quantify the X-ray photon density at the assumed acceleration distance and show that the p-gamma rate on X-rays is small compared to the UV rate for the adopted B=2000 G and L0=7e11 cm. As written, setting the X-ray contribution to zero is another parameter choice that selectively favors the UV interpretation.
minor comments (6)
  1. [Eq. (1) and Fig. 2] The notation for the angle is inconsistent: Eq. (1) uses mu_0 while Fig. 2 and the text use mu. Define the angle once and keep the notation fixed.
  2. [Introduction] The phrase 'the slope in an E2dN/dE plot' should read E^2 dN/dE.
  3. [Throughout] Typos: 'relaying on strong magnetisation' should be 'relying'; 'photo-hadronic interaction of stellar UV photons' should be 'interactions with stellar UV photons'.
  4. [References] The reference to Barrios-Jimenez (2025) is a conference talk; if a published version exists, it should be cited instead.
  5. [Fig. 2] The axis label 'R/s-1' should be typeset as 'R [s^-1]'.
  6. [Fig. 1 caption/main text] Clarify that the LHAASO data are the flaring-state, de-absorbed spectrum while the model curve is phase-averaged. If the comparison is phase-averaged, state that explicitly in the caption and in Section 'Numerical results'.

Circularity Check

1 steps flagged

Sub-PeV absorption enhancement is calibrated to the observed drop, making part of the spectral agreement pre-arranged; the PeV peak and orbital modulation remain genuine predictions.

specific steps
  1. fitted input called prediction [Numerical results, paragraph discussing Fig. 1 and enhanced absorption]
    "This rise, typical for photon fluxes below the threshold energy if there is no strong beaming, is too slow to reproduce well the measurements. As a possible explanation for the fast rise of the measured photon flux, we explore next the option that the photons produced in pγ interactions with energies below 1 PeV are (partially) absorbed, i.e. that τγγ ≃ 1 at 1 PeV. This would require that the absorption is increased by a factor of order 100-1000 compared to the standard treatment ... the density of photons in a line-driven wind is thus enhanced by the factor (N)1/2 ≃ v∞/Δv ∼ 2000."

    The unabsorbed model explicitly fails to match the data. The paper then states the observed fast rise requires an absorption enhancement of 100–1000 and adopts a random-walk estimate of ~2000, which sets τγγ ≈ 1 at 1 PeV as the data demand. The 'predicted' absorbed spectrum is therefore computed with a parameter whose magnitude is chosen to produce the observed drop; the subsequent agreement is partly an identity, not an independent test. The physical wind-scattering argument gives some independent justification, so the circularity is partial rather than total.

full rationale

The paper's central hadronic-jet interpretation has genuinely predictive elements: the pγ threshold peak near PeV and the orbital-phase modulation of the UV-interaction component follow from kinematics and geometry, and are compared to LHAASO data without tuning. The gas-interaction component also follows the assumed CR slope. The main circularity is confined to the sub-PeV absorption step: the unabsorbed UV component 'is too slow to reproduce well the measurements,' and the paper then states that the data require an absorption enhancement of order 100–1000, adopts ~2000 from a random-walk estimate, and shows the absorbed flux agrees better. Thus the sharp sub-PeV rise is explained by a parameter whose numerical target was read off the same data. The enhancement has an independent physical motivation via line-driven wind theory, which prevents the circularity from being total. The KL25 self-citation is not load-bearing in a circular sense: it cites the authors' previously published acceleration model and Monte Carlo framework, not an unsupported uniqueness claim. No other circular patterns apply. Overall score 5 reflects one partially data-calibrated ingredient while the main spectral peak and modulation predictions retain independent content.

Axiom & Free-Parameter Ledger

3 free parameters · 5 axioms · 0 invented entities

The paper relies on a small number of hand-chosen parameters (B, L0, the absorption enhancement factor) and several domain assumptions about the proton composition, jet geometry, X-ray target fields, and wind density. No new particles or forces are introduced.

free parameters (3)
  • Magnetic field strength B = 2000 G
    Chosen by hand as the lowest value compatible with a Hillas energy of 6e16 eV; affects the acceleration rate and maximal energy.
  • Size of acceleration region L0 = D = 7e11 cm
    Adopted from KL25; sets the Hillas energy and the diffusion transition scale.
  • Photon-density enhancement factor from line-driven wind = ~2000 (data requires 100-1000)
    Introduced to make τγγ ≈ 1 at 1 PeV, reproducing the observed sharp drop below 1 PeV; the value is an order-of-magnitude estimate from a random-walk argument.
axioms (5)
  • domain assumption The accelerated cosmic rays are protons.
    The paper consistently treats interactions as pγ on stellar UV photons, e.g., 'interactions of CR protons on UV photons'.
  • domain assumption The jet direction is fixed during the orbit and the counter-jet contribution is negligible.
    Used to compute the orbital modulation in Fig. 3; the text states the counter-jet is suppressed by moderately relativistic bulk motion.
  • domain assumption X-ray photons are a negligible target for photo-hadronic interactions.
    Assumed based on the LHAASO orbital modulation and IXPE observations; if wrong, the energy-loss rates and spectra would change.
  • ad hoc to paper The line-driven wind enhances the UV photon density by a factor of ~2000 within ~3 stellar radii.
    This estimate is needed to explain the observed sub-PeV absorption; it is imported from stellar wind theory and not derived in detail here.
  • domain assumption Bohm diffusion (γ=1 in Eq. 2) is used for the energy gain rate.
    Chosen for simplicity after noting Kraichnan diffusion gives a similar maximal energy; affects the DSA rate.

pith-pipeline@v1.3.0-alltime-deepseek · 4982 in / 11824 out tokens · 123853 ms · 2026-08-03T14:51:17.135525+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Cygnus X-3 as a PeVatron and the LHAASO 2025 data." pith.science (2026). https://pith.science/paper/PJ72CRW7

@misc{pith2026251218786,
  author       = {Pith},
  title        = {Pith review of: Cygnus X-3 as a PeVatron and the LHAASO 2025 data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PJ72CRW7}},
  note         = {Machine review of arXiv:2512.18786}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

We have recently argued that the high-mass X-ray binary Cygnus X-3 can accelerate cosmic rays (CR) beyond PeV energies. Meanwhile, the LHAASO collaboration published the measurement of an orbitally modulated photon flux from Cygnus X-3 extending up to 4 PeV. In this short extension of our previous work, we argue that these observations point towards CR acceleration in the jet, and secondary production in CRs scattering on gas from the wind and on stellar UV photons from the companion star. The latter channel leads naturally to a contribution to the photon flux peaking around PeV energies which is strongly orbitally modulated. The fast drop in the flux of these photons below PeV energies may be caused by absorption on an increased density of background photons in a line-driven stellar wind.

Figures

Figures reproduced from arXiv: 2512.18786 by E. Lammert, M. Kachelriess.

Figure 1
Figure 1. Figure 1: — The average photon flux as function of energy com￾pared to the LHAASO data Cao et al. (2025) from Cygnus X-3 in the flaring state (corrected for Galactic absorption); shown is also separately the contribution from scatterings on UV photons and gas. of-mass energy of the two particles, s = 2Eε(1 − µ0), and the flux factor 1 − µ0 change with the angle ϑ0 dur￾ing one orbit, leading to an orbital modulation … view at source ↗
Figure 3
Figure 3. Figure 3: — The modulation amplitude of the photon flux produced in photo-hadronic interactions as function of the orbital phase. sured by the LHAASO collaboration is in rough agree￾ment with our prediction based on CR acceleration in the jet and a two-component model based on interactions on gas and UV photons. The sharp rise of the photon flux around PeV may indicate that the density of UV photon close to the WR s… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

12 extracted references · 7 linked inside Pith

  1. [1]

    author R. J. Protheroe , journal volume 90 , pages 883 ( year 1994 )

    1994

  2. [2]

    Kachelrie and author E

    author M. Kachelrie and author E. Lammert , journal Astron. Astrophys. volume 701 , pages A22 ( year 2025 ), 2503.11448

    Pith/arXiv arXiv 2025

  3. [3]

    Cao et al

    author Z. Cao et al. ( collaboration LHAASO ) ( year 2025 ), 2512.16638

    Pith/arXiv arXiv 2025

  4. [4]

    Aleksi \'c , author L

    author J. Aleksi \'c , author L. A. Antonelli , author P. Antoranz , author M. Backes , author C. Baixeras , author J. A. Barrio , author D. Bastieri , author J. Becerra Gonz \'a lez , author W. Bednarek , author A. Berdyugin , et al. , journal volume 721 , pages 843 ( year 2010 ), 1005.0740

    Pith/arXiv arXiv 2010

  5. [5]

    Barrios-Jiménez ( collaboration MAGIC ), journal talk at TeVPA 2025 ( year 2025 )

    author L. Barrios-Jiménez ( collaboration MAGIC ), journal talk at TeVPA 2025 ( year 2025 )

    2025

  6. [6]

    author A. A. Abdo et al. , journal Science volume 326 , pages 1512 ( year 2009 )

    2009

  7. [7]

    author A. A. Zdziarski , author D. Malyshev , author G. Dubus , author G. G. Pooley , author T. Johnson , author B. de Marco , author A. Frankowski , author M. Chernyakova , and author A. R. Rao , journal Mon. Not. Roy. Astron. Soc. volume 479 , pages 4399 ( year 2018 ), 1804.07460

    Pith/arXiv arXiv 2018

  8. [8]

    Dubus , author B

    author G. Dubus , author B. Cerutti , and author G. Henri , journal Mon. Not. Roy. Astron. Soc. volume 404 , pages 55 ( year 2010 ), 1002.3888

    Pith/arXiv arXiv 2010

  9. [9]

    Veledina et al

    author A. Veledina et al. , journal Astron. Astrophys. volume 688 , pages L27 ( year 2024 ), 2407.02655

    Pith/arXiv arXiv 2024

  10. [10]

    Miku s incov \'a et al

    author R. Miku s incov \'a et al. ( year 2025 ), 2512.12879

    arXiv 2025

  11. [11]

    author P. O. Lagage and author C. J. Cesarsky , journal Astron. Astrophys. volume 125 , pages 249 ( year 1983 )

    1983

  12. [12]

    Giacinti , author M

    author G. Giacinti , author M. Kachelrie , and author D. V. Semikoz , journal JCAP volume 1807 , pages 051 ( year 2018 ), 1710.08205

    Pith/arXiv arXiv 2018