Electromagnetic radiation from Quark-Gluon Plasma at finite baryon density

Bj\"orn Schenke; Charles Gale; Chun Shen; Jean-Fran\c{c}ois Paquet; Sangyong Jeon; Xiang-Yu Wu

REVIEW 3 major objections 5 minor 13 references

Electromagnetic radiation from Quark-Gluon Plasma at finite baryon density

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read At low collision energies, thermal photons from the quark-gluon plasma can outnumber prompt photons even at high transverse momentum, and dilepton spectra can reveal the plasma's earliest moments.

desk verdict A useful finite-baryon extension of the iEBE-MUSIC EM calculations, but the marquee high-pT thermal-over-prompt result leans on an unvalidated NLO prompt-photon prediction at large x. read the letter →

arxiv 2509.03289 v1 pith:JDBRMDFD submitted 2025-09-03 nucl-th

Xiang-Yu Wu , Charles Gale , Sangyong Jeon , Jean-François Paquet , Björn Schenke , Chun Shen This is my paper · ORCID

classification nucl-th

keywords thermalphotonsdileptonsquark-gluonplasmafinitebaryondensityheavy-ioncollisionsbeamenergyscanBayesiancalibrationpre-equilibriumdynamics

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 tries to establish that electromagnetic radiation—photons and dileptons—from the quark-gluon plasma can be isolated from competing sources at the relatively low collision energies explored by the beam energy scan. Using a hydrodynamical model calibrated to hadron data through Bayesian inference, the authors compute thermal photon emission at finite baryon density and find that at low collision energies thermal photons overtake prompt photons even at high transverse momentum, making the plasma glow measurable. They also show that the intermediate-mass dilepton spectrum is sensitive to how soon the plasma forms and to the size of initial hot spots, so dileptons can serve as a probe of pre-equilibrium dynamics. If correct, these results would extend electromagnetic probes of the QGP to the high-baryon-density region of the QCD phase diagram and motivate including such observables in future model calibrations.

What carries the argument

The machinery is the multistage hydrodynamic model used here: three-dimensional dynamical initial conditions deposit energy and momentum gradually between proper times 0.5 and 1.5 fm/c, a viscous relativistic hydrodynamic code evolves the medium, a hadronic afterburner completes the evolution, and Bayesian inference fixes the parameters using hadronic data from the RHIC beam energy scan. Thermal photon and dilepton yields are then obtained by convolving in-medium emission rates—partonic and hadronic for photons, QGP-phase for dileptons, with the switch at 160 MeV—with local fluid-cell temperature, flow velocity, and baryon chemical potential, while prompt photons come from an NLO perturbative QCD calculation. This setup is what lets the authors compute the thermal-to-prompt photon ratio and the response of dilepton spectra to different initialization procedures.

What would settle it

Measure prompt photon production in proton-proton or proton-nucleus collisions at the same low nucleon-nucleon energies up to high pT and compare with the NLO pQCD prediction; if the measured prompt yield substantially exceeds the calculation in the region where the paper predicts thermal dominance (pT above about 2.5 GeV at 14.6 GeV), the claimed thermal excess in Au+Au or Pb+Pb would be an artifact of an underestimated background.

Watch

Extended reading notes

Core claim

The central result is that the balance between thermal and prompt photon production shifts with collision energy: while at 200 GeV prompt photons dominate for transverse momenta above about 2.5 GeV, at energies around 14.6–17.3 GeV the thermal component exceeds the prompt one even at high pT, because the next-to-leading-order perturbative QCD prompt yield falls steeply at large Bjorken-x. For dileptons, the paper finds that the intermediate-mass-region excess is enhanced when the plasma is initialized early (at 0.5 fm/c) rather than late (1.5 fm/c), and is larger for the posterior parameter set with smaller initial hotspots (0.1 fm) than for one with larger hotspots (0.3 fm). These comparisons with STAR and PHENIX photon data and with NA60 dilepton data indicate that electromagnetic observables are sensitive to the initial state and pre-equilibrium stage at finite baryon density.

Load-bearing premise

The high-pT thermal excess at low collision energies rests on the NLO perturbative QCD prompt-photon calculation remaining accurate at the large Bjorken-x values those collisions reach; if the true prompt background is larger there, the claimed thermal dominance disappears.

Editorial extensions

If this is right

At collision energies near 14.6–17.3 GeV, thermal photons outnumber prompt photons even for pT above 2.5 GeV, so the thermal component is not confined to low pT and its identification becomes less ambiguous.
The intermediate-mass dilepton excess is larger when the quark-gluon plasma is formed early (0.5 fm/c) than when formation is delayed to 1.5 fm/c, making the IMR a probe of pre-equilibrium dynamics.
Posterior parameter sets with smaller initial hotspots (0.1 fm) yield more intermediate-mass dileptons than sets with larger hotspots (0.3 fm), so dilepton spectra can distinguish initial-state geometry choices.
The demonstrated sensitivity of photons and dileptons motivates including electromagnetic observables in future Bayesian calibrations to tighten constraints on QGP properties at finite baryon density.

Reading between the lines

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

The steep fall of the prompt-photon calculation at large Bjorken-x implies that the thermal excess should grow further at collision energies below 14.6 GeV, a trend this paper does not itself compute.
Because dilepton yields depend on both initialization time and hotspot size, combining photon and dilepton measurements at the same energy could break degeneracies between pre-equilibrium timing and initial-state geometry in future Bayesian fits.
If the high-pT thermal excess is confirmed, high-pT photon analyses at low collision energies would need to subtract a plasma component rather than assume all high-pT photons are prompt.

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

The paper gives the heavy-ion community new calculations of thermal photons and dileptons at finite baryon density, using a Bayesian-calibrated iEBE-MUSIC framework. That is genuinely useful: it extends a mature tool to the BES/SPS regime, folds in independent hadronic calibrations, and shows clean sensitivity studies for pre-equilibrium initialization and posterior choices. No EM data are used in the fit, so the circularity burden is low, and the 200 GeV direct-photon comparison with STAR/PHENIX provides a sanity check that the framework still works in a known regime. The energy scan of the thermal-to-prompt ratio is the kind of plot experimentalists will want to see.

The soft spot is exactly where the stress-test note points. The claim that thermal photons dominate at high pT in 14.6–17.3 GeV collisions rests on INCNLO's steep fall at large Bjorken-x. That is a load-bearing part of the central conclusion, but the paper gives no validation of prompt production in that kinematic region—the left-panel comparison is at 200 GeV, where x_T is far smaller—and the right panel has no uncertainty band on the prompt component. The scale-variation discussion for 200 GeV cannot simply be assumed to apply at x_T ~ 0.4–0.55, where threshold logarithms and intrinsic-kt effects are known to matter. So the 'thermal may be measurable' statement should be read as conditional on the prompt background being right, not as a robust prediction. That does not sink the paper; it does mean the headline result is a prompt to investigate, not a settled finding.

Other concerns are minor and typical for a proceedings. The 'even better description' of NA60 data is not backed by a goodness-of-fit metric, and no code or data artifacts are provided. Neither is disqualifying here—the plots speak reasonably clearly—but they do limit how much weight a referee can put on fine distinctions.

Overall, this is a competent, honest contribution from people who know the framework. The math and rate inputs look fine, the qualitative statements align with the figures, and the authors are transparent about model choices. It deserves serious referee attention, but the referee should push for a quantitative treatment of the prompt-photon uncertainty at low energies, ideally a resummed or threshold-improved calculation, before the main claim is taken as strong evidence.

Referee Report

3 major / 5 minor

Summary. This proceedings-style paper presents electromagnetic radiation predictions from the iEBE-MUSIC hydrodynamic framework with finite-baryon-density Bayesian-calibrated initial conditions. The authors compute thermal photon and thermal dilepton yields by convolving in-medium emission rates with posterior-sampled hydrodynamic backgrounds, adding prompt photons from INCNLO. They report three main results: (i) at 200 GeV the direct photon spectrum is consistent with STAR and PHENIX data; (ii) at low collision energies (14.6-17.3 GeV) the calculated thermal photon yield exceeds the prompt yield even at high transverse momentum, which they attribute to the steep fall of NLO pQCD prompt photons at large Bjorken-x; and (iii) intermediate-mass dilepton spectra from In+In collisions at 17.3 GeV are sensitive to the pre-equilibrium initialization time and to the initial-state hotspot size encoded in different Bayesian posterior sets.

Significance. If the central claims hold, the paper identifies a potentially observable window for thermal photon emission at RHIC-BES and SPS energies, and it demonstrates that intermediate-mass dileptons can discriminate among early-time scenarios, which would motivate future Bayesian analyses that include electromagnetic probes. A notable strength is the absence of circularity: no electromagnetic data are used in the calibration, and the thermal emission rates are independent inputs. The 200 GeV direct photon comparison with STAR/PHENIX provides a useful sanity check. However, the main predictive claim about low-energy high-pT thermal photons rests on an unvalidated prompt-photon background calculation, and the dilepton sensitivity claims are not quantified with uncertainties or goodness-of-fit measures. The potential impact is real but conditional on addressing these correctness risks.

major comments (3)

[Section 3, Fig. 1 right panel] The central claim that thermal photons dominate over prompt photons at high pT for low collision energies rests entirely on the INCNLO prediction that 'our calculations of prompt photons fall steeply at large Bjorken-x.' At sqrt(sNN)=14.6-17.3 GeV and pT around 3-4 GeV, x_T = 2pT/sqrt(s) is approximately 0.4-0.55, a regime where NLO pQCD without threshold resummation or intrinsic transverse momentum is known to be less reliable. The left panel validates the prompt calculation only at 200 GeV, where x_T is much smaller for the same pT range, and the quoted scale-variation uncertainty (0.5pT to 2pT) is not shown for the low-energy ratio. Please provide the scale-variation band for the thermal-to-prompt ratio at low energies, compare the prompt predictions with existing low-energy direct photon measurements (e.g., from the SPS program), or use a resummed calculation, and state how the claimed excess depends on this background. Without this, the high-pT thermal-over-prompt observation is not established.
[Section 3, Fig. 2] The statements that instantaneous initialization at tau0=0.5 fm/c gives 'an even better description of the NA60 data' and that Posterior 1 gives 'a larger yield in the IMR' are not supported by any quantitative measure. The plotted model curves have no uncertainty bands from the posterior sampling or from the thermal emission rates, and no chi-square, likelihood, or data-to-model ratio is reported for the different scenarios. Since the sensitivity of dileptons to the pre-equilibrium stage and to the hotspot size is a central conclusion, please provide a goodness-of-fit comparison or at least show posterior/statistical uncertainties on the model curves so that the reader can assess whether the differences among the scenarios are significant.
[Section 2 and Section 3] Only thermal dileptons from the QGP phase are included in the calculation, yet the text states that the emitted dilepton spectrum 'is compatible with NA60 data' without specifying the invariant-mass range displayed in Fig. 2. If the plotted comparison includes the low-mass region (LMR), hadronic thermal dilepton sources would be non-negligible and their omission would make the comparison misleading. Please clarify the mass range shown and either justify the neglect of hadronic thermal dileptons or include them. This issue also affects the comparison of Posterior 1 and Posterior 2 in the LMR, where the two sets are claimed to give similar yields.

minor comments (5)

[Section 3, Fig. 1] The text states that the prompt photon uncertainty is estimated by scale variation from 0.5pT to 2.0pT, but the right panel of Fig. 1 shows no such uncertainty band. State explicitly that only central values are shown in that panel.
[Section 3, Fig. 2] The caption of Fig. 2 uses 'dNch/deta >30' but this quantity is not defined in the text; please define it or give a reference for the selection.
[Abstract and Section 2] The phrase 'posterior-sampled backgrounds' is not elaborated; a brief description of how many posterior samples were used and how they were propagated would help the reader understand the robustness of the results.
[Section 3, Fig. 2] The error bars on the NA60 data are not described in terms of statistical versus systematic uncertainties; please specify what is shown.
[Section 3] The sentence 'Posterior 1 [4] is the default configuration in this proceeding' is awkward; 'proceeding' appears to refer to this manuscript, and the wording should be revised for clarity.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the thermal photon and dilepton calculations use independent emission rates and hadronically calibrated hydrodynamic backgrounds, and the high-pT thermal-over-prompt excess is not a fit to EM data.

full rationale

The paper's derivation chain is self-contained. The thermal photon and dilepton yields are computed by folding independent in-medium emission rates [3, 5] with hydrodynamic backgrounds obtained from Bayesian calibrations to hadronic data [4, 11]; no electromagnetic data are used to fit any parameter entering the central results. The prompt photon background is taken from INCNLO, an independent NLO pQCD calculation [6], and is validated at 200 GeV against STAR and PHENIX data before being applied at lower energies. The claimed high-pT thermal-over-prompt excess at low collision energies follows from the calculated steep fall of the prompt photon yield at large Bjorken-x, not from any quantity defined in terms of the thermal result itself. Similarly, the dilepton sensitivity to the pre-equilibrium stage and to hotspot size is obtained by comparing different initialization procedures and posterior parameter sets against NA60 data; these are forward calculations, not fits. The paper does cite prior work by overlapping authors for the model framework and Bayesian posteriors, but those citations are load-bearing only in the sense of providing the calibrated medium, which is independent input derived from hadronic observables. Concerns about the reliability of INCNLO at large Bjorken-x, while legitimate for correctness, are not circularity because the prompt calculation is an external input rather than a renamed version of the claimed prediction.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

No new entities are introduced. The numerical inputs are inherited from previous Bayesian analyses and selected modeling choices, as listed.

free parameters (5)

Posterior 2 particlization energy density (esw) = ~0.16 GeV/fm^3
Used for the Posterior 2 dilepton curve in Fig. 2 (right); the shift to low esw is credited with changing the IMR yield.
Initial-state hotspot size = 0.3 fm (Posterior 2) vs 0.1 fm (Posterior 1)
The paper states the hotspot size difference is the main reason for the IMR dilepton difference.
Pre-equilibrium initialization proper time (tau0) = 0.5 and 1.5 fm/c
Instantaneous initialization scenarios used in Fig. 2 (left); the default dynamical deposition interval is 0.5-1.5 fm/c.
QGP-to-hadronic emission rate switch temperature = 160 MeV
The EM rates switch from QCD-based to hadronic-based at T=160 MeV.
Hadronic emission cutoff temperature = 120 MeV
The hadronic phase ends at T=120 MeV, limiting the space-time volume for emission.

assumptions (4)

domain assumption Relativistic viscous hydrodynamics with iEBE-MUSIC describes the bulk evolution of heavy-ion collisions at finite baryon density.
The entire calculation assumes the hydrodynamic description with 3D dynamical initial conditions and UrQMD afterburner is valid at RHIC-BES and SPS energies.
domain assumption The thermal photon and dilepton emission rates from refs [3,5] are valid in the relevant temperature and baryon chemical potential domain.
Rates are imported from prior literature without revalidation at finite baryon density.
domain assumption NLO pQCD prompt photon production computed with INCNLO remains reliable at the lower collision energies and large Bjorken-x considered here.
The central high-pT thermal-to-prompt enhancement depends on the prompt photon yield falling steeply at large x; if this falloff is a pQCD artifact, the conclusion inverts.
domain assumption Bayesian posterior parameters calibrated to hadronic observables can be used to predict electromagnetic observables.
The paper relies on the transferability of hadronic calibration to EM emission without using EM data in the fit.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Electromagnetic radiation from Quark-Gluon Plasma at finite baryon density." pith.science (2026). https://pith.science/paper/JDBRMDFD

@misc{pith2026250903289,
  author       = {Pith},
  title        = {Pith review of: Electromagnetic radiation from Quark-Gluon Plasma at finite baryon density},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JDBRMDFD}},
  note         = {Machine review of arXiv:2509.03289}
}

read the original abstract

Using the Bayesian calibrated iEBE-MUSIC framework, we compute the production of electromagnetic radiation from hot hadronic matter at finite baryon density. Results for thermal photon and thermal dilepton yields are obtained by folding in-medium emission rates with posterior-sampled backgrounds evolved hydrodynamically. We consider different photon sources and analyze the collision-energy dependence of the thermal-to-prompt photon ratio. The sensitivity of the dilepton spectra to the pre-equilibrium stage is explored by considering different initialization procedures. Finally, we examine the impact on dilepton spectra of choosing parameter sets stemming from different Bayesian data analyses.

Figures

Figures reproduced from arXiv: 2509.03289 by the authors.

**Figure 1.** (Left panel): Direct photon yield as a function of transverse momentum, for 0–20% centrality class Au+Au collisions at 200 GeV. The total yield and its prompt and thermal components are shown. Data are from STAR [7] and PHENIX [8] Collaborations. The lower subpanel shows the ratio between data and the calculated total yield. (Right panel): Ratio of thermal to prompt photon yields as a function of transverse momentum… view at source ↗

**Figure 2.** Dilepton invariant mass excess spectra (data minus hadronic cocktail) in In+In collisions (dNch/dη > 30) from NA60 [10]. Left panel: Comparison between dynamical deposition and instantaneous initialization with τ0 = 0.5 and 1.5 fm/c. Right panel: Comparison between different posterior parameter sets. In the left panel of [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

13 extracted references · 8 linked inside Pith

[1]

Shen and B

C. Shen and B. Schenke, Phys. Rev. C 105, no.6, 064905 (2022) doi:10.1103/PhysRevC.105.064905 [arXiv:2203.04685 [nucl-th]]

arXiv 2022
[2]

G. S. Denicol, C. Gale, S. Jeon, A. Monnai, B. Schenke and C. Shen, Phys. Rev. C 98, no.3, 034916 (2018) doi:10.1103/PhysRevC.98.034916 [arXiv:1804.10557 [nucl-th]]

arXiv 2018
[3]

J. F. Paquet, C. Shen, G. S. Denicol, M. Luzum, B. Schenke, S. Jeon and C. Gale, Phys. Rev. C 93, no.4, 044906 (2016) doi:10.1103/PhysRevC.93.044906 [arXiv:1509.06738 [hep-ph]]

arXiv 2016
[4]

S. A. Jahan, H. Roch and C. Shen, Phys. Rev. C 110, no.5, 054905 (2024) doi:10.1103/PhysRevC.110.054905 [arXiv:2408.00537 [nucl-th]]

arXiv 2024
[5]

Churchill, L

J. Churchill, L. Du, C. Gale, G. Jackson and S. Jeon, Phys. Rev. C 109, no.4, 044915 (2024) doi:10.1103/PhysRevC.109.044915 [arXiv:2311.06675 [nucl-th]]

arXiv 2024
[6]

Aurenche, M

P. Aurenche, M. Fontannaz, J. P. Guillet, B. A. Kniehl, E. Pilon and M. Werlen, Eur. Phys. J. C 9, 107-119 (1999) doi:10.1007/s100529900018 [arXiv:hep-ph/9811382 [hep-ph]]

arXiv 1999
[7]

Adamczyk et al

L. Adamczyk et al. [STAR], Phys. Lett. B 770, 451-458 (2017) doi:10.1016/j.physletb.2017.04.050 [arXiv:1607.01447 [nucl-ex]]

arXiv 2017
[8]

Adare et al

A. Adare et al. [PHENIX], Phys. Rev. C 91, no.6, 064904 (2015) doi:10.1103/PhysRevC.91.064904 [arXiv:1405.3940 [nucl-ex]]

arXiv 2015

Show all 13 references

[9]

C. Shen, A. Noble, J. F. Paquet, B. Schenke and C. Gale, PoS HardProbes2023, 042 (2024) doi:10.22323/1.438.0042 [arXiv:2307.08498 [nucl-th]]

2024 arXiv
[10]

Arnaldi et al

R. Arnaldi et al. [NA60], Eur. Phys. J. C 59, 607-623 (2009) doi:10.1140/epjc/s10052-008-0857-2 [arXiv:0810.3204 [nucl-ex]]

2009 arXiv
[11]

S. A. Jahan, H. Roch and C. Shen, [arXiv:2507.11394 [nucl-th]]
[12]

, " * write output.state after.block = add.period write newline

ENTRY address archive author booktitle chapter collaboration edition editor eid eprint howpublished institution isbn issn journal key month note number numpages organization pages publisher school series title type url doi volume year label extra.label sort.label short.list IN...
[13]

write FUNCTION output.bibitem newline

" write FUNCTION output.bibitem newline " " " " * write cite write " " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" *...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Shen and B

C. Shen and B. Schenke, Phys. Rev. C 105, no.6, 064905 (2022) doi:10.1103/PhysRevC.105.064905 [arXiv:2203.04685 [nucl-th]]

arXiv 2022

[2] [2]

G. S. Denicol, C. Gale, S. Jeon, A. Monnai, B. Schenke and C. Shen, Phys. Rev. C 98, no.3, 034916 (2018) doi:10.1103/PhysRevC.98.034916 [arXiv:1804.10557 [nucl-th]]

arXiv 2018

[3] [3]

J. F. Paquet, C. Shen, G. S. Denicol, M. Luzum, B. Schenke, S. Jeon and C. Gale, Phys. Rev. C 93, no.4, 044906 (2016) doi:10.1103/PhysRevC.93.044906 [arXiv:1509.06738 [hep-ph]]

arXiv 2016

[4] [4]

S. A. Jahan, H. Roch and C. Shen, Phys. Rev. C 110, no.5, 054905 (2024) doi:10.1103/PhysRevC.110.054905 [arXiv:2408.00537 [nucl-th]]

arXiv 2024

[5] [5]

Churchill, L

J. Churchill, L. Du, C. Gale, G. Jackson and S. Jeon, Phys. Rev. C 109, no.4, 044915 (2024) doi:10.1103/PhysRevC.109.044915 [arXiv:2311.06675 [nucl-th]]

arXiv 2024

[6] [6]

Aurenche, M

P. Aurenche, M. Fontannaz, J. P. Guillet, B. A. Kniehl, E. Pilon and M. Werlen, Eur. Phys. J. C 9, 107-119 (1999) doi:10.1007/s100529900018 [arXiv:hep-ph/9811382 [hep-ph]]

arXiv 1999

[7] [7]

Adamczyk et al

L. Adamczyk et al. [STAR], Phys. Lett. B 770, 451-458 (2017) doi:10.1016/j.physletb.2017.04.050 [arXiv:1607.01447 [nucl-ex]]

arXiv 2017

[8] [8]

Adare et al

A. Adare et al. [PHENIX], Phys. Rev. C 91, no.6, 064904 (2015) doi:10.1103/PhysRevC.91.064904 [arXiv:1405.3940 [nucl-ex]]

arXiv 2015