REVIEW 3 major objections 5 minor 37 references
Charmonium pair production in ultraperipheral collision
T0 review · 3 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash
Pith's one-line read This paper predicts that exclusive J/ψ-pair and ηc-pair production in ultraperipheral Pb-Pb collisions has measurable rates at NLO in NRQCD, with opposite-sign NLO corrections for the two channels.
desk verdict A legitimate but modest extension of an existing NLO NRQCD calculation to UPC; the central cross sections are not robust because the NLO expansion itself signals a breakdown at the chosen scale. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central machinery is NRQCD (non-relativistic QCD) factorization at next-to-leading order in QCD, applied to the color-singlet γγ→H+H amplitudes using covariant spin and color projectors, combined with the equivalent-photon approximation for ultraperipheral collisions. The NLO amplitudes are regularized dimensionally, with on-shell renormalization for the heavy quark field and mass and MS renormalization for the strong coupling. The observable UPC cross section is obtained by convolving the NLO parton-level cross section with the ion photon spectral function for heavy ions, and this factorization is what carries the argument from a QCD calculation to measurable rates.
What would settle it
Measure the rapidity-difference distribution of exclusive J/ψ-pair production in Pb-Pb UPC at the HL-LHC; if the bins with |Δy|>2 show positive rates near the leading-order size, or if the total rate differs from 28 nb by more than the quoted uncertainties, the NLO NRQCD prediction is falsified. The ηc-pair rate of 65.1 nb offers an independent check.
Extended reading notes
Core claim
The paper's central claim is that at next-to-leading order in QCD, the photon-photon fusion cross sections for exclusive J/ψ-J/ψ and ηc-ηc production in ultraperipheral Pb-Pb collisions are 28.0 nb and 65.1 nb respectively, and that the NLO corrections have opposite signs: large negative for J/ψ pairs and positive for ηc pairs. These predictions follow from NRQCD factorization with color-singlet matrix elements, and the paper argues that UPC event topologies suppress QCD backgrounds enough that the rates are measurable at the HL-LHC and FCC. The J/ψ-pair K-factor is about 0.25 and the ηc-pair about 2.35, and differential distributions in pT, mγγ, and |Δy| are provided. At the default scale, the J/ψ-pair differential cross section becomes negative for |Δy|>2, which the authors attribute to large negative loop corrections and suggest may be cured by scale resetting or resummation.
Load-bearing premise
The central numbers depend on the assumption that the next-to-leading-order QCD correction is a small, trustworthy correction; the paper's own J/ψ-pair distribution turning negative at large rapidity differences is a warning that this may not hold.
Editorial extensions
If this is right
- At the HL-LHC, the predicted yields are 140-194 J/ψ-pair events and 325-456 ηc-pair events per run from heavy-ion UPCs, with 835 and 1980 from p-p UPCs, before decay branching ratios.
- Using the branching fractions quoted in the paper, J/ψ→l+l− at 12% and ηc→K\bar Kπ at 7.3%, reconstructed candidates would be 2-12 per year for J/ψ pairs and 1-10 for ηc pairs at the HL-LHC; at the FCC the J/ψ-pair yield grows to 180-200.
- The opposite signs of the NLO K-factors, about 0.25 for J/ψ pairs and 2.35 for ηc pairs, mean the two channels respond very differently to higher-order QCD, so a simultaneous measurement would be a sensitive NRQCD test.
- The differential cross sections in pT, mγγ, and |Δy| are given without cuts, providing specific shapes that UPC experiments can compare directly with data.
- The X(6900) production cross section via two-photon fusion is estimated at 6×10^3 to 2×10^5 nb depending on the two-photon width model, implying a potentially large di-charmonium signal if the state decays predominantly into charmonium pairs.
Reading between the lines
- Not stated in the paper: if the negative NLO differential cross section at |Δy|>2 reflects a genuine breakdown of fixed-order perturbation theory rather than a scale artifact, the 28.0 nb total may be unreliable; a resummed or next-to-next-to-leading-order calculation would settle this.
- A test that follows from the sign asymmetry but is not proposed by the authors: measuring the ratio σ(ηc-ηc)/σ(J/ψ-J/ψ) as a function of pT would isolate the spin dependence of the NLO corrections and could expose missing relativistic or color-octet effects.
- The X(6900) estimates span two orders of magnitude across the three two-photon-width models; a UPC measurement of the di-J/ψ invariant mass near 6.9 GeV would effectively measure Γ(X→γγ) and discriminate among those models, going beyond the paper's tabulated estimates.
- The same NLO machinery could be applied to other exclusive channels, such as J/ψ plus ψ(2S), to test whether the opposite-sign correction pattern is specific to the spin-triplet versus spin-singlet ground states.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper computes next-to-leading-order (NLO) QCD corrections, in the NRQCD color-singlet framework, to exclusive γγ→J/ψJ/ψ and γγ→ηcηc production in ultraperipheral collisions, using the equivalent-photon approximation. It reports total cross sections of 28.0 nb for J/ψ-pair and 65.1 nb for ηc-pair production in Pb-Pb UPC at √sNN=5.52 TeV, along with differential distributions in transverse momentum, diphoton invariant mass, and rapidity difference, and estimates of event rates at the HL-LHC and FCC. The paper also estimates two-photon production of the fully charmed tetraquark X(6900) using the Low formula with several model assumptions for its diphoton width.
Significance. If the central NLO predictions were robust, this study would offer a clean test of NRQCD factorization in an exclusive two-photon process, with color-octet contributions strongly suppressed and UPC event topologies providing large rapidity gaps. The paper usefully extends the authors' earlier NLO calculation to heavy-ion UPCs, provides explicit flux-convolution formulas, and tabulates cross sections for many collision systems. The X(6900) discussion, though speculative, points to a possible two-photon production window for fully charmed tetraquarks. However, as detailed in the major comments, the scale dependence of the NLO series and the occurrence of negative differential cross sections at the chosen central scale mean that the headline numbers are not yet reliable as quantitative predictions.
major comments (3)
- [§III, Table I and Table III] The central NLO predictions are not robust because the central scale choice is both ambiguous for integrated observables and strongly scale-dependent. For the J/ψ pair, the NLO total cross section varies from 11.0 nb at μ=2mc to 83.8 nb at μ=√ŝγγ (Table I), a factor of about 7.6, while the quoted charm-mass uncertainty on the central value is only +18.4/−11.1 nb. At the central scale μ=√(4mc^2+p_T^2), the NLO correction is approximately −75% of the LO result (28.0 nb versus 111 nb), and Table III shows that the NLO differential cross section dσ/d|Δy| becomes negative for |Δy|>2, e.g., −0.47 nb in the 2.5–3 bin. The text acknowledges this and defers to scale resetting or resummation, but no such improved calculation is provided. The paper also never defines which p_T value is used when this scale is applied to the total cross section, since p_T is integrated over. These issues directly affect the abstract's headline value of 28.0 nb, so they are load-bearing rather than cosmetic.
- [§III, Table I and Abstract] The qualitative claim that NLO corrections are large and negative for J/ψ pairs and positive for ηc pairs is not scale-invariant. Table I shows that for J/ψ pairs the NLO correction is negative at μ=2mc (11.0 nb versus 120 nb LO) but positive at μ=√ŝγγ (83.8 nb versus 62.6 nb LO). The K-factors are approximately 0.25 for J/ψ and 2.35 for ηc, both far from unity, indicating that the fixed-order expansion is not a small perturbation. Therefore, the sign and magnitude of the NLO corrections, and hence the claimed qualitative NRQCD test, depend strongly on an arbitrary renormalization-scale choice. The abstract and conclusions should be revised to present these results as scale-sensitive fixed-order estimates rather than definitive predictions.
- [§III, Table I and text after Table III] The quoted theoretical uncertainties are incomplete: Table I gives only the variation with mc at fixed renormalization scale, while the scale dependence itself is an order of magnitude larger than the mass uncertainty. For a phenomenological claim based on total cross sections, the scale variation should be included as an uncertainty band, and the negative bins in Table III should be addressed by a concrete prescription rather than a statement that the problem 'could be cured' by resetting the scale. Without such a treatment, the event-rate projections derived from the central values (e.g., 140–194 J/ψ-pair events in heavy-ion UPCs) inherit an uncontrolled systematic uncertainty.
minor comments (5)
- [§III, Eq. (7) and Eq. (8)] Please clarify how the NLO-extracted radial wave function |R_NLO_J/ψ(0)|^2=0.907 GeV^3 is obtained from Eq. (8) with the two-loop running coupling, and state explicitly which value of μ0 is used in the numerical extraction beyond the statement μ0=2mc.
- [§III, Fig. 2, Fig. 3, Fig. 4] In the submitted version many axis labels and panel annotations in Figures 2–4 are unreadable. Please ensure the final figures have clearly legible labels, legends, and units.
- [§III, Table III] Table III is titled as the rapidity-difference distribution of the J/ψ pair, but the text and Figure 4 suggest the same distribution is shown for ηc pairs as well. Please clarify whether Table III applies to both channels or only to J/ψ pairs.
- [§III, Table IV] The X(6900) cross-section estimates in Table IV are orders of magnitude larger than the direct double-charmonium yields, and the text itself states this may reflect overestimated diphoton and di-charmonium decay fractions. Please label these values as model-dependent upper-bound-like estimates or remove the quantitative comparison, since the current presentation invites an unjustified quantitative interpretation.
- [§III, feasibility discussion] The feasibility conclusion is based on reconstructed event counts of 2–12 per year after branching ratios, but no background estimate or detection-efficiency assumption is given. A quantitative background assessment, even a rough one, is needed to support the claim that these channels are experimentally accessible.
Circularity Check
No significant circularity: the NLO gamma-gamma to charmonium-pair cross sections come from a previous independent calculation and are convolved with external photon fluxes, while the wave-function inputs are taken from the measured leptonic width and spin symmetry, not from the target UPC cross sections.
full rationale
The derivation chain is self-contained and not circular. The total UPC cross section in Eq. (1) is a convolution of the equivalent-photon fluxes n_i(x) with the parton-level cross section sigma_hat(gamma gamma -> H H). The latter is taken from the authors' previous NLO calculation [11], which is a separate published computation of gamma gamma -> J/psi J/psi and does not use any UPC cross section or any of the 28.0 nb / 65.1 nb results as input. The wave-function input |R_{J/psi}(0)|^2 is extracted in Eqs. (7)-(8) from the measured leptonic width Gamma(J/psi -> e+e-) = 5.55 keV, an external experimental quantity, and R_eta_c(0) is set equal to R_{J/psi}(0) by the standard heavy-quark spin symmetry relation; neither is fitted to the target processes. The scale dependence shown in Table I and the negative NLO differential cross sections at large |Delta y| in Table III are genuine limitations of the fixed-order truncation, but the paper reports them explicitly and does not hide them behind a circular argument. The X(6900) estimates use the Low formula with external decay-width estimates from VMD, NRA, and chi_c0 approximations, again independent of the double-charmonium predictions. The only self-citation that is load-bearing, Ref. [11], is an independent published calculation with stated assumptions that do not include the present target observables, so it does not constitute circularity under the stated criteria.
Assumptions & free parameters
free parameters (5)
- charm quark mass m_c =
1.5 +/- 0.1 GeV
- J/psi radial wave function at the origin |R_J/psi(0)|^2 =
LO: 0.528 GeV^3; NLO: 0.907 GeV^3
- minimum impact parameter b_min =
R_A from 7.1 fm (Pb) to 0.7 fm (proton)
- renormalization scale mu =
2 m_c, sqrt(4 m_c^2 + p_T^2), or sqrt(s_hat)
- X(6900) two-photon width Gamma(X(6900) to gamma gamma) =
67 keV (VMD), 10 keV (NRA), 2 keV (chi_c0 approximation)
assumptions (6)
- domain assumption NRQCD factorization: double charmonium production separates into a perturbative short-distance coefficient and long-distance matrix elements.
- domain assumption Color-singlet contributions dominate; color-octet contributions are suppressed by v^8 and neglected.
- domain assumption Heavy quark spin symmetry gives R_eta_c(0) = R_J/psi(0) at leading order in the relative velocity expansion.
- domain assumption The equivalent photon approximation with the photon flux of Eq. 2 and b_min = R_A describes UPC as purely electromagnetic photon-photon fusion.
- ad hoc to paper The truncated NLO QCD series gives physical cross sections at the chosen scale, with negative bins treated as a scale artifact.
- domain assumption The Low formula Eq. 10 relates the X(6900) photon-fusion cross section to its two-photon width and the effective photon luminosity.
Cite this review
Pith. "Pith review of Charmonium pair production in ultraperipheral collision." pith.science (2026). https://pith.science/paper/O7REIUZM
@misc{pith2026250414850,
author = {Pith},
title = {Pith review of: Charmonium pair production in ultraperipheral collision},
year = {2026},
howpublished = {\url{https://pith.science/paper/O7REIUZM}},
note = {Machine review of arXiv:2504.14850}
}
abstract
We study the exclusive double charmonium ($J/\psi \mbox{-} J/\psi$ and $\eta_c \mbox{-} \eta_c$) production through photon-photon fusion via ultraperipheral collision (UPC) at the HL-LHC and FCC with next-to-leading order (NLO) QCD predictions in the framework of non-relativistic QCD (NRQCD). Numerical results indicate that the NLO corrections for $J/\psi$ pair are large and negative, while positive for $\eta_c$ pair. The total cross section of $J/\psi \mbox{-} J/\psi$ ($\eta_c \mbox{-} \eta_c$) in Pb-Pb UPC is 28.0 (65.1) nb at nucleon-nucleon c.m. energy $\sqrt{s_{NN}} = 5.52$ TeV. Due to the backgrounds from various QCD interactions at UPC are highly suppressed and the event topologies for charmonium pair are easy to tag, the phenomenological studies at the LHC and FCC are feasible. The detailed transverse momentum $p_T$, diphoton invariant mass $m_{\gamma\gamma}$ and the rapidity difference $\Delta y$ distributions are given. The production for X(6900) is also discussed.
Figures
Reference graph
Works this paper leans on
-
[1]
G. T. Bodwin, E. Braaten and G. P. Lepage, Phys. Rev. D 51, 1125-1171 (1995) [erratum: Phys. Rev. D 55, 5853 (1997)] doi:10.1103/PhysRevD.55.5853 [arXiv:hep- ph/9407339 [hep-ph]]
arXiv 1995
-
[2]
J. Abdallah et al. [DELPHI], Phys. Lett. B 565, 76-86 (2003) doi:10.1016/S0370- 2693(03)00660-9 [arXiv:hep-ex/0307049 [hep-ex]]
arXiv 2003
-
[3]
J. P. Ma, B. H. J. McKellar and C. B. Paranavitane, Phys. Re v. D 57, 606-609 (1998) doi:10.1103/PhysRevD.57.606 [arXiv:hep-ph/9707480 [he p-ph]]
work page Pith review arXiv 1998
-
[4]
Color Octet Contribution to $J/\psi$ Production at a Photon Linear Collider
G. Japaridze and A. Tkabladze, Phys. Lett. B 433, 139-146 (1998) doi:10.1016/S0370- 2693(98)00697-2 [arXiv:hep-ph/9803447 [hep-ph]]
work page Pith review arXiv 1998
- [5]
- [6]
-
[7]
C. F. Qiao and J. X. Wang, Phys. Rev. D 69, 014015 (2004) doi:10.1103/PhysRevD.69.014015 [arXiv:hep-ph/0308244 [hep-ph]]
arXiv 2004
- [8]
Show all 37 references
-
[9]
Z. Q. Chen, L. B. Chen and C. F. Qiao, Phys. Rev. D 95, no.3, 036001 (2017) doi:10.1103/PhysRevD.95.036001 [arXiv:1608.06231 [hep -ph]]
2017 arXiv
-
[10]
Butenschoen and B
M. Butenschoen and B. A. Kniehl, Phys. Rev. D 84, 051501 (2011) doi:10.1103/PhysRevD.84.051501 [arXiv:1105.0820 [hep- ph]]
2011 arXiv
-
[11]
H. Yang, Z. Q. Chen and C. F. Qiao, Eur. Phys. J. C 80, no.9, 806 (2020) doi:10.1140/epjc/s10052-020-8390-z
2020 doi
-
[12]
C. F. von Weizsacker, Z. Phys. 88, 612-625 (1934) doi:10.1007/BF01333110
1934 doi
-
[13]
E. J. Williams, Phys. Rev. 45, 729-730 (1934) doi:10.1103/PhysRev.45.729
1934 doi
-
[14]
R. N. Cahn and J. D. Jackson, Phys. Rev. D 42, 3690-3695 (1990) doi:10.1103/PhysRevD.42.3690
1990 doi
-
[15]
G. T. Bodwin and A. Petrelli, Phys. Rev. D 66, 094011 (2002) [erratum: Phys. Rev. D 87, no.3, 039902 (2013)] doi:10.1103/PhysRevD.66.094011 [ar Xiv:hep-ph/0205210 [hep-ph]]
2002 arXiv
-
[16]
Beneke and V
M. Beneke and V. A. Smirnov, Nucl. Phys. B 522, 321-344 (1998) doi:10.1016/S0550- 3213(98)00138-2 [arXiv:hep-ph/9711391 [hep-ph]]
1998 arXiv
-
[17]
Navas et al
S. Navas et al. [Particle Data Group], Phys. Rev. D 110, no.3, 030001 (2024) doi:10.1103/PhysRevD.110.030001
2024 doi
-
[18]
Bruce, D
R. Bruce, D. d’Enterria, A. de Roeck, M. Drewes, G. R. Far rar, A. Giammanco, O. Gould, J. Hajer, L. Harland-Lang and J. Heisig, et al. J. Phys. G 47, no.6, 060501 (2020) doi:10.1088/1361-6471/ab7ff7 [arXiv:1812.07688 [hep-ph] ]
2020 arXiv
-
[19]
d’Enterria, M
D. d’Enterria, M. Drewes, A. Giammanco, J. Hajer, E. Bra tkovskaya, R. Bruce, N. Bur- masov, M. Dyndal, O. Gould and I. Grabowska-Bold, et al. J. Phys. G 50, no.5, 050501 (2023) doi:10.1088/1361-6471/acc197 [arXiv:2203.05939 [hep-ph]]
2023 arXiv
-
[20]
Dainese, U
A. Dainese, U. A. Wiedemann, N. Armesto, D. d’Enterria, J. M. Jowett, J. P. Lans- berg, J. G. Milhano, C. A. Salgado, M. Schaumann and M. van Lee uwen, et al. doi:10.23731/CYRM-2017-003.635 [arXiv:1605.01389 [hep -ph]]
2017 arXiv
-
[21]
Abada et al
A. Abada et al. [FCC], Eur. Phys. J. ST 228, no.4, 755-1107 (2019) doi:10.1140/epjst/e2019-900087-0 14
2019 doi
-
[22]
C. F. Qiao, Phys. Rev. D 64, 077503 (2001) doi:10.1103/PhysRevD.64.077503 [arXiv:hep-ph/0104309 [hep-ph]]
2001 arXiv
-
[23]
Z. G. He, X. B. Jin, B. A. Kniehl and R. Li, Chin. Phys. C 48, no.8, 083107 (2024) doi:10.1088/1674-1137/ad408f [arXiv:2404.08945 [hep-p h]]
2024 arXiv
-
[24]
J. P. Lansberg, PoS ICHEP2010, 206 (2010) doi:10.22323/1.120.0206 [arXiv:1012.2815 [hep-ph]]
2010 arXiv
-
[25]
Y. Feng, J. P. Lansberg and J. X. Wang, Eur. Phys. J. C 75, no.7, 313 (2015) doi:10.1140/epjc/s10052-015-3527-1 [arXiv:1504.00317 [hep-ph]]
2015 arXiv
-
[26]
Colpani Serri, Y
A. Colpani Serri, Y. Feng, C. Flore, J. P. Lansberg, M. A. Ozcelik, H. S. Shao and Y. Yedelkina, Phys. Lett. B 835, 137556 (2022) doi:10.1016/j.physletb.2022.137556 [arXiv:2112.05060 [hep-ph]]
2022
-
[27]
7+27. 0 − 13. 0 ) 48.7 (16.0) TABLE II: The nucleon-nucleon (NN) c.m. energy √sNN , effective charge radius RA [32], total LO (up) and NLO (down) cross sections for J/ψ -J/ψ , ηc-ηc and integrated luminosity per typical run Lint for ultraperipheral collisions at HL-LHC and FCC....
1980
-
[28]
Z. Q. Chen, L. B. Chen and C. F. Qiao, Phys. Rev. D 109, no.9, 096032 (2024) doi:10.1103/PhysRevD.109.096032 [arXiv:2402.05397 [he p-ph]]
2024 arXiv
-
[29]
Aaij et al
R. Aaij et al. [LHCb], Sci. Bull. 65, no.23, 1983-1993 (2020) doi:10.1016/j.scib.2020.08.03 2 [arXiv:2006.16957 [hep-ex]]
2020
-
[30]
Hayrapetyan et al
A. Hayrapetyan et al. [CMS], Phys. Rev. Lett. 132, no.11, 111901 (2024) doi:10.1103/PhysRevLett.132.111901 [arXiv:2306.07164 [hep-ex]]
2024 arXiv
-
[31]
F. E. Low, Phys. Rev. 120, 582-583 (1960) doi:10.1103/PhysRev.120.582
1960 doi
-
[32]
V. M. Budnev, I. F. Ginzburg, G. V. Meledin and V. G. Serbo , Phys. Rept. 15, 181-281 (1975) doi:10.1016/0370-1573(75)90009-5
1975 doi
-
[33]
H. S. Shao and D. d’Enterria, JHEP 09, 248 (2022) doi:10.1007/JHEP09(2022)248 [arXiv:2207.03012 [hep-ph]]
2022 arXiv
-
[34]
Biloshytskyi, V
V. Biloshytskyi, V. Pascalutsa, L. Harland-Lang, B. Ma laescu, K. Schmieden and M. Schott, Phys. Rev. D 106, no.11, L111902 (2022) doi:10.1103/PhysRevD.106.L11190 2 [arXiv:2207.13623 [hep-ph]]
2022 arXiv
-
[35]
Fariello, D
R. Fariello, D. Bhandari, C. A. Bertulani and F. S. Navar ra, Phys. Rev. C 108, no.4, 044901 (2023) doi:10.1103/PhysRevC.108.044901 [arXiv:2 306.10642 [hep-ph]]
2023 doi
-
[36]
Biloshytskyi, L
V. Biloshytskyi, L. Harland-Lang, B. Malaescu, V. Pasc alutsa, K. Schmieden and 15 M. Schott, EPJ Web Conf. 274, 06007 (2022) doi:10.1051/epjconf/202227406007 [arXiv:2211.10266 [hep-ph]]
2022
-
[37]
V. P. Gon¸ calves and B. D. Moreira, Phys. Lett. B 816, 136249 (2021) doi:10.1016/j.physletb.2021.136249 [arXiv:2101.03798 [hep-ph]]. 16
2021
Reviewed August 16, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.