{"id":"40102a2b-4eef-49c1-8f12-883155ad445e","arxiv_id":"2501.15575","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":10,"one_line_summary":"Color-octet channels are predicted to dominate many double heavy quarkonium production processes at future Z factories, with relativistic corrections suppressing charmonium rates by roughly a factor of two.","lead":"This paper calculates how often pairs of heavy quark-antiquark bound states (charmonium and bottomonium) are produced in electron-positron collisions at the Z boson mass, adding the color-octet mechanism that earlier calculations left out. The predicted rates suggest future Z factories such as CEPC and FCC-ee could test a long-standing puzzle in quantum chromodynamics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"CO-dominance claim depends linearly on CO LDMEs whose central ^3S1[8] values carry 100% fractional errors and come from a photoproduction fit with a known universality problem; Sec. IV.C explicitly declines to propagate this uncertainty, so the paper's central claim lacks a robustness test.","rationale":"The paper is a competent LO NRQCD calculation with standard v^2 relativistic corrections, and its CS results broadly agree with earlier work. The genuinely new physics claim is that CO channels are significant or dominant at the Z factory and that the ^3S1[8] fragmentation channel is the most important one. That claim is not an intrinsic property of the hard scattering: it is the product of CO short-distance coefficients and CO LDMEs, and the chosen LDMEs carry very large uncertainties. The paper quantifies uncertainties from quark mass, renormalization scale, and collision energy, but deliberately omits the one source of uncertainty most directly tied to its headline conclusion, stating 'we won't discuss the LDMEs uncertainty.' Because the quoted charmonium ^3S1[8] LDME has a nominal 100% error and comes from a photoproduction fit whose universality is disputed in the paper's own introduction, the CO-dominance conclusion is not yet robust. The event-number inconsistency between Section IV.B, Table III, and the abstract is real and should be corrected, but it is a reporting issue rather than the physical core of the claim. For these reasons, the reader's conditional acceptance remains appropriate, with the requested changes: propagate LDME uncertainties, unify event reporting, and clarify the status of the CO predictions.","tokens_in":34960,"tokens_out":9143,"duration_ms":81764,"concrete_test":"Recompute the cross sections in Tables I and II at √s = mZ after replacing the charmonium and bottomonium ^3S1[8] LDMEs with their 1σ lower bounds (0 and 0.0143 GeV^3, respectively) while keeping all other inputs fixed, and separately repeat with a hadroproduction-extracted LDME set cited in Sec. I (e.g., Refs. [6-8]). If the CO fractions for J/ψ+η_c, J/ψ+χ_{c1}, Υ+χ_{b1}, and Υ+χ_{b2} remain above 50% in both variants, the central claim is robust; if they fall below 50%, the abstract's 'significant or dominant' assertion is not supported by the chosen LDME input.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is a quantitative statement about CO LDMEs, not just NRQCD kinematics. The only charmonium CO LDME set used (Eq. (26)) has <O^{J/ψ}(3S1[8])> = 0.0013 ± 0.0013 GeV^3, a 100% uncertainty whose lower endpoint is zero, and the bottomonium ^3S1[8] entry is 0.0477 ± 0.0334 GeV^3; these values come from a photoproduction fit (Ref. [52]) that the paper itself lists among the inconsistent extractions in Sec. I. Since the CO cross section is linear in these LDMEs, the claimed dominance (e.g., 66.1% for J/ψ+η_c and 92.4% for Υ+χ_{b1} in Table I) would shrink or vanish if a hadroproduction-fit set, or the 1σ lower bounds, were used. Section IV.C explicitly says 'we won't discuss the LDMEs uncertainty,' and the only robustness statement is that one CS potential-model set leaves 'COM still significant,' which does not vary the CO LDMEs. The paper therefore has no demonstrated sensitivity interval for its central conclusion. A separate internal inconsistency in the event numbers (Section IV.B gives (52, 4382, 117, 86) while Table III and the abstract give (22, 570, 71, 61) for the same CEPC channels) compounds the problem, but the LDME dependence is the load-bearing issue.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies exclusive production of double heavy quarkonia (double charmonia and double bottomonia) in e+e- annihilation at the Z pole and at future CEPC/FCC-ee, working in the NRQCD factorization framework. The authors include color-singlet and color-octet channels at tree level, add relativistic O(v^2) corrections, and adopt NLO alpha_s K factors for the CS channels from the literature. Their central claim is that color-octet contributions, especially those mediated by gluon fragmentation into the ^3S1[8] state, are significant or dominant for many double quarkonium channels at Z-factory energies, and they provide event-rate estimates for CEPC and FCC-ee. The paper also presents cross sections as functions of center-of-mass energy, CO/CS ratios, differential distributions, and several uncertainty studies.","tokens_in":35368,"tokens_out":5170,"duration_ms":48870,"significance":"If the central claim is robust, the paper identifies new channels in which future Z-factory measurements could discriminate between color-singlet and color-octet mechanisms and could constrain the ^3S1[8] LDMEs. The work is useful in scope: it covers a broad set of double charmonium and bottomonium final states, includes relativistic corrections, and compares the CS part of the calculation with existing results in Refs. [12, 36, 40, 43, 74]. The main weakness is that the headline conclusion of CO dominance depends linearly on CO LDMEs whose uncertainties are not propagated; Section IV.C explicitly declines to discuss LDME uncertainties. The reader's stress-test concern about this point therefore lands. The paper is not circular in the fitting sense, since no observable is fitted and the LDMEs are taken from prior extractions, but the central quantitative claim is parameter-sensitive and lacks a demonstrated robustness interval.","major_comments":[{"comment":"The central claim that CO contributions are 'significant or dominant' (abstract and Table I) is not robust because the adopted CO LDMEs are varied at all. In particular, Eq. (26) gives <O^{J/psi}(3S1[8])> = 0.0013 +/- 0.0013 GeV^3, whose 1-sigma lower endpoint is zero, and the bottomonium value <O^{Upsilon}(3S1[8])> = 0.0477 +/- 0.0334 GeV^3 has a similarly large fractional error. Since the CO cross sections are linear in these LDMEs, the CO fractions quoted in Table I (for example 66.1% for J/psi+eta_c and 92.4% for Upsilon+chi_b1) would be drastically reduced or vanish at the lower endpoints. Section IV.C states 'we won't discuss the LDMEs uncertainty,' and the only robustness check varies the CS potential-model set, not the CO LDMEs. Please propagate the LDME uncertainties or, at minimum, show the cross sections obtained with an alternative hadroproduction-fit LDME set; without this the CO-dominance conclusion is not established.","section":"Sec. III, Eq. (26), and Sec. IV.C"},{"comment":"The event numbers are internally inconsistent. The text in Section IV.B says that the total cross sections at O(v0) are (32.7, 2738.9, 73.1, 53.8) x 10^-4 fb and that 'the final events would be (52, 4382, 117, 86) and (491, 41083, 1096, 806)' for CEPC and FCC-ee, respectively. Table III, by contrast, reports NLO(v2) events (22, 570, 71, 61) for CEPC and (206, 5343, 665, 576) for FCC-ee for the same four channels, and the abstract quotes the latter numbers. The reader cannot tell which set is the actual prediction. Please reconcile the text with Table III and the abstract and state explicitly which perturbative order is used for the final event counts.","section":"Sec. IV.B, Table III, and abstract"},{"comment":"The NLO alpha_s K factors from Ref. [40] are applied to the CS cross sections, whereas the CO cross sections are kept at tree level in alpha_s. Since the adopted CS K factors are large (for example 3.75 and 3.9 for J/psi+eta_c and J/psi+J/psi), the CO channels could receive comparably large NLO QCD corrections, and the hierarchy between CO and CS may change. The paper does state that the CO channels are treated at tree level, but given that the CO-dominance conclusion is the paper's main message, please add a discussion of the expected size of NLO corrections to the CO channels, or at least an explicit caveat that the CO predictions are leading-order in alpha_s.","section":"Sec. IV.B"}],"minor_comments":[{"comment":"The text refers to 'Appendix VI' and 'Appendix VII,' but the appendices are labeled 'APPENDIX. A' and 'APPENDIX. B'; please align the cross-references.","section":"Sec. IV.A and Fig. 11 caption"},{"comment":"The K factors quoted in Section IV.B (3.75, 3.9, 2.55, 2.5 and 1.08, 1.01, 0.775, 0.908) are presented without the corresponding m_c or m_b values; please state the quark masses used when applying the Ref. [40] results so the reader can reproduce the numbers.","section":"Sec. IV.B"},{"comment":"References [32] and [34] are the same paper (Erler et al., 'Physics impact of GigaZ') and should be merged or renumbered.","section":"References"},{"comment":"The R+ and R- ratios in Tables IV and V are computed at E_cm = 97% and 103% of m_Z, but the text says '15% to 20% of its peak values,' which is only true for some channels; please state that the reduction is channel-dependent.","section":"Sec. IV.C and Tables IV/V"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for a heavy-quarkonium/QCD journal and contains a useful set of predictions. The main blocker is the lack of any sensitivity test for the CO LDMEs on which the central claim rests, together with the conflicting event numbers in Section IV.B versus Table III. Both issues are fixable within the manuscript's scope, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is a legitimate NRQCD calculation that adds color-octet channels and v^2 corrections to double quarkonium production at the Z pole, giving cross sections, differential distributions, and event estimates for CEPC and FCC-ee. That is a real extension of the LO color-singlet results in Refs. [36,40,43]. The LO CS numbers mostly reproduce prior work, and the footnote flagging discrepancies with Ref. [43] is honest and useful.\n\nThe soft spots, in order of seriousness. First, the CO-dominance claim is not robustly quantified. The CO cross sections are linear in the CO LDMEs, and the charmonium ^3S1[8] value adopted has a 100% fractional error with a lower endpoint of zero, from a photoproduction fit that the paper's own Introduction lists as part of the universality problem. Section IV.C explicitly says the LDME uncertainty is not discussed. The only robustness check varies CS LDMEs, not CO LDMEs, so the paper never shows that the claimed CO dominance, e.g. 66% for J/psi+eta_c and 92% for Upsilon+chi_b1, survives a plausible change in the CO input. That is the load-bearing issue.\n\nSecond, the event numbers are internally inconsistent. Section IV.B gives (52, 4382, 117, 86) for CEPC, while Table III and the abstract give (22, 570, 71, 61) for the same channels. The difference seems to be that IV.B folds in NLO alpha_s K factors from Ref. [40] while Table III does not, but the paper never reconciles the two, and the abstract quotes the smaller set without explanation. This has to be fixed before publication. Third, the CO channels are left at LO in alpha_s while the CS channels get NLO alpha_s K factors. The authors acknowledge loop corrections are future work, but it means the channels they highlight are the least well-corrected.\n\nWho it is for: NRQCD phenomenologists planning quarkonium measurements at CEPC/FCC-ee. It deserves a serious referee because it is a new calculation with useful tables and figures. My recommendation: send it to peer review with major revision, requiring the authors to quantify the LDME sensitivity of the central claim, reconcile the event numbers, and state clearly that the CO predictions are LO in alpha_s.","headline":"A competent NRQCD calculation of color-octet double quarkonium production at the Z pole, but the central CO-dominance claim lacks a robustness test against LDME uncertainty and the event numbers are reported inconsistently.","tokens_in":35932,"tokens_out":6470,"would_cite":true,"duration_ms":52640,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper shows that color-octet channels can dominate exclusive double heavy quarkonium production at Z-factory energies, making future electron-positron colliders direct probes of the color-octet mechanism.","keywords":["double heavy quarkonium","color-octet mechanism","NRQCD","Z factory","CEPC","FCC-ee","relativistic corrections","gluon fragmentation"],"falsifier":"Measure $e^+e^- \\to J/\\psi+\\eta_c$ at $\\sqrt{s}=m_Z$ with 16 ab$^{-1}$ (CEPC) or 150 ab$^{-1}$ (FCC-ee). The color-singlet-only prediction is about 10 or 93 events after $v^2$ and $\\alpha_s$ corrections, while the full color-singlet plus octet prediction is 22 or 206 events. A measured rate consistent with the lower number would rule out the adopted octet matrix elements; a rate near the higher number would support them.","tokens_in":34726,"feed_emoji":"⚛️","tokens_out":12604,"duration_ms":98773,"temperature":0.7,"pith_summary":"The paper asks whether the color-octet mechanism of nonrelativistic QCD (NRQCD) shows up in exclusive double heavy quarkonium production at future Z factories. It computes cross sections for double charmonium and double bottomonium through $\\gamma^*/Z^*$ exchange, adding color-octet channels to the usual color-singlet ones. The central finding is that the octet channels are substantial or dominant for many channels at Z-pole energies, with gluon fragmentation into intermediate $^3S_1^{[8]}$ states the largest contributor. If this holds, Z-factory measurements would directly test the color-octet mechanism and constrain the corresponding long-distance matrix elements, especially $\\langle\\mathcal{O}(^3S_1^{[8]})\\rangle$. The paper also evaluates relativistic corrections and gives event-rate estimates for CEPC and FCC-ee, concluding that several channels are experimentally accessible.","feed_headline":"Color-octet channels can dominate quarkonium pairs at Z factories","feed_subtitle":"New event counts put dozens to thousands of quarkonium pairs within reach of CEPC and FCC-ee.","key_machinery":"The machinery is NRQCD factorization: each cross section is written as a sum over $Q\\bar Q$ intermediate states of short-distance coefficients times long-distance matrix elements (LDMEs), the non-perturbative quantities that describe hadronization of the pair into a quarkonium state. Color-octet channels are included at tree level for both QCD and electroweak diagrams, and the key dynamical feature is gluon fragmentation into $^3S_1^{[8]}$ pairs. Relativistic corrections are implemented by expanding amplitudes to $O(v^2)$, with $\\langle v^2\\rangle$ fixed through the Gremm-Kapustin relation.","core_discovery":"The paper's claim is that, at $\\sqrt{s}\\simeq m_Z$, the exclusive reactions $e^+e^- \\to H_1+H_2$ for double charmonium and double bottomonium receive color-octet contributions that are comparable to or larger than the color-singlet ones for most final states; for $\\eta_c+\\eta_c$ and $\\eta_b+\\eta_b$ the color-singlet contribution vanishes in the $\\gamma^*/Z^*$-propagated channels, making those processes pure octet probes. Gluon fragmentation into $^3S_1^{[8]}$ intermediate states is the dominant octet mechanism. Relativistic corrections to order $v^2$ suppress the charmonium cross sections by roughly a factor of 0.5 and bottomonium by 0.7-0.8. With NLO QCD $K$ factors and the planned luminosities, the paper predicts, for example, 22 and 206 events for $J/\\psi+\\eta_c$ at CEPC (16 ab$^{-1}$) and FCC-ee (150 ab$^{-1}$), respectively, with larger numbers for $J/\\psi+J/\\psi$.","pith_inferences":["Because the adopted octet LDMEs carry large uncertainties and come from a photoproduction fit, the absolute event counts are more fragile than the qualitative conclusion; ratios such as $\\sigma(J/\\psi+\\eta_c)/\\sigma(J/\\psi+J/\\psi)$ would cancel most of the parameter sensitivity.","The paper's differential cross sections show different angular shapes for CS and CO channels, so a $\\cos\\theta$ or $p_t$ cut could isolate the octet component experimentally.","The same gluon-fragmentation enhancement should appear in inclusive $Z\\to$ quarkonium plus light-hadron decays, allowing a cross-check of whether the extracted $\\langle\\mathcal{O}(^3S_1^{[8]})\\rangle$ is universal."],"forward_implications":["At the Z pole, several channels such as $J/\\psi+\\chi_{c1}$ and $\\eta_c+\\eta_c$ become color-octet dominated, so measuring them provides a direct test of the color-octet mechanism.","Comparing predicted and measured rates, especially for $J/\\psi+\\eta_c$, would give a direct constraint on $\\langle\\mathcal{O}(^3S_1^{[8]})\\rangle$ for charmonium and on the corresponding bottomonium matrix element.","Relativistic corrections must be included in such comparisons: for charmonium they reduce leading-order rates by about half, so omitting them would bias extracted matrix elements.","The estimated event counts, such as 22 and 206 $J/\\psi+\\eta_c$ events at CEPC and FCC-ee, indicate these measurements are feasible with the planned integrated luminosities.","Double-$J/\\psi$ production is almost purely color-singlet, so it can serve as a normalization channel when extracting octet matrix elements from other final states."],"supporting_citations":[{"why":"Defines NRQCD factorization, the velocity-scaling rules, and the LDMEs used throughout the calculation.","marker":"[1]"},{"why":"Supplies the color-octet LDMEs, in particular $\\langle\\mathcal{O}^{J/\\psi}(^3S_1^{[8]})\\rangle$, that drive the octet dominance and feed the event estimates.","marker":"[52]"},{"why":"Provides the NLO QCD and EW K factors used to convert the LO cross sections into the final event-count predictions.","marker":"[40]"},{"why":"Gives previous leading-order results for double heavy quarkonia at the super Z factory, used as a comparison baseline for the color-singlet cross sections.","marker":"[43]"},{"why":"Shows that color-octet contributions are significant in semi-exclusive $e^+e^-$ annihilation around the $Z^0$ peak, motivating the present exclusive study.","marker":"[46]"},{"why":"Establishes the relativistic-correction treatment and K-factor analysis used for the $O(v^2)$ reductions of charmonium cross sections.","marker":"[17]"}],"fun_headline_variants":["Color-octet paths dominate many double quarkonium channels at Z","Gluon fragmentation into color-octet states crucial for quarkonium pairs","Relativistic corrections halve expected charmonium pair rates at Z","Double quarkonium yields at CEPC and FCC-ee hinge on color-octet","eta_c+eta_c and eta_b+eta_b: pure color-octet probes at Z"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claimed octet dominance rests on the adopted color-octet long-distance matrix elements, especially $\\langle\\mathcal{O}^{J/\\psi}(^3S_1^{[8]})\\rangle = (0.0013 \\pm 0.0013)$ GeV$^3$ and the corresponding bottomonium values from a photoproduction fit; if those matrix elements are smaller or negative, the octet contributions shrink and the dominance could disappear.","fun_headline_variants_meta":{"raw":{"variants":["Color-octet paths dominate many double quarkonium channels at Z","Gluon fragmentation into color-octet states crucial for quarkonium pairs","Relativistic corrections halve expected charmonium pair rates at Z","Double quarkonium yields at CEPC and FCC-ee hinge on color-octet","eta_c+eta_c and eta_b+eta_b: pure color-octet probes at Z"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001045,"raw_usage":{"total_tokens":4481,"prompt_tokens":1121,"completion_tokens":3360,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":737,"completion_tokens_details":{"reasoning_tokens":3254}},"tokens_in":737,"tokens_out":3360,"duration_ms":23497,"temperature":1.0,"reasoning_tokens":3254,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T14:10:22.107579+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $e^+e^- \\to J/\\psi+\\eta_c$ at $\\sqrt{s}=m_Z$ with 16 ab$^{-1}$ (CEPC) or 150 ab$^{-1}$ (FCC-ee). The color-singlet-only prediction is about 10 or 93 events after $v^2$ and $\\alpha_s$ corrections, while the full color-singlet plus octet prediction is 22 or 206 events. A measured rate consistent with the lower number would rule out the adopted octet matrix elements; a rate near the higher number would support them.","supporting_citations":[{"cited_title":"Search for $C=+$ charmonium and XYZ states in $e^+e^-\\to \\gamma+ H$ at BESIII","cited_arxiv_id":"1310.0374","evidence_quote":"Supplies the color-octet LDMEs, in particular $\\langle\\mathcal{O}^{J/\\psi}(^3S_1^{[8]})\\rangle$, that drive the octet dominance and feed the event estimates."},{"cited_title":"One-loop corrections to the processes $e^+e^- \\to \\gamma, Z \\to J/\\psi~\\eta_c$ and $e^+e^- \\to Z \\to J/\\psi~J/\\psi$","cited_arxiv_id":"2101.01477","evidence_quote":"Provides the NLO QCD and EW K factors used to convert the LO cross sections into the final event-count predictions."},{"cited_title":"Liao and J","cited_arxiv_id":null,"evidence_quote":"Shows that color-octet contributions are significant in semi-exclusive $e^+e^-$ annihilation around the $Z^0$ peak, motivating the present exclusive study."},{"cited_title":"Zhang, Y.-J","cited_arxiv_id":null,"evidence_quote":"Establishes the relativistic-correction treatment and K-factor analysis used for the $O(v^2)$ reductions of charmonium cross sections."}],"review_version":1}