{"id":"2a82edf2-fc04-4c4c-af73-e3dc1e9a5eaf","arxiv_id":"2505.22426","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Direct LUNA measurements find 272.3 and 352.6 keV resonance strengths in 21Ne(p,γ)22Na that are about 1.5 times higher than literature, increasing the nova-temperature reaction rate by about 23%.","lead":"An underground laboratory measured five low-energy resonances in the 21Ne(p,γ)22Na reaction, which builds sodium-22 in stellar explosions. Two resonance strengths are about 1.5 times larger than previous values, raising the predicted reaction rate in novae by roughly 23 percent.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No load-bearing objection identified; the stopping-power input is unquantified but the agreement of three resonances with literature rules out a global normalization error large enough to explain the two discrepant strengths.","rationale":"The reader identified the stopping power as the weakest assumption, and I agree it is the least documented input. But the paper's internal consistency provides strong evidence that a global stopping-power error is not the cause of the 272.3/352.6 keV discrepancies: if ε_r were wrong by the ~50% needed to explain those factors, the 127.3, 271.4, and 291.5 keV strengths would also deviate, yet they match literature within 1σ. Therefore the stopping-power concern does not threaten the central claim of a real discrepancy; it only affects the absolute normalization and the stated systematic uncertainties. The paper should report the ε_r source and uncertainty, but this is a completeness issue rather than a correctness failure. The astrophysical impact is modest (23% rate increase at nova temperatures, no abundance changes), further reducing the stakes. The verdict of acceptance remains appropriate, and no condition is required beyond standard editorial requests for the stopping-power details.","tokens_in":10592,"tokens_out":18455,"duration_ms":194126,"concrete_test":"Verify the stopping-power source and values used in this work, presumably reported in the companion paper [30], and recompute the five resonance strengths of Table II using an independent stopping-power dataset (e.g., SRIM-2013 or PSTAR) for the actual neon gas mixtures and pressures. If the 272.3 keV strength shifts by more than about 5% or its ratio to the Görres et al. value drops below 1.5, the precision of the central claim would need revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the 272.3 and 352.6 keV resonance strengths are about a factor of 1.5 higher than literature. The most plausible weak point is Eq. (1), where every strength scales linearly with the adopted effective stopping power ε_r. The paper does not state the source of ε_r or its uncertainty, and the quoted systematic uncertainties (about 4.5%) appear to assume ε_r is known to that level. If ε_r were systematically too high by more than about 5%, the claimed 'more than a factor of 1.5' for the 272.3 keV resonance would become marginal. However, the same ε_r enters all five resonances, and the 127.3, 271.4, and 291.5 keV strengths agree with literature within 1σ. A global error in ε_r large enough to produce the 57% rise at 272.3 keV would also have shifted those three resonances, which are not shifted. Thus the stopping power cannot explain the selective discrepancy, and the direction of the central claim is robust. The remaining concern is purely about the absolute normalization and the completeness of the quoted uncertainty budget, not about whether the discrepancy is real.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a new direct study of five low-energy resonances in the 21Ne(p,γ)22Na reaction, measured at the LUNA underground accelerator using a windowless gas target and two HPGe detectors. Resonance strengths are extracted from thick-target yields using Eq. (1), with efficiency calibrated by radioactive sources and the 14N(p,γ)15O resonance and implemented in a GEANT4 simulation. The authors report that the strengths of the 127.3, 271.4, and 291.5 keV resonances agree with previous measurements within 1σ, while the 272.3 and 352.6 keV resonances are measured to be (129.9 ± 5.8) meV and (14.9 ± 0.8) meV, respectively, more than a factor of 1.5 above literature values and discrepant at 3.4σ and 4.3σ. New branching ratios are presented for three resonances. A revised thermonuclear rate is computed with the Monte Carlo code RATESMC and is about 23% higher than the Iliadis et al. evaluation in the 0.1–0.4 GK nova temperature range. Hydrodynamic ONe nova models and AGB nucleosynthesis calculations show negligible changes in final abundances when using the new rate.","tokens_in":10766,"tokens_out":6893,"duration_ms":71579,"significance":"If the reported strengths are correct, the paper provides an important improvement in a nuclear reaction of astrophysical interest, reducing the uncertainty on several resonance strengths from the previous 15–20% level to about 4–5% and resolving a large discrepancy for the 272.3 and 352.6 keV resonances. The three resonances that agree with literature act as a useful internal consistency check on the absolute normalization. The work is strengthened by the low-background LUNA environment, the use of two HPGe detectors with efficiency anchored to both radioactive sources and a well-known resonance, the Monte Carlo treatment of summing and detection efficiency, and the direct (not fitted) extraction of strengths from measured yields. The astrophysical impact section is appropriately cautious: despite the 23% rate increase, the stellar models show no significant abundance changes, which is a useful negative result. The manuscript would be fully convincing once a few methodological details, in particular the stopping-power input and the 271.4 keV branching-ratio dependence, are stated explicitly.","major_comments":[{"comment":"Equation (1) shows that every resonance strength is proportional to the adopted effective stopping power ε_r, yet the manuscript does not state the numerical value, the source (e.g., SRIM or ATOMIC), or the uncertainty of ε_r. The quoted systematic uncertainties (e.g., ±5.8 meV for the 272.3 keV resonance) appear to include a stopping-power component, but this is never made explicit. Please add this information and clarify how ε_r enters the systematic budget. I note that the agreement of the 127.3, 271.4, and 291.5 keV strengths with literature makes a global ε_r error an unlikely explanation for the selective 272.3/352.6 keV discrepancies, but the absolute normalization and the 23% rate comparison require the ε_r uncertainty to be stated.","section":"Data taking and analysis, Eq. (1), Table II"},{"comment":"The strength of the 271.4 keV resonance is derived from a single 2287 keV primary normalized to an unspecified literature branching ratio, and the weak 5468.7 keV primary was not observed. Please give the adopted branching ratio and its uncertainty, show explicitly how it contributes to the quoted 19% uncertainty on the strength, and describe how the contribution of the nearby 272.3 keV resonance (which is much stronger and only 1 keV away) was subtracted from the 271.4 keV yield. Without this information the 271.4 keV result cannot be fully evaluated.","section":"Data taking and analysis, 271.4 keV resonance"}],"minor_comments":[{"comment":"The word 'patternity' appears to be a typo; it should likely be 'parentage' or 'progenitor'.","section":"Introduction, first paragraph"},{"comment":"The first line of the manuscript contains a stray space in 'th e'; please fix the typo.","section":"Title/header"},{"comment":"The axis label 'NA<σv>' should use proper angle brackets or a clear notation such as N_A⟨σv⟩.","section":"Figure 5 axis label"},{"comment":"The branching-ratio table is difficult to parse because the columns for transition energies are not clearly labeled and the LUNA and literature values do not align by row; please add a header for the first column and reformat so each transition is a single row.","section":"Table I"},{"comment":"Reference [31] is listed only as 'Supplemental Material for more details'; please provide a full citation or a stable URL so that the supplemental data can be located.","section":"Reference [31]"},{"comment":"Please define λ_r explicitly as the de Broglie wavelength in the center-of-mass system and state the units used, to avoid ambiguity with the reduced wavelength.","section":"Equation (1)"}],"recommendation":"minor_revision","confidential_remarks":"This is a solid LUNA measurement with good cross-checks. The missing stopping-power details and the 271.4 keV branching-ratio input are readily fixable and do not undermine the main discrepancy claim. I recommend minor revision. No concerns about scope or novelty."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the short version: this is a clean, careful direct measurement from the LUNA collaboration, and the two resonance strengths that disagree with literature—272.3 and 352.6 keV—are probably right. The reason I trust them is internal: the same analysis reproduces the 127.3, 271.4, and 291.5 keV strengths to within 1σ. So a global normalization problem, like stopping power, would have to be selective to explain the discrepancies, which is implausible. The 3.4σ and 4.3σ deviations are the main new physics, and they survive the obvious systematic.\n\nWhat's actually new: five resonance strengths measured directly underground with 4-5% precision (except 271.4 keV at ~19%), new branching ratios for three resonances, and a Monte Carlo rate that is ~23% higher than Iliadis in the 0.1-0.4 GK range. The decay-scheme changes are useful too, especially for 127.3 and 272.3 keV.\n\nThe weak spots are minor and disclosed. The stopping power input ε_r isn't explicitly quantified in the text, although the internal consistency argument covers the main risk. The 271.4 keV strength leans on a literature branching ratio because the primaries are blended with 272.3; that's stated. The branching ratios for some transitions (e.g., 5700 keV at 272.3 keV: 2.7% vs 14.7% literature) change a lot without comment—probably worth a sentence in a revision. And the astrophysical impact is honestly null in their models: no change in 22Na abundances in ONe novae or AGB. That undercuts the 'importance' a bit, but the nuclear data improvement stands on its own.\n\nWho should read it: anyone using 21Ne(p,γ)22Na for nova nucleosynthesis, 22Na gamma-ray astronomy, or presolar grain studies. The measurement is formal and reproducible; the new rate table will be the reference for a while. I'd send this to a serious referee and accept it after the minor points above are addressed. Just ask the authors to state the stopping-power source and uncertainty explicitly and to add a remark on the large BR changes.","headline":"LUNA's new strengths for 21Ne(p,γ)22Na are well measured and the two big discrepancies with literature are credible; just don't expect the abundance changes in novae to move.","tokens_in":11631,"tokens_out":2325,"would_cite":true,"duration_ms":23795,"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":"Two low-energy resonances in the 21Ne(p,γ)22Na reaction, the 272.3 and 352.6 keV states, have strengths more than 1.5 times higher than previously measured, raising the thermonuclear rate by about 23% in the nova temperature range.","keywords":["21Ne(p,gamma)22Na","resonance strength","thermonuclear reaction rate","nova nucleosynthesis","neon-sodium cycle","underground nuclear astrophysics","branching ratios","gamma-ray spectroscopy"],"falsifier":"A direct measurement of the proton stopping power in neon gas at proton energies between roughly 130 and 350 keV, or an independent resonance-strength measurement using a target whose stopping power is known from first principles, would confirm or refute the new strengths. A re-measurement of the 272.3 and 352.6 keV resonances by another group using a different experimental technique would also decide whether the 3.4σ and 4.3σ deviations from the literature are real.","tokens_in":10369,"feed_emoji":"💥","tokens_out":9079,"duration_ms":83157,"temperature":0.7,"pith_summary":"Classical novae forge radioactive sodium-22, but the nuclear reaction that makes it—proton capture on neon-21—has carried large uncertainties because the relevant resonances sit at low energies where backgrounds are high. This paper reports direct measurements of five low-energy resonances in 21Ne(p,γ)22Na performed in a low-background underground facility. The strengths of the 272.3 keV and 352.6 keV resonances come out more than 1.5 times the previously published values, and revised branching ratios improve the decay schemes. The resulting thermonuclear reaction rate is about 23 percent higher than the standard evaluation in the temperature window of classical novae (0.1–0.4 GK), although hydrodynamic nova models show essentially unchanged sodium-22 yields. If the new strengths hold, the long-standing uncertainty in this reaction's contribution to the NeNa cycle is substantially reduced.","feed_headline":"Nova reaction rate climbs 23% after underground re-measurement","feed_subtitle":"Two low-energy resonances measure 1.5x stronger, lifting the 21Ne(p,γ)22Na rate in the nova range.","key_machinery":"The load-bearing identity is the thick-target yield formula $\\omega\\gamma = \\frac{2}{\\lambda_r^2} Y \\varepsilon_r \\frac{M}{m+M}$, which converts the measured thick-target yield $Y$ into the resonance strength $\\omega\\gamma$ using the de Broglie wavelength $\\lambda_r$ at the resonance energy, the effective stopping power $\\varepsilon_r$ of protons in the neon gas target, and the projectile and target masses $m$ and $M$. The formula is valid because all five resonances are narrow compared with the beam energy loss in the windowless gas target. The experiment exploits the low environmental background of an underground laboratory and two large high-purity germanium detectors, with detection efficiencies derived from a Monte Carlo simulation of the setup that also accounts for true coincidence summing and beam straggling. The resonance strengths are propagated into the thermonuclear rate with a Monte Carlo reaction-rate code that samples the input uncertainties.","core_discovery":"The authors claim that two of the five measured resonances in 21Ne(p,γ)22Na are substantially stronger than previously published: the 272.3 keV resonance at (129.9 ± 5.8) meV and the 352.6 keV resonance at (14.9 ± 0.8) meV, both more than a factor of 1.5 above the literature values, with deviations of 3.4σ and 4.3σ respectively. The remaining three resonances, at 127.3, 271.4, and 291.5 keV, are consistent with earlier measurements within 1σ. New branching ratios for the 127.3, 272.3, and 352.6 keV resonances update the decay schemes, and a Monte Carlo propagation of the new strengths yields a thermonuclear 21Ne(p,γ)22Na rate about 23% higher than the standard evaluation in the 0.1–0.4 GK classical nova window. Hydrodynamic simulations of oxygen-neon novae and nucleosynthesis calculations for AGB stars using the revised rate show essentially unchanged abundances of 22Ne, 22Na, and 25Mg, so the expected 1.275 MeV gamma-ray signal from 22Na decay is not affected.","pith_inferences":["Because the two newly higher resonance strengths share the same external stopping-power normalization, a systematic error in that normalization would move both in the same direction; the 23% rate increase should be read with that common-mode caveat in mind.","The 4–5% statistical precision demonstrated here suggests the same approach could be used to re-measure the 23Na(p,α)20Ne reaction, the other main unconstrained input to the NeNa cycle in nova models.","If the higher strengths are confirmed, the 21Ne(p,γ)22Na reaction's competition with other NeNa-cycle channels could change the predicted neon isotopic ratios in presolar grains, since those ratios are used to identify grain origins.","The 19% uncertainty on the 271.4 keV strength, which had to be derived from a single transition using an adopted branching ratio, points to that resonance as the next target for a dedicated measurement."],"forward_implications":["The 21Ne(p,γ)22Na thermonuclear rate is about 23% higher in the 0.1–0.4 GK range, which is the temperature window of classical novae.","Below 0.1 GK, relevant to AGB stars, the new rate is consistent with the previous evaluation but with smaller uncertainties.","The updated branching ratios and decay schemes for the 127.3, 272.3, and 352.6 keV resonances are available for future reaction-rate compilations and gamma-ray spectroscopy.","Hydrodynamic models of oxygen-neon novae and nucleosynthesis models of AGB stars show essentially unchanged abundances of 22Ne, 22Na, and 25Mg when the revised rate is used."],"supporting_citations":[{"why":"Supplies the literature resonance strengths for the 271.4, 272.3, 291.5, and 352.6 keV resonances that the new measurements are compared against.","marker":"[20]"},{"why":"Provides the prior strength for the 127.3 keV resonance and supports the assumption that all resonances are narrow compared with the beam energy loss.","marker":"[23]"},{"why":"Gives the previous recommended thermonuclear reaction rate that the new rate is compared to, yielding the 23% difference in the nova range.","marker":"[39]"},{"why":"Supplies the resonance strength data for all resonances beyond the five measured, used as inputs in the Monte Carlo rate calculation.","marker":"[38]"},{"why":"The companion paper describing the experimental setup, Monte Carlo simulations, and resonance energy determinations on which the present yields rely.","marker":"[30]"},{"why":"Provides the Monte Carlo code used to propagate the resonance strengths into the thermonuclear reaction rate.","marker":"[37]"},{"why":"Supplies the level scheme, Q-value, and literature branching ratios used as comparison for the new decay schemes.","marker":"[11]"},{"why":"Gives the standard thick-target yield formula connecting yield, stopping power, and resonance strength.","marker":"[33]"}],"fun_headline_variants":["Two key resonances 1.5x stronger in 21Ne(p,γ)22Na","Underground lab boosts nova reaction rate by 23%","New strengths raise 21Ne(p,γ)22Na rate in nova range","Subterranean data revise nova nucleosynthesis rate upward"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"All five resonance strengths scale linearly with the effective stopping power of protons in the neon gas target, which is adopted from external data rather than measured in this work, so any error in that stopping power would shift every reported strength—including the two discrepant ones—proportionally.","fun_headline_variants_meta":{"raw":{"variants":["Two key resonances 1.5x stronger in 21Ne(p,γ)22Na","Underground lab boosts nova reaction rate by 23%","New strengths raise 21Ne(p,γ)22Na rate in nova range","Subterranean data revise nova nucleosynthesis rate upward"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000173,"raw_usage":{"total_tokens":1347,"prompt_tokens":1081,"completion_tokens":266,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":697,"completion_tokens_details":{"reasoning_tokens":184}},"tokens_in":697,"tokens_out":266,"duration_ms":3490,"temperature":1.0,"reasoning_tokens":184,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T13:09:34.401129+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct measurement of the proton stopping power in neon gas at proton energies between roughly 130 and 350 keV, or an independent resonance-strength measurement using a target whose stopping power is known from first principles, would confirm or refute the new strengths. A re-measurement of the 272.3 and 352.6 keV resonances by another group using a different experimental technique would also decide whether the 3.4σ and 4.3σ deviations from the literature are real.","supporting_citations":[{"cited_title":"G¨ orres, C","cited_arxiv_id":null,"evidence_quote":"Supplies the literature resonance strengths for the 271.4, 272.3, 291.5, and 352.6 keV resonances that the new measurements are compared against."},{"cited_title":"Becker, H","cited_arxiv_id":null,"evidence_quote":"Provides the prior strength for the 127.3 keV resonance and supports the assumption that all resonances are narrow compared with the beam energy loss."},{"cited_title":"Iliadis, R","cited_arxiv_id":null,"evidence_quote":"Gives the previous recommended thermonuclear reaction rate that the new rate is compared to, yielding the 23% difference in the nova range."},{"cited_title":"Iliadis, R","cited_arxiv_id":null,"evidence_quote":"Supplies the resonance strength data for all resonances beyond the five measured, used as inputs in the Monte Carlo rate calculation."},{"cited_title":"Masha, F","cited_arxiv_id":null,"evidence_quote":"The companion paper describing the experimental setup, Monte Carlo simulations, and resonance energy determinations on which the present yields rely."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Monte Carlo code used to propagate the resonance strengths into the thermonuclear reaction rate."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the level scheme, Q-value, and literature branching ratios used as comparison for the new decay schemes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the standard thick-target yield formula connecting yield, stopping power, and resonance strength."}],"review_version":1}