{"id":"6d410b97-9e77-4bde-9d89-bd8b1f642e8d","arxiv_id":"2411.14530","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"In 40 ATLASGAL clumps, X(o-H2D+) falls about 6 times while X(N2D+) falls about 3 times with evolution, so the o-H2D+/N2D+ ratio is not confirmed as an evolutionary indicator; R_D = N2D+/N2H+ drops by nearly an order of magnitude.","lead":"A survey of 40 massive star-forming clumps shows that o-H2D+ and N2D+ abundances both drop as clumps evolve, while N2H+ rises slightly. The proposed o-H2D+/N2D+ abundance ratio is not confirmed as a reliable age indicator, and beam dilution may complicate deuterium fractionation tracers.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central claim rests on comparing separate abundance trends; the o-H2D+/N2D+ ratio itself is never directly plotted or statistically tested for stage separation, and no NLTE version of the ratio is presented.","rationale":"The reader's weakest-assumption pick (η_ff = 1) is valid and is explicitly flagged in Sect. 4.1, but it cannot be settled without new resolved observations. The more load-bearing issue is that the paper's central claim about the o-H2D+/N2D+ ratio is never actually tested as a ratio. The conclusion is derived from separately fitted abundance trends, which is not equivalent to showing that the ratio carries no evolutionary information. A direct ratio analysis using the already-tabulated LTE and NLTE column densities would settle the matter with existing data. If the ratio separates evolutionary stages, the paper's headline conclusion is wrong; if it does not, the conclusion is supported. The current omission of this test, together with the unverified assertion in Sect. 4.2.2 that abundance-ratio trends are unaffected by LTE/NLTE choice, makes the central claim under-supported. This is consistent with the reader's CONDITIONAL verdict: the paper needs revision to include the direct ratio test (and ideally an NLTE version) before the claim can be accepted. The abstract/body inconsistency about which molecule shows the largest LTE deviation is a separate, secondary issue that does not affect the central argument.","tokens_in":28849,"tokens_out":8067,"duration_ms":74535,"concrete_test":"Using Table A.2, compute X(o-H2D+)/X(N2D+) for every source with detections in both tracers, under LTE and under NLTE (using both the lower and upper n(H2) bounds). Plot the ratio versus log10(L/M) and versus evolutionary stage, and apply the same Anderson-Darling test used for R_D to compare the quiescent, protostellar, and YSO ratio distributions. If the ratio distributions are not significantly separated (P > 0.05) and the slope versus L/M is consistent with zero, the central claim is supported; if significant separation emerges in any version, the claim fails and the ratio is a viable indicator.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central negative claim—that X(o-H2D+)/X(N2D+) is not a reliable evolutionary indicator—is inferred in Sect. 4.2.3 and Sect. 5 from the separate findings that X(o-H2D+) decreases by a factor of ~6 between quiescent and YSO stages while X(N2D+) decreases by only ~3. However, the ratio itself is never computed for individual sources, plotted against L/M or evolutionary class, or subjected to a statistical separation test. The Anderson-Darling test is applied only to R_D = N2D+/N2H+ (Sect. 4.2.3), not to the o-H2D+/N2D+ ratio. Moreover, despite the assertion in Sect. 4.2.2 that 'the evolutionary trends of the abundances and abundance ratios... are not affected by the choice of analysis,' the NLTE analysis is shown only for R_D (Fig. 4b). This matters because the NLTE corrections differ substantially between the two species: o-H2D+ column densities are overestimated by factors of ~1.2–1.6, whereas N2D+ is overestimated by factors of ~1.3–5 depending on the assumed density (Table A.2). If these corrections vary by evolutionary stage, the LTE-based ratio trend could flatten, steepen, or even reverse. The beam-filling factor (η_ff = 1) is an acknowledged separate limitation, but the missing direct ratio analysis is a more immediate, internally checkable gap that bears directly on the headline conclusion.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper tests two proposed deuterium-based chemical clocks for high-mass star formation. The authors use APEX single-dish observations of N2D+ (3-2) and N2H+ (4-3) toward 40 ATLASGAL clumps within 4 kpc (5 quiescent, 8 protostellar, 18 YSO, 9 PDR), combined with the o-H2D+ (110-111) spectra already published in Sabatini et al. (2020). Column densities and H2-relative abundances are derived with MCWeeds LTE fits and RADEX non-LTE models, and the results are examined as a function of evolutionary class and of L/M. Three main results are reported. First, detection rates decline with evolution for both deuterated species (N2D+: 100% to 67%; o-H2D+: 80% to 10%), while N2H+ is detected in all 40 sources. Second, X(o-H2D+) declines by a factor of about 6 from quiescent to YSO stages, X(N2D+) by about a factor of 3, and X(N2H+) increases slightly; the authors conclude that the X(o-H2D+)/X(N2D+) ratio proposed by Giannetti et al. (2019) is not confirmed as a reliable evolutionary indicator. Third, the deuterium fraction R_D = N(N2D+)/N(N2H+) decreases by nearly an order of magnitude along the evolutionary sequence, with class separations that are statistically significant (Anderson-Darling test) and robust between LTE and non-LTE analyses, although the interpretation is complicated by the different spatial distributions of the two tracers and by beam dilution.","tokens_in":29205,"tokens_out":23818,"duration_ms":200088,"significance":"If confirmed, the paper's central negative result matters for the field: it would show that the o-H2D+/N2D+ ratio does not provide the strong discriminating power suggested by Giannetti et al. (2019), and it strengthens the case for R_D and for interferometric calibration of these clocks in high-mass clumps. The study deserves credit for several concrete strengths: it is a larger and more homogeneous test than earlier single-filament or small-sample studies; it openly corrects a sqrt(pi) error in the previously published o-H2D+ column densities (Sect. 4.2.1); it checks for distance-dependent biases in line widths (Fig. 1); it verifies the R_D class separation with Anderson-Darling tests; it demonstrates the LTE/non-LTE consistency for R_D (Fig. 4b); and Table A.2 is a complete and reusable data product. The detection rates in Table 1 are a clean, analysis-independent result. However, the headline claim about the o-H2D+/N2D+ ratio is currently inferred rather than demonstrated: the ratio is never computed per source or statistically tested, the o-H2D+ detections in the evolved stages are extremely sparse, and the analysis-independence argument is not shown for this ratio.","major_comments":[{"comment":"The central negative claim—that the X(o-H2D+)/X(N2D+) ratio is not a reliable evolutionary indicator—is never tested directly. The ratio is not computed for individual sources, not plotted against L/M or evolutionary class, and the Anderson-Darling test used to validate R_D is not applied to it; the conclusion is instead inferred from the separate abundance trends in Figs. 2 and 4a. A factor of about 6 decline in X(o-H2D+) together with a factor of about 3 decline in X(N2D+) implies a ratio change of only about a factor of 2, but whether this is too small to be a useful indicator depends on the per-source scatter of the ratio, which is not shown. Table A.2 shows that only nine sources have both o-H2D+ and N2D+ detections (four quiescent, one protostellar, two YSO, two PDR), so a detection-only test would be severely underpowered; the paper should either present the per-source ratio with an upper-limit treatment of the remaining sources and a class-separation test, or explicitly state that the censoring prevents such a test and qualify the conclusion accordingly.","section":"Sect. 4.2.3 and Sect. 5; Figs. 2 and 4a"},{"comment":"The assertion in Sect. 4.2.2 that the evolutionary trends of the abundances and abundance ratios are not affected by the choice of analysis is demonstrated only for R_D (Fig. 4b). The non-LTE corrections differ substantially between the two deuterated species: o-H2D+ column densities are overestimated by factors of 1.2-1.6, whereas N2D+ is overestimated by factors of 1.3-5, with the correction depending on the assumed volume density for each source (Table A.2). Because the corrections enter asymmetrically, the LTE trend of the X(o-H2D+)/X(N2D+) ratio could flatten, steepen, or reverse under non-LTE assumptions, and the non-LTE version of this ratio is never shown. The paper should present the non-LTE version of the ratio (as it does for R_D in Fig. 4b) to substantiate the analysis-independence claim; the necessary quantities are already in Table A.2.","section":"Sect. 4.2.2 and Fig. 4b"},{"comment":"The quantitative comparison between the X(o-H2D+) decline (factor about 6) and the X(N2D+) decline (factor about 3) rests on very unequal detection completeness. In the YSO class, o-H2D+ is detected in only about 10% of the sources (roughly 2 of 18) while N2D+ is detected in 67%; non-detections are shaded in Fig. 2, but no upper-limit treatment is described, and the class medians quoted in Sect. 6 and shown in Fig. 4a are computed from detections only. The YSO median of X(o-H2D+) is therefore based on a handful of detections, and the amplitude of the abundance decline—and hence the implied factor of about 2 dynamic range of the abundance ratio—is not robust. A censored-data treatment of the non-detections and a statement of the number of detections entering each median are needed before the factor-6 versus factor-3 comparison can carry the weight it is given in the conclusions.","section":"Table 1; Sect. 4.2.3; Fig. 4a"},{"comment":"There is an internal tension between the beam-dilution assumption and the spatial-distribution interpretation. Section 4.1 argues that o-H2D+ and N2D+ are chemically linked and should have similar emission extensions, so that eta_ff < 1 would cause a comparable underestimation of the two column densities. Section 5, by contrast, explains the weaker decline of X(N2D+) by its emission arising not only near the YSO but also from the outer envelope—that is, by a different spatial distribution of the two tracers. If the distributions differ, and differ in a stage-dependent way, the comparable-underestimation argument cannot hold simultaneously, and differential beam dilution would bias the very ratio whose reliability is being tested. The manuscript should reconcile these two statements and present the ratio test under explicit assumptions for eta_ff (e.g., equal eta_ff for the two deuterated species versus stage-dependent eta_ff), since the paper currently moves between a chemical-clock conclusion and an observational-bias conclusion.","section":"Sect. 4.1 vs Sect. 5"}],"minor_comments":[{"comment":"The symbol N_LTE is used for the LTE-derived column density while N_NLTE denotes the non-LTE value; since 'NLTE' conventionally means non-LTE, expressions such as 'Figure 3a compares NLTE(N2H+) with NLTE(N2D+)' in Sect. 4.2.3 are easily misread. Please rename the LTE column density (e.g., N_fit) and relabel the corresponding table columns.","section":"Sect. 4.2.1 and Table A.2"},{"comment":"In the arXiv abstract, the molecular formula 'N$_2$d$^+$' appears with a lowercase 'd'; it should read N2D+.","section":"Abstract"},{"comment":"The caption states that the orange and magenta lines represent the median FWHM values of 'N2H+ and N2H+'; the second species should be N2D+.","section":"Fig. 1 caption"},{"comment":"The phrase 'formation and destruction of deutereted molecules' contains a typo; it should read 'deuterated molecules.'","section":"Sect. 5"},{"comment":"The sentence 'The final mean T_sys lies between ~600 and 1200' is missing the unit of temperature (K).","section":"Sect. 3.2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is an honest and useful contribution, and I do not question the care with which the observations were reduced. My recommendation of major revision rests on a single but central point: the headline claim about the o-H2D+/N2D+ ratio is presented as a tested result when it is actually a qualitative inference from two separately fitted abundance trends, and the o-H2D+ detections in the most evolved classes are too sparse for the quoted amplitude of the effect. All requested additions (per-source ratio with upper limits, AD test on the ratio, NLTE version of the ratio) are feasible with data already in Table A.2 and would convert a plausible negative result into a demonstrated one. One context note for the editor: the o-H2D+ data and the tested ratio both originate from the same group's earlier work (Sabatini et al. 2020; Giannetti et al. 2019); I do not see this as a substantive flaw because the N2D+/N2H+ dataset is new and independent, but the authors could frame the study more explicitly as an independent verification test."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the paper ships a genuinely useful dataset — APEX N2D+ (3-2) and N2H+ (4-3) toward 40 ATLASGAL clumps, plus a cleaned-up re-analysis of the o-H2D+ column densities from Sabatini et al. (2020), including a typo correction. The R_D = N2D+/N2H+ result looks solid: it declines by about an order of magnitude from quiescent to YSO stages, holds in both LTE and non-LTE treatments, and the Anderson-Darling test gives clean separation between classes. That part deserves to be published and will be cited.\n\nThe soft spots are around the headline negative claim. The paper says the X(o-H2D+)/X(N2D+) ratio is not a reliable evolutionary indicator, but it never actually plots or tests that ratio. It infers the failure from the separate findings that o-H2D+ drops by ~6 and N2D+ by ~3. That is suggestive, but not the same as showing the ratio is useless. The ratio could still separate early and late stages even when both species decline. If the authors want to kill the indicator, they need to compute the ratio per source, put it against L/M or class, and run the same AD test they used for R_D.\n\nRelated gap: they claim the evolutionary trends are unaffected by the choice of LTE vs non-LTE analysis, but only show the NLTE version for R_D, not for the o-H2D+/N2D+ ratio. Given that the N2D+ LTE/NLTE discrepancy can be a factor of 5 depending on density, and that the corrections likely vary with stage, the ratio trend could shift. The paper should either present the NLTE ratio or drop the general claim.\n\nThe beam-filling issue is real but honestly flagged. They assume eta_ff=1, note they can't quantify it, and argue the two deuterated species should have similar extents. That's an assumption, not a demonstration, but they don't pretend otherwise.\n\nMinor: there's an abstract/body mismatch about whether N2D+ or N2H+ shows the largest LTE deviation. Easy fix.\n\nBottom line: this is a solid, incremental calibration paper, not a paradigm shift. The R_D result is the valuable part. The negative claim about o-H2D+/N2D+ is under-supported as written. Send it to a good referee, but the authors need to add the direct ratio test before it's done.","headline":"Useful new dataset and a robust R_D trend, but the paper's central negative claim about the o-H2D+/N2D+ ratio is inferred rather than directly tested.","tokens_in":29778,"tokens_out":2919,"would_cite":true,"duration_ms":27216,"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 X(o-H2D+)/X(N2D+) ratio does not hold up as an evolutionary indicator across 40 massive clumps.","keywords":["deuterium fractionation","chemical clocks","high-mass star formation","o-H2D+","N2D+","N2H+","ATLASGAL","beam dilution"],"falsifier":"Spatially resolve o-H2D+ and N2D+ emission in a sample of the same clumps with an interferometer: if the ratio X(o-H2D+)/X(N2D+) becomes monotonic with L/M once both emissions are mapped at matched resolution, the paper's negative result would be a beam-dilution artifact; if the ratio stays flat, the chemical counter is truly unreliable. A simpler check is measuring N(N2D+) column densities with resolved maps to see whether the flat N(N2D+)-L/M trend is caused by increasing emitting area.","tokens_in":28676,"feed_emoji":"🔭","tokens_out":4100,"duration_ms":34709,"temperature":0.7,"pith_summary":"The paper tests whether abundance ratios of deuterated molecules can order high-mass star-forming clumps by age. It finds that o-H2D+ abundance drops about six times from quiescent to YSO stages, while N2D+ drops only about three times, so the proposed X(o-H2D+)/X(N2D+) ratio is not a reliable clock. The deuteration fraction RD = N2D+/N2H+ does decline by nearly an order of magnitude, but its interpretation is complicated because the two molecules trace different spatial regions. The authors conclude that beam dilution and observational biases must be addressed before these tracers can calibrate evolutionary stage.","feed_headline":"Proposed chemical clock fails on massive clumps","feed_subtitle":"o-H2D+/N2D+ doesn't track stage; N2D+/N2H+ still declines as clumps age.","key_machinery":"The central object is the abundance ratio X(o-H2D+)/X(N2D+), proposed by earlier work as a chemical clock for massive clumps. The paper measures its ingredients from APEX spectra: o-H2D+ (110-111), N2H+ (4-3), and N2D+ (3-2) lines fitted with MCWeeds under LTE and cross-checked with RADEX non-LTE models. The luminosity-to-mass ratio L/M and a four-class evolutionary sequence (quiescent, protostellar, YSO, PDR) provide the time axis; the argument is that a reliable clock should show monotonic, class-separated abundance changes, which the ratio does not.","core_discovery":"In a sample of 40 ATLASGAL clumps observed with APEX, the paper finds that X(o-H2D+) decreases by about a factor of 6 from quiescent to YSO stages, whereas X(N2D+) decreases by about a factor of 3. Consequently, the recently proposed X(o-H2D+)/X(N2D+) ratio is not confirmed as a reliable evolutionary indicator for high-mass star formation. The deuteration fraction RD = N2D+/N2H+ decreases by almost an order of magnitude with evolution, making it a potential indicator, but the different spatial distribution of N2D+ and N2H+ complicates interpretation.","pith_inferences":["If beam-filling factors differ systematically between stages, the factor-of-six vs factor-of-three contrast could shrink or grow; resolved maps of the two ions in the same clumps would test whether the ratio is truly flat.","The flat N(N2D+) column density with L/M, in contrast to the declining abundance, suggests the molecule is spread over a larger area in evolved clumps; this could be checked by comparing line widths and spatial extents.","The paper's negative result for the ratio does not rule out that the ratio works at core scales; single-dish beams average over clumps, so interferometric follow-up may recover a sharp trend.","The increase of X(N2H+) with evolution, if confirmed, offers a complementary path: pairing N2H+ with a deuterated counterpart may separate temperature and density effects."],"forward_implications":["If correct, the o-H2D+/N2D+ ratio should be avoided as a standalone evolutionary indicator for massive clumps.","X(o-H2D+) remains a sensitive tracer of early, cold stages, dropping by a factor of about 6 with evolution.","RD = N2D+/N2H+ could still serve as an evolutionary indicator, decreasing by nearly an order of magnitude from quiescent to YSO stages, though with spatial-distribution caveats.","Detection rates of both o-H2D+ and N2D+ decline with evolution (80% to 10%; 100% to 67%), matching the picture that deuterated ions vanish as sources heat up.","Correcting for beam dilution with spatially resolved observations is required before quantitative age determinations can be trusted."],"supporting_citations":[{"why":"Proposed the X(o-H2D+)/X(N2D+) ratio as an evolutionary indicator and reported an anti-correlation between the two species; this paper directly tests that proposal.","marker":"Giannetti et al. (2019)"},{"why":"Provided the o-H2D+ APEX observations and earlier abundance trends that this study extends and re-analyzes with updated classifications.","marker":"Sabatini et al. (2020)"},{"why":"Reported resolved o-H2D+ and N2D+ emission and corrected a MCWeeds factor, supporting the extended-emission assumption and the recalibrated column densities.","marker":"Redaelli et al. (2022)"},{"why":"Measured N2D+ abundances in similar massive environments, providing the comparison baseline for the observed N2D+ decline.","marker":"Fontani et al. (2011)"},{"why":"Documented deuterium fractionation trends with temperature and density in star-forming regions, used to interpret the RD behavior.","marker":"Chen et al. (2011)"},{"why":"Tested the o-H2D+/N2D+ anti-correlation in Orion B9 cores and found an unclear N2D+ trend, a prior result this study corroborates on a larger sample.","marker":"Miettinen (2020)"},{"why":"Reported higher N2D+ detection rates in protostellar than prestellar cores, informing the interpretation of N2D+ as a later-stage tracer.","marker":"Li et al. (2022)"},{"why":"Defined the evolutionary classification of the TOP100 sample from which the 40 clumps are drawn.","marker":"König et al. (2017)"},{"why":"Provided updated dust temperatures, L/M ratios, and evolutionary classes used here as the time axis.","marker":"Urquhart et al. (2022)"},{"why":"Delivered the ATLASGAL 870 um survey that identified the clump sample and its physical parameters.","marker":"Schuller et al. (2009)"}],"fun_headline_variants":["Chemical clock fails on massive star-forming clumps","Proposed o-H2D+/N2D+ ratio not a reliable clock","Deuterium tracers decline but not as simple clock","High-mass star formation defies deuterium clock"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"That the molecules fill the telescope beam (beam-filling factor ηff = 1) for all tracers across all evolutionary stages, so that measured column densities are not systematically biased by the source size.","fun_headline_variants_meta":{"raw":{"variants":["Chemical clock fails on massive star-forming clumps","Proposed o-H2D+/N2D+ ratio not a reliable clock","Deuterium tracers decline but not as simple clock","High-mass star formation defies deuterium clock"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000221,"raw_usage":{"total_tokens":1553,"prompt_tokens":1152,"completion_tokens":401,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":768,"completion_tokens_details":{"reasoning_tokens":333}},"tokens_in":768,"tokens_out":401,"duration_ms":4140,"temperature":1.0,"reasoning_tokens":333,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T15:10:36.154739+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Spatially resolve o-H2D+ and N2D+ emission in a sample of the same clumps with an interferometer: if the ratio X(o-H2D+)/X(N2D+) becomes monotonic with L/M once both emissions are mapped at matched resolution, the paper's negative result would be a beam-dilution artifact; if the ratio stays flat, the chemical counter is truly unreliable. A simpler check is measuring N(N2D+) column densities with resolved maps to see whether the flat N(N2D+)-L/M trend is caused by increasing emitting area.","supporting_citations":[{"cited_title":"2019, A&A, 621, L7","cited_arxiv_id":null,"evidence_quote":"Proposed the X(o-H2D+)/X(N2D+) ratio as an evolutionary indicator and reported an anti-correlation between the two species; this paper directly tests that proposal."},{"cited_title":"2022, ApJ, 936, 169","cited_arxiv_id":null,"evidence_quote":"Reported resolved o-H2D+ and N2D+ emission and corrected a MCWeeds factor, supporting the extended-emission assumption and the recalibrated column densities."},{"cited_title":"2011, ApJ, 743, 196","cited_arxiv_id":null,"evidence_quote":"Documented deuterium fractionation trends with temperature and density in star-forming regions, used to interpret the RD behavior."},{"cited_title":"S., Wells, M","cited_arxiv_id":null,"evidence_quote":"Provided updated dust temperatures, L/M ratios, and evolutionary classes used here as the time axis."},{"cited_title":"M., Contreras, Y ., et al","cited_arxiv_id":null,"evidence_quote":"Delivered the ATLASGAL 870 um survey that identified the clump sample and its physical parameters."}],"review_version":1}