{"id":"a66c91bc-948b-4057-849c-632b44dd1422","arxiv_id":"2502.10123","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"Across five astrochemical codes, methanol ice toward Chamaeleon I's NIR38 and J110621 forms predominantly (>99%) via CO hydrogenation, not via CH3+OH or C+H2O routes.","lead":"This paper runs five astrochemical codes to model the ices observed by JWST toward two stars behind the Chamaeleon I cloud. It finds that methanol ice forms almost entirely by hydrogenation of frozen carbon monoxide, while two recently proposed alternative routes contribute negligibly.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claim that alternative methanol routes are negligible rests on unvaried literature rate constants; a sensitivity test is needed before the >99% CO-hydrogenation share can be considered robust.","rationale":"Both the reader and I identify the same two fragile premises: line-of-sight homogeneity and unvaried rate constants for the added methanol routes. I argue the rate constants are the more load-bearing for the central claim as stated. The >99% CO-hydrogenation share is a quantitative statement about reaction fluxes inside the models; its quantitative accuracy depends directly on the rate coefficients of the competing routes. The models' systematic underproduction of CH3OH (by factors of roughly 2–10 for several codes, per Section 3.3 and Figure 5) means the observed methanol is not fully explained; a plausible resolution is that the alternative routes are more efficient than the adopted constants imply. A sensitivity test scaling these constants up and down is the cheapest decisive check and would settle whether the 'negligible' conclusion is an artifact of input choices. The line-of-sight concern is important for the best-fit physical conditions and for assigning the observed CH3OH to a specific cloud layer, but it is less likely to overturn the qualitative dominance of CO hydrogenation, because the models' low-density and early-time phases produce very little CH3OH (Section 3.1, Figure 1), and the paper explicitly argues atomic C is locked into CO beyond AV~2 mag (Section 4.8). I therefore retain the reader's CONDITIONAL verdict rather than moving it, because the paper should add the missing sensitivity analysis before the central claim is used as a quantitative benchmark.","tokens_in":40364,"tokens_out":10711,"duration_ms":106004,"concrete_test":"Rerun the Nautilus and UCLCHEM best-fit models from Table 5 with the rate coefficients of CH4+OH→CH3+H2O, CH3+OH→CH3OH, and C+H2O→H2CO multiplied by 0.01, 1, 100, and 10^4, and recompute the methanol formation flux fraction and final CH3OH/H2O ratio. If the CO-hydrogenation contribution remains above 99% and CH3OH is still underproduced, the central claim survives; if either changes materially, the conclusion must be qualified or revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2.6 adds the water-rich methanol routes CH4+OH→CH3+H2O, CH3+OH→CH3OH, and C+H2O→H2CO using rate constants from Qasim et al. (2018), Molpeceres et al. (2021), and Potapov et al. (2021). Section 4.2 then reports that all codes find >99% (often >99.99%) of CH3OH forms via CO hydrogenation, with CH3+OH contributing at most 1.4% even in UCLCHEM. However, no sensitivity run varies these rate constants, and the models systematically underproduce CH3OH ice (Section 3.3, Figure 5), leaving the observed methanol abundance unexplained. If the true rates of CH3+OH or C+H2O are higher than assumed, these routes could both raise CH3OH toward the observed values and reduce the CO-hydrogenation share below 99%. The quantitative 'negligible' claim therefore depends on accurate values for reactions that the paper does not test against plausible uncertainty ranges.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper models the ice chemistry toward two highly extincted background stars in the Chamaeleon I cloud observed by JWST, using five gas-grain codes with different architectures (MAGICKAL, MONACO, Nautilus, UCLCHEM, and KMC). The models follow a translucent-cloud phase and then a free-fall collapse over a grid of final densities, collapse timescales, dust temperatures, and visual extinctions, with the chemical networks augmented by the alternative, water-rich methanol routes CH4+OH, CH3+OH, and C+H2O. A two-step median absolute error (MAE) comparison to the JWST column densities selects best-fit models for each code. The paper reports that high gas densities and long collapse timescales best match the observations, that CH3OH ice forms predominantly (>99%) via CO hydrogenation with negligible contribution from the added alternative routes, that CO2 formation proceeds via CO+OH or CO+O depending on the code, that CH4 is underproduced except in UCLCHEM, and that the difference between diffusive and non-diffusive chemistry explains the weak temperature dependence at Tdust<12 K.","tokens_in":40583,"tokens_out":6452,"duration_ms":62642,"significance":"The multi-code comparison is a genuine strength: five independently developed codes with different treatments of surface chemistry agree on the dominance of CO hydrogenation for methanol formation under the studied conditions, and the KMC simulations give microscopic support for considering both CO+OH and CO+O routes to CO2. The paper does not fit microphysical parameters to the JWST data, so the mechanistic conclusion is not circular; observational uncertainties are propagated into the MAE errors. The main limitations are that the quantitative >99% share claim rests on unvaried literature rate constants and that the abstract's density preference conflicts with one code's best-fit model for one source. If these points are addressed, the paper will provide a robust benchmark for ice-chemistry modelling in the JWST Ice Age era.","major_comments":[{"comment":"The central quantitative claim that CH3OH forms predominantly (>99%) via CO hydrogenation relies on the literature rate constants adopted for the added alternative routes (CH4+OH→CH3+H2O, CH3+OH→CH3OH, and C+H2O→H2CO), but no sensitivity runs vary these values. Since the models systematically underproduce CH3OH (Fig. 5), plausible upward revisions of the CH3+OH or C+H2O rates could both raise CH3OH toward the observed values and reduce the CO-hydrogenation share below 99%. The UCLCHEM add/remove test described in §3.2.4 is a partial check, but it only shows that inclusion at the adopted rates does not change the final ice abundances; it does not test the rate values. Please add sensitivity runs, for example scaling the alternative-route rates by factors of 10 and 100 or varying them within their reported uncertainties, and report the resulting CH3OH formation share for all codes, or explicitly delimit the claim to the adopted rate set.","section":"§2.6, §4.2"},{"comment":"The abstract and conclusions state that the JWST ice observations are better reproduced for gas densities ≥10^5 cm^-3 and collapse times ≥10^5 yr, but the Nautilus best-fit model for NIR38 has nH=2×10^4 cm^-3 and tcol=10^6 yr (Table 5), with an MAE of 0.54, which is lower than the MAE of 0.95 for the corresponding high-density model. This is an internal inconsistency: either the Nautilus low-density model should be treated as an exception, or the summary should be reworded to say that most codes prefer high densities or that the high-density preference is driven by the other codes and by the H2O column-density filter. Please clarify how the NIR38 Nautilus result is reconciled with the general density claim.","section":"§3.3, Table 5, Abstract"}],"minor_comments":[{"comment":"The stated 'rate constant of 10^10 s^-1' for the indirect HOCO route appears to be a typo, as a bimolecular surface reaction rate of 10^10 s^-1 is implausible; please correct the value and specify units.","section":"§4.3"},{"comment":"For the KMC finite-size check, the paper states that 'a few short test simulations' with a 100×100 grain give the same results as the 50×50 grain; please report the test conditions and the quantified comparison so readers can judge the statement.","section":"§3.2.5, §4.10"},{"comment":"The first step of the model selection uses H2O column densities that agree 'within a factor of a few'; please state the numerical threshold used and how it was applied uniformly across codes, since it affects the set of models entering the MAE ranking.","section":"§3.3, Appendix D"},{"comment":"There are typographical artifacts such as 'ysosmolecular' and 'di fferent' that should be corrected in a careful proofreading pass.","section":"Introduction and throughout"},{"comment":"The claim that CO+O and CO+OH 'contributed roughly equally' to CO2 in the KMC simulations is presented without a quantitative table or run-to-run scatter; please report the flux percentages and, ideally, the standard deviation over multiple KMC realisations.","section":"§4.3"},{"comment":"For the Nautilus NIR38 best model, the preference for the low-density model over the high-density model appears driven by the complete absence of CH4 in the latter; showing which species dominate each MAE contribution would improve transparency.","section":"Table 5 and Figure 5"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is well within the scope of A&A and the multi-code comparison is a strong feature. The two major points above should be addressed before publication: the quantitative dominance claim needs a sensitivity analysis of the added alternative-route rate constants, and the abstract's density claim needs to be reconciled with the Nautilus NIR38 best fit. I see no fundamental flaw in the methodology, and I do not think rejection is warranted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis paper is the first dedicated multi-code test of the CH3+OH and C+H2O methanol routes under the JWST Ice Age conditions toward Chamaeleon I. Five independent codes (MAGICKAL, MONACO, Nautilus, UCLCHEM, KMC) give a consistent qualitative answer: CO hydrogenation dominates methanol ice formation, and the alternative routes contribute little. That consistency across rate-equation and stochastic architectures is the paper's real strength. The KMC finding that both CO+O and the indirect HOCO route contribute comparably to CO2 ice is also a useful mechanistic insight, despite the nominal rate constant for CO+O being very low.\n\nThe MAE-based model selection is transparent, and the paper is honest about the line-of-sight homogeneity assumption and other simplifications in Section 4.8. The connection to grain-growth timescales (nH≥1e5 cm^-3, tcol≥1e5 yr) is a nice cross-check.\n\nThe main soft spot is exactly what the stress-test note says: the 'negligible' verdict for CH3+OH and C+H2O rests on literature rate constants that are never varied. Since the models underproduce CH3OH ice (often by a factor of a few to ten), a plausible upward revision of those rates could both improve the CH3OH match and shift the balance away from pure CO hydrogenation. The paper's physical arguments (heavy radicals diffuse slowly; little atomic C is available) suggest CO hydrogenation would still dominate, but the >99% quantitative claim is not yet robust. A simple sensitivity run varying the CH3+OH and C+H2O rates over, say, an order of magnitude would settle it.\n\nMinor issues: no code or data are released, which limits the use of this as a benchmark; the KMC CH4 abundance fluctuates due to finite-size effects and is not trustworthy; and the line-of-sight integration issue is acknowledged but not quantified.\n\nOverall, this is a solid community paper that deserves a serious referee. The central direction of the result is likely correct, but the quantitative 'negligible' claim needs the sensitivity test before it can be used as a benchmark. I'd send it to review, and I'd cite it for the multi-code comparison. Want to bring it to reading group?","headline":"First multi-code test of the new methanol routes under Ice Age conditions; qualitative dominance of CO hydrogenation is plausible, but the >99% 'negligible' claim needs a rate-constant sensitivity run.","tokens_in":41269,"tokens_out":3352,"would_cite":true,"duration_ms":29337,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Five astrochemical codes find methanol ice forms predominantly (>99%) via CO hydrogenation, not the water-rich CH3+OH and C+H2O routes.","keywords":["interstellar ices","methanol formation","CO hydrogenation","grain-surface chemistry","astrochemical modelling","JWST Ice Age program","Chamaeleon I molecular cloud"],"falsifier":"Resolve the ice along the NIR38 and J110621 sightlines in space or in kinematic components and measure whether CH3OH ice (9.8 micron feature) appears in regions where CO has not frozen out (gas-phase depletion factor near 1). Since the paper's dominance claim requires CO on the grains for hydrogenation, a sightline or sub-region with little CO depletion but methanol ice at the observed 4-9% of water would falsify the claim that CO hydrogenation is the sole significant route.","tokens_in":40149,"feed_emoji":"🧊","tokens_out":8709,"duration_ms":79886,"temperature":0.7,"pith_summary":"The paper asks which chemical route builds methanol ice in the dense, heavily obscured gas that JWST's Ice Age program samples toward two background stars in Chamaeleon I. By running five independent astrochemical codes over a grid of densities, dust temperatures, extinctions, and collapse times, it claims that more than 99% of $\\mathrm{CH_3OH}$ ice forms by successive hydrogenation of CO on dust grains, while the recently proposed water-rich routes $\\mathrm{CH_3 + OH \\rightarrow CH_3OH}$ and $\\mathrm{C + H_2O \\rightarrow H_2CO}$ are negligible. It also claims that matching the observed ice column densities requires gas densities $\\gtrsim 10^5\\,\\mathrm{cm^{-3}}$ and collapse times $\\gtrsim 10^5\\,\\mathrm{yr}$, and that non-diffusive surface chemistry taking over below about 12 K may explain the nearly constant ice composition toward dense cores with very different extinctions. If true, methanol in these prestellar ices is a tracer of CO freeze-out rather than of a water-rich environment, which shifts how the 3 $\\mu$m ice band and the organic budget delivered to planet-forming disks are interpreted.","feed_headline":"Models: >99% of methanol ice forms via CO hydrogenation","feed_subtitle":"Five codes agree the water-rich CH3+OH and C+H2O routes are negligible toward Chamaeleon I's dense sightlines.","key_machinery":"The load-bearing machinery is a competition between methanol formation routes on interstellar dust grains: the CO hydrogenation chain ($\\mathrm{CO \\rightarrow HCO \\rightarrow H_2CO \\rightarrow CH_3O \\rightarrow CH_3OH}$) versus the two water-rich alternatives $\\mathrm{CH_4 + OH \\rightarrow CH_3 + H_2O}$ / $\\mathrm{CH_3 + OH \\rightarrow CH_3OH}$ and $\\mathrm{C + H_2O \\rightarrow H_2CO}$. Around these reactions the paper builds a grid of free-fall collapse models spanning gas density, dust temperature, extinction, and collapse time, executed with five codes that differ in whether chemistry is solved by rate equations or stochastically and whether non-diffusive (in-place) surface reactions are included. The kinetic Monte Carlo lattice simulation is the decisive instrument for the $\\mathrm{CO_2}$ sub-question: because it follows individual surface species, it shows that both the $\\mathrm{CO+O}$ channel and the HOCO-mediated $\\mathrm{CO+OH}$ channel contribute comparably despite the tiny rate constant of the former, a conclusion rate-equation codes cannot reach cleanly.","core_discovery":"On the paper's own terms, the central discovery is that $\\mathrm{CH_3OH}$ ice toward the highly extinguished lines of sight NIR38 and J110621 forms predominantly (>99%) via the classical grain-surface hydrogenation chain $\\mathrm{CO \\rightarrow HCO \\rightarrow H_2CO \\rightarrow CH_3O \\rightarrow CH_3OH}$. The alternative formation paths proposed for water-rich ices—$\\mathrm{CH_4 + OH \\rightarrow CH_3 + H_2O}$ followed by $\\mathrm{CH_3 + OH \\rightarrow CH_3OH}$, and $\\mathrm{C + H_2O \\rightarrow H_2CO}$ with subsequent hydrogenation—contribute negligibly in all five codes, and even when $\\mathrm{CH_4}$ is abundant (as in the Uclchem models) the $\\mathrm{CH_3 + OH}$ route stays below about 1.4% of the methanol produced. The paper further reports that the models best matching the JWST ice abundances have $n_{\\mathrm H} \\gtrsim 2\\times10^5\\,\\mathrm{cm^{-3}}$ and collapse times $\\gtrsim 10^5\\,\\mathrm{yr}$; that $\\mathrm{CO_2}$ ice forms via $\\mathrm{CO+OH}$ or $\\mathrm{CO+O}$ depending on the code, with KMC simulations showing both are efficient because atomic O is highly mobile on the surface; that $\\mathrm{CH_4}$ is underproduced except in Uclchem, where extra atomic C survives the translucent phase; and that at $T_{\\mathrm{dust}}<12$ K non-diffusive chemistry replaces diffusive chemistry, which is proposed as the reason ice compositions look similar across dense cores.","pith_inferences":["If the >99% CO hydrogenation result transfers to other clouds, then the 3 $\\mu$m red-wing excess toward dense sightlines may not require ammonia-hydrate mixtures; grain-growth interpretations like the one this paper cites could become the default reading.","A testable extension: map $\\mathrm{CH_3OH}$ and CO ice absorption along resolved sub-structures in Chamaeleon I; if $\\mathrm{CH_3OH}$ ice appears where CO is not depleted, the homogeneous-collapse assumption and the dominance claim would both be testable.","If non-diffusive chemistry indeed buffers ice composition below 12 K, then laboratory work on binding-energy distributions and in-place reaction probabilities becomes the key unknown for predicting ice inventories in cold cores."],"forward_implications":["Ice observations toward high-extinction background stars should be read as evidence of CO freeze-out and hydrogenation, not of methanol formation in water-rich ice.","Dense cores with $n_{\\mathrm H}\\gtrsim 2\\times10^5\\,\\mathrm{cm^{-3}}$ and collapse timescales $\\gtrsim 10^5\\,\\mathrm{yr}$ are the environments whose ices match JWST; fast, low-density collapse is ruled out.","At dust temperatures below about 12 K, non-diffusive surface chemistry buffers ice abundances against temperature differences, explaining the similar ice compositions seen in Chamaeleon I and other cores despite different extinctions.","Chemical models of $\\mathrm{CO_2}$ ice should include both $\\mathrm{CO+O}$ and the HOCO-mediated $\\mathrm{CO+OH}$ routes, since both can be efficient even when one has a low rate constant.","Methanol and methane ice abundances track the availability of atomic carbon during the translucent phase, so reproducing $\\mathrm{CH_4}$ requires models that carry enough atomic C into the dense phase."],"supporting_citations":[{"why":"Supplies the JWST ice column densities and ratios toward NIR38 and J110621 that all models are compared against.","marker":"McClure et al. 2023"},{"why":"Reinterprets the 3 micron red wing as grain growth and revises extinctions, motivating the re-examination of water-rich methanol routes.","marker":"Dartois et al. 2024"},{"why":"Laboratory basis for the CO hydrogenation chain that the paper finds dominant.","marker":"Watanabe & Kouchi 2002"},{"why":"Proposed the CH4+OH/CH3+OH methanol route in water-rich ices that is added to the networks and found negligible.","marker":"Qasim et al. 2018"},{"why":"Quantum-chemical source for the C+H2O route that is added to the networks and found negligible.","marker":"Molpeceres et al. 2021"},{"why":"Experimental companion for the C+H2O route tested in the models.","marker":"Potapov et al. 2021"},{"why":"Provides the modified-rate treatment and binding-energy assumptions that set the temperature behaviour of CO2 and methanol formation in several codes.","marker":"Garrod & Pauly 2011"},{"why":"Establishes the KMC lattice model used to resolve the CO2 formation channels.","marker":"Cuppen & Herbst 2007"},{"why":"Supplies the KMC CO-network reactions through which CH3OH forms and the alternative CH3O/H2CO channels are evaluated.","marker":"Simons et al. 2020"}],"fun_headline_variants":["Five codes: methanol ice is >99% CO hydrogenation","CO hydrogenation wins for methanol ice in all five models","JWST methanol ice: CO hydrogenation dominates, other routes faint","Methanol ice from CO hydrogenation, not water-rich chemistry","High density and slow collapse reproduce JWST methanol ice"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The models treat each observed sightline as a single homogeneous parcel collapsing at a constant dust and gas temperature, with no grain growth, ice morphology, or line-of-sight structure, and they take the literature rate constants for the alternative methanol routes as fixed.","fun_headline_variants_meta":{"raw":{"variants":["Five codes: methanol ice is >99% CO hydrogenation","CO hydrogenation wins for methanol ice in all five models","JWST methanol ice: CO hydrogenation dominates, other routes faint","Methanol ice from CO hydrogenation, not water-rich chemistry","High density and slow collapse reproduce JWST methanol ice"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000856,"raw_usage":{"total_tokens":3898,"prompt_tokens":1303,"completion_tokens":2595,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":919,"completion_tokens_details":{"reasoning_tokens":2511}},"tokens_in":919,"tokens_out":2595,"duration_ms":18996,"temperature":1.0,"reasoning_tokens":2511,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T19:18:23.780534+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Resolve the ice along the NIR38 and J110621 sightlines in space or in kinematic components and measure whether CH3OH ice (9.8 micron feature) appears in regions where CO has not frozen out (gas-phase depletion factor near 1). Since the paper's dominance claim requires CO on the grains for hydrogenation, a sightline or sub-region with little CO depletion but methanol ice at the observed 4-9% of water would falsify the claim that CO hydrogenation is the sole significant route.","supporting_citations":[{"cited_title":"J., Fedoseev , G., et al","cited_arxiv_id":null,"evidence_quote":"Proposed the CH4+OH/CH3+OH methanol route in water-rich ices that is added to the networks and found negligible."},{"cited_title":"2021, The Journal of Physical Chemistry Letters, 12, 10854","cited_arxiv_id":null,"evidence_quote":"Quantum-chemical source for the C+H2O route that is added to the networks and found negligible."},{"cited_title":"A., J \\\"a ger , C., & Henning , T","cited_arxiv_id":null,"evidence_quote":"Experimental companion for the C+H2O route tested in the models."},{"cited_title":", & Cuppen, H","cited_arxiv_id":null,"evidence_quote":"Supplies the KMC CO-network reactions through which CH3OH forms and the alternative CH3O/H2CO channels are evaluated."}],"review_version":1}