{"id":"c81174ab-a904-4c6d-ba06-d75e977aba07","arxiv_id":"2502.09384","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"New infrared band strengths for CS2 and S8, combined with synthetic JWST spectra, yield detection thresholds showing H2S and SO2 are plausible ice detections while S8 is not.","lead":"Using new lab measurements and simulated JWST spectra, the authors assess whether sulphur-bearing molecules in interstellar ices could be detected. They conclude that H2S and SO2 are viable targets in favourable regions, while S8 is not, which helps plan future observations of the missing solid-phase sulphur.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Detection thresholds for H2S/SO2 assume a perfectly known continuum and band shapes; without an end-to-end injection-recovery test on real JWST data, the 'favourable conditions' claim is not yet secured.","rationale":"The paper's most valuable and robust result is the first laboratory band strength for S8 and the demonstration that S8 is undetectable even at the cosmic sulphur limit; that conclusion is insensitive to the synthetic-spectrum simplifications because the feature is isolated and the shortfall is an order of magnitude in column density. The conditional part of the paper is the positive claim about H2S and SO2. I read the threshold derivation carefully. In Sect. 4, the 5σ detection limit is found by adding white noise to the synthetic spectrum containing the S species and dividing by the noiseless, S-free 'clear' spectrum. This is a matched-filter-like estimate: it assumes the observer knows the exact band shape, position, and baseline. Real JWST ice spectroscopy requires subtracting a stellar continuum, dealing with gas-phase lines, scattering wings, and ice-band profile variations with composition. The authors are explicit about this (Sect. 3.2, Sect. 6), so the paper is honest. But the conclusion 'detection should be possible' requires not just honesty about the limitation but evidence that the limitation does not overturn the conclusion. For OCS, with A~1.2×10^-16 and an isolated 4.9 μm band, the margin is large. For H2S, the margin is much thinner: at the 1 mJy dense-cloud threshold, the synthetic peak optical depth is ~0.025, superimposed on a methanol combination mode with ~0.4 optical depth and, in real clouds, on the scattering red wing of water. A continuum error of even a few percent of the methanol band would be comparable to the H2S signal. The reader's conditional verdict already encodes this risk; my proposed injection-recovery test would turn the condition into a quantitative requirement. If the test shows thresholds are robust, the paper's central claim stands; if not, the positive claim should be downgraded to 'upper limits may be obtainable' rather than 'detection possible.' I therefore recommend no change to the conditional verdict, but the condition should be made explicit: the published thresholds should be labeled as ideal-noise, perfect-template limits until an end-to-end test on real data is done.","tokens_in":38164,"tokens_out":5850,"duration_ms":58216,"concrete_test":"Take a real JWST NIRSpec spectrum of a highly extincted background star from the Ice Age programme (e.g., NIR38 in Chamaeleon I). Inject synthetic H2S absorption with the Table A.1 Gaussian profile and column density equal to the paper's dense-cloud 5σ threshold for a 1 mJy continuum (4.6×10^16 cm^-2), plus the same for SO2 into a MIRI spectrum at its 7.5 μm threshold. Reduce the resulting data with the standard continuum-subtraction pipeline used for ice features, and measure the recovered SNR. If the recovered SNR is below 5, or if the recovered column density is biased by more than 50%, the published thresholds are optimistic and the 'favourable conditions' detectability claim for H2S/SO2 is not established. A cheaper analytic check is to recompute thresholds after adding a low-order polynomial continuum slope/curvature and correlated (1/f) noise instead of white noise.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that H2S and SO2 should be detectable rests on the 5σ thresholds computed in Sect. 4. Those thresholds are obtained by ratioing a noisy synthetic spectrum against a noiseless 'clear' spectrum with perfectly known Gaussian band parameters and a flat continuum. Sect. 3.2 explicitly states SynthIceSpec neglects scattering, radiative transfer, and continuum complexity and 'cannot be used to be compared with observations directly'; Sect. 6 adds that continuum subtraction and stellar photosphere were not treated. The concern is not that the authors hid this, but that the step from these idealized ratios to 'detection should be possible in favourable conditions' is unquantified. This matters most for H2S: its 3.92 μm band sits on the methanol combination mode and, in dense clouds, on the scattering red wing of the 3 μm water band (Sect. 5.1). A residual from continuum placement or band-shape mismatch can easily mimic or mask a feature whose synthetic peak optical depth is only ~0.025 (H2S at the 1 mJy dense-cloud threshold) against a methanol feature of optical depth ~0.4. The same logic applies, less severely, to SO2 at 7.5 μm in the crowded 7–8 μm region. If real-data reduction noise is correlated rather than white, or if continuum fitting absorbs some of the broad S-band flux, the true thresholds could be several times higher, potentially above the model/cometary abundances that make the conclusion interesting.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper combines new laboratory IR band-strength measurements of S8 (first reported value) and amorphous CS2 with literature values for H2S, OCS, and SO2, and estimates S3/S4 band strengths from DFT scaling of O3/O4 intensities. Using the SynthIceSpec tool, the authors generate synthetic NIRSpec and MIRI spectra for three representative environments (dense cloud, LYSO, MYSO) with ice compositions from Boogert et al. (2015), add white noise at a level derived from McClure et al. (2023), and compute 5σ detection thresholds for H2S, OCS, SO2, CS2 at three constant continuum fluxes (1, 0.1, 0.04 mJy), plus separate estimates for S8, S3, and S4. The main conclusions are that H2S and SO2 detections should be possible under favourable conditions, that CS2 and OCS remain challenging but plausible in some regimes, and that S8 would remain undetectable even if all cosmic sulphur were locked in that allotrope.","tokens_in":38530,"tokens_out":5472,"duration_ms":48136,"significance":"If the thresholds are robust, the paper provides concrete, falsifiable abundance targets for JWST searches for solid-phase sulphur carriers, which would directly address the missing-sulphur problem in dense interstellar environments. The new S8 band-strength measurement and the amorphous CS2 band strengths fill clear laboratory gaps, and the explicit comparison of thresholds with cometary abundances and gas-grain model predictions (Fig. 8) is a useful benchmark for future observations. The authors are commendably transparent about the limitations of SynthIceSpec in Sect. 3.2, stating that the tool cannot yet be compared with observations directly. The paper also makes its code repository available, which enhances reproducibility.","major_comments":[{"comment":"The dense-cloud CS2 threshold at 0.04 mJy is internally inconsistent: the listed column density of 7.4e17 cm^-2 divided by the water column density of 9.3e18 cm^-2 is approximately 7.96%, not the 0.8% printed in Table 2 and repeated in §4.1, §5.1, and the Conclusions. This factor-of-ten error changes the interpretation of CS2 detectability at faint continuum levels: instead of the stated range 0.06%–0.8% relative to water, the faint-flux threshold would be about 8%, making CS2 essentially undetectable at 0.04 mJy under the assumed dense-cloud composition. The authors should correct the table and all derived discussion, or explain the discrepancy if the percentage was computed with a different water column density.","section":"§4, Table 2; also §4.1, §5.1, §6"},{"comment":"The 5σ detection thresholds in Table 2 are derived by ratioing a noisy synthetic spectrum, built from known Gaussian band parameters and a perfectly flat continuum, against a noiseless 'clear' spectrum. The paper's own §3.2 states that SynthIceSpec neglects scattering, radiative transfer, continuum complexity, and source photosphere, and that it 'cannot be used to be compared with observations directly'; §6 further concedes that continuum subtraction and stellar photosphere were not treated. Because the central claim that H2S and SO2 are detectable rests on these thresholds, the idealized ratios must be validated with an end-to-end injection-recovery test on real JWST data or a sensitivity analysis that quantifies how correlated noise, continuum placement, and band-shape/position mismatches inflate the thresholds. This matters especially for H2S at 3.92 μm, where the synthetic peak optical depth at the 1 mJy threshold is only ~0.025 against a methanol combination-mode optical depth of ~0.4 and a possible scattering red wing of the water band (Sect. 5.1), and for SO2 at 7.5 μm in the crowded 7–8 μm region (Sect. 5.3). Without such quantification, the comparison with model and cometary abundances in Fig. 8 is likely to overstate the true detectability.","section":"§4, Eq. (1) and Figs. A.1–A.3; §3.2; §6"},{"comment":"The S3 and S4 band strengths used to derive the detection thresholds of 2e16 and 1e16 cm^-2 in §5.4 are estimated from DFT harmonic-approximation intensities scaled to O3/O4, and the paper acknowledges 'large uncertainties.' Additionally, the Gaussian widths for S3 and S4 are assumed equal to the measured S8 width. The statement in §5.4 that 'these smaller S allotropes could be detected if they contain about 20% to 10% of the sulphur budget' is therefore not robust to plausible variations in the assumed band strength or width. This is a secondary claim relative to the main H2S/SO2 conclusions, but the wording should be softened or supplemented with a short sensitivity test (e.g., varying the band strength and width by a factor of two) before publication.","section":"§2.3 and §5.4"}],"minor_comments":[{"comment":"The text reading '1.0%, 72.9%, and 15.5%' for the MYSO H2S thresholds at 0.1 mJy is a typo; Table 2 lists 7.9%, which is also consistent with the column density listed (3.9e17 cm^-2 over 5e18 cm^-2).","section":"§4.3"},{"comment":"The sentence listing the constant continuum values says '(1, 0.1 and 0.01 mJy)', but Section 4 and Table 2 use 0.04 mJy as the faintest continuum; the 0.01 value appears to be a typo.","section":"§3.1"},{"comment":"In Table A.1, the H2O band at 1655.0 cm^-1 is labelled 'Libration', but this frequency corresponds to the H2O bending mode; the text in §3.3 refers to a 'water libration mode' overlapping CS2, which should be checked for consistency.","section":"§3.3, Table A.1"},{"comment":"The abstract contains an awkward construction: 'includes vibrational absorption features of the S-carriers H2S, OCS, SO2, CS2, SO, CS, and S8 are found.' Also, §3.3 contains the typo 'prevens us' instead of 'prevents us'.","section":"Abstract and §3.3"},{"comment":"The reference list duplicates the entry for Calmonte et al. (2016), which appears twice; the duplicate should be removed.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of A&A and the lab measurements are a genuine contribution. The main concern is not the acknowledged simplifications but the absence of any quantitative test of how those simplifications affect the headline thresholds; the CS2 percentage error in Table 2 is an objective error that must be fixed. I see no citation or novelty concerns beyond the minor duplication in the reference list."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: the paper gives us two genuinely new lab numbers — the first S8 IR band strength at 20.3 µm and an updated amorphous CS2 band strength — and uses them, with existing literature values, to produce 5σ detection thresholds for H2S, OCS, CS2, and SO2 in dense cloud, LYSO, and MYSO ices. The ranking that comes out (H2S and SO2 are plausible targets, S8 is hopeless) is sensible and, I think, robust.\n\nThe lab work is careful enough. The S8 measurement is a room-temperature pellet rather than a cryogenic ice mixture, so I wouldn't treat it as a precise interstellar value, but it is ballpark right for the conclusion. The CS2 value is derived by scaling from the crystalline 77 K measurement, which is a reasonable way to get an amorphous-phase number.\n\nThe main soft spot is the detection threshold methodology. SynthIceSpec is a sum of Gaussians with known parameters and a flat continuum; the noise is white; there is no scattering, radiative transfer, or continuum-subtraction error. The authors say this clearly (Sect. 3.2 and Conclusions), and they explicitly caution that the tool isn't ready for direct comparison with observations. But then the abstract makes the 'should be possible' claim without the same weight of caveats. The stress-test's point about H2S is fair: at 3.92 µm, with a methanol combination mode nearby and the scattering wing of water in real dense clouds, the true detection threshold could be several times higher. That does not undo the qualitative conclusion — H2S is still the best candidate — but it does mean the quantitative thresholds, especially at 0.04 mJy, should be read as lower bounds on what observers can achieve.\n\nThere's also a concrete internal inconsistency: Sect. 4.3 quotes a 72.9% H2S threshold for MYSO at 0.1 mJy, while Table 2 gives 7.9%. Not a big deal, but it should be fixed.\n\nOverall: this is an honest feasibility study, with new lab data and a useful comparison context (models and comets). The central claims are hedged well enough that I would not want the simplified spectral model alone to sink it. What I would want from a referee is a request for (a) the typo fix, (b) an explicit statement that thresholds are idealized lower limits, and (c) ideally an injection-recovery demonstration on real JWST spectra in a follow-up.\n\nI'd send to peer review. Would cite for the S8 band strength, at least.","headline":"A useful lab-plus-model feasibility study with honest caveats; the new S8 and CS2 band strengths are the real goods, but the detection thresholds should be treated as idealized.","tokens_in":39065,"tokens_out":3714,"would_cite":true,"duration_ms":33684,"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":"The paper argues that JWST should be able to detect hydrogen sulphide (H2S) in dense interstellar ices and sulphur dioxide (SO2) around young stars under favourable conditions, while the S8 allotrope would remain undetectable.","keywords":["interstellar ices","sulphur-bearing molecules","JWST NIRSpec and MIRI","synthetic ice spectra","infrared band strengths","dense clouds","young stellar objects","S8 allotrope"],"falsifier":"A JWST NIRSpec and MIRI observation toward a dense cloud such as TMC-1 or Chamaeleon I that finds the H2S feature at 3.92 micrometres at an abundance below 0.5% relative to water, or that places an upper limit below that threshold, would directly test the dense-cloud claim; for S8, any credible detection of a band at 20.3 micrometres would overturn the paper's conclusion, since even the full cosmic sulphur budget yields a column density roughly ten times below the computed 5-sigma threshold.","tokens_in":38001,"feed_emoji":"🔭","tokens_out":4876,"duration_ms":45912,"temperature":0.7,"pith_summary":"The paper asks whether JWST can find the missing solid-phase sulphur in interstellar ices, beyond the one firmly confirmed carrier, OCS. It combines new laboratory measurements of CS2 and S8 infrared band strengths with synthetic spectra built from literature ice abundances for dense cores, low-mass young stellar objects, and massive young stellar objects, and computes 5-sigma detection thresholds at three continuum brightness levels. The central conclusion is that H2S in dense clouds and SO2 in young stellar objects should be detectable in favourable conditions, CS2 remains doubtful, and S8 will stay invisible even if all cosmic sulphur were locked in it. If right, JWST can deliver the first detections, or stringent upper limits, of H2S and SO2 in ices, giving astrochemical models their first solid-phase sulphur benchmarks.","feed_headline":"JWST could spot H2S and SO2 in interstellar ices","feed_subtitle":"Simulated spectra say the missing solid sulphur may hide in these two molecules, while S8 stays invisible.","key_machinery":"The load-bearing tool is SynthIceSpec, a synthetic ice spectrum generator that represents each solid-phase vibrational mode as a Gaussian absorption band whose optical depth follows N = (1/A) times the integrated band absorbance, with band strengths, positions, and widths taken from laboratory ice mixtures. For each target molecule, the generator adds the band to a simulated JWST spectrum, adds white noise at the rms level of recent JWST ice observations, and iteratively raises the molecular column density until the band reaches a 5-sigma signal-to-noise ratio, producing the detection thresholds for three continuum fluxes of 1, 0.1, and 0.04 mJy. The new laboratory work supplies the two missing inputs: the first measurement of the S8 band strength at 21.36 micrometres and updated band strengths for amorphous CS2 at 6.65 and 4.65 micrometres, calibrated against the crystalline value at 77 K.","core_discovery":"The paper establishes that the long-missing sulphur reservoir in dense interstellar ices could be partially visible to JWST, but only through specific carriers. Using the synthetic ice spectrum generator SynthIceSpec, which represents solid-phase absorption as sums of Gaussian bands with laboratory-derived positions, widths, and strengths, the authors simulated NIRSpec and MIRI spectra for dense clouds, low-mass young stellar objects, and massive young stellar objects, using observed mean ice abundances from the literature. New measurements provide the first reported S8 band strength of 1.5e-19 cm/molecule at 21.36 micrometres, and update the amorphous CS2 band strengths to 1.1e-16 cm/molecule at 6.65 micrometres and 2.1e-18 cm/molecule at 4.65 micrometres. From these spectra, the authors derive 5-sigma detectability thresholds: H2S at 3.92 micrometres requires roughly 0.5-17.5% relative to water depending on environment and continuum brightness, SO2 at 7.5 micrometres requires 0.08-2.1%, OCS at 4.9 micrometres requires 0.02-1.2%, and CS2 at 6.7 micrometres requires 0.06-1.6%. The paper concludes that H2S and possibly SO2 should be detectable in favourable regions, while S8 would require a column density near 3e18 molecules per square centimetre, an order of magnitude above what the full cosmic sulphur budget could provide.","pith_inferences":["If JWST upper limits on H2S and SO2 fall below the thresholds computed here, the main solid-phase sulphur reservoir would have to be distributed among species with very weak infrared bands or no accessible vibrational modes, such as S-chains, NH4SH salts, or refractory sulphur on grains.","Because the H2S band overlaps the methanol combination mode and the red wing of water, detection prospects depend strongly on line-of-sight ice composition; surveys across many lines of sight within one cloud would be needed to find the favourable patches the paper predicts.","The same threshold machinery could be extended to other undetected ice species, such as phosphorus-bearing or complex organic molecules, once laboratory band strengths for realistic ice mixtures become available.","A direct test of the synthetic spectrum approach would be to compare SynthIceSpec output against already-published JWST ice spectra for lines of sight with known compositions, checking whether Gaussian-sum predictions reproduce the observed continuum-divided band shapes before using the thresholds to claim new detections."],"forward_implications":["H2S at 3.92 micrometres should be detectable in dense clouds with bright background sources at abundances of about 0.5% relative to water, within the range predicted by gas-grain models and observed in comets.","SO2 at 7.5 micrometres could be detected toward low-mass and massive young stellar objects at thresholds of roughly 0.1-2% relative to water, provided overlapping bands from OCN- and complex organic molecules can be disentangled.","OCS at 4.9 micrometres remains the easiest sulphur carrier to detect, with thresholds on the order of 0.02-1.2% relative to water, consistent with existing detections and upper limits.","CS2 at 6.7 micrometres, despite its strong band strength, sits between the water and methanol bands and is deformed when mixed with water, making its detection doubtful unless its abundance exceeds roughly 0.06-1.6% relative to water.","S8 at 20.3 micrometres is effectively undetectable: even assuming all cosmic sulphur is locked in S8, the column density is about ten times below the computed 5-sigma threshold, and the band lies in an inefficient MIRI channel.","The non-detection of these molecules would imply that none of them is the main solid-phase sulphur reservoir, pointing instead to species with weaker bands, such as sulphur chains, salts, or refractory forms."],"supporting_citations":[{"why":"Supplies the observed mean column densities of major ice species for dense clouds, low-mass young stellar objects, and massive young stellar objects that define the three synthetic spectra.","marker":"Boogert et al. (2015)"},{"why":"Provides the rms noise levels used in the synthetic spectra and the existing upper limits on H2S and SO2 toward Chamaeleon I that the thresholds must beat.","marker":"McClure et al. (2023)"},{"why":"Provides the reference band strength for crystalline CS2 at 77 K used to calibrate the new amorphous CS2 band strengths.","marker":"Yamada & Person (1964)"},{"why":"Provides the H2S in water ice mixture parameters used in SynthIceSpec and the laboratory upper limits for H2S in protostellar ices.","marker":"Jiménez-Escobar & Muñoz Caro (2011)"},{"why":"Provides the OCS band strength and the H2O:OCS mixture parameters for the 4.9 micrometre feature.","marker":"Hudgins et al. (1993)"},{"why":"Provides the H2S and SO2 band strengths and evidence that sulphur band strengths in amorphous water are not strongly changed at 10 K.","marker":"Yarnall & Hudson (2022)"},{"why":"Supplies the 10 K H2O:CH3OH:CO2 ice spectrum from which the Gaussian parameters for the major ice species are fitted.","marker":"Ehrenfreund et al. (1999)"},{"why":"Provides the band position, shape, and width of SO2 in a water-ice mixture used for the SO2 detectability analysis.","marker":"Schriver-Mazzuoli et al. (2003)"},{"why":"Provides gas-grain model predictions for TMC-1 that place H2S as the main icy sulphur reservoir, used both for comparison and as model-driven synthetic spectra.","marker":"Navarro-Almaida et al. (2020)"}],"fun_headline_variants":["JWST could unmask missing sulphur hiding in interstellar ices","H2S and SO2: JWST's best shots at icy sulphur","Missing sulphur in ices: JWST may see H2S, not S8","JWST's icy sulphur hunt: H2S and SO2 are prime suspects"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole threshold calculation rests on the assumption that Gaussian band parameters measured for a handful of laboratory ice mixtures represent real interstellar ices without extra absorption, scattering, or baseline effects, an assumption the paper itself flags as preventing direct comparison with observations.","fun_headline_variants_meta":{"raw":{"variants":["JWST could unmask missing sulphur hiding in interstellar ices","H2S and SO2: JWST's best shots at icy sulphur","Missing sulphur in ices: JWST may see H2S, not S8","JWST's icy sulphur hunt: H2S and SO2 are prime suspects"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000602,"raw_usage":{"total_tokens":2955,"prompt_tokens":1236,"completion_tokens":1719,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":852,"completion_tokens_details":{"reasoning_tokens":1636}},"tokens_in":852,"tokens_out":1719,"duration_ms":12864,"temperature":1.0,"reasoning_tokens":1636,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T21:40:18.374085+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A JWST NIRSpec and MIRI observation toward a dense cloud such as TMC-1 or Chamaeleon I that finds the H2S feature at 3.92 micrometres at an abundance below 0.5% relative to water, or that places an upper limit below that threshold, would directly test the dense-cloud claim; for S8, any credible detection of a band at 20.3 micrometres would overturn the paper's conclusion, since even the full cosmic sulphur budget yields a column density roughly ten times below the computed 5-sigma threshold.","supporting_citations":[{"cited_title":"K., Rocha , W","cited_arxiv_id":null,"evidence_quote":"Provides the rms noise levels used in the synthetic spectra and the existing upper limits on H2S and SO2 toward Chamaeleon I that the thresholds must beat."},{"cited_title":"& Person , W","cited_arxiv_id":null,"evidence_quote":"Provides the reference band strength for crystalline CS2 at 77 K used to calibrate the new amorphous CS2 band strengths."},{"cited_title":"M., Sandford , S","cited_arxiv_id":null,"evidence_quote":"Provides the OCS band strength and the H2O:OCS mixture parameters for the 4.9 micrometre feature."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the H2S and SO2 band strengths and evidence that sulphur band strengths in amorphous water are not strongly changed at 10 K."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the band position, shape, and width of SO2 in a water-ice mixture used for the SO2 detectability analysis."}],"review_version":1}