{"id":"cc4ca293-7caa-4c34-8a85-ca4a78ebd541","arxiv_id":"2506.05186","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"First detections of CH and H3+ ro-vibrational emission in a protoplanetary disk, with H3+ attributed to UV-driven chemistry.","lead":"Using JWST's NIRSpec instrument, astronomers detected the CH radical and, likely, the H3+ molecular ion in the UV-irradiated protoplanetary disk d203-506 in Orion. If confirmed, these are the first detections of both molecules in a protoplanetary disk, and they support the idea that UV radiation, not just cosmic rays, can drive H3+ formation.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"H3+ detection rests on residual subtraction after a simplified LTE model; systematic residual artifacts at blended positions could mimic the stacked feature.","rationale":"The CH detection is well supported (S/N=48, resolved lines, spatial distribution consistent with PDR models), so the central claim reduces to the H3+ detection. The reader's weakest assumption correctly identifies the global-model residual subtraction as the load-bearing premise. My stress-test refines this: the stacked H3+ spectrum includes a CH-blended line (R(1,0)) and relies on LTE line ratios that the paper itself argues are likely non-LTE. A permutation test directly quantifies the false-alarm probability of the stacked feature, which is not reported in the paper. If the test fails, the H3+ detection would be downgraded to an upper limit, removing the UV-formation claim. If it passes, the detection is validated. This is a specific, feasible check with existing data, so the conditional verdict remains appropriate.","tokens_in":13774,"tokens_out":4426,"duration_ms":48884,"concrete_test":"Perform a bootstrap permutation test on the residual spectrum: take the H3+ line list and shift all line positions by random velocity offsets (e.g., sampled uniformly between -500 and +500 km/s), excluding offsets that place lines within a few resolution elements of the true positions; re-stack using the same windows and weighting, and compute the stacked S/N. Repeat 10^3 times. If >=5% of random offsets yield S/N >= 8, the H3+ detection is not significant. Also recompute the stack excluding the R(1,0) line (blended with CH) and any line with a residual spike; if the feature disappears, the detection is not robust.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central new claim is the likely H3+ detection, which underpins the UV-chemistry interpretation. The detection is at S/N=8, just above the paper's own threshold, and the authors state the lines are 'weak and lie near the detection threshold' and 'open to interpretation' (Sect. 4.2). The S/N is measured against residuals from PAHTATmol, a global LTE model that the authors note leaves residual spikes from strong lines due to simplified excitation (Sect. 3.2). The stacked detection (Appendix B) includes the R(1,0) line, which is partially blended with CH; if the CH LTE model is slightly off at that wavelength, a residual can bias the stack. More generally, the stack weights lines by their model-predicted intensities, so any systematic error in the LTE excitation model (non-LTE is argued later) would distort the expected line ratios and could create a feature at zero velocity even in the absence of H3+. No control test against random line positions is reported, so the statistical significance of the stacked feature is not established beyond the nominal S/N=8.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript presents an analysis of the JWST/NIRSpec spectrum of the irradiated protoplanetary disk d203-506 using a new multi-molecule fitting tool, PAHTATmol. The authors report a clear detection of the CH radical (S/N = 48) and a likely detection of H3+ (S/N = 8) via the ν2 band, with LTE column densities implying abundances of a few 10^-7 and ~10^-8, respectively. They interpret the CH emission as a product of hydrocarbon photochemistry in the disk PDR, and argue that the hot H3+ emission, if real, would favour UV-driven H3+ formation rather than cosmic-ray ionization. A companion photochemical model predicts N(H3+) ≲ 10^13 cm^-2, in agreement with the LTE estimate.","tokens_in":13997,"tokens_out":6528,"duration_ms":71200,"significance":"If confirmed, the H3+ detection would be the first in a protoplanetary disk and would provide direct evidence for UV-driven H3+ formation in strongly irradiated dense gas, a process of broad astrochemical interest. The CH detection is robust and is a valuable new tracer of hydrocarbon chemistry in disk PDRs. The paper introduces PAHTATmol, a fitting tool that integrates state-of-the-art laboratory spectroscopic data (ExoMol and literature constants) and simultaneously fits 19 molecular species; this tool will be useful for future JWST disk and PDR studies. The manuscript also makes a falsifiable prediction of the H3+ column density from a companion model rather than fitting the model to the H3+ lines, a strength of the paper.","major_comments":[{"comment":"The H3+ detection rests entirely on the residual spectrum after subtracting the PAHTATmol global LTE model, yet the manuscript provides no control test to quantify the false-positive rate of the stacking procedure. The stack includes R(1,0), which is partially blended with CH (Sect. 4.2 and Table B.1), and the residual spectrum exhibits spikes from strong lines due to simplified excitation modeling (Sect. 3.2). Because the stacked feature is the primary evidence for H3+, the authors should demonstrate its robustness by (i) stacking an equal number of off-line positions to estimate the noise statistics, (ii) repeating the stack without the CH-blended R(1,0) line, and (iii) explicitly quantifying how uncertainties in the subtracted CH and other molecular models propagate into the H3+ residual. Without such tests, the nominal S/N = 8 does not establish a detection at the claimed confidence.","section":"Sect. 4.2, Appendix B, Fig. 4"},{"comment":"The detection significance defined in Eq. (A.1) uses the standard deviation of the residuals over ±10 pixels as the noise. These residuals are not white noise; they contain correlated structure from imperfect modeling of strong lines and PAH bands, as acknowledged in Sect. 3.2. Consequently, the S/N value is not a statistical significance in the usual sense, and the threshold S/N > 5 does not control the false-discovery rate of the multi-molecule search. The analysis should be supplemented with a Monte Carlo or bootstrap evaluation, or the noise should be estimated from line-free spectral regions, before H3+ can be claimed at the stated confidence.","section":"Eq. A.1, Appendix A.2"},{"comment":"The conclusion states 'We claim the detection of two new molecules in a protoplanetary disk: ... CH and ... H3+', but the abstract and Section 4.2 describe H3+ as a 'likely detection' whose lines are 'weak and lie near the detection threshold' and whose presence is 'open to interpretation' because of blends. Given that the H3+ case depends on the residual-subtraction tests requested above, the conclusion should either be revised to state explicitly that H3+ is tentatively detected pending further modelling, or the manuscript should provide the control tests that would justify the stronger wording. This distinction matters because the UV-formation interpretation is predicated on the H3+ detection.","section":"Section 5, abstract, Section 4.2"}],"minor_comments":[{"comment":"Equation (A.1) is typeset incorrectly in the manuscript (the square root symbol renders as 'vtX'), making the formula difficult to read; please correct the typesetting.","section":"Eq. A.1"},{"comment":"In the abstract, the verb 'underscore' should be 'underscores' to agree with the singular subject 'Our detection'.","section":"Abstract"},{"comment":"The chemical formula 'C 2H +2' appears with awkward spacing; please check the formatting (likely C2H2+).","section":"Introduction"},{"comment":"The caption contains 'di fferent' with a spurious space; it should read 'different'.","section":"Fig. 1 caption"},{"comment":"The reference to Berné et al. (2023) is given in the list with a URL rather than a standard journal citation; it should be formatted consistently with the other references.","section":"References"},{"comment":"The software name 'pgopher' should be capitalized consistently (PGOPHER) when referring to the program.","section":"Appendix A.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is a good fit for A&A Letters. The CH detection alone is solid and likely publishable. The H3+ claim is intriguing but requires the requested robustness tests; without them, the current wording overstates the result. I would encourage the editor to allow a revision that either adds the control tests or downgrades the H3+ claim to a tentative candidate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: this is a solid CH detection and a genuinely tantalizing but not yet secure H3+ candidate. The paper's value is in the CH detection and the tool, not in the H3+ claim as it stands.\n\nWhat's new: CH ro-vibrational emission is clearly detected (S/N=48) in the irradiated disk d203-506. That's a new molecule for a disk, and the spatial distribution (peaking deeper than CH+ and CH3+) is a nice observational result that ties into the group's photochemical models. The PAHTATmol fitting tool, while an extension of PAHTAT, is a useful framework for systematic searches across JWST spectra. Credit where due: they use state-of-the-art laboratory line lists (ExoMol and literature constants), and the companion paper's predicted H3+ column density is a real, independent check that matches their estimate.\n\nThe soft spot is the H3+ detection. It sits at S/N=8, just above their own threshold, and is derived from residuals after subtracting a global LTE model that they admit leaves spikes from strong lines. The R(1,0) line is blended with CH, so the stacked feature can be biased by a small error in the CH model. There's no control test against random line positions, which makes the stacked significance hard to assess. The LTE assumption is especially questionable for H3+ given its large Einstein A coefficients — they acknowledge this, but the column density and excitation temperature are still quoted from that model. So the central claim — first H3+ in a disk, UV-driven formation — is not yet established. The paper itself is candid about this ('open to interpretation'), which I respect.\n\nIn proportion: the CH detection alone would be a nice but minor incremental result. The H3+ claim, if true, is important. Given that importance, the paper deserves a serious referee, but the referee should demand either non-LTE excitation modeling, deeper observations, or at minimum a control test and code release. I'd bring it to a reading group, and I'd cite the CH detection, but I'd hold off on citing the H3+ as a detection until it's confirmed.\n\nRecommendation: send to peer review. The authors have been honest; the flaws are addressable.","headline":"Solid CH detection, but the headline H3+ claim is still a candidate, not a secure detection.","tokens_in":14677,"tokens_out":2217,"would_cite":true,"duration_ms":26411,"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":"This paper claims that JWST observations of the irradiated Orion disk d203-506 reveal the CH radical and, likely, the H$_3^+$ molecular ion, with H$_3^+$ formed by ultraviolet-driven chemistry rather than cosmic-ray ionization.","keywords":["CH radical","H3+ molecular ion","protoplanetary disk","JWST NIRSpec","photodissociation region","UV-driven chemistry","molecular ions","PAHTATmol"],"falsifier":"Compute the $\\nu_2$ band spectrum of H$_3^+$ with a non-LTE excitation model that includes chemical formation pumping and FUV radiative pumping at the densities and radiation field of d203-506 ($n_{\\rm H} \\simeq 10^7$ cm$^{-3}$, $G_0 = 2\\times 10^4$), and compare its predicted ratios among R(1,0), Q(1,0), Q(5,G), R(7,0), and R(7,6)u with the observed residual spectrum; if no single physical model reproduces the stacked feature and its line ratios, the identification fails. A complementary check is higher-resolution spectroscopy of those lines to verify the feature sits at zero velocity with the disk's line profile rather than at the position of a residual spike from a modeled species.","tokens_in":13564,"feed_emoji":"🪐","tokens_out":12578,"duration_ms":122560,"temperature":0.7,"pith_summary":"This paper reports the first claimed detections of the methylidyne radical (CH) and the trihydrogen cation (H$_3^+$) in a protoplanetary disk, using a JWST-NIRSpec spectrum of the strongly UV-irradiated Orion disk d203-506. CH is clearly detected at signal-to-noise 48 in its 3.3 $\\mu$m ro-vibrational emission, while H$_3^+$ is likely detected at signal-to-noise 8 through several $\\nu_2 = 1 \\to 0$ lines and a stacked residual feature at zero velocity. The authors estimate CH and H$_3^+$ abundances of a few $10^{-7}$ and about $10^{-8}$ relative to hydrogen nuclei. CH fits standard hydrocarbon photochemistry in the disk's photodissociation region, but H$_3^+$ is usually attributed to cosmic-ray ionization of H$_2$; here the authors argue instead for a UV-driven formation route in dense, strongly irradiated gas. If correct, these detections show that ultraviolet radiation alone can drive the chemistry of key molecular ions in planet-forming disks.","feed_headline":"JWST finds CH radical and H3+ ion in a planet-forming disk","feed_subtitle":"New detections in Orion's d203-506 suggest ultraviolet starlight, not cosmic rays, can forge H3+ around young stars.","key_machinery":"The load-bearing tool is PAHTATmol (Polycyclic Aromatic Hydrocarbons Toulouse Astronomical Templates with Molecules), a fitting engine that models a JWST spectrum as a linear combination of PAH band Gaussians, stellar and dust continuum blackbodies, atomic and H$_2$ lines, and LTE ro-vibrational emission from 19 small molecules, radicals, and ions. Molecular templates are precomputed with a spectroscopic simulation program at 300, 500, 1000, 1500, and 2000 K from laboratory line lists, and the fit uses non-negative least squares. Detection is judged by a signal-to-noise ratio computed from the fitted line peaks divided by the local residual scatter, with a threshold of S/N > 5; CH scores 48 and H$_3^+$ scores 8. For the weakest species the paper adds a stacking step in which residual spectra around the predicted H$_3^+$ lines are co-added, converting a set of near-threshold line candidates into a single visible emission feature at zero velocity.","core_discovery":"The paper's central claim, stated in the conclusion, is that two molecules are newly detected in a protoplanetary disk: the CH radical and the H$_3^+$ molecular ion. In the NIRSpec f290lp spectrum of d203-506, fitted with the multi-molecule LTE tool PAHTATmol, CH appears as clear ro-vibrational emission around 3.3 $\\mu$m with $T \\simeq 2000$ K, $N(\\mathrm{CH}) = 4.5\\times 10^{14}$ cm$^{-2}$, and a derived abundance of a few $10^{-7}$; its spatial emission peaks slightly deeper in the disk than CH$^+$ and CH$_3^+$, consistent with formation by dissociative recombination of CH$_3^+$ and reactions of carbon atoms with vibrationally excited H$_2$. H$_3^+$ is detected as a likely set of $\\nu_2 = 1\\to 0$ lines, including R(1,0), Q(1,0), Q(5,G), R(7,0), and R(7,6)u, with $T \\gtrsim 1000$ K, $N(\\mathrm{H}_3^+) = 1.1\\times 10^{13}$ cm$^{-2}$, and an abundance near $10^{-8}$; the stacked residual spectrum shows a clear feature at the expected zero-velocity position. Because H$_3^+$ emission peaks at the same bright spot as FUV-pumped H$_2$ and photodissociated OH, and because the companion photochemical model predicts $N(\\mathrm{H}_3^+) \\lesssim 10^{13}$ cm$^{-2}$ from UV-driven chemistry involving a hot water vapor reservoir, the authors favor a formation route largely independent of cosmic-ray ionization. This would be the first identification of H$_3^+$ in a protoplanetary disk.","pith_inferences":["A reader should treat the H$_3^+$ identification as conditional on the global model subtraction; the decisive test, beyond stacking, is a dedicated non-LTE model of H$_3^+$ excitation, which the paper itself says is needed.","If CH ro-vibrational lines contribute at 3.3–3.4 $\\mu$m, measurements of the 3.3 $\\mu$m PAH band in irradiated disks may need to subtract CH, which could affect PAH abundance estimates in such objects.","The same fitting approach could be run on other targets in the observing program; a correlation between CH and H$_3^+$ emission strength and incident $G_0$ would test whether the UV-driven formation route scales as predicted.","H$_3^+$ emission co-located with OH from water photodissociation suggests the ion could serve as a marker for UV-illuminated water destruction in disks, connecting molecular-ion chemistry to the water reservoir available for planet formation."],"forward_implications":["CH and H$_3^+$ should be sought in the JWST spectra of other strongly irradiated protoplanetary disks, where the same UV-driven chemistry is expected to operate.","A confirmed H$_3^+$ detection in d203-506 would break the usual assumption that interstellar H$_3^+$ traces cosmic-ray ionization; in dense, strongly FUV-irradiated gas, UV-driven routes through hot water vapor can dominate instead.","The vibrationally hot H$_3^+$ lines require a non-LTE explanation, since radiative rates near 100 s$^{-1}$ cannot be thermalized at densities of about $10^7$ cm$^{-3}$; this motivates statistical-equilibrium models that include chemical and radiative pumping.","The CH column density inferred from LTE emission exceeds the photochemical model prediction by up to a factor of about five, pointing to additional CH formation routes or to radiative pumping of the observed levels.","The stacked H$_3^+$ spectrum provides a template that can be used to search for H$_3^+$ in other JWST data sets."],"supporting_citations":[{"why":"Discovers CH$_3^+$ in d203-506, establishes the disk as strongly irradiated, and supplies the CH$_3^+$ spectrum used in PAHTATmol.","marker":"Berné et al. (2023)"},{"why":"Provides the CH$_3^+$ ro-vibrational line list used to compute the molecular template in PAHTATmol.","marker":"Changala et al. (2023)"},{"why":"Gives the disk's H$_2$ rotational temperature, physical size, mass, and the context of CH$^+$ and excited H$_2$ chemistry used for abundance and temperature estimates.","marker":"Berné et al. (2024)"},{"why":"Detects hot OH from water photodissociation co-located with the H$_3^+$ peak, supporting the UV-driven picture.","marker":"Zannese et al. (2024)"},{"why":"Companion model predicting H$_3^+$ formation in strongly irradiated dense gas via UV-driven hot-water chemistry, with column density matching the observed value.","marker":"Goicoechea et al. (2025)"},{"why":"Source of H$_3^+$ spectroscopic constants and line positions used to generate the $\\nu_2$ band LTE template.","marker":"Mizus et al. (2017)"},{"why":"Supplies the H$_3^+$ line list used in the spectroscopic simulation from which the PAHTATmol template is built.","marker":"Bowesman et al. (2023)"},{"why":"Establishes the linear-fitting PAHTAT approach that PAHTATmol extends with molecular templates.","marker":"Foschino et al. (2019)"},{"why":"Previous JWST-NIRSpec detection of H$_3^+$ in galaxies, used as the comparison showing d203-506's higher-energy lines.","marker":"Pereira-Santaella et al. (2024)"}],"fun_headline_variants":["JWST detects CH radical and first H3+ in a planet-forming disk","UV starlight may explain hot H3+ in young planet-forming disk","First detection of H3+ in a protoplanetary disk by JWST","Hot H3+ in a planet-forming disk points to UV-driven chemistry","JWST finds CH radical and hot H3+ in irradiated disk d203-506"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The H$_3^+$ detection stands only if the PAHTATmol model of everything else in the spectrum — PAH bands, continuum, and the other 18 molecules — leaves clean residuals at the H$_3^+$ wavelengths; any error in that global model at those positions could manufacture the weak, S/N=8 features.","fun_headline_variants_meta":{"raw":{"variants":["JWST detects CH radical and first H3+ in a planet-forming disk","UV starlight may explain hot H3+ in young planet-forming disk","First detection of H3+ in a protoplanetary disk by JWST","Hot H3+ in a planet-forming disk points to UV-driven chemistry","JWST finds CH radical and hot H3+ in irradiated disk d203-506"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00082,"raw_usage":{"total_tokens":3798,"prompt_tokens":1359,"completion_tokens":2439,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":975,"completion_tokens_details":{"reasoning_tokens":2337}},"tokens_in":975,"tokens_out":2439,"duration_ms":18821,"temperature":1.0,"reasoning_tokens":2337,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T10:23:04.965329+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the $\\nu_2$ band spectrum of H$_3^+$ with a non-LTE excitation model that includes chemical formation pumping and FUV radiative pumping at the densities and radiation field of d203-506 ($n_{\\rm H} \\simeq 10^7$ cm$^{-3}$, $G_0 = 2\\times 10^4$), and compare its predicted ratios among R(1,0), Q(1,0), Q(5,G), R(7,0), and R(7,6)u with the observed residual spectrum; if no single physical model reproduces the stacked feature and its line ratios, the identification fails. A complementary check is higher-resolution spectroscopy of those lines to verify the feature sits at zero velocity with the disk's line profile rather than at the position of a residual spike from a modeled species.","supporting_citations":[{"cited_title":"2024, Nat Astronomy, 8, 577","cited_arxiv_id":null,"evidence_quote":"Detects hot OH from water photodissociation co-located with the H$_3^+$ peak, supporting the UV-driven picture."},{"cited_title":"I., Alijah, A., Zobov, N","cited_arxiv_id":null,"evidence_quote":"Source of H$_3^+$ spectroscopic constants and line positions used to generate the $\\nu_2$ band LTE template."},{"cited_title":"A., Mizus, I","cited_arxiv_id":null,"evidence_quote":"Supplies the H$_3^+$ line list used in the spectroscopic simulation from which the PAHTATmol template is built."},{"cited_title":"2019, Astron","cited_arxiv_id":null,"evidence_quote":"Establishes the linear-fitting PAHTAT approach that PAHTATmol extends with molecular templates."},{"cited_title":"2024, A&A, 689, L12","cited_arxiv_id":null,"evidence_quote":"Previous JWST-NIRSpec detection of H$_3^+$ in galaxies, used as the comparison showing d203-506's higher-energy lines."}],"review_version":1}