{"id":"bbf26d4c-a23d-4c95-a4bb-59301990534f","arxiv_id":"2411.16998","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Thorium III absorption near 18,000 angstroms should be observable in kilonova spectra when lanthanides are scarce and actinides are relatively abundant, making thorium the heaviest element identifiable in such events.","lead":"This paper predicts that thorium, the heaviest element yet studied in kilonova spectra, should create a broad absorption feature near 18,000 angstroms if the ejecta is lanthanide-poor and actinide-rich. That matters because confirming thorium would prove that neutron star mergers forge the heaviest r-process nuclei, and would tell observers when and where to look with JWST or high-altitude telescopes.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The detectability threshold scales linearly with Th III NIR gf-values, yet these are calibrated from one Boltzmann temperature with factor-of-3 scatter; an uncertainty of that size can push the required actinide enhancement outside the allowed factor-3 range.","rationale":"The reader's weakest assumption correctly identifies the Th III NIR gf-value calibration as the most load-bearing element of the paper's central claim. The paper's novelty is the specific, quantitative detectability criterion, and that criterion is directly proportional to gf-values that are estimated from a single-temperature fit with factor-of-3 scatter. The authors are transparent about this limitation—they label it 'the main uncertainty' in Section 4—but they still present the X(lan)≲6×10⁻⁴ threshold as a headline result. Because a factor-of-3 uncertainty in either direction can shift the threshold to the boundary of the physically motivated actinide-boost range, the claim cannot be considered fully quantitative. No internal inconsistency or methodological error was found; the radiative transfer treatment and candidate screening are sound. The paper makes a valuable, falsifiable prediction that motivates both improved atomic data and observational follow-up with JWST or high-altitude facilities. Therefore the CONDITIONAL verdict remains appropriate; no change to the reader's verdict is needed.","tokens_in":25683,"tokens_out":6007,"duration_ms":56423,"concrete_test":"Recompute the NIR Th III gf-values for the lines contributing to the 18000 Å blend using independent theoretical atomic calculations (e.g., Silva et al. 2022, Flörs et al. 2023, Fontes et al. 2023) or new laboratory emission measurements. If the gf-values for the dominant NIR lines differ from the values in Table 2 by more than a factor of 3, re-run the L-model radiative transfer with the corrected gf-values and with X(Z≥82)/X(lan) boosted by a factor of 3; if the 18000 Å feature no longer appears at X(lan)=5.8×10⁻⁴, the stated detectability criterion is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim—that Th III absorption near 18000 Å appears when X(lan) ≲ 6×10⁻⁴ and the actinide-to-lanthanide ratio exceeds the solar r-process value—rests on the NIR gf-values of Th III lines estimated in Section 2.1. These gf-values are not measured; they are derived from Eq. (1) using a single excitation temperature T=6000 K obtained by least-squares fitting to optical lines with known gf-values (Biémont et al. 2002). Figure 1 shows a factor-of-3 scatter between measured and calculated intensities, and the NIR lines have upper-level energies up to ~28000 cm⁻¹, well beyond the range of the calibration lines. The paper itself states (Section 4) that the required Th mass fraction scales linearly with the gf-values, so a factor-of-3 overestimate of the true transition strengths would raise the detectability threshold by the same factor, placing it at the edge of the physically motivated 'actinide-boost' factor-3 range invoked in Section 3.2. The authors acknowledge the factor-of-3 scatter but do not propagate this uncertainty into the X(lan) threshold, nor do they test whether the same Boltzmann temperature applies to the high-lying levels involved in the NIR transitions. A secondary contributor is the uncertainty in the Ce III gf-values near 20000 Å (theoretical, uncalibrated), which affect the background against which the Th feature is identified. Because both the abundance threshold and the feature's visibility depend linearly on these atomic data, the quantitative detectability criterion is not yet robust.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper investigates whether elements beyond the third r-process peak can be identified in kilonova photospheric spectra. The authors construct and update line lists for Ra II, Ac III, and Th III, combine them with their existing hybrid line list, compute Sobolev optical depths under LTE, and run Monte Carlo radiative transfer for two abundance models (Light and Solar). They conclude that Th III is the most promising candidate, producing broad absorption near 18000 Å when the lanthanide mass fraction X(lan) is ≲ 6×10^-4 and the actinide-to-lanthanide ratio is at least the solar r-process value. They argue such conditions are plausible in mergers of unequal-mass neutron stars or black hole-neutron star binaries, and recommend space- or high-altitude observations in the telluric-affected NIR region.","tokens_in":25924,"tokens_out":6755,"duration_ms":74120,"significance":"If the central claim holds, the paper provides a concrete, falsifiable spectroscopic prediction for actinide production in neutron star mergers, with a specific wavelength, epoch window (roughly 1 to 2.5 days), and abundance condition. The forward modeling is transparent and systematic: the authors compute Sobolev optical depths for candidate heavy species, identify why Th III is special (dense low-lying levels, favorable ionization balance), and present with/without-Th III spectral comparisons that isolate the predicted feature. They also make the estimated Th III line list available in Appendix A. The main weakness is the atomic-data basis: the NIR Th III gf-values are not measured but recovered from relative intensities via a single Boltzmann temperature, and the Ce III gf-values near 20000 Å are theoretical. These uncertainties propagate linearly into the abundance condition that is the paper's principal quantitative result, so the criterion should be presented with an explicit uncertainty range.","major_comments":[{"comment":"The central detection criterion is not yet robust as stated because the Th III NIR gf-values are calibrated from a single least-squares Boltzmann temperature T=6000 K applied to optical lines, while Figure 1 shows a factor-of-three scatter between measured and calculated intensities and the NIR lines reach upper-level energies up to roughly 28000 cm^-1, outside the calibration range. The authors acknowledge in Section 4 that the required Th mass fraction scales linearly with the gf-values, but they do not propagate the factor-of-three uncertainty into the X(lan) threshold or the actinide-to-lanthanide ratio. A factor-of-three overestimate of the true NIR gf-values would raise the required Th abundance by the same factor and push the criterion to the boundary of the factor-of-three actinide-boost range invoked in Section 3.2. I request a sensitivity calculation with the Th III NIR gf-values scaled by 1/3 and 3, and a restatement of the detection threshold as an allowed range rather than a single value.","section":"Section 2.1, Eq. (1), and Figure 1"},{"comment":"The visibility of the Th feature is set against the Ce III opacity near 20000 Å, and the gf-values of these Ce III lines are theoretical and uncalibrated except for the strongest lines near 16000 Å, as the authors state in Section 4. Because the threshold X(lan) is effectively the value at which the Ce III background drops below the Th III opacity, an uncertainty in the Ce III gf-values translates directly into an uncertainty in the threshold. The paper should include a sensitivity study with the Ce III gf-values varied by a factor of 2–3 and should state whether the conclusion survives those variations.","section":"Sections 3.2 and 4"},{"comment":"The observational claim that JWST/NIRSpec or high-altitude telescopes 'will give sufficient signal to noise ratio' to detect the Th feature is asserted without a quantitative estimate. Given that the feature sits in a region of strong telluric absorption and the synthetic fluxes in Figures 10 and 11 are only shown for an AT2017gfo-like event, a simple S/N estimate for a GW170817-like event at ~200 Mpc with a proposed instrument configuration would make the detectability statement much stronger. This is not the central modeling result, but it is part of the paper's stated practical conclusion.","section":"Section 4"}],"minor_comments":[{"comment":"The phrase 'solar r-process ratio' should be defined more carefully: the L model's Th-to-lanthanide ratio is about a factor of two higher than the present solar r-residual ratio because of 232Th decay over ~13 Gyr, so the criterion is not directly comparing to the present solar photospheric ratio.","section":"Abstract and Section 3.2"},{"comment":"The cross-reference 'Table 2.1' should be 'Table 2', and the footnote referring to 'gl-values' appears to mean 'gf-values' or should be rewritten for clarity.","section":"Section 2.1"},{"comment":"There are several typographical errors: 'kilonvoae' in the Introduction, 'Soboelv' in Section 2.3, 'λ /greaterorsimilar20000' in Section 3.2, and 'faction' in Section 4. These should be corrected.","section":"Throughout"},{"comment":"The axes 'Calculated intensity' and 'Relative intensity (NIST)' would be clearer with explicit labels and a note that intensities are in arbitrary units; the red/blue circle legend should also be explained more fully in the caption.","section":"Figure 1"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid forward-modeling study with a clear and testable prediction, and I see no scope or novelty concerns. The central issue is that the quantitative abundance threshold rests on estimated gf-values with a factor-of-three scatter. The authors already acknowledge the issue in Section 4, but the paper's headline criterion needs to be expressed as a function of that atomic-data uncertainty. I would not require new laboratory measurements as a condition of publication, but I would require the authors to show how the threshold shifts under plausible gf scaling and to state the resulting allowed range."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the short version: this is a careful forward-modeling paper that turns the Th hint from Domoto et al. 2022 into a concrete, testable prediction—Th III absorption near 18000 Å in kilonova photospheric spectra under a specific set of abundance conditions. It deserves a serious referee. But the quantitative boundary (X(lan) ≲ 6e-4, actinide/lanthanide ratio ≳ solar) is shakier than the abstract suggests, because the estimated gf-values carry a factor-of-3 uncertainty that shifts the threshold by the same factor.\n\nWhat's actually new: the systematic comparison of Ra II, Ac III, and Th III; the updated NIR gf-values for Th III estimated from measured line intensities; and the parameter study that defines the detectability conditions. The atomic-physics argument is solid—Th III has dense low-lying energy levels and relatively strong NIR lines, making it the most plausible actinide to produce a photospheric feature. The radiative transfer is internally consistent: the with/without-Th comparison cleanly isolates the feature, and the temperature dependence (Th III ionizes to Th II below ~5000 K) explains why it fades after ~2.5 days.\n\nThe soft spot is exactly what the stress-test flags. The NIR gf-values are derived from a single Boltzmann temperature (6000 K) with factor-of-3 scatter in Figure 1, and the NIR lines have upper levels up to ~28000 cm^-1, beyond the calibration lines. The authors acknowledge that the required Th mass fraction scales linearly with the gf-values, but they don't propagate this uncertainty into the X(lan) threshold. If the true gf-values are lower by a factor of 3, the required actinide enhancement is 3× larger, pushing it to the edge of the actinide-boost range they invoke. That doesn't kill the qualitative claim—there are plausible conditions under which Th III should be visible—but the specific threshold is not yet robust. The missing-opacity question (other actinides like Pa affecting the NIR) is also not fully closed, and the LTE assumption is an acknowledged limitation. No code or data are released, which limits independent verification.\n\nOverall: this is an honest, careful study with a clear observational target. I'd send it to peer review. The referee should push for a sensitivity analysis that propagates the gf-value uncertainty into the detectability condition, and ideally better atomic data or at least a range of plausible thresholds. The qualitative prediction—Th III as the most promising actinide diagnostic in kilonova spectra—holds up.","headline":"Careful forward-modeling paper that turns a hint into a testable Th III prediction worth refereeing, but the factor-of-3 uncertainty in the input gf-values makes the quantitative threshold softer than the abstract claims.","tokens_in":26624,"tokens_out":3875,"would_cite":true,"duration_ms":35231,"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":"Thorium is the likely heaviest element whose signature can appear in kilonova spectra, showing up as a broad absorption trough near 18,000 Å under actinide-boosted conditions.","keywords":["thorium","kilonova spectra","r-process nucleosynthesis","actinides","neutron star merger","radiative transfer","Th III","near-infrared spectroscopy"],"falsifier":"Measure the oscillator strengths of the Th III lines near 18,000 Å in the laboratory. If the true values are more than a factor of three smaller than the estimates used here, the required thorium fraction would exceed the actinide-boost range and the predicted trough would not appear, while a high signal-to-noise, space-based spectrum of a suitable kilonova at 1–2.5 days that shows no trough at 18,000 Å despite an inferred actinide-boost composition would also rule the prediction out.","tokens_in":25381,"feed_emoji":"🔭","tokens_out":11721,"duration_ms":98486,"temperature":0.7,"pith_summary":"This paper argues that thorium is the heaviest element whose signature can realistically appear in the photospheric spectra of a kilonova, and that its fingerprint is a broad absorption trough in the near-infrared near 18,000 Å. The authors build an updated line list for the actinide ions Ra II, Ac III, and Th III, then run radiative transfer simulations of neutron star merger ejecta. They find that Th III stands out as the only candidate with strong near-infrared lines, and that those lines form a visible absorption feature when the actinide-to-lanthanide mass ratio exceeds the solar r-process value and the lanthanide mass fraction is $X_{\\rm lan}\\lesssim 6\\times10^{-4}$. If real, a space-based or high-altitude spectrum taken one to two and a half days after a merger would give direct, unambiguous evidence that neutron star mergers synthesize elements heavier than the third r-process peak.","feed_headline":"Heaviest element fingerprint: thorium absorbs at 18,000 Å","feed_subtitle":"A space-based spectrum of a neutron-star merger could confirm the r-process forges actinides.","key_machinery":"The central machinery is a hybrid atomic line list for ions beyond the third r-process peak, anchored by experimentally calibrated data for Ra II and Ac III and by a set of near-infrared Th III transition probabilities estimated from measured relative intensities. The Th III estimates assume a single Boltzmann excitation temperature of $T=6000$ K, fixed by least-squares comparison between measured and calculated intensities of known optical lines. Under LTE, Saha–Boltzmann population statistics feed Sobolev optical depths for each bound-bound transition, and a Monte Carlo radiative transfer code converts those opacities into synthetic spectra. The decisive property is the temperature window: doubly ionized thorium exists over a narrower temperature range than doubly ionized cerium, so the near-infrared trough appears only while the photosphere stays near 5000–7000 K, which sets both the early observing epoch and the strict upper limit on the lanthanide fraction.","core_discovery":"On the paper's own terms, the discovery is that Th III is the most promising actinide for kilonova spectroscopy because its dense low-lying $5f$-shell energy levels place many allowed transitions at 10,000–25,000 Å with relatively large estimated oscillator strengths. In the authors' radiative transfer models these lines blend into a broad absorption feature near 18,000 Å. The feature appears when the line-forming region has a lanthanide mass fraction $X_{\\rm lan}\\lesssim 6\\times10^{-4}$ and an actinide-to-lanthanide ratio at least as large as the solar r-process ratio, and it fades as the ejecta cools below the temperature window where Th III survives. The authors therefore conclude that detecting this feature would be unambiguous evidence of actinide nucleosynthesis in neutron star mergers, and that the most likely hosts are mergers of unequal-mass neutron stars or black hole-neutron star binaries.","pith_inferences":["Because the required thorium mass fraction scales linearly with the estimated transition strengths, laboratory measurements of the Th III near-infrared lines would shrink the main uncertainty and sharpen the detectability criterion.","Non-LTE effects, which the paper notes can keep some heavy elements in doubly ionized states for weeks, may extend the visible phase of the thorium feature well past the 3.5-day limit found under LTE, making late-time spectra worth examining as well.","The same line-list calibration recipe could be applied to protactinium and heavier actinides once experimental data exist, potentially revealing additional near-infrared actinide features beyond the thorium trough.","Archival space-based near-infrared spectra of GW170817/AT2017gfo, if they cover the 1.8 micron region with sufficient quality, could provide an immediate test of the prediction, although the paper itself does not attempt that identification."],"forward_implications":["A space-based or high-altitude spectrum of a neutron star merger at 1–2.5 days, showing a broad absorption trough near 18,000 Å, would be direct spectroscopic evidence that actinides are synthesized in the event.","The feature is short-lived in the models: it is present from about one day after the merger and fades by about 3.5 days, so early near-infrared follow-up is essential.","The detectability criterion doubles as a composition diagnostic: the line-forming region must be dominated by relatively light r-process elements, with only a small admixture of very neutron-rich material that raises the actinide-to-lanthanide ratio.","Unequal-mass neutron star mergers and black hole–neutron star binaries, which numerical nucleosynthesis models associate with high actinide-to-lanthanide ratios, are the most promising observational targets.","A non-detection does not mean actinides are absent: if the lanthanide fraction is too high or the ratio is only solar-like, the thorium feature stays hidden even when thorium itself is abundant."],"supporting_citations":[{"why":"Provides the measured wavelengths and relative intensities of Th III lines from which the near-infrared gf-values are estimated.","marker":"Engleman 2003"},{"why":"Supplies the semi-empirical gf-values for optical Th III lines used to calibrate the excitation temperature T = 6000 K.","marker":"Biémont et al. 2002"},{"why":"Source for Ra II and Ac III experimental transition data, energy levels, and relative intensities used in the line list.","marker":"Kramida et al. 2023"},{"why":"Original hybrid line list and the first report of a possible Th III feature near 18,000 Å that this paper updates and extends.","marker":"Domoto et al. 2022"},{"why":"Theoretical line list for elements Z = 30–88 that supplies the baseline opacities in the radiative transfer simulations.","marker":"Tanaka et al. 2020"},{"why":"Multicomponent free-expansion model used to construct the Light and Solar abundance patterns adopted in the simulations.","marker":"Wanajo 2018"},{"why":"Defines the r-process residual pattern that the Light and Solar abundance models are tuned to match.","marker":"Prantzos et al. 2020"},{"why":"Documents actinide-boost stars whose thorium-to-lanthanide excess, up to a factor of three, sets the range of actinide enhancement explored in the models.","marker":"Siqueira Mello et al. 2013"},{"why":"Shows that unequal-mass neutron star mergers produce very neutron-rich ejecta with high actinide-to-lanthanide ratios, the conditions the paper argues are needed for detection.","marker":"Fujibayashi et al. 2023"}],"fun_headline_variants":["Thorium's spectral fingerprint could reveal actinide origin in mergers","Kilonova spectra may expose thorium, heaviest detectable element","JWST could spot thorium absorption in neutron-star merger light","Thorium's 18,000 Å feature: a beacon for actinide nucleosynthesis","Detecting thorium in kilonova would confirm r-process actinides"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The detection threshold rests on near-infrared transition strengths for Th III that have never been directly measured; they are recovered from laboratory relative intensities by assuming a single excitation temperature, and the calibration scatter is about a factor of three.","fun_headline_variants_meta":{"raw":{"variants":["Thorium's spectral fingerprint could reveal actinide origin in mergers","Kilonova spectra may expose thorium, heaviest detectable element","JWST could spot thorium absorption in neutron-star merger light","Thorium's 18,000 Å feature: a beacon for actinide nucleosynthesis","Detecting thorium in kilonova would confirm r-process actinides"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000214,"raw_usage":{"total_tokens":1464,"prompt_tokens":1020,"completion_tokens":444,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":636,"completion_tokens_details":{"reasoning_tokens":359}},"tokens_in":636,"tokens_out":444,"duration_ms":4473,"temperature":1.0,"reasoning_tokens":359,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T12:39:12.515908+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the oscillator strengths of the Th III lines near 18,000 Å in the laboratory. If the true values are more than a factor of three smaller than the estimates used here, the required thorium fraction would exceed the actinide-boost range and the predicted trough would not appear, while a high signal-to-noise, space-based spectrum of a suitable kilonova at 1–2.5 days that shows no trough at 18,000 Å despite an inferred actinide-boost composition would also rule the prediction out.","supporting_citations":[{"cited_title":"2003, JQSRT, 78, 1, doi: 10.1016/S0022-4073(02)00173-5","cited_arxiv_id":null,"evidence_quote":"Provides the measured wavelengths and relative intensities of Th III lines from which the near-infrared gf-values are estimated."},{"cited_title":"Ralchenko, Reader, J., & and NIST ASD Team","cited_arxiv_id":null,"evidence_quote":"Source for Ra II and Ac III experimental transition data, energy levels, and relative intensities used in the line list."}],"review_version":1}