{"id":"c1eb5d02-37fd-4468-9fa5-a3d2eb6dfd4f","arxiv_id":"2501.14272","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Spectral fits to three neutron-star X-ray binaries suggest the corona sits close to the neutron star, because a fitted cover fraction is large for stellar seed photons and small for disk seed photons, and NS coronae avoid the pair-production region.","lead":"Using simultaneous NICER and NuSTAR spectra, the authors fit three neutron-star X-ray binaries with reflection models that let seed photons come either from the star or the disk. They report that a fitted cover fraction is high for neutron-star seed photons and low for disk seed photons, and they place neutron-star coronae on a pair-production compactness diagram for the first time.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The new l-theta diagram fixes corona radius R=10 Rg for all sources; using the fitted lamp-post heights h~2-3 Rg shifts the points 3-5x right, so the 'all left of the forbidden region' conclusion is not tested against the paper's own compact corona.","rationale":"The reader's weakest-assumption identification matches my own reading. The paper is a serious spectral analysis: it uses simultaneous NICER and NuSTAR data, modern relxill-family reflection models, and an explicit nthratio cross-check in Appendix A that supports the fcov asymmetry and the seed-photon dependence of the blackbody normalization. The circularity of inferring a lamp-post geometry from a lamp-post-fitted height is real, but it is secondary: the fcov asymmetry already favours a corona close to the neutron star. The decisive weakness is the unvaried corona radius in the new l-theta diagram. Since l scales as Rg/R, replacing the assumed 10 Rg with the fitted 2-3 Rg corona scale is a 3-5x shift, which is large relative to the margin shown in Figure 7 and can move at least the hot literature point toward the pair-production forbidden region. The proposed test—recomputing the diagram at R=h or R=2.5/5 Rg and checking for crossings—would settle whether the 'all NS systems left of the forbidden region' claim survives. Until that sensitivity is shown, the paper's central compactness conclusion is not fully established; the reader's conditional verdict is appropriate, and I do not move it.","tokens_in":23400,"tokens_out":10624,"duration_ms":100588,"concrete_test":"Recompute Figure 7 with R set to the fitted lamp-post height h for the three new sources (h=2.2-3.0 Rg) and, for the literature sample, to R=2.5 Rg and R=5 Rg (boundary-layer scale), leaving all other inputs unchanged; re-plot the slab, hemisphere, and sphere pair-runaway lines and check whether any source crosses into the shaded forbidden region. If any point crosses, the paper must present l as a range or explicitly state that the 'left side' conclusion depends on the R=10 Rg assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The first-time compactness conclusion in Section 5.3 and Figure 7 uses l = 4π (mp/me) (Rg/R) (L/LE) with R assumed to be 10 Rg for every source. This is not a harmless convention: for the three sources analyzed here the same paper fits lamp-post heights h = 2.2-3.0 Rg, and Section 5.1 estimates boundary-layer radii of ~5-7 Rg. If the corona size is comparable to these scales, l increases by a factor of 10/R, i.e. by 1.4-5 (up to ~0.7 dex in log l). The paper does not quote any compactness range or sensitivity check, and Figure 7 shows no error bars. On a log l versus log theta plot such a shift is large compared with the observed separation from the slab pair-runaway forbidden region; in particular, the literature point for 4U 1636-53 with Ecut=135.9 keV (theta ~0.13) sits at high theta and would move substantially toward or into the forbidden area if R were taken as the fitted corona height. The claim that all nine NS systems lie safely on the left side of the pair-production forbidden region is therefore not established for physically smaller coronae; it is contingent on the unvaried R=10 Rg choice. This is the load-bearing support for the 'additional cooling from NS photons' or hybrid-plasma interpretation, so it must be tested before the conclusion can be accepted.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes simultaneous NICER and NuSTAR spectra of three neutron star low-mass X-ray binaries (4U 1636-53, XTE J1739-285, MAXI J1816-195) using reflection models with seed photons supplied either by the neutron star or by the accretion disk. It reports that in the hard spectral state the thcomp cover fraction is high (~50-80%) when the seed photons originate from the NS and low (~3-5%) when they originate from the disk, that the fitted lamp-post height is small (~2-3 Rg), and that the inferred NS blackbody normalization depends strongly on the seed-photon assumption. It also presents, for the first time, a compactness-temperature diagram for nine NS LMXBs, concluding that their coronae all lie to the left of the pair-production forbidden region and are therefore not pair-dominated, possibly because of cooling by NS seed photons, or alternatively contain hybrid thermal/non-thermal electrons.","tokens_in":23729,"tokens_out":8083,"duration_ms":75074,"significance":"If established, the cover-fraction asymmetry would provide a useful observational constraint on the location and geometry of the corona in NS LMXBs, and the compactness diagram would be a novel extension of a tool previously applied mainly to AGN and BH binaries. The data reduction and spectral fitting are generally careful: the fits are statistically acceptable (reduced chi-square 1.02-1.24), the model comparison is clearly presented in Tables 3-6, and Appendix A provides a welcome consistency check using the nthratio correction, which confirms the fcov results. However, the headline compactness conclusion rests on an unvaried assumption about the corona radius that the paper's own fitted scales call into question, and the interpretation of the fitted lamp-post height as independent geometric evidence is partly circular. These issues must be addressed before the central claims can be accepted.","major_comments":[{"comment":"The compactness is computed as l = 4*pi*(mp/me)*(Rg/R)*(L/LE) with the corona radius R fixed to 10 Rg for all nine sources. This is not a harmless convention for NS LMXBs: the same paper fits lamp-post heights h = 2.2-3.0 Rg for the three new sources (Tables 3-5) and estimates boundary-layer radii of ~5-7 Rg (Section 5.1). Since l scales as 1/R, adopting R = 3 Rg would increase l by a factor ~3.3 (about 0.5 dex) for the three new sources, and the literature point for 4U 1636-53 with Ecut = 135.9 keV (theta ~0.13) would move substantially toward or into the pair-production forbidden region. The paper presents no sensitivity analysis, no error bars on Figure 7, and no physical justification for choosing 10 Rg in NS systems. Because the claim that all nine NS systems lie safely to the left of the forbidden region is the load-bearing support for the 'not pair-dominated' and 'additional cooling from NS photons' interpretation, this assumption must be tested against the smaller scales inferred in this work before that conclusion can be accepted.","section":"Section 5.2 and Tables 3-5"},{"comment":"The small corona height h is a fitted parameter of relxilllpCp, a model that already assumes a lamp-post point-source geometry. Using this fitted height as independent evidence favoring the lamp-post geometry or a boundary-layer scenario is partly circular: the model cannot distinguish a point source at height h from an alternative extended geometry, and the boundary-layer scenario is not directly described by relxilllpCp at all. The cover-fraction result from thcomp already supports a compact corona close to the neutron star without this geometric prior; the h value should be presented as a model-dependent scale, not as an independent confirmation. A comparison with an extended-corona or slab reflection model, or an explicit statement of this limitation, would remove the circularity concern.","section":"Section 5.2 and Tables 3-5"}],"minor_comments":[{"comment":"No error bars are shown on the l and theta values in Figure 7; given the uncertainties in distances, fluxes, and kTe/Ecut listed in Table 7, error bars are essential for judging whether points are indeed 'safely' left of the forbidden region.","section":"Figure 7"},{"comment":"The conversion theta = Ecut/(2 mec^2) for sources without a fitted kTe is model-dependent; the relation between Ecut and kTe depends on the Comptonization model used in the original references. The authors should justify this factor or use a consistent model-dependent conversion.","section":"Figure 7 caption"},{"comment":"The literature sample mixes total unabsorbed fluxes with power-law fluxes and uses approximate distances (some marked with *); this heterogeneity should be stated more prominently in Section 5.3, since it directly affects the computed l values.","section":"Table 7"},{"comment":"The statement that 'more than ~50% of the NS photons enter into the corona' refers to the fitted thcomp parameter fcov; the abstract should explicitly note that this is a model-dependent cover fraction, not a direct measurement.","section":"Abstract"},{"comment":"Item (2) of the Conclusion says the corona height is '~3 Rg', while the fits give 2.2-3.0 Rg across the three sources; the range should be quoted consistently.","section":"Conclusion"},{"comment":"There are several typographical issues, including 'bina ry' in the title, 'di erent' for 'different', and 'in the all fits' for 'in all fits'; these should be corrected during revision.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":""},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this is a competent spectral analysis with a genuinely new application—the first l-theta diagram for NS LMXBs—but the two headline claims rest on parameters that are partially circular and partially sensitive to an untested assumption. Worth sending to a referee, not because the conclusions are secure, but because the underlying data and comparison are useful.\n\nWhat's new: the paper fits NICER+NuSTAR spectra of three NS LMXBs with relxillCp/relxilllpCp, comparing NS vs. disk seed photons. The cover-fraction asymmetry (fcov >50% vs. 3-5%) is stable across models for the two hard-state sources. The l-theta diagram with nine NS systems is a useful compilation, and the appendix's nthratio check shows the fcov result is robust to the low-energy cutoff issue. That is good practice.\n\nSoft spots: (1) The small height h from relxilllpCp is used as evidence for a lamp-post geometry, but h is a fitted parameter of a model that already assumes a point source above the disk; that's internally circular. The paper does acknowledge the boundary-layer alternative, so it's not entirely uncritical, but the abstract overstates. (2) The l-theta conclusion assumes R=10 Rg for all sources. The paper itself fits h~2-3 Rg; using those values shifts points 3-5x in compactness, which could move them toward the forbidden region. No sensitivity test is shown. This is the load-bearing support for the 'additional cooling' interpretation, so it should be tested. (3) There's an internal inconsistency: Section 3 says log(xi) for relxillns is fixed at 19, but Table 6 reports a fitted log(xi)=2.00±0.05. Too sloppy to ignore. (4) No scripts or systematic-error propagation, and the literature fluxes in Table 7 are heterogeneous; the diagram has no error bars.\n\nWho it's for: X-ray binary spectral modelers, especially those working on NS accretion and corona geometry. A serious referee should engage with it; the l-theta diagram is worth having in the literature, and the fcov result, even if model-dependent, is a useful constraint. The authors need to test the R sensitivity, fix the log(xi) inconsistency, and soften the circular claim.\n\nRecommendation: accept for peer review, expect major revision.","headline":"Useful first l-theta diagram for NS LMXBs and a stable cover-fraction asymmetry, but the compactness conclusion rests on an untested R=10 Rg assumption and the small-corona evidence is partly circular.","tokens_in":24294,"tokens_out":2097,"would_cite":true,"duration_ms":19301,"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":"In hard-state neutron-star X-ray binaries, over half of the star's seed photons feed a corona just 2–3 Rg out.","keywords":["neutron star low-mass X-ray binaries","X-ray reflection spectroscopy","corona geometry","seed photons","compactness-temperature diagram","pair production","lamp-post model","boundary layer"],"falsifier":"Measure the actual size of the corona in one of these systems, for instance from the energy-dependent time delay between continuum and reflected photons (reverberation), and recompute the compactness with that radius. If the radius is close to the fitted lamp-post height of 2–3 $R_g$, the points would move into or near the pair-production forbidden region, which would falsify the claim that neutron-star coronae are not pair-dominated; if the measured radius is near 10 $R_g$, the paper's placement holds.","tokens_in":23179,"feed_emoji":"🌟","tokens_out":11417,"duration_ms":88151,"temperature":0.7,"pith_summary":"This paper analyzes simultaneous NICER and NuSTAR spectra of three neutron star low-mass X-ray binaries — 4U 1636–53, XTE J1739–285, and MAXI J1816–195 — using reflection models that let the seed photons for Compton scattering come either from the neutron star surface or from the accretion disk. For the two sources in the hard spectral state, the fits imply that more than half of the neutron star's thermal photons are intercepted by the corona when the neutron star supplies the seed photons, whereas only about 3–5% of the disk photons reach the corona when the disk supplies them. Combined with the fitted lamp-post height of only 2–3 $R_g$, the paper concludes that the corona hugs the central neutron star, favoring a lamp-post or boundary-layer geometry rather than an extended corona. It also builds the first compactness–temperature ($\\ell$–$\\theta$) diagram for neutron star systems and finds all nine sources on the safe side of the pair-production forbidden region, indicating that the corona is not pair-dominated — either because neutron star seed photons cool it, or because the corona contains both thermal and non-thermal electrons.","feed_headline":"Half of neutron star photons feed the corona in hard state","feed_subtitle":"NICER+NuSTAR fits place the hot corona 2–3 Rg from the star and keep it clear of pair-production.","key_machinery":"The argument rests on two coupled modeling tools. The first is the relativistic reflection model family relxill, specifically relxillCp and relxilllpCp — where 'lp' denotes a lamp-post geometry with the corona as a point source at height $h$ above the compact object — together with relxillns for a disk illuminated by the neutron-star blackbody, and the thermal Comptonization component thcomp, whose fractional coverage parameter $f_{\\rm cov}$ measures what fraction of the seed photons (from the neutron star or the disk) are intercepted by the corona. The second is the compactness–temperature ($\\ell$–$\\theta$) diagram, in which $\\ell = 4\\pi (m_{\\rm p}/m_{\\rm e})(R_g/R)(L/L_{\\rm Edd})$ measures the radiation compactness of the corona and $\\theta = kT_{\\rm e}/m_{\\rm e} c^2$ its electron temperature; the paper computes $\\ell$ assuming a corona radius $R = 10 R_g$ and compares the points with pair-production runaway limits for slab, hemisphere, and spherical geometries. The small fitted $h$ and the large $f_{\\rm cov}$ for neutron-star seed photons locate the corona near the star, and the position of the points relative to the pair-production forbidden region supports the not-pair-dominated conclusion.","core_discovery":"The central discovery is that choosing the seed-photon source changes the inferred corona and neutron-star emission in a way that pins the corona close to the star. In the hard-state sources 4U 1636–53 and XTE J1739–285, the fractional coverage parameter $f_{\\rm cov}$ is larger than about 0.5 when the seed photons come from the neutron star but only about 0.03–0.05 when they come from the disk; for MAXI J1816–195 the values are about 0.8 and 0.3, respectively. The lamp-post height $h$ is small, roughly 2–3 $R_g$, and the boundary-layer radius estimated from the accretion rate is comparable, supporting a compact corona near the stellar surface. The paper further claims, from the first $\\ell$–$\\theta$ diagram assembled for neutron star LMXBs, that all nine systems lie to the left of the pair-production runaway line, so the coronae are not pair-dominated; the preferred explanation is extra cooling by neutron star surface photons, with a hybrid thermal/non-thermal corona as the alternative. A corollary is that the seed-photon source strongly affects how significant the neutron-star blackbody appears, which the authors use to explain why the neutron-star radiation in MAXI J1816–195 looked weak in earlier work.","pith_inferences":["Editorial inference: the paper computes the compactness with a corona radius of $10 R_g$ for all sources; if the corona is as compact as the fitted height of $2$–$3 R_g$, then $\\ell$ is larger by a factor of roughly 3–5, shifting the points rightward and potentially toward the pair-production boundary, a sensitivity the paper does not test.","Editorial inference: if neutron-star seed photons cool the corona, then sources with brighter neutron-star blackbody emission should sit further from the pair-production line; this correlation could be tested by expanding the sample.","Editorial inference: applying the same seed-photon-aware reflection models to higher-resolution spectra from upcoming X-ray missions could break the degeneracy between a static lamp-post and a rotating boundary-layer corona, for example through iron-line reverberation."],"forward_implications":["For hard-state neutron star LMXBs, a corona within a few gravitational radii of the star means the boundary layer or a jet-base lamp-post should dominate the hard X-ray emission, and reflection models of the disk must include neutron-star illumination alongside coronal illumination.","Because the inferred size of the neutron star's emitting region depends on which seed-photon source is assumed, identifying that source is a prerequisite for measuring blackbody-emitting areas on the neutron star surface.","Neutron-star seed photons provide additional cooling that keeps the corona far from pair production, giving a physical reason why neutron-star LMXB coronae differ from some black-hole binary coronae in the compactness diagram.","For MAXI J1816–195, the low electron temperature, small inner disk radius, and relatively high blackbody temperature together place the source in a soft or transitional state rather than the typical hard state of accreting millisecond X-ray pulsars."],"supporting_citations":[{"why":"Supplies the relxill reflection model family used to fit the disk reflection in all four spectral models.","marker":"García et al. 2014"},{"why":"Provides the relxilllpCp lamp-post variant whose height parameter $h$ gives the corona distance.","marker":"Dauser et al. 2016"},{"why":"Provides relxillns, the model for a disk illuminated by the neutron-star blackbody, used for MAXI J1816–195.","marker":"García et al. 2022"},{"why":"Defines the compactness–temperature diagram and the $\\ell = 4\\pi(m_{\\rm p}/m_{\\rm e})(R_g/R)(L/L_{\\rm Edd})$ scaling used to place the sources.","marker":"Fabian et al. 2015"},{"why":"Proposes that a corona below the pair limit can still be pair-dominated if non-thermal electrons are present; the paper's hybrid-plasma alternative draws on this.","marker":"Fabian et al. 2017"},{"why":"Gives the pair-production runaway lines and forbidden region used to judge whether coronae are pair-dominated.","marker":"Stern et al. 1995"},{"why":"Supplies the boundary-layer radius estimate used to compare with the fitted corona height.","marker":"Popham & Sunyaev 2001"},{"why":"Earlier NuSTAR study of 4U 1636–53 that also derived a small corona height of 2–3 Rg, supporting the same geometry.","marker":"Wang et al. 2017"},{"why":"Contributes most of the literature NS LMXB samples used to build the nine-source compactness diagram.","marker":"Ludlam et al. 2019"},{"why":"Previous NICER+NuSTAR analysis of MAXI J1816–195 that found weak neutron-star radiation, the result the paper reinterprets via seed-photon choice.","marker":"Li et al. 2023a"}],"fun_headline_variants":["Corona hugs neutron star in three LMXB spectra","Neutron star photons dominate corona feeding","Compact coronae in neutron star binaries avoid pair production","Corona near star, fed by star: new spectral fits"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that every corona is 10 gravitational radii across when computing the compactness; if a corona is as small as the fitted height of 2–3 gravitational radii, the same data points shift several-fold toward the pair-production boundary, and the paper does not test that shift.","fun_headline_variants_meta":{"raw":{"variants":["Corona hugs neutron star in three LMXB spectra","Neutron star photons dominate corona feeding","Compact coronae in neutron star binaries avoid pair production","Corona near star, fed by star: new spectral fits"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000725,"raw_usage":{"total_tokens":3348,"prompt_tokens":1138,"completion_tokens":2210,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":754,"completion_tokens_details":{"reasoning_tokens":2143}},"tokens_in":754,"tokens_out":2210,"duration_ms":16273,"temperature":1.0,"reasoning_tokens":2143,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T15:15:35.220831+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the actual size of the corona in one of these systems, for instance from the energy-dependent time delay between continuum and reflected photons (reverberation), and recompute the compactness with that radius. If the radius is close to the fitted lamp-post height of 2–3 $R_g$, the points would move into or near the pair-production forbidden region, which would falsify the claim that neutron-star coronae are not pair-dominated; if the measured radius is near 10 $R_g$, the paper's placement holds.","supporting_citations":[{"cited_title":"E., Poutanen , J., Svensson , R., Sikora , M., & Begelman , M","cited_arxiv_id":null,"evidence_quote":"Gives the pair-production runaway lines and forbidden region used to judge whether coronae are pair-dominated."},{"cited_title":"& Sunyaev , R","cited_arxiv_id":null,"evidence_quote":"Supplies the boundary-layer radius estimate used to compare with the fitted corona height."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Earlier NuSTAR study of 4U 1636–53 that also derived a small corona height of 2–3 Rg, supporting the same geometry."},{"cited_title":"M., Miller , J","cited_arxiv_id":null,"evidence_quote":"Contributes most of the literature NS LMXB samples used to build the nine-source compactness diagram."}],"review_version":1}