{"id":"ba8856e2-e64b-4ae7-9ca0-4c03eddf321d","arxiv_id":"1908.08462","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"The rms amplitude of kHz QPOs in 4U 1636-53 shows a two-dimensional structure in frequency-energy space, with the lower QPO peaking at ~760 Hz and breaking at ~12 keV, and the upper QPO extrapolating to zero at ~1400 Hz.","lead":"Using RXTE data, this paper maps how the amplitude of kilohertz X-ray oscillations in the neutron-star binary 4U 1636-53 changes with both oscillation frequency and photon energy, the first such two-dimensional view for any source. The lower oscillation peaks near 760 Hz and drops above 12 keV, while the upper oscillation fades toward zero near 1400 Hz, offering a neutron-star mass constraint.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Abstract's M_NS ≤ 1.6 M⊙ cannot be derived from the paper's own Eq. 5, which yields 1.77–2.19 M⊙ for the j range adopted in §4.3.","rationale":"The paper's empirical core is solid and deserves credit: the 2D rms map is a genuinely new presentation, built from 580 RXTE observations with explicit background-systematics checks (the factor-of-two background test in §3.1 leaves the trends unchanged), and the tied conditional fit in §3.2 (ν_i = 1398 ± 24 Hz, χ²_ν = 0.6) provides a statistically acceptable description of the data. The energy-independence of the zero-crossing is supported by the claim that ν_i is consistent across bands when left free. The load-bearing problem is the abstract's 1.6 M⊙. It cannot be derived from Eq. 5 with the j range the text adopts; 1.77 M⊙ (j = 0.17) and 2.19 M⊙ (j = 0.52) are the paper's own limits, and 1.6 M⊙ requires j = 0. This matches the reader's strongest_claim, and I regard it as more load-bearing than the reader's chosen weakest_assumption (the sonic-point/ISCO identification) because the arithmetic inconsistency is demonstrable from the text alone and unacknowledged, whereas the sonic-point assumption is explicitly conditional in the abstract ('under the assumption that this QPO reflects the Keplerian frequency at the inner edge of the accretion disc'). The sonic-point caveat is real — if the upper QPO frequency is not the ISCO Keplerian frequency, no mass constraint follows — but it is a flagged model choice rather than an internal error. The fix is minor and does not threaten the paper's contribution: report the correct range (or state j = 0 explicitly), and keep the van Doesburgh et al. (2018) caveat that the limits are likely underestimated for a fast rotator. The CONDITIONAL verdict stands unchanged.","tokens_in":125,"tokens_out":16461,"duration_ms":187789,"concrete_test":"Recompute Eq. 5 (§4.3): M_NS = 2.2 × (1000/ν_i) × (1 + 0.75 j) M⊙ with ν_i = 1398 Hz. The results are 1.77 M⊙ for j = 0.17 and 2.19 M⊙ for j = 0.52, matching the text; no j ≥ 0.17 yields 1.6 M⊙, which requires j = 0. As a second step, check the journal version of the paper to see whether the abstract's 1.6 M⊙ was corrected; if it remains, the overclaim stands and the abstract must be revised to report the range 1.77–2.19 M⊙ together with the van Doesburgh et al. (2018) caveat that these limits are likely underestimated for a fast rotator.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim as published — M_NS ≤ 1.6 M⊙ in the abstract — cannot be obtained from the paper's own analysis. In §4.3, Eq. 5 is evaluated with ν_max_QPO = ν_i = 1398 ± 24 Hz and spin parameters j = 0.17 and j = 0.52 from Morsink & Stella (1999), yielding M_NS ≤ 1.77 M⊙ and M_NS ≤ 2.19 M⊙. The abstract's 1.6 M⊙ corresponds to j ≈ 0, a non-rotating star, which is outside the adopted range. This is an internal inconsistency: the headline number overstates the constraint derived in the paper. Moreover, the paper itself notes (citing van Doesburgh et al. 2018) that Eq. 5 neglects oblateness and internal structure for fast rotators such as 4U 1636−53 (burst-oscillation spin ≈ 581 Hz), so the true limits are likely weaker than even the text values. The empirical content — the 2D rms map and the energy-independent zero-crossing of the linear fit at ~1398 Hz — is not affected; what fails is the support for the specific number in the abstract.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes 580 RXTE observations of 4U 1636-53 and constructs, for the first time for any kHz QPO source, the two-dimensional distribution of the fractional rms amplitude of the lower and upper kHz QPOs as a function of QPO frequency and photon energy. The main reported findings are: the lower QPO rms peaks near 760 Hz and drops above roughly 12 keV; the upper QPO rms decreases with frequency and has a local hump near 770 Hz; and a linear fit to the upper-QPO rms versus frequency yields a common zero-crossing at 1398 +/- 24 Hz in all energy bands. Interpreting this zero-crossing as the Keplerian frequency at the ISCO, the paper claims a neutron-star mass upper limit. The paper also fits an approximate Comptonisation model to the lower-QPO rms and lag spectra and discusses the implications for the radiative mechanism behind the QPOs.","tokens_in":20653,"tokens_out":5982,"duration_ms":55841,"significance":"If the empirical results hold, the two-dimensional rms maps are a valuable new observational product for spectral-timing models of kHz QPOs, and the energy-independent zero-crossing of the upper-QPO rms amplitude is a novel and falsifiable constraint. The analysis is careful in several respects: QPO identification criteria are consistent, the background-systematic effect is checked by varying the assumed background rate between zero and twice the pcabackest value, and upper limits are reported transparently. The main weakness is that the headline mass limit in the abstract does not follow from the paper's own calculation in Section 4.3, and the zero-crossing itself inherits a specific linear-model assumption. Neither issue undermines the empirical maps, but both must be addressed before the central mass claim can be accepted as stated.","major_comments":[{"comment":"The abstract headline mass limit M_NS <= 1.6 M_sun is not supported by the paper's own calculation. In Section 4.3, Eq. (5) is evaluated with nu_max_QPO = nu_i = 1398 +/- 24 Hz and the extreme spin parameters j = 0.17 and j = 0.52 from Morsink & Stella (1999), giving M_NS <= 1.77 M_sun and M_NS <= 2.19 M_sun, respectively. The value 1.6 M_sun corresponds to j ~ 0, which is outside the adopted range, and the text's own caveat that Eq. (5) neglects oblateness for fast rotators implies the true limits are weaker, not stronger. The abstract must be corrected to report the values actually derived, or the j = 0 choice must be justified.","section":"Abstract and Section 4.3, Eq. (5)"},{"comment":"The zero-crossing nu_i = 1398 +/- 24 Hz, which drives the mass interpretation, is the zero of a linear function fitted to the rms-versus-frequency relation of the upper kHz QPO and extrapolated about 175 Hz beyond the highest measured average frequency (approximately 1223 Hz in Table 1). Because Table 3 contains several upper limits that are treated in the fit as zero rms with the 95% upper limit as the error bar, the exact value of nu_i is sensitive to both the functional form and the treatment of upper limits. Please add a robustness test with a different decay shape, such as a power law or an exponential, or excluding the 1025-1165 Hz interval, and report how nu_i and the resulting mass limit change, or explicitly state in the abstract that the mass limit depends on the linear-model assumption.","section":"Section 3.2 and Eq. (3)"}],"minor_comments":[{"comment":"The channel-selection cell for the 4th epoch in the 6.0 keV row reads 11-05, which appears to be a typo for 11-15; please verify and correct.","section":"Table 2"},{"comment":"In the upper kHz QPO row for the 930-1025 Hz interval, the entry 21.1 +/- 6.5(28.7* is missing its closing parenthesis; the asterisk should be followed by ')' as in the other upper-limit entries.","section":"Table 3"},{"comment":"The text says the linear function used to fit the upper kHz QPO 'decreases with energy', but the independent variable in Fig. 4 is QPO frequency; this should read 'decreases with frequency'.","section":"Section 3.2, page 6"},{"comment":"The Comptonisation-model comparison is explicitly approximate, with six model parameters and only five data points after excluding two energy bins, and the two shown solutions are degenerate; the abstract's statement that the properties of the kHz QPOs 'can constrain the thermodynamic properties of the inner accretion flow' is stronger than this demonstration supports and should be softened or qualified.","section":"Section 4.5 and Fig. 10"},{"comment":"The caption contains the doubled article in 'using the the model of Kumar & Misra (2014)'; this should be corrected.","section":"Section 4.5, caption of Fig. 10"}],"recommendation":"major_revision","confidential_remarks":"The empirical analysis appears sound and the main defect is the abstract/text mass-limit inconsistency, which is fixable. If the authors correct the headline value, add the requested robustness test for the linear zero-crossing, and soften the Comptonisation-model claim, I would be satisfied with the paper. I see no reason to question the integrity of the analysis."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Plain take: the two-dimensional rms map is genuinely new, and the analysis behind it is careful. But the abstract's headline claim — M_NS ≤ 1.6 M_sun — does not follow from the paper's own equation. The text gives 1.77–2.19 M_sun for the adopted spin range, and the authors themselves note that those numbers are likely underestimates because the formula ignores oblateness. This needs fixing before the paper is in its final form.\n\nWhat is new: earlier work measured rms amplitude versus energy or versus frequency separately, marginalizing over the other variable. This paper builds the joint map for 4U 1636−53 and shows that the slope of the rms-energy relation changes with QPO frequency — the lower-kHz-QPO slope peaks around 750 Hz and then declines, while the upper-kHz-QPO slope decreases monotonically with frequency. The 2D maps in Figures 8 and 9 are a useful new diagnostic for any model of kHz QPOs. The data handling is solid: consistent QPO detection criteria, attention to the background model via the pcabackest tool with a zero-to-2× systematic test, and transparent reporting of upper limits instead of forcing detections. I believe the empirical core.\n\nSoft spots: the mass constraint is the main one. The zero-crossing ν_i = 1398 ± 24 Hz comes from a linear fit to the upper-QPO rms decay; if the true decay is curved, the zero-crossing moves. The identification of that frequency with the ISCO is an assumption from the sonic-point model, not something derived in this paper. The abstract's 1.6 M_sun is not consistent with Eq. (5) evaluated for the j values adopted in §4.3; it appears to come from j=0. The paper should either quote the range 1.77–2.19 M_sun with the appropriate caveats, or drop the mass claim from the abstract. The Kumar & Misra model comparison in §4.5 is explicitly approximate and degenerate; the authors pick parameter sets by hand and say a proper fit is left to a follow-up. That is acceptable for a qualitative discussion but no more. Minor: the high-energy behavior of the upper QPO rms-energy relation is ambiguous ('continues increasing or levels off'), and several of the highest-energy points are upper limits; the paper says as much.\n\nBottom line: the empirical map is worth publishing and should be refereed, but the mass constraint as advertised is overstated. The fix is straightforward: correct the abstract and separate the robust observational result from the model-dependent interpretation. Send it to review.","headline":"Genuinely new 2D rms map, careful empirical work, but the abstract's 1.6 M_sun mass limit is not supported by the paper's own Eq. (5).","tokens_in":21230,"tokens_out":3053,"would_cite":true,"duration_ms":28572,"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":"The rms amplitude of the upper kilohertz QPO in 4U 1636−53 extrapolates to zero at the same frequency in every energy band, which the paper uses to bound the neutron-star mass.","keywords":["kHz quasi-periodic oscillations","neutron-star X-ray binaries","accretion discs","rms amplitude","frequency-energy space","4U 1636−53","Comptonisation"],"falsifier":"A detection of the upper kHz QPO in 4U 1636−53 above about 1400 Hz with significantly nonzero rms in any energy band would falsify the zero-crossing claim; an independent neutron-star mass measurement above the derived 1.77–2.19 solar-mass range would contradict the bound if the Keplerian identification is kept.","tokens_in":20172,"feed_emoji":"🛰️","tokens_out":15485,"duration_ms":120899,"temperature":0.7,"pith_summary":"This paper presents, for the first time for any source of kilohertz quasi-periodic oscillations (kHz QPOs), the two-dimensional behaviour of the fractional rms amplitude of the kHz QPOs as a function of both QPO frequency and photon energy, using 580 archival RXTE observations of the neutron-star low-mass X-ray binary 4U 1636−53. The lower kHz QPO's rms amplitude peaks near 760 Hz and near 12 keV before falling, while the upper kHz QPO's rms amplitude declines with frequency and is consistent with reaching zero at the same frequency, about 1400 Hz, in all energy bands. Under the sonic-point assumption that this frequency marks the inner edge of the accretion disc reaching the innermost stable circular orbit, the paper converts the zero-crossing into an upper limit on the neutron-star mass: the text's Equation (5) gives $M_{\\rm NS} \\le 1.77\\,M_\\odot$ for dimensionless spin $j = 0.17$ and $M_{\\rm NS} \\le 2.19\\,M_\\odot$ for $j = 0.52$. The abstract states $M_{\\rm NS} \\le 1.6\\,M_\\odot$, a value not reproduced by the calculation in the text. The paper also ties the slope of the rms-energy spectrum to the changing QPO frequency and shows that a Comptonising-cloud model can reproduce the rms and time-lag spectra.","feed_headline":"Upper kHz QPO's amplitude vanishes near 1400 Hz in all energy bands","feed_subtitle":"If 1400 Hz marks the disc's inner edge, the neutron star in 4U 1636−53 weighs at most about 2 solar masses.","key_machinery":"The central object is the two-dimensional rms-amplitude surface, $\\mathrm{rms}(\\nu_{\\rm QPO}, E)$, assembled by fitting Lorentzians to averaged power spectra in nine frequency intervals for the upper kHz QPO, eleven intervals for the lower kHz QPO, and seven energy bands. The argument hinges on the intercept $\\nu_i$ of the linear fit to the upper QPO's rms-versus-frequency relation, with the intercept tied across energy bands, and on the sonic-point mass formula that converts $\\nu_i$ into a neutron-star mass limit. The model of a Comptonising cloud, with parameters such as cloud size, optical depth, electron temperature, seed-photon temperature, feedback fraction, and heating-rate oscillation amplitude, is used to connect the rms and time-lag spectra.","core_discovery":"The central claim is that the upper kHz QPO's fractional rms amplitude, measured in seven energy bands from about 3 to 20 keV, decreases with QPO frequency and extrapolates to zero at the same frequency, $\\nu_i = 1398 \\pm 24$ Hz, in every energy band. Because the QPO frequency is the same in all energy bands, the authors interpret this common zero-crossing as the frequency at which the inner edge of the accretion disc reaches the innermost stable circular orbit, so the upper QPO can no longer be excited. Using the sonic-point mass relation $M_{\\rm NS} \\le 2.2\\,(1000\\,\\mathrm{Hz}/\\nu_{\\max})\\,(1 + 0.75\\,j)\\,M_\\odot$, they derive $M_{\\rm NS} \\le 1.77\\,M_\\odot$ for $j=0.17$ and $M_{\\rm NS} \\le 2.19\\,M_\\odot$ for $j=0.52$. The same analysis shows that the lower kHz QPO's rms amplitude peaks near 760 Hz and near 12 keV and then declines, and that both QPOs share a Gaussian feature near 750–770 Hz, which the paper reads as evidence of a common radiative mechanism with an additional component acting only on the upper QPO.","pith_inferences":["The abstract's $1.6\\,M_\\odot$ limit does not follow from the paper's own formula and fitted $\\nu_i$; a reader should quote the text's $1.77\\text{--}2.19\\,M_\\odot$ range instead.","If the upper kHz QPO is produced by a beat-frequency or resonance mechanism rather than being the Keplerian frequency at the inner edge, the same ~1400 Hz zero-crossing would constrain those models rather than the neutron-star mass.","Re-fitting the published rms-versus-frequency table with a curved decay would test whether the zero-crossing, and therefore the mass bound, is an artifact of the linear model.","Applying the same frequency-energy analysis to other sources with long RXTE archives could map the implied zero-crossing frequency across the neutron-star population, providing a population-level test of the mass interpretation."],"forward_implications":["If the extrapolation is correct, the upper kHz QPO in 4U 1636−53 has a hard frequency ceiling near 1400 Hz, set by the inner edge of the accretion disc.","The lower kHz QPO's rms amplitude drops above about 12 keV, meaning the radiative mechanism behind this QPO becomes less efficient at energies where only the Comptonising component contributes.","The slope of the rms-energy relation changes with QPO frequency for both QPOs, so rms amplitude should be treated as a function of both frequency and energy, not just one of them.","The shared Gaussian feature near 750–770 Hz supports a common radiative mechanism acting on both QPOs, with an extra frequency-dependent component acting only on the upper QPO.","If the Comptonising-cloud model applies, the fitted parameters constrain the size, optical depth, temperature, and feedback properties of the inner accretion flow."],"supporting_citations":[{"why":"It supplies the sonic-point model in which the maximum QPO frequency is the Keplerian frequency at the inner edge of the accretion disc, the assumption that converts the zero-crossing into a mass limit.","marker":"Miller et al. 1998"},{"why":"It gives the mass upper-limit formula $M_{\\rm NS} \\le 2.2(1000/\\nu_{\\max})(1+0.75j)\\,M_\\odot$ used to evaluate the bound.","marker":"Kluzniak et al. 1990"},{"why":"It provides the two values of the dimensionless spin $j$ ($0.17$ and $0.52$) used to compute the mass limits.","marker":"Morsink & Stella 1999"},{"why":"It discusses corrections for neutron-star oblateness and rotation, cited by the paper when warning that its mass limits are likely underestimated.","marker":"van Doesburgh et al. 2018"},{"why":"It is the earlier rms-versus-frequency analysis of this source that supplies the frequency intervals, the QPO identification method, and the 'hump' result extended here.","marker":"Ribeiro et al. 2017"},{"why":"It is the Comptonising-cloud model used to reproduce the lower QPO's rms and time-lag spectra.","marker":"Kumar & Misra 2014"},{"why":"It supplies the time-lag spectra and the energy-band definitions used alongside the rms spectra in the model fits.","marker":"de Avellar et al. 2013"}],"fun_headline_variants":["kHz QPO amplitude vanishes at same frequency across all energies","Common QPO cutoff frequency in every energy band limits neutron star mass","Upper kHz QPO fades at 1400 Hz in all bands, capping the neutron star mass","Energy-independent cutoff of kHz QPO sets neutron star mass limit","Frequency-energy map of kHz QPOs shows same cutoff in all bands"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The mass constraint collapses if the upper kHz QPO frequency is not the Keplerian orbital frequency at the inner edge of the accretion disc, or if the straight-line fit that puts the zero of its amplitude at about 1400 Hz is not the correct shape.","fun_headline_variants_meta":{"raw":{"variants":["kHz QPO amplitude vanishes at same frequency across all energies","Common QPO cutoff frequency in every energy band limits neutron star mass","Upper kHz QPO fades at 1400 Hz in all bands, capping the neutron star mass","Energy-independent cutoff of kHz QPO sets neutron star mass limit","Frequency-energy map of kHz QPOs shows same cutoff in all bands"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000828,"raw_usage":{"total_tokens":3725,"prompt_tokens":1160,"completion_tokens":2565,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":776,"completion_tokens_details":{"reasoning_tokens":2468}},"tokens_in":776,"tokens_out":2565,"duration_ms":21608,"temperature":1.0,"reasoning_tokens":2468,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:38:53.638527+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A detection of the upper kHz QPO in 4U 1636−53 above about 1400 Hz with significantly nonzero rms in any energy band would falsify the zero-crossing claim; an independent neutron-star mass measurement above the derived 1.77–2.19 solar-mass range would contradict the bound if the Keplerian identification is kept.","supporting_citations":[{"cited_title":"M., M \\' e ndez M., Zhang G., Sanna A., 2017, @doi [ ] 10.1093/mnras/stx1686 , 471, 1208","cited_arxiv_id":null,"evidence_quote":"It is the earlier rms-versus-frequency analysis of this source that supplies the frequency intervals, the QPO identification method, and the 'hump' result extended here."}],"review_version":1}