{"id":"f1f9f2a1-3e70-4b52-bf25-d82cf31dbb41","arxiv_id":"1908.00655","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"New simulations of unequal-mass neutron star mergers find that only the softest equation of state ejects more than 0.01 solar masses of dynamical ejecta, and that ejecta speed and electron fraction drop as the mass ratio becomes more unequal.","lead":"Twelve new computer simulations of colliding neutron stars with unequal masses show how much matter gets ejected and how the neutrino flash depends on the neutron star equation of state. The results help predict the light and element output of cosmic mergers like the one that produced gravitational wave GW170817.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Bernoulli-criterion systematic (up to 2x ejecta mass) is omitted from the error budget and can move SFHo models across the 0.01 Msun threshold.","rationale":"The reader's weakest assumption correctly identifies the Bernoulli criterion as the key vulnerable step. My independent reading confirms this is the most load-bearing concern: it directly threatens the paper's quantitatively stated headline result, and the paper's own error budget (Section II K, Eq. 64) does not include this known systematic. The concern is concrete and testable from existing data, requiring only a post-processing change of the unbound criterion. The paper is otherwise careful and candid, with a valuable simulation set and useful trend analysis, so the conditional verdict remains appropriate. I therefore do not change the reader's verdict. I considered the mass-ratio/total-mass degeneracy as an alternative concern, but the Bernoulli systematic is more tightly tied to the specific headline claim and is more directly testable; the degeneracy, while real, is partially acknowledged in the neutrino section and affects interpretive strength rather than a single quantitative assertion.","tokens_in":30952,"tokens_out":8589,"duration_ms":89108,"concrete_test":"Recompute the ejecta masses in Table III from the stored simulation output, replacing the unbound condition in Eq. 60 with ut < -1 instead of hut < -1, and report Mej for all twelve models. Then compute the ratio Mej(hut < -1)/Mej(ut < -1) for each model. If any non-SFHo model exceeds 0.01 Msun under ut < -1, or if the two SFHo models above 0.01 Msun fall below, the headline claim fails; if the ratio is near unity and the ordering is preserved, the claim is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that total dynamical ejecta exceeds 0.01 Msun only for SFHo rests entirely on the Bernoulli criterion hut < -1 in Eq. 60. The authors themselves state in Section II J that this criterion can yield up to twice the ejected mass of the ut < -1 criterion (citing Kastaun & Galeazzi 2015), yet the quoted error estimate Eq. 64 (0.5 Mej + 1e-4 Msun) covers only resolution and regridding losses, not this systematic. In Table III, the three non-collapsing SFHo models have Mej = 1.574e-2, 0.839e-2, and 1.802e-2 Msun. If the Bernoulli criterion overestimates by a factor of two, the two models above 0.01 Msun fall below it, and the 'only SFHo' claim becomes 'no model reaches 0.01 Msun.' Conversely, applying the reported 50% resolution error to S12144 (0.839e-2 + 0.5*0.839e-2 = 1.26e-2) moves it above the threshold. The abstract's first finding is therefore not robust under the uncertainties the paper itself reports, and this directly weakens a headline quantitative conclusion.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents twelve new general-relativistic binary neutron star merger simulations performed with the SpEC code, using the SFHo, LS220, and DD2 equations of state and a gray two-moment neutrino transport scheme that evolves neutrino number density in addition to energy and flux. The binary mass ratios range from about 0.76 to 1.0, and the simulations are evolved up to 7.5 ms post-merger or to collapse. The authors report the mass, electron fraction, and asymptotic velocity of the dynamical ejecta, the disk masses, and the neutrino luminosities and angular distributions. Their central claims are that total dynamical ejecta exceeds 0.01 solar masses only for the SFHo equation of state, that ejecta electron fraction and average velocity decrease with increasing mass asymmetry, that disk mass increases with mass asymmetry and equation-of-state stiffness, and that neutrino emission is strongest for the softest equation of state.","tokens_in":31205,"tokens_out":3541,"duration_ms":39349,"significance":"If the reported trends hold, this simulation set is a valuable addition to the effort of connecting binary parameters to kilonova-relevant ejecta properties. The strengths of the paper include the use of an established code (SpEC), a neutrino transport scheme with improved lepton-number conservation, a systematic variation of equation of state and mass ratio, and a candid discussion of numerical limitations. The authors explicitly quantify their resolution-error estimate, state that the disk mass has not converged, and describe the known systematic uncertainty in the Bernoulli unbound criterion. Those admissions are commendable, but they also directly bear on the headline quantitative claims, and the current presentation does not fully reconcile the headline statements with the stated uncertainties.","major_comments":[{"comment":"The headline claim that total dynamical ejecta exceeds 0.01 solar masses only for SFHo is not robust under the uncertainties stated in the paper. The Bernoulli criterion hut<-1 is acknowledged in Section II J to yield as much as twice the ejected mass of the ut<-1 criterion, citing Kastaun and Galeazzi 2015, yet this systematic is not included in the error estimate Delta Mej = 0.5 Mej + 1e-4 solar masses in Eq. (64), which covers only resolution and regridding effects. In Table III, S12132 and S12156 have total ejecta masses of 1.574e-2 and 1.802e-2 solar masses; if the Bernoulli criterion overestimates by a factor of two, both fall below 0.01 solar masses, while applying the 50 percent resolution error to S12144 raises it above 0.01 solar masses. The abstract and conclusion should either be reworded to present this as a tentative statement, or the analysis should report both unbound criteria and incorporate the criterion choice into the error budget.","section":"Section II J, Eq. (60), Eq. (64), Table III"},{"comment":"The claimed monotonic decrease of ejecta electron fraction and average asymptotic velocity with increasing mass asymmetry is inferred from three SFHo models (S12132, S12144, S12156) and from even fewer non-collapsing runs for the other equations of state, with no measure of statistical significance. Because the lower-mass neutron star is held fixed while the higher mass is varied, total binary mass increases with mass asymmetry, so the attribute 'mass asymmetry' is not cleanly separated from total mass; Section III C correctly notes this confounding for neutrino luminosity, but the ejecta trends are presented without a similar caveat. Given the admitted non-monotonic ejecta-mass behavior in Section III B 1 and the 50 percent error estimate, the paper should either provide explicit error bars on the plotted distributions or reframe these statements as suggestive trends rather than robust findings.","section":"Section III B 2, Figs. 8 and 9, Table I"},{"comment":"The conclusion that disk mass increases with mass asymmetry and equation-of-state stiffness is based on a disk-mass definition with an arbitrary density threshold, and the paper itself states that Mdisk has not converged by 7.5 ms post-merger. Since the ordering of disk masses could change if the simulations were evolved longer, and since this trend is used in the discussion of kilonova powering, the conclusion should be labeled as provisional or accompanied by a convergence check, such as a comparison at multiple post-merger times or at a second resolution.","section":"Section III B 3, Eq. (65), Fig. 10"}],"minor_comments":[{"comment":"The word 'preformed' should be 'performed'; the same typo appears in both the abstract and the introduction.","section":"Abstract and Section I"},{"comment":"In the sentence introducing the number-density evolution equation, 'te evolution' should be 'the evolution'.","section":"Section II D"},{"comment":"The subscript 'Mon' in Eq. (60) appears to be a typo for 'Mej'.","section":"Section II J"},{"comment":"The text says 'we can make predictions on expected trends' and later describes results that do not fit the narrative; some transition between expectation and outcome would improve readability.","section":"Section III B 1"}],"recommendation":"major_revision","confidential_remarks":"The simulation data appear valuable and the paper is generally careful, but the headline threshold claim is sensitive to the unbound-matter criterion and to resolution uncertainty. I do not see a fundamental flaw in the simulation methodology; rather, the presentation overstates the robustness of several conclusions. A major revision that adds explicit caveats, reports the alternative unbound criterion, and quantifies the confound between mass ratio and total mass would make the paper suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a genuinely useful simulation survey. Twelve new runs, three EOS, mass ratios down to 0.76, and a transport scheme that evolves number density so lepton number is conserved. The trends in Ye, velocity, disk mass, and neutrino emission are the kind of thing the kilonova community wants, and the paper is refreshingly candid about single-resolution runs, 50% mass errors, and non-converged disk masses. That honesty is real and it earns credit.\n\nThe soft spot is exactly where the abstract is strong. The first finding—dynamical ejecta exceeds 0.01 Msun only for SFHo—rests entirely on the Bernoulli criterion hut<-1. Section II J tells us this criterion can give up to twice the mass of the ut<-1 criterion, citing Kastaun & Galeazzi, and then Section II K assigns a 50% error that covers resolution and regridding only. Those two statements interact badly. Applying the factor-two systematically to the SFHo numbers moves two of the three non-collapsing models below 0.01 Msun; applying the 50% resolution error to S12144 moves it above. So \"only SFHo\" is not a robust statement at the paper's own stated uncertainties. The authors may be right that Bernoulli is more accurate in SpEC, but the abstract should not sell a threshold that the error budget does not support.\n\nSecondary concerns: the mass-ratio trends are partly mass-ratio and partly total-mass trends, because the total mass is not fixed; the paper acknowledges this but the abstract phrases things as pure mass-ratio dependence. The neutrino luminosity null results have no error bars or statistical comparison; three runs per EOS is enough to report trends, not to assert no dependence. The non-monotonic ejecta values in Table III, e.g. D12144 lower than D12132 and D12156, also suggest scatter that a three-point trend cannot resolve.\n\nNone of this kills the paper. The new data and the transport scheme are valuable, the presentation is clear, and the limitations are mostly stated somewhere. But the headline claim and the neutrino non-dependence should be softened or quantified before I'd sign off.\n\nRecommendation: send to peer review. A competent referee can handle the Bernoulli-systematic issue, and the community will cite this simulation set either way.","headline":"Useful new simulation set for unequal-mass BNS mergers, but the headline 'only SFHo exceeds 0.01 Msun' is not robust to the Bernoulli/ut systematic the authors themselves cite.","tokens_in":31742,"tokens_out":2290,"would_cite":true,"duration_ms":23097,"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":"Twelve unequal-mass neutron-star merger simulations find that dynamical ejecta exceed 0.01 solar masses only for the softest equation of state, with ejecta electron fraction and velocity falling as mass asymmetry grows.","keywords":["binary neutron star mergers","unequal mass ratio","neutrino transport","dynamical ejecta","electron fraction","kilonova","numerical relativity","r-process nucleosynthesis"],"falsifier":"Recompute the ejected mass with the geodesic criterion $u_t < -1$ instead of $h u_t < -1$; if no SFHo model then exceeds $0.01\\,M_\\odot$, or a DD2 or LS220 model does, the central ejecta-mass claim fails. Evolving the non-collapsing remnants beyond 7.5 ms would also test whether matter flagged unbound near the grid edge is truly escaping.","tokens_in":30732,"feed_emoji":"💥","tokens_out":9585,"duration_ms":89102,"temperature":0.7,"pith_summary":"This paper tries to establish reliable relations between the parameters of a merging neutron-star binary and what the merger emits: how much mass is thrown off, how neutron-rich and fast that material is, and how many neutrinos are radiated. The authors run twelve simulations spanning mass ratios from about 0.76 to 1 and total masses from about 2.5 to 2.9 solar masses, using three nuclear-theory equations of state and a transport scheme that tracks neutrino number density so lepton number is conserved. They find that dynamical ejecta (material thrown off in the first milliseconds of the merger) exceed 0.01 solar masses only for the softest equation of state, SFHo, and that average ejecta electron fraction and velocity both decrease as binary asymmetry increases. These trends matter because they connect gravitational-wave-inferred binary properties to the kilonova light curves and r-process nucleosynthesis expected after such mergers.","feed_headline":"Only the softest neutron-star models eject significant mass","feed_subtitle":"Twelve merger simulations map how mass asymmetry and nuclear stiffness shape ejecta and neutrino emission.","key_machinery":"The central machinery is a gray two-moment neutrino transport scheme that evolves, for each neutrino species, the energy and momentum densities together with the neutrino number density, giving an on-the-fly estimate of the average neutrino energy and exact conservation of total lepton number. The unbound-ejecta accounting uses the relativistic Bernoulli condition $h u_t < -1$ applied to matter more than 50 solar masses from the remnant, integrated up to 7.5 ms after merger.","core_discovery":"Across twelve simulations, the paper reports that the total dynamical ejecta mass exceeds $0.01\\,M_\\odot$ only for the SFHo equation of state, with a weak dependence on mass ratio; LS220 and DD2 eject less. Ejecta electron fractions span roughly $0.06$ to $0.48$ with a mean near $0.2$, increase over time as neutrino irradiation acts on the outflow, and decrease with binary asymmetry. Asymptotic ejecta velocities span roughly $0.05c$ to $0.7c$, with averages between $0.2c$ and $0.3c$, and also decrease with asymmetry. Remnant disk mass at 7.5 ms post-merger increases with both binary asymmetry and stiffness of the equation of state, while the disk electron fraction is higher for softer equations of state. Neutrino emission is strongest for the softest equation of state; electron-neutrino luminosity and its angular distribution show no significant mass-ratio dependence, while heavy-lepton neutrino luminosity increases with asymmetry in these models.","pith_inferences":["Because the paper's unbound-mass criterion can overestimate ejecta by up to a factor of two, the absolute dynamical masses may be systematically high even if the ranking across equations of state survives.","I infer from the velocity trend that core-bounce ejecta, not tidal ejecta, set the average speed in near-equal-mass mergers; if so, highly asymmetric binaries may lack the fast blue kilonova component.","These trends imply that a gravitational-wave event with measured mass ratio could be used to predict whether its kilonova is lanthanide-rich mainly from the compactness of the neutron stars, a testable prediction once more events with broadband light curves are observed.","A natural extension would be to run the same transport scheme beyond 7.5 ms, since the paper notes its disk mass is still growing; if disk-driven winds dominate the total ejecta, the dynamical-ejecta trends in this paper would constrain only the early kilonova peak."],"forward_implications":["Kilonova models for unequal-mass binaries should use dynamical ejecta masses around $10^{-3}$ to $10^{-2}\\,M_\\odot$, exceeding $0.01\\,M_\\odot$ only when the equation of state is soft.","Greater mass asymmetry lowers both the average velocity and electron fraction of the ejecta, which should push kilonova emission redder and shift r-process yields toward heavier, more neutron-rich elements.","Disk masses at 7.5 ms grow with asymmetry and stiffness, and all but the promptly collapsing equal-mass LS220 model leave a disk large enough to supply the GW170817 kilonova if 25 to 50 percent of the disk is later ejected.","Electron-neutrino luminosity is insensitive to mass ratio, so neutrino-driven wind properties may be approximately universal across mass ratio for a fixed equation of state.","Heavy-lepton neutrino luminosity rises with mass asymmetry in this model set, although the paper attributes part of that dependence to the accompanying increase in total mass."],"supporting_citations":[{"why":"The equal-mass predecessor simulations this work extends, using the same transport method and providing an error calibration for ejecta measurements.","marker":"[30]"},{"why":"Introduces the gray two-moment neutrino transport scheme with number-density evolution used in all twelve simulations.","marker":"[61]"},{"why":"Earlier two-moment transport simulations of unequal-mass mergers with SFHo and DD2, providing the main comparison for ejecta and neutrino-luminosity trends.","marker":"[31]"},{"why":"Leakage-scheme unequal-mass simulations that found no equation of state above 0.01 solar masses, supplying the contrasting baseline for the SFHo result.","marker":"[32]"},{"why":"Viscous-hydrodynamics and leakage simulations used as the reference for ejecta masses and disk-mass estimates.","marker":"[46]"},{"why":"Shows the Bernoulli criterion can give up to twice the ejected mass of the $u_t<-1$ criterion, quantifying the main uncertainty on the ejecta-mass claim.","marker":"[78]"},{"why":"Supplies the resolution-error estimate for unequal-mass ejecta that underlies the paper's roughly 50 percent error bar.","marker":"[79]"},{"why":"Provides the prompt-collapse threshold masses used to classify which remnants collapse and therefore suppress mass ejection.","marker":"[74]"}],"fun_headline_variants":["Soft nuclear EOS alone yields significant neutron star ejecta","Only soft equation of state ejects >0.01 solar mass in mergers","Soft EOS: the only neutron-star mergers with massive ejecta","Neutron-star ejecta mass hinges on nuclear equation of state"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claim depends on identifying unbound matter with the Bernoulli condition $h u_t < -1$, which the authors note can overestimate the ejected mass by as much as a factor of two relative to the alternative $u_t < -1$ criterion.","fun_headline_variants_meta":{"raw":{"variants":["Soft nuclear EOS alone yields significant neutron star ejecta","Only soft equation of state ejects >0.01 solar mass in mergers","Soft EOS: the only neutron-star mergers with massive ejecta","Neutron-star ejecta mass hinges on nuclear equation of state"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000643,"raw_usage":{"total_tokens":3020,"prompt_tokens":1072,"completion_tokens":1948,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":688,"completion_tokens_details":{"reasoning_tokens":1873}},"tokens_in":688,"tokens_out":1948,"duration_ms":15602,"temperature":1.0,"reasoning_tokens":1873,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:40:21.752127+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the ejected mass with the geodesic criterion $u_t < -1$ instead of $h u_t < -1$; if no SFHo model then exceeds $0.01\\,M_\\odot$, or a DD2 or LS220 model does, the central ejecta-mass claim fails. Evolving the non-collapsing remnants beyond 7.5 ms would also test whether matter flagged unbound near the grid edge is truly escaping.","supporting_citations":[{"cited_title":"Foucart, E","cited_arxiv_id":null,"evidence_quote":"Introduces the gray two-moment neutrino transport scheme with number-density evolution used in all twelve simulations."},{"cited_title":"Bauswein, T","cited_arxiv_id":null,"evidence_quote":"Provides the prompt-collapse threshold masses used to classify which remnants collapse and therefore suppress mass ejection."}],"review_version":1}