{"id":"a104042a-69df-4e63-86d4-6b588cac62f6","arxiv_id":"2508.15624","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A MESA grid of 338 helium-star plus 1.4 solar-mass neutron star binaries at solar metallicity produces only 0.01% of double neutron stars with total mass at or above 3 solar masses, implying that the unstable mass-transfer channel cannot explain GW190425.","lead":"The paper models how double neutron star binaries form from helium star plus neutron star pairs at solar metallicity, using a grid of 338 MESA binary models. It finds that heavy systems like the gravitational-wave event GW190425 are extremely rare in this channel, and argues that the proposed unstable mass-transfer formation route is excluded by the models.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claim that unstable mass transfer is ruled out is not supported: the grid starts after CE and never simulates an unstable second RLO, so 'none undergo unstable MT' is built in, not derived.","rationale":"The paper presents a substantial new grid of 338 MESA models and a clear population-synthesis post-processing, and the calibrated result that heavy DNSs are rare at solar metallicity is credible. However, the strongest claim--that the unstable mass-transfer channel for GW190425 is ruled out--is not a derived result of the models. The initial conditions exclude the pre-CE phase, and the binary-evolution scheme never includes a dynamical-instability treatment for the second RLO. Thus 'none of the heavy DNS system progenitors undergo unstable mass transfer' is an artifact of the assumed setup rather than a physical conclusion. This is exactly the load-bearing concern identified in the reader's weakest_assumption, so my read does not change the conditional verdict. The paper should either soften the 'rule out' wording to 'not tested by these models' or add the explicit stability analysis described in concrete_test. The tuning of mu2b and the natal-kick modifications to match the observed population further weakens the quantitative 0.01%/0.02% fractions, but the unstable-MT gap is the more fundamental issue.","tokens_in":22588,"tokens_out":7145,"duration_ms":85358,"concrete_test":"For every grid model whose post-processing yields a DNS with M_tot >= 3 Msun, compute the adiabatic mass-radius exponent zeta_ad = (d ln R / d ln M)_ad of the He donor at the instant of first Roche-lobe contact, and compare it with the critical value zeta_crit from the orbital response under the same non-conservative mass-loss mode used in Section 2. If any such heavy-DNS progenitor has zeta_ad < zeta_crit, the second RLO is dynamically unstable, and the paper's central claim is falsified. If all heavy-DNS tracks are stable, the concern is resolved for the computed grid.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim, repeated in the Abstract and Section 4, is that the models 'rule out the formation of massive DNSs like GW190425 via unstable MT at solar metallicity.' This is not supported by the computation. Section 2.2 initializes every binary after the common-envelope phase as a naked helium star with a fixed 1.4 Msun NS, and Section 2 uses the Ritter (1988) mass-transfer scheme with Eddington-limited, completely non-conservative accretion. No dynamical-instability criterion for the second RLO is ever applied, so the grid contains only quasi-static Case BB RLO tracks. Section 3.2's statement that 'none of the heavy DNS system progenitors undergo unstable mass transfer' is therefore a property of the initial grid and assumed MT scheme, not a physical exclusion of the unstable branch. The same section reports that ~80% of massive systems undergo stable Case BB RLO and ~20% avoid RLO; none can be unstable by construction. Consequently, the 'rule out' conclusion is overclaimed. The 0.01%/0.02% fractions are additionally generated by a prescription calibrated to the observed mass distribution (mu2b = 0.1, USSN/ECSN kicks reduced, kick boost for CO cores > 1.66 Msun; Sec. 3.1.2) and are quoted without uncertainties or sensitivity runs.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a grid of 338 MESA binary models of helium stars (2.5–9.8 Msun) with a fixed 1.4 Msun neutron star companion at solar metallicity, evolved from the post-common-envelope phase through Case BB mass transfer and the second supernova. Using the Mandel & Müller (2020) remnant/kick prescription, first in a standard form and then with a modified calibration (μ2b=0.1, reduced USSN/ECSN kicks, and a factor-2 kick boost for CO cores >1.66 Msun), the authors construct synthetic DNS populations and compare their P_orb–e and total-mass distributions to the Galactic DNS sample. They find that the modified model broadly reproduces the observed distributions and that only 0.01% of all systems (0.02% of Hubble-time mergers) have total mass ≥3 Msun. The paper concludes that this channel cannot explain the rate of GW190425-like mergers and claims to rule out formation of such systems via the second unstable mass transfer.","tokens_in":22799,"tokens_out":2545,"duration_ms":32525,"significance":"If the central conclusion were fully supported, the paper would be a valuable constraint on formation channels for GW190425 and on the isolated binary evolution channel for heavy double neutron stars. The MESA grid and the comparison to the current Galactic DNS sample are useful, and the authors are transparent about their MESA version and provide inlists and post-processing scripts, which is a strength. However, the most prominent claim—that the models rule out unstable mass transfer as a formation route for GW190425-like systems—is not supported by the computation itself, because the grid never simulates unstable mass transfer. The paper's calibrated fractions also lack uncertainty/sensitivity analysis. The work is therefore a useful exploratory study, but its headline conclusion needs substantial reframing or additional modeling.","major_comments":[{"comment":"The claim in the Abstract and in Section 3.2 that the models 'rule out the formation of massive DNSs like GW190425 via the second unstable mass transfer' is not supported by the computation. All binaries are initialized after the common-envelope phase as naked helium stars (Section 2.2), and the mass-transfer scheme is the Ritter (1988) quasi-static prescription with Eddington-limited, fully non-conservative accretion. No dynamical-instability criterion for the second RLO is applied anywhere, so by construction none of the heavy-DNS progenitors can undergo unstable mass transfer. The statement in Section 3.2 that 'none of the heavy DNS system progenitors undergo unstable mass transfer' is therefore a built-in property of the grid, not a physical exclusion. The correct conclusion is that the modeled stable Case BB channel produces very few heavy DNSs; to rule out the unstable channel, the","section":"Section 2 and Section 3.2"},{"comment":"The modified model is calibrated to the observed Galactic DNS population using four adjustable choices: μ2b=0.1, the USSN kick reduction factor, the ECSN kick of 30 km/s, and the factor-2 kick boost for CO cores ≥1.66 Msun. The kick boost in particular is stated to be chosen 'by making sure that we do not disrupt the GW population'—i.e., the same population that is used later to infer the heavy-DNS fraction. This makes the subsequent 0.01%/0.02% heavy-DNS fractions post-hoc outcomes of the calibration rather than independent predictions. No sensitivity study is presented to show how these fractions vary when μ2b, the kick factors, or the ECSN threshold are changed within reasonable ranges. Without such robustness tests, the quantitative claim that only 0.01% of DNSs are heavy is not established.","section":"Section 3.1.2"},{"comment":"The absolute fractions (0.01% of all systems, 0.02% of Hubble-time mergers) are computed with an assumed initial distribution of He-star masses and orbital periods: a power-law IMF with α=2.35 and a flat-in-log period distribution (Section 3.1.1). These choices are reasonable but not unique, and the weighting scheme directly determines the fractions. The paper quotes these percentages without uncertainties and without a sensitivity analysis to the assumed weighting or to the binary fraction/normalization. As a result, the statement that the model is 'unable to explain the high rate of GW190425' is stronger than the calculations justify. A rate estimate or an explicit statement that only relative comparisons are intended is needed.","section":"Section 3.1.1 and Section 3.2"}],"minor_comments":[{"comment":"The phrase 'via the second unstable mass transfer' is used in the Abstract and Section 3.2, but the acronym 'MT' is defined earlier; please ensure the first use in the Abstract is explicit (e.g., 'unstable mass transfer (MT)').","section":"Abstract and Section 3.2"},{"comment":"The sentence 'We enable this to kick the slow merging systems visible in radio to higher velocities, simultaneously by making sure that we do not disrupt the GW population' is awkwardly worded; the adverb 'simultaneously' seems to modify an incomplete clause. Please rephrase.","section":"Section 3.1.2"},{"comment":"Typo: 'These systems recieve kicks' should be 'receive'. Also, 'Mtot≥3' appears without a space in multiple places; please fix consistency.","section":"Section 4"},{"comment":"The table lists some systems under 'Globular Clusters' whose total masses (e.g., J0514-4002E at 3.88 Msun) are outside the 2.3–2.9 Msun range quoted in the text. The text acknowledges this, but it would be helpful to explicitly state in Section 1 that the quoted range refers only to the 24 field DNSs, not the full table.","section":"Table 1"},{"comment":"The mapping from ONeMg core mass (1.37–1.43 Msun) to CO core mass (1.43–1.65 Msun) is asserted without a derivation or citation for the CO-core mapping. A reference or a one-sentence justification would improve reproducibility.","section":"Section 2.3"}],"recommendation":"major_revision","confidential_remarks":"The paper has a useful MESA grid and a careful comparison to Galactic DNS data, and the authors are transparent about their methods. However, the headline claim about ruling out unstable mass transfer is not supported by the models, because the grid never simulates unstable MT. This is not a matter of phraseology: the conclusion in Section 3.2 is a logical consequence of the initial conditions and MT prescription. I also share the concern that the calibrated kick modifications are tuned to the same population used for inference, and the quoted fractions have no uncertainty. A major revision that reframes the conclusion as applying only to the stable Case BB channel and adds sensitivity tests would bring the paper in line with its evidence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The useful core of this paper is the 338-model MESA grid of helium-star/neutron-star binaries with a careful post-processing pipeline. The authors calibrate their kick and remnant-mass prescriptions to the observed Galactic DNS population in both the Porb-e plane and the total-mass distribution, and the resulting 0.01% fraction of heavy (Mtot >= 3 Msun) DNSs is a concrete, useful prediction. The comparison to the observed mass distribution and the discussion of where the standard Mandel & Muller (2020) prescription fails are genuinely informative. The paper is a solid piece of binary-evolution work.\\n\\nThe problem is the headline claim. The abstract and conclusions say the model rules out formation of GW190425-like systems via unstable mass transfer at solar metallicity. The grid does not test that channel. Every binary is initialized after common envelope as a naked helium star with a fixed 1.4 Msun neutron star, and the Ritter mass-transfer scheme with Eddington-limited non-conservative accretion is applied throughout. There is no dynamical-instability criterion for the second RLO, so 'none of the heavy DNS progenitors undergo unstable mass transfer' is true by construction. That statement is a property of the assumed model, not a physical exclusion. The stress-test note is right, and I think the authors need to soften the conclusion to something like 'within our quasi-static Case BB RLO grid, we find no unstable MT pathways for massive DNSs.'\\n\\nThe second issue is that the modified kick prescription is heavily post-hoc. mu2b, the USSN/ECSN kick reductions, and the extra kick boost for CO cores above 1.66 Msun are all tuned to reproduce the observed Galactic population, and the kick boost is explicitly chosen to avoid disrupting the GW population. The 0.01% heavy-DNS fraction is therefore a product of that calibration, not an independent prediction. The paper gives no error bars or sensitivity runs around these parameters, which makes it hard to see how robust the fraction is. That is a real weakness, though not fatal if framed as a calibrated model rather than a discovery.\\n\\nMinor issues: the data availability section promises a GitHub repository but gives no URL, and the fixed 1.4 Msun first-born NS plus solar-metallicity-only scope mean the conclusion is narrow. The authors acknowledge these and promise Part II.\\n\\nWho is this for? People working on DNS formation, GW190425, and population synthesis. It is worth engaging with, but the central claim needs revision. I would send it to a serious referee: a good referee can push for a softened conclusion, a sensitivity analysis, and the missing repository link. As it stands, the overclaim in the abstract is likely to mislead readers, and that should be fixed before publication. Would I cite it? Probably yes, for the calibrated 0.01% number, but with a caveat about the unstable-MT claim. I would bring it to a reading group, not to praise the headline, but to discuss the calibration and what exactly the grid can and cannot say.","headline":"A solid MESA grid and calibrated population model showing heavy DNSs are rare at solar metallicity, but the 'rules out unstable mass transfer' claim is overclaimed because the grid never actually simulates an unstable RLO.","tokens_in":23411,"tokens_out":2479,"would_cite":true,"duration_ms":32138,"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":"Heavy double neutron stars are not produced by unstable mass transfer, according to a 338-model binary grid.","keywords":["double neutron stars","GW190425","binary stellar evolution","common envelope","Case BB mass transfer","supernova kicks","neutron star masses","gravitational waves"],"falsifier":"Observe one Galactic double neutron star with total mass above 3 solar masses in a short, recycled orbit consistent with a solar-metallicity isolated progenitor, or run a hydrodynamical binary model that includes the post-common-envelope hydrogen layer and a heavier first-born neutron star and still produces GW190425-like systems through a dynamically unstable second mass transfer; either would break the paper's exclusion.","tokens_in":22357,"feed_emoji":"💫","tokens_out":8595,"duration_ms":87139,"temperature":0.7,"pith_summary":"Using a grid of 338 detailed binary models, the paper asks whether the standard isolated binary channel can produce heavy double neutron stars like GW190425, whose 3.4 solar-mass total exceeds every confirmed Galactic system. It finds that at solar metallicity, with a 1.4 solar-mass first-born neutron star and Eddington-limited accretion, almost no double neutron stars reach 3 solar masses, and none of the heavy progenitors undergo the dynamically unstable second mass-transfer phase proposed as the fast-merger channel. After calibrating supernova remnant and kick prescriptions to the observed Galactic population, the models reproduce the total-mass distribution but still leave an unexplained gap in orbital eccentricity. If the paper is right, GW190425-like systems require lower metallicity, a more massive first-born neutron star, super-Eddington accretion, or dynamical assembly rather than unstable mass transfer in solar-metallicity isolated binaries.","feed_headline":"Heavy neutron-star pairs don't form via unstable mass transfer","feed_subtitle":"A 338-model grid at solar metallicity finds almost no double neutron stars as massive as GW190425.","key_machinery":"The load-bearing machinery is a grid of 338 MESA binary evolution models, using the 1D stellar evolution code MESA. Each binary starts just after the common-envelope phase as a naked helium star with a 1.4 solar-mass point-mass neutron star companion; the helium star is evolved through core helium burning and Case BB Roche-lobe overflow, with accretion onto the neutron star capped at the Eddington rate and treated as fully non-conservative. The resulting pre-supernova carbon-oxygen core and helium envelope masses are then fed into the Mandel & Müller (2020) remnant-mass and natal-kick prescription, in standard and modified forms; the modified form flattens the slope of the carbon-oxygen core","core_discovery":"The paper's central claim is a constraint: for helium stars between 2.5 and 9.8 solar masses paired with 1.4 solar-mass neutron stars at solar metallicity, the second phase of mass transfer in the models that matter is always stable Case BB Roche-lobe overflow, and none of the heavy double neutron stars they produce goes through a dynamically unstable mass transfer. With the modified supernova prescription calibrated to the Galactic field population, only about 0.01% of all systems and 0.01% of recycled systems have total mass at or above 3 solar masses, far too few to explain the inferred rate of massive DNS mergers. The paper therefore rules out the unstable-mass-transfer formation channel","pith_inferences":["The exclusion is conditional on the naked-helium-star starting point, a fixed 1.4 solar-mass first-born neutron star, and Eddington-limited accretion; if post-common-envelope hydrogen retention or a heavier first-born neutron star shifts the stability boundary of the second mass transfer, the unstable channel could re-open.","A direct extension the paper leaves for its Part II, rerunning the grid at lower metallicity and with first-born neutron star masses near 2 solar masses, is the cleanest test: if heavy double neutron star yields remain below one percent, GW190425 needs a fundamentally different formation site.","The predicted roughly 20% branch of non-recycled heavy double neutron stars from massive helium stars that avoid mass transfer gives a concrete gravitational-wave-only population whose chirp-mass distribution and merger rate can be checked against current detector catalogs.","The calibration suggests rapid population-synthesis codes should not use the unmodified carbon-oxygen core to remnant mass relation when estimating heavy double neutron star merger rates, since it produces a high-mass tail that the Galactic field population does not show."],"forward_implications":["If the exclusion holds, the unstable-mass-transfer 'fast-merger' channel contributes essentially nothing to the GW190425-class merger rate at solar metallicity.","The observed rate of massive double neutron star mergers would have to be dominated by something else: lower-metallicity progenitors, more massive first-born neutron stars, super-Eddington accretion, or dynamical assembly in dense clusters.","Calibrated models match the observed total-mass distribution of Galactic double neutron stars, so the high-mass tail in uncalibrated population synthesis is likely an artifact of the carbon-oxygen core to remnant mass relation and kick treatment.","Heavy double neutron stars that do form should be compact, mostly recycled through stable Case BB Roche-lobe overflow, with a minority non-recycled and thus detectable only by gravitational waves.","The predicted eccentricity gap (roughly 0.3 to 0.6) remains unexplained by isolated evolution, pointing to either unseen radio systems or missing kick physics such as distinct kick modes inferred from Be X-ray binaries."],"supporting_citations":[{"why":"Reports GW190425, the 3.4 solar-mass double neutron star merger whose formation channel this paper tests and rules out for the unstable path.","marker":"Abbott et al. (2020)"},{"why":"Defines ultra-stripped supernovae and the Case BB Roche-lobe overflow framework used to model the second mass-transfer phase.","marker":"Tauris et al. (2015)"},{"why":"Supplies the probabilistic supernova remnant-mass and natal-kick prescription that is the paper's standard model.","marker":"Mandel & Müller (2020)"},{"why":"Provides the observational calibration of kick parameters and the reduced ultra-stripped and electron-capture supernova kicks adopted in the modified model.","marker":"Kapil et al. (2023)"},{"why":"Proposed the unstable mass-transfer fast-merger channel for GW190425 that the grid explicitly tests and excludes.","marker":"Romero-Shaw et al. (2020)"},{"why":"Proposed the massive-helium-star channel that avoids mass transfer, matching the roughly 20% non-recycled heavy double neutron stars in the models.","marker":"Vigna-Gómez et al. (2021)"},{"why":"Proposed stable Case BB mass transfer with a massive first-born neutron star as a GW190425 route; the paper contrasts its fixed 1.4 solar-mass assumption with this scenario.","marker":"Qin et al. (2024)"},{"why":"Shows a small hydrogen layer may survive the common envelope; the paper deliberately neglects it, defining the starting point that bounds the exclusion.","marker":"Nie et al. (2025)"}],"fun_headline_variants":["Heavy neutron-star pairs ruled out via unstable mass transfer","No heavy DNS from unstable mass transfer","Unstable mass transfer can't make heavy neutron-star pairs","Heavy neutron stars not born via unstable mass transfer"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The models assume that after the common-envelope phase the neutron star's companion is a bare helium core with no hydrogen left, that the first-born neutron star is exactly 1.4 solar masses, and that mass transfer is capped at the Eddington rate; the unstable mass-transfer channel is never actually simulated, so the exclusion only applies within these starting assumptions.","fun_headline_variants_meta":{"raw":{"variants":["Heavy neutron-star pairs ruled out via unstable mass transfer","No heavy DNS from unstable mass transfer","Unstable mass transfer can't make heavy neutron-star pairs","Heavy neutron stars not born via unstable mass transfer"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000989,"raw_usage":{"total_tokens":4069,"prompt_tokens":824,"completion_tokens":3245,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":568,"completion_tokens_details":{"reasoning_tokens":3183}},"tokens_in":568,"tokens_out":3245,"duration_ms":24640,"temperature":1.0,"reasoning_tokens":3183,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T17:46:38.879174+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Observe one Galactic double neutron star with total mass above 3 solar masses in a short, recycled orbit consistent with a solar-metallicity isolated progenitor, or run a hydrodynamical binary model that includes the post-common-envelope hydrogen layer and a heavier first-born neutron star and still produces GW190425-like systems through a dynamically unstable second mass transfer; either would break the paper's exclusion.","supporting_citations":[{"cited_title":"P., et al., 2017a, Phys","cited_arxiv_id":null,"evidence_quote":"Reports GW190425, the 3.4 solar-mass double neutron star merger whose formation channel this paper tests and rules out for the unstable path."}],"review_version":1}