{"id":"40ec4044-cd13-4b21-8e32-b80cfcfe24de","arxiv_id":"2607.26205","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Disk torques can drive stellar binaries around a massive black hole to tidal disruption, and the Milky Way's young stellar disk likely caused ~10^2 such events ~5 Myr ago.","lead":"This paper shows that massive gas disks around black holes can torque binary stars into the Hills disruption mechanism, flinging out hypervelocity stars and implanting S-cluster stars. Applying this to the Milky Way's 5-million-year-old central stellar disk, it estimates roughly one hundred binary disruptions occurred, possibly producing the fastest-known hypervelocity star.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"N_d hinges on the step-function evaporation model: the paper's own 'long-lived hard binaries' variant (Fig. 7, purple) swings N_d by over an order of magnitude for old SFHs, so '~10^2' is set by an untested population assumption.","rationale":"The paper is honest and careful: the orbital dynamics are validated with N-body integrations, the analytic scalings are internally consistent, and the authors explicitly flag large systematic uncertainties. The mechanism—disks torquing binaries into the Hills loss cone—is supported by the numerical validation. However, the quantitative claim for the Galactic Center is a convolution of that validated dynamics with a toy binary population model. The weakest point is the evaporation model: eq. (12) gives a timescale, and §4.3 item 5 converts it to a step function—binaries vanish at t_evap. Realistic evaporation is gradual and leaves hard binaries intact, and the paper's own Fig. 7 purple points show that this choice changes N_d by an order of magnitude for old SFHs (e.g., Chen+23). Thus the ~10^2 estimate is substantially determined by a population-level assumption that is not tested by the N-body runs. This does not invalidate the paper—the authors are transparent and the central mechanistic claim stands—but it means the headline number should be read as a rough expectation under a specific survival model, not a robust prediction. The proposed concrete test would directly quantify how much of N_d is dynamical versus assumed, and would either support or weaken the '~10^2' claim without changing the conclusion that ADDD is a viable and possibly important process.","tokens_in":30552,"tokens_out":7164,"duration_ms":69674,"concrete_test":"Re-run the Monte Carlo population model of Fig. 7 with continuous evaporation: replace the step-function condition 't_age < t_evap' in §4.3 item 5 with an exponential survival probability S(t) = exp[-∫_0^t dt'/t_evap(a_in, m_b, r(t'))] (or a scattering-calibrated binary-single loss prescription), keeping all other parameters fixed. Record the median N_d for each of the three SFHs. If any column changes by more than a factor of 3 relative to the fiducial step-function result, the headline '~10^2' is not robust to the evaporation treatment.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 6's headline estimate of ~10^2 binary disruptions from the CWD-forming disk is the quantitative payload. The integral in eqs. (7)-(8) inherits every assumption of the binary population model in §4.3. The most fragile is item 5 / eq. (12): a binary is kept unchanged until t_age = t_evap and then instantaneously removed. Real evaporation is gradual and hard binaries survive; the paper itself demonstrates the consequence in Fig. 7, where the 'long-lived hard binaries' variant (purple points) drastically raises N_d for the Chen+23 all-5-Gyr-old SFH. Thus the central number is not pinned by the validated orbital dynamics (Section 5 checks which orbits reach r_t, not how many binaries survive to the disk epoch). With f_b,h = 1 and λ = 0 at their extremes, the plausible N_d range spans more than an order of magnitude, so '~10^2' is a conditional estimate under a specific, uncalibrated survival model rather than a robust prediction.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper extends the Kaur & Stone (2025) disk-driven disruption mechanism from single stars to binaries. It characterizes the outer-orbit conditions for binary disruptions (loss wedge and librating island), proposes a fitting formula for the separatrix action (Eq. 4), constructs a toy binary population (§4.3), and derives semi-analytical plus Monte Carlo estimates of the number of disruptions (Eqs. 7–8). These estimates are compared with N-body integrations in Fig. 6. The application to the Galactic Center's ~5-Myr-old clockwise disk yields a burst of ~10^2 binary disruptions, which the authors connect to S-cluster stars, hypervelocity stars, and possibly S5-HVS1. They also discuss implications for TDEs in post-starburst galaxies and argue that ADDD produces near-isotropic, bursty HVS ejections.","tokens_in":30896,"tokens_out":6727,"duration_ms":63955,"significance":"If the central estimate holds, ADDD would be a new, observationally relevant channel for Hills-mechanism disruptions, with a distinctive combination of bursty and near-isotropic ejection. The paper's strengths include a generic semi-analytical framework that extends KS25 to arbitrary disk/cluster density slopes, a careful Monte Carlo-vs-N-body comparison in Fig. 6, and falsifiable predictions for S-star magnitudes, HVS sky positions, and TDE delay timescales. The S5-HVS1 comparison is post-hoc but not used to fit parameters. The main weakness is that the headline number ~10^2 depends on a step-function binary-evaporation model whose sensitivity the authors themselves demonstrate in Fig. 7. The mechanism is well founded; the quantitative population-level prediction is not yet robust.","major_comments":[{"comment":"The step-function evaporation model — binaries are kept unchanged until t_age = t_evap and then removed — is the single most load-bearing population assumption in the paper. Equations (7)–(8) inherit it through r_m in Eq. (10), and Eq. (12) sets t_evap. The authors themselves show in Fig. 7 (purple points) that replacing this by 'long-lived hard binaries' raises N_d by more than an order of magnitude for the Chen+23 SFH, and the text acknowledges that this 'improved physics can cause a drastic increase.' Since the N-body validation in §5 only checks which orbits reach r_t for a prescribed binary population, it does not constrain this survival assumption. The headline '~10^2' is therefore not pinned by the validated orbital dynamics. I request either a calibration of the survival model against the observed binary-fraction profile, or a reframing of the central claim as a conditional estim","section":"§4.3 (items 5–6), Eq. (12), Fig. 7"},{"comment":"The binary population model takes f_b,h=1 and λ=0 'throughout this work', yet the observational constraints are admittedly sparse. N_d is linear in f_b,h (see the d f_b,h/dain factor in Eq. (14)) and λ enters the radial exponent Λ in Eq. (B11); taking f_b,h~0.3–0.5 and λ>0 would reduce N_d by factors of 2–5 or more. The paper does not explore this in Fig. 7, so the plotted range understates the model uncertainty. The central estimate should be accompanied by a calculation of its sensitivity to these parameters.","section":"§4.3, Eq. (10), Eq. (14)"},{"comment":"The 'excellent agreement' between Monte Carlo and N-body validates the orbit-averaged loss-wedge/librating-island selection, not the full binary-disruption count. The simulations assign a_in and m_b from the population model and record whether the outer orbit reaches r_t; they do not evolve the binary population, evaporate binaries, or model binary-single interactions. Thus the validation is narrower than the abstract's 'validated with numerical simulations' implies. Please state explicitly that the population-level assumptions in §4.3 are not validated by Fig. 6.","section":"§5, Fig. 6"},{"comment":"The text states estimates range 'from a few to a few hundred' and 'our most detailed models suggest the number was 50–100,' while the abstract and conclusions state '~10^2.' Given the factor-of-several spread from β, μ, and IMF, plus the >10 effect of the long-lived-hard-binary variant, calling this 'order-of-magnitude-accurate' is not supported by the model's own sensitivity. I recommend a more cautious summary of the quantitative payload, and a discussion of which future observations (e.g., HVS ejection-time distribution, low-mass S-star population) could discriminate among the variants.","section":"§6.2, Fig. 7"}],"minor_comments":[{"comment":"The piecewise scaling for f_d(a) is stated without derivation in the main text; the Appendix B derivation is clear, but a one-sentence justification in §3.3.1 would help. Also, Fig. 4's dashed lines are arbitrarily normalized; please state the normalization in the caption.","section":"Eq. (6) and Fig. 4"},{"comment":"The text says 'our most detailed models suggest the number was 50–100' while the Fig. 7 caption says 'between ~10 and ~10^2'; make these consistent.","section":"§6.2 vs Fig. 7 caption"},{"comment":"'von Ziepel, H. V. 1909' should be 'von Zeipel, H. V. 1909' (standard spelling of the ZLK mechanism).","section":"References"},{"comment":"The 'factor 10' in p(shrink|enter) is introduced as the only fine-tuning; since it is set to match the Dodici et al. (2026) simulations, please report its sensitivity (e.g., how N_d changes if this factor is 3 or 30).","section":"Appendix C"}],"recommendation":"major_revision","confidential_remarks":"The dynamical mechanism is sound and the authors are transparent about uncertainties. My main reservation is that the central number ~10^2 is not robust to the authors' own plausible alternative survival model; this is fixable with reframing and sensitivity tests, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, here's my take on arXiv:2607.26205. The core new result is real: the disk torque that KS25 showed drives single-star TDEs also drives binaries into the Hills mechanism. This paper characterizes that process with a clean semi-analytical framework, adds a useful separatrix-action fitting formula (eq. 4), and validates the orbital dynamics with N-body integrations. The Monte Carlo and N-body curves in Fig. 6 agree well, which is the right kind of evidence. The paper also makes a genuinely novel observational point: ADDD ejections are nearly isotropic and bursty, distinguishing them from other HVS-producing mechanisms. The comparison with Penoyre et al. (2025) is carefully argued in Appendix E.\n\nThe soft spots are real but don't break the mechanism. The binary population model sets f_b,h = 1 and lambda = 0 with thin observational justification, and the evaporation treatment in eq. (12) is a step function: binaries survive unchanged until t_evap and then vanish. The paper's own Fig. 7 shows how much this matters — the \"long-lived hard binaries\" variant changes N_d by more than an order of magnitude for the Chen+23 star-formation history. So the headline ~10^2 disruptions from the CWD-forming disk is not a robust prediction; it is a conditional estimate under a specific survival model. The plausible range spans a few to a few hundred. That is not fatal, but the abstract states ~10^2 as if it were firmer than the model supports. A referee should ask for a propagation of the evaporation uncertainty or an explicit framing of the central value as model-dependent. Also, no code or parameter files are provided for the 22k-orbit simulations, which would help others check the population-level integrals.\n\nThe S5-HVS1 connection is post-hoc, but the authors do not tune parameters to match it; they just note their models produce analogues in a few realizations. That is acceptable, though it should not be oversold. The paper is honest about the large systematic uncertainties in disk recurrence and cluster anisotropy, and it does not fit the target result to any observation.\n\nWho should read it: anyone working on Galactic Center dynamics, hypervelocity stars, or nuclear TDE rates. The orbital dynamics half is in good shape; the population half needs a careful review of the evaporation model and a reproducibility pass. I would send it to peer review, expect moderate revision, and cite it once the population-level assumptions are either tightened or explicitly framed as a range.","headline":"Well-validated new mechanism for disk-driven binary disruptions; the central ~10^2 Galactic Center estimate is soft because the evaporation model is a step function.","tokens_in":31381,"tokens_out":3128,"would_cite":true,"duration_ms":31243,"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":"The paper argues that a massive accretion disk torques binary stars into the Hills mechanism, producing bursts of S-stars and hypervelocity stars, and that the Galactic Center's young star disk drove about a hundred such disruptions.","keywords":["accretion disks","massive black holes","Hills mechanism","binary disruptions","hypervelocity stars","S-stars","Galactic Center","tidal disruption events"],"falsifier":"Measuring the ejection-age distribution of hypervelocity stars: the model predicts a burst of ejections about 5 Myr ago, roughly isotropic, whereas steady-state mechanisms predict a flat age distribution. If a complete sample of HVSs shows no burst at that epoch, the ADDD burst estimate would be contradicted.","tokens_in":30432,"feed_emoji":"⭐","tokens_out":3783,"duration_ms":36657,"temperature":0.7,"pith_summary":"The paper proposes that when a massive accretion disk forms around a supermassive black hole in a galactic nucleus, its non-spherical gravity torques binary stars on nearly perpendicular orbits into extremely eccentric paths. This can fling the binary members apart through the Hills mechanism, implanting one star close to the black hole and ejecting the other as a hypervelocity star. Applying this to the Milky Way's ~5-million-year-old disk of young stars, the paper estimates a burst of roughly 100 binary disruptions, producing an excess of S-cluster stars and hypervelocity stars. This makes disk-driven disruptions a candidate explanation for the fastest known hypervelocity star and a new source of delayed tidal disruption events in other galaxies.","feed_headline":"Disk formation at the Galactic Center likely broke up ~100 binary stars","feed_subtitle":"The burst would have created S-stars and hypervelocity stars, possibly including the fastest one known.","key_machinery":"The key object is the 'librating island' of phase space, a set of orbits in the loss wedge whose argument of pericenter librates rather than circulates under the disk potential. Orbits in this island are the ones the disk torque can drive to the loss cone. The paper uses a fitting formula for the separatrix action of this island, which generalizes previous results to arbitrary disk and cluster density slopes, to write the disrupted-binary distribution as an integral over the loss wedge and island.","core_discovery":"The central claim is that accretion-disk-driven disruption (ADDD) is a generic, rich channel for binary destruction around massive black holes. For binaries whose outer orbits are nearly perpendicular to the disk, the disk torque drives their angular momentum down while conserving its component along the disk axis, pushing many of them into the loss cone on a timescale comparable to disk growth. The paper derives a semi-analytical formula for the fraction of binaries disrupted as a function of distance from the black hole, validates it numerically, and estimates that the disk that formed the Galactic Center's young stars disrupted on the order of 100 binaries. The byproducts—implanted S-star","pith_inferences":["If ~100 binaries were disrupted in a burst 5 Myr ago, the present-day S-star cluster should contain a population of older, low-mass stars with a distinctive age gap; future faint-object observations could test this by detecting the low-mass tail of implanted stars.","Because ADDD ejects stars nearly isotropically, surveys that look for a planar or axial signature might miss this channel; joint analysis of ejection times and velocities may be more discriminating than sky position alone.","The paper's evaporation model may be the main lever on the rate: if hard binaries survive many evaporation times, the number of disruptions in old nuclear clusters could be an order of magnitude higher than the fiducial estimate, making ADDD competitive with two-body relaxation as a steady HVS source."],"forward_implications":["If a massive disk formed in the Galactic Center ~5 Myr ago, the model predicts a burst of ~100 disrupted binaries, producing an excess of S-stars and hypervelocity stars above the steady background.","Hypervelocity stars from ADDD are ejected roughly isotropically and in temporal bursts, a combination unique among known mechanisms; this signature can be searched for in future surveys.","In other galaxies, the S-star analogues implanted by ADDD can later be driven to tidal disruption, producing (possibly repeating) TDEs in the ~0.1–1 Gyr after an AGN phase, which may help explain TDE excesses in post-starburst galaxies.","The disk torque also drives secular chaos in binaries that are not disrupted, potentially shrinking stellar binaries and merging compact-object binaries via gravitational-wave emission."],"fun_headline_variants":["Disk torques around black holes shred binary stars","Galactic disk may have ripped apart ~100 binaries","Disk-driven binary disruptions spark star bursts","Massive disks can break binaries, eject fast stars"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The estimate that ~100 binaries were disrupted assumes that binaries survive unchanged until a single evaporation time and then vanish completely; the paper's own variant that lets hard binaries live longer increases the count dramatically, so the number is sensitive to this idealized binary depletion model.","fun_headline_variants_meta":{"raw":{"variants":["Disk torques around black holes shred binary stars","Galactic disk may have ripped apart ~100 binaries","Disk-driven binary disruptions spark star bursts","Massive disks can break binaries, eject fast stars"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000672,"raw_usage":{"total_tokens":2894,"prompt_tokens":739,"completion_tokens":2155,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":483,"completion_tokens_details":{"reasoning_tokens":2108}},"tokens_in":483,"tokens_out":2155,"duration_ms":13532,"temperature":1.0,"reasoning_tokens":2108,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T00:29:13.228178+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measuring the ejection-age distribution of hypervelocity stars: the model predicts a burst of ejections about 5 Myr ago, roughly isotropic, whereas steady-state mechanisms predict a flat age distribution. If a complete sample of HVSs shows no burst at that epoch, the ADDD burst estimate would be contradicted.","supporting_citations":[],"review_version":1}