{"id":"049fe6d0-47e9-4153-9cd5-456cb7e20749","arxiv_id":"2607.29075","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"Supernova remnants inside AGN discs can re-collapse into seed clouds that grow and spawn new massive stars, with yields up to hundreds per event for 10^8-solar-mass black holes.","lead":"This paper models what happens after a massive star explodes inside the dense gas disc around a supermassive black hole. It argues the cooling remnant can collapse into a new cloud and trigger a second generation of massive stars, possibly dozens to hundreds per explosion in the most massive systems.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Hill-capture growth in Eqs. (14)-(16) requires a self-gravitating point mass, but the seed cloud is pressure-confined and not self-gravitating; its Bondi radius is far smaller than its Hill radius, so the exponential growth and Table 4 yields rest on an untested unit-capture assumption.","rationale":"The reader identified f_seed as the weakest assumption, noting it is uncalibrated. I agree f_seed is uncertain, but the more vulnerable link is the growth law itself. Eq. (14) treats the Hill sphere as a capture cross-section for gas around the seed. That is valid only if the cloud is sufficiently self-gravitating that gas entering R_H becomes bound. The authors explicitly state the seed is pressure-confined and not fully self-gravitating. For a non-self-gravitating cloud the relevant scale is the Bondi radius, which in the fiducial cases is orders of magnitude smaller than R_H; gas on shear orbits will stream past rather than accrete unless some unspecified dissipation binds it. The paper's own Sec. 4 says it isolates the dynamic feasibility of Hill-capture growth and leaves full treatment to 3D radiation-MHD simulations, which is precisely the missing validation. This concern also amplifies the reader's f_seed point: in the Bondi-limited regime the growth rate scales as M_cl^2 rather than M_cl, so the seed mass matters much more than in Eq. (16). A shearing-box experiment or a recalculation with Bondi-limited accretion would settle the issue. This does not overturn the paper as a speculative pilot model, so I keep the reader's CONDITIONAL verdict, but the accretion prescription should be flagged as a primary unresolved condition.","tokens_in":22459,"tokens_out":13942,"duration_ms":157747,"concrete_test":"Run local shearing-box radiation-hydrodynamic simulations of a pressure-confined gas cloud of mass M_seed in the SG disc at R_I = 10^4.5 Rg, with the adopted background density/temperature and no imposed gravitational softening. Measure the net accretion rate dM/dt onto gas inside the Hill radius as a function of M_cl. If dM/dt is more than an order of magnitude below pi fcav rho_d Omega R_H^3, or scales as M_cl^2 rather than M_cl, replace Eq. (14) with the measured (or Bondi-limited) rate and recompute the collapse times and Table 4 yields. If collapse is then delayed beyond the local shear/depletion time or absent for small f_seed, the central claim is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The weakest link is not f_seed alone but the accretion prescription that turns any seed into a collapsing cloud. In Sec. 3.2, Eq. (14) sets Mdot_acc = pi fcav rho_d Omega R_H^3, i.e., all gas entering the Hill sphere is captured at the shear rate. This is appropriate for a self-gravitating point mass whose Bondi radius exceeds its Hill radius. The seed cloud, however, is described in Sec. 3.1 as 'pressure-confined and tidally limited' and 'not ... fully self-gravitating at birth.' For such a cloud the gravitational capture radius is the Bondi radius r_B = G M_cl / c_s,d^2. In the fiducial Mdot=10^8 Msun, RI=10^4.5 Rg case, r_B is orders of magnitude smaller than R_H (for M_seed ~0.1-1 Msun and c_s,d ~ a few km/s, r_B/R_H ~ 1e-3 to 1e-1), so Eq. (14) overestimates the bound-gas flux by roughly (R_H/r_B)^2. No cooling, angular-momentum loss, or sticking mechanism is specified to make Hill-sphere gas become bound to a non-self-gravitating pressure-confined cloud. The sensitivity study in Sec. 3.3 varies f_cav and a but not the accretion law, so the exponential growth M_cl = M_seed exp((t - t_implosion)/t_growth) and the collapse masses in Table 4 are conditional on unit capture efficiency. The paper's own Sec. 4 says it 'isolate[s] the dynamic feasibility of Hill-capture growth' and defers full treatment to 3D radiation-MHD simulations; that is exactly the unresolved step required for the central claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a semi-analytic channel for second-generation massive-star formation in AGN discs. After a core-collapse supernova explodes inside the disc, the remnant is argued to cool efficiently, enter radiative snowplow phases, and then undergo partial shell backflow that seeds a compact, pressure-confined cloud. The cloud is assumed to grow by shear-limited Hill capture (exponential growth) until it simultaneously satisfies Jeans, virial, tidal, shear, photoionization, and magnetic-supercriticality criteria. Using a Sirko-Goodman disc model, the authors compute collapse masses for three SMBH masses and three explosion radii, convert them into expected stellar populations under a top-heavy IMF, and find event yields from below one to several hundred massive stars. The paper explicitly frames the mechanism as a localized recycling channel rather than a global star-formation mode.","tokens_in":22890,"tokens_out":4740,"duration_ms":50690,"significance":"If the proposed pathway holds, it would identify a novel, localized gas-recycling loop in AGN discs: each embedded CCSN could re-process swept-up and re-filled gas into a new collapsing cloud, potentially contributing to the top-heavy massive-star population and metal enrichment in galactic nuclei. The manuscript has clear strengths: the SNR stage equations in Sec. 3.1 are standard and internally consistent; the diffusion check in Table 3 supports efficient radiative cooling for the fiducial 10^8 M_sun cases; the collapse mass is an emergent outcome of the growth ODEs rather than an input; and the authors are transparent about the uncalibrated nature of f_seed and f_star. However, the central quantitative claim depends on an untested unit-capture accretion prescription and on an uncalibrated seed-formation efficiency, so the event-level yields in Table 4 are not yet robust enough to establish the mechanism's importance without further work.","major_comments":[{"comment":"The accretion rate Mdot_acc = π fcav ρd Ω R_H^3 assumes that all gas entering the Hill sphere is captured at the shear rate, which is appropriate for a self-gravitating point mass whose Bondi radius exceeds its Hill radius. The seed cloud, however, is explicitly described as 'pressure-confined and tidally limited' and not fully self-gravitating (Sec. 3.1, after Eq. 13). For such a cloud, the gravitational capture radius is the Bondi radius r_B = G M_cl / c_s,d^2, which for the fiducial seed masses is orders of magnitude smaller than R_H, so Eq. (14) overestimates the bound-gas flux by roughly (R_H/r_B)^2. No cooling, angular-momentum loss, or sticking mechanism is specified to make Hill-sphere gas become bound to a non-self-gravitating pressure-confined cloud. The sensitivity study in Sec. 3.3 varies fcav and a but not the accretion law, so the exponential growth (Eq. 16) and all Table 4","section":"Sec. 3.2, Eq. (14)-(16)"},{"comment":"The seed-cloud mass M_seed = f_seed M_shell with f_seed = 1e-3 is explicitly uncalibrated, as the authors state that the existing simulations 'do not provide a calibrated seed-formation efficiency for AGN-disc conditions.' Because the growth is exponential (Eq. 16) and the collapse time depends logarithmically on the initial mass, the resulting collapse masses and hence the event-level yields in Table 4 are directly sensitive to this assumption. If f_seed is orders of magnitude smaller, collapse is delayed or does not occur within the relevant timescale; if larger, yields increase. The paper should at minimum provide a sensitivity study over f_seed, and ideally calibrate it against the Romano et al. (2024) simulations or bound it with physical arguments about backflow mass and shell fragmentation. As it stands, the headline quantitative range ('less than one' to 'several hundred' massive","section":"Sec. 3.1, f_seed and Table 4"},{"comment":"The event-level yields (M_star,event, N_total, N_0.1-1, etc.) are presented as single-valued predictions with no uncertainty ranges. They depend on f_star = 0.3 (an assumed integrated cloud-to-star efficiency), on f_seed, and on the unvaried Hill-capture accretion law. The authors do state that f_star is assumed, but the table does not allow the reader to see how the conclusions change if f_star or f_seed take other plausible values. At minimum, propagating the uncertainty in f_seed and f_star, or presenting yields as functions of these parameters, would make the claims in Sec. 4 ('from less than one... to several hundred') appropriately qualified. Without this, the quantitative comparison across SMBH masses overstates the model's predictive power.","section":"Table 4 and Eq. (24)"}],"minor_comments":[{"comment":"The backflow velocity uses the maximum of three speeds, including a gravitational infall speed (2GM_shell/R_MCS)^{1/2}. For the fiducial parameters M_shell is likely small and the sound speed dominates; it would be helpful to state which term dominates in the quoted t_backflow = 5.5-233 yr, since this affects the assumed seed initialization time.","section":"Sec. 3.1, Eq. (12)"},{"comment":"The axis label 'AU' is used for radius but the x-axis is time in years. Please clarify the units in the caption to avoid confusion.","section":"Figure 3"},{"comment":"The stage abbreviations FREE, ST, PDS, MCS are defined in the text but not in the table caption. Adding the full names in the caption or a footnote would improve readability.","section":"Table 1"},{"comment":"The terms 'reincarnations' and 'second-generation SF' are used interchangeably. The paper would benefit from a brief definition of 'second-generation' in the introduction to connect the metaphor with the quantitative model.","section":"Sec. 1 and Sec. 4"},{"comment":"The magnetic field estimate B_d = (2 Mdot c / R_I^2)^{1/2} is appropriate for a globally ordered field, but the paper notes that small-scale dynamo and reconnection diffusion could change λ_B. Given that the magnetic criterion is close to critical in some cases (Figure 6), adding a short discussion of how λ_B would shift under a turbulent field would be useful.","section":"Sec. 4, magnetic criterion"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is honest about its limitations and the semi-analytic derivation is mostly transparent. The main concern is not the style but the central physics: the Hill-capture growth law is applied to a pressure-confined seed cloud without a capture-efficiency criterion, and the yields scale with uncalibrated f_seed. These are fixable in principle — e.g., by adding a Bondi-limited growth stage, a capture efficiency factor, and a f_seed sensitivity analysis — but without them the quantitative conclusions are not yet supported. I recommend major revision rather than rejection, because the underlying SNR evolution and the proposed channel are plausible and the authors have already identified the key uncertainties."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth a read if you work on star formation or SNR physics in AGN discs. What's new is the quantitative end-to-end channel: an embedded CCSN cools radiatively, the shell backflows, a seed cloud forms, grows by shear-limited accretion, and collapses into a second generation of massive stars. The authors do a solid job with the SNR stage equations, check the radiative diffusion time, and are honest that this is a localized recycling channel, not the dominant mode. The yield tables as a function of SMBH mass and radius give a concrete benchmark that people will quote.\n\nThe soft spots are real and in proportion. First, the seed-formation efficiency f_seed = 1e-3 is explicitly uncalibrated, and the stellar yields scale directly with f_star = 0.3. Those are acknowledged and addressable. More concerning is the Hill-capture prescription in Eq. (14). The seed cloud is described as pressure-confined and not fully self-gravitating, but the accretion rate assumes a self-gravitating point mass whose Hill radius is the capture radius. For a pressure-confined cloud with M_seed ~ 0.1-1 Msun and sound speed a few km/s, the Bondi radius is orders of magnitude smaller than the Hill radius. The formula thus overestimates the bound-gas flux in the early growth phase, and the exponential growth that drives the collapse masses in Table 4 rests on an untested unit-capture assumption. The sensitivity study varies fcav and a, but not the accretion law itself. The paper's own Section 4 defers this to future 3D radiation-MHD simulations, which is exactly the unresolved step required for the central claim.\n\nThat said, the paper is not incoherent. The authors engage seriously with the prior simulations and flag the major simplifications. The issue is a load-bearing approximation that needs to be justified or replaced with a Bondi-limited or pressure-assisted capture prescription. It's a fixable problem, but it changes the headline numbers.\n\nFor peer review: yes, send it out. A good referee can push for the accretion-law treatment, and the paper's framework will be a useful reference in the literature. I would not cite the yield tables as-is, but I would bring this to a reading group to discuss how easily semi-analytic models can hide assumptions in accretion prescriptions.","headline":"A competent semi-analytic feasibility study of SNR-triggered star formation in AGN discs, but the Hill-capture growth step is inconsistent with the seed cloud's own non-self-gravitating state, so the quoted yields are probably optimistic.","tokens_in":23424,"tokens_out":3490,"would_cite":false,"duration_ms":41142,"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":"This paper proposes that a core-collapse supernova embedded in an AGN accretion disc can, through shell backflow and shear-limited Hill accretion, seed a compact gas cloud that grows and collapses into a second generation of massive stars.","keywords":["AGN discs","core-collapse supernovae","supernova remnants","star formation","Hill accretion","seed cloud collapse","top-heavy initial mass function","black hole accretion discs"],"falsifier":"A 3D radiation-MHD simulation of a 10^51 erg core-collapse supernova embedded in a standard self-gravitating AGN disc at M• = 10^8 solar masses and R = 10^4.5 Rg, run past the snowplow phase, that fails to produce any bound central over-density with mass ≳ 10^-3 of the shell mass would falsify the seed step; equivalently, a measured f_seed below ~10^-5 would push the exponential growth window past the local dynamical time in low-mass systems.","tokens_in":22262,"feed_emoji":"💥","tokens_out":6739,"duration_ms":59337,"temperature":0.7,"pith_summary":"The paper builds a semi-analytical model of a core-collapse supernova exploding inside the dense gas of an active galactic nucleus disc. It argues that efficient radiative cooling compresses or skips the classic adiabatic phase, driving the remnant into a radiative snowplow; once the hot interior loses pressure support, partial backflow of cooled shell material can assemble a small seed cloud. That cloud then grows by shear-limited Hill capture from the surrounding disc gas until gravitational, tidal, shear, photoionization, and magnetic criteria are simultaneously met, at which point it collapses into a top-heavy population of massive stars. The yield depends sharply on the black hole mass and explosion radius: from less than one massive star per event for a 10^6 solar-mass black hole to several hundred for 10^8 solar masses. If correct, embedded supernovae act as a localized gas-recycling channel for second-generation star formation in AGN discs.","feed_headline":"One supernova in a black-hole disc can spark new massive stars","feed_subtitle":"The remnant's shell backflow can seed a cloud that collapses into dozens to hundreds of new stars.","key_machinery":"The argument is carried by a four-stage SNR evolution (free expansion, Sedov-Taylor, pressure-driven snowplow, momentum-conserving snowplow) in which efficient cooling compresses or bypasses the adiabatic stage; the termination point is when shell pressure equals ambient disc pressure. The seed cloud is initialized with f_seed = 10^-3 of the cooled shell mass after a 'backflow' time (not followed hydrodynamically). Growth is via shear-limited Hill accretion—capture of disc gas within the cloud's Hill sphere, whose radius is set by the black hole's tidal field—giving an accretion rate proportional to cloud mass and hence exponential growth with timescale t_growth = 3M•/(π fcav ρd Ω RI^3). Col","core_discovery":"The central claim: a core-collapse supernova embedded in an AGN disc can, after radiative cooling drives the remnant into the snowplow stages, seed a compact cloud through partial backflow of cooled shell material (seed mass = 10^-3 of shell mass). The cloud grows by shear-limited Hill capture at a rate proportional to its own mass, so its mass increases exponentially until six criteria (Jeans, virial, tidal, shear, photoionization, magnetic mass-to-flux) are simultaneously satisfied. The yield depends strongly on SMBH mass and radius: below one massive star per event for 10^6 solar masses, and several to several hundred per event for 10^8 solar masses.","pith_inferences":["Inference: if the recycling channel operates, a single AGN disc patch could sustain successive generations of massive stars from one initial supernova, effectively amplifying the local core-collapse rate before the gas reservoir is exhausted.","Inference: the exponential growth prescription makes the high-yield end of the model sensitive to any process that truncates growth (orbital migration, tidal stripping, or photoevaporation), so the largest quoted yields likely act as upper bounds rather than typical expectations.","Inference: a direct numerical measurement of the seed-formation efficiency f_seed in disc conditions—not yet available—would sharply constrain this channel; values much below 10^-3 would suppress collapse in low-mass SMBH systems while preserving it at high masses.","Inference: observationally, repeated local starbursts of this kind could imprint multiple young stellar populations with correlated ages in the AGN broad-line region, offering a possible spectroscopic signature of recycled star formation."],"forward_implications":["For a 10^8-solar-mass black hole, each embedded supernova is expected to produce roughly 14 to 1000 stars (about 5 to 370 of them above 8 solar masses), depending on explosion radius; the stellar mass produced per initial shell mass reaches ~36.","For a 10^6-solar-mass black hole, the expected number of massive stars per event is below one, so this recycling channel cannot by itself explain metal enrichment in low-mass AGNs.","Because the Hill-capture accretion rate is proportional to cloud mass, the final collapse mass grows exponentially with time, making the outcome exponentially sensitive to the ratio of available growth time to the growth timescale.","With a top-heavy IMF (slope Γ=1), stars above 8 solar masses account for ~37% of stars by number but ~92% of the stellar mass, so the second-generation population's feedback and supernova budget are dominated by massive stars."],"fun_headline_variants":["Supernova in black-hole disc seeds new massive stars","AGN supernovae can ignite stellar reincarnation","Exploded stars in AGN discs can birth hundreds of successors","How one supernova in a galactic core can create many"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is the uncalibrated seed-formation step: after the remnant loses pressure support, partial backflow must assemble a bound, pressure-confined seed cloud with mass at least f_seed = 10^-3 of the shell mass; the paper explicitly notes that simulations do not yet provide a calibrated seed-formation efficiency for AGN-disc conditions.","fun_headline_variants_meta":{"raw":{"variants":["Supernova in black-hole disc seeds new massive stars","AGN supernovae can ignite stellar reincarnation","Exploded stars in AGN discs can birth hundreds of successors","How one supernova in a galactic core can create many"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000709,"raw_usage":{"total_tokens":3032,"prompt_tokens":750,"completion_tokens":2282,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":494,"completion_tokens_details":{"reasoning_tokens":2215}},"tokens_in":494,"tokens_out":2282,"duration_ms":16246,"temperature":1.0,"reasoning_tokens":2215,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T14:04:00.294666+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A 3D radiation-MHD simulation of a 10^51 erg core-collapse supernova embedded in a standard self-gravitating AGN disc at M• = 10^8 solar masses and R = 10^4.5 Rg, run past the snowplow phase, that fails to produce any bound central over-density with mass ≳ 10^-3 of the shell mass would falsify the seed step; equivalently, a measured f_seed below ~10^-5 would push the exponential growth window past the local dynamical time in low-mass systems.","supporting_citations":[],"review_version":1}