{"id":"d9e770ee-681f-43e6-8b99-15479547c64c","arxiv_id":"2504.19884","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Using MESA simulations, the authors show that a massive companion star accreting during a giant eruption stays hot and compact at low accretion rates but inflates and brightens by about an order of magnitude at rates above roughly 0.01 solar masses per year.","lead":"This paper simulates what happens to a massive companion star when it swallows gas from a partner star's giant eruption. The result is a map of when the companion stays compact versus when it swells and brightens, with implications for interpreting stellar eruptions like Eta Carinae.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 0.01 Msun/yr threshold rests on a cold-accretion deposition prescription (added gas put at photosphere entropy, Appendix Eq. 4); no test shows the shock is cooling enough at the highest rates, so the qualitative split's threshold is not yet secured.","rationale":"The reader's weakest-assumption analysis already identified the same core issue: the results depend on how MESA deposits the accreted gas, specifically the claim that the thermal state of the incoming material is irrelevant and that added gas adopts photosphere entropy. My stress test sharpens this into a concrete, testable concern: the cold-accretion limit is assumed, not demonstrated, and it controls the threshold that the central claim is built around. The proposed MESA variant with a surface energy-deposition fraction would settle whether the threshold moves. I do not think this requires changing the reader's CONDITIONAL verdict, because the paper is otherwise a clean parameter study with reproducible inlists and a plausible qualitative split; the concern is exactly the kind of physical simplification that CONDITIONAL verdicts are meant to flag. I also note that the wind-off choice is a secondary issue, but it is less likely to change the threshold substantially because the inflated stars are cool and their Dutch-wind mass loss over 20 yr is probably small. The entropy deposition issue is more fundamental because it directly sets the outer boundary condition during accretion. If the proposed test shows no shift, the central claim is robust; if it shows a shift, the quantitative threshold in the abstract must be revised.","tokens_in":20238,"tokens_out":8682,"duration_ms":101808,"concrete_test":"Rerun the 30 Msun grid at Mdot = 1e-3, 1e-2, 3e-2, and 1e-1 Msun/yr with the same Zenodo inlists, but use MESA's other_energy hook to deposit a fraction f of L_acc = G M Mdot / R into the outermost zone, for f = 0, 0.1, and 0.5. Record the final accretion-phase radius, log L, and log Teff at t = 20 yr. If the high-inflation branch first appears below 1e-2 Msun/yr for any f > 0, the cold-accretion assumption is load-bearing and the abstract's 'approximately 0.01 Msun/yr' threshold needs revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the threshold near 0.01 Msun/yr separating a hot, mildly brightened companion from an inflated, cooled, order-of-magnitude-brighter companion. The physical input that sets this threshold is how MESA deposits the accreted gas. The Appendix assumes the thermal state of the incoming material is irrelevant and assigns it the photosphere entropy, with the accretion luminosity L_acc = G M Mdot / R (Eq. 4) radiated outward without affecting the star. Section 3.3 goes further, stating that the added material has the same thermodynamic properties as the pre-existing outer layers. This is the ideal 'cold accretion' limit. The paper does not check whether the post-shock gas can actually cool on the relevant timescale at the highest rates. At Mdot = 0.1 Msun/yr and the initial radius, L_acc is several times the stellar luminosity; if even a modest fraction of that energy is trapped below the photosphere, the outer layers acquire higher entropy than the photosphere, so inflation and cooling should begin at a lower accretion rate. Because a shift in the threshold by even a factor of a few would change the interpretation of the opposite behaviors in Figure 2, and because recovery tracks also depend on the outer-layer entropy at the end of accretion, this deposition assumption is the most load-bearing part of the argument. The absence of convergence tests or an entropy-deposition sensitivity study leaves this unquantified.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses MESA (r23.05.1) to evolve 20–60 Msun companion stars that accrete at constant rates between 1e-4 and 0.1 Msun/yr for 20 yr after a Giant Eruption, with stellar winds switched off during accretion and re-enabled during recovery. For the fiducial 30 Msun star, the authors find a qualitative dichotomy: at 1e-4 and 1e-3 Msun/yr the star stays on the hot side of the HR diagram with only a mild luminosity increase, while at 1e-2 and 1e-1 Msun/yr it inflates to radii ~10^3 Rsun, cools, and brightens by roughly 0.7 dex. The paper also reports an additional luminosity fluctuation for accretion rates above about 8e-2 Msun/yr, attributed to compression of the outer layers during accretion. The abstract states the central claim as a threshold near 0.01 Msun/yr separating these two behaviors, and the recovery tracks are presented as depending on that split.","tokens_in":20535,"tokens_out":5531,"duration_ms":56938,"significance":"If the dichotomy is robust, the paper provides concrete, falsifiable predictions for companions of erupting massive stars: transient inflation and order-of-magnitude brightening at high accretion rates, a comparatively mild response at lower rates, and mass- and rate-dependent recovery paths. The grid over five initial masses and four accretion rates is a useful first systematic study of this problem. The manuscript ships reproducible MESA inlists and input files on Zenodo, and the high-rate behavior is consistent with the 'hamster' models of Lau et al. (2024) and with the order-of-magnitude companion brightening estimated by Kashi & Soker (2010). The main significance, if the results hold, is in guiding interpretation of giant eruption and supernova impostor light curves and in motivating more detailed binary-parameter studies.","major_comments":[{"comment":"The central threshold near 0.01 Msun/yr rests on the cold-accretion deposition assumption: the added gas is assigned the photosphere entropy, and the accretion luminosity L_acc = Gm Mdot/R is assumed to radiate outward without affecting the star. At Mdot = 0.1 Msun/yr and R ~ 10 Rsun, L_acc exceeds the stellar luminosity by a large factor; if even a modest fraction of that energy is trapped below the photosphere, the outer layers acquire higher entropy than assumed, and inflation and cooling would begin at a lower accretion rate. The paper does not provide a sensitivity study varying the entropy of the added gas or the fraction of L_acc deposited, nor does it quantify the post-shock cooling timescale. Because the qualitative split in Figure 2 and the recovery tracks both depend on the outer-layer entropy at the end of accretion, this assumption is load-bearing for the stated threshold. I request a deposition-sensitivity test (e.g., adding material with different entropies, or explicitly depositing a fraction of L_acc) to show that the threshold location is not an artifact of this choice.","section":"Appendix, Eq. (4); Section 3.3"},{"comment":"The Dutch wind prescription is switched off during the 20-yr accretion phase and re-enabled only during recovery. At high accretion rates the star inflates and its surface temperature drops, so the assumption of zero wind mass loss is favorable to retaining the accreted envelope; if winds continued, outer layers could be partially stripped, reducing the inflation and shifting both the threshold and the recovery tracks. The statement in Section 3.2 that winds are insufficient to remove material is not a test of this, since the wind term is disabled by construction. A quantitative sensitivity run with winds on during accretion, or a justification based on the actual Eddington ratio along the inflated track, is needed to secure the recovery-phase conclusions.","section":"Section 2; Figure 4"},{"comment":"The abstract's threshold '≲0.01 vs ≳0.01 Msun/yr' is not bracketed by the simulation grid: the only rates below the claimed threshold are 1e-3 and 1e-4 Msun/yr, and the only rates at or above it are 1e-2 and higher, a factor of ten gap. The additional runs in Section 3.2 (2e-2 through 8e-2 Msun/yr) probe only the high-rate regime and do not constrain where the transition occurs. Additional runs at intermediate rates (e.g., 2e-3, 5e-3, and 8e-3 Msun/yr) are necessary to place the critical accretion rate; without them, the quantitative value 0.01 Msun/yr is an interpolation, not a measured result.","section":"Section 3.1; Figure 2; abstract"},{"comment":"There is a tension between the text and the appendix on how the accretion energy is handled. Section 3.3 states that the added material 'has the same chemical composition and thermodynamic properties as the material that was present before the accretion' and later says 'we do not specify the fraction of thermal energy released during the accretion phase,' while the Appendix states that the thermal state of the incoming material is irrelevant and that L_acc radiates outward without impacting the added material's entropy. These statements should be reconciled, and the precise MESA mass_change implementation should be described, so the reader can see whether Eq. (4) is actually used or whether the code's default mass deposition is what sets the entropy.","section":"Section 3.3; Appendix"}],"minor_comments":[{"comment":"The phrase 'companion stars companion stars' appears to be duplicated in the first paragraph; please fix.","section":"Section 3.1"},{"comment":"The sentence 'the star with a lower accretion rate experiences radial inflation (as shown by the track from point B to D in Figure 1)' likely refers to Figure 2, not Figure 1; Figure 1 shows the no-accretion tracks.","section":"Section 3.1.2"},{"comment":"The entries for Eint,B, Eint,C, and ΔEint appear to be internally inconsistent (e.g., Eint,C = 3.70 versus Eint,B = 37.0 for the no-accretion row, with ΔEint = 0.6×10^-5). Please check the decimal points and units.","section":"Table 2"},{"comment":"The table caption reads 'T able 1' with an extra space; please fix the spacing.","section":"Section 2, Table 1"},{"comment":"The reference 'Mukhija & Kashi 2024, Submitted' should be updated with the publication status or clearly marked as in preparation, since it is cited as previous work.","section":"Reference list"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the journal's scope and I do not see a citation or novelty problem. The main risk is the untested deposition prescription and the un-bracketed threshold; both are addressable with additional MESA runs, so I do not recommend rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe thing to know: this is a clean MESA parameter study that delivers the first systematic map of how 20–60 Msun companions respond to 20 years of accretion at 1e-4 to 0.1 Msun/yr, including the post-accretion recovery. The qualitative split is real and sensible: below about 0.01 Msun/yr the star stays hot and near equilibrium; above it, the star inflates, cools, and brightens by roughly an order of magnitude. That split is the paper's main takeaway, and it is consistent with earlier work by Lau et al. and Kashi & Soker.\n\nWhat is genuinely new is the grid itself and the recovery tracks. They follow the star for several Myr after accretion, which is rarely done, and they identify a compression-driven luminosity dip near 8e-2 Msun/yr. The MESA work looks competently set up, and they ship inlists on Zenodo, so the results are reproducible. The appendix lays out the accretion energetics clearly, and the literature placement is fair, including the relevant caveats about spherical symmetry and rotation.\n\nThe soft spots are real but not fatal. The quantitative threshold at 0.01 Msun/yr rests on the cold-accretion assumption in the appendix: incoming gas is placed at photosphere entropy, with L_acc radiated away. The authors argue the local thermal timescale is short enough, but they do not test what happens if some accretion luminosity is trapped below the photosphere. At the highest rates this could shift the threshold by a factor of a few. That is a load-bearing assumption for the exact number, though not for the qualitative story. Similarly, they switch off the Dutch wind during accretion and do not present a convergence test or resolution study. These are the sorts of things a referee should ask for, not reasons to reject.\n\nThe paper deserves a serious peer review. Send it out, ask for an entropy-deposition sensitivity run and a resolution check, and it will be a useful reference for giant-eruption and supernova-impostor work.","headline":"A clean, reproducible MESA grid showing how massive companions respond to giant-eruption accretion, with a plausible 0.01 Msun/yr split that needs a deposition-sensitivity check before the threshold is quoted.","tokens_in":21092,"tokens_out":2604,"would_cite":true,"duration_ms":26749,"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":"A companion star accreting from an erupting massive star inflates to about a thousand solar radii and brightens an order of magnitude only when the accretion rate exceeds roughly 0.01 solar masses per year; below that, it stays hot and…","keywords":["giant eruptions","massive star binaries","stellar accretion","LBV eruptions","supernova impostors","stellar inflation","accretion luminosity","thermal timescale"],"falsifier":"A concrete check is to repeat the same numerical experiment while varying what the paper assumes: assign the accreted gas a different entropy (for example, the donor's surface entropy instead of the photosphere's), or leave the stellar wind prescription on during the accretion phase. If the boundary between the two regimes moves away from $\\sim0.01\\,M_\\odot\\,\\mathrm{yr}^{-1}$, or if the high-rate inflation disappears, the claimed dichotomy is an artifact of those assumptions. Observationally, monitoring the companion of $\\eta$ Car or a similar system through its eruptive cycle and detecting whether it reaches $\\log T_{\\rm eff}\\simeq 3.5$-$3.7$ K and $R\\simeq 10^3\\,R_\\odot$ while $L$ jumps by $\\sim0.7$ dex would settle whether the predicted high-rate regime occurs in nature.","tokens_in":20004,"feed_emoji":"🔭","tokens_out":14736,"duration_ms":122335,"temperature":0.7,"pith_summary":"This paper asks what happens to the companion star in a binary when the primary undergoes a Giant Eruption and dumps mass onto it. Running a grid of stellar models from 20 to 60 solar masses at accretion rates of $10^{-4}$ to $0.1\\,M_\\odot\\,\\mathrm{yr}^{-1}$ for 20 years, it finds a sharp behavioral threshold near $0.01\\,M_\\odot\\,\\mathrm{yr}^{-1}$. Below the threshold the companion stays hot, barely expands, and brightens only slightly; above it, the star suddenly brightens by about an order of magnitude, inflates to roughly a thousand solar radii, and cools. The paper then follows the recovery, showing that high-rate accretors contract back toward the main sequence in under a thousand years and eventually evolve as more massive stars, while low-rate accretors lose wind mass and then expand. If correct, this means companions of erupting massive stars are not inert mass-gainers: their transient inflation and brightening matter for interpreting giant eruption and supernova impostor light curves.","feed_headline":"Companion stars balloon when accretion exceeds 0.01 solar masses a year","feed_subtitle":"Above ~0.01 solar masses per year, the companion star inflates and brightens tenfold; below, it barely changes.","key_machinery":"The mechanism carrying the argument is the competition between the accretion timescale and the local thermal timescale in the star's outer layers. Near the surface, the local thermal time $\\tau_{\\rm th}$ is shorter than the accretion time $\\tau_{\\rm acc}$, so the added gas has time to settle to the photosphere's entropy and the thermal state of the incoming material is irrelevant; the star radiates the accretion luminosity $L_{\\rm acc}=GM\\dot M_{\\rm acc}/R$. When the accretion rate rises enough, the outer layers can no longer radiate the added gravitational energy fast enough, the star leaves thermal equilibrium, and the envelope inflates and cools. The numerical treatment adds gravitational energy to the outermost cell and includes a compression term that, at the very highest rates ($\\gtrsim 8\\times10^{-2}\\,M_\\odot\\,\\mathrm{yr}^{-1}$), temporarily suppresses the outward luminosity and produces the early luminosity dip. Recovery is governed by the same physics: once accretion stops, the bloated star radiates away the excess gravitational energy and contracts back toward the main sequence.","core_discovery":"The paper claims that the response of a massive companion to accretion from a giant-erupting binary partner is bimodal, with a transition near $0.01\\,M_\\odot\\,\\mathrm{yr}^{-1}$. For lower rates, the star remains on the hot side of the HR diagram, its luminosity rises only slightly, and it does not expand, because the accretion timescale exceeds the thermal timescale by a larger factor. For higher rates, the companion's luminosity jumps by about one order of magnitude, the star inflates and cools, and it leaves thermal equilibrium. The inflated star is two-component: the original stellar structure underneath plus a distinct accreted layer with different density, temperature, pressure, and entropy. In recovery, high-rate accretors contract back toward the hotter side in roughly $7\\times10^2$ years and then evolve as more massive stars, while low-rate accretors first shed wind mass and then expand. Eventually the accreted material mixes with the inner layers and the star continues as a more massive object.","pith_inferences":["A natural extension the paper does not pursue: binary population synthesis should treat the accreted fraction as a step function in accretion rate near $0.01\\,M_\\odot\\,\\mathrm{yr}^{-1}$; a fixed fractional accretion assumption will misassign retained mass for systems on either side of the threshold.","One observational consequence to test: in a known erupting massive binary, a companion that inflates to $\\sim10^3\\,R_\\odot$ should appear as a cool, red, luminous source with $\\log T_{\\rm eff}\\lesssim 3.7$ K during and shortly after the eruption, while a low-rate companion should show no such redness; time-resolved spectroscopy can distinguish the two.","The compression-induced luminosity dip at very high accretion rates suggests that early light-curve plateaus or dips in giant eruptions could be inverted to estimate the accretion rate onto the companion, providing a new diagnostic if calibrated with more detailed boundary-layer physics."],"forward_implications":["For accretion rates above roughly $0.01\\,M_\\odot\\,\\mathrm{yr}^{-1}$, a companion of an erupting massive star is a temporarily inflated, cool object, so light-curve and spectral models of giant eruptions should include its bloated photosphere rather than treating it as an unchanged main-sequence star.","Below that threshold, the companion's brightness changes are minor, so any order-of-magnitude brightening seen in a giant-eruption event implies high accretion rates onto a companion.","High-rate accretors recover by contracting back to the hot side of the HR diagram in roughly $7\\times10^2$ years, then continue as more massive stars; low-rate accretors first shed wind mass and then expand, so the two regimes leave different observational signatures in the months and years after the eruption.","At accretion rates above about $8\\times10^{-2}\\,M_\\odot\\,\\mathrm{yr}^{-1}$, compression in the outer layers produces an early luminosity dip, which could be recognized as a fingerprint of extreme accretion rather than a primary eruption feature.","The accreted mass (up to $2\\,M_\\odot$ for the highest rate) is eventually mixed into the star, meaning final masses and evolutionary fates of companions should be revised upward by the amount actually retained."],"supporting_citations":[{"why":"Sets the high accretion-rate range (about $0.1$-$0.3\\,M_\\odot\\,\\mathrm{yr}^{-1}$) and the order-of-magnitude luminosity response for the companion in Eta Carinae, against which the grid's high-rate results are compared.","marker":"Kashi & Soker 2010"},{"why":"Provides the rapidly-accreting-star ('hamster') framework and the thermal-acceptance and accretion-luminosity formulas used to interpret inflation along the Hayashi line and the later contraction back toward the main sequence.","marker":"Lau et al. 2024"},{"why":"Establishes the near-surface timescale hierarchy (thermal time shorter than accretion time) that justifies adding accreted gas at photosphere entropy.","marker":"Nomoto 1982"},{"why":"Supports the same thermal-versus-accretion timescale argument and the treatment of the accreted material's entropy.","marker":"Townsley & Bildsten 2004"},{"why":"Supplies the stellar evolution code whose accretion implementation (mass change, gravitational energy added to the outer cell, and wind prescription) generates the grid of models.","marker":"Paxton et al. 2011, 2013, 2015, 2018, 2019"}],"fun_headline_variants":["Below 0.01 solar masses per year, companion stays hot; above, it puffs up","Companion stars either sip or gulp: 0.01 solar masses per year is the dividing line","Massive companion stars balloon when accretion rate tops 0.01 solar masses yearly","Giant eruption accretion: low rates keep companion hot, high rates swell it"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The results depend on the assumption that the accreted gas adopts the surface temperature structure of the star immediately, so how hot or cold it was before landing does not matter, and that winds are turned off during the accretion phase; a different boundary-layer behavior or continuing winds would move the threshold and change the recovery.","fun_headline_variants_meta":{"raw":{"variants":["Below 0.01 solar masses per year, companion stays hot; above, it puffs up","Companion stars either sip or gulp: 0.01 solar masses per year is the dividing line","Massive companion stars balloon when accretion rate tops 0.01 solar masses yearly","Giant eruption accretion: low rates keep companion hot, high rates swell it"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000981,"raw_usage":{"total_tokens":4220,"prompt_tokens":1057,"completion_tokens":3163,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":673,"completion_tokens_details":{"reasoning_tokens":3067}},"tokens_in":673,"tokens_out":3163,"duration_ms":21746,"temperature":1.0,"reasoning_tokens":3067,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T05:41:09.613796+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete check is to repeat the same numerical experiment while varying what the paper assumes: assign the accreted gas a different entropy (for example, the donor's surface entropy instead of the photosphere's), or leave the stellar wind prescription on during the accretion phase. If the boundary between the two regimes moves away from $\\sim0.01\\,M_\\odot\\,\\mathrm{yr}^{-1}$, or if the high-rate inflation disappears, the claimed dichotomy is an artifact of those assumptions. Observationally, monitoring the companion of $\\eta$ Car or a similar system through its eruptive cycle and detecting whether it reaches $\\log T_{\\rm eff}\\simeq 3.5$-$3.7$ K and $R\\simeq 10^3\\,R_\\odot$ while $L$ jumps by $\\sim0.7$ dex would settle whether the predicted high-rate regime occurs in nature.","supporting_citations":[],"review_version":1}