{"id":"5f0df50f-34ac-46f7-b7e3-b867a63d56f3","arxiv_id":"2608.10750","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Simulations show wide primordial binaries are disrupted as efficiently as if no binaries formed, while close binaries survive, indicating close and equal-mass binary properties are imprinted at the star formation stage.","lead":"This study uses computer simulations of collapsing molecular clouds to track how pairs of newborn stars, or binaries, evolve while their birth cluster forms. It finds that only initially close binaries survive the chaotic environment, implying that close and equal-mass twin binaries are born during star formation, not created later by interactions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim leans on the q=1 primordial-binary prescription: the twin excess is partly an input, and equal-mass binaries are maximally hard, so both the multiplicity comparison and the twin-fossil inference need a realistic-q test to stand.","rationale":"The reader's weakest-assumption analysis is exactly right: the fixed q=1 prescription is the most load-bearing idealized choice. Equal-mass binaries are the hardest possible binaries for a given mass, so the close-binary model is systematically favored to survive, and the final q=1 excess is an input that survives rather than an emergent prediction. The authors are transparent about this in Section 4.3, but the caveat weakens the quantitative force of the central claim. No internal numerical inconsistency is apparent: the methods are clearly described, the code is established, and the qualitative dynamical picture (wide binaries disrupted, close hard binaries retained) is consistent with prior work. The main issue is that the headline inference to primordial close binaries and primordial twins has not been tested against the one assumption that most directly controls it. Because the reader already issued a CONDITIONAL verdict on these grounds, my stress-test does not move the verdict; it reinforces the need for a realistic-q follow-up as the decisive test.","tokens_in":16288,"tokens_out":3034,"duration_ms":34867,"concrete_test":"Run CB and WB variants with the same separation distributions but sample q from a realistic or flat distribution, e.g., q ∈ [0.1,1] uniform or the Moe & Di Stefano (2017) q-distribution, keeping all other parameters and random seeds fixed. If the final low-mass multiplicity in the close-binary model falls to within a factor of ~2 of the SS model, the quantitative conclusion fails; if it remains well above SS and produces a high-q but not delta-function q=1 excess, the qualitative primordial-close-binary claim survives but the twin-fossil claim does not. Also compare the final q distribution: a sharp q=1 peak appearing only in the q=1-input run would confirm that the twin excess is an input-output artifact.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline conclusion—that close binaries must form at the star formation stage and that the observed twin excess is set there—is most directly threatened by the fixed q=1 assumption in Section 2.2.2. Equal-mass binaries maximize binding energy for fixed total mass, so the CB model gives close binaries their best possible chance of surviving cluster formation; a realistic mass-ratio distribution would make many of them softer and more fragile, lowering the final multiplicity as the authors concede in Section 4.3. More importantly, the q≈1 excess in Fig. 10 is generated by construction: the simulation injects 100% q=1 binaries, and the residual excess is simply the fraction that survives. Inferring from this that the observed twin excess must be primordial is therefore partly circular. The qualitative ordering CB > WB ≈ SS may survive, but the quantitative claim that observed low-mass multiplicity and the twin excess require primordial close binaries has not been isolated from this favorable input. What is missing is a model that draws q from an observed or flat distribution; without it, the central inference is overdetermined.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper uses self-consistent N-body/SPH simulations (ASURA+BRIDGE with PETAR and SDAR) to follow the collapse of an isolated 5e3 Msun, 2 pc molecular cloud under three star-formation prescriptions: close binary formation (CB, a = 1-100 au), wide binary formation (WB, a = 100-1e4 au), and single-star formation (SS). All binary-forming models inject 100% binaries with q=1 and uniform eccentricity and semi-major-axis distributions, with three random seeds per model. The study shows that the multiplicity fraction declines in all models, that the WB model drops to roughly the SS level (~1%) while the CB model retains tens of percent, that high-mass stars maintain a relatively high multiplicity fraction in all models, that a q=1 excess survives in CB but is diluted in WB and absent in SS, and that the global cluster density structure is largely insensitive to the binary prescription. The authors conclude that close binaries must form at the star formation stage to reproduce the observed low-mass multiplicity fraction and the equal-mass-binary excess.","tokens_in":16486,"tokens_out":9969,"duration_ms":102234,"significance":"The controlled comparison of an otherwise identical cloud across CB, WB, and SS is a strength: the differential survival of hard versus soft primordial binaries is clearly shown with minimum-to-maximum ranges over three seeds, and the result that wide primordial binaries are dynamically erased to the single-star level while close binaries retain tens of percent is likely robust to the adopted simplifications. The simulations are technically demanding, coupling gas dynamics, stellar feedback, and regularized few-body dynamics, and the paper is honest about several caveats. However, because all injected binaries have q=1, the CB model assigns close binaries their maximal binding energy, and the q=1 excess in the final mass-ratio distribution is partly an input rather than an emergent prediction. The quantitative connection to observed low-mass multiplicity and twin fractions is therefore not yet established. With a realistic-mass-ratio test or suitably weakened conclusions, the paper would be a valuable step toward linking binary formation prescriptions to observed multiplicity in young clusters.","major_comments":[{"comment":"The inference that the observed q≈1 twin excess must be primordial is at least partly circular. The CB and WB models inject 100% of binaries with q=1 ('we fix q=1', §2.2.2), so the excess at q=1 in Fig. 10 is a survival count of injected twins, not evidence that a realistic primordial population would produce the observed twin excess. In addition, q=1 maximizes the binding energy for a given total mass, making the CB model the most favorable case for survival; the authors' own §4.3 caveat that a realistic mass-ratio distribution would lower the multiplicity fraction confirms that the quantitative level (20-30%) is not robust. I request either a model set with q drawn from a flat or observed distribution (e.g., Moe & Di Stefano 2017) with the same semi-major-axis ranges, or a substantive weakening of the §5 conclusion that reproducing the twin excess requires a non-negligible population of primordial twins.","section":"§2.2.2, Fig. 10, §5"},{"comment":"The claim that close binaries must form at the star formation stage 'to reproduce the observed multiplicity fraction of low-mass stars' is not quantitatively grounded. Fig. 8 compares the three models only; no observed low-mass multiplicity value or range is given, and the CB model combines a 100% primordial binary fraction with q=1, so its final multiplicity is not a prediction for a realistic star-forming population. Either overplot an observational benchmark (e.g., the field or ONC low-mass multiplicity from Raghavan et al. 2010 or Duchêne et al. 2018) and state the resulting margins, or explicitly reformulate the conclusion as a qualitative requirement for some close primordial binaries rather than a reproduction of observed levels.","section":"§3.2, Fig. 8, §5"}],"minor_comments":[{"comment":"The same symbol r is used for the scalar separation and the relative position vector; please use boldface for the vector to avoid confusion.","section":"§2.2.2, Eqs. (4)-(5)"},{"comment":"Table 1 should include the fixed values q=1 and the 100% binary formation fraction, since these are the key assumptions of the binary models.","section":"Table 1"},{"comment":"Please state explicitly whether the gas-mass enclosure check from §2.2.1 is applied to the total binary mass (m_p + m_s) or only to the primary mass.","section":"§2.2.2"},{"comment":"The multiplicity fraction of high-mass stars is based on only a handful of objects per run; please add sample sizes or a caution in the text.","section":"Fig. 11 and §4.1"},{"comment":"The final CB multiplicity is quoted as '20-30%' while Fig. 5 shows a time-dependent decline; please specify that this refers to t = 2.5 Myr.","section":"§5"},{"comment":"The three Kolmogorov-Smirnov tests are not corrected for multiple comparisons; the WB p-value of 0.057 would be even less significant under such a correction, so please temper the wording in that paragraph.","section":"§4.1"}],"recommendation":"major_revision","confidential_remarks":"The q=1 concern raised by the stress-test is real and is the main obstacle to accepting the headline conclusion. I would be willing to see a revised version with realistic-q simulations or with significantly softened claims about reproducing the observed multiplicity and twin excess. The paper otherwise fits the scope of Astronomy & Astrophysics and the simulation campaign is well executed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear X,\n\nWorth a look. The paper does something simple and useful: same turbulent 5000 M_sun cloud, same N-body/SPH code, and the only change is whether newly formed stars enter as close binaries (1–100 au), wide binaries (100–10^4 au), or singles. Three seeds for each; min/max ranges shown. The headline result is robust and not surprising: wide binaries get shredded down to the single-star level (~1% final multiplicity), close binaries hold on at tens of percent, and the host cluster structure barely changes. That last null result is clean and worth having.\n\nThe novelty is the controlled three-way comparison inside self-consistent cluster formation simulations, not just pre-specified cluster initial conditions. The code is well-established (ASURA+BRIDGE with PETAR/SDAR), parameters are explicit, and the caveats section is honest. This is a good, reproducible simulation paper.\n\nWhere it gets soft is the q=1 assumption. Equal-mass binaries are maximally hard for a fixed total mass, so the CB model's survival rate is an upper bound. The paper concedes in Section 4.3 that with realistic mass ratios the overall multiplicity fraction would likely decrease. That matters most for the twin-excess claim. The q about 1 excess in Fig. 10 is an input—they inject 100% twins and then show that twins survive, which is consistency, not evidence that observed twins must be primordial. So the headline \"close binaries need to form at the star formation stage\" is well supported qualitatively, but the quantitative multiplicity levels and the \"twin excess is a fossil\" inference are overdetermined.\n\nOther soft spots are minor: small high-mass samples (two dozen stars at most, 5/4/1 mergers), a single cloud realization (same turbulent seed), an artificial 100% binary fraction, and a uniform semi-major axis distribution. None of these sink the qualitative result; they do mean the paper is not the final word on matching observed field/ONC numbers.\n\nVerdict: send to a good referee. The fix is a realistic-q run (or at least a flat q distribution) and a clear separation of \"we show twins can survive\" from \"observed twins must be primordial.\" The core physics of hard/soft binary survival does not depend on q=1, so I expect the qualitative result to hold. I'd cite it for the wide-versus-close survival contrast.","headline":"A clean controlled experiment showing close primordial binaries survive cluster formation while wide ones do not, though the fixed q=1 prescription makes the quantitative twin-excess claim partly an input.","tokens_in":17062,"tokens_out":2733,"would_cite":true,"duration_ms":29285,"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":"Close binaries must form at the star formation stage, not by later dynamics","keywords":["binaries: close","stars: formation","galaxies: star clusters: general","primordial binaries","N-body/SPH simulations","multiplicity fraction","mass-ratio distribution","twin binaries"],"falsifier":"A survey of very young, still-embedded clusters (age ≲ 1 Myr) that finds no excess of equal-mass companions among close binaries would falsify the claim that the twin excess is set at birth. Equivalently, re-running the cluster-formation simulations with a realistic mass-ratio distribution and checking whether the close-binary model still keeps a 20–30 percent multiplicity fraction would test the quantitative conclusion directly.","tokens_in":16035,"feed_emoji":"🌟","tokens_out":8518,"duration_ms":72882,"temperature":0.7,"pith_summary":"This paper asks whether the binary stars we observe in young clusters are fossils of the star formation process or products of later dynamical encounters. The authors simulate the collapse of a molecular cloud into a star cluster with three extreme prescriptions for how stars are born: close binaries only, wide binaries only, or single stars only. They find that wide binaries are almost entirely destroyed during cluster formation, driving the multiplicity fraction down to about one percent, the same level as the single-star model, while close binaries survive and keep the multiplicity at 20–30 percent. Their central conclusion is that reproducing the observed multiplicity of low-mass stars and the excess of equal-mass 'twin' binaries requires that close binaries be formed at the star formation stage itself, not assembled afterwards by dynamics.","feed_headline":"Close binaries must be born, not made, simulations show","feed_subtitle":"Matching observed low-mass multiplicity and twin excess requires primordial close pairs; wide pairs get shredded.","key_machinery":"The central machinery is a suite of self-consistent N-body/smoothed-particle-hydrodynamics (SPH) simulations, run with the ASURA+BRIDGE code, that evolve a 5000 solar-mass, turbulent molecular cloud through collapse into a star cluster over about 2.5 Myr. A stochastic star formation module converts gas particles into stars, and a newly added binary formation module turns each eligible gas particle into a pair of stars with a prescribed semi-major axis, eccentricity, and mass ratio; the models differ only in whether the semi-major axis is drawn from 1–100 au (close) or 100–$10^{4}$ au (wide), with all binaries formed as equal-mass twins. Stellar dynamics is handled by the PETAR N-body code with slow-down algorithmic regularization (SDAR), which integrates hard binaries accurately. The physical axis that carries the argument is the hard–soft boundary near 100 au: hard binaries resist disruption and harden, soft binaries are shredded, and this separation is why the close-binary model leaves an observable imprint while the wide-binary model does not.","core_discovery":"On the paper's own terms, its central discovery is that the primordial binary population controls the final binary properties even though it barely affects the host cluster's structure. In the close-binary model the multiplicity fraction levels off at 20–30 percent, whereas in the wide-binary model it collapses to about one percent, comparable to what remains when stars form single. Because the simulation prescribes equal-mass birth companions, the surviving close binaries show a sharp peak at $q=1$, while the dynamically assembled binaries of the single-star model show a flat mass-ratio distribution; the authors take this contrast as evidence that the observed twin excess is imprinted at birth. The paper also shows that high-mass stars end up with high multiplicity in all models, so dynamical processing erases the formation imprint for massive primaries, but low-mass stars preserve it only if born close.","pith_inferences":["Because the equal-mass assumption maximizes binding energy, the quantitative multiplicity levels (20–30 percent for close binaries) are upper limits; a realistic mass-ratio distribution would lower them, as the authors concede, but the qualitative ordering of the models should persist.","One testable extension is that the close-binary fraction of low-mass stars should depend only weakly on the density of the natal environment, while the wide-binary fraction should drop sharply in dense clusters; comparing clusters formed from clouds of different densities would discriminate.","The inside/outside difference in multiplicity suggests that ejected runaway and walkaway stars in young clusters should preferentially be single or have different companion properties than cluster members, a prediction checkable with Gaia astrometry and radial-velocity surveys."],"forward_implications":["The observed low-mass multiplicity fraction in young clusters cannot be reproduced if stars form as singles or as wide binaries; a significant population of close primordial binaries is required.","The mass-ratio excess at $q=1$ in close binaries is a fossil of star formation: dynamically formed binaries show a flat mass-ratio distribution, so the twin excess must be set at birth.","Wide binaries are preferentially destroyed during cluster formation, so the deficiency of wide binaries in open clusters and the Orion Nebula Cluster is expected even when stars initially form in binaries.","The global structure of the forming cluster is insensitive to the primordial binary population because the gas potential dominates, so binary heating plays a minor role during the embedded phase."],"supporting_citations":[{"why":"Earlier magnetohydrodynamic + N-body simulations that first followed primordial binary evolution during cluster formation and showed the binary fraction decreases; this paper extends that work by systematically comparing binary prescriptions.","marker":"Cournoyer-Cloutier et al. 2021"},{"why":"Simulations showing binary properties are more strongly affected in denser and more massive clouds; used to contextualize the hard/soft survival trend.","marker":"Cournoyer-Cloutier et al. 2024b"},{"why":"Establishes the hard–soft binary classification that determines which primordial binaries survive dynamical encounters.","marker":"Heggie 1975"},{"why":"Observational survey reporting a twin excess among close OB binaries; the target the simulations aim to explain.","marker":"Moe & Di Stefano 2017"},{"why":"Field-star survey providing the reference multiplicity fraction and twin excess for solar-type stars.","marker":"Raghavan et al. 2010"},{"why":"Finds no strong correlation between close binary fraction and cluster age, supporting the interpretation that close binaries survive cluster formation.","marker":"Kounkel et al. 2019"},{"why":"Provides the stellar initial mass function used to sample primary masses in the star formation model.","marker":"Kroupa 2001"}],"fun_headline_variants":["Close binaries are born, not made, in star cluster simulations","Equal-mass twins are imprinted at birth, not by dynamics","Wide binaries get shredded; only close binaries survive to match data","Primordial close binaries needed to explain twin excess","Twin stars are born together, not dynamically assembled"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that all primordial binaries form as equal-mass twins, which gives each binary the maximum binding energy for its total mass; the authors themselves note that a realistic spread of mass ratios would make binaries more fragile and would likely lower the overall multiplicity fraction.","fun_headline_variants_meta":{"raw":{"variants":["Close binaries are born, not made, in star cluster simulations","Equal-mass twins are imprinted at birth, not by dynamics","Wide binaries get shredded; only close binaries survive to match data","Primordial close binaries needed to explain twin excess","Twin stars are born together, not dynamically assembled"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001493,"raw_usage":{"total_tokens":6030,"prompt_tokens":1019,"completion_tokens":5011,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":635,"completion_tokens_details":{"reasoning_tokens":4929}},"tokens_in":635,"tokens_out":5011,"duration_ms":30263,"temperature":1.0,"reasoning_tokens":4929,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T18:04:25.542964+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A survey of very young, still-embedded clusters (age ≲ 1 Myr) that finds no excess of equal-mass companions among close binaries would falsify the claim that the twin excess is set at birth. Equivalently, re-running the cluster-formation simulations with a realistic mass-ratio distribution and checking whether the close-binary model still keeps a 20–30 percent multiplicity fraction would test the quantitative conclusion directly.","supporting_citations":[{"cited_title":"2021, MNRAS, 501, 4464","cited_arxiv_id":null,"evidence_quote":"Earlier magnetohydrodynamic + N-body simulations that first followed primordial binary evolution during cluster formation and showed the binary fraction decreases; this paper extends that work by systematically comparing binary prescriptions."},{"cited_title":"2019, AJ, 157, 196","cited_arxiv_id":null,"evidence_quote":"Finds no strong correlation between close binary fraction and cluster age, supporting the interpretation that close binaries survive cluster formation."}],"review_version":1}