{"id":"6cbc44ed-51d2-472f-a186-06ec550f82f4","arxiv_id":"2502.00895","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"In simulations, tulip flame formation in tubes is driven by a rarefaction wave from the decelerating flame, with laminar flame velocity controlling the speed and depth of the tulip.","lead":"Using high-resolution two-dimensional simulations of hydrogen/air and methane/air flames, the authors argue that tulip flame formation in tubes is driven by a rarefaction wave created when the flame decelerates, not by flame-front instabilities. The work matters for hydrogen safety because it explains why faster flames form deeper tulips faster, a behavior relevant to explosion development in pipes and ducts.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Attribution of faster/deeper tulip formation to laminar flame velocity is confounded: H2/air vs CH4/air differ in reaction order, flame thickness, expansion ratio, and sound speed, so the reaction-order and U_f conclusions are not independently identified.","rationale":"The reader's weakest assumption is the same one I identify: the reaction-order conclusion is confounded. I agree with the reader's CONDITIONAL verdict. This is the most load-bearing concern because it targets the paper's causal attribution—not the existence of the rarefaction wave (which the simulations do support) but the claim that laminar flame velocity, rather than reaction order or flame thickness, is the main factor. A controlled variation of n would settle it. I also note the separate quantitative slip flagged by the reader: the printed ratio τ_H/τ_CH ≈ U_H/U_CH ≈ 6.3 is directionally inconsistent with faster flames forming tulips sooner; under the paper's own Eq. (12) the time ratio should be the inverse. That is a corrigendum-level issue but not the primary structural weakness. Because the conditional verdict already requires softening or justifying the reaction-order claim, my read leaves the verdict unchanged.","tokens_in":16164,"tokens_out":6450,"duration_ms":63557,"concrete_test":"Run 2D DNS with a synthetic one-step (or reduced) mechanism in which the global pressure exponent / reaction order n is varied (e.g., 1.0, 1.5, 2.0, 2.5) while transport coefficients and heat release are tuned so that U_f, L_f, and Θ match the CH4/air reference values at 1 atm; measure time from ignition to inversion and tulip depth. If inversion time/depth are nearly unchanged across n, the reaction-order claim holds; if they move substantially, the conclusion must be revised to say only that the combined mixture-property differences correlate with tulip formation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing weakness is in Section 4 and Appendix B. The paper concludes that reaction order has little effect on tulip formation and that laminar flame velocity plus flame acceleration are the main controlling factors. The evidence, however, is a two-point comparison: H2/air (n≈2, U_f=2.43 m/s, L_f=0.0325 cm, Θ=8.34, a_s=408 m/s) versus CH4/air (n≈1.1, U_f=0.38 m/s, L_f=0.0451 cm, Θ=7.2, a_s=332 m/s). These systems differ simultaneously in at least five flame/mixture properties; no simulation or analysis varies reaction order while holding U_f, L_f, Θ fixed. The theoretical support, Eq. (12)–(13), isolates U_f only under an assumed Clanet–Searby geometry at fixed Θ and D, so it cannot exclude flame thickness or reaction-order effects. Therefore the stated conclusion that U_f is the main factor, and the corollary that reaction order is unimportant, is an underdetermined attribution rather than a demonstrated result. This is the weakest link in the paper's causal chain from mixture chemistry to rarefaction-wave intensity to tulip depth/time.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports 2D direct numerical simulations of stoichiometric H2/air and CH4/air flames in closed tubes with aspect ratios 6 and 12 and in a semi-open tube, using the AMReX-based PeleC solver with detailed chemical mechanisms and adaptive mesh refinement. The central claim is that the rarefaction wave generated when the flame decelerates after the flame skirt touches the sidewall is the principal physical mechanism of tulip flame formation, with faster flames producing stronger rarefaction waves and hence faster, deeper tulip flames. The paper compares H2/air (laminar flame velocity 2.43 m/s, reaction order ~2) with CH4/air (0.38 m/s, reaction order ~1.1) and concludes that laminar flame velocity and flame acceleration control tulip formation while reaction order has little effect.","tokens_in":16349,"tokens_out":4828,"duration_ms":47534,"significance":"If the mechanism is correct, this is a useful contribution to a long-standing debate about tulip flame formation. The simulations are physically detailed, cover several tube geometries and boundary conditions, and yield a concrete, experimentally testable prediction: faster flames form tulips more quickly and with deeper petals, with amplification in closed tubes compared with semi-open tubes. The paper also gives explicit time-scale estimates suggesting that Darrieus-Landau and thermal-diffusive instabilities are too slow to be responsible for the initial flame inversion. The main weakness is that the causal attribution to laminar flame velocity is supported only by a two-point comparison of mixtures that differ in several properties simultaneously.","major_comments":[{"comment":"The conclusion that \"the reaction order has little effect on the formation of tulip flames\" is not supported by the presented evidence. The H2/air and CH4/air cases differ simultaneously in laminar flame velocity, flame thickness, expansion ratio, sound speed, and reaction order (Tables 1 and 2). No simulation varies reaction order while holding the other flame properties fixed, so the observed differences cannot be uniquely attributed to reaction order. The fitted n values in Fig. B1 only characterize the two chosen mixtures; they are not a controlled variation. This claim should be removed, explicitly reframed as a hypothesis, or supported by targeted simulations in which n is varied at fixed Uf, Lf, and expansion ratio.","section":"Section 4 and Appendix B"},{"comment":"The attribution of faster and deeper tulip formation to laminar flame velocity as \"the main factor\" is underdetermined by the presented comparison. The theory in Eqs. (12)-(13) is derived for fixed expansion ratio and channel width under the Clanet-Searby geometric model; it shows that acceleration increases with Uf when the other parameters are held fixed, but it cannot exclude flame-thickness, sound-speed, or Lewis-number effects when comparing H2/air with CH4/air. A more honest statement is that the simulations are consistent with the rarefaction-wave mechanism and with a stronger rarefaction wave for the faster flame, not that laminar flame velocity has been isolated as the unique controlling parameter.","section":"Section 3.1, Section 4, Eqs. (12)-(13)"},{"comment":"The manuscript states in Section 3.1 that tulip flame formation for H2/air occurs \"about ten times faster\" than for CH4/air, but Section 4 later claims that the ratio of tulip formation times is approximately equal to the laminar-flame-velocity ratio of 6.3. These two statements are inconsistent as written. The quantitative claim needs a clear definition of tau_tulip (from which figure and which time marker), a reported value, and an estimate of uncertainty; otherwise the proportionality between tulip time and laminar flame velocity is not established.","section":"Section 3.1 vs Section 4"}],"minor_comments":[{"comment":"The adiabatic flame temperature for H2/air is given as 2350 K in the text and 2503 K in Table 1; please reconcile these values.","section":"Section 2.2 and Table 1"},{"comment":"The sentence beginning \"On the contrary, in the case of a methane/air flame...\" appears twice almost verbatim in the same section; one occurrence should be deleted.","section":"Section 3.1"},{"comment":"The numerical value lambda_max approximately 0.2 cm does not appear consistent with the formula lambda = (2 pi / 0.3) Lf using the tabulated flame thickness values; please check the formula, the units, or the quoted number.","section":"Section 4, Eq. (17)"},{"comment":"Reference [50] is missing its closing bracket, and several inline equations in Section 4 (e.g., the acceleration values) are garbled by the typesetting; these should be corrected in the final manuscript.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The confounded two-mixture comparison is the main technical weakness, and it is fixable within the scope of the paper by softening the reaction-order and 'main factor' claims or by adding controlled parametric simulations. The paper relies heavily on the authors' own prior work [26], but it is not circular: the new simulations independently reproduce the rarefaction-wave mechanism for a slower mixture and in a semi-open tube. The novelty relative to [26] should nevertheless be stated more crisply. On balance, the manuscript is within scope and publishable after a major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper is a useful comparative DNS study of tulip flame formation in H2/air versus CH4/air, and it does a reasonable job of showing that the rarefaction-wave mechanism, established in the authors' earlier paper [26], carries over to a fast and a slow mixture across three tube geometries. That comparison is the actual new contribution, and it is worth having.\n\nThe simulations look credible: AMR with detailed chemistry, a convergence check in Appendix A, and a clear qualitative difference in tulip depth and formation speed between the two fuels. The schlieren sequences and velocity profiles are consistent with the piston/rarefaction-wave picture. The claim that faster flames produce stronger rarefaction waves and therefore deeper, faster tulips is plausibly supported by the simulations.\n\nThe soft spots are in the interpretive claims. The conclusion that reaction order has little effect (Section 4) is not supported by the evidence. You have two mixtures differing simultaneously in laminar flame speed, flame thickness, expansion ratio, sound speed, and reaction order. Nothing in the paper isolates reaction order. The scaling in Eq. (12)–(13) only shows acceleration growing with U_f under the Clanet–Searby geometry at fixed expansion ratio and tube width; it says nothing about reaction order or flame thickness. So the 'little effect' conclusion is an underdetermined attribution, not a demonstrated result.\n\nThere is also an internal inconsistency in the time-ratio statement. Section 3.1 says H2/air forms a tulip about ten times faster than CH4/air. Section 4 says the ratio of tulip formation times is approximately the ratio of laminar flame speeds, 6.3. These do not agree, and the paper does not report measured tulip formation times from the simulations. That needs to be reconciled or the phrasing softened. Not having code or input files also makes it harder to independently check the simulation details, though the resolution study helps.\n\nWho is this for? Combustion researchers interested in flame dynamics in tubes, especially hydrogen safety. It deserves a serious referee because the simulations are carefully done and the comparative data are useful. With revisions to fix the reaction-order claim and the time-ratio inconsistency, it could be a solid paper.","headline":"Useful comparative DNS of tulip flames in H2/air versus CH4/air that supports the rarefaction-wave mechanism, but overclaims on reaction order and has an internal inconsistency in the time-ratio scaling.","tokens_in":16944,"tokens_out":2627,"would_cite":false,"duration_ms":25209,"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":"The paper argues that tulip flame formation is driven by a rarefaction wave created when the flame decelerates, not by flame-front instabilities.","keywords":["tulip flame","rarefaction wave","laminar flame velocity","flame acceleration","flame front inversion","hydrogen/air","methane/air","compressible reactive flow"],"falsifier":"Run two simulations with the same laminar flame velocity and flame thickness but with global reaction orders near n=2 and n=1.1, for example by tuning the pressure dependence of the burning rate, and measure the time from ignition to flame inversion in a closed tube of aspect ratio 6; the rarefaction-wave claim predicts nearly identical inversion times, whereas reaction-order control predicts a measurable shift.","tokens_in":15885,"feed_emoji":"🔥","tokens_out":6558,"duration_ms":64941,"temperature":0.7,"pith_summary":"This paper aims to establish what physically turns a propagating flame front inside out to form the classic tulip flame in a tube. Using fully compressible reactive Navier-Stokes simulations with detailed chemistry for hydrogen/air and methane/air mixtures, it concludes that the rarefaction wave generated when the flame decelerates as its skirt touches the sidewalls is the primary mechanism. The strength of that wave is set mainly by the laminar flame velocity and the associated flame acceleration, which explains why faster flames form tulips sooner and with a deeper concave shape. The authors also argue that standard hydrodynamic flame-front instabilities act too slowly to cause the initial inversion.","feed_headline":"Rarefaction wave, not instability, flips flame into tulip shape","feed_subtitle":"Hydrogen/air and methane/air simulations trace the tulip inversion to a rarefaction wave born during flame deceleration.","key_machinery":"The load-bearing mechanism is the simple rarefaction wave generated by a decelerating flame, treated as a convex piston withdrawing from the unburned gas. The paper couples this piston analogy to a thin-flame geometric model in which the flame tip advances during the finger phase according to dX_tip/dt = Theta U_fL X_tip/D, so the flame acceleration scales with the square of the laminar flame velocity. When the finger-shaped flame skirt contacts the sidewalls, the resulting deceleration produces a rarefaction wave whose axial velocity profile is nonuniform across the tube, and that nonuniform reverse flow is what turns the flame front concave. The paper also uses time-scale estimates for the rarefaction wave versus instability growth to argue that hydrodynamic instabilities are not responsible for the initial tulip formation.","core_discovery":"The central claim is that tulip flame inversion is a purely hydrodynamic piston effect. During acceleration the flame acts like a semi-transparent accelerating piston; when its lateral parts touch the sidewalls the flame surface shrinks, the flame decelerates, and in the unburned gas this is equivalent to a withdrawing piston that generates a rarefaction wave. In a channel this reverse flow is fastest near the tube axis and slower near the walls, imprinting a mirror-tulip axial velocity profile on the unburned gas just ahead of the flame. Since each flame element moves with the sum of the local laminar burning velocity and the local gas velocity, the front inverts. The paper shows this inversion happens faster than the characteristic times of Darrieus-Landau or thermal-diffusive instabilities, and it attributes the intensity of the rarefaction wave primarily to laminar flame velocity and flame acceleration, concluding that reaction order has little effect on tulip formation itself.","pith_inferences":["If reaction order really is secondary for tulip formation, simplified one-step chemistry may be adequate for predicting tulip timing in engineering models, provided it reproduces the laminar flame velocity and flame thickness.","For hydrogen/methane blends, tulip formation time and depth should interpolate with the blended laminar flame velocity; this follows from the mechanism but is not simulated in the paper.","The piston analogy implies an acoustic tuning effect: in a closed tube, changing tube length shifts when reflected pressure waves arrive, which should either reinforce or distort the tulip, and that dependence could be mapped experimentally.","A natural test is to vary the thermal expansion ratio or flame thickness while holding laminar flame velocity fixed in simulations, to see which parameters actually control rarefaction intensity."],"forward_implications":["Faster-burning mixtures will form tulip flames sooner and with a deeper concave pocket, because the deceleration-born rarefaction wave is stronger.","In semi-open tubes, tulip formation still occurs but later and more smoothly, since no reflected pressure waves reinforce the initial rarefaction wave.","In longer closed tubes, reflected pressure waves reach the flame later, so the transition to a distorted tulip flame is delayed compared with shorter tubes.","Flame-front instabilities such as Darrieus-Landau and thermal-diffusive instabilities are not the trigger for the initial flame-front inversion, because the rarefaction-wave time scale is much shorter.","At sufficiently low laminar flame speeds, the rarefaction wave may be too weak to form a tulip before buoyancy or instability effects intervene."],"supporting_citations":[{"why":"Establishes the rarefaction-wave mechanism for tulip flame formation and supplies prior resolution and convergence tests.","marker":"[26]"},{"why":"Supplies the geometric finger-flame model and the four characteristic times used to compare flame propagation regimes.","marker":"[19]"},{"why":"Provides the thin-flame model and the pressure dependence of laminar flame velocity and thickness used in the argument.","marker":"[44]"},{"why":"Treats the flame as a gasdynamic discontinuity, which is the basis for the piston analogy.","marker":"[45]"},{"why":"Provides the classical solution for a rarefaction wave generated by a withdrawing piston.","marker":"[46]"},{"why":"Gives experimental evidence that flame front inversion occurs faster than instability growth.","marker":"[27]"},{"why":"Analyzes how flame collisions with reflected pressure waves enhance the initial rarefaction effect and later produce distorted tulips.","marker":"[53]"}],"fun_headline_variants":["Tulip flame flip traced to deceleration's rarefaction wave","Flame inversion: rarefaction wave beats instability","Hydrodynamic piston effect shapes tulip flame in tubes","Tulip flame forms from deceleration wave, not instability"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper attributes the hydrogen/air versus methane/air difference mainly to laminar flame velocity, even though the two mixtures also differ in flame thickness and reaction order, so the conclusion that reaction order has little effect is inferred rather than isolated.","fun_headline_variants_meta":{"raw":{"variants":["Tulip flame flip traced to deceleration's rarefaction wave","Flame inversion: rarefaction wave beats instability","Hydrodynamic piston effect shapes tulip flame in tubes","Tulip flame forms from deceleration wave, not instability"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000552,"raw_usage":{"total_tokens":2624,"prompt_tokens":927,"completion_tokens":1697,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":543,"completion_tokens_details":{"reasoning_tokens":1628}},"tokens_in":543,"tokens_out":1697,"duration_ms":11733,"temperature":1.0,"reasoning_tokens":1628,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T17:20:11.217772+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run two simulations with the same laminar flame velocity and flame thickness but with global reaction orders near n=2 and n=1.1, for example by tuning the pressure dependence of the burning rate, and measure the time from ignition to flame inversion in a closed tube of aspect ratio 6; the rarefaction-wave claim predicts nearly identical inversion times, whereas reaction-order control predicts a measurable shift.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the rarefaction-wave mechanism for tulip flame formation and supplies prior resolution and convergence tests."},{"cited_title":"Clanet, G","cited_arxiv_id":null,"evidence_quote":"Supplies the geometric finger-flame model and the four characteristic times used to compare flame propagation regimes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the thin-flame model and the pressure dependence of laminar flame velocity and thickness used in the argument."},{"cited_title":"Matalon, B.J","cited_arxiv_id":null,"evidence_quote":"Treats the flame as a gasdynamic discontinuity, which is the basis for the piston analogy."},{"cited_title":"Landau, E.M","cited_arxiv_id":null,"evidence_quote":"Provides the classical solution for a rarefaction wave generated by a withdrawing piston."},{"cited_title":"Ponizy, A","cited_arxiv_id":null,"evidence_quote":"Gives experimental evidence that flame front inversion occurs faster than instability growth."},{"cited_title":"Qian and M","cited_arxiv_id":null,"evidence_quote":"Analyzes how flame collisions with reflected pressure waves enhance the initial rarefaction effect and later produce distorted tulips."}],"review_version":1}