{"id":"f41bfff5-b417-4ff2-943c-6dafe07a5bcc","arxiv_id":"2607.16367","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"On grooved superhydrophobic surfaces a droplet's rebound splits into a reciprocal transverse/axial mode pair, making contact time Weber-dependent and selectable by groove width.","lead":"A simulation study shows that a groove under a bouncing droplet splits its rebound into two linked time scales, so contact time now depends on impact speed, unlike on a flat surface. The result offers a design rule: on grooved water-repellent surfaces, choose the groove width to select which rebound mode controls lift-off.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Axial-mode evidence for the reciprocal pair is too weak: Eq. (13) rests on a single proxy over N^{1/2}∈[1.10,1.25], so the central two-mode claim is not yet established.","rationale":"The reader's weakest assumption is the geometric closure Eq. (15), and that is a legitimate concern: it invalidates the exact imposed-parameter forms Eqs. (17)–(20) when the fitted width exponent is p=1.51 rather than 1. However, that failure does not destroy the two-mode picture itself; it only weakens the mapping from N to (We, W). The more load-bearing gap for the central claim is the axial mode. The manuscript itself concedes in Sec. V B that the τ∥ proxy is imperfect, that the window is narrow, that post-detachment dynamics contaminate the signal, and that high-We data are lost to Rayleigh–Plateau breakup. Those statements are in-scope evidence of missing support, and I weigh them heavily. With only N^{1/2}∈[1.10,1.25], the N^{1/2} scaling and the reciprocal product Eq. (14) are not independently established. The mode-selection transition in Sec. V C then rests on an unvalidated input, Eq. (23b). This does not change the reader's conditional verdict—the transverse scaling and experimental validation are real, so rejection would be too harsh—but it does mean the central reciprocal-pair claim should remain conditional pending stronger axial-mode data. A focused simulation campaign extending the N range and measuring τ∥ more directly would settle the point.","tokens_in":21940,"tokens_out":12215,"duration_ms":145100,"concrete_test":"Run a targeted non-wetting groove series that extends the in-regime blob number to at least N≈3–4 (e.g., W/R0=2.0–2.5 with We up to 50–70 in an enlarged/refined domain to postpone Rayleigh–Plateau breakup), and measure τ∥ from the power-spectral peak of K_y(t) over at least two full axial oscillations, separated into in-contact and post-detachment windows. Then fit τ∥/τ0 = C N^{1/2} with C fixed by the transverse-mode prefactor from Fig. 5, and check whether τ⊥τ∥/τ0^2 is constant across all N. If the N^{1/2} scaling fails or the product drifts, Eqs. (13)–(14) are not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central novelty is the reciprocal pair, not just the transverse scaling. The transverse leg is well supported by the measured-N collapse (Fig. 5), but the axial leg—Eq. (13), τ∥/τ0 ∼ N^{1/2}—is the load-bearing part that makes the response a pair, and it is tested almost entirely through Fig. 10. That test has three weaknesses. (i) The observable is a proxy: twice the interval between the first and second K_y peaks (events d and i, Fig. 4), which spans detachment and therefore includes free-drop Rayleigh oscillation; the authors themselves note post-detachment relaxation toward radially symmetric modes. (ii) The usable N-range is tiny: the abscissa N^{1/2} spans only about 1.10–1.25 (N≈1.2–1.6), so a straight line with free slope/intercept cannot distinguish N^{1/2} from a constant or a weak logarithmic trend. (iii) No fit coefficients, residuals, or per-width collapse are reported for Fig. 10, and high-We points are missing due to breakup. If τ∥ is actually N-independent—the free-drop Rayleigh time—then Eq. (14) is an artifact and the 'reciprocal pair' reduces to one groove-imposed mode. The wetting mode-selection transition inherits this because Eq. (23b) uses τ∥∼N^{1/2}, and χ⋆ is calibrated from the same branch-separation data, so Figs. 16–17 do not independently confirm the axial mode.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports entropic multi-relaxation-time lattice Boltzmann simulations of droplet impact on grooved superhydrophobic surfaces, validating the method against Chantelot et al. and then extending the blob model from a single transverse mode to a reciprocal pair: τ⊥/τ0 ∼ N^{-1/2} and τ∥/τ0 ∼ N^{1/2}, so that τ⊥τ∥ ∼ τ0^2. It further proposes imposed-parameter forms τ⊥/τ0 ∝ (W/R0) We^{-1/4} and τ∥/τ0 ∝ We^{1/4}/(W/R0), and interprets a two-branch response on partially wetting grooves as a mode-selection transition controlled by χ ∼ We^{1/2}/(W/R0).","tokens_in":22528,"tokens_out":5008,"duration_ms":62477,"significance":"If the reciprocal-pair picture is correct, it is a meaningful conceptual extension: anisotropic confinement would not merely shorten one rebound mode but would replace the degenerate flat-plate mode with two conjugate inertio-capillary modes, making the contact time Weber- and groove-width-dependent. The paper has genuine strengths: the measured-N collapse of the transverse contact time across five groove widths (Fig. 5) is a non-trivial, parameter-light result; the flat-plate validation is quantitative; and the authors are unusually explicit about the limitations of their geometric closure. The axial mode, however, is the load-bearing element for the central claim, and its evidence is currently too weak. The manuscript's own caveats in Sec. VI support a major revision rather than acceptance as-is.","major_comments":[{"comment":"The imposed-parameter prediction fails quantitatively on groove-width dependence. Equation (18) gives N ∼ We^{1/2}/(W/R0)^2, hence τc/τ0 ∝ (W/R0) We^{-1/4}, but the fitted exponent in Eq. (21) is p=1.51 with leave-one-out range 1.45–1.68, not p=1. The paper correctly attributes this to the geometric closure, and Figs. 8–9 show an additional W-dependence in Smax (p≈0.94 in Eq. 22) and ℓmax (width exponent ≈−1.63, and nearly coincident values for the two widest grooves). Because Eq. (18) is used in Eqs. (19), (20), and also in the mode-selection parameter χ of Eq. (24), all imposed-parameter predictions built on this closure are only directional unless the closure is revised. The manuscript should either repair the closure or state more forcefully that Eqs. (19)–(20) are not predictions but approximate heuristics. The current text presents them as predictions and then reports the fitted p","section":"Sec. V B, Eq. (19)"},{"comment":"The axial-mode scaling τ∥/τ0 ∼ N^{1/2} is not established by the presented data. The proxy is twice the interval between the first and second Ky peaks (events d and i in Fig. 4), which spans detachment and therefore includes post-detachment free-drop Rayleigh oscillations; the text itself notes that after detachment the oscillation relaxes toward radially symmetric modes. The usable N range is tiny: in Fig. 10 the abscissa N^{1/2} covers only about 1.10–1.25 (N ≈ 1.2–1.6), so a straight line with free slope and intercept cannot distinguish N^{1/2} from a constant or a weak logarithmic trend. No fit coefficients, residuals, per-width collapse, or uncertainty estimates are reported for this fit. Since the reciprocal relation Eq. (14) and the two-mode interpretation depend on Eq. (13), this is a load-bearing gap. I would like to see either (i) a more direct in-contact measurement of the axi","section":"Sec. V B, Eq. (13) and Fig. 10"},{"comment":"The mode-selection analysis is partly circular as presented. The transition parameter χ uses τ∥∼N^{1/2} from Eq. (23b), whose validity is not independently confirmed (see previous comment), and the final relation in Eq. (24) uses Eq. (18), whose width dependence is contradicted by the data. Moreover, the threshold χ*≈1.61 is calibrated from the same branch-separation data that the model is then said to explain, and the paper admits that We* ∝ (W/R0)^2 cannot be tested quantitatively because the Weber-number sampling is too coarse. The two-branch structure is interesting and plausibly consistent with the model, but the current evidence does not independently confirm the axial mode or the transition law. The authors should reframe Sec. V C as a consistency check and clearly separate calibrated thresholds from falsifiable predictions.","section":"Sec. V C, Eqs. (23)-(25), Figs. 16-17"}],"minor_comments":[{"comment":"Typo: 'The precise precise form' should be 'The precise form'.","section":"Sec. II, after Eq. (2)"},{"comment":"Report the slope, intercept, R², and residuals of the linear fit. Show per-width symbols and indicate the excluded high-We breakup points explicitly.","section":"Fig. 10"},{"comment":"Provide confidence intervals or standard errors for C, b, and p. The leave-one-out spread of p is informative but not a substitute for fit uncertainty.","section":"Sec. V B, Eq. (21)"},{"comment":"Typo: 'exlcluded' should be 'excluded'.","section":"Fig. 12 caption"},{"comment":"Define Ur, Rv, and τ∥ clearly at the point of introduction; the notation Rv is used before it is defined.","section":"Sec. V C, paragraph before Eq. (24)"},{"comment":"The N>1.2 cutoff is selected from the observed departure from linearity, not from an a priori criterion. Please show sensitivity of the fitted exponents and R² to this cutoff (e.g., N>1.3 or N>1.15).","section":"Sec. V B, threshold N>1.2"}],"recommendation":"major_revision","confidential_remarks":"The transverse scaling is the solid core of the paper, and the simulation study is competently executed. The main risk is overclaiming the reciprocal pair: the axial mode is supported by a narrow, post-detachment proxy, and the imposed-parameter predictions fail quantitatively. I would not reject, but I would require either substantially stronger axial-mode evidence or a clear reframing of the claims as exploratory/directional. The authors' own limitation statements in Sec. VI are consistent with this assessment."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, two things you should know about this paper. First, the transverse mode is in good shape: contact-times from five groove widths collapse onto tau_c/tau0 ~ N^{-1/2} once N is measured from the simulation (Fig. 5), and the We^{-1/4} trend is robust. Second, the companion axial mode, tau_parallel ~ N^{1/2}, is the load-bearing half of the 'reciprocal pair' and it is not established. The test in Fig. 10 uses a proxy that spans detachment into free flight, over a tiny N^{1/2} window (about 1.10 to 1.25), with no reported fit statistics. That window cannot distinguish N^{1/2} from a constant. If tau_parallel is actually the N-independent Rayleigh time, the reciprocal product in Eq. (14) is an artifact.\n\nWhat the paper does well: it is transparent. The authors flag the flat-plate area closure as a strong assumption, show that the fitted groove-width exponent (p=1.51) disagrees with their predicted p=1, and discuss the restricted validity window. The LB method is carefully calibrated: surface tension is measured from Laplace fits, wetting angles from static columns, and the flat-plate contact time comes out at 2.63 tau0 with the experimental Chantelot groove data reproduced reasonably. The wetting-groove two-branch structure is interesting, and the mode-selection parameter chi is a plausible organizing idea, but the transition chi* ~1.61 is fitted from the same branch-separation data and inherits the axial-mode uncertainty.\n\nThe soft spots are real but not fatal to everything. The measured-N transverse scaling stands on its own and is a useful result. What does not stand is the imposed-parameter prediction: Eq. (19) fails quantitatively on W-dependence, and the paper admits this. The axial mode and the wetting transition need better evidence: either a wider N-range, a cleaner observable, or direct experimental confirmation.\n\nWho should read this: people working on droplet rebound over textures. It deserves a serious referee, but the referee should ask for code/data, a re-analysis of Fig. 10 with fixed exponents and residuals, and ideally an experimental test of the two-branch transition. I would not cite it yet for the reciprocal pair; I would cite it for the measured-N collapse if that is all I needed.","headline":"A credible transverse-mode collapse and an honest but under-supported axial-mode claim; the paper deserves refereeing, with data and better axial evidence requested.","tokens_in":22904,"tokens_out":1995,"would_cite":false,"duration_ms":21987,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["76D45","76M28"],"pacs":["47.55.D-"],"model":"deepseek-v4-flash","headline":"A groove that fixes a droplet's spread in one direction splits rebound into a fast transverse mode and a slow axial mode whose time scales multiply to the square of the flat-plate inertio-capillary time, making contact time depend on Weber","keywords":["droplet rebound","superhydrophobic surfaces","contact time","inertio-capillary scaling","anisotropic confinement","grooved substrates","blob model","Weber number"],"falsifier":"Measure, at fixed Ohnesorge number and a single non-wetting groove width, both contact time τ_c and the axial recoil half-period across Weber numbers from about 3 to 30. If τ_c does not fall as We^{-1/4} once the droplet is well confined (N>1.2), or if the product of the transverse and axial times drifts away from τ0² by more than experimental scatter, the reciprocal-mode claim fails. A second check on a wetting groove: the predicted collapse of the two branches at χ≈1.61 and the scaling We*∝(W/R0)² are directly testable, and their absence would falsify the mode-selection mechanism.","tokens_in":21839,"feed_emoji":"💧","tokens_out":8084,"duration_ms":82988,"temperature":0.7,"pith_summary":"On a flat superhydrophobic surface, a bouncing droplet's contact time is set by one inertio-capillary time and is independent of impact energy, because spreading is radially symmetric. The paper argues that a grooved surface breaks that symmetry: the groove fixes the spreading length in one direction and leaves the other free, so the rebound splits into two modes with inversely related time scales. The fast transverse mode, set by recoil between the groove walls, scales as the inverse square root of a blob number; the slow axial mode, set by retraction of the elongated droplet, scales as its square root, so their product is the flat-plate time squared. The visible contact time then becomes a mode-selection outcome: on non-wetting grooves it follows the fast transverse mode and falls with Weber number, while on mildly wetting grooves a slow axial branch and a fast transverse branch compete, with the transition set by a single parameter. If this picture is right, contact-time reduction on textured surfaces becomes a matter of choosing between two conjugate inertio-capillary modes rather than accelerating a single rebound mode.","feed_headline":"A groove splits droplet rebound into two linked times","feed_subtitle":"Fast transverse recoil and slow axial recoil trade off, coupling contact time to impact energy and groove width.","key_machinery":"The load-bearing object is the blob number N = ℓ_max/W, the ratio of the droplet's axial length at maximum surface area to the groove width. The paper extends the blob model by pairing the usual transverse mode with a second axial mode, derived both from thin-sheet rim retraction and from a capillary-force/constant-acceleration argument, so the exponent is not tied to one mechanism. Converting N into imposed parameters requires two geometric closures: volume conservation for a rectangular prism, W ℓ_max h ~ R0³, and a flat-plate area law Smax/R0² ~ We^{1/2} assumed to hold independent of groove width. The mode-selection picture on wetting grooves is carried by the parameter χ, which compares","core_discovery":"The central claim is that anisotropic confinement, realized by a groove of width W, imposes a fixed transverse length and thereby breaks the radial degeneracy of flat-plate rebound. Extending the blob model from one transverse scale to two, the paper represents the spread droplet as a rectangular prism of width W, axial length ℓ_max, and thickness h, partitionable into N=ℓ_max/W isotropic blobs. The transverse recoil mode has time τ⊥/τ0 ~ N^{-1/2}, and the axial retraction mode has time τ∥/τ0 ~ N^{1/2}, so that τ⊥ τ∥ ~ τ0². Using flat-plate spread-area scaling to close the model gives τ⊥ ∝ (W/R0) We^{-1/4} and τ∥ ∝ (We^{1/4})/(W/R0). Simulations with a fully non-wetting boundary collapse con","pith_inferences":["A general design rule implied by the paper: any texture that imposes a fixed length along one axis (ridges, fibers, defects) introduces a conjugate slow mode, so minimizing contact time requires suppressing or outrunning that mode, not just shrinking blob mass.","A direct experimental extension would measure τ⊥ and τ∥ simultaneously at fixed groove width across a wide Weber range; the product law τ⊥ τ∥ ≈ τ0² is a sharp, quantitative prediction that current data only test indirectly.","The super-linear fitted width exponent (p≈1.51 for non-wetting grooves) suggests that Smax itself carries a weak groove-width dependence; replacing the flat-plate area closure with a fitted W-dependent closure could yield an exact imposed-parameter scaling.","Sweeping contact angle on wetting grooves should move the transition parameter χ* monotonically; the paper predicts the direction of that shift, so a contact-angle series would sharpen the mode-selection picture beyond the single wetting condition studied."],"forward_implications":["Contact time on grooved superhydrophobic surfaces becomes a designable function of Weber number and groove width rather than the fixed 2.6 τ0 flat-plate value.","The two modes form a reciprocal pair, so shortening the transverse recoil by stronger confinement lengthens axial recoil by the same factor; their product is invariant.","On mildly wetting grooves, the framework predicts a mode-selection transition at a Weber number proportional to (W/R0)², meaning a surface can be operated on either the slow axial branch or the fast transverse branch by changing impact energy.","The measured-N collapse of the contact-time data means the blob decomposition itself is the robust core; predictions tied to exact spread morphology are directionally correct but quantitatively approximate.","A single non-dimensional parameter χ organizes fast/slow branch selection on wetting grooves, offering a compact design variable for textured repellent surfaces."],"fun_headline_variants":["Grooves split droplet rebound into two linked time scales","Anisotropic confinement creates conjugate rebound modes","Droplet rebound on grooves: two times instead of one","Contact time on grooves depends on impact energy","Grooved surfaces force droplet to choose between two recoil modes"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that a grooved surface reshapes the spreading droplet without changing the maximum surface area set by flat-plate inertio-capillary balance, Smax/R0² ~ We^{1/2}; the paper itself flags this as a strong assumption, and its simulations show Smax and ℓ_max carry extra groove-width dependence, so the imposed-parameter scalings (though not the measured-N scaling) rest on this closure.","fun_headline_variants_meta":{"raw":{"variants":["Grooves split droplet rebound into two linked time scales","Anisotropic confinement creates conjugate rebound modes","Droplet rebound on grooves: two times instead of one","Contact time on grooves depends on impact energy","Grooved surfaces force droplet to choose between two recoil modes"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000168,"raw_usage":{"total_tokens":1147,"prompt_tokens":846,"completion_tokens":301,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":590,"completion_tokens_details":{"reasoning_tokens":225}},"tokens_in":590,"tokens_out":301,"duration_ms":3440,"temperature":1.0,"reasoning_tokens":225,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T21:46:01.017286+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure, at fixed Ohnesorge number and a single non-wetting groove width, both contact time τ_c and the axial recoil half-period across Weber numbers from about 3 to 30. If τ_c does not fall as We^{-1/4} once the droplet is well confined (N>1.2), or if the product of the transverse and axial times drifts away from τ0² by more than experimental scatter, the reciprocal-mode claim fails. A second check on a wetting groove: the predicted collapse of the two branches at χ≈1.61 and the scaling We*∝(W/R0)² are directly testable, and their absence would falsify the mode-selection mechanism.","supporting_citations":[],"review_version":1}