{"id":"5586b888-847f-469d-b57e-5daace26e4c0","arxiv_id":"1908.06380","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Three adhesion theories give stickiness criteria with the same main ingredients: rms roughness, longest roughness wavelength, and work of adhesion divided by elastic modulus.","lead":"The paper derives two simple formulas for when soft materials stick to rough surfaces and shows they nearly match an existing criterion. If right, stickiness depends on large-scale roughness and material stiffness, not fine surface details.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"BAM stickiness criterion (Eq. 13) appears to invert the adhesion-length ratio; correcting it shifts the BAM threshold by roughly a factor of 3 and weakens the claimed BAM/PT agreement.","rationale":"The Reader's verdict is CONDITIONAL, with the weakest assumption identified as the arbitrary BAM threshold and reliance on a single H=0.8 power-law PSD. My independent reading found a more specific and more damaging technical issue: the published Eq. (13) appears to invert the normalized adhesion-length ratio in the repulsive term of Eq. (7). If this is confirmed, the fitted BAM criterion (16) is not a valid consequence of the BAM model as stated, and the quantitative closeness between BAM and Persson-Tosatti in Fig. 4 is substantially weakened: a factor of about 3 in the threshold hrms, or about an order of magnitude in the prefactor β. This does not by itself destroy the broader qualitative claim that q0, hrms, and la = Δγ/E* control stickiness — the PT and Violano criteria still have that form — but it removes one of the paper's two newly derived 'universal' criteria from the quantitative comparison. The concern is concrete and falsifiable by re-derivation and numerical recomputation, so I do not think the paper should be rejected outright; it needs a corrected BAM derivation and updated fit. Because the Reader's verdict is already CONDITIONAL, my finding reinforces that verdict rather than changing it. I disagree only partially with the Reader's weakest_assumption: the arbitrary threshold matters, but the apparent algebra error in Eq. (13) is a more immediate load-bearing risk and should be checked first.","tokens_in":9664,"tokens_out":15987,"duration_ms":160054,"concrete_test":"Independently re-derive Eq. (13) from Eq. (7), substituting E*/sigma0 = eps/la and q0 = 2π/λL, and check whether the first coefficient should be 1/(la/eps) rather than la/eps. Then, using the corrected form, numerically solve for (hrms/eps)_thresh at the 1e-8 threshold for the specific case la/eps = 0.05, λL/eps = 10^7. If the solution is approximately 180 instead of approximately 550, repeat the fits in Figs. 2-3 with the corrected equation and recompute β_BAM and the comparison in Fig. 4. A second useful check: derive the asymptotic zero-pull-off balance for large λL/eps directly from Eq. (7); it gives hrms/eps ≈ sqrt(C la λL/(2π eps^2)) with C ≈ 0.28–0.40, i.e. β ≈ 0.045–0.064, not 0.6.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the BAM, Persson-Tosatti, and Violano criteria are quantitatively close depends directly on the new BAM threshold, Eq. (16), derived from the fitted Eq. (14)-(15). That fit is anchored to Eq. (13), which appears to contain an algebraic error. In Eq. (7), the repulsive term is q0 hrms (E*/sigma0) exp(...). Since E*/sigma0 = eps/la, this term in normalized variables is 2π (hrms/eps)/[(λL/eps)(la/eps)]. Equation (13), however, prints 2π (la/eps)(hrms/eps)/(λL/eps), multiplying by la/eps rather than dividing by it. For a representative Lennard-Jones value la/eps = 0.05, this changes the repulsive coefficient by a factor of (la/eps)^{-2} = 400. Re-solving the zero-threshold balance with the correct coefficient for la/eps = 0.05 and λL/eps = 10^7 gives a stickiness boundary near hrms/eps ≈ 180, not the value ≈ 550 implied by Eq. (16). Equivalently, the fitted β_BAM = 0.6 appears to be an artifact of the misprinted equation; a correct fit would yield β_BAM ≈ 0.05–0.06. This is not merely a question of the arbitrary 10^-8 threshold: the displacement of the pull-off zero is set by Eq. (7). If the printed Eq. (13) is wrong, the BAM criterion cannot support the paper's claim of surprising quantitative universality, since BAM would then sit a factor of 2–3 below PT in the hrms threshold. The same correction may also affect the claimed disappearance of eps from the final BAM criterion.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes two new \"stickiness\" criteria for soft adhesion of rough surfaces, one derived from the Persson-Tosatti energy balance and one from the bearing-area model (BAM), and compares them with the earlier DMT-based criterion of Violano et al. For a power-law roughness spectrum with Hurst exponent H = 0.8, the Persson-Tosatti criterion reduces to h_rms < sqrt(0.24 l_a lambda_L), the BAM criterion is fitted to numerical solutions and written as h_rms < sqrt(0.6 l_a lambda_L), and Violano's criterion is h_rms < epsilon^{-1/5} (0.358 l_a lambda_L)^{3/5}. The paper argues that the closeness of these three criteria is surprising and that stickiness is controlled by macroscopic quantities (the long-wavelength cutoff, the rms height, and the ratio of work of adhesion to plane-strain modulus) rather than by atomic-scale details. The conclusion is applied to the recent Dalvi et al. adhesion experiments, for which all surfaces are predicted to be sticky.","tokens_in":10044,"tokens_out":17378,"duration_ms":146974,"significance":"If correct, the paper would provide a practical and simple design rule for soft adhesion and would strengthen the Persson-Tosatti energy-balance picture against magnification-dependent criteria such as Pastewka-Robbins and Muser. The PT derivation in Section 2.1 is transparent and easy to reproduce, and the comparison with the Dalvi et al. experiments is a useful quantitative check. The BAM criterion, however, is essential to the claimed universality and is not derived in closed form: it is obtained by solving a numerical equation at an arbitrary threshold and then fitting power laws. Its validity therefore depends entirely on the correctness of Eq. (13). Because the central conclusion of quantitative agreement among the three criteria rests on this BAM criterion, the significance of the paper hinges on whether that equation is correct and on how sensitive the fitted exponent and prefactor are to the threshold choice.","major_comments":[{"comment":"The normalized repulsive term in Eq. (13) appears to be algebraically inverted. From Eq. (7), the repulsive term is q0 h_rms (E*/sigma0) exp(-u/(gamma h_rms)). With q0 = 2 pi / lambda_L and E*/sigma0 = epsilon / l_a, the term in the reduced variables used in Eq. (13) is 2 pi (h_rms/epsilon) / [(lambda_L/epsilon)(l_a/epsilon)], so it must be divided by l_a/epsilon. As printed, Eq. (13) places l_a/epsilon in the numerator, making the repulsive coefficient too small by a factor (l_a/epsilon)^{-2}, which is 400 for the representative Lennard-Jones value l_a/epsilon = 0.05. Since the power-law fits (14)-(15) and the final criterion (16) are extracted from Eq. (13), the BAM branch in Fig. 4 and the central claim of quantitative agreement between BAM and Persson-Tosatti in Section 5 are directly affected; for example, re-solving the zero-threshold balance for l_a/epsilon = 0.05 and lambda_L/epsilon = 10^7 gives a stickiness threshold near h_rms/epsilon ~ 180, not the value near 550 implied by Eq. (16).","section":"Section 2.2, Eq. (13)"},{"comment":"The BAM stickiness criterion is not derived in closed form but is obtained by solving Eq. (13) at the arbitrary threshold -sigma_min/sigma0 = 10^-8 and then fitting the results to power laws (14)-(15). The value beta_BAM = 0.6 in Eq. (16) therefore encodes both the threshold choice and the fitting range. A sensitivity analysis, for example using thresholds 10^-6 or 10^-10 and different ranges of lambda_L/lambda_L0, is needed before the close agreement with the Persson-Tosatti criterion can be presented as a universal feature rather than as a consequence of the chosen numerical definition of stickiness.","section":"Section 2.2, Eqs. (13)-(16)"}],"minor_comments":[{"comment":"The caption writes lambda_L0 = q0/2 pi = 2048 epsilon, which is dimensionally inconsistent; since q0 = 2 pi / lambda_L, the intended expression is lambda_L0 = 2 pi / q0 = 2048 epsilon.","section":"Fig. 1 caption"},{"comment":"The surname of the experimental group is spelled \"Dalvi\" in the abstract and references but appears as \"Davli\" in several places in Sections 3 and 4; please correct.","section":"Throughout"},{"comment":"The comparison of the three criteria is presented only for H = 0.8 (D = 2.2); the Persson-Tosatti prefactor depends on H through (2H-1)/(pi H), and the Violano criterion is stated for D ~ 2.2. The abstract and conclusions should avoid implying universality over all Hurst exponents until other values of H are examined.","section":"Section 3 and Fig. 4"},{"comment":"The symbol delta_gamma_rrs should presumably be delta_gamma_rss to match Eq. (29); please check the notation.","section":"Section 4.2, Eq. (30)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript's main new result is the BAM stickiness criterion, and the central claim of universality depends on it. Given the apparent algebraic error in Eq. (13), I recommend that the editor ask the author to provide the derivation of Eq. (13) explicitly and to recompute the fits and the comparison in Fig. 4. If the error is confirmed, the BAM criterion will sit well below the Persson-Tosatti curve and the claimed surprising quantitative agreement will not hold."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper's headline result does not survive a check of its central equation. Eq. (13) writes the repulsive term of the BAM force balance as 2π (la/ε)(hrms/ε)/(λL/ε), but from Eq. (7) with E*/σ0 = ε/la the correct factor is 2π (hrms/ε)/[(λL/ε)(la/ε)]. The printed version multiplies by la/ε where it should divide. For la/ε = 0.05 this is a factor 400 on the repulsive coefficient. Re-solving the threshold with the correct expression for λL/ε = 10^7 gives hrms/ε ≃ 180, not the ≃ 550 implied by Eq. (16), and the fitted power law yields β_BAM ≈ 0.06 instead of 0.6. So BAM does not sit 'extremely close' to Persson-Tosatti; it sits a factor of about two lower in the admissible roughness amplitude. The universality claim is therefore substantially weakened.\n\nWhat is genuinely useful: the Persson-Tosatti criterion (12) is a clean, explicit threshold and the three-way framing (PT, BAM, Violano) plus the contrast with Pastewka-Robbins/Müser is a helpful synthesis of where the criteria converge and diverge. The qualitative message—stickiness controlled by λL, hrms, and la rather than atomic-scale details—survives the correction, since the corrected BAM still has the same scaling form hrms < sqrt(β la λL). But the 'surprising quantitative closeness' is an artifact of the equation error.\n\nOther soft spots: the BAM threshold is defined at an arbitrary −σmin/σ0 = 10^-8 and extracted by fitting, which would already limit the claim; the comparison uses one PSD shape and H=0.8 only; the Violano et al. reference is missing from the list and several citations carry wrong years (Ciavarella et al. 2018 is 2019; Müser et al. 2018 is 2017). These are minor relative to the Eq. (13) issue but should be fixed.\n\nBottom line: the paper deserves a serious referee, not a desk reject, because the PT criterion and the comparison framework are worth having in the literature and the BAM error is fixable. But in its current form the central quantitative claim should not be accepted. I would not cite the paper until the BAM section is corrected.","headline":"The BAM stickiness criterion rests on an algebraic error (Eq. 13 multiplies by la/ε instead of dividing), and once corrected the claimed near-universality with Persson-Tosatti largely disappears.","tokens_in":10608,"tokens_out":10068,"would_cite":false,"duration_ms":86199,"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 shows that three stickiness criteria for rough soft contacts collapse to the same macroscopic law, so stickiness can be predicted from large-scale roughness and material properties alone.","keywords":["adhesion","stickiness","rough surfaces","power spectral density","bearing area model","JKR model","DMT model","soft matter"],"falsifier":"Take a soft material and two surfaces with the same longest wavelength $\\lambda_L$ and the same rms height $h_{\\mathrm{rms}}$ but different short-wavelength content, e.g. one filtered below a few nanometres and one retaining atomic-scale roughness. The paper's three criteria predict identical stickiness. If pull-off measurements, or simulations with band ratio up to $10^7$, show stickiness changing when only the atomic-scale tail is altered, the universal claim is false.","tokens_in":9433,"feed_emoji":"📏","tokens_out":15292,"duration_ms":127560,"temperature":0.7,"pith_summary":"Soft surfaces stick to rough counterfaces only when the roughness is mild enough. This paper shows that three stickiness criteria derived from different starting points—an energy-balance argument, a bearing-area geometric model, and a DMT-type adhesion theory—give nearly the same threshold when the roughness spectrum is a power law. In all three, stickiness is controlled by the longest wavelength of roughness, the rms amplitude of roughness, and the ratio of the work of adhesion to the plane-strain modulus; atomic-scale details drop out. If this is right, predicting stickiness requires only macroscopic surface and material properties, and the elastic modulus is the main practical control, consistent with the old empirical rule that tacky materials need a low modulus.","feed_headline":"Stickiness law unifies three rough-surface adhesion criteria","feed_subtitle":"Atomic-scale roughness drops out; the longest wavelength, rms height, and adhesion energy set the threshold.","key_machinery":"The machinery is a pure power-law power-spectral-density model $C(q)=Z q^{-2(1+H)}$ with Hurst exponent $H=0.8$ (fractal dimension $D=2.2$), cut off at the low wavevector $q_0=2\\pi/\\lambda_L$. Because $H>0.5$, the elastic energy per unit area needed to flatten the surface converges to $E^*\\pi h_{\\mathrm{rms}}^2/\\lambda_L \\cdot H/(2H-1)$ as the short-wavelength cutoff is removed. Setting the effective work of adhesion $\\Delta\\gamma - E^*\\pi h_{\\mathrm{rms}}^2/\\lambda_L \\cdot H/(2H-1)$ to zero yields the energy-balance criterion. The bearing-area criterion instead computes the adhesive area from a constant-traction adhesive law (constant adhesive traction up to range $\\varepsilon$) and defines stickiness as the pull-off traction falling to $10^{-8}$ of the theoretical strength; after power-law fits the dependence on $\\varepsilon$ cancels and the same square-root form appears. All three are then compared on the axis $(\\Delta\\gamma/(E^*\\varepsilon))(\\lambda_L/\\varepsilon)$ versus $h_{\\mathrm{rms}}/\\varepsilon$, where $\\Delta\\gamma/E^*$ sets the adhesion length.","core_discovery":"Stickiness—whether a soft elastic body pressed against a hard rough surface holds on at pull-off—is claimed to be governed by three macroscopic quantities: the low-wavevector cutoff $q_0=2\\pi/\\lambda_L$, the rms roughness amplitude $h_{\\mathrm{rms}}$, and the ratio $\\Delta\\gamma/E^*$ of the work of adhesion to the plane-strain elastic modulus. Two new criteria are derived: an energy-balance criterion that sets the effective surface energy to zero gives $h_{\\mathrm{rms}} < \\sqrt{0.24\\,(\\Delta\\gamma/E^*)\\,\\lambda_L}$, and a bearing-area-model criterion with a constant-traction adhesive law and a numerical pull-off threshold gives $h_{\\mathrm{rms}} < \\sqrt{0.6\\,(\\Delta\\gamma/E^*)\\,\\lambda_L}$. A third, DMT-based criterion gives $h_{\\mathrm{rms}} < \\varepsilon^{-1/5}\\,[0.36\\,(\\Delta\\gamma/E^*)\\,\\lambda_L]^{3/5}$, differing only by a weak dependence on the range of attraction. For power-law roughness with $H=0.8$, the three curves nearly coincide over many decades of wavelength. The paper concludes that small-scale roughness features such as local slopes and curvatures do not affect stickiness, while a numerical-interpolation criterion that does depend on them is regarded as an artifact of its limited roughness band.","pith_inferences":["The closeness of the prefactors 0.24 and 0.6 hints at an asymptotic identity between the energy-balance and bearing-area routes that the paper does not derive; finding it would remove the remaining factor-of-two uncertainty.","The universality is established only for $H=0.8$ power-law surfaces. A testable extension is whether it survives for $H\\le0.5$, where the elastic flattening energy diverges and the same derivation breaks down.","Because the criteria depend only on $h_{\\mathrm{rms}}$ and $\\lambda_L$, experimental data across materials and roughnesses could be collapsed onto a single master curve of $h_{\\mathrm{rms}}^2/((\\Delta\\gamma/E^*)\\lambda_L)$ against stickiness; the paper does not perform this test.","The comparison with the cited sphere experiments is indirect because the criteria are derived for nominally flat contacts; a flat-punch adhesion experiment with controlled power-law roughness would be a cleaner check."],"forward_implications":["Predicting stickiness no longer requires resolving roughness at or near atomic scale: the threshold is fixed by the longest wavelength $\\lambda_L$, the rms height $h_{\\mathrm{rms}}$, and the adhesion length $\\Delta\\gamma/E^*$.","The elastic modulus becomes the dominant practical control: lowering $E^*$ raises the stickiness threshold, in quantitative agreement with the empirical rule that soft materials around 1 MPa are tacky.","For surfaces whose roughness spans seven or more decades of wavelength, the three convergent criteria should be preferred over the numerical-interpolation criterion, which predicts a residual short-wavelength dependence.","The adhesion experiments on soft polymer hemispheres against rough diamond cited by the paper are consistent with the criteria: the estimated roughness parameter lies below the predicted stickiness threshold, so the surfaces are expected to stick.","Surface engineering for stickiness can focus on long-wavelength topography, increasing $\\lambda_L$ by polishing, rather than on removing nanoscale roughness."],"supporting_citations":[{"why":"Supplies the broad-band power-law roughness measurements and adhesion experiments that motivate the comparison and anchor the claim that the energy-balance idea works.","marker":"Dalvi et al. (2019)"},{"why":"Gives the energy-balance idea and the full-contact elastic-energy integral from which the first new stickiness criterion is derived.","marker":"Persson (2002)"},{"why":"Provides the effective surface energy formula and the roughness-induced area increase that brings the energy criterion closer to the bearing-area one.","marker":"Persson & Tosatti (2002)"},{"why":"Introduces the bearing-area model from which the second new criterion is derived.","marker":"Ciavarella (2017)"},{"why":"Supplies the third stickiness criterion, used as the DMT-based comparison point.","marker":"Violano et al. (2018)"},{"why":"Provides the DMT adhesion theory underlying the third criterion.","marker":"Persson and Scaraggi (2014)"},{"why":"Gives the numerical-interpolation stickiness criterion that the paper contrasts, supporting the claim that its magnification dependence is a finite-band artifact.","marker":"Pastewka and Robbins (2014)"},{"why":"Supplies the adhesiveless repulsive pressure relation used inside the bearing-area equation.","marker":"Persson (2007)"}],"fun_headline_variants":["Rough-surface stickiness boils down to three big quantities","Adhesion stickiness: three macroscopic criteria match","Small-scale roughness doesn't decide stickiness","Universal stickiness law: only macroscale matters"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The agreement rests on representing real surfaces as a pure power-law roughness spectrum with Hurst exponent $H=0.8$ and on defining stickiness as the pull-off traction falling below $10^{-8}$ of the theoretical strength; if real surfaces deviate from that spectrum or the threshold is set differently, the quantitative coincidence among the criteria may not persist.","fun_headline_variants_meta":{"raw":{"variants":["Rough-surface stickiness boils down to three big quantities","Adhesion stickiness: three macroscopic criteria match","Small-scale roughness doesn't decide stickiness","Universal stickiness law: only macroscale matters"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000757,"raw_usage":{"total_tokens":3359,"prompt_tokens":937,"completion_tokens":2422,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":553,"completion_tokens_details":{"reasoning_tokens":2361}},"tokens_in":553,"tokens_out":2422,"duration_ms":17611,"temperature":1.0,"reasoning_tokens":2361,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:46:39.531329+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a soft material and two surfaces with the same longest wavelength $\\lambda_L$ and the same rms height $h_{\\mathrm{rms}}$ but different short-wavelength content, e.g. one filtered below a few nanometres and one retaining atomic-scale roughness. The paper's three criteria predict identical stickiness. If pull-off measurements, or simulations with band ratio up to $10^7$, show stickiness changing when only the atomic-scale tail is altered, the universal claim is false.","supporting_citations":[],"review_version":1}