{"id":"b900575a-32b2-4cfb-8326-0a201248b8e3","arxiv_id":"1908.02522","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Using a Scheutjens-Fleer self-consistent field model with non-local interactions, the authors find a positive mean bending rigidity for symmetric liquid-liquid interfaces and a negative Gaussian bending rigidity at strong segregation.","lead":"This paper computes the bending stiffness of the interface between two liquids using a lattice-based polymer theory and finds that it takes a positive sign. It also predicts a sign change in a related elastic constant as the liquids become more immiscible.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The sign results are not shown to be independent of the chosen dividing-surface gauge; the equal-density pinning (Eq. 2) together with the non-local interaction (Eq. 4) produces positive κ, but no gauge-invariant justification is given.","rationale":"The reader's CONDITIONAL verdict is appropriate. The paper's numerical work is internally consistent: the SI's reproduction of Matsen's results when non-local interactions are removed is strong evidence that the SCF implementation is correct and that the difference indeed comes from the non-local term. My concern is not numerical but conceptual: the Helfrich moduli are not observables independent of the dividing surface. The authors explicitly acknowledge this when discussing Blokhuis's results (different interface definitions give different κ), yet they do not demonstrate that their equal-density pinning is the correct or the physically relevant one. Since the sign is the headline result, and since the sign is known to be sensitive to both the non-local term and the gauge, the claim needs a robustness check across gauges. If the sign survives, CONDITIONAL becomes ACCEPT; if not, the headline should be revised. Therefore I do not change the reader's verdict.","tokens_in":13809,"tokens_out":13710,"duration_ms":156644,"concrete_test":"Using the same converged SCF fields for representative state points (N=20, Δχ=0.02 and 0.3; N=200, Δχ=0.01 and 0.1), recompute γ_c and γ_s but with the interface position r0 chosen by three different prescriptions: (a) equal density φ_A=φ_B (current), (b) Gibbs equimolar surface defined by the zero of ∫_0^{r0} [φ_A(r)-φ_B(r)] L(r) dr, and (c) surface of tension defined by the mechanical Laplace condition. Extract κ and κ̄ in each gauge. If the sign of κ or the sign-switch threshold of κ̄ changes between (a)-(c), the central claim is gauge-dependent and the conclusions as stated are not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Central claim: κ>0 and κ̄ sign-switch for symmetric L/L interfaces. These are extracted from γ_c(R)-γ_p and γ_s(R)-γ_p using the Helfrich expansion (Eq. 1), with the interface pinned at the coordinate where φ_A=φ_B via the Lagrange term Δ_{r;r0} in Eq. 2. This pinning is a gauge choice: the Helfrich moduli are defined only relative to a chosen dividing surface, and the literature cited by the authors themselves (Refs. [3-6], [34]) shows that κ can change sign with the choice of surface (surface of tension vs. equal-density). The SI shows that replacing Eq. 4 with the local approximation ⟨φ_B⟩=φ_B reproduces Matsen's negative κ and positive κ̄; hence the positive-κ result is jointly caused by the non-local Cahn-Hilliard term and the equal-density pinning. No test is presented showing that the sign survives under a different legitimate pinning (e.g., Gibbs equimolar or surface of tension). Until such a check is done, the claim 'κ is strictly positive' is not a robust conclusion of the model; it is contingent on a specific, unvalidated gauge.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript presents Scheutjens–Fleer self-consistent field (SF-SCF) calculations of the mean and Gaussian bending rigidities of symmetric AN–BN liquid–liquid interfaces in the grand canonical ensemble. The interface is pinned at the equimolar surface through the Lagrange term Δ_{r;r0} in Eq. (2), and the non-local Cahn–Hilliard contact interaction is retained via Eq. (4). Bending moduli are extracted from the grand-potential differences between cylindrical/spherical and planar interfaces using the Helfrich expansion, Eq. (1). The central claims are: κ>0 for all interactions studied; κ̄>0 near the critical point and negative at strong segregation, with a sign-switch boundary described by (Δχ)^{1/2}=0.2+0.5/√N; the scaling of γ, κ, and κ̄ is ∝N(Δχ)^{3/2} in weak segregation and ∝(Δχ)^{1/2} in strong segregation; and the cross-over length λ=√(κ/γ) is of order the segment size, with a maximum that grows linearly with N. The SI shows that setting ⟨φ_B⟩=φ_B in Eq. (4) reproduces the earlier negative-κ results of Matsen, and hence that the sign difference is caused by the non-local interaction term.","tokens_in":14007,"tokens_out":6405,"duration_ms":68992,"significance":"If the results are robust, they would resolve a long-standing controversy about the sign of κ for symmetric polymer interfaces, provide a quantitative prediction for the sign change of κ̄, and yield falsifiable scaling laws and a sign-switch diagram. The paper has notable strengths: the SCF solutions are internally consistent; the careful benchmark in the SI against Matsen’s classical SCF demonstrates that the numerical machinery is accurate and isolates the physical ingredient (non-local interactions) responsible for the difference; and κ and κ̄ are computed directly from the grand potential rather than fitted to a target. The main caveat is that the central sign result is tied to a specific dividing-surface pinning and to the treatment of non-local interactions; the manuscript does not yet show that the sign is independent of those choices.","major_comments":[{"comment":"The extraction of κ and κ̄ uses the Helfrich expansion for an interface pinned by the Lagrange term Δ_{r;r0} to the coordinate where φ_A=φ_B. This pinning is a gauge choice: the Helfrich moduli are not invariant under a change of dividing surface. The authors themselves note (paragraph around Refs. [34]–[36]) that Blokhuis found κ to depend on the interface-position convention and to be negative at the surface of tension. The SI benchmark shows that the same pinning without Eq. (4) yields Matsen’s negative κ, but no calculation is provided for another legitimate pinning (surface of tension, Gibbs equimolar surface, or a local external field) within the full SF-SCF model. Unless such a test is supplied, the statement “κ is strictly positive for L/L interfaces” remains contingent on a specific, unvalidated gauge.","section":"Eq. (2) and the interface-pinning procedure"},{"comment":"The entire sign of κ and the sign switch of κ̄ rest on the gradient term (1/6)∇²φ_B in Eq. (4). The SI demonstrates that the local approximation ⟨φ_B⟩→φ_B exactly reproduces Matsen’s results, including negative κ and positive κ̄. That is an informative contrast, but it also means the positive-κ conclusion is not robust across two closely related SCF treatments. The manuscript gives no independent check that Eq. (4) is the correct non-local interaction for bending moduli—for example, a comparison with a different lattice discretization, an off-lattice calculation, or a continuum-limit check. Because this term is the load-bearing difference from all previous negative-κ results, this missing validation is a central concern.","section":"Eq. (4) and the SI final section"},{"comment":"The scaling exponents in Figs. 1(a) and 1(b), the sign-switch boundary in Fig. 3(b), and the linear fit λ_max=0.02N+6.54 in Fig. 2(b) are presented without error bars or convergence checks (e.g., with respect to the lattice discretization, the system size, or the radius r0). The paper states that SCF solutions are accurate to nine significant digits, but that does not quantify the error in the fitted slopes or in the location of the sign switch. Since the scaling laws and the sign switch are central quantitative claims, the authors should provide standard errors for the extracted exponents and fit parameters, at least for representative cases.","section":"Figs. 1 and 2 and Fig. 3(b)"}],"minor_comments":[{"comment":"The supplementary information title contains a typo: “Elastic Poperties” should read “Elastic Properties.”","section":"SI title"},{"comment":"The sentence beginning “small maximum and the height of this maximum is a weakly linear function of the chain length N” appears to be an incomplete fragment; it should be completed or merged with the previous sentence.","section":"SI, section on finite chain length effects"},{"comment":"Reference [27] is incompletely specified (“which includes Refs. [4, 13, 20 & 28]”); the full citation information should be provided.","section":"References"},{"comment":"The wording “We have proved that the fluctuations from L/L interface away from the planar interface indeed cost free energy” overstates the evidential status of a numerical mean-field calculation; “shown” or “demonstrated” would be more appropriate.","section":"Conclusion"}],"recommendation":"major_revision","confidential_remarks":"The paper’s internal consistency and the Matsen benchmark are strong; the main concern is the breadth of the advertised claim. In my view the authors should either (a) add the alternative-pinning and non-local-discretization tests and error bars, or (b) explicitly reframe the central claim as conditional on the equal-density pinning and the Cahn–Hilliard non-local term. As written, the abstract’s unqualified “κ is strictly positive” is broader than the evidence. The paper is within scope for a statistical-mechanics journal; a major revision is appropriate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear Colleague,\n\nThe paper to know for this week's discussion: it is the first molecular SCF calculation I have seen that returns a positive mean bending rigidity kappa for a symmetric liquid-liquid interface, plus a sign switch of the Gaussian bending rigidity kappa_bar. The calculation is clean, and the benchmark is convincing: when the non-local Cahn-Hilliard term is removed from their Scheutjens-Fleer model, they reproduce Matsen's negative kappa and positive kappa_bar almost exactly. That internal check gives me confidence the result is not a numerical artifact.\n\nWhat I find genuinely useful: the ratio kappa_bar/kappa is 1/2 in weak segregation and -3/2 in strong segregation, with a diagram of states separating the signs. The prediction that the Gaussian rigidity changes sign when the interfacial width is three to four segment lengths is concrete and falsifiable. The scaling exponents (3/2 near critical, 1/2 at strong segregation) are internally consistent, though read off slopes rather than derived. The citation pattern is honest: Matsen, Blokhuis, and the authors' own microemulsion work are discussed in context.\n\nThe soft spot is the interface-pinning gauge. The Helfrich moduli are dividing-surface dependent, and the positive kappa comes out when the interface is pinned at the equal-density point via the Lagrange term in Eq. (2). That is a natural convention for a symmetric interface, but it is still a convention. The authors know Blokhuis's fixed-mu calculation gives negative kappa on the surface of tension; they argue their choice is the correct one, but they never test an alternative pinning inside their own model. Without that test, the statement 'kappa is strictly positive' is a conditional result of this particular model plus this particular convention, not a theorem. A serious revision should either show the sign survives under a different legitimate division or explain why the equal-density surface is the physically controlling one for fluctuations.\n\nMinor: no convergence checks with respect to lattice spacing, no released code, and the 'speculation' that interfacial width controls the sign switch is honestly flagged as speculation. These are not deal-breakers for a letter.\n\nMy recommendation: send it to peer review. The novel result and the reproducible benchmark justify referee time. I would ask for the gauge test and a more nuanced abstract before acceptance.\n\nBest.","headline":"A clean SCF calculation that for the first time finds positive mean bending rigidity from a molecular model, but the sign is tied to a specific interface-pinning convention that the authors do not test.","tokens_in":14575,"tokens_out":6149,"would_cite":false,"duration_ms":62018,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["68.05.n","68.35.Md","05.70.Np","31.15.Ne"],"model":"deepseek-v4-flash","headline":"The paper claims that for symmetric liquid-liquid interfaces the mean bending rigidity is positive everywhere, the Gaussian rigidity turns negative at strong segregation, and these signs control interface fluctuations and droplet formation.","keywords":["liquid-liquid interface","bending rigidity","Gaussian bending rigidity","self-consistent field theory","grand canonical ensemble","interfacial tension scaling","polymer interface","non-local interactions"],"falsifier":"Compute the same grand potentials with a different discretization of the Laplacian, for instance by including next-nearest-neighbor lattice sites in the gradient stencil, and check whether $\\kappa$ stays positive and $\\bar{\\kappa}$ still changes sign. Alternatively, simulate a symmetric polymer blend at fixed chemical potential and extract the capillary-wave spectrum: the predicted positive $\\kappa$ and $\\lambda$ of order a few segment sizes would be confirmed or refuted by the wavelength dependence of the height fluctuations.","tokens_in":13556,"feed_emoji":"💧","tokens_out":10212,"duration_ms":91016,"temperature":0.7,"pith_summary":"The paper claims to settle a long-disputed sign question: for a symmetric interface between two immiscible liquid-like phases of equal chain length, does bending the interface cost or release free energy? Using a lattice mean-field (self-consistent field) model in the grand canonical ensemble, it reports that the mean bending rigidity $\\kappa$ is positive for all interaction strengths, so a flat interface is a local free-energy minimum. It further finds that the Gaussian bending rigidity $\\bar{\\kappa}$ is positive near the critical point and becomes negative at strong segregation, when the interfacial width is only about three or four segment sizes. If these results hold, they imply that short-wavelength height fluctuations are damped, that strongly segregated interfaces resist saddle-shaped deformations, and that the crossover length $\\lambda = \\sqrt{\\kappa/\\gamma}$ is a molecular-scale quantity accessible to experiments and simulations.","feed_headline":"Bending a liquid-liquid interface always costs free energy","feed_subtitle":"Mean-field model with non-local interactions also finds a sign switch in the Gaussian rigidity at strong segregation.","key_machinery":"The load-bearing device is the non-local average $\\langle \\phi_B(r)\\rangle = \\phi_B(r) + \\frac{1}{6}\\nabla^2 \\phi_B(r)$ appearing in the polymer-solution interaction free energy, together with a Lagrange multiplier that pins the interface at the equimolar surface. The grand potentials of planar, cylindrical, and spherical interfaces are computed at fixed chemical potentials, and the standard curvature expansion of the tension yields $\\kappa$ from the cylinder and $2\\kappa + \\bar{\\kappa}$ from the sphere. The non-local term is decisive: when it is replaced by the local approximation $\\langle\\phi\\rangle \\to \\phi$, the same numerical machinery reproduces earlier self-consistent-field results with negative $\\kappa$ and positive $\\bar{\\kappa}$.","core_discovery":"The central discovery is that in a mean-field treatment that keeps non-local corrections to the contact interaction, the elastic moduli of symmetric $A_N$-$B_N$ interfaces have definite signs: $\\kappa > 0$ for all $\\chi$ above the critical value, while $\\bar{\\kappa} > 0$ near the critical point and $\\bar{\\kappa} < 0$ at strong segregation. The ratio $\\bar{\\kappa}/\\kappa$ takes plateau values of $1/2$ in weak segregation and $-3/2$ in strong segregation, and both rigidities scale with $\\Delta\\chi = \\chi - \\chi_c$ exactly as the interfacial tension does, as $N(\\Delta\\chi)^{3/2}$ near criticality and $(\\Delta\\chi)^{1/2}$ far from it. The sign switch of $\\bar{\\kappa}$ occurs when the interfacial width shrinks to a few segment lengths, and the paper maps this switch in the $(\\sqrt{\\Delta\\chi}, 1/\\sqrt{N})$ plane.","pith_inferences":["A testable extension of the paper's logic: measuring $\\lambda$ from the capillary-wave spectrum of a symmetric polymer blend should show a non-monotonic dependence on $\\chi$, peaking at the weak-to-strong segregation crossover.","The identification of interfacial width as the control parameter suggests the sign switch is not unique to symmetric blends; any interface whose width can be tuned through a few segment sizes may show the same $\\bar{\\kappa}$ sign switch.","If the non-local gradient term is as decisive as claimed, polymer self-consistent-field codes that drop it may still give accurate interfacial tensions but should not be trusted for elastic coefficients, at least not without including such a term."],"forward_implications":["Short-wavelength height fluctuations of a symmetric liquid-liquid interface are damped: every deviation from the planar state raises the grand potential, so the interface is an elastic sheet down to the crossover length $\\lambda = \\sqrt{\\kappa/\\gamma}$.","The Gaussian rigidity sign switch means strongly segregated interfaces resist saddle deformations, making pinch-off and droplet formation more difficult, while near-critical interfaces promote such deformations.","Simulations that extract $\\kappa$ from capillary-wave spectra should find their best signal near the maximum of $\\lambda$, which grows linearly with chain length as $\\lambda_{\\max} \\simeq 0.02N + 6.54$ in segment units.","Near the critical point the elastic constants inherit the tension's scaling, so the universal ratios $\\bar{\\kappa}/\\kappa = 1/2$ and $-3/2$ in the weak- and strong-segregation regimes are testable predictions."],"supporting_citations":[{"why":"defines the crossover length $\\lambda=\\sqrt{\\kappa/\\gamma}$ and states that it should be of molecular size.","marker":"[1]"},{"why":"gives the earlier self-consistent-field calculation with negative $\\kappa$ that the authors reproduce when non-local interactions are suppressed; this is the main baseline.","marker":"[3]"},{"why":"provides the gradient expansion of the interaction free energy whose non-local term is the key addition.","marker":"[13]"},{"why":"gives the curvature expansion of the tension used to read $\\kappa$ and $\\bar{\\kappa}$ off cylindrical and spherical grand potentials.","marker":"[15]"},{"why":"supplies the lattice propagator equations that form the numerical backbone of the model.","marker":"[21, 22]"},{"why":"defines the mean-field free energy functional that is extremized.","marker":"[23]"},{"why":"reports the same $\\bar{\\kappa}$ sign switch in microemulsion systems, the precedent the authors extend to liquid-liquid interfaces.","marker":"[32]"},{"why":"analyzes how the extracted rigidities depend on the bending protocol and explains why earlier treatments found negative $\\kappa$.","marker":"[34]"}],"fun_headline_variants":["Rigidity signs flip at strong segregation in L/L interfaces","Gaussian rigidity flips sign at strong segregation","Rigidity ratio plateaus at 1/2 and -3/2","Both rigidities positive near critical, Gaussian flips sign"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire sign result rests on the non-local gradient term $\\frac{1}{6}\\nabla^2\\phi$ in the interaction energy; if the physically correct contact interaction in a density gradient is instead the purely local average, the predicted positive $\\kappa$ and the $\\bar{\\kappa}$ sign switch disappear.","fun_headline_variants_meta":{"raw":{"variants":["Rigidity signs flip at strong segregation in L/L interfaces","Gaussian rigidity flips sign at strong segregation","Rigidity ratio plateaus at 1/2 and -3/2","Both rigidities positive near critical, Gaussian flips sign"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000659,"raw_usage":{"total_tokens":3002,"prompt_tokens":924,"completion_tokens":2078,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":540,"completion_tokens_details":{"reasoning_tokens":2007}},"tokens_in":540,"tokens_out":2078,"duration_ms":16052,"temperature":1.0,"reasoning_tokens":2007,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:41:23.244213+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the same grand potentials with a different discretization of the Laplacian, for instance by including next-nearest-neighbor lattice sites in the gradient stencil, and check whether $\\kappa$ stays positive and $\\bar{\\kappa}$ still changes sign. Alternatively, simulate a symmetric polymer blend at fixed chemical potential and extract the capillary-wave spectrum: the predicted positive $\\kappa$ and $\\lambda$ of order a few segment sizes would be confirmed or refuted by the wavelength dependence of the height fluctuations.","supporting_citations":[{"cited_title":"Laradji and O","cited_arxiv_id":null,"evidence_quote":"defines the crossover length $\\lambda=\\sqrt{\\kappa/\\gamma}$ and states that it should be of molecular size."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the earlier self-consistent-field calculation with negative $\\kappa$ that the authors reproduce when non-local interactions are suppressed; this is the main baseline."},{"cited_title":"Laradji and R","cited_arxiv_id":null,"evidence_quote":"provides the gradient expansion of the interaction free energy whose non-local term is the key addition."},{"cited_title":"Helfrich, Z","cited_arxiv_id":null,"evidence_quote":"gives the curvature expansion of the tension used to read $\\kappa$ and $\\bar{\\kappa}$ off cylindrical and spherical grand potentials."},{"cited_title":"Merkl, T","cited_arxiv_id":null,"evidence_quote":"defines the mean-field free energy functional that is extremized."},{"cited_title":"Helfand, The Journal of Chemical Physics 62, 999 (1975)","cited_arxiv_id":null,"evidence_quote":"reports the same $\\bar{\\kappa}$ sign switch in microemulsion systems, the precedent the authors extend to liquid-liquid interfaces."},{"cited_title":"He identified the so-called equi- librium bending mode where µ controls the curvature","cited_arxiv_id":null,"evidence_quote":"analyzes how the extracted rigidities depend on the bending protocol and explains why earlier treatments found negative $\\kappa$."}],"review_version":1}