{"id":"2bf15b09-6fa9-464e-961d-3d47e0ab701e","arxiv_id":"2607.26920","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Surface evaporation of a passive component drives anisotropic global coarsening in an otherwise isotropic conserved ternary mixture, while bulk regions retain Model B t^{1/3} scaling.","lead":"Boundary evaporation alone can make phase-separating mixtures coarsen at different rates along different directions, even when the bulk dynamics stay isotropic. That gives a simple external knob for morphology control in coatings, blends, and thin films.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"Global anisotropic exponents may be superposition artifacts of staggered layers rather than true direction-dependent coarsening laws.","rationale":"The reader correctly isolated the weakest link: interpretability of full-sample directional lengths when layers are morphologically staggered. That concern is load-bearing for the strongest claim (“different effective global growth laws… induced by boundary fluxes rather than altered intrinsic bulk dynamics”). The paper already supplies the evidence that makes the concern acute (layer-wise morphology sequence, passive profiles, and the explicit composite-exponent discussion), so no stronger internal inconsistency is needed. The continuum corroboration and bulk Model B recovery are real strengths but do not remove the ambiguity about what the global exponents mean. A composition-matched re-analysis is a decisive, low-cost check that either rehabilitates the anisotropic-law language or forces the claim to be restated as “boundary-driven morphological gradients produce elongated domains and effective composite exponents.” This keeps the verdict CONDITIONAL, aligned with the reader, pending that clarification plus basic ensemble statistics.","tokens_in":9945,"tokens_out":572,"duration_ms":12326,"concrete_test":"Recompute the two-point correlation lengths after restricting the average to a sliding window of fixed passive-fraction (or fixed local active density) rather than the full lattice; if the direction-dependent exponents collapse to a common Model B value once composition is controlled, the claimed anisotropic growth laws are superposition artifacts. Report ensemble means and standard errors over ≥10 independent runs for both full-sample and composition-matched lengths.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim requires that full-sample two-point correlation lengths along x/y versus z report genuinely different coarsening laws induced by boundary fluxes. The paper’s own layer-wise analysis (Figs. 3–4) shows that passive fraction is strongly z-inhomogeneous, so successive planes sit at different morphological stages and cross over to binary-like Model B growth at different times. The authors explicitly note that this staggered progression “can produce an effective transverse growth exponent smaller than the local Model B value,” which is precisely how the intermediate ~1/4 exponents for \\alpha=10^{-2} and 10^{-1} are explained. Consequently the headline anisotropic scaling (different effective global growth laws parallel and perpendicular to the boundary) risks being a composite measurement artifact of an inhomogeneous sample rather than a directionally distinct dynamical process. Bulk cubes recovering isotropic t^{1/3} further weakens the claim that the system as a whole obeys anisotropic coarsening laws; the anisotropy may be only a global averaging effect. The continuum model shows similar elongation but does not resolve whether its extracted exponents are likewise composite.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript argues that boundary forcing alone can break the usual isotropic dynamical scaling of conserved phase separation. Using a three-state Blume–Capel lattice model with Kawasaki dynamics and stochastic evaporation of the passive species at one boundary, together with a continuum nonlocal Model-B-like system with a Robin flux for the passive component, the authors report direction-dependent effective growth of characteristic lengths extracted from two-point correlations: different power laws parallel and perpendicular to the evaporation direction. Layer-wise sections and passive-fraction profiles show macroscopic z-gradients and staggered morphological stages, while bulk sub-volumes away from the boundaries recover isotropic t^{1/3} Model B scaling. The anisotropy is therefore attributed to boundary-induced concentration gradients and fluxes rather than to a change of the intrinsic bulk dynamics. A simple diffusion-length argument is offered for the longitudinal t^{1/2}-like regime at weak evaporation.","tokens_in":10178,"tokens_out":1526,"duration_ms":36182,"significance":"If the anisotropic global scaling is cleanly established and not an averaging artifact, the result is a useful conceptual contribution: it identifies purely external boundary mass loss as a generic route to direction-dependent coarsening in otherwise isotropic conserved systems, without built-in anisotropic interactions or shear. That bulk cubes retain standard Model B while the full sample does not is a clear and falsifiable distinction. The dual lattice/continuum evidence, the layer-resolved passive profiles, and the explicit no-evaporation baseline from prior work strengthen the case. The finding is relevant to evaporative morphology control in polymer blends and thin films. The main scientific value hinges on whether the reported global exponents are genuine direction-dependent coarsening laws or composites of staggered layers; resolving that point determines how far the claim travels.","major_comments":[{"comment":"The central claim of distinct effective global growth laws parallel vs. perpendicular to the boundary is undercut by the paper’s own layer-wise analysis (Figs. 3–4 and surrounding text). Passive fraction is strongly inhomogeneous in z; successive planes sit at different morphological stages and cross over to binary-like Model B growth at different times. The authors explicitly state that this staggered progression “can produce an effective transverse growth exponent smaller than the local Model B value,” which is how the intermediate ~1/4 exponents for α=10^{-2} and 10^{-1} are explained. Full-sample correlation lengths along x/y versus z are therefore at risk of being superposition measures of an inhomogeneous sample rather than directionally distinct dynamical processes. Bulk cubes recovering isotropic t^{1/3} reinforce this reading. The manuscript needs either (i) a quantitative decom","section":"Figs. 2–4 and discussion of effective 1/4 exponents"},{"comment":"Power-law identification rests on visual guide lines and selected collapse windows (Fig. 2 top/bottom; Fig. 7) rather than documented fitting ranges, uncertainties, or sensitivity to the definition of the characteristic length from G. Prefactors are free (0.12 t^{1/4}, 0.45 t^{1/3}, 0.12 t^{1/2}, etc.), and the continuum exponents shift with α' in a way that is only loosely tied to the lattice cases. For the weak-evaporation longitudinal t^{1/2} claim and the fast-evaporation transverse t^{1/4} claim to support the abstract’s “different effective global growth laws,” the paper should report how the exponents are extracted (fit windows, alternative length definitions such as first zero or half-height of G, run-to-run variation) and whether the collapses remain stable under those choices.","section":"Fig. 2 and Fig. 7"},{"comment":"The continuum counterpart (Eqs. 2–3, Figs. 6–7) is used to argue that the anisotropy is not a lattice artifact, but the analysis is thinner: fewer diagnostics, no layer-wise passive profiles or bulk-cube checks analogous to the lattice, and evaporation imposed at both top and bottom. A parallel bulk-versus-global comparison in the continuum model would substantially strengthen the robustness claim; without it, the continuum evidence mainly shows morphological elongation, not the same scaling dichotomy.","section":"Eqs. (2)–(3), Figs. 6–7"}],"minor_comments":[{"comment":"Notation for the initial passive fraction switches between c_0 (lattice) and 1−ϕ_0 (continuum); a single convention or an explicit dictionary would help.","section":null},{"comment":"Evaporation is said to act only at the “top” boundary in the lattice description, yet Fig. 4 and the continuum BC (3) indicate loss at both ends; clarify the lattice implementation (one face vs two).","section":"Methods paragraph and Eq. (3)"},{"comment":"Inverse temperature is given as 0.80 for the lattice and β=1 for continuum without relating the two scales; a brief remark on corresponding reduced temperatures would aid comparison.","section":null},{"comment":"Fig. 1 caption plane labels “(x,0,z)”, “(L−1,y,z)”, “(x,y,L−1)” are slightly ambiguous; stating which face is the evaporating boundary in each panel would improve readability.","section":"Fig. 1"},{"comment":"Typos/style: “J¨ averg ˚ ard”, “att=” spacing, “conponent”, “scillating”, and repeated “in this case” in the abstract; standard copy-edit pass needed.","section":null},{"comment":"The simple ℓ∼(Dt)^{1/2} argument is plausible for weak evaporation but is not checked against the measured passive-gradient width versus time; a single panel of gradient width vs t would make the scaling argument quantitative.","section":"Scaling argument paragraph"}],"recommendation":"major_revision","confidential_remarks":"The skeptic’s superposition concern is real and load-bearing; the authors already half-acknowledge it when explaining the 1/4 exponents. I do not see a fatal inconsistency, but the abstract and title currently sell “anisotropic coarsening” and “different effective global growth laws” more strongly than the layer-wise evidence supports. Major revision with either stronger diagnostics or more careful framing should be sufficient; rejection is not warranted. Fit to cond-mat.stat-mech / PRE-style journals is appropriate. Prior self-citations supply baseline and well-posedness and do not appear to inflate novelty improperly."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The useful takeaway is simple: in a standard conserved ternary mixture, surface evaporation of the passive species alone is enough to break isotropic Model B scaling at the system level. You get different effective growth laws parallel and perpendicular to the free surface, while bulk cubes still look like ordinary t^{1/3}. That is a clean mechanistic point, and it is new relative to classical surface-directed spinodal decomposition (which needs a preferential surface field) and to the authors’ own no-evaporation baselines.\n\nWhat they did well: both lattice Kawasaki–Blume–Capel and a nonlocal continuum Model-B-like system show the same qualitative elongation and direction-dependent collapses. The layer-wise cuts and passive-fraction profiles (Figs. 3–4) are the strongest part of the paper; they make the mechanism visible and stop the authors from claiming a new local dynamical class. The bulk recovery of isotropic scaling is also honest and strengthens rather than weakens the boundary-flux story. Citation pattern is appropriate: Bray, Hohenberg–Halperin, Puri–Binder, plus their prior ternary work as baseline.\n\nSoft spots are real but proportionate. Global exponents, especially the intermediate ~1/4 transverse ones at faster evaporation, are partly composites of staggered layer crossovers—the paper says so explicitly. So the headline “anisotropic coarsening laws” should be read as effective global scaling under macroscopic gradients, not as two distinct bulk growth laws living side by side. Statistics are thin (single realizations, no ensemble errors, guide-line prefactors), continuum numerics are lightly documented, and no code is released. None of that kills the result; it just means the exponents are heuristic.\n\nThis is for people who care about morphology control in drying films, coatings, and nonequilibrium conserved dynamics. It is not a theorem paper and not a materials recipe, but it is a solid computational clarification. I would send it to referees; ask for ensemble uncertainties, sharper language on composite versus local exponents, and parameter/code release. Worth engaging.","headline":"Boundary evaporation can produce direction-dependent effective coarsening in an otherwise isotropic conserved ternary system; the claim is real as a global phenomenon, with the main caveat already flagged by the authors themselves.","tokens_in":10822,"tokens_out":505,"would_cite":true,"duration_ms":9966,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Boundary forcing alone can break isotropic scaling in conserved phase separation, producing different coarsening laws parallel and perpendicular to the boundary.","keywords":["phase separation","ternary mixture","evaporation","anisotropic coarsening","Model B","Blume–Capel model","boundary forcing","morphology formation"],"falsifier":"Restrict correlation analysis to many independent bulk subvolumes far from the evaporating faces across a range of evaporation rates: if those subvolumes still show isotropic t^{1/3} scaling while full-sample x/y versus z lengths keep splitting into the reported anisotropic exponents, the boundary-flux claim holds; if the split disappears once layers are synchronized or mass is held fixed, the global laws are composites rather than true anisotropic coarsening.","tokens_in":10742,"feed_emoji":"📉","tokens_out":851,"duration_ms":23272,"temperature":0.7,"pith_summary":"In conserved phase separation, domain growth is usually expected to be isotropic, with one universal law (Model B, size growing like t to the one-third) in every direction. This paper argues that forcing only at the boundary is enough to break that symmetry. In a ternary mixture where a passive component evaporates from the surface, mass loss builds macroscopic concentration gradients that make domains coarsen at different effective rates along and across the evaporation direction, while interior bulk regions still look like ordinary isotropic Model B. The claim matters because it separates boundary-driven anisotropy from any change in the intrinsic bulk dynamics, and it suggests that surface conditions alone can be used to steer morphology in nonequilibrium phase-separating materials.","feed_headline":"Boundary evaporation alone splits coarsening into two growth laws","feed_subtitle":"Surface mass loss drives anisotropic domain growth while the bulk still follows ordinary Model B scaling.","key_machinery":"A ternary Blume–Capel lattice model (two conserved active species plus a passive evaporating component) and its continuum nonlocal drift–diffusion counterpart: Kawasaki bulk exchanges plus stochastic passive loss at the free surface generate macroscopic composition gradients that couple to active-species coarsening and yield direction-dependent two-point correlation lengths.","core_discovery":"Boundary forcing by surface evaporation of a passive species is sufficient to break dynamical scaling symmetry in an otherwise isotropic conserved system. Macroscopic concentration gradients produced by progressive mass loss drive anisotropic coarsening, with different effective global growth laws parallel and perpendicular to the boundary, while bulk regions retain standard Model B scaling—so the anisotropy is attributed to boundary fluxes rather than altered intrinsic dynamics.","pith_inferences":["The same boundary-gradient mechanism should appear in other conserved multi-component models (not only Blume–Capel) whenever one species is removed at a face and the others remain locally conserved.","If evaporation rate and sample thickness can be tuned so the composition front sweeps uniformly, one might switch continuously between isotropic Model B and strongly anisotropic global growth.","Thin-film device processing that already uses solvent evaporation may already be operating in this anisotropic regime without recognizing the scaling split."],"forward_implications":["Surface evaporation or analogous boundary mass fluxes can be used as a control knob to set different coarsening rates along chosen axes without redesigning bulk interactions.","Morphology design in drying films and ternary coatings can exploit gradient-driven elongation rather than only composition and interfacial energy.","Claims of universal isotropic Model B scaling in conserved systems must be checked against boundary conditions whenever open or evaporating surfaces are present.","Layer-resolved measurements become necessary: global exponents alone can mix staggered local crossovers into apparent intermediate powers."],"fun_headline_variants":["Boundary evaporation alone splits coarsening into two growth laws","Surface mass loss breaks isotropic scaling in conserved phase separation","Evaporation-driven gradients yield distinct parallel and perpendicular growth","Boundary fluxes alone induce anisotropic coarsening while bulk stays Model B","Passive-species evaporation suffices to break dynamical scaling symmetry"],"cache_read_input_tokens":128,"weakest_assumption_plain":"That full-sample correlation lengths along each axis remain clean direction-dependent coarsening laws even when the passive fraction is macroscopically uneven and different layers sit at different morphological stages, so global exponents are not mainly superposition artifacts.","fun_headline_variants_meta":{"raw":{"variants":["Boundary evaporation alone splits coarsening into two growth laws","Surface mass loss breaks isotropic scaling in conserved phase separation","Evaporation-driven gradients yield distinct parallel and perpendicular growth","Boundary fluxes alone induce anisotropic coarsening while bulk stays Model B","Passive-species evaporation suffices to break dynamical scaling symmetry"]},"model":"grok-4.5","effort":"low","cost_usd":0.003823,"raw_usage":{"total_tokens":1179,"prompt_tokens":704,"num_sources_used":0,"completion_tokens":81,"cost_in_usd_ticks":38228000,"prompt_tokens_details":{"text_tokens":704,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":394,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":704,"tokens_out":81,"duration_ms":8376,"temperature":1.0,"reasoning_tokens":394,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-30T16:59:01.484954+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Restrict correlation analysis to many independent bulk subvolumes far from the evaporating faces across a range of evaporation rates: if those subvolumes still show isotropic t^{1/3} scaling while full-sample x/y versus z lengths keep splitting into the reported anisotropic exponents, the boundary-flux claim holds; if the split disappears once layers are synchronized or mass is held fixed, the global laws are composites rather than true anisotropic coarsening.","supporting_citations":[],"review_version":1}