{"id":"1db2e9bb-6d1b-4377-bb68-f4eea71aab17","arxiv_id":"1908.04443","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A local-virtual-density scheme for imposing contact angles at curved boundaries in pseudopotential lattice Boltzmann simulations removes the unphysical mass-transfer layer and keeps spurious currents low.","lead":"This paper investigates ways to set the contact angle where a droplet meets a curved solid wall in lattice Boltzmann flow simulations, and proposes a simpler scheme based on a local imaginary wall density. The proposed scheme avoids the spurious thick layer of fluid that appears near the wall with the standard constant-density method, while keeping spurious currents small.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No proof that the improved scheme avoids the mass-transfer layer in the high-contact-angle regime; its local virtual density can degenerate to the constant-density case that causes the layer.","rationale":"The paper is a useful methods contribution: the local virtual-density construction is simple, the comparisons to the geometric-formulation and solid-fluid schemes are informative, and the reduction of spurious currents at moderate angles is credible. My concern is narrower. The original scheme's mass-transfer artifact is worst when the constant wall density is close to rho_g; the improved scheme can reproduce that condition whenever a high contact angle forces rho_w = rho_l - DeltaRho down to rho_g. In a single-phase region the average in Eq. (17) is uniform, so the improved rule reduces exactly to the constant-density rule; the paper does not show density profiles for the improved scheme at theta approximately 158 degrees, only velocity profiles, which are insensitive to a thin low-density layer. Thus the central claim is not false on the evidence, but it is unverified in the regime most relevant to the motivation. This matches the reader's weakest assumption about the heuristic, under-validated scalar parameterization, and supports the CONDITIONAL verdict rather than a full acceptance or rejection.","tokens_in":15368,"tokens_out":12431,"duration_ms":130906,"concrete_test":"Run the Poiseuille channel test of Sec. IV B with the improved virtual-density scheme at theta approximately 158 degrees (liquid-filled channel, DeltaRho set as in the paper) and plot the fluid density profile across the channel; measure the thickness of the near-wall density deviation. Repeat the static cylindrical droplet at theta approximately 158 degrees (R=70) and plot the density profile along the central vertical line through the cylinder as in Fig. 7. If the density deviates from rho_l over more than two lattice units in either test, the headline claim that the improved scheme does not suffer a thick mass-transfer layer fails in the high-contact-angle regime.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the local virtual-density rule, Eqs. (16)-(17), removes the thick mass-transfer layer while retaining simplicity and low spurious currents. The load-bearing condition is that the local rule remains effective precisely where the original constant-density scheme fails. This is not demonstrated. For increasing contact angle, Eq. (16) sets rho_w(x)=ave(rho(x))-DeltaRho; in a single-phase domain, ave(rho) is uniform, so rho_w is constant and the scheme degenerates to the original virtual-density scheme. To reach theta approximately 158 degrees, DeltaRho must be large enough that rho_w near the liquid approaches rho_g, the exact condition that produces the four-lattice mass-transfer layer for the original scheme in Fig. 7. The paper reports only velocity profiles for the improved scheme at theta approximately 158 (Fig. 12), not density profiles; a thin low-density wall layer can leave the bulk velocity nearly parabolic while still constituting the artifact the scheme claims to remove. The static-droplet validations (Figs. 5 and 14) cover only theta approximately 34, 88, 125 in 2D and 53, 88, 145 in 3D, and contact angles are read visually, so the high-theta regime is not assessed with density diagnostics.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper investigates contact-angle implementation in pseudopotential lattice Boltzmann simulations with curved boundaries. It compares four schemes: the solid-fluid interaction scheme, the constant virtual-density scheme, a curved geometric-formulation scheme extended from the phase-field method, and a newly proposed improved virtual-density scheme. In the improved scheme, Eqs. (16)-(17), the virtual density at a solid point is obtained from a weighted average of neighboring fluid densities, multiplied by a factor phi or reduced by an offset DeltaRho, so that the virtual density is local rather than constant. The paper claims this improved scheme removes the thick mass-transfer layer produced by the constant virtual-density scheme while retaining its simplicity and low spurious currents. Validation is provided through static contact-angle tests on a cylindrical surface (Figs. 4-8), spurious-current comparisons (Figs. 9-10), density extrema comparisons (Fig. 11), Poiseuille flow between parallel plates (Fig. 12), droplet impact on a cylinder (Fig. 13), and static contact angles on a spherical surface in 3D (Fig. 14). The central claim is that the improved scheme is a simple, low-artifact contact-angle treatment that works in both 2D and 3D curved geometries.","tokens_in":15585,"tokens_out":7543,"duration_ms":79227,"significance":"If the central claim holds, this is a practically useful contribution to the pseudopotential LB community: it offers a simple local virtual-density formula that cures the mass-transfer-layer artifact of the standard virtual-density scheme without the implementation complexity of the geometric-formulation scheme and with much smaller spurious currents than the solid-fluid interaction scheme. The paper contains several direct quantitative comparisons that support the central claim, including density profiles showing the mass-transfer layer (Fig. 7), maximum spurious currents across a range of contact angles (Fig. 10), a Poiseuille-flow benchmark against the analytical solution (Fig. 12), and a 3D demonstration on a spherical surface (Fig. 14). The manuscript is clearly written and the numerical tests are reproducible from the description, although no code is provided. The main weakness is that the absence of the mass-transfer layer is not demonstrated with density diagnostics at high contact angles, where the improved scheme's local virtual density can degenerate to a constant value in uniform density regions, potentially reproducing the artifact it claims to eliminate.","major_comments":[{"comment":"The Poiseuille-flow test shows that the improved virtual-density scheme matches the analytical solution for theta = 158 degrees, but no density profiles normal to the wall are reported for this high-contact-angle case. Because the mass-transfer layer is a density artifact, a velocity profile that matches the analytical solution is only indirect evidence of its absence: a thin low-density wall layer can in principle leave the bulk velocity nearly parabolic. The authors should provide density profiles for the improved scheme at theta around 145-158 degrees, for both the flat channel and the curved cylinder, to directly demonstrate the absence of the mass-transfer layer in the high-theta regime.","section":"Sec. IV.B, Fig. 12"},{"comment":"In a region where the fluid density is uniform, rho_ave in Eq. (17) is constant, so the improved virtual density rho_w = rho_ave - DeltaRho in Eq. (16) reduces to a constant value. For large positive contact angles, DeltaRho is large enough that rho_w approaches rho_g, which is exactly the condition under which the original virtual-density scheme produces the four-lattice mass-transfer layer shown in Fig. 7. The paper offers no mechanistic explanation for why the improved scheme suppresses the layer in this limiting case. The authors should either provide an analysis of this degeneracy or present numerical evidence (e.g., density profiles) showing that the local averaging prevents the artifact even when rho_w is effectively constant and close to rho_g.","section":"Sec. III.B, Eq. (16)"},{"comment":"The density-profile comparison in Fig. 7 covers only the improved scheme at theta approximately 34 and 125 degrees. The claim that the improved scheme 'does not suffer from a thick mass-transfer layer' is thereby established only for moderate contact angles. The high-contact-angle regime on a curved boundary, which is where the original scheme behaves worst, is not assessed with density diagnostics. The 3D test in Fig. 14(c) reaches theta approximately 145 degrees but shows only density contours, not a quantitative profile. The central claim of the abstract should be supported by density profiles for the improved scheme at high contact angles on the cylindrical surface.","section":"Sec. IV.A, Figs. 5 and 7"}],"minor_comments":[{"comment":"The phrases 'for decreasing' and 'for increasing' in Eq. (16) are ambiguous; they should be clarified as decreasing or increasing the contact angle relative to 90 degrees, and the sign convention should be stated explicitly.","section":"Sec. III.B, Eq. (16)"},{"comment":"The contact angles are reported as approximate values, but no description is given of how they are measured from the simulation results. A brief statement of the measurement procedure (e.g., fitting the interface to a circle and measuring the tangent angle) would improve reproducibility.","section":"Captions of Figs. 4, 5, 6, 8, and 14"},{"comment":"The relative error E_r for the Poiseuille flow is introduced in the text but the equation appears garbled in the typeset version; it should be presented in a clearly formatted equation.","section":"Sec. IV.B, relative error definition"},{"comment":"The spurious-current comparison is made at contact angles that are only approximately matched across schemes (e.g., 119-125 degrees in Fig. 9). The authors should state whether the small differences in the target angle affect the quantitative comparison.","section":"Sec. IV.A, Fig. 10"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and presents a useful practical scheme. The major concern is not that the central claim is likely false, but that the evidence for the absence of the mass-transfer layer at high contact angles is incomplete, and the limiting behavior of Eq. (16) in uniform density regions raises a specific correctness risk that the authors should address. With additional density diagnostics and a discussion of the degeneracy, the paper could be acceptable. No concerns about novelty or citation patterns; the authors cite relevant prior work, including their own, appropriately."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The core idea is the local virtual-density rule in Eqs. (16)-(17): instead of a constant wall density, you use a density-weighted average of neighboring fluid nodes, scaled by φ or shifted by Δρ. That is genuinely new in this context, and it is simple enough that practitioners will actually use it. The paper also extends the 2D curved geometric-formulation scheme to pseudopotential LB, which is useful as a reference even if it cannot go to 3D.\n\nWhat the paper does well is the comparison. The density profiles in Fig. 7 clearly show the thick mass-transfer layer of the original virtual-density scheme, and the improved scheme's layer is much thinner there. The spurious-current comparison (Fig. 10) is quantitative and convincing: the solid-fluid interaction scheme is orders of magnitude worse. The Poiseuille flow test (Fig. 12) is a nice independent benchmark because it compares against an analytical solution, and the droplet-impact test shows that the improved scheme tracks the geometric-formulation result while the original scheme deviates. The 3D sphere tests demonstrate that the scheme works in three dimensions, which is where many practical LB simulations live. All of this is real evidence, and the presentation is clear.\n\nThe main soft spot is the one the stress-test note flags. The improved scheme is supposed to fix the mass-transfer layer, but the mechanism that causes the layer in the original scheme is a wall density close to the gas density. In the improved scheme, achieving a very high contact angle (θ around 158°) requires a large Δρ, which for a liquid-filled region pushes ρ_ave - Δρ down near ρ_g. That is exactly the dangerous regime. The paper's only high-angle test is the Poiseuille flow at θ≈158°, and it reports velocity profiles only. A thin low-density wall layer can leave the velocity profile nearly parabolic, so this is not a sensitive diagnostic. The static-droplet tests stop at 125° in 2D and 145° in 3D, so there is no density-profile check in the regime where the scheme might degenerate. This is a genuine gap, though not a fatal one: the improved scheme is still a clear improvement at moderate angles, and the missing test is easy to add.\n\nTwo minor complaints: contact angles are read visually from contour plots without any uncertainty, and no code or data are shipped. Both are common in this literature and do not undermine the central claim.\n\nWho is this for? Anyone doing pseudopotential LB simulations with curved solid boundaries, especially in 3D. It is a useful engineering-style contribution, not a deep theoretical one. It deserves a serious referee; the high-angle density diagnostic should be requested in revision.","headline":"A solid, incremental contribution to pseudopotential LB contact-angle modeling; the local virtual-density idea is genuinely new and clearly explained, but the paper leaves a real gap at very high contact angles where the improved scheme may quietly revert to the behavior it claims to fix.","tokens_in":16127,"tokens_out":2934,"would_cite":true,"duration_ms":31515,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["47.11.-j"],"model":"deepseek-v4-flash","headline":"The paper proposes an improved virtual-density scheme that replaces the constant virtual wall density in pseudopotential lattice Boltzmann simulations with a locally averaged virtual density, eliminating the unphysical thick mass-transfer…","keywords":["pseudopotential lattice Boltzmann","contact angle","curved boundaries","virtual density","mass-transfer layer","spurious currents","multiphase flow","wetting"],"falsifier":"Fix one value of $\\varphi$ (or $\\Delta\\rho$), lattice resolution, and equation of state, and measure the static contact angle on cylindrical or spherical surfaces of several radii; if the angle drifts systematically with curvature, or a concave boundary redevelops a density layer thicker than one or two lattices near the wall, the local-average premise is insufficient.","tokens_in":15113,"feed_emoji":"💧","tokens_out":10630,"duration_ms":91608,"temperature":0.7,"pith_summary":"The paper tackles a practical obstacle in pseudopotential lattice Boltzmann simulations of multiphase flow: imposing a contact angle on a curved solid boundary without creating artifacts. It shows that the usual constant-virtual-density scheme produces a thick unphysical mass-transfer layer near the wall, while the solid-fluid interaction scheme produces much larger spurious currents. The proposed fix is an improved virtual-density scheme in which each solid node gets a virtual density obtained as a weighted average of neighboring fluid densities, multiplied or shifted by one scalar to tune the angle (Eqs. 16 and 17). On cylindrical and spherical surfaces the scheme reproduces static and dynamic wetting, removes the mass-transfer layer, keeps maximum spurious currents below 0.006, and remains simple to implement in two and three dimensions.","feed_headline":"Local wall density fixes wetting artifacts on curved surfaces","feed_subtitle":"Switching to a locally averaged wall density removes the mass layer and keeps spurious currents low.","key_machinery":"The central object is the local virtual density $\\rho_{\\mathrm{ave}}(\\mathbf{x})$ defined by Eq. (17): at each solid lattice point, the density assigned to the solid phase is a weighted average over neighboring fluid nodes, using the same weights as the pseudopotential interaction force. This local value is then scaled by $\\varphi$ or reduced by $\\Delta\\rho$ through Eq. (16) to set the contact angle, and clipped to the liquid-gas density range. The local averaging is the mechanism that carries the argument: it makes the solid's pseudopotential respond to the local fluid distribution, so the boundary does not impose the artificial density jump that produces the mass-transfer layer, while the scheme keeps the simplicity and low spurious currents of the original virtual-density treatment.","core_discovery":"The central claim is that the thick mass-transfer layer seen with the virtual-density scheme comes from using a constant virtual density $\\rho_w$ in the solid, and that replacing it with a local virtual density $\\rho_w(\\mathbf{x})$ built from the weighted average $\\rho_{\\mathrm{ave}}(\\mathbf{x})$ of neighboring fluid densities removes the layer. The proposed update is Eq. (16): for decreasing contact angle, $\\rho_w(\\mathbf{x})=\\varphi\\rho_{\\mathrm{ave}}(\\mathbf{x})$ with $\\varphi\\ge 1$; for increasing contact angle, $\\rho_w(\\mathbf{x})=\\rho_{\\mathrm{ave}}(\\mathbf{x})-\\Delta\\rho$ with $\\Delta\\rho\\ge 0$, then the value is clipped to $[\\rho_g,\\rho_l]$. With $\\varphi=1$ or $\\Delta\\rho=0$ the scheme reduces to the standard near-90-degree case. The paper validates the scheme with static contact angles on a circular cylinder and on a sphere, shows density profiles where the original scheme's layer is about four lattices thick and the improved one has no comparable layer, and reports maximum spurious currents below 0.006 for the virtual-density-based schemes versus about 0.1 for the solid-fluid interaction scheme. In a droplet-impact test at Reynolds number 600 and a 60-degree contact angle, the improved scheme matches the geometric-formulation scheme, which the paper also extends to curved pseudopotential boundaries but notes is complicated and not directly usable in three dimensions.","pith_inferences":["Beyond the paper, the same local-average prescription could plausibly cure the fictitious-density mass-transfer artifact reported in color-gradient multiphase lattice Boltzmann schemes, since the mechanism of a constant wall density is common to both.","Because the improved scheme only averages neighboring fluid densities with fixed weights, it should transfer naturally to moving or deformable solid boundaries, where the curved geometric-formulation scheme becomes impractical.","The mapping from $\\varphi$ or $\\Delta\\rho$ to the contact angle is demonstrated on one cylinder and one sphere; a systematic sweep over surface curvature would show whether the mapping is universal or needs recalibration.","The heuristic scalar in Eq. (16) invites a derivation from a local force balance or Young's-law argument that would turn a calibrated parameter into a predictive one."],"forward_implications":["Pseudopotential lattice Boltzmann simulations can impose contact angles on curved solids with a one-parameter local-density formula and without the thick mass-transfer layer near the boundary.","The scheme works in three dimensions on a spherical surface, whereas the curved geometric-formulation scheme cannot be applied directly in three dimensions.","Maximum spurious currents stay below 0.006 for the improved scheme, roughly two orders of magnitude below the order-0.1 values of the solid-fluid interaction scheme.","In a droplet-impact test at Reynolds number 600 and a 60-degree contact angle, the improved scheme agrees with the geometric-formulation scheme, showing that removing the mass layer restores agreement for dynamic wetting.","Because the improved scheme keeps the interaction-force structure of the original virtual-density scheme, it preserves ease of implementation and remains compatible with the thermodynamic-consistency forcing used by the model."],"supporting_citations":[{"why":"It introduced the solid-fluid interaction scheme used as the comparison baseline; that scheme produces large spurious currents near curved walls.","marker":"[12]"},{"why":"It introduced the constant virtual wall density for contact angles in pseudopotential lattice Boltzmann, the treatment the improved scheme replaces.","marker":"[18]"},{"why":"It reported the thick unphysical mass-transfer layer generated by the virtual-density scheme, the artifact the paper removes.","marker":"[21]"},{"why":"It demonstrated the analogous fictitious-density mass-transfer artifact in color-gradient lattice Boltzmann, supporting the diagnosis that a constant wall density is the cause.","marker":"[20]"},{"why":"It proposed the geometric-formulation contact-angle scheme for flat phase-field surfaces, which the paper extends to curved pseudopotential boundaries.","marker":"[23]"},{"why":"It devised the characteristic moving contact-line model for curved surfaces, providing the basis for the curved geometric-formulation scheme used as a benchmark.","marker":"[26]"},{"why":"It supplies the Peng-Robinson equation of state and pseudopotential form used in all numerical tests.","marker":"[35]"},{"why":"It provides the improved forcing scheme that realizes thermodynamic consistency in the MRT pseudopotential model used here.","marker":"[28]"},{"why":"It provides the D3Q19 pseudopotential MRT lattice Boltzmann model used for the three-dimensional spherical-surface validation.","marker":"[33]"}],"fun_headline_variants":["Local wall density removes wetting layer on curved LB boundaries","Improved virtual-density scheme for curved LB contact angles","Local wall density eliminates thick mass layer in LB wetting","Cut spurious currents and wetting layers with local wall density","Locally averaged wall density fixes LB wetting on curved walls"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that a single scalar multiplier or offset applied to a weighted local average of neighboring fluid densities, Eqs. (16) and (17), is enough to represent the wetting condition at every point of a curved solid boundary, so the same parameter gives the same contact angle regardless of local curvature and surrounding flow.","fun_headline_variants_meta":{"raw":{"variants":["Local wall density removes wetting layer on curved LB boundaries","Improved virtual-density scheme for curved LB contact angles","Local wall density eliminates thick mass layer in LB wetting","Cut spurious currents and wetting layers with local wall density","Locally averaged wall density fixes LB wetting on curved walls"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000483,"raw_usage":{"total_tokens":2466,"prompt_tokens":1104,"completion_tokens":1362,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":720,"completion_tokens_details":{"reasoning_tokens":1291}},"tokens_in":720,"tokens_out":1362,"duration_ms":9393,"temperature":1.0,"reasoning_tokens":1291,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:42:41.219105+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fix one value of $\\varphi$ (or $\\Delta\\rho$), lattice resolution, and equation of state, and measure the static contact angle on cylindrical or spherical surfaces of several radii; if the angle drifts systematically with curvature, or a concave boundary redevelops a density layer thicker than one or two lattices near the wall, the local-average premise is insufficient.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It introduced the solid-fluid interaction scheme used as the comparison baseline; that scheme produces large spurious currents near curved walls."},{"cited_title":"Benzi, L","cited_arxiv_id":null,"evidence_quote":"It introduced the constant virtual wall density for contact angles in pseudopotential lattice Boltzmann, the treatment the improved scheme replaces."},{"cited_title":"Huang, Z","cited_arxiv_id":null,"evidence_quote":"It reported the thick unphysical mass-transfer layer generated by the virtual-density scheme, the artifact the paper removes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It demonstrated the analogous fictitious-density mass-transfer artifact in color-gradient lattice Boltzmann, supporting the diagnosis that a constant wall density is the cause."},{"cited_title":"Ding and P","cited_arxiv_id":null,"evidence_quote":"It proposed the geometric-formulation contact-angle scheme for flat phase-field surfaces, which the paper extends to curved pseudopotential boundaries."},{"cited_title":"Liu and H","cited_arxiv_id":null,"evidence_quote":"It devised the characteristic moving contact-line model for curved surfaces, providing the basis for the curved geometric-formulation scheme used as a benchmark."},{"cited_title":"Yuan and L","cited_arxiv_id":null,"evidence_quote":"It supplies the Peng-Robinson equation of state and pseudopotential form used in all numerical tests."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It provides the improved forcing scheme that realizes thermodynamic consistency in the MRT pseudopotential model used here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It provides the D3Q19 pseudopotential MRT lattice Boltzmann model used for the three-dimensional spherical-surface validation."}],"review_version":1}