Pith. sign in

REVIEW 3 major objections 6 minor 42 references

A unified energy-stable finite element approximation for evolving fluidic biomembranes

T0 review · 3 major / 6 minor · reviewed 2026-07-08 · glm-5.2

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read First energy-stable finite element scheme for fluidic biomembranes

desk verdict First fully discrete energy-stable scheme for coupled bulk-surface Navier-Stokes with Willmore forces; stability proofs are sound, well-posedness rests on an unverified LBB condition. read the letter →

arxiv 2607.05998 v1 pith:CT3HTA6U submitted 2026-07-07 math.NA cs.NA

Harald Garcke , Robert Nürnberg , Quan Zhao This is my paper · ORCID
classification math.NAcs.NA MSC 65M6065M1276D0576T99
keywords fluidicbiomembranefiniteelementmethodenergystabilitysurfaceNavier–StokesWillmorearbitraryLagrangian–Eulerianbulk-surfacecouplingunfittedmesh
topics Navier-Stokes Regularity
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper presents what the authors identify as the first fully discrete finite element scheme for simulating fluidic biomembranes — lipid bilayer surfaces that are fluid in-plane, coupled to surrounding viscous fluid, and resistant to bending — that provably preserves the energy dissipation structure of the continuous model at the discrete level. The physical system couples bulk Navier–Stokes flow, surface Navier–Stokes flow on the evolving membrane, and fourth-order bending forces from the Willmore energy. The central difficulty has been that prior discretizations could maintain stability only at the semi-discrete (continuous-in-time) level, leaving the fully discrete case — what one actually computes — without a stability guarantee. The authors close this gap by introducing a surface arbitrary Lagrangian–Eulerian (ALE) formulation that decouples the mesh velocity from the fluid velocity and permits a free tangential surface velocity. This decoupling, combined with a curvature evolution equation treated via the ALE framework, allows the discrete energy identity to close exactly. The resulting schemes are linear at each time step, unconditionally energy-stable (no time-step restriction), and work in both fitted-mesh (mesh follows the interface) and unfitted-mesh (interface cuts through a fixed bulk mesh) settings. The authors prove existence, uniqueness, and stability for both variants and demonstrate the methods on vesicle relaxation, tank-treading, tumbling, and flow through a constriction.

What carries the argument

Surface ALE weak formulation; curvature evolution equation (3.13); skew-symmetric convective forms A and A_Γ; mass-lumped inner products; vertex normals; LBB_Γ inf-sup condition (4.21); P2-(P1+P0) Taylor–Hood elements for fitted; P2-P1 for unfitted; Schur complement solver

What would settle it

Construct a mesh configuration — e.g., a highly distorted interface mesh with near-degenerate vertex normals or a bulk-surface element pairing that violates LBB_Γ — on which the linear system is singular or the discrete energy increases, demonstrating that the stability guarantee is not unconditional in practice.

Watch

Extended reading notes

Core claim

The core mechanism is a unified weak formulation built on three interlocking ideas. First, the bulk mesh velocity and the interface tangential velocity are freed from the fluid velocity via an ALE framework, so the mesh can move independently while geometric consistency is maintained. Second, the bending force — a nonlinear fourth-order geometric quantity — is handled through an evolution equation for the mean curvature, where the ALE technique absorbs the tangential convection naturally. Third, skew-symmetric forms for the convective terms in both bulk and surface Navier–Stokes equations ensure that these terms contribute zero net energy in the discrete test, allowing the energy dissipation

Load-bearing premise

The stability and well-posedness proofs assume an inf-sup condition (LBB_Γ, equation 4.21) for the coupled bulk-surface pressure spaces and nondegeneracy of vertex normals (4.22). These are standard for the element choices made but are not proven for the specific coupled discretization with mass-lumped inner products; if they fail on certain mesh configurations, the guarantees do not hold.

Editorial extensions

If this is right

  • Simulations of vesicle dynamics in complex geometries — microfluidic channels, flow through constrictions, confinement — can now be run with a stability guarantee rather than heuristic time-step tuning, which is the practical barrier for biological and engineering applications.
  • The ALE framework for the curvature evolution equation is not specific to the Willmore energy; it could extend to other interfacial energies with fourth-order geometric forces, such as those involving spontaneous curvature, area-difference elasticity, or phase-separated membranes with line tension.
  • The unification of fitted and unfitted approaches under one weak formulation means practitioners can switch between mesh strategies depending on the problem geometry without changing the mathematical structure or re-deriving stability.
  • The linear-in-each-step property means the scheme avoids nonlinear solvers at every time step, reducing computational cost for long-time simulations of membrane dynamics.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The stability estimate is independent of the interface update step (Remark 4.4), meaning the geometric evolution is energetically safe but geometrically driven by a separate update. This separation could be exploited for adaptive or multirate time-stepping: the Navier–Stokes–curvature system could use a different step size than the interface position update.
  • The LBB_Γ inf-sup condition (4.21) couples bulk and surface pressure spaces in a non-standard way. If one could construct enriched finite element pairs that satisfy this condition constructively — rather than assuming it — the well-posedness and stability guarantees would become fully self-contained.
  • The curvature reset every N=1000 steps in the constriction example (Section 6.2) suggests that curvature error accumulation is a practical bottleneck even with energy stability. A curvature error estimator and adaptive reset strategy would be a natural extension.
  • The framework handles d∈{2,3} uniformly in formulation, but all numerical examples are 2D (planar curves). The 3D case — actual biomembrane vesicles in 3D fluid — would test whether the mesh quality strategies (vertex normals, remeshing) scale.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. This paper presents a unified finite element framework for simulating the dynamics of fluidic biomembranes, coupling bulk Navier–Stokes equations with surface Navier–Stokes equations on an evolving interface, including Willmore-type bending forces. The key innovation is a surface ALE formulation that decouples the bulk mesh velocity from the fluid velocity and permits a free tangential surface velocity, enabling a unified treatment of both fitted (ALE) and unfitted mesh approaches. The authors propose linear, fully discrete schemes and prove unconditional energy stability (Theorems 4.3 and 6.2) and well-posedness (Theorems 4.2 and 6.1) for both approaches. Numerical experiments in 2D demonstrate convergence, volume/area preservation, and physically meaningful vesicle dynamics including tank-treading and tumbling.

Significance. The paper makes a substantial contribution by providing, to the authors' knowledge and consistent with the literature surveyed, the first fully discrete finite element scheme for fluidic biomembranes with a proven unconditional energy stability estimate. The energy stability proofs are carefully constructed: the test function choices are explicitly stated, the skew-symmetric convective terms are correctly handled, and the key cancellations between the bending-force terms and curvature evolution equations are verified. The unified ALE framework that accommodates both fitted and unfitted discretizations is a notable structural contribution. The numerical experiments demonstrate the practical viability of the approach, including a vesicle-through-constriction example that showcases the advantages of the unfitted method. The work builds naturally on the authors' prior contributions to Willmore flow and two-phase flow, extending the surface ALE technique to the coupled bulk-surface Navier–Stokes setting.

major comments (3)
  1. §4.4, Eq. (4.21) and Theorem 4.2: The coupled LBB inf-sup condition (4.21) is assumed, not proven, for the specific element combination used (P2-(P1+P0) bulk with P1 surface pressure via the interpolation operator π¹ₘ). The surface incompressibility constraint (4.18c) involves ⟨∇s·[π¹ₘ ξh], qΓ⟩, and since {π¹ₘ ξ : ξ ∈ [S²ₘ(Ω)]ᵈ} = [S¹ₕ(Γᵐ)]ᵈ, this creates a P1–P1 velocity–pressure pairing on the surface, which is individually not inf-sup stable. The coupled condition (4.21) must compensate through the bulk–surface coupling, but this has not been verified for the specific mass-lumped inner products and element combination employed. If the inf-sup constant C₀ is small or zero on certain mesh configurations (particularly distorted interface meshes), Theorem 4.2 fails and the scheme may not have unique solutions. The authors should either (a) provide a proof or reference establishing (4.21),
  2. §6.1, Theorems 6.1 and 6.2: The unfitted scheme uses P2-P1 elements (4.7a) rather than P2-(P1+P0), and the surface velocity in the viscous term is replaced by π¹ₘ U^{m+1} (6.2a). The proofs of Theorems 6.1 and 6.2 are stated to be analogous to the fitted case but are omitted. Given that the element pair differs and the unfitted discretization involves additional complications (interface elements with averaged coefficients per (6.1), the I⁰ₘ projection of density), the authors should at minimum sketch the proof of Theorem 6.2 to confirm that the same cancellations hold—particularly the interaction between the π¹ₘ-interpolated surface viscous term and the surface incompressibility constraint—and that the inequality a·b − ½|b|² ≤ ½|a|² still applies correctly. The energy estimate (6.6) involves Ē with the mass-lumped inner product ⟨·,·⟩^h_{Γ^m} for the surface kinetic energy, differing from
  3. §4.4, Theorem 4.3, Eq. (4.30): The stability estimate bounds E(ρ^m, U^{m+1}, Γ^m, κ^{m+1}) by E(ρ^{m-1}, U^m, Γ^{m-1}, κ^m), but the interface update (4.20) that produces Γ^{m+1} does not enter the estimate. Remark 4.4 acknowledges this, but the practical implication is that the energy stability is proven at the old geometry Γ^m, not the updated geometry Γ^{m+1}. This means the scheme's stability guarantee is one step behind the geometry. The authors should clarify whether this introduces any consistency concerns for long-time simulations, and whether the energy at the updated geometry Γ^{m+1} can be bounded. The numerical experiments (e.g., Figs. 2–3) do show energy decay, but the theoretical estimate does not directly cover the post-update configuration.
minor comments (6)
  1. Table 1: The convergence rates are not explicitly reported. The errors e_{h,Δt} decrease by roughly a factor of 2 per refinement level, suggesting first-order convergence, but the spatial and temporal contributions are not separated. A brief comment on the expected convergence order and whether the observed rate is consistent would strengthen this table.
  2. §5.1, Eq. (5.1): The matrix block structure uses notation like B^m_Ω, C^m, N^m_{Γ,Ω} without full definitions. While the Schur complement approach is sketched, the precise definitions of these blocks would help readers reproduce the solution strategy.
  3. §4.3, Eq. (4.18a): The term P^{m+1}_{sing} ⟨ν^m_p, ξh⟩^h_{Γ^m} appears with the vertex normal ν^m_p rather than the element normal ν^m used elsewhere in the same equation (e.g., in the bending force term α⟨F^{m+1}_Γ ν^m, ξh⟩^h_{Γ^m}). A brief remark explaining why the vertex normal appears where and why would improve clarity.
  4. §3.3, Remark 3.2: The BGN tangential velocity (3.21) and MDR approach (3.22) are mentioned as alternatives, but the paper only uses the direct fluid velocity update (3.17b). A brief discussion of when these alternatives would be preferred would be helpful.
  5. §6.2, Example 5: The curvature reset every N=1000 steps is mentioned as a practical remedy for error accumulation, but no analysis or heuristic is provided for choosing N. A brief comment on the sensitivity to this parameter would be useful.
  6. References: The paper [25] (Garcke, Nürnberg, arXiv:2508.19198, 2025) appears to be a preprint. If it has been published by the time of revision, the reference should be updated. Similarly, [36] is listed as 'to appear.'

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the careful and constructive report. The referee correctly identifies the main contributions of the paper and raises three substantive points regarding the inf-sup condition, the omitted proofs for the unfitted case, and the geometry lag in the stability estimate. We address each below.

read point-by-point responses

  1. Referee: §4.4, Eq. (4.21) and Theorem 4.2: The coupled LBB inf-sup condition (4.21) is assumed, not proven, for the specific element combination used (P2-(P1+P0) bulk with P1 surface pressure via the interpolation operator π¹ₘ). The surface incompressibility constraint (4.18c) involves ⟨∇s·[π¹ₘ ξh], qΓ⟩, and since {π¹ₘ ξ : ξ ∈ [S²ₘ(Ω)]ᵈ} = [S¹ₕ(Γᵐ)]ᵈ, this creates a P1–P1 velocity–pressure pairing on the surface, which is individually not inf-sup stable. The coupled condition (4.21) must compensate through the bulk–surface coupling, but this has not been verified for the specific mass-lumped inner products and element combination employed. If the inf-sup constant C₀ is small or zero on certain mesh configurations (particularly distorted interface meshes), Theorem 4.2 fails and the scheme may not have unique solutions. The authors should either (a) provide a proof or reference establishing (4.21),

    Authors: The referee is correct that the coupled inf-sup condition (4.21) is assumed rather than proven, and that the surface P1–P1 pairing arising from the interpolation operator π¹ₘ is individually unstable. We acknowledge this gap honestly. The key structural point is that (4.21) is a *coupled* bulk–surface inf-sup condition: the bulk pressure space (P1+P0) and the singularity constraint together with the surface pressure must satisfy a joint inf-sup condition, and the bulk coupling is what provides stability in practice. This is analogous to the coupled inf-sup conditions studied in [7, 8] for related bulk–surface Stokes systems. However, we agree that a rigorous verification for the specific mass-lumped inner products and the P2-(P1+P0)/P1 combination employed here has not been published. We will add a detailed remark clarifying that (4.21) is an assumption, explaining its structural role, and citing the closest existing results from [7, 8] and [5] that support its plausibility. We will also note that the numerical experiments, including those on distorted interface meshes, consistently yield unique solutions, which provides empirical evidence that the inf-sup constant is bounded away from zero in practice. A full proof of (4.21) for the specific element combination is an important open problem that we plan to address in future work, but it is beyond the scope of the present paper. revision: partial

  2. Referee: §6.1, Theorems 6.1 and 6.2: The unfitted scheme uses P2-P1 elements (4.7a) rather than P2-(P1+P0), and the surface velocity in the viscous term is replaced by π¹ₘ U^{m+1} (6.2a). The proofs of Theorems 6.1 and 6.2 are stated to be analogous to the fitted case but are omitted. Given that the element pair differs and the unfitted discretization involves additional complications (interface elements with averaged coefficients per (6.1), the I⁰ₘ projection of density), the authors should at minimum sketch the proof of Theorem 6.2 to confirm that the same cancellations hold—particularly the interaction between the π¹ₘ-interpolated surface viscous term and the surface incompressibility constraint—and that the inequality a·b − ½|b|² ≤ ½|a|² still applies correctly. The energy estimate (6.6) involves Ē with the mass-lumped inner product ⟨·,·⟩^h_{Γ^m} for the surface kinetic energy, differing from

    Authors: The referee's observation is correct: the unfitted case involves the P2-P1 element pair (rather than P2-(P1+P0)), the π¹ₘ-interpolated surface viscous term, averaged coefficients on interface elements, and the I⁰ₘ projection of density. We agree that these differences are non-trivial and that stating the proofs are 'analogous' without verification is insufficient. Upon careful re-examination, we confirm that the key cancellations do carry over. The critical observation is that the surface incompressibility constraint (6.2c) involves ∇s·[π¹ₘ U^{m+1}], and the surface viscous term in (6.2a) is 2µΓ⟨D^m_s(π¹ₘ U^{m+1}), D^m_s(π¹ₘ ξh)⟩_Γm. When ξh = Δt U^{m+1} is chosen as the test function, the surface pressure term becomes −⟨P^{m+1}_Γ, ∇s·[π¹ₘ(Δt U^{m+1})]⟩_Γm, which cancels against the constraint (6.2c) tested with q^h_Γ = Δt P^{m+1}_Γ. The π¹ₘ interpolation is consistent throughout: it appears identically in both the viscous term and the constraint, so the cancellation mechanism from the fitted case is preserved. The inequality a·b − ½|b|² ≤ ½|a|² applies in the same manner because the right-hand side of the energy identity involves terms of the form (I⁰ₘ ρ^{m-1} Ĩ²ₘ U^m, ξh) and ⟨Ĩ²ₘ U^m, Π^{m-1}_m ξh⟩^h_{Γ^{m-1}}, which have the same structure as in the fitted case. The mass-lumped inner product ⟨·,·⟩^h_{Γ^m} in Ē does not affect the algebraic structure of the Young's inequality argument. We will add a proof sketch of Theorem 6.2 in the revised manuscript, explicitly verifying these cancellations and the application of the Young's inequality, as the referee requests. revision: yes

  3. Referee: §4.4, Theorem 4.3, Eq. (4.30): The stability estimate bounds E(ρ^m, U^{m+1}, Γ^m, κ^{m+1}) by E(ρ^{m-1}, U^m, Γ^{m-1}, κ^m), but the interface update (4.20) that produces Γ^{m+1} does not enter the estimate. Remark 4.4 acknowledges this, but the practical implication is that the energy stability is proven at the old geometry Γ^m, not the updated geometry Γ^{m+1}. This means the scheme's stability guarantee is one step behind the geometry. The authors should clarify whether this introduces any consistency concerns for long-time simulations, and whether the energy at the updated geometry Γ^{m+1} can be bounded. The numerical experiments (e.g., Figs. 2–3) do show energy decay, but the theoretical estimate does not directly cover the post-update configuration.

    Authors: The referee correctly identifies a structural feature of the stability estimate: (4.30) bounds the energy at Γ^m, not at the updated geometry Γ^{m+1}. This is a deliberate consequence of the operator-splitting structure of the scheme, where the Navier–Stokes/curvature subsystem (4.18)–(4.19) is solved first (yielding the stability estimate), and the interface update (4.20) is performed in a second stage. We note that this is a standard feature of energy-stable splitting schemes for geometric evolution equations; see, e.g., the analogous structure in our prior work on Willmore flow [24, Theorem 4.3]. Regarding consistency concerns for long-time simulations: the energy at Γ^{m+1} differs from the energy at Γ^m by terms of order O(Δt) arising from the geometric update, which are controlled by the regularity of the discrete solution. In practice, the numerical experiments (Figs. 2–3, 8–10) consistently show energy decay at the updated geometry, confirming that the one-step lag does not introduce instabilities. A formal bound of E at Γ^{m+1} in terms of E at Γ^m would require controlling the change in curvature and surface measure under the update (4.20), which introduces additional geometric terms. We will expand Remark 4.4 to clarify these points: (i) the one-step lag is inherent to the splitting and is consistent with the continuous energy law up to O(Δt) perturbations, (ii) the numerical evidence confirms stability at the updated geometry, and (iii) a formal bound at Γ^{m+1} is possible under additional regularity assumptions but would complicate the presentation, so we leave the estimate at Γ^m as the primary theoretical guarantee. revision: partial

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found — energy stability is proven from first principles; self-citations provide building blocks but are not load-bearing in a circular way.

full rationale

The paper's central claim — unconditional energy stability of the fully discrete scheme — is proven directly within the paper (Theorems 4.3 and 6.2) by choosing specific test functions, combining the discrete equations, and applying the algebraic inequality a·b − ½|b|² ≤ ½|a|². The key cancellations between (4.18a) and (4.19a)–(4.19c) are verified explicitly in the proof of Theorem 4.3 (equation (4.31)→(4.32)→(4.30)). No step reduces to its own input by construction. Self-citations [23,24] (same authors) provide the curvature evolution equation (3.13) and the surface ALE idea, but these are used as starting points from which the present paper independently derives its stability result. The weak formulation (Section 3) and its energy law (Theorem 3.1) are derived self-containedly via Reynolds transport theorem and integration by parts. The LBB inf-sup condition (4.21) is assumed (not proven) for well-posedness (Theorem 4.2), but this is a standard assumption for mixed finite elements and does not affect the stability proof, which does not depend on LBB. The unfitted case (Theorem 6.2) is stated to follow analogously. No fitted parameter is renamed as a prediction; no uniqueness theorem is invoked to forbid alternatives. The derivation chain is self-contained against external benchmarks.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities or postulated particles. All mathematical objects (ALE maps, vertex normals, discrete inner products) are standard finite element constructs. The free parameters are standard discretization and physical parameters, plus two ad hoc numerical thresholds for mesh quality and curvature reset.

free parameters (5)
  • Time step size = varies (e.g., 5e-4, 1e-3)
    Standard discretization parameter, not fitted to data.
  • Mesh size h = varies (e.g., 1/16, 1/128)
    Standard discretization parameter.
  • Physical parameters (rho, mu, alpha, etc.) = varies per example
    Physical constants set by the problem, not by the method.
  • Remeshing threshold R_buk = pi/18
    Chosen ad hoc for mesh quality control in Example 2.
  • Curvature reset interval N = 1000
    Chosen ad hoc in Example 5 to reduce curvature error accumulation.
assumptions (4)
  • domain assumption LBB inf-sup stability condition (4.21) holds for the chosen finite element spaces with mass-lumped inner products.
    Invoked in Theorems 4.2 and 6.1 to guarantee unique solvability. Standard for Taylor-Hood elements but not proven here for the specific coupled bulk-surface formulation.
  • domain assumption Vertex normals are nondegenerate: nu_p^m(q) != 0 for all vertices q.
    Invoked in Theorems 4.2 and 6.1. A geometric condition on the discrete interface mesh.
  • domain assumption The sharp-interface model (2.1), (2.7) with Willmore energy (2.13) accurately represents fluidic biomembrane dynamics.
    Standard model from the literature [7,9], not questioned in this paper.
  • standard math The ALE map Phi[t] is invertible with W^{1,infinity} regularity.
    Stated in (3.1) as a regularity requirement for the weak formulation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A unified energy-stable finite element approximation for evolving fluidic biomembranes." pith.science (2026). https://pith.science/paper/CT3HTA6U

@misc{pith2026260705998,
  author       = {Pith},
  title        = {Pith review of: A unified energy-stable finite element approximation for evolving fluidic biomembranes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CT3HTA6U}},
  note         = {Machine review of arXiv:2607.05998}
}
read the original abstract

We present a unified finite element method for the dynamics of fluidic biomembranes. The model is governed by the Navier--Stokes equations in the bulk coupled to the surface Navier--Stokes equations on the evolving biomembrane surface, with bending forces arising from the Willmore energy. By allowing the bulk mesh velocity to be independent of the fluid velocity and permitting a free tangential surface velocity, we are able to derive a unified weak formulation of the coupled bulk-surface Navier--Stokes system. To address the bending force, we consider an evolution equation for the curvature and propose a surface arbitrary Lagrangian--Eulerian (ALE) weak formulation. Discretization with either fitted or unfitted finite elements leads to well-posed fully discrete linear schemes that are unconditionally energy stable. We present a variety of numerical examples to demonstrate the favourable properties of the proposed methods.

Figures

Figures reproduced from arXiv: 2607.05998 by the authors.

Figure 1
Figure 1. A schematic illustration of the two-phase flow in Ω [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Evolution for a smooth letter “C” in the case of [PITH_FULL_IMAGE:figures/full_fig_p025_2.png] view at source ↗
Figure 3
Figure 3. Evolution for a smooth letter “C” in the case of [PITH_FULL_IMAGE:figures/full_fig_p026_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: The time history of the discrete quantities [PITH_FULL_IMAGE:figures/full_fig_p027_4.png]
Figure 5
Figure 5. Figure 5: Several snapshots of the computational meshes. Up [PITH_FULL_IMAGE:figures/full_fig_p027_5.png]
Figure 6
Figure 6. Figure 6: The tank treading example in shearing flow for [PITH_FULL_IMAGE:figures/full_fig_p029_6.png]
Figure 7
Figure 7. Figure 7: The tumbling example in shearing flow for [PITH_FULL_IMAGE:figures/full_fig_p030_7.png]
Figure 8
Figure 8. Figure 8: Evolution for a smooth letter “C” in the case of [PITH_FULL_IMAGE:figures/full_fig_p033_8.png]
Figure 9
Figure 9. Figure 9: Evolution for a smooth letter “C” in the case of [PITH_FULL_IMAGE:figures/full_fig_p034_9.png]
Figure 10
Figure 10. Figure 10: Simulation of a vesicle flowing through a constric [PITH_FULL_IMAGE:figures/full_fig_p036_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 42 canonical work pages

  1. [1]

    Aland, S

    S. Aland, S. Egerer, J. Lowengrub, A. Voigt, Diffuse interface m odels of locally inextensible vesicles in a viscous fluid, J. Comput. Phys. 277 (2014) 32–47

    work page 2014

  2. [2]

    Arroyo, A

    M. Arroyo, A. DeSimone, Relaxation dynamics of fluid membranes, Phys. Rev. E 79 (2009) 031915

    work page 2009

  3. [3]

    Barrett, H

    J.W. Barrett, H. Garcke, R. N¨ urnberg, On the parametric finit e element approximation of evolving hypersurfaces in R3, J. Comput. Phys. 227 (2008) 4281–4307

    work page 2008

  4. [4]

    Barrett, H

    J.W. Barrett, H. Garcke, R. N¨ urnberg, Parametric approxima tion of Willmore flow and related geometric evolution equations, SIAM J. Sci. Comput. 31 (2008) 225 –253

    work page 2008

  5. [5]

    Barrett, H

    J.W. Barrett, H. Garcke, R. N¨ urnberg, A stable parametric fin ite element discretization of two-phase Navier–Stokes flow, J. Sci. Comput. 63 (2015) 78–117

    work page 2015

  6. [6]

    Barrett, H

    J.W. Barrett, H. Garcke, R. N¨ urnberg, Computational param etric Willmore flow with spontaneous curvature and area difference elasticity effects, SIAM J. Numer. A nal. 54 (2016) 1732–1762

    work page 2016

  7. [7]

    Barrett, H

    J.W. Barrett, H. Garcke, R. N¨ urnberg, A stable numerical met hod for the dynamics of fluidic membranes, Numer. Math. 134 (2016) 783–822. 37

    work page 2016

  8. [8]

    Barrett, H

    J.W. Barrett, H. Garcke, R. N¨ urnberg, Finite element approximation for the dynamics of asymmetric fluidic biomembranes, Math. Comput. 86 (2017) 1037–1069

    work page 2017

Show all 42 references
  1. [9]

    Barrett, H

    J.W. Barrett, H. Garcke, R. N¨ urnberg, Parametric finite eleme nt approximations of curvature driven interface evolutions, Handb. Numer. Anal. (Andrea Bonito and Rica rdo H. Nochetto, eds.) 21 (2020) 275–423

    2020

  2. [10]

    D. Boffi, N. Cavallini, F. Gardini, L. Gastaldi, Local mass conserva tion of Stokes finite elements, J. Sci. Comput. 52 (2012) 383–400

    2012

  3. [11]

    Bonito, R.H

    A. Bonito, R.H. Nochetto, M.S. Pauletti, Parametric FEM for geo metric biomembranes, J. Comput. Phys. 229 (2010) 3171–3188

    2010

  4. [12]

    Bonito, R.H

    A. Bonito, R.H. Nochetto, M.S. Pauletti, Dynamics of biomembran es: Effect of the bulk fluid, Math. Model. Nat. Phenom. 6 (2011) 25–43

    2011

  5. [13]

    Brezzi, M

    F. Brezzi, M. Fortin, Mixed and Hybrid Finite Element Methods, vo lume 15 of Springer Series in Computational Mathematics , Springer-Verlag, New York, 1991

    1991

  6. [14]

    Droske, M

    M. Droske, M. Rumpf, A level set formulation for Willmore flow, In terfaces Free Bound. 6 (2004) 361–378

    2004

  7. [15]

    Q. Du, C. Liu, X. Wang, A phase field approach in the numerical st udy of the elastic bending energy for vesicle membranes, J. Comput. Phys. 198 (2004) 450–468

    2004

  8. [16]

    B. Duan, B. Li, Z. Yang, An energy diminishing arbitrary Lagrang ian–Eulerian finite element method for two-phase Navier–Stokes flow, J. Comput. Phys. 461 (2022) 111215

    2022

  9. [17]

    Dziuk, Computational parametric Willmore flow, Numer

    G. Dziuk, Computational parametric Willmore flow, Numer. Math. 111 (2008) 55–80

    2008

  10. [18]

    Esedoglu, S.J

    S. Esedoglu, S.J. Ruuth, R. Tsai, Threshold dynamics for high or der geometric motions, Interfaces Free Bound. 10 (2008) 263–282

    2008

  11. [19]

    Franke, R.H

    T. Franke, R.H. Hoppe, C. Linsenmann, L. Schmid, C. Willbold, A. W ixforth, Numerical simulation of the motion of red blood cells and vesicles in microfluidic flows, Comput . Vis. Sci. 14 (2011) 167– 180

    2011

  12. [20]

    Franken, M

    M. Franken, M. Rumpf, B. Wirth, A phase field based PDE constr ained optimization approach to time discrete Willmore flow, Int. J. Numer. Anal. Model. 10 (2013) 116 –138

    2013

  13. [21]

    Garcke, R

    H. Garcke, R. N¨ urnberg, Q. Zhao, Structure-preserving d iscretizations of two-phase Navier-Stokes flow using fitted and unfitted approaches, J. Comput. Phys. 489 ( 2023) 112276

    2023

  14. [22]

    Garcke, R

    H. Garcke, R. N¨ urnberg, Q. Zhao, Unfitted finite element app roximations of the axisymmetric two- phase flow, J. Sci. Comput. 97 (2023) 14

    2023

  15. [23]

    Garcke, R

    H. Garcke, R. N¨ urnberg, Q. Zhao, Stable fully discrete finite e lement methods with BGN tangential motion for Willmore flow of planar curves, J. Sci. Comput. 105 (2025) 45

    2025

  16. [24]

    Garcke, R

    H. Garcke, R. N¨ urnberg, Q. Zhao, An energy-stable parame tric finite element method for Willmore flow with normal-tangential velocity splitting, SIAM J. Sci. Comput. 4 8 (2026) A1235–A1259

    2026

  17. [25]

    Garcke, R

    H. Garcke, R. N¨ urnberg, A parametric finite element method f or the incompressible Navier–Stokes equations on an evolving surface, arXiv:2508.19198, 2025

    2025

  18. [26]

    J. Hu, N. Lei, B. Li, R. Tang, Energy dissipating ALE-MDR method for Navier–Stokes free boundary problems with moving contact line, SIAM J. Sci. Comput. 48 (2026) A1 284–A1311. 38

    2026

  19. [27]

    J. Hu, B. Li, Evolving finite element methods with an artificial tang ential velocity for mean curvature flow and Willmore flow, Numer. Math. 152 (2022) 127–181

    2022

  20. [28]

    W.F. Hu, Y. Kim, M.C. Lai, An immersed boundary method for simulat ing the dynamics of three- dimensional axisymmetric vesicles in Navier–Stokes flows, J. Comput . Phys. 257 (2014) 670–686

    2014

  21. [29]

    Kemmochi, Y

    T. Kemmochi, Y. Miyatake, K. Sakakibara, Structure-preser ving numerical methods for constrained gradient flows of planar closed curves with explicit tangential velocit ies, Japan J. Ind. Appl. Math. 42 (2025) 575–603

    2025

  22. [30]

    Kov´ acs, B

    B. Kov´ acs, B. Li, C. Lubich, A convergent evolving finite elemen t algorithm for Willmore flow of closed surfaces, Numer. Math. 149 (2021) 595–643

    2021

  23. [31]

    Krause, A

    V. Krause, A. Voigt, A numerical approach for fluid deformable surfaces with conserved enclosed volume, J. Comput. Phys. 486 (2023) 112097

    2023

  24. [32]

    Kr¨ uger, M

    T. Kr¨ uger, M. Gross, D. Raabe, F. Varnik, Crossover from t umbling to tank-treading-like motion in dense simulated suspensions of red blood cells, Soft Matter 9 (201 3) 9008–9015

  25. [33]

    Laadhari, P

    A. Laadhari, P. Saramito, C. Misbah, Computing the dynamics of biomembranes by combining conservative level set and adaptive finite element methods, J. Com put. Phys. 263 (2014) 328–352

    2014

  26. [34]

    Masud, T.J

    A. Masud, T.J. Hughes, A space-time Galerkin/least-squares fi nite element formulation of the Navier-Stokes equations for moving domain problems, Comput. Met hods Appl. Mech. Engrg. 146 (1997) 91–126

    1997

  27. [35]

    Rodrigues, R.F

    D.S. Rodrigues, R.F. Ausas, F. Mut, G.C. Buscaglia, A semi-implicit fi nite element method for viscous lipid membranes, J. Comput. Phys. 298 (2015) 565–584

    2015

  28. [36]

    Rumpf, J

    M. Rumpf, J. Sassen, C. Smoch, A hybrid minimizing movement and neural network approach to Willmore flow, Interfaces Free Bound. (2026) to appear

    2026

  29. [37]

    Rusu, An algorithm for the elastic flow of surfaces, Interf aces Free Bound

    R.E. Rusu, An algorithm for the elastic flow of surfaces, Interf aces Free Bound. 7 (2005) 229–239

    2005

  30. [38]

    Salac, M.J

    D. Salac, M.J. Miksis, Reynolds number effects on lipid vesicles, J. F luid Mech. 711 (2012) 122–146

    2012

  31. [39]

    Seifert, Configurations of fluid membranes and vesicles, Adv

    U. Seifert, Configurations of fluid membranes and vesicles, Adv . Phys. 46 (1997) 13–137

    1997

  32. [40]

    Shi, T.W

    L. Shi, T.W. Pan, R. Glowinski, Three-dimensional numerical simula tion of red blood cell motion in Poiseuille flows, Internat. J. Numer. Methods Fluids 76 (2014) 397 –415

    2014

  33. [41]

    Willmore, Riemannian Geometry, Oxford University Press, 19 93

    T.J. Willmore, Riemannian Geometry, Oxford University Press, 19 93

  34. [42]

    Q. Zhao, W. Jiang, W. Bao, An energy-stable parametric finite e lement method for simulating solid-state dewetting, IMA J. Numer. Anal. 41 (2021) 2026–2055. 39

    2021

Pith tools

Reviewed July 8, 2026 · model on record in the stance chip above.