We propose a finite-element method to determine the dynamics of a fluid interacting with a membrane—a fluid layer equipped with a Helfrich elastic free energy [W. Helfrich, Z. Naturforschung C, 28, 693, (1973)]. Our analysis is based on a partitioned approach, where the state of the fluid, of the membrane, and of the mesh, are solved for at staggered steps. We show the predicting capabilities of the method by considering a macroscopic example of an elastomer membrane, which bulges as it interacts with air injected through a channel. We demonstrate that our method allows for describing large membrane deformations which are not accessible by perturbative methods. We discuss the limitations of our method, by characterizing its added-mass instability. Finally, we show how our analysis allows for understanding the physical mechanism of bulge formation, which is due to a combination of elastic stresses and membrane boundary conditions.
Introduction
Membranes interacting with a surrounding, bulk fluid are central to a wide range of physical systems. These include biological membranes in contact with the cell aqueous solution or actin network [1], cell cortices interacting with the cytoplasm [2], and soap films in air [3]. For instance, membrane-fluid interfaces in biology, such as the cell cortex or lipid bilayers, play a key role in cell motility [4], morphogenetic processes [2], and signaling [5]. Modeling such membrane-fluid interactions presents substantial challenges due not only to geometric nonlinearities [6], but also to the dynamical nature of the fluid spatial domain [7].
Here, we present a numerical method for the interaction between a fluid and a Helfrich membrane—a deformable fluid layer whose material elements can flow, are subject to tension, and whose elastic energy is given by the Helfrich model [8]. In the latter, the bending energy is proportional to the square of the mean curvature. Originally introduced for lipid bilayers [8], this model has become the standard description of the bending elasticity of biological membranes and vesicles [9], and it describes physical objects across different scales, e.g., thin elastic plates [10].
State of the art
A number of existing studies rely on the hypothesis that the fluid surrounding the Helfrich membrane is inertialess. For instance, Arroyo et al. study the interaction between a Helfrich membrane and bulk fluid in the Stokes approximation, for simple membrane geometries that can be described by a single parameter [11]. Boedec and coauthors combined finite element (FE) and boundary element (BE) methods describe an Helfrich membrane interacting with an inertialess fluid [12]. In there, the membrane surface rheology, such as viscous effects and flows induced by surface-tension gradients, is neglected. A Helfrich membrane coupled to a fluid has been recently studied by Zhou and co-authors with BE and FE methods, under the assumption that the fluid is inertialess and that Helfrich membrane is inextensible [13]. Barakat and Shaqfeh modelled vesicles in a circular tube by means of BE methods and lubrication theory, by neglecting Helfrich membrane rheology and describing fluid as inertialess [14]. Finally, Rodrigues et al., studied the interaction between an inertialess, closed Helfrich membranes and fluids with FE methods, by adopting a minimal description for the fluid, which is solely modelled as a volume constraint [15].
Other studies go beyond the inertialess approximation, and rely on simplifying assumptions for the Helfrich membrane. For instance, Bonito et al. propose a FE model for a Helfrich membrane interacting an fluid, the latter being in the inertial regime [16, 17]. In this study the membrane is treated as a purely elastic body, characterized solely by bending rigidity, area and volume constraints—its fluid features, such as tangential and normal flows, are neglected. Laadhari extended the analyses above to the interaction between a Helfrich membrane and a non-Newtonian fluid, in which membrane rheology is still neglected [18]. The interaction between a Helfrich membrane and a fluid—both in the inertial regime—has been studied by Barrett and co-authors, [19], under the assumption that the membrane is perfectly inextensible. Such analysis is based on a fluidic approach, in which the membrane does not obey its own, intrinsic rheology—see below—but it is advected by the surrounding fluid. As a result, the membrane tangential and normal velocities are not determined as the result of the surfacial Navier-Stokes (NS) equations, but they match the respective components of the fluid velocity.
In this study, we propose a novel FE study of the interaction between a fluid and a Helfrich membrane. Our approach is based on the arbitrary Lagrangian-Eulerian (ALE) method [7, 20, 21]. Our analysis includes a complete description of the physical features of both the membrane and the fluid, it can describe inertial regimes of both, and it accounts for a detailed description of membrane elasticity, rheology, and extensibility. Unlike fluidic approaches where the membrane dynamics is determined by the fluid, here both the flow and shape of the membrane are determined by intrinsic physical evolution equations. In addition, our study can account for both closed and open surfaces, by leveraging the flexibility of its FE formulation [22].
We solve for the system dynamics by implementing a partitioned solution-scheme [23], applied to the membrane, the fluid, and the fluid domain. As for the membrane shape, we generalize the arc length (AL) gauge—previously proposed to describe the steady state of membrane tubes [24]—to the system dynamics, by means of a time-dependent gauge fixing. Overall, this implementation allows for handling strongly turbulent membrane and fluid flows, large membrane deformations, and membrane-shape overhangs [6].
The paper is structured as follows. In Section II we discuss the ALE method, and present two illustrative examples: The interaction between a fluid and a rigid body, Section II A, and that between the fluid and the elastic body, Section II B. In Section II C we present our approach for the interaction between a fluid and a Helfrich membrane. In particular, in Sections II C 1 to II C 3, we discuss the splitting scheme and gauge-fixing procedure for the membrane, fluid, and fluid domain, and the temporal discretization is detailed in Section II C 4. Finally, Section III is devoted to the discussion of the results, and future directions. All used acronyms are listed in Section IV.
Results
Following the general ALE framework [7], for all systems that we will consider we introduce the reference (r) and current (c) configuration [10, 25, 26], see Fig. A.1. All quantities relative to the r and c configuration will be denoted by the superscript r and c, respectively. In the r and c configuration, the region occupied by the fluid is described by Cartesian coordinates xr,xc, respectively.
The r and c configurations are related as follows [25], see Fig. A.1. At time t, the point xc corresponds to a point xr in the r configuration through the deformation field [27]
uDt(xr)≡xc−xr.(1)
The deformation-gradient tensor is defined as [10]
Fαβ(u)≡δαβ+∂xrβ∂uα.(2)
In what follows, Greek indices α,β,⋯ will be used to denote vectors and tensors in two-dimensional Euclidean space—their position as upper or lower indices is thus immaterial [10, 28]. We will characterize the mapping between xc and xr configurations in Eq. (1) through the fields
xc=φt(xr),(3)
xr=ψt(xc).(4)
We denote the c fluid velocity and surface tension, i.e., the negative of the pressure, by vFc(xc) and sFc(xc), respectively. For each field vFc(xc),sFc(xc), we introduce its counterpart in the r configuration:
vFr(xr,t)≡vFc(φt(xr),t),(5)
sFr(xr,t)≡sFc(φt(xr),t).(6)
The solution of the interaction dynamics between the fluid and the rigid body, elastic body or membrane, is involved because the fluid domain has a dynamical, moving boundary, cf. Figs. A.1, B.1 and 3, respectively. Following the ALE procedure, we will bridge between the Lagrangian description used to describe the rigid body, elastic body or membrane, and the Eulerian description of the fluid [7, 25]. To achieve this, we will consider the equations of motion for the rigid body, elastic body or membrane and the fluid—which are naturally formulated the c configuration—and rewrite them in the r configuration [25].
We will first discuss this procedure in the illustrative examples of interaction between a fluid and a rigid body and that between a fluid and an elastic body, and then consider a fluid interacting with a Helfrich membrane.
Fluid and rigid body
In this Section, we will discuss the interaction between a fluid and the simplest structure—a rigid body which is only allowed to turn about one point. Only the main results will be presented here; details can be found in Sections A1 to A3, respectively, and the temporal discretization is discussed in Section A 4.
As shown in Fig. A.1, in the r configuration the axis of the ellipse, i.e., the body, lies parallel to the xc1 axis, while in the c configuration, such axis forms an angle θ (red arc) with respect to such axis.
An example of the system dynamics is shown in Figure 1, where we study the interaction between air (the fluid) and an aerogel (the rigid body)—an ultra light solid material involving a large porous structure [29]. The channel dimension are L=2.2m, h=0.41m, which have been taken from the Finite Element Analysis Tools 2D (FEAT2D) DFG 2D-3 benchmark example for a fluid flow around a cylinder [30, 31], see Fig. A.1. Given that the materials, such as the fluid and the body, are two dimensional, in what follows we will estimate their parameter values by considering their three-dimensional counterparts, and assume a unit thickness in the out-of-plane direction. The fluid parameters have been estimated from the physical properties of air: ρF=1kg/m2, ηF=10−5Pams[32]. The inflow velocity profile is chosen as a Poiseuille flow
Figure 1. Interaction between a fluid (F) and a rigid body. The fluid material properties roughly correspond to air, and those of the body to an aerogel. A) Temporal snapshot of the mesh (black lines) and fluid velocity field (arrows) at the current time t, shown on top. The direction of the velocity field is indicated by the arrows, and its norm by their color; the direction of arrows with velocity close to zero is not defined. The body (ellipse) is allowed to pivot about its left focal point (green dot); the related angle with respect to the xc1 direction is denoted by a black dashed arc. The Reynolds number is R∼2×102. B) Color map of the surface tension at the current time t.
vF(xc)=h26v0xc2(xc2−h),(7)
with v0=1m/s. The mesh of the fluid domain deforms in time, and its rigidity is determined by the stiffness exponent ζ[25], which has been chosen to be ζ=3, see Section II C 2 for details. Finally, the rigid body is an ellipse with semi-axes a=0.1m, b=0.075m, center located at c=(0.25m,0.2m), and moment of inertia I=10−3kgm2, which corresponds to an aerogel density of ρB=6kg/m3[33]. In Section A 5 we discuss a consistency check of the ALE scheme above, showing that the oscillation period of the body matches the prediction obtained with scaling arguments—see Section A 5.
Fluid and elastic body
In this Section, we will build on the analysis of Section II A and put the fluid into interaction with a more complex physical object—an elastic body; only the main results will be presented here.
The r and c configurations are shown in Fig. B.1. We allow the elastic-body boundary ∂Ω⊙ to deform, but the inner body boundary ∂Ω∘ is pinned to the xc1,xc2 plane. The equations of motion are reported in Section B 1, and the system dynamics is discussed in Section B 2.
An example of such dynamics is depicted in Figure 2, where we show the interaction between a fluid and an hydrogel (the elastic body). Here, ∂Ω⊙r is an ellipse with semi-axes a=0.2m, b=0.1m and center located at c=(0.4m,0.2m), and ∂Ω⊙r has radius r=1.25×10−2m and center located at c. Channel dimensions are the same as in Figure 1, and fluid parameters have been taken from the FEAT2D DFG 2D-3 benchmark example for a fluid flow around a cylinder [30, 31]: ρF=1kg/m2, ηF=10−3Pams. The body density is ρF=103kg/m2[34]. The storage modulus, which approximates the shear modulus μ[35] is ∼10kg/s2[36]. Given an elastic-body Poisson ratio [10] of νP∼0.49[37], we obtain K∼5×102kg/s2 from the relation K=2μ(1+νP)/(3(2νP−1))[10]. The fluid inflow velocity profile is a Poiseuille flow, Eq. (7), with v0=1m/s. Finally, the fluid-domain mesh stiffness exponent has been chosen to be ζ=3[25].
Figure 2. Interaction between a fluid (F) and an elastic body (a hydrogel). The notation is the same as in Figure 1, and the body is depicted as a red mesh. The elastic-body deformation is not symmetric under the mirror symmetry xc2→h−xc2, where h is the height of the fluid domain, because the center of the body is not located at xc2=h/2. The Reynolds number is R∼2×102.
Fluid and Helfrich membrane
In this Section, we will build on the analysis of Section II B, and complexify the structure of the body with which the fluid interacts. We consider, instead of an elastic material, a Helfrich membrane. The latter is a deformable fluid layer whose elements can flow, equipped with an elastic, bending energy given by the Helfrich model [8, 24].
We will consider the geometry depicted in Figure 3. A vertical channel with mirror symmetry about its mid axis is considered. The left boundary of the channel is ∂ΩL, and the right boundary ∂ΩR lies on the symmetry axis. The fluid is injected on ∂ΩI, from bottom to top. The membrane lies on the top boundary of the fluid, ∂Ω−, which coincides with the membrane manifold ∂Ω−. Finally, we denote by Ω the region enclosed by the boundaries, and all the quantities above will be considered in both the r and c configuration.
Figure 3. Reference (r) and current (c) configuration for a fluid and a Helfrich membrane. The Figure shows a vertical channel with mirror symmetry about its mid axis (red dashed line), see Fig. A.1. The right, mirrored part of the channel is not shown, cf. Figure 4A. A)r configuration. The top edge of the fluid domain, ∂Ω□r, coincides with the one-dimensional manifold Ω□r of the membrane. We also show the other boundaries of the fluid domain, ∂Ω■r, ∂Ω■r, ∂Ω■r, the bottom-left point of the fluid domain ∂Ω■,■r, and the membrane boundaries, ∂Ω■r and ∂Ω■r. The r unit normal n^r is also shown; its plotted length has been adjusted for clarity. B) Same as A, for the c configuration. The membrane profile is purposely, strongly deformed without remeshing, so as to illustrate mesh deformation.
The geometry of Figure 3 may represent the fluid being confined in a reservoir, lying below the channel, where the channel is a vertical neck of the reservoir, covered on top by the membrane.
Membrane motion
Both the physical nature and mathematical description of a Helfrich membrane are significantly more complex than that of an elastic body. First, unlike an elastic body, the membrane has a fluid structure: Its material elements can flow tangentially to the membrane shape, which also moves subjected to bending-rigidity forces, and to the force exerted on it by the fluid. While the tangential motion is naturally suited to an Eulerian description [27], the shape motion—reminiscent of that of an elastic body [8]—is best described with a Lagrangian approach [38]. Also, such shape motion needs to be connected to the Lagrangian approach used to describe the fluid domain. Second, the membrane is described by fourth-order partial differential equations (PDEs) [39]. From the mathematical and numerical point of view, such PDEs are significantly more complex than the second-order PDEs which describe an elastic body [6]. On top of this, from the physical standpoint, such fourth-order PDEs require more boundary conditions (BCs) than the second-order ones for an elastic body: Such BCs may specify a number of physical features of the membrane system, such as forces, positions, torques, or stresses. Third, the geometrical description of the membrane requires fixing an appropriate gauge. In fact, a general parametrization of the membrane shape is given by two functions—the two Cartesian components of the membrane shape xcM(x1). Both of these functions depend on the curvilinear coordinate x1 which is defined in the r domain of the membrane, ∂Ω−r, see Figure 3—note that here xi differs from the Cartesian coordinate xα. Despite this pair of functions, a single PDE determines the membrane shape [39]. To understand this redundancy, consider the stretching field ν>0, defined by
ν2≡e1⋅e1,(8)
ei≡∂xi∂xcM.(9)
Two parametrizations with different νs can yield the same physical shape. Such redundancy may be removed by a gauge fixing, i.e., by adding a constraint on xcM, or its derivatives. Two common gauge choices are the Monge [40, 41], or the arc-length parametrization [24], in which one sets
ν=1.(10)
The arc-length gauge has been used to describe, for instance, the steady state for membrane tubules [24]. However, while a static constraint such as Eq. (10) is suited to static shapes, it cannot describe the membrane dynamics, such as the one of a membrane interacting with a fluid. In fact, as the membrane is deformed, it will also stretch, and Eq. (10) must be replaced by a time-dependent gauge—see below.
We describe the membrane with the Lagrangian approach, along the lines of the description of the fluid domain in Section II A and an elastic body in Section II B. The membrane displacement field is
uM(x1)≡xcM(x1)−xrM(x1).(11)
At the beginning of the dynamics, we choose the r configuration as a straight line
xrM(x1)≡(x1,h),0≤x1≤L,(12)
see Figure 3A. The mesh will then be deformed with time: As the deformation becomes large, we will remesh the fluid domain. The c configuration right before remeshing will then become the r configuration after remeshing, which will be, in general, different from (12).
In addition, we introduce the angle ψ formed by the tangent to the membrane with the x1 axis [24], defined by
References
[1]P. Bassereau, R. Jin, T. Baumgart, M. Deserno, R. Di-mova, V. A. Frolov, P. V. Bashkirov, H. Grubmüller, R. Jahn, H. J. Risselada, L. Johannes, M. M. Kozlov, R. Lipowsky, T. J. Pucadyil, W. F. Zeno, J. C. Sta-chowiak, D. Stamou, A. Breuer, L. Lauritsen, C. Simon, C. Sykes, G. A. Voth, and T. R. Weikl, The 2018 biomem-brane curvature and remodeling roadmap, J. Phys. Appl. Phys. 51, 343001 (2018).DOI
[2]G. Salbreux, G. Charras, and E. Paluch, Actin cortex mechanics and cellular morphogenesis, Trends Cell Biol. 22, 536 (2012).
[3]R. J. Pugh, Soap bubbles and thin films, in Bubble and Foam Chemistry (Cambridge University Press, Cambridge, 2016) pp. 84–111.DOI
[4]P. Sens and J. Plastino, Membrane tension and cytoskeleton organization in cell motility, J. Phys. Condens. Matter 27, 273103 (2015).DOI
[5]A. Diz-Muñoz, D. A. Fletcher, and O. D. Weiner, Use the force: Membrane tension as an organizer of cell shape and motility, Trends Cell Biol. 23, 47 (2013). [6] D. Wörthmüller, G. Ferraro, P. Sens, and M. Castellana, IRENE: A fluid layer Finite-Element software, Comput. Phys. Commun. 330, 110366 (2027).
[7]J. Donea, A. Huerta, and A. Rodriguez-Ferran, Chapter 14 Arbitrary Lagrangian–Eulerian Methods, in The Encyclopedia of Computational Mechanics, Vol. 1 (2004) pp. 413–437.
[8]W. Helfrich, Elastic Properties of Lipid Bilayers: Theory and Possible Experiments, Z. Naturforschung C 28, 693 (1973).pubmed.ncbi.nlm.nih.gov/4273690
[9]U. Seifert, Configurations of fluid membranes and vesicles, Adv. Phys. 46, 13 (1997).DOI
[10]L. D. Landau and E. M. Lifshitz, Theory of Elasticity (Elsevier, 1986).
[11]M. Arroyo and A. DeSimone, Relaxation dynamics of fluid membranes, Phys. Rev. E Stat. Phys. Plasmas Fluids Relat. Interdiscip. Top. 79, 031915 (2009).pubmed.ncbi.nlm.nih.gov/19391979
[12]G. Boedec, M. Leonetti, and M. Jaeger, Isogeometric FEM-BEM simulations of drop, capsule and vesicle dynamics in Stokes flow, J. Comput. Phys. 342, 117 (2017).DOI
[13]H. Zhou, Y.-N. Young, and Y. Mori, Modeling and Simulation of Open Membranes in Stokes Flow with Mixed-Dimensional Coupling (2026), arXiv:2504.16823 [math].DOI
[14]J. M. Barakat and E. S. G. Shaqfeh, Stokes flow of vesicles in a circular tube, J. Fluid Mech. 851, 606 (2018).DOI
[15]D. S. Rodrigues, R. F. Ausas, F. Mut, and G. C. Buscaglia, A semi-implicit finite element method for viscous lipid membranes, J. Comput. Phys. 298, 565 (2015).DOI
[16]A. Bonito, R. Nochetto, and M. Pauletti, Dynamics of Biomembranes: Effect of the Bulk Fluid, Math. Model. Nat. Phenom. 6, 25 (2011).DOI
[17]A. Bonito, R. H. Nochetto, and M. Sebastian Pauletti, Parametric FEM for geometric biomembranes, J. Comput. Phys. 229, 3171 (2010).DOI
[18]A. Laadhari, Finite-Element Method for the Simulation of Lipid Vesicle/Fluid Interactions in a Quasi-Newtonian Fluid Flow, Mathematics 11, 1950 (2023).DOI
[19]J. W. Barrett, H. Garcke, and R. Nürnberg, Finite element approximation for the dynamics of axisymmetric fluid biomembranes, Math. Comput. 86, 1037 (2016).
[20]L. E. Malvern, Introduction to the Mechanics of a Continuous Medium (Prentice Hall, Englewood Cliffs, NJ, 1969).
[21]P. Howell, G. Kozyreff, and J. Ockendon, Applied Solid Mechanics (Cambridge University Press, 2008).
[22]A. Logg, K.-A. Mardal, and G. Wells, eds., Automated Solution of Differential Equations by the Finite Element Method: The FEniCS Book, Lecture Notes in Computational Science and Engineering, Vol. 84 (Springer Berlin Heidelberg, Berlin, Heidelberg, 2012).DOI
[23]C. A. Felippa, K. Park, and C. Farhat, Partitioned analysis of coupled mechanical systems, Comput. Methods Appl. Mech. Eng. 190, 3247 (2001).DOI
[24]I. Derényi, F. Jülicher, and J. Prost, Formation and Interaction of Membrane Tubes, Phys. Rev. Lett. 88, 238101 (2002).DOI
[25]D. Kamensky, Lecture notes for MAE 207: Finite element analysis for coupled problems (2022).
[26]W. S. Slaughter, The Linearized Theory of Elasticity (Birkhäuser Boston, Boston, MA, 2002).DOI
[27]L. D. Landau and E. M. Lifshitz, Fluid Mechanics (Pergamon, 1987).
[28]S. Marchiafava, Appunti di Geometria Differenziale, Vol. I, II, III (Edizioni Nuova Cultura, 2005).
[29]S. S. KISTL, Coherent Expanded Aerogels and Jellies, Nature 127, 741 (1931).
[30]H. Blum, J. Harig, S. Müller, S. Schreiber, and S. Turek, FEAT2D, Finite Element Analysis Tools User Manual (1995).
[31]M. Schafer and S. Turek, Benchmark Computations of Laminar Flow Around a Cylinder (1996).DOI
[32]J. Hilsenrath, Tables of Thermal Properties of Gases: Comprising Tables of Thermodynamic and Transport Properties of Air, Argon, Carbon Dioxide, Carbon Monoxide, Hydrogen, Nitrogen, Oxygen, and Steam, Vol. 564 (US Department of Commerce, National Bureau of Standards, 1955).
[33]A.-G. Niculescu, D.-I. Tudorache, M. Bocioacă, D. E. Mihaiescu, T. Hadaribata, and A. M. Grumezescu, An Updated Overview of Silica Aerogel-Based Nanomaterials, Nanomaterials 14, 469 (2024).DOI
[34]S. Xu, Z. Zhou, Z. Liu, and P. Sharma, Concurrent stiffening and softening in hydrogels under dehydration, Sci. Adv. 9, eade3240 (2023).DOI
[35]G. Rehaege, Elastic properties of crosslinked polymers, in Macromolecular Chemistry–9, edited by L. Cross (Butterworth-Heinemann, 1974) pp. 161–178.
[36]C. Licht, J. C. Rose, A. O. Anarkali, D. Blondel, M. Roccio, T. Haraszti, D. B. Gehlen, J. A. Hubbell, M. P. Lutolf, and L. De Laporte, Synthetic 3D PEG-Anisogel Tailored with Fibronectin Fragments Induce Aligned Nerve Extension, Biomacromolecules 20, 4075 (2019).
[37]U. Chippada, B. Yurke, and N. A. Langrana, Simultaneous determination of Young’s modulus, shear modulus, and Poisson’s ratio of soft hydrogels, J. Mater. Res. 25, 545 (2010).DOI
[39]O.-Y. Zhong-can and W. Helfrich, Bending energy of vesicle membranes: General expressions for the first, second, and third variation of the shape energy and applications to spheres and cylinders, Phys. Rev. A 39, 5280 (1989).pubmed.ncbi.nlm.nih.gov/9901091
[40]C.-C. Hsiung, A First Course in Differential Geometry (John Wiley & Sons, New York, USA, 1981).
[41]M. Deserno, Fluid lipid membranes: From differential geometry to curvature stresses, Chem. Phys. Lipids 185, 11 (2015).DOI
[42]S. C. Al-Izzi, P. Sens, and M. S. Turner, Shear-Driven Instabilities of Membrane Tubes and Dynamin-Induced Scission, Phys. Rev. Lett. 125, 018101 (2020).DOI
[43]M. Deserno, Notes on Differential Geometry, https://www.cmu.edu/biolphys/deserno/pdf/diff_geom.pdf (2004).
[44]P.-G. De Gennes, F. Brochard-Wyart, and D. Quéré, Capillarity and Wetting Phenomena (Springer New York, New York, NY, 2004).DOI
[45]J. E. Marsden and T. J. R. Hughes, Mathematical Foundations of Elasticity (Dover, New York, 1994).
[46]J. Crank and P. Nicolson, A practical method for numerical evaluation of solutions of partial differential equations of the heat-conduction type, Math. Proc. Camb. Philos. Soc. 43, 50 (1947).DOI
[47]A. J. Chorin, Numerical solution of the Navier-Stokes equations, Math. Comput. 22, 745 (1968).DOI
[48]L. Formaggia and F. Nobile, A stability analysis for the Arbitrary Lagrangian Eulerian formulation with finite elements, East-West J. Numer. Math. 7, 105 (1999).
[49]D. Boffi and L. Gastaldi, Stability and geometric conservation laws for ALE formulations, Comput. Methods Appl. Mech. Eng. 193, 4717 (2004).DOI
[50]J. San Martín, L. Smaranda, and T. Takahashi, Convergence of a finite element/ALE method for the Stokes equations in a domain depending on time, J. Comput. Appl. Math. 230, 521 (2009).DOI
[51]H. Reall, General Relativity, https://www.damtp.cam.ac.uk/user/hsr1000/teaching.html (2022).
[52]L. Feng, S. Li, and S. Feng, Preparation and characterization of silicone rubber with high modulus via tension spring-type crosslinking, RSC Adv. 7, 13130 (2017).DOI
[53]A. Müller, M. C. Wapler, and U. Wallrabe, A quick and accurate method to determine the Poisson’s ratio and the coefficient of thermal expansion of PDMS, Soft Matter 15, 779 (2019).DOI
[54]L.-H. Cai, T. E. Kodger, R. E. Guerra, A. F. Pegoraro, M. Rubinstein, and D. A. Weitz, Soft Poly(dimethylsiloxane) Elastomers from Architecture-Driven Entanglement Free Design, Adv. Mater. 27, 5132 (2015).DOI
[55]K. V. Pochivalov, A. N. Shilov, T. N. Lebedeva, A. N. Ilyasova, R. Y. Golovanov, A. V. Basko, and Y. V. Kudryavtsev, Development of vibration damping materials based on butyl rubber: A study of the phase equilibrium, rheological, and dynamic properties of compositions, J. Appl. Polym. Sci. 138, 50196 (2021).
[56]C. Förster, W. A. Wall, and E. Ramm, Artificial added mass instabilities in sequential staggered coupling of nonlinear structures and incompressible viscous flows, Comput. Methods Appl. Mech. Eng. 196, 1278 (2007).DOI
[57]P. Causin, J. Gerbeau, and F. Nobile, Added-mass effect in the design of partitioned algorithms for fluid–structure problems, Comput. Methods Appl. Mech. Eng. 194, 4506 (2005).DOI
[58]M. Heil, An efficient solver for the fully coupled solution of large-displacement fluid–structure interaction problems, Comput. Methods Appl. Mech. Eng. 193, 1 (2004).DOI
[59]C. Simon, R. Kusters, V. Caorsi, A. Allard, M. Abou-Ghali, J. Manzi, A. Di Cicco, D. Lévy, M. Lenz, J.-F. Joanny, C. Campillo, J. Plastino, P. Sens, and C. Sykes, Actin dynamics drive cell-like membrane deformation, Nat. Phys. 15, 602 (2019).
[60]M. P. do Carmo, Differential geometry of curves and surfaces (Prentice-Hall, Englewood Cliffs, N.J., 1976).
[61]R. Vasan, S. Rudraraju, M. Akamatsu, K. Garikipati, and P. Rangamani, A mechanical model reveals that non-axisymmetric buckling lowers the energy barrier associated with membrane neck constriction, Soft Matter 16, 784 (2020).DOI
[62]H. Goldstein, C. P. Poole, and J. L. Safko, Classical Mechanics (Pearson, 2004).
[63]K. Goda, A multistep technique with implicit difference schemes for calculating two- or three-dimensional cavity flows, J. Comput. Phys. 30, 76 (1979).DOI
[64]E. Hairer, S. P. Nørsett, and G. Wanner, Solving Ordinary Differential Equations I, Springer Series in Computational Mathematics, Vol. 8 (Springer Berlin Heidelberg, Berlin, Heidelberg, 1993).
[65]Y. Bazilevs, V. M. Calo, T. J. R. Hughes, and Y. Zhang, Isogeometric fluid-structure interaction: Theory, algorithms, and computations, Comput. Mech. 43, 3 (2008).DOI