Active Particles Destabilize Passive Membranes

David A. King, Thomas P. Russell, Ahmad K. Omar · 2026

Read Active Particles Destabilize Passive Membranes by David A. King, Thomas P. Russell, Ahmad K. Omar (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-16430.txt.

Introduction

David King, Thomas Russell and Ahmad Omar show how live grains shake a soft sheet. Their 2026 preprint deals with a sheet that bends with ease. Live grains swim or crawl and push on the sheet. The team solves for the push those grains place on it. That push turns out to cut the pull tight of the sheet.

It also cuts the bend stiff of the sheet as a whole. New far reach terms join spots on the sheet through the grains. Such terms have no match in dead sheet maths. Soft sheets in cells may act by these same rules. The work gives a first theory for the pair.

Catalog record

CREATORDavid A. King, Thomas P. Russell, Ahmad K. Omar
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE37 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-16430

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Active Particles Destabilize Passive Membranes David A. King, Thomas P. Russell, Ahmad K. Omar Abstract We present a theory for the interaction between active particles and a passive flexible membrane. By explicitly solving for the pressure exerted by the active particles, we show that they reduce the membrane tension and bending modulus and introduce novel non-local contributions to the membrane mechanics. This theory predicts activity-induced instabilities and their morphology are in agreement with recent experimental and simulation data. Cells are intrinsically active systems, maintained far from equilibrium by continuous energy consumption. Two hallmark behaviors, reproduction by mitosis and feeding by phagocytosis, require major morphological manipulations mediated by active cytoskeletal forces on the cell membrane May and Machesky 2001 ; Pollard and Cooper 2009 ; Rottner and Schaks 2019 ; Welf et al. 2020 . Understanding how localized active stresses induce large-scale deformations in flexible membranes is key to unraveling these processes. Recently, synthetic model systems, such as vesicles loaded with bacteria Takatori and Sahu 2020 , Janus colloids Vutukuri et al. 2020 or magnetic rollers Kim et al. 2025 , have allowed the interplay between active mechanics and membrane response to be examined. Through interactions with the active particles, the vesicles can exhibit non-equilibrium fluctuation spectra Takatori and Sahu 2020 and extreme shape instabilities Vutukuri et al. 2020 , including division Kim et al. 2025 . In this Letter, we present a theoretical framework to describe these instabilities. As a canonical example, we consider spherical active Brownian particles (ABPs) with radius b confined to one side of an infinite membrane. This geometry is appropriate for exceedingly large vesicles and enables analytical progress while capturing the essential physics. By explicitly solving for the pressure exerted by ABPs on the fluctuating membrane, we show that their influence can be captured by renormalizing the membrane’s physical properties: they introduce non-local interactions and can reduce both the surface tension and the bending modulus. Crucially, when these effective parameters become negative, a membrane stable in equilibrium can undergo an activity-induced instability , the qualitative thresholds and morphology of which agree well with recent simulations and experiments. We begin with the dynamics of a passive, impermeable membrane interacting with a suspension, active or passive Sahu et al. 2017 ; Sahu et al. 2020 . The membrane is immersed in a d -dimensional solvent and is flat at rest (a line in d=2 or a plane in d=3 with no spontaneous curvature). The normal coordinate to the flat membrane is z , and \mathbf{x} denotes the (d-1) coordinates along it. Deviations from flat are described by the height function h(\mathbf{x},t) Kamien 2002 . The suspension occupies z\geq h(\mathbf{x},t) and exerts a pressure on the membrane that depends on its internal mechanics and how the two are coupled. Our goal is to understand how this interaction affects the membrane’s dynamics in the overdamped (no inertia) and linear (small deviation {|\bm{\nabla}h|\ll 1} ) limit, governed by Granek 1997 : \partial_{t}\tilde{h}(\mathbf{q},t)=\tilde{M}(\mathbf{q})\left[\tilde{\xi}(\mathbf{q},t)-\delta\mathcal{H}/\delta\tilde{h}(\mathbf{q},t)-\tilde{P}(\mathbf{q},t)\right]. (1) Here, {\tilde{h}(\textbf{q})=\int d\mathbf{x}e^{i\mathbf{q}\cdot\mathbf{x}}h(\mathbf{x})} , denotes the Fourier transform (FT) in \mathbf{x} and introduces the wave vector \mathbf{q} . The left-hand side is the FT of the membrane’s local z -velocity. Energy dissipation as the membrane moves through the solvent gives rise to the mobility, \tilde{M}(\mathbf{q}) , which is the FT of the zz -component of the Oseen tensor if hydrodynamic interactions between the suspension and membrane are ignored Happel and Brenner 1983 ; Kim and Karrila 2005 ; Doi and Edwards 1986 ; Note1 . The bracketed terms on the right-hand side are the non-dissipative, \hat{\mathbf{z}} -directed forces per unit (d-1) -dimensional area on the membrane: The first, \tilde{\xi} , represents thermal fluctuations as a Gaussian, zero-mean, temporally white noise satisfying the fluctuation-dissipation theorem. The second encodes the membrane’s internal mechanics through the standard Helfrich Hamiltonian {\mathcal{H}=\int d\mathbf{q}(\gamma q^{2}+\kappa q^{4})\tilde{h}^{2}/2} Helfrich 1973 ; Canham 1970 , where {q=|\mathbf{q}|} , \gamma is the surface tension and \kappa is the bending modulus. The final term is the pressure exerted by the suspension (in FT) which, if the particle-membrane interactions are short-ranged contact forces, is: {P(\mathbf{x},t)=-\hat{\mathbf{n}}\cdot\bm{\sigma}(\mathbf{x},z=h(\mathbf{x},t))\cdot\hat{\mathbf{n}}} Doi and Edwards 1986 , where \bm{\sigma} is the suspension stress and {\hat{\mathbf{n}}=(\hat{\mathbf{z}}-\bm{\nabla}h)/(1+|\bm{\nabla}h|^{2})^{1/2}} is the membrane normal Kamien 2002 ; Note2 . Takatori and Sahu used Eq. ( 1 ) to study non-equilibrium fluctuations of vesicles loaded with bacteria Takatori and Sahu 2020 . However, instead of deriving P from the stress tensor, they modeled it as an athermal stochastic variable caused by bacteria-membrane collisions. This captured the experimentally observed fluctuation spectrum scaling \sim q^{-4} as q\to 0 (the equilibrium scaling is q^{-2} Fisher 2004 ) but cannot predict activity induced instabilities of otherwise stable membranes, because P was assumed to be independent of h . From Eq. ( 1 ), this implies h always fluctuates around the stable minimum, h=0 . The pressure from a passive ideal gas is independent of h , but this is generally not the case Kamien 2007 ; Dinsmore et al. 1996 ; Kaplan et al. 1994 ; Blokhuis 2013 ; Urrutia 2014 . ABPs, ideal or interacting, are known to accumulate, and exert more pressure, on valleys in a boundary than its hills Wensink and Löwen 2008 ; Elgeti and Gompper 2013 ; Lee 2013 ; Ezhilan et al. 2015 ; Yan and Brady 2015a ; Yan and Brady 2018 ; Duzgun and Selinger 2018 ; Granek et al. 2024 . Intuitively, this can lead to an instability for flexible membranes: the excess pressure grows the valleys, drawing in more active particles to grow them further. This mechanism has qualitatively described instabilities in flexible filaments Nikola et al. 2016 ; Granek et al. 2024 . Here, we develop a quantitative theory to understand activity induced membrane instabilities. The key quantity is the suspension stress at the membrane surface. For a general overdamped suspension, this is found from the mechanical force balance: {\bm{\nabla}\cdot\bm{\sigma}+\mathbf{b}=\bm{0}} , where \mathbf{b} are the body forces. We consider a system of ABPs acted on by: a Brownian force resulting in translational diffusion with constant D_{T}=k_{B}T/\zeta , where \zeta is the translational drag coefficient associated with the solvent drag force, a fixed propulsion force \zeta v_{0} , where v_{0} is the “swim speed”, in a direction \hat{\mathbf{u}} that undergoes rotational diffusion with re-orientation time {\tau_{r}=D_{r}^{-1}} , and conservative, pairwise, repulsive inter-particle forces. Mechanical balance in this case gives Omar et al. 2023 \bm{\nabla}\cdot\bm{\sigma}+\mathbf{F}-\zeta\mathbf{J}_{\rho}=\bm{0}, (2) where \mathbf{J}_{\rho} is the number density current. The non-conservative active and drag force densities appear as the body forces, {\mathbf{F}=\zeta v_{0}\mathbf{m}} and \zeta\mathbf{J}_{\rho} respectively, where \mathbf{m} is the polarization (the local average of \hat{\mathbf{u}} ). The Brownian (ideal-gas) and conservative inter-particle forces contribute to \bm{\sigma} Yan and Brady 2015b ; Epstein et al. 2019 ; Omar et al. 2020 . The force exerted by the membrane enforces no-flux at its surface. Generally, \bm{\sigma} includes contributions from density gradients. However, when \ell_{0}/b\gtrsim 1 where {\ell_{0}=v_{0}\tau_{r}} is the ABP “run length”, these are negligible in comparison to similar terms appearing in F Omar et al. 2023 , allowing \bm{\sigma} to be approximated by the “conservative pressure”: {\bm{\sigma}\approx-P_{\rm c}(\rho)\mathbf{I}} , where \mathbf{I} is the identity. Here, P_{\rm c} is specified by an “equation of state” that depends only on \rho , and may be obtained

Read the complete text