Analysis of a one-dimensional biofilm model

Patrick Guidotti, Christoph Walker · 2025

Read Analysis of a one-dimensional biofilm model by Patrick Guidotti, Christoph Walker (2025) 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-2505-23545.txt.

Introduction

Guidotti and Walker analyzed a one-dimensional biofilm model in twenty twenty-five. A biofilm is a layer of microbes growing on a surface. Their reduced model tracks how a biofilm grows when a needed substance, consumed by reaction, limits its growth.

The model becomes a moving boundary problem, where the biofilm's edge shifts over time. The authors prove the problem is well-posed, meaning solutions exist and behave properly, in a strong mathematical sense. They prove the same for a simpler near-steady version of the model. They also study the shape of solutions over time. In a special case they show steady states exist, and that the non-trivial steady state is unique and stable.

Catalog record

CREATORPatrick Guidotti, Christoph Walker
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE55 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-23545

Browse the full library · About the Tower of Babel

Full text

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

Analysis of a one-dimensional biofilm model Patrick Guidotti, Christoph Walker Abstract . In this paper a reduced one-dimensional moving boundary model is studied that describes the evolution of a biofilm driven by the presence of a reaction limiting substrate. Global well-posedness is established for the resulting parabolic free boundary value problem in strong form in Sobolev spaces and for a quasi-stationary approximation in spaces of classical regularity. The general existence results are complemented by results about the qualitative properties of solutions including the existence, in general, and, additionally, the uniqueness and stability of non-trivial equilibria, in a special case. 2020 Mathematics Subject Classification 35Q92,35R35 1. Introduction Microbial systems are the focus of an increasing body of research due to their prevalence in many important fields of application. Microbes tend to live in localized communities called biofilms. Most of the research dedicated to biofilms is confined to the biological realm or to the specific area of application. The systematic development of mathematical models and their analysis is still quite limited. Inspired by the review article [ 4 ] , we consider a one-dimensional toy model for the evolution of a single biofilm in a single reaction limiting susbtrate (e.g. oxygen) introduced in that article. Even in its simplicity the model takes the form of a non-standard moving boundary problem with interesting features. The goal of this paper is to provide a rigorous analysis of the system that include well-posedness as well as qualitative properties such as global existence and long-time behavior. The model tracks the substrate, via its concentration c(t,z) everywhere in the upper half-space (or half-line since we shall assume homogeneity in the other variables). Its concentration can be assumed constant in the bulk fluid region far enough from the biofilm, where a well mixed regime is typically observed. The biofilm is simply modeled by the height (depth) h(t)\geq 0 to which it has grown. Within the biofilm and in a diffusive boundary layer on top of it, the substrate can freely diffuse with (slightly) different diffusivities. It is assumed, based on observations, that the diffusive boundary layer have a given thickness L>0 . At the bottom a wall confines the system and a no-flux condition is enforced. At the interface z=h(t) continuity of the substrate concentration and flux is imposed along with a dynamic boundary condition for the evolution of the biofilm thickness. The latter captures the modeling assumption that the speed of the interface coincides with the one generated by the pressure change due to the biofilm growth or decay. More specifically, let L,c_{*}>0 , \kappa,\kappa_{L}>0 , \varepsilon_{L},\varepsilon>0 , and consider the system (1.1a) \displaystyle\varepsilon_{L}c_{t} \displaystyle=\kappa_{L}c_{zz}, \displaystyle h(t)<z<h(t)+L,\quad t>0, (1.1b) \displaystyle\varepsilon c_{t} \displaystyle=\kappa c_{zz}-r(c), \displaystyle 0<z<h(t), (1.1c) \displaystyle c(t,h(t)+L) \displaystyle=c_{*},\quad c_{z}(t,0)=0, \displaystyle t>0, (1.1d) \displaystyle\llbracket c\rrbracket \displaystyle=\llbracket\varkappa c_{z}\rrbracket=0, \displaystyle z=h(t), (1.1e) \displaystyle c(0,z) \displaystyle=c_{0}(z),\> \displaystyle 0<z<h_{0}+L, (1.1f) \displaystyle h_{t} \displaystyle=\int_{0}^{h}g\big(r(c(t,z))\big)\,\mathrm{d}z, \displaystyle t>0, (1.1g) \displaystyle h(0) \displaystyle=h_{0}>0, for the unknown biofilm height h=h(t) and substrate c=c(t,z) . Here we have set \varkappa(z)=\begin{cases}\kappa_{L},&z\in\bigl(h(t),h(t)+L\bigr)\\ \kappa,&z\in\bigl(0,h(t)\bigr),\end{cases} for the substrate diffusivities in the diffusive layer [h(t)<z<h(t)+L] and in the biofilm [0<z<h(t)] , respectively. The function r models the substrate consumption that drives the biofilm growth, itself captured by a function g of r . A common choice for the function g is given by g(r)=\alpha(r-b) , where \alpha>0 represents a yield coefficient and b the basic substinence level of the biofilm. Minimal assumptions on the function r are that r(0)=0 , modeling the fact that no substrate is consumed if none is present, and that r(c)>0 for c>0 indicating that available substrate is always used. The function r is also typically made to reflect a saturation at increasing levels of the substrate concentration. We shall not make use of this additional structure in this analysis but we will assume that r^{\prime}(c)>0 for c>0 , which does not contradict the saturation assumption and simply encodes the fact the consumption increases with substrate concentration. The assumptions about the substrate beyond the top diffusive layer and the bottom wall are reflected in the boundary conditions ( 1.1c ), while continuity across the interface is encoded in the conditions ( 1.1d ). The kynetic condition ( 1.1f ) for h is derived from an incompressibility assumption for the fluid (mostly water) and Darcy’s law connecting velocity and pressure. Full details can be found in [ 4 ] or the earlier [ 2 ] . The remaining equations ( 1.1e ) and ( 1.1g ) fix an arbitrary initial configuration. If diffusion in the diffusive biofilm layer is assumed to take place at a much faster time scale than biofilm growth, it can be assumed that \varepsilon_{L}=0 in ( 1.1a ). In that case, c is affine on (h,h+L) and it follows from ( 1.1c ) and \llbracket c\rrbracket=0 on z=h(t) that c is determined on (h,h+L) by c\bigl(t,h(t)\bigr) through (1.2) c(t,z)=\frac{c_{*}-c\bigl(t,h(t)\bigr)}{L}\big[z-h(t)\big]+c\bigl(t,h(t)\bigr). Using that \llbracket\varkappa c_{z}\rrbracket=0 on z=h(t) then yields c_{z}\bigl(t,h(t)-\bigr)=\frac{\kappa_{L}}{\kappa}c_{z}\bigl(t,h(t)+\bigr)=\frac{\kappa_{L}}{\kappa}\frac{c_{*}-c\bigl(t,h(t)\bigr)}{L} and one obtains the Robin type boundary condition (1.3) c\bigl(t,h(t)\bigr)+\frac{L\kappa}{\kappa_{L}}c_{z}\bigl(t,h(t)\bigr)=c_{*}, for c\big|_{[0,h(t)]} where here and in the following c_{z}\bigl(t,h(t)\bigr)=c_{z}\bigl(t,h(t)-\bigr) . In summary, assuming that \varepsilon_{L}=0 we obtain the following reduced model (1.4a) \displaystyle\varepsilon c_{t} \displaystyle=\kappa c_{zz}-r(c), \displaystyle 0<z<h(t),\quad t>0, (1.4b) \displaystyle c\bigl(t,h(t)\bigr)+\frac{L\kappa}{\kappa_{L}}c_{z}\bigl(t,h(t)\bigr) \displaystyle=c_{*},\quad c_{z}(t,0)=0, \displaystyle t>0, (1.4c) \displaystyle c(0,z) \displaystyle=c_{0}(z), \displaystyle z\in(0,h_{0}), (1.4d) \displaystyle h_{t} \displaystyle=\int_{0}^{h}g\big(r(c(t,z))\big)\,\mathrm{d}z, \displaystyle t>0, (1.4e) \displaystyle h(0) \displaystyle=h_{0}. on which we shall focus in the following. Throughout, if not stated otherwise, we shall assume that (1.5) g\in{\rm C}^{1-}(\mathbb{R}),\qquad r\in{\rm C}^{1}(\mathbb{R})\ \text{ with }\ sr(s)>0\ \text{ for }\ s\not=0, where {\rm C}^{1-} refers to Lipschitz continuity. In fact, in all of our analysis only the restrictions of r to [0,c_{*}] and of g to [0,r(c_{*})] play a role. Note that ( 1.5 ) implies r(0)=0 . 2. The Evolution Problem In this section we focus on the full evolutionary system ( 1.4 ) and establish the existence of a unique global strong solution. We then derive some qualitative properties of the solution including the asymptotic extinction of the biofilm when the growth rate is negative, the boundedness of the substrate gradient for initial data with bounded gradient as well as the preservation of susbstrate monotonicity during the evolution. 2.1. Well-Posedness of the Evolution Problem We consider the evolution problem ( 1.4 ) with \varepsilon=1 for notational simplicity. Introducing dimensionless variables via (2.1) y=\frac{z}{h(t)}\quad\text{ and }\quad v(t,y):=c_{*}-c\bigl(t,yh(t)\bigr), the moving boundary problem ( 1.4 ) with \varepsilon=1 is equivalent to the following initial (fixed) boundary value problem with homogeneous boundary conditions (2.2a) \displaystyle v_{t}=\frac{\kappa}{h^{2}(t)}v_{yy}+\frac{h_{t}(t)}{h(t)}yv_{y}+r\bigl(c_{*}-v(t,y)\bigr), \displaystyle 0<y<1,\quad t>0, (2.2b) \displaystyle\frac{\kappa}{h(t)^{2}}v_{y}(t,0)=0, \displaystyle t>0, (2.2c) \displaystyle\frac{\kappa_{L}}{Lh(t)}v(t,1)+\frac{\kappa}{h(t)^{2}}v_{y}(t,1)=0, \displaystyle t>0, (2.2d) \displaystyle v(0,y)=v_{0}(y):=c_{*}-c_{0}(yh_{0}), \displaystyle 0<y<1, (2.2e) \displaystyle h_{t}(t)=h(t)\int_{0}^{1}g\big(r\bigl(c_{*}-v(t,y)\bigr)\big)\,\mathrm{d}y, \displaystyle t>0, (2.2f) \displaystyle h(0)=h_{0}, where we write the boundary conditions ( 2.2b )-( 2.2c ) in this particular form to be consistent with the notation used in [ 1 ] . Theorem 2.1 . Assume ( 1.5 ) and let p\in(1,\infty) . Given initial values h_{0}\in(0,\infty) and v_{0}\in\operatorname{W}_{p}^{1}\bigl((0,1)\bigr) with v_{0}(y)\in[0,c_{*}] for y\in(0,1) , there is a unique global solution (v,h) of ( 2.2 ) such that h\in{\rm C}^{1}\big([0,\infty),(0,\infty)\big) and v\in{\rm C}\bigl([0,\infty),\operatorname{W}_{p}^{1}\bigl((0,1)\bigr)\bigr)\cap{\rm C}\big((0,\infty),\operatorname{W}_{p}^{2}\big((0,1)\bigr)\big)\cap{\rm C}^{1}\big([0,\infty),{\rm L}_{p}\big((0,1)\bigr)\big) with v(t,y)\in[0,c_{*}] for t\geq 0 and y\in(0,1) . Proof. Step 1. In order to establish the local existence of a solution we recast ( 2.2 ) as a quasilinear equation for w=(v,h) to which the theory described in

Read the complete text