Active Particle Diffusion in Convection Roll Arrays
Pulak K. Ghosh 1, Fabio Marchesoni 2,3, Yunyun Li 2, Franco Nori 4,5 · 2025
Read Active Particle Diffusion in Convection Roll Arrays by Pulak K. Ghosh 1, Fabio Marchesoni 2,3, Yunyun Li 2, Franco Nori 4,5 (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-2508-09924.txt.
Introduction
Tiny self driven particles, called Janus particles, swim through flows in ways plain drifting grains do not. Pulak Ghosh and three co-authors simulated their motion in 2025 inside a row of counter turning flow rolls, like a line of small whirlpools. The swimmers were given a range of self speeds and left to wander.
Like plain grains, the swimmers spread faster than in still fluid, but only up to a set self speed. Faster swimmers instead rode the edges of the array, spreading by a different rule. Their positions also clumped unevenly, unlike the even spread of plain grains. The authors tie these quirks to the mix of self drive, noise, and tight space.
Catalog record
| CREATOR | Pulak K. Ghosh 1, Fabio Marchesoni 2,3, Yunyun Li 2, Franco Nori 4,5 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-09924
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Active Particle Diffusion in Convection Roll Arrays Pulak K. Ghosh 1, Fabio Marchesoni 2,3, Yunyun Li 2, Franco Nori 4,5 Abstract We numerically investigated the Brownian motion of active Janus particles in a linear array of planar counter-rotating convection rolls at high Péclet numbers. Similarly to passive particles, active microswimmers exhibit advection enhanced diffusion, but only for self-propulsion speeds up to a critical value. The diffusion of faster Janus particles is governed by advection along the array’s edges, whereby distinct diffusion regimes are observed and characterized. Contrary to passive particles, the relevant spatial distributions of active Janus particles are inhomogeneous. These peculiar properties of active matter are related to the combined action of noise and self-propulsion in a confined geometry and hold regardless of the actual flow boundary conditions. I Introduction The diffusion of a tracer (organic or artificial, alike) in a suspension fluid is a standard problem of classical transport theory Redner . This paper combines two distinct aspects of this phenomenon, which recently attracted widespread interdisciplinary interest, each for its own merit: (i) the persistent (or time-correlated) random motion of self-propelling particles and (ii) colloidal dispersion in laminar flows. The most tractable example of persistent Brownian motion is represented by artificial micro-swimmers, namely tiny Brownian particles capable of self-propulsion in an active medium Granick ; Muller . A class of artificial swimmers widely investigated in the current literature is the so-called Janus particles (JP), mostly spherical colloidal particles with two differently coated hemispheres, or “faces” Marchetti ; Gompper . Recently, artificial micro- and nano-swimmers of this class have been the focus of pharmaceutical (e.g., smart drug delivery smart ) and medical research (e.g., robotic microsurgery Wang ). These peculiar Brownian particles change direction randomly as usual, but with finite time scale; persistence makes their diffusion extremely sensitive to geometric confinement and other constraints ourPRL ; sperm . Technological applications involving sub-millimeter artificial swimmers thus require accurate control of their diffusive properties in non-homogeneous environments Redner ; BechRMP . On the other hand, Brownian diffusion in an advective medium is also a nanotechnological issue, for instance, in the design and operation of microfuidic devices Kirby ; Tabeling ; PNAS or chemical reactors Moffatt . Let us consider a Brownian tracer of free diffusion constant D_{0} , advected by the free-boundary stationary laminar flow of Fig. 1 (a) with stream function Chandra ; Child2 , \psi(x,y)=({U_{0}L}/{2\pi})\sin({2\pi x}/{L})\sin({2\pi y}/{L}). (1) On combining the two constants, L , the flow’s spatial period, and U_{0} , the maximum advection speed, one defines the advection diffusion scale, D_{L}=U_{0}L/2\pi , and the maximum roll vorticity, \Omega_{L}=2\pi U_{0}/L (Appendix A). Figure 1: Spatial distributions of a Janus particle in the laminar flow of Eq. ( 1 ), sketched in (a)-(b), for v_{0}=0.1 (c) and 0.5 (d). The chart levels are color-coded on natural logarithmic scales as indicated. Other simulation parameters are: D_{0}=0.01 , D_{\theta}=0.01 , U_{0}=1 and L=2\pi . According to Eq. ( 3 ), here v_{c}=0.4 . A practical realization of a linear convection array is represented by the Rayleigh-Bénard rolls sketched in (b); the JP self-propulsion model of Eq. ( 2 ) is illustrated in (a). At high Péclet numbers, {\rm Pe}=D_{L}/D_{0}\gg 1 , a passive tracer undergoes normal diffusion with enhanced diffusion constant D=\kappa\sqrt{D_{L}D_{0}} with \kappa=1.065 , that is D>D_{0} Rosen . This advection effect, termed advection enhanced diffusivity (AED), has been explained Rosen ; Soward ; Shraiman ; Pomeau by noticing that for D_{0}<D_{L} an unbiased particle jumps between convection rolls while being advected along their separatrices. Narrow flow boundary layers (FBL) of estimated width \delta=({D_{0}/\Omega_{L}})^{1/2} , form a network of advection channels centered around the \psi(x,y) cell separatrices, thus enabling a large-scale particle’s diffusion. Peculiar effects due to the combination of self-propulsion and advection are expected to emerge when one considers an active JP suspended in a one dimensional (1D) array of counter-rotating convection rolls. An ideal experimental setup is sketched in Fig. 1 (b). An array of stationary Rayleigh-Bénard cells can occur in a plane horizontal layer of fluid heated from below RB1 ; RB2 . Assuming that they are counter-rotating cylinders parallel to the z -axis, the z coordinate of a suspended tracer is ignorable; hence the reduced two dimensional (2D) flow pattern of Eq. ( 1 ). Advection enhanced diffusivity of passive colloidal particles in arrays of Rayleigh-Bénard rolls has already been demonstrated experimentally Gollub1 ; Gollub3 . Experimental data on the dispersion of self-propelling microswimmers in convective laminar flows are scarse. In this regard, active JPs are ideal tracers for this kind of measurements because self-propulsion speed can be conveniently tuned with respect to the advection drag established in the convection cell. This paper is organized as follows. In Sec. II we present our model and briefly discuss the dynamical significance of the relevant parameters. Our derivation of the relevant time scales is detailed in Appendix A. Our main numerical results are analysed in Secs. III and IV , where we show that: (i) The interplay of advection and self-propulsion causes the nonuniform spatial distribution of a confined active JP. For self-propulsion speeds below a certain threshold, its distribution tends to accumulate along the roll boundaries (Sec. III and Appendix B); (ii) Under these conditions, its self-propulsion and advection velocities tend to line up, so that, contrary to the 2D case of Ref. RR2 , the large-scale diffusion of an active JP is insensitive to self-propulsion itself (Sec. IV ); (iii) Active tracers with self-propulsion speeds larger than the above threshold, attain a maximum diffusion constant for an optimal persistence time, which we relate to advection at the array’s edges (Sec. IV and Appendix C). In Sec. V we stress the role of geometric confinement on the diffusion properties of an active JP in a convection array and show that the picture above holds also for rigid (i.e., no-slip) edge flows. II Model By (linear) convection array we mean here a stationary laminar flow with periodic stream function like \psi(x,y) of Eq. ( 1 ), confined between two parallel edges, y=0 and y=L/2 , which act as dynamical reflecting boundaries. The unit cell of the array consists of two counter-rotating convection rolls [Fig. 1 (a)]. The dynamics of an overdamped active JP can then be formulated by means of two translational and one rotational Langevin equation (LE), \displaystyle\dot{\mathbf{r}} \displaystyle= \displaystyle{\mathbf{v}}_{\psi}+{\mathbf{v}}_{0}+\sqrt{D_{0}}~{\bm{\xi}}(t) (2) \displaystyle\dot{\theta} \displaystyle= \displaystyle({\alpha}/{2})~\nabla\times{\mathbf{v}}_{\psi}+\sqrt{D_{\theta}}~\xi_{\theta}(t), where {\mathbf{r}}=(x,y,) , {\mathbf{v}}_{\psi}=(\partial_{y},-\partial_{x})\psi denotes the advection velocity and the self-propulsion vector, {\mathbf{v}}_{0}=v_{0}(\cos\theta,\sin\theta) , has constant modulus, v_{0} , and is oriented at an angle \theta with respect to the longitudinal x -axis. The translational (thermal) noises in the x and y directions, {\bm{\xi}}(t)=(\xi_{x}(t),\xi_{y}(t)) , and the rotational noise, \xi_{\theta}(t) , are stationary, independent, delta-correlated Gaussian noises, \langle\xi_{i}(t)\xi_{j}(0)\rangle=2\delta_{ij}\delta(t) , with i,j=x,y,\theta . D_{0} and D_{\theta} are the respective noise strengths, which for generality we assume to be unrelated ourPRL . To avoid uncontrolled hydrodynamic effects, the particle is taken to be pointlike PNAS . Other effects due to its actual geometry and chemical-physical characteristics are encoded in the model dynamical parameters. The reciprocal of D_{\theta} coincides with the angular persistence (or correlation) time, \tau_{\theta} , of {\mathbf{v}}_{0} ; accordingly, l_{\theta}=v_{0}/D_{\theta} quantifies the persistence length of the particle’s self-propelled random motion. The flow shear