Hostname: page-component-7b7567c8c-wxvwp Total loading time: 0 Render date: 2026-11-16T11:19:56.562Z Has data issue: false hasContentIssue false

Slip-enhanced Rayleigh–Plateau instability of a liquid film on a fibre

Published online by Cambridge University Press:  09 January 2023

Chengxi Zhao
Affiliation:
Department of Modern Mechanics, University of Science and Technology of China, Hefei 230026, PR China
Yixin Zhang
Affiliation:
Physics of Fluids Group and Max Planck Center Twente for Complex Fluid Dynamics, MESA+ Institute and J. M. Burgers Centre for Fluid Dynamics, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands
Ting Si*
Affiliation:
Department of Modern Mechanics, University of Science and Technology of China, Hefei 230026, PR China
*
†Email address for correspondence: tsi@ustc.edu.cn

Abstract

Boundary conditions at a liquid–solid interface are crucial to dynamics of a liquid film coated on a fibre. Here, a theoretical framework based on axisymmetric Stokes equations is developed to explore the influence of liquid–solid slip on the Rayleigh–Plateau instability of a cylindrical film on a fibre. The new model not only shows that the slip-enhanced growth rate of perturbations is overestimated by the classical lubrication model, but also indicates a slip-dependent dominant wavelength, instead of a constant value obtained by the lubrication method, which leads to larger drops formed on a more slippery fibre. The theoretical findings are validated by direct numerical simulations of Navier–Stokes equations via a volume-of-fluid method. Additionally, the slip-dependent dominant wavelengths predicted by our model agree with the experimental results provided by Haefner et al. (Nat. Commun., vol. 6, issue 1, 2015, 7409).

Information

Type
JFM Papers
Copyright
© The Author(s), 2023. Published by Cambridge University Press
Figure 0

Figure 1.Schematic of a liquid film on a fibre.

Figure 1

Figure 2.The dispersion relation between the growth rate $\omega$ and the wavenumber $k$ on the no-slip boundary condition ($l_{s}=0$) for various dimensionless fibre radii: (a) $\alpha =0.7$ (black), $\alpha =0.6$ (red), $\alpha =0.5$ (blue); (b) $\alpha =0.3$ (black), $\alpha =0.2$ (red), $\alpha =0.1$ (blue). The predictions of the Stokes model (3.16) are plotted in solid lines, and results from the lubrication model (3.18) are illustrated in dashed lines. The triangles and circles are the long-wave and Stokes results of Craster & Matar (2006), employed here to verify (3.16) and (3.18), respectively.

Figure 2

Figure 3.The dispersion relation between the growth rate $\omega$ and the wavenumber $k$ on different boundary conditions of various dimensionless fibre radii: (a) $\alpha =0.9$, (b) $\alpha =0.7$, (c) $\alpha =0.5$, (d) $\alpha =0.3$. The solid and dashed lines are the predictions of the Stokes model (3.16) and the lubrication model (3.18), respectively, on three boundary conditions with $l_{s}=0$ (black), $l_{s}=0.25$ (blue) and $l_{s}=0.5$ (red).

Figure 3

Figure 4.Variations of the dominant wavenumber $k_{max}$ and corresponding growth rate $\omega _{max}$ with slip lengths. The solid and dashed lines are the predictions of the Stokes model (3.16) and the lubrication model (3.18), respectively, for different fibre thicknesses, i.e. $\alpha =0.9$ (black), $\alpha =0.7$ (blue), $\alpha =0.5$ (red) and $\alpha =0.3$ (purple).

Figure 4

Figure 5.Film thinning simulation results of one perturbation wave in an axially symmetric domain. For this case, $L=10$, $l_{s}=0$ and $\alpha =0.7$. Interface profiles $h$ (upper half) and contours of the axial velocity $u(x,r,t)$ inside the film (lower half) are shown at three time instants: (a) $t_1=300$, (b) $t_2=600$, (c) $t_3=900$. Here, $h_{min}$ represents the minimum radius of the film.

Figure 5

Figure 6.Axial velocity fields at $t=100$ for different slip lengths: (a,b) $l_{s} = 0$, (c,d) $l_{s} = 0.25$, (e, f) $l_{s} = 0.5$. Here, the fibre radius is $\alpha = 0.7$. (a,c,e) Contours of the entire configuration (upper plots) and velocity vectors of the local field near $x=6$ (lower plots). (b,d, f) Velocity profiles at four local positions $x$: $5.5$ (black), 6.0 (purple), 6.5 (blue) and 7.0 (red), shown by dashed lines with arrows in (a,c,e).

Figure 6

Figure 7.Linear time evolution of the minimum radii of films $h_{min}(t)$ on two fibre radii: (a) $\alpha =0.7$, (b) $\alpha =0.5$. The numerical solutions (dotted lines with symbols) are compared to the predictions of both the lubrication model (dashed lines) and Stokes model (solid lines) for three slip lengths: $l_{s} = 0$ (black), $l_{s}=0.25$ (red) and $l_{s}=0.5$ (blue). The inset illustrates $h_{min}(t)$ in the uniform grids. Here, $t$ is scaled by $\mu h_0/\gamma$.

Figure 7

Figure 8.Ratio of the velocity gradients at $t=100$ for different slip lengths: (a) $l_{s} = 0$, (b) $l_{s} = 0.25$, (c) $l_{s} = 0.5$. Here, the fibre radius is $\alpha = 0.7$, and $u_{xx}$ and $(r u_r)_r/r$ represents the second-order axial and radial derivatives of the axial velocity in (A24), respectively.

Figure 8

Figure 9.(a) Space–time plot of interface positions $h(x,t)$ on a no-slip fibre of the radius $\alpha =0.7$. Darker (brighter) colours correspond to the smaller (larger) values of $h(x,t)$. (b,c) Interface profiles extracted from (a) at early and late stages of the instability.

Figure 9

Figure 10.(a,b) The root mean square (r.m.s.) of non-dimensional perturbation amplitude versus non-dimensional wavenumber on fibres with different slip lengths, at four time instants: (a) $2\times 10^3$ (green), $4.5\times 10^3$ (black), $6\times 10^3$ (red) and $7\times 10^3$ (blue); (b) $1.2\times 10^2$ (green), $3.3\times 10^2$ (black), $4.2\times 10^2$ (red) and $4.8\times 10^2$ (blue). The inset shows the time history of the dominant wavenumber. (c) Variations of the dominant wavelengths with slip lengths: a comparison between the theoretical predictions of (3.16) and numerical solutions for two fibres of different radii.

Figure 10

Figure 11.(a) Interface profiles of liquid films on fibres of radius $\alpha =0.7$. The results for two slip lengths are presented by a solid blue line for $l_{s}=0$, and a red dash-dotted line for $l_{s}=3$. (b) Distributions of the droplet size on the no-slip (upper plot) and slip (lower plot).

Figure 11

Figure 12.Influence of the geometry on the dominant wavelength of the instability: a comparison between the theoretical predictions and experimental data. The experimental data come from the work of Haefner et al. (2015).

Figure 12

Figure 13.Influence of the geometry on the growth rate of the dominant mode on no-slip and slip fibres. The solid lines and dashed lines are the predictions of the Stokes model and the lubrication model, respectively. Blue lines represent no-slip results, and red lines are the slip ones. The experimental data come from the work of Haefner et al. (2015).