Hostname: page-component-7b7567c8c-7pwfn Total loading time: 0 Render date: 2026-11-15T16:48:46.560Z Has data issue: false hasContentIssue false

Shear dispersion of multispecies electrolyte solutions in the channel domain

Published online by Cambridge University Press:  06 September 2023

Lingyun Ding*
Affiliation:
Department of Mathematics, University of California Los Angeles, CA 90095, USA
*
†Email address for correspondence: dingly@g.ucla.edu

Abstract

In multispecies electrolyte solutions, even in the absence of an external electric field, differences in ion diffusivities induce an electric potential and generate additional fluxes for each species. This electro-diffusion process is well-described by the advection Nernst–Planck equation. This study aims to analyse the long-time behaviour of the governing equation under electroneutrality and zero current conditions, and to investigate how the diffusion-induced electric potential and shear flow enhance the effective diffusion coefficients of each species in channel domains. The exact solutions of the effective equation with certain special parameters, as well as the asymptotic analyses for ions with large diffusivity discrepancies, are presented. Furthermore, there are several interesting properties of the effective equation. First, it is a generalization of the Taylor dispersion, with a nonlinear diffusion tensor replacing the scalar diffusion coefficient. Second, the effective equation exhibits a scaling relation, revealing that the system with a weak flow is equivalent to the system with a strong flow under scaled physical parameters. Third, in the case of injecting an electrolyte solution into a channel containing well-mixed buffer solutions or electrolyte solutions with the same ion species, if the concentration of the injected solution is lower than that of the pre-existing solution, then the effective equation simplifies to a multi-dimensional diffusion equation. However, when introducing the electrolyte solution into a channel filled with deionized water, the ion–electric interaction results in several phenomena not present in the advection–diffusion equation, including upstream migration of some species, spontaneous separation of ions, and non-monotonic dependence of the effective diffusivity on Péclet numbers. Finally, the dependence of effective diffusivity on concentration and ion diffusivity suggests a method to infer the concentration ratio of each component and ion diffusivity by measuring the effective diffusivity.

Information

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

Figure 1.The schematic shows the set-up for the special case of a quadratic shear flow in the parallel-plate channel domain. A multispecies electrolyte in water exists in the form of dissolved ions. Ions of like charges repel, while ions of opposite charges attract, due to electrostatic forces. The interplay between the flow and the ion–electric interactions has a crucial role in determining the behaviour and dynamics of the system.

Figure 1

Figure 2.Experimental photo showing the advection of a sodium fluorescein solution in a pipe through pressure-driven flow. The purple regions represent the presence of ultraviolet tube lights. Upon exposure to these lights, the sodium fluorescein solution emits a vibrant green light. The experimental set-up includes a pump attached to the left-hand end of the pipe, which generates a steady pressure-driven flow to transport the sodium fluorescein solution towards the right. For the detailed procedure and set-up of the laminar channel flow experiments, we refer to Aminian et al. (2018).

Figure 2

Figure 3.Numerical solution to (a,c,e) (2.4) and (b,d, f) (3.42), at $t=0.2$ with ${Pe}=2$. The initial condition, diffusivities and valences are provided in (5.1). The black dashed lines depict the shape of the shear flow profile.

Figure 3

Figure 4.(a–c) Numerical solutions to (2.4) at $t=2$ with ${Pe}=2$. The initial condition, diffusivities and valences are provided in (5.1). (d) The thicker curves represent the cross-sectional average of concentration fields presented in (a–c), with $\bar {c} (x,t) = ({1}/{| \varOmega |}) \int _{\varOmega }c (x,\boldsymbol {y},t) \,\mathrm {d}\boldsymbol {y}$. The thinner curve represents the solution of effective equation (3.23). The complete dynamics of the simulation in (a–c) can be observed in supplementary movie 1, available at https://doi.org/10.1017/jfm.2023.626.

Figure 4

Figure 5.The numerical solution to (2.4), with ${Pe}=8$ at (a,c,e) $t=0.2$, and (b,d, f) $t=1$. The initial condition, diffusivities and valences are provided in (5.1). The complete dynamics of the simulation can be observed in supplementary movie 2.

Figure 5

Figure 6.(a) Comparison between the variance of the numerical solutions for (2.4) (thicker curves) and their theoretical asymptotics (thinner curves), in log-log scale. The numerical solutions at $t=2$ are plotted in figure 4. The asymptotics of the variance is provided in (2.8). In this case, $\mathrm {Var}(\bar {c}_{i})\approx 2 t\,\kappa _{{eff},i} + \sigma ^{2}$, where $\sigma = \frac {1}{4}$ and $\kappa _{{eff},i}$ is defined in (3.32). (b) The infinity norm of $\partial _{\tau }\boldsymbol {C}$ as a function of $\tau$, where $\boldsymbol {C}$ solves (3.27).

Figure 6

Figure 7.(a) The effective diffusivities as functions of the Péclet number ${Pe}$. The inset shows the functionsfor Péclet numbers from 1 to 10. The three curves intersect at Péclet number approximately 3. (b) The relative difference between the effective diffusivity and its approximation provided in (5.2). The solid red curve represents the first ion species, the dashed blue curve represents the second ion species, and the dotted black curve represents the third ion species.

Figure 7

Figure 8.(a) The upper plot shows the numerically solved self-similarity solution of the effective equation (3.28). The initial condition, diffusivities and valences are provided in (5.1). The lower plot shows the ratio of each component, $C_{i}/\sum _{i=1}^{3}C_{i}$. (b) The blue solid curve represents $M_{1} (\xi ^{*})$ and is associated with the left-hand $y$-axis. The red dashed curve represents $M_{2}(\xi ^{*})/M_{1}(\xi ^{*})$ and is associated with the right-hand $y$axis.

Figure 8

Figure 9.(a) The upper plot shows the self-similarity solution of the effective equation (3.28) with ${Pe}=2$. The initial condition, diffusivities and valences are provided in (5.1). The lower plot shows the ratio of each component, $C_{i}/\sum _{i=1}^{3}C_{i}$. (b) The blue solid curve represents $M_{1} (\xi ^{*})$ and is associated with the left-hand $y$-axis. The red dashed curve represents $M_{2}(\xi ^{*})/M_{1}(\xi ^{*})$ and is associated with the right-hand $y$-axis.

Figure 9

Figure 10.(a) The upper plot shows the self-similarity solution of the effective equation (3.28) with ${Pe}=100$. The initial condition, diffusivities and valences are provided in (5.1). The lower plot shows the ratio of each component, $C_{i}/\sum _{i=1}^{3}C_{i}$. (b) The blue solid curve represents $M_{2} (\xi ^{*})$ and is associated with the left-hand $y$-axis. The red dashed curve represents $M_{1}(\xi ^{*})/M_{2}(\xi ^{*})$ and is associated with the right-hand $y$-axis.

Figure 10

Figure 11.The normalized self-similarity solution for effective equation (3.28) with $\widetilde {Pe}=100$ and the diffusivities $\tilde {\kappa }_{1}=1$, $\tilde {\kappa }_{2}=10$, $\tilde {\kappa }_{3}=1$, valences $\tilde {z}_{1}=1$, $\tilde {z}_{2}=0.1$, $\tilde {z}_{3}=-2$ (or equivalently, $\tilde {z}_{1}=10$, $\tilde {z}_{2}=1$, $\tilde {z}_{3}=-20$), and the initial condition $\tilde {c}_{I,1}=({1}/{\sigma \sqrt {2{\rm \pi} }})\,\textrm {e}^{-{1}/{2} ( {x}/{\sigma } )^{2}}$, $\tilde {c}_{I,2}=({10}/{\sigma \sqrt {2{\rm \pi} }})\,\textrm {e}^{-{1}/{2} ( {x}/{\sigma } )^{2}}$, where $\sigma =\frac {1}{4}$.

Figure 11

Table 1.Effective diffusivity for different mass ratios of the second and first ion species, $\int _{-\infty }^{\infty } c_{2}\,\mathrm {d}x/ \int _{-\infty }^{\infty } c_{1}\,\mathrm {d}x$. The parameters are $\kappa _{1}=1$, $\kappa _{2}=0.1$, $\kappa _{3}=1$, $z_{1}=1$, $z_{2}=1$, $z_{3}=-2$.

Supplementary material: Image

Ding Supplementary Movie 1

See "Ding Supplementary Movie Captions"

Download Ding Supplementary Movie 1(Image)
Image 4.4 MB
Supplementary material: Image

Ding Supplementary Movie 2

See "Ding Supplementary Movie Captions"

Download Ding Supplementary Movie 2(Image)
Image 1.2 MB
Supplementary material: PDF

Ding Supplementary Movie Captions

Ding Supplementary Movie Captions

Download Ding Supplementary Movie Captions(PDF)
PDF 105.6 KB