Diffusioosmosis of electrolyte solutions in axisymmetric channels

Technical Summary: Diffusioosmosis of Electrolyte Solutions in Axisymmetric Channels

Problem Statement
The paper addresses the theoretical challenge of describing steady-state diffusio-osmotic flow of electrolyte solutions in long, axisymmetric channels (e.g., cylindrical, conical, or periodically modulated pores) subject to concentration and pressure gradients. While diffusio-osmosis in planar slits with neutral solutes is well-understood, the theory for charged channels containing electrolytes remains underdeveloped. In such systems, the flow is driven not only by solute concentration gradients but also by tangential non-uniform electric fields (electro-osmotic contributions) arising from the differential diffusion of cations and anions. Previous analytical treatments often assumed constant concentration gradients or focused solely on planar geometries, failing to account for the non-linear coupling between fluid flow, ion flux, and variable channel cross-sections found in realistic porous media and microfluidic devices.

Methodology
The authors develop a non-linear analytical theory based on the coupled Nernst-Planck, Poisson, and Stokes equations. The model considers a charged axisymmetric channel of length $L$ and local radius $R(X)$ connecting reservoirs of different salinities ($C_0$ and $C_1$) and hydrostatic pressures ($P_0$ and $P_1$).

Key methodological features include:

• Thin Double Layer Approximation: The theory assumes the electrostatic diffuse layers are thin compared to the local channel radius ($\lambda_D \ll R$), allowing the channel to be treated as electro-neutral in the "bulk" region with slip boundary conditions at the walls.
• Dimensionless Formulation: The governing equations are non-dimensionalized to derive closed-form expressions for the global fluid flow rate ($Q$) and ionic flux ($J$).
• Self-Consistent Solution: Unlike linear theories, the authors solve for the concentration profile $c_m(x)$ and potential $\psi_m(x)$ self-consistently, acknowledging that the local driving forces depend on the global flow rate and vice versa.
• Geometric Modeling: Variable cross-sections are modeled using a family of curves $r(x) = (1 + A\omega \sin(\omega x))^{-1/2}$, encompassing expanding, contracting, and sinusoidally modulated geometries.
• Cylinder Approximation: To interpret results for variable cross-sections, the authors introduce an "effective cylinder" approximation, relating the hydrodynamic resistivity of complex shapes to an equivalent cylindrical radius.

Key Contributions and Results

1. Flow Rate in Cylinders vs. Slits: The authors demonstrate that the magnitude of the pure diffusio-osmotic flow rate ($Q_{DO}$) in a cylindrical channel is identical to that in a planar slit of equivalent thickness. However, the response to an applied pressure drop ($\Delta p$) differs quantitatively; the pressure-driven contribution to the total flow is smaller in cylinders than in slits due to geometric factors.
2. Non-Linear Coupling: The study confirms that for large concentration drops, the relationship between flow rate and driving forces is non-linear. The global flow rate $Q$ and ionic flux $J$ are coupled through the local bulk concentration profile, precluding a simple linear matrix representation of mobility.
3. Effect of Geometry on Flow:
   • Expanding Channels: Channels that widen in the direction of flow (toward the salty reservoir) accelerate the diffusio-osmotic flow.
   • Contracting Channels: Channels that narrow in the direction of flow retard the diffusio-osmotic flow.
   • Linearity of $\Delta p(Q)$: The relationship between pressure drop and flow rate is found to be nearly linear for all tested geometries, though the slope (hydrodynamic resistivity) is highly sensitive to the channel shape.
4. Ionic Flux Tuning: The paper derives an equation linking ionic flux to the total flow rate. It shows that while the sign of the diffusio-osmotic flow is generally toward the fresh reservoir (equalizing concentrations), the net ionic flux can be tuned to be positive or negative depending on the channel geometry and the applied pressure. Specifically, the geometry can shift the point where the net ionic flux is zero.
5. Cylinder Approximation Validity: The authors propose a simple analytical approximation where a complex channel is replaced by an imaginary cylinder with an effective radius $R_{eff}$. This radius is determined by the channel's hydrodynamic resistivity ($W$). This approximation accurately predicts the $\Delta p(Q)$ curves and the zero-flow pressure drop, providing a practical tool for interpreting data.

Significance and Claims
The paper claims to provide a framework for interpreting experimental and numerical data regarding diffusio-osmosis in axisymmetric geometries, which are more representative of real materials (e.g., porous membranes, biological pores) than planar slits.

The authors emphasize that their analysis reveals how channel geometry acts as a control parameter for fluid flow rates, either enhancing or retarding the flow compared to a standard cylinder. They conclude that while the diffusio-osmotic flow in thick channels inherently tends to equalize salt concentrations (limiting its direct use for desalination in this regime), the ability to tune the ionic flux and flow rate via channel shape and pressure drop offers significant potential for the design of micro- and nanofluidic devices. The work extends previous findings on planar slits to axisymmetric systems, offering a more realistic theoretical basis for applications in separation technologies, energy harvesting, and biosensing.