Hydraulic Permeability Estimation (Poiseuille Model)

Reference ID: MET-F99F | Process Engineering Reference Sheets Calculation Guide

Introduction & Context

Hydraulic permeability, denoted as L_p, is a fundamental transport property in membrane science and process engineering. It quantifies the intrinsic ability of a porous medium to transmit fluid under a pressure gradient. In the context of microfiltration (MF) and ultrafiltration (UF) processes, L_p serves as the proportionality constant between the applied transmembrane pressure and the resulting volumetric flux. This calculation is essential for membrane characterization, process design, and predicting the performance of separation systems where Newtonian fluids are processed through porous structures.

Methodology & Formulas

The estimation of hydraulic permeability is derived from the Hagen-Poiseuille equation, which describes laminar, viscous flow through cylindrical conduits. The model assumes the membrane consists of a series of parallel, non-interconnecting cylindrical pores of uniform radius. The governing equation for the hydraulic permeability is defined as:

\[ L_{p} = \frac{\epsilon \cdot r^{2}}{8 \cdot \mu \cdot z} \]

To ensure physical accuracy, all input parameters must be converted to SI base units prior to calculation. The variables are defined as follows:

L_p: Hydraulic permeability \(\left[ \frac{\text{m}}{\text{Pa} \cdot \text{s}} \right]\)
ε: Membrane surface porosity [dimensionless]
r: Pore radius [m]
μ: Dynamic viscosity of the fluid [Pa·s]
z: Membrane thickness [m]

The validity of this model is constrained by the physical characteristics of the membrane and the flow regime. The following table outlines the empirical thresholds required for the model to remain physically meaningful:

Parameter	Validity Range / Condition
Porosity (ε)	0.05 < ε < 0.80
Pore Radius (r)	0.01 μm ≤ r ≤ 10.0 μm
Dynamic Viscosity (μ)	μ > 0 (Newtonian fluid)
Membrane Thickness (z)	z > 0
Flow Regime	Re_pore << 1 (Creeping flow)

The Poiseuille model allows calculation of hydraulic permeability (L_p) directly from the known structural properties of a porous membrane, assuming it can be modeled as a bundle of parallel, non-interconnecting cylindrical pores. The required inputs and procedure are:

Determine the membrane surface porosity (\(\varepsilon\)), often obtained from manufacturer data or image analysis.
Measure or obtain the nominal pore radius (r), typically via microscopy or bubble-point tests.
Know the membrane thickness (z).
Use the dynamic viscosity (\(\mu\)) of the Newtonian fluid of interest at the process temperature.
Apply the Hagen-Poiseuille equation rearranged for permeability: \[ L_{p} = \frac{\varepsilon \cdot r^{2}}{8 \cdot \mu \cdot z} \]
Ensure all parameters are in consistent SI units before calculation.

The model is valid only when the following conditions are met:

Flow is laminar and fully developed within each pore.
Pores can be approximated as straight, circular cylinders.
The fluid behaves as a Newtonian liquid with constant viscosity.
No slip occurs at the pore walls.
Pore size distribution is narrow enough that an average radius is representative.

When dealing with shear-dependent fluids, replace the constant viscosity (\(\mu\)) with an apparent viscosity (\(\mu_{\text{app}}\)) that reflects the shear rate in the pore:

Estimate the wall shear rate (\(\dot{\gamma}\)) for Poiseuille flow: \(\dot{\gamma} = \frac{\Delta P \cdot r}{2 \cdot \mu_{\text{app}} \cdot z}\).
Use the fluid’s rheological model (e.g., Power-law: \(\mu_{\text{app}} = K \cdot \dot{\gamma}^{\,n-1}\)) to calculate \(\mu_{\text{app}}\) at that shear rate.
Insert \(\mu_{\text{app}}\) into the Poiseuille equation. This often requires an iterative solution because the permeability (and thus flow rate for a given \(\Delta P\)) depends on \(\mu_{\text{app}}\), which itself depends on the shear rate set by the flow.

Validation can be performed through the following procedure:

Conduct a series of steady-state pure water (or Newtonian fluid) flux experiments at different transmembrane pressures (\(\Delta P\)).
Record the corresponding volumetric flux values (\(J\)) and calculate the experimental permeability using Darcy’s law: \(L_{p,\text{exp}} = J / \Delta P\).
Compare the experimental \(L_{p,\text{exp}}\) values with those predicted by the Poiseuille model (\(L_{p,\text{calc}}\)).
Assess deviations; if systematic and significant, consider correcting the model for factors like pore tortuosity or a non-uniform pore size distribution.
Document the range of conditions where the model provides acceptable accuracy (e.g., ±10% error).

Worked Example: Hydraulic Permeability Estimation for a Microfiltration Membrane

A process engineer is designing a water purification system using a polymeric microfiltration membrane. To predict the clean water flux under an applied pressure, the intrinsic hydraulic permeability \( L_p \) must be estimated using the ideal Poiseuille model for parallel cylindrical pores.

Known Input Parameters:

Porosity, \( \varepsilon = 0.40 \) (dimensionless)
Nominal pore radius, \( r = 0.20 \, \mu\text{m} \)
Membrane thickness, \( z = 100.0 \, \mu\text{m} \)
Dynamic viscosity of pure water at 20°C, \( \mu = 1.00 \, \text{cP} \)

Step-by-Step Calculation:

Convert all parameters to SI base units.
- Pore radius: \( r = 0.20 \, \mu\text{m} \times 1.000 \times 10^{-6} \, \frac{\text{m}}{\mu\text{m}} = 2.000 \times 10^{-7} \, \text{m} \) (using \( \text{UM_TO_M} = 1.000 \times 10^{-6} \))
- Membrane thickness: \( z = 100.0 \, \mu\text{m} \times 1.000 \times 10^{-6} \, \frac{\text{m}}{\mu\text{m}} = 1.000 \times 10^{-4} \, \text{m} \) (using \( \text{UM_TO_M} = 1.000 \times 10^{-6} \))
- Dynamic viscosity: \( \mu = 1.00 \, \text{cP} \times 1.000 \times 10^{-3} \, \frac{\text{Pa} \cdot \text{s}}{\text{cP}} = 1.000 \times 10^{-3} \, \text{Pa} \cdot \text{s} \) (using \( \text{CP_TO_PAS} = 1.000 \times 10^{-3} \))
Apply the ideal Poiseuille equation. \[ L_p = \frac{\varepsilon \cdot r^2}{8 \cdot \mu \cdot z} \] Substitute the converted values: \[ L_p = \frac{0.40 \cdot (2.000 \times 10^{-7} \, \text{m})^2}{8 \cdot (1.000 \times 10^{-3} \, \text{Pa} \cdot \text{s}) \cdot (1.000 \times 10^{-4} \, \text{m})} \]
Compute the square of the pore radius. \[ r^2 = (2.000 \times 10^{-7} \, \text{m})^2 = 4.000 \times 10^{-14} \, \text{m}^2 \]
Calculate the numerator. \[ \varepsilon \cdot r^2 = 0.40 \cdot 4.000 \times 10^{-14} \, \text{m}^2 = 1.600 \times 10^{-14} \, \text{m}^2 \]
Calculate the denominator. \[ 8 \cdot \mu \cdot z = 8 \cdot 1.000 \times 10^{-3} \, \text{Pa} \cdot \text{s} \cdot 1.000 \times 10^{-4} \, \text{m} = 8.000 \times 10^{-7} \, \text{Pa} \cdot \text{s} \cdot \text{m} \]
Solve for \( L_p \). \[ L_p = \frac{1.600 \times 10^{-14} \, \text{m}^2}{8.000 \times 10^{-7} \, \text{Pa} \cdot \text{s} \cdot \text{m}} = 2.000 \times 10^{-8} \, \frac{\text{m}}{\text{Pa} \cdot \text{s}} \]

Final Answer: The hydraulic permeability of the membrane is \( L_p = 2.000 \times 10^{-8} \, \frac{\text{m}}{\text{Pa} \cdot \text{s}} \). This value can be used in the flux equation \( J = L_p \cdot \Delta P \) to estimate volumetric flux for a given transmembrane pressure.

"Un projet n'est jamais trop grand s'il est bien conçu." — André Citroën

"La difficulté attire l'homme de caractère, car c'est en l'étreignant qu'il se réalise." — Charles de Gaulle