Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Cake Layer Formation and Cake Filtration of Soft Particles : Numerical Simulation of Flow in Porous Media: Permeability of Micropillar Arrays in a Microfluidic Membrane

Al-Ariqi, Amgad Khaled Mansoor Shaef LU (2025) METM01 20251
Chemical Engineering (M.Sc.Eng.)
Abstract
Membrane fouling, particularly by soft, deformable particles, is a significant challenge in microfiltration (MF) processes, leading to reduced efficiency and increased operational costs. This thesis investigates the formation of filter cake layers by numerically simulating fluid flow through microfluidic membranes designed with random, non-overlapping cylindrical micropillar arrays. The study aims to understand how geometric parameters of the porous structure, specifically pillar diameter, minimum pore throat size, and porosity, influence the hydrodynamic properties of the membrane. Using a custom Python-based Monte Carlo (MC) algorithm, 27 distinct 2D micropillar configurations were generated. Computational Fluid Dynamics (CFD)... (More)
Membrane fouling, particularly by soft, deformable particles, is a significant challenge in microfiltration (MF) processes, leading to reduced efficiency and increased operational costs. This thesis investigates the formation of filter cake layers by numerically simulating fluid flow through microfluidic membranes designed with random, non-overlapping cylindrical micropillar arrays. The study aims to understand how geometric parameters of the porous structure, specifically pillar diameter, minimum pore throat size, and porosity, influence the hydrodynamic properties of the membrane. Using a custom Python-based Monte Carlo (MC) algorithm, 27 distinct 2D micropillar configurations were generated. Computational Fluid Dynamics (CFD) simulations were then performed using COMSOL Multiphysics to solve the steady-state Navier-Stokes equations for water flow. The permeability of each configuration was calculated using Darcy's law, and the results were analysed as a function of pillar geometry and applied inlet pressure (10-40 mbar). The study also includes the fabrication of corresponding physical molds and polydimethylsiloxane (PDMS) microfluidic devices via 3D printing and soft lithography, laying the groundwork for future experimental validation. The simulation results demonstrate that permeability is strongly dependent on the microstructural geometry, increasing with larger pillar diameters, pore throat sizes, and higher porosity, with an increase in porosity of just 0.1 resulting in a more than two-fold increase in permeability. A key finding is the pressure-dependent nature of permeability; in all configurations, permeability decreased near-linearly with increasing inlet pressure. This deviation from classical Darcy's Law indicates the onset of non-Darcian flow, where inertial effects become significant even at low Reynolds numbers (Re < 10).

Analysis of the velocity and pressure fields revealed that the disordered pillar arrangements create highly tortuous flow paths, leading to channelling and the formation of eddies in the wake of pillars. These recirculating zones increase the local residence time of particles, raising the likelihood of their adhesion to pillar surfaces, which can initiate fouling. The study concludes that the overall flow resistance and fouling potential are governed by the global pillar network structure rather than solely by localized constrictions. Direct numerical simulation (DNS) proves to be an essential tool for accurately predicting flow behaviour in such complex geometries, as established theoretical models showed significant discrepancies. These findings provide critical insights for optimizing membrane design to minimize fouling by soft particles. (Less)
Popular Abstract
In industries from water purification to food production, a major problem called "membrane fouling" can increase energy consumption by over 50% and significantly raise operational costs. This clogging of filters is a major challenge, especially when the particles being filtered are soft and squishy, like tiny gels or biological cells, because they can deform and compact to create a dense, impermeable layer.
This research explores how to design better microfilters to prevent this kind of clogging. Instead of a simple mesh, modern microfilters can be designed with an intricate micropillar arrays. The way these pillars are arranged, their size, the space between them, and how densely they are packed can dramatically change how fluid flows... (More)
In industries from water purification to food production, a major problem called "membrane fouling" can increase energy consumption by over 50% and significantly raise operational costs. This clogging of filters is a major challenge, especially when the particles being filtered are soft and squishy, like tiny gels or biological cells, because they can deform and compact to create a dense, impermeable layer.
This research explores how to design better microfilters to prevent this kind of clogging. Instead of a simple mesh, modern microfilters can be designed with an intricate micropillar arrays. The way these pillars are arranged, their size, the space between them, and how densely they are packed can dramatically change how fluid flows through the filter.
Using powerful computer simulations, this study investigated how water flows through different arrangements of these micropillars. We discovered that the filter's design has a significant impact on its performance. For instance, larger pillars and gaps between them allow water to flow more easily.

More importantly, we found that the filter's permeability, a measure of how easily fluid can pass through, is not a fixed value. When we increased the pressure, the permeability went down. This is because the complex paths the water takes around the pillars create eddies in the flow. These eddies create low-flow zones where soft particles are more likely to settle and accumulate, which can lead to clogging.
By understanding how the filter's microscopic geometry creates these flow patterns, we can start to design smarter filters and devices that guide particles through smoothly, minimizing the dead zones where they can get trapped. This research provides a blueprint for developing the next generation of filtration systems that are more effective and resistant to fouling. (Less)
Please use this url to cite or link to this publication:
author
Al-Ariqi, Amgad Khaled Mansoor Shaef LU
supervisor
organization
course
METM01 20251
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Porous media, Permeability, Micropillar Array, Microfluidics, Membrane fouling, Soft particles, Cake filtration, CFD, Membrane technology
language
English
id
9210774
date added to LUP
2025-09-01 13:16:47
date last changed
2025-09-01 13:16:47
@misc{9210774,
  abstract     = {{Membrane fouling, particularly by soft, deformable particles, is a significant challenge in microfiltration (MF) processes, leading to reduced efficiency and increased operational costs. This thesis investigates the formation of filter cake layers by numerically simulating fluid flow through microfluidic membranes designed with random, non-overlapping cylindrical micropillar arrays. The study aims to understand how geometric parameters of the porous structure, specifically pillar diameter, minimum pore throat size, and porosity, influence the hydrodynamic properties of the membrane. Using a custom Python-based Monte Carlo (MC) algorithm, 27 distinct 2D micropillar configurations were generated. Computational Fluid Dynamics (CFD) simulations were then performed using COMSOL Multiphysics to solve the steady-state Navier-Stokes equations for water flow. The permeability of each configuration was calculated using Darcy's law, and the results were analysed as a function of pillar geometry and applied inlet pressure (10-40 mbar). The study also includes the fabrication of corresponding physical molds and polydimethylsiloxane (PDMS) microfluidic devices via 3D printing and soft lithography, laying the groundwork for future experimental validation. The simulation results demonstrate that permeability is strongly dependent on the microstructural geometry, increasing with larger pillar diameters, pore throat sizes, and higher porosity, with an increase in porosity of just 0.1 resulting in a more than two-fold increase in permeability. A key finding is the pressure-dependent nature of permeability; in all configurations, permeability decreased near-linearly with increasing inlet pressure. This deviation from classical Darcy's Law indicates the onset of non-Darcian flow, where inertial effects become significant even at low Reynolds numbers (Re < 10).

Analysis of the velocity and pressure fields revealed that the disordered pillar arrangements create highly tortuous flow paths, leading to channelling and the formation of eddies in the wake of pillars. These recirculating zones increase the local residence time of particles, raising the likelihood of their adhesion to pillar surfaces, which can initiate fouling. The study concludes that the overall flow resistance and fouling potential are governed by the global pillar network structure rather than solely by localized constrictions. Direct numerical simulation (DNS) proves to be an essential tool for accurately predicting flow behaviour in such complex geometries, as established theoretical models showed significant discrepancies. These findings provide critical insights for optimizing membrane design to minimize fouling by soft particles.}},
  author       = {{Al-Ariqi, Amgad Khaled Mansoor Shaef}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Cake Layer Formation and Cake Filtration of Soft Particles : Numerical Simulation of Flow in Porous Media: Permeability of Micropillar Arrays in a Microfluidic Membrane}},
  year         = {{2025}},
}