Understanding Gas Flow Through Porous Materials
Principal Investigator:
Prof. Dr. Francesca di Mare
Affiliation:
Ruhr-Universität Bochum, Bochum Germany
Local Project ID:
pn73gi
HPC Platform used:
SuperMuc-NG PH1-CPU at LRZ
Date published:
Gas flow through porous materials is crucial in various industrial and environmental systems, such as chemical reactors, filtration units, cooling systems, and natural environments. These materials, with void spaces called pores, create complex pathways for gas flow, especially under turbulent conditions. Researchers at the Chair of Thermal Turbomachines and Aeroengines used advanced numerical simulations to study these effects. They replicated industrial packed beds and used the SuperMUC-NG supercomputing system to model gas flow through structured arrays of particles. Their simulations enhance the current understanding of gas flow in porous media and impact the current mathematical models used to study such systems.
Gas flow through porous materials plays a critical role in many industrial and environmental systems. These include chemical reactors filled with catalyst pellets, filtration and separation units, cooling systems, and natural environments such as forests or porous rock layers. Porous materials are naturally occurring or engineered solids with void spaces—pores—through which a gas can flow. Although gases are able to move freely in open space, their behaviour becomes far more complex when passing through or over dense arrangements of solid structures.
When a gas flows through such media, it interacts with the solid structures in ways that are often unpredictable, as it is forced into tortuous pathways determined by the pore structure of a given porous substance. Complexity grows even greater in cases involving turbulence—a swirling, chaotic motion of the gas itself. Predicting how gases behave in such environments is essential for improving efficiency, reducing energy consumption, and ensuring safety across a range of applications.
To study these effects, researchers at the Chair for Thermal Turbomachines and Aeroengines of Ruhr University Bochum, developed and applied advanced numerical simulation techniques capable of resolving fine-scale details of turbulent flow through porous materials. The research activities were carried out by Wojciech Sadowski and Mohammed Sayyari, under the supervision of prof. Francesca di Mare. The main focus was placed on capturing the interaction between the gas in the free-flowing region above the porous material and the gas moving within the porous medium itself.
The research involved replicating a packed bed similar to those used in industrial chemical processes. Such an approach requires immense computation effort as both the geometrical complexity and the swirling flow features have to be accurately represented in the simulations. To fullfil this objective, the computing power of SuperMUC-NG supercomputing system was used in the context of the project „Effect of turbulence modeling on the fuel mixing and reaction in an array of cubes. Work in the context of Subproject C6, SFB/TRR287 BULK-REACTION” (id number: pn73gi).
Figure 1: A model packed bed reactor consisting of an array of spherical particles stacked on top of each other (both pictures differ by viscosity of the flowing gas, with the left being more viscous; higher value of Reynolds number Rep denotes lower velocity). The magnitude of the velocity of a gas flowing through the reactor is visualized by a color mapping at three different planes. As the Reynolds number increases the patterns generated by the streamlines of the flow (the lines showed on the planes) are becoming more intricate, indicating a stronger swirling motion present in the flowfield. The large recirculation regions present for the lower Rep are also getting smaller, as the unsteady motion above the packed bed increases mixing and redistributes the velocity. Source: [1]
In the simulated setup [1], gas was passed through a structured array of spherical particles. The numerical results were compared with experimental measurements using optical imaging methods. This allowed the researchers to validate their models and confirm that the simulations accurately captured the gas behaviour both inside the packed bed and in the region above it.
Figure 2. The visualisation of vortical motion in the turbulent flow over a porous, permeable wall.
The porous wall is modelled by an array of cubes, allowing the fluid flowing above it to permeate
in between them, however still exerting resistance on the flow by the viscous forces. The eddies
(vortical structures) are represented here by the iso-surfaces of so-called Q criterion, visible as
the „colourful” worms. Each elongated structure identifies one vortex line, which is generated in
the turbulent fluid.
In another simulation considering the gas flowing parallel to a porous substrate [2], the gas was observed to penetrate into the porous material rather than simply flowing over it, with highly fluctuating flow penetrating deep into the porous material. These interactions at the surface of the material were found to influence the overall flow regime, including how momentum and energy are transferred across the surface. Even at moderate flow velocities, the internal structure of the material had a strong influence on the flow patterns [3].
In addition to studying the physics, the research also addressed the mathematical models used to represent gas flow in those systems [2]. When engineers simulate porous media on a large scale, they often use averaged versions of the flow equations. While this simplifies calculations, it can introduce errors if the averaging is not applied carefully, especially in the regions where the morfology of the porous medium varies in space. Newly developed mathematical framework, helps to ensure that models remain reliable even when applied to complex or irregular porous structures.
The results of the research highlight several important points. First, it is essential to consider how the gas interacts with the solid parts of a porous material—not just how it behaves in the open regions. Flow cannot be accurately predicted without accounting for what happens inside the material itself. Second, simplified models must be applied with caution. Errors introduced by averaging or by neglecting certain interactions can lead to inaccurate predictions, especially when designing systems where precision is critical. Third, high-resolution simulations, when validated against experimental data, offer a powerful tool for studying systems that are otherwise too difficult to analyse directly.
Understanding how gases move through porous media is important for a wide range of technologies. It affects reaction efficiency and heat transfer, influences how pollutants spread through soils or how wind flows in urban areas. In aerospace applications, it impacts the performance of thermal protection materials and noise-reduction designs. The ability to model and predict gas flow accurately in these contexts can contribute to more efficient, safe, and sustainable solutions.
[1] Sadowski, W. et al. (2024). Particle-resolved simulations and measurements of the flow through a uniform packed bed. Physics of Fluids, 36, 023330. https://doi.org/10.1063/5.0188247
[2] Sadowski, W. et al. (2023). Large eddy simulation of flow in porous media: Analysis of the commutation error of the double-averaged equations. Physics of Fluids, 35, 055121. https://doi.org/10.1063/5.0148130
[3] Sadowski, W., and di Mare, F. (2024). Turbulent Flow Over a Permeable Wall Under a Range of Reynolds Numbers. Conference paper, Ruhr University Bochum.