Multiscale Simulation of Adsorption Filters for Industrial Applications

Adsorption is increasingly important in modern filtration because it enables the removal of pollutants, odors, volatile compounds, and chemical contaminants from gases and liquids. For companies such as MANN+HUMMEL GmbH, relevant applications include water and wastewater treatment, air purification, solvent recovery, carbon capture, respiratory protection, and process-gas filtration. Filter performance depends on both contaminant transport through the porous medium and interactions with the adsorbent surface. Accurate modeling of adsorption, mass transfer, flow, and material saturation is therefore essential for predicting efficiency, breakthrough behavior, service life, and pressure drop.

To support simulation-based analysis and design, a new adsorption capability has been integrated into GeoDict. It enables adsorption to be studied in detailed 3D models of porous filter structures and reveals the spatial and temporal distribution of adsorbed species. The approach evaluates the effects of pore geometry, flow conditions, transport properties, adsorbent characteristics, and operating parameters, helping reduce physical prototyping and improve filter design.

The methodology couples fluid flow and mass transport across multiple length scales. Molecular transport is modeled with an Euler–Lagrange particle-tracing approach, while adsorption equilibria are described using Langmuir, Toth, and Freundlich isotherms. A tracer-based method predicts breakthrough in large, non-resolved structures such as cabin air filters and honeycombs, whereas a field-based method represents transient solute concentration in smaller, fully resolved microstructures such as packed beds.

For validation, MANN+HUMMEL GmbH provided a µCT scan of an activated-carbon filter medium, along with material parameters, adsorption isotherms for toluene, and experimental data. The scan was processed in GeoDict to create a 3D model for adsorption and desorption simulations under experimental conditions. Tracer-based toluene adsorption results showed very good agreement with measured breakthrough data, while desorption results were consistent with physical expectations.

The validated approach helps identify critical transport and adsorption effects, including hot spots and poorly utilized regions. By providing insight from the microstructure to the component scale, it supports systematic optimization and more efficient digital development of advanced adsorption-based filtration technologies.