Transport and Reaction Processes in Porous Media
Some of future's key-technologies (fuel-cells, fluidized beds, transport in bio-materials, e.g.) and most demanding environmental problems (sub-surface contaminant transport, CO2-storage) can only be solved when detailed understanding and prediction of flow, transport, and reaction in porous media can be achieved. For example, it is not yet understood how colloids facilitate or hinder contaminant transport in aquatic subsurface systems.
Current interest is on multi-phase flow (gaseous/liquid) including surface tension, evaporation/condensation, colloidal transport, chemical conversion and biological degradation. Additional complexities are involved by precipitation that may result in modifying the pore space itself. Description from first principles is possible only on the pore-scale while it is typically sought on a macro-scale. However, the geometric complexity of the pore space in combination with the non-linearity of the involved sub-processes prohibits easy up-scaling from the pore space to the scales interesting for engineering prediction. Research is needed in the fields of detailed studies of sub-processes, systematic up-scaling methods and physics-based or stochastic multi-scale formulation of relevant conservation laws.
The proposed project aims at developing a consistent simulation framework for transport and reaction processes in porous media. To reach this goal the following steps are planned:
- an accurate quantitative description of multi-phase transport and reaction processes on the pore scale by direct simulation techniques
- experimental validation by micromodels (reaction-on-a-chip) and magnetic resonance imaging allowing for space-time resolved measurements and
- up-scaling to macro-scale using stochastic and suitable localisation techniques that are able to resolve the spatial inhomogeneities.
Two scenarios will serve as test beds for identification of relevant processes, their mathematical-numerical description on pore-scale and identification of suitable up-scaling procedures to the macroscopic scale:
- colloidal transport and dynamics in ground water flow (Manhart) and
- multi-phase flow with phase change in fuel-cells (Rüde).
Accurate and efficient flow solvers are already available to describe single-phase flow in porous media. Extensions will be made for multiphase flows including phase change for scenario 2 and Lagrange particle transport for scenario 1. The results will be assessed by experiments-on-a-chip available at the chemistry department. For scenario 1 up-scaling will be achieved by a stochastic Lagrange-method that allows for spatial inhomogeneities of concentrations (Rank), consistent formulation of mobility of various colloids, and efficient formulation of reaction and precipitation processes (Rentrop).
In subproject Manhart a basic random sphere pack generator has been developed .Pore scale simulations on a large number of random sphere pack domains of different sizes have been performed using in house code MGLET. In each case the distribution of velocity in the domain has been studied and analysed. The probability density of the velocity field has been obtained and fitted to certain functions. In order to validate results an intensive grid study was performed. The project continues with the study of velocity distributions in larger and more complex domains.
|Prof. Dr.-Ing. habil. Michael Manhart (coordination)||Hydromechanics|
|Prof. Dr. Hans-Joachim Bungartz||Scientific Computing in Computer Science|
|Prof. Dr. Ernst Rank||Computing in Engineering|
|Prof. Dr. Peter Rentrop||Numerical Analysis|
|Dr.-Ing. Sheema Kooshapur|
|Dr.-Ing. Quanji Cai|
Partners: FAU Erlangen-Nürnberg (Rüde), TUM chemistry department
Poster (PDF) - Sheema Kooshapur
- Quanji Cai, Finite Cell Method for Transport Problems in Porous Media, Dissertation, 2013
- Sheema Kooshapur, Modelling dispersion on the pore scale based on the velocity distribution function, Dissertation, 2016