zur Startseite

Research - Dept. of Hydromechanics and Modelling of Hydrosystems

Multiphase-, multicomponent processes within gas diffusion layers in fuel cells and their interactions with channel flow
Project manager:apl. Prof. Dr.-Ing. Holger Class, Prof. Dr.-Ing. Rainer Helmig
Research assistants:Dipl.-Inf. Andreas Lauser
Duration:1.10.2008 - 30.9.2011
Funding:International Research Training Group NUPUS
Continuity project of: Modellierung der Mehrphasenströmung auf der Kathodenseite der Wasserstoff-Brennstoffzelle

This project is part of the research area:
Model coupling and complex structures


Non-fossil energy carriers are important prerequisite to make the dawning age of renewable energy sources sustainable. A promising energy vector for a wide range of applications are hydrogen powered polymer electrolyte membrane (PEM) fuel cells. PEM fuel cells are currently under intense investigation, and many break-thoughs have already been achieved. Despite this, it is still necessary to improve their performance in terms of power density, durability and manufacturing costs before they are competitive with conventional fossil based technologies in mainstream applications.

For the power density, the relatively slow kinetics of oxygen reduction at the cathode is the limiting factor. Thus, optimal oxygen supply of the cathodic reaction layer is essential for performance. On one hand, oxygen supply is constrained by liquid product water, but on the other hand, some liquid water is required to ensure sufficiently high proton conductivity of the polymer membrane. For this reason, optimizing PEM fuel cell in terms of power density requires a detailed understanding of the physical phenomena which influence the water management trade-off.

Many fundamental questions regarding the cathodic diffusion layer are still to be answered. Here, a crucial issue is the wettability of the porous skeleton: Commonly used materials are often based on carbon fiber structures which are partially hydrophobized using PTFE (Teflon) because the hydrophobic material enhance drainage of liquid product water. However, it has been observed that under operating conditions, parts of the diffusion layer are hydrophilic and retain liquid water. This effect significantly changes hydraulic properties and is extremely challenging for numerical models.

Another difficulty for numerical models is describing the boundary conditions at the interface between the gas channels (i.e. gas distributors) and the diffusion layer. The purpose of the gas distributors are two-fold: On one hand, the gas channels supply oxygen to the diffusion layer, on the other hand they take up product water in form of steam and as a liquid. The design of the gas distributors can be conventional or inter-digitated: A conventional gas distributor provides an equally pressurized gas flow along the interface to the diffusion layer. Consequently, the gas flow is mainly diffusive. In contrast, in an inter-digitated configuration, pressure of the gas on both sides of the shoulder is different and the gas is advectively forced through the diffusion layer. In both cases, water saturations within the diffusion layer as well as water flux into the gas channel over its interface are dynamically depending on the cell's operation conditions.

This is why an important aspect of numerical models for water management in fuel cells are adequate interface conditions for the fluxes between the porous gas diffusion layer and the gas distributor. The primary goal of this project is thus to study the suitability of various continuum scale approaches for the description of the interface conditions.