Pretty Porous

In-depth text

In the context of mobility change, the fuel cell is becoming more and more important in the automotive world. This technique allows for emission-free driving on a local level. For heavy vehicles in particular, it represents an attractive alternative to battery-powered vehicles.

The PEM (Polymer Electrolyte Membrane) fuel cell consists of multiple porous layers with different characteristics. To remain competitive, high efficiency and durability must be achieved at low cost. Simulations can help us to research the porous layers.

Efficient methods must be developed to describe and model the porous materials, the transport, and interactions of the porous layers. A fuel cell stack, used for example in a vehicle, can be divided into multiple scales (stack scale, full cell scale, representative volume inside the cell, pore scale). The transport processes in the fuel cell takes place on different scales. However, this method is not suitable for the simulation of complete cells or even entire fuel cell stacks due to the computing effort involved. For a precise representation, however, it is important to consider the relevant effects of the small-scale phenomena also on coarser scales. Therefore, intelligent methods must be found. However, to do this, the relevant effects must first be identified. To describe the interaction of the two-phase water-air-transport at the interface between the GDL (gas diffusion layer) and the gas distributor, in this project a pore network model is used. The transport processes are described on the pore scale. A major challenge here is the change of wettability from the hydrophobic GDL to the hydrophilic gas distributor and the resulting occurring phenomena. In addition to modelling on the pore scale, the frequency of the phenomena occurring must also be analyzed. With this information, a stochastically based model can finally be transferred to the full cell scale.