Investigation of intrusion effects of a gas diffusion layer into channel cross sections depending on channel parameters of metallic bipolar plates

Abstract

The constructive design of a flow field layout and the channel cross section parameters from a metallic half- or bipolar plate can have a significant influence on the performance characteristics of a fuel cell. Especially the intrusion effect of Gas Diffusion Layers (GDL) under preload conditions leads to a narrowing of the effective flow channel cross section at the active area. In this work, it is assumed that the intrusion effect is dependent on the channel parameters from metallic channel plates. This aspect is investigated experimentally. First, the developed GDL test bench, as well as the test methodology, is explained. Further, a general definition of the channel parameters is shown. Two general equations will be derived, with this it is possible to calculate the channel parameters and channel cross section area analytically. Furthermore, an analytic-empirical intrusion model is derived and validated based on the measurement results. With the intrusion model, it is possible to calculate the effective cross-sectional area with a good approximation. These results can be used for further investigations, for example for flow simulations. Finally, the investigation results of two variation series are listed. At the first series the parameter channel width (CW) and at the second series the channel-to-rib-ratio (CRR) is varied. The considered two measurement parameters are the intrusion in the channel middle and the channel width reduction. It turns out that the investigated both channel parameters influences the GDL intrusion behavior. Particularly, between the channel width parameter and the intrusion in channel middle, a significant dependency can be determined. In the presented experimental series, a certain channel width can be detected, over that the intrusion does not increase any further. Moreover, in the experimental series the dependency between CRR and intrusion in the middle of the channel as well as between CRR and the channel width reduction can be demonstrated. As the CRR increases, both results tend to decrease. In a concluding error analysis, the test results are critically examined and evaluated. In principle, however, the assumption of the dependence between intrusion effect and channel parameters can be sufficiently proved by means of the developed test bench and the test method.