Gas flow in fuel cell porous electrodes is usually modelled with Darcy’s law, which requires the definition of a resistance constant for the material. This can be done directly via experimentation or indirectly via numerical tuning to fit experimental data on cell behaviour. Both methods lack generality, as they do not take into account the particular porous structure of each electrode. In the present work, a numerical procedure for calculating the resistance constant for a given porous structure is presented. This procedure is based on Lattice Boltzmann models, which treat the problem from a microscopic point of view, reproducing collisions between fluid molecules and solid particles. It can be demonstrated that under certain hypotheses, these models yield Navier-Stokes equations on a macroscopic scale, hence obeying fluid mechanics laws. Here the flow in a set of thirty randomly generated porous structures was analyzed, thus obtaining a distribution of values for Darcy’s constant. The analysis was repeated for ten different pressure gradients applied to a portion of the electrode and for three different volume porosities. The results showed that, for a given volume porosity, the value of Darcy’s constant is strongly affected by the material porous structure. On the other hand, the mean value of resistance remained almost constant while varying the applied load, thus correctly reproducing the linear dependence between velocity and pressure gradient, as stated by Darcy’s law. As fuel cell models are a great help in designing and predicting component operation, a further analysis was carried out in order to study the influence of the electrode resistance constant on cell performance prediction. The Lattice Boltzmann model was used to obtain resistance data characteristic of fuel cell electrodes, and the results were implemented in a one dimensional fuel cell model. The simulations showed that the variation of Darcy’s constant does not significantly affect the prediction of the cell polarization curve, while a significant effect was found on the prediction of the exact operating point on the polarization curve. In conclusion, if accurate modelling of a fuel cell is required, great care must be taken in evaluating the electrodes resistance constant. The procedure presented here, coupled with a non destructive tomography scan of the electrode structure could greatly help in refining existent fuel cell models.

This content is only available via PDF.
You do not currently have access to this content.