Sönke Gößling

Freeze start is a challenge in the commercialization of PEM fuel cells. In this study, ice formation in cell layers is investigated through experiments and simulations. Segmentation of the fuel cell on the test bench allows to determine the local distributions of current density and high frequency resistance over the active cell area. The location and timing of ice formation are analyzed in the experiments. It is shown that the formation of ice lenses can be detected by local measurements of the high frequency resistances. Then, a multiphysical CFD model is built and validated with the measurements and the commonalities and differences between the model results and the experiments are studied. It is shown that the model determines the freeze start behavior very well in wide operating ranges. Together with the findings from the experimental investigations, the model will finally be used to investigate local ice formation in detail.

The polymer electrolyte membrane (PEM) fuel cell model of a commercial software package is presented. The basic performance model is extended by two chemical degradation effects: ionomer degradation and carbon corrosion including platinum oxidation. The ionomer degradation model describes the ionomer mass loss due to hydrogen peroxide formation and subsequent attack of the ionomer by radicals. The carbon corrosion model calculates the carbon mass loss caused by carbon oxidation and the active area reduction due to platinum oxidation. The degradation models are coupled with an agglomerate model of the catalyst layer. The model is validated against measurements on an industrial cell. For these measurements, the cell is equipped with a segmented measuring board, which is used to measure the current density distribution and high frequency resistance of every segment. In order to test the predictability of the model under different operating conditions, measurements for stoichiometry and pressure variations are carried out. Calculated and measured current density distributions of the cell, aged by an accelerated stress test, are compared for the validation of the degradation model. Moreover, 3D simulation results of the fresh and aged cells are analyzed in detail and the influence of operating conditions on fuel cell aging is pointed out.