Anders Christian Olesen

In this work, a non-equilibrium formulation of a compression dependent water uptake model has been implemented in a two-dimensional, two-phase, multi-component and non-isothermal high pressure PEM electrolysis model. The non-equilibrium formulation of the water uptake model was chosen in order to account for interfacial transport kinetics between each fluid phase and the perfluorinated sulfonic acid membrane. Besides modeling water uptake, the devised membrane model accounts for water transport through diffusion and electro-osmotic drag in the electrolyte phase, and hydraulic permeation in the liquid phase. Charge transport and electrochemistry are likewise included. The obtained model is validated against experimental measurements. In order to investigate the effect of membrane compression, a parametric study is carried. Results underline that the predicted water uptake and cell voltage are highly dependent on the applied water uptake model depending on operation conditions like temperature and the specific surface area of the ionomer in the catalyst layer.

The Fickian diffusivity of water in Nafion has been shown to exhibit a local maximum. In the present research effort it is shown that this spike vanishes if the equilibrium water content and chemical diffusivity are modeled carefully. Further, it is shown that permeation experiments falsely can generate a local maximum, if a wrong equilibrium water content relation is used. Finally, in order to study the obtained diffusivity model a parameter variation is carried out using a one-dimensional steady-state model. The effect of diffusivity model, surface roughness and water content driving force is studied.

A fully coupled three-dimensional, steady-state, two-fluid, multi-component and non-isothermal DMFC model has been developed in the commercial CFD package CFX 13 (ANSYS inc.). It accounts for the presence of micro porous layers, non-equilibrium phase change, and methanol and water uptake in the ionomer phase of the catalytic layer, and detailed membrane transport of methanol and water. In order to verify the models ability to predict methanol crossover, simulation results need to be compared with experimental measurements under different current densities along with air and methanol stoichiometries. Methanol crossover be indirectly measured based on the combined anode and cathode exhaust CO2 mole fraction and by accounting for the CO2 production at the anode as a function of current density. This approach is simple and assumes that all crossed over methanol is oxidized. Moreover, it takes CO2 crossover into account.

A three-dimensional, multicomponent, two-fluid model developed in the commercial CFD package CFX 13 (ANSYS Inc.) is used to investigate the effect of porous media compression on water transport in a proton exchange membrane fuel cell (PEMFC). The PEMFC model only consist of the cathode channel, gas diffusion layer, microporous layer, and catalyst layer, excluding the membrane and anode. In the porous media liquid water transport is described by the capillary pressure gradient, momentum loss via the Darcy-Forchheimer equation, and mass transfer between phases by a nonequilibrium phase change model. Furthermore, the presence of irreducible liquid water is taken into account. In order to account for compression, porous media morphology variations are specified based on the gas diffusion layer (GDL) through-plane strain and intrusion which are stated as a function of compression. These morphology variations affect gas and liquid water transport, and hence liquid water distribution and the risk of blocking active sites. Hence, water transport is studied under GDL compression in order to investigate the qualitative effects. Two simulation cases are compared; one with and one without compression.

A three-dimensional, multi-component, two-fluid model developed in the commercial CFD package CFX 13 (ANSYS inc.), is used to investigate the effect of porous media compression on transport phenomenon of a PEM Fuel cell (PEMFC). The PEMFC model only consist of the cathode channel, gas diffusion layer, micro-porous layer and catalyst layer, excluding the membrane and anode. In the porous media liquid water transport is described by the capillary pressure gradient, momentum loss via the Darcy-Forchheimer equation and mass transfer between phases by a non-equilibrium phase change model. Furthermore, the presence of irreducible liquid water is taken into account. In order to account for compression, porous media morphology variations are specified based on the GDL throughplane strain and intrusion which are stated as a function of compression. These morphology variations affect gas and liquid water transport, and hence liquid water distribution and the risk of blocking active sites. Hence, water transport is studied under GDL compression, in order to investigate the qualitative effects. Two simulation cases are compared; one with and one without compression.