The absolute gas permeability of several common gas diffusion layer (GDL) materials for polymer electrolyte membrane fuel cells was measured.Measurements were made in three perpendicular directions to investigate anisotropic properties. Most materials were found to display higher in-plane permeability than through-plane permeability. The permeability in the two perpendicular in-plane directions was found to display significant anisotropy.Materials with the most highly aligned fibers showed the highest anisotropy and the permeability could differ by as much as a factor of 2. In-plane permeability was also measured as the GDL was compressed to different thicknesses.Typically, compression of a sample to half its initial thickness resulted in a decrease in permeability by an order of magnitude. Since the change in GDL thickness during compression can be converted to porosity, the relationship between measured permeability and porosity was compared to various models available in the literature, one of which allows the estimation of anisotropic tortuosity. The effect of inertia on fluid flow was also determined and found to vary inversely with permeability, in agreement with available correlations. The results of this work will be useful for 3D modeling studies where knowledge of permeability and effective diffusivity tensors is required.

No abstract

A pore network model of the gas diffusion layer (GDL) in a polymer electrolyte membrane fuel cell is developed and validated. The model idealizes the GDL as a regular cubic network of pore bodies and pore throats following respective size distributions. Geometric parameters of the pore network model are calibrated with respect to porosimetry and gas permeability measurements for two common GDL materials and the model is subsequently used to compute the pore-scale distribution of water and gas under drainage conditions using an invasion percolation algorithm. From this information, the relative permeability of water and gas and the effective gas diffusivity are computed as functions of water saturation using resistor-network theory. Comparison of the model predictions with those obtained from constitutive relationships commonly used in current PEMFC models indicates that the latter may significantly overestimate the gas phase transport properties. Alternative relationships are suggested that better match the pore network model results. The pore network model is also used to calculate the limiting current in a PEMFC under operating conditions for which transport through the GDL dominates mass transfer resistance. The results suggest that a dry GDL does not limit the performance of a PEMFC, but it may become a significant source of concentration polarization as the GDL becomes increasingly saturated with water.3

The condition of liquid water breakthrough at the cathode of polymer electrolyte fuel cells (PEMFC) is studied experimentally and data on corresponding water saturation and capillary pressure are provided for gas diffusion layers (GDL) with and without a microporous layer (MPL). The data demonstrate that the GDL saturation at water breakthrough is drastically reduced from ca. 25% to ca. 5% in the presence of MPL. This observation is consistent with considerations of invasion percolation in finite-size lattices and suggests an explanation for the role of MPL in improving PEMFC performance at high current densities.

The relationship of capillary pressure to liquid saturation for the water-air fluid pair in two different types of gas diffusion media (GDM) used in polymer electrolyte membrane fuel cell (PEMFC) electrodes is elucidated. It is experimentally demonstrated that GDM samples with and without treatment with poly(tetrafluoroethylene) (PTFE) ubiquitously display permanent capillary pressure hysteresis. Water does not imbibe spontaneously into a dry GDM, neither is it ejected spontaneously from a water-saturated GDM.Rather, positive displacement pressure is required to force both water and air into GDMs, whereas the main effect of adding PTFE is to increase the amount of work required for forcing water into the GDM, and to decrease the work required for water removal.Irrespective of PTFE content, the GDM samples tested are generally shown to behave as materials of intermediate (neutral) wettability. The US Bureau of Mines (USBM) wettability index nevertheless shows that water is the preferentially non-wetting phase in PTFE-treated GDMs and the preferentially wetting phase in untreated GDMs. Water-air capillary pressure curves are found to depend on sample thickness, clearly demonstrating that finite-size effects are important. Finally, compression of the GDM is found to increase the capillary pressures for water injection and decrease the capillary pressures required for water withdrawal. These results should aid the design of GDMs with improved water management properties and the modeling of PEMFC electrodes in general.3