An essential part of proton exchange membrane fuel cells (PEMFCs) is the gas diffusion layer (GDL), which provides pathways for by-products to be removed from PEMFCs. One of the main properties of GDLs is porosity. The two widely used experimental methods for finding the porosity of GDLs are mercury intrusion porosimetry (MIP) and method of standard porosimetry (MSP). In addition to these methods, the porosity of GDLs can be calculated based on the high resolution 3D images that are acquired using X-ray microtomography (μXCT) as shown in recent studies (e.g., [7,12]). Despite the general success of using μXCT to measure GDL porosity, different porosity values have been reported for similar GDLs.
These variations are due to different assumptions made for determining the surface of the sample, and hence, its external dimensions. In this research, current methods used for calculating porosity of GDLs from μXCT images are discussed, and a new surface identification method based on a rolling ball algorithm is introduced. The main advantage of this new method is that variations in surface topology or roughness are taken into account when calculating porosity. The new method is not only applicable to GDLs, but can be applied to characterize a wide range of highly porous media.