PTFE Distributions in Diffusion Media of PEFCs and Related Effects on Water Transport

Abstract

The diffusion media (DM) of polymer electrolyte fuel cells (PEFCs) typically consists of a macro-substrate (or gas diffusion layer, GDL) coated with a micro-porous layer (MPL). The GDL portion is manufactured from carbon fibers/binder and is often treated with polytetrafluoroethylene (PTFE) to increase the hydrophobicity. Similarly, the MPL is manufactured from a slurry containing carbon nanoparticles and PTFE. While the distribution of PTFE undoubtedly influences water transport within the DM, the actual 3-D distribution of PTFE is extremely difficult to capture using a single, standalone device. Therefore, the objective of this work is to examine the impact of PTFE distribution on water transport by generating various PTFE distributions within the measured 3-D structures of DM materials.A commercial DM sample was previously analyzed using X-ray computed tomography (XCT) and focused ion beam-scanning electron microscopy (FIB-SEM) to capture the 3-D structures of the GDL and MPL regions, respectively [1]. In this study, three different PTFE distribution cases are generated on the internal surface of the measured GDL structure: i) random PTFE surface pixels; ii) random PTFE surface clusters; and iii) wetting of PTFE into GDL crevices to mimic the PTFE treatment process used during manufacturing. Similarly, cases (i) and (ii) are generated in the measured MPL structure. In each case, PTFE is assigned to existing solid pixels so that the structure of the measured sample is preserved.The well-established pore morphology or full morphology (FM) model [1,2], used for determining capillary pressure-saturation behavior in uniformly wetting pore structures, was recently extended to account for locally different wetting surfaces in a structure [3,4]. The extended-FM approach is applied here in 3-D to capture the impact of the PTFE distributions on water transport in the DM, by assuming one contact angle for the carbon fiber/binder surface and a different contact angle for the hydrophobic PTFE surface. Two uniform surface conditions (i.e., traditional FM approach) are also considered for comparison: one in which an average contact angle for carbon fiber/binder and PTFE is assumed, and another in which PTFE is neglected (i.e., untreated GDL sample). The capillary pressure-saturation results are compared with experimental and modeling studies available in literature.