One of the major problems of current proton exchange membrane (PEM) fuel cells is water management. The gas diffusion layer (GDL) of the fuel cell plays an important role in water management since humidification and water removal are both achieved through the GDL. Various numerical models developed to illustrate the multiphase flow and transport in the fuel cell require the accurate measurement of the GDL properties (wettability and surface energy). In a recent study, the capillary penetration technique has been used to measure indirectly the wettability of the GDL based on the experimental height penetration of the sample liquid into the porous sample. In essence, a high resolution microscope/camera was used to detect the rate of penetrated height into the sample GDL. The shortcoming of this type of visualization is that it can only be used for thin hydrophilic GDL samples with zero Teflon loadings. For the thick and high Teflon loading GDLs, there is not enough contrast to detect the height of the penetrated liquid. In the real fuel cells, the GDLs are made of the micro-porous and macro-porous layers with an optimum Teflon loading. Therefore, it is required to develop a new experimental methodology capable of detecting the rate of penetration and as a result the wettability of GDLs samples used in fuel cells. In this paper, the fluorescence microscopy technique is integrated into the experimental setup of the capillary penetration method to improve the contrast between the wetted and non-wetted area. The fluorescence setup uses a powder die, dissolved in the test fluid, which is excited by a concentrated ultraviolet light illuminated in the vertical manner. To acquire the profile images of the penetrated liquid, an optical mirror was used. This new setup has the added advantage of providing a visual representation of the different regimes of penetration (e.g., the fingering effect reported for the pathways of the liquid penetrated into the GDLs) that are defined by the capillary number and mobility ratio of each fluid. Since the GDL samples used in this study are relatively hydrophobic (e.g., with 40% Teflon loadings), the pattern of liquid penetration is not uniform. Thus, an image analysis program was developed to determine the average height of penetration based on the area under the entire wetted area. The general Washburn equation was then used to fit the extracted height data and provide the average internal contact angle for each test liquid.
