(768a) Effect of Hydrophobic Polymer Treatments on the Capillary Properties of Gas Diffusion Layers

One of the principle appeals of hydrogen energy and the associated polymer electrolyte fuel cell technology is that only pure water is produced as exhaust. From a technical perspective, however, the production of water inside the cell is actually one of the principle challenges. As current densities are pushed higher the production of water leads to the formation of liquid water. The presence of liquid water in the porous electrode and gas diffusion layer backing (GDL) presents a major obstacle for mass transfer to the catalyst layer, leading mass transfer limited operation and a reduction in the power output of the cell. A typical remedy for this problem is to impregnate the GDL with a hydrophobic polymer, which is supposed to promote water rejection from the electrode and maintain clear gas pathways. Although this treatment is known to improve performance, the actual results of these treatments on the capillary properties of the GDL are not well understood as no suitable characterization technique has been available. Consequently multi-physics modeling of the coupled transport processes inside the cell has been struggling to incorporate the two-phase flow phenomena and optimization of the electrode wetting properties has been limited to trial and error approaches. To address this shortcoming an apparatus has been developed to study the air-water capillary curves of GDL, which are very thin (<0.5mm) and highly porous (> 80%) papers made from carbon fibers (~ 10 micron diameter). The developed apparatus allows the injection of water into a completely dry sample, as well as repeated full injection and withdrawal loops and partial or internal loops. This array of tests has yielded a wealth of information about the behavior of water inside these materials, such as history dependent wetting and estimates of the residual phase saturations. A wide variety of carbon papers have been tested with varying amounts of hydrophobic polymer additive (0 ? 30 wt%). The results suggest that increased polymer content significantly increases the breakthrough pressure of some materials, while tests on other materials have shown no effect of increasing polymer content, suggesting that application technique may be critical to achieve good hydrophobicity. Access to such detailed information about the capillary properties of the electrode will open many new avenues for optimizing and studying GDLs.