, the water flow continued until almost the entire water column was drained. Moreover, it was observed that in a constant pressure head, the flow rate increased by time till a steady condition was reached. In an actual fuel cell, water is produced constantly As such, in this experiment the flow rate of 2 ml min−1 was maintained and the maximum pressure, Pmax, was recorded. To investigate any hysteresis effects, the tested samples were dried as described previously. The same experiment is then repeated on the dried GDLs, the average values of the permeation threshold are listed in Table 2.
The measured values for the permeation threshold are plotted as a function of GDL thickness in Fig. 4. It can be seen that thinner GDLs pass water at a lower pressure, as expected. The samples with 5% PTFE loading were dried several times and retested using the pressurized membrane test setup. The relative reduction in the permeation threshold in the first and the second trials, (Pc2 − Pc1)/Pc1, for all of the samples was almost similar. Interestingly, the number of trials for drying and testing more than twice did not affect our results significantly and the hysteresis effect was more pronounced in the first two trials. The reason for this hysteresis is not clearly known. However, based on previous studies, surface degradation, microstructural deformation, and existence of nano-size water droplets are the main reasons for changes in the water permeation properties of GDL