Fuel cells are being increasingly deployed in various applications, however like all emerging technologies, its cost still needs to be lowered for wider commercialization. The most expensive component is the noble platinum (Pt) catalyst, so reducing the amount of Pt remains an ongoing effort. There are two ways to approach this problem: find a cheaper alternative^1,2 or use less Pt.^3,4 Regarding the former, most candidates are less active than Pt, so require higher loading and thicker catalyst layers, while the latter case, the lower loading must be compensated by higher accessibility and utilization. In both cases it is essential to design improved catalyst layer (CL) structures to provide maximum accessibility of reactant to the catalytic sites.⁵ Characterizing the nanoporous structure of the CL in terms of transport processes is essential to understanding how to design a better layer. The key transport parameter in fuel cell operation is the effective diffusivity through the CL structure since diffusion is the primary mode of reactant transport. Despite the importance of characterizing the effective diffusivity of the CL, the thinness of the CL combined with the fact that it’s not self-supporting, has hindered the development of well-established, easy to apply, and standardized tool and only a limited number of technique is available.⁶

There are several requirements in designing the proper ex-situ effective diffusivity measurement experiment for the CL. First, any surface in contact with the sample must be scratch-free to eliminate additional diffusion pathways. Second, the need for sealing should be avoided as it is nearly impossible to properly seal such thin (~10µm) layer. Lastly, the experiment should be designed so that it is performed within a reasonable time. This work presents a new method building on previous work^7,8, but using a radial geometry instead. The main advantage of the method is that it requires no sealing around the edge of the sample, therefore easily applied even to ultrathin porous layers. Inside a cylindrical chamber, the catalyst layer sample is placed between two circular pedestals and the oxygen concentration is measured at the center as shown in Figure 1(a). The sample is initially flushed with air, then nitrogen gas is flowed past the sample perimeter at high flow rate to ensure instant change in boundary condition. The depletion of the oxygen concentration at the center of the sample is measured and recorded as a function of time. The effective diffusivity is determined by fitting the analytical solution of the Fick’s second law in cylindrical coordinates to the transient oxygen concentration profile.⁹ The technique was validated against open air and known GDL materials.^7,8 In the present study, this novel experimental technique was applied to ultrathin porous layers fabricated with different ink formulae to explore the impact of morphology, and strong differences were seen.

References:

1. M. Lefèvre, E. Proietti, F. Jaouen, and J.-P. Dodelet, Science (80-. )., 324, 71–74 (2009)
2. E. Proietti et al., Nat. Commun., 2, 416 (2011)
3. S. Martin, B. Martinez-Vazquez, P. L. Garcia-Ybarra, and J. L. Castillo, J. Power Sources, 229, 179–184 (2013)
4. S. Shukla, K. Domican, and M. Secanell, ECS Trans., 69, 761–772 (2015).
5. Y. Tabe, M. Nishino, H. Takamatsu, and T. Chikahisa, J. Electrochem. Soc., 158, B1246 (2011)
6. Z. Yu and R. N. Carter, J. Power Sources, 195, 1079–1084 (2010).
7. R. Rashapov, F. Imami, and J. T. Gostick, Int. J. Heat Mass Transf., 85, 367–374 (2015)
8. R. R. Rashapov and J. T. Gostick, Transp. Porous Media, 115, 1–23 (2016).
9. J. Crank, (1975) "The Mathematics of Diffusion".