Invited: Characterization of Polymer Electrolyte Fuel Cells with Ultra-Low Catalyst Loadings

Platinum-group metal (PGM) electrocatalysts are utilized in PEFCs to catalyze the kinetics of the desired reactions.  In order to reduce cost, membrane-electrode assemblies (MEAs) with ultra-low catalyst loadings (ULCLs) are being developed (i.e., ≤ 0.1 mg/cm² of PGM on the cathode, and lower anode loadings).  However, obtaining acceptable performance with these ULCL MEAs can be problematic, especially at high current densities.

Recent observations indicate that mass-transport losses in the cathode catalyst layer of a PEFC increase as the catalyst loading is decreased; transport losses are often significantly higher with ULCL MEAs than in MEAs with conventional catalyst loadings [1].  Higher mass-transport losses are not unexpected because the required flux rate to each catalyst site necessarily increases as the total number of catalyst sites decreases.  However, the locations of the dominant transport losses, and the length scales, tend to be substantially different in ultra-thin catalyst layers than in thicker catalyst layers.  Transport losses through the thickness of the catalyst layer diminish as the amount of catalyst is reduced.  Conversely, transport losses through films on the catalyst surfaces are expected to increase.  Understanding the locations and types of transport losses in these ULCL electrodes is critical to obtaining optimal MEA performance.

Obtaining acceptable performance with ULCL MEAs over the full range of desired operating conditions can also be challenging.  For example, ULCL MEAs are typically more sensitive to the relative humidity (RH) of the reactant gases than thicker catalyst layers.  This RH sensitivity may be due to the reduction in pore volume within the electrode, and/or may result from changes in the dominant transport mechanisms, as described above.  Additionally, it has been shown that nano-structured-thin-film (NSTF) catalysts have many potential benefits [2], but PEFCs with NSTF-based electrodes are unfortunately also extremely temperature sensitive [3].  This unusual temperature sensitivity may be due to a variety of factors, including the fact that proton transport in a NSTF layer is necessarily different than in a conventional catalyst layer due to the absence of ionomer in direct contact with much of the catalyst.  Developing an understanding of the mechanisms responsible for these losses in various ULCL MEA structures over a broad range of operating conditions is a major goal of this work.

This talk will focus on characterization methods that are generally based on theoretically-predicted limiting cases for different performance-loss mechanisms.  These limiting cases are then compared to PEFC performance data measured under different conditions, or MEAs with different electrode structures, in order to provide insight into what are the dominant mechanism(s) responsible.  Changes in cell performance that result from changes in operating conditions, such as RH or temperature, can readily utilize the same diagnostic methodology developed to investigate durability losses in PEFCs; since whether the performance changes result from load hours or from operating conditions is irrelevant to this methodology [4].

In addition to these previously-described diagnostic tools, analogous methods can be utilized that are based on results from recent PEFC models, which explicitly consider the impact of varying catalyst loading and also include various transport modes, Ohmic effects, and reaction kinetics [5].  Since different transport mechanisms should have a different dependence on electrode geometry, a comparison of various limiting cases can be used to provide mechanistic insights, analogous to the dependence on operating conditions.

The understanding gained from these characterization and diagnostic methods can then be used to predict changes in the design of MEAs that should result in improved performance over a broad range of operating conditions.  This talk will therefore conclude with some recommendations on electrode designs that could be beneficial in advanced MEAs.

Acknowledgements

Thanks to the organizers of the PEFC 14 Symposia for the invitation to present on this topic.  The author would also like to thank his many fuel-cell collaborators at UTC (both past and present), and especially Dr. Robert Darling for his modeling work and discussions, which has had a major influence on the results to be presented here.  Funding from U.S. Department of Energy, EERE’s Fuel Cell Technologies Office under contract numbers DE-AC02-05CH11231 and DE-AC02-06CH11357 has enabled much of the recent work at UTRC on PEFCs with ULCLs, and is also gratefully acknowledged.

References

1. N. Nonoyama, et.al., J. Electrochem. Soc., 158, B416 (2011).
2. M. Debe, 2011 DOE Fuel Cell Program Review, FC001 (2011).
3. M. Perry, C. Shovlin, and R. Zaffou, ECS Meeting Abstract, MA2012-02: 1586 (2012).
4. M. Perry, R. Balliet, and R. Darling, Modern Topics in Polymer Electrolyte Fuel Cell Degradation, M. Mench, et.al., Eds; Elsevier, Denmark, p.335 (2011).
5. W. Yoon and A. Z. Weber, J. Electrochem. Soc., 158, B1007 (2011).