Selective Oxidation of PEMFC Catalyst Supports: Overcoming the Trade-Off between Kinetics and Mass Transport Limitations

Abstract

Widespread commercialization of PEM fuel cells and their cost parity with incumbent propulsion technologies depend in no small part on further reductions in platinum requirements (<0.125 gPt kW-1 rated). This even increases in importance as manufacturers move towards high volume production, because raw materials (such as Pt) do not benefit from economies of scale.[1] Lower cathode loadings (<0.1 mgPt cm-2), however, disparately impact high current density performance due to local mass transport limitations of oxygen and protons. Recent research efforts have demonstrated the key role of catalyst support structure and Pt nanoparticle location to mitigate these transport-related losses.[2-3] Typical high surface area carbons (HSCs) such as Ketjenblack host Pt nanoparticles in internal pores, resulting in obstructed gas diffusion towards the catalytically active sites. In contrast, Pt deposition onto carbons without appreciable mesoporosity (e.g. Vulcan) yields nanoparticles on the external surface of the primary carbon particle, where intimate contact with the ionomers’ sulfonate endgroups depresses the oxygen reduction reaction (ORR) kinetics due to ionomer poisoning. To overcome this trade-off, Yarlagadda et al. proposed an intermediate carbon morphology with accessible pores hosting Pt nanoparticles, shielding them from direct ionomer contact to avoid ionomer poisoning while impeding oxygen transport to the Pt surface as little as possible.[4] In this contribution, we systematically investigate an oxidative heat treatment suited to obtain such accessible carbon supports. We demonstrate that this methodology can selectively remove up to 14 wt% of carbon and analyze the concomitant changes to pore morphology using N2 physisorption. Catalysts with pristine and modified supports are fabricated into low-loaded membrane electrode assemblies (MEAs) and tested in 5 cm2 single cell PEM fuel cells. While our oxidative treatment leads to slightly reduced ORR mass activities, high current density performance in H2/air improves substantially compared to the pristine support (up to 90 mV at 2 A cm-2, Figure 1). We ascribe this voltage gain to significantly reduced oxygen mass transport resistance, in accordance with better pore accessibility.Lastly, we evaluate catalyst susceptibility to start-up/shut-down (SUSD) degradation. In SUSD events, H2/air fronts passing through the anode compartment induce oxidation currents in the cathode layer, leading to carbon corrosion and, ultimately, to a collapse of the electrode structure. It is thus conceivable that their selective pre-oxidation might make carbon supports more prone to this failure mode. Our experiments suggest that while higher degrees of oxidation modestly accelerate SUSD degradation, appreciable differences are only observed after both pristine and modified catalysts have already suffered SUSD-induced voltage losses that constitute end of life (50-100 mV loss versus initial performance). The performance gain conferred by the oxidative treatment is therefore not outweighed by lower resilience towards SUSD. ACKNOWLEDGMENTS We gratefully acknowledge funding from the Swiss National Foundation under the funding scheme Sinergia (project grant number 180335). We also express our gratitude to Ana Marija Damjanović for her help with the Start-Up/Shut-Down protocol and Mohammad Fathi Tovini for valuable consultations regarding thermogravimetric analysis. REFERENCES [1] C. S. Gittleman, A. Kongkanand, D. Masten, W. Gu, Current Opinion in Electrochemistry 2019, 18, 81-89.[2] G. S. Harzer, A. Orfanidi, H. El-Sayed, P. Madkikar, H. A. Gasteiger, Journal of The Electrochemical Society 2018, 165, F770-F779.[3] Y.-C. Park, H. Tokiwa, K. Kakinuma, M. Watanabe, M. Uchida, Journal of Power Sources 2016, 315, 179-191.[4] V. Yarlagadda, M. K. Carpenter, T. E. Moylan, R. S. Kukreja, R. Koestner, W. Gu, L. Thompson, A. Kongkanand, ACS Energy Letters 2018, 3, 618-621. Figure 1