Chemo-Mechanical Durability of Reinforced Fuel Cell Membranes in the Presence of Metallic Foreign Particles

Abstract

Listen

Translate article

High-throughput production is vital to achieve cost-effective proton exchange membrane fuel cells (PEMFCs) at scale1. The high-volume mass production of PEMFCs using industrial manufacturing machinery may however introduce unwanted material in the final product due to the wear and tear of the machinery. As an example, roll-to-roll processes applied to the manufacture of catalyst coated membrane (CCM) generally use Fe alloys such as stainless steel2 components that could release fine metallic particles or related debris that may affect the product quality3,4. Therefore, it is important to understand the fundamental impact of such non-uniformities on PEMFC performance and durability.This work focusses on the impact of slightly oxidized iron (Fe) and stainless steel 316L (SS316L) micro-particles entrapped between the cathode catalyst layer (CCL) and a GORE-SELECT® Membrane A. Fe cations are considered as Fenton’s catalyst/reagent which enhance harmful radical generation and ionomer attack, known to weaken its mechanical strength and ionic conductivity5. Therefore, membrane electrode assemblies (MEAs) containing solid Fe and SS316L particles were prepared to conduct in-situ PEMFC experiments and identify their impact on membrane degradation and durability. The chosen particles were placed carefully on the bare membrane at selected locations based on the flow field design of the graphite bipolar plates. Later, the assembled MEAs were tested using a small-scale fixture (SSF) fuel cell in a combined chemical and mechanical AST after beginning-of-life (BOL) conditioning and diagnostics6.X-ray computed tomography (XCT) has been shown to be a powerful and non-invasive 3D characterization tool for analyzing membrane degradation7,8. Thus, periodic same-location XCT visualization of the CCL-membrane interface was performed after every 10 AST cycles. According to the Fe Pourbaix diagram9, Fe oxidizes to Fe2+ at 0.77 V and indeed, in this work, complete dissolution of the Fe-50µm particles was observed leaving cavities at the CCL-membrane interface. This dissolution manifested as a distinct black spot in CT imaging, as highlighted by the enclosed red circle in Figure 1. In contrast, corrosion of SS316L particles is usually prevented by a native oxide layer formation during fuel cell operational conditions10. The Fe particle laden AST reached membrane failure within 30 cycles and indicates an escalation of degradation by global membrane thinning. This is substantially lower than the baseline AST without particles. The XCT images for Fe-50µm AST seen in Figure 1 show that severe global membrane thinning occurred in the Fe-50µm MEA, which contributed to a high electrochemical leak detection (ELDT) signal exceeding test failure criteria. In contrast, the SS316L MEA did not reach the threshold failure criteria for the AST duration, however, the average OCV remained lower as compared to the baseline MEA AST. XCT imaging showed that the SS316L- 50µm particles did not dissolve throughout the AST duration and the membrane thinning was more pronounced near the particle rather than globally. Additionally, we have evaluated whether chemically and mechanically mitigated GORE-SELECT Membrane® B could enable more robustness and minimize impact of multiple metallic particles. Analysis of combined chemo-mechanical AST with a larger dimension SS316L-500µm present on mitigated GORE-SELECT® Membrane B suggest that chemical and mechanical mitigation can reduce impact of large particles and enable lifetime response at the level close to baseline. In conclusion, the membrane lifetime and failure mode were found to be strongly dependent on the particle chemical composition and membrane degradation mitigation strategies, which are therefore should be considered at PEM design stages to reduce risks and improve CCM and MEA production quality. Keywords: PEM durability, quality control, cost reduction, XCT Acknowledgments This research was supported by the Natural Sciences and Engineering Research Council of Canada, Canada Foundation for Innovation, British Columbia Knowledge Development Fund, Western Economic Diversification Canada, Ballard Power Systems, and W.L. Gore & Associates. This research was undertaken, in part, thanks to funding from the Canada Research Chairs program. References H. Tsuchiya and O. Kobayashi, Int. J. Hydrogen Energy, 29, 985–990 (2004).J. Chen, H. Liu, Y. A. Huang, and Z. Yin, J. Manuf. Process., 23, 175–182 (2016).A. Phillips et al., Fuel Cells, 20, 60–69 (2020).A. Phillips, M. Ulsh, K. C. Neyerlin, J. Porter, and G. Bender, Int. J. Hydrogen Energy, 43, 6390–6399.J. G. Goodwin, K. Hongsirikarn, S. Greenway, and S. Creager, J. Power Sources, 195, 7213–7220 (2010).Y. Chen et al., J. Power Sources, 520, 230673 (2022).Y. Chen et al., J. Power Sources, 520, 230674 (2022).Y. Singh et al., J. Power Sources, 412, 224–237 (2019).R. Tolouei et al., Phys. Chem. Chem. Phys., 18, 19637–19646 (2016).N. Kumar et al., Int. J. Energy Res., 44, 6804–6818 (2020). Figure 1