Membrane failure is an important factor for the overall durability of polymer electrolyte fuel cells. Lifetime limiting failure occurs when the membrane loses its ability to reliably separate the reactant gases such that potentially combustive conditions may arise due to mixing of hydrogen and oxygen. While diffusive gas crossover is regularly present at benign rates, critical leak rates require significant convective fluxes that can only occur in the presence of large holes and cracks that span the full thickness of the membrane. Membrane degradation and damage development during fuel cell operation takes place through complex interplay between chemical degradation due to radical attack of the ionomer and mechanical degradation due to hygrothermal variations under mechanically constrained conditions. The combined action of chemical and mechanical degradation, which is difficult to avoid during dynamic fuel cell operation, is known to drastically accelerate the accumulation of membrane damage and be the primary cause of ultimate failure [1]. However, the process through which failure occurs is only partially understood.

Membrane failure analysis has historically been performed by 2D imaging techniques such as optical and electron microscopy, which is limited to surface views of the electrodes and cross-sectional snapshots of the internal MEA structure. These techniques are also destructive in nature, demand tedious sample preparation with risk of artifacts, operate under vacuum, and/or are suitable only for electrically conductive samples. Our group recently proposed the use of X-ray computed tomography (XCT) to overcome these limitations and open up a novel non-destructive 3D characterization method for imaging of membranes inside an MEA [2]. This approach leverages recent advances in laboratory-based XCT technology to non-destructively acquire 3D images of membrane damage features and to develop a unique 3D failure analysis framework for fuel cell membranes [3]. This methodology is systematically applied to fuel cells subjected to pure chemical, pure mechanical, and combined chemical and mechanical membrane degradation in order to reveal the different types of membrane failures that may occur during fuel cell operation and assign their presence to specific degradation mechanisms. Novel discoveries made using this technique include distinct identification of I, Y, and X branched cracks (Figure 1), formation of exclusive membrane cracks, interaction of membrane and catalyst layers cracks and delamination sites, and electrode shorts due to excessive chemical membrane degradation. However, regular 3D visualization of membrane failures cannot reveal the root cause of each damage feature. Therefore, our recent efforts have focused on further exploiting the non-destructive nature of XCT scans to perform 4D membrane visualization by means of 3D scans at different points in time [4]. This allows us to track the propagation of membrane damage during the degradation process and also enables back-tracking of failure modes to determine the actual root cause by inspecting same-location scans performed at earlier times. Using this methodology, we have determined under which circumstances catalyst layer cracks may propagate into membrane cracks and vice versa. Overall, 3D visualization of membrane degradation and failure by XCT is shown to be a highly promising method to help understand complex failure modes and degradation mechanisms in fuel cells.

Acknowledgements:

Funding for this research was provided by the Natural Sciences and Engineering Research Council of Canada, Canada Foundation for Innovation, British Columbia Knowledge Development Fund, and Ballard Power Systems through an Automotive Partnership Canada grant. This research was undertaken, in part, thanks to funding from the Canada Research Chairs program.

References:

1. Lim, C., Ghassemzadeh, L., Van Hove, F., Lauritzen, M., Kolodziej, J., Wang, G.G., Holdcroft, S. and Kjeang, E., J. Power Sources 257 (2014) 102-110

2. White, R.T., Najm, M., Dutta, M., Orfino, F.P. and Kjeang, E., J. Electrochem. Soc. 163 (2016) F1206-F1208

3. Singh, Y., Orfino, F.P., Dutta, M. and Kjeang, E., J. Power Sources 345 (2017) 1-11

4. White, R.T., Wu, A., Najm, M., Orfino, F.P., Dutta, M. and Kjeang, E., J. Power Sources 350 (2017) 94-102

Figure 1. Representative reconstructed false color tomogram of an MEA, showing planar, cross-sectional, and 3D virtual views of an X-shaped membrane crack and adjacent catalyst layer cracks (M = membrane; ACL = anode catalyst layer; CCL = cathode catalyst layer).