3D Analysis of PEM Fuel Cell Membrane Cracks Using X-Ray Computed Tomography
Bibliographic record
Abstract
Crack and pinhole formation in polymeric membranes of fuel cells is understood to be caused by combined chemical, mechanical, and/or thermal mechanisms that are active during operation under typical automotive duty cycles [1] . This type of damage leads to gas leakage through the membrane thereby compromising its functionality and overall durability of the fuel cell system. Due to the innermost location of membrane within the fuel cell membrane electrode assembly, it has been impossible to explore the full features and morphology of membrane cracks with conventional 2D imaging techniques such as optical and electron microscopy [2]. Moreover, these techniques are destructive in nature, demand tedious sample preparation with risk of artifacts, operate under vacuum, and/or are suitable only for electrically conductive samples. The X-ray computed tomography (XCT) technique overcomes these traditional limitations and opens up a novel non-destructive 3D characterization method for imaging interior features of an object. Synchrotron facilities have been typically used for 3D imaging but their access is limited and costly. This work leverages the XCT technology using laboratory-based systems developed by ZEISS Xradia® to image membrane cracks in their natural state within the membrane electrode assembly (MEA) and gain a comprehensive insight into their features and characteristics. An end-of-life (EOL) MEA sample, subjected to cyclic open circuit voltage (COCV) accelerated stress test (AST) protocol [3], is analyzed upon failure using the XCT technique and compared with a similar beginning-of-life (BOL) sample. Based on their reach in the through-plane direction, cracks penetrating the entire membrane thickness are categorized into: (i) Exclusive cracks that remain confined within the membrane; and (ii) Non-exclusive cracks that extend into and penetrate through the anode and/or cathode catalyst layers. The very presence of Exclusive cracks (cf.Fig. 1), without any adjoining cracks in the catalyst layers, indicates the likelihood of crack initiation within the membrane. An examination of 9 membrane cracks spread over a 0.88 mm2area reveals that more than 50% of the cracks fall into the Exclusive category. This percentage is found to be significantly higher than the 13% overall interaction of catalyst layer cracks with the membrane cracks. In the in-plane direction, some membrane cracks are found to propagate as a single entity forming a curved I-shape while the others branch out once forming a Y-shape. Distribution of the two shapes is approx. equal among the analyzed cracks. The smaller Y-cracks (<20 μm average branch length), which are likely to be in their initial phase of development, tend to propagate at equal rate in all three directions. The total in-plane crack length in the membrane is found to have a linear relationship with the maximum crack width. The crack width remains almost uniform along its length and tapers sharply at the ends indicating that the crack propagation in the membrane is mechanical in nature caused by the cyclic in-plane stresses. The effect of these stresses seems to have been exacerbated by the observed non-uniform reduction in membrane thickness resulting in stress concentration locations within the membrane. During this analysis, a strong probability for crack development is observed at such locations with membrane thickness found to be approx. 30% lower at the crack sites than its average EOL bulk value. The work reported here is unprecedented from the perspective that, for the first time, a truly 3D view of a membrane crack is achieved; thus, making this comprehensive analysis possible. These results, establish the XCT as a breakthrough approach for reliable failure analysis of fuel cell membranes. Acknowledgements This research is financially supported by Ballard Power Systems and Automotive Partnership Canada (APC). The authors thank Chan Lim, Lida Ghassemzadeh and Erin Rogers for providing samples and technical support. References [1] S. Kundu, M.W. Fowler, L.C. Simon, S. Grot, J. Power Sources. 157 (2006) 650–6. [2] F.H. Garzon, S.H. Lau, J.R. Davey, R.L. Borup, ECS Trans. 11 (2007) 1139–49. [3] C. Lim, L. Ghassemzadeh, F. Van Hove, M. Lauritzen, J. Kolodziej, G.G. Wang, et al., J. Power Sources. 257 (2014) 102. Figure 1
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".