Assessing the Unique Degradation Mechanisms of Hydrocarbon-Based Membranes in Conventional MEA Design Using 4D in-Situ X-Ray Computed Tomography
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The degradation of proton exchange membrane fuel cells (PEMFCs) stands as a critical determinant in evaluating the robustness of fuel cell systems, crucial for the commercial-scale transition to sustainable energy. For zero-emission vehicles, fuel cell stacks need to demonstrate lifespans exceeding 8,000 (light-duty vehicles) and 30,000 hours (heavy-duty vehicles)[1]. For achieving robust PEMFC systems at commercial scale, it is crucial to precisely understand the degradation pathways to effectively pinpoint and mitigate degradation issues. To achieve this, customized small-scale fuel cell fixtures are utilized for accelerated stress testing (AST) and X-ray computed tomography (XCT) in-situ and ex-situ visualization [2,3]. This approach enables comprehensive analysis of membrane electrode assembly (MEA) aging processes. Our group previously employed this approach to identify chemical, mechanical, and chemo-mechanical degradation mechanisms in commonly used perfluorosulfonic acid (PFSA) ionomer membranes[4,5]. In recent years, transitioning from PFSA membranes to hydrocarbon-based (HC) membranes in PEMFCs has gained significant attention. This shift eliminates fluorinated compounds typically found in PFSA membranes, aligning with sustainability goals, and reducing environmental impact. This transition signifies both technological advancement and innovation, driving forward the quest for efficient, cost-effective, and environmentally friendly energy solutions. Despite their promising characteristics, HC membranes are more susceptible to mechanical degradation than PFSA ones[6], presenting unique compatibility challenges due to their distinct chemical composition. Historically, PEMFC systems intentionally developed and optimized for PFSA may face performance issues and premature failure with HC membranes. This research therefore investigates the unique aspects of HC membrane degradation within a conventional fuel cell design with PFSA ionomer catalyst layers, using 4D in-situ XCT to understand chemo-mechanical degradation, enhancing fuel cell technology. Herein, ZEISS Xradia® 520 Versa micro XCT system was used to visualize samples. Repetitive scans were taken at room temperature in a dry state of the MEA. The first set of MEAs consisted of reinforced HC membranes (Pemion-PF1-HLF8-15-X) of 15 µm thickness, spray-coated with PFSA catalyst ink to form catalyst-coated membranes (CCMs), and assembled with gas diffusion layers (Freudenberg H14C15, 190 µm) featuring smooth and crack-free microporous layer. The AST protocol was custom-designed for chemo-mechanical membrane degradation and included open circuit voltage hold for chemical degradation and relative humidity cycling for mechanical degradation, which were applied consecutively. It has been demonstrated that MEAs undergo dimensional changes during wet-dry AST cycles due to alternating conditions of hydration and dehydration[5,7]. It is apparent from the degraded XCT image ( Figures 1a-d ) that cracks were initiated and propagated between the membrane and catalyst layers, with cracks prominently visible near agglomerated electrode particles. Differential swelling between the membrane and catalyst layer, compounded by varying rates of expansion or contraction, resulted in magnified mechanical stresses at the membrane-catalyst layer interfaces. The second set of cells consisted of the same reinforced HC membrane assembled with commercial gas diffusion electrodes (GDEs; 0.5 mg/cm² 60% Pt on Vulcan, Sigracet 22 BB, 215 µm). The existence of interfacial voids ( Figure 1e ) was observed in the pristine sample at both anode and cathode membrane-GDE interfaces, likely due to interfacial adhesion fatigue (IAF)[8]. Uneven compression and interfacial incompatibility during fuel cell assembly results in IAF. When these GDE based MEAs were subjected to wet-dry AST cycles, mechanical stress was induced at the membrane-GDE interfaces[9, 10]. The uneven mechanical stress can aggravate creep propagation in the membrane. Due to creep propagation, membrane thinning at the edges was evident from the XCT images, and resulted in electrode shorting, as shown in Figure 1(f and g) . It is believed that constraints during creep failure hindered membrane's return to original thickness after swelling, causing permanent dimensional changes. The combined effects of cyclic dimensional changes and membrane thinning due to the creep mechanism contributed to performance degradation and reduced durability of GDE-based MEAs. Acknowledgements This work was supported by the Natural Sciences and Engineering Research Council of Canada, Ionomr Innovations Inc., Canada Foundation for Innovation, British Columbia Knowledge Development Fund, Pacific Economic Development Canada, and Canada Research Chairs. References C.S. Gittleman et al., Joule , 5 (2021) 1660–1677. Y. Singh et al., J Power Sources , 412 (2019) 224–237. D. Ramani et al., Electrochim Acta , 380 (2021) 138194. Y. Chen et al., J Power Sources , 520 (2022) 230674. Y. Chen et al., J Electrochem Soc , 170 (2023) 114526. S.H. Mirfarsi et al., Int J Hydrogen Energy , 47 (2022) 13460–13489. A. Sadeghi Alavijeh et al., J Power Sources , 427 (2019) 207–214. X. Huang et al., J Polym Sci B Polym Phys , 44 (2006) 2346–2357. V.A. Sethuraman et al., J Electrochem Soc , 155 (2008) B50. S.H. Mirfarsi et al., Int J Hydrogen Energy , 50 (2024) 1507–1522. 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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 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 | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it