With a recent shift in focus from light-duty to heavy-duty vehicle applications, the durability and performance targets for polymer electrolyte fuel cells (PEFCs) have become even more stringent and challenging to achieve; for example, a net power output of 2.5 kW/g-PGM (1.07 A/cm2 current density) at 0.7 V is required after 25,000 hour-equivalent accelerated stress test (AST) [1]. Generally, different ASTs are designed to evaluate Pt electrocatalyst and support durability; for example, a square-wave (SW) AST with H2/N2 between 0.6 V to 0.95 V vs. reversible hydrogen electrode (RHE) targets Pt catalyst and a separate triangular-wave (TW) AST with a higher potential range (1 – 1.5 V vs. RHE) assesses the durability of carbon-based supports [2]. Recently, many studies [3]–[7] have revealed the heterogeneous nature of cathode catalyst layer degradation under different ASTs. Fuel cell electric vehicles are typically subjected to an enormous number of load/unload and start/stop cycles during regular operation. Though there is a considerable body of literature discussing the isolated effects of load/unload and start/stop cycles on durability and performance, there is a dearth of studies exploring the effects of cascading load/unload and start/stop cycles, a scenario more relevant to actual automotive conditions. In the present study, a segmented cell along with ex-situ Pt particle size mapping (through micro-X-ray diffraction) is used to understand the effects of cascading a DoE square-wave AST (0.6 V to 0.95 V vs. RHE) and a triangular-wave (1 – 1.5 V vs. RHE) on the durability and performance of a PEFC.The membrane electrode assemblies (MEAs) are subjected to two cascade ASTs. The first one involves 15k cycles of SW AST followed by 2.5k cycles of TW AST (referred to as Pt-followed-by-carbon, PtfCC, AST) and the other one involves 2.5k cycles of TW AST followed by 15k cycles of SW AST (referred to as Carbon-followed-by-Pt, CCfPt, AST). Multiple in-situ and post-mortem ex-situ techniques are used to obtain particle size distribution and spatial degradation profiles. Preliminary results indicate that the performance losses at the end-of-life (EOL) depend on the cascade order; PtfCC AST leads to higher losses compared to CCfPt AST in wet polarization conditions. The local current distributions are also strongly impacted by the cascade order during the aging process. Figure 1 shows the polarization performance of the samples at the beginning, middle (following the first AST), and the end-of-life of cascade ASTs.
Read full abstract