Influence of Flow Rate on Catalyst Layer Degradation in Polymer Electrolyte Fuel Cells

With the U.S. Department of Energy (DOE) 2020 durability target for transportation applications of 5000 hours in mind [1], several studies have been performed to understand the factors influencing load cycle durability of the cathode catalyst layer subjected to various operating conditions. In general, high upper potential limit (> 1 V vs RHE), high temperature (> 90 °C) and high relative humidity (RH) during potential cycling have been shown to exacerbate degradation [2]. It is known that Electrochemical surface area (ECSA) loss occurs by Pt dissolution attributed to particle size growth via modified Ostwald ripening, crystal migration, detachment from carbon support due to carbon corrosion, and precipitation in the membrane by chemical reduction due to hydrogen crossover [3]. Accelerated stress tests (ASTs) are often used to mimic material component degradation similar to real automotive driving conditions. Different cycle profiles have been shown to cause varying rates of degradation depending on the operating conditions. The influence of flow rate of reactant gases on degradation, however, has received little attention. Stariha et al. [4], showed that a relatively high flow rate AST had the highest degradation acceleration factor and resulted in non-uniform decay rates when compared to low flow rate conditions of nitrogen gas (N 2 ) at the cathode. However, this phenomenon has not been further studied in detail. The current study is aimed at elucidating the effect of flow rate on cathode catalyst layer degradation subjected to a standard DOE square-wave AST protocol. Single cell studies with 5-cm 2 active area were performed using Nafion® XL membrane (Ion Power Inc.) and SGL-22 BB gas diffusion electrodes (GDEs) as MEA materials and 1-serpentine flow field. All the MEA samples used in the present study had nominally identical Pt loading and were subjected to a consistent conditioning procedure prior to AST. A DOE square-wave cycling protocol [4] was executed: H 2 /N 2 flows at anode/cathode, voltage cycling between 0.6 V and 0.95 V, and 3 seconds hold at each voltage. 30,000 cycles (total test time of 50 hours) were performed at 100% RH, 80°C, and atmospheric pressure. Three different cathode N 2 flow rates of 20 sccm, 40 sccm and 80 sccm were used during the AST with fixed H 2 flow of 40 sccm at the anode. ECSA loss was tracked using cyclic voltammograms at several intervals during the AST. Complete electrochemical characterization at beginning and th