Cathode Loading Impact on PEM Fuel Cell Performance Losses in Voltage Cycling Based Accelerated Stress Tests

Abstract

In order to reach the target of carbon neutrality by the end of 2050 set by the European Union, hydrogen technologies will play an important role in supporting the transition to a renewable energy economy, especially for heavy duty applications, such as trucks and airplanes.1 For the latter, hydrogen is a viable alternative fuel compared to kerosene, offering a higher specific energy density as well as zero in-flight CO2 emissions.2 However during operation, proton exchange membrane fuel cells (PEMFCs) suffer from significant performance losses due to the instability of Pt-based catalysts for the oxygen reduction reaction (ORR) via the loss of electrochemically active surface area (ECSA).3 This induces substantial decay in the H2/air performance mainly due to the increasing impact of the local O2 transport resistance (RO2,local) with decreasing cathode roughness factor (rf, in units of cm2 Pt cm-2 MEA), which is especially pronounced for low Pt-loadings.4 Thus, for heavy duty applications requiring long-term durability, higher Pt loadings are required to reduce the impact of RO2,local.In this study, we analyse the H2/air performance loss contributions during voltage cycling based accelerated stress tests (ASTs) using cathodes with different Pt-loadings. ASTs were performed using 5 cm2 MEAs with cathode Pt loadings of 0.2, 0.4, and 0.8 mgPt cmMEA -2 (48.6 wt%, TEC10V50E, Tanaka Kikinzoku K.K. (TKK)). Full characterization of the MEA performance characteristics after different numbers of voltage cycles, was performed by measuring: i) H2/O2 and H2/air polarization curves; ii) cathode rf, determined by cyclic voltammetry and CO stripping; iii) O2 transport resistances (RO2,total) via limiting current measurements; and, iv) proton conduction resistances in the cathode catalyst layer (RH+,cath) via electrochemical impedance spectroscopy (EIS). Voltage cycling was performed under H2/N2 (200/75 nccm) at 80 °C, 95% RH, and ambient pressure using square wave profiles with lower and upper potential limits of 0.60 and 0.95 V, respectively, with hold times of 1 s at each potential.A strong H2/air performance dependency on the cathode rf is visible for the same Pt/C cathode catalyst with varying Pt loading. As shown in Figure 1, the same H2/air performance is reached at similar cathode rf values, independent of number of voltage cycles. This is due to the observation that the RO2,local values only depend on the cathode rf and not on the electrode’s history.Finally, when choosing an end-of-life (EoL) criterion of a more than 20 mV deviation of the measured cell voltages at a given current density from the theoretical ORR kinetics losses (c.f. red line in Figure 1 and gray areas for accepted cell voltage range), this EoL criterion is reached for the differently loaded MEAs at the same cathode rf but after different amounts of aging cycles. For example, at 1.5 A cm2 MEA this 20 mV deviation is reached at a cathode rf of 50-70 mgPt 2 cmMEA -2. However, for the 0.8 mgPt cmMEA -2 Pt/C cathode more than 200.000 cycles are needed to reach this value, compared to only 5.000 cycles for the 0.2 mgPt cmMEA -2 loaded cathode. Consequently, a 4-fold increase of the the cathode Pt loading increases its voltage cycling stability by »40-fold.