Elucidating the Voltage-Cycling Induced Degradation Mechanism of PEM Fuel Cell Catalysts with Porous Carbon Supports

Abstract

In order to reach the target of climate neutrality by the end of 2050, an increasing interest in the application of proton exchange membrane fuel cells (PEMFCs) for heavy duty vehicles (HDVs), such as trucks and buses has developed recently. However, HDVs require highly stable and durable cathode catalyst materials, which retain sufficient performance for at least 30,000 h of load-cycling during normal operation.1 It is well known that the concomitant voltage cycling of the cathode leads to severe losses in electrochemically active surface area (ECSA) due to Pt-dissolution that results in both Pt particle growth by Ostwald ripening and Pt loss into the ionomer phase.2 In general, employing high surface area carbons (HSACs) with Pt-particles located mostly in the internal pores of the primary carbon particles, reduces the loss of ECSA due to Pt-dissolution.3 Furthermore, it has been shown that catalyst degradation due to particle coalescence is mitigated using HSACs due to the larger inter-particle distance compared to catalysts with solid carbon supports.3 However, it is still not clear how the loss and growth of internal Pt-nanoparticles is promoted throughout voltage cycling, i.e., if they grow inside the pores or if they move outward to the external C-surface and get subsequently lost in the ionomer or grow due to Ostwald ripening.4, 5 Therefore, voltage-cycling based accelerated stress tests (ASTs) were performed using 5 cm2 membrane electrode assemblies (MEAs) with cathode loadings of 0.1 mgPt cmMEA -2, made with Pt/C catalysts based on different supports: i) solid Vulcan carbon (Vu, TEC10V20E, TKK); ii) porous Ketjenblack (KB, TEC10E20E, TKK); and, iii) a partially oxided Pt/KB catalyst (Pt/KBmod) in order to study the effect of the accessibility of internal Pt particles on catalyst degradation. The latter was produced by heat-treating a commercial Pt/KB catalyst under an air/Ar mixture at 250 °C for 12 h to induce Pt-catalyzed local C-support oxidation of the bottleneck pore entrance, as described by Lazaridis et al.6 A full MEA characterization was included to measure the ECSA by CO-stripping, the H2/air and H2/O2 performance, and the O2 and H+ transport resistance in the cathode by limiting current measurements and electrochemical impedance spectroscopy (EIS), respectively. Finally, Pt-particle location prior and after voltage-cycling ASTs was investigated using scanning transmission electron microscopy (STEM).It was found that after the same number of voltage cycles, a smaller loss in H2/air performance can be observed using porous carbon supports (see. Fig. 1, Vu vs KB/KBmod), concomitant with a superior ECSA retention that we ascribe to a reduced dissolution/sintering of the Pt particles located with the pores of the carbon support. Furthermore, a performance increase for the unmodified Pt/KB cathode is shown during the first 300 cycles (see. Fig. 1), which can be attributed to facilitated transport of O2 and H+ to the Pt sites. Lastly, Pt-utilization measured by CO-stripping increased significantly after extended aging, indicating that the share in external particles has risen over voltage-cycling.