The Impact of Mesopores of SnO2 Catalyst Support on the Performance of Polymer Electrolyte Fuel Cells

Abstract

Recently, the focus of future polymer electrolyte fuel cell (PEFC) application for automobiles has shifted from the class of passenger light-duty vehicles (LDVs) to that of heavy-duty vehicles (HDVs). The total operating time in the lifetime of HDVs is much longer than that of LDVs. Thus, durability is becoming more and more important for the cell components including the catalyst materials, and the compatibility between initial performance and durability at a high level is desired. Employing mesoporous carbons as catalyst supports is a promising option to enhance the initial performance [1]. In the catalyst layer of a PEFC with mesoporous carbon support, Pt-based catalyst nanoparticles deposited inside the mesopores are not directly covered by ionomer. As a result, a catalyst poisoning by sulfonic acid groups of the ionomer is hindered and a high catalytic activity is obtained. In addition, a high oxygen transport resistance thorough Pt/ionomer interfaces [2] can be mitigated by employing mesoporous carbon support. Nonetheless, from the viewpoint of thermodynamics, carbon support can oxidatively degrade during the long-term operation. Tin oxide (SnO2) is one of the promising candidates for substituting the carbon support because of its high electrochemical stability at high potentials [3]. However, to our best knowledge, the mitigation of the ionomer adsorption on the Pt surface by using mesoporous SnO2 support has not been studied in previous works. In this study, we synthesized mesoporous SnO2 supports and investigated the impact of the mesopore on the performance of the PEFC. Connected mesoporous Sb-doped SnO2 particles (CMSbTOs) were synthesized by using a mesoporous carbon as a template (Figure (a)). The size of the mesopore was successfully controlled in the range of 4 – 10 nm by adjusting the carbon template structure and calcination temperature (Figure (b)). The BET surface area and conductivity of the CMSbTOs were in the ranges of 90 – 210 m2 g-1 and ca. 10-2 S cm-1, respectively. Pt nanoparticles were deposited on the CMSbTOs via a colloidal method to obtain Pt/CMSbTO catalyst (Pt loading of 20 wt.%). It was confirmed that the Pt nanoparticles were located both on the outer surface and inside the mesopores of the CMSbTOs by scanning transmission electron microscopy-energy dispersive spectroscopy (TEM-EDS) analysis. A membrane electrode assembly (MEA) was fabricated by using a Pt/CMSbTO catalyst for the cathode catalyst layer and single cell tests were performed under dried (30%RH, 82 °C) and highly humidified (80%RH, 60 °C) conditions. In addition, a homemade Pt/Sb-SnO2 catalyst with solid-core SnO2 support and a commercial Pt/Vulcan catalyst (TEC10V30E, TKK) were also tested for comparison. Under the dried conditions, the Pt/CMSbTO catalyst showed a better I-V performance than the Pt/Sb-SnO2 (solid) catalyst and Pt/Vulcan catalyst (Figure (c)). The oxygen reduction reaction (ORR) activity of the Pt/CMSbTO catalyst at 0.84 V was higher than that of the Pt/Sb-SnO2 (solid) and Pt/Vulcan catalysts. This result suggests that catalyst poisoning by the ionomer is mitigated when the mesoporous SnO2 support is employed. Moreover, the local oxygen transport resistance (oxygen transport resistance that is independent of the total gas pressure) of the Pt/CMSbTO catalyst was lower than that of the Pt/Sb-SnO2 (solid) and Pt/Vulcan catalysts, probably because of a decrease in the oxygen transport resistance due to Pt/ionomer interfaces. In contrast, under the highly humidified conditions, the I-V performance of the Pt/CMSbTO catalyst was comparable to that of the Pt/Sb-SnO2 (solid) catalyst and slightly lower than that of the Pt/Vulcan catalyst (Figure (d)). This result can be ascribed to floodings inside the mesopores of the CMSbTO support. Accelerated stress tests (repeated potential steps between 1.0 – 1.5 V under 100%RH, 80 °C) were also performed to confirm the electrochemical stability of the Pt/CMSbTO catalyst at high potentials. The Pt/CMSbTO catalyst showed a much higher retention rate of the electrochemical Pt surface area (ECSA) than the Pt/Vulcan catalyst as expected. More details, including the effect of the pore size of the CMSbTO support on the initial performance, will be discussed in the presentation.