Hierarchical Porous Structure and ORR Properties of Pt Supported on an Ordered Mesoporous Carbon with a Network Structure

INTRODUCTION A Pt/carbon-based catalyst using an ordered mesoporous carbon with a network structure (nw-OMC) as the support was recently developed by our group, demonstrating higher ORR activity than commercial Pt/CB catalysts. In this study, the effects of the hierarchical porous structure—comprising nanopores, primary pores, and secondary pores—on the I–V characteristics of the catalyst layer were investigated. Furthermore, the effect on catalytic performance of the Pt dispersion and deposition location within the nanopores was examined in detail. EXPERIMENTAL nw-OMC powder was synthesized using resol–nonionic surfactant micelles as both the carbon source and the structure-directing agent. Resol–F127 micelles were prepared by mixing phenol, formaldehyde, water, and F127 with sodium hydroxide. After hydrothermal treatment, the resulting solid was filtered, thoroughly washed, and dried under vacuum. The obtained powder was carbonized at 700 ℃ and subsequently annealed at 1000 ℃. Pt was deposited onto the nw-OMC powder via selective deposition techniques. Cyclic voltammetry (CV) and linear sweep voltammetry (LSV) were conducted in N₂- and O₂-saturated 0.1 M HClO₄ solution at 25 ℃ using a conventional rotating disk electrode (RDE) setup. Potential step cycling measurements simulating load cycling between 0.6 V and 1.0 V vs. RHE were performed following the FCCJ protocol. Catalyst ink was applied to a glassy carbon disk substrate using an electrospray (ES) coating technique developed by our group. The morphology of Pt/nw-OMC was characterized by scanning transmission electron microscopy (STEM). Catalyst-coated membranes (CCMs) were fabricated using a pulse-spray technique. Nafion NRE212 was employed as the membrane, and TEC10E50E was used as the anode catalyst. MEA measurements were conducted using a JARI single cell (1 cm²). Morphological analyses before and after electrochemical evaluation were performed by STEM, TEM, and cross-sectional SEM. RESULTS AND DISCUSSION Previous studies have shown that the synthesized nw-OMC consists of ~80 nm OMC nanoparticles interconnected to form a network with well-developed primary and secondary pores (100–several hundred nm). On the surface of the OMC particles, nanopores (~5 nm in diameter) were found to be regularly spaced (~9 nm), and selective deposition of Pt nanoparticles inside the nanopores was achieved at a ratio of 60–70%. RDE testing at room temperature indicated that the Pt/nw-OMC catalyst exhibits superior ORR mass activity and durability compared to a commercial Pt/CB (TEC10E50E). In the present work, further improvements were achieved by decreasing the primary particle size of the OMC and enhancing the Pt dispersion. STEM measurements revealed that the average OMC particle size was below 50 nm, forming a well-developed network structure (Fig. 1). The average Pt particle size was decreased to 2.5 nm, in contrast to 4 nm in previous samples. RDE measurements showed a mass activity of 475 A g⁻¹ at 0.9 V, more than double that of the commercial Pt/CB. However, although the catalyst exhibited excellent mass activity in RDE testing, the performance in the MEA evaluation was hindered by excessive void formation within the catalyst layer. This issue was attributed to the tendency of the network-structured OMC to form secondary aggregates during ink preparation, especially under non-optimized mixing conditions. These voids negatively affected the uniformity of the pore structure and the interfacial contact between the catalyst layer and the membrane. In addition to this issue, insufficient mixing between the ionomer and the Pt/OMC catalyst resulted in non-uniform ionomer coverage on the catalyst surface during CL formation, further limiting MEA performance. To overcome this issue, the ink formulation process was carefully optimized by tuning the dispersion and mixing conditions to prevent secondary aggregation of the nw-OMC particles. This optimization effectively suppressed the formation of coarse voids, resulting in a more uniform and well-connected pore structure and enhanced interfacial adhesion between the catalyst layer and the membrane. As a result of the improved pore structure and interfacial contact, the I–V performance of the MEA surpassed that of commercial Pt/CB across the entire current density range (Fig. 2), with a particularly notable lowering of the concentration overpotential during high current density operation. Acknowledgement: This work was partially supported by the ECCEED’30-FC project from NEDO. Figure 1