Carbon supported platinum nanoparticle is utilized as the cathode catalyst for polymer electrolyte fuel cells. However, the efficiency is about 40% (60% of platinum exist inside nanoparticle) for commercial 2-3 nm diameter platinum and thus there is still room for improvement. One of the approaches to enhance the platinum efficiency is the use of smaller sized nanoparticles. However, smaller nanoparticles generally have lower specific activity and durability. Therefore, one must overcome the trade-off between high surface area and high stability.We have succeeded in the exfoliation of layered platinic acid to platinum oxide nanosheets and subsequent reduction to platinum nanosheets with an average thickness of ~0.5 nm (2 atomic layers). The carbon supported platinum nanosheets showed higher electrochemically active surface area (ECSA) and ORR mass activity than conventional carbon supported 3 nm platinum nanoparticles. Moreover, the carbon supported platinum nanosheets exhibited higher load-cycle durability owing to the high coordination number of surface atoms of platinum nanosheets. However, the nanosheets were partially aggregated and transformed to nanoparticles, resulting in platinum efficiency of about 60% of theoretical surface area. In this study, we discuss the topotactic reduction of carbon supported platinum oxide nanosheets via mild treatment such as vapor phase reduction with hydrogen gas and/or carbon monoxide for further improvement of ECSA and ORR activity.The ECSA of the carbon supported platinum nanosheets reduced by the thermal treatment under hydrogen gas at 100 ℃ (Pt(ns)/C-H2) was 100 to 120 m2 (g-Pt)-1, which is higher than 3 nm platinum nanoparticles (80 m2 (g-Pt)-1). The ECSA of the carbon supported platinum nanosheets reduced by carbon monoxide gas at 25 ℃ (Pt(ns)/C-CO) was 133 m2 (g-Pt)-1. It is considered that carbon monoxide adsorbed on reduced platinum atoms, leading to suppression of aggregation and growth of platinum.The ORR specific activity of Pt(ns)/C-H2 was 300-350 μA cm-2, while that of Pt(ns)/C-CO was 150 μA cm-2 which is half of Pt(ns)/C-H2. From cyclic voltammograms (Figure 1), the peak potential for the reduction of platinum oxide of Pt(ns)/C-CO showed lower potential than that of Pt(ns)/C-H2. This suggests that Pt(ns)/C-CO has a more unstable surface than Pt(ns)/C-H2, and Pt−OHad and/or Pt−Oad formed more on platinum surface, lowering the ORR. As a result of the high ECSA, the ORR mass activity of Pt(ns)/C-H2 was 1.7 times higher compared to that of 3 nm platinum nanoparticles, while that of Pt(ns)/C-CO was almost same as 3 nm platinum nanoparticles despite the higher ECSA.This work was supported in part by the “Polymer Electrolyte Fuel Cell Program” and the FC-Platform projects from the New Energy and Industrial Technology Development Organization (NEDO) of Japan (20001201-0, 22101136-0).[1] D. Takimoto, S. Toma, Y. Suda, T. Shirokura, Y. Tokura, K. Fukuda, M. Matsumoto, H. Imai and W. Sugimoto, Nat. Commun. 14, 19 (2023). Figure 1
Read full abstract