(PEFCE&E 21 Best Student Poster - First Place) Oxygen Reduction Reaction Properties and Microstructures of Pt/Nb:SnO2 Model Catalysts Fabricated on Pt Single Crystal Surfaces

Abstract

Introduction Considering strong acidic and severe potential fluctuation environments for operating conditions of Polymer Electrolyte Fuel Cells (PEFC), electrochemical degradation of electrode materials, such as elution and/or oxidation, is crucial for development of high-performance electrode materials, e.g., Pt-based nanoparticles (NPs) and C-based supports. Recently, replacing the C-based supports with metal-oxides aimed for improving electrochemical stabilization has attracted much attention. Among the candidates of metal-oxides used as support materials, X (=Nb, Sb, Ta etc.) doped tin dioxide (X:SnO2) shows relatively high electrical conductivity. Actually, Kakinuma et al. [1,2] demonstrated that superior oxygen reduction reaction (ORR) activity and durability of the Pt NPs supported on Nb-doped SnO2 (Pt NPs/Nb:SnO2). The results clearly show that such the metal-oxide supports are one of the powerful candidates for the next-generation cathode materials. Also, it is considered possible to use Nb:SnO2 as a core material (Pt@Nb:SnO2) for highly durable Pt-shell / metal-oxide-core type catalyst, although influences of microstructures of the Pt-based materials and metal-oxide interface on the ORR properties remains unclear. Therefore, in this study, we prepared Pt NPs/Nb:SnO2 and Pt@Nb:SnO2 model catalysts with atomically-controlled Pt/SnO2 interface under ultra-high vacuum (UHV; ~10-8 Pa) and investigated relationship between the surface and interface microstructures of the model catalysts and their ORR properties. Experimental Nb-doped Sn layer with 5 nm thickness was deposited on the surface-cleaned Pt(111) substrate (Ar+-sputtering and annealing) by arc-plasma deposition (APD) of Sn-Nb alloy target (composition ratio 96:4 (at.%)) at the substrate temperature of 300 K. Subsequently, the prepared sample was transferred to a furnace and thermally-oxidized in dry air (Nb:SnO2 / Pt(111)). Then, the Nb:SnO2 / Pt(111) sample was re-introduced into the UHV-chamber, Pt was vacuum-deposited by the APD for Pt NPs / Nb:SnO2 / Pt(111) and by e-beam evaporation for Pt-shell on the Nb:SnO2/Pt(111) (Pt shell / Nb:SnO2 / Pt(111)) (Fig1-(a) and Fig1-(b), respectively). The microstructures of the catalyst model samples were analyzed by STM, cross sectional STEM-EDS, XPS etc. The ORR properties were evaluated by CV and LSV with the RDE method conducted in N2-purged and O2-saturated 0.1 M HClO4, respectively. The ORR activity was evaluated from j k values at 0.9 V vs. RHE by using Koutecky-Levich equation. The electrochemical stability was discussed by the ORR activity changes during applying the potential cycles (PCs) of 0.6-1.0 V vs. RHE (each 3 s) in O2 saturated 0.1 M HClO4 at room temperature. Results and Discussion The STM image (Fig. 2-(a)) of the Pt NPs / Nb:SnO2 / Pt(111) for the Pt/SnO2 model catalyst showed Pt NPs with an average size of 3.6 nm were distributed over the Nb:SnO2 substrate surface. The ORR activity and durability for the Pt NPs / Nb:SnO2 / Pt(111) clearly outperformed relative to the Pt NPs formed on a highly oriented pyrolytic graphite (HOPG)[3] formed under the same Pt deposition condition: increase in Pt particle sizes and size distribution of Pt NPs by the 10000 PCs loading was suppressed compared to that of the Pt NPs / HOPG.A cross-sectional HAADF-STEM image of the Pt-shell / Nb:SnO2 / Pt(111) model core-shell sample (Fig. 2-(b)) reveals that the surface Pt layers with (111) orientation (ca. 2.5 nm in thickness) and SnO2(101) lattice of ca. 4 nm-thick Nb:SnO2 layer are grown on the Pt(111) substrate surface. This is probably due to lattice matching between the Pt(111) (0.48 × 0.56 nm) and SnO2(101) (0.48 × 0.58 nm)(Fig. 2-(c)). Such the lattice matching might correlate with the forementioned suppression of nanoparticle migration [4] in the Pt NPs / Nb:SnO2 / Pt(111) model catalyst. Furthermore, as shown in Fig. 2-(d), the Pt shell / Nb:SnO2 / Pt(111) shows that the ORR activity enhancement factor is at most × 4.1 vs. clean Pt(111) at 500 PCs and × 3.2 even after 5000 PCs loading. The ORR activity enhancement might stem from so-called the ligand effect of Sn and/or Nb on the surface Pt(111) layer through thermal diffusion such the elements into the surface Pt(111) layer during the model catalyst UHV preparation. At the meeting, I will show ORR properties of the Pt / Nb:SnO2 / Pt(110) and discuss influences of the lattice orientations of the surface Pt and underlaid SnO2 layer. Acknowledgement This study was supported by new energy and industrial technology development organization (NEDO) of Japan.Reference[1] K. Kakinuma et al., Electrochim. Acta., 110, 316-324 (2013).[2] K. Kakinuma et al., J. Electrochem. Soc., 165, 15, J3083-J3089 (2018).[3] S. Takahashi et al., Phys. Chem. Chem. Phys., 17, 18638-18644 (2015).[4] M. Wakisaka, et al., Phys. Chem. Chem. Phys., 12, 4184-4190 (2010). Figure 1