Polymer electrolyte fuel cell (PEFC) is expected to one of the candidates for automobile and residential power sources. In PEFC, the carbon-supported platinum-based nanoparticle is generally used as the electrocatalyst. The catalyst support plays the two important roles of the electron conduction between Pt nanoparticles (PtNPs) and gas diffusion layer and the homogeneous dispersion of PtNPs by its high surface area. Activated carbon, which is cheap and has high surface area, employed as the catalyst support in general. However, the activated carbon is corroded at the startup/shutdown of PEFC which is reaching the high potential region. This carbon corrosion leads to detachment and aggregation of the PtNPs, as a result, produces to drastically degradation of the electrocatalyst. Another problem is that all of the PtNPs on the activated carbon cannot be using completely since the complex pores in the activated carbon accommodate of the PtNPs partially. In order to the realization of wide-spread commercialization of PEFC, the build-up of the low-cost system is necessary by improvement of the catalytic durability and the Pt utilization.

Recently, graphite oxide (GO) has been reported as a high corrosion resistant catalyst support to PtNPs. ^[1-2] PtNPs is highly dispersed on the GO in order to the electrostatic interaction between Pt complex and surface function group of GO. However, GO supported PtNPs catalyst has the high mass diffusion resistance and the low Pt utilization. This reason is thought that the majority of PtNPs would deposit between the GO layers which are poor the substance diffusion. We thought that the key point is a shape of GO nanosheets (GOns). To expose all of the PtNPs, the ordered structure by restacked GOns, such as sphere shape which has the higher substance diffusion, should be constructed. Therefore, we have been developing a new catalyst that PtNPs are supported on the carbon-sphere formed by GOns. Past carbon-sphere supported Pt (Pt/CS) catalyst had a problem with its electron conductivity. In the reduction step of Pt complex, PMMA sphere used as the core of CS was melting. And its melt was covering on the many PtNPs. This means that the number of active surface of PtNPs and the electron conductivity of catalyst were decreased because melt species were nonconductive. In this study, the silica beads were employed as the new core material for the carbon-sphere composed by GOns wall in order to prepare the Pt/CS. And the relationship between its ORR activity and electroconductivity was investigated.

GO was synthesized by typical Hummers method. GOns colloid solution was obtained by shaking GO with 0.01 wt% TBAOH aqueous solution. Silica beads were dispersed in 0.01 wt% polyvinyl alcohol copolymer aqueous solution and washed with ultrapure water. After that, their beads were mixed with GOns colloid solution and washed. Pt/CS was synthesized by the impregnation method. A pH of a Pt complex ethanol solution was adjusted to about 4 by NaOH aqueous solution. After that, the solution was added to a carbon sphere ethanol dispersion. And then, this mixed solution was dried at 60ºC. An obtained powder was thermally treated at 200 ºC under the hydrogen flow conditions. Its morphology, crystalline structure, and electronic state were observed SEM, TEM, XRD, and XPS, respectively. The electrochemical measurement was conducted by the rotating disk electrode (RDE) method at 60°C in 0.1 M HClO₄. The mass loading of Pt on test electrode was 17.3 µg cm^-2. The counter electrode was a carbon fiber and the reference electrode was Ag/AgCl.

The shape of CS has been kept sphere after the supporting treatments of PtNPs. However, the large size of silica beads was partially covered with GOns, not completely. In contrast, small particles were covered all over a surface by GOns. From XRD results, broad and weak peaks were observed at ~40° and ~45° which are attributed to Pt (111) and Pt (200) phases. The mass activity of Pt/CS was low value. However, this value was drastically improved by the addition of the ketjen black as the conductivity agent. Therefore, partially covered CS by GOns may inhibit the electron conductivity, and lead to low mass activity of ORR.

Now, full covering CS with GOns is preparing. Using this CS as the catalyst support is reformed the electron conductivity, and expected enhancement of ORR activity. Its physical property and ORR activity will be reported at the venue.

Reference

[1] L. Xin, et al., ACS Catal., 6, 2642-2653 (2016).

[2] J. Lu, X. Liu, Y. Zhang, Y. Liu, M. Li, X. Lu, Anal. Methods, 8, 816-823 (2016).