The group 4 and 5 transition-metal oxides have been investigated as the non-Pt alternative catalysts. These oxide catalysts, which was made by the partial-oxidation of carbonitride, have high ORR activity [1]. Now, it is believed that the ORR arises at the oxygen vacancy site of oxide surface which acts as the active site. Moreover, the deposited carbon, which was residual carbon species of the organic metal precursor, would become the role of the micro electron-conduction path. However, the deposited carbon and the oxygen vacancies are not well understood whether how to be actually working on the active factors.
In order to understand for an ORR-activity fundamental-factor, this study investigated that the group 4 and 5 transition-metal oxides (TiO2, ZrO2, Nb2O5) are how the relationship between ORR activity and the crystalline structure and size.
Those samples were prepared by sol-gel methods. An organometallic compound (Ti(OC4H9)4, Zr(OC4H9)4, Nb(OC2H5)5), diethylene glycol and 1-penten-3-ol solution were mixed in 1-butanol. Gelatinization of the sol solution was conducted by addition of ultra-pure water. The obtained gel was calcination at 400 - 700ºC for 12 hours.
The crystal structure was characterized by X-ray diffraction (XRD), and the chemical surface condition was analyzed by X-ray photoelectron spectroscopy (XPS). The cyclic voltammetry measurement was conducted by the static three-electrode cell in 0.1 M HClO4. The ORR activity was evaluated by calculated the ORR currents (iORR). The iORR was calculated by subtracting the Ar saturated currents (iAr) from the O2 saturated currents (iO2). The onset potential for the ORR (EORR) was defined as the electrode potential at the starting of the iORR.
The diffraction peaks of the obtained samples at 700ºC were attributed to that of TiO2, ZrO2 and Nb2O5, respectively. The crystal structure of ZrO2 depended on the calcination temperature. In case of 700ºC, the crystal structure was only the orthorhombic. In case of calcination at 400 and 600ºC, not only the orthorhombic but also the cubic was detected. The peak intensity of the cubic structure became stronger by decreasing the calcination temperature. EORR of ZrO2 at all calcination temperature was almost same and low. Thus, the crystal structure would not affect ORR activity.
On the other hand, the crystallite size of all samples also depended on the calcination temperature. In the samples of Nb2O5, the crystallite size of 700ºC was estimated ~ 63 nm at 22.5º, whereas that of 400 and 300ºC were about several nanometers and no crystallinity, respectively. EORR of Nb2O5 was the same potential in spite of different crystallite sizes. Therefore, the crystallite size also would not be relevant to ORR activity.
The crystal structure and the crystallite size may be no relation to the ORR activity in group 4 and 5 oxides. Because EORR of ZrO2 and Nb2O5 at different calcination temperature were almost the same potential and low. Therefore, it is thought that some active sites and electronic conduction paths like such the oxygen vacancies and the residual carbons are the most impacts for ORR. However, in the XPS measurement, the residual carbon has detected all samples. The residual carbon would be also no effect ORR activity since all samples showed the low EORR. Thus, it is not as simple as that only the presence of oxygen vacancy and residual carbon lead to high ORR activity.
In rare cases, some prepared oxide indicated the high ORR activity although its reproducibility was low. There is a possibility that other factors, which have not known yet, promotes ORR. For the future work, the ORR activity will be comparing with/without oxygen vacancies.
reference
[1] A. Ishihara, M. Tamura, Y. Ohgi, M. Matsumoto, K. Matsuzawa, S. Mitsushima, H, Imai, K. Ota, J. Phys. Chem. C, 117, 18837 (2013).