1272
Determination of Oxygen-Reduction-Reaction Active Sites of Niobium Oxide-Based Cathodes Used for Polymer Electrolyte Fuel Cells

Thursday, 9 October 2014: 09:20
Sunrise, 2nd Floor, Star Ballroom 8 (Moon Palace Resort)
M. Arao, T. Asada, K. Kubobuchi, M. Mogi, M. Matsumoto (NISSAN ARC Ltd.), A. Ishihara, K. I. Ota (Yokohama National University), and H. Imai (NISSAN ARC Ltd.)
Non-platinum oxygen-reducition reaction (ORR) electrocatalysts used for polymer electrolyte fuel cells (PEFCs) have been attracted much attension because of high-cost and relatively low-durability of platinum. Among non-platinum electrocatalysts, group 4 and 5 transition-metal oxides, such as tantalum, zirconium and niobium oxides, which are synthesized by partial oxidation of carbo-nitrides of transition metals, are known to show high ORR capability comparable to platinum and higher durability than platinum [1]. However, their ORR currents are insufficient for practical use and need to be improved. A primary reason for the low ORR currents is that the density of ORR active sites is too small. In order to increase density of active sites, we then need to identify the structural features and amount of active sites.

     According to our previous study, the active sites of oxide-based catalysts are assumed to be the oxygen-vacancies [2, 3]. In order to understand the structural feature of active sites and relationship between amount of oxygen vacancies and catalytic activities, then we performed neutron diffraction measurements, which are more sensitive to oxygen than x-ray, and transmission-electron microscopy (TEM) for niobium oxide catalysts synthesized from niobium carbo-nitride (Nb-CNOs).

     Nb-CNO catalysts were synthesized by partial oxidation of Nb2CN under 2%H2/N2+0.5%O2/N2 gases at 800¢ªC. Five catalysts were prepared by changing time of partial oxidation. We also prepared commercial Nb2O5powder sample as reference. The electrochemical ORR evaluation of the samples was carried out by using similar method as previously reported [4, 5]. Neutron diffraction measurements were performed by the Ibaraki Powder Neutron Diffractometer (BL20, iMATERIA) at J-PARC (Proposal No. 2012BM0009). Electron diffraction patterns and the high-resolution micrographs were taken by using FEI Tecnai G2 F20 (acceleration voltage of 200 kV) microscope. The samples for TEM observation was prepared by the crushing method.

     Fig. 1 shows the results of ORR electrochemical measurements. As partial oxydation time got longer, ORR current at 0.8 V gradually increased and reached a peak at 15 h, then decreased until 25 h.

     We show neutron diffraction patterns of Nb-CNO catalysts together with that of reference Nb2O5 sample in Fig. 2. From the measurements, Nb-CNOs are found to consist of two types of Nb2O5, the monoclinic one and the orthorhombic one. Notable feature is that the former is increased and the latter is decreased as oxydation time gets longer. Then we understood that the monoclinic phase contribute to catalytic activity. Moreover, the relative intensities of certain peaks, as indicated by A in Fig. 2, are increased as partial oxydation time gets longer. From the comparison between these results and the simulated diffraction patterns, we found that the change in the diffraction intensities could be reproduced by the decreasing oxygen occupancies at the apex of the NbO6 octahedra, which are shared in the edge-shared manner. We then conclude that the oxygen vacancies at such atomic sites could work as active sites for oxygen-reduction-reaction.

£Ð«x«·«Ã«&hibar;; font-size: medium;" £Ð«x«·«Ã«&hibar;" size="3">

Acknowledgement:

£Ð«x«·«Ã«&hibar;; font-size: medium;" £Ð«x«·«Ã«&hibar;" size="3">

A part of this work was performed under the “Non-precious metal oxide-based cathode for PEFC Project” supported by the New Energy and Industrial Technology Development Organization (NEDO).

£Ð«x«·«Ã«&hibar;; font-size: medium;" £Ð«x«·«Ã«&hibar;" size="3">

£Ð«x«·«Ã«&hibar;; font-size: medium;" £Ð«x«·«Ã«&hibar;" size="3">

References:

£Ð«x«·«Ã«&hibar;; font-size: medium;" £Ð«x«·«Ã«&hibar;" size="3">

[1]  K. Ota, Y. Ohgi,  K-D Nam, K. Matsuzawa, S. Mitsushima, A. Ishihara, J. Power Sources, 2011, 169, 5256.

£Ð«x«·«Ã«&hibar;; font-size: medium;" £Ð«x«·«Ã«&hibar;" size="3">

[2] H. Imai, M. Matsumoto, T. Miyazaki, S. Fujieda, A. Ishihara, M. Tamura, and K. Ota,  Appl. Phys. Lett., 2010, 96, 191905.

£Ð«x«·«Ã«&hibar;; font-size: medium;" £Ð«x«·«Ã«&hibar;" size="3">

[3] Y. Ohgi, A. Ishihara, K. Matsuzawa, S. Mitsushima, K. Ota, M. Matsumoto, H. Imai, J. Electrochem. Soc., 2013, 160, F162.

£Ð«x«·«Ã«&hibar;; font-size: medium;" £Ð«x«·«Ã«&hibar;" size="3">

[4] A. Ishihara, Y. Ohgi,  K. Matsuzawa, S. Mitsushima, K. Ota, Electrochim. Acta, 2010, 55, 7581.

£Ð«x«·«Ã«&hibar;; font-size: medium;" £Ð«x«·«Ã«&hibar;" size="3"> [5] A. Ishihara, M. Tamura, K. Matsuzawa, S. Mitsushima, K. Ota, Electrochim. Acta, 2010, 55, 8005.