4120
P.I. Cowin et al. / Journal of Alloys and Compounds 509 (2011) 4117–4121
suggested in Eq. (3).
◦
600 C, 5%H /Ar
2
Co2V2O7−−−−−−−−−→Co2−xV1+xO4 + xCoO + (1 − x)VO + 2H2O (3)
Considering that vanadium monoxide was formed upon reduc-
tion, the decomposition route proposed in Eq. (2) is less probable
than that proposed in Eq. (3). Taking into account the impurity of
the sample, the magnitude of the conductivity for a pure, dense
sample of CoV2O4 should be significantly higher than the observed
conductivity. Despite the possible increased conductivity, research
into alternative transition metal spinels found significant redox
instability [27]. Due to the high probability of redox instability,
this compound is unlikely to be a suitable candidate for an anode
material for solid oxide fuel cells.
4. Conclusions
Cobalt pyrovanadate was successfully synthesised using a solid
state synthesis techniques. Impedance measurements between
300 ◦C and 700 ◦C in air determined that Co2V2O7 is an intrinsic
semiconductor with a band gap of 1.16(3) eV. The conductivity of
the sample in 5% H2/Ar also exhibits semiconductor behaviour,
albeit with a much smaller activation of 0.04(4) eV. The conduc-
tivity in air reached a maximum of 4 × 10−4 S cm−1 at 700 ◦C. In 5%
H2/Ar, albeit with a much smaller activation energy of 0.04(4) eV,
the conductivity ranged from 2.45 S cm−1 to 2.68 S cm−1, which is
approaching the magnitude required for SOFC anode materials.
Thermogravimetric analysis found a significant weight loss
upon reduction of the compound and X-ray diffraction analysis
indicated compound degradation into Co2−xV1+xO4, CoO and VO.
The redox instability and the low conductivity lead us to the con-
clusion that cobalt pyrovanadate is unsuitable for utilisation as an
anode material for SOFCs although the conductivity is reasonable in
a reducing atmosphere. Conductivity measurements on a pure sam-
ple of CoV2O4 should result in significant increases in the observed
magnitude of the conductivity.
Fig. 6. X-ray diffraction pattern of Co2V2O7 post sintering (PS) and after reducing
in 5% H2/Ar at 700 ◦C for 12 h.
ber 01-073-1633) and VO (JCPDS number 03-065-2896) through
the degradation pathway below:
◦
600 C, 5%H /Ar
2
(1)
Thermogravimetric analysis, Fig. 2, revealed a weight loss of
8.77% upon reduction of the compound. This correlates well with
the theoretical weight loss from the decomposition route given in
Eq. (1), 9.64%. The difference between the values is likely due to
incomplete reduction of the compound during thermogravimetric
analysis.
Acknowledgements
¯
Co2VO4 is cubic, space group Fd3m, with vanadium and cobalt
We would like to thank EPSRC and the ScotChem SPIRIT scheme
for funding. We would also like to thank Marian Millar for her aid
with X-ray diffraction data collection.
randomly distributed in 12-coordinate sites and edge-sharing MO6
octahedra [25]. From geometric considerations, this compound is
likely to have conduction through the mixed transition metal octa-
hedra. Cobalt in this compound is found as Co2+, which has the
electronic structure 3d7, t2g eg2, Due to electron-electron repul-
5
References
sion, conduction through cobalt octahedra is likely to have higher
activation energy for conduction that through vanadium, found as
[1] S.W. Tao, J.T.S. Irvine, Nat. Mater. 2 (2003) 320–323.
[2] J.C. Ruiz-Morales, J. Canales-Vásquez, C. Savaniu, D. Marrero-López, W. Zhou,
J.T.S. Irvine, Nature 439 (2006) 568–571.
V4+, with the electronic configuration 3d1, t2g eg
.
1
0
Previous research indicates that the activation energy observed
for this compound, 0.04 eV, correlates well to that found by
Rogers et al. in 1963 for CoV2O4, 0.07 eV, while a much higher
activation energy was observed for Co2VO4, 0.37 eV [26]. Both
Co2VO4 and CoV2O4 form with the spinel structure, although
with differing distributions of cobalt and vanadium cations and,
therefore, have similar X-ray diffraction patterns. No information
is available on the thermal properties of either of these com-
pounds, however as the DSC curve does not demonstrate any
significant peaks on cooling, no differentiation between these com-
pounds would have been available using this data. The activation
energy would suggest that the decomposition route in Eq. (2) is
more credible, despite the evidence to the contrary from XRD
analysis.
[3] Y.-H. Huang, R.I. Dass, Z.-L. Xing, J.B. Goodenough, Science 312 (2006) 254–257.
[4] S.W. Tao, J.T.S. Irvine, Solid State Ionics 179 (2008) 725–731.
[5] C.M. Grgicak, R.G. Green, J.B. Giorgi, J. Power Sources 179 (2008) 317–328.
[6] D. Simwonis, F. Tietz, D. Stöver, Solid State Ionics 132 (2000) 241–251.
[7] K.F. Chen, Z. Lu, X.J. Chen, N. Ai, X.Q. Huang, B. Wei, J.Y. Hu, W.H. Su, J. Alloys
Compd. 454 (2008) 447–453.
[8] T.S. Li, W.G. Wang, H. Miao, T. Chen, C. Xu, J. Alloys Compd. 495 (2010) 138–143.
[9] J.P.P. Huijsmans, F.P.F. van Berkel, G.M. Christie, J. Power Sources 71 (1998)
[10] F. Abraham, J.C. Boivin, G. Mairesse, G. Norogrocki, Solid State Ionics 40–41
(1990) 934–937.
[11] Z. Cheng, S. Zha, L. Aguilar, M. Liu, Solid State Ionics 176 (2005) 1921–1928.
[12] M.M. Cooper, K. Channa, R. De Silva, D.J. Bayless, J. Electrochem. Soc. 157 (2010)
B1713–B1718.
[13] Z. Xu, J. Luo, K.T. Chuang, ECS Trans. 11 (2008) 1–17.
[14] C.T.G. Petit, R. Lan, P.I. Cowin, J.T.S. Irvine, S.W. Tao, J. Mater. Chem. 21 (2010)
525–531.
[15] Y. Takita, S. Hikazudani, K. Soda, K. Nagaoka, J. Mol. Catal. A: Chem. 280 (2008)
164–172.
[16] M. Touaiher, K. Rissouli, K. Benkhouja, M. Taibi, J. Aride, A. Boukhari, B. Heulin,
Mater. Chem. Phys. 85 (2004) 41–46.
[17] G.M. Clark, A.N. Pick, J. Therm. Anal. 7 (1975) 289–300.
[18] M. Kurzawa, M. Bosacka, J. Therm. Anal. Calorim. 65 (2001) 451–455.
[19] Z. He, J.-I. Yamaura, Y. Ueda, W. Cheng, J. Solid State Chem. 182 (2009)
2526–2529.
◦
600 C, 5%H /Ar
2
Co2V2O7−−−−−−−−−→CoV2O4 + CoO + 2H2O
(2)
It is also possible that non-stoichiometric phases were formed
during decomposition, which would proceed through the route