- Email: [email protected]
Perovskite related cuprate phases as potential cathode materials for solid oxide fuel cells
Solid State Ionics 175 (2004) 99 – 102 www.elsevier.com/locate/ssi
Perovskite related cuprate phases as potential cathode materials for solid oxide fuel cells J.E.H. Sansom, H.A. Rudge-Pickard, G. Smith, P.R. Slater*, M.S. Islam Department of Chemistry, University of Surrey, Guildford, Surrey, GU2 7XH, UK
Abstract The electrical properties of the perovskite-related cuprate materials YSr2Cu2FeO7+y and YSr2Cu2CoO7+y have been investigated for possible use as cathodes in solid oxide fuel cells. Both phases were shown to be stable at high temperatures (1173 K) under N2, and their conductivities were determined with varying temperature and oxygen partial pressure. Interesting variations of the conductivity with temperature were seen for both samples. In the case of YSr2Cu2FeO7+y, the conductivity showed semiconductor behaviour up to c823 K, with a decrease in conductivity then being observed for higher temperatures. This decrease is believed to be due to a reduction in the number of charge carriers, which is supported by TGA studies, which show oxygen loss above this temperature. For YSr2Cu2CoO7+y, the sample showed semiconducting behaviour over the entire range of temperatures studied, although a steeper increase in conductivity was seen above 1073 K. High temperature XRD studies show that a phase change (orthorhombic-tetragonal) occurs in this temperature region. Chemical compatibility studies (at temperatures between 1173 and 1273 K) of these phases with SOFC electrolytes are also reported. D 2004 Elsevier B.V. All rights reserved. Keywords: Solid oxide fuel cells; Superconductor; Cuprate; Cathode
1. Introduction Solid oxide fuel cells (SOFCs) are currently attracting considerable interest as stationary power generators due to the high efficiency and low levels of pollution that they promise. There is a large amount of research being performed targeting new materials for use in such fuel cells, and in this paper we present initial results on the investigation of two cuprate phases related to the high temperature superconducting systems as possible SOFC cathode materials. YBa2Cu3O7 y is a well known material, widely studied due to its superconducting properties. It has also been investigated as a possible cathode material for SOFCs, but experiences a number of problems. At raised temperatures (N1173 K) there is significant degradation of the structure including decomposition and a reaction with YSZ. There is * Corresponding author. Tel.: +44 1483 686847; fax: +44 1483 686851. E-mail address: [email protected] (P.R. Slater). 0167-2738/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.ssi.2004.09.035
also electrochemical decomposition at fairly low current densities (20 mA cm 2) [1]. Nonetheless it is an interesting material, which has quite good electronic and oxide ion conductivity [2]. It has two copper sites, one site (Cu2) associated with the superconducting CuO2 layers, and a second site (Cu1) which link these layers along the c direction. The main stability problem is located at the Cu1 site. In the fully oxidised material, yc0, this Cu atom is in square planar coordination. If the material is reduced at high temperatures in an inert atmosphere, such as Ar, N2, oxygen is lost reducing this Cu to Cu1+ with linear coordination. Further reduction at higher temperatures leads then to decomposition of the sample. In order to reduce or remove this instability we can partially or completely replace the Cu at the Cu1 site with other cations, ranging from Al, Ga, to transition metals, and even oxyanions (SO42 , BO33 ). Co-substitution of Sr for Ba also has the effect of improving the thermal stability. We can thus prepare phases of the form YSr2Cu3 x Mx O7+y (M=Ga, Al, Co, Fe, B) [3–9]. Although these phases have
100
J.E.H. Sansom et al. / Solid State Ionics 175 (2004) 99–102
been known for several years, there have been no studies of their high temperature conductivities. In this paper we rectify this situation with an examination of the conductivities of YSr2Cu2FeO7+y and YSr2Cu2CoO7+y with temperature and p(O2) (0.2–10 5 atm) in order to evaluate their potential as SOFC cathode materials. In the case of YSr2Cu2FeO7+y, previous structural studies have shown that, although Fe predominantly substitutes for the Cu1 sites, some Fe does enter the Cu2 site. The value of y is typically c0.4–0.5 as prepared, with this bexcessQ oxygen going into vacant sites around the Cu1 position increasing the coordination for some of the Fe from 4 to 5/6 coordination. This phase, as prepared, is not a superconductor, although recent reports have shown that superconductivity can be induced by annealing at high temperatures under N2, followed by subsequent low temperature oxidation. The change is attributed to a redistribution of the Fe under the moderately reducing conditions, with all the Fe now favouring location on the Cu1 site [10]. In the case of YSr2Cu2CoO7+y, structural studies have shown that all the Co is located in the Cu1 site [4,5], leading to a complete Co layer. The preference of Co for tetrahedral coordination, rather than the square planar coordination favoured by Cu in undoped YBa2Cu3O7 y, leads to cooperative oxygen ordering and consequently an expanded unit cell. Unlike for YSr2Cu2FeO7+y, YSr2Cu2CoO7+y shows negligible oxygen excess ( yc0).
2. Experimental High purity Y2O3, SrCO3, CuO, Fe2O3, Co3O4 were used to prepare the samples YSr2Cu2FeO7+y, YSr2Cu2 CoO7+y. The starting materials were ground in the stoichiometric ratios and heated to 1273 K for 16 h. The powders were then reground and reheated under the same conditions. Conductivity measurements were made using the standard four probe d.c. technique. Sintered pieces were prepared
by pressing at 6000 kg cm 2 and sintering at 1273 K for 16 h. Conductivity measurements were then taken in air as a function of temperature (298–1173 K). Measurements of the conductivity versus oxygen partial pressure (measured via a YSZ sensor) were made at a temperature of 1173 K. For these measurements, the conductivities were initially taken in moving from oxidising to reducing conditions, where nitrogen was allowed to pass into the system. In this case the change in p(O2) proceeds quite rapidly (0.2–10 5 atm in about 1 h). The sample was then held under a N2 atmosphere for at least 15 h, in order for the system under test to reach equilibration, prior to any measurements being taken in the reverse direction (from reducing to oxidising). The N2 flow was then turned off, and oxygen was allowed to leak back into the system naturally. This typically took a period of approximately 24 h, with measurements made over this time. X-ray diffraction confirmed sample purity before and after measurement. Oxygen contents were determined by thermogravimetric analysis (reduction (5%/H2/95%N2) at 1223 K) using a Stanton Redcroft STA 780 thermal analyser. Thermogravimetric analysis was also used to investigate oxygen loss within the samples on heating between 298 and 1173 K (heating rate 10 8C/min, held at 1173 K for 20 min) in both air and N2. The chemical compatability of the phases YSr2Cu2FeO7+y and YSr2Cu2CoO7+y with potential electrolytes was examined by grinding together the sample and electrolyte in a 1:1 by mass ratio and heating at temperatures between 1173 and 1273 K for 1–3 weeks. The electrolytes examined included YSZ, CGO, La0.9Sr0.1Ga0.8Mg0.2O2.85, and Apatite-type (La9.33(Si/Ge)6O26).
3. Results The phases YSr2Cu2FeO7+y and YSr2Cu2CoO7+y were successfully prepared, with cell parameters of a=b= 3.830(3), c=11.465(9) 2 (YSr 2 Cu 2 FeO 7+y ) and a= 22.820(3), b=5.454(1), c=5.415(1) 2 (YSr2Cu2CoO7+y ).
Fig. 1. Plot of log r vs. log p(O2) at 1173 K for (a) YSr2Cu2CoO7+y, (b) YSr2Cu2FeO7+y.
J.E.H. Sansom et al. / Solid State Ionics 175 (2004) 99–102
Fig. 2. Plot of log r vs. 1/T for (a) YSr 2 Cu 2 FeO 7+y and (b) YSr2Cu2CoO7+y.
TGA studies indicated an oxygen content of 7.42(4) for YSr2Cu2FeO7+y and 7.02(4) for YSr2Cu2CoO7+y. Examination of the TGA traces for both samples in air and N2 between 298 and 1173 K showed no measureable oxygen loss for YSr2Cu2CoO7+y, while for YSr2Cu2FeO7+y significant oxygen loss was observed. In the case of air, oxygen loss began at a temperature c823 K, with the oxygen content decreasing to 7.19(4) at 1173 K. Heating in N2 resulted in a greater mass loss to 7.03(4) at 1173 K. For both samples the conductivity versus oxygen partial pressure measurements showed a decrease in conductivity with decreasing p(O2), which indicates that these materials are p-type conductors. The data showed significant hysteresis between oxidation/reduction curves (see Fig. 1). The pellet densities of these samples were 84% and 80% of the theoretical for YSr2Cu2CoO7+y and YSr2Cu2FeO7+y, respectively. Attempts were made to prepare more porous samples by sintering at lower temperature (1173 K), to examine whether the hysteresis was related to pellet density. However, severe problems were encountered in measuring pellets sintered at such lower temperatures due to their very brittle nature. It should be noted that the measured data do not represent the equilibrium state, since the oxygen partial pressure changes relatively quickly. This means that it is not possible to calculate accurate gradients for the plots of log r vs. log p(O2) to relate to the defect equations. However, at the two end points (0.2, c10 5 atm) the sample is well equilibrated, and so an approximate average gradient can be determined from the conductivity values at these two end points. Using these points, approximate gradients of 0.08 and 0.37 were determined for YSr 2Cu2CoO 7+y and YSr 2Cu2FeO7+y, respectively. The key observation from this is that the p(O2) dependence for YSr2Cu2FeO7+y is much greater than that observed for YSr2Cu2CoO7+y. This could be related to two factors. Firstly the observed large change in oxygen stoichiometry observed for YSr2Cu2FeO7+y. A consideration of the TGA data for YSr2Cu2FeO7+y shows an oxygen content change from 7.19(4) to 7.03(4) at 1173 K in going from air to N2, leading to a large change in the number of charge carriers. In contrast, TGA studies showed negligible
101
oxygen loss for YSr2Cu2CoO7+y. A second factor is the possible redistribution of the Fe over the two Cu sites under N2, with an increase in the Fe occupancy of the Cu1 site and consequent decrease in the occupancy at the Cu2 site, as suggested previously in the literature [10]. In addition to data on the p(O2) dependence of the conductivity at 1173 K, data on the variation of the conductivity with temperature in air has been collected. This data showed some interesting variations (Fig. 2). For YSr2Cu2FeO7+y, the conductivity shows semiconducting behaviour up to c823 K, above which a gradual decrease in conductivity is observed (r 823K=35 S cm 1, r 1173K=6 S cm 1). The decrease in conductivity at higher temperatures can be related to oxygen loss (as shown by TGA studies) reducing the number of charge carriers. The variation in conductivity with temperature observed for YSr2Cu2CoO7+y was particularly interesting. For this sample semiconducting behaviour was observed for the whole temperature range examined, but there was a steeper increase in the conductivity around 800 8C. We have recently collected high temperature X-ray diffraction data for this sample, which has shown that there is an orthorhombic-tetragonal phase transition at this temperature, and this phase transition is now being investigated in more detail. The chemical compatibility studies of these two samples and SOFC electrolytes showed significant reaction in all cases. For YSZ, insulating SrZrO3 was observed, while for the La9.33(Si/Ge)6O26 electrolytes, a complex mixture of phases was seen, which included (La/Sr)2CuO4 y and unidentified phases. A complex pattern was also observed for reaction with La0.9Sr0.1Ga0.8Mg0.2O2.85, while for CGO, the phases (Y/Ce) 2 Sr 2 Cu 2 MO 9+x (M=Fe, Co) were observed. The latter are related to the YSr2Cu2MO7+x (M=Fe,Co) phases by the replacement of the Y3+ by a (Y/ Ce)2O2x+ bfluorite blockQ, and we are currently investigating the high temperature conductivities of these materials.
Acknowledgements We would like to thank EPSRC and Merck Ltd for funding, and A.J. Wright (University of Birmingham) for the provision of high temperature XRD facilities.
References [1] J.G. Fletcher, J.T.S. Irvine, A.R. West, J.A. Labrinca, J.R. Frade, F.M.B. Marques, Mater. Res. Bull. 29 (1994) 1175. [2] S. Tsukui, M. Adachi, R. Oshima, H. Nakjima, F. Toujou, K. Tsukamoto, T. Tabata, Physica, C 351 (2001) 357. [3] P.R. Slater, C. Greaves, Physica, C 180 (1991) 299. [4] T. Den, T. Kobayashi, Physica, C 196 (1992) 141. [5] T.G.N. Babu, J.D. Kilgour, P.R. Slater, C. Greaves, J. Solid State Chem. 103 (1992) 472. [6] G. Roth, P. Adelmann, G. Heger, R. Knitter, T. Wolf, J. Phys. 1 (1991) 721.
102
J.E.H. Sansom et al. / Solid State Ionics 175 (2004) 99–102
[7] S.A. Sunshine, L.F. Schneemeyer, T. Siegrist, D.C. Douglas, J.V. Waszczak, R.J. Cava, E.M. Gyorgy, D.W. Murphy, Chem. Mater. 1 (1989) 331. [8] L.E. Moran, R. Rojas, M.A. Alario-Franco, J. Mater. Chem. 6 (1996) 1517.
[9] J.Q. Li, W.J. Zhu, Z.X. Zhao, D.L. Yin, Solid State Commun. 85 (1993) 739. [10] J. Shimoyama, K. Otzschi, T. Hinouchi, K. Kishio, Physica, C 341 (2000) 563.
Perovskite related cuprate phases as potential cathode materials for solid oxide fuel cells J.E.H. Sansom, H.A. Rudge-Pickard, G. Smith, P.R. Slater*, M.S. Islam Department of Chemistry, University of Surrey, Guildford, Surrey, GU2 7XH, UK
Abstract The electrical properties of the perovskite-related cuprate materials YSr2Cu2FeO7+y and YSr2Cu2CoO7+y have been investigated for possible use as cathodes in solid oxide fuel cells. Both phases were shown to be stable at high temperatures (1173 K) under N2, and their conductivities were determined with varying temperature and oxygen partial pressure. Interesting variations of the conductivity with temperature were seen for both samples. In the case of YSr2Cu2FeO7+y, the conductivity showed semiconductor behaviour up to c823 K, with a decrease in conductivity then being observed for higher temperatures. This decrease is believed to be due to a reduction in the number of charge carriers, which is supported by TGA studies, which show oxygen loss above this temperature. For YSr2Cu2CoO7+y, the sample showed semiconducting behaviour over the entire range of temperatures studied, although a steeper increase in conductivity was seen above 1073 K. High temperature XRD studies show that a phase change (orthorhombic-tetragonal) occurs in this temperature region. Chemical compatibility studies (at temperatures between 1173 and 1273 K) of these phases with SOFC electrolytes are also reported. D 2004 Elsevier B.V. All rights reserved. Keywords: Solid oxide fuel cells; Superconductor; Cuprate; Cathode
1. Introduction Solid oxide fuel cells (SOFCs) are currently attracting considerable interest as stationary power generators due to the high efficiency and low levels of pollution that they promise. There is a large amount of research being performed targeting new materials for use in such fuel cells, and in this paper we present initial results on the investigation of two cuprate phases related to the high temperature superconducting systems as possible SOFC cathode materials. YBa2Cu3O7 y is a well known material, widely studied due to its superconducting properties. It has also been investigated as a possible cathode material for SOFCs, but experiences a number of problems. At raised temperatures (N1173 K) there is significant degradation of the structure including decomposition and a reaction with YSZ. There is * Corresponding author. Tel.: +44 1483 686847; fax: +44 1483 686851. E-mail address: [email protected] (P.R. Slater). 0167-2738/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.ssi.2004.09.035
also electrochemical decomposition at fairly low current densities (20 mA cm 2) [1]. Nonetheless it is an interesting material, which has quite good electronic and oxide ion conductivity [2]. It has two copper sites, one site (Cu2) associated with the superconducting CuO2 layers, and a second site (Cu1) which link these layers along the c direction. The main stability problem is located at the Cu1 site. In the fully oxidised material, yc0, this Cu atom is in square planar coordination. If the material is reduced at high temperatures in an inert atmosphere, such as Ar, N2, oxygen is lost reducing this Cu to Cu1+ with linear coordination. Further reduction at higher temperatures leads then to decomposition of the sample. In order to reduce or remove this instability we can partially or completely replace the Cu at the Cu1 site with other cations, ranging from Al, Ga, to transition metals, and even oxyanions (SO42 , BO33 ). Co-substitution of Sr for Ba also has the effect of improving the thermal stability. We can thus prepare phases of the form YSr2Cu3 x Mx O7+y (M=Ga, Al, Co, Fe, B) [3–9]. Although these phases have
100
J.E.H. Sansom et al. / Solid State Ionics 175 (2004) 99–102
been known for several years, there have been no studies of their high temperature conductivities. In this paper we rectify this situation with an examination of the conductivities of YSr2Cu2FeO7+y and YSr2Cu2CoO7+y with temperature and p(O2) (0.2–10 5 atm) in order to evaluate their potential as SOFC cathode materials. In the case of YSr2Cu2FeO7+y, previous structural studies have shown that, although Fe predominantly substitutes for the Cu1 sites, some Fe does enter the Cu2 site. The value of y is typically c0.4–0.5 as prepared, with this bexcessQ oxygen going into vacant sites around the Cu1 position increasing the coordination for some of the Fe from 4 to 5/6 coordination. This phase, as prepared, is not a superconductor, although recent reports have shown that superconductivity can be induced by annealing at high temperatures under N2, followed by subsequent low temperature oxidation. The change is attributed to a redistribution of the Fe under the moderately reducing conditions, with all the Fe now favouring location on the Cu1 site [10]. In the case of YSr2Cu2CoO7+y, structural studies have shown that all the Co is located in the Cu1 site [4,5], leading to a complete Co layer. The preference of Co for tetrahedral coordination, rather than the square planar coordination favoured by Cu in undoped YBa2Cu3O7 y, leads to cooperative oxygen ordering and consequently an expanded unit cell. Unlike for YSr2Cu2FeO7+y, YSr2Cu2CoO7+y shows negligible oxygen excess ( yc0).
2. Experimental High purity Y2O3, SrCO3, CuO, Fe2O3, Co3O4 were used to prepare the samples YSr2Cu2FeO7+y, YSr2Cu2 CoO7+y. The starting materials were ground in the stoichiometric ratios and heated to 1273 K for 16 h. The powders were then reground and reheated under the same conditions. Conductivity measurements were made using the standard four probe d.c. technique. Sintered pieces were prepared
by pressing at 6000 kg cm 2 and sintering at 1273 K for 16 h. Conductivity measurements were then taken in air as a function of temperature (298–1173 K). Measurements of the conductivity versus oxygen partial pressure (measured via a YSZ sensor) were made at a temperature of 1173 K. For these measurements, the conductivities were initially taken in moving from oxidising to reducing conditions, where nitrogen was allowed to pass into the system. In this case the change in p(O2) proceeds quite rapidly (0.2–10 5 atm in about 1 h). The sample was then held under a N2 atmosphere for at least 15 h, in order for the system under test to reach equilibration, prior to any measurements being taken in the reverse direction (from reducing to oxidising). The N2 flow was then turned off, and oxygen was allowed to leak back into the system naturally. This typically took a period of approximately 24 h, with measurements made over this time. X-ray diffraction confirmed sample purity before and after measurement. Oxygen contents were determined by thermogravimetric analysis (reduction (5%/H2/95%N2) at 1223 K) using a Stanton Redcroft STA 780 thermal analyser. Thermogravimetric analysis was also used to investigate oxygen loss within the samples on heating between 298 and 1173 K (heating rate 10 8C/min, held at 1173 K for 20 min) in both air and N2. The chemical compatability of the phases YSr2Cu2FeO7+y and YSr2Cu2CoO7+y with potential electrolytes was examined by grinding together the sample and electrolyte in a 1:1 by mass ratio and heating at temperatures between 1173 and 1273 K for 1–3 weeks. The electrolytes examined included YSZ, CGO, La0.9Sr0.1Ga0.8Mg0.2O2.85, and Apatite-type (La9.33(Si/Ge)6O26).
3. Results The phases YSr2Cu2FeO7+y and YSr2Cu2CoO7+y were successfully prepared, with cell parameters of a=b= 3.830(3), c=11.465(9) 2 (YSr 2 Cu 2 FeO 7+y ) and a= 22.820(3), b=5.454(1), c=5.415(1) 2 (YSr2Cu2CoO7+y ).
Fig. 1. Plot of log r vs. log p(O2) at 1173 K for (a) YSr2Cu2CoO7+y, (b) YSr2Cu2FeO7+y.
J.E.H. Sansom et al. / Solid State Ionics 175 (2004) 99–102
Fig. 2. Plot of log r vs. 1/T for (a) YSr 2 Cu 2 FeO 7+y and (b) YSr2Cu2CoO7+y.
TGA studies indicated an oxygen content of 7.42(4) for YSr2Cu2FeO7+y and 7.02(4) for YSr2Cu2CoO7+y. Examination of the TGA traces for both samples in air and N2 between 298 and 1173 K showed no measureable oxygen loss for YSr2Cu2CoO7+y, while for YSr2Cu2FeO7+y significant oxygen loss was observed. In the case of air, oxygen loss began at a temperature c823 K, with the oxygen content decreasing to 7.19(4) at 1173 K. Heating in N2 resulted in a greater mass loss to 7.03(4) at 1173 K. For both samples the conductivity versus oxygen partial pressure measurements showed a decrease in conductivity with decreasing p(O2), which indicates that these materials are p-type conductors. The data showed significant hysteresis between oxidation/reduction curves (see Fig. 1). The pellet densities of these samples were 84% and 80% of the theoretical for YSr2Cu2CoO7+y and YSr2Cu2FeO7+y, respectively. Attempts were made to prepare more porous samples by sintering at lower temperature (1173 K), to examine whether the hysteresis was related to pellet density. However, severe problems were encountered in measuring pellets sintered at such lower temperatures due to their very brittle nature. It should be noted that the measured data do not represent the equilibrium state, since the oxygen partial pressure changes relatively quickly. This means that it is not possible to calculate accurate gradients for the plots of log r vs. log p(O2) to relate to the defect equations. However, at the two end points (0.2, c10 5 atm) the sample is well equilibrated, and so an approximate average gradient can be determined from the conductivity values at these two end points. Using these points, approximate gradients of 0.08 and 0.37 were determined for YSr 2Cu2CoO 7+y and YSr 2Cu2FeO7+y, respectively. The key observation from this is that the p(O2) dependence for YSr2Cu2FeO7+y is much greater than that observed for YSr2Cu2CoO7+y. This could be related to two factors. Firstly the observed large change in oxygen stoichiometry observed for YSr2Cu2FeO7+y. A consideration of the TGA data for YSr2Cu2FeO7+y shows an oxygen content change from 7.19(4) to 7.03(4) at 1173 K in going from air to N2, leading to a large change in the number of charge carriers. In contrast, TGA studies showed negligible
101
oxygen loss for YSr2Cu2CoO7+y. A second factor is the possible redistribution of the Fe over the two Cu sites under N2, with an increase in the Fe occupancy of the Cu1 site and consequent decrease in the occupancy at the Cu2 site, as suggested previously in the literature [10]. In addition to data on the p(O2) dependence of the conductivity at 1173 K, data on the variation of the conductivity with temperature in air has been collected. This data showed some interesting variations (Fig. 2). For YSr2Cu2FeO7+y, the conductivity shows semiconducting behaviour up to c823 K, above which a gradual decrease in conductivity is observed (r 823K=35 S cm 1, r 1173K=6 S cm 1). The decrease in conductivity at higher temperatures can be related to oxygen loss (as shown by TGA studies) reducing the number of charge carriers. The variation in conductivity with temperature observed for YSr2Cu2CoO7+y was particularly interesting. For this sample semiconducting behaviour was observed for the whole temperature range examined, but there was a steeper increase in the conductivity around 800 8C. We have recently collected high temperature X-ray diffraction data for this sample, which has shown that there is an orthorhombic-tetragonal phase transition at this temperature, and this phase transition is now being investigated in more detail. The chemical compatibility studies of these two samples and SOFC electrolytes showed significant reaction in all cases. For YSZ, insulating SrZrO3 was observed, while for the La9.33(Si/Ge)6O26 electrolytes, a complex mixture of phases was seen, which included (La/Sr)2CuO4 y and unidentified phases. A complex pattern was also observed for reaction with La0.9Sr0.1Ga0.8Mg0.2O2.85, while for CGO, the phases (Y/Ce) 2 Sr 2 Cu 2 MO 9+x (M=Fe, Co) were observed. The latter are related to the YSr2Cu2MO7+x (M=Fe,Co) phases by the replacement of the Y3+ by a (Y/ Ce)2O2x+ bfluorite blockQ, and we are currently investigating the high temperature conductivities of these materials.
Acknowledgements We would like to thank EPSRC and Merck Ltd for funding, and A.J. Wright (University of Birmingham) for the provision of high temperature XRD facilities.
References [1] J.G. Fletcher, J.T.S. Irvine, A.R. West, J.A. Labrinca, J.R. Frade, F.M.B. Marques, Mater. Res. Bull. 29 (1994) 1175. [2] S. Tsukui, M. Adachi, R. Oshima, H. Nakjima, F. Toujou, K. Tsukamoto, T. Tabata, Physica, C 351 (2001) 357. [3] P.R. Slater, C. Greaves, Physica, C 180 (1991) 299. [4] T. Den, T. Kobayashi, Physica, C 196 (1992) 141. [5] T.G.N. Babu, J.D. Kilgour, P.R. Slater, C. Greaves, J. Solid State Chem. 103 (1992) 472. [6] G. Roth, P. Adelmann, G. Heger, R. Knitter, T. Wolf, J. Phys. 1 (1991) 721.
102
J.E.H. Sansom et al. / Solid State Ionics 175 (2004) 99–102
[7] S.A. Sunshine, L.F. Schneemeyer, T. Siegrist, D.C. Douglas, J.V. Waszczak, R.J. Cava, E.M. Gyorgy, D.W. Murphy, Chem. Mater. 1 (1989) 331. [8] L.E. Moran, R. Rojas, M.A. Alario-Franco, J. Mater. Chem. 6 (1996) 1517.
[9] J.Q. Li, W.J. Zhu, Z.X. Zhao, D.L. Yin, Solid State Commun. 85 (1993) 739. [10] J. Shimoyama, K. Otzschi, T. Hinouchi, K. Kishio, Physica, C 341 (2000) 563.