This article was published in an Elsevier journal. The attached copyis furnished to the author for non-commercial research and

education use, including for instruction at the author’s institution,sharing with colleagues and providing to institution administration.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Renewable Energy 33 (2008) 331–335

Compositional dependence of electrical conductivityof Ce1�xTbxO2�d (0pxp1)

Fei Yea,�, Toshiyuki Moria, Ding Rong Oua, Motoi Takahashia, Jin Zoub,c, John Drennanc

aFuel Cell Materials Center, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, JapanbSchool of Engineering, The University of Queensland, St. Lucia, Brisbane, Qld. 4072, Australia

cCentre for Microscopy and Microanalysis, The University of Queensland, St. Lucia, Brisbane, Qld. 4072, Australia

Available online 20 June 2007

Abstract

Well dispersed Ce1�xTbxO2�d nano-powders were synthesized by using the carbonate coprecipitation method in an entire

compositional range of 0pxp1, which allowed the preparation of highly dense pellets by sintering at 1673K and the systematical study

of the electrical conductivity in such a wide compositional range. It was found that the conductivity increased with increasing Tb

concentration except that of x ¼ 0.80. Secondary phase were observed by using X-ray diffraction in the samples with xX0.80, which

might have negative impact on the conductivity of the samples.

r 2007 Elsevier Ltd. All rights reserved.

Keywords: Ceria; Powder; Microstructure; Conductivity

1. Introduction

Ceria doped with rare-earth elements has been investi-gated extensively as oxide ionic conducting solid electro-lytes for applications in solid oxide fuel cells (SOFCs) [1,2].Although many efforts have been focused on improving theoxygen ionic conductivity of solid electrolytes for applica-tion in SOFCs, some recent efforts began to introduceenhanced electronic conduction into ionic conductivematerials to develop mixed conductor for oxygen mem-brane application [3].

The mixed conduction can be achieved by doping amixed-valence element into a pure ionic conducting oxide.Based on this principle, mixed conduction of Pr-dopedceria has been studied [4–7]. Another rare-earth element,Tb, also has mixed valency. Shuk et al. [8] have studied theelectrical conduction of Ce1�xTbxO2�d (xp0.30). At973K, with increasing Tb concentration, the ionic con-ductivity increased and reached a maximum at x ¼ 0.25,while the electronic conductivity increased and exceededthe oxide ionic conductivity when x40.25, resulting a

highest total conductivity at x ¼ 0.25. At lower tempera-ture such as 773K, the total conductivity monotonouslyincreased with increasing Tb concentration from x ¼ 0.05to 0.30 [8]. However, the conduction behavior at higherdoping concentration (x40.30) remains unclear.To study the electrical conductivity in a wide composi-

tional range, dense sintered samples are required to beprepared from well-dispersed nano-powders with differentcompositions. Several wet-chemical methods have beenused to synthesis Tb-doped ceria powders, such ashydrothermal method [8], microemulsion method [9,10],and ammonia coprecipitation method [11,12]. However, deVries and Meng [11] reported that microcracks were oftenvisible on sintered samples with x ¼ 0.40 and 0.50 when theammonia coprecipitation method was used, which mightlimit the conductivity measurement at high Tb concentra-tion (e.g. x40.40). Recently, Li et al. [13] developed anammonium carbonate coprecipitation method to preparerare-earth doped ceria powders with a doping concentra-tion of 20 at%, which showed excellent sinterability.However, when they used this method to prepare dopedceria powders with different doping concentration, e.g.10–35 at% Y-doped ceria [14], only the powders withhigher Y concentration (15–35 at%) were round and well

ARTICLE IN PRESS

www.elsevier.com/locate/renene

0960-1481/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.renene.2007.05.014

�Corresponding author.

E-mail address: fei.ye@nims.go.jp (F. Ye).

Author's personal copy

dispersed. Therefore, the preparation conditions of theammonium carbonate coprecipitation method should beadjusted in order to synthesize well dispersed and roundpowders with different doping concentration.

In the present work, the carbonate coprecipitationmethod was used to syntheze Ce1�xTbxO2�d powders. Itwill be shown that by adjusting the preparation conditions,well-dispersed nano-powders can be synthesized in anentire compositional range of 0pxp1, and highly densesintered samples can be prepared from these powders.Then, the conductivity of Ce1�xTbxO2�d was studied in theentire compositional range in order to comprehensivelyunderstand how the composition affects the conductivity.

2. Experimental

Ce1�xTbxO2�d powders were synthesized via the ammo-nium carbonate coprecipitation method. The startingmaterials were Ce(NO3)3 � 6H2O (purity: 99.9%; KantoChemicals Co. Ltd., Japan) and Tb(NO3)3 � 6H2O (purity:99.9%; Soekawa Chemicals Co. Ltd., Japan) as cationsources. They were mixed at the predetermined molarratios and dissolved in 300mL distilled water. The totalconcentration was 0.15M for both Ce and Tb cations. Thesalt solution was dripped into 300mL of ammoniumcarbonate (AC; analytical grade; Wako Chemical Co. Ltd.,Japan) distilled water solution at a particular temperatureunder mild stirring. The resultant suspension was homo-genized for 1 h after the completion of precipitation. Thenthe precipitates were washed three times with distilledwater and two times with anhydrous alcohol beforedrying at room temperature under flowing nitrogen gas(200mL/min) over 24 h. The resultant precursors werecalcined under flowing oxygen (200mL/min) at 1073K for2 h to yield oxide powders.

The morphology of the powders with different concen-trations is greatly affected by the molar ratio R ¼ AC/(Ce3++Tb3+) and the reaction temperature for precipita-tion. Li et al. [13,14] prepared powders at 343K andR ¼ 10. Based on this experimental condition, we adjustedthe value of R and the reaction temperature (as listed inTable 1) in order to obtain well dispersed and roundpowders in the entire compositional range.

The calcined powders were isostatically pressed under200MPa pressure and the compacted bodies were heated to1673K at a rate of 5K/min and held for 6 h before the

samples were cooled to room temperature at a rate of5K/min. The densities of the sintered samples weremeasured by the Archimedes method. The morphologiesof the powders and the sintered samples were examined byscanning electron microscopy (SEM; Hitachi S-5000). SEMspecimens of the sintered samples were prepared bypolishing and thermally etching in air at 1573K for 2 hto reveal grain boundaries. The average grain sizes of thesintered samples were determined using the linear interceptmethod [15]. The crystal phases of the calcined powdersand sintered samples were investigated by X-ray diffraction(XRD; RINT2200HF). The average crystallite size D of thecalcined powders was calculated using the (5 1 1) peak ofthe fluorite structure by the Scherrer formula

D ¼0:9l

b cos y, (1)

b ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffib2m � b2s

q, (2)

where l is the wavelength of the X-ray, y is the diffractionangle, b, bm and bs are the corrected half-width, theobserved half-width and the half-width of the (5 1 1) peakin a standard sample of CeO2 (D�100 nm), respectively.Electrical conductivities of the sintered samples were

measured by DC three-point method at 673–1073K.Before the conductivity measurement, the samples werepainted with platinum and then fired at 1273K for 1 h toensure a good bond between the sample surfaces and theplatinum electrodes.

ARTICLE IN PRESS

Table 1

The molar ratio R ¼ AC/(Ce3++Tb3+) and the reaction temperature for precipitation during synthesizing Ce1�xTbxO2�d powders with different Tb

concentration x

x

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

R 3.5 6 9 10 11 9 6 6 6 6.5 7

Temperature (K) 323 333 343 348 348 343 333 323 313 303 303

Fig. 1. SEM micrograph of the calcined powder with x ¼ 0.50.

F. Ye et al. / Renewable Energy 33 (2008) 331–335332

Author's personal copy

3. Results and discussion

Fig. 1 is SEM image of the calcined powder withx ¼ 0.50, showing its spherical morphology and welldispersion. The morphologies of the powders with othercompositions are similar to that shown in Fig. 1. Theparticle sizes determined from SEM observations are about20–40 nm for the powders with xX0.1, while that of thepure ceria (x ¼ 0.0) is a bit larger (about 100 nm). From theXRD profiles of the powders, the average crystallite sizesare calculated by the Scherrer formula and given inTable 2, which are about 30–40 nm for xX0.1 and 91 nmfor the pure ceria. Since the crystallite sizes measured byXRD are consistent with the particle sizes observed bySEM, it can be concluded that each particle of the powdersis a single crystallite.

The morphologies of the sintered samples prepared fromthese powders are shown in Fig. 2. The densities of thesesintered samples are over 95% of their correspondingtheoretical densities. The average grain sizes of thesesamples were calculated and listed in Table 2. The averagegrain sizes are 0.53–1.30 mm for the samples with xX0.1,while that of the pure ceria is about 6.21 mm. Comparingwith the previous work in which microcracks could be seenin the sample with xX0.4 [11], no visible cracks can beobserved in this study for samples with xp0.60. Forsamples with xX0.70, micrometer sized cracks could beoccasionally observed. However, their fraction is too smallto have obvious effect on the conductivity measurement.

The crystal phases of the sintered samples were studiedby XRD. As shown in Fig. 3, for xp0.7, the XRD patternsof sintered samples show only simple fluorite structure,while extra peaks were observable for xX0.80 (somemarked by arrows), indicating that these samples consistof not only fluorite structure but also a secondary phase.Because the extra peaks is the most obvious in the samplewith x ¼ 0.90, this sample contains largest amount of thesecondary phase. However, the detailed structure of thesecondary phase remains unclear.

Fig. 4 shows the conductivities of the sintered samplesmeasured at different temperatures. Only the data of pureCeO2 fit well to the Arrhenius behavior. The conductivitiesof the samples with xX0.1 show slight curvatures attemperature about 973K, which may be related to thedisassociation of defect interactions [16]. In addition to thecurvatures, small kinks can be seen at about 773–873K for

ARTICLE IN PRESS

Table 2

The average crystallite sizes of the calcined powders and the average grain sizes of Ce1�xTbxO2�d sintered samples with different Tb concentration x

x

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Crystallite sizes of the powders (nm) 91.0 39.3 42.6 43.8 37.1 37.3 35.2 29.9 41.8 34.4 32.3

Grain sizes of the sintered samples (mm) 6.21 0.70 0.86 0.82 1.30 1.28 1.13 0.86 0.55 0.53 0.77

Fig. 2. SEM micrographs of sintered samples with x ¼ 0.10, 0.50

and 0.90.

F. Ye et al. / Renewable Energy 33 (2008) 331–335 333

Author's personal copy

the samples with x ¼ 0.10, 0.20, as indicated by arrows inFig. 4. The conductivities at about 1023 and 773K as afunction of Tb concentration are summarized in Fig. 5. Asthe Tb concentration increased, the conductivity increasedexcept that of x ¼ 0.80, showing an obvious decrease at773K. Such increase in the conductivity is obvious asxo0.30, which is basically consistent with the observationresult of Shuk et al. [8], while the conductivity does notincrease much as xX0.30.

As mentioned above, the average grain sizes vary in therange of 0.53–1.30 mm for the samples with xX0.1.However, by comparing the conductivities shown inFig. 5 with the grain sizes listed in Table 2, it can be seenthat such difference in the grain sizes has no considerableeffect on the compositional dependence of the conductivity.

Based on the experimental results, we believe that thesecondary phase has negative impact on the conductivity.It is interesting that although the sample with x ¼ 0.90 hasthe largest amount of secondary phase, the obviousdecrease in the conductivity was found in the sample withx ¼ 0.80. We suspect that this might be because the effectof the secondary phase on the conductivity is related to notonly the amount but also the morphology and distributionof the secondary phase. Furthermore, as shown in Fig. 5,the effect of the secondary phase on the conductivity is notobvious at 1023K as at 773K. It is possible because thesecondary phase mainly affect the ionic conductivity as

observed in other types of doped ceria, such as Y-doped[17] and Sm-doped [18] ceria. With increasing thetemperature, the electronic conductivity becomes more

ARTICLE IN PRESS

Fig. 3. XRD profiles of sintered samples.

-6

-5

-4

-3

-2

-1

0

0 1

logσ

(Sc

m-1

)

Tb concentration x

0.40.2 0.6 0.8

1023 K

773 K

Fig. 5. Conductivities at 773 and 1023K as a function of Tb concentra-

tion x.

1

logσ

T (

S cm

-1 K

)

x =1.00

x =0.90

x =0.80

x =0.70

x =0.60

x =0.50

x =0.40

x =0.30

x =0.20

x =0.10

x =0.00

1.7

1.2

0.7

0.2

-0.3

-0.8

-1.3

-1.8

-2.3

-2.80.9

1000/T (K-1)

1.51.41.31.21.1

Fig. 4. Arrhenius plot of the conductivities of Ce1�xTbxO0�d with

different Tb concentration x.

F. Ye et al. / Renewable Energy 33 (2008) 331–335334

Author's personal copy

dominating in Tb-doped ceria [19]. Therefore, the influenceof secondary phase on the ionic conductivity at highertemperature is covered by the increased electronic con-ductivity and then could be difficult to be observed as atlower temperature.

4. Conclusion

Dense Ce1�xTbxO2�d sintered samples can be preparedin an entire compositional range of 0pxp1 from nano-sized and well-dispersed powders. It was observed that theelectrical conductivity of these samples increased withincreasing Tb concentration in the entire compositionalrange except that of x ¼ 0.80 at 773K. Such increase in theconductivity is readily visible as xo0.30, but slows down asxX0.30. Furthermore, the conductivities measured atdifferent temperatures did not totally fit the Arrheniusbehavior. Curvatures and kinks were seen in thelog(sT)�1/T plot. Through XRD study, secondary phasein the samples with xX0.80 were identified. It is believedthat the secondary phase can have negative impact on theconductivity.

Acknowledgment

This work was partly supported by Japan Society for thePromotion of Science (JSPS).

References

[1] Steele BCH. Solid State Ionics 2000;134:3.

[2] Steele BCH, Heinzel A. Nature 2001;414:345.

[3] Gellings PJ, Bouwmeester HJM. Handbook of solid state electro-

chemistry. New York: CRC Press; 1997. p. 630.

[4] Takasu Y, Sugino T, Matsuda Y. J Appl Electrochem 1984;14:79.

[5] Nauer M, Fticos C, Steele BCH. J Eur Ceram Soc 1994;14:493.

[6] Shuk P, Greenblatt M. Solid State Ionics 1999;116:217.

[7] Qi X, Lin YS, Holt CT, Swartz SL. J Mater Sci 2003;38:1073.

[8] Shuk P, Greenblatt M, Croft M. Chem Mater 1999;11:473.

[9] Hungrıa AB, Martınez-Arias A, Fernandez-Garcıa M, Iglesias-Juez

A, Guerrero-Ruiz A, Calvino JJ, et al. Chem Mater 2003;15:4309.

[10] Wang X, Hanson JC, Liu G, Rodriguez JA, Iglesias-Juez A,

Fernandez-Garcıa M. J Chem Phys 2004;121:5434.

[11] de Vries KJ, Meng GY. Mater Res Bull 1998;33:357.

[12] Blanco G, Calvino JJ, Cauqui MA, Corchado P, Lopez-Cartes C,

Colliex C, et al. Chem Mater 1999;11:3610.

[13] Li JG, Ikegami T, Mori T, Wada T. Chem Mater 2001;13:2913.

[14] Li JG, Ikegami T, Wang Y, Mori T. J Solid State Chem 2002;168:52.

[15] Mendelson MI. J Am Ceram Soc 1969;52:443.

[16] Subbarao EC, Maiti HS. Solid State Ionics 1984;11:317.

[17] Ou DR, Mori T, Ye F, Takahashi M, Zou J, Drennan J. Acta Mater

2006;54:3737.

[18] Mori T, Drennan J, Lee JH, Li JG, Ikegami T. Solid State Ionics

2002;154–155:461.

[19] Tuller HL. In: Sørensen OT, editor. Nonstoichiometric oxides.

London: Academic Press; 1981. p. 271.

ARTICLE IN PRESSF. Ye et al. / Renewable Energy 33 (2008) 331–335 335