Nanocrystals of zirconia- and ceria-based solid electrolytes: Syntheses and properties

ARTICLE IN PRESS

Science and Technology of Advanced Materials 8 (2007) 524–530 www.elsevier.com/locate/stam

Review

Nanocrystals of zirconia- and ceria-based solid electrolytes: Syntheses and properties Takahisa Omata, Yuji Goto, Shinya Otsuka-Yao-Matsuo Division of Materials and Manufacturing Science, Graduate School of Engineering, Osaka University, 2-1 Yamada-oka, Suita 565-0871, Japan Received 1 June 2007; received in revised form 22 August 2007; accepted 8 September 2007 Available online 22 October 2007

Abstract Nano-effect on solid electrolytes that appears as a conductivity enhancement is expected as a way to develop or improve practical solid ionic devices, such as solid oxide fuel cells. Interfaces play a major role in the enhanced conductivity. Using nanocrystals (NCs) as the starting material, the nanostructured materials containing many interfaces, i.e., grain boundaries, can be simply made by forming the NCs with successive sintering without grain growth. The past decade has seen significant advances in the syntheses of solid electrolyte NCs and understanding the nano-effects on the ionic conductivity. In the present paper, the syntheses of zirconia- and ceria-based NCs, which are important constituent materials of solid oxide fuel cells, and their grain size-dependent conductivity due to the nano-effect are briefly reviewed. r 2007 NIMS and Elsevier Ltd. All rights reserved. Keywords: Nanocrystals; Solid oxide fuel cell; Synthesis; Electrical conductivity

Contents 1. 2.

3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syntheses of NCs and their characterization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Zirconia-based materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Ceria-based materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electrical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Zirconia-based NCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Ceria-based NCs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary and feature challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Since the discovery of the significantly higher ionic conductivity of the nanocomposite of LiI and porous Al2O3 than that of bulk LiI [1], the enhancement of the Corresponding author. Tel.: +81 6 6879 7462; fax: +81 6 6879 7464.

E-mail address: [email protected] (T. Omata).

524 525 525 526 527 527 527 529 529 530

ionic conductivity at the interfaces and/or near interface regions has been drawing the attention as nano-effects appearing in solid electrolytes. The ionic conducting properties in the nanometer region around the interfaces have been experimentally and theoretically studied, and several important developments have occurred as a result of these efforts. The discovery of the strongly enhanced ionic conductivity in the multilayered BaF2/CaF2 thin film

1468-6996/$ - see front matter r 2007 NIMS and Elsevier Ltd. All rights reserved. doi:10.1016/j.stam.2007.09.001

ARTICLE IN PRESS T. Omata et al. / Science and Technology of Advanced Materials 8 (2007) 524–530

and the development of the space charge theory, which explains the conductivity enhancement, are such examples [2]. Today, it is expected that the study of the nano-effect is the next phase, which focused on developing the strongly enhanced conductivity due to nano-effects in practical materials. Nanocrystals (NCs) are expected as suitable starting materials for developing nanostructured practical devices, in which many interfaces play a major role in the nanoeffect, because of their cost efficiency and mass productivity. Therefore, the synthesis of solid electrolyte NCs can be regarded as the starting point in order to develop practical nanostructured solid-state ionic devices. The solid oxide fuel cell (SOFC) is an attractive target for application of the nano-effect. When the conductivity of the solid electrolytes and the electrodes consisting of SOFCs is enhanced by the nano-effect, the effect on our life is significant. From the viewpoint described above, the recent advances in the syntheses of NCs of zirconia- and ceria-based materials, which are the major constituent materials of SOFCs, and the size-dependent conductivity caused by the nano-effect are briefly reviewed in this paper. 2. Syntheses of NCs and their characterization 2.1. Zirconia-based materials Various methods have been developed for the syntheses of less agglomerated zirconia-based NCs. Among them, inert gas condensation (IGC) using thermal evaporation has a merit that the as-synthesized powder has a crystalline form [3,4]. The typical experimental setup is described in Ref. [3]. For the ZrO2 NCs, zirconium monoxide evaporated in a UHV-chamber is condensed as ZrO2 on a cold plate cooled by liquid nitrogen. The average particle size of the obtained powder is approximately 10 nm. The powder is completely crystallized, and its defect concentration, such as dislocations, is very rare [3]. The particle size and its distribution depend on the cooling rate during the condensation. When the cooling rate is high, small NCs are obtained. However, the particle size and its distribution are not exactly controllable, because the crystallization, i.e., the condensation, rate is extremely high. Instead of Joule heating, gas condensation applied by DC magnetron sputtering using Zr metal as the sputtering target was reported [5]. By applying the sputtering, purer ZrO2 NCs can be obtained. Combustion syntheses also results in wellcrystallized ZrO2 NCs; however, the as-synthesized NCs are strongly agglomerated [6–8]. The chemical solution routes, such as precipitation [9–13] and sol–gel [14,15] techniques, are generally used for the zirconia-based NCs. Typical procedures for these techniques are as follows. For the precipitation technique, zirconium oxychloride (ZrOCl2) or zirconium oxynitrate (ZrO(NO3)2) is dissolved in water, and then an aqueous ammonium solution is added dropwise until the pH becomes sufficiently high. For the sol–gel technique, the

525

zirconium iso-propoxide (Zr(OPr)4) propanol complex is dissolved in dry propanol, and water is then added dropwise. For the synthesis of the Y2O3-stabilized ZrO2 (YSZ), yittrium nitrate (Y(NO3)3) or yittrium chloride (YCl3) as well as the zirconium source are dissolved in the starting solution for both techniques. The as-synthesized powders obtained by these chemical solution routes are amorphous complex oxyhydroxides, and the particle size and their distributions are not exactly controllable due to their high reaction rate. However, many researchers have employed these techniques due to the simplicity and easiness. The as-synthesized amorphous oxyhydroxide is crystallized by post-annealing above 300 1C. Crystallization is complex and usually leaves the hydroxyl-group, i.e., interstitial protonic defects, in the resulting NCs. Chadwick et al. [16] studied in detail the crystallization by X-ray absorption and NMR spectroscopies. According to their study, crystallization is basically represented as dehydroxylation and cross-linking [17], but dehydoxylation is not complete prior to crystallization; the composition is not accurately described as ZrO2 until 4500 1C. In order to obtain small NCs, the amorphous oxyhydroxides are frequently annealed at 300 1C. The properties of the obtained NCs should be described by taking the remaining protonic defects into consideration. The size of NCs is usually X10 nm after crystallization [9,10,16], while wellcrystallized NCs with a single nm particle size cannot be obtained by these techniques. The hydrothermal process results in crystalline NCs without post-annealing [18–22]. For example, 80 1C and subsequent 180 1C hydrothermal treatment of the aqueous solution of ZrO(NO3)2, Y(NO3)3 and urea results in wellcrystallized YSZ NCs with 10 nm particle size [18]. However, the obtained NCs may involve significant protonic defects, because the NCs are crystallized from the aqueous solution. In contrast to the aqueous solution route, which results in amorphous oxyhydroxides, the non-hydrolytic sol–gel reaction reported by Joo et al. [23] results in wellcrystallized NCs. The overall reaction, termed alkylhalide elimination, is described as follows: ZrðOPrÞ4 þ ZrX4 ! 2ZrO2 þ 4PrX;

(1)

where Zr(OPr)4 is a zirconium iso-propoxide and ZrX4 is a zirconium halide. The organic solvent with a high boiling temperature, e.g., tri-octylphosphineoxide (TOPO) and oleylamine, are usually used for the reaction solvent [24–27]; the starting solution, i.e., the mixture of Zr(OPr)4 and ZrX4 in the organic solvent, reacts at 300–350 1C. Unlike the sol–gel reaction using an aqueous solution, the non-hydrolytic sol–gel technique does not involve water as the solvent and/or by-product; therefore, the protonic defect, i.e., the OH-group, in the resulting NCs should be negligible. The solvent, such as the TOPO and oleylamine, behaves as a surfactant. These surfactant molecules coat the surfaces of the NCs, termed capping; therefore, the agglomeration of the NCs is suppressed, and the NCs are

ARTICLE IN PRESS T. Omata et al. / Science and Technology of Advanced Materials 8 (2007) 524–530

526

(b) Volume distribution / Arb. units

(a)

0

1

2 3 4 5 Particle size, d / nm

6

7

Fig. 1. (a) HRTEM image and (b) particle size distribution evaluated from the SAXS for ZrO2 NCs synthesized by the non-hydrolytic sol–gel technique.

RM2O2CO2R þ R0 2O2M ! RM2O2MR þ R0 2O2CO2R; (2)

where R and R0 are alkyl groups. Details of this process are described in another review paper [29]. The powder X-ray diffraction (XRD) profiles of the NCs are highly broadened due to their small crystalline size; therefore, it is very hard to detect any structural change in the zirconia NCs upon Y2O3-doping, i.e., transformation from the monoclinic to the tetragonal and cubic forms. Fig. 2(a) shows the XRD profiles of the 3 and 8 mol% Y2O3-doped ZrO2 NCs synthesized by co-precipitation and subsequent calcination at 650 1C [11]. The 3 and 8 mol% Y2O3-doped zirconias should have the tetragonal and cubic forms, respectively, under equilibrium condition. For the tetragonal form, diffractions at 2y251, and 501 and 601 become doublets; however, distinct peak splitting cannot be observed due to the broadened profile, and the profile appeared to be identical to that of the 8 mol% Y2O3-doped NCs. It is very difficult to identify by XRD whether the obtained phase is the tetragonal or cubic form. In contrast to the XRD profiles, a distinct difference in the number of peaks is observed in the Raman spectra (Fig. 2(b)) [11]. The changes in the Raman bands depending on the structural

3mol% Y2O3-doped ZrO2

8mol% Y2O3-doped ZrO2

Tetragonal zirconia

8mol% Y2O3-doped ZrO2

625

Cubic zirconia

20

30

40 50 2θ / degree

60

70

641

613

473

3mol% Y2O3-doped ZrO2

464

obtained in a colloidal solution form. Because the crystal growth is kinetically suppressed by surface capping, the size of the obtained NCs is several nanometers. Figs. 1(a) and (b) show the high-resolution transmission electron microscope (HRTEM) image and the particle size distribution evaluated from the small angle X-ray scattering (SAXS) of the ZrO2 NCs [27], which are synthesized by the non-hydrolytic sol–gel technique. These figures indicate that the obtained NCs are well-crystallized and highly monodispersed. Y2O3 can be easily doped into ZrO2 by adding Y(OPr)3 or YX3 into the starting solution [28]. The non-hydrolytic condensation reaction (1) usually results in a hydrogen halide (HX) as the by-product. Therefore, the process is not suitable for the materials, which easily react with the HX. For such cases, the ester elimination reaction described by the following overall reaction (2) is suitable because a highly reactive by-product is not generated.

400

500

600

700

Raman shift / cm-1

Fig. 2. (a) X-ray diffraction profiles and (b) Raman spectra of 3 and 8 mol% Y2O3-doped ZrO2 NCs.

change upon Y2O3-doping into ZrO2 have been established [30–34]. The 641, 613 and 473 cm1 bands observed for the 3 mol% Y2O3-doped NCs are attributable to the Raman active modes for the tetragonal form, and the asymmetric 625 cm1 band and weak 464 cm1 hump observed for the 8 mol% Y2O3-doped NCs are attributable to the Raman active and normally inactive modes for the cubic form, respectively. Thus, Raman spectroscopy gives information that can distinguish the crystal form of the zirconia-based NCs. 2.2. Ceria-based materials Ceria-based NCs, whose sizes are less than several tens of nanometers, can be synthesized by IGC [35], precipitation [36–43], combustion [44–46], hydrothermal [47,48] and non-hydrolytic sol–gel methods [27,49,50] as in the case of the zirconia-based NCs. Among them, the precipitation method of NH4OH addition into a Ce(NO3)3 aqueous solution is generally used. Unlike the zirconia-based case,

ARTICLE IN PRESS T. Omata et al. / Science and Technology of Advanced Materials 8 (2007) 524–530

3. Electrical properties 3.1. Zirconia-based NCs According to the electrical conductivity studies of the YSZ microcrystallines, the total conductivity decreased with decreasing grain size [61,62], because the grain boundary conductivity decreased with decreasing grain size. In contrast to the microcrystalline case, the enhanced conductivity upon decreasing the grain size was observed for the YSZ NCs [63–65], whose grain size was less than 100 nm (Fig. 3). When the space charge theory, which well explains the enhanced conductivity for the BaF2/CaF2 multilayered films [2], is applied to the YSZ case, the Debye length l ¼ [ee0RT/2F2CN]1/2 was calculated to be 0.2 nm, assuming that the dielectric constant, e, and the bulk carrier concentration, CN, are 30 and 1.3  1021 cm3, respectively. For this case, an enhanced conductivity

Temperature / °C 1100 900 700 1

500

300

8YSZ on Si dg=55nm

0

log [σ / S cm-1]

the as-precipitated powder crystallizes by aging of the product solution below 80 1C or by drying of the product powder. The reaction is generally understood as oxidation from the as-precipitated CeIII(OH)3 to CeIV(OH)4 by the dissolved oxygen and subsequent dehydroxylation during aging [37]. The oxidation from CeIII to CeIV was visually observed, i.e., the color of the solution changed from purple for CeIII to dark brown and finally light yellow for CeIV [51]. However, it was shown from the equilibrium calculation in the CeIII–CeIV–H2O–O2 system that hydrated CeO2 can be directly precipitated from the Ce3+ solution [52]. When (NH3)2CeIV(NO3)6 is used as a starting material [53], the CeIV state may be stabilized in the aqueous solution due to the formation of complex ions. For such a case, it is plausible that the CeO2 NCs are directly precipitated from the aqueous solution. Further studies are required in order to ascertain the precipitation reaction and its mechanism. For the doped CeO2 NCs, such as the Gd2O3-doped CeO2 (GDC), Gd(NO3)3 as well as the cerium source is dissolved in the starting solution [38]. The valence state of the cerium ions has been discussed for the undoped ceria NCs. Upon decreasing the size to less than several tens of nanometers, an anomalous lattice expansion is observed [54]. The mechanism was based on the presence of Ce3+, whose size is larger than that of Ce4+ in the ceria NCs. The presence of Ce3+ ions was supported by XRD [55], EXAFS [56] and XPS [57] data; the Ce3+ ion concentration was reported to be as high as 17% for the 30 nm NCs and 44% for the 3 nm NCs [57]. The driving force of the reduction was explained by the relaxation of the strain due to the extremely high surface area of the NCs. For the bulk cerium oxides, many intermediate IV compounds appear between the CeIII 2 O3 and Ce O2, i.e., CenO2n2 phases [58–60]; the oxygen-deficient site in the bulk CenO2n2 phases is in an ordered arrangement. It is very interesting whether or not the oxygen-deficient site in the reduced ceria NCs is ordered.

527

-1 -2

10YSZ on MgO dg=15nm 10YSZ on MgO dg=29nm

8YSZ on Al2O3 dg=20nm

-3 -4

8YSZ polycrystal dg=2400nm

-5 8

10

12 104

T

14 -1

/K

16

18

-1

Fig. 3. Arhenius plot of the electrical conductivity of YSZ NCs. Data were obtained from Refs. [63–65]. The 8YSZ on Al2O3 film was prepared by using polymer precursor spin coating and successive drying and crystallization at 800 1C. The 8YSZ on Si film was prepared by spin coating of YSZ-sol polymer and successive drying at 110 1C, firing at 400 1C and crystallization at 500 1C. The 10YSZ on MgO films was prepared by pulsed laser deposition on single crystal MgO substrate.

should appear for the NCs with a grain size less than 1 nm; the space charge theory cannot explain the enhanced conductivity for the YSZ NCs. The theoretical model, which can completely describe the enhanced conductivity in the YSZ NCs, is yet to be established, but the defect segregation in the vicinity of the grain boundary is proposed as one of the possible mechanisms [66, 67]. Details of the defect segregation will be described in the next section. Densification suppressing grain growth is an important process in order to form an electrolyte from the NC powders. High-pressure densification by applying several GPa is a possible process at this time [68,69]. It was reported that the NCs can be densified up to 92% of their theoretical density at 4.5 GPa for the YSZ NCs, and the grain growth was suppressed to less than 20 nm. However, its electrical conductivity was two orders of magnitude lower than that of the microcrystalline samples [68]. This observation is not consistent with the other studies; therefore, further study is needed. Densification applying several GPa is not suitable for the practical production because of its low productivity. To develop the densification process without grain growth with high productivity is an important issue in order to use the NCs in the practical devices. 3.2. Ceria-based NCs For the ceria-based NCs, the observed nano-effect on the electrical conductivity can be categorized into two groups

ARTICLE IN PRESS T. Omata et al. / Science and Technology of Advanced Materials 8 (2007) 524–530

528

depending on whether the NCs are undoped or heavily doped. The conductivity increases with the decreasing grain size for both the undoped and heavily doped ceria NCs as shown in Fig. 4; however, the major conduction carriers are the electrons for the undoped and slightly doped cases [70–75], but the oxide ions for the heavily doped case [76–81].

For the undoped and slightly doped cases, the enhanced electronic conductivity has been possibly explained by the two brick-layer models consisting of (i) a grain core and grain boundary core, and (ii) a grain core, grain boundary core and space charge zone as shown in Fig. 5. Kosacki, Tuller et al. discussed the enhancement from the formation enthalpy of the oxygen vacancies, DH, and implicitly used

Temperature / °C 900

0

800

Temperature / °C

600

700

500

900 700 500 0

CeO 2 dg=10nm

log [ σ/ S cm-1]

log [ σ/ S cm-1]

10mol% Gd2O3 on saphire PLD dg=46nm

-2

-4

CeO2 spray dg=230nm

100

10mol% Gd2O3 on saphire spray dg=29nm

CeO2 spray dg=37nm

-2

300

CeO 2 dg=20nm

-6

10%mol Sm2O3 microcrystalline 10mol% Gd2O3 dg=36nm dg=15nm

-4

-6 10mol% Gd2O3

dg=36nm -8 microcrystalline CeO 2 d g=5μm -10

8

9

10 104

-8 11

T

-1

10%mol Y2O3 dg=41nm microcrystalline

/K

12

13

8

-1

12

16 104

24

20 T

-1

/K

-1

Fig. 4. Arhenius plots of the electrical conductivity of (a) undoped and (b) doped CeO2 NCs. Data were obtained from Refs. [76,78,80]. The thin films of CeO2 and GDC indicated as spray in the figures were prepared by spray pyrolysis of cerium nitrate and gadolinium chloride solutions. The substrate temperature was at 310 1C. The CeO2 NC films, whose grain sizes are 10 and 20 nm, and 10 mol% GDC films, whose grain sizes are 15 and 36 nm, were prepared by spin coating a polymer precursor solution. The solution was prepared by cerium and gadolinium nitrate, ethylene glycol and nitric acid. The YDC and SDC films were prepared by fast-firing process at 700 1C for 3 min of uniaxial-pressed ultra-fine powders.

(a)

e’ VO

bulk

gc

scz

(b)

e’ VO Fig. 5. Schematic illustrations of proposed conduction path by the brick-layer model for the CeO2 NCs. (a) Model consisting of grain core (bulk) and grain boundary core (gc), and (b) model consisting of grain core, grain boundary core and the space charge zone (scz). In the former model corresponding to model (i) in the text, both electronic and ionic conductions are enhanced. In the latter model corresponding to model (ii) in the text, the space charge zone enhances the electronic conduction, but it behaves as a barrier to the ionic conduction. The latter model well explains the observation for CeO2 NCs.

ARTICLE IN PRESS T. Omata et al. / Science and Technology of Advanced Materials 8 (2007) 524–530

model (i) [71]. DH was evaluated from the oxygen pressure dependence of the conductivity and the activation energy obtained from the Arrhenius plot of the electrical conductivity. They determined a smaller DH for the NCs than that for the microcrystallines. The formation of oxygen vacancies is described by the followng reaction: Oxo ! Vo þ 2e0 þ 1=2O2 .

(3)

Therefore, a decrease in the DH of the oxygen vacancies results in an increased electron density. The increase in the electron density in the grain boundary core caused the enhanced conductivity. There is an objection to their model such that the model cannot explain the enhancement for only the electronic conduction, because a decrease in DH results in an increase in the concentration of both the electrons and oxygen vacancies [75]. However, the model cannot be disregarded due to the following reason: the decrease in the DH of the oxygen vacancies is consistent with the presence of many Ce3+ ions in the ceria NCs as described in Section 2.2 [54–57]. When the oxygen vacancies in the reduced ceria NCs are ordered as observed in the bulk CenO2n2 phases [58–60], the inferred mobility of the oxide ions is suppressed. Based on this assumption, only the enhancement of the electronic conduction due to the high electron density and the suppression of the ionic conduction due to the low mobility are plausible. Tscho¨pe [81] discussed the enhanced conductivity of undoped ceria NCs based on the model (ii) including the space charge zone using the Gouy–Chapman theory [82], which describes the deviation of the point defect concentrations from their bulk value in the vicinity of the electrically charged grain boundaries, i.e., the space charge zone. According to the model, the electronic carrier is enriched in the space charge zone; therefore, the space charge zone becomes a highly conducting path with respect to the electronic current. On the contrary, depletion of the oxygen vacancies in the space charge zone occurs. Therefore, the space charge zone behaves as a barrier to the ionic conduction as shown in Fig. 5(b). Kim, Maier et al. also explained the electrical properties of the undoped ceria NCs using model (ii), but they employed the Mott–Schottky model in order to evaluate the charge concentration profile in the vicinity of the interfaces [73–75]. For the heavily doped ceria NCs, the enhanced ionic conductivity was explained by the defect segregation in the vicinity of the grain boundaries and decrease in the activation energy, Ea, of conduction in most studies [76–79]. The enriched dopant concentration in the grain boundary with respect to the bulk was experimentally detected by several independent techniques such as electron energy-loss spectroscopy, HRTEM and nano-Auger spectroscopy [77–79]. Based on these observations, the enhanced ionic conductivity in the heavily doped ceria NCs has been described by the higher dopant concentration in the grain boundary than that in the bulk and the high grain boundary density due to the extremely small grain size. Bellino et al. [78] explained the enhanced conductivity by a

529

decrease in Ea, but they attributed the decrease in Ea to the increase in the grain boundary diffusivity. 4. Summary and feature challenges In the present paper, the synthesis and the sizedependent electrical conductivity due to the nano-effect for zirconia- and ceria-based NCs were briefly reviewed. The advances in NC syntheses have allowed an understanding of the nano-effect in these materials possessing a grain size less than 100 nm. The major interest in the nano-effect on the electrical conductivity is the mechanism of the enhanced conductivity. The space charge theory provides a powerful model to describe the enhanced conductivity. However, the mechanism still remains an open question, because some experimental results, which cannot be described by the theory, remain. On the other hand, one must still elucidate the fundamental properties of the NCs as to whether or not the oxygen vacancies in the reduced ceria NCs are ordered. In addition to these issues, the following items are very interesting challenges: (a) Grain size effect on electrical conductivity for grains possessing a single nanometer grain size: Investigations with respect to the electrical conductivity of zirconiaand ceria-based NCs have been covered for grain sizes larger than 10 nm. The nano-effect for the single nanometer grains is yet to be clarified. (b) Hetero-interfaces: In previous studies, the target materials consisted of doped zirconia or ceria NCs; therefore, the nano-effect introduced by the homointerfaces has been primarily addressed. The strong conductivity enhancement due to the nano-effect has been observed for the materials containing the heterointerfaces [1,2]. For the hetero-interfaces, conductivity enhancement due to atomic defect is also expected. Recently, an enhanced conductivity for the doped zirconia/ceria multilayered films [83] and the syntheses of core/shell-type composite NCs, such as CeO2/ZrO2 [27] and ZrO2/Al2O3 [84], have been reported. Further advances are expected for investigating composite NCs. (c) Artificial assembly of NCs: Recent developments of the artificial assembly technique of NCs are remarkable. It is expected that the artificial assembly of NCs leads a way to the high-performance solid electrolytes and electrodes.

Acknowledgment This work was supported in part by a Grant-in-Aid for Scientific Research on Priority Area, ‘‘Nanoionics (439)’’ by the Ministry of Education, Culture, Sports and Technology and a research grant from the Hosokawa Powder Technology Foundation.

ARTICLE IN PRESS 530

T. Omata et al. / Science and Technology of Advanced Materials 8 (2007) 524–530

References [1] C.C. Liang, J. Electrochem. Soc. 120 (1973) 1289. [2] N. Sata, K. Eberman, K. Ebert, J. Maier, Nature 408 (2000) 946. [3] R. Nitsche, M. Rodewald, G. Skandan, H. Fuess, H. Hahn, NanoStruct. Mater. 7 (1996) 535. [4] V.V. Srdic, M. Winterer, J. Eur. Ceram. Soc. 26 (2006) 3145. [5] H. Hahn, R.S. Averback, J. Appl. Phys. 67 (1990) 1113. [6] K.R. Venkatachari, D. Huang, S.P. Ostrander, W.A. Schulze, G.C. Stangle, J. Mater. Res. 10 (1995) 756. [7] R.E. Jua´rez, D.G. Lamas, G.E. Lascalea, N.E. Walso¨e de Reca, J. Eur. Ceram. Soc. 20 (2000) 133. [8] Z. Lei, Q. Zhu, S. Zhang, J. Eur. Ceram. Soc. 26 (2006) 397. [9] H. Liu, L. Feng, X. Zhang, Q. Xue, J. Phys. Chem. 99 (1995) 332. [10] S. Roy, J. Ghose, Mater. Res. Bull. 35 (2000) 1195. [11] A. Ghosh, A.K. Suri, M. Pandey, S. Thomas, T.R.R. Mohan, B.T. Rao, Mater. Lett. 60 (2006) 1170. [12] F. Chen, L. Huang, Z. Zhong, G.J. Gan, S.M. Kwan, F. Kooli, Mater. Chem. Phys. 97 (2006) 162. [13] G.E. Rush, A.V. Chadwick, I. Kosacki, H.U. Anderson, J. Phys. Chem. B 104 (2000) 9597. [14] B. Julian, R. Corberan, E. Cordoncillo, P. Escribano, B. Viana, C. Sanchez, Nanotechnology 16 (2005) 2707. [15] S. Shukla, S. Seal, R. Vanfleet, J. Sol–Gel Sci. Technol. 27 (2003) 119. [16] A.V. Chadwick, G. Mountjoy, V.M. Nield, I.J.F. Poplett, M.E. Smith, J.H. Strang, M.G. Tucker, Chem. Mater. 13 (2001) 1219. [17] X. Turrillas, P. Barnes, D. Gascoigne, J.Z. Turner, S.L. Jones, C.J. Norman, C.G. Pygall, A.J. Dent, Radiat. Phys. Chem. 45 (1995) 491. [18] Y.-W. Zhang, X. Sun, G. Xu, S.-J. Tian, C.-S. Liao, C.-H. Yan, Phys. Chem. Chem. Phys. 5 (2003) 2129. [19] A. Cabanas, J.A. Darr, E. Lester, M. Poliakoff, Chem. Commun. (2000) 901. [20] S. Somiya, T. Akiba, J. Eur. Ceram. Soc. 19 (1999) 81. [21] M. Yoshimura, S. Somiya, Mater. Chem. Phys. 61 (1999) 1. [22] J.D. Lin, J.G. Duh, J. Am. Ceram. Soc. 81 (1998) 853. [23] J. Joo, T. Yu, Y.W. Kim, H.M. Park, F. Wu, J.Z. Zhang, T. Hyeon, J. Am. Chem. Soc. 125 (2003) 6553. [24] J. Tang, J. Fabbri, R.D. Robinson, Y. Zhu, I.P. Herman, M.L. Steigerwald, L.E. Brus, Chem. Matter. 16 (2004) 1336. [25] J. Tang, F. Zhang, P. Zoogman, J. Fabbri, S.-W. Chan, Y. Zhu, L.E. Brus, M.L. Steigerwald, Adv. Funct. Mater. 15 (2005) 1595. [26] R.D. Robinson, J. Tang, M.L. Steigerwald, L.E. Brus, I.P. Herman, Phys. Rev. B 71 (2005) 115408. [27] T. Omata, S. Sasai, Y. Goto, M. Ueda, S. Otsuka-Yao-Matsuo, J. Electrochem. Soc. 153 (2006) A2269. [28] Y. Goto, T. Omata, S. Otsuka-Yao-Matsuo, to be published. [29] A. Vioux, Chem. Mater. 9 (1997) 2292. [30] C.M. Phillippi, K.S. Mazdiyasni, J. Am. Ceram. Soc. 54 (1971) 254. [31] M. Ishigame, T. Sakurai, J. Am. Ceram. Soc. 60 (1967) 367. [32] E. Anastassakis, B. Papanicolaou, I.M. Asher, J. Phys. Chem. Solids 36 (1975) 667. [33] V.G. Kerimidas, W.B. White, J. Chem. Phys. 59 (1973) 1561. [34] A. Geinberg, C.H. Perry, J. Phys. Chem. Solids 42 (1981) 513. [35] N. Guillou, L.C. Nistor, H. Fuess, H. Hahn, NanoStruct. Mater. 8 (1997) 545. [36] P.-L. Chen, I.-W. Chen, J. Am. Ceram. Soc. 76 (1993) 1577. [37] B. Djuricic, S. Pickering, J. Eur. Ceram. Soc. 19 (1999) 1925. [38] J. Markmann, A. Tsho¨pe, R. Birringer, Acta Mater. 50 (2002) 1433. [39] S. Kim, R. Merkle, J. Maier, Solid State Ionics 161 (2003) 113. [40] M. Kamruddin, P.K. Ajikumar, R. Nithya, A.K. Tyagi, B. Raj, Scr. Mater. 50 (2004) 417. [41] H.-I. Chen, H.-Y. Chang, Ceram. Int. 31 (2005) 795. [42] M.-S. Tsai, J. Cryst. Growth 274 (2005) 632. [43] H.-J. Choi, J. Moon, H.-B. Shim, K.-S. Han, E.-G. Lee, K.-D. Jung, J. Am. Ceram. Soc. 89 (2006) 343. [44] S. Basu, P.S. Devi, H.S. Maiti, J. Mater. Res. 19 (2004) 3162.

[45] T. Mokkelbost, I. Kaus, T. Grande, M.-A. Einarsrud, Chem. Mater. 16 (2004) 5489. [46] W. Chen, F. Li, J. Yu, Mater. Lett. 60 (2006) 57. [47] M. Hirano, E. Kato, J. Am. Ceram. Soc. 82 (1999) 786. [48] K. Kaneko, K. Inoue, B. Freitag, A.B. Hungria, P.A. Midgley, T.W. Hansen, J. Zhang, S. Ohara, T. Adschiri, Nano Lett. 7 (2007) 421. [49] T. Yu, J. Joo, Y.I. Park, T. Hyeon, Angew. Chem. 117 (2005) 7577. [50] M. Kamruddin, P.K. Ajikumar, R. Nithya, G. Mangamma, A.K. Tyagi, B. Raj, Powder Technol. 161 (2006) 145. [51] X.-D. Zhou, W. Huebner, H.U. Anderson, Chem. Mater. 15 (2003) 378. [52] P. Yu, S.A. Hayes, T.J. O’Keefe, M.J. O’Keefe, J.O. Stoffer, J. Electrochem. Soc. 153 (2006) C74. [53] A.S. Deshpande, N. Pinna, B. Smarsly, M. Antonietti, M. Niederberger, Small 1 (2005) 313. [54] S. Tsunekawa, K. Ishikawa, Z.-Q. Li, Y. Kawazoe, A. Kasuya, Phys. Rev. Lett. 85 (2000) 3440. [55] F. Zhang, S.-W. Chan, J.E. Spanier, E. Apak, Q. Jin, R.D. Robinson, I.P. Herman, Appl. Phys. Lett. 80 (2002) 127. [56] P. Nachimuthu, W.-C. Shih, R.-S. Liu, L.-Y. Jang, J.-M. Chen, J. Solid State Chem. 149 (2000) 408. [57] S. Deshpande, S. Patil, S.V. Kuchibhatla, S. Seal, Appl. Phys. Lett. 87 (2005) 133113. [58] M. Ricken, J. No¨lting, I. Riess, J. Solid State Chem. 54 (1984) 89. [59] D.J.M. Bevan, J. Kordis, J. Inorg. Nucl. Chem. 26 (1964) 1509. [60] E. Ku¨mmerle, G. Heger, J. Solid State Chem. 147 (1999) 485. [61] M.J. Verkerk, B.J. Middelhuis, A.J. Burggraaf, Solid State Ionics 6 (1982) 159. [62] A.I. Ioffe, M.V. Inozemtsev, A.S. Lipilin, M.V. Perfil’ev, S.V. Karpachov, Phys. Stat. Sol. A 30 (1975) 87. [63] I. Kosacki, T. Suzuki, V. Petrovsky, H.U. Anderson, Solid State Ionics 136–137 (2000) 1225. [64] Y.W. Zhang, S. Jin, Y. Yang, G.B. Li, J. Tian, J.T. Jia, C.S. Liao, C.H. Yan, Appl. Phys. Lett. 77 (2000) 3409. [65] I. Kosacki, C.M. Rouleau, P.F. Becher, J. Bentley, D.H. Lowndes, Solid State Ionics 176 (2005) 1319. [66] M. Aoki, Y.-M. Chiang, I. Kosacki, L.J.-R. Lee, H. Tuller, Y. Liu, J. Am. Ceram. Soc. 79 (1996) 1169. [67] P. Mondal, A. Klein, W. Jaegermann, H. Hahn, Solid State Ionics 118 (1999) 331. [68] Q. Li, T. Xia, X.D. Liu, X.F. Ma, J. Meng, X.Q. Cao, Mater. Sci. Eng. B 138 (2007) 78. [69] U. Anselmi-Tambrini, J.E. Garay, A.Z. Munir, Scr. Mater. 54 (2006) 823. [70] Y.-M. Chiang, E.B. Lavik, I. Kosacki, H.L. Tuller, J.Y. Ying, Appl. Phys. Lett. 69 (1996) 185. [71] Y.-M. Chiang, E.B. Lavik, I. Kosacki, H.L. Tuller, J.Y. Ying, J. Electroceram. 1 (1997) 7. [72] I. Kosacki, T. Suzuki, V. Petrovsky, H.U. Anderson, Solid State Ionics 136–137 (2000) 1225. [73] S. Kim, J. Maier, J. Electrochem. Soc. 149 (2002) J73. [74] S. Kim, J. Fleig, J. Maier, Phys. Chem. Chem. Phys. 5 (2003) 2268. [75] S. Kim, J. Maier, J. Eur. Ceram. Soc. 24 (2004) 1919. [76] T. Suzuki, I. Kosacki, H.U. Anderson, Solid State Ionics 151 (2002) 111. [77] Y. Lei, Y. Ito, N.D. Browning, J. Am. Ceram. Soc. 85 (2002) 2359. [78] M.G. Bellino, D.G. Lamas, N.E. Walso¨e de Reca, Adv. Funct. Mater. 16 (2006) 107. [79] H. Huang, T.M. Gu¨r, Y. Saito, F. Prinz, Appl. Phys. Lett. 89 (2006) 143107. [80] J.L.M. Rupp, J. Gauckler, Solid State Ionics 177 (2006) 2513. [81] A. Tscho¨pe, Solid State Ionics 139 (2001) 267. [82] D.F. Evans, H. Wennerstro¨m, The Collidal Domain, VCH Publishers, New York, 1994, p. 110. [83] S. Azad, O.A. Marina, C.M. Wang, L. Saraf, V. Shutthanandan, D.E. McCready, A. El-Azab, J.E. Jaffe, M.H. Engelhard, C.H.F. Peden, S. Thevuthasan, Appl. Phys. Lett. 86 (2005) 131906. [84] E. Geuzens, G. Vanhoyland, J. D’Haen, S. Mullens, J. Luyten, M.K. Van Bael, H. Van den Rul, J. Mullens, J. Eur. Ceram. Soc. 26 (2006) 3133.