Inorganic Perovskite Oxides

Part of the Springer Handbooks book series (SPRINGERHAND)


Crystal structure and important functions of inorganic perovskite oxides are introduced. Perovskite oxides comprise large families among the structures of oxide compounds, and several perovskite-related structures are currently recognized. Typical structures (ABO3) consist of large-sized 12-coordinated cations at the A-site and small-sized 6-coordinated cations at the B-site. Several complex halides and sulfides and many complex oxides have a perovskite structure. From a variety of compositions and structures, a variety of functions are observed in perovskite oxides. In particular high electronic conductivity, which is at a similar level as metal, and surface activity to oxygen dissociation, are highly attractive in this oxide. Perovskite oxide is now widely used for solid oxide fuel cells.

Oxide groups consisting of two or more different cations are called complex or mixed oxides, and many types of crystal structures are known. In some special cases, oxides consisting of single cations in different oxidation states are also classified as mixed oxides. For example, in Eu3O4, the mixed oxide consists of Eu(III) and Eu(II) in 6- or 8-coordination respectively. However, the most typical structure of a mixed oxide consists simply of two or more different cations with different oxidation states, ionic radii, and coordination numbers. This diversity, which comes from the complexity of these structures, results in a larger number of different properties as compared to those of simple oxides. One of the most well known and important complex oxide structures is the spinel structure (AB2O4), which shows important magnetic properties. The structure of such oxides displays a most interesting complexity. Since the size of the A and B ions in this structure is close, oxides of this type are typical examples of the versatility of mixed oxides.

Another important structure of mixed oxide is perovskite and a variety of related structures are classified as this oxide. The typical chemical formula of the perovskite structure is ABO3, where A and B denotes two different cations. The ilmenite structure has the same composition as the perovskite one, i. e., ABO3; however, A and B in this structure are cations of approximately the same size, which occupy an octahedral site. Therefore, in spite of the fact that they share the same general chemical formula, structures classified as ilumenite- or ilmenite-related structures (e. g., LiSbO3) are different from perovskite.

Perovskite oxides comprise large families among the structures of oxide compounds, and several perovskite-related structures are currently recognized. Typical structures consist of large-sized 12-coordinated cations at the A-site and small-sized 6-coordinated cations at the B-site. Several complex halides and sulfides and many complex oxides have a perovskite structure. In particular, (Mg,Fe)SiO3 or CaSiO3 is thought to be the predominant compound in the geosphere [59.1, 59.2]. Perovskite compounds with different combinations of charged cations in the A and B-sites, for example 1 + 5, 2 + 4 and 3 + 3, have been discovered. Even more complex combinations are observed, such as \(\mathrm{Pb(B^{\prime}_{1/2}B^{\prime\prime}_{1/2})O_{3}}\), where \(\mathrm{B^{\prime}}=\text{Sc}\), Fe and \(\mathrm{B^{\prime\prime}}=\text{Nb}\), Ta, or \(\mathrm{La(B^{\prime}_{1/2}B^{\prime\prime}_{1/2})O_{3}}\), where \(\mathrm{B^{\prime}}=\text{Ni}\), Mg, etc. and \(\mathrm{B^{\prime\prime}}=\text{Ru(IV)}\) or Ir(IV). In addition, many ABO3 compounds crystallize in polymorphic structures, which show only a small distortion from the most symmetrical form of the perovskite structure.

The ideal structure of perovskite, which is illustrated in Fig. 59.1, is a cubic lattice. Although few compounds have this ideal cubic structure, many oxides have slightly distorted variants with lower symmetry (e. g., hexagonal or orthorhombic). Furthermore, even though some compounds have ideal cubic structure, many oxides display slightly distorted variants with lower symmetry. Several examples of perovskite oxides are listed in Table 59.1, where it is clear that a large number of perovskite oxides have a rombohedral lattice. Additionally, in many compounds a large extent of oxygen or cation deficiency has been observed. Due to the large lattice energy, many compounds are classified as perovskite oxides in spite of the large cation and/or oxygen deficiencies. There are various types of distortions in the perovskite structure that are strongly related to their properties, in particular their ferromagnetic or ferroelectricity.
Fig. 59.1

Ideal cubic perovskite structure

Table 59.1

Typical peroskite compound


Lattice parameter (Å)




Cubic structure

























Tetragonal structure





















LaAlO3 type

















GdFeO3 type





















In order to understand the deviations from the ideal cubic structure, these ABO3 oxides are first regarded as purely ionic crystals. In the case of the ideal structure, the following relationship between the radii of the A, B, and O2− ions holds true
Therefore, the deviation from the ideal structure in perovskite oxides can be expressed through the following so-called tolerance factor t
In perovskite-type compounds, the value of t lies between approximately 0.80 and 1.10. It is noted that the oxides with the lower t values crystallize in the ilmenite structure, which is a polymorph of the perovskite structure. It seems superfluous to say that for the ideal cubic structure, the value of t is close to 1 or at least higher than 0.89. Figure 59.2 shows the crystal groups for \(\mathrm{A^{2+}B^{4+}O_{3}}\) and \(\mathrm{A^{3+}B^{3+}O_{3}}\) combinations, which are related to deviation from the ideal structure [59.3]. As the value of t decreases, the structure of the unit lattice is shifted from cubic to triclinic due to the increased distortions. Figure 59.3 shows chemical elements that can be accommodated within the perovskite structure. It is evident that almost all elements except for noble gases can occupy either A or B lattice positions in the perovskite structure, including dopants. The stability and the crystal group is mainly determined by the ratio of the ionic radii of the A and B cations. Indeed, the structure is dependent not only on the size but also on the nature of the A and B atoms. For example, AMnO3 compounds crystallize in the perovskite structure when A cation is La or Ce–Dy, whereas a new hexagonal structure with 5- and 7-coordination of Mn and A respectively, is formed when \(\mathrm{A}=\mathrm{Ho}{-}\mathrm{Lu}\) or Y if A = La or Ce–Dy [59.4]. Here, the attention should be paid to the nature of the B atom, where the nature of the bond is highly covalent, and therefore the coordination number is lower than 6. The typical example of this type structure is BaGeO3. In spite of the t value close to t = 1, i. e., ideal ionic size combination, BaGeO3 crystallizes not in the perovskite structure but in the silicate-related one. This is attributable to the fact that the preferred coordination number of Ge is 4. On the other hand, due to the progress in high pressure technology, the synthesis of new Ge-based perovskite oxides has been reported [59.5]. Since the coordination number of Ge increases with the pressure, the perovskite structures with higher coordination numbers are preferred, and a typical example of this is CaGeO3. Another group of interesting perovskite compounds is the oxynitrides, i. e., LaWO3−xN x , LaTaO2N, etc. Therefore, the value of t, which is determined by the ionic size, is an important index for the stability of perovskite structures; however, the contribution of the chemical nature, such as coordinating number of the constituent elements, needs to be taken into account.
Fig. 59.2a,b

The effect of ionic size of A- and B-site cations on the observed distortions of the perovskite structure; (a\(\mathrm{A^{2+}B^{4+}O_{3}}\) case (b\(\mathrm{A^{3+}B^{3+}O_{3}}\) case

Fig. 59.3

Chemical elements that can occupy cation sites in the perovskite structure

The formation of superstructures in the perovskites is discussed next. If a B-site cation is progressively replaced by a dopant, a large difference in ionic radii tends to lead to the formation of the superstructures rather than random arrangements of the two kinds of ions. The typical case of this is Ba2CaWO6, which is regarded as Ba2 ( CaW ) O6. Similarly, in the compounds with the general formula Ba3MTa2O9, there is random distribution of M and Ta ions in the octahedral positions when M is Fe, Co, Ni, Zn, or Ca, whereas the formation of the superstructure with a hexagonal lattice is observed in Ba3SrTa2O9. Another interesting type of superstructure observed in the perovskite is the ordering of cation vacancies located on A-sites: for example MNb3O9 (M = La, Ce, Pr, Nb) and MTa3O9 (M = La, Ce, Pr, Nd, Sm, Gd, Dy, Ho, Y, Er). In these oxides, there is an octahedral framework of the ReO3 type with incomplete occupancy of the 12-fold coordinated A-sites. Figure 59.4 shows the structure of LaNb3O9. The B-sites of the perovskite structure are occupied by Nb ion and two-thirds of the A-sites remain vacant.
Fig. 59.4

Structure of LaNb3O9, an A-site-deficient perovskite oxide

Other typical polymorphs of the perovskite structure are Brownmillalite (A2B2O5) and K2NiF4 structures. Brownmillalite (A2B2O5) is an oxygen-deficient type of perovskite where oxygen vacancy is ordered. The unit cell contains BO6 and BO4 units in an ordered arrangement. Due to the oxygen deficiency, the coordination number of A-site cations decreases to 8. The lattice parameter of the Brownmillalite structure relates to cubic lattice parameter (ap) of the ideal perovskite as \(a=b=\sqrt{2a_{\mathrm{p}}}\), c = 4ap. Cu-based oxides or Ni-based oxides tend to adopt these oxygen-deficient structures because of the large amount of oxygen defects.

A combination of ordered B-sites and oxygen defects is seen in K2NiF4 structures, which are well-known as they show superconducting properties. The K2NiF4 structures consist of two units; a KNiF3 perovskite unit and a KF rock salt unit (Fig. 59.5 ), which are connected in series along the c axis. Since the rock salt structure is embedded into the c axis direction, the K2NiF4 compound shows strong two-dimensional properties. Based on the intergrowth of the different number of KNiF3 and KF units, there are many structures called Ruddelsden–Popper compounds with the general formula ( ABO3) n AO (Fig. 59.6), i. e., Sr3Ti2O7 (n = 2), Sr4Ti3O10 (n = 3). It is interesting to compare the isostructural Sr2TiO4 or Ca2MnO4 with SrTiO3 or CaMnO3, which crystallize in the perovskite structures. Two different A cations forming the perovskite and the rock salt units are also possible, and LaO ⋅ nSrFeO3 is the typical example of this arrangement. Another interesting variant of these K2NiF4 structures is when two different anions occupy the two building blocks exclusively, i. e., SrFeO3 ⋅ SrF or KNbO3 ⋅ KF. In any case, it is evident that perovskite oxides comprise a large family of oxides. As a result, a variety of crystal structures and properties is expected in these compounds. For further detailed discussion on the perovskite-related oxides, the reader is referred to references [59.5, 59.6, 59.7, 59.8].
Fig. 59.5

K2NiF4 structure, a perovskite-related structure

Fig. 59.6

Ruddelsden–Popper structure , another type of perovskite-related structure

59.1 Typical Properties of Perovskite Oxides

Due to the variety of structures and chemical compositions, perovskite oxides exhibit a large variety of properties. The well-known properties of the perovskite oxides are ferroelectricity in BaTiO3-based oxides and superconductivity in Ba2YCu3O7 etc. In addition to these well-known properties, several perovskite oxides exhibit good electrical conductivity, which is close to those of metals, ionic conductivity as well as mixed ionic and electronic conductivity. Based on these variations in the electrical conducting property, perovskite oxides are chosen as the components for solid oxide fuel cells (SOFC s). It is also well known that several perovskite oxides exhibit high catalytic activity with respect to various reactions, in particular oxidation reactions [59.9]. Table 59.2 provides examples of the typical properties of perovskite oxides. In this section, several typical properties of the perovskite oxides, namely ferroelectricity, magnetism, superconductivity, and catalytic activity, are briefly overviewed.
Table 59.2

Typical properties of perovskite oxides

Typical property

Typical compound


BaTiO3, PdTiO3


Pb ( Zr , Ti ) O3, ( Bi , Na ) TiO3

Electrical conductivity

ReO3, SrFeO3, LaCoO3, LaNiO3, LaCrO3


La0.9Sr0.1CuO3, YBa2Cu3O7, HgBa2Ca2Cu2O8

Ion conductivity

La ( Ca ) AlO3, CaTiO3, La ( Sr ) Ga ( Mg ) O3, BaZrO3, SrZrO3, BaCeO3


LaMnO3, LaFeO3, La2NiMnO6

Catalytic properties

LaCoO3, LaMnO3, BaCuO3

Electrode materials

La0.6Sr0.4CoO3, La0.8Ca0.2MnO3

59.1.1 Dielectric Properties

Ferroelectricity , piezoelectricity, electrostriction, and pyroelectricity are special properties inherent to dielectric materials, and are important properties of electroceramics. The most well-known property of perovskite oxides is ferroelectric behavior, where BaTiO3, PdZrO3, and their doped compounds are representative examples. The study of the ferroelectricity in BaTiO3 has long history, and many detailed reviews have been published. Furthermore, since the ferroelectric behavior of BaTiO3 has a strong relationship with the crystal structure, detailed studies of the crystal structure have been reported for BaTiO3. BaTiO3 undergoes mainly three phase transformations from, i. e., monoclinic to tetragonal and cubic as the temperature increases. Above 303 K, BaTiO3 crystallizes in the cubic perovskite structure, which does not show ferroelectric behavior. The high dielectric constant observed in BaTiO3 can be explained by the basis of the anisotropy of the crystal structure.

59.1.2 Electrical Conductivity and Superconductivity

One of the most well-known properties of perovskite oxides is superconductivity . In 1984, superconductivity was first reported by Bednorz and Müller in La–Ba–Cu–O perovskite oxide [59.10]. After their report, much attention has been paid to new types of high-temperature oxide superconductors, mainly Cu-based oxides. As a result, several superconducting oxides with different A-site cations have been discovered. However, the presence of Cu on the B-site is found to be essential for superconductivity to occur. High-temperature oxide superconductors of the YBa2Cu3O7 system [59.11] and the Bi2Sr2Ca2Cu3O10 system [59.12] were reported in 1987 and 1988 respectively, and currently the critical temperature of the superconducting transition (Tc) has been further increased to 130–155 K in the HgBa2Ca2Cu3O8+δ system [59.13]. Since all high-temperature superconducting oxides are cuprites (Cu-based oxides), superconductivity is clearly related to the Cu–O layers. The critical temperature for superconductivity, Tc, is related to the number of Cu–O layers in the crystal structure:
  1. 1.

    Cu–O layer  Tc ≈ 30 K

  2. 2.

    Cu–O layers Tc ≈ 90 K

  3. 3.

    Cu–O layers Tc ≈ 110 K

  4. 4.

    Cu–O layers Tc ≈ 120 K.

It is expected that further increase in the number of Cu–O layers may result in the higher Tc values. However, due to the low chemical stability, synthesis of five or more Cu–O layered compounds has not been successful so far. YBa2Cu3O7 is one of the most important superconductor systems with high Tc, and detailed studies of its crystal structure have been performed. Also, the content of oxygen nonstoichiometry is an important factor for high Tc. When the value of d is smaller than 0.5, YBa2Cu3O7 crystallizes in an orthorhombic structure, which is superconductive, while for d > 0.5, YBa2Cu3O7 has a tetragonal structure, which does not exhibit superconductivity.

In addition to superconductivity, there are many perovskite oxides showing high electronic conductivity, which is close to those of metals like Cu. The typical examples of such perovskite oxides are LaCoO3, LaFeO3, and LaMnO3, which are now commonly used as cathodes in SOFCs. These perovskite oxides show superior hole conductivity, which is as high as \(\sigma={\mathrm{100}}\,{\mathrm{S/cm}}\) originated from excess oxygen [59.14, 59.15]. Doping of aliovalent cations on the A-site is also highly effective in enhancing the electrical conductivity due to the increased number of mobile charge carriers generated by the charge compensation.

59.1.3 Catalytic Activity

Because of the variety of component elements and the high chemical stability, perovskite oxides have also been extensively studied as catalysts for various reactions. Two types of research trends clearly emerged from the above reasons. The objective of the first one is the development of the oxidation catalysts or oxygen-activated catalysts as an alternative to catalysts containing precious metals. The second trend regards perovskite as a model for active sites. The stability of the perovskite structure allows preparation of compounds with unusual valence states of elements or a high extent of oxygen deficiency. It is also noted that the high catalytic activity of perovskite oxides is based partially on the high surface activity to oxygen reduction or oxygen activation due to the large number of oxygen vacancies presented.

Among the various catalytic reactions studied, the ones applicable to environmental catalysis (e. g., automobile exhaust gas cleaning catalysts) attract particular attention. Initially, it was reported that perovskite oxide consisting of Cu, Co, Mn or Fe exhibited superior activity to NO direct decomposition at higher temperature [59.16, 59.17, 59.18]. The direct NO decomposition reaction (\(2\text{NO}=\mathrm{N_{2}}+\mathrm{O_{2}}\)) is one of the dream reactions in the catalysis field. In this reaction, the ease in the removal of surface oxygen as a product of the reaction plays an important role, and due to the facility of oxygen deficiency present, perovskite oxides are active with respect to this reaction at high temperatures. It is pointed out that the doping is highly effective in enhancing NO decomposition activity. Under an oxygen enriched atmosphere (up to 5%), a relatively high NO decomposition activity was reported for Ba ( La ) Mn ( Mg ) O3 perovskite [59.19].

Recently, another interesting application of perovskite oxides as an automobile catalyst has been reported, namely the so-called intelligent catalyst  [59.20]. Up to now, three-way Pd–Rh–Pt catalysts have been widely used for the purpose of removal of NO, CO, and uncombusted hydrocarbons. In order to decrease the amount of precious metals, the catalyst consisting of fine particles with high surface-to-volume ratio is required. However, these fine particles are not stable under the operating conditions and easily sinter, resulting in the deactivation of the catalyst. In order to maintain a high dispersion state, the redox property of perovskite oxides has been proposed, i. e., under oxidation conditions, palladium is oxidized and exists as LaFe0.57Co0.38Pd0.05O3, and under reducing conditions, palladium is deposited as fine metallic particles with a radius of 1–3 nm. This cycling of the catalyst through oxidizing and reducing conditions results in the partial substitution of Pd into, and deposits from, the perovskite framework, thus maintaining a high dispersion state of Pd. This was found to be highly effective in improving the long-term stability of Pd during removal of NO x , CO, and hydrocarbons from the exhaust gas. The high dispersion state of Pd can be recovered by exposing the catalyst to oxidation and reduction environments. As a result this catalyst is called an intelligent catalyst. This unique property also originates from the high stability of the perovskite crystal structure and charge compensation is automatically done by redox couples of another cation in the lattice.

Another interesting application of perovskite oxide is as a photocatalyst for water splitting. Among the catalysts for water splitting into H2 and O2, it is reported that Ta- or Nb-based perovskite oxide shows high activity by using ultraviolet light. This will be introduced in detail in the next section.

59.2 Photocatalytic Activity

Photo-excited electrons and holes can be used for splitting water into H2 and O2 and this reaction is attracting much interest for converting solar energy to hydrogen. Various inorganic catalysts have been studied as photocatalysts for water splitting, in particular, Pt/TiO2 is a well-known inorganic semiconductor for photocatalysis. Among the various catalysts reported, in this section, photocatalysts based on perovskite structure are briefly introduced. The Ta-based oxide is generally active in the photocatalytic water splitting reaction [59.22]. In particular, the Ta-based perovskite oxide, ATaO3 (A = alkaline cation) shows high activity in water splitting [59.23]. The activity is strongly affected by the A cation and this is because crystal structure is related to the electronic configuration of the oxide. The bond angles of Ta–O–Ta are 143 (LiTaO3), 163 (NaTaO3), and 180 (KTaO3). As the bond angle is close to 180, migration of excited energy in the crystal occurs more easily and the band gap becomes smaller. Therefore, the order of the delocalization of excited energy is \(\mathrm{LiTaO_{3}}<\mathrm{NaTaO_{3}}<\mathrm{KTaO_{3}}\), while that of the band gap is reversed in the order as shown in Fig. 59.7. Table 59.3 shows photocatalytic activities for water splitting into H2 and O2 in pure water on alkali tantalite photocatalysts with and without NiO cocatalysts. NaTaO3 photocatalysts showed the highest photocatalytic activity when NiO cocatalysts were loaded. In this case, excess sodium in the starting material was indispensable for showing the high activity [59.23]. The conduction band level of the NaTaO3 photocatalyst was higher than that of NiO (−0.96 eV) as shown in Fig. 59.7 [59.23]. Moreover, the excited energy was delocalized in the NaTaO3 crystal. Therefore, the photogenerated electrons in the conduction band of the NaTaO3 photocatalyst were able to transfer to the conduction band of the NiO cocatalyst of an active site for H2 evolution, resulting in the enhancement of the charge separation. Therefore, NiO loading was effective for the NaTaO3 photocatalyst even without special pretreatment.
Table 59.3

Photocatalytic activities for water splitting into H2 and O2 in pure water on alkali tantalite photocatalysts with and without NiO cocatalysts. (After [59.21])


Ratio of alkali to Tab

Band gapc (eV)

Surface area (m2 g−1)

Activity (μmol h−1)









NiO(0.10 wt%)/LiTaO3











NiO(0.05 wt%)/NaTaO3











NiO(0.05 wt%)/NaTaO3











NiO(0.10 wt%)/KTaO3





a Catalyst: 1 g, pure water: 350 ml, 400 W high-pressure mercury lamp, inner irradiation cell made of quartzb In starting materialsc Estimated from the onset of absorption

Fig. 59.7

Band structures of alkali tantalates ATaO3 (A : Li, Na and K) with perovskite-type structures in comparison to a normal hydrogen electrode (NHE ). (After [59.21])

It is reported that the activity of photocatalytic water splitting is also much increased by additives. For example, Table 59.4 shows band gaps, surface areas and photocatalytic activities for water splitting into H2 and O2 in pure water on various lanthanide-doped NaTaO3 (denoted as NaTaO3:Ln hereafter) with NiO cocatalysts [59.21]. NaTaO3:Ln powders had larger surface areas than that of nondoped NaTaO3. Band gaps of NaTaO3:Ln are slightly higher than that of nondoped NaTaO3. The activity of the NiO/NaTaO3 photocatalyst is remarkably improved by doping of Ln, except for Eu and Yb. In particular, obviously, NiO/NaTaO3:La is the most active: H2 and O2 evolved steadily and efficiently as shown in Fig. 59.8. The optimized NiO(0.2 wt%)/NaTaO3:La(1.5%) photocatalyst evolves H2 and O2 with the rates of 14.6 and 7.2 m mol h−1 respectively. The apparent quantum yield is approximately 50% at 270 nm. Thus, it is demonstrated that the photocatalytic water splitting is able to proceed efficiently using a photocatalyst powder system.
Table 59.4

Band gap, surface areas and photocatalytic activities for water splitting into H2 and O2 in pure water on various lanthanidedoped NaTaO3 with NiO cocatalysts. (After [59.21])


Band gap (eV)

Surface area (m2 g−1)

Activity (m mol h−1)a






2.18 b

1.10 b














































a Catalyst: 1 g, pure water: 390 ml, inner irradiation cell made of quartz, 400 W high-pressure mercury lampb Initial activity

Fig. 59.8

Photocatalytic water splitting over NiO(0.05 wt%)/NaTaO3:La(1 mol%). The cell was evacuated (evac.) after each experimental run. Catalyst: 1 g, pure water: 390 ml, 400 W high-pressure mercury lamp, inner irradiation cell made of quartz. Open marks: H2, closed marks: O2. (After [59.23])

On the other hand, effects of substitution of Ta-sites of KTaO3 are also reported [59.24]. Not only dopants at A-sites, but also those at B-sites are strongly influenced by photocatalytic activity. Since Pyrex glass was used for the reactor, the catalytic activity is different from those in Table 59.4 because of negligible ultraviolet light, and it is reported that KTaO3 is more active than that of NaTaO3, which corresponds with the most narrow band gap among ATaO3. The effects of substitution of various elements on mainly Ta-sites in KTaO3 on photocatalytic activity to water splitting are shown in Table 59.5. Obviously, H2 and O2 formation rates are significantly increased by doping Ta-sites, in particular, the highest H2 formation rate can be achieved by substituting Ta partially with Zr or HF. Since positive effects are tended to be obtained by doping tetravalent cations, it seems that a decrease in carrier density is effective for increasing photocatalytic activity for water splitting.
Table 59.5

Effect of substitution of various elements on mainly Ta-sites in KTaO3 on photocatalytic activity to water splitting. (After [59.24])


Formation rate (μmol ∕ h)


























































Pt/TiO2 (0.3 wt%)



a 1 wt% NiO loaded

In addition to Ta-based perovskite, another important oxide based on the perovskite structure is SrTiO3, which has narrower band gap than that of Ta-based perovskite oxide [59.25]. Since the valence band is not low enough for the formation of O2, complete decomposition of water can be achieved by combining with oxygen-formation catalysts such as BiVO4 with Pt/SrTiO3 doped with Rh [59.22]. Among the various catalysts reported, it is obvious that perovskite oxide is an important family for photocatalysts from a unique semiconducting property viewpoint, and it is expected that research is expanded to the oxygen-deficient-type perovskite oxides such as K2NiF4-type structures.

59.3 Application for Solid Oxide Fuel Cells (SOFCs)

An important application area of inorganic perovskite oxide is solid oxide fuel cells and air electrodes of metal-air batteries because of high catalytic activity to oxygen reduction and superior mixed conductivity achieved simultaneously. In this section, the application of perovskite oxide for SOFCs is briefly mentioned. Further details are available in the another book [59.26].

Table 59.6 summarizes the important applications of perovskite oxides for solid oxide fuel cell technology. As shown in Table 59.6, LaCoO3 or LaMnO3 are promising candidates for SOFC cathodes and LaGaO3-based oxides for the electrolyte. In addition, recently there have been several reports on the application of Cr-based perovskites as anodes. Therefore, the concept of SOFCs based entirely on perovskite components, all-perovskite SOFCs, is also proposed and some preliminary results have been reported [59.27]. In contrast to the SOFCs using oxide-ion-conducting electrolytes, the development of SOFCs using high-temperature proton-conducting electrolytes is slightly delayed, particularly when compared with development of polymer electrolyte-type fuel cells. However, the Toyota group has quite successfully demonstrated a high-power SOFC using a BaCeO3-based electrolyte film on Pd foil [59.28]. Their data suggest that the proton-conducting perovskite oxides might also be an important component in real SOFCs in the near future.
Table 59.6

Important materials for perovskite oxide for solid oxide fuel cell applications


Typical Materials


La ( Sr ) MnO3, La ( Sr ) CoO3, Sm0.5Sr0.5CoO3, La ( Sr ) Fe ( Co ) O3


La ( Sr ) Ga ( Mg ) O3 ( O2− ) , BaCeO3 ( H+ ) , BaZrO3 ( H+ ) , SrZrO3 ( H+ ) , Ba2In2O5 ( O2− ) 


\(\mathrm{La_{1-\mathit{x}}Sr_{\mathit{x}}Cr_{1-\mathit{y}}M_{\mathit{y}}O_{3}}\) (M = Mn, Fe, Co, Ni), SrTiO3


La ( Ca ) CrO3

59.3.1 Cathode

The principal requirement for a cathode of a SOFC is to electrochemically reduce oxygen molecules into oxide ions, and there are several requirements for oxide applied for cathodes of SOFCs i. e., catalytic activity, thermodynamic stability and compatibility in mechanical and chemical properties. Perovskite oxide is the most suitable material satisfying these requirements for cathodes. Figure 59.9 shows the reaction route considered for cathodes of SOFCs. Oxygen reduction proceeds on the electrode surface or at the electrode/electrolyte/gas-phase interface; the so-called triple phase boundary (TPB ). The electrode material catalyzes the oxygen molecules to be dissociated into atoms, charged and incorporated into the electrolyte (Fig. 59.9). For the cathode material, the electrocatalytic activity is an important parameter to be considered. The surface reaction rate constant in oxygen isotope exchange is a good measure for the catalytic activity. Kilner et al. [59.29] compared various oxides in isotope diffusion coefficient and found a positive correlation between those parameters. A highly mixed electronic and ionic conductor may be a promising candidate in terms of the electrode performance. At the early stage of SOFC development, La0.8Ca0.2MnO3 or La0.6Sr0.4MnO3 were widely used, however, because of low oxide ion conductivity in Mn-based perovskite oxide, the reaction is limited to a three-phase boundary resulting in large cathodic overpotential. Recently, several oxides have been reported to show extremely high surface exchange rate for oxygen activation. Baumann et al. [59.30] compared several Co- and Fe-based perovskites in a controlled shape and found Ba0.5Sr0.5Co0.8Fe0.2O3 shows 100 times smaller electrochemical resistance than that of La0.6Sr0.4Co0.2Fe0.8O3 (LSCF) which is often used for the intermediate-temperature SOFC cathode. One reason for such high cathodic activity is assigned to the high mixed conductivity and large reaction area available, i. e., the two phase boundary (route 2 in Fig. 59.9) is also contributed to the oxygen dissociation reaction. Another active composition to the cathode is Sm0.5Sr0.5CoO3, which also has high oxide ion conductivity [59.31]. However, the most popular composition used for SOFC is La0.6Sr0.4Co0.2Fe0.8O3 (LSCF) due to surface activity and stability. Recently, research has shifted to the more oxygen-deficient perovskites such as double (A2BaO6), Rudrusden–Poppered (A x B y O z ), or K2NiF4 structure. Sase et al. [59.32] reported that existence of a ( La , Sr)2CoO4 phase on the ( La , Sr ) CoO3 electrode enhances the oxygen exchange reaction rate. The double perovskite phase of PrBaCo2O6 is also reported as an active phase for oxygen dissociation [59.33] and so not only perovskite but also perovskite-related phases, in particular the oxygen-deficient perovskite phase, is now attracting as much attention as the active cathode catalyst for intermediate temperature operations.
Fig. 59.9

Reaction route considered for a cathode of SOFC

On the other hand, one of another important issues for SOFC development is long-term stability and decrease in cathode performance is pointed out. There are several reasons decreasing cathodic performance is reported, i. e., chemical poisoning with Cr, S and B and sintering, and phase separation. Among them, surface segregation with Sr on the perovskite cathode has been suggested recently [59.34]. Figure 59.10 shows the change in surface composition of LSCF by low-energy ion scattering techniques, which are sensitive to the elements in the outermost surface layer. Obviously, the surface of LSCF is immediately enriched with Sr and it is considered that an Sr-enriched surface is highly reactive with Cr or S, resulting in the decrease in surface activity. Apparently, the surface composition of perovskite is slightly different from that of bulk, however, reactivity for oxygen dissociation is more strongly affected by B-site cations. Therefore, there is still some unclear points on parameters determining the activity for oxygen dissociation and this will be discussed more intensively in the future.
Fig. 59.10

Change in surface composition of La0.6Sr0.4Fe0.8Co0.2O3 (LSCF) analyzed with low-energy ion scattering spectroscopy

59.3.2 Anode

For the anode of SOFC, metal-oxide ion-conducting oxide composites named cermet have been widely used, in particular, Ni-Y2O3-stabilized ZrO2 (Y0.16Zr0.84O2, YSZ) cermet has been widely used. However, Ni is well known to be deactivated easily by aggregation and coarsening. In addition, re-oxidation of Ni is also occurs easily, resulting in the permanent failure of the cell. Therefore, recently, the application of oxide for the anode has also been proposed and among the proposed oxide anodes, perovskite oxides such as SrTiO3 doped with La, Nd, etc. [59.35] or La0.75Sr0.25Mn0.5Cr0.5O3 (LSCrM) [59.36] show interesting performance and are promising as oxide anodes.

In particular, improved performance has been obtained with complex perovskites based upon Cr and Mn at the B-sites forming compositions (La,Sr)\(\mathrm{Cr_{1-{\mathit{x}}}M_{\mathit{x}}O_{3-{\mathit{\delta}}}}\). Tao and Irvine have focused upon doped lanthanum chromite doped with Sr and Mn up to 20% dopant on the B-site, usually 5 or 10%. ( La0.75Sr0.25 ) Cr0.5Mn0.5O3 (LSCrM) exhibits comparable electrochemical performance to Ni/YSZ cermets [59.27, 59.36]. Figure 59.11 shows the I–V, I–P curves of the cell using LSCrM for the anode at 1173 and 1123 K. Apparently, the open circuit potential (OCV ) is achieved to a theoretical level and reasonable power density of 0.4 W ∕ cm2 was achieved at 1173 K. The electrode polarization resistance approaches 0.2 Ω cm2 at 1173 K in 97%H2/3%H2O. Good performance is achieved for methane oxidation without using excess steam. The anode is stable in both fuel and air conditions and shows stable electrode performance in methane. Thus both redox stability and operation in low steam hydrocarbons have been demonstrated. Catalytic studies of LSCM demonstrate that it is primarily a direct-oxidation catalyst for methane oxidation as opposed to a reforming catalyst, with the redox chemistry involving the Mn–O–Mn bonds. Although oxygen ion mobility is low in the oxidized state, the diffusion coefficient for oxide ions in reduced LSCrM is comparable to YSZ.
Fig. 59.11

Power generation property of the cell using La0.75Sr0.25Cr0.5Mn0.5O3 for the anode

Another important double perovskite is Sr2MgMoO6−δ, which has recently been shown to offer good performance, with power densities of 0.84 W ∕ cm2 in H2 and 0.44 W ∕ cm2 in CH4 at 1073 K, and good sulfur tolerance [59.37]. The molybdenum-containing double perovskite was initially prepared at 1473 K in flowing 5% H2 and then deposited on top of a lanthanum ceria buffer layer before testing. Although there is some decomposition recognized, it is also reported that La0.5Sr0.5MnO3 is active to not only the cathode but also the anode. The maximum power density of 1 W ∕ cm2 was reported at 1273 K when La0.9Sr0.1Ga0.8Mg0.2O3 (LSGM, 0.5 mm thickness) is used for the electrolyte [59.38].

59.3.3 Electrolyte

An oxide ion conductor is widely used for electrolytes of SOFCs. In the case of SOFCs, solid proton conducting ceramics such as BaZrO3 or SrCeO3 doped with various rare earth cations to B-sites have been also studied [59.39], in particular for low-temperature SOFCs; however, considering the chemical stability and power density achieved, SOFCs using oxide ion conducting electrolytes are more promising. At present, almost all SOFCs under development for commercialization use YSZ for the electrolyte, with only a few exceptions. However, for increasing power density, there is still strong demand for replacing YSZ with an alternative electrolyte showing higher oxide ion conductivity. CeO2 with fluorite structure and LaGaO3 with perovskite structure are showing promise as electrolytes for this purpose. In particular, LaGaO3 doped with Sr and Mg is the first pure oxide ion conductor with a perovskite structure [59.40]. Figure 59.12 shows a comparison of the oxide ion conductivity of LSGM with typical fluorite oxide. Apparently, oxide ion conductivity in LSGM is much higher than that of YSZ and almost comparable with Gd-doped CeO2. In the case of CeO2, partial electron conductivity appears in reducing atmospheres, however, LSGM exhibits the pure oxide ion conductivity across a wide \(p_{\mathrm{O_{2}}}\) range, i. e., from pure oxygen to pure hydrogen.
Fig. 59.12

Comparison of the oxide ion conductivity of LSGM with typical fluorite oxide

The application of LSGM for electrolytes of SOFCs is also studied and the power generation property of the cell using LSGM is shown in Fig. 59.13. Since transport number of the oxide ion in LSGM is close to unity from pure O2 to pure H2, the open circuit potential is 1.15 V, which is almost the same as that of theoretical open circuit potential. In addition, in spite of the thick electrolyte (0.5 mm), high maximum power density is achieved compared with that of the cell using YSZ. Therefore, it is expected that the operating temperature of SOFCs could be decreased by using LSGM for the electrolytes. Making thin films of LSGM is another important subject, however, it is reported that LSGM has some reactivity with Ni in the anode substrate and so the power density is not as high as expected from the film thickness and ionic conductivity. Therefore, for LSGM film cells, preventing reaction with Ni in the anode substrate is an important issue. Huang et al. reported that La-doped CeO2 is effective for preventing Ni diffusion [59.41], however, electrical conductivity of La-doped CeO2 is insufficient. Recently, Hong et al. reported that Ti- and La-codoped CeO2 is promising as a buffer layer for preventing Ni diffusion with reasonable electrical conductivity and sintering properties [59.42]. In any case, LSGM perovskite oxide has a high potential as a fast oxide ion conductor for intermediate-temperature SOFCs and as an alternative to YSZ. Another interesting oxide ion conductor is Ba2In2O5, which is a Brownmillerite structure and although oxide ion conductivity is slightly lower than that of YSZ, this new family of oxide ion conductors is also considered as a promising electrolyte for SOFCs [59.43].
Fig. 59.13

Power generation property of the cell using LSGM with 0.182 mm thickness

Another important family for electrolytes of SOFC is oxide proton conductor and perovskite oxide of ACeO3 or BZrO3(A = Ba and Sr), which shows fast proton conductivity. Protons in perovskite oxide were formed by the following equation and the interstitial protons are mobile at the interstitial position of the perovskite lattice. Figure 59.14 shows the comparison of proton conductivity in perovskite oxide [59.44]. BaCeO3 doped with Y shows reasonably high proton conductivity at low temperature. It is reported that extremely large power density is achieved by using BaCeO3 doped with a Y thin film for the electrolyte of SOFC; however, these proton-conducting perovskites show high reactivity with CO2 resulting in decomposition by formation of carbonate. Therefore, an increase in chemical stability is required for these oxide perovskite. Recently, there has been high interest on ACe1−xZr x O3-based oxides doped with Y from chemical stability; in particular it is reported that BaCe0.4Zr0.5Y0.1O3 shows reasonably high proton conductivity and reasonable chemical stability to CO2 and hydrolysis with water [59.45].
Fig. 59.14

Comparison of proton conductivity in perovskite oxide. (After [59.44].) Plots for BaCe0.4Zr0.5Y0.1O3 were added from [59.45]

59.3.4 Interconnector

The interconnector is an important component for stacking SOFCs and requires high stability in reducing and oxidizing atmospheres. The interconnector is used for connecting the SOFC single cell as shown in Fig. 59.15 and so high electrical conductivity and no oxide ion conductivity are also required for the high performance of the SOFC stack. Perovskite oxides of LaCrO3 have been also widely used for interconnectors. Although LaCrO3 is stable under reducing and oxidizing atmospheres, the electrical conductivity is still insufficient and also sintering is rather difficult. Therefore, doping Ca2+ for La-site in LaCrO3 is generally performed for increasing conductivity as well as sintering properties.
Fig. 59.15

Schematic view of the SOFC stack

Figure 59.16 shows electronic conductivity of doped LaCrO3 as a function of temperature [59.46, 59.47]. The electronic conductivity increases with increasing temperature, suggesting the semiconductor temperature dependence. An increasing of the Ca concentration in \(\mathrm{La_{1-{\mathit{x}}}Ca_{\mathit{x}}CrO_{3-{\mathit{\delta}}}}\) enhanced the electronic conductivity due to the increase of Cr4+ concentration. There are some deviations of the electrical conductivity among the examined alkaline earth elements: Ca-doped LaCrO3 shows larger electrical conductivity than Sr-doped LaCrO3. This difference was reported to be due to the difference of lattice distortion and phase stability. The activation energy for conductivity was 0.12–0.14 eV and the mobility was 0.066–0.075 cm2 ∕  ( Vs )  at 1173–1323 K.
Fig. 59.16

Electrical conductivity of \(\mathrm{La_{1-\mathit{x}}Ca_{\mathit{x}}CrO_{3-{\mathit{\delta}}}}\) (x = 0.1–0.3) in air as a function of inverse temperature. (After [59.47])

The electronic conductivity decreases with a reduction of oxygen partial pressure because of the decrease of Cr4+ concentration in a reducing atmosphere. The electrical conductivity decreases with a reduction of oxygen partial pressures. The electrical conductivity is proportional to pO21∕4, which is consistent with the defect chemistry of \(\mathrm{La_{1-{\mathit{x}}}Ca_{\mathit{x}}CrO_{3-{\mathit{\delta}}}}\) [59.47]. A doping of the B-site has been also considered by several authors [59.48]. A typical dopant cation is Mg2+, replaced into Cr3+ sites. This substitution also increases the concentration of Cr4+, and eventually increases the electrical conduction. Because of low sintering property, SrTiO3 doped with La for Sr sites or Nd for Ti sites is also studied as an interconnector. Perovskite oxide is also important for interconnectors of SOFCs.

In summary, perovskite oxide is widely used for current SOFCs and becomes important compounds for SOFC components. Therefore, the all-perovskite concept is also proposed for SOFC and the device consisting of materials with the same structure is highly interesting.

59.4 Oxygen Separating Membrane

Perovskite oxide shows high electronic and oxide ionic conductivity and these conductors are called mixed conductors. Since charge compensation by ion transport can be automatically achieved with electronic conductivity in mixed conductors, so ions can be transported in mixed conductors without outside circuits. Therefore, an important application of such mixed conductors is as a separation membrane for oxygen from air. The oxygen permeation rate is shown as the following equation if the bulk diffusion is a rate-limiting step
Here, \(J_{\mathrm{O_{2}}}\); oxygen flux, σel; electron conductivity, σion; oxide ion conductivity, F; Faraday constant, R; gas constant, t; membrane thickness, T; temperature and ph, pl means high and low oxygen partial pressure respectively.
From (59.1), the oxygen permeation rate is limited by bulk diffusivity of oxygen. Figure 59.17 shows the diffusivity of oxide ions in several perovskite oxides considered for oxygen permeation membranes [59.49]. Apparently, Fe- or Co-based perovskite oxides show fast oxide ion conductivity and so large oxygen permeation rate is expected on these perovskite oxides. In fact, a large oxygen permeation rate of 1.05 ml ∕  ( cm2 min )  at 1173 K is reported for Ba0.5Sr0.5Fe0.8Co0.2O3 (BSCF) (2 mm thickness) [59.50]. According to (59.1), the oxygen permeation rate is determined by the diffusivity of oxide ions and the thickness of the membrane when the oxygen partial pressure differential across the membrane is the same. Therefore, the oxygen permeation rate should be increased with decreasing membrane thickness; however, it is reported that the oxygen permeation rate is dependent on membrane thickness when thickness is large, but with decreasing thickness, the oxygen permeation rate becomes independent of oxygen permeation rate because of the limitation by surface reactions of oxygen dissociation. To achieve the large oxygen permeation rate, not only high diffusivity of oxide ion in bulk but also high surface activity to oxygen dissociation is required.
Fig. 59.17

Diffusivity of oxide ion in several perovskite oxides considered for an oxygen permeation membrane. (After [59.49])

Recently, perovskite-related oxides are also attracting much interest as mixed conductors, and applied use for oxygen permeation membranes. Among the perovskite-related oxides, there is strong interest for the K2NiF4-type oxide because of the large amount of oxygen deficiency. As explained in Figure 59.5, K2NiF4 oxides consist of a series of connected perovskite and rock salt blocks, and a large free volume exists in the rock salt block. Therefore, interstitial oxygen can be easily introduced into the rock salt block. For the K2NiF4 oxides, the oxide ion permeation property in La2NiO4-based oxide is now studied in detail and it is reported that doping Cu and Ga is effective for increasing oxygen permeation rate and Pr2Ni0.71Cu0.24Ga0.05O3 (denoted as PNCG) is the optimized composition for oxygen permeation and permeation rate of approximately 3 cc ∕  ( min cm2 )  was reported on PNCG with 0.5 mm thickness from air to He [59.51]. The oxygen transport route is also studied by neutron diffraction analysis and as shown in Fig. 59.18, oxide ion transports through interstitial positions in the rock salt block is clearly demonstrated. Therefore, in this PNCG, oxide ions are mainly transported through rock salt but not in the perovskite block. On the other hand, PNCG shows high hole conduction and holes are mainly transported in the perovskite block; and so in PNCG, it is interesting that two different routes for oxide ion and hole conduction are expected [59.52].
Fig. 59.18

Oxygen transport route in Pr2Ni0.81Cu0.24Ga0.05O4 estimated by neutron diffraction analysis. (After [59.52])

The most interesting use of mixed conductors is to combine the catalytic reaction and the so-called membrane reactor system; and several catalytic reactions such as NO x decomposition and partial oxidation of alkanes are reported by using mixed conductors for removing the reactant from or into reaction systems. To permeate oxygen through mixed conductor membranes, a gradient in oxygen partial pressure is required as a driving force for oxygen transport; however, in membrane reactor system, differences in oxygen partial pressure are automatically achieved. Therefore, the combination of mixed conductors with catalytic reactions is the most ideal usage. From this aspect, partial oxidation of CH4 has been studied with perovskite mixed conductors for oxygen separation from air [59.53]. The schematic image of this catalytic reactor system was shown in Fig. 59.19. Since oxygen partial pressure in CH4 is as low as approximately 10−20 atm, a large oxygen permeation rate is achieved under CH4 partial oxidation conditions and it is reported that an oxygen permeation rate of 5 cc ∕  ( min cm2 )  was achieved by using SrFe0.8Co0.2O3 perovskite [59.54]. However, the oxygen permeation rate is decreased by phase changes in a reducing atmosphere of CH4. Therefore, although Fe- or Co-based perovskites have been mainly studied, the most important issues for application of perovskite oxides to catalytic membrane reactors are stability in a wide oxygen partial pressure range and it is reported that Ni- or Fe-doped LaGaO3 shows hole and oxide ion conductivity stably over wide \(p_{\mathrm{O_{2}}}\) ranges and the oxygen permeation rate of 12 cc ∕  ( min cm2 )  was exhibited on La0.7Sr0.3Ga0.6Fe0.4O3 of 0.2 mm thick at 1273 K in CH4 partial oxidation [59.55]. As a result, obviously, perovskites are important materials for oxygen separation membranes and the application of catalytic membrane reactors is an important area; albeit the chemical stability is required to be much improved.
Fig. 59.19

Schematic image of the catalytic membrane reactor using mixed conductors

59.5 Summary

In this chapter, crystal structures of perovskites and related oxides were explained. Perovskite oxide has a variety of composition and component elements. In addition, there are many isomorphs in crystal structure. Therefore, based on variety of crystal structures, there are many functions and rich application areas. In particular, in this chapter, the application of perovskite oxides for photocatalytic properties and solid oxide fuel cells were briefly overviewed. Since high electric conductivity and surface activity to oxygen dissociation are achieved simultaneously, perovskite oxides, mainly Co-, Fe- and Mn-based oxides are widely used for SOFCs and oxygen permeation membranes. On the other hand, for the application to photocatalysts, Ta- or Ti-based perovskites are highly active. Therefore, perovskite oxide is a highly important compound for these areas.


  1. 59.1
    R.M. Hazen: Sci. Amer. 258(6), 74 (1988)CrossRefGoogle Scholar
  2. 59.2
    T. Yagi, H.K. Mao, P.M. Bell: Phys. Chem. Minerals 3, 97 (1978)CrossRefGoogle Scholar
  3. 59.3
    F. Kanamura: Perovskite related compound. In: Kikan Kigaku Sasetsu, No.32, ed. by Japanese Society of Chemistry (Japanese Society of Chemistry, Tokyo 1997) p. 9Google Scholar
  4. 59.4
    S. Geller, J.B. Jeffries, P.J. Curlander: Acta Cryst. B 31, 2770 (1975)CrossRefGoogle Scholar
  5. 59.5
    R.C. Liebermann, L.E.A. Jones, A.E. Ringwood: Phys. Earth Planet. Inter. 14, 165 (1977)CrossRefGoogle Scholar
  6. 59.6
    A.F. Well: Structural Inorganic Chemistry, 5th edn. (Oxford Univ. Press, Oxford 1984)Google Scholar
  7. 59.7
    A.F. Cotton, G. Wilkinson: Advanced Inorganic Chemistry (Wiley, New York 1988)Google Scholar
  8. 59.8
    F.S. Galasso: Perovskites and High Tc Superconductors (Gordon Breach, New York 1990)Google Scholar
  9. 59.9
    R.H. Mitchell, T. Bay: Perovskites Modern and Ancient (Almaz, Thunder Bay 2002)Google Scholar
  10. 59.10
    H. Arai, T. Yamada, K. Eguchi, T. Seiyama: Appl. Catalysis 26, 265 (1986)CrossRefGoogle Scholar
  11. 59.11
    J.B. Bednorz, K.A. Müller: Z. Phys. B 64, 189 (1986)CrossRefGoogle Scholar
  12. 59.12
    P.H. Hor, R.L. Meng, Y.Q. Wang, L. Gao, Z.J. Huang, J. Bechtold, K. Forster, C.W. Chu: Phys. Rev. Lett. 58, 1891 (1987)CrossRefGoogle Scholar
  13. 59.13
    H. Maeda, Y. Tanaka, M. Fukutomi, T. Asano: Jpn. J. Appl. Phys. 27, L209 (1988)CrossRefGoogle Scholar
  14. 59.14
    L. Gao, Y.Y. Xue, F. Chen, Q. Xiong, R.L. Meng, D. Ramirez, C.W. Chu, J.H. Eggert, H.K. Mao: Phys. Rev. B 50, 4260 (1994)CrossRefGoogle Scholar
  15. 59.15
    J. Mizusaki, M. Yoshihiro, S. Yamauchi, K. Fueki: J. Solid State Chem. 58, 257 (1985)CrossRefGoogle Scholar
  16. 59.16
    J. Mizusaki, I. Yasuda, J. Shimoyama, S. Yamaguchi, K. Fueki: J. Electrochem. Soc. 140, 467 (1993)CrossRefGoogle Scholar
  17. 59.17
    S. Shin, H. Arakawa, Y. Hatakeyama, K. Ogawa, K. Shimomura: Mater. Res. Bull. 14, 633 (1979)CrossRefGoogle Scholar
  18. 59.18
    Y. Teraoka, T. Harada, S. Kagawa: J. Chem. Soc. Faraday Trans. 1998, 94 (1887)Google Scholar
  19. 59.19
    H. Yasuda, T. Nitadori, N. Mizuno, M. Misono: Bull. Chem. Soc. Jpn. 66, 3492 (1993)CrossRefGoogle Scholar
  20. 59.20
    H. Iwakuni, Y. Shinmyou, H. Yano, H. Matsumoto, T. Ishihara: Appl. Catal. B 299, 74 (2007)Google Scholar
  21. 59.21
    H. Kato, A. Kudo: Catal. Today 78, 561 (2003)CrossRefGoogle Scholar
  22. 59.22
    Y. Nishihata, J. Mizuki, T. Akao, H. Tanaka, M. Uenishi, M. Kimura, T. Okamoto, N. Hamada: Nature 418, 164 (2002)CrossRefGoogle Scholar
  23. 59.23
    K. Maeda: J. Photochem. Photobiol. C Photochem. Rev. 12, 237 (2011)CrossRefGoogle Scholar
  24. 59.24
    A. Kudo, H. Kato: Chem. Phys. Lett. 331, 373 (2000)CrossRefGoogle Scholar
  25. 59.25
    T. Ishihara, H. Nishiguchi, K. Fukamachi, Y. Takita: J. Phys. Chem. B 103, 1 (1999)CrossRefGoogle Scholar
  26. 59.26
    K. Domen, S. Naito, M. Soma, T. Onishi, K. Tamaru: J. Chem. Soc. Chem. Commun. 12, 543 (1980)CrossRefGoogle Scholar
  27. 59.27
    T. Ishihara (Ed.): Perovskite Oxide for Solid Oxide Fuel Cells (Springer, New York 2009)Google Scholar
  28. 59.28
    S. Tao, J.T.S. Irvine, J.A. Kilner: Adv. Mater. 17, 1734 (2005)CrossRefGoogle Scholar
  29. 59.29
    N. Ito, M. Iijima, K. Kimura, S. Iguchi: J. Power Source 152, 200 (2005)CrossRefGoogle Scholar
  30. 59.30
    J.A. Kilner, R.A. De Souza, I.C. Fullarton: Solid State Ion. 86–88, 703 (1996)CrossRefGoogle Scholar
  31. 59.31
    F.S. Baumann, J. Fleig, G. Cristiani, B. Stuhlhofer, H.U. Habermeier, J. Maier: J. Electrochem. Soc. 154, B931 (2007)CrossRefGoogle Scholar
  32. 59.32
    T. Ishihara, M. Honda, T. Shibayama, H. Nishiguchi, Y. Takita: J. Electrichem. Soc. 145(9), 3177 (1998)CrossRefGoogle Scholar
  33. 59.33
    M. Sase, K. Yashiro, K. Sato, J. Mizusaki, T. Kawada, N. Sakai, K. Yamaji, T. Horita, H. Yokokawa: Solid State Ion. 178, 1843 (2008)CrossRefGoogle Scholar
  34. 59.34
    M. Burriel, J. Pena-Martinez, R.J. Chater, S. Fearn, A.V. Berenov, S.J. Skinner, J.A. Kilner: Chem. Mat. 24, 613 (2012)CrossRefGoogle Scholar
  35. 59.35
    J. Druce, H. Tellez, M. Burriel, M. Sharp, L. Fawcett, S. Cook, D. McPhail, T. Ishihara, H.H. Brongersma, J.A. Kilner: Energy Environ. Sci. 7, 3593 (2014)CrossRefGoogle Scholar
  36. 59.36
    J.T.S. Irvine, P.R. Slater, P.A. Wright: Ionics 2, 213 (1996)CrossRefGoogle Scholar
  37. 59.37
    S.W. Tao, J.T.S. Irvine: Nat. Mater. 2, 320 (2003)CrossRefGoogle Scholar
  38. 59.38
    Y.H. Huang, R.I. Dass, Z.L. Xing, J.B. Goodenough: Science 312, 254 (2006)CrossRefGoogle Scholar
  39. 59.39
    T. Ishihara, S. Fukui, M. Enoki, H. Matsumoto: J. Electrochem. Soc. 153, A2085 (2006)CrossRefGoogle Scholar
  40. 59.40
    H. Iwahara, T. Esaka, H. Uchida, N. Maeda: Solid State Ion. 3–4, 359 (1981)CrossRefGoogle Scholar
  41. 59.41
    T. Ishihara, H. Matsuda, Y. Takita: J. Am. Chem. Soc. 116, 3801 (1994)CrossRefGoogle Scholar
  42. 59.42
    K. Huang, R. Tichy, J.B. Goodenough, C. Milliken: J. Am. Ceram. Soc. 81, 2581 (1998)CrossRefGoogle Scholar
  43. 59.43
    J.E. Hong, T. Inagaki, S. Ida, T. Ishihara: J. Am. Ceram. Soc. 95(11), 3588 (2012)CrossRefGoogle Scholar
  44. 59.44
    J.B. Goodenough, J.E. Ruiz-Diaz, Y.S. Zhen: Solid State Ion. 44, 21 (2000)CrossRefGoogle Scholar
  45. 59.45
    H. Iwahara: Solid State Ion. 86–88, 9 (1996)CrossRefGoogle Scholar
  46. 59.46
    I. Higuchi, T. Tsukamoto, N. Sata, S. Yamaguchi, S. Shin, I. Hattori: Solid State Ion. 176(39/40), 2963 (2005)CrossRefGoogle Scholar
  47. 59.47
    W.J. Weber, C.W. Griffin, J.L. Bates: J. Am. Ceram. Soc. 70, 265 (1987)CrossRefGoogle Scholar
  48. 59.48
    I. Yasuda, T. Hikita: J. Electrochem. Soc. 140, 1699 (1993)CrossRefGoogle Scholar
  49. 59.49
    J.W. Fergus: Solid State Ion. 171, 1 (2004)CrossRefGoogle Scholar
  50. 59.50
    J.A. Kilner: Solid State Ion. 129, 13 (2000)CrossRefGoogle Scholar
  51. 59.51
    L.A. Tan, X.H. Gu, L. Yang, W.Q. Jin, L.X. Zhang, N.P. Xu: J. Membrane Sci. 212, 157 (2003)CrossRefGoogle Scholar
  52. 59.52
    T. Ishihara, S. Miyoshi, T. Furuno, O. Sanguanruang, H. Matsumoto: Solid State Ion. 177, 3087 (2006)CrossRefGoogle Scholar
  53. 59.53
    M. Yashima, H. Yamada, S. Nuansaeng, T. Ishihara: Chem. Mater. 24, 4100 (2012)CrossRefGoogle Scholar
  54. 59.54
    B. Ma, U. Balachandran, J.H. Park, C.U. Segre: J. Electrochem. Soc. 143, 1736 (1996)CrossRefGoogle Scholar
  55. 59.55
    T. Ishihara, Y. Tsuruta, T. Todaka, H. Nishiguchi, Y. Takita: Solid Sate Ion. 152/153, 709 (2002)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Dept. of Applied ChemistryKyushu UniversityFukuokaJapan

Personalised recommendations