Perovskite-Structured Compounds

The energy bands along a [100] direction for four perovskite-structure compounds as obtained by Mattheiss (1972b), and his interpretation of the atomic origin of the bands. The oxygen 2s bands lie at about — 16 eV, as would be suggested by the atomic sp-splitting. 

Zirconates and hafnates can be prepared by firing appropriate mixtures of oxides, carbonates or nitrates. None of them are known to contain discrete [M04]" or [MOs] ions. Compounds M ZrOs usually have the perovskite structure whereas M2Zr04 frequently adopt the spinel structure. 

Perovskites are compounds of the ABC3- type where C is often oxygen, but not always. Figure 11.6 shows two versions of the perovskite crystal structure... 

Zener, C. 1951. Interaction between the (/-shells in the transition metals. II. Ferromagnetic compounds of manganese with perovskite structure. Physical Review 82 403-405. 

The same analysis can be applied to compounds with a more complex formula. For example, the oxide LaCoCL, which adopts the cubic perovskite structure, usually shows a large positive Seebeck coefficient, of the order of +700 jjlV K-1, when prepared in air (Hebert et al., 2007). This indicates that there are holes present in the material. The La ions have a fixed valence, La3+, hence the presence of holes must be associated with the transition-metal ion present. Previous discussion suggests that LaCo03 has become slightly oxidized to LaCoCL+j, and contains a population of Co4+ ions (Co3+ + h or Coc0)- Each added oxygen ion will generate two holes, equivalent to two Co4+ ... 

Table 11.1 lists the resulting low-temperature phases calculated for this set of compounds. Where experimental data are available (marked with a star) the predicted structures are those observed at low temperatures. Inverse denotes a perovskite structure in which a large divalent ion is 12-coordinate and a smaller univalent ion 6-coordinate. Unit cell dimensions are predicted to within 1% of the measured values. 

RfLl phases The existence of these phases (cubic AuCu3 type) had been reported for R = La, Ce, Pr, Nd and Sm. Subsequently, however, Buschow and van Vucht (1967) found that many of the R3A1 phases do not form, unless some carbon is present. C atoms occupy the body-centred site, which in the AuCu3 type structure is normally vacant, while the Au atoms occupy the corners, and the Cu atoms the face-centred positions. When the face-centred position is completely filled, the structure is known as the anti-perovskite structure. This occurs for R = Nd, Sm, Gd, Tb, Dy Ho and Er. They also noted that neither N nor O would stabilize these compounds. Notice that Ce3Al and Pr3Al are truly binary compounds and C is not required for these two phases to form. Independently, Nowotny (1968) found that the anti-perovskite structure could be formed by C and N additions to R3M alloys, where M = Al, Ga, In, Tl, Sn and Pb. 

The perovskite structure and its variant and derivative structures, and superstructures, are adopted by many compounds with a formula 1 1 3 (and also with more complex compositions). The ideal, cubic perovskite structure is not very common, even the mineral CaTi03 is slightly distorted (an undistorted example is given by SrTi03). 

Perovskites and related compounds also have a three-dimensional structure. In perovskites of formula ABOj, the octahedra of BOj lie on a cubic lattice, and are joined at the corners. Between these octahedra are large sites for the A atoms. In ReOj, the A atoms are missing, so guests can be added to the A positions. Because adjacent octahedra are joined together by only one oxygen they can rotate relative to one another, changing the shape of the A site. In LijReOj, the rotation splits the large A sites into two smaller sites more suitable for Li ions (Cava et al, 1982). Bronzes of WO3 also have a perovskite structure. 

The structures of the compounds AMeFs are closely related to each other and can be derived from the well known perovskite structure. Therefore they may be generalizing referred to as fluoroperovskites, although some deformations of the cubic perovskite t e may occxir orthorhombic, tetragonal and hexagonal structures have been observed in ternary fluorides, in addition to the basic cubic type. 

A special case of distorted perovskite structures is reported of compounds KxFeF3 of the bronze type (243). There appear close relationships to the similar bronzes of tungsten, KxW03, which will not be discussed further in this paper. 

The same orthorhombic deformation of the perovskite structure, that Geller 111) reported of the ternary oxide GdFeOs, is according to Rudorff et al. 268, 270, 271), also present in the sodium compounds NaMeFs. The unequal sizes of the Na+- and fluoride-ions bring about a considerable distortion of their common close-packing. To describe the structure a... 

This type of orthorhombic perovskite structure appears, if the tolerance factor of Goldschmidt is smaller than t — 0.88. The example of the compound NaMnFs [t = 0.78), showing doubled lattice constants a and h (287), is likely to mark the lower limit of the field in which orthorhombic fluoro-perovskits of the GdFe03-t3q>e may occur. Fluoroperovskites which have a smaller tolerance factor than t = 0.78 never have been observed so far, nor do fluoride structures of the ilmenite type seem to exist, which might be expected for ya = Me, corresponding to 1=1/1/2=0.71. 

These copper-oxide compounds crystallize in the perovskite structure and superconductivity is based on the (hole or electron) doping in the copper-oxide planes. This is the reason why these materials can be regarded as being 2D. The first compound of the family was La2 i Sr i Cu04 with Tc 38 K, which soon led to YBa2Cu307 5 with Tc — 92 K for 5 < 1 (Bums, 1993). The non-copper oxide electron-doped perovskite Bai-jcK cBiOa exhibits superconductivity near 30 K for 0.3 < X < 0.5 (Cavaeta/., 1988). 

ASOa-perovskites of Table 6 extends from Dza through Can to Ci. Distinct band-splittings are not observed, however, 13) (Fig. 2). As is demonstrated for the YCra Ali-a 03-system, the Oo-transition is situated at 13600 to 13700 cm i independent of the Cr3+ concentration (Fig. 2) leading to a B55 of 705 cm (Table 3), which is more than 5% lower than for the corundum and spinel compounds discussed. This is in agreement with the possibility of relatively strong Ji-bonds in the perovskite structure, which could be inferred from spectroscopic and crystallographic results 3) as well as from NMR-data and MO-calculations 33) for Ni2+,... 

The Perovskite Structure, ABXS Systems. Cubic Pm3m (Space Group 221) A cubic structure was assigned to the mineral perovskite, CaTiOj, but this particular compound was later found to actually possess orthorhombic symmetry. Today, however, we refer to the perovskite structure in its idealized form as having cubic symmetry and it is normally represented by a simple unit cell (Figure 10). 

Ordered Perovskite-type Compounds, A2(BB )06 Systems Cubic Fmim A feature of the perovskite structure is that, with the proper substitutions, many types of ordered structures can readily be formed. This can be accomplished by the substitution of two suitable metal ions (with different oxidation states) in the octahedral sites of the structure. In this case the unit ceil is doubled along the three cubic axes to generate an 0.8 A unit ceil (Figure 15). Partial substitution of different transition metal ions in the octahedral sites is also possible the general formulation for these compounds would be A2(B2 xB x)06. The parentheses in this formulation enclose atoms occupying the octahedral sites in the structure. 

---