Ruddlesden-Popper phase

Many examples of oxide-based Ruddlesden-Popper analogues have 

The structures can be described in a number of ways, each of which is suited to an explanation of differing properties. Each entire perovskite-like slab, including the edges of the sheets, has a formula but if the region between these 

In practice, many phases are disordered and contain random or partly ordered wider or narrower bands of perovskite-like stracture corresponding to different n values. Non-stoichiometry is common for phases containing Co, Mn or Fe as the B-site cation. In addition, octahedral tilting and distortion are present in the perovs-kite slabs, and these both lower the symmetry of the structures and influence their physical properties. 

A number of A BO phases are isostractural to NdjCuO, including the lantha-noid cuprates Pr CuO and Sm CuO (Table 4.3). It has also been found that 

LajCuO can adopt this structure as a metastable form that reverts to the K NiF type on heating. Rather surprisingly, the phase Sr CuO F also adopts the Nd CuO structure, although the fluorine-rich compound Sr CuPjF takes on the LapuO structure type. 

---

Fig. 1. (a) The ideal cubic perovskite structure and (b) the n = 1 Ruddlesden-Popper phase AO AMO3. 

Compounds Containing Perovsldte Layers. A second class of layered oxides have structures related to the three-dimensional perovsldte lattice and include the Auriv-iUius phases, the Ruddlesden Popper phases and the Dion-Jacobson phases. The general composition can be written M [A iB 03 +i] where A is an alkaline or rare earth metal, and B is niobium or titanium. In the AurivUhus phases M = Bi202 +, whereas M is an aUcah metal cation in the ion-exchangeable Ruddlesden Popper a = 2) and Dion-Jacobson a = 1) phases. The relationships between the three structure types is shown in Figure 14. The intercalation chemistry of the Dion Jacobson phases was the first to be studied. 

It is not surprising that perovskite oxides continue to attract unusual levels of attention. In this category, we include the Ruddlesden Popper phases of compositions (A0)(AT03) where A is a large cation, either a divalent alkaline earth or a trivalent lanthanide, and T is a transition element. The index, n, enumerates the number of perovskite ATO3 units intergrown with rock salt AO layers. 

Idealizations of the crystal stmctures of both the Ruddlesden-Popper phases from n = 1, 2, 3, and 00 (the parent perovskite) are shown in Fignre 15. 

Figure 15 Idealized crystal structures of the Ruddlesden-Popper phases, (A0)(AB03) , n = 00, ABO3 n = 1, A2BO4 n = 2, A3B2O7 n = 3, A4B3O10. The comer-sharing BO3 octahedrals are shown in polyhedral representation while the A atoms are shown as spheres...

Figure 28 Phase diagram for n = 2 Ruddlesden-Popper phases illustrated hy Sr2 j Lai+j Mn207. (Reprinted from Ref 51, 1999, with permission from Elsevier)...

The Ruddlesden-Popper phase has the general formula A 2[A ,B 03 +i], where [A iB 03 +i] donates perovskite-like slabs of n octahedra in thickness, formed by slicing the perovskite structure along one of the cubic directions, and A indicates the interleaved cations. Sr2Ti04, Sr3Ti207, Sr4Ti30,o, and A 2[Ln2Ti30,o] (A = K, Rb Ln = lanthanide) are included in this series. 

One problem with any material with a layered structure is that these materials often exhibit stacking faults. As the structure is the same in two dimensions, it can easily be disrupted to produce the wrong stacking sequence. As the layer sequence becomes more complex, the hkeh-hood of stacking faults increases. Many Ruddlesden-Popper phases exist for = 1, but as n increases the number of known phases decreases. This is likely to be related to the increased formation of stacking faults, which prevents isolation of the perfectly ordered phase for characterization purposes. 

If the pair of atoms at the boundaries of the perovskite-like sheets in the Ruddlesden-Popper phases are replaced with just one A atom, the series takes the formula AXA jB Oj j), and the materials are called Dion-Jacobson phases (Table 4.4). The perovskite slabs are cut parallel to the [100] planes of an ideal perovskite parent and have a composition (A As with the... 

Type 111 structures have a displacement of (ap+bp)/2 between the successive perovskite layers (Figure 4.4e and f), which is the same as in the Ruddlesden-Popper phases. These form for the smallest A cations, Li, Na and Ag, and are represented by LiCajTajOjg and NaCa TajOj. The ideal structures are tetragonal with a=b a, with the c-axis taken as perpendicular to the perovskite slabs. The oxygen coordination of the A cations is tetrahedral. 

As with the Ruddlesden-Popper phases, non-stoichiometry is common for phases containing Co, Mn or Fe as the B-site cation. 

The Aurivillius phases again contain slabs of perovskite sliced along the ideal [100] direction. They are formed by replacement of the interlayer A structures in the Ruddlesden-Popper phases and A in the Dion-Jacobson phases with a layer of composition Bi O - This gives the series a general formula (Bip XA jB j, ), sometimes written in an ionic form (Bip ) (A jBPj j) " (Table 4.5). As before, the perovskite slabs have the formula (A where A is a large cation nom-... 

The phases related to Ca Nb O, as in the phases described previously, are built from slabs of the perovskite stmcture, this time cut into slabs parallel to ideal per-ovskite [llOJp planes. The first materials of this type were found in the system bounded by the end members Ca Nb O and perovskite NaNbOj. The structure of CUjNbjO is, in reality, composed of slabs four octahedra in thickness and for this reason is better written Ca Nb Oj (Figure 4.6a). The perovskite slabs are stacked along the c-axis of the idealised unit cell, and each slab is displaced from its neighbours by (ap+bp)/2, as in the Ruddlesden-Popper phases. The BOg octahedra have a crenellated appearance viewed down and comer-linked rows when viewed along bp (Figure 4.6b). 

Oxygen loss is common in phases in which the B cations can take on a variable valence. In some phases these are distributed at random over the normal oxygen positions, thereby transforming some BO octahedra into BO square pyramids. This occurs in the n=2 Ruddlesden-Popper phase LajNi O when reduced in hydrogen to a composition of La3Ni20g3j. In the reduced stmcture, almost 2/3 of the nickel atoms are in square pyramidal coordination, which is well suited to the nominal charge state of NF (Ni j), thus generating... 

The oxygen vacancies appear to be disordered in this phase. Similarly, in the n=3 Ruddlesden-Popper phase Sr Fep in which 5 lies between 0 and 0.73, the oxygen vacancies appear to be disordered. The oxygen loss is balanced by the Fe " /Fe " ratio, as nominally StjFe +Og gradually transformations to StjFe Fe Og j. The reaction for the oxidation of La,Fe,0. is... 

A similar ordering is found within each of the perovskite layers in the reduced n=3 Ruddlesden-Popper phase SrjMn Og, derived from the fuUy oxidised Sr Mn O,. Under normal preparation methods the ordering is confined to the perovskite layers and does not extend to three dimensions in the macroscopic crystal, although electron microscopy suggests that microdomains of such three-dimensionally ordered stmctures do exist. These structures are similar to those of Mn-containing brownmiUerite-related compounds (Section 2.5.1). 

Three-dimensional ordering of square pyramidal coordination does occur in a number of Fe-containing phases. An example is provided by the n=3 Ruddlesden-Popper phase Sr FejOjg in which the Sr is substituted by La. In the series so formed, replacement of Sr + by La + in the parent phase Sr Fe +Oj generates an equal number of balancing Fe " ions to give La Sr La (Fej Fe ) 0, with counterbalanced defects La and Fe (Fe "). 

There are large numbers of these reactions described in the literature. The transformation of Dion-Jacobson phases into the corresponding Ruddlesden-Popper phases is often the first step in making new compounds. Ruddlesden-Popper phases can, in most cases, be made by heating the corresponding Dion-Jacobson solid in the vapour of an alkali metal, lithium being the exception. These Ruddlesden-Popper structures can then be transformed into modular structures separated by A and OH- or A and Cl- (Figure 4.16). 

---