Crystal structure of post-perovskite-type CaIrO3 reinvestigated: new insights into atomic thermal vibration behaviors

Single crystals of CaIrO3 were grown from a CaCl2 flux at atmospheric pressure and crystallized with the post-perovskite type of structure. The crystal structure is reinvestigated on the basis of single-crystal X-ray diffraction data measured using a high-power X-ray source, and the atomic thermal vibration behavior is discussed in terms of the coordination environments.

The crystal structure of the post-perovskite-type CaIrO 3 was first proposed by Rodi & Babel (1965) on the basis of a single-crystal X-ray diffraction experiment, but incorrect atomic positions were reported. Recently, we have successfully grown single crystals of the post-perovskite-type CaIrO 3 and refined the crystal structure of this compound on the basis of single-crystal X-ray diffraction data measured using a sealed X-ray tube (40 kV, 30 mA) as the radiation source (Sugahara et al., 2008). However, the measured intensity data were rather weak and their accuracy was rather low, because thin needlelike crystals were obtained and the selected crystal for the intensity measurements had a poor grade of crystallinity. This resulted in rather large reliability indices [R(F) = 0.064, wR(F) = 0.065 for 377 reflections] and in structural parameters with rather large uncertainties. In particular, the resulting displacement ellipsoids were unusually elongated or flattened. Subsequently, Hirai et al. (2009) reinvestigated the crystal structure of the post-perovskite-type CaIrO 3 by single-crystal X-ray diffraction and conducted structure refinements for two different crystals using two different types of diffractometers. The two independent refinements showed convergent results with much better reliability indices [R(F 2 ) = 0.013, wR(F 2 ) = 0.031 for 365 reflections; R(F 2 ) = 0.007, wR(F 2 ) = 0.008 for 149 reflections] and structural parameters with reasonably smaller uncertainties. Consequently, Hirai et al. (2009) concluded that the displacement ellipsoids had no significant anisotropies in contradiction to our previous report (Sugahara et al., 2008), but provided no further details of the atomic thermal vibration behaviors. Their X-ray diffraction experiments were conducted under the operating conditions of 2 max = 80 at 45 kV/40 mA for one crystal and 2 max = 55 at 50 kV/85 mA for the other crystal. These operating conditions with a low X-ray power and a relatively low 2 max value may be insufficient for the determination of reliable atomic displacement parameters (ADPs).
In the present study, the crystal structure of the postperovskite-type CaIrO 3 was reinvestigated on the basis of single-crystal X-ray diffraction data measured over a much wider 2 range using a high-power rotating-anode X-ray generator (60 kV, 250 mA). Special attention to exclude the influence of multiple scattering effects and secondary extinction effects on ADPs was paid as far as possible during data reduction and structure refinement procedures, as will be described in Section 5. On the basis of the resulting structural parameters, the validity of the crystal structure proposed by Hirai et al. (2009) is examined and the detailed atomic thermal vibration behaviors are discussed.

Structural commentary
The post-perovskite-type phase of CaIrO 3 crystallizes in the space group Cmcm. The crystal structure consists of IrO 6 octahedral layers and CaO 8 hendecahedral layers stacked alternately along [010] (Fig. 1)

Atomic thermal vibration behaviors
In the present structure refinement using a high-power X-ray source, the accuracy of the refined structural parameters was considerably improved compared with our previous report (Sugahara et al., 2008), being comparable to those reported by Hirai et al. (2009). The resulting positional parameters also show excellent consistency with those reported by Hirai et al. (2009). On the other hand, the present displacement ellipsoids (Fig. 4) are different from those given by Hirai et al. (2009). They considered that the thermal vibrations of the Ir and Ca atoms exhibited no significant anisotropies, but in fact the reported displacement ellipsoids of both atoms were somewhat elongated parallel to the (100) plane. In contrast, the mean-square displacements (MSDs) of both atoms obtained from the present refinement are as follows: Ir, 0.00316 (5) Å 2 along the shortest ellipsoid axis, 0.00319 (5)  Polyhedral view of an IrO 6 octahedral layer projected on (010). Symmetry codes: (i) x + 1 2 , y + 1 2 , z; (ii) x + 1 2 , Ày + 1 2 , Àz; (iii) x + 1, Ày + 1, z À 1 2 ; (iv) x + 1, y, Àz + 1 2 ; (v) x, Ày + 1, z À 1 2 ; (vi) x, y, Àz + 1 2 .

Figure 4
Unit cell of the CaIrO 3 post-perovskite with displacement ellipsoids drawn at the 80% probability level.

Synthesis and crystallization
The reagents Ca(OH) 2 and Ir were employed as the starting materials, and mixed together with CaCl 2 in the molar ratio Ca(OH) 2 :Ir:CaCl 2 = 1:1:10. The mixture was heated in air at 1100 K for 8 h, and then cooled gradually to 600 K at a rate of 10 K h À1 . Dark reddish-brown crystals of the post-perovskitetype CaIrO 3 with a thin needle shape were grown from the CaCl 2 flux. The crystals were isolated by dissolving the solidified CaCl 2 melt with distilled water.

Refinement
A total of 2593 intensity data up to 2 max = 100 were collected. After the intensity data were corrected for Lorentzpolarization factors and absorption effects ( -scan method; North et al., 1968), they were averaged in Laue symmetry mmm to give 692 independent reflections. Of these, independent reflections with F o 3(F o ) were eliminated. Even if independent reflections had intensities of F o > 3(F o ) after averaging, those averaged from a data set of equivalent reflections including reflection(s) with F o 3(F o ) were also discarded since these reflections were potentially affected by multiple scattering. Moreover, independent reflections with (sin )/ < 0.334 Å À1 were eliminated to reduce secondary extinction effects and to avoid dependence on atomic charge as far as possible in the choice of atomic scattering factors. Finally, 412 independent reflections were used in the present refinement. Several correction models for the secondary extinction effects were attempted during the refinement, and the isotropic correction of Type II (Becker & Coppens, 1974a,b) with a Gaussian particle-size-distribution model yielded the best fit. The reliability indices converged to R(F) = 0.0108 and wR(F) = 0.0104 for 412 reflections, comparable to those of Hirai et al. (2009), and were dramatically improved in comparison with those of our previous report (Sugahara et al., 2008). Crystal data, data collection and structure refinement details are summarized in Table 2  RADY (Sasaki, 1987); molecular graphics: ATOMS for Windows (Dowty, 2000); software used to prepare material for publication: publCIF (Westrip, 2010).