Zr₇Ni₁₀ AND Zr₉Ni_11.5, REINVESTIGATED AND RELATED

Radovan Cerný¹ and Jean-Marc Joubert²

¹- Laboratoire de Cristallographie, Université de Genève, 24, quai Ernest-Ansermet, CH-1211 Genève 4, Suisse,e-mail: Radovan.Cerny@cryst.unige.ch
²- Laboratoire de Chimie Métallurgique des Terres Rares, CNRS - UPR 209, 2-8, rue Henri Dunant, F-94320 Thiais Cedex, France

Keywords: intermetallic compound, crystal structure, single crystal x-ray diffraction

Zr₇Ni₁₀ : Two crystal structures of that phase were found in [1] based on single crystal photographic data: the nickel-rich phase (referred to as stoichiometric, space group Aba2) and the zirconium-rich phase (space group Pbca). We have shown in [2] using single crystal diffractometer data that the true space group of the so-called stoichiometric phase is Cmca, a = 12.38(1), b = 9.185(5), c = 9.221(5) Å. The structure is composed of pseudo-tetragonal buckled layers which are stacked along the a axis. The interatomic distances between Zr and its nearest Ni neighbours (2.60 - 2.75 Å) are significantly shorter than the sum of the metallic radii (2.84 Å) thus indicating strong Zr - Ni interactions. The Ni - Ni distances (2.62 - 2.63 Å) are longer than, and the Zr - Zr distances (3.19 - 3.35 Å) similar to the sum of metallic radii (Ni - Ni: 2.48 Å; Zr - Zr: 3.20 Å).

Zr₉Ni_11.5 : First structural model of that phase was proposed in [3] based on powder diffraction data. It is isostructural with Zr₉Pt₁₁ [4], I4/m. In both works the close relation to the CsCl structure type was discussed. We have found using single crystal diffractometer data that the true symmetry is P4, a = 9.898(2), c = 6.622(2) Å. The lowering of the symmetry compared to the older model is due to the substitution of some Ni atoms in the 0,0,z and 1/2,1/2,z channels by Ni-Ni dumbbells. Two types of atoms chains along these channels were found in the crystal structure :

A : Zr ... Ni-Ni ... Zr ... Ni ... Zr
B : Zr ... Ni-Ni ... Zr ... Ni ... Zr ... Ni ... Zr

The periodicity of the chain A is 2c, that of the chain B is 3c. The positions of these two chains are not ordered in the c direction, because no superstructure reflections corresponding to the periodicity of 6c were observed. Ordering in the a (b) direction is not probable either, however it was difficult to check it, due to not sufficient resolution of the diffraction data. The same probability of the occurrence of both chains in the structure was therefore accepted giving the composition Zr₉Ni_11.5, which was confirmed by the electron micro-probe analysis and the density measurement. The interatomic distances follow the same rules as in Zr₇Ni₁₀, however they cover broader distributions : Zr - Ni (2.53 - 2.85 Å), Ni - Ni (2.59 - 2.69 Å, except the distances in the dumbbells), Zr - Zr (3.10 - 3.37 Å). The Ni - Ni distances in the dumbbells (2.47 - 2.53 Å) are similar to the sum of metallic radii.

Relation to CsCl-type : Both crystal structures can be derived from the CsCl structure type by substituting some Zr atoms by pairs of Ni atoms (Zr₇Ni₁₀) or by single Ni atoms (Zr₉Ni_11.5). In addition to that some Ni atoms are replaced by the Ni-Ni dumbbells in the Zr₉Ni_11.5.

[1] Kirkpatrick, M.E., Smith, J.F. & Larsen, W.L., Acta Cryst. 15 (1962) 894-903
[2] Joubert, J.-M., Èerný, R., Yvon, K., Latroche, M. & Percheron-Guégan, A., Acta Cryst. C53 (1997) 1536-1538
[3] Glimois, J.L., Becle, C., Develey, G. & Moreau, J.M., J. Less-Com. Met. 64 (1979) 87-90
[4] Panda, S.C. & Bhan, S., J. Less-Com. Met. 34 (1974) 344-347