BOND LENGTHS AND BOND VALENCES IN (Ba0.875Sr0.125)RuO3, BaRuO3 AND Ba3MRu2O9 (M=Fe,Co, Ni, Zn, In)

A. Santoro1 , Q. Huang1,2

1 NIST Center for Neutron Research, National Institute of Standards and Technology, Gaithersburg, MD 20899, USA.
2 Department of Materials and Nuclear Engineering, University of Maryland, College Park, MD 20742, USA.

Keywords: crystal chemistry, bond length predictions, bond valence and distortion, valence determinations.

The compounds (Ba0.875Sr0.125)RuO3, BaRuO3 and Ba3MRu2O9 (M=Fe, Co, Ni, Zn, In) crystallize, respectively, with the 4-, 9- and 6-layer structures, formed by layers RX3 of close-packed spheres of (almost) equal size, stacked according to the sequences ...ABAC..., ...ABABCBCAC..., and ...ABCACB... . As a consequence of these packings, the RuO6 octahedra share either corners or faces with sequences ...cfcf..., ...cffcffcff..., and ...cfccfc..., where c indicates a layer on which the octahedra share corners and f one on which they share faces. Bond valence sums calculated from the observed bond distances (D. Altermatt and I.D.Brown, Acta Cryst. B41, 240(1985); ibid., B41, 244(1985)) show that in every compound the apparent valence of Ba is significantly larger than 2, and that of Ru lower than the expected values (4, 4.5, and 5 v.u.). Bond length predictions, obtained from bond valence sum equations and ring equations (I.D.Brown, Acta Cryst. B48, 553(1992); M.O'keeffe, Structure and Bonding, 71, 161(1989)) show that the Ba-O and Ru-O distances which satisfy valence requirements of the atoms are incommensurate under the constraints imposed by the space group symmetry observed for the three structural types. In these conditions, the presence of strains is unavoidable.

Following the criteria used by Brown in his studies of YBa2Cu3Ox (J.Solid State Chem. 82, 122(1989), ibid., 90, 199(1991)) and La2NiO4 (Zeit. f. Kristall, 199, 255(1992)), it is possible to model quite accurately the 4- and 9-layer structures, and thus discover that these structures accommodate the imcommensurability of the Ba-O and Ru-O bonds by adjusting the lattice parameters until a best compromise is reached between the bond valence requirements of the Ba and Ru atoms. Further relaxation of the structures is provided by shifts of the Ru and O atoms and by a re-adjustment of the c-parameter, until "reasonable" Ru-Ru, Ru-O, and O-O distances are achieved, with a correspondingly acceptable value of the Ru valence. The final models of the 4- and 9-layer structures are quite close to the observed values and thus completely explain the experimental results. In the 6-layer compounds, the valence of Co and Ni(2+) and that of Fe (3+) are derived from a comparison of the Ru-O distances in these compounds and in those containing Zn and In. The analysis of these structures, however, is not yet complete, because accurate values of the bond valence parameters for Ru3+-O and Ru5+-O are not available at this time.