Crystal structure of the sodium cobaltate deuterate superconductor NaxCoO2⋅4xD2O (x≈13)

Abstract

Neutron and x-ray powder diffraction have been used to investigate the crystal structures of a sample of the newly-discovered superconducting sodium cobaltate deuterate compound with composition ${\mathrm{Na}}_{0.31\left(3\right)\mathrm{}}{\mathrm{CoO}}_2\cdot 1.25\left(2\right)\mathrm{}{\mathrm{D}}_2\mathrm{O}$ and its anhydrous parent compound ${\mathrm{Na}}_{0.61\left(1\right)\mathrm{}}{\mathrm{CoO}}_2.$ The anhydrous parent compound ${\mathrm{Na}}_{0.61\left(1\right)\mathrm{}}{\mathrm{CoO}}_2$ has two partially occupied Na sites sandwiched, in the same plane, between ${\mathrm{CoO}}_2$ layers. When Na is removed to make the superconducting composition, the Na site that experiences the strongest Na–Co repulsion is emptied while the occupancy of the other Na site is reduced to about one third. The deuterate superconducting compound is formed by coordinating four ${\mathrm{D}}_2\mathrm{O}$ molecules (two above and two below) to each remaining Na ion in a way that gives Na–O distances nearly equal to those in the parent compound. One deuteron of the ${\mathrm{D}}_2\mathrm{O}$ molecule is hydrogen bonded to an oxygen atom in the ${\mathrm{CoO}}_2$ plane and the oxygen atom and the second deuteron of each ${\mathrm{D}}_2\mathrm{O}$ molecule lie approximately in a plane between the Na layer and the ${\mathrm{CoO}}_2$ layers. This coordination of Na by four ${\mathrm{D}}_2\mathrm{O}$ molecules leads in a straightforward way to ordering of the Na ions and ${\mathrm{D}}_2\mathrm{O}$ molecules consistent with the observation of additional shorter-range scattering features in the diffraction data. The sample studied here, which has $T_c=4.5\mathrm{}\mathrm{K},$ has a refined composition of ${\mathrm{Na}}_{0.31\left(3\right)\mathrm{}}{\mathrm{CoO}}_2\cdot 1.25\left(2\right)\mathrm{}{\mathrm{D}}_2\mathrm{O},$ in agreement with the expected 1:4 ratio of Na to ${\mathrm{D}}_2\mathrm{O}.$ These results show that the optimal superconducting composition should be viewed as a specific hydrated compound, not a solid solution of Na and ${\mathrm{D}}_2\mathrm{O}$ $\left({\mathrm{H}}_2\mathrm{O}\right)$ in ${\mathrm{Na}}_x{\mathrm{CoO}}_2\cdot {\mathrm{D}}_2\mathrm{O}.$ The hydrated superconducting compound may be stable over a limited range of Na and ${\mathrm{D}}_2\mathrm{O}$ concentration, but studies of $T_c$ and other physical properties vs Na or ${\mathrm{D}}_2\mathrm{O}$ composition should be viewed with caution until it is verified that the compound remains in the same phase over the composition range of the study.