Intercalation of sodium heteroclusters into the C60 lattice

• DOI: 10.1038/360568a0
• Publication Date: 2003-06-12T23:50:11Z
• Authors: T. Yildirim; O. Zhou; J. E. Fischer; N. Bykovetz; R. A. Strongin; M. A. Cichy; A. B. Smith III; C. L. Lin; R. Jelinek
• Keywords: 01 natural sciences; 0104 chemical sciences

INTERCALATION of sodium into C60 has been shown1 to yield a range of compounds Nax C60 (2<x<6). Unlike the other alkali-metal-doped compounds, there is no evidence of superconductivity within this range of doping. Furthermore, Na6C60 retains the face-centred-cubic (f.c.c.) structure of undoped C60 whereas the analogous x = 6 phases with K, Rb or Cs are body-centred cubic (b.c.c.)2. Here we report the preparation of sodium-doped C60 with x up to about 10; the structure remains f.c.c. throughout. Rietveld refinement of X-ray diffraction data yields a structure with ideal composition Na11C60, in which the octahedral interstitial site contains a nine-atom body-centred cluster of sodium atoms while the tetrahedral sites are singly occupied. Clusters of four to nine sodium atoms in the octahedral site provide sufficient 'chemical pressure' to prevent the low-temperature lattice distortion that occurs when this site is singly occupied, as in Na3C60 (ref. 1). This distortion is thought to suppress superconductivity in the latter compound. Thus we suggest that new sodium-doped superconduct-ing phases may be found in the range 4< x< 11.