Gold cluster formation on a fullerene surface.

The growth of Au clusters on a fullerene thin film was investigated by in situ photoelectron spectroscopy in the ultraviolet (UPS) and x-ray (XPS) regime. Due to its highly corrugated surface fullerene films provide a wide range of bonding sites which could be exploited as molecular templates and serve to create a cluster superstructure. To gain insight into the fullerene-Au interaction two types of experiments were performed: (i) the deposition of Au on a fullerene surface, and (ii) the deposition of fullerenes on a Au surface. In both experiments an island growth mode is observed. The deposition of submonolayer amounts of C60 onto a gold film showed that the main interaction of the two species is due to chemisorption of the first C60 monolayer. In addition a constant band bending in the fullerene film is detected, but the UPS valence-band spectra show that there is no charge transfer from the Au to the C60 lowest unoccupied molecular orbital. In the reverse experiment, the cluster growth of Au on the corrugated C60 surface, the analysis of the Au core level does not reveal a specific bonding or nucleation site for Au atoms and clusters. This is in contrast to observations with Si clusters, which prefer to reside in the troughs between the fullerene molecules. The Au clusters grow continually from a size of about 55 atoms for the early stages of growth up to 150 atoms for the deposition of a nominal coverage of 1.5 nm. These data are derived from an analysis of the d-band splitting and the Au 4f core-level shift due to delayed photohole relaxation. The thermal stability of the Au-clusters-covered fullerene film was investigated by annealing in situ up to temperatures of 650 degrees C. For temperatures up to 450 degrees C a continuous growth of the clusters is detected, which is accompanied by a slight drop in Au concentration in the range of XPS for annealing temperatures higher than 350 degrees C. This may be due to a ripening of the clusters. The presence of Au apparently delays fullerene sublimation. The film shows a very good thermal stability and even after annealing at 650 degrees C there is still a fullerene film detectable in the photoelectron spectroscopy spectra.