|
|
Fullerenes offer the exciting possibility of trapping atoms in the carbon cage. A
variety of applications for these novel materials is anticipated, ranging from molecular
container compounds used in organ imaging to organic superconductors with high critical
magnetic fields. Some of these predictions are based on known properties of metal doped
fullerides. However, these externally doped solids suffer from sensitivity to
oxygen and humidity. Internally doped, so-called endohedral complexes are
inherently more stable, because the carbon cage protects the encapsulated atoms.
We are preparing metallofullerenes by bombarding C60 with metal ions. A surface ionization source generates these metal ions with a low kinetic energy spread. Two different approaches are being pursued: Ion implantation into a film of fullerenes, and gas-phase collisions.
A fullerene film is grown by vapor deposition onto a rotating target; metal ions are co-implanted during film growth (see Figures above). The metal-to-fullerene ratio is of the order of 1:1 or less. A beam from a pulsed Nd:YAG laser is used to desorb and ionize the material into a time-of-flight mass spectrometer. The film is kept in vacuum throughout the experiment. Laser desorption time-of-flight mass spectra are shown in the Figure to the right. For sodium ion implantation at 20 eV we observe a C60 peak at 60 x 12 = 720 amu, but no heavier ions. The substructure of the peak is due to isotopomers, i.e. molecules containing one or more of the naturally occurring C-13 isotopes. For higher implantation energies, we observe NaC60 ions at 743 amu. Sodium (23 amu) is mono-isotopic; therefore the peak exhibits the same substructure as the C60 peak. The yield of sodium-fullerenes complexes is amazingly high, about 40 % relative to that of C60 at an implantation energy of 100 eV.
In another experimental approach, the metal ion beam is crossed with a thermal beam of fullerenes; ensuing metallofullerene ions are identified in a time-of-flight mass spectrometer (Figures 4 and 5). This approach is better suited for an accurate determination of the threshold energy for formation and dissociation of metallofullerenes. However, with the C60 source operating at 600 to 700 °C, only one out of about 107 metal ions will successfully collide with a fullerene; therefore the NaC60+ ion signal is very small. Research in collaboration with scientists in Austria and Denmark has also been devoted to fullerenes. In particular, two questions which had been controversial for a number of years have finally been answered: First, the nature of the reaction channel by which energetically excited fullerene ions lose a large number of carbon atoms is strictly sequential; C2 units rather than larger carbon complexes are ejected. Second, the activation energy for loss of C2 from C60 is approximately 11 eV. This value is in excellent agreement with theory, but in striking disagreement with a large number of earlier publications which had placed this important value near 5 to 8 eV.
Links to other fullerene research groups Eberhardt &
Kessler @ Jülich |