Search

Unlike the addition of a second equivalent of diazoalkane to C60, which does not show marked selectivity, the addition of azides to C60 shows chemo and regio selectivity [4]. Utilization of this unique selectivity, led us to the preparation of the first heterofullerene ‘C59N’ in bulk quantity [5, 6]. Herein we report on its synthesis and mechanism of formation. We also describe the application of this procedure for the formation of higher heterofullerene analogs, i.e. ‘C69N’ isomers [7] and introduce methodology that opens a viable route to the functionalization of azafullerenes [8]. The azafullerene (C59N)2 can be synthesized from C60 in three steps. The initial step in the formation of azafullerenes is the 1,3-dipolar cycloaddition of 2methoxyethoxymethyl azide (MEM-azide) [9]. Not unlike the addition of diazoalkane, the addition of an alkyl azide to C60 occurs across the [6,6] bond. At room temperature, the initially formed triazoline can be isolated in 35-40% yield. Thermolysis of the triazoline, followed by concomitant loss of nitrogen, gives rise to azafulleroids and fulleroaziridines. Fulleroaziridines and azafulleroids can also be obtained directly from C60 at higher reaction temperatures (160-180°C).

Using a combination of results from photoemission, electron energy-loss spectroscopy in transmission and gradient-corrected density functional calculations, we have examined the effects of chemical substitution of N for C in the dimerised heterofullerene (C59N)2. We show that there is a new HOMO state, which lies close in energy to the hu-states of C60 and has a significant degree of N-character. In contrast, the lowest lying unoccupied levels have predominantly C-character. The photoemission data of the alkali metal intercalated salts AxC59N reveal signs of at least one additional phase other than the undoped (C59N)2 and an alkali metal saturated phase. At no stage during the intercalation is experimental evidence for metallic behaviour (in the form of a density of states at the Fermi level) observed.

We report the results of a detailed study of the occupied and unoccupied electronic structure of dimers of the new heterofullerene C59N by means of photoemission and electron energy-loss spectroscopy. A close similarity is found between the electronic structures of pristine (C59N)2 and C60 with an additional broadening of the spectra in the former due to the distortion of the fullerene cage caused both by dimerization and the chemical substitution. Both the occupied and unoccupied electronic states, as well as the interband transitions between them, attest to the high degree of molecular character retained in the solid state. Comparison of the shake-up structures in the C1s and N1s X-ray photo emission spectra confirm that the highest lying occupied states in the heterofullerene have a strong degree of N character, whereas the lowest lying unoccupied states have mainly C character. We also present the optical conductivity of the heterofullerene (derived from the loss function), which shows an optical gap of 1.4 eV, some 0.4 eV smaller than that of C60.

We present electron energy-loss spectroscopy studies in transmission of the electronic structure of (C59N)2. The overall broadening observed for the π-derived structures in the low energy-loss function of (C59N)2 when compared to C60 is due to the reduced symmetry of the heterofullerene. The optical gap of (C59N)2 is about 1.4 eV, slightly smaller than that of C60 (1.8 eV). Using a Kramers-Kronig analysis the dielectric function in the low energy region was calculated. A static dielectric function at zero energy of ∼ 5.6 is found.