Fullerenes nitrene addition

Cycloadditions to [6,6]-double bonds of Cjq are among the most important reactions in fullerene chemistry. For a second attack to a [6,6]-bond of a C q monoadduct nine different sites are available (Figure 10.1). For bisadducts with different but symmetrical addends nine regioisomeric bisadducts are, in principle, possible. If only one type of symmetrical addends is allowed, eight different regioisomers can be considered, since attack to both e - and e"-positions leads to the same product. Two successive cycloadditions mostly represent the fundamental case and form the basis for the regioselectivity of multiple additions. In a comprehensive study of bisadduct formations with two identical as well as with two different addends, nucleophilic cyclopropanations, Bamford-Stevens reactions with dimethoxybenzo-phenone-tosylhydrazone and nitrene additions have been analyzed in detail (Scheme 10.1) [3, 9, 10]. 

In contrast to the N-MEM (N-alkyl) analogue (see Sect. 2.2.2), N-aryl-aza-fiilleroids, obtained from the thermal reaction of aryl azides with 50, can be photochemically isomerized to azirenofullerenes [51]. This closely resembles the difference in the photochemical behavior of C-aryl- and C-alkyl fulleroids obtained from the reaction of diazo compounds with Qo (e.g. [5,6] CgiHj is photochemically stable). After some initial studies (see for example [52-55]), the addition of azides and nitrenes to C o has been investigated and used for the preparation of a series of functional fullerene derivatives by a number of other investigators, but their work is not relevant in relation to the preparation of aza[60] fullerene. The addition of azides to fullerenes has been brie% reviewed previously [56-58]. 

---