Opening and closure of the fullerene cage in cis-1-bisimino adducts of C-60: The influence of the addition pattern and the addend

The synthesis, isolation, and spectroscopic characterization of the bisimino[60]fullerenes C-60(NCOOR)(2) (1 a: R = Et, 1b. R = tBu) with a cis-1 addition pattern as well as of their regioisomeric analogues 2-7 with different addition patterns are described. Whereas compounds 2-7 are typical fulleroaziridines and obey the rule of the minimization of [5,6] double bonds, the cis-1 isomers 1 a and 1 b represent the first examples of fullerene derivatives with open transannular [6,6] bonds. Characteristic features within the fullerene framework of these valence isomers VI are the presence of a doubly bridged open 14-membered ring with a phenanthrene perimeter as well as of an 8-membered 1,4-diazocine heterocycle. Moreover, it is shown that, by transforming cis-1-C60(NCOOtBu)2 (1 b) into cis-1-C60(NH)2 (1c), the fullerene cage can be closed in an intraring 2 pi --> 2 sigma isomerization to valence isomer V. These are the first chemical modifications of the fullerene core that allow the synthesis of both open and closed valence isomers with the same addition pattern. Density-functional as well as AM1 calculations corroborate the experimental findings that it depends on the addition pattern as well as on the nature of the addends whether or not the transannular [6,6] bonds are closed. Only in the cis-1 adducts C-60(NR)(2) that prefer planar imino bridges (e.g., carbamates or amides) are the open forms VI more stable than the closed isomers V.