Pyrene is not an appropriate model for miniature graphene in edge-functionalization

We claim, on the basis of experimental and theoretical grounds, that pyrene cannot serve as a model for graphene for the purpose of edge-functionalization, as previously reported on the basis of direct Friedel–Crafts acylation of pyrene in polyphosphoric acid. Pyrene has been shown to undergo, at the most, Friedel–Crafts tetraacylation, at positions 1, 3, 6, and 8. Furthermore, pyrene has not undergone any Friedel–Crafts acylations at positions 2, 4, 5, 7, 9 and 10. We show, using DFT calculations (B3LYP/6–31G(d) and M06/6–31G(d)) that the regioselective Friedel–Crafts acylation of pyrene is kinetically controlled and is not thermodynamically controlled. The relative stabilities (ΔG 298, kJ/mol) of the σ-complexes of monobenzoylpyrenes (σ-BzPYH+) are 1σ-BzPYH+ (0.0) >4σ-BzPYH+ (37.5) >2σ-BzPYH+ (53.1). The most stable σ-complexes of dibenzoylpyrenes are also protonated at position 1 (1σ,3Z-, 1σ,4Z-, 1σ,6Z-, 1σ,8Z-, 1σ,9Z- and 1σ,10Z-Bz2PYH+). 2-BzPY and 2,7-Bz2PY, which are the most stable isomers in their respective series, are not formed in the Friedel–Crafts benzoylation of pyrene. The introduced electron-withdrawing acyl groups at positions 1, 3, 6, and 8 deactivate further acylation(s) of the pyrene ring system at ortho and peri positions, preventing the formation of, e.g., Bz n PY with n > 4. DFT calculations of bisanthene (BA) show that the preferred kinetically controlled products of mono- and dibenzoylations of bisanthene are 7-BzBA and 7,14-Bz2BA, respectively. The formation of neighboring peri and ortho acyl substituents at pyrene and bisanthene may be prohibited, due to considerable overcrowding.