Fullerene recognition with molecular tweezers made up of efficient buckybowls: a dispersion-corrected DFT study

In 2007, Sygula and co-workers introduced a novel type of molecular tweezers with buckybowl pincers that have attracted the substantial interest of researchers due to their ideal architecture for recognizing fullerenes by concave–convex π⋯π interactions (A. Sygula et al., J. Am. Chem. Soc., 2007, 129, 3842). Although in recent years some modifications have been performed on these original molecular tweezers to improve their ability for catching fullerenes, very few improvements were achieved to date. For that reason, in the present work a series of molecular tweezers have been devised and their supramolecular complexes with C60 studied at the B97-D2/TZVP//SCC-DFTB-D and B97-D2/TZVP levels. Three different strategies have been tested: (1) changing the corannulene pincers to other buckybowls, (2) replacing the tetrabenzocyclooctatetraene tether by a buckybowl, and (3) adding methyl groups on the molecular tweezers. According to the results, all the three approaches are effective, in such a way that a combination of the three strategies results in buckycatchers with complexation energies (with C60) up to 2.6 times larger than that of the original buckycatcher, reaching almost −100 kcal mol−1. The B97-D2/TZVP//SCC-DFTB-D approach can be a rapid screening tool for testing new molecular tweezers. However, since this approach does not reproduce correctly the deformation energy and this energy represents an important contribution to the total complexation energy of complexes, subsequent higher-level re-optimization is compulsory to achieve reliable results (the full B97-D2/TZVP level is used herein). This re-optimization could be superfluous when quite rigid buckycatchers are studied.