Energy Barriers of Be and B in Passing Through the C60 Fullerene Cage

We have studied the potential barriers for the penetration of atomic beryllium or boron inside the C60 fullerene by performing ab initio density functional theory (DFT) calculations with three variants for the exchange and correlation: B3LYP (hybrid functional), PW91 and PBE. Four principal trajectories to the inner part of C60 for the penetrating atom have been considered: through the center of six-member-carbon ring (hexagon), five-member-carbon ring (pentagon), and also through the center of the double C-C bond (D-bond) and the center of the single C-C bond (S-bond). Averaging over the three DFT variants yields the following barriers for beryllium penetrating inside a deformable fullerene: 3.2 eV (hexagon), 4.8 eV (S-bond), 5.3 eV (D-bond), 5.9 eV (pentagon). These barriers correspond to the slow and adiabatic penetration of Be, in contrast to the fast (non-adiabatic) penetration through the rigid cage of C60 resulting in 5.6 eV (hexagon), 16.3 eV (pentagon), 81.8 eV (S-bond) and 93.4 eV (D-bond). The potential barriers for the boron penetrating inside deformable/rigid C60 are: 3.7/105.4 eV (D-bond), 4.0/86.8 eV (S-bond), 4.7/7.8 eV (hexagon), 6.8/14.0 eV (pentagon). The potential barriers for Be and B escaping from the inner part of C_60 are higher by the value of 0.84 eV for Be and 0.81 eV for B. The considerable reduction of the potential barriers for the deformable fullerene is ascribed to the formation of the Be-C and B-C bonds. We discuss the difference between Be and B, compare three variants of DFT, and analyze the role of the dispersion interaction.