The performance of B3-LYP density functional theory in describing S(N)2 reactions at saturated carbon

The performance of the B3-LYP variant of density functional theory when used in conjunction with the 6-31G(d) and 6-311 + G(3df, 2p) basis sets in describing the prototypical gas-phase S(N)2 reactions of Cl- + CH3Cl and Cl- + CH3Br has been examined in detail. Reasonable values of the complexation energies (Delta H-comp) for the ion-molecule complexes formed in these reactions are obtained. However, the overall (Delta H-ovr(#)) and central (Delta H-cent(#)) barriers for these reactions calculated using the B3-LYP functional are significantly underestimated when compared with G2(-t) or experimental results. This implies that the B3-LYP energies for the Cl(H3CCl- (D-3h) and Cl(H3CBr- (C-3v) transition structures are relatively too low. The B3-LYP errors appear to be systematic, with similar errors being found for corresponding quantities for the two reactions examined.