On the Choice of Optimal Methodology for Calculation of 13C and 1H NMR Isotropic Chemical Shifts in Cagelike Systems. Case Studies of Adamantane, 2-Adamantanone, and 2, 4-Methano-2, 4-dehydroadamantane

The (13)C and (1)H isotropic chemical shift values computed at HF, BLYP, B3LYP, and MPW1PW91/6-311+G(2d,p) levels of theory, for the BLYP and B3LYP/6-31G(d,p) optimized geometries of adamantane, 2-adamantanone, and 2,4-methano-2,4-dehydroadamantane ([3.1.1] propellane) are reported and compared with the experimental data. Except for the "inverted" carbon atoms and some of their nearest neighbors, the HF values are superior over the DFT ones, when the isotropic shifts with respect to TMS are in question. However, in case of the relative shifts computed with respect to the most deshielded center within the molecule, the DFT methods yield significantly better agreement with the experimental data than the HF method, the hybrid DFT methods being superior over "pure" DFT ones. The most probable reason for these findings may be the cancellation of errors arising from the inappropriate description of the paramagnetic contributions to the overall shielding tensor within the Kohn-Sham approach when an internal standard (within a molecule) is chosen, instead of an external one. Almost excellent linear correlation was found between the calculated and experimental relative shift values, which is significantly superior at DFT levels than at HF level, further proving the more systematical nature of errors in predicting the second-order magnetic response properties at DFT levels of theory. Among all DFT methods employed, the MPW1PW91 showed the best performance, in line with the significantly improved long-range behavior of this functional, as compared to the B3LYP one.