The question of planarity and the validity of the amide resonance model have been investigated in formamide on the basis of high-level quantum chemical calculations. Complete geometry optimizations were performed for the equilibrium structure and for the 90°-rotated transition state at the MBPT(2), MBPT(4), CCSD, and CCSD(T) electron correlation levels, with basis sets up to cc-PVTZ. While electron correlation tends to give nonplanar equilibrium, the final result at the CCSD(T)/PVTZ level is an exactly planar structure, as proven by the absence of imaginary vibrational frequencies. The crucial parameter in the geometry, the C - N bond length is calculated at 1.354 Å. For the barrier to internal rotation around the C - N bond our best estimate, including the zero-point-energy correction, is 15.2 ± 0.5 kcal/mol. To check predictions of the resonance model, we have analyzed geometric changes, charge shifts from Mulliken population analysis, and the nature of relevant valence orbitals and also calculated NMR chemical shieldings as a function of internal rotation. In contrast to recent suggestions by Wiberg et al. (J. Am. Chem. Soc. 1987, 109, 5935; 1992, 114, 831; Science 1991, 252, 1266) that π-resonance would not play a significant role in explaining the rotational barrier in formamide, we have found no compelling evidence to doubt the validity of the amide resonance model.
ASJC Scopus subject areas
- Physical and Theoretical Chemistry