Double halogenation controls the conformation of benzyl alcohol: does DFT fail to predict it even qualitatively?

Abstract

Benzyl alcohol and its derivatives can occur in chiral and in more elusive mirror-symmetric structures. ortho-Halogenation (X) couples this structural duality with distorted OH⋯X vs. very weak CH⋯O/CH⋯X hydrogen bonding. We show by a combination of infrared and rotational spectroscopy in supersonic jets that additional para-halogenation introduces a particularly interesting halogen-sensitive conformational preference. While ortho-para-difluorobenzyl alcohol prefers the chiral OH⋯X arrangement, ortho-para-dichlorobenzyl alcohol favors the achiral structure with CH⋯O/CH⋯X contacts. Density functional (exemplified by dispersion corrected B3LYP) calculations fully support the infrared and rotational spectra of the competing conformations, but fail in predicting the energetic switch from F to Cl substitution, even when harmonic zero point energy is added or the electronic energy is replaced by its DLPNO-CCSD(T) counterpart. Only the combined effect of vibrational zero point energy and correlated wavefunction energetics is able to correctly predict the switch to mirror symmetry for Cl. Is there a density functional approach out there that predicts the correct structural switch on its own?

Abstract as supplied by the indexing source. © the authors / publisher.