DFT/B3LYP is used to calculate the gas-phase absolute and relative phenolic O-H bond dissociation enthalpies (BDEs) in hydroxy/methoxy ortho substituted phenols. The PCM and SCIPCM continuum models are applied to calculate the liquid-phase BDEs. This is the first theoretical determination of liquid-phase BDEs of phenols, the corresponding experimental data of which is rare. The solvated-phase optimized structures of both the parent phenols and their respective radicals are also presented for the first time. A systematic study on a series of 17 different basis sets on phenol, 2-hydroxyphenol (catechol), and 2-methoxyphenol (guaiacol) leads to the optimum 6-31+G(,3pd) basis set. Derived BDEs are among the most accurate of any gas-phase ones (deviations of the absolute gas-phase BDEs do not exceed 0.20 kcal/mol, relative to experiment, and those of the relative ones do not exceed 0.24 kcal/mol). Use of the optimum basis set to obtain the absolute gas-phase BDEs of 2,6-dimethoxyphenol (syringol) and 2,6-dihydroxyphenol (pyrogallol), the liquid-phase BDEs, the solvent, and substituent effects of phenols shows the usefulness of this approach. Seven solvents, differing in their H-bonding ability and polarity, n-heptane, benzene, acetone, acetonitrile, ethanol, methanol, and water, are used to model different environmental situations. Only the PCM model describes well the "bulk" solvent effects, which, depending on the E N T and/or R polarity parameter values of the solvent, modify the structure of the solute. Calculated liquid-phase BDEs are in close agreement with the experimental ones, where available, exceeding those in the gas-phase by as much as ca. 8 kcal/mol in some media. Solvent effects are common for catechol and phenol and different for guaiacol. Close agreement is derived between the theoretical and the experimental solvent effects for known phenolic antioxidants, namely, ubiquinols and flavonoids. The different ortho groups in catechol and guaiacol lead to different substituent effects in accordance with experimental findings.

The paper describes a DFT/B3LYP study, in the liquid phase, [using the PCM continuum model] on the O-H bond dissociation enthalpy (BDE) and ionization energy (IE) parameter values of the 2-monosubstituted phenols (2-X-ArOH), related to the H-atom transfer (HAT) and single-electron transfer (SET) mechanisms. The solvent and substituent effects on the conformers, the BDEs, and the IEs were studied using four electron-donating (EDG) and five electron-withdrawing (EWG) groups, in seven different solvents. In both the EDG- and/or EWG-substituted species of the parent compounds, radicals, and/or cation radicals, the most stable conformer is varied, depending on the medium and the substitution. The EWG-substituents increase IEs, resulting in a weaker antioxidant activity than the EDG ones; the effect appears stronger on the IEs than on BDEs. However, although the liquid-phase IEs, which are related to solution-phase oxidation potentials, decrease with the polarity and/or the hydrogen-bonding ability of the solvent, the opposite holds true for the BDEs, exhibiting a weaker effect. The gas-phase-calculated IE for benzene is among the most accurate ones in the field, compared to the experiment, that for phenol being the most accurate. In addition, calculated IEs for the 2-X-ArOH are in close agreement with the very few existing experimental ones. It is shown that the oxidation potentials are (a) highly correlated with the gas-phase ones, and (b) strongly solvent dependent. The stabilization/destabilization of the cation radical (SPC) contribution, in all media, is the decisive factor in the DeltaIE calculation. The reasonable correlations found between the DeltaBDE and DeltaIE could account well for the assumption of the simultaneous action of both mechanisms in the 2-X-ArOH, in both the gas and the liquid phase. It seems, however, that the presence of a particular solvent by itself is not sufficient enough for the HAT to SET transition. The involvement of specific ED and/or EW groups in the 2-X-ArOH seems also necessary. It appears that our theoretical approach is not only generally applicable to the set of substituents important to antioxidant activity but also useful in (a) the rational design of phenolic antioxidants and (b) affording accurate BDE and IE parameter values related to both possible antioxidant mechanisms.