Calculations on phenol and a large number of phenols substituted with methyl, methoxyl, and amino groups have yielded reliable gas-phase O−H bond dissociation energies, BDE(ArO−H)^gas. Geometries for the phenol, ArOH, and aryloxyl radical, ArO, were optimized at the (semiempirical) AM1 level followed by single point density functional theory (DFT) calculations using a 6-31G basis set supplemented with p-functions on the hydrogen atom and the B3LYP density functional. This gave BDE(PhO−H)^gas = 86.46 kcal/mol, which is in good agreement with the experimental value of 87.3 ± 1.5 kcal/mol. All but one of the compounds and conformations examined had weaker O−H BDE's than phenol, the exception being o-methoxyphenol with the O−H group pointing toward this substituent (BDE = 87.8 kcal/mol). Where comparison was possible, calculated differences in O−H BDE's were in excellent agreement with experiment (better than 1 kcal/mol). A simple group additivity scheme also gave excellent agreement with calculated BDE (ArO−H)^gas values. Some potential new leads to phenolic antioxidants more active than vitamin E have been uncovered.