Tautomerism in 1-phenylazo-4-naphthols: Experimental results vs quantum-chemical predictions

The reliability in the description of the tautomerism of 1-phenylazo-4-naphthol by using of HF and MP2 ab initio levels of theory and DFT methods with variety of pure GGA (OLYP), hybrid (B3LYP and B3PW91), long range corrected (LC-BLYP) and double-hybrid (B2PLYP and mPW2PLYP) functionals with large number of basis sets was estimated. In this evaluation three criteria were used: reproduction of the bond lengths in the structures of the individual tautomers, description of the non-planarity of the enol tautomer and prediction of the position of the tautomeric equilibrium (∆G value) at 298 K. The results show that in substantial number of cases HF reasonably covers all requirements. The tested pure (OLYP) and hybrid functionals (B3LYP and B3PW91) fail in the prediction of the position of the equilibrium independent on the basis set. The situation is slightly better at the long range corrected functional (LC-BLYP), which give predominance of the enol tautomer at 6-31+G** and D95++**. The double hybrid functionals give very good description with D95++** basis set, but at substantial computational costs.

► Most sensitive tautomeric fragment as bond lengths is determined by X-ray analysis.
► DFT methods very well describe the bond lengths at middle to large basis sets.
► HF in most of the cases is suitable for correct description of the tautomeric equilibrium.
► Only two DFT methods provide accurate prediction: B2PLYP and mPW2PLYP, at D95++**.