FLUORESCENCE-DIP INFRARED-SPECTROSCOPY OF THE TROPOLONE-H2O COMPLEX

Fluorescence dip infrared spectroscopy (FDIRS) is used to probe the ef fect of a solvent water molecule on intramolecular H-atom tunneling in tropolone. As with the bare molecule discussed in paper I, the FDIR s pectrum of the tropolone-H2O complex is recorded in the O-H and C-H st retch regions. Three OH stretch fundamentals are observed in the spect rum, and can be assigned nominally to a free OH stretch of the;water m olecule (3724 cm(-1)), a hydrogen bonded OH stretch of water (3506 cm( -1)), and the OH stretch of tropolone (similar to 3150 cm(-1)). The br eadth and complexity of the bands is highly mode specific. The free OH stretch transition is sharp (1.8 cm(-1) FWHM) and has weak combinatio n bands built on it at +73 and +1600 cm(-1). The former is assigned to a combination band with the in-plane bending mode of the tropolone-H2 O hydrogen bond, while the latter is the free OH/intramolecular water bend combination band. The water hydrogen-bonded OH fundamental is als o a sharp transition which, after correction for the decreased infrare d power at its frequency, is clearly the strongest transition in the s pectrum. It is flanked by three close-lying satellite bands 13, 23, an d 34 cm(-1) above it, and also supports a weak combination band at +69 cm(-1) due to the in-plane intermolecular bending mode. The tropolone OH absorption is in the same frequency region as in the bare molecule ,but broadened to over 10 cm(-1) in TrOH-H2O. Distinct substructure in the band is present, with spacings reminiscent of those in the water H-bonded OH stretch region. Ab initio calculations on tropolone-H2O ar e carried out at both the MP2 and Becke3LYP levels of theory. Two isom ers with similar binding energies and vibrational frequencies are iden tified. In one isomer (isomer I), the water molecule serves as a hydro gen-bonded bridge between the tropolone OH and keto groups. In the oth er (isomer II), the water molecule is exterior to the tropolone and hy drogen bonded to the keto oxygen. The experimental evidence does not c onclusively distinguish between these two possibilities, though the ex terior structure seems somewhat more in keeping with the data as a who le. (C) 1996 American Institute of Physics.