Theoretical simulations are performed for the vibrational spectra of selected isomers of OH−(H2O)3 and OH−(H2O)4 clusters to understand the origin of the difference in the experimental OH stretching fundamental spectra between these clusters reported by Robertson et al. [Science, 2003, 299, 1367]. Vibrational excitation energies and intensities are calculated by combining the adiabatic separation treatment of water and OH− units, and the least order inter unit coupling correction. Moreover, to directly simulate the shape of the experimental spectra, both the homogeneous and inhomogeneous widths of the spectra are calculated using on-the-fly quasi-classical trajectories and rotational constants information. Through these simulations, we show that the dominant isomer of OH−(H2O)4 should be the one with a second solvation shell water as suggested by Robertson et al. to explain the spectra in the 3200–3700 cm−1 range. In particular, rather than the peaks of the second shell water OH bands themselves, the peak corresponding to the weakly hydrogen bonded OH of the first solvation shell water is essential for the assignment of the dominant isomer. We also discuss the power law relation between the homogeneous width and the red-shift by the hydrogen bond, the limitation of B3LYP for the accurate description of strong hydrogen bonded OH peak positions, and the dependence of the inter unit coupling effects on the structure and the size of the clusters.
You have access to this article
Please wait while we load your content...
Something went wrong. Try again?