Calculations have been performed on 10 structures of the cluster H+(H20)5. It is shown that the most stable ones are an open (Eigen) and a cyclic four-membered-ring structure very close in energy and possibly degenerate. This can explain that different structures were proposed by experimentalists. The easy evolution of some structures into others is likely related to the nature of the first solvation shell in larger clusters or solutions. Vibrational frequencies, useful to interpret experimental data, are computed for the two most stable structures.
The Problem: Structure of H+(HzO)s and the First Solvation ShellIn 1954, Eigen et al.' proposed as a hydration model for the proton in aqueous solution and ice an oxonium ion H30+ surrounded by three water molecules in a first solvation shell. In some early theoretical work (1956) the presence of a fourth water molecule in the first solvation shell has also been suggested;* three water molecules are hydrogen bonded with the three hydrogen atoms of the ion, while the fourth water molecule is located above the oxygen atom (Figure 1, protondonor structure 1). Since then, there were many proposal^^-^^ on the structure and coordination number, and presently the issue is not settled, neither theoretically nor experimentally. Some of these studies are concemed with the cluster H30+(H20)4, others with larger systems.The first attempt by Newton et aL3 in 1971 to study such systems with ab initio quantum mechanical calculations was restricted to small hydrates involving, at best, four water molecules and the oxonium ion. Using a 4-31G basis set at the Hartree-Fock level, the authors optimized a few structures for the system &O+(H*0)3 and then added to it a fourth molecule. Within these limitations the most stable structure has three water molecules in the Eigen-like structure first shell and a fourth water molecule in the second shell, hydrogen bonded to one water molecule of the first shell with an 0. * 0 distance arbitrarily chosen (Figure 1, structure 2). The interpretation of an infrared spectrum published some time later is based on such a structure.8 However, two experimental paper^^.^ suggested that the fist solvation shell has four water molecules. In particular, from X-ray and thermal neutron studies of hydrochloric acid solutions at 20 "C, Triolo et a1.6 proposed a charge-dipole complex (Figure 1, structure 3). In a further work, ' Newton (1977) extended his studies to structures 1 and 3 in order to check this assumption. He found that neither of these two structures were stabilized with respect to H30+(H20)3 f Hz0, with a more favorable situation for the charge-dipole complex structure 3 than for the hydrogen bonded structure 1. A complete optimization of these structures with ab initio calculations was unfortunately not possible at the time, and the true minima might have been missed. Furthermore, it must be pointed out that the structure of the first shell may be different 'Abstract published in Advance ACS Abstracts, April 15, 1995.