Calculations ab initio of the potential surfaces and geometry of nonrigid molecules III. The nonrigid complex molecule NaBH4

Abstract

The study of the structure of nonrigid molecules (nonrigid molecules are understood to be systems for which the potential surfaces have either fiat minima or a few or many minima having the same or similar energies, corresponding to different nuclear configurations and separated by small or zero barriers; a characteristic feature of these molecules is the existence of large mean-square amplitudes of the vibrations) has recently been attracting increasing attention from experimental and theoretical workers. The reasons for this interest are on the one hand the increased level of experimental work, in which it has now become difficult to ignore the dynamic structure of molecules, and on the other hand the increased possibilities provided by quantum-chemical methods. The theoretical study of the problem consists of two parts: the construction of sufficiently complete and accurate potential surfaces of the systems, and examination of the dynamics of the system. The present series of papers is devoted to a study of the first part of the problem, which is being carried out by quantum-chemical calculations using MO LCAO SCF methods. What information can be given by a sufficiently correct quantum-chemical calculation of potential surfaces ? 1. The positions of the absolute and relative minima of the system and the differences in their energies. 2. The minimum energy path (MEP) for the migration of a mobile atom (or group of atoms) relative to the remaining ("stationary") part of the molecule. 3. The profile of the MEP, the height and shape of its barriers, the curvature of the walls of the "valley" along the MEP, etc. The construction of the potential surface is a necessary preliminary step for the subsequent examination of the dynamic problem, in which not only the energetic but also the momentum and other characteristics are important, and where the optimum migration path may not coincide with the MEP. This Luformation gives less unambiguous evidence for or against conclusions regarding the "nonrigidity" of the bonding in a given clas~ of molecules, compared with the simple experimental fact that the amplitudes of the vibrations are large. Not only the MEP, but also the wave functions and the nature of the electron density distribution in the nonrigid systems are important. Up to the present it is still not completely clear what "polytopic, bonding essentially involves, what happens to it when a cation moves relative to an anion, how extensive is the polarization of the electron density of the anion and cations, etc.