Nature of transition-metal cluster structures

Abstract

The theory of cluster structure presented in this article, which is based on a fundamental classification of ligands as having weak or strong fields, has a sufficiently general character. The concepts developed permit reliable interpretation of the most important experimentally established patterns of cluster-compound structure and prediction of the energy and stereochemical characteristics of different cluster groupings. Thus, it is possible for the first time to explain why cluster structures are generally unstable for metals at the beginning of the transition series, while cluster structures for transition metals in the central groups are very stable only for lower compounds with weak-field ligands (halogens, carboxylic acids, etc.) and cluster structures occur for transition metals at the end of the period only for formally low-valence compounds with strong-field ligands (CO, C5H5, etc.). Moreover, it is possible to give a concise interpretation of the characteristic tendency toward formation of metal-metal bonds of different multiplicities (extending as far as quadruple dicentric M-M bonds) in chloride and carboxyl clusters and to explain the redominantly single σ-character of the M-M bonds in carbonyl clusters, where one of the main factors limiting formation of multiple M-M bonds is the destabilizing nature of the metal-metal π interaction. Finally, our concepts of the relative energy of the filled metallic molecular orbitals enable us to explain why the energy of two-electron M-M bonds in carbonyl clusters is materially reduced (by a factor of almost 2) in cornparison with that in chloride clusters and thus give us an understanding of the unusual strength of halide clusters (and clusters containing weakfield ligands in general) and the exceptional diversity of forms in which they exist.