Metal Clusters: Between Atom and Bulk

Abstract

The basic question of metal cluster research is the following: How do the macroscopic metallic properties – conductivity, optical lustre, chemical and catalytical reactivity, magnetism, malleability and ductility – evolve as a function of cluster size starting from the electronic states of the atom? It turns out that in a cluster of 20-40 atoms most of what we acknowledge as «metallic» is already fully developed.This is the reason why the beautiful chemistry of metal cluster complexes has not contributed to an understanding of the transitionfrom the atom to the bulk: The chemical interaction of the ligands (e.g. CO, C₅H₅, PR₃) with the core of metal atoms dominates electronic and structural properties to the extent that the underlying laws of metallic conduct are completely masked. We have, instead, to investigate bare metal clusters. This is, however, impossible with the traditional tools of inorganic chemistry. High temperature methods, molecular supersonic beam technology, manipulation in vacuum systems, laser spectroscopy, mass spectrometry, and low (and high) temperature matrix isolation experiments are necessary. Furthermore guidance of the experimental work by quantumchemical calculations, molecular dynamics simulations, and the combination of the two are important because «chemical intuition» does not lead far in a field where chemical thinking has barely begun to develop: The molecular science of metals! – In this article we present the experimental environment of metal cluster work and then concentrate on the results available: stability of cluster sizes, selectivity of the heterometallic bond, electronic and magnetic properties from the atom to the bulk, insulator-metal transitions, size-dependent chemical and catalytic properties. Global, quantumchemical and ab initio molecular dynamics models are discussed. This leads to the last question of chemical interest: Why do metal clusters not have shape, or if they do, how can it be determined?