Temporal stability of magic-number metal clusters: beyond the shell closing model.

The anomalous stability of magic-number metal clusters has been associated with closed geometric and electronic shells and the opening of HOMO-LUMO gaps. Despite this enhanced stability, magic-number clusters are known to decay and react in the condensed phase to form other products. Improving our understanding of their decay mechanisms and developing strategies to control or eliminate cluster instability is a priority, to develop a more complete theory of their stability, to avoid studying mixtures of clusters produced by the decay of purified materials, and to enable technology development. Silver clusters are sufficiently reactive to facilitate the study of the ambient temporal stability of magic-number metal clusters and to begin to understand their decay mechanisms. Here, the solution phase stability of a series of silver:glutathione (Ag:SG) clusters was studied as a function of size, pH and chemical environment. Cluster stability was found to be a non-monotonic function of size. Electrophoretic separations showed that the dominant mechanism involved the redistribution of mass toward smaller sizes, where the products were almost exclusively previously known cluster sizes. Optical absorption spectra showed that the smaller clusters evolved toward the two most stable cluster sizes. The net surface charge was found to play an important role in cluster stabilization although charge screening had no effect on stability, contrary to DLVO theory. The decay mechanism was found to involve the loss of Ag(+) ions and silver glutathionates. Clusters could be stabilized by the addition of Ag(+) ions and destabilized by either the addition of glutathione or the removal of Ag(+) ions. Clusters were also found to be most stable in near neutral pH, where they had a net negative surface charge. These results provide new mechanistic insights into the control of post-synthesis stability and chemical decay of magic-number metal clusters, which could be used to develop design principles for synthesizing specific cluster species.