Nature uses metal ± ion coordination complexes to undertake many tasks. Frequently it imparts function by modifying the conventional behavior of a metal complex by incorporation within a protein superstructure. [1] For example, by isolating heme groups in the center of globular proteins oxidative dimerization is prevented and reversible oxygen binding is achieved. Steric isolation owes much to the dimensions of such globular proteins which are much larger than the small molecules prepared by synthetic chemists. Dendrimers represent a class of monodisperse synthetic macromolecules [2] which approach the dimensions of proteins and have been proposed as potential protein and enzyme mimics. [3] For example, porphyrins have been encapsulated at the center of spherical organic dendrimers to create compounds analogous to globular electron-transfer proteins [4] and then used as oxidation catalysts. [5] However, in nature, coordination sites are often not completely encapsulated in the core of a protein. For example, in carbonic anhydrase (CA) the N 3 zinc(ii)-coordination site is situated at the foot of a cleft accessible to the solvent. The essentially hydrophobic protein cleft plays an important role in the catalytic cycle, by binding the substrate, creating a hydrogen-bonded solvent network to facilitate proton transfer, and expelling the hydrophilic product. [6] Equally importantly this cleft modifies and controls the chemistry of the coordination site. Investigations of zinc(ii) species with tripodal N 3 ligands (L) reveal a strong tendency to form [ZnL 2 ] 2 complexes or hydroxy-bridged dimers. [7,8] In the enzyme the protein cavity sterically prevents the formation of these species and discrete mononuclear 1:1 complexes [ZnL(OH)] and [ZnL(OH 2 )] 2 result. We were interested to see if we could imitate this natural approach to modify and control the behavior of a transition metal complex. Rather than simply isolating a complex within a globular dendritic structure we have focused on the design of systems in which a dendritic unit could be used to generate a cavity around a metal complex.
We reasoned that blocking one direction of ªgrowthº of a dendrimer would lead to a large cleft in the structure around the site of the ªblockageº. If the metal binding site forms the blocking group, or is attached to it, then at high generation