Pure Appl. Chem., Vol. 75, No. 4, pp. 461–481, 2003. © 2003 IUPAC 461 Organometallic chemistry at the nanoscale. Dendrimers for redox processes and catalysis* Didier Astruc Laboratoire de Chimie Organique et Organométallique, UMR CNRS No 5802, Université Bordeaux I, 33405 Talence Cedex, France Abstract: An overview of the metal-mediated synthesis and use of nanosized metalloden- drimers is given with emphasis on electron-transfer processes (molecular batteries consisting in dendrimers decorated with a large number of equivalent redox-active centers) and catalytic reactions (electron-transfer-chain catalytic synthesis of dendrimers decorated with ruthenium carbonyl clusters, redox catalysis of nitrate and nitrite electroreduction in water by star-shape hexanuclear catalysts). INTRODUCTION Although most rules of stoichiometric and catalytic organometallic activation are now known [1], there remains a wide-open field of research for development and applications using the exploration of the best metal-ligand combination toward a given reaction. Present examples can now be found in metathesis that remained restricted to nonfunctional olefins for a long time until the recent successful efforts of Grubbs’ [2] and Schrock’s groups [3]. Their commercial catalysts are now of everyday use for organic and polymer chemists. A previous success story of this kind started in 1970 when Kagan first published efficient asymmetric catalysis with the powerful concept of chiral chelating phosphine [4]. This idea led many chemists, including other prestigious ones, to find optically active drugs with close to 100 % ee, an essential requirement for safe use by the public [5]. What now remains to be done? Indeed, the task is enormous for 21 st century organometallic and molecular chemists. Mononuclear homogeneous catalysts [6] cannot be easily removed, thus they con- taminate products and cannot be re-used. Heterogeneous catalysts often have poor selectivity, their mechanism of action is not well defined, and they also suffer from damage from one batch to the next [7]. Finally, supported catalysts do not always work as well or in the same fashion as homogeneous analogs, their distribution in the polymer is not well defined, they suffer from leaching, and by no means represent a definitive solution as witnessed by the poor extension to the industrial level [8]. An analo- gous reasoning can be made in other fields such as molecular electronics and sensors, in which the ab- sence of impurities can be crucial. Organometallic chemistry is a fantastic tool to activate molecules, make bonds and introduce functions, but at the same time many challenges remain in order to fill the gaps toward applications. Therefore, we have engaged in a large research program aimed at carrying organometallic chem- istry at the nanoscale level. The goals are to design nanoscale organometallic catalysts, sensors, and components for molecular electronics that will not only be easily separated from other products, but whose molecular definition should be as precise as that of monometallic compounds and yet be ad- dressable by macrocomponents such as monodisperse polymers, electrodes, and surfaces. In the pres- ent review, we provide a few examples in this direction. Dendrimers, a young field of molecular chem- istry [8,9], are the first perfectly defined synthetic macromolecules (i.e., of polydispersity 1.0) that are * Plenary lecture presented at the XX th International Conference on Organometallic Chemistry (ICOMC), Corfu, Greece, 7–12 July 2002. Other presentations are published in this issue, pp. 421–494.