Gold colloids: from quasi-homogeneous to heterogeneous catalytic systems.

Ruby red colloids of gold have been used for thousands of years and in the past have attracted much attention due to their optical properties. Surface plasmon resonance (SPR) bands are responsible for gold colloid colors and typically appear for nanometer-sized gold nanoparticles (GNPs). These lie in the visible range and their position (and intensity) depends on the size, distribution of size, and shape of GNPs but also on their interaction with other materials (i.e., support). Scientists consider colloids as quasi-homogeneous systems, but because of their intrinsic thermodynamic instability, they need different capping agents providing sufficient stability. The strength and the nature of the interaction between the protective (or functionalizing) molecule and the GNP surface is of utmost importance. It can determine the catalytic properties of the nanoparticles, as they mainly interact with the active sites, thus interfering with reactant. Therefore, the protective layer should contribute to the colloid stability, but at the same time, it should not be irreversibly adsorbed on the active site of the GNP surface providing convenient accessibility to reactant. From a catalytic point of view, the milder the interaction is between the particle surface and the capping agent, the more the activity increases. Unfortunately, the reaction conditions often do not allow the required stability of GNPs, which constitutes a fundamental prerequisite for stable catalytic activity. Anchoring GNPs on suitable supports can circumvent the problem, and this technique is now considered a valuable alternative to classical methods to produce highly dispersed gold catalysts. In this Account, we describe the advantages in using this technique to produce gold heterogeneous catalysts of high metal dispersion on a large variety of supports with the possibility of tuning to a large extent the size and (even partially) the shape of GNPs. We also review our recent progress on the sol-immobilization technique. Specifically, we highlight how, depending on its nature, the protective agent not only mediates the activity of GNPs in alcohol oxidation process but also actively participates in the anchoring process and to the stability of GNPs depending on the support surface. We can also use the modification of the metal surface operated by the capping agent to prepare bimetallic species and influence the surface potential, which modifies the intrinsic activity of the GNP. In conclusion, this technique implies many contributions (sometimes not yet clarified factors) that are not simply concerning dimension and dispersion of GNPs or type of support. Chemists should make careful selection of the protective agent and reaction parameters depending on which support is used in which reaction.