Gold Atoms Stabilized on Various Supports Catalyze the Water–Gas Shift Reaction

Abstract

For important chemical reactions that are catalyzed by single-site metal centers, such as the water-gas shift (WGS) reaction that converts carbon monoxide and water to hydrogen and carbon dioxide, atomically dispersed supported metal catalysts offer maximum atom efficiency. Researchers have found that for platinum metal supported on ceria and doped ceria in the automobile exhaust catalyst, atomic Pt-Ox-Ce species are the active WGS reaction sites. More recently, preparations of gold at the nanoscale have shown that this relatively "new material" is an active and often more selective catalyst than platinum for a variety of reactions, including the WGS reaction. The activity of gold is typically attributed to a size effect, while the interface of gold with the support has also been reported as important for oxidation reactions, but exactly how this comes about has not been probed satisfactorily. Typical supported metal catalysts prepared by traditional techniques have a heterogeneous population of particles, nanoclusters, subnanometer species, and isolated atoms/ions on the support surfaces, making the identification of the active sites difficult. Both we and other researchers have clearly shown that gold nanoparticles are spectator species in the WGS reaction. Evidence has now amassed that the gold active site for the WGS reaction is atomic, that is, Au-Ox species catalyze the reaction, similar to Pt-Ox. In this Account, we review the relevant literature to conclude that the intrinsic activity of the Au-Ox(OH)-S site, where S is a support, is the same for any S. The support effect is indirect, through its carrying (or binding) capacity for the active sites. Destabilization of the gold under reducing conditions through the formation of clusters and nanoparticles is accompanied by a measurable activity loss. Therefore, it is necessary to investigate the destabilizing effect of different reaction gas mixtures on the gold atom sites and to consider regeneration methods that effectively redisperse the gold clusters into atoms. For gold catalysts, we can remove weakly bound clusters and nanoparticles from certain supports by leaching techniques. Because of this, we can prepare a uniform dispersion of gold atoms/ions strongly bound to the support surface by this two-step (loading followed by leaching) approach. Presently, one-step preparation methods to maximize the number of the single atom sites on various supports need to be developed, specific to the type of the selected support. Often, it will be beneficial to alter the surface properties of the support to enhance metal ion anchoring, for example, by shape and size control of the support or by the use of light-assisted deposition and anchoring of the metal on photoresponsive supports. Because of their importance for practical catalyst development, synthesis methods are discussed at some length in this Account.