Photocatalytic processes have attracted significant interest since the 1970s and are still very active research areas today. The degradation of organic pollutants and the production of hydrogen and oxygen from water by using semiconductor photocatalysts such as TiO 2 , ZnO, and CdS have been extensively studied. However, few studies on photocatalytic reactions under visible light with catalysts such as gold nanoparticles (AuNPs) have been reported in the area of organic chemistry. Although fine gold particles have been used for centuries in stained glass windows, researchers have only recently discovered two of the most significant properties of AuNPs. Firstly, AuNPs can catalyze many reactions of organic compounds at elevated temperatures, including the oxidation of various substrates and the reduction of nitrobenzene. Secondly, AuNPs can strongly absorb visible light because of the surface plasmon resonance (SPR) effect. This effect is the collective oscillation of conduction electrons, which resonate with the electromagnetic field of the incident light to result in a significant enhancement of the local electromagnetic fields near the rough surfaces of the AuNPs. As a result of the SPR effect, light absorption for typical spherical AuNPs is generally observed between 520 and 550 nm, which can generate excited electrons in the AuNPs and also cause rapid heating. Together, these two properties offer an interesting hypothesis: the molecules on the AuNPs can be activated for reaction by their proximity to the heated AuNPs and their interaction with the oscillating electrons in the AuNPs. Thus, we may be able to drive reactions on the AuNPs by using visible light at ambient temperature. The AuNPs also exhibit significant ultraviolet UV absorption that causes an interband excitation of electrons from 5d to 6sp; this excitation may also be used to drive chemical reactions. Given that visible light (wavelength 400-700 nm) and UV irradiation (wavelength < 400 nm) constitute around 43 % and 4 % of the solar energy emitted by the sun, respectively, a photocatalytic process with supported AuNP photocatalysts has the potential to use sunlight to efficiently drive chemical reactions. Light absorption by AuNPs is also a local effect that is limited to the gold particles, so that the light heats only the AuNPs, which generally account for a few percent of the overall catalyst mass. [4-6, 16, 17] Therefore, such a process, if successfully realized, can be conducted at ambient temperature; thus this photocatalytic process can produce compounds that would have been unstable intermediates in a thermal reaction at high temperature.
Aromatic azo compounds are widely used in the production of dyes, food additives, and pharmaceutical products. Currently, the synthesis of these compounds is often conducted under high pressures and at high temperatures using transition-metal reducing agents. The by-products formed from the reducing agent are harmful to the environment. Recently, it was reported that aromatic azo compounds could be synthesized from the corresponding nitroaromatic compounds through a two-step, one-pot reaction with catalysts comprising AuNPs on TiO 2 or CeO 2 at 100 8C or higher. Firstly, nitrobenzene was over-reduced to aniline on the gold catalysts by hydrogen (9 bar). The reaction was continued by flushing out the H 2 and introducing O 2 at 5 bar and 100 8C to oxidize aniline to azobenzene. Azobenzene was formed as an intermediate in the thermal hydrogenation of nitrobenzene to form aniline, but, in this case, the azobenzene was unstable under the reaction conditions and was rapidly reduced to aniline. A subsequent oxidation step was thus required to produce the target azobenzene. As photocatalytic reactions are mostly conducted at ambient temperature and atmospheric pressure, many intermediates are stable under such conditions and would not react further. If the direct reduction of nitroaromatic compounds to their corresponding azo aromatic compounds can be realized by a photocatalytic process, the synthesis of aromatic azo compounds would be a much more controlled, simplified, and greener process.
To test the possibility of driving the reduction of nitrobenzene with light, AuNPs were supported on zirconia powder by reducing HAuCl 4 with NaBH 4 in the presence of ZrO 2 powder (full details of the synthesis and characterization are given in the Experimental Section). Photocatalysts with three gold loadings (1.5, 3.0, and 5.0 wt % of the overall catalyst mass) on the ZrO 2 powder were prepared. TEM images of these samples showed that the gold existed on ZrO 2