Oxidation Reactions Catalyzed by Atomically Well-Defined Gold Clusters

National @ +CLA:ADM:ATUInternational audienceGold nanoparticles have demonstrated interesting catalytic activity in many chemical processes.[1] However, fundamental investigations on the structure-catalytic activity relationships still lag behind, partly due to the polydispesity issues of gold nanoparticles. Sub-2 nm gold clusters have attracted more attention since atomically well-defined and monodisperse thiolate capped ones have been successfully isolated.[2] These clusters, with Aun(SR)m formula, are formed by a precise number n of gold atoms comprised between ~10 and few hundred and stabilized by m thiolate molecules. What renders these clusters unique is the discrete electronic structure of the gold core due to quantum size effects, leading to original properties intermediate between those of molecular and bulk states. In addition, the total structure determination by single crystal X-ray crystallography permits to know exactly the position of the atoms forming the gold core and the organic layer (Fig 1).[3] More importantly, their catalytic performance have been demonstrated, thus, gold nanoclusters hold promises as a new generation of catalysts which permit in-depth studies on the subtle correlation of structure and catalytic activity.[4] In this communication we investigate the influence of the number of gold atoms, the type of ligand at the surface and also the support effect on different aerobic oxidation reactions. 2 Results and discussion In a first part, we will present the synthesis of clusters made of different thiolate ligands, such as 2- phenylethanethiol (HSC2H4Ph) or functionalized thiophenol (HSPh-X with X = H, NH2 or CO2H). Two synthetic approaches are used to get pure atomically well-defined thiolate gold clusters, which are by exchange reaction from Aun(SC2H4Ph)m or by direct synthesis starting with HAuCl4. To obtain the exact Aun(SR)m formula of the clusters, electrospray mass spectrometry is a technique of choice to prevent their fragmentation. In addition all the synthesized clusters are characterized by small and large angle X-ray scattering, X-ray photoelectron, infra-red and UVvisible spectroscopies, thermogravimetric and elemental analyses. Among those clusters, an unprecedented Au25(SPhNH2)17 cluster has been isolated and the different characterizations in solid and liquid phases show that the 18th position is occupied by an amine or DMF molecules to close its electronic shell.[5] This well-defined cluster is really appealing for further experiments, because it offers possibilities of post-functionalization or deposition on substrates by coupling with the amino groups. Once synthesized, the clusters have been deposited by impregnation on silica supports, such as the mesoporous SBA-15 and the hydrophobic ~16 nm SiO2 particles. These choices have been made to enhance the dispersability of the so-prepared catalysts in the solvents (methylcyclohexane and toluene) and be able to compare them directly with more commonly used supported gold catalysts. We studied the catalytic activity of the gold nanoclusters in two different aerobic oxidation catalytic tests: the epoxidation of t-stilbene and the oxidation of benzyl acohol to form benzaldehyde. The first observation for trans-stilbene epoxidation is that the bare Au25(SCH2CH2Ph)18 and Au25(SPhNH2)17 are precipitating, resulting in a lack of accessibility on the gold catalyst. This is why these nanoclusters are deposited on hydrophobic silica (Amorphous silica from Degusa, AerosilŽ R972). This way, the dispersion is enhanced and the high selectivity of 97 % obtained is, to our knowledge, the best reported up to now. For comparison the catalytic test carried out in the same conditions with the reference gold nanoparticles from the World Gold Council (3.5 ą 0.9 nm) deposited on TiO2 gives a selectivity of 75 %. This observation shows two major points (i) the hybrid clusters are chemically stable for catalytic oxidation reaction, (ii) the presence of the ligands around the clusters does not prevent the catalytic reaction and may even be responsible for such high selectivity. This conclusion is supported by the comparison between the activity of two Au25 clusters surrounded by different thiolate molecules: Au25(SCH2CH2Ph)18 and Au25(SPhNH2)17 . Both show a selectivity for the formation of the epoxide > 97 %, but the activity is much higher for clusters surrounded by phenylethanethiolate that aminothiophenol. This may be due to the affinity of the gold nanoclusters with the support or to the flexibility of SCH2CH2Ph, which would lead to a better access of the substrates to the gold atoms. At the opposite when the clusters are deposited inside hydrophilic mesoporous silica (SBA-15) for oxidation of benzyl alcohol, there is no activity. The ligands have to be burned at 300°C to observe the formation of benzaldehyde. This observation shows a third point (iii) the catalytic activity of gold nanoclusters is dependent of the presence of the ligands as for the type of reaction. Fig. 2. Au25(SPhNH2)17 nanoclusters used as catalyst for aerobic t-stilbene epoxidation. 3 Conclusions We investigated the catalytic properties in aerobic oxidations of sub-2 nm gold compounds. By comparing the results for the different gold clusters and supported nanoparticules, we pointed out that while still being functionalized; gold nanoclusters are active in catalytic oxidation of trans-stilbene. For the oxidation of benzyl acohol, we showed that the ligand presence is inhibiting the catalytic activity. This ligand effect present with functionalized gold clusters makes possible to tune the overall catalytic selectivity and specificity to have chimio specific reaction on multi-modal sites. We showed here that thiolated gold nanoclusters compounds hold great interests and promises as a new class of catalysts thanks to their well-defined structure and to the influence of the ligands in the catalysis. References 1 A. S. K. Hashmi; G. J. Hutchings, Angew. Chem.-Int. Edit. 45 (2006) 7896. 2 R. Jin; Y. Zhu, H. Qian, Chem. Eur. J. 17 (2011) 6584; T. Tsukuda, Bull. Chem. Soc. Jpn 85 (2012) 151. 3 M. Zhu; W. T. Eckenhoff; T. Pintauer; R. Jin, J. Phys. Chem. C 112 (2008) 14221. 4 Y. Zhu, H. Qian, R. Jin, J. Mater. Chem. 21 (2011) 6793; J. Oliver-Meseguer; J. R. Cabrero-Antonino; I. Dominguez; A. Leyva-Perez; A. Corma Science 338 (2012) 1452. 5 C. Lavenn, F. Albrieux, G. Bergeret, R. Chiriac, A. Tuel, A. Demessence, Nanoscal