Zusammenfassung
In this thesis, nano-design has been used as a powerful tool in order to obtain materials with very specific properties and functions. Catalysis by gold was identified as investigation field because perhaps, it represents the best example in which catalytic properties are strongly dependent on the metal size. Thus, the proper tuning of any parameter in this length-scale should lead to a wide range of possible materials and, consequently, catalytic results. Furthermore, because of the easy functionalization and tuning, gold colloids were chosen as starting material in all the preparations.
With the aim of studying the size dependence of catalysts performance and in order to prove that nano-design can be applied to a broad range of situations, different targets were selected. First of all, gold based catalysts were prepared via immobilization of colloids on different supports in order to study the factors controlling activity in the oxidation of carbon monoxide. Afterwards, selected gold catalysts prepared with the above mentioned route were tested in the methanol partial oxidation. This reaction was selected because of the interesting implication on the hydrogen production for fuel cells.
Finally gold colloids were used for the fabrication of a novel thermostable catalysts. This is basically a system in which gold nanoparticles are entrapped in a porous matrix characterized by pores narrower than the size of the metal clusters. Thus, these particles are not free to meet and sinter, while reactants due to the smaller size can easily reach the active center, react and retrodiffuse.
It has been found that by using the colloidal deposition method a series of very active and reproducible gold catalysts for the low temperature CO oxidation can be synthesized. The activity measured for a Au/TiO2 was as high as that of the best material prepared via deposition-precipitation. Furthermore, this catalysts show a much better thermal stability than the analogous ones prepared via conventional routes (i.e. deposition-precipitation and impregnation). Using different protecting agents, different gold particle size distributions were obtained on the same support. The obtained data confirm the widely spread notion in the literature, that the activity for CO oxidation is strongly dependent on the gold particle size. The most important finding of this study is the fact that almost identical gold particle size distributions on different supports result in different activities for CO oxidation. This clearly demonstrates that the metal support interaction influences the catalytic properties of the final materials. However, this influence does not follow the reducibility and oxygen activation ability of the supports. Active catalysts were prepared based on titania (supposedly active) and alumina (supposedly inactive), while much less active samples were synthesized with ZnO (supposedly active) and zirconia (supposedly inactive). However, although the gold particle sizes are not changed by the deposition, the support influences the shape of the deposited particles, leading to faceting and possibly creating defect sites. The exact mechanism how the different supports influence the activity is still unclear, but different faceting may be related to the catalysts activity.
Moreover, the colloidal deposition method was successfully employed for the preparation of highly active catalysts for the production of hydrogen by partial oxidation of methanol. The activity of the prepared materials was strongly dependant on the support used. Very low reaction light off temperature (i.e. 80 °C) and high hydrogen yield (ca. 75 % at 440 °C) were observed when gold was supported on CeO2. Remarkably, this material was stable for 100 h under a stream composed of 20 vol % of methanol (MeOH : O2 = 1 : 1 mol mol-1) at a space velocity as high as 220000 cm3 gcat-1 h-1. From an application point of view, it is also interesting that under these conditions the reaction operates autothermally. However, reaction catalyzed by Au/CeO2 was extremely difficult to study, since it proceeds oscillatory over a wide range of experimental conditions. Due to this reason, the activity of gold on titania and gold on zinc oxide catalysts prepared using the same route was investigated. Among them, Au/ZnO resulted in the most promising material. In fact, a hydrogen yield close to 100 % could be achieved, at a temperature as low as 360 °C.
Finally, a new core-shell Au, @ZrO2 catalyst has been built as proof of the concept. In this material, gold nanoparticles are incorporated into mesoporous zirconium oxide shells. A very well controlled and reproducible synthesis procedure has been developed, which allows obtaining one gold particle per shell. Gold nanoparticles are thus isolated in a porous matrix characterized by pores which are smaller than the metal nanoparticle. Consequently, even if gold particles are free to move inside the porous cage, they can not migrate outside and meet other particles. Sintering via particle coalescence can thus be avoided. The produced system has been then tested in the CO oxidation and compared to a benchmark obtained by destruction of the shells. The results obtained indicate that isolation of metal catalyst particles by hollow sphere encapsulation indeed allows stabilization of the catalyst against sintering.
