NOVEL APPROACHES TOWARDS THE OPTIMISATION OF METAL NANOPARTICLE BASED CATALYSTS

The main goal of my Ph.D. project has been the development of novel approaches for the optimisation of supported noble metal nanoparticles, which are well-established catalysts for liquid phase oxidation reactions. The partial oxidation of oxygen-containing compounds (alcohols, aldehydes, carbohydrates) is a profitable process, the corresponding products (aldehydes, ketones, epoxides, carboxylic acids, esters and lactones) being key intermediates in the synthesis of fine chemicals and commodity. In the perspective of biomass valorisation these processes are recently assuming an increasing relevance. Indeed, many biomass-derived platform molecules contain oxidizable functional groups and therefore can be easily converted in value-added compounds by oxidation. The oxidation of organic compounds can be carried out in the gas phase through continuous-flow reactors using air or oxygen as oxidant. Nevertheless, these processes require high temperatures and their application is restricted to volatile and thermally stable reactants and products. From this point of view working in the liquid phase seems to be more suitable for energy saving, since milder conditions can be adopted, compared to gas phase. The main drawback of the current industrial technologies for liquid phase oxidation processes is the use of stoichiometric inorganic oxidants, such as dichromate and permanganate, which are toxic and corrosive. The employment of these reactants therefore entails environmental issues (production of high volumes of toxic wastes), handling difficulties and reactor maintenance problems (corrosion, plating out on reactor walls). According to green chemistry principles, the replacement of toxic stoichiometric processes with catalytic and environmentally benign routes is then heartily recommended. Noble metal unsupported and supported nanoparticles have been extensively explored as heterogeneous catalysts for the liquid phase oxidation of oxygen-containing organic compounds in the presence of molecular oxygen, air or hydrogen peroxide as sole oxidants. In particular platinum group metals (Pt, Pd, Ru, Rh) have shown to be able to oxidize alcohols to the corresponding carbonyl or carboxylic compounds under mild conditions (close to ambient conditions) . However, these systems rapidly undergo deactivation by over-oxidation or metal dissolution into solution (leaching). Otherwise, nano-sized gold exhibits a remarkable activity and it possesses unexpected advantages over platinum group metals in terms of selectivity control and resistance to deactivation. The strongest limitation in using gold NPs as catalysts is the compulsory use of a basic environment . Recent studies showed that alloying gold with a second metal (platinum or palladium) is possible to obtain effective catalytic systems in terms of activity, durability and selectivity even in the absence of a base. Besides the use of bi- or multimetallic systems (e.g. AuPt or AuPd), the catalytic performances are strongly affected by many factors, including the addition of promoters (e.g. Bi), the influence of support and the preparation route. A simultaneous fine tuning of all these parameters is not a straightforward task, therefore the design of catalysts is a still challenging research target. A multidisciplinary approach seems to be the better strategy for facing this challenge. In this view, the development of a catalyst should be the result of the combination of three main aspects (preparation, characterization, testing), which can be investigated on different levels (from atomic to macroscopic level) and with several tools (from in situ characterization to computational modelling). During my Ph.D. project a similar approach has been adopted. In the first section (Chapter 2) my research focused on the possible strategies for tuning the selectivity in base-free glycerol oxidation, a reaction of industrial interest, which has attracted significant attention in the last decades, due to the need for the valorisation of this bio-platform molecule. The reaction pathway of glycerol oxidation is complex and leads to a large number of valuable organic compounds (glyceric acid, tartronic acid, dihydroxyacetone, lactic acid, etc.). Therefore directing the reaction to the desired target product represents a key-point. Two main features were investigated in the detail: the role of support and the addition of promoters. It has been observed that the support greatly affected the selectivity of AuPt based catalysts in base-free glycerol oxidation. In particular, using an acidic support, H-Mordenite, an enhancement in the selectivity to C3 products (glyceric and tartronic acid) has been obtained.4a Also basic supports such as hydroxyapatite and MgO have been shown to be useful supports.4b In order to investigate more in the detail the effect of support acidity, I extended these studies to a series of supports with different acid-base properties, namely H-Mordenite, SiO2, MCM-41, sulfated ZrO2, Activated Carbon (AC X40S) as representative of acidic supports, and MgO and NiO as references for basic supports. Acidic surface properties of these materials have been full characterised by means of Infrared Spectroscopy and microcalorimetry, and their influence on the catalytic behaviour of alloyed AuPt nanoparticles was highlighted. An high selectivity to C3 compounds was observed, using acidic supports, glyceric acid and glyceraldehyde being the main products. Both these compounds are obtained by the oxidation of the primary hydroxyl function, on the other hand the main product of secondary hydroxyl group oxidation, dihydroxyacetone (DHA), is economically the most interesting oxidation product due to its use as tanning agent in cosmetic industries. In the literature it has been reported that the addition of Bi as promoter for Pt catalysts enhances the yield of DHA. On the other hand Bi-Pt catalysts suffer from heavy deactivation during reaction, due to the leaching of metals. On the basis of these considerations, we decided to modify AuPt and AuPd alloyed catalysts with Bi to investigate the effect of the addiction of this promoter, not only in terms of selectivity to DHA, but also for the durability of the catalyst. In the second part (Chapter 3) Operando Attenuated Total Reflectance Infrared (ATR-IR) spectroscopy and catalytic batch reactor experiments were performed in parallel to elucidate the different catalytic performance of Au, Pd, and AuPd supported on TiO2 and Al2O3 in the liquid-phase oxidation of benzyl alcohol. In particular the development of different surface species and the role of the protective agent (polyvinyl alcohol, PVA) in catalysts prepared by a sol immobilization route were examined. Finally, in the last part of the thesis (Chapter 4) a periodic Density Functional Theory (DFT) study of the adsorption and activation of ethanol on different surfaces (13 atom Au cluster, oxygenated 13 atom Au cluster, TiO2 rutile surface and Au ribbon on TiO2 surface) is proposed, in order to unravel the presence of preferential sites for the adsorption of alcohol on catalyst surfaces. Simulations were carried out using the plane wave basis set code VASP and the Perdew, Burke and Ernzerhof (PBE) functional.