Development of advanced mesostructured catalytic coatings on different substrates for fine chemical synthesis

Abstract

Catalytic microstructured reactors are becoming widely recognized for their unique properties, such as high surface–to–volume ratios, isothermal conditions due to high heat transfer rates, enhanced safety, and potential applications in chemistry and in chemical industry. The efficient use of microreactors requires shaping of the catalyst usually by deposition of thin catalytic films on microchannel walls. Thin films with large internal surface area can considerably enhance the catalytic surface in the absence of internal mass transfer limitations in the porous structure. As the diffusion of large organic molecules in the porous network is much slower as compared to that observed in gas phase reactions, the maximum catalytic film thickness to avoid internal mass transfer limitations is estimated to be ca. 3 µm. Mesoporous titania thin films doped with metal nanoparticles attract considerable attention in catalysis due to their large surface areas, narrow pore (and particle) size distribution, and favorable surface properties. To obtain the mesostructured materials with required properties it is important to control accurately the synthesis conditions during the preparation of mesoporous catalytic films. Mesoporous films doped with metal nanoparticles were successfully used in a number of reactions, such as reduction of hydrogen peroxide, hydrogenation of acetylene alcohols and ??,??-unsaturated aldehydes and others. Due to a strong metal support interaction between catalytic nanoparticles (Au, Ag, Pd, Pt) and the support, mesostructured titania can prevent metal nanoparticles aggregation and thus improve catalyst stability. This thesis is devoted to preparation of noble metal doped titania films, deposited on the flat plates and within the fused silica microcapillaries. The mesoporous titania coatings were synthesized via the sol–gel method combined with evaporation induced self assembly, which is based on the self–assembling of the structures in the presence of ionic surfactants or block–copolymers. The catalytic films were obtained via direct adding of metal nanoparticles into the titania sol or by adsorption of the metal-organic compounds onto the mesostructured titania films. The catalytic films properties, such as thickness, porosity, particle and pore sizes, and structure, were investigated by a number of techniques (TEM, SEM, ellipsometry, IR, XRD, XPS). The particle size in a nanostructured catalyst, the pore size and the thickness of the mesoporous support can be independently controlled via a one-pot evaporation induced self assembly synthesis. Titania films doped with Au or Pt-Sn nanoparticles were tested in a model reaction of citral hydrogenation into unsaturated alcohols (UCOL). Citral (3,7-dimethyl-2,6-octadienal) is considered as an interesting model molecule since it has three hydrogenation targets: a carbonyl group, a C=C bond conjugated with the carbonyl group, and an isolated C=C bond. However, the citral hydrogenation reaction attracts not only scientific, but also practical interest, since the hydrogenation is an important step in many fine chemical processes. The synthesis of mesoporous TiO2 and Au/TiO2 films is described in Chapter 2. The influence of synthesis parameters, such as the ratio of templating agent/Ti-precursor and pH of initial sol on the structure of the films was studied. The films have been investigated by a number of physico-chemical methods in order to characterize the properties of the mesoporous films. It was found that highly ordered hexagonal mesostructures can be obtained at a certain range of pH values (1.5-1.8) and template/Ti ratios (0.006-0.009). The thickness of the mesoporous titania films was increased from 300 to 1000 nm by a deposition of multi-layer coatings. Gold nanoparticles with an average size of 4.5 nm were embedded in the mesoporous titania films by condensation of metal oxide species via self-assembly in the presence of a known amount of stabilized gold colloids. The complete surfactant removal was achieved at 573 K under a residual pressure of 10 mbar within 4 hrs without destroying the film structure. Under these conditions the Ti4+ sites were reduced to Ti3+ which might be responsible for a high selectivity to UCOL in the hydrogenation of citral over Au-containing supported catalysts. A 300 nm thick Au-containing film showed an initial TOF of 1.4 s-1 and selectivity towards UCOL as high as 90% in the hydrogenation of citral. Thicker films demonstrated a high selectivity towards the saturated aldehyde (above 55%) and a lower intrinsic catalytic activity (initial TOF of 0.7-0.9 s-1) in the absence of internal diffusion limitations. A new approach for predicting the pore size distribution of mesoporous thin films using ellipsometric porosimetry is described in Chapter 3. The method has been developed taking into account multilayer adsorption and capillary condensation phenomena. The improved Derjaguin - Broekhoff - de Boer model (IDBdB) was applied for determination of the mesopore size on titania and silica films with a thickness of ca. 200 nm deposited on a silicon substrates. This approach eliminates uncertainties related to the application of the statistical film thickness determined via t-plots in previous versions of the DBdB model. The deviation in the surface tension of ethanol in the mesopores from that of a flat interface was described by the Tolman parameter. An empirical expression for the disjoining pressure isotherm has been applied to represent the interaction between the adsorbate and the adsorbent. The parameters in the empirical expression have been obtained by fitting the multilayer region of an experimental isotherm on titania and silica films. Using these parameters, the adsorption and desorption isotherms of ethanol were predicted and adjusted to the experimental data by fitting the Tolman parameter of the model, characterizing the surface tension for a curved surface. The IDBdB desorption model with a positive value of the Tolman parameter of 0.2 nm was found to describe accurately (±1%) the pore diameter in a wide range of mesopores from 2.1 to 8.3 nm what was in a good agreement with results obtained by TEM and XRD. The synthesis and characterization of Pt-Sn/TiO2 powder catalysts and catalytic films is presented in Chapter 4. The processing parameters, such as Pt-Sn precursor types, impregnation techniques, adsorption time, were varied. The resulting catalysts were tested in citral hydrogenation reaction. The selectivity to the desired product (UCOL) was found to be ca. 90% at a citral conversion above 95%. A reaction rate in terms of TOF of 0.2 – 3.3 min-1 was observed in the citral hydrogenation over the Pt-Sn/TiO2 powder catalysts. The highest activity was observed in the case of catalysts prepared by co-impregnation in comparison with the catalysts prepared via organometallic route and successive impregnation, what could be attributed to an intimate contact between the metals. The Pt-Sn/TiO2 films were prepared via i) one-pot synthesis using Pt-Sn colloids or clusters, and ii) adsorption of Pt-Sn clusters into the mesoporous titania films. The activity and selectivity of Pt-Sn/TiO2 films strongly depend on the precursor type and adsorption time (in case of clusters). The highest UCOL yield (72%) was observed in the case of a two layer catalytic film obtained from Pt-Sn clusters solution with adsorption time of 5 hrs. Chapter 5 is focused on the demonstration of the potential of capillary microreactors by the results obtained in a selective hydrogenation of citral to UCOL. The Au/TiO2 sol was impregnated by dip-coating into the fused silica capillary with a length of 10 m and internal diameter of 250 µm. A Pt-Sn/TiO2 wall-coated capillary was made by an impregnation of Pt-Sn clusters into the capillary preliminary coated with a titania layer. Both capillaries were tested in the citral hydrogenation reaction. The influence of the reaction conditions, such as liquid flow, reaction temperature, hydrogen partial pressure, on the catalytic properties was investigated. The detailed analysis of kinetics is presented. The Pt-Sn/TiO2 coated capillary shows better catalytic properties in comparison with the Au/TiO2 coated one. The maximal yield of UCOL was found to be 79 and 29% for Pt-Sn/TiO2 and Au/TiO2 catalysts, respectively. The reaction has first and zero order in hydrogen and citral, respectively. The synthesis and characterization of a new type of catalyst – ZnO nanowires doped with Pd nanoparticles – is presented in Chapter 6. The ZnO nanowires were made by the atomic layer deposition technique. Pd nanoparticles with average sizes of 2.7 and 4.5 nm were impregnated from the Pd colloids solutions. The catalysts were tested in the selective hydrogenation of 2-methyl-3-butyn-2-ol (MBY) to 2-methyl-3-buten-2-ol (MBE). The effect of the solvent, hydrogen pressure and reaction temperature on the MBE yield was investigated. ZnO nanowires possess 88% selectivity to MBE at 99% MBY conversion at 5 bar hydrogen pressure and 323K and show no deactivation after 20 runs of the reaction. The Pd/ZnO nanowires catalyst shows 94% yield of MBE in the first run of the reaction. Catalytic tests performed in water and in a methanol–water (3:1 vol.) mixture led to sintering of the Pd particles decreasing both the activity and selectivity. The ZnO NWs were rather stable in water which is promising for further optimization of their activity and stability using highly polar solvents. A kinetic model is proposed to explain activity and selectivity of the ZnO and Pd/ZnO nanowires catalysts. Finally, in Chapter 7 the main conclusions of the previous chapters are summarized.