|dc.description.abstract||When metal nanoparticles are supported on the surface of an inert carrier, the resulting materials can become powerful heterogeneous catalysts capable of accelerating a diverse range of chemical reactions. Though there are a wide variety of synthetic routes by which these materials can be prepared, many offer little control over the textural properties of the resulting catalyst. Conversely, methods using pre-formed colloidal nanoparticles as a metal source offer the potential for precise control of properties like particle size and shape. This is both a topical class of syntheses, and the subject of the current body of work. Three different types of colloids were therefore prepared using different stabilising agents, and subsequently used to produce an array of different catalysts.
As support materials for the catalysts, a series of four commercially available materials were obtained, including silica, alumina, hydroxyapatite, and microcrystalline cellulose. Four novel analogues were also obtained, including halloysite, perlite, bioapatite, and macrofibrillar cellulose. In order to understand their physicochemical properties, they were characterised by a combination of Fourier transform infrared microspectroscopy, solid-state ¹H, ¹³C, ²⁷Al, ²⁹Si, and ³¹P nuclear magnetic resonance spectroscopy, x-ray diffraction, x-ray fluorescence spectroscopy, scanning and transmission electron microscopy, laser diffraction analysis, and zeta potential measurements. Additional experiments were performed to investigate the wetting properties of the materials using the Washburn capillary rise method, which was shown to be a valuable method for rationalising solvent choice for reactions involving these materials.
An initial batch of catalysts was produced by immobilising electrostatically-stabilised rhodium, palladium, and platinum colloids onto these support materials. Instrumental characterisation of the catalysts showed no major perturbation to the chemistry of the materials following the immobilisation, though some subtle surface effects were observed. These catalysts were tested in the hydrogenation of several olefins, and were found to have excellent catalytic activity, up to 64,400 hr⁻¹, under standard conditions.
A second batch of catalysts was produced by immobilising polymer-stabilised metal colloids onto the supports. Characterisation of the materials showed the polymers and the nanoparticles they contained were associated strongly with the support materials, though the spatial distribution varied significantly between systems. Activation of the catalysts by calcination was shown to initiate their thermal decomposition and oxidation, which in turn affected their catalytic activity. The catalysts containing supported rhodium or platinum were inactive, while those containing supported palladium showed good activity, up to 41,900 hr⁻¹.
A third batch of catalysts was produced by immobilising solvent-stabilised metal colloids onto the supports. Amide-type solvents were shown to be the most effective for this purpose, though spectroscopic evidence suggested the samples contained residual solvent and related decomposition products. These catalysts displayed modest activities of up to 8,100 hr⁻¹.
Finally, a method was developed for quantifying the amount of metal in the catalysts using laser-induced breakdown spectroscopy. A series of matrix-matched standards were prepared, and their elemental compositions measured accurately using neutron activation analysis. The method was shown to be rapid and inexpensive to implement, with linearity up to 2.0 wt% metal and a precision of over 95%.||