Lucica Soare
The aim of this thesis was to synthesise and control a nanostructured composite of copper/cobalt. Both the copper and cobalt oxalate exhibit a nanostructure. Attempts to produce nanocomposite particles in the size range of 10-70 nm were made via an oxalate co-precipitation route followed by the appropriate thermal treatment to finally obtain the metal nanocomposite. Much attention was focussed on the understanding of the copper oxalate precipitation and its further transformation into metal before the investigation of the co-precipitation system. One major challenge of this thesis work was to achieve a better understanding of the copper oxalate precipitation mechanism. This was made by following kinetic parameters in order to shed light on the various steps of the precipitation from a supersaturated solution (nucleation, growth and aggregation of nanocrystals). Following the pH as a function of time and using the thermodynamic solubility data it was possible to propose a kinetic model of copper oxalate precipitation, with the co-precipitation of slight amounts (around 0.40% wt) of malachite (CuCO3·Cu(OH)2). The copper oxalate nanostructured particle growth mechanism, from the self-assembly of nanosized buildings blocks, was confirmed for intermediate precipitation times (1-15 minutes). Evidence for such organisation of the particles was shown by a combination of XRD diffraction, SEM and AFM measurements showing the presence of steps at the particle surface with a height that corresponds to a multiple of the mean crystallite size in that particular crystallographic orientation. Further investigations were performed for the early steps of the precipitation by SAXS but either the precipitated volume fraction was too low for detection of the particles or nucleation and growth kinetics were to fast (less than one second) to be followed. The TEM cross-section analyses showed a possible core-shell assembly mechanism. The core showed a random organisation of the crystallites with a size of around 25 nm, while the crystallite in the shell with a size of around 40 nm presented certain order along the 110 axis, particularly towards the particle surface. All these details provided the opportunity to propose a new and more detailed mechanism of the copper oxalate polycrystalline particle formation. In order to master the conditions of the Cu/Cu oxalate decomposition, a good understanding of the simple copper oxalate decomposition was necessary. All along this thesis, much attention was paid to the transformation of the copper oxalate cubic particles into the metal. The objective was thus to conserve the particles cubic macrostructure morphology and their internal nanoscale spatial organisation. To this goaltwo routes were investigated: a direct transformation and another, an indirect one, that required the formation of an oxide. Both the copper oxalate and the oxide showed an anisotropic behaviour during the transformation into the metal. It is shown experimentally that the anisotropy, nanostructure and inhomogeneity of the initial nanocrystallites of both the oxalate and the oxide have an important influence on the mechanism and evolution of transformation into the metal. The particle morphology was shown to be lost for a transformation yield of α>0.80 in the case of the direct transformation from the oxalate whereas the morphology was kept up to the metallic state when passing via the oxide. A kinetic model was proposed for both systems using the method of the sudden change in temperature and pressure. The kinetics analysis did not permit a total understanding of the transformations studied, as several stages were shown to have complex and concurrently competing mechanisms. However, a geometrical model was proposed using the ex-situ analysis of the samples as a function of the reaction yield for both routes. For the initial stages of the copper oxide reduction under He/H2 atmospheres, the kinetic analysis showed hydrogen dissociation as a rate-limiting step. With a view to producing a cobalt/copper composite, preliminary experiments were carried out for the co-precipitation of the Co/Cu oxalate with different cationic ratios. Thermodynamic calculations showed the formation of the two solids was possible independent of the ratio Co/Cu. Experimentally, however a co-precipitation was obtained only for a ratio Co/Cu of 1, whereas for other ration only one single phase (either copper oxalate or cobalt oxalate) was formed. A second route was investigated using cobalt oxalate or cobalt oxide seeds. The amount of the cobalt detected by TEM in the precipitate using either the oxalate or from the oxide seeds was around 3%wt, which is lower than the value of 10%wt necessary to provide desired magnetic properties in the resulting precipitate. The exact nature and spatial distribution of the cobalt was not ascertained and further analysis of the nanostructure by TEM needs to be carried out to confirm the premise of the seed route. This thesis has shown that it is possible to get to a deeper understanding of the kinetics and the mechanism of the copper oxalate precipitation using the appropriate techniques. The cubic macrostructure can be conserved from the initial CuO nanoparticles (13 nm) to metallic Cu (42 nm) by a controlled transformation in a reducing atmosphere. Preliminary experiments made on the possible formation of a Co/Cu composite via the use of the cobalt oxide seeds as heterogeneous nuclei for the copper oxalate precipitation seems most promising.
EPFL2004
