In this work 23Na MAS NMR was validated as a successful quantitative method for studies of exchanging sodium in bentonites useful, in particular, for studies of ion-exchange kinetics. Na-enriched bentonites equilibrated in a re-circulated process water at iron-oxide pelletizing plants may acquire properties of Ca-bentonites after already 20 minutes of the equilibration time, since >50 % of sodium ions will be exchanged by calcium ions during first minutes of bentonite placed in contact with the process water. It was shown that all sodium activated bentonites used in this study exchange >50% of sodium in Na+/Ca2+ and ca 20 % of sodium in binary Na+/Mg2+ systems with the same bentonite/solution ratio and same concentrations of these ions in aqueous solutions as in the process water at a pelletizing plant. In total, approximately 50 % of the exchangeable sodium in original bentonites was exchanged after equilibrating of bentonites in the process water already after 20 minutes. Experimental Na+/Ca2+ exchange curves for ‘model’ Ca2+(aq) solutions and for process water are very similar as Ca2+ is the dominant constituent in the process water. Since bivalent ions (Ca2+ and Mg2+) that present in the process water readily replace Na+ ions, Na-bentonite transforms into Ca- or Mg- bentonite, which have worse rheological, swelling and, therefore, binding properties. This ion-exchange process can influence the binder performance in the pelletizing process. Taking into account that fluorapatite is one of the components present in a blend of minerals processed, possible interactions between orthophosphate (the principal anionic component of apatites) and bentonites in aqueous suspensions are considered. It was found that sorption of orthophosphate on Ca-montmorillonite follows a different pattern from sorption of orthophosphate on aluminum oxides and kaolinite. While there is a small amount of sorption below pH 7, which may involve inner-sphere complexation and precipitation of AlPO4 to Al-OH edge sites on the montmorillonite crystals, most sorption of orthophosphate occurs at higher pH. Both macroscopic sorption measurements and solid-state 31P MAS NMR suggest that above pH 7 there is precipitation of proton depleted calcium phosphate phases. Based on both 31P chemical shifts and 31P chemical shift anisotropies it was concluded that the principal precipitated phased are most likely ‘brushite-like’ phases. Very short spin-lattice T2(31P) relaxation times (≤100 μs) for the orthophosphate/bentonite systems can possibly be explained by the presence of paramagnetic Fe in bentonites. Since there are insufficient concentrations of soluble Fe species in the supernatant solution that may give rise to the observed effects, it is likely that orthophosphate is precipitated as thin layers on the surfaces of montmorillonite crystals, where phosphorus may interact with Fe atoms present in the crystal lattice. PO4-tetrahedra in sorbed species can be also distorted giving rise to a larger 31P CSA than for pure ‘apatite-like phases’. 29Si MAS and 1H-29Si CP/MAS NMR experiments on bentonite samples also performed in this work provide information about impurities of quartz in bentonites, a level of substitution of aluminum by iron atoms in the structure of montmorillonite and about the degree of hydration of montmorillonite. 29Si NMR experiments on bentonite incubated with waterglass in aqueous suspensions at concentrations of sodium silicates as in the process water demonstrated that one can follow the process of polymerization of waterglass in solutions and also detect sodium silicates polymerized on surfaces of bentonites already after 1 hour of incubation. Polymerized waterglass sorbed on bentonite surfaces may also alter rheological, swelling and, therefore, binding properties of sodium-activated bentonites used in pelletization of iron-oxide ores.

L'objectif de ce travail de thèse est double : d'une part mesurer les propriétés mécaniques de couches hybrides à base d'organosiloxanes et de silice colloïdale et ensuite de lier ces caractéristiques mécaniques à leur structure.<br />La partie bibliographique est divisée en deux chapitres : le premier porte sur la description de la technique de nanoindentation et son utilisation pour obtenir les propriétés mécaniques des matériaux élastoplastiques. Le second est consacré aux matériaux hybrides organiques-inorganiques, à leur élaboration sous forme de couches et enfin à leur caractérisation structurale par spectroscopie dans le domaine de l'infrarouge et Résonance Magnétique Nucléaire du 29Si.<br />L'étude expérimentale est également divisée en deux chapitres : le chapitre I est consacré à l'élaboration des sols permettant la fabrication des couches et à la caractérisation structurale à la fois des sols (par RMN du 29Si en phase liquide) et des dépôts (par spectroscopie IR et RMN du 29Si en phase solide). Chaque système étudié est composé d'un organosiloxane différent R'nSi(OR)4-n et de silice colloïdale. Le chapitre II présente les caractérisations mécaniques par nanoindentation de ces différents systèmes qui sont analysées en relation avec leurs structures. Le but est ainsi de mettre en évidence plusieurs effets : celui de la fonctionnalité de l'alcoxyde, celui de la nature du groupement R' non polymérisable, l'effet de charge et d'encombrement stérique et enfin celui de la teneur et de la taille des particules de silice.

New complex silicon-containing molecules were made by B(C6F5)3-catalyzed hydrosilation, dehydrocoupling, and dealkylative coupling reactions starting from Si-H reagents. The scope of reactions starting from disilane was expanded to include the formation of silicon-sulfur1, silicon-oxygen and silicon-alkyl side-chains. Reaction inhibition was found with some heteroatom substrates, such as phenols and imines, that strongly bound to B(C6F5)3, and was consistent with the proposed mechanism (Chapter 2). B(C6F5)3 was found to be selective for Si-H activation in reactions of disilane and no competing Si-Si bond cleavage side-reactions were observed. This result will guide future studies and application of B(C6F5)3-catalyzed reactions with polysilanes. A different type of selectivity, the competing B(C6F5)3-catalyzed over-reduction, is evaluated and discussed in Chapter 3. This over-reduction reaction was classified into two distinct cases: alkyl groups for which over-reduction reaction was dependent on the steric bulk of the alkyl group and benzylic groups for which over-reduction was dependent on having an alpha-aryl group. These reactions are consistent with the proposed Piers-Oestreich mechanism (see Chapter 3) and suggest the rate-determining step for over-reduction is the nucleophilic attack of the alkoxysilane (R -O-SiR3) to the R3Si•••H•••B(C6F5)3 complex. Benzylic side-chains were over-reduced regardless of the steric bulk of the aryl groups. Literature precedents suggest that benzyl over-reductions must undergo an alternative mechanism to the Piers-Oestreich mechanism. A number of mechanisms have been proposed in the literature and in Chapter 3, suggesting conventional heteroatom substrate borane or silane-borane complexation. Furthermore, over-reduction of benzylic sulfur containing side-chains was found and this reaction was exploited in the B(C6F5)3-catalyzed synthesis of unique silicon-sulfur silicon-containing products. These over-reduction reactions highlighted the role of the silane for over-reduction and the challenges associated with the post-polymerization modification of poly(phenylsilane). The advances in B(C6F5)3-catalyzed synthesis of small silane molecules suggested reaction conditions and gave spectroscopic benchmarks that were applied to the post polymerization modification of poly(phenylsilane) (Chapter 4). New X-modified poly(phenylsilane) derivatives with thiolato (sulfur), alkoxy/aryloxy (oxygen), amido (nitrogen) and alkyl(carbon) side-chains were prepared with 10-40% incorporation of the ‘X’ group into poly(phenylsilane). These new polysilanes were characterized by the following methods: 1H/13C/29Si NMR, IR, MALS-GPC, EA, and UV-vis absorption spectroscopy. Together, these characterization methods showed that the polysilane had not undergone Si-Si cleavage and thus demonstrated the utility of B(C6F5)3 for the selective activation of Si-H bonds. Thermal decomposition of X-modified poly(phenylsilane) derivatives and parent poly(phenylsilane) showed interesting redistribution pathways (Chapter 5). The thermal decomposition products of poly(phenylsilane) were identified: volatile monosilanes, a structurally complex not-yet-identified phenylsilicon-containing material generated at 500 °C, and a mixture of silicon carbide (SiC) and elemental carbon generated at 800 °C. The B(C6F5)3-catalyzed post-polymerization method (Chapter 4) was evaluated based on the substitution percentage for X-functionalized poly(phenylsilane) derivatives. Reactions of highly electron-donating substrates gave a low amount of X incorporation (10%, e.g. aryloxy side-chains derived from phenol). Aryloxy groups were alternatively introduced via demethanative coupling, which gave a polymer with a greater substitution percentage (25%). The overall impact of the H-to-X substitution reactions was gauged by UV-vis absorption spectra and desirable UV absorption properties would require the modified poly(phenylsilane) to have a high degree of substitution. / Graduate / 2017-09-02